Electrostatically assembled fluorescent array sensor element library and construction method and application thereof

Abstract

The application discloses a static electro-combination fluorescent array sensor element library and a construction method and application thereof. The sensor can be applied to rapid detection of bacteria, can identify and distinguish a plurality of common bacteria in clinical infectious diseases in 2 hours, and realizes simple, rapid and accurate bacterial identification. The fluorescent array sensor has high sensitivity and fast detection time, has important significance for distinguishing and identifying bacteria, and is expected to realize early diagnosis of diseases, provide guidance for clinical medication, and reduce the death rate of bacterial infection.

Description

Technical Field

[0001] This invention relates to fluorescent array sensors and their applications, and particularly to a library of fluorescent array sensor elements based on electrostatic complexes formed from amino acid carbon quantum dots and antimicrobial peptides, as well as their construction and application. Background Technology

[0002] In recent years, bacterial infections have become a leading cause of death, seriously threatening human health. Currently, broad-spectrum antibiotics remain the primary treatment for bacterial infections. The emergence of drug-resistant bacteria exacerbates this situation and increases mortality rates. Timely and effective identification of pathogens in the early stages of bacterial infection helps guide the clinical use of antibiotics and significantly improves patient survival rates.

[0003] To date, widely used methods for bacterial identification include plate culture, morphological observation, and genetic and immunological characterization. Plate culture is the gold standard for traditional pathogen identification; however, it takes 3-5 days. Other techniques, such as polymerase chain reaction (PCR), enzyme-linked immunosorbent assay (ELISA), surface-enhanced Raman scattering (SERS), and mass spectrometry, often rely on expensive specialized equipment and procedures. Importantly, none of these methods can simultaneously detect multiple bacteria. Therefore, developing a rapid, accurate, and timely pathogen detection technology is crucial for reducing antibiotic overuse and lowering mortality rates.

[0004] Fluorescent sensors have attracted widespread attention due to their high sensitivity, strong discrimination ability, and real-time detection. Traditional fluorescent sensors mostly employ a "lock-and-key" approach, enabling highly selective identification of specific analytes but not the identification of a class of analytes. Therefore, cross-responsive fluorescent array sensors offer unique advantages in identifying similar compounds and complex samples. Array-based sensing systems, mimicking the olfactory and gustatory systems of mammals, provide another method for high-throughput detection and identification of bacterial species. Based on the differential interactions between sensing elements in the sensor array and bacteria, unique response patterns for the identification of various bacteria can be obtained through linear discriminant analysis (LDA). Generally, the distinguishing patterns are achieved through non-specific interactions between the analyte and the sensing unit. Bacterial identification is mainly achieved through differential electrostatic interactions between bacteria and the sensor array.

[0005] However, most of the fluorescent array sensors reported in the literature are designed based on the non-specific interaction between the sensing element and bacteria, which limits their discrimination ability and sensitivity, and also has poor anti-interference ability and requires complex synthesis steps. Therefore, there is still a need in the field to develop a fluorescent array sensor that is simple to operate, highly sensitive, and has a wide detection range for bacterial detection. Summary of the Invention

[0006] Objectives of this invention: The objective of this invention is to provide a library of electrostatic composite fluorescence array sensor elements, utilizing an electrostatic composite system of amino acid-functionalized carbon quantum dots and antimicrobial peptides to distinguish and detect bacteria, thereby enabling early diagnosis of clinical bacterial infections. Another objective of this invention is to provide a method for constructing an electrostatic composite fluorescence array sensor. A further objective is to provide an application of this electrostatic composite fluorescence array sensor in bacterial detection or bacterial infection detection.

[0007] Technical Solution: This invention provides an electrostatic composite fluorescence array sensor element library, which includes several sensing elements with different compositions. Each sensing element is composed of an amino acid-functionalized negatively charged carbon quantum dot and a positively charged antimicrobial peptide through electrostatic interaction. The amino acid-functionalized negatively charged carbon quantum dot is selected from a first set containing N kinds of amino acid-functionalized negatively charged carbon quantum dots; the positively charged antimicrobial peptide is selected from a second set containing M kinds of positively charged antimicrobial peptides.

[0008] In this process, based on the bacteria to be identified, superior sensing elements are selected from the element library for the identification and classification of the bacteria. The selection method is as follows: a set of bacteria to be identified is selected; based on the number of bacterial species to be identified in the set, a corresponding number of element libraries are constructed; the bacteria to be identified are added to each sensing element in the corresponding element library; and then the fluorescence response of each sensing element to the bacteria is measured.

[0009] Based on the fluorescence response value, several sensing elements that rank in the top 10-50% of the fluorescence response values ​​for all bacteria to be identified are selected as initial screening sensing elements; then, based on the factor contribution, sensing units that rank in the top 10-50% of the contribution are selected, and the final sensing elements obtained are the superior sensing elements.

[0010] In the above scheme, the sensing element is composed of carbon quantum dots with different amino acid functionalizations and antimicrobial peptides interacting together. Differential fluorescence output is achieved by leveraging the differential interaction between the amino acid carbon quantum dot-antimicrobial peptide composite system and bacteria, thereby enabling the identification of pathogenic bacteria.

[0011] Furthermore, in the first component, the amino acids used for functionalization in the negatively charged carbon quantum dots are selected from two or more of the following amino acids: alanine (Ala), phenylalanine (Phe), glutamic acid (Glu), diaminopropionic acid (Dap), tryptophan (Trp), tyrosine (Tyr), serine (Ser), histidine (His), aspartic acid (Asp), valine (Val), lysine (Lys), leucine (Leu), arginine (Arg), isoleucine (Ile), methionine (Met), proline (Pro), cysteine ​​(Cys), threonine (Thr), asparagine (Asn), and glutamine (Gln).

[0012] Furthermore, in the first component, the carbon source in the amino acid-functionalized negatively charged carbon quantum dots is selected from one or more of the following carbon sources: diammonium citrate, citric acid, urea, fish scale powder, ethylenediaminetetraacetic acid, sugars, folic acid, ethylenediamine, phenylenediamine, silk, milk, wool, and silkworm.

[0013] Furthermore, in the second component, the antimicrobial peptides are selected from two or more of the following sequences:

[0014]

[0015]

[0016] Furthermore, in the second component, the N-terminus and C-terminus of the antimicrobial peptide sequence are modified by one or both of the fluorophores, which are independently selected from: FAM, FITC, Cy3, TRITC, Texas Red, Cy5, Alexa Fluor 405, Bodipy, Cy7, Rodamine B, and AMC.

