Metabolomic Analysis Identifies Inflammatory and Noninflammatory

Publication Date (Web): February 9, 2010. Copyright © 2010 American Chemical Society ... The New Zealand Institute for Plant & Food Research Limited...
0 downloads 0 Views 1MB Size
Download PDF
Metabolomic Analysis Identifies Inflammatory and Noninflammatory Metabolic Effects of Genetic Modification in a Mouse Model of Crohn’s Disease Hui-Ming Lin,†,‡,| Matthew P.G. Barnett,§,| Nicole C. Roy,§,| Nigel I. Joyce,‡,| Shuotun Zhu,†,| Kelly Armstrong,§,| Nuala A. Helsby,† Lynnette R. Ferguson,†,| and Daryl D. Rowan*,‡,| School of Medical Sciences, University of Auckland, New Zealand, The New Zealand Institute for Plant & Food Research Limited, New Zealand, AgResearch, New Zealand, and Nutrigenomics, New Zealand Received December 9, 2009

Interleukin-10 is an immunosuppressive cytokine involved in the regulation of gastrointestinal mucosal immunity toward intestinal microbiota. Interleukin-10-deficient (IL10-/-) mice develop Crohn’s diseaselike colitis unless raised in germ-free conditions. Previous gas chromatography-mass spectrometry (GC-MS) metabolomic analysis revealed urinary metabolite differences between IL10-/- and wildtype C57BL/6 mice. To determine which of these differences were specifically associated with intestinal inflammation arising from IL10-deficiency, urine samples from IL10-/- and wildtype mice, housed in either conventional or specific pathogen-free conditions, were subjected to GC-MS metabolomic analysis. Fifteen metabolite differences, including fucose, xanthurenic acid, and 5-aminovaleric acid, were associated with intestinal inflammation. Elevated urinary levels of xanthurenic acid in IL10-/mice were attributed to increased production of kynurenine metabolites that may induce T-cell tolerance toward intestinal microbiota. Liquid chromatography-mass spectrometry analysis confirmed that plasma levels of kynurenine and 3-hydroxykynurenine were elevated in IL10-/- mice. Eleven metabolite differences, including glutaric acid, 2-hydroxyglutaric acid, and 2-hydroxyadipic acid, were unaffected by the severity of inflammation. These metabolite differences may be associated with residual genes from the embryonic stem cells of the 129P2 mouse strain that were used to create the IL10-/- mouse, or may indicate novel functions of IL10 unrelated to inflammation. Keywords: metabolite profiling • metabolomic • metabonomic • interleukin-10-deficient mouse • Crohn’s disease • genetic modification • kynurenine

Introduction Interleukin-10 (IL10) is an immunosuppressive cytokine that is expressed by various innate and adaptive immune cells such as macrophages, dendritic cells, and T-cells.1,2 The immunosuppressive functions of IL10 include the suppression of proinflammatory cytokine production by immune cells, downregulation of major histocompatibility complex and co-stimulatory molecules on antigen-presenting cells, and inhibition of T-cell proliferation.1,2 The important role of IL10 in the regulation of intestinal mucosal immunity toward intestinal microbiota is demonstrated by the spontaneous development of chronic intestinal inflammation in interleukin-10-deficient (IL10-/-) mice, unless these mice are reared in germ-free conditions.3,4 Inflammation is less severe when the mice are raised in specific pathogen-free (SPF) conditions.3 * To whom correspondence should be addressed. E-mail: daryl.rowan@ plantandfood.co.nz. Fax: +64-9537701. Address: Plant & Food Research Palmerston North, Private Bag 11600, Palmerston North 4442, New Zealand. † University of Auckland. ‡ The New Zealand Institute for Plant & Food Research Limited. | Nutrigenomics. § AgResearch. 10.1021/pr901130s

 2010 American Chemical Society

Crohn’s disease in humans is an inflammatory bowel disorder attributed to a dysregulated mucosal immune response toward intestinal microbiota.5,6 The exact disease etiology is unknown but involves genetic and environmental factors.5 Despite the role of IL10 in intestinal mucosal immunity, the association between IL10 gene variants and Crohn’s disease is weak, with inconsistent findings by genome-wide association and candidate gene studies.7 However, IL10 is likely to be involved in Crohn’s disease pathogenesis, as peripheral and mucosal immune cells from Crohn’s patients produce lower amounts of IL10 than healthy controls.8-10 Furthermore, the NOD2 mutant protein (3020insC) associated with Crohn’s disease inhibits IL10 transcription,11 suggesting a possible disease mechanism in Crohn’s patients with this mutation. Recently, loss-of-function mutations in subunits of the IL10 receptor were discovered in four patients with severe early onset enterocolitis,12 demonstrating that loss of IL10 signaling in humans can result in inflammatory bowel disease. The IL10-/- mouse displays some physiological and biochemical similarities to Crohn’s disease, such as a Th1-mediated immune response,13 intermittent transmural inflammation lesions,13 and an increased intestinal permeability.14 Therefore, research Journal of Proteome Research 2010, 9, 1965–1975 1965 Published on Web 02/09/2010

research articles -/-

on the IL10 mouse may provide insights into the role of IL10 in Crohn’s disease pathogenesis. The IL10-/- mouse (B6.129P2-Il10tm1Cgn) available from The Jackson Laboratory (Bar Harbor, Maine) is a congenic strain of C57BL/6 background that was created using embryonic stem cells derived from the 129P2 mouse strain. Phenotypic effects due to 129P2-derived genetic material have been reported, and genotyping analyses indicate that as much as 30 centi-Morgan (cM) of the chromosomal region flanking the modified IL10 gene in IL10-/- mice is of 129 strain origin.15,16 Bolivar et al. (2001) reported that IL10-/- and wildtype C57BL/6 mice behaved differently in an open field, with IL10-/- mice displaying less exploratory activity.15 This behavioral trait was mapped to the 129P2-derived chromosomal region.15 Turner et al. (2009) discovered that the resistance of IL10-/- mice to infection by the pathogen Yersinia pestis KIM15 is an inherent trait of 129 mouse substrains, which was mapped to the 129P2-derived chromosomal region in IL10-/- mice.16 These studies demonstrate that phenotype differences between IL10-/- and wildtype mice can arise from C57BL/6 and 129P2 genetic differences, which could be misinterpreted as functions of the IL10 protein. Previously we have identified urinary metabolite differences between IL10-/- and wildtype C57BL/6 mice by nontargeted metabolite profiling,17 also known as metabolomic analysis.18,19 Metabolites with higher levels in the urine of IL10-/- mice compared with wildtype included xanthurenic acid, fucose, and 5-aminovaleric acid, whereas those with decreased levels included glutaric acid, 2-hydroxyglutaric acid, 2-hydroxyadipic acid, hexanoylglycine, cis-aconitic acid, isocitric acid, cytosine, and glucose.17 As intestinal inflammation is less severe in specific pathogen-free (SPF) IL10-/- mice,3 metabolite changes associated with intestinal inflammation should be minimal in SPF IL10-/- mice, whereas metabolite changes associated with noninflammatory effects of genetic modification should be observed as persistent differences between IL10-/- and wildtype mice regardless of the severity of inflammation. We used this rationale in the present study to determine if the urinary metabolite differences between IL10-/- and wildtype mice are associated with intestinal inflammation or other metabolic effects of genetic modification of the IL10-/- mouse.

