Effects of Probiotic Lactobacillus Paracasei Treatment on the Host Gut Tissue Metabolic Profiles Probed via Magic-Angle-Spinning NMR Spectroscopy Francois-Pierre J. Martin,† Yulan Wang,† Norbert Sprenger,‡ Elaine Holmes,† John C. Lindon,† Sunil Kochhar,§ and Jeremy K. Nicholson*,† 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, UK, and Department of Nutrition and Health and Bioanalytical Sciences, Nestle´ Research Center, P. O. Box 44, Vers-chez-les-Blanc, CH-1000 Lausanne 26, Switzerland Received November 13, 2006
We have used a simplified gnotobiotic mouse model to evaluate the effects of single bacterial species, Lactobacillus paracasei NCC2461, on the metabolic profiles of intact intestinal tissues using highresolution magic-angle-spinning 1H NMR spectroscopy (HRMAS). A total of 24 female gnotobiotic mice were divided into three groups: a control group supplemented with water and two groups supplemented with either live L. paracasei or a γ-irradiated equivalent. HRMAS was used to characterize the biochemical components of intact epithelial tissues from the duodenum, jejunum, ileum, proximal, and distal colons in all animals and data were analyzed using chemometrics. Variations in relative concentrations of amino acids, anti-oxidant, and creatine were observed relating to different physiological properties in each intestinal tissue. Metabolic characteristics of lipogenesis and fat storage were observed in the jejunum and colon. Colonization with live L. paracasei induced region-dependent changes in the metabolic profiles of all intestinal tissues, except for the colon, consistent with modulation of intestinal digestion, absorption of nutrients, energy metabolism, lipid synthesis and protective functions. Ingestion of γ-irradiated bacteria produced no effects on the observed metabolic profiles. 1H MAS NMR spectroscopy was able to generate characteristic metabolic signatures reflecting the structure and function of intestinal tissues. These signals acted as reference profiles with which to compare changes in response to gut microbiota manipulation at the tissue level as demonstrated by ingestion of a bacterial probiotic. Keywords: colon • duodenum • HRMAS 1H NMR spectroscopy • ileum • intact tissue • intestine • jejunum • Lactobacillus paracasei • Metabolomics • Metabonomics • Chemometrics • O-PLS-DA
Introduction The gastrointestinal system has a histologically diverse epithelial structure superimposed on a complex longitudinal and cross sectional ordering of multiple connective and muscular cell types. The functional complex of the gastrointestinal tract is responsible for digestion, absorption, and propulsion of dietary components through the digestive tract and the ejection of waste. The intestinal surface is an important host organism-environment boundary and interactions of gut microbes in the gastrointestinal tract are important in immune system development.1 Symbiotic microbial species in the gastrointestinal tract also help to protect against opportunistic pathological infection, and the gut environment allows the intimate coexistence of dissimilar species that exert mutually * To whom correspondence should be addressed. Prof. J. Nicholson (
[email protected]), Tel, +44 (0)20 7594 3195; Fax, +44 (0) 20 7594 3226. † Imperial College London. ‡ Department of Nutrition and Health, Nestle ´ Research Center. § Bioanalytical Sciences, Nestle´ Research Center. 10.1021/pr060596a CCC: $37.00
2007 American Chemical Society
beneficial effects on the host organism.1 This is well illustrated in a recent study on gnotobiotic mice colonized with Bacteroides thethaiotaomicron, where these micro-organisms were shown to devote a significant proportion of their genomic resource to allow an otherwise poorly accessible source of nutrients to be utilized by host metabolism.2 In addition, the presence of enteric gut microbiota affect intestinal morphology, physiology and modulate gut motility.3,4 Hence, understanding the interactions that occur between gut microbiota and their host may provide new insights into the aetiologies of many gastrointestinal diseases.5 As a consequence, much attention has recently been focused on the use of probiotics as food supplements, as a means of preventing allergies and inflammatory disease and for promoting gut health. One approach to aiding the understanding of the interactions between micro-organisms and their hosts is to utilize a simplified mammalian ecosystem in the absence of bacteria and to evaluate the effects on host functions after colonizing the gut with a single microbial species.5 In the current study, the effects of Lactobacillus paracasei, a promising probiotic for human Journal of Proteome Research 2007, 6, 1471-1481
1471
Published on Web 02/23/2007
research articles health-care, have been investigated in animals that were raised without any resident micro-organisms. It has been shown in vitro and in vivo that L. paracasei is able to shape the intestinal physiology, aid digestion and absorption of nutrients, as well as improve intestinal microbial balance and prevent infection of pathogenic bacteria.6,7 Moreover, L. paracasei has been shown to produce antagonist metabolites and antioxidants, to stimulate cells of the immune system in vitro,8 to induce a homeostatic micro-environment that favors tolerance rather than allergy in gnotobiotic mice, and to contribute to the normalization of gastrointestinal disorders.9,10 Recently, gene expression profiling approaches have been applied to elucidate the molecular mechanisms underlying symbiotic host-bacterial relationships.2,11 Here, a complementary metabonomic approach has been employed for the biochemical characterization of metabolic changes triggered by gut microbiota. Metabonomics involves the study of the dynamic multi-parametric metabolic response of complex cellular organisms to biological stimuli or genetic modification,12 using in this case high resolution 1H magic-anglespinning nuclear magnetic resonance (HRMAS NMR) spectroscopy coupled with multivariate statistical methods. MAS averages major line broadening factors including dipole-dipole interaction, chemical shift anisotropy and magnetic field inhomogeneities and results in narrow line width.13 HRMAS NMR spectroscopy is nondestructive and can provide molecular structural information as well as relative quantitative information on small intact tissue samples by generating a profile that contains physiological and pathological information.14 Recently, a HRMAS NMR-based approach has been successfully applied to characterize the biochemistry of rat intestine development15 and healthy human gastric mucosa,16 and to study multi-organ effects of drug-induced toxicity in animal models,17,18 metabolic compartmentation in the rat heart and cardiac mitochondria19 and to investigate the “microbe-mammalian metabolic axis” interactions in a Trichinella spiralis-induced Irritable Bowel Syndrome mouse model.20 In the present study, we have extended our metabolic studies to characterize the biochemical composition of intact intestinal mouse tissues in the absence of microbiota and then to evaluate the impact of colonization of gnotobiotic animals with both live and γ-irradiated L. paracasei.
