Oral microbial gene markers for distinguishing cholangiocarcinoma from healthy people and their application

Abstract

The present invention belongs to the field of biomedicine technology, and specifically relates to an oral microbial gene marker for distinguishing cholangiocarcinoma from healthy people and its application. The present invention provides an oral microbial gene marker for distinguishing cholangiocarcinoma from healthy people, which is composed of three microbial genes shown in SEQ ID NO: 1-3, and the microbial genes are significantly enriched in the human oral cavity. Therefore, the microbial gene discrimination model of the present invention achieves good discrimination ability between cholangiocarcinoma and healthy people. The feasibility and applicability of the discrimination model are also verified.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine technology, and specifically relates to an oral microbial gene marker for distinguishing cholangiocarcinoma from healthy people and its application. Background Art

[0002] Liver cancer is one of the most common malignancies worldwide. Primary liver cancer includes hepatocellular carcinoma (HCC) (accounting for 75%-85%) and cholangiocarcinoma (CCA) (accounting for 10%-15%), as well as other rare types. In recent years, the prevalence and mortality of cholangiocarcinoma have increased. Cholangiocarcinoma is sporadic, and there are regional differences in risk factors. Cholangiocarcinoma is difficult to diagnose due to its anatomical location, growth pattern, and lack of clear diagnostic criteria. Although new diagnostic biomarkers are frequently discovered and reported, there are many limitations. Liver biopsy is the gold standard for diagnosing CCA, however, it is an invasive procedure and is sometimes not possible to perform. Therefore, there is an urgent need to find new, non-invasive diagnostic biomarkers for CCA.

[0003] The oral microecology system is closely related to health and disease. It is well known that dysbiosis of oral microflora not only leads to oral diseases such as caries, periodontitis, and oral mucosal diseases, but also involves systemic diseases, including gastrointestinal diseases such as inflammatory bowel disease, cirrhosis, and pancreatic cancer, nervous system diseases such as Alzheimer's disease, endocrine system diseases such as diabetes and polycystic ovary syndrome, immune system diseases such as rheumatoid arthritis, and cardiovascular system diseases such as atherosclerosis.

[0004] Key functional bacteria in the human microbiome can become new biomarkers for human diseases. Disease diagnostic models based on the intestinal microbiome have been widely used in multiple diseases. Qin J et al. first used metagenomics to report the correlation between intestinal microecology and type 2 diabetes, and pointed out that 23 bacterial species may become intestinal microbial markers for distinguishing type 2 diabetes. Qin N, Li A et al. (Nature magazine in 2014) analyzed the intestinal microecological structure of patients with cirrhosis in the Chinese population through metagenomic sequencing technology, identified 15 markers specific to cirrhosis at the microbial gene and functional levels, and created a highly accurate index for distinguishing patients with cirrhosis. The characteristics of the oral microecology or the differentiation model based on oral microorganisms are increasingly being widely reported and recognized as a tool for distinguishing specific diseases or tumors. Cui J et al. studied 78 gastritis patients and 50 healthy individuals and proposed that oral microecology can be used as a non-invasive marker for gastritis and its precancerous lesions. Lu H, pointed out that two bacterial genera (Protobacter and Clostridium) may become oral microbial markers for distinguishing liver cancer, which is a non-invasive diagnostic model established based on the oral microbiome. Therefore, oral microbiota may be a powerful tool for the diagnosis of different diseases. However, oral microbiota models for distinguishing cholangiocarcinoma from healthy subjects have not been reported. Summary of the invention

[0005] The present invention provides an oral microbial gene marker for distinguishing cholangiocarcinoma from healthy people, which consists of three microbial genes shown in SEQ ID NO: 1-3, and the microbial genes are significantly enriched in the human oral cavity.