Methods Mice. Experimental procedures were reviewed and approved by the Crown Research Institute Animal Ethics Committee (Hamilton, New Zealand), according to the New Zealand Animal Welfare Act (1999). IL10-/- (B6.129P2-Il10tm1Cgn, stock no. 2251) and wildtype C57BL/6 mice (stock no. 664) were purchased from The Jackson Laboratory (Bar Harbor, Maine) which certified these mice as specific pathogen-free (SPF). Mice were all male, weaned, and 4.4 weeks old (31 ( 3 days old) upon arrival at the Small Animal Facility of AgResearch, Ruakura (Hamilton, New Zealand). Mice were individually housed either in shoebox-style cages under conventional conditions (conventional mice), or in cages individually ventilated with HEPA-filtered air (Technoplast) under SPF conditions (SPF mice). Room temperature was 22 °C with a 14 h light10 h dark cycle. All mice were fed in-house prepared powdered AIN76A diet that was sterilized by gamma-irradiation (ScheringPlough Animal Health, Wellington, New Zealand). Drinking water consisted of tap water for conventional mice and autoclaved water for SPF mice. SPF mice were handled in a laminar flow biohazard hood. After six days of acclimatization in the facility, conventional mice were orally dosed with 200 1966

Journal of Proteome Research • Vol. 9, No. 4, 2010

Lin et al. µL (equivalent to 1.5 × 10 colony forming units) of a mixture of Enterococcus faecalis, E. faecium, and complex intestinal flora as described by Roy et al. (2007),20 to ensure that all conventional mice received the same microbial exposure, and that conventional IL10-/- mice developed intestinal inflammation. The day of dosing was considered as the first day of the experiment (average age 5.3 weeks old). 9

Spot urine samples were collected from all mice by manual handling as described previously17 between 10 a.m. and 12 noon, when the average ages were 6.3, 8.0, 9.4, and 10.4 weeks old. Half of the conventional mice were sacrificed after three weeks (22nd and 23rd day) of the experiment when the average age was 8.3 weeks old. The remaining mice were sacrificed after 6 weeks (40th to 42nd day) when the average age was 11 weeks old. Mice were subjected to a fasting regime as described previously prior to euthanasia by carbon dioxide asphyxiation with cervical dislocation.17 Blood was removed by cardiac puncture with EDTA as the anticoagulant and centrifuged to obtain plasma. Serum amyloid-A levels in plasma were measured by ELISA (murine serum amyloid A ELISA kit, Tridelta Development Ltd.). Intestinal tissues were dissected, rinsed with saline solution, and immersed in phosphate-buffered formalin for histology analysis. GC-MS Analysis of Urine. Urine samples were analyzed by gas chromatography-mass spectrometry (GC-MS) as described previously,17 but using a HP-5 ms GC column (30 m length, 0.25 mm i.d., 0.25 µm film thickness, Agilent Technologies) with the following modifications. Sample injection was splitless (1 µL volume) and injection temperature was 250 °C. Sampling time was 0.5 min, helium carrier gas flow was 1.2 mL/min, and column inlet pressure was 70 kPa. The temperature gradient of GC separation started with 60 °C for 4 min, increased from 60 to 120 at 10 °C/min, 120 to 200 at 2.5 °C/ min, and 200 to 300 at 10 °C/min, and was held at 300 °C for 8 min. A total of 157 urine samples were analyzed in 10 analytical batches. Samples from the same sampling time points were analyzed within the same or consecutive analytical batches. Within each analytical batch, samples were randomized according to treatment groups. Triplicate quality control samples were included in each batch. The quality control sample was created by pooling 71 urine samples from this experiment. Metabolites were identified by mass spectra matches to previously identified metabolites or GC-MS mass spectral libraries. The identities of metabolites were confirmed by GC-MS analysis of authentic chemical standards (listed in Supporting Information, Table S1) purchased from SigmaAldrich. The exception was for 3,4-dihydroxybutyric acid which was synthesized in-house by hydrolysis of β-hydroxy-γ-butyrolactone (Sigma-Aldrich) with sodium hydroxide.21 Data Preprocessing and Analysis. GC-MS datafiles were preprocessed by XCMS (version 1.6.0), and the collated data set was normalized and subjected to PCA (prcomp package, R software version 2.5.1), multiple t-tests (multtest package, R), and ANOVA (General Linear Model, Minitab version 15.1.0.0) as described previously.17 The “minfrac” parameter for peak detection by XCMS was 0.25. P-value adjustment was not applied as explained previously.17 Instead, a metabolite was considered to be significantly different if all its XCMS peaks were significantly different (P < 0.05) between treatments at most of the urine sampling time points, with confirmation from t-tests and ANOVA of normalized peak areas from the original GC-MS datafiles. Metabolite differences between IL10-/- and wildtype mice in conventional conditions were validated using

Effects of Genetic Modification in Crohn’s Disease Mouse

research articles

Figure 1. Histology scores of colon tissue (A) and serum amyloid-A levels (B) in individual mice. (“n” ) number of mice assessed. The histology scoring method is described in Supporting Information, section S3). Table 1. Number of Spot Urine Samples Collected number of urine samples collected number of mice per groupa

6.3 weeks old

8.0 weeks old

9.4 weeks old

conventional IL10-/wildtype

20 20

18 16

19 16

8 10

7 8

IL10-/wildtype total urine samples

10 10

8 9 51

5 8 48

7 8 33

5 5 25

SPF

10.4 weeks old

a Half of the conventional mice were sacrificed after urine sampling at 8.0 weeks old.

data from previous experiments.17 Metabolites associated with intestinal inflammation should be at significantly different levels between conventional IL10-/- and conventional wildtype mice, as well as between SPF IL10-/- and conventional IL10-/mice due to lower inflammation severity in SPF IL10-/- mice. Such metabolites should not be at significantly different levels between SPF wildtype and conventional wildtype mice in order to rule out SPF effects on a nondisease state. Metabolites unassociated with intestinal inflammation should be at significantly different levels between IL10-/- and wildtype mice regardless of housing conditions. Heat map visualization and hierarchical clustering of normalized peak areas (MultiExperiment

Viewer software, version 4.4)22 were performed with Pearson correlation as the clustering distance and average linkage clustering as the linkage method. All P-values, fold differences, and graphs presented in the results were calculated using normalized peak areas of GC-MS datafiles unless stated otherwise. LC-MS Analysis of Vitamin B6 and Tryptophan Metabolites. A 50 µL aliquot of plasma was mixed with 100 µL of 0.1 g/mL trichloroacetic acid, left on ice for 1 h, and centrifuged at 10956g for 10 min. Then 120 µL of the supernatant was analyzed by liquid chromatography-mass spectrometry (LC-MS) with electrospray ionization in positive mode on a Thermo-Finnigan LC system coupled to a LTQ Thermo-Finnigan ion trap mass spectrometer. The column was an Alltima C18 AQ. A 2 µL aliquot of each plasma extract was separated with a mobile phase of 3% formic acid in water (solvent A) and acetonitrile (solvent B) with a flow rate of 200 µL/min and the following gradient (time ) solvent A/solvent B (v/v)): 0 min ) 100/0, 4 min ) 100/0, 10 min ) 70/30, 13 min ) 45/55, 14 min ) 5/95, and then cycled to start conditions. Analytes were detected by selective reaction monitoring (SRM), with the chromatogram split into three segments containing either three or four SRM scan events. A collision energy of 35 units was applied to selected protonated molecular ions to produce product ions (Supporting Information, Table S2) which were quantitated using aqueous external calibration standards.

