Unique Microbial Patterns in the Mouth, Throat, and Intestines of Crohn's Disease Patients

All the study participants were enrolled from the Shanghai Tenth People’s Hospital from July 2021 to August 2021, including patients in the active stage, remission stage, and healthy controls. The diagnosis of CD was based on a comprehensive evaluation of clinical symptoms, laboratory examinations, endoscopic features, radiological findings, and histological features by two experienced physicians (Ding et al., 2020). Patients with CD in the active stage all underwent ileostomy, and the patients in the remission stage came to the hospital for the reversal of ileostoma. The exclusion criteria included severe oral and respiratory diseases diagnosed within 1 month, such as dental caries, periodontitis, oral malignant tumor, oral CD, peri-oral CD, pneumonia, and lung cancer. Meanwhile, a history of probiotics or prebiotic consumption and antibiotic use within 1 month and other acute or chronic gastrointestinal disorders were also excluded. People who came to the hospital for physical examination were recruited as healthy controls. Among them, people with a history of chronic inflammatory conditions including IBD and current gastrointestinal or respiratory symptoms were excluded. The Ethics Committee of Shanghai Tenth People’s Hospital approved this study protocol (Ethical Approval Number: SHSY-IEC-4.1/21-145/01). Informed consent was obtained from all patients. All procedures were performed following the Declaration of Helsinki and its later amendments. This study has been registered in ClinicalTrial.gov (NCT04965584↗).

The nutrition risk screening score (2002) (Kondrup et al., 2003) and Crohn’s disease activity index (CDAI) (Yoshida, 1999) were recorded for each recruited patient. A CDAI score over 150 was regarded as the active stage, whereas a CDAI score of 150 or below was defined as the remission stage.

Saliva samples from the tongue base of the recruited patients and healthy controls were collected with a sterile cotton swab and immediately placed into a 2-ml sterile container. Before sputum sample collection, the participants were instructed to gargle three times and inhale 5 ml hypertonic sodium chloride solution by atomization for 10 min, then coughed hard to obtain sputum samples. Fecal samples from the ileal stoma were collected with a sterile cotton swab and placed into a 2-ml sterile container. In addition, 2 g of cefmetazole was routinely used intraoperatively, and there was no history of antibiotic use before surgery for those patients (Figure 1A). All samples were transported to the laboratory with an ice pack within 2 h, frozen immediately, and stored at −80°C before analyses.

Microbial DNA was extracted from all samples using the E.Z.N.A.^® soil DNA Kit (Omega Bio-tek, Norcross, GA, USA) according to the manufacturer’s instructions. The final DNA concentration and purification were determined by a NanoDrop 2000 UV-vis spectrophotometer (Thermo Scientific, Wilmington, USA), and the DNA quality was checked by 1% agarose gel electrophoresis. The V3–V4 hypervariable regions of the bacteria 16S ribosomal RNA (rRNA) gene were amplified with primers 338F (5′- ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′) with the thermocycler PCR system (GeneAmp 9700, ABI, USA). The PCR reactions were conducted using the following procedure: 3 min of denaturation at 95°C, 27 cycles of 30 s each at 95°C, 30 s for annealing at 55°C, and 45 s for elongation at 72°C, and a final extension at 72°C for 10 min. PCR reactions were performed in triplicate, with 20 μl mixture containing 4 μl of 5× FastPfu buffer, 2 μl of 2.5 mM deoxyribonucleotide triphosphates (dNTPs), 0.8 μl of each primer (5 μM), 0.4 μl of FastPfu polymerase, and 10 ng of template DNA. The resulting PCR products were extracted from a 2% agarose gel, further purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) and quantified using QuantiFluor™-ST (Promega, USA) according to the manufacturer’s protocol.

Purified amplicons were pooled in equimolar ratios and paired-end sequenced (2 × 300) on an Illumina MiSeq platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

Raw fastq files were demultiplexed, quality filtered by Trimmomatic, and merged by FLASH with the following criteria: (i) the reads were truncated at any site receiving an average quality score <20 over a 50-bp sliding window; (ii) primers were exactly matched, allowing two nucleotide mismatching, and reads containing ambiguous bases were removed; and (iii) sequences whose overlap was longer than 10 bp were merged according to their overlap sequence.

