Animal immune evaluation method based on cat flea infection model

Abstract

The application provides an animal immune evaluation method based on a Lutzomyia longipalpis infection model, and comprises the following steps: S1, experimental animal preparation and screening; S2, Lutzomyia longipalpis preparation and quality verification; S3, Lutzomyia longipalpis infection of experimental animals; S4, immune related index detection; S5, animal immune function evaluation; and S6, repeatability verification and data summarization. The application realizes systematic and objective evaluation of the effect of immune candidate drugs.

Description

Technical Field

[0001] This invention relates to the field of parasite immunology and pharmacology evaluation technology, and in particular to an animal immune evaluation method based on a feline cephalopod infection model. Background Technology

[0002] Traditional animal immune evaluation methods have many limitations. On the one hand, the standardization of infection models is low, and the species purity, viability, and physiological consistency of the source of infection and the host lack strict control, resulting in insufficient model stability. On the other hand, immune evaluation often relies on a single indicator or subjective judgment, making it difficult to capture the nonlinear correlation between immune indicators. This leads to one-sided evaluation and poor reliability of results, and cannot provide accurate and systematic support for the development of immune candidate drugs. Summary of the Invention

[0003] This invention provides an animal immune evaluation method based on a *Ctenophora catina* infection model. First, healthy experimental cats with consistent physiological states are selected to construct a standardized host model. Second, *Ctenophora catina* strains are prepared and validated through both morphological and molecular biological methods to ensure the purity and viability of the infection source. Then, the experimental cats are grouped and a standardized infection model is constructed by co-hospitalizing them with positive cats, while a control group is also set up. Next, samples are collected at key time points to detect multidimensional indicators related to humoral immunity, cellular immunity, and infection control. Then, a four-algorithm fusion approach—PCA dimensionality reduction, surface fitting nonlinear correction, AHP objective weighting, and SVM classification prediction—is used to transform the original indicators into a comprehensive immune score and immune level. Finally, the stability of the results is verified through multiple repeatable experiments, and the data are summarized to form a standardized evaluation report, achieving a systematic and objective assessment of the efficacy of candidate immune drugs.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] An animal immune evaluation method based on a feline ctenoid flea infection model, comprising:

[0006] S1. Preparation and screening of experimental animals: Healthy domestic cats aged 1-3 years were selected. Qualified individuals were screened for external parasites, assessed for clinical symptoms, and tested for complete blood count, blood biochemistry, and Toxoplasma gondii antibodies. Baseline health data were recorded, and a standardized host model was constructed.

[0007] S2. Preparation and quality verification of cat fleas: screen positive cats with ≥4 live fleas on their bodies and negative for Toxoplasma gondii antibodies. Morphological and molecular biological identification of fleas collected from the bodies of positive cats is performed. Female adult fleas that meet the standards for motility, morphological integrity and blood-sucking potential are selected to form a standardized source of infection and a quality verification report is issued.

[0008] S3. Infect experimental animals with Ctenocephalides felis: Randomly divide the healthy cats qualified in step S1 into a blank control group, an infected control group, and an immune candidate drug group. The blank control group is separately raised in an environment free of flea contamination. The infected control group is co-housed with the positive cats in step S2. The immune candidate drug group is given the immune candidate drug according to body weight 24 hours before co-housing with the positive cats and then co-housed; during co-housing, monitor the number of live fleas, clinical symptoms, and blood indexes of each group of cats every week to verify whether the infection model meets the preset standards;

[0009] S4. Detection of immune-related indexes: Collect blood samples of each group of cats at the preset time points after infection and immune organ samples at some time points, detect humoral immune indexes, cellular immune indexes, and infection control effect indexes, record the original detection data and perform quality control;

[0010] S5. Evaluation of animal immune function: First, extract the principal components with a cumulative variance contribution rate ≥ 85% from the detection data in step S4 through principal component analysis to reduce dimensions and remove noise; then use a quadratic polynomial surface fitting model to perform non-linear correction on the extracted principal components to obtain the comprehensive principal component score; objectively assign weights to the corrected comprehensive principal component score and the extracted principal components through the analytic hierarchy process based on the variance contribution rate and correlation coefficient, and calculate the corrected immune comprehensive score; finally, use a support vector machine model to map the corrected immune comprehensive score to an immune grade;

[0011] S6. Reproducibility verification and data summary: Conduct 3 independent repeated experiments according to steps S1 - S5, control variables such as experimental animals, Ctenocephalides felis, reagents, instruments, and operators, perform consistency tests on the immune grade results of the 3 experiments, and summarize the whole-process data to form an evaluation report.

[0012] In this specification, in step S1, the indexes for blood routine detection include red blood cells, hemoglobin, white blood cells, lymphocyte ratio, hematocrit, mean red blood cell volume, mean red blood cell hemoglobin content, mean red blood cell hemoglobin concentration, neutrophil count and ratio, monocyte count and ratio, eosinophil count and ratio, basophil count and ratio, and platelets, a total of 14 items; the indexes for blood biochemical detection include total protein, albumin, glucose, urea, alanine aminotransferase, aspartate aminotransferase, and creatinine, a total of 7 items; the detection of toxoplasma antibody uses a commercial ELISA kit, and the result is considered qualified when the OD value < 0.3.

[0013] In this instruction manual, step S2, molecular biological identification, involves PCR detection of the mitochondrial COI gene in *Ctenophora catina*. Specifically, this includes: extracting flea genomic DNA using a blood / cell / tissue genomic DNA extraction kit; constructing a 25 μL PCR reaction system using upstream primer 5'-AGA ATT AGG TCA ACC AGG A-3' and downstream primer 5'-GAA GGG TCA AAG AAT GAT GT-3'; performing PCR according to a program of 35 cycles: 94℃ pre-denaturation for 5 minutes, 94℃ denaturation for 30 seconds, 55℃ annealing for 30 seconds, 72℃ extension for 40 seconds, and a final extension at 72℃ for 10 minutes; performing agarose gel electrophoresis on the products; and confirming species homozygosity if a band of approximately 600 bp appears and the sequencing result shows ≥99% homology with the *Ctenophora catina* COI gene sequence in GenBank.

[0014] In this instruction manual, in step S3, after grouping, it is necessary to verify the differences between the three groups of cats in baseline weight and baseline blood indicators (white blood cell and lymphocyte ratio, red blood cell and hemoglobin in blood routine tests and total protein, glucose and urea in blood biochemistry). One-way ANOVA is used to determine that the differences are not statistically significant (P>0.05) and the groups are considered balanced. The infection model validation criteria include: the infection rate of the infection control group is ≥75%, 87.5%, 100% and 100% respectively from week 4 to week 7; the average number of live fleas is ≥14 in week 7; ≥70% of the cats show itching and ≥25% of the cats show rashes in week 7; at least 3 blood indicators are abnormal in week 7; and the blank control group has no flea infection and the indicators are normal throughout the process.

[0015] In this instruction manual, in step S4, the humoral immune indicator is the specific IgG titer of *Ctenophora catina*, which is detected using an indirect ELISA method: *Ctenophora catina* is lysed by sonication with RIPA lysis buffer, and the supernatant after centrifugation is used as the antigen. It is diluted with carbonate buffer to 2 μg / mL to coat the ELISA plate. After blocking, plasma samples are added in a gradient of 1:100 to 1:12800, and then HRP-labeled goat anti-cat IgG secondary antibody is added. After color development, the maximum dilution factor when the OD value is ≥0.2 and twice that of the negative control is taken as the IgG titer.

[0016] In this specification, during step S5 of the principal component analysis, the five immune-related indicators are first constructed into an original matrix and standardized. The covariance matrix of the standardized indicators is then calculated, and the eigenvalues ​​and corresponding eigenvectors of the covariance matrix are solved. Principal components are selected based on a cumulative variance contribution rate ≥85%. The five immune-related indicators include feline ctenoid flea-specific IgG titer, CD4+, and other parameters. + / CD8 + T cell ratio, IFN-γ concentration, flea survival rate, and blood-feeding rate.

[0017] In this specification, in step S5, the quadratic polynomial expression of the surface fitting model includes a constant term, a first-order term of the principal components, a quadratic term of the principal components, and a cross term between the principal components. The model coefficients are solved using the least squares method to determine the coefficients R. 2 A value ≥0.85 indicates a satisfactory fit.

[0018] In this specification, during step S5, when assigning weights using the analytic hierarchy process (AHP), the overall principal component score and the variance contribution rate and Pearson correlation coefficient of each principal component with the true value of the immune effect are calculated separately. The variance contribution rate and Pearson correlation coefficient are then standardized. Finally, the standardized variance contribution rate and Pearson correlation coefficient are merged with equal weights to obtain the objective weights of each evaluation object. The true value of the immune effect is calculated based on the standardized values ​​of five immune-related indicators and the weights determined by information entropy.

