Single-vesicle membrane protein marker combinations for assessing embryo implantation potential, uses and methods of assessing embryo implantation potential

Abstract

This invention discloses a combination of single vesicle membrane protein markers for assessing embryo implantation potential, its application, and a method for assessing embryo implantation potential, relating to the field of biomedical testing. The marker combination includes the single vesicle membrane proteins ITGA8 and L1CAM. ITGA8 has been found to predict implantation success, while L1CAM can predict implantation failure. The combination of these two markers achieves an AUC value of 0.90, providing molecular markers for accurate assessment of embryo implantation potential. Compared to existing methods that rely solely on morphological scoring, this invention provides molecular-level assessment indicators, significantly improving the accuracy of embryo implantation potential assessment. It can effectively identify low-potential embryos that are morphologically "high-quality" but have "abnormal" protein signals, overcoming the shortcomings of existing technologies that cannot accurately predict the implantation potential of high-quality embryos. Furthermore, this invention only detects EVS in the culture medium, without contacting the embryo itself, making it a safe and non-invasive assessment method for embryo quality.

Description

Technical Field

[0001] This invention relates to the field of biomedical testing, specifically to a combination of single vesicle membrane protein markers for assessing embryo implantation potential, its application, and a method for assessing embryo implantation potential. Background Technology

[0002] With the development of assisted reproductive technology, improving embryo transfer success rates has become a clinical focus. Currently, embryo quality assessment mainly relies on morphological scoring systems, including indicators such as blastocyst expansion, inner cell mass, and trophectoderm cell quality. However, morphological scoring has drawbacks such as strong subjectivity and limited predictive accuracy; even embryos with high morphological scores can have significantly different implantation success rates.

[0003] In recent years, extracellular vesicles (EVs) have played a crucial role in embryonic development and implantation as important mediators of intercellular communication. Studies have shown that EVs in embryo culture media carry various biomolecules, including proteins, nucleic acids, and lipids, which may reflect the developmental status and implantation potential of the embryo. CN115851888A discloses a method for counting multi-subpopulation extracellular vesicles using single vesicle membrane protein expression profiling. This method achieves the separation and analysis of EVs through antibody-conjugated nucleic acid complexes, enabling digital absolute quantification of EVs. However, this method mainly focuses on EV counting and fails to deeply analyze the relationship between EV membrane protein composition and embryo implantation potential.

[0004] CN120905357A proposes a method for annotating vesicle cell subsets based on single vesicle membrane proteomics. This method achieves the identification and annotation of core exosome subsets by encoding and labeling antibody probes. This technology provides a new approach for studying EV heterogeneity, but it has not been applied to the field of embryo quality assessment.

[0005] Regarding embryo quality assessment, CN115128278A discloses a set of biomarkers and screening methods for assessing the rate of high-quality embryos in the PCOS population. This method screens for differentially expressed proteins related to oocyte quality by assessing the proteomic expression profile of follicular fluid. However, this method is mainly targeted at the PCOS population and focuses on assessing oocyte quality rather than embryo implantation potential.

[0006] CN105112491A proposes a method for screening differential metabolites in goat transgenic clone embryos based on gas chromatography-mass spectrometry and morphology. This method combines embryo morphology assessment to identify metabolic markers related to embryo quality. However, this method mainly targets animal embryos and does not involve membrane proteomics analysis.

[0007] CN116254225B discloses a method for isolating and enriching membrane protein double-positive specific extracellular vesicle (EV) subsets. This method can achieve non-destructive isolation and enrichment of membrane protein double-positive specific EV subsets. This technology provides a new approach for EV subset isolation, but it has not been applied to the field of embryo quality assessment.

[0008] In summary, existing technologies have the following shortcomings: First, current embryo quality assessment methods mainly rely on morphological standards and lack objective evaluation indicators at the molecular level, leading to significant differences in implantation outcomes among morphologically similar embryos. Second, although some studies have focused on the role of endothelial cells (EVs) in embryonic development, there is a lack of methods for systematically analyzing the composition of EV membrane proteins at the single vesicle level, failing to fully reveal the relationship between EV heterogeneity and embryo implantation potential. Third, current technologies have not established an interaction model between embryonic EV membrane proteins and endometrial receptivity factors, lacking a deep understanding of the embryo-endometrial interaction mechanism. Fourth, existing biomarker screening methods often focus on single indicators and lack research on protein combination markers, limiting the accuracy and specificity of the assessment.

[0009] Therefore, there is an urgent need to develop a combination of single vesicle membrane protein markers for assessing embryo implantation potential, as well as to apply and evaluate embryo implantation potential, in order to improve the accuracy of embryo implantation potential prediction. Summary of the Invention

[0010] To address the technical problem that existing technologies primarily rely on morphological scoring for assessing embryo implantation potential, yet significant differences remain in the implantation success rate of high-quality embryos, and the lack of effective molecular-level assessment methods to predict embryo implantation outcomes, this invention provides a combination of single-vesicle membrane protein markers for assessing embryo implantation potential, its application, and a method for assessing embryo implantation potential. This invention aims to achieve the technical effect of identifying the heterogeneity of extracellular vesicles originating from embryos at the single-vesicle level, gaining a deeper understanding of the embryo-endomembrane interaction mechanism, and accurately predicting the implantation potential of high-quality embryos.

[0011] To achieve the above-mentioned technical objectives, the present invention mainly adopts the following technical solutions:

[0012] In a first aspect, the present invention discloses a combination of single vesicle membrane protein markers for assessing embryo implantation potential, the combination of markers including single vesicle membrane proteins ITGA8 and L1CAM.