Operational taxonomic units (OTUs) were clustered with 97% similarity cutoff using UPARSE (version 7.1 http://drive5.com/uparse/↗), and chimeric sequences were identified and removed using UCHIME. The taxonomy of each 16S rRNA gene sequence was analyzed by the RDP Classifier algorithm (http://rdp.cme.msu.edu/↗) against the Silva (SSU123) 16S rRNA database using a confidence threshold of 70%.

Alpha diversity was analyzed using the Mothur software (version 1.30.2 https://www.mothur.org/wiki/Download_mothur↗). To evaluate alpha diversity, the Chao index was calculated, representing the abundance of microbiota, while the Shannon index represented microbiota diversity. A Venn diagram was drawn by Venn Diagram of software R (version 3.3.1) to show the similarity and overlap of microbiota composition of each group. In addition, the Community Bar diagram was drawn by software R (version 3.3.1) to show the abundance of microbiota in multi-ecological sites. The principal coordinates analysis (PCoA) is a multivariate statistical analysis method that reduces the dimensionality of multidimensional data to extract the most important elements. The differences between groups are presented in a two-dimensional coordinate graph (Liu et al., 2020). Furthermore, the Wilcoxon rank-sum test (R package “stats” and python package “scipy”) was used to analyze the inter-group differences of microbiota. A correlation heatmap was drawn with the “pheatmap” package of software R (version 3.3.1), visually reflecting the association between clinical factors and dominant microbiotas in the corresponding ecological sites. In addition, the PICRUSt software (version 1.1.0 http://picrust.github.io/picrust/↗) was used to predict the function of 16S in the amplicon-sequencing results. The Benjamini–Hochberg method was used to correct multiple comparisons, and an false discovery rate (FDR) value <0.05 was considered statistical significant (Gao et al., 2021). p < 0.05 at two sides was considered significant in all statistical analyses Figure 1B.

The 16S rRNA gene sequence data sets have been deposited in the NCBI Sequence Read Archive (SRA) under the study accession number PRJNA773617.

A total of 43 patients (23 patients in the active stage and 20 patients in the remission stage) and 18 healthy controls were recruited in this study (Table 1). Patients with CD were at higher nutritional risk (p < 0.001) than their healthy counterparts. Most patients with CD had a regular diet and had no history of smoking, alcohol, or hematochezia, similar to the healthy controls. Moreover, the CD lesions were mainly located in the ileocolon. The levels of hematological indexes, including C-reactive protein (CRP), white blood cell (WBC), hemoglobin, albumin, prealbumin, and erythrocyte sedimentation rate (ESR), were significantly different between the active and remission stages (all p < 0.05). However, the levels of inflammatory biomarkers, including procalcitonin (PCT), IL-1, IL-6, IL-8, and IL-10, were similar between the active and remission stages of patients with CD (Table 1).

In this study, the Chao and Shannon indexes were used to analyze the alpha diversity of microbiota. The Chao index was used to reflect the abundance of microbiota in the samples, and the Shannon index was used to indicate microbiota diversity. Patients in the remission stage had a significantly lower Chao index of oral microbiota than healthy controls on both phylum and genus levels (p = 0.006 and p = 0.002) (Figures 2A, B). Similar results were observed in sputum samples (Figures 2C, D). In addition, the Shannon index of the oral microbiota in healthy controls was significantly higher than that in both groups of patients with CD (Figures 2E, F). However, there was no significant difference in sputum microbiota in patients with CD and healthy controls (Figures 2G, H). As for the ileal microbiota, the Chao and Shannon indexes of the active stage were higher than those of the remission stage, but the differences were not statistically significant (Supplementary Figures S1A–D).

Principle coordinate analysis (PCoA) using the unweighted UniFrac distances also demonstrated that the microbiota in the oral cavity (p = 0.013 and p = 0.001) and sputum (p = 0.043 and p = 0.001) microbiota differed significantly among different stages of CD and healthy controls on phylum and genus levels, respectively (Figures 2I–L). In terms of ileal microbiota, a notable difference was observed between active and remission stages on the genus level (p = 0.001) but not on the phylum level (Figures 2M, N).