Materials and Methods Animal Handling Procedure and Sample Collection. All animal studies were carried out under appropriate national guidelines at the Nestle´ Research Center. A total of 24 gnotobiotic C3H female mice, age 6 weeks, were randomly assigned to 3 groups of 8 mice and all received the same standard diets. L. paracasei, strain NCC2461, was obtained from the Nestle´ Culture Collection. At day 0, one group of gnotobiotic mice was kept as control and received water only by oral gavage, one group received a single dose of live bacteria (log10 cells ) 8.78) and one group received γ-irradiated bacteria (log10 cells ) 9.18) both in their spent culture medium. The maintenance of pathogen-free conditions and the intestinal establishment of L. paracasei were evaluated by bacterial counting using single freshly collected fecal pellet at timed intervals over the duration of the experiment. Briefly, for each mouse 1 fecal pellet was homogenized in 0.5 mL Ringer solution (Oxoid, UK) supplemented with 0.05% (w/v) L-Cystein (HCl) and 2 times 100 µL thereof were plated on 2 TSS plates (Trypcase soy agar with 5% sheep blood; BioMerieux, France). One plate was incubated 1472
Journal of Proteome Research • Vol. 6, No. 4, 2007
Martin et al.
aerobically for 24 h at 37 °C, and the second plate was incubated anaerobically for 48 h at 37 °C. At euthanasia, parts of the intestine were removed aseptically and immediately transferred into preweighed tubes with Ringer solution supplemented with 0.05% (w/v) L-cystein (HCl), weighed, homogenized, and plated onto MRS (Man Rogosa Sharpe) agar plates (Oxoid). Plates were incubated anaerobically for 48 h at 37 °C. As a secondary objective of the study that is not evaluated here, all mice also received a single intra-gastric inoculation of whey proteins at day 15 after the start of the experiment for induction of tolerance to β-lactoglobulin. Five days after oral tolerance induction, at day 19, the animals were immunized with specific β-lactoglobulin antigen via subcutaneous injections. Animals were euthanized at day 30 by isoflurane anaesthesia followed by exhaustive bleeding by sampling the cardiac aorta. After sacrifice, the intestine was removed from each animal. The first 1 cm of gut after the stomach was designated as duodenum and the rest of the intestine to the caecum 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. Microbiological tests were carried out on jejunum, ileum and colon tissues to verify absence of contamination and the L. paracasei population, as described by Guigoz et al.21 Each sample was flushed using an iso-osmotic phosphate buffer solution (pH 7) and then snap-frozen in liquid nitrogen and then preserved at -80 °C prior to NMR spectroscopic analysis. Sample Preparation and 1H NMR Spectroscopic Analysis. Intact intestinal tissue samples were bathed in 0.9% saline D2O solution. A portion of each tissue (approximately 15 mg) was inserted into a zirconium oxide 4 mm outer diameter rotor, using an insert to make a spherical sample volume of 25 µL. A drop of deuteriated isotonic saline solution was added to provide a field-frequency lock for the NMR spectrometer. All 1H NMR spectra were recorded on a Bruker AV-600 NMR spectrometer (Rheinstetten, Germany) operating at 600.11 MHz for 1H, equipped with a high-resolution magic-angle-spinning probe at a spin rate of 5000 Hz. Sample temperature was regulated using cooled N2 gas to 283 K during the acquisition of spectra to minimize biochemical degradation. Three different types of 1H NMR spectra were collected for each sample, a standard one-dimensional (1D) spectrum with water suppression, a 1D Carr-Purcell-Meiboom-Gill (CPMG) spin-echo spectrum, and a diffusion-edited spectrum. Because the CPMG spin-echo experiment gave the clearest signature of metabolic changes following intervention, with little extra information contained in the higher molecular weight components observed in the other types of spectra, only the results derived from the CPMG experiments are presented here. CPMG spin-echo spectra were acquired using the pulse sequence [RD-90°-(t-180°-t)n - acquire FID], with a spin-spin relaxation delay, 2nτ, of 200 ms.22 The 90° pulse length was 9.0-12 µs. A total of 128 transients were collected into 32 K data points. The recycle delay (RD) was 2 s. For assignment purposes, 2D correlation spectroscopy (COSY)23 and total correlation spectroscopy (TOCSY)24 NMR spectra were acquired on selected samples. In both cases, 48 transients per increment and 256 increments were collected into 2 K data points. The spectral width in both dimensions was 10 ppm. The TOCSY NMR spectra were acquired by using the MLEV-1724 spin-lock scheme with a spin-lock power of 6 kHz. COSY spectra were recorded with gradient selection. In both 2D NMR experiments,
research articles
L. Paracasei Treatment on Gut Tissue
Figure 1. 600 MHz 1H MAS NMR CPMG spectra of the aliphatic region (δ 4.6-0.5) and aromatic region (δ 8.3-5.2) from gnotobiotic tissues: duodenum (A), jejunum (B), ileum (C), proximal colon (D), and distal colon (E). The vertical scale for the aromatic region is 8 times higher than the aliphatic region. Key: a full assignment of metabolites is given in Table 1.
the water signal was irradiated with a selective pulse (∼50 Hz) during the recycle delay. These data were zero-filled to 2 K data points in the evolution dimension. Before Fourier transformation, an unshifted sine-bell and a shifted sine-bell apodization function were applied to the free induction decays of the COSY and TOCSY spectra, respectively. Further assignment of the metabolites was also accomplished with the use of Statistical TOtal Correlation SpectroscopY (STOCSY) on 1D spectra.25 Data Processing. 1H MAS NMR spectra of tissues were manually phased and baseline corrected by using XwinNMR 3.5 (Bruker Analytik, Rheinstetten, Germany). The 1H NMR spectra were referenced to 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, The Mathworks inc, Natwick, MA), using 22 000 data points.26 The water resonance signal (δ 4.50-5.19) was removed. Due to contamination of some samples by ethanol during dissection, the regions containing ethanol peaks were also removed at δ 3.64 and δ 1.17. Normalization to a constant sum of the spectrum was carried out before statistical analyses. Multivariate Statistical Data Analysis. The multivariate pattern recognition techniques used are based on the orthogonal projection to latent structure (O-PLS) approach.26,27 In the O-PLS algorithm, the variation in the X matrix (NMR spectral data) is decomposed in three parts: first, the variation in the X matrix related to the Y matrix (treatment class), and the two last parts contain the specific systemic variation in X and any residuals, respectively.26 This leads to a model with a minimal number of predictive components defined by the number of degrees of freedom between group variance. Here, O-PLS discriminant analysis (O-PLS-DA)26 was used. Variables that influence and are correlated to the predictive components were identified from the variable coefficients, which are presented visually using our recently published back-scaling method.26 This allows the correlation of each variable with the classes to be plotted in combination with its covariance. The covariance and correlation correspond to the shape and color code, respectively. The coefficient cutoff value (r ) 0.57) was determined according to the test for the significance of the Pearson product-moment correlation coefficient, and corresponds to
a discrimination significance between classes at the level of p < 0.03 for pairwise comparisons and p < 0.0002 for analysis based on five classes. The standard 7-fold cross validation method (repeatedly leaving out a seventh of the samples and predicting them back into the model)28 was applied to establish the robustness of the model.
Results Microbiological Status of the Animals. An absence of bacterial development was observed in the gnotobiotic mice and in the group of mice supplemented with irradiated bacteria. For the mice supplemented with probiotics, an L. paracasei population was established in both the small intestine and the colon. The populations of L. paracasei were significantly greater in the colon than in the jejunum or ileum with values of log 10 (Colony Forming Unit ( standard deviation) of 9 ( 0.4 (colon), 6.6 ( 0.9 (jejunum), and 6.3 ( 1.4 (ileum). NMR Spectra of Intestinal Tissues from Gnotobiotic Mice. Representative metabolic profiles of each intestinal tissue, i.e., duodenum, jejunum, ileum, proximal, and distal colons, were obtained using 1H MAS NMR spectroscopy. Typical examples of 1H MAS CPMG NMR spectra of gut sections obtained from a gnotobiotic mouse are illustrated in Figure 1. From these spectra, 49 metabolites were unambiguously assigned, and their chemical shifts and peak multiplicity are given in Table 1 along with the corresponding 1H NMR chemical shifts and signal multiplicities. Assignment of metabolites was made by comparison with published literature15,16 and confirmed by 2D COSY, TOCSY, and STOCSY methods.25 The spectra from all tissues contained resonances from a number of amino acids, organic acids, triglycerides, fatty acids, and glycerol as well as choline, inosine, myo-inositol, scylloinositol, ethanolamine, and membrane components. Pyrimidine metabolites such as uracil and cytosine were also detected. Other assigned metabolites including creatine, glucose, ethanol, glutathione, trimethylamine-N-oxide (TMAO), betaine, and N-acetylglutamate were identified in some of the tissues. Visual inspection of the 1H NMR spectra shown in Figure 1 reveals that each tissue produces its own specific metabolite Journal of Proteome Research • Vol. 6, No. 4, 2007 1473
research articles
Martin et al.