[0017] On the other hand, the present invention provides a method for constructing an electrostatic composite fluorescence array sensor, the method comprising the following steps:

[0018] (1) Construct a component library including N*M sensing elements: Select N kinds of amino acid-functionalized negatively charged carbon quantum dots to form the first component set, and select M kinds of antimicrobial peptides to form the second component set; The N kinds of components in the first component set and the M kinds of components in the second component set are combined to obtain N*M sensing elements with different compositions.

[0019] (2) Screening of sensing elements: Select a set of bacteria to be identified, and construct a corresponding number of element libraries based on the number of bacterial species to be identified in the bacterial set; add the bacteria to be identified to each sensing element in the corresponding element library, and then measure the fluorescence response of each sensing element to the bacteria.

[0020] Based on the fluorescence response value, select several sensing elements whose fluorescence response values ​​for all bacteria to be identified account for the top 10-50% as initial screening sensing elements.

[0021] (3) Then, based on the machine learning algorithm, the sensor elements with a large factor contribution are selected from the above-mentioned initial screening of sensor elements, and the sensor elements finally obtained constitute the electrostatic composite fluorescence array sensor.

[0022] The criterion for judging a large factor contribution is: based on the factor contribution, the sensing units ranked in the top 10 to 50 are judged as sensing elements with a large factor contribution.

[0023] Further, in step (1), the preparation method of amino acid functionalized negatively charged carbon quantum dots is as follows: carbon source and amino acid are synthesized into carbon quantum dots by bottom-up method. The solvents used include, but are not limited to, water, N,N-dimethylformamide, ethanol, glycerol, etc., followed by purification and drying to obtain the target product.

[0024] Further, in step (1), the method for preparing the sensing element is as follows: carbon quantum dots are prepared into a solution of 0.1 μg-100 μg / mL using a Na2HPO4-NaH2PO4 (20 mM) buffer solution with pH=6-8, and then an antimicrobial peptide aqueous solution of 0.001 μg-1000 μg / mL is added and mixed evenly to obtain the sensing unit. A fluorescent array sensor can be obtained by combining carbon quantum dots functionalized with different amino acids with different antimicrobial peptides.

[0025] Furthermore, in step (3), the machine learning algorithm is a neural network, support vector machine, decision tree, K-nearest neighbor algorithm, random forest, Gaussian process, branch and bound, logistic regression, principal component analysis, or linear discriminant analysis.

[0026] On the other hand, the present invention provides an application of the fluorescence array sensor constructed by the above method in bacterial detection or bacterial infection disease detection.

[0027] Furthermore, the application method is as follows: different sensing units of the fluorescence array sensor are mixed with bacteria and incubated at 37°C. The fluorescence intensity of each sensing unit is tested from 0h to 24h. The fluorescence data is processed and the data matrix is ​​classified to achieve the identification and differentiation of pathogenic bacteria.

[0028] Furthermore, the fluorescence data were processed and analyzed using the statistical analysis software SYSTAT (version 13.0). Linear discriminant analysis (LDA) was used to classify the data matrix and visualize the data, enabling the differentiation and detection of various bacteria.

[0029] Furthermore, if the bacteria to be detected are aqueous bacteria, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP60.

[0030] Furthermore, if the bacteria to be detected are serum system bacteria, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP3.

[0031] Furthermore, if the bacteria to be detected are urine bacteria, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, and Val-CDots-AMP60.

[0032] Furthermore, if the bacteria to be detected are Staphylococcus aureus of different concentrations, the sensing units included in the electrostatic complex fluorescence array sensor are: Val-CDots-AMP7, Tyr-CDots-AMP7, Trp-CDots-AMP7, Val-CDots-AMP2, and Dap-CDots-AMP7.

[0033] Furthermore, if the bacteria to be detected are different concentrations of Pseudomonas aeruginosa, the sensing units included in the electrostatic complex fluorescence array sensor are: Trp-CDots-AMP7, His-CDots-AMP4, Val-CDots-AMP7, Val-CDots-AMP60, and Dap-CDots-AMP7.

[0034] Furthermore, if the bacteria to be detected are a mixture of Escherichia coli and Enterococcus faecalis, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP7, Val-CDots-AMP7, Dap-CDots-AMP7, Dap-CDots-AMP3, and Val-CDots-AMP2.

[0035] Furthermore, if the bacteria to be detected are bacteria in the serum sample of a clinical sepsis patient, the sensing units included in the electrostatic complex fluorescence array sensor are His-CDots-AMP4, Tyr-CDots-AMP3, His-CDots-AMP2, Val-CDots-AMP60, and Dap-CDots-AMP3, respectively.

[0036] Furthermore, if the bacteria to be detected are bacteria in a urine sample from a clinical urinary tract infection, the sensing units included in the electrostatic complex fluorescence array sensor are: Val-CDots-AMP60, Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, and His-CDots-AMP4.

[0037] As a preferred option:

[0038] The method for constructing the fluorescence array sensor as described above is as follows: using diammonium citrate as the carbon source, and D-amino acid-functionalized carbon quantum dots...

[0039] The solution was diluted to a final concentration of 0.1 μg–100 μg / mL using a Na₂HPO₄–NaH₂PO₄ (20 mM) buffer solution with a pH of 7.0–7.6. Antimicrobial peptides (final concentrations of 0.001 μg–1 μg / mL) were then added to construct the sensing units of the array sensor. The selected D-amino acid-functionalized carbon quantum dots were Ala-CDots, Phe-CDots, Glu-CDots, Trp-CDots, Tyr-CDots, Ser-CDots, His-CDots, Asp-CDots, Val-CDots, and Dap-CDots; the selected antimicrobial peptides were AMP1, AMP2, AMP3, AMP4, AMP5, AMP6, AMP7, AMP58, AMP59, and AMP60. The sensing units were obtained by combining the D-amino acid-functionalized carbon quantum dots with the antimicrobial peptides.

[0040] The method for distinguishing different types of bacteria, as described above, involves combining the D-amino acid-functionalized carbon quantum dots obtained in the above construction with an antimicrobial peptide to create a sensing unit. 50 μL of sterilized MH broth medium and 50 μL of different types of bacteria (final bacterial concentration OD) are added to the sensing unit. 600=0.001), including Klebsiella pneumoniae, Proteus mirabilis, Enterobacter hormaechei, Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus lugdunensis, Klebsiella oxytoca, Enterococcus faecium, Enterococcus faecalis, Serratia marcescens, Enterobacter aerogenes, Acinetobacter baumannii, and Salmonella. The samples were shaken for 10 seconds, and each group of bacteria was tested six times. Fluorescence intensity data were recorded at 0h, 2h, 4h, 6h, and 12h.

