A high-throughput, automated electrochemically addressed sensor array and methods and systems for screening anticancer drugs
By constructing a high-throughput electrochemical addressing sensor array and machine learning algorithms, the high cost and low efficiency of anticancer drug screening were solved, enabling rapid and accurate drug screening, reducing systematic errors, and providing support for the development of novel anticancer drugs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUBEI UNIV
- Filing Date
- 2023-09-28
- Publication Date
- 2026-06-02
Smart Images
Figure CN117288822B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of intelligent high-throughput electrochemical sensors and anticancer drug screening, specifically involving a high-throughput, automated electrochemical addressing sensor array and a method and system for screening anticancer drugs. Background Technology
[0002] Despite significant breakthroughs in the treatment of malignant tumors in clinical medicine, cancer remains the leading cause of death worldwide, necessitating the development of novel and highly effective anticancer drugs to reduce cancer mortality. However, anticancer drug development primarily involves stages such as candidate drug selection, preclinical research, clinical trials, and post-marketing surveillance, all of which are characterized by high risk, long duration, and high costs (Cancer Discov. 2021, 11, 822-837). Therefore, improving the efficiency and accuracy of drug screening in preclinical studies is crucial for the development of novel anticancer drugs. Traditional drug screening techniques typically rely on complex and costly animal models to assess the efficacy and toxicity of candidate drugs, which severely limits the throughput of drug screening. As an alternative strategy, using low-cost and readily available cancer cells as targets to evaluate drug efficacy has attracted widespread attention. However, common cell-based drug screening methods, such as MTT assays, CCK-8 assays, cell transfection, or target sequencing, are all endpoint assays, providing only limited static data on cellular responses and pharmacokinetics. Furthermore, these cell-based optical drug screening strategies suffer from drawbacks such as cumbersome procedures (cell staining and DMSO dissolution of formazan) and long processing times, which significantly limit drug screening efficiency. Therefore, the development of simple, efficient, and dynamic cell sensing technologies is urgently needed to achieve high-throughput and efficient screening of anticancer drugs.
[0003] Electrochemical sensors offer advantages such as simple equipment, low cost, ease of operation, high sensitivity, rapid response, and ease of automation, leading to their widespread application in in vitro and in vivo detection. Currently, researchers have reported several biocompatible electrochemical sensing systems capable of in-situ, non-invasive, and label-free screening of certain cancer cells or anticancer drugs (Adv. Mater. 2018, 30, 1707442). However, these methods primarily rely on single-channel electrochemical sensing electrodes and significant manual labor, making them unsuitable for high-throughput, accurate screening of anticancer drugs. Summary of the Invention
[0004] The technical problem this invention aims to solve is to propose a high-throughput, automated electrochemical addressing sensor array and its method for screening anticancer drugs, addressing the shortcomings of the existing technology, so as to achieve efficient and accurate screening of different types of anticancer drugs.
[0005] The technical solution adopted by the present invention to solve the above-mentioned problems is as follows:
[0006] The main steps of fabricating a high-throughput electrochemically addressed sensing array are as follows:
[0007] (1) Laser-induced graphene electrode film (LIG electrode film) was prepared on polyimide (PI) tape by laser engraving;
[0008] (2) A microporous array is constructed on a polymer tape (such as non-conductive polymers like polyvinyl chloride) by laser cutting, and the microporous array is then attached and assembled onto the surface of the LIG electrode film to obtain a LIG-based electrode array with micropores.
[0009] (3) Modify titanium dioxide (TiO2) single crystals into the micropores of the LIG-based electrode array to obtain a LIG-TiO2 microporous electrode array;
[0010] (4) Platinum (Pt) nanoparticles were uniformly deposited in the micropores of the LIG-TiO2 microporous electrode array by constant potential deposition to obtain the LIG-TiO2-Pt microporous electrode array.
[0011] (5) After sterilizing the LIG-TiO2-Pt microporous electrode array by ultraviolet irradiation, laminin (Lam) was modified into its micropores to obtain the LIG-TiO2-Pt-Lam microporous electrode array.
[0012] (6) Using the LIG-TiO2-Pt-Lam microporous electrode array as the working electrode, it is a high-throughput electrochemical addressing sensor array that can be assembled with the reference electrode and the counter electrode to form a high-throughput electrochemical addressing sensor for high-throughput screening of anticancer drugs.
[0013] The high-throughput electrochemically addressed sensing array described above uses multiple micropores to incubate tumor cells of the same or different types. Candidate anticancer drugs are then added for incubation, and phorbol ester (PMA) is added to stimulate the tumor cells to release H2O2. The survival status of these cancer cells can be monitored in situ based on the transient current response (i.e., the electrical addressing pulse signal) generated in each micropore that is linearly related to the H2O2 concentration therein, thereby achieving rapid and high-throughput screening of anticancer drugs.