[0013] Secondly, this invention discloses the application of single vesicle membrane proteins ITGA8 and L1CAM as markers in the preparation of kits for evaluating embryo implantation potential.

[0014] Thirdly, this invention discloses a method for assessing embryo implantation potential, which evaluates embryo quality by detecting the expression levels of ITGA8 and L1CA proteins in extracellular vesicles of embryo culture medium.

[0015] In a preferred embodiment of the present invention, the following steps are included:

[0016] Step 1: Extract the raw signal values ​​of the corresponding probes of ITGA8 and L1CAM in the sample to be tested, and summarize them according to the sample to obtain the sample-level raw signal values ​​of ITGA8 and L1CAM in each sample.

[0017] Step 2: Calculate the standardized signal value as follows:

[0018] ITGA8 std =log2(R ITGA8 +1)−log2(R ITGA8,ref +1);

[0019] L1CAM std =log2(R L1CAM +1)−log2(R L1CAM,ref +1);

[0020] Among them, R ITGA8 and R L1CAM These are the original signal values ​​of ITGA8 and L1CAM in the sample under test, respectively; R ITGA8,ref and R L1CAM,ref These are the original signal values ​​of the corresponding proteins in the same batch of positive reference control samples; the positive reference control samples are commercial EV reference materials or EV samples from stable cell lines.

[0021] Step 3: Use the formula Score = 0.8 × ITGA8 std -0.6×L1CAM std The results were interpreted to assess the quality of the embryos.

[0022] Furthermore, in step 1, the raw signal values ​​of the corresponding probes of ITGA8 and L1CAM in the sample to be tested are extracted, including the following steps:

[0023] Step 11: Collect the blastocyst culture medium on day 5 of the IVF-ET cycle after thawing and recovery. After the blastocyst recovery is completed, perform routine embryo transfer. Collect the remaining single embryo culture medium under aseptic conditions and extract EVs from the embryo culture medium.

[0024] Step 12: Using the proximity coding technique PBA, single-particle level membrane proteomics detection was performed on EVs in a single embryo culture medium to obtain sequencing data;

[0025] Step 13: Perform bioinformatics analysis on the sequencing data to obtain the raw signal values ​​of each protein in each single vesicle;

[0026] Step 14: Extract the original signal values ​​of the corresponding probes for ITGA8 and L1CAM from the original signal values ​​of each protein, and summarize them by sample to obtain the sample-level original signal values ​​of ITGA8 and L1CAM in each sample.

[0027] Furthermore, in step 11, size exclusion chromatography is used to extract EVs from the embryo culture medium.

[0028] Furthermore, step 12 specifically includes the following steps:

[0029] Preparation of conjugated nucleic acid antibodies: Antibodies are conjugated to oligonucleotide sequences to prepare DNA-antibody probes. The oligonucleotides include the following functional regions: 1. An antibody-specific protein tag region for distinguishing different antibody probes; 2. A molecular tag region for subsequent molecule counting and deduplication analysis; 3. A hybridization region complementary to the RCA product for sequence recognition and extension in subsequent PBA reactions. The total length of the oligonucleotides is 40-80 nt, of which the protein tag region is 8-12 nt long; the molecular tag region is 8-12 nt long; and the hybridization region complementary to the RCA product is 15-25 nt long.

[0030] Preparation of Rolling Circle Amplification (RCA) Products: Rolling circle amplification is performed using a circular DNA template to prepare an RCA product that can be used to recognize EVs. The RCA product contains the following functional regions: 1. a recognition region complementary to the DNA-antibody probe; 2. a 15nt composite tag region (EV tag) for recognizing EVs; 3. a fixed sequence region for subsequent extension reactions and PCR library construction; the circular DNA template is 60-120nt in length, of which the EV tag is 15nt in length.

[0031] EV immobilization: An affinity capture strategy based on the cholera toxin B subunit for GM1 gangliosides on the EV membrane was used to immobilize the EVs to be tested on the bottom surface of the well.

[0032] DNA-antibody probe incubation: Add the DNA-antibody probe mixture to the wells of the immobilized EVs to allow it to bind to the EV surface membrane proteins;

[0033] PBA reaction: The RCA product is added to the well of an EV that has been bound to a DNA-antibody probe, so that the probe DNA hybridizes with the complementary sequence in the RCA product; then an extension reaction occurs under the action of DNA polymerase, so that multiple adjacent probes on the same EV can share the same complex tag.

[0034] PCR amplification, library construction and sequencing: The DNA fragments obtained from the extension reaction are amplified by PCR to enhance the signal and construct a next-generation sequencing library. The DNA fragments contain: 1. a protein tag; 2. a molecular tag UM; 3. an EV tag; and 4. a PCR primer binding immobilization region.

[0035] Preferably, step 13 specifically includes the following steps: splitting, quality-controlling and comparing the raw sequencing data obtained in step 12, identifying the corresponding protein barcode, molecular tag UMI and EV tag in each sequence, deduplicating and counting the effective reads belonging to the same single vesicle, the same protein probe and different UMIs, to obtain the original signal value of each protein in each single vesicle.

[0036] Preferably, step 14 specifically includes the following steps:

[0037] Raw signal extraction: For each sample, the valid sequencing reads corresponding to ITGA8 and L1CAM are counted by removing duplicates from the unique molecular tag UMI to obtain the raw signal value of the target protein; at the same time, the total valid signal value, total UMI count and number of detected protein entries corresponding to all protein probes in the sample are counted as the basis for subsequent sample quality control and standardization analysis.

