Exosome colorimetric detection method based on MnO2 / CeO2 heterojunction nano-enzyme

Abstract

The invention belongs to the technical field of crossing of nano materials, biosensing and medical detection, and particularly relates to an exosome colorimetric detection method based on MnO2 / CeO2 heterojunction nano enzyme. The MnO2 / CeO2 heterojunction structure is constructed, a stable Mn-O-Ce synergistic interface is formed, electron transfer and interface charge separation are effectively promoted, and the peroxidase-like catalytic ability of the material is improved. The color development reaction is completed through the paper-based platform, an HSV color space analysis system is integrated, and a numeralization result can be obtained by using a smart phone for photographing. Paper-based colorimetric analysis, an intelligent mobile phone auxiliary chromaticity identification system and micro-plate reader quantitative verification are combined. Compared with a traditional RGB method, the HSV model can more accurately reflect hue and brightness changes, the recognition stability and the anti-jamming capability are improved, and visual preliminary screening of high-risk samples in a non-professional environment is achieved.

Description

Technical Field

[0001] The present invention belongs to the interdisciplinary technical field of nanomaterials, biosensors and medical detection, and specifically relates to a colorimetric detection method for exosomes based on MnO2 / CeO2 heterojunction nanozymes. Background Art

[0002] In recent years, tumor liquid biopsy technology has gradually developed, and new biomarkers represented by circulating free DNA and exosomes in the blood have shown important application value in early disease screening. Exosomes are a type of nanoscale extracellular vesicles derived from cells, with a size of 30-150nm. They are widely present in body fluids such as blood, saliva, and urine, and can carry a variety of functional molecules including proteins, mRNA, and miRNA. Studies have shown that specific proteins (such as CD63, FGG, and FGB) in tumor-derived exosomes have clear diagnostic and typing value. Among them, exosomal proteins such as CD63, FGG, and FGB have shown high diagnostic value in distinguishing lung cancer from benign lesions.

[0003] On the other hand, artificial biomimetic nanozymes are a class of nanomaterials that possess the catalytic properties of natural enzymes. In particular, materials with peroxidase (POD)-like activity are widely used in colorimetric detection as an alternative to natural enzymes such as HRP. Some heterojunction nanozymes are believed to possess enhanced catalytic efficiency and stability due to their interfacial structure, which promotes electron transfer. However, achieving ultrasensitive and highly selective detection of exosomes in complex biological fluids remains a major technical challenge.

[0004] Currently, there are reports on the use of nanozymes combined with colorimetric substrates (such as TMB) to construct visual sensing platforms for preliminary screening of exosomes. This type of method generally involves the following steps: ① Preparation of a single-component nanomaterial with POD activity; ② Grafting with a capture molecule (such as an aptamer or antibody) to form a composite probe; ③ Adding the nanozyme to the exosome sample to be tested, catalyzing color development through substrate catalysis; ④ Using a spectrophotometer or mobile phone to capture the color change for qualitative or semi-quantitative analysis.

[0005] In addition, some studies have attempted to combine paper-based platforms with mobile phone image analysis (such as RGB reading) to improve the portability of testing. However, most of these studies have not incorporated standardized color model processing algorithms, still relying on manual interpretation, and most have not involved subsequent quantitative validation or the establishment of stratified diagnostic models.

[0006] Although existing technologies have achieved colorimetric detection of exosomes to a certain extent, they still have the following major shortcomings:

[0007] 1. The lack of high-performance nanozyme materials with optimized structural design results in limited catalytic efficiency and low detection sensitivity;

[0008] 2. Most methods remain at the "detection" level and lack a systematic closed-loop strategy of "initial screening + precise quantification + intelligent diagnosis";

[0009] 3. Image recognition is limited to RGB reading, without the introduction of HSV color space and standardization processing algorithm, making it impossible to achieve sample stratification judgment;

[0010] 4. The initial screening samples were not further quantified using a microplate reader, and a logical model (such as logistic regression) was not constructed to distinguish between benign and malignant nodules;

[0011] 5. The overall detection system is difficult to deploy and promote in primary medical institutions.

[0012] In summary, existing technologies are unable to take into account catalytic performance, visualization, initial screening efficiency and quantitative diagnostic capabilities. There is an urgent need for a new, integrated, and field-deployable detection system to meet the actual needs of early exosome stratified screening for lung cancer. Summary of the Invention

[0013] The present invention aims to overcome the shortcomings of the existing technology and has successfully developed a novel MnO2 / CeO2 heterojunction nanoenzyme colorimetric immunosensor for detecting plasma-derived exosomes, thereby accurately distinguishing benign from malignant pulmonary nodules. By combining the high oxidase-mimicking activity of the MnO2 / CeO2 heterojunction with specific recognition of the exosomal proteins CD63, FGG, and FGB, the proposed sensor exhibits excellent sensitivity (minimum detection limit of approximately 2 particles / μL), a wide detection range, high specificity, and good repeatability and reusability.

[0014] In order to achieve the above object, the present invention discloses the following technical solutions:

[0015] The first aspect of the present invention discloses a MnO2 / CeO2 composite nanozyme for exosome detection. MnO2 has a one-dimensional nanorod structure and CeO2 is an irregular polyhedral nanoparticle. After the two construct a heterojunction, a stable Mn-O-Ce interface bonding is formed.

[0016] The hydrothermal method is used to synthesize the MnO2 / CeO2 composite material. The specific steps are as follows:

[0017] (1) Dissolve Ce(NO3)3·6H2O and urea in 60 mL of deionized water to form a Ce source precursor solution;

[0018] (2) Dissolve MnSO4·H2O and KMnO4 in 30 mL of deionized water to form a Mn source solution;

[0019] (3) Ce source and Mn source were mixed in a volume ratio of 1:1, and then hydrothermally reacted at 160 °C for 24 h. After cooling, the mixture was centrifuged, washed, dried at 60 °C, and calcined at 300 °C for 3 h to obtain MnO2 / CeO2 composite nanozyme.

[0020] The second aspect of the present invention discloses a Glycine-MnO2 / CeO2@Apt bioconjugate, which is composed of the above-mentioned MnO2 / CeO2 composite nanozyme and glycine.

[0021] The specific preparation method of Glycine-MnO2 / CeO2@Apt bioconjugate is as follows:

[0022] (1) Aptamer activation: The carboxylated CD63 aptamer was reacted with EDC and NHS at 37°C for 1 hour to obtain an activated aptamer mixture;

[0023] (2) Preparation of Glycine-MnO2 / CeO2: Glycine was dissolved in DDW, stirred at room temperature, and the MnO2 / CeO2 composite nanozyme solution was added dropwise, and stirring was continued for 24 hours to obtain the Glycine-MnO2 / CeO2 solution;

[0024] (3) Bioconjugation: Glycine-MnO2 / CeO2 solution was added to the activated aptamer mixture and incubated at 37°C for 1 hour;

[0025] (4) Purification: Centrifuge at 16099×g for 10 min, retain the precipitate, wash with DDW and resuspend to obtain Glycine-MnO2 / CeO2@Apt.

[0026] The third aspect of the present invention discloses the use of the above-mentioned Glycine-MnO2 / CeO2@Apt bioconjugate for detecting exosomes in a paper-based colorimetric sensor platform.

[0027] Preferably, the paper-based colorimetric sensor platform uses hydrophilic filter paper as a substrate, wherein a three-channel exosome detection system is constructed, which is loaded with anti-CD63, anti-FGG, and anti-FGB antibodies, respectively, combined with chitosan modification and glutaraldehyde cross-linking fixation.

[0028] The construction and detection methods of the paper-based colorimetric sensor platform are as follows:

[0029] (1) 0.1 mL of glacial acetic acid was dissolved in 9.9 mL of deionized water to prepare a 1% acetic acid solution, and 1 g of chitosan was added to prepare a 1% chitosan solution. 50% glutaraldehyde stock solution was diluted with deionized water to prepare a 2.5% glutaraldehyde solution. 4 μL of 1% chitosan was added dropwise to the surface of the printed carbon electrode and dried at room temperature. Subsequently, 6 μL of 2.5% glutaraldehyde solution was added dropwise and incubated at room temperature for 4 hours.

