Topographical Variation in Murine Intestinal Metabolic Profiles in Relation to Microbiome Speciation and Functional Ecological Activity Francois-Pierre J. Martin,* Yulan Wang, Ivan K. S. Yap, Norbert Sprenger, John C. Lindon, Serge Rezzi, Sunil Kochhar, Elaine Holmes, and Jeremy K. Nicholson* Nestle´ Research Center, P. O. Box 44, Vers-chez-les-Blanc, CH-1000 Lausanne 26, Switzerland, Department of Biomolecular Medicine, Division of Surgery, Oncology, Reproductive Biology and Anaesthetics, Faculty of Medicine, Imperial College London, Sir Alexander Fleming Building, South Kensington, London SW7 2AZ, United Kingdom, and State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Centre for Magnetic Resonance, Wuhan Institute of Physics and Mathematics, The Chinese Academy of Sciences, Wuhan, 430071, PR China Received February 7, 2009
Symbiotic gut microbes can have a significant influence on host health and disease etiology. Here, we assessed the effects of inoculating germfree mice with human baby microbiota (HBM, n ) 17) on the biochemical composition of intact intestinal tissues (duodenum, jejunum, ileum, proximal and distal colon) using magic-angle-spinning 1H NMR spectroscopy. We compared the HBM tissue metabolite profiles with those from conventional (n ) 9) and conventionalized (n ) 10) mice. Each topographical intestinal region showed a specific metabolic profile that was altered differentially by the various microbiomes, especially for osmolytes. In each animal model, duodenum had higher ethanolamine and myo-inositol, and ileum higher taurine and betaine than other gut regions. HBM mice showed lower taurine and myo-inositol in the colon, and all ex-germfree animals had higher taurine, choline and ethanolamine in the jejunum. Interestingly, the jejunum of HBM mice was marked by a higher glutathione level and lower concentrations of its precursor methionine when compared to other groups. Proximal and distal colon tissues were differentiated in the different microbiome models by the concentrations of bacterial products (higher in conventional animals). These studies show the depth of gut microbiome modulations of the intestinal biochemistry. Keywords: high resolution magic angle spinning proton nuclear magnetic resonance spectroscopy • human baby microbiota • metabolomics • microbiome • intestine • metabolic profiling • symbiosis
Introduction Gut microbial species, their activities and their metabolic capabilities vary between individuals due to differences in diet or colonization history. Such variations can impact upon health and predisposition to various diseases such as inflammatory bowel diseases and colon cancer.1-3 Human microbial communities are spread across a wide range of habitats ranging from the entire body to specific mucosal surfaces (e.g., gut, stomach, vagina), or even on undigested food particles in the distal intestine.2 The intestinal tract is also an important organism-environment interface, with a complex web of metabolic interactions between the thousands of microbe species inhabiting the intestine and the host.4 The complex microbial community differs in composition along the length of the intestine (with an increasing gradient of indigenous microbes from the stomach to the colon) and across the * To whom correspondence should be addressed. Prof. J. Nicholson: e-mail,
[email protected]; tel, +44 (0)20 7594 3195; fax, +44 (0) 20 7594 3226. Dr. F.-P. Martin: e-mail,
[email protected]; tel, + 41 (0) 21 785 8771; fax, +41 (0) 21 785 9486.
