Altered intestinal microbiota associated with colorectal cancer

Abstract

The gut microbiota plays an important role in the development and progression of colorectal cancer (CRC). To learn more about the dysbiosis of carcinogenesis, we assessed alterations in gut microbiota in patients with CRC. A total of 23 subjects were enrolled in this study: 9 had CRC (CRC group) and 14 had normal colons (normal group). The microbiome of the mucosal--luminal interface of each subject was sampled and analyzed using 16S rRNA gene amplicon sequencing. We also used Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) to predict microbial functional profiles. The microbial composition of the mucosal lumen differed between the groups, and the presence of specific bacteria may serve as a potential biomarker for colorectal carcinogenesis. We identified a significant reduction in Eubacterium, which is a butyrate-producing genera of bacteria, and a significant increase in Devosia in the gut microbiota of CRC patients. Different levels of gut microflora in healthy and CRC samples were identified. The observed abundance of bacterial species belonging to Eubacterium and Devosia may serve as a promising biomarker for the early detection of CRC.

Keywords

colorectal cancer (CRC) / gut microbiota / intestinal / Eubacterium / Devosia

Introduction

Colorectal cancer (CRC) is one of the most common malignant tumors in the Western world and, second to lung cancer, causes the greatest number of deaths from cancer [¹,²]. The incidence of CRC in China has recently increased.

Epidemiological studies have shown that the occurrence and development of sporadic CRC is a multifactorial and multi-step process that is jointly affected by environmental and genetic factors [³]. However, certain populations are at higher risk of developing the disease than others, including those with a family history of CRC and colon polyps, a personal history of CRC or polyps, patients with chronic inflammatory bowel disease, and those over 50 years of age. Accumulating evidence shows links between microbial populations and different stages of tumor development. For example, Helicobacter pylori is correlated with the development of gastric cancer [⁴].

Alterations in the gut microbiota, whether caused by lifestyle, diet, environmental factors, or infection, can change the symbiotic relationship between a host organism and its environment and have increasingly been associated with the occurrence of CRC [⁵,⁶]. No single bacterial species has yet been determined to be a risk factor for CRC, but pathogenic bacterial species are directly responsible for approximately 15% of all CRC cases [⁷]. Fusobacterium is a common bacterium significantly enriched in the gut microbiota of CRC patients compared with that in normal controls [⁸–¹⁰]. The Clostridium leptum and Clostridium coccoides subgroups are specific to CRC in fecal microbiota [¹¹]. The pathogenesis of CRC is not well understood, and whether specific intestinal bacteria are associated with the dysbiosis of CRC remains unclear. Microbiota from mucosal samples represent the underlying dysbiosis and seem to be more appropriate for detecting shifts in microbial composition than fecal samples [¹¹]. However, previous studies mainly focus on differences in fecal microbiota between CRC and healthy controls. Therefore, whether the intestinal microbiome plays a major role in the early stages of CRC remains unclear. Moreover, most of the relevant studies are conducted in Western populations, the lifestyles and dietary habits of which are quite different from those of Chinese populations.

In this study, we determined the microbial composition of samples from normal and CRC-diagnosed subjects. Most published studies use fecal samples to detect gut microbiota because these samples are convenient to collect; however, in the present work, we used colonic mucosal-luminal interface samples and analyzed them using 16S rRNA sequencing. The role of butyrate-producing bacteria, which present distinct interactions with the host at mucosal surfaces, in the dysbiosis of CRC is a current area of research. Some butyrate-producing bacteria, such as Eubacterium and Faecalibacterium, are well suited to colonize intestinal mucosal surfaces, thereby indicating that mucosal-luminal interface samples, rather than stool samples, may be more suitable for detecting these important bacteria. Here, we compared the bacterial community structures of CRC with those of normal colons and identified several potential bacterial genera and species associated with the dysbiosis of CRC.

