Remission in Crohn’s disease is accompanied by alterations in the gut microbiota and mucins production

Previous studies have demonstrated that patients with Crohn’s disease (CD) in remission do not exhibit an improvement in gut microbiota composition, which might trigger relapses. The present study investigated the dysbiosis and mucins production in CD patients during remission. We performed an analytical cross-sectional single center study, which recruited 18 CD patients and 18 healthy controls (CG) residing in the same home, meaning that all of the participants experienced the same environmental factors, with similar hygiene status, diet, pollution and other common lifestyle characteristics that may influence the composition of the gut microbiota. When compared to healthy controls, the CD patients exhibited lower microbial α-diversity (p = 0.047), a greater abundance of the Proteobacteria phylum (p = 0.037) and a reduction in the Deltaproteobacteria class (p = 0.0006). There was also a reduction in the Akkermansia (p = 0.002) and Oscillospira (p = 0.024) genera and in the proportion of the yeast Saccharomyces cerevisiae (p = 0.01). Additionally, CD patients in remission presented increased neutral (p = 0.001) and acid mucin (p = 0.002) concentrations. The reductions in the proportions of Oscollospira and Akkermansia genera, sulfate-reducing bacteria and Saccharomyces cerevisiae, observed in the CD group, may account for the increased mucins production observed in these patients.

importance of mucins in the GI tract, the role of these proteins during the CD remission phase has not yet been deeply investigated.
Several studies, over the last 10 years, have shown that dysbiosis occurs in Inflammatory Bowel Diseases (IBD) [8][9][10][11][12] . More specifically, in IBD, the global composition of the gut microbiota contains specific pathogens that may be relevant to the etiology and pathogenesis of the disease. With regards to the gut microbiota of CD patients, it has been reported that there is an overall reduction in microbial diversity 13 , evidenced by alterations in the relative abundance of specific bacterial taxa 6 and fungal communities, when compared to healthy individuals 14 .
Predominant changes described in literature regarding the gut microbiota composition of individuals with IBD and CD, include: alterations in the proportion of Bacteroides and Firmicutes, an increase in the percentage of Gammaproteobacteria 15,16 and Enterobacteriales (i.e. Escherichia coli in CD with ileal disease), and a decrease in Clostridiales (i.e. Faecalibacterium prausnitzii) 17 . Furthermore, it has been observed that the amount of the fungi Candida albicans is increased in patients with CD 18 , whereas Saccharomyces cerevisiae is more abundant in non-inflamed mucosa 18 .
In CD, there are frequent relapses after periods of remission which are not entirely well understood. In fact, it is plausible that clinical remission is not accompanied by a reestablishment in the microbial balance of the GI tract, which might trigger future relapses. It is important to point out that there are no previous studies have evaluated of the gut microbiota and mucins production during CD remission. Therefore, the present study sought to compare and contrast the GI tract microbiomes and mucin expression patterns in CD patients during remission and healthy controls.

Results
Study population characteristics. Between June 2016 and May 2017, 36 subjects were studied, 18 patients with CD and 18 healthy controls. The subjects in the control group (CG) resided in the same house to the corresponding CD patient, sharing the same environmental status, like hygiene habitus, diet, pollution and other factors. The baseline characteristics of groups are show in detail in Table 1. According to Montreal classification, A2, L2 and B1 phenotypes were the commonest. Perianal involvement was present in 44.4%. and endoscopic remission were present in 72.3% of the CD patients. Regarding CG, one individual used probiotic for seven days, between 2-3 months prior to study.
Histological and histochemical analysis. Both mucins concentrations were increased in CD group.

Microbial patterns in crohn's disease group and control group.