3464 Journal of Proteome Research 2009, 8, 3464–3474 Published on Web 06/03/2009
diameter of the gastrointestinal tract, and is comprised of both rapidly transiting and relatively persistent components.3 The gut symbionts communicate with surrounding tissues to shape a host environment that fosters their implantation and persistence.5 In particular, bacteria influence angiogenesis, development of intestinal villi, renewal of gut epithelial cells,6 fortification of the mucosal barrier, and promotion of host epithelial cell production of fucosylated glycans on which many gut bacteria feed.7,8 In addition, the subversion of host-cell pathways by bacterial pathogens is reported as a determinant factor in the etiology of disease and onset of deregulation of metabolic homeostasis.9 Other functional contributions of the gut microbiota include energy recovery from poorly digestible nutrients,10 and altered lipid metabolism in the host through modification of bile acids.11,12 It is also known that the microbiota are involved in the development and maintenance of certain gut disorders,13 as well as production of neurologically active compounds.14 One of the main aims of modern nutritional science is to develop new functional foods to optimize health through modulation of gut symbionts, and this approach includes the 10.1021/pr900099x CCC: $40.75
2009 American Chemical Society
Microbiome and Topographical Intestinal Metabotype
research articles
13,15,16
use of probiotic and prebiotic strategies. Probiotic supplementation aims at replacing or reducing the number of potentially harmful bacteria in the intestine by enriching the populations of gut microbiota that ferment carbohydrates.15 Prebiotics are generally oligosaccharides that are resistant to digestion in the upper gastrointestinal tract, and selectively stimulate the growth and/or activity of beneficial members of the gut microbiota in the colon.17 In particular, the combined use of prebiotics and probiotics, that is, synbiotic intervention, may offer superior synergistic effects in health maintenance.18 Therefore, understanding how the distribution, activities and evolution of gut microbes determine the response of an organism to environmental stimuli is key to optimizing nutrition with respect to promoting health.19 Nutrimetabonomics offers a novel strategy for measuring changes in the metabolic end-points of the physiological regulatory processes of an organism after specific nutritional interventions.1,20 We have recently described the modification of the calorific bioavailability in the host via transplantation of a simplified human baby microbiota (HBM) into mice,11 and we have also shown how the use of pro- and prebiotics modulates the gut microbiota.21,22 These studies formed the basis of a strategy aiming to define the extent and importance of the microbiome symbiotic metabolic exchanges with the mammalian host. Deciphering the interactions between gut microbial residents and the physiology of distinct intestinal compartments at the molecular level will provide new strategies for restoring and/or maintaining human health. For instance, we have determined typical metabolic signatures reflecting the structure and function of the intestine via spectroscopic profiling of intestinal biopsies from humans,23 rats,24 and gnotobiotic mice,25 which has provided a set of reference metabolite profiles that can be used to assess gut microbial impact at the tissue level.25 Moreover, metabolic profiling of blood plasma and intact gut cross sections has been successfully applied to explore further the molecular mechanisms associated with the probiotic-induced normalization of gastrointestinal disorders in a Irritable Bowel Syndrome mouse model.26 In the present study, we sought to capture the metabolic impact of the gut microbiota on the biochemical composition of intact intestinal tissues (duodenum, jejunum, ileum, proximal and distal colon) obtained from four different microbiome mouse models (conventional, conventionalized, HBM colonized with probiotic supplementation, and HBM colonized with synbiotic intervention). Figure 1 provides a schematic of the experimental design. This investigation aims at the characterization of the effects of different gut microbiomes on the metabolic profiles of intestinal tissues to promote understanding of how these changes correlate with system metabolic changes, such as altered dietary fat recovery and lipid metabolism.11
Materials and Methods Animal Handling Procedure. All animal studies were carried out under appropriate national guidelines at the Nestle´ Research Center (Lausanne, Switzerland) and are part of a larger study previously published.11 The experimental design is given in Figure 1. A total of 36 C3H female mice (9 conventional and 27 germfree animals, in-house breeding), aged 6 weeks, were fed with a standard semisynthetic AIN93G rodent diet consisting of 50% cornstarch, 20% casein, 10% sucrose, 7% soybean oil, 5% cellulose, 0.25% choline bitartrate, 0.3% cystine, and
Figure 1. Schematic of the experimental study design.