Materials and methods

Subject enrolment

A total of 23 subjects, 9 with CRC and 14 with healthy colons, were selected randomly from Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, China. Written informed consent was obtained from all participants, and the study protocol was approved by the Research Ethics Board of the Hospital. Endoscopy is the gold standard for the diagnosis of colorectal diseases. All eligible adult patients who were scheduled to undergo colonoscopies and were not diagnosed with other diseases except CRC were included in the study. Patients with intact colons served as normal controls. The exclusion criteria were related to known conditions affecting the intestinal microbiota composition including: (1) irritable bowel syndrome, (2) use of any antibiotics or probiotics within the past 30 days before colonoscopy, or (3) infectious gastroenteritis within the past 60 days.

Sample collection

Mucosal-luminal interface samples were collected from the ascending colon using an operation method similar to that previously published [¹²]. Preparations for the colonoscopy were completed within a day based on standard protocols. During colonoscopy, when the cecum and proximal ascending colon were reached, any loose fluid and debris that were present were aspirated. Sterile water was flushed onto the mucus to remove the mucus layers from the mucosal epithelial cells, and the mixture of water, mucus, and intestinal cells was aspirated into sterile containers through the colonoscope. These mucosal-luminal interface samples were instantly frozen in liquid nitrogen and then stored at -80 °C for analysis.

DNA extraction and 16S rRNA sequencing

DNA extraction from the mucosal-luminal interface samples was performed using the QIAamp DNA Mini Kit (Qiagen, Germany) according to the manufacturer’s instructions. DNA integrity was assessed using 1% agarose gel electrophoresis, and purity and concentration were assessed using a NanoDrop2000 UV spectrophotometer (Thermo Scientific, USA). Barcoded amplicons were generated covering variable region 6 (V6) of the 16S rRNA gene using universal primers (1048F, 5′-GTGSTGCAYGGYYGTCGTCA-3′; 1194R, 5′-ACGTCRTCCMCNCCTTCCTC-3′) and incorporating the Illumina paired-end sequencing adapters and barcode sequences. PCR was conducted using a GeneAmp® PCR System 9700 thermal cycler (Applied Biosystems) with the following parameters: 3 min of initial denaturation at 94 °C, followed by 25 cycles of 94 °C for 10 s, 55 °C for 15 s, and 72 °C for 30 s, and a final extension at 72 °C for 7 min. Each PCR reaction mixture (25 µL) contained 10 ng of genomic DNA, 2.5 µL of 10×Ex Taq Buffer (Takara), 0.5 µL of each primer, 1 µL of dNTP (2.5 mmol/L each), and 0.1 µL of Ex Taq DNA polymerase (Takara). The amplicons of each sample were separated on a 1% agarose gel and purified using an Agencourt AMPure XP kit (Beckman Coulter, USA). The concentration of purified DNA for each reaction was measured using the Qubit dsDNA HS Assay Kit (Invitrogen, Carlsbad, CA, USA). Finally, amplicon libraries from all samples were pooled at equal molar concentrations and then sequenced using a 500-cycle MiSeq reagent kit via the paired-end (2 × 251 bp) method on the Illumina MiSeq platform.

Bioinformatics analysis of sequencing data

Raw sequences were assigned to different samples according to their barcodes. After removing adapters and reads with an average quality value lower than 20 or any sequence where the longest homopolymer was greater than 8 nt, only reads longer than 350 bp were considered. Unique clean reads were aligned in accordance with the SILVA database [¹³] using k-mer-based methods. The resulting sequences were processed and analyzed using Mothur software (v.1.39.5) [¹⁴]. Potential chimeric sequences were screened using the chimera.uchime command and deleted by the remove.seqs command of Mothur. The sequences were then classified using the Bayesian classifier with the classify.seqs command, and undesirables were removed by the remove.lineage command.

