Article Cite This: J. Agric. Food Chem. 2018, 66, 8864−8875
pubs.acs.org/JAFC
Microbiome−Metabolomics Analysis of the Impacts of Long-Term Dietary Advanced-Glycation-End-Product Consumption on C57BL/6 Mouse Fecal Microbiota and Metabolites Wanting Qu,† Chenxi Nie,† Jinsong Zhao,† Xiyang Ou,† Yingxiao Zhang,† Shanchun Yang,† Xue Bai,† Yong Wang,‡,§ Jiawei Wang,§ and Juxiu Li*,†
Downloaded via UNIV OF GOTHENBURG on January 25, 2019 at 01:10:41 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
College of Food Science and Engineering, Northwest A&F University, 22 Xinong Road, Yangling, Shaanxi Province 712100, PR China ‡ Shaanxi Research Institute of Agricultural Products Processing Technology, Xi’an, Shaanxi Province 710016, PR China § Shaanxi University of Science and Technology, Xi’an, Shaanxi Province 710016, PR China S Supporting Information *
ABSTRACT: Thermally processed diets are widely consumed, although advanced-glycation end products (AGEs) are unavoidably formed. AGEs, clusters of protein-cross-linking products, become less digestible because they impair intestinal peptidase proteolysis. We characterized the impacts of dietary AGEs on gut microbiota through a microbiome-to-metabolome association study. C57BL/6 mice were fed a heat-treated diet (high-AGE diet, H-AGE) or a standard AIN-93G diet (low-AGE diet, L-AGE) for 8 months. Fecal-microbiota composition was examined by 16S rDNA sequencing, and fecal-metabolome profile was evaluated by gas chromatography−tandem time-of-flight mass spectrometry (GC-TOF-MS). Reduced α-diversity and altered microbiota composition with elevated Helicobacter levels were found in the H-AGE group, and among the 57 perturbed metabolites, protein-fermentation products (i.e., p-cresol and putrescine) were increased. Major dysfunctional metabolic pathways were associated with carbohydrate and amino acid metabolism in two groups. Moreover, high correlations were found between fluctuant gut microbiota and metabolites. These findings might reveal the underlying mechanisms of the detrimental impacts of dietary AGEs on host health. KEYWORDS: advanced-glycation end products (AGEs), gut microbiota, metabolome, short-chain fatty acids (SCFAs), protein fermentation of early oral-CML-administration studies.4,5 Recently, researchers have taken an interest in the roles of AGEs that are not absorbed. These unabsorbed AGEs are deemed to change the gut-microbiota composition, with an adverse effect on gastrointestinal (GI)-tract health.6 However, these issues have not been investigated completely to date, and there is a limited knowledge regarding the influence of diet-derived AGEs on the gut microbiota and metabolome. Dietary composition undoubtedly plays a key role in the intestinal ecosystem, long-term dietary habits undoubtedly have crucial effects on human-gut microbiota. Many studies have demonstrated relationships between an imbalanced microbiota structure and inflammatory disorders,7−9 and microbiota metabolites are one of the most predominant connections.9 During food digestion, the intestinal microbiota coproduces a large array of small molecules, which can enter the bloodstream through absorption, enterohepatic circulation, or a leaky gut.10 These small molecules play critical roles in shuttling information between the microbial symbionts and their host’s cells.11 Nicholson et al.12defined the metabolic
1. INTRODUCTION Modern diets have embraced excessive amounts of processed foods, primarily heat-treated foods such as grilled and fried foods, which are often featured in Western diets. Although diets differ among nations and regions, eating habits have changed remarkably as a result of changes in income, urbanization, and globalization. One of the most profound results of those changes is that many Western-style fast-food outlets have become widely distributed worldwide, and their rate of distribution is increasing.1 Western diets are largely heat-processed, which results in the inevitable formation of extensive advanced-glycation end products (AGEs). In the Maillard reaction, reducing sugars react with free amino groups in