Intestinal type scoring model for undifferentiated spondyloarthritis and construction method and application thereof

Abstract

The application discloses an intestinal type scoring model for undifferentiated spondyloarthropathy and a construction method and application thereof, and belongs to the technical field of medicines.The 16s rRNA intestinal flora amplicon of 100 uSpA subjects is detected by using high-throughput second-generation sequencing.The intestinal type classification is explored, the intestinal type scoring is constructed, and the clinical characteristics of different intestinal types are researched.The research mainly includes the following four parts:(1) collecting the stool samples and peripheral blood of the uSpA subjects, and detecting the peripheral lymphocytes and cytokines by using flow cytometry (FCM) and CBA method;(2) performing intestinal flora 16s sequencing by using the lllumina high-throughput sequencing technology;(3) analyzing the 16s sequencing data by using the QIIME2 platform;(4) describing the variability of the microbiome data by using the Dirichlet multinomial mixture unsupervised clustering, classifying the intestinal flora of the uSpA, and mining the characteristics of the intestinal flora of the subjects.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of medicine, and particularly relates to an intestinal type scoring model for undifferentiated spondyloarthritis and a construction method and application thereof. BACKGROUND

[0002] Undifferentiated spondyloarthritis (uSpA) refers to a group of diseases that have certain clinical and / or radiological features of spondyloarthritis (SpA), but have not yet reached the diagnostic criteria of any established spondyloarthritis. uSpA may be an early manifestation of certain definite spondyloarthritis, and may be transformed into one of them. The cause of uSpA is unknown, and may be related to genetics, infection, etc. So far, its pathogenesis is not very clear and the immune characteristics are not clear.

[0003] Stratification is an effective method that helps better understand complex biological problems such as human physical and mental health. When this method is applied to intestinal microbes, different community compositions can be defined as "intestinal types". The proportions of major groups such as Bacteroides and Prevotella are stable in the adult intestine. Considering the importance and complexity of the intestinal system, it is necessary to determine the structural patterns of microbial community composition and the assembly mechanisms behind them, which can help us better understand the disease state of uSpA. Classification methods based on community composition structure will strengthen microbe-based disease diagnosis, treatment or prevention. SUMMARY

[0004] In view of the unclear pathogenesis and unclear immune characteristics of undifferentiated spondyloarthritis, the application provides an intestinal type scoring model for undifferentiated spondyloarthritis and a construction method and application thereof.

[0005] The purpose of the application is to classify uSpA test samples into "intestinal types", construct intestinal type scores, and study the clinical characteristics of different intestinal types, thereby enhancing the understanding of uSpA disease and providing new ideas and basis for the diagnosis and treatment of uSpA.

[0006] In order to achieve the above purpose, the application adopts the following technical solutions:

[0007] The application discloses an intestinal type scoring model of undifferentiated spondyloarthropathy, which is characterized in that the intestinal flora of uSpA samples is excavated, the variability of the microbiome data is described by adopting a Dirichlet multinomial mixed unsupervised clustering, the intestinal flora of the uSpA is classified into E1 type and E2 type, a classification marker of intestinal microorganisms based on a genus level is screened out between different intestinal types by taking LDA SCORE log10>2 as a boundary, the principal components PC1 of dominant bacteria of the two intestinal types are extracted respectively by principal component analysis PCA, the principal components PC1 are taken as characteristic scores PC1i and PC1j of the classification marker, and the intestinal type scoring is constructed, and the specific calculation formula is as follows:

[0008] Intestinal type scoring = ∑PC1i - ∑PC1j;