Table 1. Assignment of the Metabolites in Gut Tissues key
isoleucine
RCH, βCH, γCH2, γCH3,δCH3
2
leucine
RCH, βCH2, γCH, δCH3,δCH3
3 4 5 6 7
valine ethanol lactate alanine arginine
8
lysine
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
acetate threonine proline glutamate methionine glutamine aspartic acid asparagine creatine ethanolamine choline phosphorylcholine GPC taurine glycine glutathione
25
lipids (triglycerides and fatty acids)
27 28 29 30 31 32 33 34 35 36 37 38
trimethylamine-Noxide (TMAO) tyrosine phenylalanine oxaloacetate myo-inositol scyllo-inositol β-glucose R-glucose formate uracil fumarate cytosine inosine
39 40 41
propionate unknown N-acetyl-glutamate
42
serine
43
cysteine
44 45 46 47 48 49
phosphorylethanolamine GPE acetoacetate betaine glycerol tryptophan
50
cystamine
δ 1H (ppm) and multiplicitya
moieties
1
26
a
metabolites
3.65(d), 1.95(m), 1.25(m)-1.45(m), 0.99(t), 1.02(d) 3.72(t), 1.96(m),1.63(m), 1.69(m), 0.91(d), 0.94(d) 3.6(d), 2.26(m), 0.98(d), 1.04(d) 1.18(t), 3.65(q) 4.11(q), 1.32(d) 3.77(q), 1.47(d) 3.76(t), 1.89(m), 1.63(m), 3.23(t)
RCH, βCH, γCH3 CH2, CH3 RCH, βCH3 RCH, βCH3 RCH, βCH2, γCH2, δCH2 RCH, βCH2, γCH2, δCH2, CH2 CH3 RCH, βCH, γCH3 RCH, βCH2, γCH2, δCH2 RCH, βCH2, γCH2 RCH, βCH2, γCH2, δCH3 RCH, βCH2, γCH2 RCH, βCH2 RCH, βCH2 N-CH3, CH2 CH2NH2, CH2OH N-(CH3)3, RCH2, βCH2 N(CH3)3, NCH2, OCH2 N-(CH3)3, OCH2, NCH2 N-CH2, S-CH2 CH2 CH2, CH2, S-CH2, N-CH, CH CH3, (CH2)n, CH2-CdC, CH2-CdO, dC-CH2-Cd, -CHdCHCH3
1.91(s) 3.59(d), 4.25(m), 1.32(d) 4.11(t), 2.02(m)-2.33(m), 2.00(m), 3.34(t) 3.75(m), 2.08(m), 2.34(m) 3.78(m), 2.14(m), 2.6(dd), 2.13(s) 3.77(m), 2.15(m), 2.44(m) 3.89(m), 2.68(m), 2.82 (m) 3.90(m), 2.86(m) 2.94(m) 3.03(s), 3.92(s) 3.15 (t), 3.83 (t) 3.2(s), 4.05(t), 3.51(t) 3.22(s), 3.61(t), 4.21(t) 3.22(s), 4.32(t), 3.68(t) 3.26(t), 3.40(t) 3.55(s) 2.17(m), 2.55 (m), 2.95 (m), 3.83(m), 4.56(m), 0.89(m), 1.27(m), 2.0(m), 2.3(m), 2.78 (m), 5.3(m) 3.26(s)
CH, CH 2,6-CH, 3,5-CH, 4-CH CH2 5-CH, 4,6-CH, 1,3-CH, 2-CH CH 2-CH, 1-CH 1-CH CH CH, CH CH CH, CH 2-CH, 8-CH, 2′-CH, 4′-CH, 5′-CH, CH2, CH2 CH2, CH3 CH3CH, CH, CH2, CH2, CH2, CH3, CH3 RCH, βCH2 RCH, βCH2 CH2, CH2
7.16(m), 6.87(m) 7.40(m), 7.33(m), 7.35(m) 2.38 (s) 3.29(t), 3.63(t), 3.53(dd), 4.06(t) 3.35 (s) 3.25 (m), 4.67 (d) 5.24(d) 8.45 (s) 5.78(d), 7.52(d) 6.51(s) 7.87(d), 5.92(d) 8.34(s), 8.24(s), 6.1(d), 4.44(t), 4.28(q), 3.92(dd), 3.85(dd) 2.19(q), 1.06 (t) 4.41(s) 1.33(d), 4.12(m), 2.06(m), 1.89(m), 2.23(t), 2.04(s), 1.95(s) 3.84(m), 3.96(m) 3.99(m), 3.10(m), 3.13(m) 3.23(m), 4.00 (m)
CH2, CH2 CH3, CH2 CH3, CH2 CH2, CH 4-CH, 7-CH, 2-CH, 5-CH, 6-CH, CH, CH2, CH2 CH2, CH2
3.30(m), 4.12 (m) 2.29(s), 3.45(s) 3.27(s), 3.90(s) 3.57(m), 3.66 (m), 3.81(m) 7.74(d), 7.55(d), 7.33(s) 7.29(t), 7.21(t), 4.06(dd), 3.49(dd), 3.31(dd) 3.06(t), 3.41(t)
3.77(t), 1.89(m), 1.72(m), 1.47(m), 3.01(t)
s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; dd, doublet of doublets.
