Site, primer, kit and method for detecting cervical cancer DNA methylation

Abstract

The invention discloses a site, a primer, a kit and a method for detecting cervical cancer DNA methylation, a DNA methylation site LRFN5 is obtained through library construction, hybrid capture, high-throughput sequencing and qMSP detection on a cervical cancer tissue sample through screening, and it is found that the LRFN5 site has differentiated expression in the early stage (1-2 stages) of CIN compared with cervical cancer samples in other stages, so that the DNA methylation site LRFN5 can be used for detecting cervical cancer DNA methylation. Therefore, the site can be used for early recognition of the cervical cancer, and early diagnosis and early treatment of the cervical cancer are facilitated.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology, specifically relating to a site, primer, reagent kit, and method for detecting DNA methylation in cervical cancer. Background Technology

[0002] Cervical cancer, also known as uterine cancer, is a malignant tumor of the female reproductive tract that occurs in the cervix. The most common type is squamous cell carcinoma, primarily caused by human papillomavirus (HPV) infection, especially HPV types 16 and 18. Globally, an estimated 570,000 new cases were diagnosed in 2018, accounting for 7.5% of all cancer deaths in women. Of these, 85... 90% of cases are squamous cell carcinoma (SCC), while the remaining 10% are... Of the 15% of cases, the majority are adenocarcinomas. The clinical symptoms of cervical cancer are often not obvious, or the HPV test is negative and the diagnosis is missed, or it only presents with symptoms similar to cervicitis. This leads to difficulties in diagnosis and the risk of missed diagnosis. Once obvious symptoms appear, the disease has often progressed to the middle or late stage, and the prognosis is poor, missing the best time for treatment.

[0003] Currently, commonly used cervical cancer screening methods include cervical cytology testing and HPV nucleic acid testing. Cervical cytology testing is a cytological screening method, requiring a strict operating environment, lacking universality, and exhibiting extremely low sensitivity and poor reliability. While HPV nucleic acid testing has high sensitivity and specificity, its laboratory processing requirements and costs are high, and the result acquisition cycle is long. Abnormal DNA methylation usually occurs in the cervical intraepithelial neoplasia (CIN) stage caused by persistent HPV infection, earlier than cellular morphological changes, enabling early identification of precancerous lesions (such as CIN2 / 3). Compared to the high sensitivity but low specificity of HPV testing (susceptible to interference from transient infections), methylation markers (such as PAX1, SOX1, and ZNF582) are more strongly associated with high-grade lesions, reducing the risk of excessive referral for colposcopy. Therefore, human gene methylation testing has become a research hotspot in the field of cervical cancer screening technology due to its early warning value and high specificity. To date, more than 100 types of human gene methylation have been found to be closely related to the occurrence and development of cervical cancer.

[0004] Chinese patent CN118755838A discloses a composition, kit, and uses for detecting cervical cancer and high-grade cervical lesions. It discloses a detection reagent for detecting methylation levels in the JAM3, PCDHGB7, and SORCS1 gene regions, comprising primers and nucleic acid probes. This reagent can reliably achieve high sensitivity and specificity in detecting high-grade cervical lesions and cervical cancer with fewer tests, avoiding overdiagnosis and potentially alleviating the current problem of excessive resource consumption and low efficiency in cervical cancer screening. However, it primarily targets the triage of detection for cervical intraepithelial neoplasia grade 2 and above (CIN2+) and cervical intraepithelial neoplasia grade 3 and above (CIN3+).

[0005] Chinese patent CN113549694A discloses a novel method for detecting methylation genes in cervical cancer, including primers for probe-based quantitative real-time PCR, a detection probe, and a specific blocking probe. The detection probe specifically binds to the methylation sequence target genes PAX1, FAM19A4, and hsa. mir124 2. ATP10A, HAS1, and internal reference ACTB (β) The actin gene, based on PAX1, FAM19A4, and hsa mir124 2. The methylation value of the sample to be tested was determined by the relative fluorescence CT values ​​of the PCR amplification results of ATP10A, HAS1, and ACTB genes. The methylation values ​​of PAX1, FAM19A4, and hsa were calculated respectively. mir124 2. Methylation values ​​of ATP10A and HAS1 relative to the ACTB gene. This improves the sensitivity and specificity of cervical cancer methylation gene detection. However, it requires the use of probes and blocking probes to achieve accurate detection, and is mainly used to differentiate between cervical cancer patients and non-cervical cancer controls.