[0014] According to the above scheme, in step (1), the laser power is 3-5W, and the laser engraving depth, i.e., the thickness of the LIG electrode film, is 15-25μm. Polyimide tape has excellent heat resistance, excellent mechanical properties, good dielectric properties, good chemical stability, and resistance to damp heat.
[0015] According to the above scheme, in step (3), the single crystal titanium dioxide (TiO2) is a rhomboid or spindle-shaped nanoparticle with a size generally in the range of 50 to 80 nm.
[0016] According to the above scheme, in step (2), the micro-hole array is generally set with N rows × M columns, where N and M are both positive integers, generally not less than 2. The diameter of each micro-hole is 2 to 4 mm, and the spacing between micro-holes is 0.5 to 1.5 mm. The depth of the micro-holes is 0.15 to 0.3 mm, which is the same as the height of the micro-hole array. Among them, when preparing the micro-hole array by laser cutting, the laser power is set to 8 to 9 W.
[0017] According to the above scheme, the specific process of step (3) is as follows: TiO2 single crystals are dispersed in ultrapure water and ultrasonically treated to obtain a uniform TiO2 dispersion with a concentration of 0.4-1 mg / mL. -1 Then, 5-10 μL of TiO2 dispersion was drop-coated into the micropores of the LIG-based electrode array and dried to obtain the LIG-TiO2 microporous electrode array.
[0018] According to the above scheme, the specific process of step (4) is as follows: the above LIG-TiO2 microporous electrode array is placed in a chloroplatinic acid (H2PtCl6) solution, and Pt nanoparticles are uniformly modified into the micropores of the LIG-TiO2 microporous electrode array by a constant potential deposition method. After rinsing with ultrapure water and drying, the LIG-TiO2-Pt microporous electrode array is obtained. Among them, the pH of the H2PtCl6 solution is 2-5 (pH adjusted by HCl), the concentration of H2PtCl6 is 3-9 mM; the applied potential for electrodeposition is -0.5-0.1 V (vs Ag / AgCl), and the electrodeposition time is 1-10 min.
[0019] According to the above scheme, the specific process of step (5) is as follows: Lam solution is added dropwise into the micropores of the LIG-TiO2-Pt microporous electrode array after ultraviolet sterilization, and incubated at room temperature for 80-100 min; then, excess Lam is washed with phosphate buffered saline (PBS) to obtain the LIG-TiO2-Pt-Lam microporous electrode array; wherein, the Lam solution is prepared using PBS with a concentration of 30-50 μg / mL. -1 Add 5–10 μL to each microwell.
[0020] The main steps of the high-throughput electrochemically addressed sensor array-based method for screening anticancer drugs are as follows:
[0021] A three-electrode system was used to drop tumor cell (Tcell) suspensions into multiple microwells of the high-throughput electrochemical addressing sensor array in parallel. The cells were incubated in a constant temperature and humidity incubator at 37±1℃ for 2–6 h. Then, candidate anticancer drugs were added and incubated at 37℃ for 4–12 h under constant temperature and humidity conditions of 5%±0.5% CO2 and 95%±0.5% Air (volume fraction). The microwells were then rinsed with PBS solution. PMA solution was added to stimulate the tumor cells to produce H2O2. The generated electro-addressing pulse signal in each microwell was linearly correlated with the H2O2 concentration, indicating that the survival rate of tumor cells in each microwell was related to the electro-addressing pulse signal, reflecting the inhibitory level of the candidate anticancer drugs on tumor cells. The electro-addressing pulse signal was collected using an electrochemical workstation, thus achieving rapid and high-throughput screening of anticancer drugs.
[0022] In the above method for screening anticancer drugs, the tumor cell suspension is 1×10 5 Tcells mL -1 The drop volume is 5–10 μL; the concentration of the candidate drug solution is 5–15 μg / mL. -1 Add 5–10 μL; the concentration of the PMA solution is 40–60 μg / mL. -1 Add 1 to 5 μL.
[0023] In the above method for screening anticancer drugs, the tumor cells in multiple microwells can be of the same type or different types.
[0024] To construct an automated electrochemical addressing platform, this invention, in the aforementioned method for screening anticancer drugs based on a high-throughput electrochemical addressing sensor array, connects the electrochemical addressing sensor array as the working electrode to a robotic arm integrating a reference electrode and a counter electrode. The robotic arm's movement trajectory is set to collect the electrochemical addressing pulse signals generated after incubation with added candidate anticancer drugs in each microwell. Data is stored and processed using a computer or instrument, and machine learning (ML) algorithms are employed for data statistics and analysis. This further eliminates potential systematic or random errors that may exist during the sensing process of the electrochemical addressing sensor array, thereby enabling more efficient, accurate, automated, and high-throughput screening of anticancer drugs.