[0041] Principal component analysis and other algorithms were used to process the fluorescence data, and the sensing units were screened based on the identified bacteria. Five sensing units were obtained and formed into a sensor array, and their fluorescence response to the corresponding bacteria was tested. The data matrix was classified to obtain a two-dimensional array discrimination fingerprint of bacteria, enabling visual identification. The obtained two-dimensional array discrimination fingerprint was used as a model to detect unknown samples, and the accuracy of the prediction for unknown samples was calculated, thus achieving the differentiation and detection of bacteria.

[0042] Beneficial Effects: This invention utilizes electrostatic interactions to combine D-amino acid-functionalized carbon quantum dots with antimicrobial peptides, integrating specific and non-specific interactions to form a novel semi-specific and highly selective fluorescent array sensor. This sensor can simultaneously distinguish and detect fourteen bacterial types, exhibiting high accuracy and sensitivity, short detection time, low cost, high repeatability, and requiring no specialized technicians. Furthermore, it demonstrates excellent distinguishing effects in complex systems such as serum and urine, showing promising results in the detection of serum samples from clinical sepsis and urine samples from patients with clinical urinary tract infections. Attached Figure Description

[0043] Figure 1 These are the infrared spectra of 10 carbon quantum dots;

[0044] Figure 2 This is a hydration particle size distribution diagram of 10 types of carbon quantum dots;

[0045] Figure 3 The results are shown in Example 11; (A) is the fluorescence response signal of five sensing units to 14 kinds of bacteria in water; (B) is the visualization (LDA) of the recognition of bacteria in water by an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides.

[0046] Figure 4 The results are shown in Example 12; (A) is the fluorescence response signal of five sensing units to 14 bacteria in the serum system; (B) is the visualization (LDA) of the bacteria recognition in the serum system by the array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides.

[0047] Figure 5 This is a graph showing the measurement results in Example 13; where (A) is the fluorescence response signal of the five sensing units to 14 types of bacteria in the urine system;

[0048] (B) is a visualization (LDA) of bacteria recognition in a urine system by an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides.

[0049] Figure 6 This is a graph showing the measurement results in Example 14; where (A) is the fluorescence response signal of the five sensing units to different concentrations of Staphylococcus aureus in water;

[0050] (B) is a visualization (LDA) of an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides recognizing different concentrations of Staphylococcus aureus in water.

[0051] Figure 7 The graph shows the measurement results in Example 15; where (A) is the fluorescence response signal of the five sensing units to different concentrations of Pseudomonas aeruginosa in water;

[0052] (B) is a visualization (LDA) of different concentrations of Pseudomonas aeruginosa in water by an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides.

[0053] Figure 8 The graph shows the measurement results in Example 16; where (A) is the fluorescence response signal of the five sensing units to different proportions of Escherichia coli and Enterococcus faecalis in water;

[0054] (B) is a visualization (LDA) of an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides recognizing different proportions of Escherichia coli and Enterococcus faecalis in water.

[0055] Figure 9 The results are shown in Examples 17-18; where (A) is a visualization (LDA) of bacterial identification in clinical serum samples by an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides.

[0056] (B) is a visualization (LDA) of bacteria identification in clinical urine samples by an array sensor composed of amino acid-functionalized carbon quantum dots and antimicrobial peptides. Detailed Implementation

[0057] The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0058] Example 1

[0059] The synthesis steps of D-glutamic acid-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-glutamic acid were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0060] Example 2

[0061] The synthesis steps of D-histidine-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-histidine are dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution is added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue is dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0062] Example 3

[0063] The synthesis steps of D-valine-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-valine were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0064] Example 4

[0065] The synthesis steps of D-diaminopropionic acid-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-diaminopropionic acid were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0066] Example 5

[0067] The synthesis steps of D-tyrosine-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-tyrosine were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0068] Example 6

[0069] The synthesis steps of D-serine-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-serine were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0070] Example 7

[0071] The synthesis steps of D-aspartic acid-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-aspartic acid were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0072] Example 8

[0073] The synthesis steps of D-alanine-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-alanine were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0074] Example 9

[0075] The synthesis steps of D-phenylalanine-functionalized carbon quantum dots were as follows: 500 mg of diammonium citrate and 500 mg of D-phenylalanine were dissolved in 25 mL of DMF and transferred to a high-pressure reactor. The mixture was reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated three times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0076] Example 10

[0077] The synthesis steps of D-tryptophan-functionalized carbon quantum dots are as follows: 500 mg of diammonium citrate and 500 mg of D-tryptophan were dissolved in 25 mL of DMF, transferred to a high-pressure reactor, and reacted at 180 °C for 12 h. The resulting solution was added to 25 mL of ethyl acetate and centrifuged at 8000 r / min for 15 min, repeated 3 times. Finally, the brown residue was dispersed in water, poured into a dialysis bag with a molecular weight cutoff of 3500, dialyzed for 48 h, and freeze-dried to obtain a brownish-black powder for later use. The infrared spectrum and hydrated particle size distribution are shown in [reference needed]. Figure 1 and Figure 2 .

[0078] Example 11

[0079] Screening and construction process of fluorescence array sensors for water systems and bacterial detection in water systems

[0080] Here, we selected 10 amino acid-functionalized carbon quantum dots and 10 broad-spectrum antimicrobial peptides as components of the sensing units. Ala-CDots, Phe-CDots, Glu-CDots, Trp-CDots, Tyr-CDots, Ser-CDots, Asp-CDots, His-CDots, Val-CDots, and Dap-CDots were prepared using a 20 mM NaH₂PO₄-Na₂HPO₄ buffer solution at pH 7, with concentrations of 2.5 μg / mL, 2 μg / mL, 5 μg / mL, 1.5 μg / mL, 3 μg / mL, 3 μg / mL, 2 μg / mL, 2 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively. Then, 10 AMP solutions at 100 μg / mL were added to each of the 10 amino acid-functionalized carbon quantum dot solutions. This yielded a library of 100 sensing units. We selected 14 pathogenic bacteria with high mortality rates and tested the sensor unit's response to bacteria (OD). 600 The response is 0.05.