[0030] (2) After rinsing the electrode surface with PBS, 4 μL of 1 mg / mL anti-FGG antibody, anti-FGB antibody, and anti-CD63 antibody were added dropwise in sequence and incubated overnight in a humidified environment at 4°C. After incubation, the surface was washed with PBS, and then 1% BSA was added to block nonspecific sites. The surface was incubated at room temperature for 1 hour and then washed again.

[0031] (3) Add different concentrations of exosomes and incubate at 37°C for 90 minutes. The antibodies recognize the corresponding proteins and capture the exosomes. The unbound part is washed with PBS. Then, 6 μL of Glycine-MnO2 / CeO2@Apt bioconjugate is added dropwise, incubated at 37°C for 60 minutes, and washed with PBS. 5 mM TMB substrate (pH 3.5) is prepared with acetic acid-sodium acetate buffer, 50 μL of substrate solution is added, and the reaction is carried out in the dark for 10 minutes.

[0032] (4) The solution was transferred to a 96-well plate and 50 μL of 2 M sulfuric acid was added to terminate the reaction. After the reaction, the color of the solution was identified by colorimetry, and suspected positive samples were detected using an enzyme-labeled instrument.

[0033] Preferably, the specific operation of the colorimetric identification is:

[0034] (1) Take photos of the color-developed area of ​​the paper base, extract the RGB value of the colorimetric channel and convert it into an HSV model;

[0035] (2) HSV value conversion uses the OpenCV standard library to obtain the Hue, Saturation, and Value components of each detection point;

[0036] (3) Preliminary screening with V value (brightness) as the main indicator;

[0037] (4) The system has a built-in standardization algorithm to eliminate interference from light and background colors;

[0038] (5) If the V value is lower than the preset threshold, the system will output a “high risk warning”, mark it as a suspected positive sample, and guide quantitative analysis.

[0039] Preferably, the specific steps of the quantitative analysis are as follows:

[0040] The suspected positive samples were transferred to a 96-well plate, and TMB reaction system was added to the 96-well plate. After 10 minutes, 2M H2SO4 was added to terminate the reaction, and the absorbance was measured at a wavelength of 450 nm.

[0041] The absorbance values ​​correspond to the three targets CD63, FGG, and FGB;

[0042] Construct a logistic regression model, use the OD values ​​of the three channels as independent variables, input the OD values ​​of the CD63, FGG, and FGB channels into the logistic regression model, obtain the regression coefficient and model prediction probability; output the benign or malignant prediction result of the sample;

[0043] The training samples in the Logistic regression model are clinical plasma samples.

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

[0045] (1) Heterojunction interface structure significantly enhances the catalytic performance of nanozymes

[0046] By constructing a MnO2 / CeO2 heterojunction structure, a stable Mn-O-Ce synergistic interface is formed, which effectively promotes electron transfer and interfacial charge separation, thereby improving the peroxidase-like catalytic ability of the material. Experimental data show that the catalytic activity of MnO2 / CeO2 reaches 63.19U / μg, which is significantly higher than that of MnO2 (35.63U / μg) and CeO2 (10.76U / μg). Its catalytic reaction rate for TMB is 1.93μmol / (L·min), which is 1.55 times and 8.04 times that of single-component MnO2 and CeO2. This improvement is attributed to the introduction of a synergistic electron transfer pathway in the composite structure, which increases the reaction active sites.

[0047] (2) Paper-based colorimetry + image recognition system realizes portable primary screening

[0048] Unlike traditional colorimetric detection, which relies on a microplate reader or spectrophotometer, this method uses a paper-based platform to develop the color reaction and integrates an HSV color space analysis system, allowing for digital results to be obtained simply by taking a photo with a smartphone. This method combines paper-based colorimetric analysis with a smartphone-assisted colorimetric recognition system and quantitative verification using a microplate reader. The HSV model more accurately reflects hue and brightness variations than traditional RGB methods, improving recognition stability and anti-interference capabilities, enabling visual initial screening of high-risk samples in non-professional settings.

[0049] (3) Logic gate judgment + enzyme reader verification to build an intelligent hierarchical screening process

[0050] Traditional detection platforms are mostly single-channel and linear in response, making it difficult to provide stratified diagnostic information. This paper-based detection system introduces a logic gate screening mechanism for the first time, setting an HSV-V threshold to identify "high-risk" samples, thus avoiding misjudgments and wasted resources. High-risk samples are then accurately quantified using an enzyme-linked microplate reader for absorbance. Finally, a logistic regression model is used to integrate multi-channel signals to achieve classification of benign and malignant lung nodules. The model validation AUC was 0.996, demonstrating excellent diagnostic efficacy.

[0051] (4) The platform is low-cost and easy to promote, and is particularly suitable for primary medical scenarios

[0052] Compared with detection methods that rely on electrochemical equipment or complex separation processes, the present invention only requires filter paper, nanozyme solution, TMB and a smartphone to complete the initial screening. The equipment is simplified, the operation is easy, and the materials used are cheap and stable. It is particularly suitable for primary medical institutions with limited resources and weak detection capabilities to carry out early exosome screening for lung cancer.

[0053] (5) Wide detection range and high sensitivity

[0054] The minimum detection limit of this platform for exosomes is 2 particles / μL, and the detection linear range is 102-108 particles / μL, which can meet the screening needs of samples at different stages. It also supports the joint detection of three markers: CD63, FGG, and FGB, improving specificity while ensuring sensitivity. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1A Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of CeO2, MnO2, and MnO2 / CeO2 composites are shown.

[0056] Figure 1B Elemental mapping analysis corresponding to the prepared samples (CeO2, MnO2, and MnO2 / CeO2 composites) is shown: Ce (yellow), O (red), and Mn (green).

[0057] Figure 1B E1, E2, and E3 show the intensity distribution diagrams of the content of each element in CeO2, MnO2, and MnO2 / CeO2 composite materials.

[0058] Figure 2 A is the X-ray diffraction spectrum (XRD) of CeO2, MnO2 and MnO2 / CeO2 composite materials; Figure 2 B in the middle is Fourier transform infrared spectroscopy (FTIR); Figure 2 CD are nitrogen adsorption isotherm and pore size distribution, respectively; Figure 2 EF in the middle is the high-resolution XPS spectrum of MnO2; Figure 2 GH in the middle is the high-resolution XPS spectrum of CeO2; Figure 2 IK in the figure is the high-resolution XPS spectrum of the MnO2 / CeO2 composite material; Figure 2 L in the figure is the selected area electron diffraction (SAED) pattern of the MnO2 / CeO2 composite material.

[0059] Figure 3Middle AC shows that CeO2, MnO2 and MnO2 / CeO2 composites catalyze TMB to blue products with maximum absorption wavelengths ranging from 330 to 780 nm; Figure 3 Middle D shows the color changes of the prepared samples after incubation with CeO2, MnO2 and MnO2 / CeO2 composites at different concentrations (25 μg / mL, 50 μg / mL, 100 μg / mL, 500 μg / mL, 1 mg / mL and 2 mg / mL); Figure 3 E in the middle is the quantitative analysis of POD-like activity of the prepared samples; Figure 3 F in the middle is the time-dependent absorbance of oxidized TMB of the synthesized samples (CeO2, MnO2, and MnO2 / CeO2 composites) at 652 nm; Figure 3 G in the middle is the POD reaction rate of the synthesized nanozymes (CeO2, MnO2 and MnO2 / CeO2 composites) under different concentrations of TMB; Figure 3 H is the steady-state kinetic measurement of the synthesized samples (CeO2, MnO2 and MnO2 / CeO2 composites) with TMB.