[0025] Compared with the prior art, the beneficial effects of the present invention are:
[0026] This invention first obtains a LIG electrode film of a specific thickness through precise control of laser engraving technology, which is a prerequisite for ensuring sensitive and accurate detection by the sensor. Then, by adjusting the laser cutting power and the spacing, depth, and size of the micropore array, low-cost, high-performance LIG-based electrode arrays are rapidly and in batches. Furthermore, single-crystal titanium dioxide (TiO2) is used as a sensitizing material in the LIG-based electrode array, and its structure and conductivity are crucial, directly affecting the electron transport rate. Pt nanoparticles are then combined as a specific electrocatalytic layer to promote the selective recognition and oxidation reaction of H2O2 and to accurately monitor changes in H2O2 concentration. Finally, laminin is modified for cell adhesion and growth, thereby realizing the preparation of a sensitive and highly selective electrochemically addressed sensing array.
[0027] Building upon this foundation, this invention utilizes an electrochemically addressed sensing array as the working electrode, connected to a robotic arm integrating a reference electrode and a counter electrode, to construct a high-throughput, automated electrochemically addressed sensing system. By adding candidate anticancer drugs to the high-throughput electrochemically addressed sensing array for incubation, the change in the oxidation current of H2O2 released by tumor cells is measured using electrochemical methods to capture information on the interaction between tumor cells and anticancer drugs, thus enabling a more efficient and accurate assessment of the therapeutic effect of anticancer drugs. This invention uses tumor cells as sensing targets to evaluate the efficacy of candidate anticancer drugs, offering advantages over traditional animal models such as ease of acquisition, low cost, and high throughput, providing a new approach for the efficient screening of anticancer drugs.
[0028] In addition, this invention can combine electrochemically addressed cell sensing technology with powerful ML algorithms, which can effectively eliminate random or systematic errors in high-throughput, automated electrochemically addressed sensing systems (AEAC). Furthermore, it can deeply mine the potential correlations of the collected data, intelligently acquire and analyze the collected electrochemically addressed pulse signals, thereby more efficiently and accurately evaluating the therapeutic effects of anticancer drugs and providing strong support for the clinical research and development of anticancer drugs. Attached Figure Description
[0029] Figure 1 This is a schematic diagram of an electrochemically addressed sensor array used for anticancer drug screening in an embodiment; wherein, the XGBoost algorithm is a type of ML algorithm.
[0030] Figure 2 The following are schematic diagrams of the fabrication of LIG-based electrode arrays (A), LIG-TiO2-Pt microporous electrode arrays (B), and LIG-TiO2-Pt-Lam microporous electrode arrays (C) in the embodiments.
[0031] Figure 3 The following are SEM (A) and TEM (B) characterizations of the LIG electrode film in the examples.
[0032] Figure 4 The following are the SEM (A, B), TEM (C, D), and EDS (E, F) characterizations of the LIG-TiO2 microporous electrode arrays (A, C, E) and LIG-TiO2-Pt microporous electrode arrays (B, D, F) in the examples; the XRD (G), Raman (H), and XPS (I) characterizations of LIG, TiO2, LIG-TiO2, LIG-Pt, and LIG-TiO2-Pt; and the XPS peak characterizations of the C1s (J), Ti 2p (K), and Pt 4f (L) states in LIG-TiO2-Pt.
[0033] Figure 5 The following are examples: cyclic voltammetry (A), AC impedance (B), chronoamperometry (C), electrochemical active area (D), conductivity (E), and radar chart comparison of the above results (F) for LIG(a), LIG-TiO2(b), LIG-Pt(c), and LIG-TiO2-Pt(d); optimized structural model (G) and H2O2 binding energy (H) for H2O2 adsorption sites of LIG-TiO2(G1), LIG-Pt(G2), and LIG-TiO2-Pt(G3); and robot arm movement trajectory diagram, i.e., electrochemical addressing route diagram (I).
[0034] Figure 6 The feasibility (A), array reproducibility (B), independence (C), linearity (D), selectivity (E), and batch reproducibility (F) of the LIG-TiO2-Pt-Lam electrode array for H2O2 sensing in the embodiments are tested.