[0038] Initial screening at the sample level: Preliminary quality control is performed on the samples. Samples meeting the following criteria are retained, while those not meeting the criteria are discarded: 1. The total number of valid reads is not less than 1 × 10⁻⁶. 4 2. The total UMI count is not less than 5 × 10 3 3. The number of detected protein entries shall not be less than 20% of the total number of protein probes; 4. The total effective signal value of the negative blank control well shall be less than 5% of the median of the same batch of samples; The samples retained after the above screening shall be recorded as cleaned samples;

[0039] Protein entry-level screening: Protein entries in each sample are further screened, and those meeting the following criteria are retained: 1. The original UMI count in the sample is ≥3; 2. The protein is detected in at least 10% of all valid samples; 3. The protein signal value is more than twice the mean protein signal value of the corresponding negative control. Protein entries that do not meet the above criteria are considered low-confidence signals and are removed from subsequent normalization analysis. The data retained after screening are recorded as cleaneddata.

[0040] Preliminary library size correction: Using the total effective signal value or total UMI count of each sample as an indicator of sample sequencing scale, preliminary correction is made for differences in sequencing depth between samples;

[0041] Inter-sample normalization: After the initial library size correction, the trimmed mean normalization method (TMM) is used to normalize the inter-samples. Finally, the original signal values ​​of ITGA8 and L1CAM are corrected by combining the library size and the normalization factor to obtain the normalized protein signal values.

[0042] Batch quality control: Each batch of testing includes both a negative blank control and a positive reference control. The negative blank control, which does not include EV samples, is used to assess background noise. The positive reference control uses commercially available EV reference materials or stable cell line EV samples to monitor probe binding efficiency, amplification efficiency, and batch-to-batch stability. The positive reference control also serves as a calibration reference, unifying protein signals from different batches of test samples to the same benchmark. A batch of data is considered valid when all of the following conditions are met: 1. The total number of valid reads in the positive reference control sample reaches 80%-120% of the preset reference range; 2. The standardized signal values ​​of ITGA8 and L1CAM in the positive reference control sample deviate from the historical mean by no more than ±20%; 3. The signal values ​​of ITGA8 and L1CAM in the negative blank control are both less than 10% lower than the corresponding values ​​in the positive reference control. If the above conditions are not met, the batch of data is considered invalid and must be retested.

[0043] Data acquisition after calibration: The data that has passed sample-level quality control, protein item-level screening and batch quality control are used to obtain the final cleaned data that can be used for analysis.

[0044] In a preferred embodiment of the present invention, in step 3, when the result is ≥ 2.60, it indicates that the embryo has high implantation potential and can be preferentially transplanted; when the result is < 1.80, it indicates that the embryo has low implantation potential and is not recommended for preferential transplantation; when the result is between 1.80 and 2.59, it is determined to be a borderline state, and a comprehensive judgment is made based on morphological and clinical factors.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] This invention successfully identified two differentially expressed protein markers (ITGA8, L1CAM) associated with embryo implantation outcome, and the combination of these two markers achieved an AUC value of 0.90. Compared to existing methods that rely solely on morphological scoring, this invention provides molecular-level assessment indicators, significantly improving the accuracy of embryo implantation potential assessment.

[0047] This invention discovers that ITGA8 can be used to predict implantation success, and L1CAM can be used to predict implantation failure, providing molecular markers for the accurate assessment of embryo implantation potential. These markers can effectively identify low-potential embryos that are morphologically "high-quality" but have "abnormal" protein signals, overcoming the shortcomings of existing technologies that cannot accurately predict the implantation potential of high-quality embryos. Furthermore, this invention only examines EVS in the culture medium, without contacting the embryo itself, thus providing a safe and non-invasive method for embryo quality assessment. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the oligonucleotide structure provided in Embodiment 1 of the present invention;

[0049] Figure 2 A comparison chart of the area under the ROC curve (AUC) of the ITGA8 single-index method, L1CAM single-index method, and ITGA8-L1CAM joint scoring method provided by the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1

[0052] A method for assessing embryo implantation potential, the method comprising the following steps:

[0053] Step 1: Collection of embryo culture medium samples:

[0054] Culture medium from thawed and revived blastocysts on day 5 of the IVF-ET cycle was collected. After blastocyst thawing, routine embryo transfer was performed. The culture medium from any remaining single embryos was collected under aseptic conditions for subsequent extracellular vesicle (EV) isolation. Each sample corresponded to a unique transfer cycle and embryo information, and blastocyst morphology scores and key clinical data were recorded simultaneously. All patients were followed up according to standard procedures after embryo transfer, and clinical pregnancy outcomes were recorded. Clinical pregnancy was defined as the presence of an intrauterine gestational sac and primitive cardiac tube pulsation on ultrasound examination, and patients were grouped accordingly into a pregnant group and a non-pregnant group.

[0055] Step 2: Extraction of EVs from embryo culture medium:

[0056] Size exclusion chromatography (SEC) was used to extract EVs from embryo culture medium. The specific steps included:

[0057] (1) Sample pretreatment: The embryo culture medium was centrifuged sequentially at 4°C to remove cells and debris. First, it was centrifuged at 1500×g for 20 minutes, and then at 3000×g for 15 minutes. The supernatant was used for subsequent purification. Then, the supernatant was filtered through a 0.8μm filter membrane, and the filtered sample was made up to 1mL with PBS.

[0058] (2) Equilibration of exclusion column: After the exclusion column is taken out from 4℃, it is equilibrated at room temperature (18-25℃) for 30 minutes. The exclusion column is pre-washed with sterile PBS with a washing volume of 10mL.

[0059] (3) Sample loading and collection: Slowly add 1 mL of pretreated sample to the exclusion column. After the sample has completely entered the packing material, add PBS in steps for elution. Collect the first 5 fractions in 500 μL for EV enrichment.

