Per me è molto difficile risalire a cosa mangiassero i miei nonni, vivendo nella povertà e campagna comunque penso mangiassero prodotti propri, brodi, cerali pane, latte di capra e mucca, verdure e quando disponibile la carne.
Nonna materna ha avuto 2 ictus, in seguito è morta. Nonno materno è morto di infarto abbastanza giovane (sulla sessantina).
Nonno paterno non ricordo, mentre nonna paterna ha avuto un ictus ed in seguito è morta di tumore all'intestino.

Beh sicuramente non seguirei il loro stile alimentare. Comunque e'sempre un terno al lotto e lo capite bene. Si puo vedere che i cereali, vino, e in genere legumi non sono quindi .."la buona dieta mediterranea"...alla faccia. Come dicevamo e'difficile capire cosa ha giocato male nel caso dei vostri avi. Lo stress, la poverta...poca carne e molti cereali/legumi, i pastori no sono solamente noti per la loro humile vita ma anche per il trascurarsi, mangiare male e come quando capita.
che ne pensate se di aprire un post sulla paleodieta/mediterranea italiana e familiare.

Discussione interessante. Entrambe le mie nonne sono morte ultranovantenni, ed entrambi i nonni appena sessantenni. Eppure credo (pur non essendone sicuro) che mangiassero in modo uguale, perciò la situazione sarebbe un po' difficile da spiegare. Secondo me bisognerebbe risalire alla loro infanzia, infatti poi la povertà del dopoguerra ha fatto si che la dieta si uniformasse su pochissima carne, alcuni derivati da cereali (farina bianca e farina di mais), latte e vino.
Mio papà è del '25, è ancora in formissima e in ottima salute, eppure la sua dieta degli ultimi 20 anni è basata su pasta, pane, un sacco di dolciumi, zero verdure, poca carne e pochi grassi animali, chissà che il segreto non sia l'alimentazione che ha avuto nella sua infanzia, allattato al seno fino a 3 anni e zero vaccinazioni.

Discussione interessante. Entrambe le mie nonne sono morte ultranovantenni, ed entrambi i nonni appena sessantenni. Eppure credo (pur non essendone sicuro) che mangiassero in modo uguale, perciò la situazione sarebbe un po' difficile da spiegare. Secondo me bisognerebbe risalire alla loro infanzia, infatti poi la povertà del dopoguerra ha fatto si che la dieta si uniformasse su pochissima carne, alcuni derivati da cereali (farina bianca e farina di mais), latte e vino.
Mio papà è del '25, è ancora in formissima e in ottima salute, eppure la sua dieta degli ultimi 20 anni è basata su pasta, pane, un sacco di dolciumi, zero verdure, poca carne e pochi grassi animali, chissà che il segreto non sia l'alimentazione che ha avuto nella sua infanzia, allattato al seno fino a 3 anni e zero vaccinazioni.

interessante vero. e che ne pensi del fatto che non hanno mai visto un antibiotico? nella loro infanzia non hanno distrutto la loro flora intestinale/con conseguente rimodellamento genetico con l'uso di antibiotici. inoltre i cibi erano sicuramente piu sani. si, si moriva di brucellosi, ma il latte era latte. il pesce era pescato e non allevato con l'uso di coloranti per far apparire le carni piu appetibili, non c'era il mercurio etc. insomma non hanno avuto quell'impatto che noi da piccoli abbiamo avuto. i polli che abbiamo mangiato da piccoli erano zeppi di ormoni e antibiotici. il numero di ginecomastie nei giovani e' elevato. il mais non era OGM, ne veniva ultrabombardato con pesticidi. e cosi via. Il grano non aveva 4 volte il codice genetico. ieri ho visto in un filmato sulla televisione olandese che i filetti di pollo vegono analizzati su un nastro trasportatore con un grande sistema Xray per evitare che pezzetti di ossa finisero nei filetti. i batteri amano i raggi X ...sai che belle mutazioni.
lo stile di vita/alimentare e'cambiato troppo in fretta. Oggi non trovi niente che non abbia qualcosa di aggiunto.

Si difatti inoltre ho una mia teoria personale sul grano creso che è stata la vera rovina di questi ultimi 30 anni. Fortunati quelli che non lo hanno mangiato nella loro gioventù.

Si difatti inoltre ho una mia teoria personale sul grano creso che è stata la vera rovina di questi ultimi 30 anni. Fortunati quelli che non lo hanno mangiato nella loro gioventù.

che ne dici se si fa un post sulla paleodieta/epigenetica italiana. Partendo da quello che si sa dal tempo degli etruschi si puo avere una idea della dieta piu consona al nostro organismo. la paleo e' la dieta locale..alla Weston Price. non una per tutti.
Si il creso e' in effetti il primo OGM ottenuto per bombardamento con raggi gamma.

Si difatti inoltre ho una mia teoria personale sul grano creso che è stata la vera rovina di questi ultimi 30 anni. Fortunati quelli che non lo hanno mangiato nella loro gioventù.

Questo appunto e'interessante

The biological standard of living in Europe during the last two millennia
Nikola Koepke and Joerg Baten
+ Author Affiliations
University of Tuebingen, Department of Economic History, Mohlstraβe 36, D-72074 Tübingen, Germany Abstract
This study offers the first anthropometric estimates of the biological standard of living in Europe during the first millennium AD, and extends the literature on the second millennium. The overall picture drawn from our data is one of stagnant heights. There was no large-scale progress in European nutritional status over the period studied, not even for the period between 1000 and 1800, for which recent GDP per capita estimates indicate increasing development. We find that heights stagnated in Central, Western and Southern Europe during the Roman imperial period, while astonishingly increasing in the fifth and sixth centuries. Noteworthy also is the similarity of height development in the three large regions of Europe. In an exploratory regression analysis of height determinants, population density turns out to have been an economically (not statistically) significant and negative factor, indicating the relevance of decreasing marginal product theories and Malthusian theory for the pre-1800 period. Of marginal significance, however, were climate (warmer temperatures being favourable for a good nutritional status), social inequality and gender inequality (both reducing average height). Lastly, we also discuss the limitations of our approach.

L'ostruzione delle vene mi pare vada per la maggiore tra infarti ed ictus. La "mediterranea" dei nostri nonni non era tutto sto granchè.

Questo e'un buon articolo. PDF da scaricare gratis su PubMed.
Qui si chiarisce interazione tra cibi processati e ...per cosi dire..innaturali (sembro un vegano) e la flora batterica.

Comparison with ancestral diets suggests dense acellular carbohydrates promote an inflammatory microbiota, and may be the primary dietary cause of leptin resistance and obesity.
Spreadbury I.
SourceGastrointestinal Diseases Research Unit, Queen's University, Kingston, Ontario, Canada.

Abstract
A novel hypothesis of obesity is suggested by consideration of diet-related inflammation and evolutionary medicine. The obese homeostatically guard their elevated weight. In rodent models of high-fat diet-induced obesity, leptin resistance is seen initially at vagal afferents, blunting the actions of satiety mediators, then centrally, with gastrointestinal bacterial-triggered SOCS3 signaling implicated. In humans, dietary fat and fructose elevate systemic lipopolysaccharide, while dietary glucose also strongly activates SOCS3 signaling. Crucially however, in humans, low-carbohydrate diets spontaneously decrease weight in a way that low-fat diets do not. Furthermore, nutrition transition patterns and the health of those still eating diverse ancestral diets with abundant food suggest that neither glycemic index, altered fat, nor carbohydrate intake can be intrinsic causes of obesity, and that human energy homeostasis functions well without Westernized foods containing flours, sugar, and refined fats. Due to being made up of cells, virtually all "ancestral foods" have markedly lower carbohydrate densities than flour- and sugar-containing foods, a property quite independent of glycemic index. Thus the "forgotten organ" of the gastrointestinal microbiota is a prime candidate to be influenced by evolutionarily unprecedented postprandial luminal carbohydrate concentrations. The present hypothesis suggests that in parallel with the bacterial effects of sugars on dental and periodontal health, acellular flours, sugars, and processed foods produce an inflammatory microbiota via the upper gastrointestinal tract, with fat able to effect a "double hit" by increasing systemic absorption of lipopolysaccharide. This model is consistent with a broad spectrum of reported dietary phenomena. A diet of grain-free whole foods with carbohydrate from cellular tubers, leaves, and fruits may produce a gastrointestinal microbiota consistent with our evolutionary condition, potentially explaining the exceptional macronutrient-independent metabolic health of non-Westernized populations, and the apparent efficacy of the modern "Paleolithic" diet on satiety and metabolism.

no,se vi legge bressanini vi bolla come complottisti incompetenti e vi ride pure addosso come è nel suo stile da presuntuoso so tutto io,in realtà probabilmente sa tutto ma dice solo quello che gli fa comodo

provate a chiedergli del grano creso(OGM) collegato ai problemi che porta a lungo andare... come minimo scatta come una molla per sostenere che gli OGM sonosicuribuonifavolosienonhannodifetti... pensa che quando non conoscevo la monsanto e lessi un suo articolo pensai che fosse una associazione benefica che veniva criticata ingiustamente da deliranti complottisti

no,se vi legge bressanini vi bolla come complottisti incompetenti e vi ride pure addosso come è nel suo stile da presuntuoso so tutto io,in realtà probabilmente sa tutto ma dice solo quello che gli fa comodo

provate a chiedergli del grano creso(OGM) collegato ai problemi che porta a lungo andare... come minimo scatta come una molla per sostenere che gli OGM sonosicuribuonifavolosienonhannodifetti... pensa che quando non conoscevo la monsanto e lessi un suo articolo pensai che fosse una associazione benefica che veniva criticata ingiustamente da deliranti complottisti

La discussione sul glutine è partita parlando di Bressanini https://www.mangiaconsapevole.com/forum/T-GLUTINE.html ma rimaniamo in tema qui.

no,se vi legge bressanini vi bolla come complottisti incompetenti e vi ride pure addosso come è nel suo stile da presuntuoso so tutto io,in realtà probabilmente sa tutto ma dice solo quello che gli fa comodo

provate a chiedergli del grano creso(OGM) collegato ai problemi che porta a lungo andare... come minimo scatta come una molla per sostenere che gli OGM sonosicuribuonifavolosienonhannodifetti... pensa che quando non conoscevo la monsanto e lessi un suo articolo pensai che fosse una associazione benefica che veniva criticata ingiustamente da deliranti complottisti

Ciao,
lo scopo dell'articolo non era certo resuscitare...morti ( Berssanini....spero mi scusi) ma continuare sulla relazione dieta, cibo, flora intestinale ed epigenetica.
del grano creso e delle diavolerie della Monsanto non mi esprimo, ma sappiate che io ho studiato all'universita di scienze Agrarie e di quello che la Monsanto fa ne so molto. Fosse solo il grano...il mais..le carote...mica sono nate arancioni..etc. La Monsanto e'il diavolo per le verdure cosi come c'e'ne sono altri per carne, pesce etc.

il punto da non perdere di vista e' l'effetto della dieta sulla flora batterica. ho trovato un articolo dove si dimostra sperimentalmente, quindi non epidemiologicamente, che la dieta non altera la flora batterica in se ma altera il genotipo dei batteri. in altre parole non e'che la ripartizione dei generi/specie batteriche cambi ma avvengono molte mutazioni nel loro genoma. Il che fa vedere come anche i batteri, cosa che gia sapevamo, vadano incontro alla epigenetica. Mutazioni nel codice genetico dei batteri implicano produzione di altri acidi grassi, altre proteine, altre attivita metaboliche e diversa attivita sulla restante flora intestinale. ma anche la relazione con l'ospite cambia

Ciao
stavamo parlando della relazione cibo/dieta/epigenetica umana/animale. quindi avevo introdotto il tema dell'effetto della dieta sulla microflora intestinale. ebbene ci sono molti studi, e quello che ho copiato ne e'un esempio, in cui si fa vedere come la dieta non cambi tanto l'assetto della microflora ma molto di piu crea mutazioni/variazioni genetiche nella microflora presente. si potrebbe argumentare che tali mutazioni potrebbero essere la base delle variazioni epigenetiche che si evidenziano nell'uomo. ora e'come parlare di cosa c'e'stato prima..l'uovo o la gallina..insomma la dieta induce mutazioni genetiche nei batteri che a sua volta conportano variazioni nel tipo di acidi grassi e peptidi i quali attraversando le pareti intestinali aiutano a creare le variazioni epigenetiche...o e'il contrario.