[0035] Figure 7 Images (A) of HCM, HUVEC, MCF-7, Hela, HepG2, and U87 cells cultured at the LIG-TiO2-Pt-Lam interface under confocal microscopy in the examples; chronoamperometry curves (B) of MCF-7 cells after the addition of PMA, DMSO, and peroxidase to LIG-TiO2-Pt-Lam; histogram (C) of the relative current response of H2O2 in MCF-7 cells 1-6 hours after treatment with the anticancer drug DOX, and a comparison (D) of cells without drug treatment; electrochemical addressing pulse signals (E) and their heatmaps (F) of MCF-7, Hela, HepG2, and U87 cells screened for anticancer drugs using the AEAC sensing system; and a heatmap (G) of MCF-7, Hela, HepG2, and U87 cells screened for anticancer drugs using the MTT assay.
[0036] Figure 8The following are illustrations of anticancer drug screening using the ML algorithm in the example (A); eigenvalue correlation analysis of the ML model (B); prediction accuracy of different algorithm classifiers based on random forest (RF), k-nearest neighbor (KNN), logistic regression (LR), extreme gradient boosting (XGBoost), support vector machine (SVM) and decision tree (DT) for candidate anticancer drug screening (C); receiver operating characteristic (ROC) curve (D).
[0037] Figure 9 The confusion matrix for anticancer drug screening predicted by the ML algorithm is shown in (A); the classifier prediction of the effectiveness of candidate anticancer drugs by the XGBoost algorithm is shown in (B), where "red circle" represents effective drugs and "blue circle" represents ineffective drugs. Detailed Implementation
[0038] To more clearly illustrate the objectives, technical solutions, and advantages of this invention, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the following embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.
[0039] In the following examples, the method for preparing TiO2 single crystals is as follows: 20 mL of isopropanol and 5 mL of acetone are vigorously stirred in a beaker for 5-10 min to mix, then 1 mL of ethylenediamine is slowly added and stirring is continued for 5-10 min; subsequently, 1 mL of tetrabutyl titanate is added to the beaker and stirring is continued for 15-20 min; finally, the above mixed solution is transferred to a 50 mL high-pressure autoclave lined with polytetrafluoroethylene, heated at 200 °C for 20 h, cooled to room temperature, and the white TiO2 product is washed several times with ethanol and then dried for later use.
[0040] Example
[0041] (1) Preparation of LIG electrode thin film
[0042] A 4.5×4.5cm PI tape was attached to a 5×6cm PET substrate. The surface of the PI tape was cleaned with 95% ethanol. Then, the laser power of the laser engraving equipment was set to 4W and the laser engraving depth was set to 60%. A 4cm×4cm LIG electrode film with a thickness of 21μm was engraved. The film was rinsed with ultrapure water and dried at 60℃ for 30 minutes for later use.
[0043] Figure 3 A and B show that the LIG electrode film exhibits a three-dimensional (3D) porous ultrathin graphene nanosheet structure, and typical graphene corrugated wrinkles can be observed.
[0044] (2) Fabrication of LIG-based electrode arrays with micropores:
[0045] A 6-row × 6-column micropore array was constructed on a 4×4cm PVC tape using laser cutting (laser power set to 8-9W, micropore diameter 3mm, hole spacing 1mm, micropore depth 0.15-0.3mm, totaling 36 micropores). Then, the PVC micropore array and the aforementioned LIG electrode film were glued together to form a stacked LIG-based electrode array, i.e., a LIG-based electrode array with micropores.
[0046] (3) Fabrication of LIG-TiO2 microporous electrode array
[0047] 0.6 mg of spindle-shaped TiO2 single crystals were dispersed in 1 mL of ultrapure water and ultrasonically treated for 10 min to obtain a uniform TiO2 dispersion. Then, the dispersion was drop-coated into the micropores of a LIG-based electrode array and dried in an electric oven at 60 °C for 30 min to achieve the modification of TiO2 single crystals in the micropores of the LIG-based electrode array, thus obtaining a LIG-TiO2 microporous electrode array.
[0048] (4) Fabrication of LIG-TiO2-Pt microporous electrode array
[0049] The LIG-TiO2 microporous electrode array was placed in a 0.1M HCl solution of 5mM H2PtCl6 and electrodeposited at a potential of -0.1V for 2 min using a constant potential deposition method. During this process, Pt nanoparticles were uniformly modified on the surface of the LIG-TiO2 microporous electrode array. Subsequently, it was rinsed with ultrapure water and dried at 60℃ for 30 min to obtain the LIG-TiO2-Pt microporous electrode array.
[0050] (5) Fabrication of LIG-TiO2-Pt-Lam microporous electrode array
[0051] The LIG-TiO2-Pt microporous electrode array was sterilized by irradiation under ultraviolet light for 60 min. Then, Lam solution (5 μL, 40 μg mL) was added dropwise into the micropores of the sterilized LIG-TiO2-Pt microporous electrode array. -1 Lam modification was achieved by incubation at room temperature for 90 min, followed by washing with PBS solution to obtain LIG-TiO2-Pt-Lam microporous electrode array.