We combined the nearest neighbor algorithm in Mothur with a 0.03 distance unit cut-off to cluster the reads. For high-quality sequences, operational taxonomic units (OTUs) were clustered at 97% similarity. The bacterial community composition of each sample was counted at different taxonomic levels by comparison with the Greengenes 13.5 database. Thereafter, we analyzed the richness estimator (Chao), diversity index (Shannon), and species composition at various taxonomic levels (i.e., phylum, class, order, family, genus, and species). In the present data, sequences>97% identical to each other were considered to correspond to the same OTUs, representing a group of reads presumably belonging to the same species.

We also performed unweighted UniFrac principal coordinate analysis (PCoA) based on the matrix of distance. By using a nonparametric Kruskal–Wallis rank sum test, linear discriminant analysis effect size (LEfSe) [¹⁵] analysis allows for a quick comparison between multiple groups; species with significant differences in abundance were identified as biomarkers.

We performed Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) [¹⁶] analysis to predict the functional potential of bacteria communities based on the 16S rRNA data. To do so, pick_closed_reference_otus.py script was used to cluster the reads into a collection of OTUs sharing 97% sequence identity. The OTUs were normalized based on 16S rRNA gene copy numbers by employing the normalize_by_copy_number.py script. The normalized OTU table was used as input data to predict microbial community metagenomes with the predict_metagenomes.py script, and metagenome prediction was further categorized into Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways at levels 2 and 3. The edgeR package in R was applied to identify KEGG pathways that were significantly different between groups.

Statistical analyses

Data were analyzed using the Kruskal–Wallis test or Mann–Whitney test (two-sided) for continuous variables and Fisher’s exact test for categorical variables using GraphPad Prism 6 or R (version 3.4.0). The threshold of statistical significance was set at P≤0.05.

Data access

The 16S rRNA gene sequencing data generated in this study were submitted to the NCBI Sequence Read Archive under accession number SRP139052.

Results

Eligible subjects and samples

A total of 23 subjects were included in this study: 14 healthy individuals in the normal group (NG) and 9 individuals in the CRC group (CG). The mean age of the subjects was 62.6±8.9 years in CG and 44.1±15 years in NG. Seven males (50%) were in NG, while six males (67%) were in CG (Table 1). CG had more males and older participants than NG. Details of the clinical information of the patients are shown in Supplementary Table 1.

Characteristics of 16S rRNA gene sequencing

After removing sequences shorter than 350 bp and sequences with ambiguous bases or homopolymers longer than 8 nt, a total of 540 119 high-quality sequences were obtained from 23 samples. Samples showed an average of 23 483 sequences, and no significant difference between NG and CG was found (Fig. 1A, P = 0.27). The Chao community richness estimator is shown in Fig. 1B. Although no significant difference in community richness between the two groups was found, CG tended to have lower richness than NG (P = 0.08). The Shannon index (Fig. 1C) suggested that CG has lower bacterial diversity than NG. To compare the types and amounts of microflora between the two groups, we conducted unweighted UniFrac PCoA based on the OTUs of each sample. The overall microbial communities of the CRC and normal groups were similar according to PC1 and PC2 (CG= 8.2% of the variance explained; NG= 6.52% of the variance explained; Supplementary Fig. 1); however, CG exhibited lower diversity than NG.