fingerprint that allowed differentiation between tissues. Among the most striking observations of intestinal composition, the duodenum tissues showed high levels of amino acids compared 1474
Journal of Proteome Research • Vol. 6, No. 4, 2007
to other tissues, whereas the jejunum contained higher levels of lipids. In addition, high levels of taurine, glutathione, creatine, and oxaloacetate were observed in both the jejunum
research articles
L. Paracasei Treatment on Gut Tissue
and the ileum. An abundance of choline and other membrane components was also observed in the jejunum and ileum. The colon tissues showed higher levels of lipids, phosphorylcholine and uracil than the small intestinal tissues. In particular, the distal colon manifested high levels of myo-inositol and scylloinositol and low levels of cytosine. Multivariate Statistical Analysis of NMR Spectra from Gnotobiotic Murine Tissues. To better characterize the metabolic profile of each intestinal section with respect to the others, O-PLS-DA was performed on 1H MAS CPMG NMR spectra obtained from the 5 classes of gut tissues from gnotobiotic mice. The cross-validated scores plot showed statistically significant separations between the different gut compartments as presented in the type of plot that we term a “differential metabogram” (Figure 2 A), with the strongest differences between the small and large intestine. Discrimination of tissues within the small bowel and colon was also achieved, but with some overlap between adjacent compartments (for example, duodenum and jejunum). The corresponding O-PLS-DA coefficients identified the specific metabolites associated with each gut section (positive values) with respect to the rest of the intestinal tissues (negative values) (Figure 2 B-F). The weight of a variable in the discrimination is given by the square of its correlation coefficient (r2), which is color coded from zero in blue to high values (0.6-0.7) in red. In addition, further characterization of the intestinal metabolic profiles was achieved using O-PLS-DA with pairwise comparisons between intestinal compartments to better identify subtle differences. Statistically significant discrimination between each tissue was achieved, and Table 2 summarizes the characteristics of each model. To better visualize the longitudinal metabolic differences along the intestine, the sequential statistical comparison of each segment to its adjacent segment from proximal to distal intestine is depicted in Figure 3. The coefficients plots (Figure 3) show the dominant metabolites that influence the differentiation between adjacent tissues, and these are listed in Table 3. In addition, relative changes displayed using box-and-whiskers plots showed dispersion of the relative concentrations among the animals which are in a agreement with the multivariate data analysis (see Supplementary Figure 1). The duodenum was differentiated from all the other intestinal tissues by relatively high levels of observable glutathione, ethanolamine, glycerophosphorylethanolamine (GPE), and amino acids including alanine, valine, isoleucine, leucine, threonine, methionine, lysine, arginine, tyrosine, phenylalanine, and cysteine. In addition, relative higher levels of choline and glycerophosphorylcholine (GPC) were observed when comparing the duodenum to the adjacent jejunum. Jejunal tissue showed relative lower contents of glutamine and glutamate, and higher levels of phosphorylcholine and phosphorylethanolamine when compared to all other intestinal tissues. Relative elevated triglycerides, taurine, creatine and in particular glutathione were additionally observed in the jejunal tissues when compared to duodenum. Moreover, distinction between the gnotobiotic jejunum and ileum was also observed due to relative higher levels of triglycerides and amino acids including alanine, glycine, glutamate, aspartate, and asparagine in the jejunum (Figure 3). The ileal profile was associated with relative high levels of glutathione, taurine, oxaloacetate, phosphorylcholine, and betaine when compared to other tissues. Pairwise O-PLS-DA also revealed that the ileal tissue contained relative higher levels
of acetoacetate and GPC when compared to the jejunum. In addition, levels of lactate, glycine, choline, GPC, ethanolamine, and GPE were relatively higher in ileal tissue than in the colon. The proximal colon showed relative higher levels of triglycerides, uracil, acetate, and propionate than other tissues whereas the distal colon showed relative high concentrations of creatine, myo-inositol, scyllo-inositol, N-acetyl-glutamate, and lower levels of ethanolamine. When comparing the two regions of the colon, OPLS-DA showed that the distal colon contained relative higher levels of phosphorylcholine, myoinositol and scyllo-inositol, creatine, N-acetyl-glutamate, and uracil (Figure 3), whereas the proximal colon contained relative higher levels of propionate, acetate, oxaloacetate, alanine, formate, and cytosine. The variations in longitudinal metabolic profiles of the gut are likely to have strong functional significance and will be considered later. Metabolic Changes Induced by L. paracasei in Intestinal Tissues. Visual inspection of the 1H NMR spectra shown in Supplementary Figure 2 does not reveal specific metabolite fingerprint that allowed differentiation between tissues from controls and probiotic supplemented mice. To characterize the effects of live and irradiated L. paracasei in the intestinal metabolic profiles, O-PLS-DA was applied to 1H CPMG MAS NMR spectra of gut tissues obtained from germ-free mice and those supplemented with live or irradiated probiotics. The comparison of the metabolic profiles from control gnotobiotic mice and those supplemented with live L. paracasei indicated that the probiotics induced changes in the small intestine (duodenum, jejunum, and ileum) as described by the statistical model descriptors shown as coefficients plots (Figure 4). These show the metabolites exerting the strongest influence on the separation of the two animal classes, i.e., germ-free and live probiotic supplemented. These significant metabolites are listed in Table 4. Moreover, no probiotics-induced metabolic variation was observed in the colon as indicated by the negative values of the model Q2 parameter meaning that the discriminant function has no statistical power. In contrast with live probiotics supplementation, metabolic profiling established that supplementing mice with irradiated bacteria did not induce significant changes in the biochemical composition of the tissues, as also indicated by the negative values of the Q2 parameter obtained for the statistical model generated. In the duodenum, establishment of L. paracasei increased the relative level of tissue choline, but decreased the amounts of serine and glutamine (Figure 4). In the jejunum and ileum, relative levels of glutamate, glutamine, glutathione, methionine, acetate, alanine, glycine, creatine, choline, and GPC were decreased with the probiotic supplementation (Figure 4). In addition, the probiotic gavage was correlated with relative decreased levels of cysteine and an unknown compound in the ileal tissues and relative increased concentrations of lactate in the jejunum (Figure 4).
Discussion The aims of the present study were to define a metabolic profile for each intestinal tissue from female germ-free mice and to evaluate the impacts of L. paracasei on intestinal tissue physiology and host metabolism. The metabolic variations observed between tissues are discussed in relation to functional biochemical properties of the gut wall pertaining to digestion, contractility and protection from oxidative stress. Journal of Proteome Research • Vol. 6, No. 4, 2007 1475
research articles
Martin et al.
Figure 2. O-PLS-DA cross-validated scores plot using five classes (A) and differential metabogram coefficient plots (B, C, D, E, F) obtained by analysis of 5 tissues from each gut compartment including duodenum (b, black), jejunum (9, dark red), ileum ((, blue), proximal colon (2, red), distal colon (1, green). The model was generated with 5 predictive and 1 orthogonal components and was characterized by Q2 ) 0.61 and R2 ) 0.91. The discrimination was obtained by comparing each individual tissue with all others. Positive O-PLS coefficients (+) indicate higher levels of a metabolite and negative values (-) lower concentrations of a metabolite in the specific tissue. The color code corresponds to the correlation between the spectral data points and the discriminant axis between classes. Keys to metabolite assignment are given in Table 1.
Digestive Functions. A major source of the intestinal metabolites comes from further processing of dietary nutrients by intestinal cells after digestion and absorption, and this occurs mainly in the first quarter of the small intestine 1476
Journal of Proteome Research • Vol. 6, No. 4, 2007
(duodenum and jejunum).3 The importance of these nutrients in the tissue metabolic profiles were consistent with previous reports29 (Figures 2 and 3). For instance, the final digestion of proteins and lipids that produces absorbable products takes
L. Paracasei Treatment on Gut Tissue
research articles
Table 2. Summary of O-PLS-DA Models for the Different Discriminations among 1H MAS CPMG NMR Spectra of Intestinal Tissuesa
a O-PLS-DA models were generated with 1 predictive component and 2 orthogonal components to discriminate between 2 groups of mice. The R2 value shows how much of the variation in the dataset X is explained by the model. The Q2 value represents the predictability of the models and relates to its statistical validity.