[0052] Figure 4 A and C indicate that TiO2 is a rhombic or spindle-shaped crystal with a diameter of 50–80 nm, and is an anatase single phase. It was successfully modified on the LIG electrode film to form a LIG-TiO2 composite material. Figure 4B and D indicate that 100 nm spherical Pt nanoparticles are also tightly fixed on the surface of LIG-TiO2, forming a unique 3D structure with LIG and TiO2. That is, the microporous inner surface of the LIG-TiO2-Pt microporous electrode array is actually a LIG-TiO2-Pt composite material. Figure 4 E and F indicate that the C, Ti, O, and Pt elements in the energy-dispersive X-ray (EDX) characterization are uniformly distributed, corresponding to LIG-TiO2 and LIG-TiO2-Pt, respectively, proving that LIG-TiO2 and LIG-TiO2-Pt composite materials were successfully prepared in the micropores of the electrode array. Figure 4 GI indicates that X-ray diffraction (XRD), Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS) characterization results further confirm that TiO2 and Pt were successfully modified on the LIG surface. Figure 4 JL indicates that the XPS peak spectra of the C1s, Ti 2p, and Pt 4f states in the composite material LIG-TiO2-Pt further prove the successful fabrication of the LIG-TiO2-Pt microporous electrode array.
[0053] Figure 5 A indicates that, in cyclic voltammetry (CV) testing, the redox peak current (Io) of the LIG-TiO2-Pt composite material... pa The larger size indicates that it has a stronger electron transfer capability. Figure 5 B indicates that, in the electrochemical impedance spectroscopy (EIS) test, the charge transfer resistance (R) of the LIG-TiO2-Pt composite material is... ct The smaller value indicates that it has a stronger charge transfer capability. Figure 5 C indicates that the LIG-TiO2-Pt composite material has a stronger catalytic ability for H2O2 in the chronoamperometry (it) test. Figure 5 D indicates that, in the electrochemical surface area (ECSA) test, the double-layer capacitance (C) of the LIG-TiO2-Pt composite material is... dl The smaller size indicates that it has more active sites and higher electrochemical activity. Figure 5 E indicates that, in the conductivity (G) test, the LIG-TiO2-Pt composite material has a faster electron transfer rate and the highest electron transfer efficiency. Figure 5 F indicates that the test results for CV, EIS, it, ECSA, and G are consistent. Figure 5G and H indicate that, based on density functional theory (DFT), the binding energy of H2O2 at the sensing interfaces of LIG, LIG-TiO2, and LIG-TiO2-Pt was calculated, and the electrochemical sensing performance of the three materials for H2O2 was evaluated. Among them, LIG-TiO2-Pt has a higher binding energy for H2O2, which can accelerate the adsorption of H2O2 on its surface. Pt nanoparticles not only act as important active centers for the selective adsorption of H2O2, but also activate the surrounding Ti active sites to enhance its electrochemical non-enzymatic H2O2 catalytic and sensing performance.
[0054] (6) In order to construct an automated electrochemical addressing platform, the LIG-TiO2-Pt-Lam microporous electrode array is used as the working electrode and assembled with a robot arm that integrates the reference electrode and the counter electrode (Ag / Pt electrode pair, reference electrode-Ag wire and counter electrode-Pt wire). The robot arm’s movement trajectory is set to collect the electrochemical addressing pulse signal in each micropore in sequence. The collected electrochemical addressing pulse signal is connected to a computer for data storage and processing. It can be used as a high-throughput, automated electrochemical addressing anticancer drug screening system and platform (AEAC).
[0055] The tumor cell suspension (10 μL, 1×10) 5 Tcells mL -1 Tumor cells were added in parallel to 36 microwells of a LIG-TiO2-Pt-Lam electrode array (using MCF-7, HeLa, HepG2, and U87 cells respectively) and incubated at 37°C for 2 hours to obtain the LIG-TiO2-Pt-Lam / Tcells microwell electrode array. Then, candidate anticancer drugs (D1-Control, D2-DMSO, D3-Genistein, D4-Naringenin, D5-Cisplatin, D6-Taxol, D7-Quercitrin, D8-DOX, D9-Docetaxel, D1-Taxol, D2-Doxetine, D3-Doxetine, D4-Naringenin, D5-Cisplatin, D6-Taxol, D7-Quercitrin, D8-DOX, D9-Doxetine, D1-Doxetine) were incubated. 10 -Apigenin), capecitabine (D 11 -Capecitabine), isoliquiritin (D) 12 -Isoliquiritigenin, resveratrol (D) 13 -Resveratrol), Kaempferol (D) 14 -Kaempferol), Sorafenib (D 15 -Sorafenib) and epigallocatechin (D 16-Epigallocatechin), the concentration of the candidate anticancer drug solution was 10 μg / mL. -1 10 μL of PMA solution was added to the microwells and incubated for 4 h under constant temperature and humidity conditions of 5% CO2 and 95% Air at 37 °C. Subsequently, the microwells containing the anticancer drug were rinsed with PBS solution to remove excess drug residue. Then, 1 μL of PMA solution (50 μg / mL) was added to the microwells. -1 This stimulates tumor cells to produce H2O2. The electrical addressing pulse signal generated in each microwell is linearly correlated with the H2O2 concentration therein, meaning that the survival rate of tumor cells in each microwell is linearly correlated with the electrical addressing pulse signal, indirectly reflecting the therapeutic effect of candidate anticancer drugs on different types of tumor cells.