[0006] Currently, widely recognized gene loci include the Cervical Cancer Methylation Gene Detection Kit from Shanghai GenenoBio, but its development is primarily based on European and American women and mainly targets patients who have progressed to the cancer stage. Studies have shown that while the currently popular GynTect and QIAsure kits can detect all CIN3 and cervical cancer cases, the detection rate of the two tests is not significantly different across different cervical lesion grades (Dippmann et al., 2020). Therefore, it is essential to screen for cervical cancer methylation loci more suitable for Chinese women, and to identify differentially expressed methylation loci in early CIN (stages 1-2). This facilitates early diagnosis and treatment, improving the survival and cure rates of cervical cancer. Furthermore, current hospital checkups still rely on traditional TCT and HPV tests. Screening for cervical cancer methylation loci suitable for local conditions and developing a kit can effectively reduce medical labor and economic costs. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a site, primers, kit, and method for detecting DNA methylation in cervical cancer. A cervical cancer methylation site suitable for Chinese women and capable of differential expression in early CIN (stages 1-2) has been screened, facilitating early identification of cervical cancer.

[0008] To achieve the above objectives, the present invention provides a site for detecting DNA methylation in cervical cancer, wherein the DNA methylation site is LRFN5 and / or LHX8, the sequence of which is SEQ ID NO:1, and the sequence of LHX8 is SEQ ID NO:4.

[0009] The present invention also provides primers for detecting cervical DNA methylation, wherein the primers are LRFN5 F, LRFN5R and / or LHX8 F, LHX8 R, wherein the sequence of LRFN5 F is SEQ ID NO:2, the sequence of LRFN5 R is SEQ ID NO:3, the sequence of LHX8 F is SEQ ID NO:5, and the sequence of LHX8 R is SEQ ID NO:6.

[0010] The present invention also provides a kit for detecting DNA methylation in cervical cancer, the kit comprising the primer pairs described above.

[0011] Preferably, the kit further includes at least one of PCR reaction solution, template DNA, polymerase, and nuclease-free water.

[0012] More preferably, the template DNA is a mixture comprising human unmethylated whole-genome DNA, human genomic DNA treated with bisulfite, and untreated human genomic DNA.

[0013] The present invention also provides a method for DNA methylation in cervical cancer, comprising the following steps: (1) Extracting DNA from cervical exfoliated cells; (2) DNA was subjected to bisulfite conversion treatment to obtain bis-DNA; (3) Using bis-DNA as a template, qMSP detection was performed, fluorescence signals were collected, and Ct values ​​were read based on the fluorescence signals; (4) Construct a model using Ct and Tm values, and use the model to directly output the cervical cancer staging results.

[0014] Preferably, the Nanodrop / A260 / A280 of the DNA in step (1) is >1.8 and has good integrity.

[0015] Preferably, the qMSP detection in step (3) uses BS-GUSB as the internal reference gene.

[0016] Preferably, the primers used for qMSP detection in step (3) are LRFN5 F, LRFN5 R and / or LHX8 F, LHX8R.

[0017] Preferably, the model described in step (4) is the CIN 2+ split model.

[0018] The beneficial effects of this invention are as follows: the LRFN5 site was screened and found to be differentially expressed in the early stage (1-2) of CIN, which can be used for cervical cancer staging, thereby realizing the early identification of cancerous changes in cervical cancer, which facilitates early diagnosis and treatment of cervical cancer. Attached Figure Description

[0019] Figure 1 This is a feature importance distribution map of CpG sites in Example 1.

[0020] Figure 2 The figures show the distribution maps of ΔCt and Tm values ​​for the five sites in Example 3. In the figures, a is the distribution map of Tm value for LRFN5, b is the distribution map of Tm value for LHX8, c is the distribution map of Tm value for ASCL1, d is the distribution map of Tm value for HAS1, e is the distribution map of Tm value for PCDHGB7, f is the distribution map of ΔCt value for LRFN5, g is the distribution map of ΔCt value for LHX8, h is the distribution map of ΔCt value for ASCL1, i is the distribution map of ΔCt value for HAS1, and j is the distribution map of ΔCt value for PCDHGB7.