[0019] In this specification, in step S5, the support vector machine model uses a radial basis kernel function. The penalty coefficient C and kernel parameter γ are optimized by combining grid search with 5-fold cross-validation to determine the optimal parameters with a classification accuracy of ≥90%. A binary classifier is constructed using a "one-to-one" strategy, and the final immune level is determined by voting. The immune level is divided into four levels: excellent, good, medium, and poor. Sample labels are labeled based on objective thresholds of five immune-related indicators.

[0020] In this manual, in step S6, the consistency test of repeatability experiments uses the Kappa coefficient. When the Kappa coefficient is ≥0.80, the consistency of the three experimental results is considered to be excellent. If the experimental results of a certain batch are inconsistent with those of other batches, two more repeated experiments are added, and the majority result is taken as the final evaluation result. The reagents used in the three repeated experiments are from the same batch. The instrument is calibrated before each use, and the core operations are performed by designated personnel.

[0021] In summary, the present invention has at least the following beneficial effects:

[0022] A standardized host model and source of infection were constructed, ensuring the consistency and stability of the experimental basis and reducing the interference of irrelevant factors on the evaluation results.

[0023] By integrating multi-dimensional immune indicator detection and combining multiple algorithms, a quantitative evaluation of immune function has been achieved, solving the problems of subjectivity and one-sidedness in traditional evaluation.

[0024] A complete process solution of "experimental preparation - infection modeling - indicator detection - algorithm evaluation - repeatability verification" has been formed, which improves the accuracy, repeatability and traceability of immune evaluation.

[0025] It provides a scientific and reliable technical means for evaluating the efficacy of candidate immune drugs, and helps the research and development process of parasite-related immune drugs. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the animal immune evaluation method based on the feline cephalopod infection model involved in this invention.

[0027] Figure 2 This is a schematic diagram illustrating the morphological identification of the cat flea involved in this invention.

[0028] Figure 3 This is a schematic diagram of PCR identification of the cat flea involved in this invention. Detailed Implementation

[0029] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] like Figure 1 As shown, this embodiment provides an animal immune evaluation method based on a feline ctenoid flea infection model, including:

[0031] S1. Preparation and screening of experimental animals: Healthy domestic cats aged 1-3 years were selected. Qualified individuals were screened for external parasites, assessed for clinical symptoms, and tested for complete blood count, blood biochemistry, and Toxoplasma gondii antibodies. Baseline health data were recorded, and a standardized host model was constructed.

[0032] S2. Preparation and quality verification of cat fleas: screen positive cats with ≥4 live fleas on their bodies and negative for Toxoplasma gondii antibodies. Morphological and molecular biological identification of fleas collected from the bodies of positive cats is performed. Female adult fleas that meet the standards for motility, morphological integrity and blood-sucking potential are selected to form a standardized source of infection and a quality verification report is issued.

[0033] S3. Experimental animals infected with *Ctenopharynx felis*: Healthy cats that passed the screening in step S1 were randomly divided into a blank control group, an infection control group, and an immunization candidate drug group. The blank control group was housed alone in an environment free of flea contamination. The infection control group was housed together with the positive cats from step S2. The immunization candidate drug group was given the immunization candidate drug according to body weight 24 hours before being housed with the positive cats. During the housed period, the number of live fleas, clinical symptoms, and blood indicators of each group of cats were monitored weekly to verify whether the infection model met the preset criteria.

[0034] S4. Detection of immune-related indicators: Blood samples and immune organ samples were collected from each group of cats at preset time points after infection. Humoral immune indicators, cellular immune indicators and infection control effect indicators were detected. Raw test data were recorded and quality control was performed.

[0035] S5. Animal immune function evaluation: Principal component analysis was first used to extract principal components with a cumulative variance contribution rate ≥85% from the detection data in step S4 for dimensionality reduction and noise reduction. Then, a quadratic polynomial surface fitting model was used to perform nonlinear correction on the extracted principal components to obtain a comprehensive principal component score. Based on the variance contribution rate and correlation coefficient, the analytic hierarchy process was used to assign objective weights to the corrected comprehensive principal component score and the extracted principal components to calculate the corrected comprehensive immune score. Finally, a support vector machine model was used to map the corrected comprehensive immune score to the immune level.

[0036] S6. Repeatability verification and data summary: Perform three independent replicate experiments according to steps S1-S5, controlling variables such as experimental animals, cat fleas, reagents and instruments, and operators. Verify the consistency of the immunization grade results of the three experiments, and summarize the data of the entire process to form an evaluation report.

[0037] In some embodiments, in step S1, the indicators for routine blood tests include 14 items: red blood cells, hemoglobin, white blood cells, lymphocyte percentage, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin content, mean corpuscular hemoglobin concentration, neutrophil count and percentage, monocyte count and percentage, eosinophil count and percentage, basophil count and percentage, and platelets; the indicators for blood biochemistry tests include 7 items: total protein, albumin, glucose, urea, alanine aminotransferase, aspartate aminotransferase, and creatinine; the Toxoplasma gondii antibody test uses a commercially available ELISA kit, and a result with an OD value <0.3 is considered acceptable.

[0038] In some embodiments, step S2, molecular biological identification, specifically PCR detection of the mitochondrial COI gene of *Ctenophora catina*, includes: extracting flea genomic DNA using a blood / cell / tissue genomic DNA extraction kit; constructing a 25 μL PCR reaction system using upstream primer 5'-AGA ATT AGG TCA ACC AGG A-3' and downstream primer 5'-GAA GGG TCA AAG AAT GAT GT-3'; performing PCR according to a program of 35 cycles: 94℃ pre-denaturation for 5 minutes, 94℃ denaturation for 30 seconds, 55℃ annealing for 30 seconds, 72℃ extension for 40 seconds, and a final extension at 72℃ for 10 minutes; performing agarose gel electrophoresis on the product; and confirming species homozygosity if a band of approximately 600 bp appears and the sequencing result shows ≥99% homology with the *Ctenophora catina* COI gene sequence in GenBank.

[0039] In some embodiments, in step S3, after grouping, it is necessary to verify the differences among the three groups of cats in baseline weight and baseline blood indicators (white blood cell and lymphocyte ratio, red blood cell and hemoglobin in blood routine tests and total protein, glucose and urea in blood biochemistry). When the difference is not statistically significant (P>0.05) by one-way ANOVA, it is considered that the groups are balanced. The infection model validation criteria include: the infection rate of the infection control group is ≥75%, 87.5%, 100% and 100% respectively from week 4 to week 7; the average number of live fleas is ≥14 in week 7; ≥70% of the cats have itching and ≥25% of the cats have rashes in week 7; at least 3 blood indicators are abnormal in week 7; and the blank control group has no flea infection and the indicators are normal throughout the process.

[0040] In some embodiments, in step S4, the humoral immune indicator is the specific IgG titer of *Ctenophora catina*, which is detected by indirect ELISA: *Ctenophora catina* is lysed by sonication with RIPA lysis buffer, and the supernatant after centrifugation is used as the antigen. It is diluted with carbonate buffer to 2 μg / mL to coat the ELISA plate. After blocking, plasma samples are added in a gradient of 1:100 to 1:12800, and then HRP-labeled goat anti-cat IgG secondary antibody is added. After color development, the maximum dilution factor when the OD value is ≥0.2 and twice the OD value of the negative control is used as the IgG titer.

[0041] In some embodiments, during step S5, the principal component analysis first constructs an original matrix of five immune-related indicators and performs standardization. The covariance matrix of the standardized indicators is then calculated, and the eigenvalues ​​and corresponding eigenvectors of the covariance matrix are solved. Principal components are selected based on a cumulative variance contribution rate ≥85%. The five immune-related indicators include feline flea-specific IgG titer, CD4+, and other parameters. + / CD8 + T cell ratio, IFN-γ concentration, flea survival rate, and blood-feeding rate.

[0042] In some embodiments, in step S5, the quadratic polynomial expression of the surface fitting model includes a constant term, a first-order term of the principal components, a quadratic term of the principal components, and a cross term between the principal components. The model coefficients are solved using the least squares method to determine the coefficients R. 2 A value ≥0.85 indicates a satisfactory fit.

[0043] In some embodiments, during step S5, when assigning weights using the analytic hierarchy process, the overall principal component score and the variance contribution rate and Pearson correlation coefficient of each principal component with the true value of the immune effect are calculated separately. The variance contribution rate and Pearson correlation coefficient are standardized respectively, and then the standardized variance contribution rate and Pearson correlation coefficient are merged with equal weights to obtain the objective weight of each evaluation object. The true value of the immune effect is calculated based on the standardized values ​​of five immune-related indicators and the weights determined by information entropy.

[0044] In some embodiments, in step S5, the support vector machine model uses a radial basis kernel function, and optimizes the penalty coefficient C and kernel parameter γ by combining grid search with 5-fold cross-validation to determine the optimal parameters with a classification accuracy of ≥90%; a binary classifier is constructed using a "one-to-one" strategy, and the final immune level is determined by voting. The immune level is divided into four levels: excellent, good, medium, and poor. Sample labels are labeled based on objective thresholds of five immune-related indicators.