[0056] Figure 6 A shows that when the reference electrode-Ag wire and the counter electrode-Pt wire come into contact with the electrolyte droplets in the micropores of the LIG-TiO2-Pt-Lam microporous electrode array, the three electrodes form a complete electrochemical reaction cell, immediately generating an instantaneous current response (electrical addressing pulse signal) that is linearly related to the H2O2 concentration in the droplets. This proves that the LIG-TiO2-Pt-Lam electrode array of the present invention has excellent sensitivity to H2O2. Figure 6 B indicates that the different sensing regions (i.e., different micropores) of the LIG-TiO2-Pt-Lam microporous electrode array described in this invention exhibit good reproducibility (RSD = 3.79%). Figure 6 C indicates that, despite the absence of additional washing treatment for this LIG-TiO2-Pt-Lam microporous electrode array, the H2O2 detection in each sensing region is almost unaffected, demonstrating the excellent independence of the LIG-TiO2-Pt-Lam microporous electrode array of the present invention. Figure 6 D indicates that as the H2O2 concentration increases, the electrical addressing pulse signal also gradually increases, indicating a clear linear relationship between H2O2 concentration and current response signal. The linear equation is I(μA)=0.7665+0.00376C(μM)(R 2 =0.9984). Figure 6 E indicates that the drug screening process based on tumor cell models is subject to numerous interfering factors, such as sodium ions (Na+). + ), calcium ions (Ca 2+ Even with high concentrations (100 μM) of interfering substances, such as glucose (Glu), uric acid (UA), folic acid (FA), and glycine (Gly), the current response signal of the LIG-TiO2-Pt-Lam sensor array to H2O2 is basically unaffected, proving that the LIG-TiO2-Pt-Lam electrode array of the present invention has good selectivity. Figure 6F indicates that when six electrode arrays were selected for parallel testing with H2O2 solutions of the same concentration, the measurement results showed small fluctuations (RSD = 3.37%), demonstrating that the LIG-TiO2-Pt-Lam microporous electrode array of the present invention has good batch-to-batch reproducibility.
[0057] Figure 7 A shows that normal cells (HCM, HUVEC) and tumor cells (MCF-7, HeLa, HepG2, U87) incubated on the LIG-TiO2-Pt-Lam surface exhibited excellent viability (wherein Calcein-AM was used to detect live cells and propidium iodide PI was used to detect dead cells), demonstrating that the LIG-TiO2-Pt-Lam microporous electrode array of the present invention has excellent biocompatibility. Figure 7 B shows that when MCF-7 cells are present, the current signal increases significantly after the addition of PMA; after the addition of peroxidase, the current signal decreases significantly, proving that the sensor can monitor the release of H2O2 from tumor cells in situ in real time. Figure 7 C and D indicate that after 6 hours of treatment with the anticancer drug DOX, the relative current response of MCF-7 cells to PMA stimulation was significantly reduced (from 100% to 38%), indicating that DOX can effectively kill MCF-7 cells. In contrast, the relative current response rate of MCF-7 cells without DOX treatment to PMA stimulation was slightly reduced (from 100% to 96.9%), indicating that MCF cells do not die naturally within 6 hours on the LIG-TiO2-Pt-Lam microporous array surface. Figure 7 EG indicates that the test results of the AEAC sensing system described in this embodiment for anticancer drug screening are basically consistent with the test results of the MTT method for anticancer drug screening, proving that the AEAC sensing system can screen anticancer drugs rapidly and with high throughput.