[0021] Figure 3 The figure shows the performance of the CIN2+ shunt model in Example 4. In the figure, a is the AUV distribution, b is the positive rate distribution of each lesion stage, c is the calibration curve, and d is the decision curve.

[0022] Figure 4 The figure shows the validation results of the CIN2+ triage model in Example 5. In the figure, a is a bar chart of the validation results of ROC / AUC for different feature sets, b is a validation curve for different numbers of sites, and c is a distribution chart of the importance of SHAP features of the final model. Detailed Implementation

[0023] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings and specific embodiments. It is worth noting that the following embodiments are only preferred embodiments of the present invention and should not be construed as limiting the present invention. The scope of protection of the present invention should be determined by the contents of the claims. Modifications and substitutions made by those skilled in the art to the technical solution of the present invention without creative effort all fall within the scope of protection of the present invention.

[0024] Example 1: Screening of methylation sites (1) Sample preparation: 686 cervical cancer CC tissue samples and 163 normal cervical tissue samples were collected (tissue samples refer to cervical samples obtained by formalin fixation after live sampling). Genomic DNA was extracted using QIAamp DNA Mini Kit to ensure that the Nanodrop / A260 / A280 of the genomic DNA was >1.8 and that the integrity was good. Among them, the cervical cancer CC tissue samples covered 5 histological categories, namely CIN1 (113 cases), CIN2 (155 cases), CIN2-3 (180 cases), CIN3 (111 cases) and invasive carcinoma (127 cases). Samples with a history of total hysterectomy, immunodeficiency or pregnancy at the time of sampling were excluded. (2) Genomic DNA acquisition and bisulfite treatment: Genomic DNA was extracted from the sample described in step (1) using the QIAamp DNA Mini kit, and then the genomic DNA was bisulfite treated using the EZ DNA Methylation-Lightning Kit (ZymoResearch) to convert unmethylated cytosine (C) to uracil (U), while methylated C remained unchanged, to obtain bis-DNA; (3) Hybrid capture probe design: Literature review yielded cervical cancer-related genes MIR124-2, ZIC1, ASCL1, SOX17, HAS1, DLX1, RXFP3, FAM19A4, POU4F3, ZNF582, and PCDHGB7. The region 2000 bp upstream of the gene was used as the promoter region, and the CpG islands of the gene were screened using UCSC Genome Browser. The CpG islands were used as the target regions, and the fragments within ±500 bp upstream and downstream of the target region were captured using RocheSeqCap EZ. Probes of 80-120 bp were designed to avoid repetitive sequences. (4) Library construction: The bis-DNA was repaired at the end, A tail was added and adapter was ligated using the Sino methylation library capture kit (Cat. JZ9651B); then the library was amplified and enriched using PCR technology; the PCR reaction system is shown in Table 1 and the reaction procedure is shown in Table 2. Table 1 PCR reaction system

[0025] Table 2 PCR reaction procedure

[0026] (5) Hybridization capture: Take 400-500 ng of library and concentrate to 3.4 µL, add 5.6 µL of blocking buffer, deform at 95 °C for 5 min, and then keep at 65 °C; add 20 µL of hybridization buffer and biotin-labeled capture probe (customized by Sinomics), hybridize for 24 h, capture probe-library complex with streptavidin beads, and elute unbound fragments with 21 µL of 0.1 M NaOH solution; perform CT transformation using a sulfite conversion kit and purify to obtain 20 µL of transformation product.

[0027] (6) PCR amplification and enrichment of captured fragments: The transformation product was amplified and enriched by PCR, and then the PCR product was purified by magnetic beads to obtain 20µL of product. After amplification, the Qubit quantitative library was used and the fragment size was detected by Agilent Bioanalyzer. The PCR system is shown in Table 3 and the PCR program is shown in Table 4. The size of the qualified library should be between 250bp and 500bp and free from adapter-dimer contamination. Sequencing was performed using Illumina or BGI sequencers, and the sequencing data of each sample should be greater than 0.5G.

[0028] Table 3 PCR reaction system

[0029] Note: The PCR amplification mixture is the product enriched, transformed and purified in step (6); the sequence of the Indexing universal primer is SEQ ID NO:15 and / or SEQ ID NO:16.