[0006] >OTU20

[0007] CCTACGGGTGGCTGCAGTGGGGAATTTTGGACAATGGGCGAAAGCCTGATCCAGCAATGCCGCGTgtgtgATGAAGGCCTTCGGGTTGTAAAGCACTTTTGGCGGGAACGAAAAGGACTGTGCCAATACCACGGTTCGATGACGGTACCCGCAGAATAAGCACCGGCTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGAGTGCGCAGGCGGTTTTGCAAGACCGATGTGAAATCCCCGGGCTTAACCTGGGAACTGCATTGGTGACTGCAAGGCTAGAGTgtgtCAGAGGGAGGTGGAATTCCGCATGTAGCAGTGAAATGCGTAGATATGCGGAGGAACACCGATGGCGAAGGCAGCCTCCTGGGATAACACTGACGCTCATGCACGAAAGCGTGGGGAGCAAACAGGATTAGATACCCTAGTAGTC

[0008] >OTU30

[0009] CCTACGGGTGGCTGCAGTGAGGAATATTGGTCAATGGGCGAGAGCCTGAACCAGCCAAGTAGCGTGCAGGATGACGGCCCTCCGGGTTGTAAACTGCTTTTAGTTGGGAATAAaaaaaGGGACTTGTCCCTTCTTGTATGTACCTTCAGAAaaaGGACCGGCTAATTCCGTGCCAGCAGCCGCGGTAATACGGAAGGTCCAGGCGTTATCCGGATTTATTGGGTTTAAAGGGAGCGTAGGCGGATTGTTAAGTCAGCGGTTAAAGGGTGTGGCTCAACCATGCATTGCCGTTGAAACTGGCGATCTTGAGTGCAGACAGGGATGCCGGAATTCGTGGTGTAGCGGTGAAATGCTTAGATATCACGAAGAACTCCGATCGCGAAGGCAGGTGTCCGGGCTGCAACTGACGCTGAGGCTCGAAAGTGTGGGTATCAAACAGGATTAGATACCCTAGTAGTC

[0010] >OTU51

[0011] CCTACGGGTGGCTGCAGTGGGGAATATTGCACAATGGGCGCAAGCCTGATGCAGCGACGCCGCGTGAGGGATGGAGGCCTTCGGGTTGTAAACCTCTTTCGCCAGTGAAGCAGGCCTGTCCCTTGTGGGTGGGTTGACGGTAGCTGGATAAGAAGCGCCGGCTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGGCGCGAGCGTTGTCCGGAATTATTGGGCGTAAAG AGCTCGTAGGCGGCTGGTCGCGTCTGTCGTGAAATCCTCTGGCTTAACTGGgggCTTGCGGTGGGTACGGGCCGGCTTGAGTGCGGTAGGGGAGACTGGAACTCCTGTGTAGCGGTGGAATGCGCAGATATCAGGAAGAACACCGGTGGCGAAGGCGGGTCTCTGGGCCGTTACTGACGCTGAGGAGCGAAAGCGTGGGGAGCGAACAGGATTAGATACCCTAGTAGTC

[0012] In addition, the present invention also provides a detection reagent, including primers for detecting the three microbial genes shown in SEQ ID NOs: 1-3. The primer sequences are SEQ ID NOs: 4-5, and the primer sequences are as follows:

[0013] Primers

[0014] Sequencing region V3+V4: 338F-806R

[0015] Upstream primer: 338F ACTCCTACGGGAGGCAGCA

[0016] Downstream primer: 806R GGACTACHVGGGTWTCTAAT

[0017] Saliva samples were collected from the subjects, total microbial DNA was extracted, and 16S rDNA Miseq sequencing of the microbial DNA was completed to detect the presence of the three microbial genes shown in SEQ ID NO: 1-3.

[0018] Furthermore, saliva samples were collected from the subjects, total microbial DNA was extracted, and 16SrDNA Miseq sequencing of the oral flora was performed. Based on high-throughput sequencing data, a microbial differentiation model between cholangiocarcinoma and healthy people was established, and the probability of disease (POD) index of cholangiocarcinoma was established; the POD index can be used to calculate its differentiation ability for verification.