Figure 2. PCA score plots of XCMS peak areas (1163 peaks) for all urine samples (Table 1), labeled according to sample type (A) or housing conditions (B): IL10 ) IL10-/- mouse, WT ) wildtype, SPF ) specific pathogen-free, conv ) conventional. Journal of Proteome Research • Vol. 9, No. 4, 2010 1967

research articles

Lin et al.

Results Inflammation Status. Histological examination of intestinal tissues and serum amyloid-A (SAA) levels confirmed that the severity of intestinal inflammation was generally reduced in SPF IL10-/- mice relative to conventional IL10-/- mice (Figure 1). Intestinal inflammation was detected only in the colon and mainly characterized by infiltration of immune cells. Histology scores and SAA levels for 11 week-old SPF IL10-/- mice were, respectively, 1.7 and 11-fold lower than those of 11 week-old conventional IL10-/- mice (2-tail t-test P-values < 0.03). Urinary Metabolite Profiles. Visual examination of GC-MS chromatograms of urine samples did not reveal obvious distinguishing features between IL10-/- and wildtype mice, or between housing conditions (Supporting Information, Figure S4). XCMS peak detection and alignment of GC-MS datafiles of all urine samples (Table 1) and replicate quality control

samples produced a data set of 1701 peaks. Removal of 538 artifacts from trimethylsilyl derivatives and diet contamination of urine samples reduced the number of peaks to 1163. These peaks were considered to represent the urinary metabolite profiles and consisted of 194 groups of peaks with the same retention time (rounded to the nearest 0.1 min). PCA scoreplots of the peak areas for these 1163 peaks showed that replicate quality controls clustered closely together and that urine samples did not cluster according to analytical batches (Supporting Information, Figure S5), indicating no major anomalies during analytical processing. The PCA scoreplots also showed no separation of samples according to sampling time points (Supporting Information, Figure S5) or treatment groups (Figure 2A) on PC1 or PC2. Instead, urine samples clustered according to housing conditions on PC1 (Figure 2B), suggesting that the urinary metabolite profiles were

Table 2. Urinary Metabolite Differences among the Groups of Micea comparisons of urinary levelsc Metaboliteb

IL10 SPF:IL10 conv

WT SPF:WT conv

Associated with Intestinal Inflammation expected profile for metabolite of inflammation Sig diff N.S. xanthurenic acid 2283 407 or 408 v:V v:V fucose 1740 204 v:V v:V unknown RT861 1395 230 v:V N.S. 5-aminovaleric acid 1640 174 v:V N.S. unknown RT1683 1799 260 v:V N.S. uracil 1348 241 v:V N.S. unknown RT910n 1410 61 V:v N.S. unknown RT954n 1435 218 V:v N.S. unknown RT1177n 1552 247 V:v N.S. unknown RT1295n 1609 158 v:V N.S. unknown RT1369 1660 231 v:V N.S. unknown RT1451n 1680 161 v:V V:vd unknown RT1524 1730 188 v:V N.S. unknown RT2081n 1956 172 V:v N.S. unknown RT3242n 3170 174 v:V N.S.

Sig diff V:v V:v V:v V:v V:v V:v v:V v:V v:V V:v V:v V:v V:v v:V V:v

N.S. N.S. N.S. N.S. V:vd N.S. V:ve N.S. N.S. N.S. N.S. V:vd N.S. N.S. v:Vd N.S.

Associated with the IL10-/- Genotype expected profile for metabolite of IL10-/- genotype Sig diff Sig diff glutaric acid 1415 55 or 158 V:v V:v 2-hydroxyglutaric acid 1597 247 V:v V:v 2-hydroxyadipic acid 1697 171 or 261 V:v V:v unknown RT1933 1907 98 V:v V:v 3,4-dihydroxybutyric acid 1450 233 v:V v:V cis-aconitic acid 1776 229 V:v V:v isocitric acid 1862 273 V:v V:v hexanoylglycine 1647 158 or 172 V:v V:v unknown RT1891n 1872 287 V:v V:v 4-hydroxyphenyllactic acid 1923 179 V:v V:v 5-hydroxyindoleacetic acid 2243 290 V:v V:v

N.S. N.S. V:vd V:vd N.S. N.S. N.S. N.S. V:vd V:vd V:ve N.S.

N.S. N.S. N.S. N.S. N.S. N.S. v:Ve v:Ve N.S. N.S. N.S. N.S.

V:v v:V V:v v:V V:v N.S. N.S.

V:v N.S. V:v v:V V:v N.S. N.S.

unknown RT657 succinic acid cytosine glucose unknown RT2800 unknown RT1461n unknown RT1874

R.I.

1240 1321 1534 2037 2410 1684 1885

m/z

172 247 254 191 375 245 292

IL10 conv:WT conv

Complex Responses v:V v:V v:V V:v V:v v:V v:V

IL10 SPF:WT SPF

v:V v:V N.S. V:v v:V N.S. N.S.

a Abbreviations: R.I. ) retention index, m/z ) quantitated ion, IL10 ) IL10-/- mice, WT ) wildtype, conv ) conventional, SPF ) specific pathogen-free, Sig diff ) significant difference, N.S. ) not significantly different, RT ) retention time. b Unknowns are named according to XCMS retention time (seconds). Unknowns marked with ‘n’ suffix were first discovered in this study but confirmed to be present and significantly different in GC-MS data of previous study.17 c Interpretation: v:V for A:B means metabolite levels for A were higher than B and vice versa. All comparisons were significantly different according to ANOVA of all time points (P-value < 0.05) except for those with N.S. XCMS ions, ANOVA, t-test P-values, and fold differences are in Supporting Information, Tables S6-S8. d t-Tests showed significant difference (P-value < 0.05) at one time point only. e t-Tests showed no significant difference (P9-value > 0.05) at any time point.