[0081] First, the 20 sensor units with the best response values ​​for each type of bacteria were selected. Then, statistical analysis was performed on the 280 sensor units, and the 20 sensor units were selected based on their frequency of occurrence. For sensor units with the same frequency, a random selection method was used. Subsequently, PCA analysis was used to select 10 sensor units based on the sum of the contributions of each sensor unit to Factors 1-Factor 9 (Table 1).

[0082] Table 1. PCA analysis of the bacterial response of 20 sensing units

[0083]

[0084] An array sensor consisting of 10 sensing units was tested for its ability to detect bacteria (OD). 600 The fluorescence response was 0.001 (Table 2). We found that pathogens could be distinguished using only five sensing units.

[0085] Table 2. Fluorescence response data matrix of 10 sensing units for different bacteria after 2 hours of incubation (final concentration of 14 bacteria is OD). 600 =0.001)

[0086]

[0087]

[0088]

[0089] Furthermore, using PCA analysis (Table 3), a sensor array consisting of 5 sensor units was selected based on the contribution value of each sensor unit to Factor 1. We first selected the sensor unit that contributes the most to Factor 1 (His-CDots-AMP2). The sensor unit that contributes the most to Factor 2 (His-CDots-AMP2) has already been selected. The sensor unit that contributes the most to Factor 3 (His-CDots-AMP4) is then selected. Next, we selected the sensor unit that contributes the second most to Factor 1 (Tyr-CDots-AMP3). The sensor unit that contributes the second most to Factor 2 (Tyr-CDots-AMP3) has already been selected. We then selected the sensor unit that contributes the second most to Factor 3 (Dap-CDots-AMP60). Finally, we selected the sensor unit that contributes the third most to Factor 1 (Val-CDots-AMP60), thus completing the construction of the sensor array consisting of 5 sensor units.

[0090] Table 3. PCA analysis of the bacterial response of the 10 sensing units

[0091]

[0092] The synthesized D-amino acid-functionalized carbon quantum dots (Tyr-CDots, His-CDots, Val-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted with a Na2HPO4-NaH2PO4 buffer solution (pH=7) to final concentrations of 3 μg / mL, 2 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively, in 30 mL solutions. Water-soluble antimicrobial peptides (AMP1, AMP2, AMP3, AMP60) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. 30 μL of the antimicrobial peptide solution was added to the carbon quantum dot solution and shaken well to obtain five composite systems (Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, Dap-CDots-AMP60), which were then set aside.

[0093] The method for distinguishing different types of bacteria is as follows: Take 100 μL of each of the five complexes Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP60 and add them to the stock solutions of different types of bacteria (final concentration OD). 600=0.001) 50 μL and 50 μL MH broth culture medium, shake for 10 seconds. The fluorescence intensity of Tyr-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the first set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the second set of signals. The fluorescence intensity of His-CDots-AMP4 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the third set of signals. The fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the fourth set of signals. The fluorescence intensity of Dap-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 350 nm and an emission wavelength of 460 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria at 37°C. Each bacterium and each complex was tested 6 times, obtaining 5 sets of signals each time. A total of 14 bacteria and 5 complexes were used, yielding 6 × 14 × 5 = 420 fluorescence signals. These 420 fluorescence signals were then used to construct a data matrix (Table 4). The 5 sensor units showed different fluorescence responses to different bacteria. Figure 3 A).

[0094] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 14 bacteria × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, 84 test samples were successfully divided into 14 groups. The LDA plot shows that even with low bacterial concentrations and a high variety of bacteria, different bacteria can still be distinguished. Figure 3 B), the Jackknifed Classification Matrix shows that the accuracy in distinguishing bacteria reaches 100% (Table 5).

[0095] Table 4. Fluorescence response data matrix of the fluorescence array sensor for different bacteria after 2 hours of incubation (final concentration of 14 bacteria is OD). 600 =0.001)

[0096]

[0097]

[0098] Table 5. Classification matrix of different bacteria by fluorescence array sensor after 2 hours of incubation.

[0099]

[0100] The array sensor's ability to distinguish unknown samples: Fourteen types of bacteria were randomly and sequentially subjected to blind testing, resulting in 56 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 6). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 56 unknown samples tested, 52 were correctly detected, achieving an accuracy of 93%.

[0101] Table 6. Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with unknown samples.

[0102]

[0103]

[0104] Example 12

[0105] Construction of a fluorescence array sensor and detection of bacteria in a serum system

[0106] Following the screening procedure in Example 11, five sensing units targeting the serum system were ultimately selected. The synthesized D-amino acid-functionalized carbon quantum dots (Tyr-CDots, His-CDots, Val-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of solutions with Na2HPO4-NaH2PO4 buffer (pH=7) to final concentrations of 3 μg / mL, 2 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively. Water-soluble antimicrobial peptides (AMP2, AMP3, AMP4, AMP60) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. Add 30 μL of antimicrobial peptide solution to carbon quantum dot solution, shake well, and obtain 5 composite systems (Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, Dap-CDots-AMP3) for later use.

[0107] The method for distinguishing different types of bacteria is as follows: Take 100 μL of each of the five complexes Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP3 and add them to the stock solutions of different types of bacteria (final concentration OD). 600=0.001, containing 10% serum) 50 μL and 50 μL MH broth medium, shake for 10 seconds. The fluorescence intensity of Tyr-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the first set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the second set of signals. The fluorescence intensity of His-CDots-AMP4 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the third set of signals. The fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the fourth set of signals. The fluorescence intensity of Dap-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 350 nm and an emission wavelength of 460 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria at 37°C. Each bacterium and each complex was tested 6 times, obtaining 5 sets of signals each time. A total of 12 bacteria and 5 complexes were used, yielding 6 × 14 × 5 = 420 fluorescence signals. These 420 fluorescence signals were then used to construct a data matrix (Table 7). The 6 sensor units showed different fluorescence responses to different bacteria. Figure 4 A).

[0108] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 14 bacteria × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, 84 test samples were successfully divided into 14 groups. The LDA plot shows that even with low bacterial concentrations and a high variety of bacteria, different bacteria can still be distinguished. Figure 4 B), the Jackknifed Classification Matrix shows an accuracy of 98% in distinguishing bacteria (Table 8).

[0109] Table 7. Fluorescence response data matrix of different bacteria in serum system after 2 hours of incubation by fluorescence array sensor (final concentration of 12 bacteria is OD). 600 =0.001)

[0110]

[0111]

[0112]

[0113] Table 8. Classification matrix of different bacteria in serum system after 2 hours of incubation by fluorescence array sensor.