[0019] Specifically include:

[0020] (1) Saliva samples were collected from the subjects (cholangiocarcinoma patients and healthy people), and the total microbial DNA in the saliva samples was extracted according to the standard DNA extraction method. The 16S rDNA high-throughput sequencing of the oral flora was completed on the Illumina MiSeq platform;

[0021] (2) Based on high-throughput sequencing data, in the cohort of the microbial discrimination model, between 50 cases of cholangiocarcinoma and 100 healthy people, a random forest model was used to identify the best three microbial gene markers for the model through a five-fold cross-validation algorithm.

[0022] (3) Based on three microbial gene markers, the probability of disease (POD) index of cholangiocarcinoma was calculated by using the ratio of randomly generated decision trees.

[0023] (4) The microbial discrimination model had a discrimination ability of 99.22% between 50 cholangiocarcinoma patients and 100 healthy controls. The POD index was significantly increased in cholangiocarcinoma patients, and there was a significant difference between the two groups (p < 0.0001).

[0024] (5) In the validation set, the microbial discrimination model had a discrimination ability of 98.08% between 24 patients with cholangiocarcinoma and 50 healthy controls; the POD index was significantly increased in patients with cholangiocarcinoma, and there was a significant difference between the two groups (p < 0.0001).

[0025] Therefore, the microbial gene discrimination model of the present invention has achieved good discrimination ability between cholangiocarcinoma and healthy people, and also verified the feasibility and applicability of the discrimination model.

[0026] In addition, a kit for distinguishing oral microbial models of cholangiocarcinoma and healthy people is also provided, comprising primers for detecting three microbial genes shown in SEQ ID NO: 1-3.

[0027] The present invention also provides an application of an oral microbial gene marker for distinguishing cholangiocarcinoma from healthy people in establishing an oral microbial model for distinguishing cholangiocarcinoma from healthy people.

[0028] The microbial differentiation model is suitable for distinguishing cholangiocarcinoma patients from healthy people.

[0029] The saliva of the subject is tested to determine whether the sample contains the microbial gene and whether a microbial gene model can be established to distinguish between cholangiocarcinoma patients and healthy people.

[0030] An oral microbial model for distinguishing cholangiocarcinoma from healthy people, comprising three microbial genes shown in SEQ ID NO:1-3.

[0031] The specific operation steps of the present invention are as follows:

[0032] (1) According to the design principles of prospective clinical trials, the research design of the present invention is as follows: Figure 1 The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University. All enrolled patients signed the informed consent for the study protocol and the informed consent for clinical sample collection.

[0033] (2) Each enrolled cholangiocarcinoma patient and healthy person provided a fresh saliva sample. The researchers divided the sample into 10 portions of 500ul each and immediately stored them in a -80°C refrigerator. The extraction method of total saliva bacterial DNA was carried out according to the instructions of the kit.

[0034] (3) Complete the amplification of total DNA samples of salivary bacteria and the construction of DNA libraries, and complete 16S rDNA sequencing on the Illumina Miseq sequencing platform. All output sequences complete basic preprocessing and basic bioinformatics analysis.

[0035] (4) Randomly select an equal number of sequences from all samples and splice them into corresponding 16S rDNA gene sequence classification units (Operational Taxonomy Units, OTUs) according to the UPARSE pipeline. Collect and organize the OTUs gene sequences of all samples. Based on the microbial gene sequences, use the RDP classifier version 2.6 for annotation.

[0036] (5) Based on the representative sequences generated by high-throughput sequencing data, the OTUs frequency files of microbial gene markers were calculated. These OTUs were used in a correlation study to identify the abundance of OTUs between cholangiocarcinoma patients and healthy people. The Wilcoxon test method was used to statistically analyze the differences in microbial gene markers between cholangiocarcinoma patients and healthy people.