The total number of OTUs across all samples was 65 on the phylum level, among which 29 were identified in the active stage, 19 in the remission stage, and 17 in the healthy controls. Fifteen of these OTUs were common to all three groups (Figure 3A). The relative abundances of Firmicutes and Actinobacteria were the highest in the active stage and were lower in the remission stage of CD and lowest in the healthy controls. On the contrary, Bacteroidetes and Fusobacteria were predominant in the remission stage of CD and healthy controls (Figure 3C).

We then aimed to investigate the stage-specific oral microbial profiles. Regarding the oral microbiota, Fusobacteria was more abundant in the remission stage of CD (p = 0.026), while Chloroflexi was more abundant in the active stage on the phylum level (p = 0.032) (Figure 4A). The relative abundances of Actinobacteria and Synergistota were higher in the active stage of CD (p = 0.047 and p = 0.036), while Fusobacteria and Spirochaetota were more abundant in the healthy controls (p = 0.006 and p < 0.001) (Figure 4B). In addition, the relative abundances of Spirochaetota and Synergistota were higher in the healthy controls than in the remission stage of CD (p = 0.036 and p < 0.001, respectively) (Figure 4C).

On the genus level, 999 OTUs were identified in total, 468 were identified in the active stage, 255 in the remission stage, and 276 in the healthy controls. A total of 178 OTUs were shared across all three groups. The active stage group and the remission stage group had 33 overlapping OTUs, while the active stage group and the control group shared 38 OTUs. In contrast, the remission stage group and the control group had nine OTUs in common (Figure 3B). The relative abundances of Veillonella, Actinomyces, Megasphaera, and Atopobium decreased gradually from the active stage and remission stage of CD to healthy controls. However, the relative abundances of Fusobacterium, Haemophilus, and Porphyromonas increased gradually along the same sequence. Furthermore, Campylobacter was abundant in the oral cavity of patients with CD but less predominant in the healthy controls (Figure 3D). The only oral microbiota with a significant difference between the active and remission stages was Leptotrichia (p = 0.018), which was more abundant in the remission stage (Figure 4D). The relative abundances of Fusobacterium, Haemophilus, Porphyromonas, and Neisseria demonstrated a higher prominence in healthy controls than in the active stage of CD (p < 0.001, p < 0.001, p < 0.001, and p = 0.010) (Figure 4E). Comparing the remission stage to healthy controls, Porphyromonas, Haemophilus, Neisseria, Fusobacterium, and Alloprevotella were more abundant in the healthy controls, whereas Megasphaera was more abundant in the remission stage. These differences in microbiotas were statistically significant (all p < 0.05) (Figure 4F).

The total number of OTUs across all samples was 78 on the phylum level; 34 were identified in the active stage, 22 in the remission stage, and 22 in the healthy controls, respectively. Seventeen OTUs were common to all three groups (Figure 5A). The relative abundances of Firmicutes and Actinobacteria were highest in the active stage of CD and lowest in healthy controls, while the contrary was observed for Fusobacteria (Figure 5C).

Regarding the sputum microbiota, Chloroflexi was the only microbiota that was significantly enriched in the active stage (p = 0.006) (Figure 6A). The relative abundances of Fusobacteria, Spirochaetota, and Synergistota were significantly higher in the healthy controls, especially compared to the active stage of CD (p = 0.022, p = 0.034, and p = 0.017, respectively) (Figure 6B). Furthermore, Actinobacteria was more abundant in the remission stage of CD (p = 0.022), whereas Patescibacteria and Synergistota were more abundant in the healthy controls (p = 0.031 and p = 0.040) (Figure 6C).

On the genus level, the total number of OTUs across all samples was 1,258. The three groups shared 214 OTUs, active stage and remission stage groups shared 52 OTUs, the active stage and control groups shared 63 OTUs, and the remission stage and control groups shared 4 OTUs (Figure 5B). The relative abundance of Streptococcus declined gradually from the active stage, remission stage of CD to healthy controls. On the contrary, the relative abundances of Prevotella, Fusobacterium, Porphyromonas, and Haemophilus were lowest in the active stage of CD and highest in the healthy controls (Figure 5D). The only sputum microbiota with a significant difference between the active and remission stages was Neisseria (p = 0.014), as it was more abundant in the remission stage (Figure 6D). In contrast, the relative abundances of Fusobacterium, Haemophilus, and Porphyromonas were much higher in the healthy controls (p < 0.001, p = 0.049, and p = 0.001) (Figure 6E). Furthermore, the relative abundances of Fusobacterium and Porphyromonas were significantly higher in the healthy controls than in the remission stage group (p = 0.009 and p = 0.004) (Figure 6F).