To evaluate the gut microbiota diversity between CD and CG groups, PERMANOVA analysis was carried out. The results showed that the distribution and abundances of two groups (Supplementary Table 1S), are different (p < 0.01) In addition, α diversity among all samples were compared, therefore CD showed more species diversity variability profile than CG, based on Shannon index (Fig. 2). In general, the individuals residing in different houses (CG) group have a range Shannon index (approximately 3.2 up to 4) while CD group showed a range Shannon index (approximately 1.8 up to 4.6) (Fig. 2). Patients with CD showed a lower microbial α diversity compared with the CG as reflected by Shannon indexes (p = 0.047). Regarding α diversity, between groups, it was observed that the control group (CG) have a very similar alpha diversity while CD group do not show this homogeneity (Fig. 2B). The analysis of the gut microbiota showed no significant change in the proportion of the two most abundant microbial phyla: Firmicutes and Bacteroidetes between CD and CG groups (Fig. 3). Contrarily, the CD group showed greater abundance of the phylum Proteobacteria (CG 3.5% ± 0.47 vs CD 7.4% ± 1.4; p = 0.037 unpaired test and p = 0.016 paired test) (Fig. 4A) and lower of the phylum Verrucomicrobia (CG 0.40% ± 0.19 vs CD 0.02% ± 0.02; p = 0.014 unpaired test and p = 0.003 paired test) (Fig. 4B).
Deltaproteobacteria class, which contains most of the Sulfate-reducing bacteria, was reduced in the CD group. The relative abundance of Deltaproteobacteria class was high in CG group compared with CD (CG 1.23% ± 0.33 vs CD 0.4% ± 0.21; p = 0.0006 unpaired test and p = 0.007 paired test) (Fig. 4C). In addition, there was reduction of the beneficial genera Akkermansia (CG 0.68% ± 0.27 vs CD 0.04% ± 0.02; p = 0.002 unpaired test and p = 0.005 paired test) (Fig. 4D) and Oscillospira (CG 1.20% ± 0.33 vs CD 0.55% ± 0.15; p = 0.024 unpaired test and p = 0.025 paired test) (Fig. 4E) when compared to CG 19 . Although the analysis of the gut microbiota in this study showed a reduction in the genera Dialister and Bifidobacterium in the CD group (Supplementary Figure 1S), we did not find significant changes in the proportion of the species that have already been correlated with active CD like Bifidobacterium adolescents, Dialister invisus and Faecalibacterium prausnitzii 20 .
Saccharomyces cerevisiae quantification. Saccharomyces cerevisiae may exhibit a protective role in the inflammatory process. Our data showed that the proportion of Saccharomyces cerevisiae was significantly lower in CD group when compared with CG group (CG 0.02% ± 0.007 vs CD 0.005% ± 0.002 p = 0.01 unpaired test and p = 0.008 paired test) (Fig. 5) by quantification using qPCR.
The significant findings of this study are summarized in Fig. 6.

Discussion
The results of the present study demonstrated that the GI tract of CD patients in remission still display dysbiosis, that is characterized by reduced microorganism diversity. Additionally, when compared to the control group, which resided in the same home, there was an observed increase in the amount of Proteobacteria and a decrease in Verrucomicrobia. The CD group also presented increased mucins production. According to Rothschild et al. 21 , individuals that experience the same environmental factors tend to have a similar gut microbiota compositions. Additionally, it is known that microbiota vary according to geography, www.nature.com/scientificreports www.nature.com/scientificreports/ cultural habits, age, and lifestyle factors that include diet, smoking, physical activity as well as others. Thus, in order to avoid potential biases when comparing CD patients to control group all of the study subjects must be exposed to the same environment 8,[22][23][24] . In the present study, our control group consisted of people who were considered healthy, resided in the same home, had a similar hygiene status, consumed identical diets, and were, in general, exposed to equivalent environmental and other common lifestyle factors that could influence the composition of the gut microbiota.
The inflammation caused by CD, frequently results in damage to the intestinal wall 7 . The colonic mucus barrier provides a protective layer against potential antigens, pathogenic bacteria and metabolites produced by microorganisms present in the intestinal lumen. Additionally, the goblet cells produce mucins, which are connected by disulfide bridges and impede bacteria from penetrating the intestinal lumen 6 . It is important to mention that individuals with CD produce higher levels of mucins, that could attenuate the inflammatory response. Interestingly, the results herein showed that, even during remission CD patients also exhibit augmented mucin production, despite most of these patients not displaying any signs of inflammation, as revealed through colonoscopies.