[0060] Figure 4 A in the figure is the cyclic voltammetry analysis of the electrochemical behavior of electrodes modified with different materials; Figure 4 B is the electrochemical impedance spectra of electrodes modified with different materials; Figure 4 C in the middle is the UV-visible absorption spectrum characterizing the gradual modification process of the sensor; Figure 4 Figure 2D shows the time-absorbance curves of different materials (a: bare SPE; b: bare SPE / chitosan; c: bare SPE / chitosan / glutaraldehyde; d: bare SPE / chitosan / glutaraldehyde / anti-FGG antibody; e: bare SPE / chitosan / glutaraldehyde / anti-FGG antibody / BSA; f: bare SPE / chitosan / glutaraldehyde / anti-FGG antibody / BSA / exosomes; g: bare SPE / chitosan / glutaraldehyde / anti-FGG antibody / BSA / exosomes / Gly-MnO2 / CeO2@Apt).

[0061] Figure 5 Figures AB are fluorescent immunocytochemical staining of FGG and FGB proteins in normal lung epithelial cells (BEAS-2B) and lung cancer cells (A549) seeded on coverslips. DAPI (blue) stains the cell nuclei, FGG (green), and FGB (red). Merge (merged image after DAPI, FGG, and FGB staining). Figure 5 Middle C is Western blot analysis of FGB (Ca) and FGG (Cb) proteins in exosomes isolated from two normal lung cell lines (BEAS-2B and NHLF) and three lung cancer cell lines (A549, H1299, and PC9); Figure 5Middle D is the transmission electron microscopy (TEM) image of BEAS-2B, NHLF, A549, H1299, PC9, and human plasma-derived exosomes; Figure 5 Figure E shows nanoparticle tracking analysis (NTA) of the size distribution of exosomes from BEAS-2B, NHLF, A549, H1299, PC9, and human plasma.

[0062] Figure 6 Middle A shows the HSV-V values ​​of CD63, FGG, and FGB markers in plasma samples of benign (n=50) and malignant (n=60) lung nodules; Figure 6 Middle B shows the smartphone-based color analysis results: (a) negative samples above the threshold, (b) high-risk samples with at least one V value below the threshold; Figure 6 Middle C is the logic gate diagram for high-risk classification.

[0063] Figure 7 Figure A shows the clinical applicability of the colorimetric sensor developed using CD63 as an exosome marker in differentiating benign and malignant lung nodules, revealing a statistically significant difference between malignant and benign samples (P<0.0001), with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.833. Figure 7 Figure B shows the clinical applicability evaluation of the colorimetric sensor developed using FGG as a marker in distinguishing benign and malignant lung nodules, revealing a statistically significant difference between malignant and benign samples (P<0.0001) with an AUC of 0.874. Figure 7 Figure C shows the clinical applicability evaluation of the colorimetric sensor developed using FGB as a marker in distinguishing benign and malignant lung nodules, revealing a statistically significant difference between malignant and benign samples (P<0.0001) with an AUC of 0.881.

[0064] Figure 8 Center A shows the heat map of the absorbance values ​​of CD63, FGG, and FGB in the training cohort (30 benign cases and 30 malignant cases), and the ROC curve of the logistic regression model in the training cohort, with an AUC of 0.987 (P < 0.0001); Figure 8 Middle B is the absorbance value heat map of CD63, FGG, and FGB in the validation cohort (67 benign cases and 67 malignant cases) and the ROC curve of the logistic regression model in the validation cohort, with an AUC of 0.996 (P<0.0001).

[0065] Figure 9 A in the middle shows the preparation of Glycine-MnO2 / CeO2@Apt bioconjugate; Figure 9 The establishment of a screening platform for B-type display; Figure 9 Center C shows an antibody-functionalized screen-printed carbon electrode; Figure 9Middle D shows the construction of a lung nodule characteristic diagnostic model by combined detection of exosomal proteins.

[0066] Figure 10 This is an enlarged view of MnO2 / CeO2 and the electron diffraction (SAED) pattern of the corresponding area of ​​the MnO2 / CeO2 composite material.

[0067] Figure 11 Optimized data for sensor construction conditions. Figure 11 Center A shows the incubation time of exosomes with capture antibodies. Figure 11 Center B shows the incubation temperature of exosomes. Figure 11 Center C shows the concentration of Apt-CD63. Figure 11 Middle D shows the incubation time of exosomes with glycine-MnO2 / CeO2@Apt bioconjugate.

[0068] Figure 12 To study the sensitivity of the sensor, different concentrations (10 2 , 10 3 , 10 4 , 10 5 , 10 6 , 10 7 and 10 8 Particles / μL) of A549 cell-derived exosomes were analyzed separately ( Figure 12 A and B are CD63 proteins), ( Figure 12 C and D are FGG proteins), and ( Figure 12 Figures E and F represent FGB proteins. Error bars are calculated based on three replicate experiments. The inset shows the color change of the constructed sensor under natural light when detecting different exosome concentrations. The wavelength range was set from 320 nm to 600 nm, with a wavelength interval of 10 nm.

[0069] Figure 13 To study the selectivity, reproducibility and reusability of colorimetric sensors. Figure 13 Center A shows a selectivity test using potential interfering substances (glucose, bovine serum albumin, immunoglobulin G); Figure 13 Middle B is displayed at 10 5 Reproducibility evaluation of 15 independently prepared sensors at the same particle / μL concentration; Figure 13 Center C shows a 21-day reusability test. DETAILED DESCRIPTION

[0070] The technical solutions of the present invention are described in detail below with reference to the accompanying drawings and examples, but the present invention is not limited to the scope of the examples. Experimental methods and techniques in the following examples where specific conditions are not specified are generally performed in accordance with conventional conditions in the art or the conditions recommended by the manufacturer.

[0071] Reagents and materials

[0072] The CD63-specific aptamer sequence used for exosome detection is:

[0073] 5'-COOH-CACCCCACCTCGCTCCCGTGACACTAATGCTA-3' was synthesized by Shanghai Shenggong Bioengineering Co., Ltd. Potassium permanganate (KMnO4), cerium oxide hexahydrate (CeO2·6H2O), chitosan, and glutaraldehyde were purchased from Aladdin Reagent Co., Ltd. 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) and N-hydroxysuccinimide sodium salt (NHS) were purchased from MacLean Biochemical Technology Co., Ltd. FGB and FGG antibodies were purchased from Affinity Biosciences. A plasma exosome extraction kit was purchased from Dalian Meilun Biotechnology Co., Ltd. Cerium nitrate (Ce(NO3)3·6H2O), urea (CO(NH2)2), potassium permanganate (KMnO4), manganese sulfate (MnSO4·H2O), potassium hydroxide (KOH), anhydrous ethanol (analytical grade), and conductive graphite (chemical grade) were purchased from Sinopharm Chemical Reagent Co., Ltd. All solutions were prepared using double distilled water (DDW).

[0074] Instruments and Equipment

[0075] All tumor cells were in New BrunswickTM Cells were cultured in a 170S cell culture incubator (Germany). Exosome transmission electron microscopy (TEM) imaging was performed using a Hitachi-7650 transmission electron microscope (Japan). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed on a Bio-Rad electrophoresis system, and images were acquired using a Bio-Rad ChemiDoc™ MP imaging system (USA). All electrochemical measurements were performed on a CHI760E electrochemical workstation (China). Photometry was performed using a Thermo Varioskan TM LUX multifunctional microplate reader (USA). The crystal structure of the sample was tested by Bruker D8 high-speed X-ray diffractometer (XRD) (CuKa radiation Tube voltage 40 kV, current 40 mA, scan range 10–80°). Fourier transform infrared spectroscopy (FTIR) was performed using a Bruker Tensor 27 spectrometer (4000–400 cm -1, KBr as dispersant). The morphology and microstructure of the materials were observed using a Hitachi S-4800II field emission scanning electron microscope (FESEM, 20 kV), a Philips TECNAI 12TEM, and a FEI Tecnal G2F30 high-resolution transmission electron microscope (HRTEM). Energy dispersive spectroscopy (EDS) and elemental distribution maps were collected using HRTEM. The specific surface area and pore structure parameters of the samples were measured using a TriStar 3000 nitrogen adsorption analyzer. X-ray photoelectron spectroscopy (XPS) was performed using a Thermo Fisher Scientific PHI5700ESCA (USA) to test the structure-activity relationship of the samples.