Comparisons of gut microbiota at different levels

We analyzed the bacterial communities in the mucosal-luminal interface samples taken from NG and CG participants. The overall microbial structures at the phylum, class, order, and family levels were assessed by taxonomic assignment for each group (Fig. 2). The dominant phyla in both groups were Firmicutes, Proteobacteria, Bacteroidetes, and Fusobacteria (Fig. 2A). Firmicutes was the highest contributor to the bacterial populations in NG (50.48%) and CG (45.08%). Proteobacteria was the second highest contributing phylum and accounted for 33.55% and 43.76% of the bacteria in NG and CG, respectively. Bacteroidetes was the third highest contributing phylum and accounted for 13.05% and 6.95% of the bacteria in NG and CG, respectively. Fusobacteria was the fourth most abundant phylum, contributed 1.45% of the bacteria in NG and 3.06% of those in CG. Within individual study participants, the composition of the microbes was highly variable. For example, Firmicutes accounted for 5.09%–85.99%, Bacteroidetes contributed 0.47%–45.17%, and Fusobacteria provided 0.11%–15.52% of the microbial composition in all study participants (Supplementary Fig. 2). In addition, the numbers of Bacteroidetes were higher whereas the numbers of Proteobacteria were lower in the gut microbiota of the NG samples compared with those of the CG samples. Although the distribution of these phyla did not differ significantly between the groups, Fusobacteria increased in CG relative to that in NG. This trend was also observed at the class, order, and family levels (Fig. 2B–2D). Clostridia was the dominant class and order, and a trend, although not statistically significant, toward decreasing numbers from NG to CG was found. The compositions of the dominant genera in the samples are shown in Supplementary Fig. 3. Blautia, Salmonella, Faecalibacterium, Bacteroides, Dorea, Hemophilus, Coprococcus, Prevotella, Neisseria, and Streptococcus were the 10 most abundant genera in NG. In CG, the 10 most abundant genera were Salmonella, Faecalibacterium, Blautia, Bacteroides, Coprococcus, Hemophilus, Dorea, Neisseria, Fusobacterium, and Streptococcus. These genera constituted over 60% of the total bacteria in each group, and nine of these genera were common to both groups. Salmonella was the most abundant genus in CG (21.7% vs. 10.9% in NG, P = 0.07), while Prevotella was higher in NG than in CG. Only the 20 most abundant species in each group are shown in Supplementary Fig. 4. Although not statistically significant, Clostridium perfringens was higher in NG (0.16%) than in CG (0.0042%).

Identification of key contributors for structural segregation of the groups

LEfSe software presents a powerful identification function that can identify high-dimensional biomarkers and reveal genomic features through statistically significant biological differences. The algorithm emphasizes statistical significance and biological correlations, allowing researchers to identify different features and related categories. In this study, LEfSe was used to identify specific bacterial phylotypes for which the abundance was significantly different between the groups (Fig. 3); the details are shown in Supplementary Table 2. Devosia (class Alphaproteobacteria, order Rhizobiales, family Hyphomicrobiaceae) was enriched in the CG. The species stercorea (genus Prevotella, family Prevotellaceae), copri (genus Prevotella, family Prevotellaceae), and producta (genus Blautia, family Lachnospiraceae; the genera Eubacterium (family Erysipelotrichaceae), 02d06 (family Clostridiaceae), Phascolarctobacterium (family Veillonellaceae), Leptotrichia (family Leptotrichiaceae), and Klebsiella (family Enterobacteriaceae); and the family Desulfovibrionaceae (class Deltaproteobacteria, order Desulfovibrionales) were enriched in NG (Table 2 and Fig. 3).

Functional properties predicted by PICRUSt

We performed PICRUSt analysis of the 16S rRNA sequences to determine whether the taxonomic differences between the groups corresponded to functional changes (Table 3). Compared with normal group, the CRC group showed a remarkably larger abundance of KEGG pathways affiliated with cancers (small cell lung cancer, colorectal cancer, P<0.05), infectious diseases (influenza A, toxoplasmosis, P <0.05), cardiovascular diseases (viral myocarditis, P<0.05), cell growth and death (p53 signaling pathway), transport and catabolism (endocytosis, P <0.05), immune system (FcγR-mediated phagocytosis, P<0.05), and endocrine system (GnRH signaling pathway, P <0.05), as well as a significantly lower abundance of the KEGG pathway affiliated with Vibrio cholerae infection (P <0.05).