place among the microvilli, and lipid-soluble components are able to penetrate in the epithelial membranes and diffuse into the mucosal cells.29 The digestion of alimentary phospholipids produces a variety of intestinal lipids, free fatty acids, phosphoethanolamine, and GPE.30 Furthermore, dietary phosphatidylcholine is hydrolyzed in the intestinal lumen to 1-acylglyceryl-phosphorylcholine, which on entering the mucosal cell is mainly converted to phosphatidylcholine and GPC.31 The observation of relative elevated concentrations of lipids, phosphoethanolamine, GPE, choline, phosphorylcholine, and GPC in the duodenum and jejunum from germ-free mice is consistent with these reports (Figures 2 and 3). In particular, the jejunum contained relative high levels of phosphorylcholine and lipids when compared to the duodenum and ileum (Figure 3). This is consistent with previous studies showing that jejunum is primary location for digested fat storage.32,33 Also, the essential role of choline, phosphorylcholine, and GPC in cell metabolism, signaling processes and membrane constitution, as well as the osmolyte function of taurine and betaine in the intestine have been described previously.34-36 In addition, choline, GPC, and phosphorylcholine also play a crucial role in lipid cholesterol transport and metabolism.34 Hence, their relative high concentrations in the overall composition of the intestinal tissues are consistent with their known functions. Furthermore, relative higher level of taurine was observed in the ileum (Figure 3) when compared to other compartments. The ileum is known to actively reabsorb 90-95% of all the luminal tauroconjugated bile acids.37 Similarly to jejunum, colon is characterized by its lipid biosynthetic ability,32,38-40 which is consistent with its relative higher lipid content when compared to the small intestine (Figure 3). Moreover, the distal colon of germ-free mice also showed relative lower levels of propionate, acetate, formate, oxaloacetate, and cytosine when compared to the proximal colon (Figure 3). Acetate and propionate are a significant carbon source for the synthesis of lipids, and in particular for phospholipids, whereas formate and oxaloacetate are intermediates in glyceroneogenesis to synthesize triacylglycerols.38 In germ-free models, it has been suggested that the relative reduced concentrations of these short chain fatty acids and
Figure 3. Differential metabogram coefficient plots showing the discrimination between adjacent intestinal compartments derived from O-PLS-DA of 1H MAS CPMG NMR spectra of gut tissues. (A) Duodenum vs jejunum, (B) jejunum vs ileum, (C) ileum vs proximal colon, and (D) proximal colon vs distal colon. The vertical axis is the discriminant axis between the two tissues. Positive O-PLS coefficients (+) indicate higher levels of a metabolite in the first tissue whereas negative values (-) correspond to higher concentrations of a metabolite in the other tissue. Keys to metabolite assignment is given in Table 1.
cytosine are related to requirements for intestinal lipid biosynthesis.38,39 Unlike jejunum, the colon does not transport or absorb lipids efficiently.41 Both the proximal and distal colons of germ-free mice showed relative lower levels of choline, GPC and phosphorylcholine as compared to the small intestine (Figures 2 and 3) consistent with reduced lipid transport as suggested by previous workers.34,42 Recent studies established that animals with a normal microbiota have a 30% greater “calorific bioavailability” than their germ-free counterparts.3 This is due to the conversion of many dietary substances into absorbable nutrients by microbial metabolism.3 The investigation of the metabolic profiles from germ-free mice re-colonized with L. paracasei revealed a relative elevation of choline in the duodenum and lactate in the jejunum (Figure 4). Lactate is a major product of Lactobacillus metabolism,43 and data suggest an extended microbialinduced processing of dietary nutrients, including bacterial fermentation, leading to higher absorption of lactate.43 Recently, Journal of Proteome Research • Vol. 6, No. 4, 2007 1477
research articles
Martin et al.
Table 3. Key Observed Metabolic Differences between Sequential Intestinal Compartments correlation coefficients (rankingc) metabolite (keya)
δ (ppm)b
acetate (9) acetoacetate (46) acetyl-glutamate (41) alanine (6) arginine (7) asparagine (16) aspartic acid (15) betaine (47) choline (19) creatine (17) cysteine (43) cytosine (37) unknown (40) ethanol (4) ethanolamine (18) formate (34) glutamate (12) glutamine (14) glutathione (24) glycine (23) GPC (21) GPE (45) isoleucine (1) lactate (5) leucine (2) lipids (25) lysine (8) methionine (13) myoinositol (30) oxaloacetate (29) phenylalanine (28) phosphorylcholine (20) phosphorylethanolamine (44) proline (11) propionate (39) scyllo-inositol (31) serine (42) taurine (22) threonine (10) TMAO (26) tyrosine (27) uracil (35) valine (3)
1.91(s) 2.29(s) 1.95(s) 1.47(d) 1.63(m) 2.94(m) 2.68(m) 3.90(s) 3.2(s) 3.03(s 3.10(m), 5.92(d) 4.41(s) 1.18(t) 3.15 (t) 8.45 (s) 2.34(m) 2.44(m) 2.55 (m) 3.55(s) 3.22(s) 4.12 (m) 1.02(d) 4.11(q) 0.91(d) 1.27(m) 1.72(m) 2.13(s) 4.06(t) 2.38 (s) 7.40(m) 3.22(s) 4.00 (m) 2.00(m), 1.06 (t) 3.35 (s) 3.96(m) 3.40(t) 4.25(m) 3.26(s) 6.87(dd) 5.78(d) 1.04(d)
duodenum vs jejunum
jejunum vs ileum
-0.73 (5) 0.82 (7) 0.84 (6) 0.84 (6) 0.85 (5)
ileum vs proximal colon
-0.73 (6)
-0.79 (10) 0.78 (11)
0.85 (1) 0.76 (4)
0.82 (7) 0.88 (2) -0.74 (13) 0.68 (16) 0.82 (7) 0.81 (8) 0.87 (3)
-0.71 (5) 0.76 (3) 0.67 (7)
-0.65 (9) 0.67 (17) 0.8 (9)
0.58 (12) -0.7 (6) 0.66 (8)
-0.71 (6) -0.76 (4) -0.77 (3) -0.83 (2)
proximal vs distal colon
0.79 (3) 0.63 (11)
-0.84 (1) -0.7 (7) -0.7 (7) -0.67(8)
0.74 (5) 0.64 (9) 0.68 (7) 0.68 (7) 0.6 (11)
0.88 (2) -0.76 (12) 0.86 (4) 0.87 (3) 0.81 (8) -0.84 (6) -0.7 (15) 0.84 (6) 0.81 (8) -0.72 (14) 0.9 (1)
0.67 (8)
-0.62 (10) -0.77 (2) 0.58 (12) 0.82 (2) -0.66 (8)
-0.64 (10) 0.74 (4) -0.8 (1)
0.57 (13) 0.57 (12)
0.8 (9)
-0.66 (8)
-0.65 (9)
0.9 (1)
a Key numbers correspond to those shown on the 600 MHz spectra in the Figures. b s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; dd, doublet of doublets. c Ranking is computed from the selected variables for each model according to their O-PLS weights.