[0030] Table 4 PCR reaction procedure

[0031] Example 2 (1) High-throughput sequencing: The PCR amplification enrichment and capture fragments obtained in Example 1 were sequenced using Illumina NovaSeq to obtain the original FASTQ file; wherein each sample had ≥10M reads with a coverage depth of ≥50×. (2) Bioinformatics analysis: FastQC (v0.12.1) was used to perform an overall quality assessment of all original FASTQ files. Then, fastp (v0.23.4) was used to remove adapter sequences, cut low-quality bases (sliding window Q20, window 4 bp), remove PCR repeats, and filter reads with a length <50bp. Reads that passed quality control were selected and aligned to the human reference genome GRCh38 using Bismark (v0.24.2). CpG site methylation information was then extracted using bismark_methylation_extractor, retaining only sites with a coverage depth ≥10× to reduce random noise, resulting in CpG site samples ( Figure 1 This ensures that the sample meets the requirements of an alignment rate ≥ 70% and a duplication rate ≤ 20%. (3) Differential methylation analysis: Under R 4.3.2 environment, differential methylation sites (DMC) / regions (DMR) were screened for CpG site samples obtained in step (2) using methylKit (v1.24.0). Fisher exact test was used and Benjamini–Hochberg method was used to correct multiple comparisons. The decision threshold was: |Δβ|≥0.20 (cervical cancer vs normal) and FDR<0.05; (4) Priority scoring: The DMC / DMR is weighted and scored based on the following three indicators, and the top 5% of high-priority sites are selected: ① High-frequency methylation: ≥70% of cervical cancer samples showed methylation rates higher than the median within the group; ② Functional relevance: Located in the promoter (TSS±2 kb) or annotated enhancer region, and intersects with KEGG "cervicalcancer pathway" or HPV-related pathways; ③ Literature / Patent Support: ≥5 hits in PubMed and Lens.org searches; (5) ROC analysis was used to evaluate the diagnostic value of individual sites in the top 5% of high-priority sites in step (4), and only sites with AUC>0.80 were retained; then glmnet (v4.1.8) was used to perform LASSO logistic regression, and the most explanatory sites were selected to construct the final feature set (Table 5). Table 5 Final Feature Set

[0032] (6) Validation: MethPrimer was used to design bisulfite-specific primers / probes for each feature site in the final feature set, so that the fragment length of the amplified product was 80-120bp, the Tm value of the primers was between 58-62℃, and the GC value was 40-60%; then OligoAnalyzer was used to screen for dimers and hairpin structures, and finally 5 CpG sites were obtained: LRFN5, LHX8, ASCL1, HAS1, and PCDHGB7.

[0033] Example 3 (1) CpG sites obtained from Example 2: LRFN5, LHX8, ASCL1, HAS1, PCDHGB7, and the internal reference gene was determined to be GUSB as the target detection gene; (2) Using the bisulfite-specific primers / probes designed for each characteristic site in Example 2 as primers, and the bis-DNA in step (2) of Example 1 as a template, quantitative methylation-specific PCR (qMSP) was performed. The primer information is shown in Table 6, the reaction system is shown in Table 7, and the reaction procedure is shown in Table 8. Table 6 Primer Information

[0034] Table 7 qMSP reaction system

[0035] Table 8 qMSP Reaction Procedure

[0036] (3) Collect qMSP fluorescence signal, read Ct value based on fluorescence signal, and calculate ΔCt value by subtracting Ct value from internal reference gene GUSB, and calculate Tm for each sample (each sample is repeated three times).

[0037] The results showed that as the histological severity of cancer progressed from normal epithelial tissue to invasive carcinoma, the ΔCt value gradually decreased, indicating a gradual increase in promoter methylation level; except that the Tm value gradually increased with the severity of the lesion. Figure 2 ).

[0038] LRFN5 showed differential expression in early CIN (stages 1-2) compared to samples from other stages. Figure 2 f) indicates that LRFN5 can identify cancerous changes in the early stages of cervical cancer, facilitating early diagnosis and treatment of cervical cancer.

[0039] Example 4 (1) Take the ΔCt value and Tm value of the 5 sites obtained in Example 3, and collect the cytological examination results, high-risk HPV status and age status of different tissue samples, and perform benchmark testing through algorithms; the algorithms include logistic regression, random forest, custom table converter and TabPFN; (2) TabPFN was used to further optimize the information of the five sites, and the possible combinations of all sites were evaluated by nested cross-validation using the cytological examination results, ΔCt value and Tm value of the five sites. Finally, two sites were selected: LRFN5 and LHX8. (3) Using the ΔCt and Tm values ​​of LRFN5 and LHX8 and the results of cytological examination, the patient-level CIN2+ risk is output, and a CIN2+ triage model is constructed accordingly.