[0114]

[0115] The array sensor's ability to distinguish unknown samples: Fourteen types of bacteria were randomly and sequentially subjected to blind testing, resulting in 56 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 9). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 56 unknown samples tested, 52 were correctly detected, achieving an accuracy of 93%.

[0116] Table 9. Fluorescence response data matrix of fluorescence array sensor for unknown serum samples after 2 hours of incubation.

[0117]

[0118]

[0119] Example 13

[0120] Construction of a fluorescence array sensor and detection of bacteria in a urine system

[0121] Following the screening process in Example 11, five sensing units targeting the urine system were ultimately selected. The synthesized D-amino acid-functionalized carbon quantum dots (Tyr-CDots, His-CDots, Val-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of solutions with final concentrations of 3 μg / mL, 2 μg / mL, and 1.5 μg / mL using a Na2HPO4-NaH2PO4 buffer solution (pH = 7). Water-soluble antimicrobial peptides (AMP1, AMP2, AMP3, AMP4, AMP60) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. 30 μL of the antimicrobial peptide solution was added to the carbon quantum dot solution and shaken well to obtain five composite systems (Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60) for later use.

[0122] The method for distinguishing different types of bacteria is as follows: Take 100 μL of each of the five complexes Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, Val-CDots-AMP60, and His-CDots-AMP4 and add them to the stock solutions of different types of bacteria (final concentration OD). 600=0.001, 50 μL of MH broth medium (containing 50% artificial urine) and 50 μL of MH broth medium, shaken for 10 seconds. The fluorescence intensity of Tyr-CDots-AMP1 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the first set of signals. The fluorescence intensity of Tyr-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the second set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the third set of signals. The fluorescence intensity of His-CDots-AMP4 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the fourth set of signals. The fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria in the complex system at 37°C. Each bacterium and each complex was tested 6 times, obtaining 5 sets of signals each time. A total of 14 bacteria and 5 complexes were used, yielding 6 × 14 × 5 = 420 fluorescence signals. These 420 fluorescence signals were then used to construct a data matrix (Table 10). The 6 sensor units showed different fluorescence responses to different bacteria. Figure 5 A).

[0123] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 14 bacteria × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, 84 test samples were successfully divided into 14 groups. The LDA plot shows that even with low bacterial concentrations and a high variety of bacteria, different bacteria can still be distinguished. Figure 5 B), the Jackknifed Classification Matrix shows an accuracy of 98% in distinguishing bacteria (Table 11).

[0124] Table 10. Fluorescence response data matrix of different bacteria in urine system after 2 hours of incubation by fluorescence array sensor (final concentration of 14 bacteria is OD). 600 =0.001)

[0125]

[0126]

[0127] Table 11. Classification matrix of different bacteria in urine system by fluorescence array sensor after 2 hours of incubation.

[0128]

[0129] The array sensor's ability to distinguish unknown samples: A total of 56 unknown samples of 14 bacterial species were randomly and sequentially tested in a blind test. The tests were conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 12). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 56 unknown samples tested, 54 were correctly detected, achieving an accuracy of 96%.

[0130] Table 12 Fluorescence response data matrix of fluorescence array sensor for unknown urine samples after 2 hours of incubation.

[0131]

[0132]

[0133] Example 14

[0134] Construction of a fluorescence array sensor and detection of different concentrations of Staphylococcus aureus

[0135] Based on the response values ​​of the sensing units to Staphylococcus aureus, five optimal response units were selected. The synthesized D-amino acid-functionalized carbon quantum dots (Val-CDots, Tyr-CDots, Trp-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of Na₂HPO₄-NaH₂PO₄ buffer solution (pH = 7) to final concentrations of 1.5 μg / mL, 3 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively. Water-soluble antimicrobial peptides (AMP2, AMP7) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. Add 30 μL of antimicrobial peptide solution to carbon quantum dot solution, shake well, and obtain 5 composite systems (Val-CDots-AMP7, Tyr-CDots-AMP7, Trp-CDots-AMP7, Val-CDots-AMP2, Dap-CDots-AMP7) for later use.

[0136] The method for distinguishing different types of bacteria is as follows: Take 100 μL of each of the five complexes Val-CDots-AMP7, Tyr-CDots-AMP7, Trp-CDots-AMP7, Val-CDots-AMP2, and Dap-CDots-AMP7 and add them to different concentrations of Staphylococcus aureus stock solution (final concentration OD). 600 =0.0001, OD 600 =0.001, OD 600 =0.01, OD 600=0.05, OD 600 =0.1, OD 600 =0.2) 50 μL and 50 μL MH broth medium, shaken for 10 seconds. The fluorescence intensity of Val-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 555 nm at an excitation wavelength of 470 nm, obtaining the first signal; the fluorescence intensity of Tyr-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 550 nm at an excitation wavelength of 470 nm, obtaining the second signal; the fluorescence intensity of Trp-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 490 nm at an excitation wavelength of 440 nm, obtaining the third signal; the fluorescence intensity of Val-CDots-AMP2 bound to bacteria was measured at an emission wavelength of 555 nm at an excitation wavelength of 470 nm, obtaining the fourth signal; the fluorescence intensity of Dap-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 460 nm at an excitation wavelength of 350 nm, obtaining the fifth signal; the relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria at 37°C. Each bacterial concentration and each complex was tested 6 times, yielding 5 sets of signals each time. For a total of 6 bacterial concentrations and 5 complexes, 6×6×5=180 fluorescence signals were obtained. These 180 fluorescence signals were then used to construct a data matrix (Table 13). The 6 sensor units showed different fluorescence responses to different bacteria. Figure 6 A).

[0137] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 6 bacterial concentrations × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, the 36 test samples were successfully divided into 6 groups. The LDA plot shows that even at low bacterial concentrations, different concentrations of bacteria can be distinguished. Figure 6 B), the Jackknifed Classification Matrix shows that the accuracy in distinguishing bacteria reaches 100% (Table 14).

[0138] Table 13 Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with different concentrations of Staphylococcus aureus.

[0139]

[0140]

[0141] Table 14. Classification matrix of Staphylococcus aureus at different concentrations after 2 hours incubation by fluorescence array sensor.

[0142]

[0143] The array sensor's ability to distinguish unknown samples: Six concentrations of bacteria were randomly and sequentially subjected to blind testing, resulting in 24 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 15). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. All 24 unknown samples were correctly detected, achieving an accuracy of 100%.

[0144] Table 15. Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with Staphylococcus aureus sample of unknown concentration.