[0045] In some embodiments, in step S6, the consistency test of the repeatability experiment uses the Kappa coefficient. When the Kappa coefficient is ≥0.80, the consistency of the three experimental results is considered to be excellent. If the experimental results of a certain batch are inconsistent with those of other batches, two more repeated experiments are added and the majority result is taken as the final evaluation result. The reagents used in the three repeated experiments are from the same batch. The instrument is calibrated before each use, and the core operations are performed by fixed personnel.

[0046] The technical concept of this invention is as follows:

[0047] The animal immune evaluation method based on the feline flea infection model achieves a systematic and objective assessment of the efficacy of candidate immune drugs through a complete process design of "experimental preparation - infection modeling - index detection - algorithm evaluation - replication validation". The specific steps (S1-S6) are as follows:

[0048] S1: Preparation and screening of laboratory animals (construction of standardized host models)

[0049] The core purpose of this step is to select healthy cats with consistent physiological conditions and no underlying diseases, laying the foundation for the stability of the subsequent infection model and the reliability of immune indicators. The specific operation is as follows:

[0050] 1.1 Determination of Experimental Animal Species and Specifications

[0051] Breed selection: Select healthy domestic cats aged 1-3 years (male or female), with preference given to short-haired breeds (such as American Shorthair and British Shorthair) to facilitate subsequent flea counting and inspection;

[0052] Source requirements: All animals must be purchased from units with laboratory animal production licenses and laboratory animal use licenses, and must provide animal pedigree certificates (including date of birth and genetic background) and health quarantine reports (excluding four core infectious diseases such as feline panleukopenia, feline coronavirus, feline herpesvirus type I, and feline calicivirus).

[0053] Physiological specifications: Weight should be controlled between 3.0 and 4.5 kg (to avoid poor tolerance to infection due to being underweight and uneven immune response due to being overweight), and the birth dates should not differ by more than 1 month (to reduce the impact of age differences on immune function).

[0054] 1.2 Adaptive Feeding and Environmental Control

[0055] Rearing conditions: Individual cage rearing (cage size 80cm×60cm×50cm), with sterile absorbent bedding material laid in the cage (changed once a day), and a dedicated food bowl and water bowl provided (cleaned and disinfected daily).

[0056] Environmental parameters: Temperature 22-25℃ (daily fluctuation ≤±1℃), relative humidity 45%~55% (automatically controlled by dehumidifier / humidifier), light cycle 12h light (08:00-20:00) / 12h darkness (to avoid light disturbance affecting immune rhythm);

[0057] Feeding and management: Feed commercial complete cat food (crude protein ≥28%, crude fiber ≤5%, batch number must be consistent), free drinking water (purified water disinfected by ultraviolet light), acclimatization feeding for 1 week, during which time observe the animal’s activity level (such as active feeding, climbing frequency), coat condition (smooth and close-fitting without shedding) and fecal form (formed without diarrhea) daily, record abnormal individuals and remove them in time.

[0058] 1.3 Multi-dimensional health screening

[0059] After the acclimatization period, all cats were screened as follows, and only fully qualified individuals were retained:

[0060] Screening for external parasites: Two laboratory personnel cooperated, one restraining the cat (wearing anti-scratch gloves) and the other combing the entire body against the grain with a fine-toothed comb (in the order of "head → neck → back → abdomen → limbs → tail"), while examining with a dermatoscope (magnification ×40) to confirm the absence of external parasites such as fleas, mites, and ticks.

[0061] Clinical symptom assessment: Observe mental status (active and not lethargic), mucosal color (conjunctiva and gums are light pink), and respiratory rate (15-30 breaths / minute). Exclude individuals with abnormal symptoms such as sneezing, runny nose, and skin redness and swelling.

[0062] Blood test results:

[0063] Blood collection method: Blood was collected from the hind limb veins. 0.5 mL was collected in an anticoagulant tube (containing EDTA-K2) for complete blood count, and 1 mL was collected in a non-anticoagulant tube for blood biochemistry.

[0064] Complete blood count (CBC) test (instrument model: Mindray BC-6800Vet): Detects red blood cells (RBC, normal range 5.5-10.0×10⁻⁶). 12 / L), hemoglobin (HGB, 90-150 g / L), white blood cell count (WBC, 5.5-19.5 × 10⁻⁶ g / L), and white blood cell count (WBC, 5.5-19.5 × 10⁻⁶ g / L). 9The following parameters were measured: red blood cell count ( / L), lymphocyte percentage (LYM%, 20%–50%), hematocrit (HCT, 30%–50%), mean corpuscular volume (MCV, 40–55 fL), mean corpuscular hemoglobin (MCH, 12–18 pg), mean corpuscular hemoglobin concentration (MCHC, 300–360 g / L), and neutrophil count (NEU, 2.0–12.0 × 10⁻¹⁰). 9 (Neu percentage) and proportion (NEU%, 30%–70%), monocyte count (MONO, 0.1–1.0 × 10⁻⁶ / L), monocyte count (MONO, 0.1–1.0 × 10⁻⁶ / L). 9 / L) and proportion (MONO%, 2%–8%), eosinophil count (EOS, 0.1–1.5 × 10 9 / L) and proportion (EOS%, 1%~10%), basophil count (BASO, 0-0.1×10 9 / L) and proportion (BASO%, 0%~1%), platelets (PLT, 100-500×10 9 ( / L), a total of 14 indicators, all of which are considered qualified if they are within the reference range;

[0065] Blood biochemistry test (instrument model: Hitachi 7180): Detects total protein (TP, 55-85 g / L), albumin (ALB, 25-45 g / L), glucose (GLU, 3.9-8.3 mmol / L), urea (UREA, 2.5-10.0 mmol / L), alanine aminotransferase (ALT, 10-100 U / L), aspartate aminotransferase (AST, 15-40 U / L), and creatinine (CREA, 44-159 μmol / L), a total of 7 indicators. All indicators within the reference range are considered qualified.

[0066] Toxoplasma gondii antibody detection: Serum Toxoplasma gondii IgG antibodies were detected using a commercial ELISA kit (brand: IDEXX). A negative result (OD value <0.3) was considered acceptable (to avoid cross-infection of the infection model by pathogens).

[0067] 1.4 Recording of health baseline data

[0068] Select qualified experimental cats (e.g., C01-C24), record the following baseline data, and form an experimental animal health baseline table:

[0069] Basic information: ID number, gender, age, weight, breed;

[0070] Blood parameters: specific values ​​of complete blood count (RBC, HGB, WBC, LYM%, HCT, MCV, MCH, MCHC, NEU, NEU%, MONO, MONO%, EOS, PLT) and blood biochemistry (TP, GLU, UREA, ALT, AST, CREA);

[0071] Body surface condition: hair integrity, skin color, presence or absence of dandruff.

[0072] S2: Preparation and quality verification of *Ctenopharynx felis* (construction of a standardized source of infection)

[0073] The core objective of this step is to obtain cat fleas that are highly pure, viable, and of a clearly defined species, in order to avoid instability in the infection model due to flea quality issues. The specific procedures are as follows:

[0074] 2.1 Screening and confirmation of flea-positive cats

[0075] Screening scope: In conjunction with stray animal shelters and veterinary hospitals, flea tests were conducted on 60 stray cats and cats awaiting adoption.

[0076] Initial screening method: Use a fine brush to comb the cat's body surface, collect the fallen fleas, and place them in a centrifuge tube containing 75% alcohol. If the number of live fleas on the body surface of a single cat is ≥4, it is considered a "suspected positive cat".

[0077] Toxoplasmosis exclusion: For suspected positive cats, venous blood (0.5 mL) was collected and Toxoplasma gondii IgG antibody was detected using an ELISA kit. Only individuals with negative antibodies were retained (a total of 4 positive cats were screened out, numbered F01-F04, with 4, 8, 9, and 16 live fleas respectively).

[0078] Cats with positive fleas should be kept separately (cage as in S1.2), with bedding changed daily and cat food from the same batch provided. Stress should be avoided to prevent fleas from surviving.

[0079] 2.2 Flea species identification (morphological + molecular biological dual verification)

[0080] 2.2.1 Morphological identification

[0081] Sample processing: Three live fleas were collected from the body surface of each positive cat and fixed in 75% alcohol for 24 hours, and then transferred to 50% glycerol solution for clearing treatment for 48 hours (to make the flea structure clear).

[0082] Microscopic observation: After the fleas have become transparent, transfer them to a glass slide, add 1 drop of glycerin, cover with a coverslip, and observe the following characteristics under an optical microscope (×100 magnification):

[0083] Female flea: The head tip forms an acute angle with the lower edge of the cheek. The cheek comb contains 8 bristles (the first bristle is 4 / 5 the length of the second bristle). The metathorax has 1-2 bristles. The last cut of the hind tibia is only a shallow cut. The head of the spermatheca is broad and round.