[0058] The specific steps for screening anticancer drugs using the MTT assay are as follows: tumor cells are seeded into 96-well plates (10... 5 Cells / well, 100 μL DMEM), and the well plate was incubated in a 37°C incubator for 24 h. Subsequently, the culture medium was replaced with fresh DMEM (10 μg / mL) containing different anticancer drugs. -1 (100 μL). After incubation for 24 h, wash twice with PBS and incubate at 37 °C with MTT reagent (5 mg / mL). -1 After incubating for 4 hours, the culture medium was removed, and DMSO (150 μL) was added to dissolve the formazan. Finally, the absorbance at 590 nm was measured using a microplate reader to obtain the anticancer drug screening results.
[0059] (7) Machine learning (ML) algorithms assist AEAC sensing systems in the efficient and accurate screening of anticancer drugs.
[0060] Based on the electrically addressed pulse signals acquired using the above scheme, the ML algorithm is used for data statistics and analysis to deeply mine potential correlation information in the data, further eliminating possible systematic or random errors in the AEAC sensing process, thereby screening anticancer drugs more efficiently and accurately. The main steps are as follows:
[0061] (a) Preprocess the acquired electrical addressing pulse signal dataset to remove abnormal data.
[0062] (b) Based on the data information, determine the characteristic values and extract the characteristic values (A1-peak current, A2-midpoint current, A3-endpoint current, A4-difference between peak and midpoint current, A5-difference between peak and endpoint current, A6-difference between midpoint and endpoint current, A7-ratio of peak to midpoint current difference and peak to endpoint current difference, A8-ratio of midpoint to endpoint current difference and peak to endpoint current difference).
[0063] (c) Randomly split the dataset into a training set (70%) and a test set (30%).
[0064] (d) Use the training set to train the Random Forest (RF), k-Nearest Neighbor (KNN), Logistic Regression (LR), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM) and Decision Tree (DT) algorithms.
[0065] (e) Validate the trained ML model using a test set and output the data analyzed by the ML algorithm.
[0066] (f) Transform the output data into a visual graph based on the evaluation indicators.
[0067] Figure 8 A is a schematic diagram of ML algorithm-assisted screening of anticancer drugs. Figure 8 B indicates that using the Pearson correlation coefficient to quantify the mutual information among the eight features, most features show low linear correlation, indicating strong predictive power for the target variable. Figure 8 C indicates that the XGBoost algorithm exhibits the highest prediction accuracy (98%) compared to RF (96.5%), KNN (96.5%), LR (84.2%), SVM (89.6%), and DT (94.1%) algorithms. Figure 8 D indicates that the receiver operating characteristic (ROC) curves of the XGBoost model in the MCF-7, Hela, HepG2, and U87 tests were 0.95, 0.98, 0.98, and 1.00, respectively, demonstrating that the selected XGBoost model has a strong ability to distinguish between "effective drugs" and "ineffective drugs" in binary classification. Similarly, Figure 9The confusion matrix results show that the XGBoost model can accurately predict anticancer drug data. Figure 9 B indicates that the XGBoost algorithm can accurately predict the test dataset and regress it to its respective clusters (red circles represent effective drug clusters, and blue circles represent ineffective drug clusters).
[0068] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements and modifications without departing from the inventive concept of the present invention, and these all fall within the protection scope of the present invention.
Claims
1. A method for high-throughput, automated screening of anticancer drugs using electrochemically addressed sensor arrays, characterized in that, By constructing a microporous array on a laser-induced graphene electrode film, a laser-induced graphene-based electrode array with micropores is obtained. Then, titanium dioxide single crystals, platinum nanoparticles and laminin are sequentially modified in the micropores to obtain a LIG-TiO2-Pt-Lam microporous electrode array, which is a high-throughput electrochemical addressing sensing array. The LIG-TiO2-Pt-Lam microporous electrode array serves as the working electrode. Multiple micropores in the array are used to incubate tumor cells of the same or different types. After incubation with added candidate anticancer drugs, the cells release H2O2, generating an electrical addressing pulse signal. The electrical addressing pulse signal generated in each micropore is linearly correlated with the H2O2 concentration therein. A three-electrode system is used to collect the electrical addressing pulse signal using an electrochemical workstation, thereby achieving automated, high-throughput screening of anticancer drugs. The reference electrode and counter electrode of the three-electrode system are integrated with a robotic arm, and the electrical addressing pulse signal in each micropore is collected sequentially by setting the movement trajectory of the robotic arm.
2. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 1, characterized in that, The method for fabricating the high-throughput electrochemically addressed sensing array includes the following steps: (1) Laser-induced graphene electrode film, i.e. LIG electrode film, is prepared on polyimide tape by laser engraving; (2) A microporous array is constructed on a polymer tape by laser cutting, and the microporous array is then bonded and assembled on the surface of the LIG electrode film to obtain a LIG-based electrode array with micropores; (3) Modify single-crystal titanium dioxide in the micropores of the LIG-based electrode array to obtain a LIG-TiO2 microporous electrode array; (4) Platinum Pt nanoparticles were deposited in the micropores of the LIG-TiO2 microporous electrode array by potentiostatic deposition to obtain the LIG-TiO2-Pt microporous electrode array. (5) After sterilizing the LIG-TiO2-Pt microporous electrode array by ultraviolet irradiation, the laminin Lam is modified in the micropores to obtain the LIG-TiO2-Pt-Lam microporous electrode array. (6) Using the LIG-TiO2-Pt-Lam microporous electrode array as the working electrode, a high-throughput electrochemical addressing sensing array is obtained.
3. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 2, characterized in that, In step (1), the laser power is 3~5 W and the laser engraving depth is 15~25 μm; in step (2), the micro-hole array is set with N rows × M columns, where N and M are both positive integers and not less than 2, the diameter of each micro-hole is 2~4 mm, the spacing between micro-holes is 0.5~1.5 mm, and the depth of the micro-hole is 0.15~0.3 mm, which is the same as the height of the micro-hole array.
4. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 2, characterized in that, In step (3), single-crystal titanium dioxide (TiO2) is rhomboid or spindle-shaped nanoparticles with a size in the range of 50~80 nm.
5. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 2, characterized in that, The specific process of step (3) is as follows: TiO2 single crystals are dispersed in ultrapure water and ultrasonically treated to obtain a uniform TiO2 dispersion with a concentration of 0.4~1 mg / mL. -1 Then, 5-10 μL of TiO2 dispersion was drop-coated into the micropores of the LIG-based electrode array and dried to obtain the LIG-TiO2 microporous electrode array.
6. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 2, characterized in that, The specific process of step (4) is as follows: The above-mentioned LIG-TiO2 microporous electrode array is placed in H2PtCl6 solution, and Pt nanoparticles are uniformly deposited in the micropores of the LIG-TiO2 microporous electrode array by constant potential deposition method. After rinsing with ultrapure water and drying, LIG-TiO2-Pt microporous electrode array is obtained; wherein, the pH of H2PtCl6 solution is 2~5, the concentration of H2PtCl6 is 3~9 mM; the applied potential for electrodeposition is -0.5~0.1 V, and the electrodeposition time is 1~10 min.
7. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 2, characterized in that, The specific process of step (5) is as follows: Lam solution is added dropwise into the micropores of the LIG-TiO2-Pt microporous electrode array after UV sterilization, and incubated at room temperature for 80-100 min; then, excess Lam is washed with phosphate-buffered saline (PBS) to obtain the LIG-TiO2-Pt-Lam microporous electrode array; wherein, the Lam solution is prepared using PBS with a concentration of 30-50 μg / mL. -1 Add 5-10 μL to each microwell.
8. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 1, characterized in that, Tumor cell suspension was added dropwise in parallel to each microwell of the electrochemical addressing sensor array and incubated in an incubator at 37±1℃ and constant temperature and humidity for 2-6 h. Then, candidate anticancer drugs were added and incubated at 37±1℃ and constant temperature and humidity for 4-12 h under conditions of 5%±0.5% CO2 and 95%±0.5% air. The microwells were then rinsed with PBS solution. Phloride solution was then added to stimulate tumor cells to produce H2O2. The generated electrochemical addressing pulse signal in each microwell was linearly correlated with the concentration of H2O2 therein.
9. The method for high-throughput, automated screening of anticancer drugs using an electrochemically addressed sensor array according to claim 8, characterized in that, The tumor cell suspension was 1×10 5 Tcells mL -1 The drop volume in each microwell is 5–10 μL; each candidate anticancer drug is prepared to a concentration of 5–15 μg / mL. -1 The solution was prepared by adding 5–10 μL of each candidate anticancer drug solution to each microwell; the concentration of phorbol ester solution was 40–60 μg / mL. -1 The addition volume for each microwell is 1~5 μL.
10. A high-throughput, automated system for electrochemically addressed sensing screening of anticancer drugs, characterized in that, A three-electrode system is adopted, with a high-throughput electrochemical addressing sensor array as the working electrode. The working electrode is connected to a robotic arm that integrates a reference electrode and a counter electrode. The micropore array of the working electrode is used to incubate the same or different types of tumor cells and anticancer drugs. The movement trajectory of the robotic arm is set to collect the electro-addressing pulse signal generated after adding the candidate anticancer drug to each micropore for incubation, and the signal is connected to a computer for data storage and processing. The high-throughput electrochemical addressing sensing array is obtained by constructing a microporous array on a laser-induced graphene electrode film to obtain a laser-induced graphene-based electrode array with micropores, and then sequentially modifying titanium dioxide single crystals, platinum nanoparticles and laminin within the micropores.