[0076] Example 1

[0077] 1. Preparation of CeO2

[0078] 0.26 g of cerium nitrate (Ce(NO₃)₃·6H₂O) and 2.25 g of urea (CO(NH₂)₂) were dissolved in 60 mL of deionized water and stirred thoroughly. The mixture was then transferred to a stainless steel autoclave and reacted at 160°C for 24 hours. After the reaction, the mixture was cooled to room temperature. The white precipitate was separated by centrifugation (7104 × g) and washed sequentially with deionized water and anhydrous ethanol until the supernatant was colorless and transparent. The resulting product was dried at 60°C for 12 hours, ground in an agate mortar, and calcined in a muffle furnace at 300°C for 3 hours to obtain a light yellow powder.

[0079] 2. Preparation of MnO2

[0080] Dissolve 0.448 g of manganese sulfate (MnSO4·H2O) and 1 g of potassium permanganate (KMnO4) in 30 mL of deionized water respectively, and slowly pour the MnSO4·H2O solution into the KMnO4 solution. The subsequent operations are consistent with the CeO2 preparation process.

[0081] 3. Preparation of MnO2 / CeO2

[0082] Mix 30 mL of cerium nitrate and urea mixture with 30 mL of manganese sulfate and potassium permanganate suspension. The subsequent operations are the same as those for the preparation of CeO2. The obtained sample is named MnO2 / CeO2. The other raw material ratios and preparation processes remain the same.

[0083] 4. Synthesis of Glycine-MnO2 / CeO2@Apt Bioconjugate

[0084] (1) Aptamer activation: Carboxylated CD63 aptamer (1.25 μM) was reacted with EDC (400 mM) and NHS (100 mM) at 37°C for 1 h;

[0085] (2) Preparation of Glycine-MnO2 / CeO2: Dissolve 5 mg of glycine in 20 mL of DDW, stir at room temperature, add 10 mL of MnO2 / CeO2 solution (0.12 mg / mL) dropwise, and continue stirring for 24 h.

[0086] (3) Bioconjugation: 200 μL of Glycine-MnO2 / CeO2 solution was added to the activated aptamer mixture and incubated at 37°C for 1 h;

[0087] (4) Purification: Centrifuge at 16099 × g for 10 min, retain the precipitate, wash with DDW and resuspend to obtain Glycine-MnO2 / CeO2@Apt for subsequent experiments.

[0088] 5. Material Characterization

[0089] 5.1 Microstructure morphology and element distribution

[0090] Figure 1A The following are scanning electron microscope (SEM) and transmission electron microscope (TEM) images of the sample. CeO2 exhibits an irregular morphology, appearing as aggregates of various spherical, ellipsoidal and rod-shaped particles. Some of these particles range from nanoscale to submicron scale, with regular shapes, such as cubes or cuboids, and relatively smooth surfaces. Combined with the TEM image of CeO2, it can be seen that there are certain pores or spaces between the particles, which is beneficial to improving the specific surface area and capacitance performance of CeO2. MnO2 is in the shape of regular one-dimensional nanorods with a diameter of about 80nm, different lengths and a smooth surface. However, due to its high surface tension and high surface energy, the MnO2 nanorods are severely agglomerated, and multiple nanorods aggregate together, significantly reducing the specific surface area of ​​the sample.

[0091] In addition, CeO2, MnO2 and MnO2 / CeO2 samples were observed by high-resolution transmission electron microscopy (HRTEM). Figure 1A As shown in Figure 2, CeO2 nanoparticles successfully grew along the surface of MnO2 nanorods and were evenly distributed. The distribution of oxygen in the three samples (CeO2, MnO2, and MnO2 / CeO2) was studied by elemental mapping and energy dispersive spectroscopy (EDS). Figure 1B The results show that Ce element is evenly distributed in CeO2 and MnO2 / CeO2 samples ( Figure 1B D1, D3, F1 and F3 in ), while Mn element was also detected in MnO2 and MnO2 / CeO2 samples ( Figure 1B D2, D3, F2 and F3 in the .

[0092] Figure 1AThe scanning electron microscope (SEM) and transmission electron microscope (TEM) images of CeO2, MnO2 and MnO2 / CeO2 composite materials are shown in Figure 2. Figure 1A A1, A2, A3, B1, B2, and B3 are SEM images. Figure 1A C1, C2, and C3 are TEM images;

[0093] Figure 1B D1, D2, and D3 are the element mapping analyses corresponding to the prepared samples (CeO2, MnO2, and MnO2 / CeO2 composite materials): Ce (yellow), O (red), and Mn (green); Figure 1B E1, E2, and E3 are the intensity distribution diagrams of the element contents in CeO2, MnO2, and MnO2 / CeO2 composite materials, respectively.

[0094] 5.2 Crystal structure, functional groups and pore structure characteristics

[0095] like Figure 2 As shown in Figure A, the diffraction peaks appearing in the XRD pattern of CeO2 correspond to the (111), (220) and (311) crystal planes, respectively, which can well match the standard CeO2 diffraction pattern (JCPDS 81-0792), indicating that the prepared sample is a fluorite-type CeO2. The CeO2 has a face-centered cubic structure and belongs to the Fm3m space group. The diffraction peak intensity is high and the peak shape is sharp, indicating that the CeO2 has good crystallinity. No obvious diffraction peaks of Ce(OH)CO3 (orthorhombic JCPDS: 41-0013 or hexagonal) are observed in the spectrum.

[0096] JCPDS 32-0189), indicating that the Ce(OH)CO3 content in CeO2 samples is extremely low.

[0097] The diffraction peaks of MnO2 appear at 2θ = 12.8, 18.2, 25.8, 28.9, 36.8, 37.6, 41.3, 42.0, 49.8, 56.4, 60.3 and 69.7. Their peaks are sharp and have high intensity, indicating that it is a typical α-MnO2 with good crystallinity (JCPDS card number: 44-0141). At the same time, the XRD pattern of the MnO2 / CeO2 hydrothermal composite is almost the superposition effect of CeO2 and MnO2, indicating that the two still retain their respective crystal structures after composite. However, it is obvious that the diffraction peaks appear broadened and the intensity weakened after composite, indicating that the composite material tends to be amorphous; the diffraction angle shifts toward the low angle direction, indicating that the grains are refined and the crystallinity is reduced.

[0098] Figure 2The FTIR spectrum of the composite material in Figure B further verifies the successful composite of CeO2 and MnO2, because it has the characteristic absorption peaks of two single components at the same time. For CeO2 and its composite, 3250–3600 cm -1 The broad weak absorption band in the range is attributed to the stretching vibration of hydroxyl (-OH). -1 The obvious absorption peak at 2- Ion v3 vibration (triplet absorption peak); and at 1064.5, 846.6 and 727.5 cm -1 The other weak peaks that appear are CO3 2- The FTIR spectrum of CeO2 shows that the hydrothermal synthesis sample contains a small amount of -OH and CO3 2- , which is mainly attributed to the presence of Ce(OH)CO3. These results indicate that the prepared samples (CeO2, MnO2 and MnO2 / CeO2) were successfully prepared.

[0099] Figure 2 Figure C shows the N2 adsorption-desorption isotherms of the samples. The isotherms of CeO2 and MnO2 are between type I and type IV, indicating that their pore sizes are mainly distributed in the micropore and small pore range. Figure 2 The pore size distribution curves (D) also show that both materials are primarily composed of micropores and mesopores smaller than 5 nm. The weak adsorption / desorption hysteresis loops indicate a limited number of mesopores. When the relative pressure (P / P0) exceeds 0.8, the N2 adsorption capacity increases significantly, indicating that CeO2 and MnO2 contain a certain number of macropores. Figure 2 Figure D also shows that CeO2 and MnO2 have fewer macropores with diameters larger than 15 nm. The N2 adsorption capacity of MnO2 is significantly lower than that of CeO2, indicating its lower porosity, which is closely related to its severe aggregation and smooth one-dimensional nanorod structure.