A core gut microbiome in obese and lean twins
Peter J. Turnbaugh,1 Micah Hamady,3 Tanya Yatsunenko,1 Brandi L. Cantarel,5 Alexis Duncan,2 Ruth E. Ley,1 Mitchell L. Sogin,6 William J. Jones,7 Bruce A. Roe,8 Jason P. Affourtit,9 Michael Egholm,9 Bernard Henrissat,5 Andrew C. Heath,2 Rob Knight,4 and Jeffrey I. Gordon1
1Center for Genome Sciences, Washington University School of Medicine, St Louis MO 63108, USA2Department of Psychiatry, Washington University School of Medicine, St Louis MO 63108, USA3Department of Computer Science, University of Colorado, Boulder, CO 80309, USA4Department of Chemistry and Biochemistry, University of Colorado, Boulder, CO 80309, USA5CNRS, UMR6098 Marseille, France6Josephine Bay Paul Center, Marine Biological Laboratory, Woods Hole, MA 02543, USA7Environmental Genomics Core Facility, University of South Carolina, Columbia, SC 29208, USA8Department of Chemistry and Biochemistry and the Advanced Center for Genome Technology, University of Oklahoma, Norman, OK 73019, USA9454 Life Sciences, Branford, CT 06405, USA.*Correspondence and requests for materials should be addressed to J.I.G. (Email: [email protected])
Author information ► Copyright and License information ►
Copyright notice and Disclaimer
The publisher's final edited version of this article is available at Nature
See other articles in PMC that cite the published article.
Go to:Abstract.The human distal gut harbors a vast ensemble of microbes (the microbiota) that provide us with important metabolic capabilities, including the ability to extract energy from otherwise indigestible dietary polysaccharides1–6. Studies of a small number of unrelated, healthy adults have revealed substantial diversity in their gut communities, as measured by sequencing 16S rRNA genes6–8, yet how this diversity relates to function and to the rest of the genes in the collective genomes of the microbiota (the gut microbiome) remains obscure. Studies of lean and obese mice suggest that the gut microbiota affects energy balance by influencing the efficiency of calorie harvest from the diet, and how this harvested energy is utilized and stored3–5. To address the question of how host genotype, environmental exposures, and host adiposity influence the gut microbiome, we have characterized the fecal microbial communities of adult female monozygotic and dizygotic twin pairs concordant for leanness or obesity, and their mothers. Analysis of 154 individuals yielded 9,920 near full-length and 1,937,461 partial bacterial 16S rRNA sequences, plus 2.14 gigabases from their microbiomes. The results reveal that the human gut microbiome is shared among family members, but that each person’s gut microbial community varies in the specific bacterial lineages present, with a comparable degree of co-variation between adult monozygotic and dizygotic twin pairs. However, there was a wide array of shared microbial genes among sampled individuals, comprising an extensive, identifiable ‘core microbiome’ at the gene, rather than at the organismal lineage level. Obesity is associated with phylum-level changes in the microbiota, reduced bacterial diversity, and altered representation of bacterial genes and metabolic pathways. These results demonstrate that a diversity of organismal assemblages can nonetheless yield a core microbiome at a functional level, and that deviations from this core are associated with different physiologic states (obese versus lean).

We characterized gut microbial communities in 31 monozygotic (MZ) twin pairs, 23 dizygotic (DZ) twin pairs, and where available their mothers (n=46) (Supplementary Tables 1–5). MZ and DZ co-twins and parent-offspring pairs provided an attractive paradigm for assessing the impact of genotype and shared early environment exposures on the gut microbiome. Moreover, genetically ‘identical’9 MZ twin pairs gain weight in response to overfeeding in a more reproducible way than do unrelated individuals10 and are more concordant for body mass index (BMI) than DZ twin pairs11.

Twin pairs who had been enrolled in the Missouri Adolescent Female Twin Study (MOAFTS12) were recruited for this study (mean period of enrollment in MOAFTS, 11.7±1.2 years; range, 4.4–13.0 years). All twins were 25–32 years old, of European or African ancestry (EA and AA, respectively), were generally concordant for obesity (BMI≥30 kg/m2) or leanness (BMI=18.5–24.9 kg/m2) [1 twin pair was lean/overweight (overweight defined as BMI ≥25 and 800 km apart. Since fecal samples are readily attainable and representative of interpersonal differences in gut microbial ecology7, they were collected from each individual and frozen immediately. The collection procedure was repeated again with an average interval between sampling of 57±4 days.

To characterize the bacterial lineages present in the fecal microbiotas of these 154 individuals, we performed 16S rRNA sequencing, targeting the full-length gene with an ABI 3730xl capillary sequencer. Additionally, we performed multiplex pyrosequencing with a 454 FLX instrument to survey the gene’s V2 variable region13 and it’s V6 hypervariable region14 (Supplementary Tables 1–3).

Complementary phylogenetic and taxon-based methods were used to compare 16S rRNA sequences among fecal communities (see Methods). No matter which region of the gene was examined, individuals from the same family (a twin and her co-twin, or twins and their mother) had a more similar bacterial community structure than unrelated individuals (Fig. 1A and Supplementary Fig. 1A,B), and shared significantly more species-level phylotypes (defined as sharing ≥97% identity in their 16S rRNA sequences) [G=55.2, p<10−12 (V2); G=12.3, p<0.001 (V6); G=11.3, p<0.001 (full-length)]. No significant correlation was seen between the degree of physical separation of family members’ current homes and the degree of similarity between their microbial communities (defined by UniFrac15). The observed familial similarity was not due to an indirect effect of the physiologic states of obesity versus leanness; similar results were observed after stratifying twin-pairs and their mothers by BMI category (concordant lean or concordant obese individuals; Supplementary Fig. 2). Surprisingly, there was no significant difference in the degree of similarity in the gut microbiotas of adult MZ versus DZ twin-pairs (Fig. 1A). However, in the present study we could not assess whether MZ and DZ twin pairs had different degrees of similarities at earlier stages of their lives.

Figure 116S rRNA gene surveys reveal familial similarity and reduced diversity of the gut microbiota in obese individualsMultiplex pyrosequencing of V2 and V6 amplicons allowed higher levels of coverage compared to what was feasible using Sanger sequencing, reaching on average 3,984±232 (V2) and 24,786±1,403 (V6) sequences per sample. To control for differences in coverage, all analyses were performed on an equal number of randomly selected sequences [200 full-length, 1,000 V2, and 10,000 V6]. At this level of coverage, there was little overlap between the sampled fecal communities. Moreover, the number of 16S rRNA gene sequences belonging to each phylotype varied greatly between fecal microbiotas (Supplementary Tables 6–8).

Since this apparent lack of overlap could reflect the level of coverage (Supplementary Tables 1–3), we subsequently searched all hosts for bacterial phylotypes present at high abundance using a sampling model based on a combination of standard Poisson and binomial sampling statistics. The analysis allowed us to conclude that no phylotype was present at more than ~0.5% abundance in all of the samples in this study (see Supplementary Results). Finally, we subsampled our dataset by randomly selecting 50–3,000 sequences/sample; again, no phylotypes were detectable in all individuals sampled within this range of coverage (Supplementary Fig. 3).

Samples taken from the same individual at the initial collection point and 57±4 days later were consistent with respect to the specific phylotypes found (Supplementary Figs. 4,5), but showed variations in relative abundance of the major gut bacterial phyla (Supplementary Fig. 6). There was no significant association between UniFrac distance and the time between sample collections. Overall, fecal samples from the same individual were much more similar to one another than samples from family members or unrelated individuals (Fig. 1A), demonstrating that short-term temporal changes in community structure within an individual are minor compared to inter-personal differences.Analysis of 16S rRNA datasets produced by the three PCR-based methods, plus shotgun sequencing of community DNA (see below), revealed a lower proportion of Bacteroidetes and a higher proportion of Actinobacteria in obese versus lean EA and AA individuals (Supplementary Table 9). Combining the individual p-values across these independent analyses using Fisher’s method disclosed significantly less Bacteroidetes (p=0.003), more Actinobacteria (p=0.002), but no significant difference in Firmicutes (p=0.09). These findings are in agreement with previous work showing comparable differences in both taxa in mice2 and a progressive increase the representation of Bacteroidetes when 12 unrelated obese humans lost weight after being placed on one of two reduced calorie diets6.Across all methods, obesity was associated with a significant decrease in the level of diversity (Fig. 1B plus Supplementary Fig. 1C–F). This reduced diversity suggests an analogy: the obese gut microbiota is not like a rainforest or reef, which are adapted to high energy flux and are highly diverse, but rather may be more like a fertilizer runoff where a reduced diversity microbial community blooms with abnormal energy input16.

We subsequently characterized the microbial lineage and gene content of the fecal microbiomes of 18 individuals representing 6 of the families (3 lean EA MZ twin-pairs and their mothers plus 3 obese EA MZ twin pairs and their mothers) through shotgun pyrosequencing (Supplementary Tables 4,5) and BLASTX comparisons against a number of databases [KEGG17 (v44) and STRING18] plus a custom database of 44 reference human gut microbial genomes (Supplementary Figs. 7–10 and Supplementary Results). Our analysis parameters were validated using control datasets comprised of randomly fragmented microbial genes with annotations in the KEGG database17 (Supplementary Fig. 11 and Supplementary Methods). We also tested how technical advances that produce longer reads might improve these assignments by sequencing fecal community samples from one twin pair using Titanium pyrosequencing methods [average read length of 341±134 nt (SD) versus 208±68 nt for the standard FLX method]. Supplementary Fig. 12 shows that the frequency and quality of sequence assignments is improved as read length increases from 200 to 350 nt.

The 18 microbiomes were searched to identify sequences matching domains from experimentally validated Carbohydrate-Active enZymes (CAZymes). Sequences matching 156 total CAZy families were found within at least one human gut microbiome, including 77 glycoside hydrolase, 21 carbohydrate-binding module, 35 glycosyltransferase, 12 polysaccharide lyase, and 11 carbohydrate-esterase families (Supplementary Table 10). On average 2.62±0.13% of the sequences in the gut microbiome could be assigned to CAZymes (total of 217,615 sequences), a percentage that is greater than the most abundant KEGG pathway (‘Transporters’; 1.20±0.06% of the filtered sequences generated from each sample), and indicative of the abundant and diverse set of microbial genes directed towards accessing a wide range of polysaccharides.

Category-based clustering of the functions from each microbiome was performed using Principal Components Analysis (PCA) and hierarchical clustering19. Two distinct clusters of gut microbiomes were identified based on metabolic profile, corresponding to samples with an increased abundance of Firmicutes and Actinobacteria, and samples with a high abundance of Bacteroidetes (Fig. 2A). A linear regression of the first principal component (PC1, explaining 20% of the functional variance) and the relative abundance of the Bacteroidetes showed a highly significant correlation (R2=0.96, p<10−12; Fig. 2B). Functional profiles stabilized within each individual’s microbiome after ~20,000 sequences had been accumulated (Supplementary Fig. 13). Family members had more similar profiles than unrelated individuals (Fig. 2C), suggesting that shared bacterial community structure (who’s there based on 16S rRNA analyses) also translates into shared community-wide relative abundance of metabolic pathways. Accordingly, a direct comparison of functional and taxonomic similarity (see Supplementary Methods) disclosed a significant association: individuals with similar taxonomic profiles also share similar metabolic profiles (p<0.001; Mantel test).

Figure 2Metabolic pathway-based clustering and analysis of the human gut microbiome of MZ twinsFunctional clustering of phylum-wide sequence bins representing microbiome reads assigned to 23 human gut Firmicutes and 14 Bacteroidetes reference genomes showed discrete clustering by phylum (Supplementary Figs. 14A,15). Bootstrap analyses of the relative abundance of metabolic pathways in the microbiome-derived Firmicutes and Bacteroidetes sequence bins, disclosed 26 pathways with a significantly different relative abundance (Supplementary Fig. 14A). The Bacteroidetes bins were enriched for a number of carbohydrate metabolism pathways, while the Firmicutes bins were enriched for transport systems. The finding is consistent with our CAZyme analysis, which revealed a significantly higher relative abundance of glycoside hydrolases, carbohydrate-binding modules, glycosyltransferases, polysaccharide lyases, and carbohydrate esterases in the Bacteroidetes sequence bins (Supplementary Fig. 14B).

One of the major goals of the International Human Microbiome Project(s) is to determine whether there is an identifiable ‘core microbiome’ of shared organisms, genes, or functional capabilities found in a given body habitat of all or the vast majority of humans1. Although all of the 18 gut microbiomes surveyed showed a high level of beta-diversity with respect to the relative abundance of bacterial phyla (Fig. 3A), analysis of the relative abundance of broad functional categories of genes (COG) and metabolic pathways (KEGG) revealed a generally consistent pattern regardless of the sample surveyed (Fig. 3B and Supplementary Table 11): the pattern is also consistent with results we obtained from an meta-analysis of previously published gut microbiome datasets from nine adults20,21 (Supplementary Fig. 16). This consistency is not simply due to the broad level of these annotations, as a similar analysis of Bacteroidetes and Firmicutes reference genomes revealed substantial variation in the relative abundance of each category (see Supplementary Fig. 17). Furthermore, pair-wise comparisons of metabolic profiles obtained from the 18 microbiomes in this study revealed an average R2 of 0.97±0.002 (Fig. 2A), indicating a high level of functional similarity.