[0040] The results show that the AUC of the CIN2+ shunt model reaches 0.947 at the pre-set operating point ( Figure 3 a) The sensitivity was 87.9%, the specificity was 85.1%, the Brill score was 0.079, the slope was 1.15, and the intercept was -0.05, indicating that the model calibration was effective. Figure 3 c); Decision curve analysis shows that, within the screening-related threshold range, this model has a higher net benefit compared to the "full referral / no referral" and "cytology-only" strategies. Figure 3 d).

[0041] Example 5 Using the CIN2+ triage model constructed in Example 4, cross-validation was performed on cytological examination results, ΔCt and Tm values ​​to explore the optimal feature model.

[0042] The results showed that adding Tm to ΔCt significantly improved the model's discrimination, with AUC increasing by 0.072 (from 0.857 to 0.929); while adding cytological examination results to ΔCt+Tm only slightly increased AUC by 0.018 (reaching 0.947). Figure 4 a). For reference, the AUC for cytology alone was 0.712. Gene-scale analysis showed that the model performance plateaued when the number of genes was 2; therefore, LRFN5 and LHX8 were selected as the final selection sites. Figure 4b). In the final five-feature model (cytological examination results + ΔCt and Tm of the two genes), feature attribution analysis showed that the two Tm features were the most important, exceeding ΔCt and cytological examination results. Figure 4 c) indicates that LRFN5 and LHX8 are used as detection sites, and their Tm and ΔCt values ​​are used to differentiate cervical cancer stages in tissues at different stages for early identification and differentiation.

Claims

1. A method for detecting DNA methylation sites in cervical cancer, characterized in that: The DNA methylation sites are LRFN5 and / or LHX8, with the sequence of LRFN5 being SEQ ID NO:1 and the sequence of LHX8 being SEQ ID NO:

4.

2. A primer for detecting cervical DNA methylation, characterized in that: The primers are LRFN5 F, LRFN5 R and / or LHX8 F, LHX8 R, wherein the sequence of LRFN5 F is SEQ ID NO:2, the sequence of LRFN5 R is SEQ ID NO:3, the sequence of LHX8 F is SEQ ID NO:5, and the sequence of LHX8 R is SEQ ID NO:

6.

3. A kit for detecting DNA methylation in cervical cancer, characterized in that: The kit includes the primers as described in claim 2.

4. A kit for detecting DNA methylation in cervical cancer according to claim 3, characterized in that: The kit also includes at least one of the following: PCR reaction solution, template DNA, polymerase, unmethylated primer pair, and nuclease-free water.

5. A kit for detecting DNA methylation in cervical cancer according to claim 4, characterized in that: The template DNA is a mixture of unmethylated whole-genome DNA, bisulfite-treated human genomic DNA, and human genomic DNA.

6. A method for detecting DNA methylation in cervical cancer, characterized in that: Includes the following steps: (1) Extract DNA from cervical tissue or exfoliated cervical cells; (2) DNA was subjected to bisulfite conversion treatment to obtain bis-DNA; (3) Using bis-DNA as a template, qMSP detection was performed with primers, and fluorescence signals were collected. Based on the fluorescence signals, Ct and Tm values ​​were read. (4) Construct a model using Ct and Tm values, and use the model to directly output the cervical cancer staging results.

7. A method for cervical cancer DNA methylation according to claim 6, characterized in that: The DNA described in step (1) has a Nanodrop / A260 / A280 ratio > 1.8 and good integrity.

8. A method for DNA methylation in cervical cancer according to claim 6, characterized in that: The internal reference gene for qMSP detection in step (3) is BS-GUSB.

9. A method for DNA methylation in cervical cancer according to claim 6, characterized in that: The primers used for qMSP detection in step (3) are LRFN5 F, LRFN5 R and / or LHX8 F, LHX8 R as described in claim 2.

10. A method for DNA methylation in cervical cancer according to claim 6, characterized in that: The model described in step (4) is the CIN 2+ split model.

Images