[0060] (4) Ultrafiltration concentration: Transfer all collected EVs eluent to a 100kDa molecular weight cutoff ultrafiltration tube and concentrate by centrifugation at 4000g. For miRNA extraction and sequencing, concentrate to approximately 60μL; for routine characterization or subsequent experiments, concentrate to approximately 200μL.

[0061] Step 3: Single vesicle membrane proteomics detection:

[0062] The proximity coding (PBA) technique was used to perform single-particle-level membrane proteomics detection on EVs in individual embryo culture media, covering 260 membrane protein biomarkers. Specific steps included:

[0063] (1) Preparation of conjugated nucleic acid antibodies: Antibodies are conjugated with oligonucleotide sequences to prepare DNA-antibody probes. The oligonucleotide structure is shown below. Figure 1 As shown, it includes the following functional regions: 1. Antibody-specific protein tag region, used to distinguish different antibody probes; 2. Molecular tag region, used for subsequent molecule counting and deduplication analysis; 3. Hybridization region complementary to RCA product, used for sequence identification and extension in subsequent PBA reaction.

[0064] Specifically, the total length of the oligonucleotide is 40-80 nt, of which the protein tag region is 8-12 nt long; the molecular tag region is 8-12 nt long; and the hybridization region complementary to the RCA product is 15-25 nt long.

[0065] Oligonucleotide design must meet the following requirements: GC content of 40%-60%; at least 3 base sites must be different for any two protein tag sequences of the same length; no more than 4 identical bases must appear consecutively; and no significant self-complementary or hairpin structures must be formed.

[0066] Oligonucleotides are coupled to antibodies using cleavable cross-linking agents or electrochemical methods.

[0067] The molar ratio of antibody to oligonucleotide is 1:3-1:10, meaning the molar ratio can be 1:3, 1:4, 1:5, 1:6, 1:7, 1:8, 1:9, or 1:10. Conjugation is completed by incubating in a conjugation buffer at pH 7.2-7.8 at room temperature for 1-2 hours or at 4°C overnight. After the conjugation reaction, unconjugated free oligonucleotides are removed by centrifugation, ultrafiltration, or molecular sieve chromatography to obtain the DNA-antibody probe working solution.

[0068] The final concentration of a single probe in the DNA-antibody probe working solution is 0.1-0.5 μg / mL.

[0069] (2) Preparation of rolling circle amplification (RCA) products: Rolling circle amplification (RCA) is performed using a circular DNA template to prepare RCA products that can be used to recognize EVs. The RCA products contain the following functional regions: 1. a recognition region complementary to the DNA-antibody probe; 2. a 15nt complex tag region (EV tag) for recognizing EVs; 3. a fixed sequence region for subsequent extension reactions and PCR library construction.

[0070] The circular DNA template is 60-120 nt in length, with the EV tag being 15 nt in length. The RCA reaction system includes: 1. Circular DNA template: 1 nM; 2. Phi29 DNA polymerase: 10 U / reaction; 3. dNTPs: final concentration 0.25 mM; 4. RCA reaction buffer: prepared according to the enzyme preparation instructions.

[0071] The RCA reaction conditions were: 30℃ for 60 min; followed by treatment at 65℃ for 10 min to terminate the reaction.

[0072] After the reaction, unreacted template and small molecule residues were removed using a purification column or magnetic beads to obtain the RCA product working solution. The final concentration of the RCA product was adjusted to 2 nM for later use.

[0073] (3) EV fixation: The EVs to be tested were fixed on the bottom surface of the well by adopting an affinity capture strategy based on choleratoxin sub unit B (CTB) for GM1 gangliosides on the EV membrane.

[0074] A 96-well plate pre-coated with CTB can be used; add 100 μL of CTB per well at a concentration of 5 μg / mL to the plate, incubate overnight at 4°C, wash the plate three times with PBS the next day, and block with 1% BSA / PBS at room temperature for 30 min.

[0075] The EV sample extracted and concentrated by SEC was added to the coated well at 100 μL / well. The fixation conditions were: incubation at room temperature for 1 hour; after incubation, the plate was washed 3 times with PBST (0.05% Tween-20), 200 μL / well each time, and the number of EVs added to each well was not less than 1×10^4.

[0076] (4) DNA-antibody probe incubation: Add the DNA-antibody probe mixture to the wells of the immobilized EVs to allow it to bind to the EV surface membrane proteins. The DNA-antibody probe mixture was prepared using the commercially available ExoSeek kit. ® Panel260 (Secretech, Shenzhen, China) can simultaneously detect 260 EV membrane protein markers.

[0077] The incubation system was as follows: DNA-antibody probe mixture: 100 μL / well; probe dilution buffer: PBS containing 1% BSA; final concentration of a single probe was 1 M.

[0078] The incubation conditions are: incubate at room temperature for 2 hours or at 37°C for 1 hour.

[0079] After incubation, wash the plate 3-5 times with PBST to remove unbound probes.

[0080] (5) PBA reaction: The RCA product is added to the EV well that has been bound to the DNA-antibody probe, so that the probe DNA hybridizes with the complementary sequence in the RCA product; then an extension reaction occurs under the action of DNA polymerase, so that multiple adjacent probes on the same EV share the same complex tag.

[0081] The PBA reaction system includes: RCA product: final concentration 2 nM; DNA polymerase: 5 U / reaction; dNTP: final concentration 0.25 mM; extension buffer: prepared according to the enzyme preparation instructions.