Figure 3Comparison of taxonomic and functional variations in the human gut microbiomeOverall functional diversity was compared using the Shannon index22, a measurement that combines diversity (the number of different types of metabolic pathways) and evenness (the relative abundance of each pathway). The human gut microbiomes surveyed had a stable and high Shannon index value (4.63±0.01), close to the maximum possible level of functional diversity (5.54; see Supplementary Methods). Despite the presence of a small number of abundant metabolic pathways (listed in Supplementary Table 11), the overall functional profile of each gut microbiome is quite even (Shannon evenness of 0.84±0.001 on a scale of 0 to 1), demonstrating that most metabolic pathways are found at a similar level of abundance. Interestingly, the level of functional diversity in each microbiome was significantly linked to the relative abundance of the Bacteroidetes (R2=0.81, p<10−6); microbiomes enriched for Firmicutes/Actinobacteria had a lower level of functional diversity. This observation is consistent with an analysis of simulated metagenomic reads generated from each of 36 Bacteroidetes and Firmicutes genomes (Supplementary Fig. 18): on average, the Bacteroidetes genomes have a significantly higher level of both functional diversity and evenness (Mann-Whitney, p95% were found after 26.11±2.02 Mb of sequence was collected from a given microbiome, whereas the ‘variable’ groups continue to increase substantially with each additional Mb of sequence. Of course, any estimate of the total size of the core microbiome will be dependent upon sequencing effort, especially for functional groups found at a low abundance. On average, our survey achieved greater than 450,000 sequences per fecal sample, which, assuming an even distribution, would allow us to sample groups found at a relative abundance of 10−4. To estimate the total size of the core microbiome based on the 18 individuals, we randomly sub-sampled each microbiome in 1,000 sequence intervals (Supplementary Fig. 19D). Based on this analysis, the core microbiome is approaching a total of 2,142 total orthologous groups (one site binding hyperbola curve fit to the resulting rarefaction curve, R2=0.9966), indicating that we have identified 93% of functional groups (defined by STRING) found within the core microbiome of the 18 individuals surveyed. Of these core groups, 71% (CAZy), 64% (KEGG), and 56% (STRING) were also found in the 9 previously published but much lower coverage datasets generated by capillary sequencing of adult fecal DNA20,21 (average of 78,413±2,044 bidirectional reads/sample; see Supplementary Methods).

Metabolic reconstructions of the ‘core’ microbiome revealed significant enrichment for a number of expected functional categories, including those involved in transcription and translation (Fig. 4). Metabolic profile-based clustering indicated that the representation of ‘core’ functional groups was highly consistent across samples (Supplementary Fig. 20), and includes a number of pathways likely important for life in the gut, such as those for carbohydrate and amino acid metabolism (e.g. fructose/mannose metabolism, aminosugar metabolism, and N-Glycan degradation). Variably represented pathways and categories include cell motility (only a subset of Firmicutes produce flagella), secretion systems, and membrane transport (e.g. phosphotransferase systems involved in the import of nutrients, including sugars; Fig. 4 and Supplementary Fig. 20).

Figure 4KEGG categories enriched or depleted in the core versus variable components of the gut microbiomeThe distribution of CAZy glycoside hydrolase and glycosyltransferase families was compared between each pair of microbiomes (see Supplementary Table 10 for CAZy families with a relative abundance >1%). This analysis revealed that all individuals have a similar profile of glycosyltransferases (R2=0.96±0.003), while the profiles of glycoside hydrolases were significantly more variable, even between family members (R2=0.80±0.01; p<10−30, paired Student’s t-test). This suggests that the number and spectrum of glycoside hydrolases is probably affected by ‘external’ factors such as diet more than the glycosyltransferases.

To identify metabolic pathways associated with obesity, only non-core associated (variable) functional groups were included in a comparison of the gut microbiomes of lean versus obese twin pairs. A bootstrap analysis23 was used to identify metabolic pathways that were enriched or depleted in the variable obese gut microbiome. For example, similar to a mouse model of diet-induced obesity4, the obese human gut microbiome was enriched for phosphotransferase systems involved in microbial processing of carbohydrates (Supplementary Table 12). All gut microbiome sequences were compared against the custom database of 44 human gut genomes: an odds ratio analysis revealed 383 genes that were significantly different between the obese and lean gut microbiome (q-value < 0.05; 273 enriched and 110 depleted in the obese microbiome; Supplementary Tables 13,14). By contrast, only 49 genes were consistently enriched or depleted between all twin-pairs (see Supplementary Methods).

These obesity-associated genes were representative of the taxonomic differences described above: 75% of the obesity-enriched genes were from Actinobacteria (vs. 0% of lean-enriched genes; the other 25% are from Firmicutes) while 42% of the lean-enriched genes were from Bacteroidetes (vs. 0% of the obesity-enriched genes). Their functional annotation indicated that many are involved in carbohydrate, lipid, and amino acid metabolism (Supplementary Tables 13,14). Together, they comprise an initial set of microbial biomarkers of the obese gut microbiome.

Our finding that the gut microbial community structures of adult MZ twin pairs had a degree of similarity that was comparable to that of DZ twin pairs, and only slightly more similar compared to their mothers, is consistent with an earlier fingerprinting study of adult twins24, and with a recent microarray-based analysis, which revealed that gut community assembly during the first year of life followed a more similar pattern in a pair of DZ twins compared to 12 unrelated infants25. Intriguingly, another fingerprinting study of MZ and DZ twins in childhood showed a slightly reduced similarity profile in DZ twins26. Thus, comprehensive time-course studies, comparing MZ and DZ twin pairs from birth through adulthood, as well as intergenerational analyses of their families’ microbiotas, will be key to determining the relative contributions of host genotype and environmental exposures to (gut) microbial ecology.

The hypothesis that there is a core human gut microbiome, definable by a set of abundant microbial organismal lineages that we all share, may be incorrect: by adulthood, no single bacterial phylotype was detectable at an abundant frequency in the guts of all 154 sampled humans. Instead, it appears that a core gut microbiome exists at the level of metabolic functions. This conservation suggests a high degree of redundancy in the gut microbiome and supports an ecological view of each individual as an ‘island’ inhabited by unique collections microbial phylotypes: as in actual islands, different species assemblages converge on shared core functions provided by distinctive components. Our findings raise the question of how core functionality is assembled in this body habitat. Understanding the underlying principles should provide insights about microbial adaptation to, and perhaps mutualistic community assembly within, a wide range of environments.

Go to:METHODS SUMMARY.Fecal samples were collected from each individual. Community DNA was prepared and used for pyrosequencing (454 Life Sciences), as well as for PCR and sequencing of bacterial 16S rRNA genes. Shotgun reads were mapped to reference genomes using the NCBI ‘non-redundant’ database, KEGG17, STRING18, CAZy ( http://www.cazy.org/), and a 44-member human gut microbial genomes database. Metabolic reconstructions were performed based on CAZy, KEGG, and STRING annotations. The relative abundance of KEGG metabolic pathways is referred to as a ‘metabolic profile.’

Go to:METHODS.Community DNA preparation
Fecal samples were frozen immediately after they were produced. De-identified samples were stored at −80°C before processing. 10–20g of each sample was pulverized in liquid nitrogen with a mortar and pestle. An aliquot (~500mg) of each sample was then suspended, while frozen, in a solution containing 500 µl of extraction buffer [200 mM Tris (pH 8.0), 200 mM NaCl, 20 mM EDTA], 210 µl of 20% SDS, 500 µl of a mixture of phenol:chloroform:isoamyl alcohol (25:24:1, pH 7.9), and 500 µl of a slurry of 0.1 mm-diameter zirconia/silica beads (BioSpec Products, Bartlesville, OK). Microbial cells were subsequently lysed by mechanical disruption with a bead beater (BioSpec Products) set on high for 2 min at room temperature, followed by extraction with phenol:chloroform:isoamyl alcohol, and precipitation with isopropanol. DNA obtained from three separate 10 mg frozen aliquots of each fecal sample were pooled (≥200µg DNA) and used for pyrosequencing (see below).

16S rRNA gene sequence-based surveys
Complementary phylogenetic and taxon-based methods were used to compare 16S rRNA sequences among fecal communities. Phylogenetic clustering with UniFrac15 is based on the principle that communities can be compared in terms of their shared evolutionary history, as measured by the degree to which they share branch length on a phylogenetic tree. We complemented this approach with taxon-based methods27, which disregard some of the information contained in the phylogenetic tree of the taxa in question, but have the advantage that specific taxa unique to, or shared among, groups of samples can be identified (e.g., those from lean or obese individuals). Prior to both types of analyses, we grouped 16S rRNA gene sequences into Operational Taxonomic Units (OTUs/phylotypes) using both cd-hit28 and the furthest-neighbor-like (FNL) algorithm, with a sequence identity threshold of 97%, which is commonly used to define ‘species’-level phylotypes. Taxonomy was assigned using the best-BLAST-hit against Greengenes29 (E-value cutoff of 1e-10, minimum 88% coverage, 88% percent identity) and the Hugenholtz taxonomy (downloaded from http://greengenes.lbl.gov/Download/Sequence_Data/Greengenes_format/ on May 12, 2008, excluding sequences annotated as chimeric).

Selection of operational taxonomic units (OTUs)
16S rRNA gene-derived pyrosequencing data were pre-processed to remove sequences with low quality scores, sequences with ambiguous characters, or sequences outside of the length bounds (V6 < 50nt, V2 < 200nt), and binned according to sample-specific barcode (e.g. ref. 13). Similar sequences were identified using Megablast30 and cd-hit, with the following parameters: E-value 1e−10 (Megablast only); minimum coverage, 99%; and minimum pairwise identity, 97%. Candidate OTUs were identified as sets of sequences connected to each other at this level using a maximum of 4000 hits per sequence. Each candidate OTU was considered valid if the average density of connection was above threshold; otherwise it was broken up into smaller connected components27.

Tree building and UniFrac clustering for PCA analysis
A relaxed neighbor-joining tree was built from one representative sequence per OTU using Clearcut31, employing the Kimura correction (the PH Lane mask was applied to V2 and full-length data), but otherwise with default comparisons. Unweighted UniFrac15 was run using the resulting tree. PCA was performed on the resulting matrix of distances between each pair of samples. To determine if the UniFrac distances were on average significantly different for pairs of samples (i.e. between twin-pairs, between twins and their mother, or between unrelated individuals), we performed a t-test on the UniFrac distance matrix, and generated a p-value for the t-statistic by permutation of the rows and columns as in the Mantel test, regenerating the t-statistic for 1,000 random samples, and using the distribution to obtain an empirical p-value.

Rarefaction and phylogenetic diversity (PD) measurements
To determine which individuals had the most diverse communities of gut bacteria, rarefaction plots and Phylogenetic Diversity (PD) measurements, as described by Faith32, were made for each sample. PD is the total amount of branch length in a phylogenetic tree constructed from the combined 16S rRNA datasets, leading to the sequences in a given sample. To account for differences in sampling effort between individuals, and to estimate how far we were from sampling the diversity of each individual completely, we plotted the accumulation of PD (branch length) with sampling effort, in a manner analogous to rarefaction curves. We generated the PD rarefaction curve for each individual by applying custom python code ( http://bmf2.colorado.edu/unifrac/about.psp) to the Arb parsimony insertion tree27.

Pyrosequencing of total community DNA
Shotgun sequencing runs were performed on the 454 FLX pyrosequencer from total fecal community DNA. Two samples were also analyzed in a single run employing Titanium extra long read pyrosequencing technology (see Supplementary Table 4,5). Sequencing reads with degenerate bases (“Ns”) were removed along with all duplicate sequences, as sequences of identical length and content are a common artifact of the pyrosequencing methodology. Finally, human sequences were removed by identifying sequences homologous to the H.sapiens reference genome (BLASTN e-value75, and score>50).

CAZyme analysis
Metagenomic sequence reads were searched against a library of modules derived from all entries in the Carbohydrate-Active enZymes (CAZy) database (www.cazy.org using FASTY33 e-value<10−6). This library consists of ~180,000 previously annotated modules (catalytic modules, carbohydrate binding modules (CBMs) and other non-catalytic modules or domains of unknown function) derived from ~80,000 protein sequences. The number of sequencing reads matching each CAZy family was divided by the number of total sequences assigned to CAZymes and multiplied by 100 to calculate a relative abundance. An R2 value was calculated for each pair of CAZy profiles. We then compared the distribution of glycoside hydrolase similarity scores to the distribution of glycosyltransferase similarity scores.