[0009] E1-type is characterized by Turicibacter, GCA_900066575, Lachnospiraceae_FCS020_group, _Ruminococcus__gauvreauii_group, Lachnospiraceae_UCG_001, _Eubacterium__ventriosum_group, Coprococcus, _Eubacterium__xylanophilum_group, Oscillibacter, Bilophila, Lachnospiraceae_UCG_003, CAG_56, Lachnospiraceae_UCG_010, Collinsella, Monoglobus, Parabacteroides, Dorea, _Eubacterium__hallii_group, Desulfovibrio, UCG_003, Lachnospiraceae_ND3007_group, Butyrivibrio, _Eubacterium__siraeum_group, Odoribacter, RF39, Romboutsia, NK4A214_group, Erysipelotrichaceae_UCG_003, Akkermansia, Fusicatenibacter, Barnesiella, _Eubacterium__ruminantium_group, UCG_005, CAG_352, _Eubacterium__eligens_group, Coprococcus, Agathobacter,Christensenellaceae_R_7_group, Lachnospiraceae_NK4A136_group, UCG_010, Roseburia, UCG_002, Dialister, Ruminococcus, Subdoligranulum, _Eubacterium__coprostanoligenes_group, Clostridia_UCG_014, Alistipes, Prevotella, Faecalibacterium;,

[0010] E2 type is characterized by the following bacteria: Bacteroides, Escherichia Shigella, Lachnoclostridium, _Ruminococcus__gnavus_group, Veillonella, Fusobacterium, Enterococcus, Phascolarctobacterium, _Clostridium__innocuum_group, Hafnia_Obesumbacterium, Lachnospira, Flavonifractor, Erysipelatoclostridium, Clostridium_sensu_stricto_1, Colidextribacter, Sutterella, Clostridioides, Hungatella.

[0011] Compared with the prior art, the present application has the following advantages:

[0012] This invention utilizes 16S rRNA gut microbiota sequencing data to analyze the enterotype of uSpA samples. The results show that uSpA samples can be divided into two currently recognized enterotypes: Prevotella-driven and Bacteroides-driven, with differences in species diversity, species composition, and clinical characteristics between the two types. Furthermore, we constructed an enterotype scoring model that can effectively distinguish between the two types. Enterotype is closely related to dietary habits. Because the Prevotella type has a strong ability to ferment plant fiber, vegetarians or people who consume more fiber tend to have a Prevotella enterotype; the Bacteroides type has a strong ability to ferment protein and fat, and people whose diet mainly consists of meat and protein are more likely to have a Bacteroides enterotype. This invention represents basic research on uSpA disease, providing a theoretical basis for a deeper understanding of uSpA and offering a reference for the prevention and treatment of uSpA disease. Attached Figure Description

[0013] Figure 1 This is a line graph of the intestinal type classification of the present invention;

[0014] Figure 2 This is a box plot representing the intestinal type characterization of this invention;

[0015] Figure 3 This invention provides LEfSe, a marker for gut microbiome classification between different gut types.

[0016] Figure 4 This is a dilution curve plot of different alpha diversity indices for the bacterial communities of different enterotype test samples according to the present invention. The horizontal axis represents sequencing depth, and the vertical axis represents the alpha diversity index of the samples. The figure below shows a comparative analysis of alpha diversity between the two groups.

[0017] Figure 5 This is an alpha diversity analysis diagram of bacterial communities in different enterotype test samples of this invention;

[0018] Figure 6 This is the PcoA diagram of beta diversity analysis of gut microbiota in test samples of different enterotypes according to the present invention;

[0019] Figure 7 This is a stacked diagram of the phylum-level species distribution of different enterocoloid groups in this invention;

[0020] Figure 8 This is a stacked diagram of the genus-level species distribution of different enterotypes in this invention;

[0021] Figure 9 This invention presents a box plot of clinical characteristics between different intestinal types.

[0022] Figure 10The ROC curve of the intestinal type score constructed by the present application. DETAILED DESCRIPTION

[0023] The 16s rRNA intestinal flora amplicon of 100 uSpA subjects was detected by high-throughput second-generation sequencing. The classification of intestinal type was explored to construct the intestinal type score, and the clinical characteristics of different intestinal types were studied. The research mainly includes the following four parts:

[0024] (1) Collect the fecal samples and peripheral blood of uSpA subjects, and detect the peripheral lymphocytes and cytokines by flow cytometry (FCM) and CBA method. (2) Use lllumina high-throughput sequencing technology for intestinal flora 16s sequencing. (3) Use QIIME2 platform to analyze 16s sequencing data. (4) Use Dirichlet multinomial mixture unsupervised clustering to describe the variability of microbiome data, classify uSpA intestinal flora, and mine the characteristics of the intestinal flora of the subjects.

