Article Cite This: J. Agric. Food Chem. 2019, 67, 916−926
pubs.acs.org/JAFC
Impact of Cyanocobalamin and Methylcobalamin on Inflammatory Bowel Disease and the Intestinal Microbiota Composition Xuan Zhu,†,‡ Shasha Xiang,†,‡ Xiao Feng,‡ Huanhuan Wang,§ Shiyi Tian,‡ Yuanyuan Xu,‡ Lihua Shi,‡ Lu Yang,§ Mian Li,∥ Yubiao Shen,⊥ Jie Chen,‡ Yuewen Chen,‡ and Jianzhong Han*,‡
J. Agric. Food Chem. 2019.67:916-926. Downloaded from pubs.acs.org by IOWA STATE UNIV on 01/23/19. For personal use only.
‡
School of Food Science and Bioengineering, Zhejiang Gongshang University, 18 Xuezheng Street, Hangzhou, Zhejiang 310018, People’s Republic of China § School of Medicine, Hangzhou Normal University, Hangzhou, Zhejiang 310018, People’s Republic of China ∥ Zhejiang Huakang Pharmaceutical Company, Limited, Kaihua, Quzhou, Zhejiang 324302, People’s Republic of China ⊥ Yangtze Delta Region Institute of Tsinghua University, Zhejiang, Jiaxing, 314000, China S Supporting Information *
ABSTRACT: Patients with inflammatory bowel disease (IBD) are usually advised to supplement various types of vitamin B12, because vitamin B12 is generally absorbed in the colon. Thus, in the current study, the influence of cyanocobalamin (CNCBL) or methylcobalamin (MECBL) ingestion on IBD symptoms will be investigated. Then, whether and how the application of various cobalamins would modify the taxonomic and functional composition of the gut microbiome in mice will be examined carefully. Dextran-sulfate-sodium-induced IBD mice were treated with MECBL or CNCBL; disease activity index (DAI) scores and intestinal inflammatory conditions of mice were evaluated. Fecal samples were collected; microbiota composition was determined with a 16s rRNA analysis; functional profiles were predicted by phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt); and short-chain fatty acids were measured. The consequence of higher relative abundances of Enterobacteriaceae and isomeric short-chain fatty acids by cobalamin treatment revealed that a high concentration of CNCBL but not MECBL supplementation obviously aggravated IBD. A microbial ecosystem rich in Escherichia/ Shigella and low in Lactobacillus, Blautia, and Clostridium XVIII was observed in IBD mice after a high concentration of CNCBL supplementation. In cobalamin-dependent enzymes, CNCBL was more efficient in the adenosylcobalamin system than MECBL and vice versa in the MECBL system. The distinct effects of various cobalamins were associated with the distribution and efficiency of vitamin-B12-dependent riboswitches. CNCBL had a strong inhibitory effect on all riboswitches, especially on btuB and pocR riboswitches from Enterobacteriaceae. CNCBL aggravated IBD via enhancing the proportion of Enterobacteriaceae organisms through riboswitch and enzyme systems. The present study provides a critical reference for offering a suitable amount and type of cobalamin during a symbiotic condition. KEYWORDS: cyanocobalamin, methylcobalamin, gut microbiome, inflammatory bowel disease, Enterobacteriaceae
■
INTRODUCTION Inflammatory bowel disease (IBD), including ulcerative colitis and Crohn’s disease, is an uncontrolled chronic and recurring inflammatory response in the gastrointestinal tract that has been linked to mucosal immunity dysregulation and alterations in the intestinal microbiota.1 IBD commonly involves the terminal ileum, which is the major site of vitamin B12 absorption. Vitamin B12 abnormalities are common in IBD patients with a prior ileal or ileocolonic resection.2 Therefore, IBD patients, especially those who are elderly, are usually advised to supplement vitamin B12. Cobalamin (vitamin B12), one of the watersoluble vitamins, is the general name for naturally occurring organometallic compounds containing cobalt and works as a cofactor for two important human enzymes. Cobalamin deficiency is believed to be related with disturbances in cell division, neuropathy, nervous system disease, and pernicious anemia.3 To prevent such fatal diseases caused by cobalamin deficiency, daily intake of 2.4 μg of cobalamin is recommended.4 Cobalamin includes four bioactive forms. The adenosyl group could be replaced by a methyl group, a hydroxyl group, and a cyano group to form methyl-, hydroxo-, and cyanocobalamin, respectively. © 2018 American Chemical Society