[0037] (6) In the training set of the microbial discrimination model, including 50 patients with cholangiocarcinoma and 100 healthy people, the screened OTUs abundance files were used to screen microbial gene markers in a random forest model (R software 3.4.1 and random forest package 4.6–12) using a five-fold cross-validation algorithm (software parameters were default except for setting "importance=TRUE"). Five trials of five-fold cross-validation were used to obtain a cross-validation error curve, in which the minimum cross-validation error point was used as the cut-off value. The minimum cross-validation error value plus the standard deviation of the corresponding value was the cut-off value. A set of OTUs markers with an error rate of less than 3 and less than the cut-off value was screened out, and the set with the minimum number of OTUs was selected as the best set of microbial gene markers. Finally, the best 3 microbial gene markers for this model were identified ( Figure 2 ). The gene sequences of the three selected microbial OTUs markers are shown in SEQ ID NO: 1-3.

[0038] (7) The probability of disease (POD) index is calculated by using the ratio of randomly generated decision trees. The decision tree prediction sample is "CCA", and the parameter prediction is set as: proximity=T, norm.votes=T, predict.all=TRUE. The random forest model constructed in LOO mode is used to predict the POD index of each sample in the cohort, and finally the average predicted POD index of each sample is calculated.

[0039] (8) The pROC tool in the R 3.3.0 package was used to calculate the receiver operating characteristic (ROC) curve to evaluate the microbial discrimination model, and the area under the curve (AUC) was used to specify the effect size of the ROC.

[0040] (9) The microbial discrimination model had a discrimination ability of 99.22% between 50 cases of cholangiocarcinoma and 100 healthy subjects. Figure 3 ), the POD index was significantly increased in patients with cholangiocarcinoma, and there was a significant difference between the two groups (p < 0.0001) ( Figure 4 ).

[0041] (10) In the validation set, the POD index was significantly increased in 24 cases of cholangiocarcinoma, and there was a significant difference between the two groups (p < 0.0001). Figure 5 ), the microbial discrimination model had a discrimination ability of 98.08% between 24 cases of cholangiocarcinoma and 50 healthy controls ( Figure 6 ).

[0042] Therefore, the microbial gene differentiation model of the present invention achieved good differentiation ability between cholangiocarcinoma and healthy people, and also verified the feasibility and applicability of the differentiation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 . Study design and application of an oral microbiome model to distinguish cholangiocarcinoma from healthy subjects.

[0044] Figure 2 . The best oral microbiome gene markers identified using a five-fold cross-validation method based on a random forest model.

[0045] Figure 3. Difference in the prevalence index (POD) between the two groups in a cohort of 50 cholangiocarcinoma patients and 100 healthy subjects.

[0046] Figure 4 . The discriminatory power achieved by the microbiome-gene discrimination model in a cohort of 50 cholangiocarcinoma patients and 100 healthy subjects.

[0047] Figure 5. In the validation set, the prevalence index (POD) was significantly increased in 24 cholangiocarcinoma samples compared with 50 healthy controls.

[0048] Figure 6. The ability of the prevalence of disease (POD) index to discriminate between 50 healthy controls and 24 cholangiocarcinoma cases in the validation set. DETAILED DESCRIPTION

[0049] The present invention will be further described below in conjunction with embodiments, but the protection content of the present invention is not limited to these embodiments.

[0050] The methods used in the following examples are all conventional methods unless otherwise specified. The materials or reagents required in the following examples are all obtained from public commercial channels unless otherwise specified.

[0051] The present invention collects saliva samples from the subjects, extracts total microbial DNA, and performs 16SrDNA Miseq sequencing of the oral flora. Based on high-throughput sequencing data, a microbial differentiation model between cholangiocarcinoma and healthy people is established in the training set, and a cholangiocarcinoma prevalence (probability of disease, POD) index is established; the POD index can be used to calculate its differentiation ability and is verified in the validation set.

[0052] The operation steps are as follows:

[0053] (1) According to the design principles of prospective clinical trials, the research design of the present invention is as follows: Figure 1The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University. All enrolled patients signed the informed consent for the study protocol and the informed consent for clinical sample collection.

[0054] (2) Each enrolled cholangiocarcinoma patient and healthy person provided a fresh saliva sample. The researchers divided the sample into 10 portions of 500ul each and immediately stored them in a -80°C refrigerator. The extraction method of total saliva bacterial DNA was carried out according to the instructions of the kit.