vitamin and mineral mixtures.27 All groups were supplied with ad libitum saline (0.9% NaCl) solution. A group of conventional mice was kept as control (n ) 9). One group of germfree mice was conventionalized by removal from their germfree isolator and exposure to normal environment for a period of 4 weeks (n ) 10, Figure 1). A second group of germfree mice received a single dose of simplified HBM by oral gavage (n ) 17, Figure 1). At 8 weeks of age, HBM mice were given Lactobacillus paracasei probiotic bacteria in Man, Rogosa and Sharpe (MRS) culture medium (108 cfu/mL) daily for a period of 2 weeks and were separated into two groups. From 8 to 10 weeks of age, conventional, conventionalized and one group of HBM mice supplemented with probiotics (n ) 7) were fed with a basal mix diet containing in composition 2.5% of a glucose-lactose mix (1.25% each). Conventional and conventionalized animals received a saline drink ad libitum containing MRS culture medium as a placebo for probiotic intervention. A second group of HBM mice supplemented with probiotics was fed with a diet containing 3 g per 100 g diet of an in-house preparation of galacto-oligosaccharides for a period of 2 weeks (n ) 10, defined as synbiotic treatment). The in-house preparation of prebiotics is composed of 80% of commercially available galactosyl-oligosaccharides (VivinalGOS, Borculo Domo Ingredients, Netherlands) and 20% of a mixture containing other in-house galactosyl-oligosaccharide structures. The Vivinal-GOS comprises 75% dry matter in syrup and a dry matter basis composed of 23% lactose, 22% glucose, 0.8% galactose and 54.2% galactosyl-oligosaccharides with Degree of Polymerization (DP) ranging between 3 and 9, and primarily composed of β-1,4 linkages. The relative sugar composition of Vivinal-GOS was previously reported, with DP 3 oligomers accounting for 37% of total carbohydrates.28 The proprietary oligosaccharides are primarily composed of DP 3 oligomers with β-1,3 and β-1,6 linkages. The control diet was supplemented with lactose and glucose to control for the lactose and glucose that were introduced into the experimental diets by the galactosyl-oligosaccharide preparations. The human baby microbiota is composed of a total of 7 bacterial strains, isolated from stool of a 20 day old female baby who was naturally delivered and breast-fed, namely, Escherichia coli, Bifidobacterium breve, Bifidobacterium longum, Staphylococcus epidermidis, Staphylococcus aureus, Clostridium perfringens, and Bacteroides distasonis. Bacterial cell mixtures Journal of Proteome Research • Vol. 8, No. 7, 2009 3465
research articles
Martin et al. 10
contained approximatively 10 cells/mL for each strain and were kept in frozen aliquots until use. L. paracasei NCC2461 probiotics were obtained from the Nestle´ Culture Collection (Lausanne, Switzerland) and grown under anaerobic conditions in MRS culture medium. Sample Collection. Animals were euthanized at 10 weeks age and the intestine was removed from each animal. The first centimeter of gut after the stomach was designated as duodenum and the rest of the intestine to the cecum was divided into three sections, the first 2/3 were designated as jejunum and remaining 1/3 as ileum, and 3 cm samples were excised from the middle of each section. The colon was separated into two parts, defined as proximal and distal colon. Each sample was flushed using 1 mL of an iso-osmotic phosphate buffer solution (0.2 M Na2HPO4/0.04 M NaH2PO4, pH ) 7.4) using a sterile syringe. Intact tissue samples were placed in Eppendorf tubes and then preserved at -80 °C prior to NMR spectroscopic analysis. Microbial Profiling of Fecal Contents. Microbiological tests were carried out on jejunum tissues and on the fecal pellets as described previously.11 Briefly, for each mouse, 1 fecal pellet and the jejunal content were individually homogenized in 0.5 mL of Ringer solution (Oxoid, U.K.) supplemented with 0.05% (w/v) L-Cystein (HCl) and were plated on selective and semiselective media for the enumeration of specific microorganisms: B. breve and B. longum on Eugon Tomato medium (Chemie Brunschwig, CH), L. paracasei on MRS medium (Chemie Brunschwig, CH) with antibiotics (phosphomycin, sulfamethoxazole and trimethoprim) medium (Sigma, CH), C. perfringens on NN-agar medium (Chemie Brunschwig, CH), E. coli on Drigalski medium (BioRad, CH), Bacteroides distasonis on Shaedler Neo Vanco medium (BioMe´rieux, CH), and S. aureus and S. epidermidis on Chapman medium (BioMe´rieux, CH). The bacterial cultures of E. coli, S. aureus and S. epidermidis were incubated at 37 °C under aerobic conditions for 24 h and those of B. breve, B. longum, L. paracasei, B. distasonis, and C. perfringens were incubated under anaerobic conditions for 48 h. Sample Preparation and 1H MAS NMR Spectroscopic Analysis. Intact intestinal tissue samples were bathed in 0.9% saline D2O solution. A portion of each tissue (approximately 15 mg) was packed individually into 4 mm diameter zirconia rotors and a drop of deuteriated isotonic saline solution was added to provide a field-frequency lock for the NMR spectrometer. Rotors were