Cyanocobalamin (CNCBL) does not exist naturally but, nowadays, is used as a supplementary nutrient for humans and stocks. Vitamin B12 is the only vitamin produced exclusively by bacteria and archaea but used by many domains of life. Synthesis of vitamin B12 is energetically costly, requiring nearly 30 different enzymes.5 Only 20% of prokaryotes have the genetic capacity to produce it, such as Aerobacter, Agrobacterium, Alcaligenes, Azotobacter, Bacillus, Clostridium, Corynebacterium, Flavobacterium, Micromonospora, Mycobacterium, Norcardia, Propionibacterium, Protaminobacter, Proteus, Pseudomonas, Rhizobium, Salmonella, Serratia, Streptomyces, Streptococcus, and Xanthomonas.5,6 In gut microbial ecology, the limited sources of vitamin B12 are recognized as a precious commodity. As such availability, vitamin B12 has been suggested to impart a fundamental contribution, as not only nutrition but also a signal, to the spatial and functional organization of gut Received: Revised: Accepted: Published: 916
October 18, 2018 December 14, 2018 December 14, 2018 December 20, 2018 DOI: 10.1021/acs.jafc.8b05730 J. Agric. Food Chem. 2019, 67, 916−926
Article
Journal of Agricultural and Food Chemistry microecology.5,6 However, the IBD symptoms of some patients are aggravated after supplementation with different kinds of vitamin B12, as observed in our studies and other investigations.7 Multiple causes can be attributed in this deregulation, but many studies have shown that dysbiosis of the gut microbiota8 and genes involved in interactions of host− microorganisms are linked to IBD.9 A relative abundance of species such Escherichia coli, Ruminococcus gnavus, Alistipes putredinis, and Clostridium difficile were reported to be associated with IBD.10 Meanwhile, both the human and mouse intestinal microbiota are dominated by the same members of 15 000 species and only varied between 1 and 160 species, which can be considered to be similar in broad terms.11 Nevertheless, we lack an understanding of how different cobalamins shape the composition of the gut microbiome, how cobalamin mediates the growth of Enterobacteriaceae by enzymes or riboswitches, and further, how it influences IBD. Riboswitch is a 5′-untranslated leader sequence of the correspondent mRNA, which regulates translation initiation and gene expression by binding to a specific molecule.12 VitaminB12-dependent riboswitch may act as a regulator during IBD development. Expression of cobalamin biosynthetic cob operon and transporter btuB gene was repressed by the presence of CNCBL, methylcobalamin (MECBL), and adenosylcobalamin (ADCBL).13 It has been reported that ADCBL directly binds to the riboswitch region, leading to a conformational change in the secondary structure of mRNA, which masks the ribosomebinding site (RBS), thus inhibiting gene expression.14 Degnan et al. demonstrated that 313 genomes of gut microbiota include 809 vitamin B12 riboswitches predicted to regulate 3868 genes, most of which are related to vitamin-B12-related enzyme, transporter, and isoezyme, however, part of which have not been previously associated with vitamin B12.6 Consequently, those results suggest that vitamin B12 riboswitches may shape gut microbiota ecology as a result of its wide association with a relative abundance range of vitamin-B12-dependent and/or -regulated protein expression. Thus, in the current study, we used 60 mice to dissect the reasons for IBD aggravation after supplementation with various cobalamins. We experimentally examined whether the application of cobalamin would aggravate IBD and modify the taxonomic and functional composition of the gut microbiome in mice.