[0055] (3) Complete the amplification of total DNA samples of salivary bacteria and the construction of DNA libraries, and complete 16S rDNA sequencing on the Illumina Miseq sequencing platform. All output sequences complete basic preprocessing and basic bioinformatics analysis.

[0056] (4) Randomly select an equal number of sequences from all samples and splice them into corresponding 16S rDNA gene sequence classification units (Operational Taxonomy Units, OTUs) according to the UPARSE pipeline. Collect and organize the OTUs gene sequences of all samples. Based on the microbial gene sequences, use the RDP classifier version 2.6 for annotation.

[0057] (5) Based on the representative sequences generated by high-throughput sequencing data, the OTUs frequency files of microbial gene markers were calculated. These OTUs were used in a correlation study to identify the abundance of OTUs between cholangiocarcinoma patients and healthy people. The Wilcoxon test method was used to statistically analyze the differences in microbial gene markers between cholangiocarcinoma patients and healthy people.

[0058] (6) In the cohort of the microbial discrimination model, including 50 cholangiocarcinoma patients and 100 healthy subjects, the screened OTUs abundance files were used to screen microbial gene markers in a random forest model (R software 3.4.1 and random forest package 4.6–12) using a five-fold cross-validation algorithm (software parameters were default except for setting "importance=TRUE"). Five trials of five-fold cross-validation were used to obtain a cross-validation error curve, in which the minimum cross-validation error point was used as the cut-off value. The minimum cross-validation error value plus the standard deviation of the corresponding value was the cut-off value. A set of OTUs markers with an error rate of less than 3 and less than the cut-off value was screened out, and the set with the minimum number of OTUs was selected as the best set of microbial gene markers. Finally, the best three microbial gene markers for this model were identified ( Figure 2). The gene sequences of the three selected microbial OTUs markers are shown in SEQ ID NO: 1-3.

[0059] (7) The probability of disease (POD) index is calculated by using the ratio of randomly generated decision trees. The decision tree prediction sample is "CCA", and the parameter prediction is set as: proximity=T, norm.votes=T, predict.all=TRUE. The random forest model constructed in LOO mode is used to predict the POD index of each sample, and finally the average predicted POD index of each sample is calculated.

[0060] (8) The pROC tool in the R 3.3.0 package was used to calculate the receiver operating characteristic (ROC) curve to evaluate the microbial discrimination model, and the area under the curve (AUC) was used to specify the effect size of the ROC.

[0061] (9) The microbial discrimination model had a discrimination ability of 99.22% between 50 cases of cholangiocarcinoma and 100 healthy subjects. Figure 4 ), the POD index was significantly increased in patients with cholangiocarcinoma, and there was a significant difference between the two groups (p < 0.0001) ( Figure 3 ).

[0062] (10) In the validation set, the POD index was significantly increased in 24 cases of cholangiocarcinoma, and there was a significant difference between the two groups (p < 0.0001). Figure 5 ), the microbial discrimination model had a discrimination ability of 98.08% between 24 cases of cholangiocarcinoma and 50 healthy controls ( Figure 6 ).