[0025] Example 1

[0026] 100 test samples of patients with undifferentiated spondyloarthritis (uSpA) diagnosed in the Rheumatology Department of the Second Hospital of Shanxi Medical University from January 2018 to June 2019 were collected. About 10g of the test sample was taken as the first morning mid-fecal sample, and 250mg of the fecal sample was accurately weighed within 1min and placed in a new 2mL sterile container. The sample was stored in a -80℃ refrigerator for standby, and then genomic DNA extraction and sequencing were performed. All test samples required complete clinical data, active cooperation, on-time stool sample collection, and exclusion of other immune-related diseases or intestinal diseases, recent 6-month history of chronic infection, combined with malignant tumors, etc. that may affect the test results, recent 1-month participation in other drug observation groups, or recent use of immunomodulatory drugs, immunosuppressive drugs, antibiotics, and probiotics, prebiotics, etc. microecological regulators, and test samples in pregnancy or lactation period.

[0027] Peripheral blood of all test samples was collected to detect the absolute count of total T, B, CD4+T, CD8+T, NK cells and Th1, Th2, Th17, Treg cells and other CD4+lymphocyte subsets by flow cytometry (FCM). The levels of IL-2, IL-4, IL-6, IL-10, INF-γ and TNF-α and other cytokines were detected. The markers of each immune cell are as follows: T cells: CD45+CD3+; B cells: CD45+CD3-CD19+; CD4+T cells: CD45+CD3+CD4+; CD8+T cells: CD45+CD3+CD8+; NK cells: CD45+CD3-CD16+CD56+. Th1 cells: CD4+INFγ+; Th2 cells: CD4+IL-4+; Th17 cells: CD4+IL-17+; Treg cells: CD4+CD25+Foxp3+.

[0028] CTAB method was used to extract fecal microbial DNA. We amplified the V3-V4 region of bacterial 16S gene. 16S forward primer 5'-ACTCCTACGGGACCCAGCAG-3' and 16S reverse primer 5'-GGACTACHVGGGTWTCTAAT-3' were used to amplify 16S ribosomal DNA (16S rDNA) gene. KAPA HiFi HotStart Ready Mix (Promega, Madison, WI) was used for PCR amplification reaction according to the manufacturer's method. The PCR cycle conditions were as follows: denaturation at 95°C for 2 min, annealing at 55°C for 30 s, extension step at 72°C for 30 s, 25 cycles of amplification, and extension step at 72°C for 5 min. The quality of the amplified product was checked by 2% agarose gel. The PCR product was purified by Zymoclean Gel DNA Recovery Kit (Zymoclean Research). The gel-purified product was quantified by 4.0 fluorometer (Invitrogen) with 4 nM of DNA as the starting amount for all libraries. 2x300 paired-end sequencing was used according to the description of Novaseq6000. 4.0 fluorometer (Invitrogen) with 4 nM of DNA as the starting amount for all libraries. 2x300 paired-end sequencing was used according to the description of Novaseq6000.

[0029] The raw sequence data was quality controlled by fastqc, and the primer position information was obtained by usearch and vsearch software. The DADA2 software of QIIME2 was used to splice the paired-end (PE) reads obtained by Hiseq / Miseq sequencing into a sequence, filter low-quality and repeated reads, remove chimeras, correct sequencing errors of amplicons, and finally obtain optimized sequences (ASV). Based on the optimized sequences, the silva database was used as the reference database for species annotation to perform species classification annotation.