[0082] The hybridization conditions were: 37℃ for 30 min;

[0083] The extension conditions were: reaction at 37°C for 30 min; followed by treatment at 65°C for 10 min to terminate the reaction.

[0084] After the reaction was completed, the plate was washed three times with PBST, and the immobilized composite product in the wells was retained for subsequent amplification.

[0085] (6) PCR amplification, library construction and sequencing: The DNA fragments obtained from the extension reaction are subjected to PCR amplification to enhance the signal and construct a next-generation sequencing library. The DNA fragments contain: 1. protein tag; 2. molecular tag (UMI); 3. complex tag / EV tag; 4. PCR primer binding immobilization region.

[0086] The PCR reaction system includes: template DNA: 5 μL; 2× high-fidelity PCR Mix: 12.5 μL; forward primer: final concentration 0.4 μM; reverse primer: final concentration 0.4 μM; total reaction volume: 25 μL.

[0087] The PCR amplification program was as follows: 95℃ pre-denaturation for 3 min; 95℃ for 15 s, 60℃ for 15 s, 72℃ for 20 s, for a total of 15 cycles; extension at 72℃ for 5 min.

[0088] The PCR products were purified with magnetic beads and then subjected to library quality control.

[0089] A library is considered valid if it meets the following conditions: 1. The length of the main peak fragment is between 200-500 bp; 2. The library concentration is not less than 1 ng / μL; 3. There are no obvious primer dimer peaks.

[0090] Sequencing was performed using the Illumina platform in PE150 mode. After obtaining the raw sequencing sequences, the data analysis process proceeded to step 4.

[0091] It should be noted that the antibody-conjugated oligonucleotides, circular DNA templates, EV tags, and PCR primers involved in step 3 all use the probe system, tag sequences, and fixed sequence regions provided by commercial kits, and no separate design is required; in specific implementation, preparation, incubation, amplification, and library construction should be carried out in accordance with the instructions accompanying the commercial kits.

[0092] Step 4: Proteomics Data Analysis:

[0093] Bioinformatics analysis of sequencing data includes the following workflow:

[0094] The raw sequencing sequences obtained in step 3 are split, quality controlled, and aligned to identify the corresponding protein barcodes, molecular tags (UMIs), and vesicle complex tags in each sequence. Repeated readings belonging to the same single vesicle and the same protein probe but with different UMIs are counted to obtain the raw signal values ​​of each protein in each single vesicle.

[0095] To screen candidate membrane proteins associated with embryo implantation outcomes, this invention first performs sample-level quality control, protein-level screening, library size correction, and TMM normalization on the single vesicle membrane proteomics data obtained from PBA detection, obtaining cleaned data suitable for subsequent analysis. Subsequently, among candidate proteins meeting the detection frequency and background threshold requirements, the expression differences between the clinical pregnancy group and the non-pregnancy group were compared, and their discriminatory power for embryo implantation outcomes was evaluated using single-indicator ROC analysis. Furthermore, considering the stability of the candidate protein's differential direction, detection repeatability, single-indicator discriminative power, and the performance improvement after combined modeling, ITGA8 and L1CAM were ultimately selected as target proteins. Specifically, ITGA8 showed a relatively increasing trend in implantation success-related samples, while L1CAM showed a relatively increasing trend in implantation failure-related samples, demonstrating good discriminatory power for embryo implantation outcomes.

[0096] The raw signal values ​​of the corresponding probes for ITGA8 (integrin α8) and L1CAM were further extracted and summarized by sample to obtain the sample-level raw expression values ​​of ITGA8 and L1CAM in each sample. The specific process is as follows:

[0097] (1) Extraction of raw signal

[0098] For each sample, the valid sequencing reads corresponding to ITGA8 and L1CAM are deduplicated and counted using unique molecular identifiers (UMIs) to obtain the original signal values ​​of the target proteins. At the same time, the total valid signal values, total UMI counts, and number of detected protein entries corresponding to all protein probes in the sample are counted as the basis for subsequent sample quality control and standardization analysis.

[0099] (2) Initial screening at the sample level

[0100] Preliminary quality control was performed on the samples. Samples meeting the following criteria were retained, while those not meeting the criteria were discarded:

[0101] The total number of valid reads is no less than 1 × 10 4 ;

[0102] The total UMI count is not less than 5 × 10 3 ;

[0103] The number of detected protein entries shall not be less than 20% of the total number of protein probes;

[0104] The total effective signal value of the negative blank control wells was less than 5% of the median of the same batch of samples.

[0105] The samples retained after the above screening are denoted as cleaned samples.

[0106] (3) Protein item-level screening

[0107] The protein entries in each sample were further screened. Protein entries that met the following criteria were retained:

[0108] The original UMI count in this sample is ≥3;

[0109] The protein was detected in at least 10% of all valid samples;

[0110] The protein signal value was twice as high as the mean protein signal value of the negative control.

[0111] Protein entries that do not meet the above conditions are considered low-confidence signals and are removed from subsequent normalization analysis. The data retained after screening is denoted as cleaned data. In this invention, cleaned data refers to data that has been retained after sample-level quality control, protein entry-level screening, and batch quality control, and can be used for subsequent standardization and joint scoring calculations.

[0112] (4) Preliminary correction of library size

[0113] The total effective signal value or total UMI count for each sample was used as an indicator of the sample sequencing scale to preliminarily correct for differences in sequencing depth between samples.