[0100] The MnO2 / CeO2 hydrothermal composite exhibits typical type IV isotherms, with N2 adsorption capacity far exceeding that of the individual components. At low relative pressures, the adsorption capacity increases slowly, with a subtle bump at the beginning of the adsorption branch, indicating a low micropore content. The N2 adsorption capacity increases linearly with increasing P / P0. When P / P0 > 0.75, a significant hysteresis loop appears between the adsorption and desorption branches, attributed to capillary condensation of nitrogen in the mesopores. Figure 2 The pore size distribution curve of the D-composite is relatively wide, indicating that it is not only rich in small pores <5 nm, but also has a high content of macropores of 15–45 nm, presenting a typical hierarchical porous structure, which is suitable as a sensing material.

[0101] The specific surface area and pore structure parameters of the samples are summarized in Table 1. The BET specific surface area of ​​CeO2 is 45.689 m 2 / g, MnO2 is only 20.876m 2 / g, while the BET of MnO2 / CeO2 composite material reaches 59.264m 2 / g, which is much higher than that of a single component. The mesoporous structure is conducive to the rapid migration of protons, thereby improving the rate performance and power characteristics of the material. The mesopore diameter (Dp) and total pore volume (Vp) of MnO2 / CeO2 are also significantly improved, reaching 15.89nm and 0.22cm respectively. 3 / g, compared with MnO2 (17.31nm, 0.08cm 3 / g) and CeO2(21.94nm,0.04cm 3 / g) were improved, indicating that the hydrothermal compounding effectively improved the pore structure and stacking state of the particles.

[0102] The MnO2 / CeO2 sample has the largest BET, Dp, and Vp, making it the optimal sensor material. These results are highly consistent with the SEM and TEM analyses.

[0103] Table 1 Structural parameters of CeO2, MnO2 and MnO2 / CeO2

[0104]

[0105] In order to further reveal the structure-activity relationship of CeO2, MnO2 and MnO2 / CeO2, X-ray photoelectron spectroscopy (XPS) tests were carried out. Figure 2 The Mn 2p spectra in E, F, and K show that the Mn species exhibits multiple oxidation states, corresponding to binding energies of 644.7 eV and 641.92 eV, respectively. 4+ and Mn 3+ For Ce 3d spectra see Figure 2 All samples can be decomposed into eight peaks, corresponding to Ce 3+ and Ce 4+ That is, the formation of MnO2 / CeO2 heterojunction creates active Mn-O-Ce bonds and through the “Mn 3+ +Ce 4+ →Mn 4+ +Ce 3+ The oxygen reduction reaction enhances the electron transfer between Mn and Ce atoms. This conjecture can be verified by the slight shift of the peaks at approximately 888.2eV, 897.9eV and 907.1eV in the Ce 3d spectrum of the MnO2 / CeO2 catalyst, indicating that the chemical environment of the Ce atoms has changed. Therefore, it is reasonable to speculate that some Mn atoms have entered the lattice structure of Ce oxide.

[0106] All the above results show that the abundant Mn on the surface of MnO2 / CeO2 4+ and Ce 3+ ions contribute to its excellent catalytic activity. Subsequently, the selected area electron diffraction (SAED) pattern ( Figure 2 Zhong L and Figure 10 ) showed good crystallinity of the synthesized samples (CeO2, MnO2 and MnO2 / CeO2), indicating the presence of high levels of metal content particles.

[0107] 5.3 Evaluation of POD-like catalytic activity and kinetic behavior

[0108] Figure 3 Figures AC show the results of POD-like catalytic activity. As shown in the figure, compared with CeO2 and MnO2, the characteristic absorption peak in the UV-Vis spectrum of the MnO2 / CeO2 composite material decreases rapidly, and the intensity of the absorption peak gradually weakens with the increase of sample concentration. As the sample concentration increases, the color of TMB disappears, indicating that all three groups (CeO2, MnO2 and MnO2 / CeO2 composite material) can effectively catalyze the conversion of TMB molecules ( Figure 3 According to the definition of enzyme activity, the POD-like activity value of MnO2 / CeO2 was calculated to be 63.19 U / μg, while that of MnO2 and CeO2 was 35.63 U / μg and 10.76 U / μg, respectively. This indicates that the catalytic activity of the MnO2 / CeO2 composite material is approximately 1.77 times that of MnO2 and 5.87 times that of CeO2 ( Figure 3 In addition, after 60 minutes of reaction, the absorbance of the blue product generated by TMB catalyzed by MnO2 / CeO2 was 2.33, while that of MnO2 and CeO2 was 1.50 and 0.16, respectively, which were 1.55 times and 14.56 times that of MnO2 / CeO2 ( Figure 3 F). The reaction rate of TMB oxidation catalyzed by the prepared nanozyme was further calculated. The results showed that when the reactant concentration was 100 μg / mL, the reaction rate of MnO2 / CeO2 was 1.93 μmol / (L×min -1 ), while the reaction rates of MnO2 and CeO2 were 1.24 μmol / (L×min -1 ) and 0.24 μmol / (L×min -1 ), which are 1.55 times and 8.04 times that of MnO2 and CeO2 respectively ( Figure 3 In addition, the POD-like activity of MnO2 / CeO2 exhibited typical Michaelis-Menten kinetics ( Figure 3The Km and Vmax values ​​further indicated that the POD activity of MnO2 / CeO2 was higher than that of MnO2 and CeO2 (Table S1). All the above results indicate that MnO2 / CeO2 has higher POD-like activity and a stronger ability to catalyze the oxidation of substrates than MnO2 and CeO2.

[0109] Example 2 Cell culture and exosome extraction

[0110] 1. Cell culture

[0111] A549 cells were cultured in F12K complete medium; BEAS-2B, PC9, and NHLF cells were cultured in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin; and H1299 cells were cultured in 1640 complete medium. All cell lines were cultured at 37°C in a humidified atmosphere with 5% CO2.

[0112] 2. Isolation of exosomes from cell culture supernatant

[0113] The cells were cultured in exosome-depleted FBS medium for 48 hours, and the supernatant was collected: centrifuged at 300×g for 10 minutes to remove dead cells; centrifuged at 2000×g for 10 minutes to remove cell debris; filtered through a 0.22μm filter to remove large vesicles; and centrifuged at 100,000×g for 70 minutes to collect exosomes. The pellet was resuspended in 100μL PBS. save.

[0114] 3. Extraction of exosomes from plasma samples

[0115] Blood samples were centrifuged at 3000 × g for 10 min to obtain cell-free plasma;

[0116] Add 0.5 mL of PBS and shake to mix, then add 0.5 mL of proteinase K and incubate at 37°C for 10 min; then add 0.4 mL of exosome extraction reagent and incubate at 2-8°C for 30 min; finally, centrifuge at 4°C, 10,000 × g for 5 min, collect the exosome precipitate, resuspend in PBS and save.

[0117] Example 3 Analysis based on HSV color space

[0118] This paper-based detection platform was developed and implemented using the WeChat mini-program "High-Precision Color Recognition System" to analyze HSV color space. Combining computer vision with mathematical modeling, the platform collects color data from camera or uploaded images in real time, converts RGB to HSV space using the OpenCV library, and accurately extracts hue (H), saturation (S), and value (V) components using normalization and geometric decomposition algorithms.

[0119] Example 4 Sensor Construction

[0120] Prepare a 1% acetic acid solution by dissolving 0.1 mL of glacial acetic acid in 9.9 mL of deionized water. Add 1 g of chitosan to prepare a 1% chitosan solution. Dilute a 50% glutaraldehyde stock solution with deionized water to prepare a 2.5% glutaraldehyde solution. Add 4 μL of the 1% chitosan solution dropwise to the surface of the printed carbon electrode (SPE) and allow to dry at room temperature. Then, add 6 μL of the 2.5% glutaraldehyde solution dropwise and incubate at room temperature for 4 hours.

[0121] After rinsing the electrode surface with PBS, 4 μL of 1 mg / mL anti-FGG antibody, anti-FGB antibody, and anti-CD63 antibody were added dropwise, followed by overnight incubation at 4°C in a humidified atmosphere. After incubation, the surface was washed with PBS, followed by addition of 1% BSA to block nonspecific sites. The surface was incubated at room temperature for 1 hour and then washed again.