The total number of OTUs across all samples was 56 on the phylum level, 30 were identified in the active stage, and 26 in the remission stage. Twenty-two OTUs were found in common in the active stage and remission stage groups (Supplementary Figure S1E). The amount of OTUs on the genus level was 1,105, and the two groups shared 353 OTUs. 321 were exclusively identified in the active stage and 78 in the remission stage (Supplementary Figure S1F). The relative abundances of Bacteroidetes and Desulfobacterota were much higher in the active stage on the phylum level (all p < 0.001) (Figure 7C). However, the relative abundances of Firmicutes and Patescibacteria were higher in the active stage than in the remission stage of CD, but the differences between these microbiotas were not statistically significant (Figures 7A, C). On the genus level, the relative abundance of Bacteroides was much higher in the active stage of CD (p < 0.001). On the contrary, the relative abundances of Streptococcus, Veillonella, and Lactococcus were higher in the remission stage of CD (p = 0.033, p < 0.001, and p = 0.004) (Figures 7B, D).

To identify the correlations between dominant microbiota and several clinical factors, the Spearman’s rank correlation coefficient was used. Regarding the oral microbiota, ESR levels (R = 0.308, p = 0.045) were positively associated with Firmicutes, whereas IL-1 (R = −0.324, p = 0.034) and hemoglobin levels (R = −0.317, p = 0.038) in the blood of patients with CD showed negative correlations (Figure 3E). Similarly, the relative abundance of Firmicutes in the sputum samples was also positively correlated with ESR levels (R = 0.343, p = 0.030), while it was negatively correlated with hemoglobin levels (R = −0.375, p = 0.017) (Figure 5E). However, no significant correlations between the relative abundance of Firmicutes and clinical factors were observed in the ileal samples (Figure 7E). On the genus level, PCT levels in the blood of patients with CD were found to be negatively correlated with the relative abundance of Prevotella in the oral and sputum samples (R = −0.407, p = 0.007 and R = −0.321, p = 0.043, respectively) (Figures 3F, 5F). Moreover, IL-10 (R = −0.360, p = 0.018), hemoglobin (R = −0.377, p = 0.013), and albumin (R = −0.369, p = 0.015) concentrations were also negatively correlated with the relative abundance of Prevotella in the oral samples (Figure 3F). The relative abundance of Streptococcus in the sputum samples was negatively correlated with white blood cell count (R = −0.334, p = 0.035) and hemoglobin concentration (R = −0.313, p = 0.049), whereas no significant correlations in the oral samples were observed (Figures 3F, 5F). Regarding the ileal microbiota, hemoglobin (R = 0.414, p = 0.010) and albumin (R = 0.367, p = 0.023) concentrations showed positive correlations with the relative abundance of Escherichia–Shigella, whereas CDAI score (R = −0.353, p = 0.030) and ESR levels (R = −0.333, p = 0.041) in the blood were negatively associated with the microbiota. In addition, CDAI score (R = −0.390, p = 0.016) and IL-6 concentration (R = −0.335, p = 0.040) were negatively associated with the relative abundance of Streptococcus (Figure 7F).

To characterize the microbial functions alterations in patients with CD, we analyzed clusters of orthologous group (COG) functional classifications for microbiota and characterization of different metagenomes of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways levels. Oral microbial gene processes, such as translation, ribosomal structure and biogenesis, general function prediction only, amino acid transport and metabolism, replication, recombination and repair, and cell wall/membrane/envelope biogenesis pathways were highly abundant in all samples (Supplementary Figure S2A). In addition, the microbiota of sputum samples showed similar functional characteristics (Supplementary Figure S2B). However, the most active COG functional categories in the ileum were amino acid transport and metabolism, followed by general function prediction only, carbohydrate transport, and metabolism and transcription (Supplementary Figure S2C). Regarding the relative abundance of different metagenomes of KEGG pathways levels, the predicted metabolic pathways in the oral cavity, sputum, and ileum were similar, predominantly involving metabolism, environmental and genetic information processing (Figures 8A–C).

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/↗, PRJNA773617.