[0030] The Dirichlet multinomial mixture unsupervised clustering was used to describe the variability of the microbiome data, and the uSpA gut microbiota was classified into E1 and E2 types. Based on the LDA SCORE (log10) > 2, the intestinal microbial classification markers at the genus level between different intestinal types were screened by LEfSe. Through principal component analysis (PCA), the principal components PC1 of the dominant bacteria of the two intestinal types were extracted as the characteristic scores PC1i and PC1j of the classification markers, and the intestinal type score was constructed. The specific calculation formula is as follows:

[0031] Intestinal type score = ∑PC1i - ∑PC1j

[0032] Among them, the characteristic bacteria of E1 type are: Turicibacter, GCA_900066575, Lachnospiraceae_FCS020_group, _Ruminococcus__gauvreauii_group, Lachnospiraceae_UCG_001, _Eubacterium__ventriosum_group, Coprococcus, _Eubacterium__xylanophilum_group, Oscillibacter, Bilophila, Lachnospiraceae_UCG_003, CAG_56, Lachnospiraceae_UCG_010, Collinsella, Monoglobus, Parabacteroides, Dorea, _Eubacterium__hallii_group, Desulfovibrio, UCG_003, Lachnospiraceae_ND3007_group, Butyrivibrio, _Eubacterium__siraeum_group, Odoribacter, RF39, Romboutsia, NK4A214_group, Erysipelotrichaceae_UCG_003, Akkermansia, Fusicatenibacter, Barnesiella, _Eubacterium__ruminantium_group, UCG_005, CAG_352, _Eubacterium__eligens_group, Coprococcus, Agathobacter,Christensenellaceae_R_7_group, Lachnospiraceae_NK4A136_group, UCG_010, Roseburia, UCG_002, Dialister, Ruminococcus, Subdoligranulum, _Eubacterium__coprostanoligenes_group, Clostridia_UCG_014, Alistipes, Prevotella, Faecalibacterium, Bacteroides, Escherichia_Shigella, Lachnoclostridium, _Ruminococcus__gnavus_group, Veillonella, Fusobacterium, Enterococcus, Phascolarctobacterium, _Clostridium__innocuum_group, Hafnia_Obesumbacterium, Lachnospira, Flavonifractor, Erysipelatoclostridium, Clostridium_sensu_stricto_1, Colidextribacter, Sutterella, Clostridioides, Hungatella.

[0033] Based on the above two intestinal types, alpha diversity index analysis was performed using the MicrobiotaProcess package, Bray-curtis distance was used for beta diversity analysis, and Anosim analysis (Analysis of similarities), a non-parametric test method based on permutation test and rank sum test, was used to test whether the difference between groups was significantly greater than the difference within groups, so as to determine whether the typing was meaningful.

[0034] To better understand the complex characteristics of the gut microbiota in the test samples, we performed gut type characterization analysis on the uSpA test samples. For example... Figure 1 As shown, a clear inflection point appears on the curve when K=2, indicating that the gut microbiota can be divided into two types (enterotype 1 and enterotype 2). Testing revealed that enterotype 1 is driven by *Prevotella*, while enterotype 2 is driven by *Bacteroides*, and there is a difference between enterotype 1 and enterotype 2 (P<0.01). Figure 2 As shown. Our LEfSe analysis revealed that *Prevotella* was the dominant spore type 1 bacteria, while *Bacteroides* was the dominant spore type 2 bacteria, and differences existed between the two types, such as... Figure 3 As shown. All the above analyses are based on genus-level classification, because the genus level can better reflect niche changes.

[0035] To better characterize the two enterotypes, the following analysis was conducted: The sequencing depth of the randomly sampled samples was gradually increased. It was observed that as the sequencing depth increased, the Rarefaction Curve gradually increased, and the curve slope gradually smoothed out, indicating sufficient sequencing volume and that the sample alpha diversity index reached stability. Figure 4 Alpha diversity analysis revealed that the species Observe index (P<0.001), Chao1 index (P<0.001), ACE index (P<0.001), Shannon index (P<0.001), and Simpson index (P<0.001) of enterotype 1 gut microbiota were all higher than those of enterotype 2. Figure 5 Through Beta diversity analysis, it was found that, for example... Figure 6 As shown, enterotype 1 and enterotype 2 flora were significantly separated, and the Bray-curtis distance from Anosim to their Beta diversity index was 0.152, indicating significant differences in Beta diversity among the different enterotypes (P = 0.001).