Statistical analyses
Xipe23 (version 2.4) was employed for bootstrap analyses of pathway enrichment and depletion, using the parameters sample size=10,000 and confidence level=0.95. Linear regressions were performed in Excel (version 11.0, Microsoft). Mann-Whitney and Student’s t-tests were utilized to identify statistically significant differences between two groups (Prism v4.0, GraphPad; Excel version 11.0, Microsoft). The Bonferroni correction was used to correct for multiple hypotheses. The Mantel test was used to compare distance matrices: the matrix of each pairwise comparison of the abundance of each reference genome, and the abundance of each metabolic pathway, were compared (Mantel program in Python using PyCogent34; 10,000 replicates). Data are represented as mean±SEM unless otherwise indicated.

Microbiome sequences were compared against the custom database of 44 gut genomes (BLASTX e-value50, and %identity>50). A gene by sample matrix was then screened to identify genes ‘commonly-enriched’ in either the obese or lean gut microbiome (defined by an odds ratio greater than 2 or less than 0.5 when comparing the pooled obese twin microbiomes to the pooled lean twin microbiomes and when comparing each individual obese twin microbiome to the aggregate lean twin microbiome, or vice versa). The statistical significance of enriched or depleted genes was then calculated using a modified t-test (q-value2 or <0.5): they represent a variety of taxonomic groups, including Firmicutes, Bacteroidetes, and Actinobacteria, and did not show any clear functional trends.

Go to:Supplementary Material.1
Supplementary Information is linked to the online version of the paper at www.nature.com/nature.

Click here to view.(11M, pdf)
Go to:Acknowledgements.We thank Sabrina Wagoner and Jill Manchester for technical support; Stacey Marion and Deborah Hopper for recruitment of participants and sample collection, Andrew Goodman, Brian Muegge, and Michael Mahowald for helpful suggestions, plus Sue Huse (Marine Biological Laboratory), Faheem Niazi and Sayid Attiya (454 Life Sciences), Chris Markovic, Lucinda Fulton, Bob Fulton, Elaine Mardis and Richard Wilson (Washington University Genome Sequencing Center), and Simone Macmil, Graham Wiley, Chunmei Qu, and Ping Wang (University of Oklahoma) for their assistance with sequencing, and Pedro M. Coutinho (Université de Provence, France) for help with the CAZy analysis. Deep draft assemblies of reference gut genomes were generated as part of an NHGRI-sponsored human gut microbiome initiative (HGMI, http://genome.wustl.edu/pub/organism/Microbes/Human_Gut_Microbiome/). This work was supported in part by the NIH (DK78669/ES012742/AA09022/HD049024), the NSF (OCE0430724), the W.M. Keck Foundation, and the Crohn’s and Colitis Foundation of America.

Go to:Footnotes.
Author Information – PJT, ACH, RK, and JIG designed the experiments. PJT, TY, AD, REL, MLS, WJJ, BAR, JPA, and ME generated the data. PJT, MH, MLS, BLC, AD, BH, ACH, RK, and JIG analyzed the data. PJT, ACH, RK, and JIG wrote the manuscript with input from the other members of the team. This Whole Genome Shotgun project has been deposited at DDBJ/EMBL/GenBank under accession number 32089. 454 pyrosequencing reads have been deposited in the NCBI Short Read Archive. Nearly full-length 16S rRNA gene sequences are deposited in GenBank under the accession numbers FJ362604-FJ372382. Annotated sequences are also available for further analysis in MG-RAST ( http://metagenomics.nmpdr.org/). The authors declare no competing financial interests.

Go to:REFERENCES.1. Turnbaugh PJ, et al. The human microbiome project. Nature. 2007;449:804–810. [PubMed]
2. Ley RE, et al. Obesity alters gut microbial ecology. Proc. Natl. Acad. Sci. USA. 2005;102:11070–11075. [PMC free article] [PubMed]
3. Turnbaugh PJ, et al. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature. 2006;444:1027–1031. [PubMed]
4. Turnbaugh PJ, Bäckhed F, Fulton L, Gordon JI. Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome. Cell Host Microbe. 2008;3:213–223. [PubMed]
5. Bäckhed F, et al. The gut microbiota as an environmental factor that regulates fat storage. Proc. Natl. Acad. Sci. USA. 2004;101:15718–15723. [PMC free article] [PubMed]
6. Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Human gut microbes associated with obesity. Nature. 2006;444:1022–1023. [PubMed]
7. Eckburg PB, et al. Diversity of the human intestinal microbial flora. Science. 2005;308:1635–1638. [PMC free article] [PubMed]
8. Frank DN, et al. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proc. Natl. Acad. Sci. USA. 2007;104:13780–13785. [PMC free article] [PubMed]
9. Bruder CE, et al. Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am. J. Hum. Genet. 2008;82:763–771. [PMC free article] [PubMed]
10. Bouchard C, et al. The response to long-term overfeeding in identical twins. N. Engl. J. Med. 1990;322:1477–1482. [PubMed]
11. Maes HH, Neale MC, Eaves LJ. Genetic and environmental factors in relative body weight and human adiposity. Behav. Genet. 1997;27:325–351. [PubMed]
12. Heath AC, et al. Ascertainment of a mid-western US female adolescent twin cohort for alcohol studies: assessment of sample representativeness using birth record data. Twin Research. 2002;5:107–112. [PubMed]
13. Hamady M, Walker JJ, Harris JK, Gold NJ, Knight R. Error-correcting barcoded primers for pyrosequencing hundreds of samples in multiplex. Nat. Methods. 2008;5:235–237. [PMC free article] [PubMed]
14. Sogin ML, et al. Microbial diversity in the deep sea and the underexplored "rare biosphere". Proc. Natl. Acad. Sci. USA. 2006;103:12115–12120. [PMC free article] [PubMed]
15. Lozupone C, Hamady M, Knight R. UniFrac-an online tool for comparing microbial community diversity in a phylogenetic context. BMC Bioinformatics. 2006;7:371. [PMC free article] [PubMed]
16. Watson SB, McCauley E, Downing JA. Patterns in phytoplankton taxonomic composition across temperate lakes of differing nutrient status. Limnology and Oceanography. 1997;42:487–495.
17. Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori M. The KEGG resource for deciphering the genome. Nucleic Acids Res. 2004;32:D277–D280. [PMC free article] [PubMed]
18. von Mering C, et al. STRING 7-recent developments in the integration and prediction of protein interactions. Nucleic Acids Res. 2007;35:D358–D362. [PMC free article] [PubMed]
19. de Hoon MJ, Imoto S, Nolan J, Miyano S. Open source clustering software. Bioinformatics. 2004;20:1453–1454. [PubMed]
20. Gill, et al. Metagenomic analysis of the human distal gut microbiome. Science. 2006;312:1355–1359. [PMC free article] [PubMed]
21. Kurokawa, et al. Comparative metagenomics revealed commonly enriched gene sets in human gut microbiomes. DNA Res. 2007;14:169–181. [PMC free article] [PubMed]
22. Dinsdale EA, et al. Functional metagenomic profiling of nine biomes. Nature. 2008;452:629–632. [PubMed]
23. Rodriguez-Brito B, Rohwer F, Edwards RA. An application of statistics to comparative metagenomics. BMC Bioinformatics. 2006;7:162. [PMC free article] [PubMed]
24. Zoetandal EG, Akkermans ADL, Akkermans-van Vliet WM, de Visser JA, de Vos WM. The host genotype affects the bacterial community in the human gastrointestinal tract. Microb. Ecol. Health Disease. 2001;13:129–134.
25. Palmer C, Bik EM, Digiulio DB, Relman DA, Brown PO. Development of the human infant intenstinal microbiota. PLoS Biol. 2007;5:e177. [PMC free article] [PubMed]
26. Stewart JA, Chadwick VS, Murray A. Investigations into the influence of host genetics on the predominant eubacteria in the faecal microflora of children. J. Med. Microbiol. 2005;54:1239–1242. [PubMed]

Il messaggio di Andrea Bertocchi e le relative risposte sono state spostate nel topic del glutine https://www.mangiaconsapevole.com/forum/T-GLUTINE.html?page=5

Qui rimaniamo in tema, grazie per la collaborazione.

EPIGENETICA. COME L'AMBIENTE CI MODIFICA
http://www.aipro.info/page.php?id=92
n.b. per vedere i pdf sottostanti bisogna essere loggati su google.
Transgenerational Inheritance of an Acquired Small RNA-Based Antiviral Response in C. elegans
[pdf] http://www.aipro.info/drive/File/11_CELL_Transgenerational_Inheritance.._Acquired_Small_RNA%5B1%5D.pdf [/pdf]

Acute Exercise Remodels Promoter Methylation in Human Skeletal Muscle
[pdf] http://www.aipro.info/drive/File/12%20CELL%20Acute%20Exercise%20Remodels%20Promoter%20Methylation.pdf [/pdf]

Chronic Pain: Emerging Evidence for the Involvement of Epigenetics
[pdf] http://www.aipro.info/drive/File/12_NEURON_Chronic_Pain..the_Involvement_of_Epigenetics%5B1%5D.pdf [/pdf]

p.s. Fabietto hai ancora problemi col visualizzatore PDF?