The novelty of our study comes from integrating the gut microbiota composition and mucin production and evaluating how these changes influence the pathology of CD during remission. In this sense, it is tempting to speculate that the modulation of the gut microbiota may stimulate mucin production. In fact, we observed that the Oscillospira and Akkermansia genera were reduced in the CD group, when compared to the controls.   Interestingly, the bacteria belonging to these two genera have been associated with the degradation of mucins and reductions in the proportions of these genera result in increased intestinal permeability 25,26 . Moreover, both neutral and acid mucin concentrations were increased in the CD group during remission.  www.nature.com/scientificreports www.nature.com/scientificreports/ With regards to other types of microorganisms, sulfate-reducing bacteria (SRB) obtain their energy through the reduction of sulfate to hydrogen sulfide (H 2 S). The SRB produced H 2 S can induce cytotoxity in human intestinal epithelial cells, promoting cellular damage and even cell death 25,27 . Most SRB are members of the Desulfovibrio genus in the Deltaproteobacteria class, and were found to be reduced in the CD group (p = 0.0166) of the present study.
The reduction in the proportion of the Oscollospira; Akkermansia genera and SRBs detected in the CD group during remission, may account for the increased mucin production, and may represent a compensatory mechanism for handling the CD-mediated inflammation. The CD group during remission presented a different gut microbiota when compared to both the healthy subjects (HS) and patients with active CD, apparently representing an intermediate microbiota composition.
In addition, a previous study reported that CD patients also have higher proportions of fungi in their intestines, with increased C. albicans and decreased S. cerevisiae content 28 . Herein we showed that CD patients in remission have reduced S. cerevisiae levels when compared with the control group. This result is similar to what has been observed in patients with active CD 29 . This is also relevant to inflammation, since a previous study with mice demonstrated that S. cerevisiae supplementation promoted a significant reduction in TNF-α expression and an increase in IL-10 production 30 . Through this anti-inflammatory effect S. cerevisiae can contribute to the regulation of the inflammatory process in patients with CD 29 , and its reduced presence may make these individuals more susceptible to relapses.
Our study has some limitations. First, it is a cross-sectional study and these results are observational. Secondly, we have not studied patients with active CD. Thirdly, the fungal analysis was limited to just one species (S. cerevisiae).
In conclusion, during remission CD patients present increased amounts of the neutral and acid mucins that are accompanied by a global dysbiosis. In particular, we identified a reduction in the Deltaproteobacteria class, Oscillospira and Akkermansia genera, and S. cerevisiae. While these results are encouraging future studies need to further investigate the relationship between the gut microbiota, and mucin production in CD. Furthermore, studies focusing on patients with active CD, CD in remission and healthy subjects need to be undertaken to further validate the results presented here.

Methods
Study design and population. We did an analytical cross-sectional single center study (IBD clinics of Campinas State University -Unicamp -Brazil), with CD patients and healthy controls. Inclusion criteria were CD patients with diagnosis confirmed by means of clinical, endoscopic and histological criteria and for control group adults residents in the same house, without previous history of chronic disease. Exclusion criteria included individuals that used antibiotics or probiotics during the previous 2 months. Disease activity in CD patients was assessed by the CDAI score 31 and endoscopic findings. CDAI score under 150 were considered inactive disease (clinical remission).
Clinical data, disease classification, medications and comorbidities were collected on the same day of colonoscopy procedure. Fecal samples were collected by the subjects at home using Sarstedt tubes (Sarstedt, Nümbrecht, Germany) filled with a preservative buffer and brought to the IBD clinics within 24 hours after defecation. Stool samples were stored in frozen at -80 °C for microbiota analyses.
Metagenome profile. Total DNA of fecal samples was extracted with the Stool PSP Spin DNA kit (STRATEC Biomedical AG, Germany), an integrated system for collecting, transporting and storing feces samples and subsequent DNA purification.