the gut microbiota have been implicated as an important environmental factor that regulates fat synthesis and storage in the host cells.33 In particular, the jejunum was proposed to be a major site for dietary fat accumulation.32 Interestingly, mice supplemented with L. paracasei showed specific metabolic changes in both the jejunum and the ileum, with relative lower concentrations of choline, GPC and acetate (Figure 4). These metabolites have been described as key intermediates of lipid synthesis,38 suggesting that L. paracasei has the capacity to modulate intestinal fat metabolism. Dietary supplementation with live L. paracasei reduced the levels of glutamine and serine in both the duodenum and jejunum (Figure 4), which were specifically characterized as crucial metabolic fuels44 used by intestinal enterocytes for rapid division,45 repairing of epithelial cells,46 nutrient transport, and protein turnover (Figures 2 and 3).29 Nutrient utilization is known to be less efficient in gnotobiotic as compared to conventional mice.3 Therefore, germ-free intestines generally show changed gut morphology with increased surface area to 1478
Journal of Proteome Research • Vol. 6, No. 4, 2007
compensate for the less efficient nutrient usage.47 These malformations include atypical epithelial structure,47 longer villi, and shorter crypts containing fewer cells.48 As a consequence, energy requirements for cell division and growth are augmented in gnotobiotic intestines as suggested by the relative reduction of glutamine and serine seen in the intestinal tissues from germ-free mice supplemented with live L. paracasei. The results presented here are consistent with the reported contribution of symbiotic bacteria to the amino acid homeostasis and nutritional status of the host.49 Metabolites Related to Muscle Contractility Variation Along the Gut. The intestinal tract is rich in smooth muscles that are essential for the gut movements and the intestinal transit. It has been reported that the gut is capable of synthesizing creatine, a metabolite found in relative high concentrations in muscles and used for the production of ATP (Figure 3).50 Here, relative elevated amounts of creatine are found in the duodenum and colon of germ-free mice when compared to the overall composition of the intestine, which suggests a more
research articles
L. Paracasei Treatment on Gut Tissue
Table 4. Principal Metabolic Effects of L. Paracasei on Intestinal Tissues correlation coefficients
metabolite (keya)
chemical shift δ (ppm)
duodenum monoassociated
jejunum monoassociated
ileum monoassociated
serine (42) choline (19) glutamine (14) acetate (9) alanine (6) glutamate (12) glutathione (24) methionine (13) glycine (23) GPC (21) creatine (17) taurine (22) lactate (5) cysteine (43) unknown (40)
3.96(m) 3.2(s) 2.44(m) 1.91(s) 1.47(d) 2.34(m) 2.55 (m) 2.14 (s) 3.55(s) 3.22(s) 3.92 (s) 3.43 (t) 1.31 (d) 3.15 (t) 4.41(s)
-0.67 0.74 -0.57 -
-0.53 -0.59 -0.58 -0.57 -0.62 -0.59 -0.61 -0.57 -0.57 -0.57 -0.57 0.64 -
-0.59 -0.60 -0.65 -0.57 -0.62 -0.59 -0.57 -0.65 -0.66 -0.67 -0.67 -0.64 -0.59
a
Figure 4. Differential metabogram coefficient plots derived from O-PLS-DA of 1H MAS NMR CPMG spin-echo spectra of gut tissues, illustrating the discrimination between gnotobiotic (-) and live L. paracasei re-colonized tissues (+). (A) Duodenum (Q2 ) 0.56, R2 ) 0.51), (B) jejunum (Q2 ) 0.67, R2 ) 0.80), and (C) ileum (Q2 ) 0.67, R2 ) 0.78). To enhance interpretability, the color scale has been adjusted to plot all the variables with a correlation coefficient higher than 0.57 in red. Keys to metabolite assignments are given in Table 1. O-PLS-DA models were generated with one predictive component and two orthogonal components to discriminate between two groups of mice.
developed muscular morphology in these sections of the intestinal tract. The adjacent position of the duodenum to the stomach, an organ rich in muscular tissues, could explain the relative higher creatine concentrations in the duodenum. The distal colon tissues exhibited relative higher levels of creatine and N-acetyl-glutamate, an intermediate in creatine synthesis.51 These changes were associated with relative increased amount of myo-inositol, which is part of the phosphatidylinositol pathway, as well as higher concentrations of scyllo-inositol (Figures 3 and 4). Previous studies reported the specific muscular metabolism of the circular smooth muscles in the colon to be coupled to the phosphatidylinositol pathway.52 The investigation of the metabolite profiles of jejunum and ileum of L. paracasei supplemented mice suggested perturbed intestinal muscular physiology and motility, as indicated by relative decreased levels of taurine and creatine, which are important component of muscle tissues.20,36 Furthermore, there is evidence that taurine modulates several calcium-dependent mechanisms including muscular contractility, excitability, and tissue osmolality.36,53,54 In addition, relative reduced concentrations of alanine and glutamine suggested fewer interconversions of amino acids to produce energy via Cori cycle.55 These
See Table 3.
results are in agreement with other reports of the crucial role of L. paracasei in regulating hyper-contractility of intestinal muscle20 and microbial importance in modulating gastrointestinal motility.4,56,57 Possible Protective Mechanisms against Gut Inflammation. Glutathione was observed in the mucosal cells, and conventional mice have been previously shown to have relative higher concentrations in the duodenal cells when compared to ileal and colon cells.58 The primary function of glutathione is to protect cells from oxidative stress, and a relation between glutathione concentration and mucosal damage has been established in diverse gastrointestinal diseases.58 Glutathione plays a important anti-oxidant role and may prevent gut disorders because it neutralizes reactive oxygen species that are possible mediators of inflammatory bowel disease.59 Here, the metabolic profile of germ-free intestinal tissues showed that the relative highest concentrations of glutathione were in the duodenum and ileum (Figure 2). When mice were colonized with L. paracasei, relative lower levels of glutathione and its precursors glutamate, methionine, cysteine, and glycine60,61 were observed in the jejunum and ileum (Figure 4). These observations suggest a probioticinduced effect on the γ-glutamyl cycle leading to a modulation of the antioxidant ability of the gut and this might be associated with altered sulfur and cysteine metabolism.62 Additionally, bacterial colonization led to relative decreased levels of taurine and creatine in the small intestine when compared to gnotobiotic controls, which is consistent with previous observations. Other studies showed that microbiota decrease intestinal levels of glutathione, which is in agreement with the work presented here.58 The probiotic-induced ability to modulate antioxidant enzymes in intestinal tissue may be of important interest in understand the underlying mechanism of gastrointestinal cancers.63 Because intestinal amino acid catabolism plays an important role in modulating dietary amino acid availability to extra-intestinal tissues and intestinal synthesis of glutathione, this work suggest that L. paracasei may affect enterocyte glutathione metabolism due to its known antioxidant ability6 and by regulation of the catabolism of dietary nutrients29 Effects of Ingestion of Live and Irradiated Bacteria. An observation requiring further discussion is the absence of live Journal of Proteome Research • Vol. 6, No. 4, 2007 1479
research articles
Martin et al.
bacterial-induced changes in the metabolic profile of the colon. Different gut microbiota were shown to modulate the host intestinal transcriptome in both the small intestine and colon and also the systemic metabolism and immune responses as compared to gnotobiotic controls.2,33,64,65 We have also shown the influences of L. paracasei supplementation to conventional mice infected with Trichinella spiralis and have demonstrated intricate metabolic changes in the jejunal wall and longitudinal myenteric muscular tissues using NMR-based metabonomics.20 The results presented here show intestinal region-dependent metabolic changes triggered by ingestion of live L. paracasei in initially germ-free mice. Distinct Lactobacillus strains have been suggested to colonize the gut differently and hence to generate divergent host responses.8 Colon tissues from animals supplemented with L. paracasei were found to be metabolically similar to those from germ-free animals, even though there were ten times more bacteria in the colon than in the small intestine. Interestingly, transcriptomic data have indicated that gut microbiota modulate limited changes in colon when compared to the small intestine. The reported transcriptional changes are related mainly to the biological functions affecting water absorption.64 Therefore, it might not be surprising that using NMR-based techniques, no metabolic changes in colon tissues were found upon re-colonization of colon with bacteria. However, an elevation in short chain fatty acids, including lactate and acetate was observed in colon flushes of bacterial associated colon (data not shown). These indicate bacterial fermentation and do not reflect colon metabolic status.
of the transgenomic interactions between L. paracasei and the host, at the metabolic level to modulate the gut development.47 We have also demonstrated that 1H MAS NMR spectroscopy produces a stable metabolic signature for each intestinal compartment with respect to their structure/function and has also generated reference profiles with which to compare changes in response to gut microbiotal manipulation. Abbreviations: CFU, colony forming unit; COSY, 1H-1H correlated spectroscopy; CPMG, Carr-Purcell-Meiboom-Gill; GPC, glycerophosphorylcholine; GPE, glycero-phosphorylethanolamine; HRMAS, high resolution magic-angle-spinning; MRS, Man-Rogosa-Sharpe; NMR, nuclear magnetic resonance; OPLS-DA, orthogonal projection to latent structure discriminant analysis; STOCSY, statistical total correlation spectroscopy; TOCSY, 1H-1H total correlation spectroscopy.