1968

Journal of Proteome Research • Vol. 9, No. 4, 2010

Effects of Genetic Modification in Crohn’s Disease Mouse

research articles

mainly influenced by the metabolic effects of intestinal microbiota as SPF mice were not orally dosed with bacteria. Metabolite Differences. Metabolite differences between IL10-/- and wildtype mice that are not associated with intestinal inflammation should distinguish the IL10-/- genotype from wildtype irrespective of housing conditions. Thus we refer to these metabolites as associated with the IL10-/- genotype. Statistical comparisons of peak areas among the groups of mice revealed 33 metabolite differences between conventional IL10-/- and wildtype mice, of which 15 were associated with intestinal inflammation, 11 were associated with the IL10-/genotype, and 7 showed complex effects (Table 2). Among these 33 metabolite differences, 12 were not reported in our previous study17 but inspection of the previous GC-MS datafiles confirmed their differences between conventional IL10-/- and wildtype mice. Some metabolites were not detected by XCMS data preprocessing in that previous study, presumably due to differences in chromatogram profiles affecting peak detection and alignment. Likewise, some metabolites detected previously17 were not present in the XCMS data set of the current study but were present in the GC-MS chromatograms. Additionally, some metabolite differences were significantly different only at a few time points in the previous study and thus had not been selected for further examination. Hierarchical clustering analysis using the levels of all the metabolites associated with intestinal inflammation or the IL10-/- genotype for urine samples from conventional mice generally resulted in the clustering of these samples according to genotype (Figure 3). Metabolites Associated with Intestinal Inflammation. Xanthurenic acid and fucose may be considered as the main metabolites associated with the early stages of intestinal inflammation, as their levels were significantly elevated from the first sampling time point for conventional IL10-/- relative to conventional wildtype mice (Figure 4, P-values and fold differences in Supporting Information, Tables S6-S8). Furthermore, their levels ceased to be significantly different between SPF IL10-/- and conventional IL10-/- mice at later sampling time points, consistent with SPF IL10-/- mice slowly developing intestinal inflammation (Figure 4). Xanthurenic acid levels appeared to be the earliest to change, as they were significantly elevated from the first sampling time point for SPF IL10-/relative to SPF wildtype mice, unlike fucose levels which were elevated at later time points (Figure 4). 5-Aminovaleric acid, unknown RT861, and unknown RT1683 may be considered as the main metabolites associated with the later stages of intestinal inflammation, due to their significantly different levels and consistently large fold differences at later time points between conventional IL10-/- and wildtype mice (Figure 4, P-values and fold differences in Supporting Information, Tables S6-S8). Furthermore, the levels of these metabolites were not significantly different between SPF IL10-/and conventional IL10-/- mice at any of the time points unlike the levels of fucose and xanthurenic acid. The levels of the other 10 metabolites associated with intestinal inflammation (Table 2) were also different at much later time points but more variable, thus representing minor metabolite differences at later stages of inflammation (Table 2, P-values and fold differences in Supporting Information, Tables S6-S8). Metabolites Associated with the IL10-/- Genotype. Glutaric acid, 2-hydroxyglutaric acid, and 2-hydroxyadipic acid may be considered as the main metabolites associated with the IL10-/genotype, as indicated by their statistically significant different

Figure 3. Hierarchical clustering and heat map of the levels of differential metabolites in 69 urine samples from conventional IL10-/- and wildtype mice. (IL10 ) IL10-/- mice, WT ) wildtype mice, tp ) time point, m/z ) quantitated ion, TMS ) trimethylsilyl derivative).

levels and consistently large fold differences between IL10-/and wildtype mice at all of the time points (Figure 5, P-values and fold differences in Supporting Information, Tables S6-S8). The levels of these metabolites were decreased in IL10-/- mice relative to wildtype, irrespective of the housing conditions. Metabolites with Complex Responses. The levels of some metabolites such as succinic acid, cytosine, and glucose were significantly different between IL10-/- and wildtype mice, as well as between conventional and SPF conditions for each genotype (Table 2, P-values and fold differences in Supporting Information, Tables S6-S8). Others such as unknown RT1461n and RT1874 were significantly different between conventional IL10-/- and wildtype mice only (Table 2). Thus it was not possible to conclude if the levels of these metabolites were associated with intestinal inflammation or the IL10-/- genotype. Vitamin B6 Status and Tryptophan Metabolites. The plasma levels of 11 metabolites associated with vitamin B6 status and Journal of Proteome Research • Vol. 9, No. 4, 2010 1969

research articles

Lin et al.

Figure 4. Individual (left) and average levels (right) of main metabolites associated with intestinal inflammation. Metabolite levels are relative to the mean of quality control samples: m/z ) quantitated ions, TMS ) trimethylsilyl derivative, WT ) wildtype, IL10 ) IL10-/mice, QC ) quality control, conv ) conventional, SPF ) specific pathogen-free.

tryptophan metabolism were measured by LC-MS to determine the biological significance of the higher urinary excretion of xanthurenic acid by IL10-/- mice. Among the three treatment groups (housed in conventional conditions until 8.3 weeks old, housed in conventional conditions until 11 weeks old, housed in SPF conditions until 11 weeks old), the levels of tryptophan, kynurenine, and 3-hydroxykynurenine were significantly different between IL10-/- and wildtype mice of at least two groups, whereas the others were only significantly different in one group or none (Table 3). Tryptophan levels were lower in 8.3 and 11 week-old 1970

Journal of Proteome Research • Vol. 9, No. 4, 2010

conventional IL10-/- mice relative to wildtype (Figure 6). Kynurenine levels were significantly higher in IL10-/- mice of all treatment groups relative to wildtype, whereas 3-hydroxykynurenine levels were significantly higher in 11 week-old conventional and SPF IL10-/- mice than wildtype (Figure 6). Plasma levels of xanthurenic acid were detected only for 56% of the mice, and were not significantly different between IL10-/- and wildtype mice. The plasma levels of kynurenine correlated well with urinary levels of xanthurenic acid (Pearson correlation coefficient, R ) 0.76, P-values < 0.001, Supporting Information, Figure S11).

research articles

Effects of Genetic Modification in Crohn’s Disease Mouse

Figure 5. Individual (left) and average levels (right) of main metabolites associated with the IL10-/- genotype. Metabolite levels are relative to the mean of quality control samples; m/z ) quantitated ions, TMS ) trimethylsilyl derivative, WT ) wildtype, IL10 ) IL10-/mice, QC ) quality control, conv ) conventional, SPF ) specific pathogen-free. Table 3. Fold Differences and P-Values from Comparisons of Plasma Levels between IL10-/- and Wildtype Mice for Metabolites Associated with Vitamin B6 Status and Tryptophan Metabolisma fold difference IL10-/-/wildtype group

housing/age (weeks):

pyridoxal-5′-phosphate pyridoxal 4-pyridoxic acid tryptophan kynurenine kynurenic acid 3-hydroxykynurenine xanthurenic acid 3-hydroxyanthranilic acid quinolinic acid

conv/8.3

conv/11

1.0 1/1.1

1.2 1.1

1/1.3 1.8 1/1.7 1.4 2.4 1/2.5 1/1.5

1/1.3 2.0 1/1.8 1.7 1/1.6 1/1.4 1/2.0

SPF/11

t-test P-values IL10-/- versus wildtype conv/8.3

1.3 0.801 1.1 0.265 levels below detection 1.0 0.052 1.5 0.000 1.6 0.015 1.7 0.061 1/1.6 0.099 2.6 0.083 1.0 0.052

conv/11

SPF/11

0.146 0.628

0.046 0.514

0.018 0.000 0.299 0.006 0.721 0.298 0.073

0.650 0.007 0.207 0.051 0.800 0.042 0.946

a Boldface values are not statistically significant (P > 0.05). Mean concentrations of metabolites per group are listed in Supporting Information, Table S9: conv ) conventional, SPF ) specific pathogen-free.

Discussion Comparisons of and wildtype mice tions revealed that and wildtype mice

-/-

the urinary metabolite profiles of IL10 housed under conventional or SPF condi15 metabolite differences between IL10-/were associated with intestinal inflamma-

tion, and 11 were not influenced by the inflammation status of the mice which was represented by the housing conditions. Xanthurenic acid was associated with the early stages of intestinal inflammation in IL10-/- mice, and is a side product of the tryptophan catabolism pathway (Figure 7). Increased Journal of Proteome Research • Vol. 9, No. 4, 2010 1971

research articles

Lin et al.

Figure 6. Plasma levels of tryptophan, kynurenine, and 3-hydroxykynurenine. (Plots for the other metabolites are in Supporting Information, Figure S10).

Figure 7. Tryptophan metabolism pathway that produces kynurenine metabolites and xanthurenic acid.43

urinary excretion of xanthurenic acid following a tryptophan loadisanindirectmarkerofvitaminB6(pyridoxine)deficiency.23,24 Decreased plasma levels of pyridoxal-5-phosphate have been observed in patients with inflammatory conditions such as arthritis25 and inflammatory bowel diseases,26 suggesting an increased requirement for pyridoxal-5-phosphate as an enzyme cofactor in inflammatory pathways. However, the plasma levels of the B6 vitamers, pyridoxal-5-phosphate, and pyridoxal, in IL10-/- mice were not lower than wildtype, indicating that IL10-/- mice did not have a decreased vitamin B6 status. Presumably the standard AIN76A diet provided adequate vitamin B6 for the mice. Thus, the increased urinary levels of xanthurenic acid in IL10-/- mice did not appear to be related to vitamin B6 deficiency. Microarray analysis of intestinal tissues from our previous study17 showed that the gene expression of indoleamine-2,31972

Journal of Proteome Research • Vol. 9, No. 4, 2010

dioxygenase (IDO), the first enzyme which catabolizes tryptophan (Figure 7), was significantly upregulated by 11-fold in IL10-/- mice relative to wildtype (personal communication, Shelley Edmunds, The University of Auckland). IDO expression is induced by pro-inflammatory stimuli such as bacterial lipopolysaccharide and the cytokines interferon-γ and TNFR.27 The induction of immune tolerance in T-cells is attributed to the action of kynurenine metabolites derived from IDO catabolism of tryptophan (Figure 7) by dendritic cells.27,28 Indeed, the plasma levels of kynurenine and 3-hydroxykynurenine were significantly higher in IL10-/- mice relative to wildtype, whereas tryptophan levels were slightly lower in IL10-/- mice relative to wildtype. The plasma levels of xanthurenic acid were below the limit of detection for approximately half of the mice and were not significantly different between IL10-/- and wildtype mice. Thus, the transformation of plasma kynurenine and 3-hydroxykynurenine into xanthurenic acid may be occurring in the kidneys, resulting in the high urinary levels of xanthurenic acid. Alternatively, the transfer of xanthurenic acid from plasma to urine may be extremely rapid. The metabolism of plasma kynurenine into urinary xanthurenic acid was supported by the high correlation of their levels. Naı¨ve T-cells differentiate into T-regulatory cells that produce IL10 when treated with kynurenine metabolites.29 Thus, the elevated plasma levels of kynurenine and 3-hydroxykynurenine in IL10-/- mice may be involved in the induction of immune tolerance toward intestinal microbiota by inducing T-cells to differentiate into T-regulatory cells that produce IL10. The downstream metabolites 3-hydroxyanthranilic acid and quinolinic acid were not correspondingly higher in IL10-/- mice relative to wildtype, even though these latter metabolites have also been shown to influence T-cell proliferation and differentia-

Effects of Genetic Modification in Crohn’s Disease Mouse

research articles

Figure 8. Chromosome map of mouse Chromosome 1 showing that D2HGDH and ACMSD genes are located within the 129P2-derived chromosomal region flanking the IL10 gene (A). D1Mit19 is one of the markers used to determine the length of the chromosomal region derived from the 129P2 strain.15 D-2-Hydroxyglutaric acid is the substrate of the D2HGDH enzyme, whereas glutaric acid and 2-hydroxyadipic acid are produced downstream of the reaction catalyzed by the ACMSD enzyme in the glutarate pathway (B).

tion.28-30 The disparity in these levels may be a mechanism of regulating the type of immune response. For example, 3-hydroxyanthranilic acid but not the other kynurenine metabolites induced apoptosis in monocyte-derived cells.31 Thus, the synthesis of 3-hydroxyanthranilic acid and quinolinic acid may be suppressed to promote the regulation of T-cells rather than monocytes. Elevated levels of urinary fucose were also associated with the early stages of intestinal inflammation in IL10-/- mice. Fucose is a common sugar component in the carbohydrate chains of glycoproteins with cell recognition and cell signaling fuctions.32 As previously discussed,17 changes in fucosylation of proteins or increased urinary fucose levels have been observed during the acute phase response and in pathological conditions such as liver disease, cancer, and inflammatory bowel disease. Thus, it is not surprising that elevated urinary fucose levels in IL10-/- mice were associated with intestinal inflammation. The elevated fucose levels may be related to leukocyte trafficking, which requires fucosylation of cell adhesion molecules.33 Elevated levels of urinary 5-aminovaleric acid were associated with the later stages of intestinal inflammation in IL10-/mice. 5-Aminovaleric acid is produced from the metabolism of the polyamine cadaverine, which is synthesized by bacteria by decarboxylation of lysine.34 Higher levels of 5-aminovaleric acid in IL10-/- mice may be caused by inflammation-induced tissue damage that increases the levels of lysine for intestinal microbiota, resulting in enhanced synthesis of cadaverine. The increase in urinary levels of 5-aminovaleric acid at later time points would be consistent with increasing tissue damage during the later stages of inflammation. Alternatively, inflammation-induced changes in the intestinal microbiota may be favoring the growth of cadaverine-producing bacteria. Inflammation-induced changes in intestinal microbiota populations have been demonstrated in mice with chemically induced or pathogen-induced intestinal inflammation, and in IL10-/mice.35,36 There are no reported associations of abnormal 5-aminovaleric acid or cadaverine levels with Crohn’s disease,

but differences in the intestinal microbial population have been observed in Crohn’s patients,37 indicating the association of the disease with dysbiosis. Metabolomic analyses of urinary and fecal samples from Crohn’s patients have also revealed differential metabolites that are related to intestinal microbial metabolism.38,39 However, it is unclear if dysbiosis contributes to Crohn’s disease pathogenesis or is a result of the chronic intestinal inflammation. Also, the influence of the altered intestinal microbiota and their metabolites on the disease progression is unknown. The functions of IL10 as an immunosuppressive cytokine are well-characterized.1,2 Thus, IL10-deficiency is expected to produce phenotypes related to the immune response, as demonstrated by the development of intestinal inflammation in the IL10-/- mouse. The discovery of metabolite differences between IL10-/- and wildtype mice that were not associated with intestinal inflammation suggests that these metabolites may be related to noninflammatory effects of genetic modification of the IL10-/- mouse. In particular, the main metabolite differences, glutaric acid, 2-hydroxyglutaric acid, and 2-hydroxyadipic acid, may be associated with residual genes from the embryonic stem cells of the 129P2 mouse strain. The genes for the enzymes associated with the production of these three metabolites, D-2-hydroxyglutarate dehydrogenase (D2HGDH)40 and aminocarboxymuconate semialdehyde decarboxylase (ACMSD),41 are located within the 129P2-derived chromosomal region flanking the IL10 gene (Figure 8). Thus, 129P2 and C57BL/6 differences in the activities of these enzymes may be responsible for IL10-/- and wildtype differences in the levels of these three metabolites. Interestingly, glutaric acid, 2-hydroxyglutaric acid, 2-hydroxyadipic acid, and some of the other metabolites associated with the IL10-/- genotype can be linked to energy metabolism, particularly lipid metabolism. Glutaric acid, 2-hydroxyglutaric acid, and 2-hydroxyadipic acid may be produced from β-oxidation of longer-chain dicarboxylic acids derived from fatty acid ω-oxidation.42 Hexanoylglycine may be produced from the Journal of Proteome Research • Vol. 9, No. 4, 2010 1973

research articles

Lin et al. 43

hexanoyl-CoA intermediate of fatty acid β-oxidation, whereas 3,4-dihydroxybutyric acid may be produced from fatty acid β-oxidation of γ-hydroxybutyric acid (GHB).44 The production of 2-hydroxyglutaric acid may also be linked to GHB metabolism, as GHB and R-ketoglutaric acid are transformed into succinic semialdehyde and 2-hydroxyglutaric acid by D-2-hydroxyglutarate transhydrogenase.45 cis-Aconitic acid and isocitric acid are products of the tricarboxylic acid cycle. The association of several of these metabolites with energy metabolism thus suggests that IL10 may be involved in the regulation of energy metabolism under noninflammatory circumstances.

GC-MS technical assistance. Nutrigenomics New Zealand is a collaboration between The New Zealand Institute for Plant & Food Research Limited, AgResearch Limited, and The University of Auckland. This research was funded by the Foundation for Research, Science and Technology under contract CO6X702.

Current evidence for the involvement of IL10 in energy metabolism includes decreased IL10 levels in patients with Type 2 diabetes or metabolic syndrome,46,47 promotion of insulin sensitivity by IL10 in mice,48,49 antiatherogenic effects of IL10 in mice,50 and increased hepatic steatosis in IL10-/mice fed a high fat diet.51 The metabolic effects of IL10 in some of these situations appear to be related to its immunosuppressive function. For example, prevention of diet-induced insulin resistance may be associated with IL10 reducing the production of pro-inflammatory cytokines that adversely affect insulin signaling and glucose metabolism.49 However, the antiatherogenic effects of IL10 appear to involve nonimmunosuppressive mechanisms, as IL10 enhanced macrophage uptake and efflux of cholesterol by increasing the expression of scavenger receptors and a cholesterol transporter.52 The mechanism for increased hepatic steatosis in IL10-/- mice fed a high fat diet is unknown but did not appear to be related to inflammation.51

References

Various cytokines are known to regulate normal physiological sleep by affecting neuromodulatory systems. TNFR or IL1 treatment increases nonrapid eye movement (NREM) sleep,53,54 whereas IL10 treatment reduces NREM sleep, presumably by inhibiting TNFR and IL1 production.55,56 IL10-/- mice spend more time in the slow wave sleep (SWS) stage of NREM sleep compared to wildtype mice during the dark phase of the dark-light cycle.57 Among the metabolites associated with the IL10-/- genotype, 3,4-dihydroxybutyric acid and 5-hydroxyindoleacetic acid may be related to the metabolism of neurotransmitters from neuromodulatory systems associated with sleep regulation. 3,4-Dihydroxybutyric acid is produced from GHB, which is a product of γ-aminobutyric acid (GABA) metabolism,58 whereas 5-hydroxyindoleacetic acid is produced from serotonin metabolism.43 GABA signaling is associated with promotion of sleep,53 whereas serotonin signaling is associated with promotion of wakefulness.54,59 Thus, the elevated levels of 3,4-dihydroxybutyric acid and decreased levels of 5-hydroxyindoleacetic acid in urine of IL10-/- mice may be associated with increased GABA signaling and decreased serotonin signaling, which are consistent with increased NREM sleep in IL10-/- mice. In conclusion, this study highlights the potential of metabolomic analysis to distinguish between metabolic effects of genetic modification that are related to the function of the ablated gene and those that are not. The discovery of metabolites associated with intestinal inflammation in IL10-/- mice provides new insights into the role of IL10 in the regulation of the intestinal mucosal immune response. The discovery of metabolites that are not associated with intestinal inflammation indicates genetic modification effects that are unrelated to the function of the ablated gene and provides new insights into functions of IL10 that are unrelated to inflammation.

Acknowledgment. We thank Ric Broadhurst for assistance with the mouse experiment and Martin Hunt for 1974

Journal of Proteome Research • Vol. 9, No. 4, 2010

Supporting Information Available: Supporting Figures and Tables. This material is available free of charge via the Internet at http://pubs.acs.org.

(1) Murray, P. J. Understanding and exploiting the endogenous interleukin-10/STAT3-mediated anti-inflammatory response. Curr. Opin. Pharmacol. 2006, 6, 379–86. (2) Maynard, C. L.; Weaver, C. T. Diversity in the contribution of interleukin-10 to T-cell-mediated immune regulation. Immunol. Rev. 2008, 226, 219–33. (3) Kuhn, R.; Lohler, J.; Rennick, D.; Rajewsky, K.; Muller, W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 1993, 75, 263–74. (4) Sellon, R. K.; Tonkonogy, S.; Schultz, M.; Dieleman, L. A.; Grenther, W.; Balish, E.; Rennick, D. M.; Sartor, R. B. Resident enteric bacteria are necessary for development of spontaneous colitis and immune system activation in interleukin-10-deficient mice. Infect. Immun. 1998, 66, 5224–31. (5) Baumgart, D. C.; Carding, S. R. Inflammatory bowel disease: Cause and immunobiology. Lancet 2007, 369, 1627–40. (6) Baumgart, D. C.; Sandborn, W. J. Inflammatory bowel disease: Clinical aspects and established and evolving therapies. Lancet 2007, 369, 1641–57. (7) Amre, D. K.; Mack, D. R.; Morgan, K.; Israel, D.; Lambrette, P.; Costea, I.; Krupoves, A.; Fegury, H.; Dong, J.; Grimard, G.; Deslandres, C.; Levy, E.; Seidman, E. G. Interleukin 10 (IL-10) gene variants and susceptibility for paediatric onset Crohn’s disease. Aliment Pharmacol. Ther. 2009, 29, 1025–31. (8) Correa, I.; Veny, M.; Esteller, M.; Pique, J. M.; Yague, J.; Panes, J.; Salas, A. Defective IL-10 production in severe phenotypes of Crohn’s disease. J. Leukoc. Biol. 2009, 85, 896–903. (9) Hvas, C. L.; Kelsen, J.; Agnholt, J.; Hollsberg, P.; Tvede, M.; Moller, J. K.; Dahlerup, J. F. Crohn’s disease intestinal CD4+ T cells have impaired interleukin-10 production which is not restored by probiotic bacteria. Scand. J. Gastroenterol. 2007, 42, 592–601. (10) Netea, M. G.; Kullberg, B. J.; de Jong, D. J.; Franke, B.; Sprong, T.; Naber, T. H.; Drenth, J. P.; Van der Meer, J. W. NOD2 mediates anti-inflammatory signals induced by TLR2 ligands: Implications for Crohn’s disease. Eur. J. Immunol. 2004, 34, 2052–9. (11) Noguchi, E.; Homma, Y.; Kang, X.; Netea, M. G.; Ma, X. A Crohn’s disease-associated NOD2 mutation suppresses transcription of human IL10 by inhibiting activity of the nuclear ribonucleoprotein hnRNP-A1. Nat. Immunol. 2009, 10, 471–9. (12) Glocker, E. O.; Kotlarz, D.; Boztug, K.; Gertz, E. M.; Schaffer, A. A.; Noyan, F.; Perro, M.; Diestelhorst, J.; Allroth, A.; Murugan, D.; Hatscher, N.; Pfeifer, D.; Sykora, K. W.; Sauer, M.; Kreipe, H.; Lacher, M.; Nustede, R.; Woellner, C.; Baumann, U.; Salzer, U.; Koletzko, S.; Shah, N.; Segal, A. W.; Sauerbrey, A.; Buderus, S.; Snapper, S. B.; Grimbacher, B.; Klein, C. Inflammatory bowel disease and mutations affecting the interleukin-10 receptor. N. Engl. J. Med. 2009, 361, 2033–45. (13) Berg, D. J.; Davidson, N.; Kuhn, R.; Muller, W.; Menon, S.; Holland, G.; Thompson-Snipes, L.; Leach, M. W.; Rennick, D. Enterocolitis and colon cancer in interleukin-10-deficient mice are associated with aberrant cytokine production and CD4(+) TH1-like responses. J. Clin. Invest. 1996, 98, 1010–20. (14) Madsen, K. L.; Malfair, D.; Gray, D.; Doyle, J. S.; Jewell, L. D.; Fedorak, R. N. Interleukin-10 gene-deficient mice develop a primary intestinal permeability defect in response to enteric microflora. Inflammatory Bowel Dis. 1999, 5, 262–70. (15) Bolivar, V. J.; Cook, M. N.; Flaherty, L. Mapping of quantitative trait loci with knockout/congenic strains. Genome Res. 2001, 11, 1549–52. (16) Turner, J. K.; Xu, J. L.; Tapping, R. I. Substrains of 129 mice are resistant to Yersinia pestis KIM5: Implications for interleukin-10deficient mice. Infect. Immun. 2009, 77, 367–73. (17) Lin, H. M.; Edmunds, S. J.; Helsby, N. A.; Ferguson, L. R.; Rowan, D. D. Nontargeted urinary metabolite profiling of a mouse model of Crohn’s disease. J. Proteome Res. 2009, 8, 2045–2057.

research articles

Effects of Genetic Modification in Crohn’s Disease Mouse (18) Fiehn, O. MetabolomicssThe link between genotypes and phenotypes. Plant Mol. Biol. 2002, 48 (1-2), 155–71. (19) Nicholson, J. K.; Lindon, J. C.; Holmes, E. Metabonomics: Understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica 1999, 29, 1181–9. (20) Roy, N.; Barnett, M.; Knoch, B.; Dommels, Y.; McNabb, W. Nutrigenomics applied to an animal model of inflammatory bowel diseases: Transcriptomic analysis of the effects of eicosapentaenoic acid- and arachidonic acid-enriched diets. Mutat. Res. 2007, 622, 103–16. (21) Shinka, T.; Inoue, Y.; Ohse, M.; Ito, A.; Ohfu, M.; Hirose, S.; Kuhara, T. Rapid and sensitive detection of urinary 4-hydroxybutyric acid and its related compounds by gas chromatography-mass spectrometry in a patient with succinic semialdehyde dehydrogenase deficiency. J. Chromatogr. B 2002, 776, 57–63. (22) Saeed, A. I.; Sharov, V.; White, J.; Li, J.; Liang, W.; Bhagabati, N.; Braisted, J.; Klapa, M.; Currier, T.; Thiagarajan, M.; Sturn, A.; Snuffin, M.; Rezantsev, A.; Popov, D.; Ryltsov, A.; Kostukovich, E.; Borisovsky, I.; Liu, Z.; Vinsavich, A.; Trush, V.; Quackenbush, J. TM4: A free, open-source system for microarray data management and analysis. Biotechniques 2003, 34, 374–8. (23) Leklem, J. E. Vitamin B-6: a status report. J. Nutr. 1990, 120, 1503–7. (24) Ogasawara, N.; Hagino, Y.; Kotake, Y. Kynurenine-transaminase, kynureninase and the increase of xanthurenic acid excretion. J. Biochem. 1962, 52, 162–6. (25) Roubenoff, R.; Roubenoff, R. A.; Selhub, J.; Nadeau, M. R.; Cannon, J. G.; Freeman, L. M.; Dinarello, C. A.; Rosenberg, I. H. Abnormal vitamin B6 status in rheumatoid cachexia. Association with spontaneous tumor necrosis factor alpha production and markers of inflammation. Arthritis Rheum. 1995, 38, 105–9. (26) Saibeni, S.; Cattaneo, M.; Vecchi, M.; Zighetti, M. L.; Lecchi, A.; Lombardi, R.; Meucci, G.; Spina, L.; de Franchis, R. Low vitamin B(6) plasma levels, a risk factor for thrombosis, in inflammatory bowel disease: Role of inflammation and correlation with acute phase reactants. Am. J. Gastroenterol. 2003, 98, 112–7. (27) Mellor, A. L.; Munn, D. H. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat. Rev. Immunol. 2004, 4, 762–74. (28) Moffett, J. R.; Namboodiri, M. A. Tryptophan and the immune response. Immunol. Cell Biol. 2003, 81 (4), 247–65. (29) Fallarino, F.; Grohmann, U.; You, S.; McGrath, B. C.; Cavener, D. R.; Vacca, C.; Orabona, C.; Bianchi, R.; Belladonna, M. L.; Volpi, C.; Santamaria, P.; Fioretti, M. C.; Puccetti, P. The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor ζ-chain and induce a regulatory phenotype in naive T cells. J. Immunol. 2006, 176, 6752–61. (30) Weber, W. P.; Feder-Mengus, C.; Chiarugi, A.; Rosenthal, R.; Reschner, A.; Schumacher, R.; Zajac, P.; Misteli, H.; Frey, D. M.; Oertli, D.; Heberer, M.; Spagnoli, G. C. Differential effects of the tryptophan metabolite 3-hydroxyanthranilic acid on the proliferation of human CD8+ T cells induced by TCR triggering or homeostatic cytokines. Eur. J. Immunol. 2006, 36, 296–304. (31) Morita, T.; Saito, K.; Takemura, M.; Maekawa, N.; Fujigaki, S.; Fujii, H.; Wada, H.; Takeuchi, S.; Noma, A.; Seishima, M. 3-Hydroxyanthranilic acid, an L-tryptophan metabolite, induces apoptosis in monocyte-derived cells stimulated by interferon-gamma. Ann. Clin. Biochem. 2001, 38, 242–51. (32) Becker, D. J.; Lowe, J. B. Fucose: Biosynthesis and biological function in mammals. Glycobiology 2003, 13, 41R–53R. (33) Maly, P.; Thall, A.; Petryniak, B.; Rogers, C. E.; Smith, P. L.; Marks, R. M.; Kelly, R. J.; Gersten, K. M.; Cheng, G.; Saunders, T. L.; Camper, S. A.; Camphausen, R. T.; Sullivan, F. X.; Isogai, Y.; Hindsgaul, O.; von Andrian, U. H.; Lowe, J. B. The R(1,3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L-, E-, and P-selectin ligand biosynthesis. Cell 1996, 86, 643–53. (34) Osborne, D. L.; Seidel, E. R. Gastrointestinal luminal polyamines: Cellular accumulation and enterohepatic circulation. Am. J. Physiol. 1990, 258, G576–84. (35) Bibiloni, R.; Simon, M. A.; Albright, C.; Sartor, B.; Tannock, G. W. Analysis of the large bowel microbiota of colitic mice using PCR/ DGGE. Lett. Appl. Microbiol. 2005, 41, 45–51. (36) Lupp, C.; Robertson, M. L.; Wickham, M. E.; Sekirov, I.; Champion, O. L.; Gaynor, E. C.; Finlay, B. B. Host-mediated inflammation disrupts the intestinal microbiota and promotes the overgrowth of enterobacteriaceae. Cell Host Microbe 2007, 2, 119–29. (37) Sokol, H.; Lay, C.; Seksik, P.; Tannock, G. W. Analysis of bacterial bowel communities of IBD patients: What has it revealed. Inflamm. Bowel Dis. 2008, 14, 858–67.

(38) Jansson, J.; Willing, B.; Lucio, M.; Fekete, A.; Dicksved, J.; Halfvarson, J.; Tysk, C.; Schmitt-Kopplin, P. Metabolomics reveals metabolic biomarkers of Crohn’s disease. PLoS One 2009, 4, e6386. (39) Williams, H. R.; Cox, I. J.; Walker, D. G.; North, B. V.; Patel, V. M.; Marshall, S. E.; Jewell, D. P.; Ghosh, S.; Thomas, H. J.; Teare, J. P.; Jakobovits, S.; Zeki, S.; Welsh, K. I.; Taylor-Robinson, S. D.; Orchard, T. R. Characterization of inflammatory bowel disease with urinary metabolic profiling. Am. J. Gastroenterol. 2009, 104, 1435–44. (40) Achouri, Y.; Noel, G.; Vertommen, D.; Rider, M. H.; Veiga-DaCunha, M.; Van Schaftingen, E. Identification of a dehydrogenase acting on D-2-hydroxyglutarate. Biochem. J. 2004, 381, 35–42. (41) Fukuoka, S.; Ishiguro, K.; Yanagihara, K.; Tanabe, A.; Egashira, Y.; Sanada, H.; Shibata, K. Identification and expression of a cDNA encoding human R-amino-β-carboxymuconate-ε-semialdehyde decarboxylase (ACMSD). A key enzyme for the tryptophan-niacine pathway and “quinolinate hypothesis”. J. Biol. Chem. 2002, 277, 35162–7. (42) Yu, S.; Rao, S.; Reddy, J. K. Peroxisome proliferator-activated receptors, fatty acid oxidation, steatohepatitis and hepatocarcinogenesis. Curr. Mol. Med. 2003, 3, 561–72. (43) Kyoto Encylopedia of Genes & Genomes. (http://www.genome.jp/ kegg). (44) Lee, C. R. Evidence for the beta-oxidation of orally administered 4-hydroxybutyrate in humans. Biochem Med 1977, 17, 284–91. (45) Struys, E. A.; Verhoeven, N. M.; Jansen, E. E.; Ten Brink, H. J.; Gupta, M.; Burlingame, T. G.; Quang, L. S.; Maher, T.; Rinaldo, P.; Snead, O. C.; Goodwin, A. K.; Weerts, E. M.; Brown, P. R.; Murphy, T. C.; Picklo, M. J.; Jakobs, C.; Gibson, K. M. Metabolism of γ-hydroxybutyrate to d-2-hydroxyglutarate in mammals: Further evidence for d-2-hydroxyglutarate transhydrogenase. Metabolism 2006, 55, 353–8. (46) van Exel, E.; Gussekloo, J.; de Craen, A. J.; Frolich, M.; BootsmaVan Der Wiel, A.; Westendorp, R. G. Low production capacity of interleukin-10 associates with the metabolic syndrome and type 2 diabetes: The Leiden 85-Plus Study. Diabetes 2002, 51, 1088–92. (47) Esposito, K.; Pontillo, A.; Giugliano, F.; Giugliano, G.; Marfella, R.; Nicoletti, G.; Giugliano, D. Association of low interleukin-10 levels with the metabolic syndrome in obese women. J. Clin. Endocrinol. Metab. 2003, 88, 1055–8. (48) Kim, H. J.; Higashimori, T.; Park, S. Y.; Choi, H.; Dong, J.; Kim, Y. J.; Noh, H. L.; Cho, Y. R.; Cline, G.; Kim, Y. B.; Kim, J. K. Differential effects of interleukin-6 and -10 on skeletal muscle and liver insulin action in vivo. Diabetes 2004, 53, 1060–7. (49) Hong, E. G.; Ko, H. J.; Cho, Y. R.; Kim, H. J.; Ma, Z.; Yu, T. Y.; Friedline, R. H.; Kurt-Jones, E.; Finberg, R.; Fischer, M. A.; Granger, E. L.; Norbury, C. C.; Hauschka, S. D.; Philbrick, W. M.; Lee, C. G.; Elias, J. A.; Kim, J. K. Interleukin-10 prevents diet-induced insulin resistance by attenuating macrophage and cytokine response in skeletal muscle. Diabetes 2009, 58, 2525–35. (50) Kleemann, R.; Zadelaar, S.; Kooistra, T. Cytokines and atherosclerosis: A comprehensive review of studies in mice. Cardiovasc. Res. 2008, 79, 360–76. (51) den Boer, M. A.; Voshol, P. J.; Schroder-van der Elst, J. P.; Korsheninnikova, E.; Ouwens, D. M.; Kuipers, F.; Havekes, L. M.; Romijn, J. A. Endogenous interleukin-10 protects against hepatic steatosis but does not improve insulin sensitivity during high-fat feeding in mice. Endocrinology 2006, 147, 4553–8. (52) Han, X.; Kitamoto, S.; Lian, Q.; Boisvert, W. A. Interleukin-10 facilitates both cholesterol uptake and efflux in macrophages. J. Biol. Chem. 2009, 284, 32950–8. (53) Krueger, J. M.; Obal, F. J.; Fang, J.; Kubota, T.; Taishi, P. The role of cytokines in physiological sleep regulation. Ann. N.Y. Acad. Sci. 2001, 933, 211–21. (54) Imeri, L.; Opp, M. R. How (and why) the immune system makes us sleep. Nat. Rev. Neurosci. 2009, 10, 199–210. (55) Kushikata, T.; Fang, J.; Krueger, J. M. Interleukin-10 inhibits spontaneous sleep in rabbits. J Interferon Cytokine Res. 1999, 19, 1025–30. (56) Opp, M. R.; Smith, E. M.; Hughes, T. K., Jr. Interleukin-10 (cytokine synthesis inhibitory factor) acts in the central nervous system of rats to reduce sleep. J. Neuroimmunol. 1995, 60, 165–8. (57) Toth, L. A.; Opp, M. R. Cytokine- and microbially induced sleep responses of interleukin-10 deficient mice. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2001, 280, R1806–14. (58) Wong, C. G.; Bottiglieri, T.; Snead, O. C. GABA, γ-hydroxybutyric acid, and neurological disease. Ann. Neurol. 2003, 54, S3–12. (59) Monti, J. M.; Jantos, H. The roles of dopamine and serotonin, and of their receptors, in regulating sleep and waking. Prog. Brain Res. 2008, 172, 625–46.

PR901130S Journal of Proteome Research • Vol. 9, No. 4, 2010 1975