closed with an insert and Kel-F cap to make a spherical sample volume of 25 µL. All 1H NMR experiments were carried out on a Bruker DRX-600 spectrometer (Bruker Biospin, Rheinstetten, Germany), operating at a 1 H frequency of 600.11 MHz and equipped with a tripleresonance (1H, 13C, 31P), high-resolution, magic-angle spinning (MAS) probe with a magic-angle gradient. Samples were spun at 5000 Hz at the magic angle (54.7°) and sample temperature was regulated to 283K using cooled N2 gas during the acquisition of spectra to minimize biochemical degradation.29 A total of 10 min was allowed for the temperature to reach equilibrium for each sample before a spectrum was acquired. For all the samples, three types of 1H MAS NMR spectra were acquired for each sample, (a) a standard one-dimensional pulse sequence with water suppression,30 (b) a Carr-PurcellMeiboom-Gill (CPMG),31 pulse sequence with water suppression, and (c) a diffusion-edited pulse sequence.32 Since the CPMG spin-echo experiment gave the clearest signature of metabolic changes in relation to gut microbiota and intestinal 3466
Journal of Proteome Research • Vol. 8, No. 7, 2009
compartments, only the CPMG results are analyzed in detail here. The CPMG spin-echo experiment attenuates signals from macromolecules with short spin-spin relaxation times using the pulse sequence D1-90°-(τ-180°-τ-)n-acquisition with water suppression.31 CPMG spin-echo spectra were measured using a spin-echo loop time (2nτ) of 200 ms and a relaxation delay D1 of 2.0 s. A total of 128 transients were collected into 32 k data points for each spectrum with a spectral width of 20 ppm. For assignment purposes, 2D COrrelation SpectroscopY (COSY)33 and TOtal Correlation SpectroscopY (TOCSY)34 were acquired for selective samples as detailed previously.25,35 Data Processing. Free induction decays were multiplied by an exponential function equivalent to a 0.3 Hz linebroadening factor prior to Fourier transformation. 1H MAS NMR spectra of tissues were manually phased and baselinecorrected using XwinNMR 3.5 (Bruker Analytik, Rheinstetten, Germany). The 1H NMR spectra were referenced to the chemical shift of the methyl resonance of alanine at δ 1.47. The spectra over the range of δ 0.2-10.0 were imported into Matlab (Version 7.0, The Mathworks, Inc, Natwick, MA) using 22K data points. The water resonance signal (δ 4.50-5.19) was removed to avoid the effects of imperfect water suppression. Normalization of each spectrum to a constant sum was carried out prior to conducting statistical analysis. Multivariate Statistical Analyses. The multivariate pattern recognition techniques used in this study were based on Principal Component Analysis (PCA),36 Projection to Latent Structure (PLS)37 analysis and the Orthogonal Projection to Latent Structure (O-PLS) approach.38 NMR variables were subjected to unit variance scaling after mean centering. PCA was carried out using SIMCA-P 11.5 software (Umetrics, Umeå, Sweden) in order to detect the presence of inherent similarities between metabolic profiles. Data were visualized by means of principal component scores, where each point represents an individual biochemical profile of a sample. Biochemical components responsible for any detected differences between samples in the scores plot can be extracted from the corresponding loadings plot, where each coordinate represents a single NMR spectral region. The PCA loadings (covariance) plots were analyzed using a back-scaling transformation, which preserves the original spectral structure appearance for NMR data.39 Supervised methods such as PLS and OPLS discriminant analysis approaches operating in a MATLAB environment were applied to maximize the discrimination of experimental groups (i.e., the specific tissue or the gut microbiome model) and focus on metabolic variations related to the changes in gut microbiota. For visualization purposes, the O-PLS coefficients indicating influential variables contributing to the discrimination in the model were processed according to the method previously described.39 The standard 7-fold cross validation method was used to test the validity of the model.39 The classification accuracy of the O-PLS-DA model was established from the prediction-set samples in the 7-fold cross-validation, using a decision-rule based on the largest predicted Y-value. The statistical total correlation spectroscopy (STOCSY)40 method was also applied to aid elucidation of the structure of metabolites, whereby a correlation matrix was calculated between the data point at the apex of each candidate biomarker signal and the rest of the spectral variables.
Results Microbiological Status. Microbiological profiles were generated from the fecal samples for each animal to assess the
research articles
Microbiome and Topographical Intestinal Metabotype a
Table 1. Microbial Species Counts in Mouse Feces at the End of the Experiment groups /log
10
cfu
Lactobacilli Enterobacteria Bifidobacteria Staphylococcus C. perfringens Bacteroides
conventional (n ) 9)
conventionalized (n ) 10)
HBM-L. paracasei (n ) 7)
HBM-synbiotic (n ) 10)
6.4 ( 1.5 6.5 ( 1.0 7.0 ( 1.2 4.8 ( 0.7