[0122] Exosomes at varying concentrations were added and incubated at 37°C for 90 minutes. Antibodies recognized the corresponding proteins and captured the exosomes. Unbound portions were then washed with PBS. Six microliters of Glycine-MnO2 / CeO2@Apt bioconjugate was then added dropwise, incubated at 37°C for 60 minutes, and washed with PBS. A 5 mM TMB substrate (pH 3.5) was prepared in acetic acid-sodium acetate buffer, and 50 microliters of the substrate solution was added. The reaction was allowed to proceed in the dark for 10 minutes, during which time the solution turned from colorless to blue.

[0123] The blue solution was transferred to a 96-well plate and the reaction was terminated by adding 50 μL of 2M sulfuric acid, which caused the color to change from blue to yellow. Preliminary results were read using a smartphone deep learning interface, and suspected positive samples were further tested using a microplate reader.

[0124] 1. Characterization of the electrochemical behavior of electrode modified materials by cyclic voltammetry

[0125] The electrochemical behavior of electrodes modified with different materials was studied using cyclic voltammetry (CV). Figure 4 In A), such as Figure 4 As shown in Figure A (using an anti-FGG antibody-modified electrode as an example), after chitosan was modified on a bare screen-printed electrode (SPE), the current intensity increased, which was attributed to the excellent conductivity of chitosan. Similarly, after treatment with glutaraldehyde, the current intensity was further enhanced, which is due to the good conductivity and solubility of glutaraldehyde. Subsequently, when anti-FGG antibodies were modified on the glutaraldehyde-treated chitosan electrode, the peak current decreased, which can be attributed to the poor conductivity of the protein. Similarly, the subsequent binding of BSA and exosomes to the anti-FGG antibody / chitosan modified electrode surface caused the peak current to gradually decrease. These results indicate that the antibodies and exosomes were successfully immobilized on the chitosan-modified SPE surface.

[0126] When the Gly-MnO2 / CeO2@Apt bioconjugate was further modified on the exosome / anti-FGG antibody / chitosan-modified SPE, a further decrease in the peak current was observed, indicating that the bioconjugate was successfully bound to the exosomes and subsequently immobilized on the electrode surface. The figure also shows the electrochemical impedance spectroscopy (EIS) results, where a larger semicircle diameter corresponds to a higher resistance and lower conductivity ( Figure 4 Middle (B). The semicircle diameter gradually decreases after the electrode surface is modified with chitosan and glutaraldehyde. Further modification with antibodies, BSA, and exosomes increases the diameter. Finally, after incubation with the Gly-MnO2 / CeO2@Apt bioconjugate, the diameter decreases again. These trends are consistent with the CV results, further confirming the successful stepwise construction of the sensor.

[0127] 2. Stepper sensor construction verified by UV-visible absorption spectroscopy

[0128] UV-visible absorption spectroscopy was used to characterize the step-by-step construction of the sensor ( Figure 4 (C) 50 μL of a 5 mM TMB solution was added to the electrode surface and allowed to react in the dark at room temperature for 10 minutes. The reaction was then terminated by the addition of 50 μL of 2 M sulfuric acid. The absorbance at 450 nm of the electrodes modified with different materials was measured. As shown in the figure, no significant color change or absorbance change was observed after the addition of TMB to the electrodes: bare SPE, chitosan-modified SPE, bare SPE / chitosan / glutaraldehyde, bare SPE / chitosan / glutaraldehyde / anti-FGG antibody, bare SPE / chitosan / glutaraldehyde / anti-FGG antibody / BSA, and bare SPE / chitosan / glutaraldehyde / anti-FGG antibody / BSA / exosomes. In contrast, the electrode modified with the Gly-MnO2 / CeO2@Apt bioconjugate exhibited a blue color change, then a bright yellow color, upon addition of sulfuric acid, with strong absorbance detected by a microplate reader.

[0129] In addition, after applying 50 μL of 5 mM TMB solution, the change in absorbance at 630 nm was monitored for 3 minutes ( Figure 4 (D). As shown in the figure, the previously modified electrode showed no significant absorbance change. However, the solution on the Gly-MnO2 / CeO2@Apt bioconjugate-modified electrode exhibited a gradual color change from colorless to light blue to dark blue over 3 minutes, accompanied by a simultaneous increase in absorbance. These results confirm that the color change of TMB is due to the oxidase-like catalytic activity of MnO2 / CeO2, which is absent in other components such as antibodies and chitosan.

[0130] 3. Cell Verification

[0131] Since the colorimetric sensor is based on the specific recognition and capture of FGG and FGB proteins, cell validation experiments were performed to confirm the presence of these proteins in exosomes derived from lung cancer cells. Cell climbing experiments were performed using A549 (lung cancer cells) and BEAS-2B (normal bronchial epithelial cells). Figure 5 The results showed that the expression levels of FGG and FGB proteins in A549 cells were significantly higher than those in BEAS-2B cells.

[0132] Subsequently, Western blot analysis was performed to detect FGG and FGB proteins in exosomes from two normal lung cell lines and three lung cancer cell lines ( Figure 5 Middle C). The expression levels of FGG and FGB proteins in exosomes from normal lung cell lines were significantly lower than those in exosomes from lung cancer cell lines. These findings suggest that FGG and FGB proteins may serve as potential biomarkers for detecting exosomes derived from lung cancer.

[0133] Furthermore, based on previous studies that found CD63, a known exosome marker protein, to be significantly upregulated in lung cancer exosomes, CD63 protein was also included in the sensor design in addition to FGG and FGB to improve the sensitivity and accuracy of the colorimetric sensor.

[0134] The exosomes used in the present invention were further characterized by physical methods. Transmission electron microscopy (TEM) images ( Figure 5 Middle D) shows that exosomes present a typical cup-shaped vesicle morphology. Nanoparticle tracking analysis (NTA) results ( Figure 5 Middle E) shows that the size distribution of exosomes ranges from 30 to 150 nm, confirming the successful isolation of exosomes in the present invention.

[0135] 4. Optimization of detection conditions

[0136] The sensor is based on a sandwich-type immunocomplex consisting of an exosome-specific capture antibody immobilized on chitosan, exosomes, and a glycine-MnO2 / CeO2@Apt bioconjugate. These bioconjugates have dual functions: (1) mimicking oxidase activity to catalyze the colorimetric oxidation of TMB; and (2) acting as nanocarriers to immobilize the aptamer. To improve detection sensitivity and shorten detection time, key parameters in the detection process were systematically optimized.

[0137] In order to determine the optimal incubation time of exosomes with capture antibodies (using anti-FGG modified electrode as an example), we monitored the absorbance changes of TMB solution at different incubation times. Figure 11As shown in Figure 1A, the absorbance gradually increases with incubation time, likely due to the gradual binding of more exosomes to the capture antibody, thereby recruiting more glycine-MnO2 / CeO2@Apt bioconjugates. The absorbance peaks at 90 minutes, indicating saturation of binding, with no further increase. Therefore, 90 minutes was selected as the optimal incubation time for exosomes with the capture antibody.

[0138] Temperature is another key factor affecting the performance of the sensor. While keeping other conditions unchanged, exosomes were incubated at different temperatures. The absorbance increased with increasing temperature and reached a peak at 37°C ( Figure 11 (B) This is likely due to the optimal enzymatic activity and stability of key biomolecules, such as transport proteins and enzymes, at physiological temperatures. Temperatures exceeding or falling below these temperatures can lead to protein denaturation or enzyme inactivation. Therefore, 37°C was selected as the optimal incubation temperature.

[0139] The concentration of aptamer Apt-CD63 also affects the detection performance. Figure 11 As shown in Figure C, when the Apt-CD63 concentration exceeded 1.25 μM, the absorbance signal decreased. This may be because the excessive aptamer concentration interfered with the binding efficiency of the aptamer and glycine-MnO2 / CeO2 to exosomes. Therefore, 1.25 μM was determined to be the optimal concentration of Apt-CD63.