[0145]

[0146] Example 15

[0147] Construction of a fluorescence array sensor and detection of different concentrations of Pseudomonas aeruginosa

[0148] Based on the response values ​​of the sensing units to *Pseudomonas aeruginosa*, five optimal response units were selected. The synthesized D-amino acid-functionalized carbon quantum dots (Trp-CDots, His-CDots, Val-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of Na₂HPO₄-NaH₂PO₄ buffer solution (pH = 7) to final concentrations of 1.5 μg / mL, 2 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively. Water-soluble antimicrobial peptides (AMP4, AMP7, AMP60) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. Add 30 μL of antimicrobial peptide solution to carbon quantum dot solution, shake well, and obtain 5 composite systems (Trp-CDots-AMP7, His-CDots-AMP4, Val-CDots-AMP7, Val-CDots-AMP60, Dap-CDots-AMP7) for later use.

[0149] The method for distinguishing different types of bacteria is as follows: 100 μL of five complexes Trp-CDots-AMP7, His-CDots-AMP4, Val-CDots-AMP7, Val-CDots-AMP60, and Dap-CDots-AMP7 are added to different concentrations of Pseudomonas aeruginosa stock solution (final concentration OD). 600 =0.0001, OD 600 =0.001, OD 600 =0.01, OD 600 =0.05, OD 600 =0.1, OD600 =0.2) 50 μL and 50 μL MH broth medium, shaken for 10 seconds. The fluorescence intensity of Trp-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 490 nm after excitation at 440 nm, obtaining the first signal; the fluorescence intensity of His-CDots-AMP4 bound to bacteria was measured at an emission wavelength of 490 nm after excitation at 440 nm, obtaining the second signal; the fluorescence intensity of Val-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 555 nm after excitation at 470 nm, obtaining the third signal; the fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an emission wavelength of 555 nm after excitation at 470 nm, obtaining the fourth signal; the fluorescence intensity of Dap-CDots-AMP7 bound to bacteria was measured at an emission wavelength of 460 nm after excitation at 350 nm, obtaining the fifth signal; the relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation of the complex system with bacteria at 37°C. Each bacterium and each complex was tested 6 times, obtaining 5 sets of signals each time. With 6 bacterial concentrations and 5 complexes, a total of 6×6×5=180 fluorescence signals were obtained. These 180 fluorescence signals were then used to construct a data matrix (Table 16). The 6 sensor units showed different fluorescence responses to different bacteria. Figure 7 A).

[0150] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 6 bacterial concentrations × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, the 36 test samples were successfully divided into 6 groups. The LDA plot shows that even with low bacterial concentrations and a high variety of bacterial species, different bacteria can still be distinguished. Figure 7 B), the Jackknifed Classification Matrix shows that the accuracy in distinguishing bacteria reaches 100% (Table 17).

[0151] Table 16. Fluorescence response data matrix of fluorescence array sensor for different concentrations of Pseudomonas aeruginosa after 2 hours of incubation.

[0152]

[0153] Table 17. Classification matrix of different concentrations of Pseudomonas aeruginosa after 2 hours of incubation with a fluorescence array sensor.

[0154]

[0155]

[0156] The array sensor's ability to distinguish unknown samples: Six bacterial concentrations were randomly and sequentially tested in a blind test, resulting in 24 unknown samples. The tests were conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 18). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 24 unknown samples tested, 23 were correctly detected, achieving an accuracy of 96%.

[0157] Table 18. Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with unknown concentrations of Pseudomonas aeruginosa samples.

[0158]

[0159] Example 16

[0160] Construction of a fluorescence array sensor and detection of mixed bacteria in water at different ratios

[0161] Based on the response values ​​of the sensing units to *Escherichia coli* and *Enterococcus faecalis*, five sensing units were selected respectively. Based on the frequency distribution of the ten sensing units, five sensing units were finally selected to construct a fluorescence array sensor. The synthesized D-amino acid-functionalized carbon quantum dots (Tyr-CDots, Val-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted with a Na2HPO4-NaH2PO4 buffer solution (pH=7) to final concentrations of 1.5 μg / mL and 10 μg / mL, respectively, in 30 mL solutions. Water-soluble antimicrobial peptides (AMP2, AMP3, AMP7) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. Add 30 μL of antimicrobial peptide solution to carbon quantum dot solution, shake well to obtain 5 composite systems (Tyr-CDots-AMP7, Val-CDots-AMP7, Dap-CDots-AMP7, Dap-CDots-AMP3, Val-CDots-AMP2), for later use.

[0162] The method for distinguishing different types of bacteria is as follows: Take 100 μL of each of the five complexes Tyr-CDots-AMP7, Val-CDots-AMP7, Dap-CDots-AMP7, Dap-CDots-AMP3, and Val-CDots-AMP2 and add 50 μL of different mixing ratios of Escherichia coli: Enterococcus faecalis stock solutions (ratios of 100%; 0%, 75%; 25%, 50%; 50%, 25%; 75%, 0%; 100%) and 50 μL of MH broth medium, and shake for 10 seconds. The fluorescence intensity of Tyr-CDots-AMP7 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the first set of signals. The fluorescence intensity of Val-CDots-AMP7 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the second set of signals. The fluorescence intensity of Dap-CDots-AMP7 bound to bacteria was measured at an excitation wavelength of 350 nm and an emission wavelength of 460 nm, yielding the third set of signals. The fluorescence intensity of Dap-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 350 nm and an emission wavelength of 460 nm, yielding the fourth set of signals. The fluorescence intensity of Val-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria in the complex system at 37°C. Each bacterium and each complex was tested 6 times, yielding 5 sets of signals each time. With 5 different bacterial ratios and 5 different complexes, 6 × 5 × 5 = 150 fluorescence signals were obtained. These 150 fluorescence signals were then used to construct a data matrix (Table 19). The 5 sensor units showed different fluorescence responses to different bacteria. Figure 8 A).

[0163] The fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 5 bacterial mixture ratios × 6 replicates) was converted into 5 canonical factors. In the LDA plot using factors 1 and 2, the 30 test samples were successfully divided into 5 groups. The LDA plot shows that even with low bacterial concentrations and a high variety of bacterial species, different bacteria can still be distinguished. Figure 8 B), the Jackknifed Classification Matrix shows that the accuracy in distinguishing bacteria reaches 100% (Table 20).

[0164] Table 19 Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with mixed bacteria of different proportions.

[0165]

[0166] Table 20 Classification matrix of folding knives by fluorescence array sensor after incubation of mixed bacteria of different proportions for 2 hours.