[0084] Male fleas: The end of the clasper stalk is not enlarged or is slightly enlarged, and the structure of the clasper at the end of the abdomen is consistent with the characteristics of the cat flea;

[0085] Morphological observation results as follows Figure 2 As shown, all the fleas obtained exhibited the typical morphological characteristics of the aforementioned cat flea;

[0086] Identification criteria: Based on the Chinese Atlas of Morphological Classification of Livestock and Poultry Parasites (Huang Bing et al., 2006) and Small Animal Parasitology (Zhu Xingquan, 2006), if all observed characteristics match, it is preliminarily identified as Ctenocephalides felis.

[0087] 2.2.2 Molecular biological identification (PCR detection of mitochondrial COI gene)

[0088] DNA extraction: Select morphologically qualified fleas (one flea per positive cat), cut the flea body into pieces with sterile scissors, and use the blood / cell / tissue genomic DNA extraction kit (brand: Tiangen DP304) to extract genomic DNA according to the instructions. Measure the OD260 / OD280 ratio (1.8-2.0 is considered qualified).

[0089] PCR reaction system (25 μL): DNA template 2 μL, upstream primer (Cff-F: 5'-AGA ATT AGG TCA ACCAGG A-3', 10 μmol / L) 1 μL, downstream primer (Cff-R: 5'-GAA GGG TCA AAG AAT GAT GT-3', 10 μmol / L) 1 μL, 2×Taq PCR MasterMix 12.5 μL, ddH2O 8.5 μL;

[0090] PCR reaction procedure: 94℃ pre-denaturation for 5 minutes; 94℃ denaturation for 30 seconds, 55℃ annealing for 30 seconds, 72℃ extension for 40 seconds, for a total of 35 cycles; 72℃ final extension for 10 minutes, store at 4℃.

[0091] Result determination: Take 5 μL of PCR product and perform 1.5% agarose gel electrophoresis (120V, 20 minutes). If a band of about 600 bp appears (consistent with the expected target fragment of 552 bp), and the sequencing result shows ≥99% homology with the COI gene sequence of *Ctenophora catinatum* in GenBank (accession number KP231121.1), then the species is confirmed to be homozygous.

[0092] 2.3 Screening and Preservation of Flea Viability

[0093] Viability screening: All live fleas were collected from the bodies of 4 positive cats under a stereomicroscope (×40). Screening criteria:

[0094] Locomotion: After being lightly touched, it will quickly crawl or jump within 3 seconds;

[0095] Morphological integrity: The body wall is intact, and the appendages (antennae, feet) are complete and without deformities;

[0096] Blood-sucking potential: The female flea's abdomen is moderately plump (not excessively shriveled or swollen).

[0097] Screening results: Only female adult fleas were retained (male adult fleas have a low blood-sucking frequency and weak immune stimulation to the host), and the number after screening was prepared according to "20 qualified female fleas per experimental cat";

[0098] Short-term storage: Divide qualified female fleas into 1.5mL centrifuge tubes (20 fleas per tube), place moist sterile filter paper at the bottom of the tube (to maintain humidity), and store temporarily in a refrigerator at 4℃ (storage time ≤2 hours to avoid decrease in viability).

[0099] 2.4 Flea Quality Verification Report

[0100] The following data should be compiled to form a quality verification report for cat fleas. Only after confirming that the fleas are qualified can they be used for infection testing:

[0101] Species identification results: morphological characteristic matching table, PCR electrophoresis image (e.g.) Figure 3 (as shown) and sequencing alignment results;

[0102] Viability screening data: number of qualified female fleas, viability pass rate (number of qualified fleas / total number collected × 100%, must be ≥80%);

[0103] Pathogen exclusion: Negative report for positive feline toxoplasmosis antibody test.

[0104] S3: Infection of experimental animals with *Ctenopharynx felis* (construction of a standardized infection model)

[0105] The core objective of this step is to establish a stable infection rate of feline ctenoid fleas in the experimental cats through a "mixed-species rearing" method, while simultaneously setting up a control group to provide a benchmark for subsequent immunization evaluation. The specific procedures are as follows:

[0106] 3.1 Grouping of experimental animals (randomized design)

[0107] The 24 healthy cats that passed the S1 screening were randomly divided into 3 groups (8 cats in each group) using a random number table method to ensure that the initial conditions of the groups were consistent:

[0108] Blank control group (G1): Cats were kept alone in isolation cages free from flea contamination, without contact with positive cats and fleas, and only underwent routine feeding and testing to exclude the interference of environmental factors on immune indicators.

[0109] Infection control group (G2): Cats were housed together with flea-positive cats selected by S2 (2 experimental cats were paired with 1 positive cat, for a total of 4 mixed-breeding cages), without any immune intervention, to reflect the immune response under natural infection conditions;

[0110] Immunization candidate drug group (G3): Cats were housed together with positive cats (same grouping as G2). Twenty-four hours prior to housekeeping, an immunization candidate drug (such as recombinant feline ctenoid flea antigen vaccine) was administered orally or intraperitoneally at 10 mg / kg of body weight. The efficacy of the drug intervention was then evaluated. The relationship between the body weight of the experimental cats in each group and the dosage of the drug is shown in Table 1.

[0111] Table 1. Correspondence between the weight of experimental cats in each group and the dosage of medication.

[0112] ;

[0113] 3.2 Group Balance Verification

[0114] After grouping, the following indicators of the three groups of cats were tested and compared to confirm that there were no statistically significant differences between the groups (P>0.05):

[0115] Basal body weight: mean ± standard deviation for each group, one-way ANOVA;

[0116] Baseline blood parameters: intergroup mean comparison of complete blood count (WBC, LYM%, RBC, HGB) and blood biochemistry (TP, GLU, UREA);

[0117] Body surface condition: No parasites or skin abnormalities were found in any group.

[0118] 3.3 Infection procedures and dynamic monitoring (for 7 consecutive weeks)

[0119] 3.3.1 Environmental Control for Mixed Culture

[0120] Mixed-species cage: 120cm×80cm×60cm, with built-in scratching post and cat bed (to be changed twice a week), and good ventilation;

[0121] Positive cat management: Each positive cat should be used in only one mixed-species cage. If skin redness and swelling occur (indicating a host immune response), decreased flea activity (slow crawling), a sudden decrease in flea numbers (≥5 fewer fleas per day), or hair loss in the host cat (localized area ≥1cm²), the cat will be reported immediately. 2 In case of any of the following four abnormal situations, immediately replace the cat with a new positive cat (selected from the S2 backup positive cats);

[0122] Monitoring frequency: Starting from the first week of mixed breeding, “flea count + clinical symptom observation + blood sampling” will be carried out once every Monday morning (recorded as D0 before mixed breeding, and W1-W7 after mixed breeding).

[0123] 3.3.2 Live flea counting (standardized procedure)

[0124] Restraint method: One person wraps the cat's body with a towel (leaving its head and limbs exposed) to prevent it from scratching;

[0125] Counting areas: Divide the area into 6 regions according to "head (forehead, behind the ears) → neck → back → abdomen → limbs (armpits, groin) → tail". Comb each region against the direction of hair 3 times and collect the live fleas that have fallen off.

[0126] Counting standard: Placed under a stereomicroscope, count only the fleas that are "upright and movable", and record the number of live fleas for each cat;

[0127] The results of live flea counts on the bodies of the cats are shown in Table 2. Starting from the 4th week of mixed rearing, the infection rate and intensity of the infection in the experimental cats increased significantly, with an average of 14.13 live fleas per cat by the 7th week.

[0128] Table 2. Results of live flea counts on the bodies of various cats.

[0129] ;

[0130] Data calculation: The "infection rate" (number of infected cats / total number of cats in the group × 100%) and "average infection intensity" (total number of live fleas in all cats in the group / number of infected cats) are calculated for each group weekly.

[0131] 3.3.3 Clinical symptom observation

[0132] The following symptoms were recorded for each cat to form a clinical symptom record table. The clinical symptom observation results of each cat at different times after infection are shown in Table 3. Itching symptoms began to appear in the 4th week, red rashes appeared in the 6th week, and the symptoms worsened further in the 7th week:

[0133] Table 3. Clinical symptoms observed in different cats at different time points after infection.

[0134] ;

[0135] Skin reactions: Itching (frequent scratching or licking of a certain area, ≥5 times per day is considered positive), erythema (red raised bumps ≥0.5cm in diameter, ≥1 in number is considered positive), hair loss (localized hair loss area ≥1cm). 2 (considered positive)

[0136] Overall condition: mental state (active / listless), appetite (a decrease of ≥20% in daily cat food consumption compared to baseline is considered abnormal), water intake (a change of ≥30% in daily water intake compared to baseline is considered abnormal), mucous membrane color (pale / yellow is considered abnormal).