Finally, irradiation-killed bacteria did not change the metabolic profiles of the intestinal tissues when compared to the gnotobiotic controls (Table 3). This indicates that the metabolic changes observed with live bacteria are probably due to genuine host-bacterial interactions rather than to the simple digestion of bacterial cells or to purely sensing of bacterial components by the host. Moreover, although no measurement of animal weight and food intake were recorded, it remains possible that the variations of the tissue phenotypes result specifically from probiotics-induced effect on food intake. Ley et al.66 and Turnbaugh et al.67 reported recently that the gut microbiota is a contributing factor to the pathophysiology of obesity by stimulating energy recovery from the diet, while reducing nutritional consumption. Our findings also illustrate the influences of gut microbe variations on mammalian biology and the complex metabolic interactions involving more than one genome, which has been termed “the microbial-mammalian metabolic axis”.12,68
(1) Tannock, G. W. Gastroenterol. Clin. North Am. 2005, 34, 361-82, vii. (2) Xu, J.; Bjursell, M. K.; Himrod, J.; Deng, S.; Carmichael, L. K.; Chaing, H. C.; Hooper, L. V.; Gordon, J. I. Science 2003, 299, 20742076. (3) Hooper, L. V.; Midtvedt, T.; Gordon, J. I. Annu. Rev. Nutr. 2002, 22, 283-307. (4) Husebye, E.; Hellstrom, P. M.; Sundler, F.; Chen, J.; Midtvedt, T. Am. J. Physiol. Gastrointest. Liver Physiol. 2001, 280, G368-G380. (5) Falk, P. G.; Hooper, L. V.; Midtvedt, T.; Gordon, J. I. Microbiol. Mol. Biol. Rev. 1998, 62, 1157-1170. (6) Annuk, H.; Shchepetova, J.; Kullisaar, T.; Songisepp, E.; Zilmer, M.; Mikelsaar, M. J. Appl. Microbiol. 2003, 94, 403-412. (7) Sarker, S. A.; Sultana, S.; Fuchs, G. J.; Alam, N. H.; Azim, T.; Brussow, H.; Hammarstrom, L. Pediatrics 2005, 116, e221-e228. (8) Ibnou-Zekri, N.; Blum, S.; Schiffrin, E. J.; von der, W. T. Infect. Immun. 2003, 71, 428-436. (9) Verdu, E. F.; Bercik, P.; Bergonzelli, G. E.; Huang, X. X.; Blennerhasset, P.; Rochat, F.; Fiaux, M.; Mansourian, R.; CorthesyTheulaz, I.; Collins, S. M. Gastroenterology 2004, 127, 826-837. (10) Prioult, G.; Fliss, I.; Pecquet, S. Clin. Diagn. Lab. Immunol. 2003, 10, 787-792. (11) Xu, J.; Chiang, H. C.; Bjursell, M. K.; Gordon, J. I. Trends Microbiol. 2004, 12, 21-28. (12) Nicholson, J. K.; Holmes, E.; Lindon, J. C.; Wilson, I. D. Nat. Biotechnol. 2004, 22, 1268-1274. (13) Lowe, I. J. Phys. Rev. Lett. 1959, 2, 285-287. (14) Nicholson, J. K.; Wilson, I. D. Nat. Rev. Drug Discov. 2003, 2, 668676. (15) Wang, Y.; Tang, H.; Holmes, E.; Lindon, J. C.; Turini, M. E.; Sprenger, N.; Bergonzelli, G.; Fay, L. B.; Kochhar, S.; Nicholson, J. K. J. Proteome. Res. 2005, 4, 1324-1329. (16) Tugnoli, V.; Mucci, A.; Schenetti, L.; Calabrese, C.; Di, F. G.; Rossi, M. C.; Tosi, M. R. Int. J. Mol. Med. 2004, 14, 1065-1071. (17) Garrod, S.; Bollard, M. E.; Nicholls, A. W.; Connor, S. C.; Connelly, J.; Nicholson, J. K.; Holmes, E. Chem. Res. Toxicol. 2005, 18, 115122. (18) Wang, Y.; Bollard, M. E.; Nicholson, J. K.; Holmes, E. J. Pharm. Biomed. Anal. 2006, 40, 375-381. (19) Bollard, M. E.; Murray, A. J.; Clarke, K.; Nicholson, J. K.; Griffin, J. L. FEBS Lett. 2003, 553, 73-78. (20) Martin, F. P.; Verdu, E. F.; Wang, Y.; Dumas, M. E.; Yap, I. K.; Cloarec, O.; Bergonzelli, G. E.; Corthesy-Theulaz, I.; Kochhar, S.; Holmes, E.; Lindon, J. C.; Collins, S. M.; Nicholson, J. K. J. Proteome. Res. 2006, 5, 2185-2193.