[0167]

[0168] The array sensor's ability to distinguish unknown samples: Five bacterial species were randomly mixed in varying proportions and blindly tested on 20 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 21). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. All 20 unknown samples were correctly detected, achieving an accuracy of 100%.

[0169] Table 21 Fluorescence response data matrix of fluorescence array sensor after 2 hours of incubation with mixed bacterial samples of unknown proportion.

[0170]

[0171]

[0172] Example 17

[0173] Construction of a fluorescence array sensor and detection of clinical sepsis serum samples

[0174] A fluorescence array sensor constructed using a serum system was used for the detection of serum samples from clinical sepsis. Synthesized D-amino acid-functionalized carbon quantum dots (Tyr-CDots, His-CDots, Val-CDots, Dap-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of Na₂HPO₄-NaH₂PO₄ buffer (pH = 7) to final concentrations of 3 μg / mL, 2 μg / mL, 1.5 μg / mL, and 10 μg / mL, respectively. Water-soluble antimicrobial peptides (AMP₂, AMP₃, AMP₄, AMP₆₀) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. Add 30 μL of antimicrobial peptide solution to carbon quantum dot solution, shake well, and obtain 5 composite systems (His-CDots-AMP4, Tyr-CDots-AMP3, His-CDots-AMP2, Val-CDots-AMP60, Dap-CDots-AMP3) for later use.

[0175] The method for distinguishing different types of bacteria is as follows: Take 100 μL of the five complexes His-CDots-AMP4, Tyr-CDots-AMP3, His-CDots-AMP2, Val-CDots-AMP60, and Dap-CDots-AMP3, add 50 μL of the stock solution (containing 10% clinical serum) of 25 clinical serum samples, and 50 μL of MH broth medium, and shake for 10 seconds. The fluorescence intensity of His-CDots-AMP4 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the first set of signals. The fluorescence intensity of Tyr-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the second set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the third set of signals. The fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the fourth set of signals. The fluorescence intensity of Dap-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 350 nm and an emission wavelength of 460 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria. Each clinical sample was tested 6 times with each complex, obtaining 5 sets of signals each time. A total of 25 clinical samples and 5 complexes were tested, resulting in 6×25×5=750 fluorescence signals. These 750 fluorescence signals were then used to form a data matrix (Table 22). The 5 sensor units showed different fluorescence responses to different bacteria.

[0176] Fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 25 clinical samples × 6 replicates) was converted into four canonical factors. In the LDA plots using factors 1 and 2, 150 test samples were successfully divided into 5 groups. The LDA plots showed that even at low bacterial concentrations, different bacteria could be distinguished. Figure 9 A) The Jackknifed Classification Matrix shows an accuracy of 92% in distinguishing bacteria (Table 23).

[0177] Table 22 Fluorescence response data matrix of fluorescence array sensor for different clinical serum samples after 6 hours of incubation.

[0178]

[0179]

[0180]

[0181] Table 23 Classification matrix of folding knives based on fluorescence array sensor after 6 hours of incubation of clinical serum samples.

[0182]

[0183] The array sensor's ability to distinguish unknown samples: Five types of clinical serum samples were randomly and sequentially subjected to blind testing, resulting in a total of 100 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 24). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 100 unknown samples tested, 71 were correctly detected, achieving an accuracy rate of 71%.

[0184] Table 24. Fluorescence response data matrix of clinical sepsis serum samples incubated for 6 hours by the fluorescence array sensor.

[0185]

[0186]

[0187]

[0188] Example 19

[0189] Construction of a fluorescence array sensor and detection of urine samples from patients with clinical urinary tract infections.

[0190] A fluorescence array sensor constructed using a urine system was used to detect urine samples from patients with clinical urinary tract infections. Synthesized D-amino acid-functionalized carbon quantum dots (Val-CDots, Tyr-CDots, His-CDots) were dissolved in distilled water to prepare a 1 mg / mL stock solution. This stock solution was then diluted to 30 mL of Na₂HPO₄-NaH₂PO₄ buffer (pH = 7) to final concentrations of 3 μg / mL, 2 μg / mL, and 1.5 μg / mL, respectively. Water-soluble antimicrobial peptides (AMP1, AMP2, AMP3, AMP4, AMP60) were dissolved in distilled water and diluted to a concentration of 100 μg / mL. 30 μL of the antimicrobial peptide solution was added to the carbon quantum dot solution and shaken well to obtain five composite systems (Val-CDots-AMP60, Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4) for later use.

[0191] The method for distinguishing different types of bacteria is as follows: Take 100 μL of the five complexes Val-CDots-AMP60, Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, and His-CDots-AMP4, add 50 μL of clinical urine sample stock solution (containing 50% clinical urine) and 50 μL of MH broth medium, and shake for 10 seconds. The fluorescence intensity of Val-CDots-AMP60 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 555 nm, yielding the first set of signals. The fluorescence intensity of Tyr-CDots-AMP1 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the second set of signals. The fluorescence intensity of Tyr-CDots-AMP3 bound to bacteria was measured at an excitation wavelength of 470 nm and an emission wavelength of 550 nm, yielding the third set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the fourth set of signals. The fluorescence intensity of His-CDots-AMP2 bound to bacteria was measured at an excitation wavelength of 440 nm and an emission wavelength of 490 nm, yielding the fifth set of signals. The relative fluorescence intensity change was used as the detection signal (I-I0) / I0. Fluorescence intensity was measured at 37°C for 0h, 2h, 4h, and 6h after incubation with bacteria in the complex system. Each clinical sample was tested 6 times with each complex, obtaining 5 sets of signals each time. A total of 15 clinical urine samples and 5 complexes were collected, yielding 6×15×5=450 fluorescence signals. These 450 fluorescence signals were then used to construct a data matrix (Table 25). The 5 sensor units showed different fluorescence responses to different bacteria.

[0192] Fluorescence data were processed and analyzed using statistical analysis software. Linear discriminant analysis (LDA) was used to convert the fluorescence response patterns into canonical patterns. The training matrix (5 channels × 15 clinical urine samples × 6 replicates) was converted into two canonical factors. In the LDA plots using factors 1 and 2, 90 test samples were successfully divided into 3 groups. The LDA plots show that even at low bacterial concentrations, different bacteria can ultimately be distinguished. Figure 9 B), the Jackknifed Classification Matrix shows an accuracy of 91% in distinguishing bacteria (Table 26).