[0114] (5) Normalization between samples

[0115] After initial library size correction, the trimmed mean of M-values ​​(TMM) method is used for inter-sample normalization. This includes steps such as reference sample selection, expression ratio calculation, outlier trimming, and normalization factor determination. Preferably, samples with total effective signal values ​​close to the median of all samples are selected as reference samples. For each protein, the logarithmic expression ratio (M-value) and mean expression level (A-value) of the test sample relative to the reference sample are calculated. After removing low-expression protein entries, the M-values ​​are trimmed by 20% on each side, and the A-values ​​are trimmed by 5% on each side. Then, the M-values ​​of the retained entries are weighted and averaged to obtain the sample normalization factor. Finally, the original signal values ​​of ITGA8 and L1CAM are corrected based on the library size and normalization factor to obtain standardized protein signal values.

[0116] (6) Batch quality control

[0117] Each batch of tests includes both negative blank control and positive reference control samples.

[0118] The negative control, which does not include EV samples, is used to assess background noise.

[0119] Positive reference controls, using commercially available EV reference materials or stable cell line EV samples, are used to monitor probe binding efficiency, amplification efficiency, and batch-to-batch stability. These positive reference control samples also serve as calibration references, standardizing protein signals from different batches of test samples to a common benchmark.

[0120] The batch of data is considered valid if all of the following conditions are met:

[0121] The total number of valid reads in the positive reference control sample reached 80%-120% of the preset reference range;

[0122] The standardized signal values ​​of ITGA8 and L1CAM in the positive reference control samples deviated from the historical mean by no more than ±20%;

[0123] In the negative control, the signal values ​​of ITGA8 and L1CAM were both 10% lower than the corresponding values ​​in the positive reference control.

[0124] If the above conditions are not met, the batch of data is deemed invalid and needs to be retested.

[0125] (7) Data acquisition after calibration

[0126] The data that has passed sample-level quality control, protein item-level screening, and batch quality control are used to obtain the final cleaned data that can be used for analysis.

[0127] Furthermore, positive reference control EV samples from the same batch were used as calibration references. The raw signal values ​​of ITGA8 and L1CAM in the test sample and the positive reference control sample were read respectively, and the standardized signal values ​​were calculated as follows:

[0128] ITGA8 std =log2(R ITGA8 +1)−log2(R ITGA8,ref +1)

[0129] L1CAM std =log2(R L1CAM +1)−log2(R L1CAM,ref +1)

[0130] Among them, R ITGA8 and R L1CAM These are the original signal values ​​of ITGA8 and L1CAM in the sample under test, respectively; R ITGA8,ref and R L1CAM,ref These are the raw signal values ​​of the corresponding proteins in the same batch of positive reference control samples. Through the above standardization process, comparable protein expression signals can be obtained between different samples, and further used for joint scoring calculation.

[0131] Step 5: Result Interpretation:

[0132] The standardized protein expression results were expressed using the formula Score = 0.8 × ITGA8. std -0.6×L1CAM std

[0133]

[0134] Application examples

[0135] This case was approved by the Ethics Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, with ethics approval number

[2025] Lun Shen Zi (1363). All samples were obtained from patients undergoing assisted reproductive treatment at the Reproductive Medicine Center of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. Before sampling, the researchers fully informed the patients about the research objectives, sample usage, and potential risks, and obtained their written informed consent.

[0136] This study included high-quality blastocysts thawed and revived on day 5 of IVF-ET cycles. After blastocyst thawing, routine embryo transfer was performed. The culture medium of any remaining single embryos was collected under aseptic conditions for subsequent extracellular vesicle (EV) isolation. Each sample corresponded to a unique transfer cycle and embryo information, and blastocyst morphology scores and key clinical data were recorded simultaneously. Blastocyst morphology grading was performed using our center's standard clinical scoring system, with inclusion criteria of 20 high-quality blastocysts with 4AA or higher. All patients were followed up according to standard procedures after embryo transfer, and clinical pregnancy outcomes were recorded.

[0137] It should be noted that all high-quality blastocysts included in this comparison have undergone genetic testing, and no aneuploidy abnormalities were found. By including high-quality embryos with essentially the same genetic status in the same evaluation system, the interference of embryonic chromosomal aneuploidy factors on the transplantation outcome can be largely eliminated, thus allowing different samples to be compared on a relatively consistent baseline. This design helps reduce confounding bias caused by aneuploidy factors.

[0138] Furthermore, to minimize the impact of maternal endometrial factors on embryo implantation outcomes, all patients included in this study had no clear medical history affecting endometrial receptivity, including but not limited to endometriosis, significant endometrial lesions, intrauterine adhesions, endometrial polyps, and other intrauterine diseases affecting embryo implantation; and hysteroscopic examination confirmed that the uterine cavity morphology was basically normal, and endometrial biopsy showed no abnormal endometrial features that would significantly affect embryo implantation. Based on these conditions, implantation failure caused by significant endometrial abnormalities can be ruled out to a certain extent.

[0139] The expression levels of ITGA8 (integrin α8) and L1CAM proteins in the culture medium-derived EVs from the aforementioned 20 4AA or higher blastocysts were detected according to the method described in Example 1. First, the raw signal values ​​of ITGA8 and L1CAM in each sample were read; simultaneously, using the same batch of positive reference control EV samples as calibration references, the raw signal values ​​of ITGA8 and L1CAM in the reference control samples were recorded. It should be noted that the raw signal values ​​in this example are the effective UMI repeat counts of the corresponding protein probes in the sequencing data. The following raw signal values ​​were obtained:

[0140] Table 1 Raw signal values ​​of protein expression

[0141]

[0142] The expression values ​​of the two molecules were standardized according to the formula for calculating standardized signal values, taking V1 embryo samples as an example:

[0143] ITGA8 std =log2(16645+1)−log2(113+1)=7.19;

[0144] L1CAM std =log2(2322+1)−log2(89+1)=4.69;