Grazie Alessio,
molto interessanti questi articoli. nell'articolo Pain/epigenetic si faceva l'ipotesi, basata sulle metilazioni osservate, tra dolore cronico e variazioni epigenetiche.
Ne avevo letto uno sul Crohn.
eccolo:
Integrated Metagenomics/Metaproteomics Reveals
Human Host-Microbiota Signatures of Crohn’s Disease
Alison R. Erickson1,2., Brandi L. Cantarel3., Regina Lamendella4.¤, Youssef Darzi5,6,
Emmanuel F. Mongodin3, Chongle Pan1, Manesh Shah1, Jonas Halfvarson7, Curt Tysk7,
Bernard Henrissat8, Jeroen Raes5,6, Nathan C. Verberkmoes1, Claire M. Fraser3", Robert L. Hettich1",
Janet K. Jansson4*"
1 Chemical Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States of America, 2 Graduate School of Genome Science and Technology,
University of Tennessee, Knoxville, Tennessee, United States of America, 3 Institute for Genome Sciences, University of Maryland School of Medicine, Baltimore, Maryland,
United States of America, 4 Department of Ecology, Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States of America,
5 Bioinformatics and Eco-Systems Biology lab, Department of Structural Biology, Vrije Universiteit Brussel, Brussels, Belgium, 6 Research Group of Microbiology (MICR),
Faculty of Sciences and Bioengineering Sciences, Vrije Universiteit Brussel, Brussels, Belgium, 7 Department of Internal Medicine, Division of Gastroenterology, O¨ rebro
University Hospital and School of Health and Medical Sciences, O¨ rebro University, O¨ rebro, Sweden, 8 Architecture et Fonction des Macromole´cules Biologiques, UMR6098,
Centre national de la recherche scientifique, Universite´s Aix-Marseille I & II, Marseille, France
Abstract
Crohn’s disease (CD) is an inflammatory bowel disease of complex etiology, although dysbiosis of the gut microbiota has
been implicated in chronic immune-mediated inflammation associated with CD. Here we combined shotgun metagenomic
and metaproteomic approaches to identify potential functional signatures of CD in stool samples from six twin pairs that
were either healthy, or that had CD in the ileum (ICD) or colon (CCD). Integration of these omics approaches revealed
several genes, proteins, and pathways that primarily differentiated ICD from healthy subjects, including depletion of many
proteins in ICD. In addition, the ICD phenotype was associated with alterations in bacterial carbohydrate metabolism,
bacterial-host interactions, as well as human host-secreted enzymes. This eco-systems biology approach underscores the
link between the gut microbiota and functional alterations in the pathophysiology of Crohn’s disease and aids in
identification of novel diagnostic targets and disease specific biomarkers.
Citation: Erickson AR, Cantarel BL, Lamendella R, Darzi Y, Mongodin EF, et al. (2012) Integrated Metagenomics/Metaproteomics Reveals Human Host-Microbiota
Signatures of Crohn’s Disease. PLoS ONE 7(11): e49138. doi:10.1371/journal.pone.0049138
Editor: Bryan A. White, University of Illinois, United States of America
Received July 10, 2012; Accepted October 3, 2012; Published November 28, 2012
This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for
any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Funding: Research was funded by the National Institutes of Health Human Microbiome Project, grant UH2DK83991. The funders had no role in study design,
data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]
. These authors contributed equally to this work.
" These authors are joint Senior Authors.
¤ Current address: Juniata College, Biology Department, Huntingdon, Pennsylvania, United States of America
Introduction
Humans live in close association with communities of microorganisms
(the human microbiota) that inhabit every exposed
surface and cavity in the body [1]. The collective genetic
information of the human microbiota represents a second genome,
the human microbiome, currently the focus of intense international
sequencing and research efforts [2–7]. To date the main
focus has been on using high throughput sequencing to determine
the composition of the human microbiome in healthy individuals
(e.g. characterization of the human microbiome across different
body sites [5] and across different ages and geographic areas [7].
Several of these studies have found a large variation in the gut
microbial community composition between individuals, but
considerable functional redundancy [5], [8].
The next step is to determine how the human microbiome
varies with disease. As part of a demonstration project funded
through the NIH Human Microbiome Project (HMP) we have
focused on the impact of the inflammatory bowel disease (IBD),
Crohn’s disease on the gut microbiota. Although most human
host-microbe associations are beneficial, several studies using both
culture-dependent and molecular approaches have suggested that
there is a dysbiosis in the gut microbiota of patients with Crohn’s
disease (CD) compared to healthy subjects [9–13]. In the current
study we specifically aimed to focus on functional differences in the
gut that may account for the previously observed dysbiosis.
Although recent advances in DNA sequencing and proteomics
technologies have opened the door to investigation of the structure
and function of the gut microbiota without the necessity for
cultivation, there have been very few efforts to date that have used
a multi-‘‘omics’’ approach to study the complex ecosystem in the
human gut [14]. The ability to combine information about the
identities of microbial community members (obtained from 16S
rRNA gene-based measurements), metabolic potential (obtained from
metagenome sequence data) and expression (obtained from
PLOS ONE | www.plosone.org 1 November 2012 | Volume 7 | Issue 11 | e49138
metaproteome data) should enable exploration of the gut
microbiota at multiple molecular levels simultaneously.
This study was focused on a subset of stool samples collected
from a large Swedish twin cohort with inflammatory bowel disease
(IBD) that was previously characterized with respect to their
bacterial community composition by deep 16S rRNA pyrotag
sequencing [15] and metabolite profiling [16]. Previous data
indicated that healthy twin pairs had a similar gut microbiota,
even when they had been living separately for decades [11], as also
supported by other studies showing higher similarity between twins
than between unrelated individuals [8]. By contrast, twin pairs in
which one or both subjects had CD harbored very dissimilar gut
microbial compositions [11]. This disparity of the gut microbiota
was particularly striking for subjects with inflammation in the
ileum (ileal CD, ICD) compared to healthy subjects [11], [15],
[16] and was primarily characterized by the reduced abundance of
several key beneficial members of the community, such as
Faecalibacterium prausnitzii.
Here our aim was to further explore a subset of the same
Swedish twin cohort for functions that were correlated to CD by
applying non-targeted, shotgun metagenomics [17] and metaproteomics
[18]. Although we know from our previous studies
mentioned above that there were differences in the microbial
communities and metabolite profiles between individuals with CD
and healthy in this cohort, what is lacking is an understanding of
the reasons for the differentiation of the samples in a functional
context. By application of an eco-systems biology approach [19],
here we were able to detect and directly correlate genes, proteins,
and metabolic pathways for the first time in complex human gut
samples. It was particularly valuable to include discordant twin
pairs in the sample set, where one twin was diseased and one was
healthy, thus representing some level of internal control of host
genetics on the microbiome (Table S1 in Supporting Information
S1).
The specific questions that we set out to address in this study
were: (1) What genes are actually expressed as proteins in the gut
and could play a functional role in the gut environment? (2) Are
there specific genes and proteins that could help to explain the
previously observed differentiation of the samples according to
Crohn’s disease etiology?
Shotgun metaproteomics is a relatively new technology in its’
application to complex and highly diverse microbial communities,
such as the human gut, and only recently have there been reports
about protein compositions in the gut and from only a few healthy
subjects [18], [20–23]. Therefore, in this study we deliberately
selected samples that were previously well characterized and
shown to significantly differ between healthy and CD for
optimization of the methodology and to increase our chances of
detecting proteins that could correlate to disease etiology. The
sample cohort included one healthy twin pair, one colonic Crohn’s
(CCD) twin pair, two ICD concordant twin pairs and two ICD
discordant twin pairs (Table S1 in Supporting Information S1). To
perform these analyses we optimized a shotgun metaproteomics
pipeline with matched metagenomes to obtain the most comprehensive
coverage of human distal gut proteins to date.
Results
Data generation and sequence clustering
We generated shotgun metagenomic (Table S2 in Supporting
Information S1) and shotgun mass spectrometry (MS)-based
metaproteomic (Tables S3, S4, S5, S6) datasets from the same
stool samples for direct comparisons. Metagenomic data were used
to assess whole-community gene content and predicted functional
capabilities of the gut microbiome, while metaproteomics was used
to identify the measurable microbial and human proteins being
expressed in the system.
Assessment of expressed genes using metaproteomics
Metagenomic data does not reveal the identities and abundances
of expressed gene products (proteins) under the conditions
studied. Therefore, to directly address gene function and protein
abundance, we performed database searches with tandem mass
spectra (MS/MS) of peptides from the same samples collected via
multi-dimensional liquid chromatography tandem mass spectrometry
(2d-LC-MS/MS). These extensive MS/MS datasets were
searched either against their corresponding matched metagenome
(MM) (Table S2 in Supporting Information S1) or a representative
set of 51 sequenced human microbial isolate reference genomes
(HMRGs) (Table S7), each concatenated with the predicted
human protein database (July 2007 release, NCBI). Although 51
reference genome sequences cannot capture all of the protein
diversity within the human gut microbiota, we chose to select these
as a minimal set of reference genomes based on genera that have
been previously found in these samples [15]. By selecting only a
subset of the larger bank of human isolate reference genomes that
are being produced through the Human Microbiome Project [3],
we aimed to reduce the sequence redundancy between species/
strains that is a limitation of current MS database searching
algorithms. While the isolate genomes chosen represent about
75% of the genera estimated by 16S analysis [15], the rest of the
community is comprised of genera that represent less than 1% of
the total community, or are unknown (Figure S1A in Supporting
Information S1). The HMRGs provided complete gene sequences
for many of the most abundant genera (Figure S1A in Supporting
Information S1), in contrast to the MMs that had more
fragmented sequence data from all of the taxa in the microbiota.
However, relying solely on reference genomes for proteome
identification limits the protein families identified to those in
sequenced organisms, which is a small percentage of the total
bacteria in the gut. To address the issue of gene redundancy
between strains/species belonging to the same genera in the
metagenome data, we developed a novel method for clustering of
proteins from the MM datasets to provide a more robust method
of assigning peptide-spectrum counts for relative quantification
[23].
On average, a total of 1,250 (healthy), 850 (ICD), and 788
(CCD) orthologous protein clusters were identified with MM
searches and 2,904 (healthy), 1,928 (ICD), and 2,241 (CCD)
proteins using HMRG searches. Together, these data represent
the largest metaproteome analysis of the human gut to date
(Tables S3 and S4). To gauge the overlap in protein sequence
coverage between the MM (read-based protein spectrum matches,
PSMs) and HMRG databases, we compared the assigned, nonredundant
spectra with high mass accuracy (610ppm) with PSMs
from both searches. Of the total spectra that have peptide
assignments to microbial and human proteins, 64% and 33% of
the PSMs were unique to the MM and HMRG databases,
respectively (Figure S1B in Supporting Information S1). These
results suggest that these databases are complimentary, each
containing a large set of unique peptides that individually are a
sampling of these very complex proteomes. This approach enabled
us to take advantage of both MMs and HMRGs to identify
disease-specific proteins associated with the human gut microbiota,
including those with unknown function.
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 2 November 2012 | Volume 7 | Issue 11 | e49138
General overview of metagenomic and metaproteomic
datasets
By broad comparison of the metagenomes and metaproteomes,
CD clustered separately from healthy (Figures 1 and S2), as also
seen by prior analysis of 16S rRNA gene sequence data [15] and
metabolite data [16] from the same cohort. The distinct clustering
according to disease phenotype observed in the metaproteome
data was statistically significant (p = 0.004) (Figure 1). The
clustering of samples from discordant twin pairs into their
respective disease category, instead of with their co-twin, suggests
that the disease phenotype was a stronger discriminator than
genetics (Figure S2 in Supporting Information S1). Therefore, for
the rest of the analyses we only considered disease phenotype for
comparisons, not twin status, and the four healthy individuals and
six ICD individuals were treated as separate phenotypic groups.
Although healthy and CCD metaproteomes could be distinguished
from another, they clustered more closely together
compared to the ICD metaproteomes that were clearly distinct
(Figure 1 and S2). This also substantiates previous findings that
there is a more substantial dysbiosis of the gut microbiota
associated with ICD [11], [13], [15]. Therefore, we primarily
focused on functions that differentiated ICD from healthy, but
included comparisons to CCD when relevant.
Taxonomic profile differences
Taxonomic profiles of the metagenomic data were determined
using nucleotide alignments and compared based on disease status
(healthy, CCD, ICD). Greater than 60% of the metagenomic
sequence reads in the samples from healthy subjects could not be
assigned at the phylum, family or genus level, compared to ,40%
of the reads in ICD or CCD subjects, potentially reflecting the
reduced bacterial diversity in the gut of CD patients. Of the
metagenomic reads for which a taxonomic assignment could be
made, 396 genera were represented in all of the samples, and nine
of those were present at .5% of reads, representing the core taxa.
Some members of the Firmicutes phylum, such as Faecalibacterium,
were significantly depleted in ICD compared to healthy (p,0.05;
Figure 2A), a result consistent with 16S rRNA gene sequencing