■
CNCBL or MECBL were prepared in drinking water and administered to mice in the last 3 days together with DSS treatment. Cobalamin is stable under a pH value between 5.0 and 8.0. DSS consists of dextran and sulfate units, both of which do not react with cobalamin in those pH conditions. The high concentration of cobalamins was calculated from the recommended intake of humans (1.5 mg/day) to a dosage for mice according to body surface. See the details of IBD induction and treatment in S Figure 1 of the Supporting Information. Mice were sacrificed on day 7, and colon tissues were rapidly removed and measured for length and weight. After the feces in the colonic lumen were washing out with saline, the colons were fixed in 4% neutral paraformaldehyde solution for morphologic evaluation. The remaining portions were stored in liquid nitrogen for further investigation. Evaluation of the Disease Activity Index (DAI). Throughout the experiment, we recorded the following parameters of each mouse daily to determine the DAI according to a previously reported protocol:15,16 loss of body weight, stool consistency, and presence of occult or gross bleeding in feces. The detailed information is listed in S Table 1 of the Supporting Information. Evaluation of the Histological Score. The colon samples were collected, and histological scores were evaluated following a previous protocol.17 Briefly, colons were fixed in a 4% phosphate-buffered paraformaldehyde solution for 24 h, dehydrated via a graduated ethanol series, and embedded in paraffin blocks. Colon sections (5 mm) were cut and stained with hematoxylin and eosin (H&E). Morphological changes in the colons were evaluated by two pathologists who did not know the experimental design. Quantitative Polymerase Chain Reaction (Q-PCR). Colons were ground rapidly in liquid nitrogen. Total RNA was isolated using TRIzol reagent (Sangon Biotech, Shanghai, China) according to the protocol of the manufacturer. Total RNA were then converted to cDNA using a High-Capacity cDNA Reverse Transcription (RT) Kit (Takara, Shiga, Japan). The quantification of gene expression was determined by real-time Q-PCR followed by the SYBR Green PCR Master Mix (Roche) standard protocol. The relative expression levels were determined via the ΔΔCt method, using β-actin as an endogenous control. The primers used in the assays are listed in S Table 2 of the Supporting Information. DNA Isolation, PCR, and 16S rRNA Gene Analysis. DNA from different samples was extracted using a MicroElute Genomic DNA Kit (D3096-01, Omega Bio-tek, Inc., Norcross, GA, U.S.A.) according to a previously described method, with a small modification.18 The total DNA was eluted in 50 μL of elution buffer by a modified version of the procedure described by the manufacturer (QIAGEN), and the samples were stored at −80 °C until PCR amplification (LC-Bio Technology Co., Ltd., Hangzhou, Zhejiang, China). Bacteria 16S rRNA sequencing genes (V3−V4 regions) were amplified from the whole genome of furu samples via the primer pair (319F, 5′-ACTCCTACGGGAGGCAGCAG-3′; 806R, 5′-GGACTACHVGGGTWTCTAAT-3′) according to a previous method, with a small modification.18 All reactions were conducted in 25 μL (total volume) mixtures, including approximately 25 ng of a genomic DNA extract, 12.5 μL of PCR Premix, 2.5 μL of each primer, and PCR-grade water to adjust the volume. PCR reactions were performed in a Mastercycler gradient thermocycler (Eppendorf, Hamburg, Germany) set to the following conditions: initial denaturation at 98 °C for 30 s, 35 cycles of denaturation at 98 °C for 10 s, annealing at 54/52 °C for 30 s, and extension at 72 °C for 45 s, and the final extension at 72 °C for 10 min. The PCR products were normalized by an AxyPrep Mag PCR normalizer (Axygen Biosciences, Union City, CA, U.S.A.), which allowed for the quantification step to be skipped, regardless of the PCR volume submitted for sequencing. The amplicon pools were prepared for sequencing with AMPure XT beads (Beckman Coulter Genomics, Danvers, MA, U.S.A.), and the size and quantity of the amplicon library were assessed on the LabChip GX (PerkinElmer, Waltham, MA, U.S.A.) and with the Library Quantification Kit for Illumina (Kapa Biosciences, Woburn, MA, U.S.A.), respectively. The PhiX control library (V3, Illumina) was combined with the amplicon library (expected at 30%). The library was clustered to a
MATERIALS AND METHODS
Animals. Male C57BL/6 mice (10 weeks old) were provided by the Centre of Laboratory Animals, Hangzhou Normal University. Mice were housed under a 12 h dark/light cycle at a controlled room temperature (22 °C) and a relative humidity of 50% and were provided standard laboratory chow. Before the experimental procedures, animals were acclimatized for 5 days. The Animal Care and Use Committee of the School of Medicine, Hangzhou Normal University, approved the experimental protocols in this study. Induction of Colitis and Treatment. A total of 60 mice were randomly divided into 10 groups: normal control (NC) group (group 1), dextran sulfate sodium (DSS) + saline-treated group (group 2), DSS + CNCBL-treated group at low (0.0125 mg/L, group 3), medium (0.125 mg/L, group 4), and high (1.25 mg/L, group 5) concentrations, DSS + MECBL-treated group at low (0.0125 mg/L, group 6), medium (0.125 mg/L, group 7), and high (1.25 mg/L, group 8) concentrations, CNCBL (1.25 mg/L, Sigma-Aldrich, St. Louis, MO, U.S.A.)-treated group (group 9), and MECBL (1.25 mg/L, SigmaAldrich, St. Louis, MO, U.S.A.)-treated group (group 10). For induction of acute colitis, 5% (w/v) DSS (MP Biochemicals, Santa Ana, CA, U.S.A.) was dissolved in drinking water and administered to mice for 5 days. For cobalamin treatment, certain concentrations of 917