Discussion

Many studies have shown that the gut microbiota and their metabolites are associated with CRC; however, changes in the populations of intestinal microbes, their effect on CRC, and the related mechanism(s) remain unclear. Although no microorganism that is pathogenic to any host from a microecological point of view has yet been uncovered, specific perturbations to the intestinal microbiome are indicative of some disease states, and specific gut bacteria are known to be involved in the pathogenesis of CRC. In this work, we compared microbial populations in mucosal-luminal interface samples from study participants with CRC versus normal controls. Previous studies report that the diversity of the gut microbiota in CRC patients is reduced compared with that in healthy controls, which may be caused by inhibition of the immune response in the pathological tissue. In the current study, PCoA results exhibited a slight difference in the overall microbiota structure between CG and NG (Supplementary Fig. 1) using the first two principal component scores, which indicates that changes in the gut microbiota may contribute to the pathogenesis of CRC. In addition, no distinct clustering of the points for the two groups based on visual inspection of the PCoA score plot was found, which is likely related to the low power of the study because of its small sample size.

The human microbiome contains over 1000 microbial species; the gastrointestinal tract itself harbors up to 100 trillion bacteria [¹⁷]. Although many types of bacteria are found in the gut, the amounts of each species vary widely. Over 99% of the gut microbiome is composed of 30–40 types of bacterial species [¹⁸]. These bacteria can be divided into three broad groups according to their different physiological functions in the gut: commensal bacteria, conditional pathogens, and pathogenic bacteria. Commensal bacteria occupy more than 99% of the gut microbiota, produce beneficial substances, and protect human health [¹⁹]. Compared with commensal bacteria, fewer conditional pathogens are found in the gut; however, allowing these pathogens to reproduce under certain conditions can have deleterious effects on the body. Pathogenic bacteria, such as Salmonella, Shigella, Proteus, Escherichia coli, and Jerson, directly cause disease. Salmonella was the most abundant genus in CG and found in higher amounts in CG than in NG, thus indicating that intestinal flora disorder is potentially more serious in the former than in the latter. In both groups, the intestinal microbiome was composed mainly of Firmicutes, Bacteroidetes, and Proteobacteria, followed by far less abundant Fusobacteria and Actinobacteria. On average, participants in CG had higher amounts of Proteobacteria and Fusobacteria and lower amounts of Bacteroidetes and Firmicutes than those in NG. These results are consistent with previous studies of human samples and animal models of CRC [²⁰,²¹]. Fusobacteria is a small group of Gram-negative bacteria commonly found in the digestive tract that can cause some diseases. Large numbers of Fusobacterium are associated with CRC, but its role in disease development is unclear [⁸,²²,²³]. Although the average number of Fusobacterium in the mucosal-luminal interface samples was higher in CG than that in NG, further studies with larger sample sizes are needed to confirm this trend. Previous studies have shown that the C. leptum and C. coccoides subgroups are specific to CRC [²⁴], but neither species was detected in the present study. However, we found that Clostridium perfringens tended to decrease in CG, which is similar to a finding by Sasada et al. from mouse models [²⁵].

Firmicutes includes a large group of bacteria, most of which are Gram positive. Firmicutes is highly enriched in the intestinal lumen and can enhance energy harvesting from the diet [²⁶]; the phylum consists of several genera, including Clostridium, Blautia, Coprococcus, Dorea, and Streptococcus, and the butyrate producers Eubacterium and Faecalibacterium. Butyric acid is an important short-chain fatty acid that is an energy source for colonic epithelial cells and regulates gene expression, inhibits inflammation, and prevents tumorigenesis [²⁷]. The in vivo supply of butyrate strongly depends on butyrate-producing bacteria, which mainly exist in the cecum and colon. Here, we identified a significant reduction in Eubacterium in the gut microbiota of CRC patients; this genus was further identified as a biomarker by LEfSe analysis. Balamurugan et al. reported that Eubacterium and Faecalibacterium decreased by approximately 4-fold in CRC patients compared with those in healthy control volunteers [²⁸]. In the present study, Faecalibacterium comprised 10.5%–13.0% of the microbiota in the different patient groups with no significant differences between groups. A randomized clinical trial showed that butyrylated starch intake can prevent red meat-induced O6-methyl-2-deoxyguanosine adducts in human rectal tissue [²⁹]. In another dietary intervention study, the abundance of Eubacterium showed a strong positive correlation with fecal butyrate concentrations in response to carbohydrate intake [³⁰], thus revealing the importance of Eubacterium in the production of butyrate in vivo. Many animal experiments and clinical studies have shown that butyrate plays an important role in the repair of intestinal mucosa and the treatment of esoenteritis. According to a study by Sengupta et al., the delivery of adequate butyrate to the appropriate sites appears to protect against early tumorigenic events [³¹]. Therefore, our results, together with previous reports, suggest that Eubacterium may play an important role in protecting hosts from colorectal carcinogenesis. However, a study with greater power requires a larger sample size; further research is also needed to explore the underlying mechanisms.