gene sequencing of the same samples [15].
In the metaproteome data we also found a sigificant depletion of
proteins from members of the Firmicutes phylum in ICD,
p = 0.00025 (Figure 2B). For example, proteins from Faecalibacterium,
Roseburia, Dialister and Coprococcus were significantly less
Figure 1. Clustering of distal gut metaproteomes according to disease. Non-metric multidimensional scaling (nMDS) of distal gut
metaproteomes from CD twin cohort. The different colored square symbols represent the metaproteomic profiles for each sample (Blue = CCD, Grey
= Healthy, Red = ICD). The numbers beside the symbols refer to the specific patient ID from Dicksved et al., 2008 (proteomes were run in technical
duplicates). The axes are dimensionless: the coefficients of determination for the correlations between ordination distances and distances in the
original n-dimensional space are. 472 and. 831 for Axis 1 and 2, respectively. A matrix of normalized spectral counts per protein (HMRG database
search) from each duplicate metaproteome was imported into PCORD v5 software. nMDS was performed using the Bray-Curtis distance measure A
three-dimensional solution was found after 119 iterations. The final stress for the nMDS was 6.47458. The white spots with grey shading correspond
to individual proteins identified using HMRG database. Arrows indicate strength of correlation of specific bacterial strains to ordinated data. Pearson
correlation coefficients for Faecalibacterium prausnitzii, Anaerofustis stercorihominis, Clostridium leptum, Bacteroides ovatus, Bacteroides sp. 4_3, and
Bacteroides sp. 3_1 were 20.875, 20.851, 0.784, 0.8, 0.788, and 0.817, respectively.
doi:10.1371/journal.pone.0049138.g001
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 3 November 2012 | Volume 7 | Issue 11 | e49138
abundant in ICD relative to healthy subjects (Figure 2B;
Table S8). This finding demonstrates that the systems biology
approach used was consistent at both the gene and protein level.
Broad metagenome-metaproteome comparisons
A larger proportion of genes in the metagenomes were
expressed and identified as proteins in healthy subjects compared
to CD patients (8% H versus 2% ICD or 2% CCD) (Figure 3A).
This finding was also supported by a significant decrease in
functional richness in the metagenomes of individuals with CD,
examined comparing KEGG Orthologous groups (KOs) identified
in each sample (Figure 3B). Due to the redundancy of orthologous
genes in the HMRG and MM databases, microbial ORFs, which
shared .80% sequence identity were clustered into orthologous
clusters (OCs), reducing 890,000 ORFs to 68,000 clusters. This
generated a total of 5,692 and 3,101 orthologous clusters (OC)
from the HMRGs and MMs, respectively, across all metaproteome
datasets. Of the OCs that were identified using the MM
searches, 344 were identified across all subjects (core) and included
general housekeeping proteins (such as ribosomal proteins);
whereas 1,221, 720, and 145 OCs were unique to either the
healthy, ICD, or CCD core metaproteomes, respectively (Table
S9). Analysis of these OCs revealed that 1,017 proteins from
the MM searches were unique (i.e., they were singletons), in
contrast to all identified proteins from the HMRG search,
suggesting that there is considerable protein diversity within the
human gut microbiota that is not captured in current reference
genome sequences.
Each dataset contained a subset of genes and proteins of
unknown function. For example, ,17% of predicted ORFs in the
metagenomic data were either conserved with no known function
or were not homologous to any known proteins. Approximately
31% of the proteins identified with the HMRG database
(Table S6) and 29% of proteins identified using MM microbial
OCs (including proteins that did not cluster) had no known
functions (Table S6). Interestingly, one OC comprising 11
unknown proteins was significantly correlated with ICD, whereas
five OCs (10–100 s of unknown proteins) were significantly
correlated with healthy subjects. These findings support the need
for better coupling of phenotypic assays with -omics strategies to
aid in the characterization of potentially important unknown genes
and proteins.
Differences between ICD and healthy metaproteomes
There were significant differences in several COG categories
when comparing the metaproteomes of ICD to healthy, primarily
due to a decrease in abundance of proteins in ICD (Figure 4).
General COG categories that were significantly less represented in
ICD compared to healthy included ‘‘carbohydrate transport and
metabolism’’, ‘‘energy production and conversion’’, ‘‘amino acid
transport and metabolism’’, ‘‘lipid transport and metabolism’’,
‘‘nucleotide transport and metabolism’’, ‘‘transcription, ‘‘intracel-
Eubacterium
Ruminococcus
Prevotella
Bacteroides
Roseburia
Dialister
Coprococcus
Faecalibacterium*
Subdoligranulum*
Median Fraction of Reads
10-1
10-2
10-4
10-5
10-3
10-6
10-1
10-2
10-4
10-5
10-3
10-6
Eubacterium
Ruminococcus
Prevotella
Bacteroides
Roseburia*
Dialister*
Coprococcus*
Faecalibacterium*
Median Fraction of Assigned Spectra
A B
Figure 2. Taxonomic assignments in metagenome and metaproteome datasets. Relative abundance (log scale) of genera in (A)
metagenomic datasets, determined by reference genome alignments and (B) metaproteomic datasets, determined by HMRG PSMs. Error bars
represent standard error of the mean of the samples from Healthy (3 MG, 4 MP), ICD (5 MG, 6 MP) and CCD (2 MG/MP). Asterisks indicate genera that
were statistically lower in relative abundance in ICD compared to Healthy (q-values of 0.0030, 0.0041, 0.0041, 0.0040 for Faecalibacterium Roseburia,
Coprococcus and Dialaster, respectively). Subdolidogranulum was not included in the HMRG database, so it is not shown in the metaproteome. Grey
bars = Healthy, Blue bars = CCD, Red bars = ICD. standard error of the mean.
doi:10.1371/journal.pone.0049138.g002
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 4 November 2012 | Volume 7 | Issue 11 | e49138
lular trafficking’’, and ‘‘defense mechanisms’’; suggesting that these
general processes are deficient in ICD (Figure 4). Only one
category, ‘‘replication, recombination and repair’’, was significantly
higher in the ICD metaproteomes compared to healthy
(Figure 4).
At a finer scale of resolution, there were 116 statistically
significant differentiating specific COGs between disease categories
in the metaproteome data (spectra count difference $5 and
adjusted p-value (q-value) of #0.05; Table S10 for complete
listing). In particular there was a depletion of microbial proteins in
ICD compared to healthy. The general depletion of microbial
proteins in ICD could either result from decreased expression,
increased protein degradation, or decreased microbial diversity
(i.e. reduction of Firmicutes). However, nine COGs belonging to
‘‘translation’’, ‘‘carbohydrate metabolism’’, ‘‘amino acid metabolism’’
and ‘‘inorganic ion metabolism’’ (i.e., COG 4771, an outer
membrane receptor for ferrienterochelin and colicins), were
statistically more abundant in ICD relative to healthy metaproteomes,
suggesting that they are potential stool indicators of ICD.
Metabolic pathways that differentiate ICD and healthy
phenotypes
The metaproteome data indicated significant differences in
carbohydrate degradation pathways between ICD and healthy
(Figure 4). Similar to a recent study [24] we also found by
screening the metagenomes that the healthy subjects had a higher
abundance of genes encoding carbohydrate active enzymes
‘‘CAZymes’’ typical of those that degrade complex carbohydrates
in the plant cell wall (e.g. glycoside hydrolases: GH78, GH9,
GH30, GH28 and GH26 and polysaccharide lyase PL11),
compared to those for degradation of animal-type carbohydrates
such as starch and glycogen (e.g. glycoside hydrolases: GH33,
GH0109, GH92 and GH89) (Figure S3 in Supporting Information
S1). By contrast, the ICD subjects had lower relative amounts
of genes encoding CAZymes for degradation of both plant and
animal-type carbohydrates compared to healthy. Because IBD and
Crohn’s patients, in particular, are discouraged from eating fibrous
foods, these changes could reflect functional shifts driving these
dietary recommendations. However, we do not have detailed
metadata about the diet of these subjects. Additionally, the
abundance of the protein in CAZy family GH112, which is
involved in mucin degradation [25], was depleted in ICD
compared to healthy (p,0.01) (Figure 5B), despite more of the
corresponding genes (i.e. mucin-desulfating sulfatase (Mds) genes)
in ICD (Figure 5A). Mucin desulfation is a rate-limiting step in
mucin degradation by colon bacteria [26]. In the colon, secreted
mucins have oligoscaccharide side chains that are more heavily
sulfated than the side chains of secreted mucins in regions of the
digestive tract with lower bacterial numbers. Sulfation of mucins
could make them less susceptible to degradation by bacterial
glycosidases.
There was also a depletion of butyrate and other short-chain
fatty acid (SCFA) production pathways in ICD in both the
metagenome (Figure 5A) and metaproteome (Figure 5B) datasets;
corresponding to a depletion of members of the Firmicutes
(Figure 5C). KEGG pathway analysis of the metaproteomic
datasets also revealed that central metabolic pathways, such as
glycolysis, were under-represented in ICD compared to healthy
(Figure 6A). Butyrate is known to be a major energy source for
colonocytes, is involved in the maintenance of colonic mucosal
health and can elicit anti-inflammatory effects, thus its depletion
could be one reason for the inflammation in CD. In addition, the
reduction of proteins involved in butyrate production in
Faecalibacterium was even lower than would be expected by the
abundance of this organism (Figure 6B), suggesting that their
expression was down regulated.
Bacterial-host interactions and defense
Some specific genes and proteins had a higher relative
abundance in ICD. For example, by close examination of both
gene and protein abundance measurements we found that several
Gram-negative bacterial outer membrane proteins (e.g. OmpA,
RagB, SusC/D and TonB) had a higher representation in the ICD
microbiota compared to healthy (Figure 5). Based on matches to
the HRMG database, these proteins largely corresponded to
Bacteroides proteins (Figures 5C and 6A). These different membrane
proteins have different predicted roles. For example, TonBdependent
receptors take up large macromolecular complexes,
including iron/siderophore complexes, vitamin B12 and sulfate
esters [27]. OmpA, a pore-forming protein in the outer membrane
of many Gram-negative bacteria, harbors diverse functions
including maintenance of cell structure, binding various substances,
adhesion, and resistance to antimicrobials [28], and is
suggested to be involved in gut mucosal association [29]. One
A B
0.04
0.06
0.08
0.10
0.12
H CCD ICD
Fraction of Metagenome
10A
10B
15A
15B
18A
9A
9B
6A
16B
18B
200 250 300 350 400 450
KO Richness
H
CCD
ICD
Patient ID
Figure 3. Comparison of protein expression levels across disease categories. (A) Boxplots depicting the distribution of the fraction of the
metagenomes with PSMs. Boxes indicate 25th, 50th and 75th percentile, with whiskers representing 10th and 90th percentile points. (B) Gene family
richness as measured by the number of KEGG Orthologous group (KO) matches in the metagenomic dataset. Grey = Healthy, Blue = CCD, Red =
ICD.
doi:10.1371/journal.pone.0049138.g003
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 5 November 2012 | Volume 7 | Issue 11 | e49138
hypothesis is that because OmpA is highly represented and highly
conserved in many enteric bacteria, the immune system has
acquired the ability to recognize and to be activated by this class of
protein [30]. Because these proteins are more abundant in ICD,
the immune system may respond with a heightened immune
response. Our study also provides the first evidence of elevated
abundance of other major OMPs, such as RagB, SusC/D
associated with CD (Figures 5 and 6A). An elevated IgG response
to RagB was previously reported in subgingival samples of patients
with periodontitis [31] and virulence of the rag locus was
demonstrated in Porphyromonas gingivalis strains [32]. While the
role of RagB/Sus in the etiology of CD warrants further study, our
data suggest that there is a shift from a healthy microbiota towards
a microbial consortium that can elicit an inflammatory immune
response. This finding would support the current hypothesis that
CD is manifested by an aberrant mucosal response to otherwise
harmless bacterial antigens in genetically susceptible individuals
[33], [34]. These differences could also be due to broad shifts in
Gram-negative versus Gram-positive bacteria, since we see a
reduction in Gram-positive Firmicutes relative to Gram-negative
Proteobacteria based on 16S studies [11], [12], [15]. Although
there was no observed shift in total Bacteroides, previously we found
that there were differences in proportions of specific Bacteroides
species in individuals with ICD compared to healthy [11].
Translation, ribosomal
structure and biogenesis
Carbohydrate transport
and metabolism
Energy production
and conversion
Posttranslational modification,
protein turnover, chaperones
Amino acid transport
and metabolism
Lipid transport
and metabolism
Coenzyme transport
and metabolism
Nucleotide transport
and metabolism
Secondary metabolites
biosynthesis transport
and catabolism
Transcription
Replication, recombination
and repair
Intracellular trafficking,
secretion, and vesicular
transport
Cell cycle control, cell division
chromosome partitioning
Defense mechanisms
Signal transduction
mechanisms
Cell wall, membrane,
envelope biogenesis
Cell motility
Inorganic ion transport
and metabolism
Difference between means (H-ICD)
*
*
*
*
*
*
*
*
*
-2 -1 0 1 2 3 4 5 6 7
Figure 4. Metaproteome differences between mean Healthy and mean ICD COG frequencies. To determine statistically significant
differences between categories, White’s non-parametric t-test was used with bootstrapping and Storey FDR multiple test correction. 95% upper and
lower confidence intervals are shown. Red and grey bars indicate COG categories that are higher in ICD or Healthy metaproteomes, respectively;
Asterisks indicate COG categories that were significantly different between ICD and healthy (q-value,0.05).
doi:10.1371/journal.pone.0049138.g004
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 6 November 2012 | Volume 7 | Issue 11 | e49138
H
CCD
ICD
A
B
OmpA
TonB
SCFA
RagB,
SusC/D
MSD
OmpA
TonB
SCFA
Butyrate
RagB,
SusC/D
MSD
C
Butyrate
1e−06 5e−06 1e−05 5e−05 1e−04 5e−04 1e−03
1e−06 1e−04 1e−02
mean relative abundance
mean relative abundance
Bacteroides sp. 3_2_5
Bacteroides thetaiotamicron VPI-5842
Bacteroides vulgatus ATCC 8492
Bacteroides sp. 2_2_4
Bacteroides dorei DSM 17855
Bacteroides ovatus ATCC8483
Bacteroides uniformis
Bacteroides sp. 3_1_33FAA
Bacteroides sp. 4_3_47FAA
Bacteroides sp. 9_1_42FAA
Bacteroides sp. D4
Coprococcus comes
Clostridium sp. SS2/1
Ruminococcus gnavus ATCC 29149
Ruminococcus lactaris ATCC 29176
Ruminococcus obeum ATCC 29174
Prevotella copri CB7, DSM 18205
Dorea formicigenerans ATCC 27755
Citrobacter koseri ATCC BAA-895