[0145] The following results were obtained:

[0146] Table 2 Normalized signal values ​​of ITGA8 and L1CAM protein expression

[0147]

[0148] The results, based on the formula proposed in this invention and the actual clinical follow-up outcomes, are shown in the table below:

[0149] Table 3 Evaluation results and actual outcomes of this invention

[0150]

[0151] *TP indicates a true positive, TN indicates a true negative, FP indicates a false positive, and FN indicates a false negative;

[0152] The statistical results are shown in the table below:

[0153] Table 4 Evaluation Results of the Invention

[0154]

[0155] Considering that all selected embryos were of high quality with a morphological grade of 4AA or above, borderline embryos were statistically analyzed based on their high implantation potential, resulting in the following:

[0156] Specificity = TN / (TN + FP) = 8 / 9 = 88.89%;

[0157] Sensitivity = TP / (TP + FN) = 10 / 11 = 90.91%;

[0158] Accuracy = (TP + TN) / 20 = (10 + 8) / 20 = 90.00%;

[0159] Comparative Example 1

[0160] All other steps are the same as in the application example, but the final risk assessment standard is different, based solely on the ITGA8 standard (ITGA8). std An evaluation was conducted using a score of ≥6.20 (the standard for judging high planting potential). A score above 6.2 was considered high planting potential, and a score below 6.2 was considered low planting potential. The results are shown in the table below.

[0161] Table 5. Results of ITGA8 Single Standard Judgment

[0162]

[0163] Statistics show that:

[0164] Specificity = TN / (TN+FP) = 7 / (7+2) = 7 / 9 = 77.78%;

[0165] Sensitivity = TP / (TP+FN) = 9 / (9+2) = 9 / 11 = 81.82%;

[0166] Accuracy = (TP+TN) / (TP+FP+TN+FN) = (9+7) / 20 = 16 / 20 = 80.00%;

[0167] Comparative Example 2

[0168] All other steps are the same as in the application example, but the final risk assessment standard is different, based solely on the L1CAM standard: using L1CAM... std A value <4.95 is considered high planting potential, while a value above 4.96 is considered low planting potential. The results are shown in the table below:

[0169] Table 6 L1CAM Single Criterion Judgment Results

[0170]

[0171] Statistics show that:

[0172] Specificity = TN / (TN+FP) = 8 / (8+1) = 8 / 9 = 88.89%;

[0173] Sensitivity = TP / (TP+FN) = 9 / (9+2) = 9 / 11 = 81.82%;

[0174] Accuracy = (TP+TN) / (TP+FP+TN+FN) = (9+8) / 20 = 17 / 20 = 85.00%;

[0175] The performance metrics of the above methods are summarized in the table below:

[0176] Table 7 Performance Indicators of Each Method

[0177]

[0178] The results show that the ITGA8-L1CAM joint scoring method of the present invention is the best evaluation model in terms of accuracy, sensitivity, specificity, and AUC value.

[0179] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A combination of single vesicle membrane protein markers for assessing embryo implantation potential, characterized in that, The biomarker combination includes the single vesicle membrane protein ITGA8 and L1CAM.

2. Application of single vesicle membrane proteins ITGA8 and L1CAM as markers in the preparation of kits for evaluating embryo implantation potential.

3. A method for assessing embryo implantation potential, characterized in that, Embryo quality was assessed by detecting the expression levels of ITGA8 and L1CAM proteins in extracellular vesicles of embryo culture medium.

4. The method according to claim 3, characterized in that, Includes the following steps: Step 1: Extract the raw signal values ​​of the corresponding probes of ITGA8 and L1CAM in the sample to be tested, and summarize them according to the sample to obtain the sample-level raw signal values ​​of ITGA8 and L1CAM in each sample. Step 2: Calculate the standardized signal value as follows: ITGA8 std =log2(R ITGA8 +1)−log2(R ITGA8,ref +1); L1CAM std =log2(R L1CAM +1)−log2(R L1CAM,ref +1); Among them, R ITGA8 and R L1CAM These are the original signal values ​​of ITGA8 and L1CAM in the sample under test, respectively; R ITGA8,ref and R L1CAM,ref These are the original signal values ​​of the corresponding proteins in the same batch of positive reference control samples; the positive reference control samples are commercial EV reference materials or EV samples from stable cell lines. Step 3: Use the formula Score = 0.8 × ITGA8 std -0.6×L1CAM std The results were interpreted to assess the quality of the embryos.

5. The method according to claim 4, characterized in that, Step 1 involves extracting the raw signal values ​​of the corresponding probes for ITGA8 and L1CAM in the sample to be tested, including the following steps: Step 11: Collect the blastocyst culture medium on day 5 of the IVF-ET cycle after thawing and recovery. After the blastocyst recovery is completed, perform routine embryo transfer. Collect the remaining single embryo culture medium under aseptic conditions and extract EVs from the embryo culture medium. Step 12: Using the proximity coding technique PBA, single-particle level membrane proteomics detection was performed on EVs in a single embryo culture medium to obtain sequencing data; Step 13: Perform bioinformatics analysis on the sequencing data to obtain the raw signal values ​​of each protein in each single vesicle; Step 14: Extract the original signal values ​​of the corresponding probes for ITGA8 and L1CAM from the original signal values ​​of each protein, and summarize them by sample to obtain the sample-level original signal values ​​of ITGA8 and L1CAM in each sample.

6. The method according to claim 5, characterized in that, In step 11, size exclusion chromatography is used to extract EVs from the embryo culture medium.