[0137] 3.3.4 Dynamic monitoring of blood indicators

[0138] Sampling time: D0 (before mixed rearing), W7 (week 7 of mixed rearing);

[0139] Test items: Same as S1.3 blood routine and blood biochemistry indicators, compare changes before and after infection;

[0140] Anomaly detection: If an indicator exceeds the reference range, it is considered abnormal. Record the number of abnormal cats and the type of abnormal indicator in each group. For example, in group G2, record "3 cats with elevated WBC, 1 cat with elevated NEU, 4 cats with decreased RBC, 2 cats with decreased HGB, and 1 cat with decreased PLT".

[0141] 3.4 Infection Model Validation Criteria

[0142] The infection model is considered successfully constructed when all of the following conditions are met:

[0143] Infection rate: The infection rates of G2 group W4-W7 were ≥75%, 87.5%, 100%, and 100% respectively (e.g., W4 was 75% and W7 was 100%).

[0144] Infection intensity: The average number of live fleas in group G2 (W7) was ≥14 (14.13 in this experiment).

[0145] Clinical symptoms: In group G2, ≥70% of cats with W7 showed itching (7 / 8 cats), and ≥25% of cats showed rashes (2 / 8 cats).

[0146] Blood parameters: In the G2 group, at least 3 parameters were abnormal at W7 (e.g., 3 cats had elevated WBC, 1 cat had elevated NEU, 4 cats had decreased RBC, 2 cats had decreased HGB, and 1 cat had decreased PLT).

[0147] Blank control: Group G1 had no flea infection throughout the entire process, and all clinical symptoms and blood indicators were normal.

[0148] If the above conditions are not met, positive cats need to be re-screened or the mixed-species density needs to be adjusted, and the infection process needs to be repeated.

[0149] S4: Detection of immune-related indicators (obtaining multi-dimensional immune data)

[0150] The core purpose of this step is to detect indicators of humoral immunity, cellular immunity, and infection control in the infected experimental cats, providing raw data for subsequent algorithm evaluation. The specific steps are as follows:

[0151] 4.1 Sample Collection (Key Time Points and Methods)

[0152] 4.1.1 Sampling Time Point

[0153] Based on the immune response pattern determined in the preliminary experiments (innate immunity initiates 1 day after infection, adaptive immunity peaks 3–7 days later, and remains stable 14 days later), the following time points were selected:

[0154] Blood samples: 1 day (W0+1d), 3 days (W0+3d), 7 days (W1), 14 days (W2), 21 days (W3), 28 days (W4), 35 days (W5), 42 days (W6), 49 days (W7) after infection;

[0155] Immune organ samples: collected only at 14 days (W2) and 49 days (W7) post-infection (3 cats were randomly selected from each group and euthanized to avoid excessive consumption of samples).

[0156] 4.1.2 Sample Types and Collection Methods

[0157] Blood sample:

[0158] Blood collection site: hind limb vein or jugular vein;

[0159] Blood collection volume: 1.5 mL of blood is collected from each cat each time, of which 0.5 mL is injected into an anticoagulant tube containing EDTA-K2 (for blood cell separation) and 1 mL is injected into a non-anticoagulant tube (for plasma separation).

[0160] Sample processing: Centrifuge blood in anticoagulant tubes at 3000 rpm for 10 minutes (4℃) to separate the upper plasma layer (0.2 mL / tube) and the lower blood cell layer (0.3 mL / tube). After aliquoting, store the plasma at -80℃ (avoid repeated freeze-thaw cycles). Wash the blood cells twice with PBS buffer (pH 7.4) (centrifuge at 3000 rpm for 5 minutes), resuspend in 1 mL of RPMI-1640 medium (containing 10% fetal bovine serum), and store at 4℃ (for testing within 2 hours).

[0161] Immune organ samples:

[0162] Dissection procedure: After euthanizing the cat by cervical dislocation, the spleen and bilateral inguinal lymph nodes were obtained under aseptic conditions.

[0163] Sample processing: Weigh the spleen (normal reference range 0.5–1.0 g), grind it in RPMI-1640 medium on ice using a sterile grinder, filter through a 400-mesh sieve to remove tissue residue, stain with trypan blue and count the percentage of viable cells (≥90% is considered acceptable), and adjust the cell concentration to 1×10⁻⁶. 6 Cells / mL; lymphocytes were bound and weighed (normal reference value 0.1-0.2g), and single-cell suspensions were prepared using the same method.

[0164] 4.2 Core Indicator Testing (5 key indicators, standardized methodology)

[0165] 4.2.1 Humoral immune indicators: Cat Ctenopharynx-specific IgG titer ( )

[0166] Detection method: Indirect ELISA method;

[0167] Antigen preparation: Take 100 cat fleas that passed the S2 screening, add 1 mL of RIPA lysis buffer (containing protease inhibitor), and sonicate on ice (300W power, 3 seconds on, 5 seconds off, 30 times in total), centrifuge at 12000 rpm for 20 minutes (4℃), and take the supernatant as the flea lysis antigen. Determine the protein concentration using the BCA method (adjust to 1 μg / μL), and aliquot and store at -80℃.

[0168] ELISA operation steps:

[0169] 1. Coating: Dilute the antigen to 2 μg / mL with carbonate buffer (pH 9.6), add 100 μL to each well of a 96-well microplate, and incubate overnight at 4°C;

[0170] 2. Blocking: Discard the coating solution, add 200 μL of PBST (0.05% Tween-20) containing 5% skim milk powder to each well, and block at 37°C for 1 hour;

[0171] 3. Sample addition: Dilute plasma samples with PBST at serial dilutions of 1:100, 1:200, 1:400…1:12800, add 100 μL to each well, and incubate at 37°C for 1.5 hours (a negative control: G1 group plasma, and a positive control: known high-titer anti-Ctenopharynx IgG serum).

[0172] 4. Secondary antibody binding: Discard the sample solution, wash the plate 3 times, add 100 μL of HRP-labeled goat anti-cat IgG secondary antibody (1:5000 dilution) to each well, and incubate at 37°C for 1 hour;

[0173] 5. Color development and reading: Wash the plate 5 times, add TMB color development solution (100 μL / well), react at 37℃ in the dark for 15 minutes, stop the reaction by adding 2 mol / L H2SO4 (50 μL / well), and read the OD value at 450 nm using a microplate reader (model: Thermo Multiskan FC).

[0174] Result determination: The highest dilution factor when "OD value ≥ 0.2 and twice the OD value of the negative control" is used as the IgG titer (e.g., 1:800).

[0175] 4.2.2 Cellular immune marker 1: CD4 + / CD8 + T cell ratio ( )

[0176] Detection method: Flow cytometry;

[0177] Reagent preparation: Fluorescently labeled antibody (anti-cat CD3) + -PE, CD4 + -FITC, CD8 + -APC, brand: BDBiosciences, batch number: 20240301), PBS buffer (pH 7.4), flow cytometer sheath fluid;

[0178] Operating steps:

[0179] 1. Take 100 μL of the blood cell suspension prepared in 4.1.2, add antibody mixture (5 μL of each antibody), and stain in the dark for 30 minutes (4℃).

[0180] 2. Add 1 mL of PBS to stop staining, centrifuge at 3000 rpm for 5 minutes, discard the supernatant, and resuspend in 300 μL of PBS;

[0181] 3. Flow cytometry (model: BD FACSCanto II) was used to acquire 10,000 cell events, and CD3 was analyzed using FlowJo software (version 10.8.1). + CD4 + T cells (helper T cells) and CD3 + CD8 + The proportion of T cells (cytotoxic T cells);

[0182] Result calculation: CD4 + / CD8 + Ratio = CD3 + CD4 + T cell percentage / CD3 + CD8 + T cell ratio (normal reference range 1.2-2.0).

[0183] 4.2.3 Cellular immune marker 2: IFN-γ concentration ( )

[0184] Detection method: CBA (Cytometric Bead Array) cytokine detection method;

[0185] Reagent preparation: Feline IFN-γ CBA detection kit (brand: R&D Systems, batch number: 20240215), standards (0-1000 pg / mL), flow cytometer;

[0186] Operating steps:

[0187] 1. Take 1 mL of the spleen single-cell suspension prepared in 4.1.2, add 5 μg / mL flea lysis antigen (same as 4.2.1), and stimulate in a 37℃, 5% CO2 incubator for 24 hours;

[0188] 2. Centrifuge at 12000 rpm for 10 minutes (4℃) and collect the supernatant;

[0189] 3. Following the kit instructions, mix the supernatant with the capture microspheres and detection antibody, and incubate in the dark for 3 hours (room temperature).

[0190] 4. Flow cytometry analysis was performed to calculate the IFN-γ concentration based on the standard curve (normal reference value <50 pg / mL, which increases after infection).