Catenibacterium mitsuokai DSM 15897
Bacteroides pectinophilus ATCC 43243
Bifidobacterium adolescentis L2-32
Faecalibacterium prausnitzii A2-165
Eubacterium rectale ATCC 33656
Roseburia intestinalis L1-82
Faecalibacterium prausnitzii M21/1
Clostridium M62/1
Anareofustis stercorihominis DSM 17244
Normalized Spectral Abundance
3-hydroxybutyryl-CoA dehydrogenase
NADPH-dependent butanol dehyrdrogenase
acetate kinase
acetyl-CoA acetyltransferase
butyryl-CoA dehydrogenase
formate acetyltransferase 1
phosphotransacetylase
major outer membrane protein OmpA
SusC
Outer membrane receptor for
ferrienterochelin and colicins
SusD/RagB
0 50 100 150 200 250 300
Figure 5. Specific genes and proteins that differ in relative amounts according to disease state. Relative Abundance of mucindesulfating
sulfatase (Mds), RagB and SusC/D, Outer Membrane Protein A (OmpA), TonB, Short-Chain Fatty Acid production (SCFA) and Butyrate
production in (A) metagenomes and (B) MM metaproteomes. Error bars in (A) and (B) represent the standard error of the mean of the samples from
Healthy (3 MG, 4 MP), ICD (5 MG, 6 MP) and CCD (2 MG/MP). (C) Specific outer membrane proteins and proteins involved in SCFA pathway that
differed between disease categories. Protein abundances were calculated as normalized spectral abundance using the HMRG database search. The
presence-absence heatmap indicates which of the 51 bacterial strains each protein matched to in the HMRG database search: black = species
present, white = species absent. Grey = Healthy, Blue = CCD, Red = ICD.
doi:10.1371/journal.pone.0049138.g005
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 7 November 2012 | Volume 7 | Issue 11 | e49138
Nucleotide
Metabolism
Carbohydrate
Metabolism
Metabolism
Energy
Amino Acid
Metabolism
Bacteria
Faecalibacterium prausnitzii
Dialister invisus
Ruminococcus lactaris
Clostridiales
Bacteroides dorei
Bacteroides
Unassigned
1
2
ABC
transporters
SusD/SusC RagB Sec (secretion)
system
Entner-Doudoroff
pathway
Cystein and methionine
Metabolism
Oxaloacetate
decaboxylase
Butyrate
production
Glucose-1-phosphate
Starch and adenyltransferase
sucrose
metabolism
Glutamate
Methylmalonyl-CoA mutase dehydrogenase
Glycolysis
Glycolysis
Gluconeogenesis
RNA polymerase
Glycolysis/
Gluconeogenesis
Oxaloacetate
decaboxylase
A
−120 −100 −80 −60 −40 −20 0
−30 −20 −10 0 10 20 30
Amino acid metabolism
Central metabolism
Energy metabolism
Genetic information processing
Glycan metabolism
Lipid metabolism
Metabolism of other molecules
Nucleotide metabolism
Transport system
Ubiquitin system
Akkermansia
Alistipes
Anaerofustis
Bacteroides
Bifidobacterium
Blautia
Catenibacterium
Citrobacter
Clostridium
Collinsella
Coprococcus
Dialister
Dorea
Eubacterium
Faecalibacterium
Prevotella
Roseburia
Ruminococcus
Butyrate
Production RNA
Polymerase
Module Fold Change (ICD/H)
B Species Fold Change (ICD/H)
Figure 6. Metabolic Pathways that Differentiate Healthy and ICD phenotypes. (A) Metabolic pathways differentiating between healthy and
ICD according to metabolic module analysis (p,0.05; 5% FDR). All pathways are less abundant in ICD compared to healthy except for Bacteroides
membrane proteins (upper left box) that are more abundant in ICD. The colors reflect their phylogenetic origin that was determined using the lowest
common ancestor of their HMRG mappings. Grey highlighted areas discussed in the main text: (1) butyrate production; (2) membrane proteins. (B)
Observed metabolic module abundance shift versus its expected value based on the abundance of the host species. To separate out modules whose
fold change is higher/lower than expected by the difference in its species abundance, we used the prediction interval of a fitted linear model (blue
lines). The grey symbols are (species-separated) modules that are not significantly different between ICD and H (Wilcoxon rank-sum test; 5% FDR).
They could have a high median fold change, but this is not always significant (eg when interpersonal variation is high). The colored symbols are
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 8 November 2012 | Volume 7 | Issue 11 | e49138
Broad functional comparisons of the human proteome
Because we are able to measure both bacterial and human
proteins in the same samples using metaproteomics, a total of
1,646 human proteins were experimentally identified in addition
to the microbial proteins discussed above. Gene ontology (GO)
analysis revealed that human proteins found in all 3 subject groups
(core) are enriched in functions associated with the structural
integrity of the mucosal epithelium such as regulation and activity
of actin cytoskeletal components. Proteolysis, digestion, and
carbohydrate catabolism were also among the most abundant
‘core’ functional terms, as would be expected in the human GItract
(Figure S4A in Supporting Information S1). For human
proteins that varied in healthy compared to CD, the majority were
involved in epithelial integrity and function, as detailed below.
Impaired epithelial integrity in ICD
The observation of several human proteins detected in higher
abundance in CD supports the hypothesis that subjects with ICD,
even in remission, have a defective epithelial barrier. The higher
abundance of human proteins could also be a consequence of
surgical resection of the ileum. For example, a higher abundance
of proteins involved in inflammatory and host defense, wounding
response, intracellular transport, and epithelial development and
differentiation were enriched in ICD subjects (Figure S4B in
Supporting Information S1). Furthermore, other proteins that
function in maintaining mucosal integrity were identified as being
statistically under-represented in ICD (q-value =0.022), including
protocadherin LKC, a calcium dependent mediator of cell-cell
adhesion that associates with the mucosal actin cytoskeleton [35]
and type 1 collagen (alpha-2), the major collagen in the intestinal
extracellular matrix [36]. A depletion of these proteins might
compromise host defense at the mucosal interface.
A defective epithelial barrier is thought to result in an aberrant
host response to luminal antigens leading to an exaggerated
adaptive immune response and chronic inflammation [37].
Human alpha defensin 5, a protein implicated in regulation of
bacterial concentrations in the ileal intestinal crypt [38–40] was
also statistically more abundant in ICD (q-value =0.022), suggesting
that the host may increase expression of defensins in response
to aberrant microbiota in these subjects, or that the products are
leaking from the intestinal site of action and therefore detected in
higher amounts in the stool samples.
Impaired intestinal absorption in ICD
Several pancreatic enzymes that are largely broken down in the
small intestine: chymotrypsinogen B1 and B2, pancreatic carboxypeptidase
A1 and B1 and pancreatic lipase, were identified
with higher abundance in stool samples of the subjects with ICD.
These enzymes are synthesized in the pancreas as inactive
precursors that are activated in the intestine where they aid in
digestion. Relatively high amounts of pancreatic enzymes in stool
samples may be indicative of pancreatitis, which has been linked to
CD [41], but remains to be confirmed since the subjects in this
study did not have active pancreatitis at the time of sampling.
Discussion
In this study we used a combination of large and complementary
‘‘-omics’’ datasets to provide the most comprehensive view of
the functional role of the gut microbiota in CD to date. We studied
the same stool samples obtained from twelve individuals that were
previously characterized with respect to microbial community and
metabolite compositions as part of a large CD twin cohort [11],
[12], [15], [16]. Here our aim was to specifically gain insight into
functional differences at the gene and protein level that were
correlated to Crohn’s disease. The results of this study not only
support existing lines of evidence but also add more pieces of
information to help fill in the complex puzzle of CD etiology.
Similar to the previous studies of 16S rRNA genes [11], [12], [15]
and metabolites [16], this study also found that the proteins
extracted from the samples clustered separately according to
disease status. Together these different omics datasets provide an
enormous amount of information, with dozens of species,
thousands of metabolites and hundreds of proteins that vary in
relative amounts, particularly when comparing ICD to healthy.
The majority of the metabolites [16] and many of the proteins that
differed according to disease status have not yet been characterized
and their functions are unknown. Specifically, the unknown
proteins detected here that were expressed in higher amounts in
ICD are of particular interest for further exploration because they
were expressed and not merely hypothetical proteins predicted
from sequence data and therefore potentially play functional roles
of importance to ICD.
The value of the eco-systems biology approach used here comes
from the ability not only to examine the structure and function of
the microbiota from multiple perspectives, but also from the ability
to integrate data from the gut microbiota and the host. New
findings from this study suggest several malfunctions in ICD, both
with respect to the intestinal microbiota and the host. For
example, dysbiosis of the bacterial community in ICD resulted in
expression of higher levels of several bacterial cell surface proteins,
many of which are antigenic and could contribute to an
exaggerated immune response. This imbalance came at the
expense of loss of proteins produced by many beneficial members
of the microbiota, including proteins involved in butyrate
production and degradation of mucin, thus supporting the
previously observed decrease in abundance of the corresponding
species in the same samples using 16S rRNA gene fingerprinting
approaches [11], [12]. At the same time, there were several
preliminary indications that the host epithelial barrier was
impaired, both with respect to structural integrity of the mucosal
boundary and with respect to its ability to absorb secreted
enzymes; although these findings could also be a consequence of
ileal resection. This finding correlates to the previously reported
increase in bile acid metabolites in the same samples from the ICD
individuals [16].
Together these large omics datasets point towards several new
targets for further investigation in the pursuit for diagnosis and
therapeutic treatments for Crohns disease. This study also
highlights the value of using an eco-systems biology approach to
obtain a more complete picture of the complex interactions
between the thousands of bacterial species in the distal gut with the
(species-separated) modules that are significant between ICD and H (Wilcoxon rank-sum test; 5% FDR). Colored symbols inside the interval are
significantly different but are in line with what would be expected from the species difference. Colored symbols outside the blue lines are higher/
lower than expected. Specific Faecalibacterium proteins that are down regulated in the butyrate module (green squares) include the following:
butyryl-CoA dehydrogenase (EC 1.3.99.2), 3-hydroxyacyl-CoA dehydrogenase (EC 1.1.1.35), enoyl-CoA hydratase/carnithine racemase, and acetyl-CoA
acetyltransferases; as well as the module for lysine fermentation to acetate and butyrate (pink square). Specific Bacteroides proteins that are down
regulated in the DNA-directed RNA polymerase module are the following (red X’s): alpha and beta subunits (EC 2.7.7.6).
doi:10.1371/journal.pone.0049138.g006
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 9 November 2012 | Volume 7 | Issue 11 | e49138
human host. It will be of great value to extend these studies to
larger cohorts of CD patients and to carry out longitudinal studies
to assess i) how the composition and function of the gut microbiota
changes over time with respect to disease inflammation and ii) how
the microbiota is impacted by other factors including drug therapy
and surgery.
Materials and Methods
Patient cohort
The Swedish twin cohort was previously described in several
studies [11], [12], [15], [16], [42], [43]. For this study, we focused
on six monozygotic twin pairs including: one set of healthy twins
with existing metaproteome data [18] one set of concordant twins
with Crohn’s disease inflammation localized in the colon (CCD),
two sets of concordant twins with Crohn’s disease inflammation
localized in the ileum (ICD) and two sets of ICD discordant twins
(Table S1 in Supporting Information S1). Representatives of both
sexes were included in the study (6 females and 6 males) and the
subjects were all adults (youngest, born 1962; oldest born 1947).
None had taken antibiotics within 12 months of sampling. Three
of the subjects had gastroenteritis within 3 months prior to
sampling. Most of the patients had undergone surgery as
indicated, but all were many years prior to the sampling event
(Table S1 in in Supporting Information S1). All patients were in
endoscopic remission, or had minor inflammatory activity in the
neo-terminal ileum only, at the time of sampling. In addition, the
16S rRNA gene composition was determined for all samples
previously by 454 pyrotag sequencing [15] and the metabolite
compositions were determined from fecal water collected from the
same samples [16].
Community DNA preparation
Stool samples were shipped to the Orebro University Hospital,
Orebro, Sweden, at most one day after sample collection and
immediately frozen at 270uC upon arrival. The samples were
stored continuously frozen until use and small portions were
excised and thawed immediately prior to DNA extraction to avoid
freeze-thaw damage. DNA was extracted from 250 mg of each
stool sample in duplicate using the MoBio Power Soil DNA Kit
(MoBio, Solana Beach, CA, USA), as previously described [15],
and if necessary to get higher yields we also used an optimized
IGS-Zymo DNA extraction protocol reported previously [44].
Shotgun metagenomic sequencing
DNA isolation from stool samples yielded 3–5 ug of purified
metagenomic DNA from each of twelve samples. Each sample was
subjected to picogreen and gel-based QC assays prior to library
construction. Unpaired, shotgun fragment sequencing libraries
were constructed using our customized, automated library
construction procedure. Our method modifies the manufacturerprovided
protocol by adjusting enzymatic reaction volumes and
replacing gel-based fragment size-selection steps with AMPure
SPRI magnetic beads to enable automation of the process using
liquid-handling robotics. Following library construction, each
sample was subjected to emPCR amplification and 454 sequencing
according to manufacturer specifications. Raw sequence data
was processed using the Roche/454 run processing software to
filter short, mixed, and low-quality reads. Whole metagenomic
shotgun sequencing generated a total of 15,307,850 reads and
more than 5,428,202 kilobases (or 5 Gbp) of high-quality, passedfilter
sequence data (Table S2 in Supporting Information S1).
The metagenome sequence data can be retrieved using the
following URL for the NCBI SRA data deposit, under project ID
46321: http://www.ncbi.nlm.nih.gov/sites/entrez?db = bioproject&