[0036] Species composition analysis revealed that, at the phylum level, Firmicutes, Bacteroides, and Proteobacteria were the two dominant phyla. Figure 7 At the genus level, *Bacteroides*, *Faecalibacterium*, and *Escherichia-Shigella* are the two dominant types of bacteria. Figure 8 Based on clinical characteristics, it was found that the expression levels of IL-6 and TNF-alpha were higher in enteric type 1 than in enteric type 2, and the difference was significant (P<0.05).

[0037] To identify potential biological signatures of each uSpA test sample enterotype, 68 genus-level gut microbial taxonomic markers were determined using LEfSe, and an enterotype score was constructed. Our results showed that the enterotype score could effectively distinguish the two enterotypes with an AUC of 81.7% (95% CI: 73.2%-90.2%) ( Figure 10 ).

[0038] The content not described in detail in the specification of the present application belongs to the prior art known to the person skilled in the art. Although the above describes the specific embodiments of the present application in a demonstrative manner, so as to facilitate the understanding of the present application by the person skilled in the art, it should be clear that the present application is not limited to the scope of the specific embodiments, and for the person skilled in the art, as long as various changes are within the spirit and scope of the present application defined and determined by the appended claims, all the inventions utilizing the concept of the present application are within the protection.

Claims

1. An intestinal type scoring model for undifferentiated spondyloarthritis, characterized in that: By mining the characteristics of the gut microbiota in uSpA samples, Dirichlet multinomial mixed unsupervised clustering was used to describe the variability of the microbiome data, and the uSpA gut microbiota was classified into E1 and E2 types. Using LDA SCORE log10 > 2 as the boundary, LEfSe was used to screen for genus-level gut microbial classification markers between different enterotypes. Principal component analysis (PCA) was used to extract the principal components PC1 of the dominant bacteria in each of the two enterotypes as feature scores PC1i and PC1j for classification markers, constructing an enterotype score. The specific calculation formula is as follows: ； The characteristic bacteria of type E1 are: *Turicibacter*, *GCA_900066575*, *Lachnospiraceae_FCS020_group*, *Ruminococcus_gauvreauii_group*, *Lachnospiraceae_UCG_001*, *Eubacterium_ventriosum_group*, *Coprococcus*, *Eubacterium_xylanophilum_group*, *Oscillibacter*, *Bilophila*, *Lachnospiraceae_UCG_003*, *CAG_56*, and *UCG_*. 010 bacteria, Lachnospiraceae_UCG_010, Collinsella, Monoglobus, Parasteroids, Dorea, Eubacterium_hallii_group, Desulfovibrio, UCG\ 003 bacteria, UCG_003, Lachnospiraceae_ND3007_group, Butyrivibrio, Eubacterium_siraeum_group, Odoribacter, RF39 bacteria, Romboutsia, NK4A214_group, Erysipelothrixaceae_UCG\ 003 bacteria (Erysipelotrichaceae_UCG_003), Akkermansia, Fusicatenibacter, Barnesiella, Eubacterium ruminantium_group, UCG\ 005 bacteria (UCG_005), CAG\ 352 bacteria (CAG_352), Eubacterium eligens_group, Coprococcus, Agathobacter, Asteraceae\ R\ 7 bacteria (Christensenellaceae_R_7_group), Trichophyceae\NK4A136\bacteria Lachnospiraceae_NK4A136_group, UCG\ 010UCG_010, Roseburia, UCG\ 002bacteria UCG_002, Dialister bacteria, Ruminococcus, Subdoligranulum, Eubacterium coprostanoligenes_group, Clostridia UCG\ 014UCStridia_UCG_014, Alistipes, Prevotella, Faecalibacterium; The characteristic bacteria of type E2 are: Bacteroides, Escherichia coli, Shigella, Lachnoclostridium, Ruminococcus gnavus group, Veillonella, Fusobacterium, Enterococcus, Phascolarctobacterium, Clostridium innocuum group, Hafnia obesumbacterium, Lachnospira, Flavonoidfractor, Erysipelatoclostridium, Clostridium sensustricto, Coridextribacter, Sutterella, Clostridioides, and Hungatella.

Images