proteins; thus, AGE formation is accompanied by protein cross-linking. The most well-known representative AGEs are Nε-(carboxymethyl)-lysine (CML), carboxyethyllysine, pyrraline, and pentosidine. There is emerging evidence that chronic exposure to excessive amounts of AGEs is correlated with the pathogenesis of diabetes mellitus, cardiovascular diseases, hypertension, and nephropathy.2 AGEs exert pro-inflammatory bioactivity under the condition in which they are first absorbed. As a result of cross-linking and protein aggregation, only 10−30% of dietary AGEs are absorbed and enter circulation, according to kinetic studies.3 Additionally, the CML-recovery rate never attains 100% in excreta on the basis © 2018 American Chemical Society
Received: Revised: Accepted: Published: 8864
March 21, 2018 July 23, 2018 July 23, 2018 July 23, 2018 DOI: 10.1021/acs.jafc.8b01466 J. Agric. Food Chem. 2018, 66, 8864−8875
Article
Journal of Agricultural and Food Chemistry host−microbe axis as “a multidirectional interactive chemical communication highway between specific host cellular pathways and a series of microbial species and activities”. Some metabolites have positive impacts on host health, such as providing antioxidant and anti-inflammatory activities. One example is short-chain fatty acids (SCFAs), which are mainly produced through carbohydrate fermentation. SCFAs regulate immunity and energy metabolism, improve insulin secretion, and exert antidiabetic effects by binding to GPR41 and GPR43.9 In contrast, other metabolites are deleterious to the host and cause toxicity, cytotoxicity, and genotoxicity, such as metabolites produced by the fermentation of proteinaceous material in the distal colon.13 Hellwig et al.14 first investigated the stability of CML after 24 h of incubation with human-feces microbiota in vitro; at least 40.7 ± 1.5% of the CML was found to be degraded. Our early study confirmed that the intake of dietary AGEs for 18 weeks in rats reduced the α-diversity and richness of the cecal microbiota and adversely altered gut-microbiota composition.15 Those results suggested that mutual modulation might exist between the intestinal microflora and dietary AGEs. Additionally, AGEs have been reported to increase proteolytic metabolism, leading to the secretion of a range of putrefactive metabolites.16 The negative effects of dietary AGEs on GI-tract diseases, such as inflammatory-bowel disease (IBD) and colorectal cancer, are likely caused not only by the adverse actions of glycated amino acids themselves but also by the their associated microbial metabolites.17 However, whether other diverse metabolites are also influenced by dietary AGEs is not clear. Therefore, more research is urgently required to identify the microbial metabolites of dietary AGEs and to determine whether AGEs can influence host metabolic pathways. Untargeted metabolomics can capture variations of metabolites in biological samples to determine changes in metabolic phenotypes in nutrition intervention.18 This process could also be applied to explore the specific metabolites in feces after dietary AGE intervention. In this study, we fed C57BL/6 mice a heat-treated AIN-93G diet for 8 months and identified the specific microbiota and their metabolites in the feces by developing a multivariate strategy employing 16S rDNA gene sequencing and fecal-metabolite profiling of the gut-microbiota structure using gas chromatography−tandem time-of-flight mass spectrometry (GC-TOF-MS).
Table 1. Compositions of the Experimental Diets AIN-93G composition a
carbohydrates fata proteina caloriesa fluorescent AGEsb CML contentb CEL contentb MGO contentb GO contentb
unit
regular (L-AGE)
heated (H-AGE)
% % % kcal/kg AU/g μg/g μg/g mg/kg mg/kg
64.3 7.0 17.8 3766 2147.53 ± 61.83 142.67 ± 23.08 0.97 ± 0.33 12.09 ± 0.81 1.12 ± 0.27
64.3 7.0 17.8 3766 6968.08 ± 93.34** 271.58 ± 16.16** 6.26 ± 0.84** 49.09 ± 6.51* 28.73 ± 3.04**
a
Data provided by the Trophic Animal Feed High-Tech Company, Ltd. (Nantong, China). bMeans ± SD (n = 5). Asterisks (**) indicate significant differences with p < 0.01, as determined by Student’s t test.