7. The method according to claim 5, characterized in that, Step 12 specifically includes the following steps: Preparation of conjugated nucleic acid antibodies: Antibodies are conjugated to oligonucleotide sequences to prepare DNA-antibody probes, wherein the oligonucleotides include the following functional segments:

1. An antibody-specific protein tag region for distinguishing different antibody probes; 2. A molecular tag region for subsequent molecule counting and deduplication analysis; 3. A hybridization region complementary to the RCA product for sequence identification and extension in subsequent PBA reactions; the total length of the oligonucleotide is 40-80 nt, of which the protein tag region is 8-12 nt long; the molecular tag region is 8-12 nt long; and the hybridization region complementary to the RCA product is 15-25 nt long. Preparation of Rolling Circle Amplification (RCA) Products: Rolling circle amplification is performed using a circular DNA template to prepare an RCA product that can be used to recognize EVs. The RCA product contains the following functional regions:

1. a recognition region complementary to the DNA-antibody probe; 2. a 15nt composite tag region (EV tag) for recognizing EVs; 3. a fixed sequence region for subsequent extension reactions and PCR library construction; the circular DNA template is 60-120nt in length, of which the EV tag is 15nt in length. EV immobilization: An affinity capture strategy based on the cholera toxin B subunit for GM1 gangliosides on the EV membrane was used to immobilize the EVs to be tested on the bottom surface of the well. DNA-antibody probe incubation: Add the DNA-antibody probe mixture to the wells of the immobilized EVs to allow it to bind to the EV surface membrane proteins; PBA reaction: The RCA product is added to the well of an EV that has been bound to a DNA-antibody probe, so that the probe DNA hybridizes with the complementary sequence in the RCA product; then an extension reaction occurs under the action of DNA polymerase, so that multiple adjacent probes on the same EV can share the same complex tag. PCR amplification, library construction and sequencing: The DNA fragments obtained from the extension reaction are amplified by PCR to enhance the signal and construct a next-generation sequencing library. The DNA fragments contain:

1. a protein tag; 2. a molecular tag UM; 3. an EV tag; and 4. a PCR primer binding immobilization region.

8. The method according to claim 7, characterized in that, Step 13 specifically includes the following steps: splitting, quality-controlling, and comparing the raw sequencing data obtained in step 12; identifying the corresponding protein barcode, molecular tag UMI, and EV tag in each sequence; deduplicating and counting the effective reads belonging to the same single vesicle, the same protein probe, and with different UMIs; and obtaining the original signal values ​​of each protein in each single vesicle.

9. The method according to claim 5, characterized in that, Step 14 specifically includes the following steps: Raw signal extraction: For each sample, the valid sequencing reads corresponding to ITGA8 and L1CAM are counted by removing duplicates from the unique molecular tag UMI to obtain the raw signal value of the target protein; at the same time, the total valid signal value, total UMI count and number of detected protein entries corresponding to all protein probes in the sample are counted as the basis for subsequent sample quality control and standardization analysis. Initial screening at the sample level: Preliminary quality control is performed on the samples. Samples meeting the following criteria are retained, while those not meeting the criteria are discarded:

1. The total number of valid reads is not less than 1 × 10⁻⁶. 4 2. The total UMI count is not less than 5 × 10 3 3. The number of detected protein entries shall not be less than 20% of the total number of protein probes; 4. The total effective signal value of the negative blank control well is less than 5% of the median of the same batch of samples; the samples retained after the above screening are recorded as cleaned samples; Protein entry-level screening: Protein entries in each sample are further screened, and protein entries that meet the following conditions are retained:

1. The original UMI count in the sample is ≥3; 2. The protein was detected in at least 10% of all valid samples; 3. The protein signal value was more than twice the mean of the corresponding protein signal in the negative control; Protein entries that did not meet the above conditions were considered low-confidence signals and were removed from the subsequent normalization analysis. The data retained after filtering is denoted as cleaned data; Preliminary library size correction: Using the total effective signal value or total UMI count of each sample as an indicator of sample sequencing scale, preliminary correction is made for differences in sequencing depth between samples; Inter-sample normalization: After the initial library size correction, the trimmed mean normalization method (TMM) is used to normalize the inter-samples. Finally, the original signal values ​​of ITGA8 and L1CAM are corrected by combining the library size and the normalization factor to obtain the normalized protein signal values. Batch quality control: For each batch of testing, negative blank control and positive reference control samples are set up simultaneously. The negative blank control does not include EV samples and is used to assess background noise. Positive reference controls are commercially available EV reference materials or EV samples from stable cell lines, used to monitor probe binding efficiency, amplification efficiency, and batch-to-batch stability. Positive reference control samples also serve as calibration references, standardizing protein signals from different batches of test samples to the same benchmark. A batch of data is considered valid if all of the following conditions are met:

1. The total number of valid reads in the positive reference control sample reaches 80%-120% of the preset reference range; 2. The standardized signal values ​​of ITGA8 and L1CAM in the positive reference control sample deviate from the historical mean by no more than ±20%; 3. The signal values ​​of ITGA8 and L1CAM in the negative control are both less than 10% lower than the corresponding values ​​in the positive reference control. If the above conditions are not met, the batch of data is considered invalid and needs to be retested. Data acquisition after calibration: The data that has passed sample-level quality control, protein item-level screening and batch quality control are used to obtain the final cleaned data that can be used for analysis.

10. The method according to claim 4, characterized in that, In step 3, when the result is ≥ 2.60, it indicates that the embryo has high implantation potential and can be transferred first; when the result is < 1.80, it indicates that the embryo has low implantation potential and is not recommended for priority transfer; when the result is between 1.80 and 2.59, it is judged as a borderline state, and a comprehensive judgment is made based on morphological and clinical factors.

Images