We observed significant increases in the richness of Devosia in CG compared with that in NG to CG. Devosia is a type of mycotoxin deoxynivalenol-degrading bacterium that belongs to the family Hyphomicrobiaceae [³²]. The proportion of Devosia bacteria increases as the degree of radiation in polluted soils raises, and Devosia is one of the most abundant genera in propylene oxide saponification wastewater treatment plants [³²]. Consistent with these data, we observed high levels of Devosia in the CRC samples, as well as a negative correlation between enterobacterial and Devosia levels. These findings suggest that the genus could be a promising biomarker for tumorigenesis. Nevertheless, further investigation is required to fully understand and evaluate its role in the progress of CRC.

According to the predictive functional profiles of microbial communities determined by PICRUSt analysis, the most abundant functions were infectious diseases, cancers, transport and catabolism, biosynthesis of other secondary metabolites, and endocrine system (Table 3). Disorders of the KEGG pathways of colorectal cancer, small cell lung cancer, and the p53 signaling pathway may be related to the carcinogenic mechanism of CRC. Increases in the endocytosis pathway indicated increased communication between cells and CRC environments. Alterations in the KEGG pathways of influenza A, toxoplasmosis, viral myocarditis, the GnRH signaling pathway, and FcγR-mediated phagocytosis may explain the serious adverse health effects caused by CRC.

Because of the difficulties associated with obtaining samples from the intestinal mucosa, one limitation of the current study is the sample size of each group; this limitation inhibits the drawing of definitive conclusions related to the changes to the intestinal microbiota during the progression of CRC based solely on the results of this study. In addition, this research only focused on characterizing the observed microbiota; intestinal metabolites were not analyzed. We believe that further studies including the analysis of intestinal metabolites will provide a better understanding of the mechanisms leading to CRC.

In conclusion, in this case-control study, the abundance of gut microbiota at different levels were detected in CRC patients and healthy volunteers. The presence of exclusive microbial genera in each group indicates the existence of potential biomarkers for the disease. Our results reveal that the observed abundance of species belonging to Eubacterium and Devosia may act as a promising biomarker for the early detection of CRC. Nonetheless, more detailed information on these taxa and further exploration of intestinal lumen-associated microbiota remain essential. Further research on the relationship between different organisms and the etiology of CRC could lead to the monitoring of individual microbial strains for the early detection of intestinal cancer. Eventually, future work may establish new treatments based on the application of probiotic strains.