contents were significantly higher in the H-AGE diet. Food intake and body weights were measured twice per month. In the last week at the end of the study protocol, each mouse was housed individually in a metabolic cage for the collection of fecal samples, which were each divided into two parts for the subsequent gut-microbiota and metabolome analyses. For the microbiota analysis, feces were collected immediately in a sterile chilled tube and frozen with liquid nitrogen immediately. For the metabolome analysis, the sample was mixed with one drop of 1% (w/v) NaN3 and then quickly frozen. All of the animal-experimental procedures followed the Guide for the Care and Use of Laboratory Animals22 according to the Northwest A&F University Animal Care and Use Committee. 2.2. Fecal Microbiota: 16S rDNA Sequencing. Microbial DNA was extracted from each stool sample (200 ± 20 mg) using the EZNA Stool DNA Kit (Omega Biotek, Inc., Norcross, GA). To ensure complete cell lysis, almost 200 mg of 1 mm sterile glass beads was added with the lysis buffer, and the samples were completely homogenized with a vortexer for 10 min. To increase the extractedDNA concentration, the elution step in which the DNA from the HiBind DNA column was dissolved with the Elution Buffer was repeated twice. The integrity of the extracted-DNA sample was characterized and determined by Nano-200 nucleic acid and protein spectrophotometry. The V3−V4 region of the 16S rDNA gene was amplified using forward (5′-ACTC CTAC GGGA GGCA GCA-3′) and reverse primers (5′-GGAC TACH VGGG TWTC TAAT-3′) with dual-index barcodes. After quantification, equimolar concentrations of 16 PCR products were sequenced on the Illumina HiSeq platform at Biomarker Technologies Company, Ltd. (Beijing, China). 2.3. Bioinformatics and Statistical Analyses. Paired-end raw reads were merged from the original DNA fragments with FLASH (version 1.2.7). After quality filtering with Trimmomatic (version 0.33), a total of 1 098 978 clean reads were extracted. Then, effective tags were produced by removing the chimeric sequences with UCHIME (version 4.2). Finally, after stringent quality checking and data cleaning, high-quality effective tags with 97.70% Q20 bases and 95.48% Q30 bases (i.e., base quality greater than 20 and 30) were applied for the subsequent bioinformatics analysis. Using QIIME (version 1.8.0), all effective reads from each sample were clustered into operational taxonomic units (OTUs) on the basis of 97% sequence similarity according to UCLUST, and the representative sequence of each OTU was aligned against the Silva database for taxonomy analysis. For α-diversity analysis, rarefaction and Shannonindex curves were generated, and the ACE and Chao1 estimators and Simpson and Shannon indices were calculated using Mothur (version 1.30). The β-diversity of the microbial communities was explored with principal-coordinate-analysis (PCoA) plots based on unweighted and weighted UniFrac distances. Unweighted-pair-group method with arithmetic mean (UPGMA) was also performed using QIIME. Taxonomy-based analyses were performed to identify the significantly different phylotypes between the H-AGE- and L-AGE-treated groups. In the Mann−Whitney test, only taxa with an average abundance
2. MATERIALS AND METHODS 2.1. Animals and Diets. Four week old male C57BL/6 mice were acquired from Beijing Vital River Laboratory Animal Technology Company, Ltd. (Beijing, China) and housed with two or three animals per cage. The controlled environmental conditions were as follows: the temperature was 22 ± 1 °C, the humidity was 55 ± 15%, and there was a 12−12 h light−dark cycle. After an adaptation period of 2 weeks, 8 mice were randomly chosen for each group. The lowAGE-diet (L-AGE) group was fed a standard AIN-93G diet, and the high-AGE-diet (H-AGE) group was fed a heat-treated AIN-93G diet for 8 months. The mice were free to access rodent chow and water. The L-AGE and H-AGE diets were manufactured by Trophic Animal Feed High-Tech Company, Ltd. (Nantong, China). The specific-dietproduction process was reported previously by our group15 with the slight modification that the H-AGE diet was exposed to 175 °C for 45 min. In order to avoid nutrient differences, identical AIN-93G vitamin premixtures and mineral premixtures were added and mixed well after the heating step. Diet fluorescent AGE,19 CML, CEL,20 GO, and MGO21 contents were measured. As shown in Table 1, both diets were isocaloric and identical in terms of fiber, mineral, and vitamin contents, but the fluorescent AGE, CML, CEL, GO, and MGO 8865
DOI: 10.1021/acs.jafc.8b01466 J. Agric. Food Chem. 2018, 66, 8864−8875
Article
Journal of Agricultural and Food Chemistry >1%, a p value 1.0 and a p value 1.0 from OPLS-DA modeling and p values