[0193] Table 25 Fluorescence response data matrix of fluorescence array sensor for different clinical urine samples after 2 hours of incubation.

[0194]

[0195]

[0196] Table 26. Classification matrix of folding knives based on fluorescence array sensor incubation of clinical urine samples for 2 hours.

[0197]

[0198] The array sensor's ability to distinguish unknown samples: Three types of clinical urine samples were randomly and sequentially tested blindly, totaling 60 unknown samples. The testing was conducted according to the steps described above, and the changes in relative fluorescence intensity were recorded (Table 27). Linear discriminant analysis was then used to verify the model's ability to test unknown samples and distinguish bacterial species. Of the 60 unknown samples tested, 51 were correctly detected, achieving an accuracy of 85%.

[0199] Table 27 Fluorescence response data matrix of fluorescence array sensor for clinical urine samples after 2 hours of incubation.

[0200]

Claims

1. A library of electrostatic composite fluorescent array sensor elements, characterized in that, The element library comprises N*M sensing elements with different compositions. Each sensing element is composed of an amino acid-functionalized negatively charged carbon quantum dot and a positively charged antimicrobial peptide through electrostatic interaction. The amino acid-functionalized negatively charged carbon quantum dot is selected from a first set containing N amino acid-functionalized negatively charged carbon quantum dots. The positively charged antimicrobial peptide is selected from a second set containing M positively charged antimicrobial peptides. The amino acids are selected from two or more of the following amino acids: D -Alanine, D -Phenylanine, D -Glutamic acid, D -Diaminopropionic acid, D -Tryptophan, D -Tyrosine, D -serine, D -histidine, D -Aspartic acid D -Viarine; the carbon source in the amino acid-functionalized negatively charged carbon quantum dots is diammonium citrate; in the second component, the antimicrobial peptides are selected from two or more of the following sequences: 。 2. The electrostatic composite fluorescent array sensor element library according to claim 1, characterized in that, In the second component, the N-terminus and C-terminus of the antimicrobial peptide sequence are modified by one or both of the fluorophores, which are independently selected from: FAM, FITC, Cy3, TRITC, Texas Red, Cy5, Alexa Fluor 405, Bodipy, Cy7, Rhodamine B, and AMC.

3. A method for constructing an electrostatic composite fluorescence array sensor comprising the electrostatic composite fluorescence array sensor element library described in any one of claims 1-2, characterized in that, The construction method includes the following steps: (1) Construct a component library including N*M sensing elements: Select N kinds of amino acid functionalized negatively charged carbon quantum dots to form the first component set, and select M kinds of antimicrobial peptides to form the second component set; The N kinds of components in the first component set and the M kinds of components in the second component set are combined to obtain N*M sensing elements with different compositions. (2) Screening of sensing elements: Select the set of bacteria to be identified; add the bacteria to be identified to each sensing element in the corresponding element library, and then measure the fluorescence response of each sensing element to the bacteria. Based on the fluorescence response value, select several sensing elements whose fluorescence response values ​​account for the top 10-50% of all bacteria to be identified as initial screening sensing elements. (3) Then, based on the machine learning algorithm, the sensor elements with a large factor contribution are selected from the above-mentioned initial screening of sensor elements, and the final sensor elements constitute the electrostatic composite fluorescence array sensor.

4. The method for constructing the electrostatic composite fluorescence array sensor according to claim 3, characterized in that, In step (1), the preparation method of amino acid functionalized negatively charged carbon quantum dots is as follows: carbon source and amino acid are synthesized into carbon quantum dots by bottom-up method, and then amino acid functionalized negatively charged carbon quantum dots are obtained by dialysis, freezing and drying. And / or, in step (1), the method for preparing the sensing element is as follows: dilute the amino acid-functionalized negatively charged carbon quantum dots with a buffer solution, then add an antimicrobial peptide aqueous solution, mix evenly, and obtain the sensing unit; And / or, in step (3), the machine learning algorithm is a neural network, support vector machine, decision tree, K-nearest neighbor algorithm, random forest, Gaussian process, branch and bound, logistic regression, principal component analysis or linear discriminant analysis.

5. The application of a fluorescence array sensor constructed by the method of constructing an electrostatic complex fluorescence array sensor according to claim 3 in the detection of bacteria for non-diagnostic and therapeutic purposes.

6. The application according to claim 5, characterized in that, The application method is as follows: different sensing units of the fluorescence array sensor are mixed and incubated with the bacteria to be detected, and the fluorescence intensity of each sensing unit is tested from 0 h to 24 h. The fluorescence data is processed and classified according to the data matrix to achieve the identification and differentiation of pathogenic bacteria.

7. The application according to claim 5, characterized in that, If the bacteria to be detected are water-borne bacteria, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP60. And / or, if the bacteria to be detected are serological bacteria, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60, and Dap-CDots-AMP3. And / or, if the bacteria to be detected are bacteria in the urine system, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, His-CDots-AMP4, Val-CDots-AMP60. And / or, if the bacteria to be detected are Staphylococcus aureus of different concentrations, the sensing units included in the electrostatic complex fluorescence array sensor are: Val-CDots-AMP7, Tyr-CDots-AMP7, Trp-CDots-AMP7, Val-CDots-AMP2, and Dap-CDots-AMP7, respectively. And / or, if the bacteria to be detected are different concentrations of Pseudomonas aeruginosa, the sensing units included in the electrostatic complex fluorescence array sensor are: Trp-CDots-AMP7, His-CDots-AMP4, Val-CDots-AMP7, Val-CDots-AMP60, and Dap-CDots-AMP7, respectively. And / or, if the bacteria to be detected are a mixture of bacteria including Escherichia coli and Enterococcus faecalis, the sensing units included in the electrostatic complex fluorescence array sensor are: Tyr-CDots-AMP7, Val-CDots-AMP7, Dap-CDots-AMP7, Dap-CDots-AMP3, Val-CDots-AMP2; And / or, if the bacteria to be detected are bacteria in serum samples from clinical sepsis patients, the sensing units included in the electrostatic complex fluorescence array sensor are His-CDots-AMP4, Tyr-CDots-AMP3, His-CDots-AMP2, Val-CDots-AMP60, and Dap-CDots-AMP3, respectively. And / or, if the bacteria to be detected are bacteria in a urine sample of a clinical urinary tract infection, the sensing units included in the electrostatic complex fluorescence array sensor are: Val-CDots-AMP60, Tyr-CDots-AMP1, Tyr-CDots-AMP3, His-CDots-AMP2, and His-CDots-AMP4.

Images