References

[1] Cunningham D, Atkin W, Lenz HJ, Lynch HT, Minsky B, Nordlinger B, Starling N. Colorectal cancer. Lancet 2010; 375(9719): 1030–1047

[2] Regula J, Rupinski M, Kraszewska E, Polkowski M, Pachlewski J, Orlowska J, Nowacki MP, Butruk E. Colonoscopy in colorectal-cancer screening for detection of advanced neoplasia. N Engl J Med 2006; 355(18): 1863–1872

[3] Azcárate-Peril MA, Sikes M, Bruno-Bárcena JM. The intestinal microbiota, gastrointestinal environment and colorectal cancer: a putative role for probiotics in prevention of colorectal cancer? Am J Physiol Gastrointest Liver Physiol 2011; 301(3): G401–G424

[4] Horiuchi Y, Fujisaki J, Ishizuka N, Omae M, Ishiyama A, Yoshio T, Hirasawa T, Yamamoto Y, Nagahama M, Takahashi H, Tsuchida T. Study on clinical factors involved in Helicobacter pylori-uninfected, undifferentiated-type early gastric cancer. Digestion 2017; 96(4): 213–219

[5] Nicholson JK, Holmes E, Kinross J, Burcelin R, Gibson G, Jia W, Pettersson S. Host-gut microbiota metabolic interactions. Science 2012; 336(6086): 1262–1267

[6] Yatsunenko T, Rey FE, Manary MJ, Trehan I, Dominguez-Bello MG, Contreras M, Magris M, Hidalgo G, Baldassano RN, Anokhin AP, Heath AC, Warner B, Reeder J, Kuczynski J, Caporaso JG, Lozupone CA, Lauber C, Clemente JC, Knights D, Knight R, Gordon JI. Human gut microbiome viewed across age and geography. Nature 2012; 486(7402): 222–227

[7] Parkin DM. The global health burden of infection-associated cancers in the year 2002. Int J Cancer 2006; 118(12): 3030–3044

[8] Amitay EL, Werner S, Vital M, Pieper DH, Höfler D, Gierse IJ, Butt J, Balavarca Y, Cuk K, Brenner H. Fusobacterium and colorectal cancer: causal factor or passenger? Results from a large colorectal cancer screening study. Carcinogenesis 2017; 38(8): 781–788

[9] Yamaoka Y, Suehiro Y, Hashimoto S, Hoshida T, Fujimoto M, Watanabe M, Imanaga D, Sakai K, Matsumoto T, Nishioka M, Takami T, Suzuki N, Hazama S, Nagano H, Sakaida I, Yamasaki T. Fusobacterium nucleatum as a prognostic marker of colorectal cancer in a Japanese population. J Gastroenterol 2018; 53(4): 517–524

[10] Repass J, Maherali N, Owen K, Reproducibility Project: Cancer Biology. Registered report: Fusobacterium nucleatum infection is prevalent in human colorectal carcinoma. eLife 2016; 5:e10012

[11] Mira-Pascual L, Cabrera-Rubio R, Ocon S, Costales P, Parra A, Suarez A, Moris F, Rodrigo L, Mira A, Collado MC. Microbial mucosal colonic shifts associated with the development of colorectal cancer reveal the presence of different bacterial and archaeal biomarkers. J Gastroenterol 2015; 50(2): 167–179

[12] Mottawea W, Chiang CK, Mühlbauer M, Starr AE, Butcher J, Abujamel T, Deeke SA, Brandel A, Zhou H, Shokralla S, Hajibabaei M, Singleton R, Benchimol EI, Jobin C, Mack DR, Figeys D, Stintzi A. Altered intestinal microbiota-host mitochondria crosstalk in new onset Crohn’s disease. Nat Commun 2016; 7(1): 13419

[13] Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 2013; 41(Database issue): D590–D596

[14] Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber CF. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 2009; 75(23): 7537–7541

[15] Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, Huttenhower C. Metagenomic biomarker discovery and explanation. Genome Biol 2011; 12(6): R60

[16] Langille MG, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes JA, Clemente JC, Burkepile DE, Vega Thurber RL, Knight R, Beiko RG, Huttenhower C. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nat Biotechnol 2013; 31(9): 814–821

[17] Ley RE, Peterson DA, Gordon JI. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 2006; 124(4): 837–848