The biochemical components of the duodenum, jejunum, ileum, proximal, and distal colons in gnotobiotic mice could be characterized by the use of nondestructive HRMAS NMR spectroscopy. Chemometrics methods allowed discrimination of the effects of probiotic treatment in gut epithelial metabolism relating to topographic variation in nutrient absorption, propulsion and detoxification functions. This work also demonstrates the potential of metabolic profiling for studying properties of the “microbe-mammalian metabolic axis” and probiotic modulations of symbiont-host biochemistry. Contrary to the effects induced by live L. paracasei, no changes were seen with supplementation of irradiation-killed bacteria. These modifications of the intestinal metabolic profile were shown to be region-dependent and correlated with the intestinal functions as manifested by the effects on digestion, absorption, amino acid homeostasis, fat synthesis, and protection of oxidative stress. The results presented here demonstrate the importance 1480
Journal of Proteome Research • Vol. 6, No. 4, 2007
Acknowledgment. We acknowledge Dr. Olivier Cloarec for the use of O-PLS-DA Matlab scripts. Dr. Ivan K. S. Yap is acknowledged for useful discussion on this manuscript. We also acknowledge the various contributions of Marc Dumas, Isabelle Rochat, Catherine Murset, Gloria Reuteler, Karina Portugal, Massimo Marchesini, and Catherine Schwartz. This work received financial support from Nestle´ to F.-P.J.M. and Y.W. Supporting Information Available: Supplementary Figures 1 and 2. This material is available free of charge via the Internet at http://pubs.acs.org. References
research articles
L. Paracasei Treatment on Gut Tissue (21) Guigoz, Y.; Rochat, F.; Perruisseau-Carrier, G.; Rochat, I.; Schiffrin, E. Nutr. Res. 2002, 13-25. (22) Stejskal, E. O.; Tanner, J. E. J. Chem. Phys. 1965, 42, 288-292. (23) Nagayama, K. J. Magn. Reson. 1980, 40, 321-334. (24) Bax, A.; Davis, D. J. Magn. Reson. 1985, 65, 355-360. (25) Cloarec, O.; Dumas, M. E.; Craig, A.; Barton, R. H.; Trygg, J.; Hudson, J.; Blancher, C.; Gauguier, D.; Lindon, J. C.; Holmes, E.; Nicholson, J. Anal. Chem. 2005, 77, 1282-1289. (26) Cloarec, O.; Dumas, M. E.; Trygg, J.; Craig, A.; Barton, R. H.; Lindon, J. C.; Nicholson, J. K.; Holmes, E. Anal. Chem. 2005, 77, 517-526. (27) Trygg, J.; Wold, S. J. Chemom. 2003, 17, 53-64. (28) Holmes, E.; Cloarec, O.; Nicholson, J. K. J. Proteome. Res. 2006, 5, 1313-1320. (29) Wu, G. J. Nutr. 1998, 128, 1249-1252. (30) Diagne, A.; Mitjavila, S.; Fauvel, J.; Chap, H.; Douste-Blazy, L. Lipids 1987, 22, 33-40. (31) Parthasarathy, S.; Subbaiah, P. V.; Ganguly, J. Biochem. J. 1974, 140, 503-508. (32) Robertson, M. D.; Parkes, M.; Warren, B. F.; Ferguson, D. J.; Jackson, K. G.; Jewell, D. P.; Frayn, K. N. Gut 2003, 52, 834-839. (33) Backhed, F.; Ding, H.; Wang, T.; Hooper, L. V.; Koh, G. Y.; Nagy, A.; Semenkovich, C. F.; Gordon, J. I. Proc. Natl. Acad. Sci. U. S. A 2004, 101, 15718-15723. (34) Zeisel, S. H.; Blusztajn, J. K. Annu. Rev. Nutr. 1994, 14, 269-296. (35) Kettunen, H.; Tiihonen, K.; Peuranen, S.; Saarinen, M. T.; Remus, J. C. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 2001, 130, 759-769. (36) Huxtable, R. J. Physiol. Rev. 1992, 72, 101-163. (37) Floch, M. H. Dig. Liver Dis. 2002, 34 Suppl 2, S54-S57. (38) Zambell, K. L.; Fitch, M. D.; Fleming, S. E. J. Nutr. 2003, 133, 3509-3515. (39) Carter, J. R.; Kennedy, E. P. J. Lipid Res. 1966, 7, 678-683. (40) Marie, M. C.; Fred, G. L.; Lauretta, G.; Satchitanandum; George, V. Dig. Dis. Sci. 1985, V30, 468-476. (41) Levy, E.; Loirdighi, N.; Thibault, L.; Nguyen, T. D.; Labuda, D.; Delvin, E.; Menard, D. Am. J. Physiol 1996, 270, G813-G820. (42) Hearn, J. D.; Lester, R. L.; Dickson, R. C. J. Biol. Chem. 2003, 278, 3679-3686. (43) Tannock, G. W. Appl. Environ. Microbiol. 2004, 70, 3189-3194. (44) Stoll, B.; Henry, J.; Reeds, P. J.; Yu, H.; Jahoor, F.; Burrin, D. G. J. Nutr. 1998, 128, 606-614. (45) Labow, B. I.; Souba, W. W. World J. Surg. 2000, 24, 1503-1513. (46) Luk, G. D.; Bayless, T. M.; Baylin, S. B. J. Clin. Invest 1980, 66, 66-70. (47) Tannock, G. W. Normal Microflora; Chapman and Hall: New York, 1995.
(48) Alam, M.; Midtvedt, T.; Uribe, A. Scand. J. Gastroenterol. 1994, 29, 445-451. (49) Metges, C. C. J. Nutr. 2000, 130, 1857S-1864S. (50) de Jonge, W. J.; Marescau, B.; D’Hooge, R.; de Deyn, P. P.; Hallemeesch, M. M.; Deutz, N. E.; Ruijter, J. M.; Lamers, W. H. J. Nutr. 2001, 131, 2732-2740. (51) Wu, G.; Morris, S. M., Jr. Biochem. J. 1998, 336 ( Pt 1), 1-17. (52) Warner, F. J.; Liu, L.; Lubowski, D. Z.; Burcher, E. Clin. Exp. Pharmacol. Physiol. 2000, 27, 928-933. (53) Huxtable, R. J. Prog. Clin. Biol. Res. 1990, 351, 157-161. (54) Schanne, O. F.; Dumaine, R. J. Pharmacol. Exp. Ther. 1992, 263, 1233-1240. (55) Waterhouse, C.; Keilson, J. J. Clin. Invest. 1969, 48, 2359-2366. (56) Hooper, L. V.; Wong, M. H.; Thelin, A.; Hansson, L.; Falk, P. G.; Gordon, J. I. Science 2001, 291, 881-884. (57) DeSesso, J. M.; Jacobson, C. F. Food Chem. Toxicol. 2001, 39, 209228. (58) Loguercio, C.; Di, P. M. Ital. J. Gastroenterol. Hepatol. 1999, 31, 401-407. (59) Liu, C.; Russell, R. M.; Smith, D. E.; Bronson, R. T.; Milbury, P. E.; Furukawa, S.; Wang, X. D.; Blumberg, J. B. Int. J. Vitam. Nutr. Res. 2004, 74, 74-85. (60) Reeds, P. J.; Burrin, D. G.; Jahoor, F.; Wykes, L.; Henry, J.; Frazer, E. M. Am. J. Physiol. Endocrinol. Metab. 1996, 33, E413-E418. (61) Roth, E.; Oehler, R.; Manhart, N.; Exner, R.; Wessner, B.; Strasser, E.; Spittler, A. Nutrition 2002, 18, 217-221. (62) De La Rosa, J.; Stipanuk, M. H. Comp. Biochem. Physiol. B 1985, 81, 565-571. (63) Drew, J. E.; Arthur, J. R.; Farquharson, A. J.; Russell, W. R.; Morrice, P. C.; Duthie, G. G. Biochem. Pharmacol. 2005, 70, 888-893. (64) Mutch, D. M.; Simmering, R.; Donnicola, D.; Fotopoulos, G.; Holzwarth, J. A.; Williamson, G.; Corthesy-Theulaz, I. Physiol. Genomics 2004, 19, 22-31. (65) Schumann, A.; Nutten, S.; Donnicola, D.; Comelli, E. M.; Mansourian, R.; Cherbut, C.; Corthesy-Theulaz, I.; Garcia-Rodenas, C. Physiol. Genomics 2005, 23, 235-245. (66) Ley, R.; Turnbaugh, P.; Klein, S.; Gordon, J. Nature 2006, 444, 1023-1024. (67) Turnbaugh, P.; Ley, R.; Mahowald, M.; Magrini, V.; Mardis, E.; Gordon, J. Nature 2006, 444, 1027-1031. (68) Nicholson, J. K.; Holmes, E.; Wilson, I. D. Nat. Rev. Microbiol. 2005, 3, 431-438.
PR060596A
Journal of Proteome Research • Vol. 6, No. 4, 2007 1481