[0140] Finally, the incubation time between glycine-MnO2 / CeO2@Apt bioconjugate and exosomes was optimized. The absorbance increased with the extension of incubation time and reached the maximum value at 60 min ( Figure 11 Middle (D), indicating that the bioconjugate has fully bound to the exosomes captured on the electrode. Beyond 60 minutes, the signal decreases, likely due to nonspecific interactions or signal instability. Therefore, 60 minutes was selected as the optimal incubation time for exosomes with the glycine-MnO2 / CeO2@Apt bioconjugate.

[0141] 5. Research on the detection performance, specificity, repeatability and reproducibility of the sensor

[0142] Under the optimized experimental conditions, the absorbance response of the chromatographic sensor to different concentrations of exosomes was evaluated (e.g. Figure 12 As shown). Using A549 cell-derived exosomes as a model, a linear relationship was observed between the exosome concentration (from 102 to 108 particles / μL) and the absorbance at 450 nm. As shown in the figure, when CD63 was used as a specific exosome marker, the linear regression equation was A=0.2533logC+0.0171, and the correlation coefficient R 2=0.9953, where A represents the absorbance at 450 nm and C represents the concentration of A549-derived exosomes. Similarly, when FGG was used as an exosome marker, the regression equation was A=0.2581logC–0.0642(R 2 =0.9934), while for FGB, a linear relationship was also observed, and its equation is A=

[0143] 0.3053logC–0.3006(R 2 =0.9997).

[0144] The limits of detection (LOD) for the CD63-, FGG-, and FGB-targeted sensors were 2.10 particles / μL, 2.15 particles / μL, and 2.00 particles / μL, respectively. The LOD calculation was based on the 3σ / k criterion, where σ represents the standard deviation of blank measurements (n=3) and k is the slope of the corresponding calibration curve. This ensures reliable detection at a 99% confidence level, effectively distinguishing signal from background noise.

[0145] Furthermore, the detection performance of the proposed chromatographic sensor was compared with that of previously reported sensors. As shown in Table 2, the proposed sensor exhibited a wider detection range, which could be attributed to the high oxidase-mimicking activity of MnO2 / CeO2, which enabled the efficient catalysis of the oxidation of TMB to oxTMB.

[0146] Table 2 Comparison of different colorimetric sensors for exosome detection

[0147]

[0148] 6. Research on sensor selectivity, repeatability and regeneration

[0149] In order to evaluate the selectivity of the sensor, potential interfering substances that often coexist with exosomes in plasma, such as IgG, glucose, and BSA, were selected as negative control experiments (using the anti-FGG antibody modified electrode as an example). Figure 13 As shown in Figure 3A, the absorbance signals generated by these interfering substances are negligible, indicating that the developed sensor has excellent selectivity for exosomes.

[0150] To evaluate the reproducibility of the sensor, 15 chromatographic sensors (based on anti-FGG antibody-modified electrodes) were independently prepared using the same procedure and tested at a concentration of 10 5 Particles / μL of exosomes were tested. Figure 13 As shown in B, the relative standard deviation (RSD) of the absorbance values ​​of different batches was 2.25%, indicating that the sensor had good repeatability.

[0151] To study the regenerative properties of the sensor, chitosan was first drop-coated on the SPE surface, then treated with glutaraldehyde and subsequently cross-linked with anti-FGG antibody to form a capture antibody / chitosan modified electrode. The modified electrode was stored in PBS buffer at 4°C. The electrode was washed every three days and then specifically bound to BSA, exosomes, and glycine-MnO2 / CeO2@Apt bioconjugates in sequence. Figure 13 As shown in Figure 3C, after 21 days of storage, the absorbance signal retained 86.36% of the original value, indicating that the sensor has good regeneration ability.

[0152] 7. Screening of high-risk groups based on a high-precision color recognition system

[0153] In order to verify the positive threshold of the high-precision color recognition system, the present invention collected 110 plasma samples from patients with pathologically confirmed pulmonary nodules in Shanghai Pudong Hospital (50 of which were benign and 60 were malignant) ( Figure 6 Exosomes were isolated from all samples using a commercial plasma exosome isolation kit, followed by quantitative hue-saturation-value (HSV) analysis using a self-developed color recognition system. This analysis can be performed using the WeChat mini-program "High-Precision Chromaticity Recognition System." Each sample was tested three times to minimize technical variation, and the average value was used for threshold determination ( Figure 6 Middle B).

[0154] In the benign group, the mean HSV-V values ​​(relative standard deviation) of CD63, FGG and FGB were 25.3 (6.12%), 24.6 (4.27%) and 25.0 (5.88%), respectively. In contrast, the corresponding values ​​in the malignant group were significantly lower, at 18.7 (5.35%), 19.2 (6.40%) and 19.5 (7.13%), respectively. Based on the benign group data, the abnormal lower limit threshold was defined as the mean minus two times the standard deviation (SD = mean × RSD), covering a 95% confidence interval. The calculated thresholds for CD63, FGG and FGB were 22.2, 22.5 and 22.1, respectively. Using the "OR" logic rule, if the value of any marker was lower than its corresponding threshold, the sample would be classified as high-risk lung cancer. The preliminary positive samples were further quantitatively verified by microplate reading to enhance the reliability of the results, and a joint diagnostic model was constructed based on the two-stage data ( Figure 6 Middle C).

[0155] The present invention adopts a dual strategy approach, combining deep learning-assisted smartphone imaging for preliminary interpretation of colorimetric signals on paper-based chips, followed by quantitative confirmation by a microplate reader. The paper strip platform enables rapid and low-cost screening, making it well-suited for large-scale testing, especially in primary care settings, where specialized equipment such as microplate readers may not be available. While microplate readers provide high throughput and precise optical measurements, their limited availability in primary care facilities may limit their use alone for early screening. Therefore, the combination of smartphone-assisted preliminary screening with microplate confirmation strikes an effective balance between accessibility, efficiency, and diagnostic accuracy, providing a scalable and layered solution to aid in the early detection and prevention of lung cancer.

[0156] 8. Clinical applicability assessment

[0157] To evaluate the practical clinical applicability of the developed colorimetric sensor (based on a microplate reader), the study collected 60 clinical plasma samples from Shanghai Pudong Hospital, including 30 patients with pathologically confirmed benign pulmonary nodules and 30 patients with malignant pulmonary nodules. Exosomes were isolated from all plasma samples using a commercial exosome isolation kit and analyzed for absorbance using the established colorimetric sensor.

[0158] like Figure 7 As shown in Figure A, when CD63 protein was used as a specific exosome marker, the absorbance value of the malignant nodule group was significantly higher than that of the benign nodule group (P<0.0001). Receiver operating characteristic (ROC) curve analysis further confirmed the diagnostic accuracy of the sensor, with an area under the curve (AUC) of 0.833.

[0159] Similarly, when FGG protein was used as an exosome marker, the absorbance value of the malignant group was significantly higher than that of the benign group (P<0.0001), and the AUC was 0.874, indicating that the sensor had strong diagnostic performance ( Figure 7 Middle B).

[0160] The detection of FGB protein as a target marker also showed a significant difference in absorbance between the malignant group and the benign group (P<0.0001), and the AUC of the ROC curve was 0.881, showing a high discrimination ability ( Figure 7 Middle C).

[0161] These results suggest that CD63, FGG, and FGB proteins in plasma-derived exosomes have the potential to serve as biomarkers for distinguishing benign from malignant pulmonary nodules. However, despite their promising diagnostic performance, they performed comparably among the three markers and did not have exceptionally high diagnostic performance when used alone.

[0162] 9. Construction of a diagnostic model for lung nodules based on combined exosome protein detection

[0163] To address the limited diagnostic performance of a single biomarker, we used the absorbance data of CD63, FGG, and FGB proteins in plasma-derived exosomes from clinical samples (the 60 clinical plasma samples mentioned above served as the training set) and performed a logistic regression analysis (Table 3). The results showed that all three markers made significant contributions in distinguishing malignant from benign pulmonary nodules.

[0164] Table 3 Binary Logistic regression analysis results of the effects of CD63, FGG, and FGB on pulmonary nodule characteristics

[0165]

[0166] Results showed that the regression coefficients (B) for CD63, FGG, and FGB were 6.449, 5.788, and 8.333, respectively, indicating that elevated levels of these markers were significantly associated with the likelihood of malignant nodules. FGB had the strongest effect, with an odds ratio (Exp(B)) of 4157.649, indicating that the probability of malignant nodules increased approximately 4158-fold for each unit increase in FGB level. Similarly, CD63 and FGG also exhibited strong effects, with Exp(B) values ​​of 632.205 and 326.465, respectively. All variables were statistically significant (P < 0.05), further confirming their strong association with malignant pulmonary nodules.

[0167] Based on these results, a logistic regression model for the diagnosis of pulmonary nodules based on plasma exosomal protein markers was established. The regression equation is as follows (P = probability of malignant nodules):

[0168]

[0169] like Figure 8 As shown in center A, the logistic regression model based on exosomal CD63, FGG, and FGB levels performed well in the training cohort with an AUC of 0.987 (P < 0.0001) ( Figure 8 This model has high accuracy in distinguishing benign from malignant pulmonary nodules. However, further validation is needed to avoid the potential risk of overfitting.

[0170] To evaluate the generalizability of the model, 134 clinical plasma samples (67 benign cases and 67 malignant cases) from Shanghai Pudong Hospital were collected as a validation cohort. After exosome separation, the samples were analyzed using the developed colorimetric sensor, and the results were input into the established model. Figure 8 As shown in B, the AUC of the model in the validation cohort was 0.996 (P < 0.0001) ( Figure 8 The 95% confidence interval was 0.797 to 1.05, with a sensitivity of 98.51% and a specificity of 95.52%. These results demonstrate that the model has excellent diagnostic accuracy and strong generalization ability, highlighting its potential for clinical translation.

[0171] Example 5

[0172] The present invention constructs a four-stage detection process ( Figure 9 First, a paper-based colorimetric sensing platform enables captured exosomes to rapidly react with MnO2 / CeO2 heterojunction nanozymes in situ. Second, a smartphone-based HSV color space analysis system captures and processes color changes in real time. Third, a logic gate algorithm classifies samples with hue values ​​below a preset threshold as "high risk." Finally, these high-risk samples undergo quantitative absorbance testing using a microplate reader for further confirmation. Subsequently, a logic model is used to integrate the absorbance data, improving classification accuracy based on multiple markers.

[0173] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention shall be construed as equivalent.

[0174] Any effective replacement method also belongs to the protection scope of the present invention and is included in the protection scope of the present invention.

Claims

1. A MnO2 / CeO2 composite nanozyme for exosome detection, characterized in that: MnO2 presents a one-dimensional nanorod structure, and CeO2 is an irregular polyhedral nanoparticle. After the two construct a heterojunction, a stable Mn-O-Ce interface bonding is formed.

2. The MnO2 / CeO2 composite nanozyme for exosome detection according to claim 1, characterized in that: The MnO2 / CeO2 composite material was synthesized by hydrothermal method. The specific steps are as follows: (1) Dissolve Ce(NO3)3·6H2O and urea in 60 mL of deionized water to form a Ce source precursor solution; (2) Dissolve MnSO4·H2O and KMnO4 in 30 mL of deionized water to form a Mn source solution; (3) Ce source and Mn source were mixed in a volume ratio of 1:1, and then subjected to hydrothermal reaction at 160 °C for 24 h. After cooling, the mixture was centrifuged, washed, dried at 60 °C, and calcined at 300 °C for 3 h to obtain MnO2 / CeO2 composite nanozyme.

3. A Glycine-MnO2 / CeO2@Apt bioconjugate, characterized in that: The nanoenzyme is composed of the MnO2 / CeO2 composite nanozyme according to claim 1 or 2 and glycine.

4. The Glycine-MnO2 / CeO2@Apt bioconjugate according to claim 3, wherein: The specific preparation method of Glycine-MnO2 / CeO2@Apt bioconjugate is as follows: (1) Aptamer activation: The carboxylated CD63 aptamer was reacted with EDC and NHS at 37°C for 1 hour to obtain an activated aptamer mixture; (2) Preparation of Glycine-MnO2 / CeO2: Glycine was dissolved in DDW, stirred at room temperature, and the MnO2 / CeO2 composite nanozyme solution was added dropwise, and the stirring was continued for 24 hours to obtain the Glycine-MnO2 / CeO2 solution; (3) Bioconjugation: Add the Glycine-MnO2 / CeO2 solution to the activated aptamer mixture and incubate at 37°C for 1 hour; (4) Purification: Centrifuge at 16099×g for 10 min, retain the precipitate, wash with DDW and resuspend to obtain Glycine-MnO2 / CeO2@Apt.

5. Use of the MnO2 / CeO2 composite nanozyme according to claim 1 or 2, or the Glycine-MnO2 / CeO2@Apt bioconjugate according to any one of claims 3-4 for detecting exosomes in a paper-based colorimetric sensor platform.

6. The use according to claim 5, characterized in that The paper-based colorimetric sensor platform uses hydrophilic filter paper as the substrate, in which a three-channel exosome detection system is constructed, which is loaded with anti-CD63, anti-FGG, and anti-FGB antibodies, respectively, combined with chitosan modification and glutaraldehyde cross-linking fixation.

7. The use according to claim 6, characterized in that The construction and detection methods of the paper-based colorimetric sensor platform are as follows: (1) Prepare 1% acetic acid solution by dissolving 0.1 mL of glacial acetic acid in 9.9 mL of deionized water, and add 1 g of chitosan to prepare 1% chitosan solution; dilute 50% glutaraldehyde stock solution with deionized water to prepare 2.5% glutaraldehyde solution; add 4 μL of 1% chitosan to the surface of the printed carbon electrode and dry it at room temperature, then add 6 μL of 2.5% glutaraldehyde solution and incubate at room temperature for 4 hours; (2) After rinsing the electrode surface with PBS, 4 μL of 1 mg / mL anti-FGG antibody, anti-FGB antibody, and anti-CD63 antibody were added dropwise in sequence and incubated overnight in a humidified environment at 4 °C; after incubation, the surface was washed with PBS, and then 1% BSA was added to block nonspecific sites. The surface was incubated at room temperature for 1 hour and then washed again; (3) Exosomes of different concentrations were added and incubated at 37 °C for 90 min. The antibodies recognized the corresponding proteins and captured the exosomes. The unbound part was washed with PBS. Then 6 μL of Glycine-MnO2 / CeO2@Apt bioconjugate was added dropwise and incubated at 37 °C for 60 min. The mixture was washed with PBS. 5 mM TMB substrate (pH 3.5) was prepared with acetic acid-sodium acetate buffer, 50 μL of substrate solution was added, and the mixture was reacted in the dark for 10 min. (4) Transfer the solution to a 96-well plate and add 50 μL of 2 M sulfuric acid to terminate the reaction. After the reaction is completed, the color of the solution is identified by colorimetry, and suspected positive samples are detected using an enzyme-labeled instrument.

8. The use according to claim 7, characterized in that The specific operation of the colorimetric identification is: (1) Take photos of the color-developed area of ​​the paper base, extract the RGB value of the colorimetric channel and convert it into an HSV model; (2) HSV value conversion uses the OpenCV standard library to obtain the Hue, Saturation, and Value components of each detection point; (3) Conduct preliminary screening using V value (brightness) as the main indicator; (4) The system has a built-in standardization algorithm to eliminate interference from light and background colors; (5) If the V value is lower than the preset threshold, the system will output a "high risk prompt", mark it as a suspected positive sample and guide quantitative analysis.

9. The use according to claim 8, characterized in that The specific steps of the quantitative analysis are as follows: The suspected positive samples were transferred to a 96-well plate, and the TMB reaction system was added to the 96-well plate. After 10 minutes, 2 MH2SO4 was added to terminate the reaction, and the absorbance was measured at a wavelength of 450 nm. The absorbance values ​​correspond to the three targets CD63, FGG, and FGB; Construct a logistic regression model, use the OD values ​​of the three channels as independent variables, input the OD values ​​of the CD63, FGG, and FGB channels into the logistic regression model, obtain the regression coefficient and model prediction probability; output the benign or malignant prediction result of the sample; The training samples in the Logistic regression model are clinical plasma samples.

Images