[18] Kimura K, McCartney AL, McConnell MA, Tannock GW. Analysis of fecal populations of bifidobacteria and lactobacilli and investigation of the immunological responses of their human hosts to the predominant strains. Appl Environ Microbiol 1997; 63(9): 3394–3398

[19] Sears CL. A dynamic partnership: celebrating our gut flora. Anaerobe 2005; 11(5): 247–251

[20] Zhu Q, Jin Z, Wu W, Gao R, Guo B, Gao Z, Yang Y, Qin H. Analysis of the intestinal lumen microbiota in an animal model of colorectal cancer. PLoS One 2014; 9(6): e90849

[21] Wang T, Cai G, Qiu Y, Fei N, Zhang M, Pang X, Jia W, Cai S, Zhao L. Structural segregation of gut microbiota between colorectal cancer patients and healthy volunteers. ISME J 2012; 6(2): 320–329

[22] Tahara T, Yamamoto E, Suzuki H, Maruyama R, Chung W, Garriga J, Jelinek J, Yamano HO, Sugai T, An B, Shureiqi I, Toyota M, Kondo Y, Estécio MR, Issa JP. Fusobacterium in colonic flora and molecular features of colorectal carcinoma. Cancer Res 2014; 74(5): 1311–1318

[23] McCoy AN, Araújo-Pérez F, Azcárate-Peril A, Yeh JJ, Sandler RS, Keku TO. Fusobacterium is associated with colorectal adenomas. PLoS One 2013; 8(1): e53653

[24] Scanlan PD, Shanahan F, Clune Y, Collins JK, O’Sullivan GC, O’Riordan M, Holmes E, Wang Y, Marchesi JR. Culture-independent analysis of the gut microbiota in colorectal cancer and polyposis. Environ Microbiol 2008; 10(3): 789–798

[25] Sasada T, Hinoi T, Saito Y, Adachi T, Takakura Y, Kawaguchi Y, Sotomaru Y, Sentani K, Oue N, Yasui W, Ohdan H. Chlorinated water modulates the development of colorectal tumors with chromosomal instability and gut microbiota in APC-deficient mice. PLoS One 2015; 10(7): e0132435

[26] Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 2006; 444(7122): 1027–1031

[27] Scharlau D, Borowicki A, Habermann N, Hofmann T, Klenow S, Miene C, Munjal U, Stein K, Glei M. Mechanisms of primary cancer prevention by butyrate and other products formed during gut flora-mediated fermentation of dietary fibre. Mutat Res 2009; 682(1): 39–53

[28] Balamurugan R, Rajendiran E, George S, Samuel GV, Ramakrishna BS. Real-time polymerase chain reaction quantification of specific butyrate-producing bacteria, Desulfovibrio and Enterococcus faecalis in the feces of patients with colorectal cancer. J Gastroenterol Hepatol 2008; 23(8 Pt 1): 1298–1303

[29] Le Leu RK, Winter JM, Christophersen CT, Young GP, Humphreys KJ, Hu Y, Gratz SW, Miller RB, Topping DL, Bird AR, Conlon MA. Butyrylated starch intake can prevent red meat-induced O6-methyl-2-deoxyguanosine adducts in human rectal tissue: a randomised clinical trial. Br J Nutr 2015; 114(2): 220–230

[30] Duncan SH, Belenguer A, Holtrop G, Johnstone AM, Flint HJ, Lobley GE. Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces. Appl Environ Microbiol 2007; 73(4): 1073–1078

[31] Sengupta S, Muir JG, Gibson PR. Does butyrate protect from colorectal cancer? J Gastroenterol Hepatol 2006; 21(1 Pt 2): 209–218

[32] Sato I, Ito M, Ishizaka M, Ikunaga Y, Sato Y, Yoshida S, Koitabashi M, Tsushima S. Thirteen novel deoxynivalenol-degrading bacteria are classified within two genera with distinct degradation mechanisms. FEMS Microbiol Lett 2012; 327(2): 110–117

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature