当前位置：文档库 › microbiome–host metabolic signal

microbiome–host metabolic signal

Microbes and Metabolism

Understanding the role of gut microbiome–host metabolic signal disruption in health and disease

Elaine Holmes,Jia V.Li,Thanos Athanasiou,Hutan Ashra?an and Jeremy K.Nicholson

Department of Surgery and Cancer,Imperial College London,London,UK

There is growing awareness of the importance of the gut microbiome in health and disease,and recognition that the microbe to host metabolic signalling is crucial to understanding the mechanistic basis of their interaction. This opens new avenues of research for advancing knowledge on the aetiopathologic consequences of dys-biosis with potential for identifying novel microbially-related drug targets.Advances in both sequencing tech-nologies and metabolic pro?ling platforms,coupled with mathematical integration approaches,herald a new era in characterizing the role of the microbiome in metabolic signalling within the host and have far reaching implications in promoting health in both the developed and developing world.

Implications of host–microbiome metabolic interactions in health and disease

The symbiosis between the human host and gut microbiota can trigger speci?c biological responses both locally and systemically.Of the55bacterial divisions,only two are prominent in mammalian gut microbiota(Bacteroidetes and Firmicutes),and it is,therefore,likely that host eu-karyotic organisms have co-evolved with gut microbiota in the context of this symbiosis to achieve a physiological homeostasis[1].Individual bacterial species present unique pathological effects and,similarly,shifts in gut bacterial colonies can also prompt speci?c disease-inducing activity(dysbiosis)or disease-protective activity(probio-sis).This balance between the bacterial host defence rein-forcement and proin?ammatory pathology was?rst proposed by the1908Nobel laureate Elie Metchnikoff who noted the therapeutic bene?ts of acid-producing lactic bacilli[2].The bene?cial effects of gut microbiota include: (i)immune-cell development and homeostasis(Th1vs.Th2 and Th17),(ii)food digestion,(iii)supporting fat metabo-lism,(iv)epithelial homeostasis,(v)enteric nerve regula-tion and(vi)promoting angiogenesis.Conversely, maladapted microbial ecology can impair many of these homeostatic and physiological signals so that they result in a number of disease states that include allergy,in?amma-tory bowel disease(IBD),obesity,cancer and diabetes.We describe the relationship of the microbiome with speci?c physiological and pathological states and explore the meta-bolome as a window for monitoring the activity of the gut microbiome via microbial–mammalian cometabolism. Moreover,we identify potential avenues for exploitation of this host–microbial metabolic axis with respect to im-proving human health.

Metabolic pro?ling studies,mainly adopting mass spec-trometric(MS)and NMR spectroscopic platforms(Box1)to measure the metabolic composition of biological samples, have been instrumental in characterizing a wide variety of diseases and have been used for biomarker screening, elucidating mechanistic information relating to disease aetiology and in monitoring responses to therapeutic inter-ventions[3,4].Subtle changes in metabolic pro?les in response to physiological perturbations or environmental stimuli have also been described[5,6].In many such studies,the panel of‘diagnostic’biomarkers has included several metabolites of gut microbial or microbial–mamma-lian cometabolic origin.To probe the exact nature of the metabolic handshake between the mammalian host and the resident gut microbiota,several animal models of simpli?ed microbiota have been developed including germ-free animals,antibiotic-treated rodent models (resulting in a temporary knockout effect of selected bac-terial groups)and animals with transplanted microbiota, such as the Schaedler microbiota(consisting of eight bac-teria including Escherichia coli var.mutabilis,Streptococ-cus faecalis,Lactobacillus acidophilus,Lactobacillus salivarius,group N Streptococcus,Bacteroides distasonis, a Clostridium sp.and an extremely oxygen-sensitive fusi-form bacterium)[7]or human infant microbiota.The con-sequences of these microbial modi?cations on the host metabolism are summarized in Box2with a brief discus-sion of their advantages and limitations as a model for understanding the host–microbiome functional relation-ship.

Given the proximity of the gut microbiota with the intestine,it is unsurprising that the microbiome is impli-cated in intestinal diseases.However,the reach of the gut microbiota has been shown to extend far beyond local effects to remote organ systems such as the brain and encompasses more processes than in?ammation.Dysbiosis

Review

Corresponding author:Nicholson,J.K.(j.nicholson@https://www.wendangku.net/doc/bf2933845.html,)

of the gut microbiota has been implicated in several mod-ern epidemics in the Western world,the most notable being IBD,certain cancers,heart disease and metabolic syn-drome and associated risk factors such as obesity and hypertension(Table1).We summarize the role of the microbiome in the aetiology and development of several diseases and examine concurrent changes in the metabo-lome(Table2).Furthermore,we explore the potential for exploiting this host–microbial relationship with respect to developing new therapeutic interventions.IBD

IBD represents two disorders of chronic intestinal in?am-mation,ulcerative colitis(UC)and Crohn’s disease(CD), both of which are associated with a complex genetic sus-ceptibility together with clear evidence for the involvement of environmental triggers.Evidence implicating the role of microbiota in IBD was initially derived from observations that some antibiotics improved the disease course of patients with this disorder whereas several animal models of IBD required bacterial colonisation for in?ammation to

https://www.wendangku.net/doc/bf2933845.html,mon spectroscopic techniques for metabolic profiling

Spectroscopic technologies commonly used for the metabolic profiling of biological fluids and tissues can be used in either an untargeted mode allowing for screening of the global metabolic composition of metabolites without the necessity of preselecting a set of analytes,or alternatively,can be tailored to measure selected classes of metabolite such as lipids or bile acids.No single analytical method can measure the total set of metabolites present in a biological sample owing to analytical limitations and,therefore,the choice of analytical platform is driven by prior knowledge of the nature of metabolic disruption, platform availability,cost,robustness and biological matrix.

Mass spectrometry(MS):metabolites in complex mixtures can be measured using MS-based methods,which discriminate molecules according to their mass to charge(m/z)ratio.For the purpose of metabolic profiling of biofluids,MS methods are generally prefaced with a separation technology such as gas chromatography(GC-MS), high performance or ultra performance liquid chromatography (HPLC-MS or UPLC-MS)or capillary electrophoresis(CE-MS).These platforms tend to exhibit high sensitivity and have been widely applied to targeted metabolite classes(e.g.lipids,bile acids and SCFAs)in the context of characterizing disease.Metabolite identifica-tion is achieved by use of existing databases of chromatographic retention times and m/z values and by ion fragmentation patterns. NMR spectroscopy:exploits the property of spin that nuclei possess.These nuclei can exist in discrete orientations that relate to discrete energy states when placed in a magnetic field.Spectra are measured following perturbation of the system with radiofrequency pulses and are expressed in the frequency domain,which carries information on the molecular structure of chemical constituents based on their chemical shift with respect to a reference standard, signal intensities and splitting patterns.1H is typically the choice of nucleus for high throughput metabolic screening.NMR is an extremely robust technology and requires little or no sample preparation.For a comprehensive review of the strengths and limitations of these profiling methods,see[91].

Mathematical modelling and data integration:spectra obtained from biological samples are highly multivariate and complex with various degrees of signal overlap.Interpretation of spectra with respect to identifying patterns of metabolites relating to a given physiological or pathological condition is typically achieved using data reduction and multivariate statistical analysis methods.The general aim is to classify or predict objects by identifying inherent patterns in a set of indirect measurements and to relate these classifications to the metabolites or signals that strongly weigh the classification in order to identify candidate biomarkers.These methods operate by compressing the variation in a data set into a smaller number of components based on latent variables and can operate either with or without the incorporation of classification information in the model.As with the profiling methods themselves, each pattern recognition technique has strengths and limitations,and the choice of method is made based on considerations such as the priority of classification over biomarker identification,sensitivity of method,input data and number of missing values.It is generally appropriate to use more than one method for the purpose of validation and maximizing the extraction of https://www.wendangku.net/doc/bf2933845.html,mon methods include principal components analysis(PCA),partial least squares (PLS),PLS-discriminant analysis(PLS-DA),clustering algorithms,self-organizing maps,neural networks and genetic algorithms.Detailed descriptions of multivariate methods can be found in[92].

Typically,for ascertaining the covariation of mammalian metabo-lites with microbial composition and identifying potential biological associations between specific bacterial species or families and metabolic profiles of biofluids or tissues,relatively simple correlation methods have been adopted.Outputs of microbial composition of faeces including DGGE,FISH and454sequencing data have variously been used to generate correlation matrices with urinary,faecal,serum or intestinal tissue profiles[46,93]although bidimensional partial least squares algorithms and other more complex methods can be used to integrate two disparate-omic datasets[94].

Box2.Metabolic signatures of microbially-modulated animal models

The metabolic phenotypes of several classic models of microbial modulation have been characterized using NMR and MS methods: Germ-free:several strains of rat and mice have been profiled.NMR studies have addressed differences in urinary metabolic signatures between conventional mice(C3H/HeJ)and their germ-free counter-parts(e.g.3-hydroxylcinnamic acid,4-hydroxypropionic acid,hippu-rate and phenylacetylglucine),liver(glutathione,glycine,hypotaurine and trimethylamine N-oxide),intestine(alanine,aspartate,glutamate, lactate,taurine-conjugated bile acids,tyrosine and glycine)and kidney(betaine,choline,ethanolamine,inosine and myo-inositol) [95].Another study on Sprague Dawley germ-free rats has shown the different urinary levels of2-oxoglutarate,formate,trimethylamine N-oxide,hippurate,4-hydroxypropionic acid,3-hydroxypropionic acid and plasma levels of betaine,glucose and lactate compared with conventional rats[96].A targeted UPLC-MS method for profiling bile acids has shown that the relative proportion of taurine-conjugated bile acids in the liver,kidney,heart and plasma of germ-free rats is markedly higher than in conventional animals[97].

Germ-free conventionalization:to understand the metabolic trajec-tory in acclimatized germ-free rats,male Fischer344germ-free rats were examined and found to have increased urinary levels of phenylacetylglycine after6h following introduction to a standard laboratory environment.3-Hydroxypropionic acid was observed at day12and increased up to day17.The urinary metabolic profile was within the control range by day21,suggesting the establishment of a stable gut microbiota[98].

Germ-free colonized with simplified microbiota:NMR and UPLC-MS profiling have also been applied to investigate the global metabolic signatures of plasma,urine,faecal extracts,liver tissues and ileal flushes from humanized microbiome mice treated with probiotics(https://www.wendangku.net/doc/bf2933845.html,ctobacillus paracasei or Lactobacillus rhamnosus)[99]and prebiotics(e.g.galactosyl-oligosaccharide) [100].

Antibiotic treatment:NMR urine profiles of antibiotic treated rodents show a similarity with germ-free rat models with lower concentrations of phenolics and other microbial metabolites includ-ing hippurate,phenylacetylglycine,trimethylamine,dimethylamine and higher oligosaccharides and choline.Particularly,hippurate level shifted back to control level within19days post a single vancomycin dose in female NMRI mice[101].

350

occur[8].When compared with controls,patients with IBD demonstrate a reduced diversity of Firmicutes and Bacter-oidetes,and mucous-invading bacteria such as E.coli have been associated with disease-speci?c activity[9].Genetic studies con?rm a role for the underlying microbial–host interaction in IBD pathogenesis.These include the geno-mic regions of nucleotide-binding oligomerisation domain containing2(NOD2),which is an intracellular receptor that recognizes proteins found in bacterial cell wall species. Patients with CD demonstrate an association with the NOD2gene including three polymorphisms that weaken the host peptidoglycan response.Carriers of NOD2have a 1.75–4-fold increased risk of CD and are clinically more likely to undergo surgical gastrointestinal(GI)resection. The mechanisms linking NOD2require further investiga-tion;however,the microbial–host interaction is likely to be modulated by an underlying cytokine environment,and disease development occurs only when these animals are colonized by normal gut bacteria[10].In a study of interleukin-10(IL10)-de?cient mice,which spontaneously develop colitis,gas chromatography(GC)-MS analysis identi?ed a shift towards a higher plasma low-density lipoprotein(LDL)to very low-density lipoprotein(VLDL) ratio and higher plasma levels of lactate,pyruvate and citrate with lower levels of glucose consistent with in-creased fatty acid oxidation and glycolysis[11].Chemical-ly-induced murine models of colitis have also shown systematic differentiation in plasma and tissue pro?les from matched control animals consisting of altered succi-nate,indole-3-acetate,glutamate and glutamine,which tracked the developmental phases of colitis and its severity [12].Metabolic pro?ling studies of IBD in humans have shown systematic differentiation of CD and UC based on urinary and faecal metabolite pro?les.Although reduced concentrations of butyrate,acetate,methylamine and tri-methylamine and increased excretion of amino acids sug-gestive of malabsorption were characteristic of both conditions,the extent of metabolic perturbation was greater

Table1.Examples of disease models with modulated gut microbiota

Disease Animal model Experimental design Mechanism Refs Obesity Mice Conventionally reared vs.germ-free Increased food consumption[69] Obesity Mice Conventionally reared vs.germ-free Increased gut monosaccharide absorption

and induction of hepatic lipogenesis

[69]

Obesity Mice FIAF genetic knockouts Fat storage(increased lipoprotein

lipase activity)

[69]

Obesity Mice(genetic obese

leptin-de?cient ob/ob)Obese ob/ob vs.lean ob/+and ob+/+Genes encoding enzymes that breakdown

otherwise indigestible polysaccharides

[70]

Obesity Mice(genetic obese

leptin-de?cient ob/ob)

Obese ob/ob vs.lean ob/+and ob+/+Increased fermentation[70]

Obesity Mice Obese ob/ob vs.lean ob/+,ob+/+and

ob/+mothers Decreased Bacteroidetes,increased

Firmicutes

[50,51]

Obesity Mice Colonization of germ-free mice Increased energy extraction by bacterial

interactions

[71]

Obesity and diabetes Mice Non-mutant vs.CD14mutant mice Bacterial LPS from Gram-negative bacteria

triggers in?ammation in response to

a high fat diet,which results in metabolic

syndrome

[61]

Diabetes Mice NOD(non-obese diabetic)

MyD88knockout mice

vs.non-mutant mice A dysfunction in the microbial responsive

immune system can lead to autoimmunity

and diabetes

[41]

Diabetes Rats Biobreeding diabetes resistant

vs.biobreeding diabetes prone Administration of Lactobacillus johnsonii

isolated from biobreeding diabetes

resistant delays the onset of type

1diabetes in biobreeding diabetes prone

[72]

IBD Humans IBD patients Reduced diversity of Firmicutes and

Bacteroidetes

[9]

IBD Mice IL10-de?cient mutants

vs.non-mutants Gut in?ammation associated with

IL10de?ciency occurs in the presence

of normal gut bacteria

[10]

Cancer Mice Tgfb1-null mice vs.non-mutants GI cancer develops in the context of normal

gut bacteria with a lack of TGFB1

[20]

Cancer Mice Germ-free IL10-de?cient mice

vs.germ-free non-mutants Colitis-associated colorectal cancer can be

reversed by colonisation with normal gut microbiota

[21]

Cancer Mice Apc Min vs.non-mutants GI polyps only occur in a germ-free environment[22]

Cancer Mice Helicobacter-free C3H/HeN

mice vs.C57BL/6FL-N/35mice Enteric microbiota de?ne hepatocellular

carcinoma risk in mice exposed to carcinogenic

chemicals or hepatitis virus transgenes

[73]

Cardiovascular dyslipidaemia Mice Normal chow-fed mice

vs.high fat-fed mice

vs.high fat-oligofructose-fed mice

Endotoxaemia signi?cantly and negatively

correlates with Bi?dobacterium spp.

[59]

Cardiovascular dyslipidaemia Hamsters Grain sorghum lipid

extract-fed hamsters

Bi?dobacteria increased in grain sorghum

lipid extract-fed hamsters and positively

associated with HDL plasma cholesterol

[74]

351

in the CD group and was accompanied by increased faecal glycerol,which was not seen in the UC group[13].The differential urinary signature of IBD from CD and UC includes alteration of hippurate,4-cresyl sulfate and for-mate,all potential metabolites of gut microbial activity[14].It has recently been shown via genome-wide association studies(GWAS)that there are overlapped genetic signa-tures of CD and UC,and furthermore,that IBD is geneti-cally linked to mutations associated with a number of non-GI diseases[15].Given that these GI conditions are

Table2.Examples of metabolites associated with microbial metabolism or microbial–host cometabolism and illustrations of associations with disease

Metabolite class Sample type Metabolites Origin of metabolites and examples of association with

disease

Refs

SCFAs Plasma,

faeces,

tissues Acetate,butyrate,

propionate

Microbes ferment substances presented in the large

intestine,including indigestible oligosaccharides,dietary

plant polysaccharides or?bre,non-digested proteins and

intestinal mucin,and produce SCFAs.

[75]

Polyamines Urine,

plasma,

faeces,

tissues Putrescine,

cadaverine

Protein putrefaction by a range of anaerobic

microorganisms can result in amines such as putrescine

and cadaverine,can exert adverse impact including

genotoxicity on the host.These polyamines have been

previously found to increase resistance of Neisseria

gonorrhoeae to mediators of the innate human host

defence.Putrescine has also been proposed as a urinary

marker for diabetes in a diet-induced mouse model of

diabetes.

[76,77]

Methylamines

and products of choline degradation Urine,

plasma,

faeces,

tissues

Methylamine,

dimethylamine,

dimethylglycine,

trimethylamine,

trimethylamine

N-oxide

Dietary choline can be metabolized into methylamines by

the gut microbes and has been shown to be modulated in

dietary-induced models of obesity,diabetes and

cardiovascular disease.

[77,78]

Benzoates Urine,

plasma Benzoic acid(plasma),

hippurate(urine),

2-hydroxyhippurate

Decreased concentrations of urinary hippurate have been

consistently reported as characteristic of obesity in both

animal models and humans.2-Hydroxyhippurate was

shown to be positively correlated with colorectal cancer in

a clinical study whereas high urinary levels of hippurate

and dimethylamine were characteristic of diabetes in a rat

model.

[79–81]

Chlorogenic acids Urine,

plasma Dihydroferulic acid,

dihydroferulic acid-3-O-

sulfate,ferulic acid-4-O-

sulfate,dihydroferulic

acid glucuronide,

feruloylglycine

Chlorogenic acids are known antioxidants and have been

bene?cially linked to health,and have been shown to

lower blood pressure in a human cohort with essential

hypertension.

[82]

Protein putrefaction products

Tyrosine Urine,

plasma,

faeces 4-cresyl sulfate

4-cresyl

glucuronide

Decreased plasma concentrations of indoxylsulfate and

indoleacetate were key discriminators in a rat model of

depression.Indoxyl sulfate and phenylacetylglutamine

have been found in higher concentrations in the plasma of

diabetic individuals compared to non-diabetics.Abnormal

urinary excretion of phenylacetylglutamine,hippurate and

hydroxyhippurates has been reported in autistic children.

[63,83–85]

Tryptophan Urine,

plasma,

faeces Indoleactylacetate, indoleactylglycine, indolelactate,

3-hydroxyindole, indoxyl sulfate

Phenylalanine Urine,

plasma,

faeces Phenylacetlyglycine, phenylacetylglutamine

Bile acids Urine,

plasma,

faeces Cholic acid,hyocholic

acid,deoxycholic acid,

chenodeoxycholic acid,

hyodeoxycholic acid,

ursodeoxycholic acid,

glycocholic acid,

glycodeoxylcholic acid,

glycochenodeoxycholic

acid,taurocholic acid,

taurohyocholic acid,

taurodeoxylcholic acid,

taurochenoxycholic acid

Conjugated bile acids are cholesterol derivatives

synthesized in the liver and contain a steroid ring

conjugated with glycine or taurine.Bile acids are well

known to facilitate lipid absorption and signal systemic

endocrine functions to regulate triglyceride,cholesterol,

glucose and energy homeostasis.Gut microbiota have

been shown to modify the bile acid pro?les through a

broad range of reactions such as deconjugation

(hydrolysis of bile salt conjugates to form free bile acids),

dehydroxylation,oxidation(dehydrogenation)and

sulfation,resulting in the formation of secondary and

tertiary bile acids.Deoxycholate has been reported to be

higher in the plasma of diabetics.

[84,86]

352

also associated with gut microbial abnormalities,this begs the question as to whether a signi?cant transgenomic net-work process is in operation linking both microbial and human genes to disease aetiopathogenesis.These studies collectively support a key role for the microbiota in IBDs with contribution to the immunological component of the disease as well as exerting direct metabolic effects.Howev-er,the mechanisms by which these microbial alterations contribute to IBD pathogenesis remain unknown.Thus, speci?c studies probing key functionally-active microbiota are of importance.

Cancer

The association of the gut microbiota with cancer is most commonly observed with GI tumours,as expected,al-though there are examples of these microbiota modifying the cancer risk to other systems such as in breast tumours [16].The bacterium Helicobacter pylori has been proposed to have an aetiological relationship with gastric cancer[17] and has been applied to track the early migration of Homo sapiens suggesting that humans were already infected by H.pylori before their migrations from Africa[18]and have demonstrated a direct gut microbe–host relationship ever since[19].Colonic cancer is the third most common cancer in industrialized countries and is the second leading cause of cancer deaths.Several knockout mice models have elucidated a growing understanding of colonic cancers and gut microbiota.Transforming growth factor beta1 (Tgfb1)-null mice develop colonic cancer in the presence of conventional gut microbiota[20],whereas the germ-free Il10-de?cient mice are resistant to colitis-associated colo-rectal cancer and resistance is lost upon colonization with normal gut microbiota[21].Conversely,tumour-prone Apc Min mice demonstrate only a modest reduction in GI polyps when raised in a germ-free environment[22].These effects of the gut microbiota on cancer development re?ect a complex interplay between the gut genome and the physiological environment of the host and the microbiota. Other examples include colonic cancer models with disor-dered transforming growth factor beta(TGFB)signalling, such as Tgfb1-null,Smad3-null and Rag2-null mice,and predispose the development of cancer with pro-in?amma-tory gut bacterial species such as H.pylori and Helicobacter hepaticus[23,24].

Metabolic pro?ling studies in colon and other cancers have also drawn attention to the role of the gut microbiota in aetiogenesis showing profound modulation of lipid and ste-rol pathways,and changes in faecal amino acids,short-chain fatty acids(SCFAs)and amines[25–27].The underlying mechanisms of shifts in microbial ecology contributing to colonic tumourigenesis include dietary changes and subse-quent shifts in metabolic expression.Speci?cally,there is an epidemiological association between colorectal cancer and raised sul?de production by sulfur-reducing bacteria(SRB) such as Desulfovibrio vulgaris that is typically seen in meat-rich Western diets[28].Here,SRBs compete with methano-genic bacteria for hydrogen to produce hydrogen sul?de. This is supported by evidence that demonstrates that sul?de acts as an oxidation inhibitor of the SCFA n-butyrate in the colonic epithelium.Butyrate is a potent histone deacetylase inhibitor that has epigenic activity in colonocytes and micro-RNA-dependent p21gene expression activity leading to colonic cancer prevention[29]while also offering important epithelial regulatory activities such as ion absorption,mem-brane lipid control,cellular detoxi?cation and mucus for-mation.An association between4-cresol and colon cancer [30]has been reported.The production of4-cresol is depen-dent upon intestinal environmental factors such as the composition of the microbiota,food intake and pH of the intestinal tract[31].4-Cresol is synthesized from tyrosine and phenylalanine via4-hydroxylphenylacetate by gut microbiota.Clostridium dif?cile and certain Lactobacillus strains are known to produce p-cresol by decarboxylation of 4-hydroxyphenylacetate[32,33].Subsequently,4-cresol is excreted in the form of4-cresyl glucuronide and4-cresyl sulfate in urine[34]through glucuronidation and sulfation.

Marked changes in the urinary composition of gut micro-bial metabolites have been found to be characteristic of several other cancers.For example,quinolinate,4-hydro-xybenzoate and gentisate concentrations were higher in kidney cancer patients whereas3-hydroxyphenylacetate and4-hydroxybenzoate were anticorrelated with renal can-cer[35].Urinary hippurate,4-hydroxyphenylacetate and formate,among other metabolites,have been reported to differentiate ovarian cancer from breast cancer patients and healthy controls[36],and urinary hippurate was found to be one of the strongest discriminators of lung cancer[37]and has also been found in relatively high concentrations in

Table2(Continued)

Metabolite class Sample type Metabolites Origin of metabolites and examples of association with

disease

Refs

Lipids Plasma,

faeces Acylglycerols,

sphingomyelin,

cholesterol,

phosphatidylcholines,

phosphoethanolamines,

triglycerides

Gut microbiota regulate lipid synthesis and metabolism

directly through various microbial activities.Undigested

and non-absorbed glycerides in colon may be hydrolysed

into free fatty acids and glycerol by bacterial lipases.

Glycerol can be subsequently converted into

3-hydroxypropanal(3-HPA)and then1,3-propanediol

(1,3-PDO)by a NAD+-dependent oxidoreductase by a wide

range of colonic bacteria including the genera Klebsiella,

Enterobacter,Citrobacter,Clostridum and Lactobacillus

https://www.wendangku.net/doc/bf2933845.html,ctobacilli,particularly Lactobacillus reuteri,

are most ef?cient in accumulating3-HPA.

[87–89]

Organic acids Urine,

plasma,

faeces Lactate,formate High plasma levels of lactate and formate have been found

to be characteristic of several parasitic infections and urinary

formate has also been shown to inversely correlate with

high blood pressure.

[4,90]

353

tumour tissue itself [38].A GC-MS analysis of urine from patients with osteosarcoma also identi?ed that gut micro-bial metabolites,in particular putrescine,hippurate and 4-hydroxyphenylpropionate,comprised part of the metabolic signature of the disease [37].Further evidence of the contri-bution of the microbiota to cancer is the ?nding that one of the most discriminatory metabolites in a study of 50breast cancer patients and matched controls was 4-hydroxypheny-lacetate [39],a microbial product of tyrosine degradation.Adding to the challenge of identifying biomarkers of speci?c cancers and elucidating potential aetiological roles of the gut microbiota is the impact of diet on both the microbiome composition and the disease.For example,the Western diet alters the colonic proportions of bile acids to increase the proportion of primary bile acids transformed to secondary bile acids by intestinal bacteria.Deoxycholic acid (one of the secondary bile acids)is associated with several models of carcinogenesis and correlated with the enzymatic activity of 7a -dehydroxylating bacteria [40].Several species of Clostridium demonstrate high and low 7a -dehydroxylase activity,which could represent a novel target for modifying GI cancer risk.The central role of the gut microbiota in cancer and obesity is illustrated in Figure 1.Therefore,in addition to the direct relationship of particular microorganisms with speci?c cancers,such as H.pylori with gastric cancer,the gut microbiota also contribute indirectly via cholesterol and lipid metabolism with growing evidence that obesity is associated with cancer risk and poorer prognosis.

Diabetes

The global epidemic of obesity is associated with a dramat-ic increase in the prevalence of type 2diabetes mellitus (T2DM).T2DM is characterized by insulin resistance

whereas type 1diabetes mellitus is characterized by a loss of insulin-producing beta cells in the pancreatic islets of Langerhans,which results in insulin de?ciency.The notion that gut microbiota are key players in the onset and development of diabetes is becoming more widely accepted as the evidence base grows.Experimental evidence has shown that the non-obese diabetic (NOD)mouse lacking the innate microbial-recognition immune system receptor MyD88is resistant to type 1diabetes [41].If MyD88knockout mice were depleted of normal intestinal ?ora,type 1diabetes would ensue,whereas if NOD mice were colonized with the altered Schaedler ?ora (ASF),the dia-betes would be attenuated [41].This alluded to the role of gut microbiota and the innate immune receptor MyD88in priming the immune system in the context of type 1diabetes autoimmunity.As type 2diabetes is strongly associated with obesity,these two conditions could be linked by an underlying physiological process including the disordered regulation of gut bacterial pro?les.

From the metabolic pro?ling literature,several studies have reported modulated choline degradation products and bile acids as characteristic of insulin resistance and diabetes in animal models.Studies involving high fat diet-induced initiation of insulin resistance in animal models found alterations in the plasma lipids and in the concen-trations of urinary metabolites such as phenylacetate,hydroxyphenylacetylglycine and hydroxyindoleacetic acid,although host species and strain in?uenced the exact metabolic composition of the bio?uids in these animal models [42–45].Low urinary concentrations of hippurate and higher concentrations of dimethylamine have been found in Zucker obese and Goto-kakizaki rats in compari-son with the respective control strains [43,46].A strepto-zotocin-induced rat model of diabetes reported acetate,

Diabetes, Metabolic Syndrome,Cardiovascular Disease & Cancer

Gut microbiome

Metabolic surgery (BRAVE Effects)

Oxidative stress Lipid peroxidation Inflammatory cytokines Insulin resistance

Free fatty acids

Sex steroids

Sex steroid receptor

Sex hormone binding globulin

Bile acid metabolites

Insulin

Insulin & IGF-1

receptor

Inflammation

Adipokines

Steatosis

Metabolic syndrome

risk factors

OBESITY

TRENDS in Microbiology

Figure 1.Schematic of the gut microbial contribution to obesity and related diseases.

354

ethanol and lactate among the key discriminatory meta-bolites in addition to a modulated plasma lipid pro?le[44]. NMR spectroscopy was applied to characterize the plasma pro?les of congenic strains of rats derived from crossing a diabetic and control animal.Quantitative trait loci were used to generate a correlation matrix with the plasma metabolite pro?les and linkage to benzoate was found to be a consequence of deletion of a uridine diphosphate glucuronosyltransferase[47].The ability of the microbiota to in?uence the expression of diabetes in animal models and the clear impact upon the choline degradation path-way in diabetic animals and man emphasizes the potential of the microbes to contribute to disease aetiology and expression.

Obesity

Obesity has traditionally been considered to be a disorder of energetic and nutritional surplus,which in some cases is associated with a genetic predisposition.Recently,howev-er,the evidence for the role of the gut microbiome has offered new insight into aspects of our understanding of obesity pathogenesis.These include the association of gut microbiota with the following:intestinal permeability, systemic quantity of adipose tissue and body weight.

The initial association between gut bacterial species and weight gain was derived from studies where decreasing dietary?bre intake resulted in excess body weight and diabetes,and was hypothesized to result from a change in gut microbiota as a consequence of altered nutrient supply and digestion.Subsequently,it has been demonstrated that consumption of a high fat diet results in a decrease of total gut bacterial levels and an increase in Gram-negative bacteria.Four bacterial mechanisms have been identi?ed to result in excess bodily energy gain:(i)micro-biota increase energy bioavailability by transforming in-creased proportions of non-digestible food into biochemically absorbable nutrients;(ii)the in?uence of intrinsic bacterial metabolism to generate and raise sys-temic levels of SCFAs to activate triglyceride synthesis; (iii)high fat diets can result in a responsive bacterial metabolism resulting in pathology(such as microbial con-version of choline to methylamines leading to a choline de?cient state,which induces liver disease);and(iv)the ability of the microbiome in regulating gut gene expression to favour an obese state.This could occur through the reduction of lipoprotein lipase activity through the inhibi-tion of angiopoietin-like4(Angptl4)and fasting-induced adipocyte factor(FIAF)to increase free fatty acids and adipose levels[48,49].

Experimental evidence has revealed that germ-free mice are less obese than normal controls but gain weight and have decreased Angptl4expression following colonization by conventional gut bacteria[1].The ob/ob leptin-de?cient obese mouse has a microbiome with a50%reduction in Bacteriodetes and a concurrent increase in Firmicutes, which might be associated with increased food consumption in these animals.Obese humans also demonstrate an alter-ation of the Firmicutes to Bacteroidetes ratio that can be altered by weight loss[50,51].Furthermore,bacterial mod-ulation of obesity is also suggested by the transmission of obesogenic bacterial pro?les in ex-germ-free mice[52].The fact that consistent differences have been noted in the metabolic pro?les of animals and humans following caloric restriction also concurs with the notion that body weight and weight change is associated with alteration in microbial composition.Among these microbial signature changes in-clude increased urinary excretion of hippurate,4-hydroxy-phenylacetic acid,phenylacetylglycine and decreased acetate and lactate[53,54].

Bariatric surgery offers the most consistent method of weight loss in morbidly obese patients,and also achieves the resolution of metabolic dysfunction(including the res-olution of T2DM).Bariatric surgery procedures modify the gut microbiome to achieve a low in?ammatory and weight loss pro?le that might reverse the effects of the metabolic syndrome[55].Work on a bariatric animal model has revealed that these operations work through the BRAVE effects:(i)bile?ow alteration,(ii)reduction of gastric size, (iii)anatomical gut rearrangement and altered?ow of nutrients,(iv)vagal manipulation and(v)enteric gut hormone modulation[56].Furthermore,substantial shifts of the main gut phyla following metabolic surgery(in a rodent animal model)towards higher levels of Proteobac-teria and lower levels of both Firmicutes and Bacteroidetes have been demonstrated as compared to controls[57].The microbial effects of successful weight loss and anti-diabetic operations could lead to novel mechanisms and therapies for obesity and T2DM.In a recent study by Mutch et al.

[58],where weight loss was achieved in obese subjects following Roux-en-Y gastric bypass surgery,marked alterations in the plasma pro?le were found with many classes of metabolites changing post surgery.In addition to lipids,indoles,fatty acids and amino acids,4-cresyl sulfate was characteristic of the obese pro?les[58]and might indicate an altered clostridial pro?le.To date,investigat-ing microbial functions in obesity have been approached via both weight gain(obese patients and murine models) and weight loss(bariatric surgery in human and murine models)with good consensus that body weight in?uences or is in?uenced by the host microbiota.The exact nature of this host–microbial interplay requires further mechanistic investigation.

Cardiovascular dyslipidaemia and metabolic endotoxaemia

Dyslipidaemia is a disorder of lipoprotein metabolism, which results in a systemic increase of disease-inducing blood lipids such as cholesterol,triglycerides and LDL particles and a decrease in the levels of protective lipids such as high-density lipoprotein(HDL)particles.A sus-tained high fat diet can induce dyslipidaemia that results in a proin?ammatory response including an increase of Gram-negative bacterial outer membrane lipopolysaccha-ride(LPS)levels to result in a‘metabolic endotoxaemia’demonstrable in both rodents[59]and humans[60].An infusion of LPS into rodents caused clinical features of metabolic syndrome such as weight gain,insulin resis-tance and hepatic lipid overload.Fat has been demonstrat-ed to be a highly ef?cient transporter of LPS from the gut lumen to the bloodstream and its effects can be delayed in mice lacking the LPS-CD14receptor[61].Non-fatty diets are associated with decreased levels of LPS and lipoprotein

355

particles can buffer its proin?ammatory effects.High fat diets demonstrate a shift in gut bacterial ecology by in-creasing the Gram-negative:Gram-positive ratio whereas an increased?bre intake has been demonstrated to reverse these changes[59]and heralds a possible avenue to pro-mote dietary therapy for the prevention of metabolic endo-toxaemia and atherogenic dislipidaemia. Neuropathology

The composition of the intestinal microbiota is extremely relevant in neurogastroenterology,which deals with the interactions of the central nervous system and the gut (gut–brain axis).Several neuropathological diseases are thought to be associated with the gut microbiota.Autism is a disorder of neural development with impaired social behaviour and often involves GI symptoms.Previous stud-ies examined the faecal microbial pro?les of autistic chil-dren,which indicated10-fold higher numbers of Clostridium https://www.wendangku.net/doc/bf2933845.html,pared with healthy subjects[62]. Many species of Clostridium are known to produce neuro-toxins,which could contribute to the autism spectrum. Metabolic alterations in gut host–microbial cometabolism including higher urinary levels of hippurate and phenyla-cetylglutamine and tryptophan/nicotinic acid metabolism have been observed in autistic children[63].The bile acids have been shown to differ across various neurological conditions.For example,glycocholate(GCA),glycodeoxy-cholate(GDCA)and glycochenodeoxycholate have been shown to be altered in plasma pro?les in Alzheimer’s disease(AD)[64],whereas tauroursodeoxycholic has been shown to modulate p53-mediated apoptosis in AD and to be neuroprotective in Huntington’s disease[64].Clearly,gut microbiota not only exert a local effect on the GI tract but also impact remote organs such as the brain through chemical signalling.Further investigation of the gut–brain axis may provide valuable mechanistic insight into a range of neuropathological and developmental diseases. Concluding remarks

Advances in technology in both metabolic pro?ling and microbial phenotyping methods have improved our ability to derive correlations between the microbial and metabolic phenotypes,and mathematical modelling tools have been developed to accommodate high density data such as those generated by metabonomic and metagenomic methods. New methods of integrating these-omics datasets have been developed to extract correlations between speci?c microbes and metabolites.Several methods ranging from simple correlations to bidirectional partial least squares approaches have been explored(Box1),but more work is needed both on the preprocessing and data modelling components of this integration process.One limitation of the technology at present is that most of the microbial phenotyping tools,such as?uorescence in situ hybridiza-tion(FISH),denaturing gradient gel electrophoresis (DGGE)and454sequencing,map the content of the microbes present without giving any indication of activity.

A much needed breakthrough is to further pro?le the microbiotal metabolism by establishing which microbial products are formed directly from speci?c substrates (originating from the human diet or faecal mucins)using stable-isotope labelling(U-13C glucose)approaches[65].In the disease areas summarized in this review and beyond, there is scope for further elucidation of the role of the gut microbiota and development of potential therapeutic tar-gets based on the underpinning metabolic linkages or even on the microbes themselves.

Although the fact that there is a strong relationship between the mammalian host and its enteric microbiota, which impacts upon health,cannot be disputed,much of the literature is confusing and consensus on the exact mechanisms by which microbes modulate disease process-es has not yet been reached.For example the ratio of Firmicutes:Bacteroidetes has been reported to differ in obesity in several studies,whereas in other studies,no modulation of this ratio has been found with weight gain or loss.This may be in part due to the fact that the Gram-positive and Gram-negative bacteria in the gut are inde-pendent and thus ratios of particular bacterial groups or families may be uninformative.Sequencing analysis at a more re?ned level may provide a better understanding of the structure and function of the microbiota.A recent international study has identi?ed three robust clusters of microbiota(known as enterotypes)that are not nation or continent speci?c.Metagenomic reads from populations at different geographical locations were mapped using DNA sequence homology to1511reference genomes to reveal that intestinal microbiota variation is generally strati?ed,not continuous.This con?rmed the existence of a select series of host–microbial symbiotic relationships, where certain genes are signi?cantly correlated with age and physiological characteristics such as body mass index. These enterotypes can be applied to identify novel bio-markers or targets of disease[66].

To date,most of the research into the effect of microbial metabolism on the host has been focused on diseases that predominate in the Western world.One potential avenue for exploiting the host–micobiome metabolic crosstalk is to use the knowledge of the mechanisms involved in weight gain to promote‘thrifty’bacteria in populations in devel-oping countries where malnutrition is one of the biggest clinical problems.Studies in humans and in animal models of parasitic infection have indicated that a three-way relationship exists between the host and its enteric micro-biota and parasites,with each parasite causing speci?c changes in the gut microbial contribution to the urine, plasma and faecal metabolomes.Therefore,there is clear potential for adapting the analytical strategies and knowl-edge gained from other disease areas and applying them to promote health in developing countries.

Exploration of the gut–brain axis and the role of the microbiota in modifying behaviour is also an exciting but underdeveloped area of research.The possibility of under-standing the metabolic basis of the link between the gut and brain and,moreover,the ability to in?uence this connection is tantalizing.For example,germ-free mice have been shown to have lower neurotrophic factor expres-sion levels and to mount a more severe elevation in corti-costerone levels than speci?c pathogen free(SPF)mice. Moreover,it was shown that this response could be re-versed by colonization with Bi?dobacterium infantis or reconstitution with SPF faeces but potentiated by E.coli

356

[67].The potential for uncovering new bacterial targets or dietary strategies for treating neurological aspects of such diseases is enormous.Some speci?c probiotics such as Lactobacillus farciminis have been demonstrated to have an impact on spinal neuronal activation[68].

As the drive towards research consortia strengthens, multidisciplinary teams with the capacity for combining microbial phenotyping,metabolic pro?ling and clinical expertise become a reality and should serve to develop the current understanding of the metabolic language of mammalian–microbial communication.The bene?ts of attaining this knowledge are clear.The gut microbiome functions as a virtual organ and signi?cantly extends the metabolic capacity of the host.This transgenomic metabo-lism offers a new paradigm for developing novel therapies for many diseases and has potential to bene?cially impact upon a range of acute and chronic pathologies. References

1Backhed,F.et al.(2005)Host–bacterial mutualism in the human intestine.Science307,1915–1920

2Kaufmann,S.H.(2008)Elie Metchnikoff’s and Paul Ehrlich’s impact on infection biology.Microbes Infect.10,1417–1419

3Clayton,T.A.et al.(2006)Pharmaco-metabonomic phenotyping and personalized drug treatment.Nature440,1073–1077

4Holmes, E.et al.(2008)Human metabolic phenotype diversity and its association with diet and blood pressure.Nature453,396–400

5O’Sullivan,A.et al.(2011)Dietary intake patterns are re?ected in metabolomic pro?les:potential role in dietary assessment studies.

Am.J.Clin.Nutr.93,314–321

6Teague,C.R.et al.(2007)Metabonomic studies on the physiological effects of acute and chronic psychological stress in Sprague–Dawley rats.J.Proteome Res.6,2080–2093

7Dewhirst Fe Fau-Chien,C.C.et al.(1999)Phylogeny of the de?ned murine microbiota:altered Schaedler?ora.Appl.Environ.Microbiol.

65,3287–3292

8Barnich,N.et al.(2007)CEACAM6acts as a receptor for adherent-invasive E.coli,supporting ileal mucosa colonization in Crohn disease.J.Clin.Invest.117,1566–1574

9Frank,D.N.et al.(2007)Molecular-phylogenetic characterization of microbial community imbalances in human in?ammatory bowel diseases.Proc.Natl.Acad.Sci.U.S.A.104,13780–13785

10Kuhn,R.et al.(1993)Interleukin-10-de?cient mice develop chronic enterocolitis.Cell75,263–274

11Martin, F.P.et al.(2009)Metabolic assessment of gradual development of moderate experimental colitis in IL-10de?cient mice.J.Proteome Res.8,2376–2387

12Shiomi,Y.et al.(2011)GCMS-based metabolomic study in mice with colitis induced by dextran sulfate sodium.In?amm.Bowel Dis.DOI:

10.1002/IBD.21616

13Marchesi,J.R.et al.(2007)Rapid and noninvasive metabonomic characterization of in?ammatory bowel disease.J.Proteome Res.6, 546–551

14Williams,H.R.et al.(2009)Characterization of in?ammatory bowel disease with urinary metabolic pro?ling.Am.J.Gastroenterol.104, 1435–1444

15Lees,C.W.et al.(2011)New IBD genetics:common pathways with other diseases.Gut DOI:10.1136/gut.2009.199679

16Rao,V.P.et al.(2006)Innate immune in?ammatory response against enteric bacteria Helicobacter hepaticus induces mammary adenocarcinoma in mice.Cancer Res.66,7395–7400

17Suzuki,H.et al.(2009)Helicobacter pylori and gastric cancer.Gastric Cancer12,79–87

18Linz,B.et al.(2007)An African origin for the intimate association between humans and Helicobacter pylori.Nature445,915–918

19O’Keefe,S.J.et al.(2007)Why do African Americans get more colon cancer than Native Africans?J.Nutr.137(Suppl.1),175S–182S 20Engle,S.J.et al.(2002)Elimination of colon cancer in germ-free transforming growth factor beta1-de?cient mice.Cancer Res.62, 6362–6366

21Uronis,J.M.et al.(2009)Modulation of the intestinal microbiota alters colitis-associated colorectal cancer susceptibility.PLoS ONE4,e6026 22Dove,W.F.et al.(1997)Intestinal neoplasia in the ApcMin mouse: independence from the microbial and natural killer(beige locus) status.Cancer Res.57,812–814

23Erdman,S.E.et al.(2009)Nitric oxide and TNF-alpha trigger colonic in?ammation and carcinogenesis in Helicobacter hepaticus-infected, Rag2-de?cient mice.Proc.Natl.Acad.Sci.U.S.A.106,1027–1032 24Maggio-Price,L.et al.(2006)Helicobacter infection is required for in?ammation and colon cancer in SMAD3-de?cient mice.Cancer Res.

66,828–838

25Monleon,D.et al.(2009)Metabolite pro?ling of fecal water extracts from human colorectal cancer.NMR Biomed.22,342–348

26Scanlan,P.D.et al.(2008)Culture-independent analysis of the gut microbiota in colorectal cancer and polyposis.Environ.Microbiol.10, 789–798

27Nambiar,P.R.et al.(2010)An‘Omics’based survey of human colon cancer.Mutat.Res.693,3–18

28Christl,S.U.et al.(1992)Role of dietary sulphate in the regulation of methanogenesis in the human large intestine.Gut33,1234–1238 29Hu,S.et al.(2011)The microbe-derived short chain fatty acid butyrate targets miRNA-dependent p21gene expression in human colon cancer.PLoS ONE6,e16221

30Ling,W.H.and Ha¨nninen,O.(1992)Shifting from a conventional diet to an uncooked vegan diet reversibly alters fecal hydrolytic activities in humans.J.Nutr.122,924–930

31Smith, E.A.and Macfarlane,G.T.(1996)Enumeration of human colonic bacteria producing phenolic and indolic compounds:effects of pH,carbohydrate availability and retention time on dissimilatory aromatic amino acid metabolism.J.Appl.Bacteriol.81,288–302

32Elsden,S.R.et al.(1976)The end products of the metabolism of aromatic amino acids by Clostridia.Arch.Microbiol.107,283–288 33Yokoyama,M.T.and Carlson,J.R.(1981)Production of skatole and para-Cresol by a rumen Lactobacillus sp.Appl.Environ.Microbiol.41, 71–76

34Lesaffer,G.et al.(2003)Urinary excretion of the uraemic toxin p-cresol in the rat:contribution of glucuronidation to its metabolization.

Nephrol.Dial.Transplant.18,1299–1306

35Kim,K.et al.(2011)Urine metabolomic analysis identi?es potential biomarkers and pathogenic pathways in kidney cancer.OMICS15, 293–303

36Slupsky,C.M.et al.(2010)Urine metabolite analysis offers potential early diagnosis of ovarian and breast cancers.Clin.Cancer Res.16, 5835–5841

37Carrola,J.et al.(2011)Metabolic signatures of lung cancer in bio?uids:NMR-based metabonomics of urine.J.Proteome Res.10, 221–230

38Catchpole,G.et al.(2011)Metabolic pro?ling reveals key metabolic features of renal cell carcinoma.J.Cell.Mol.Med.15,109–118

39Nam,H.et al.(2009)Combining tissue transcriptomics and urine metabolomics for breast cancer biomarker identi?cation.

Bioinformatics25,3151–3157

40Reddy,B.S.et al.(1996)Effect of amount and types of dietary fat on intestinal bacterial7alpha-dehydroxylase and phosphatidylinositol-speci?c phospholipase C and colonic mucosal diacylglycerol kinase and PKC activities during stages of colon tumor promotion.Cancer Res.56,2314–2320

41Wen,L.et al.(2008)Innate immunity and intestinal microbiota in the development of Type1diabetes.Nature455,1109–1113

42Fearnside,J.F.et al.(2008)Phylometabonomic patterns of adaptation to high fat diet feeding in inbred mice.PLoS ONE3,e1668

43Zhao,L.C.et al.(2010)A metabonomic comparison of urinary changes in Zucker and GK rats.J.Biomed.Biotechnol.DOI:10.1155/2010/ 431894

44Zhang,S.et al.(2008)Correlative and quantitative1H NMR-based metabolomics reveals speci?c metabolic pathway disturbances in diabetic rats.Anal.Biochem.383,76–84

45Godzien,J.et al.(2011)Metabolomic approach with LC-QTOF to study the effect of a nutraceutical treatment on urine of diabetic rats.

J.Proteome Res.10,837–844

357

46Waldram,A.et al.(2009)Top-down systems biology modeling of host metabotype–microbiome associations in obese rodents.J.Proteome Res.8,2361–2375

47Dumas,M.E.et al.(2007)Direct quantitative trait locus mapping of mammalian metabolic phenotypes in diabetic and normoglycemic rat models.Nat.Genet.39,666–672

48Dumas,M.E.et al.(2006)Metabolic pro?ling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice.Proc.

Natl.Acad.Sci.U.S.A.103,12511–12516

49Backhed,F.et al.(2007)Mechanisms underlying the resistance to diet-induced obesity in germ-free mice.Proc.Natl.Acad.Sci.U.S.A.

104,979–984

50Ley,R.E.et al.(2005)Obesity alters gut microbial ecology.Proc.Natl.

Acad.Sci.U.S.A.102,11070–11075

51Ley,R.E.et al.(2006)Microbial ecology:human gut microbes associated with obesity.Nature444,1022–1023

52Turnbaugh,P.J.et al.(2008)Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome.Cell Host Microbe3,213–223

53Zhang,Y.et al.(2011)Analysis of urinary metabolic pro?le in aging rats undergoing caloric restriction.Aging Clin.Exp.Res DOI:10.3275/7519

54Veselkov,K.A.et al.(2009)Recursive segment-wise peak alignment of biological(1)h NMR spectra for improved metabolic biomarker recovery.Anal.Chem.81,56–66

55Furet,J.P.et al.(2010)Differential adaptation of human gut microbiota to bariatric surgery-induced weight loss:links with metabolic and low-grade in?ammation markers.Diabetes59,3049–3057

56Ashra?an,H.et al.(2010)Metabolic surgery:an evolution through bariatric animal models.Obes.Rev.11,907–920

57Li,J.V.et al.(2011)Metabolic surgery profoundly in?uences gut microbial host metabolic cross-talk.Gut DOI:10.1136/gut.2010.

234708

58Mutch,D.M.et al.(2009)Metabolite pro?ling identi?es candidate markers re?ecting the clinical adaptations associated with Roux-en-Y gastric bypass surgery.PLoS ONE4,e7905

59Cani,P.D.et al.(2007)Selective increases of bi?dobacteria in gut micro?ora improve high-fat-diet-induced diabetes in mice through a mechanism associated with endotoxaemia.Diabetologia50,2374–2383

60Amar,J.et al.(2008)Energy intake is associated with endotoxemia in apparently healthy men.Am.J.Clin.Nutr.87,1219–1223

61Cani,P.D.et al.(2007)Metabolic endotoxemia initiates obesity and insulin resistance.Diabetes56,1761–1772

62Sekirov,I.et al.(2010)Gut microbiota in health and disease.Physiol.

Rev.90,859–904

63Yap,I.K.et al.(2010)Urinary metabolic phenotyping differentiates children with autism from their unaffected siblings and age-matched controls.J.Proteome Res.9,2996–3004

64Greenberg,N.et al.(2009)A proposed metabolic strategy for monitoring disease progression in Alzheimer’s disease.

Electrophoresis30,1235–1239

65de Graaf,A.A.et al.(2010)Pro?ling human gut bacterial metabolism and its kinetics using[U-13C]glucose and NMR.NMR Biomed.23, 2–12

66Arumugam,M.et al.(2011)Enterotypes of the human gut microbiome.Nature473,174–180

67Sudo,N.et al.(2004)Postnatal microbial colonization programs the hypothalamic-pituitary-adrenal system for stress response in mice.J.

Physiol.558(Pt1),263–275

68Ait-Belgnaoui, A.et al.(2009)Lactobacillus farciminis treatment attenuates stress-induced overexpression of Fos protein in spinal and supraspinal sites after colorectal distension in rats.

Neurogastroenterol.Motil.21,567–573e18–9

69Backhed,F.et al.(2004)The gut microbiota as an environmental factor that regulates fat storage.Proc.Natl.Acad.Sci.U.S.A.101, 15718–15723

70Turnbaugh,P.J.et al.(2006)An obesity-associated gut microbiome with increased capacity for energy harvest.Nature444,1027–1031 71Samuel,B.S.and Gordon,J.I.(2006)A humanized gnotobiotic mouse model of host–archaeal-bacterial mutualism.Proc.Natl.Acad.Sci.

U.S.A.103,10011–1001672Valladares,R.et al.(2010)Lactobacillus johnsonii N6.2mitigates the development of type1diabetes in BB-DP rats.PLoS ONE5, e10507

73Fox,J.G.et al.(2010)Gut microbes de?ne liver cancer risk in mice exposed to chemical and viral transgenic hepatocarcinogens.Gut59, 88–97

74Martinez,I.et al.(2009)Diet-induced metabolic improvements in a hamster model of hypercholesterolemia are strongly linked to alterations of the gut microbiota.Appl.Environ.Microbiol.75, 4175–4184

75Arora,T.and Sharma,R.(2011)Fermentation potential of the gut microbiome:implications for energy homeostasis and weight management.Nutr.Rev.69,99–106

76Goytia,M.and Shafer,W.M.(2010)Polyamines can increase resistance of Neisseria gonorrhoeae to mediators of the innate human host defense.Infect.Immun.78,3187–3195

77Kim,I.Y.et al.(2010)1H NMR-based metabolomic study on resistance to diet-induced obesity in AHNAK knock-out mice.Biochem.Biophys.

https://www.wendangku.net/doc/bf2933845.html,mun.403,428–434

78Wang,Z.et al.(2011)Gut?ora metabolism of phosphatidylcholine promotes cardiovascular disease.Nature472,57–63

79Calvani,R.et al.(2010)Gut microbiome-derived metabolites characterize a peculiar obese urinary metabotype.Int.J.Obes.

(Lond.)34,1095–1098

80Qiu,Y.et al.(2010)Urinary metabonomic study on colorectal cancer.

J.Proteome Res.9,1627–1634

81Zhao,L.et al.(2010)1H NMR-based metabonomic analysis of metabolic changes in streptozotocin-induced diabetic rats.Anal.

Sci.26,1277–1282

82Watanabe,T.et al.(2006)The blood pressure-lowering effect and safety of chlorogenic acid from green coffee bean extract in essential hypertension.Clin.Exp.Hypertens.28,439–449

83Zheng,S.et al.(2010)Urinary metabonomic study on biochemical changes in chronic unpredictable mild stress model of depression.

Clin.Chim.Acta411,204–209

84Suhre,K.et al.(2010)Metabolic footprint of diabetes:a multiplatform metabolomics study in an epidemiological setting.PLoS ONE5, e13953

85Lis,A.W.et al.(1976)Pro?les of ultraviolet-absorbing components of urine from autistic children,as obtained by high-resolution ion-exchange chromatography.Clin.Chem.22,1528–1532

86Russell, D.W.and Setchell,K.D.(1992)Bile acid biosynthesis.

Biochemistry31,4737–4749

87Mackie,R.I.et al.(1991)Lipid metabolism in anaerobic ecosystems.

Crit.Rev.Microbiol.17,449–479

88De Weirdt,R.et al.(2010)Human faecal microbiota display variable patterns of glycerol metabolism.FEMS Microbiol.Ecol.

74,601–611

89Vollenweider,S.et al.(2003)Puri?cation and structural characterization of3-hydroxypropionaldehyde and its derivatives.

J.Agric.Food Chem.51,3287–3293

90Li,J.V.et al.(2008)Global metabolic responses of NMRI mice to an experimental Plasmodium berghei infection.J.Proteome Res.7,3948–3956

91Lindon,J.C.et al.(2004)Metabonomics:systems biology in pharmaceutical research and development.Curr.Opin.Mol.Ther.

6,265–272

92Madsen,R.et al.(2010)Chemometrics in metabolomics:a review in human disease diagnosis.Anal.Chim.Acta659,23–33

93Li,M.et al.(2008)Symbiotic gut microbes modulate human metabolic phenotypes.Proc.Natl.Acad.Sci.U.S.A.105,2117–2122

94Crockford, D.J.et al.(2006)Statistical heterospectroscopy,an approach to the integrated analysis of NMR and UPLC-MS data sets:application in metabonomic toxicology studies.Anal.Chem.

78,363–371

95Claus,S.P.et al.(2008)Systemic multicompartmental effects of the gut microbiome on mouse metabolic phenotypes.Mol.Syst.Biol.4, 219

96Swann,J.et al.(2009)Gut microbiome modulates the toxicity of hydrazine:a metabonomic study.Mol.Biosyst.5,351–355

97Swann,J.R.et al.(2011)Systemic gut microbial modulation of bile acid metabolism in host tissue compartments.Proc.Natl.Acad.Sci.

U.S.A.108(Suppl.1),4523–4530

358

98Nicholls,A.W.et al.(2003)NMR spectroscopic-based metabonomic studies of urinary metabolite variation in acclimatizing germ-free rats.Chem.Res.Toxicol.16,1395–1404

99Martin, F.P.et al.(2008)Probiotic modulation of symbiotic gut microbial–host metabolic interactions in a humanized microbiome mouse model.Mol.Syst.Biol.4,157100Martin,F.P.et al.(2008)Top-down systems biology integration of conditional prebiotic modulated transgenomic interactions in a humanized microbiome mouse model.Mol.Syst.Biol.4,205

101Yap,I.K.et al.(2008)Metabonomic and microbiological analysis of the dynamic effect of vancomycin-induced gut microbiota modi?cation in the mouse.J.Proteome Res.7,3718–3728

359