[0063] Therefore, the microbial gene discrimination model of the present invention achieved good discrimination ability between cholangiocarcinoma and healthy people, and also verified the feasibility and applicability of the discrimination model. Sequence Listing <110> The First Affiliated Hospital of Zhengzhou University <120> Oral microbial gene markers for distinguishing cholangiocarcinoma from healthy people and their application <160> 5 <170> SIPOSequenceListing 1.0 <210> 1 <211> 465 <212> DNA <213> Oral microorganism <400> 1 cctacgggtg gctgcagtgg ggaattttgg acaatgggcg aaagcctgat ccagcaatgc 60 cctacgggtg gctgcagtga ggaatattgg tcaatgggcg agagcctgaa ccagccaagt 60 cgcgtgtgtg atgaaggcct tcgggttgta aagcactttt ggcgggaacg aaaaggactg 120 agcgtgcagg atgacggccc tccgggttgt aaactgcttt tagttgggaa taaaaaaagg 120 tgccaatacc acggttcgat gacggtaccc gcagaataag caccggctaa ctacgtgcca 180 gcagccgcgg taatacgtag ggtgcaagcg ttaatcggaa ttactgggcg taaagagtgc 240 gcaggcggtt ttgcaagacc gatgtgaaat ccccgggctt aacctgggaa ctgcattggt 300 gactgcaagg ctagagtgtg tcagagggag gtggaattcc gcatgtagca gtgaaatgcg 360 tagatatgcg gaggaacacc gatggcgaag gcagcctcct gggataacac tgacgctcat 420 gcacgaaagc gtggggagca aacaggatta gataccctag tagtc 465 <210> 2 <211> 459 <212> DNA <213> Oral microorganism <400> 2 cctacgggtg gctgcagtga ggaatattgg tcaatgggcg agagcctgaa ccagccaagt 60 agcgtgcagg atgacggccc tccgggttgt aaactgcttt tagttgggaa taaaaaaagg 120 gacttgtccc ttcttgtatg taccttcaga aaaaggaccg gctaattccg tgccagcagc 180 cgcggtaata cggaaggtcc aggcgttatc cggatttatt gggtttaaag ggagcgtagg 240 cggattgtta agtcagcggt taaagggtgt ggctcaacca tgcattgccg ttgaaactgg 300 cgatcttgag tgcagacagg gatgccggaa ttcgtggtgt agcggtgaaa tgcttagata 360 tcacgaagaa ctccgatcgc gaaggcaggt gtccgggctg caactgacgc tgaggctcga 420 aagtgtgggt atcaaacagg attagatacc ctagtagtc 459 <210> 3 <211> 460 <212> DNA <213> Oral microorganism <400> 3 cctacgggtg gctgcagtgg ggaatattgc acaatgggcg caagcctgat gcagcgacgc 60 cgcgtgaggg atggaggcct tcgggttgta aacctctttc gccagtgaag caggcctgtc 120 ccttgtgggt gggttgacgg tagctggata agaagcgccg gctaactacg tgccagcagc 180 cgcggtaata cgtagggcgc gagcgttgtc cggaattatt gggcgtaaag agctcgtagg 240 cggctggtcg cgtctgtcgt gaaatcctct ggcttaactg ggggcttgcg gtgggtacgg 300 gccggcttga gtgcggtagg ggagactgga actcctggtg tagcggtgga atgcgcagat 360 atcaggaaga acaccggtgg cgaaggcggg tctctgggcc gttactgacg ctgaggagcg aaagcgtggg gagcgaacag gattagatac cctagtagtc <210> 7 <211> 19 <212> DNA <213> Artificial Sequence <400> 7 actcctacgg gaggcagca <210> 8 <211> 20 <212> DNA <213> Artificial Sequence <400> 8 20. ggactachvg ggtwtctaat

Claims

1. An oral microbial gene marker for distinguishing cholangiocarcinoma patients from healthy people, characterized by: The present invention is composed of three microbial genes shown in SEQ ID NO: 1-3, and the microorganisms are obviously enriched in the oral cavity.

2. Use of a detection reagent for detecting the oral microbial gene marker described in claim 1 in the preparation of a detection kit for distinguishing cholangiocarcinoma and healthy people, wherein the detection reagent is suitable for detecting the oral microbial gene described in claim 1.

3. The use according to claim 2, characterized in that: Saliva samples were collected from the subjects, total microbial DNA was extracted, 16S rDNA Miseq sequencing of the microbial DNA was completed, and the presence of the three microbial genes described in claim 1 was detected.

4. The use according to claim 2, characterized in that: Saliva samples were collected from the subjects, total microbial DNA was extracted, and 16S rDNA Miseq sequencing of the oral flora was performed. Based on the high-throughput sequencing data, a microbial differentiation model between cholangiocarcinoma patients and healthy people was established in the cohort, and the POD index of cholangiocarcinoma prevalence was established. The POD index can be used to calculate its differentiation ability for verification.

Images