To profiling microbiota composition, the hyper-variable region (V3-V4) of the bacterial 16S rRNA gene was amplified by following the Illumina 16S Metagenomic Sequencing Library Preparation guide 32 which uses the following sequence: 338F -5′-TCGTCGGCAGCGTCAG ATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG -3 and 785R -5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3′ (2 × 300 bp paired-end an insert size of ~550 bp). www.nature.com/scientificreports www.nature.com/scientificreports/ Bioinformatic and statistical data analysis. The fastq sequences were analysed using DADA2 tool as describe by Callahan et al. 33 , that allows to recover single-nucleotide resolved Amplicon Sequence Variants (ASVs) from amplicon data. The default parameters were used to improve the overall quality of the sequences, www.nature.com/scientificreports www.nature.com/scientificreports/ the reads were filtered and trimmed using the "filterAndTrim" function implemented in DADA2 as described in https://benjjneb.github.io/dada2/tutorial.html. Low quality bases at the end of the reads were removed and the truncLen option was set to 280 and 220 to trim the forward and reverse fastq files respectively. Moreover, the sequences were also trimmed at the 5′ end using the trimLeft option set to 17 and 21 for the forward and reverse reads respectively. The taxonomic assignment was subsequently performed using the naïve Bayesian classifier method implemented in DADA2 using as reference the SILVA database. Finally, the final phylogenetic tree of the ASVs was obtained using the function AlignSeq implemented in DECIPHER 34 R package to create the multiple sequence alignment and the Fast Tree program 35 .  www.nature.com/scientificreports www.nature.com/scientificreports/ Statistical analysis was performed on R (Version 3.4.4) using the following R packages: phyloseq (version 1.24.0) to facilitate the import, storage, analysis, and graphical display of microbiome census data 36 . Data were pre-processed filtering features with less than 10 read counts and present in less than 2 samples 36 . Vegan (version 2.4.2) for PERMANOVA analysis 37 . Shannon diversity indices and bar plot graphical were generated by using the R package ggplot2. The longitudinal microbiome studies was carried out from q2-longitudinal, a software plugin for the QIIME 2 microbiome analysis platform (https://qiime2.org) 38 .
Staining techniques for neutral and acid mucins. The biopsies of large bowel mucosa content of the neutral and acid mucins were determined individually to modify histochemical Periodic Acid Schiff (PAS) 39 and Alcian Blue (AB) techniques. The slides were read under an ordinary optical microscope with a final magnification of 200×. The histological parameters were analyzed qualitatively and quantitatively by a pathologist with experience of diseases, of the digestive tract who was unaware of the origin of the material and the objectives of the study. The neutral mucins stained magenta, while the acid mucins stained blue.
image processing, computer-assisted. The images selected were captured on a video camera that had been coupled to an optical microscope. These images were processes and analyzed using the NIS-Elements (Nikon Corporation. Instruments Company, Japan) software, installed in a computer with good image processing capacity. By means of colored histograms in RGB system (Red, Green, Blue) the software determined the color intensity in number of pixels in each field selected and transformed the final data into percentage expressions by analyzed fields. The final value in the segments with and without intestinal transit was the mean of the values found from evaluating three different fields. Statistical analyses. Sample size. based on pilot study data, the sample size calculation was done having as the main variable the relative contribution of Proteobacteria percentage. Considering the paired group (pilot with 6 subjects in each group, effect size = 0.90), assuming α in 5% and β in 5% (power 95%) 12 subjects were necessary in each group. The Software used for the calculation sample size was G*Power version 3.1.2. (Program written, concept and design by Franz, Universitat Kiel, Germany, freely available windows application software) 41 .
The inclusion of the CD patients versus control groups were expressed as medians and percentiles (interquartile range (IQR), 25-75%) for continuous variables and as frequency for categorical variables. For the qualitative variables, the Fischer exact test and the chi-square test (χ 2 ) were select. The Mann-Whitney U-test (non-parametric distribution) was used to compare continuous variables between categories. The significance level adopted was 5% for all statistical tests (p-value < 0.05). Statistical analyses were used according to SSPS v.20.0 software (IBM Inc., Armonk, NY, USA). ethical statement. During routine visits, subjects who agreed in participating in the study signed up an informed consent form. All methods were performed in accordance with the relevant guidelines and regulations. The study was approved by Institutional Ethics Review Board at Unicamp, in Campinas, Brazil, under reference number 885.749/14.