DOI: 10.1021/acs.jafc.8b05730 J. Agric. Food Chem. 2019, 67, 916−926
Article
Journal of Agricultural and Food Chemistry
Figure 1. Effects of MECBL and CNCBL on IBD. (A) DAI score evaluation from DSS- and cobalamin-treated mice. DAI scores were calculated on the basis of body weight, stool consistency, and occult bleed test from mice (n = 6; #, p ≤ 0.01 versus DSS). (B) Mice were sacrificed, and colon tissue were collected, measured, and weighted (n = 6; ∗, p ≤ 0.05). (C) Histopathological H&E staining analysis of colon tissue from DSS- and cobalmin-treated mice. Red boxes represent leukocyte infiltration, and red asterisks represent mucosal ulceration (scale bar = 200 μm). (D) Colon tissues were collected, and mRNA was extracted, followed by RT-PCR experiments. Inflammatory cytokines, IL-1β, IL-6, and TNFα, were evaluated (n = 6; ∗, p ≤ 0.05 versus DSS). 37 °C in liquid MRS media. Cells were grown for 48 h in corresponding media, washed, and resuspended in triplicate at an optical density at a wavelength of 600 nm (OD600) in vitamin B12 analysis media supplemented with 0.02 mg/L MECBL or 0.02 mg/L CNCBL. Triple parallel tests of each strain under various cobalamin supplementations were conducted. OD600 measurements of both strains were internally collected every 2 h for 48 h. The supernatants of E. coli E4 were collected every 6 h from 30 to 48 h. The shiga toxins were analyzed via colorimetric enzyme-linked immunosorbent assay (ELISA) immunoassay previously described by Gehring et al., with a tiny modification.22 All wells were filled with 200 μL of phosphatebuffered saline containing 0.05% Tween 20 (PBST) and immediately emptied by rapidly inverting the plate. Supernatants (100 μL) of E. coli E4 were added to wells for 30 min. The wells were washed twice with PBST. Then, 100 μL of horseradish peroxidase (HRP)− antibody conjugate cocktail was added to each well, followed by static incubation for 20 min. Wells were washed twice with PBST and then once with PBS. To each well, 50 μL of tetramethylbenzidine and 50 μL of H2O2 were separately added and reacted for 10 min. A total of 50 μL of H2SO4 was added to stop the reaction. The contents of shiga toxin were determined by OD450. Riboswitch Sensor Comparison and Construction. Six genera of gut microbiota (S Table 3 of the Supporting Information), including three Gram-negative genera and three Gram-positive genera, were chosen to scan for vitamin-B12-dependent riboswitches. We subsampled six publicly available human gut microbial genomes23 into a custom database that includes a single representative of each species. The methods of searching for riboswitches were reported by Degnan et al.24 To construct cobalamin ribosensor switches, the above determined fragments containing an inferred riboswitch and RBS from gut microbiota (S Table 3 of the Supporting Information) were amplified. The fragments containing a riboswitch and RBS were PCR-amplified. The amplicon was cloned in p519 ngfp between the pnpt2 promoter and green fluorescent protein (GFP) (S Figure 2 of the Supporting Information) and sequenced as described in a previous work.25 For
density of approximately 570 000/mm2. The libraries were sequenced on the 300PE MiSeq runs, and one library was sequenced with both protocols using the standard Illumina sequencing primers, eliminating the need for a third (or fourth) index read. 16S rRNA gene sequences were processed and modified as in previously described methods.19 The reads were controlled and confirmed by quantitative insights into microbial ecology (QIIME, http:// qiime.org/tutorials/processing_illumina_data.html), VSEARCH (version 2.3.4, https://github.com/torognes/vsearch), FastQC, and fast length adjustment of short reads (FLASH, version 1.2.8, http://ccb.jhu. edu/software/FLASH) quality filters. A Cd-hit method was introduced to select operational taxonomic units (OTUs) by generation of an OTU table.20 When the similarity of OTUs was over 97%, sequences were assigned to one unit. Ribosomal Database Project (RDP) classifiers were applied to distribute the 16S rRNA gene into distinct taxonomic categories by aligning representative sequences to taxonomically annotated sequences.21 To calculate the α diversity, we rarefied the OTU table and calculated two metrics: the Chao1 metric, which estimates the richness, and the Shannon index. Principal component analysis (PCA) was carried out by all taxa relative abundances. The heatmap of an important family of the gut microbiome was created by Mev. 4-9-0. A Spearman correlation based on the relative abundances of various genera in bacterial communities of this study was applied. Co-occurrence networks were plotted via Cytoscape 3.4.0 software. When the p value was