[0191] Quality control requirements: Standard curve R 2 ≥0.99, blank control concentration <5 pg / mL.

[0192] 4.2.4 Infection control effectiveness indicator 1: Flea survival rate ( )

[0193] Calculation method: 48 hours after infection (W0+2d), count the number of surviving fleas according to the method in S3.3.2;

[0194] Formula: Flea survival rate = (Number of surviving fleas / Number of inoculated fleas) × 100% (The number of inoculated fleas is 20 per cat prepared in S2).

[0195] 4.2.5 Infection control effectiveness indicator 2: Blood sucking rate ( )

[0196] Judgment criteria: Observe the intestines of surviving fleas under a stereomicroscope. If the intestines are ≥80% full (dark red, occupying more than 80% of the body cavity), they are judged to have sucked blood.

[0197] Formula: Blood-sucking rate = (Number of fleas that have sucked blood / Number of surviving fleas) × 100%.

[0198] 4.3 Data Recording and Quality Control

[0199] Each batch of tests includes a positive control, a negative control, and a blank control to ensure the effectiveness of the reagents.

[0200] Record the raw test values ​​of all indicators to form a raw data table of immune-related indicators, and mark the test date, instrument model and reagent batch number;

[0201] If the test result of a sample exceeds the normal range by 3 times (e.g., IFN-γ concentration > 1000 pg / mL), it is necessary to resample and test to confirm whether it is due to operational error.

[0202] S5: Evaluation of Animal Immune Function

[0203] The core purpose of this step is to transform the five original indicators of S4 into "Comprehensive Immune Score" and "Immune Level" through a collaborative algorithm of "Principal Component Analysis (PCA) - Surface Fitting (Nonlinear Correction) - Analytic Hierarchy Process (AHP) - Support Vector Machine (SVM)," thereby addressing the problems of subjectivity, bias, and insufficient capture of nonlinear correlations in traditional evaluation methods. The specific operation is as follows:

[0204] 5.1 Algorithm 1: Principal Component Analysis (PCA) – Dimensionality Reduction and Noise Reduction, Extraction of Linear Core Features

[0205] 5.1.1 Purpose

[0206] There may be linear correlations among the five original indicators (such as the negative correlation between IgG titer and flea survival rate). Direct analysis can easily lead to information redundancy. PCA transforms high-dimensional indicators into a few uncorrelated principal components through linear transformation, retaining ≥85% of the original information.

[0207] 5.1.2 Operating Procedures

[0208] 1. Construction of the original indicator matrix: Assume there are n samples in the experiment (3 repeated experiments × 3 groups × 8 cats = 72 samples), and each sample corresponds to 5 indicators ( -IgG titer, -CD4 + / CD8 + , -IFN-γ、 -Flea survival rate -Blood steal rate), construct the n×5 order original matrix X:

[0209] ;

[0210] in Let be the measured value of the j-th indicator for the i-th sample.

[0211] 2. Indicator standardization: Eliminating the influence of dimensions, the formula is:

[0212] ;

[0213] In the formula: (The sample mean of the j-th indicator is referenced from the S1 health baseline data). (Sample standard deviation of the j-th indicator); These are the standardized values ​​(mean 0, standard deviation 1).

[0214] 3. Covariance matrix calculation: describes the linear correlation between standardized indicators, and the formula is: ;in Let Z be the transpose of Z (5×n order), and C be a 5×5 symmetric matrix. ), This represents the covariance between the j-th and k-th indices. >0 indicates a positive correlation. <0 indicates a negative correlation.

[0215] 4. Principal component extraction:

[0216] Find the eigenvalues ​​of the covariance matrix C (The larger the eigenvalue, the more information content the corresponding principal component contains.)

[0217] For each eigenvalue Solve (C- ) =0 (where I is a 5th-order identity matrix), thus obtaining the eigenvectors. , The loading coefficient of the k-th principal component (reflecting the contribution of the j-th index to the k-th principal component);

[0218] The kth principal component The calculation formula is as follows: ;

[0219] Screening criteria: Extract the top p principal components with a cumulative variance contribution rate ≥ 85%. .

[0220] 5. Example of results:

[0221] Eigenvalues: =2.2, =1.5, =0.8, =0.3, =0.2 (total 5);

[0222] Cumulative variance contribution rate: =44%, =74%, =90% (≥85%), therefore the first 3 principal components were selected. ;

[0223] Load factor:

[0224] (Comprehensive immune characteristics): , , , , (Integrating humoral immunity, cellular immunity, and infection control);

[0225] (Cellular immune characteristics): , , , , (Primarily reflects cellular immunity);

[0226] (Humoral immunity-infection association characteristics): , , , , (Primarily reflects humoral immunity and infection control).

[0227] 5.2 Algorithm 2: Surface Fitting Algorithm – Correcting Principal Component Nonlinear Relationships

[0228] 5.2.1 Purpose

[0229] The principal components extracted by PCA are linear combinations, while there may be non-linear associations between immune indicators (such as...). and Synergistic enhancement effect (The marginal benefit diminishes when it is too high). Surface fitting captures this nonlinear relationship through a quadratic polynomial model and outputs the corrected comprehensive principal component score Q, providing a more accurate evaluation object for subsequent AHP.

[0230] 5.2.2 Operating Procedures

[0231] 1. Fitting function definition: A quadratic polynomial surface fitting method is used (balancing nonlinearity capture capability and model simplicity), with the PCA output as the model value. Construct a fitting function with principal components and independent variables:

[0232] ;

[0233] Q: Corrected composite principal component score (core features after incorporating nonlinear relationships); : constant term; : Linear term coefficient (reflecting the linear contribution of the principal components); : Quadratic coefficients (reflecting the nonlinear effects of the principal components themselves, such as...) (Contribution decreases when too high) Cross term coefficients (reflecting synergistic / antagonistic effects between principal components, such as...) and (synergistic effect of immune enhancement).

[0234] 2. Solving for fitting coefficients (least squares method):

[0235] Construct the objective function: using the principal component scores of historical samples ( (From PCA output) and "True Value of Immune Effect" (Based on the objective comprehensive score of the S45 original indicators, formula:) ,in Based on the standardized values ​​of the original indicators (with weights determined according to the indicator information entropy), minimize the sum of squared residuals between the fitted values ​​and the true values: Where N=72 (sample size). is the fitted value for the i-th sample.

[0236] Solving for coefficients in matrix form:

[0237] Define a design matrix X (N×10 order), where each row corresponds to a combination of independent variables for one sample: ;

[0238] Define coefficient vector True value vector ;

[0239] Solve for the coefficients using the least squares solution formula: ;

[0240] 3. Validation of fit: Calculate the coefficient of determination Assess goodness of fit ( The closer to 1, the better the fit. In the formula As the mean of the true values, this invention requires (Actual fitting results) =0.89, qualified).

[0241] 4. Coefficient Determination and Application Output:

[0242] The final fitting coefficients are calculated to be: =2.1, =0.8, =0.6, =0.3, =-0.1, =-0.05, =0, =0.2, =0.1, =0.05; Principal component score for the new sample Substitute into the fitted function to calculate (Example: If) =1.2, =0.8, =0.5, then =2.1 + 0.8 × 1.2 + 0.6 × 0.8 + 0.3 × 0.5 + (-0.1) × 1.2 2+(-0.05)×0.8 2 +0×0.5 2 +0.2×1.2×0.8+0.1×1.2×0.5+0.05×0.8×0.5=4.32). As the core evaluation object of AHP.

[0243] 5.3 Algorithm 3: Analytic Hierarchy Process (AHP) – Objective weight allocation, calculation of the corrected immune comprehensive score

[0244] 5.3.1 Purpose

[0245] Abandoning traditional expert subjective judgment, this study uses the "information content" and "correlation strength" of experimental data to determine the Q and PCA values ​​output by surface fitting. By assigning objective weights, the multi-dimensional features are weighted into a single "corrected immune comprehensive score S'", ensuring the objectivity and scientific nature of the weights.

[0246] 5.3.2 Operational Steps (Objective Weighting Based on "Variance Contribution Rate + Correlation Coefficient")

[0247] 1. Determine the objective evaluation criteria:

[0248] Basis 1: Information Content of Indicators (Variance Contribution Rate) – Reflects the explanatory power of indicators on the original data. The variance contribution rate of Q is calculated using its multiple correlation coefficients with the five original indicators (Formula: ,Right now =89%) The variance contribution rate is directly adopted from the PCA results ( =44%, =30%, =16%, total 90%.

[0249] Basis 2: The strength of the association between the indicator and the immunization effect (Pearson correlation coefficient) – reflects the degree of influence of the indicator on the final evaluation goal. Calculate Q, Correlation coefficient with the true value Y of the immune effect Example results: =0.92, =0.85, =0.78, =0.65.

[0250] 2. Standardization of evaluation criteria:

[0251] Variance contribution rate standardization (eliminating proportional differences): ;

[0252] in The variance contribution rate of the k-th indicator (k=1 corresponds to Q, k=2 corresponds to k=3 corresponds to k=4 corresponds to ), calculated to be:

[0253] ; ; ;

[0254] ;

[0255] Correlation coefficient standardization (dimension elimination): ;

[0256] The calculation yields:

[0257] ; ; ;

[0258] ;

[0259] 3. Calculate the objective portfolio weights:

[0260] The "equal-weight fusion" strategy is adopted (balancing information content and correlation strength to avoid bias from a single criterion), formula: ;

[0261] Example calculation: ; ; ; Weight normalization: ensures that the sum of the weights is 1, and the final weights are: =0.42, =0.26, =0.21, =0.11 (0.42+0.26+0.21+0.11=1).

[0262] 4. Verification of weighting rationality:

[0263] Logical verification: Q has the highest weight (0.42) because it integrates non-linear relationships and has the strongest correlation with Y. (Comprehensive immune characteristics) had the second highest weight (0.26); The (humoral immunity-infection association) had the lowest weight (0.11), which conforms to the immune logic of "non-linear correction index > linear comprehensive index > single association index".

[0264] Data validation: The correlation between weights and indicator information entropy was calculated (the lower the information entropy, the higher the indicator discrimination, and the higher the weight should be). The results showed a correlation coefficient r = -0.88 (negative correlation, as expected), which validated that the weight allocation was reasonable.

[0265] 5. Calculate the corrected immune composite score:

[0266] formula: Example: If a certain sample =4.32, =1.2, =0.8, =0.5, then =0.42×4.32+0.26×1.2+0.21×0.8+0.11×0.5=1.8144+0.312+0.168+0.055=2.3494, The higher the value, the stronger the immune function.

[0267] 5.4 Algorithm 4: Support Vector Machine (SVM) – Classification Prediction, Output Immunity Level

[0268] 5.4.1 Purpose

[0269] The "corrected immune comprehensive score S'" calculated by AHP is mapped to a specific "immunity level" (excellent / good / medium / poor), which solves the subjectivity problem of traditional manual grading. At the same time, because S' integrates non-linear relationships and objective weights, the classification accuracy is significantly improved.

[0270] 5.4.2 Operating Procedures

[0271] 1. Sample label definition:

[0272] Based on the objective threshold of the original S4 index (the optimal cutoff value was determined by ROC curve) and the measured data in Tables 4, 5, and 6, immune grade labels were assigned to 72 historical samples. :

[0273] =1 (Excellent): IgG titer ≥1:800 ( ≥1.2 (standardized value), CD4 + / CD8 + =1.5-2.0 ( ≥1.0), IFN-γ≥100pg / mL ( ≥1.5), flea survival rate ≤30% ( ≤-0.8), Bloodsucking rate ≤20% ≤-0.9 (all 5 conditions are met, corresponding to the status of the immune candidate drug group 7 days after administration in Table 6); =2 (Good): Meets 4 indicators (corresponding to the status of the candidate immune drug group 2 days after administration in Table 6); =3 (medium): Meets 3 indicators (corresponding to the status of some cats in the infection-free group in Table 6); =4 (Difference): Satisfying ≤2 indicators (corresponding to the status of most cats in the infection-free group in Table 6); The live flea count data of experimental cats before and after drug administration are shown in Table 4. The number of fleas in the immune candidate drug group decreased significantly after drug administration, while there was no significant change in the infection-free group:

[0274] Table 4. Live flea counts in experimental cats before and after drug administration.

[0275] ;

[0276] *: Within the same group, different superscript letters indicate significant differences (P<0.05), while the same superscript letters indicate no significant differences (P>0.05).

[0277] Table 5 shows the clinical symptom observation results of experimental cats before and after drug administration. In the immune candidate drug group, the symptoms gradually disappeared after drug administration, while the symptoms persisted in the infection-free group.

[0278] Table 5. Clinical symptom observation results in experimental cats before and after drug administration.

[0279] ;

[0280] Forming a training set ,in The corrected composite score calculated for AHP.

[0281] 2. Model building and parameter optimization:

[0282] Kernel function selection: Since S' contains nonlinear information, the radial basis function (RBF) is used to improve classification ability. The kernel function formula is as follows: ;in For kernel parameters (controlling nonlinear mapping capability);

[0283] Objective function optimization: Minimize structural risk (balancing classification accuracy and generalization ability):

[0284] ; ( Slack variables; C is the penalty coefficient, which controls the severity of the penalty for misclassified samples.

[0285] Parameter determination: through grid search (C∈[1,5,10,20,50], Using the 5-fold cross-validation (∈[0.1,0.5,1,2]), the optimal parameter is determined to be C=10. =0.5, at which point the model accuracy reaches 96%.

[0286] 3. Multi-classification strategies and decision-making:

[0287] A "one-to-one" strategy was used to construct six binary classifiers (1vs2, 1vs3, 1vs4, 2vs3, 2vs4, 3vs4). The efficacy evaluation results at each time point after drug administration are shown in Table 6. All immune candidate drug groups were effective 7 days after drug administration, while the infection-free group was ineffective throughout the course of treatment. This can be used as a validation basis for SVM classification.

[0288] Table 6. Results of efficacy evaluation at various time points after drug administration

[0289] ;

[0290] Corrected composite score for the new sample Each classifier outputs a prediction, and the final ranking is determined by a voting method (the ranking with the most votes is the result; example:) When the value is 3.05, the voting result is "1", which means the immunity level is "excellent".

[0291] 5.5 Four-Algorithm Fusion Closed Loop and Verification

[0292] 5.5.1 Closed-loop logic

[0293] This forms a complete closed loop: "data input → linear dimensionality reduction → nonlinear correction → objective weight quantification → classification output → effect feedback".

[0294] Input: 5 raw metrics for S4;

[0295] Intermediate process: PCA ( → Surface fitting (Q) → AHP (Objective Weights → S') → SVM (Immunity Grade);

[0296] Feedback: The immune rating is used to assess the effectiveness of candidate immune drugs (e.g., "Excellent" indicates that the drug can significantly enhance humoral immunity, cellular immunity and control infection).

[0297] Data Loop: The training data for SVM comes from S' and the previous experiments. The training set will be continuously supplemented with new experimental data to improve the model's generalization ability.

[0298] 5.5.2 Closed-loop termination condition

[0299] The classification accuracy of the SVM model after 5-fold cross-validation is ≥90% (actually reaching 96%).

[0300] The Kappa coefficient of consistency of the three repeated experiments in the evaluation of the immune grade was ≥0.80 (actually reached 0.92, which is excellent consistency).

[0301] S6: Repeatability Validation and Data Summary

[0302] The core objective of this step is to verify the stability of the S5 four-algorithm fusion evaluation results through repeated experiments, summarize the data from the entire process to form a standardized report, and provide a traceable and referable basis for the development of immune drug candidates. The specific operation is as follows:

[0303] 6.1 Repeatable experimental design (strictly controlled variables)

[0304] Perform three independent replicate experiments following steps S1-S5 (named Exp1, Exp2, and Exp3), with each experiment meeting the following control conditions:

[0305] Laboratory animals: 24 healthy cats aged 1-3 years were repurchased from qualified units each time, and the screening process of S1 was repeated to ensure that the initial conditions of each group were consistent (no significant differences in weight and blood indicators).

[0306] Cat fleas: Each time, cats that are positive for fleas are re-screened (4 cats / time), and the species identification and viability screening process of S2 is repeated to avoid the flea population generation being too high and thus causing a decrease in viability;

[0307] Reagents and instruments: The same batch of reagents (ELISA antibody, CBA kit, PCR primers) and instruments (flow cytometer, microplate reader) were used in all three experiments. The instruments were calibrated before each use (e.g., wavelength calibration of the microplate reader and fluorescence compensation calibration of the flow cytometer).

[0308] Operators: Two designated laboratory personnel are responsible for core operations (flea counting, sample collection, and indicator detection) to avoid errors caused by operational differences;

[0309] Algorithm parameters: The PCA eigenvalues, surface fitting coefficients, AHP objective weights, and SVM parameters of S5 are all fixed (e.g., surface fitting coefficients). AHP weights (Unchanged), ensuring uniformity of evaluation standards.

[0310] 6.2 Data Recording and Statistical Analysis

[0311] 6.2.1 Data Recording: The following core data were recorded for each experiment to form a repeatability experimental data table: S1: Baseline health data of experimental cats (number, weight, complete blood count, blood biochemistry); S2: Quality verification report of cat fleas (species identification results, viability pass rate); S3: Infection monitoring data (weekly infection rate, average infection intensity, incidence of clinical symptoms, abnormal rate of blood indicators); S4: Immune indicator detection data (raw test values ​​and standard deviations of 5 indicators); S5: Algorithm evaluation results (PCA principal component score, surface fitting Q value, corrected comprehensive score S', immune grade).

[0312] 6.2.2 Statistical Analysis

[0313] 1. Descriptive statistics: Calculate the mean ± standard deviation of each indicator in the three experiments (e.g., mean ± SD of IgG titer in W7 of G3 group) and assess the reproducibility of the data (standard deviation ≤ 20% of the mean is considered good reproducibility).

[0314] 2. Analysis of differences between groups: One-way ANOVA was used to compare the differences in indicators among groups G1, G2 and G3 at each time point; pairwise comparisons were performed using the LSD-t test, and P<0.05 was considered statistically significant (used to verify the effectiveness of candidate immune drugs, such as S' in group G3 being significantly higher than that in group G2, suggesting that the drug induces an immune response).

[0315] 3. Algorithm Consistency Test: Calculate the Kappa coefficient of the immune grade in group G3 (immune candidate drug group) in the three experiments. The formula is: ;in The actual consistency rate is calculated as (number of samples with the same grade in 3 experiments / total number of samples). To determine the expected consistency rate (the sum of the products of the grade distributions of each group), K ≥ 0.80 is considered to be of excellent consistency. If the immunization grade of a certain batch of experiments is inconsistent with the other two batches (e.g., Exp1 is "excellent" and Exp2 is "good"), two more repeated experiments (Exp4 and Exp5) are required, and the majority result is taken as the final grade (e.g., 3 "excellent" and 2 "good", then the final grade is "excellent").

[0316] 6.3 Final evaluation report formation: Summarize all data from S1 to S6 to form an animal immune evaluation report based on the feline cephalopod infection model.

Claims

1. A method for evaluating animal immunity based on a feline ctenoid flea infection model, characterized in that, include: S1. Preparation and screening of experimental animals: Healthy domestic cats were selected and qualified individuals were screened for external parasites, clinical symptoms were assessed, blood routine tests, blood biochemistry tests and toxoplasmosis antibody tests. Healthy baseline data were recorded and a standardized host model was constructed. S2. Preparation and quality verification of cat fleas: positive cats with ≥4 live fleas on their bodies and negative for Toxoplasma gondii antibodies were screened. The fleas collected from the bodies of the positive cats were morphologically and molecularly identified. Female adult fleas that met the standards for motility, morphological integrity and blood-sucking potential were screened to form a standardized source of infection. S3. Experimental animals infected with *Ctenopharynx felis*: Healthy cats that passed the screening in step S1 were randomly divided into a blank control group, an infection control group, and an immunization candidate drug group. The blank control group was housed alone in an environment free of flea contamination. The infection control group was housed together with the positive cats from step S2. The immunization candidate drug group was given the immunization candidate drug according to body weight 24 hours before being housed with the positive cats. During the housed period, the number of live fleas, clinical symptoms, and blood indicators of each group of cats were monitored weekly to verify whether the infection model met the preset criteria. S4. Detection of immune-related indicators: Blood samples and immune organ samples were collected from each group of cats at preset time points after infection. Humoral immune indicators, cellular immune indicators and infection control effect indicators were detected. Raw test data were recorded and quality control was performed. S5. Evaluation of animal immune function: Principal component analysis is first used to extract principal components with a cumulative variance contribution rate of ≥85% from the detection data in step S4 to reduce dimensionality and noise; then, a quadratic polynomial surface fitting model is used to perform nonlinear correction on the extracted principal components to obtain the comprehensive principal component score. Based on the variance contribution rate and correlation coefficient, objective weights are assigned to the corrected comprehensive principal component score and the extracted principal components using the analytic hierarchy process, and the corrected comprehensive immune score is calculated. Finally, a support vector machine model is used to map the corrected comprehensive immune score to the immune level. S6. Repeatability verification and data summary: Perform three independent replicate experiments according to steps S1-S5, controlling for variables such as experimental animals, cat fleas, reagents and instruments, and operators. Verify the consistency of the immunization grade results of the three experiments, and summarize the data of the entire process to form an evaluation report.

2. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S1, the indicators for routine blood tests include red blood cells, hemoglobin, white blood cells, lymphocyte percentage, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin content, mean corpuscular hemoglobin concentration, neutrophil count and percentage, monocyte count and percentage, eosinophil count and percentage, basophil count and percentage, and platelets, totaling 14 items. The indicators for blood biochemistry tests include total protein, albumin, glucose, urea, alanine aminotransferase, aspartate aminotransferase, and creatinine, totaling 7 items. The Toxoplasma gondii antibody test uses a commercially available ELISA kit, and a result with an OD value <0.3 is considered acceptable.

3. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S2, molecular biological identification involves PCR detection of the mitochondrial COI gene in *Ctenophora catina*. Specifically, this includes: extracting flea genomic DNA using a blood / cell / tissue genomic DNA extraction kit; constructing a 25 μL PCR reaction system using upstream primer 5'-AGA ATT AGG TCA ACC AGGA-3' and downstream primer 5'-GAA GGG TCA AAG AAT GAT GT-3'; performing PCR according to a program of 35 cycles: 94℃ pre-denaturation for 5 minutes, 94℃ denaturation for 30 seconds, 55℃ annealing for 30 seconds, 72℃ extension for 40 seconds, and a final extension at 72℃ for 10 minutes; and performing agarose gel electrophoresis on the products. If a band matching the target fragment size appears and the sequencing results show ≥99% homology with the *Ctenophora catina* COI gene sequence in GenBank, then the species is confirmed to be homozygous.

4. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S3, after grouping, the differences in baseline weight and blood baseline indicators among the three groups of cats were verified. One-way ANOVA was used to determine that the differences were not statistically significant, which was considered as group equilibrium. The infection model validation criteria included: the infection rate of the infection control group was ≥75%, 87.5%, 100%, and 100% respectively from week 4 to week 7; the average number of live fleas was ≥14 in week 7; ≥70% of the cats showed itching and ≥25% of the cats showed rashes in week 7; at least 3 blood indicators were abnormal in week 7; and the blank control group had no flea infection and the indicators were normal throughout the process.

5. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S4, the humoral immune indicator is the specific IgG titer of *Ctenophora catina*, which is detected by indirect ELISA: *Ctenophora catina* is lysed by sonication with RIPA lysis buffer, and the supernatant after centrifugation is used as the antigen. It is diluted to 2 μg / mL with carbonate buffer to coat the ELISA plate. After blocking, plasma samples are added in a gradient of 1:100 to 1:12800, and then HRP-labeled goat anti-cat IgG secondary antibody is added. After color development, the maximum dilution factor when the OD value is ≥0.2 and twice that of the negative control is taken as the IgG titer.

6. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S5, during principal component analysis, the five immune-related indicators are first constructed into an original matrix and standardized. The covariance matrix of the standardized indicators is calculated, and the eigenvalues ​​and corresponding eigenvectors of the covariance matrix are solved. Principal components are selected based on a cumulative variance contribution rate ≥85%. The five immune-related indicators include feline flea-specific IgG titer, CD4+, and other indicators. + / CD8 + T cell ratio, IFN-γ concentration, flea survival rate, and blood-feeding rate.

7. The animal immune evaluation method based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S5, the quadratic polynomial expression of the surface fitting model includes constant terms, linear terms of principal components, quadratic terms of principal components, and interaction terms between principal components. The model coefficients are solved using the least squares method to determine the coefficients R. 2 A value ≥0.85 indicates a satisfactory fit.

8. The method for evaluating animal immunity based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S5, when assigning weights using the analytic hierarchy process (AHP), the overall principal component score, the variance contribution rate of each principal component, and the Pearson correlation coefficient with the true value of the immune effect are calculated separately. The variance contribution rate and Pearson correlation coefficient are then standardized. Finally, the standardized variance contribution rate and Pearson correlation coefficient are merged with equal weights to obtain the objective weights of each evaluation object. The true value of the immune effect is calculated based on the standardized values ​​of five immune-related indicators and the weights determined by information entropy.

9. The method for evaluating animal immunity based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S5, the support vector machine model uses a radial basis kernel function. The penalty coefficient C and kernel parameter γ are optimized by grid search combined with 5-fold cross-validation to determine the optimal parameters with a classification accuracy of ≥90%. A one-to-one strategy is used to construct a binary classifier, and the final immune level is determined by voting. The immune level is divided into four levels: excellent, good, medium and poor. Sample labels are labeled based on the objective thresholds of five immune-related indicators.

10. The method for evaluating animal immunity based on the feline ctenoid flea infection model according to claim 1, characterized in that, In step S6, the consistency test of repeatability experiments uses the Kappa coefficient. When the Kappa coefficient is ≥0.80, the consistency of the three experimental results is considered to be excellent. If the experimental results of a certain batch are inconsistent with those of other batches, two more repeated experiments are added and the majority result is taken as the final evaluation result. The reagents used in the three repeated experiments are from the same batch. The instrument is calibrated before each use, and the core operations are performed by fixed personnel.

Images