cmd = Retrieve&dopt = Overview&list_uids = 46321.
Metagenomic taxonomic classification
Metagenomic reads were compared to publically available
human-associated bacterial reference genomes using NUCMER
(80% id, 80% coverage) for taxonomic assignment. In cases where
reads did not match reference genomes taxonomic classification
was made using sequence comparison against known proteins in
NCBI NR using BLASTX (90% id). In cases where reads had high
identity matches to multiple sequences, the taxonomic nearest
neighbor was chosen. Taxonomic classification for each MS
spectrum was determined by the protein sequence predicted from
metagenomic contig sequences, where the taxonomy of a contig is
based on the nearest neighbor classification of the read sequences
composing the contig. In cases where no classification was
obtained, the ‘human gut microbiome classification’ was given.
Family assignments are based on the NCBI taxonomic tree.
Potential 16S sequences were identified using RNA-HMM and
classified using RDP 2.0. Clustering of samples by taxonomy was
done using Ginko, with a log10(X+1) normalization, euclidean
distances and Ward’s method for hierarchical clustering.
Metagenomics gene finding and protein clustering
Sequences were assembled with the Newbler Assembler
(v2.0.01.14) and genes were predicted on contigs greater than
500 bp using METAGENE [45]. Genes on contigs less than
500 bp were searched against a database of reference genomes
using FASTX [46]. Genes were predicted from alignments to
homologous sequences. In regions where no homologous sequences
are found, METAGENE [47] was used for de novo gene
prediction and generated 594,362 genes, greater than 50 nt, across
10 metagenomic datasets.
An all-vs-all BLASTP [47] search was performed against the
human associated bacterial reference genome protein database
using thresholds of percent identity .80 and e-value ,1025,
protein clusters were created using an MCL [48] with an inflation
value of 1.5. Predicted ORFs from metagenomes were mapped to
17,408 of these clusters using BLASTP with an 80% identity
threshold; 196,002 genes did not map to a cluster.
Functional analysis
ORFs were searched against the eggNOG [49], CAZY [50] and
KEGG Orthologous groups [51] databases using NCBI-BLAST
[47] using e-value cutoff of 1026 and bits per position cutoff of 1.
COG and NOG functional assignments were assigned based on
this comparison. In addition sequences were searched against a
library of HMMs consisting of TIGRFAMS [52], and PFAM [53],
[54] using HMMPFAM [55]. Relative abundances of annotations
were determined using a random sampling of the smallest number
of reads in contigs as the sample size with 100 iterations. The
mean of this random sampling was calculated to determine the
relative abundance of a gene or function in the sample.
Cell lysis and protein extraction
Approximately 10 g portions of the same stool samples used for
DNA extractions were processed by differential centrifugation to
enrich the bacterial cell fraction as previously described [18]. The
microbial cell pellets (,100 mg) were processed via single tube cell
lysis [56] protein digestion and peptide desalting prior to 2d-LCMS/
MS analyses [18], [57]. Briefly, the cell pellet was
resuspended in 6 M Guanidine/10 mM DTT to lyse cells,
denature proteins, and reduce disulfide bonds. The guanidine
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 10 November 2012 | Volume 7 | Issue 11 | e49138
concentration was diluted to 1 Mwith 50 mM Tris buffer/10 mM
CaCl2 and sequencing grade trypsin (Promega, Madison, WI) was
added to digest proteins to peptides. Following proteome digestion,
the peptide solution was treated again with 10 mM DTT to
reduce disulfide bonds. We have found this method of double
reduction to be as effective as blocking with iodoacetamide. The
complex peptide solution was desalted via C18 solid phase
extraction, concentrated, solvent exchanged into 100% water/
0.1% formic acid, filtered (0.45 um filter), and aliquoted.
2D-LC-MS/MS
All samples were analyzed in technical duplicates via twodimensional
(2D) nano-LC MS/MS with a split-phase column
(RP-SCX-RP) [58], [59] on a LTQ Orbitrap (Thermo Fisher
Scientific) with 22 hr runs per sample. For each sample, peptide
mixtures were separated by a 12 step, multidimensional highpressure
liquid chromatographic elution consisting of eleven salt
pulses (ammonium acetate) followed by a 2 hr reverse-phase
gradient from 100% solvent A (A: 95% H2O, 5% acetonitrile,
0.1% formic acid) to 50% solvent B (B: 30% H2O, 70%
acetonitrile, 0.1% formic acid). The last salt pulse was followed
with a gradient from 100% solvent A to 100% solvent B. During a
single chromatographic separation (22 hr run), mass spectral data
acquisition was performed with Xcalibur software (version 2.0.7;
Thermo Fisher Scientific). Precursor full MS spectra (from 400–
1700 m/z) were acquired in the Orbitrap with resolution
r = 30,000 followed by five data-dependent MS/MS scans at
35% normalized collision energy in the LTQ with dynamic
exclusion enabled (repeat count 1).
Protein database construction
The first database, referred to as the matched metagenome
(MM), was created per sample by directly predicting ORFs from
raw sequencing reads to prevent loss of sequence diversity when
collapsing unrelated sequencing reads for metgenome assembly
(RMPS metagenomic processing method described in detail by
Cantarel et al. [23]. ORFs larger than 50 nt were predicted using
Metagene. Redundant protein sequences were removed, by
pairwise comparisons using 100% identity over 100% of the
shorter proteins (i.e. when aligning 2 proteins, the shorter of the
two must be covered completely by the larger one at 100%
identity), producing 491K – 1.58 M ORFs per sample. Each of
these 12 individual protein databases (6a, 6b, 9a, 9b, 10a, 10b,
15a, 15b, 16a, 16b, 18a, and 18b) included human reference
sequences (July 2007 release, NCBI; ,36,000 protein sequences)
and common contaminants (i.e., trypsin and keratin; 36 protein
sequences).
A second database, referred to as the human microbial isolate
reference genome database (HMRGs), was utilized in a complementary
database search and also contained human reference
sequences and common contaminants. While this reference
genome database is not exactly representative of each sample, it
can provide definitive species/protein identifications, which were
used to support and complement the MM searches. This database
was created by concatenating 51 human-derived reference isolate
genomes from the JGI IMG human microbiome project (IMGHMP)
into a single FASTA-formatted protein sequence database.
The criteria used to select 51 human-derived microbial isolates
were based on genera that have been previously found in the 16S
data from the same samples [15] in addition to strains that are
known to be common gut inhabitants; while avoiding representation
from similar species and strains to reduce redundancy. A list
of all 51 isolates included in this database can be found in
Table S7. All protein databases, MM and HMRG datasets, and
supporting figures and tables can also be downloaded from: http://
compbio.ornl.gov/crohns_disease_metagenomics_metaproteomics/.
Proteome informatics
All MS/MS from individual runs were searched with the
SEQUEST (v.27) algorithm [60] against a custom-made FASTA
formatted protein sequence databases described below. The
SEQUEST database search required fully tryptic (tryptic at both
termini) peptides with up to 4 miscleavages and a 3 Da mass
tolerance window around the precursor ion mass and 0.5 Da for
fragment ion masses. As previously described [23], all SEQUEST
output files were assembled and filtered using DTASelect (v1.9)
[61] at $2 peptides per protein for the HMRG database searches
and at a 1-peptide level (required minimum of $1 peptides to
confidently identify theoretical peptides from a genomic read
followed by $2 peptides to identify a protein) for the MM
database searches with the following widely accepted parameters:
cross correlation scores (XCorr) of at least 1.8, 2.5, 3.5 for +1, +2,
and +3 charge states, respectively and a minimum deltCN of 0.0
for all 12 samples (24 MS runs). A ‘‘post-database’’ search filter
was applied to the MM identifications where we used the high
mass accuracy capabilities of the Orbitrap to remove all peptides
that did not fall within 210# ppm #10 to the predicted parent
mass of the SEQUEST identified peptide. This was done to
remove the large number of false positives generated from the
minimum of $1 peptides to confidently identify a peptide from a
genomic read. This method of filtering peptides via high mass
accuracy post-SEQUEST database searches is generally an
accepted alternative to filtering during the search via mass
accuracy. Both methods have advantages and disadvantages, but
for our workflow filtering after the SEQUEST search was found to
be most effective.
The acquired mass spectrometry data (mzXML format) from this
publication have been submitted to the Proteome Commons Tranche
repository at www.proteomecommons.org and assigned the hash
identifier: rji3fAXT1XG0PxdrWWrM1M4XXznm6i7XKW2ZMVbfyYvo2G44eBimTcv4osnXHyhDvoCOA1av4EywiTFqX8Pf-
JI9SP4EAAAAAAAChfg.
False discovery rates
A target-decoy database [62], [63] was generated for the
HMRGs and the MMs for one healthy (6b, run 1), ICD (18a, run
2), and CCD (9a, run 2) subject and searched against their
corresponding MS experiments (i.e., forward-reverse database for
sample 6b was searched against spectra from run 1) to estimate the
peptide-level false discovery rate (FDR). All target-decoy SEQUEST
output files were assembled and filtered using DTASelect
(v1.9) [61] with the same XCorr filters of at least 1.8, 2.5, 3.5 for
+1, +2, and +3 charge states, respectively. The HMRGs were
filtered at a $2 peptide per protein with a deltCN 0.0 with an
empirical FDR threshold of #2.0%. The MM data was filtered at
a $1 peptide per predicted genomic read with a deltCN 0.0 and
high mass accuracy of parent peptide (210 # ppm # 10) followed
by a post-database $2 peptide per protein filter, with an empirical
FDR threshold of #2.0%. Additional metrics and results on false
discovery rates can be found in Supporting Information S1 and
Tables S11 and S12.
Proteome label-free quantification
The spectral count for a microbial protein cluster (‘‘CLST…’’)
was calculated as the number of unique peptide identifications that
can be attributed to proteins from that cluster and not from any
other cluster. Because proteins with high sequence similarity were
grouped in clusters, the majority of peptide identifications from the
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 11 November 2012 | Volume 7 | Issue 11 | e49138
metagenomic read databases (RMPS) can be uniquely attributed
to only one cluster. The spectral counts for human proteins were
calculated from both unique and non-unique peptide identifications
using DTASelect with default settings as described above.
Spectral counts for both human proteins and microbial protein
clusters from an MS/MS run were normalized by the total
numbers of tandem mass spectra (MS/MS) of this run. A scaling
factor, ai, was calculated for every run as ai = N/ni, where N is
the average number of total MS/MS spectra per run and ni is the
MS/MS spectral number of run i. The spectral counts for all
proteins in a single MS run were then normalized by multiplying
them with the run’s scaling factor. The reference isolate genome
database results were also normalized using the same scaling factor
and approach.
The 24 MS runs were grouped into the following three sample
sets for both databases (MMs and HMRGs): healthy subjects: 6a,
6b, 16b, and 18b; CCD subjects: 9a and 9b; and ICD subjects:
10a, 10b, 15a, 15b, 16a, and 18a.
Statistical analyses
The metagenomic microbial protein clusters (MM databases)
with differential expression between two sample sets were identified
using label-free quantification. We only considered microbial
protein clusters that have more than five spectral counts in four
or more of the runs in the two sets under comparison. P-values were
calculated using the Wilcoxon rank sum test. The p-values were
then used to compute q-values [64]. Proteins were considered as
differentially expressed if their q-values were less than a false
discovery rate threshold of 0.05 and the differences between their
median spectral counts of the two sets are greater than 5. Human
proteins were quantified separately using the same procedure.
The proteomics results were also analyzed using hierarchical
clustering. We only considered proteins with median absolute
deviations greater than 1. Normalized spectral counts were log2
transformed by adding a pseudo-count of one. Hierarchical
clustering on both proteins and samples were performed using the
hclust function in the R stat library and the heatmap was plotted
using the heatmap.2 function in the R gplot library.
Non-metric multidimensional scaling (nMDS) was performed
using normalized spectral abundances of proteins derived from 24
MS runs searched against 51 human-associated bacterial isolates.
nMDS was performed in PCORD v5 using the Bray-Curtis distance
measure [65]. Briefly, a matrix of normalized spectral counts for
each protein from each metaproteomic run were imported into
PCORD v5 and the indicator analysis was performed using the
randomization method. MRPP analysis was performed on the rank
transformed spectral abundances within PCORD v5 to test the null
hypothesis that there is no difference between the bacterial
metaproteomic profiles from each phenotype.
KEGG modules analysis was performed to highlight differences
in metabolism between healthy and CD. The bulk of metaproteomic
KOs were mapped to the KEGG modules reference
database in addition to the butyrate production module. Only
modules that had more than 30% coverage were considered for
downstream analysis. Then differential expression between modules
was tested using Wilcoxon’s rank-sum test in R and p-values
were corrected for multiple testing using Benjamini-Hochberg’s
false discovery rate (FDR). A module was considered significantly
different if the median difference between the two groups was
more than 5 with FDR set to 10% under a two-sided alternative
hypothesis. Modules and KOs that were significantly down
regulated in ICD were visualized within iPATH [66]. Additionally,
the phylogenetic origin of these modules and KOs, was shown
using the lowest common ancestor.
Ethics
LBNL has an approved Federal-wide assurance on file with
HHS that covers this activity: OHRP Federal-wide Assurance
Number FWA 00006253. Certification of Human Subjects
Committee review: This activity has been reviewed and approved
by the HSC in accordance with requirements sent forth in the
DHHS regulations at 45 CFR 46.103(f), which requires that each
application or proposal for HHS-supported human subject
research be reviewed and approved by the Institutional Review
Board. Date of Approval: April 30, 2010; Approval Number:
272H01-30APR11.
The consent procedure was approved by the ethical research
committee at O ¨ rebro University Hospital, where the samples were
collected. The study was approved by O ¨ rebro Lans Landsting on
December 17, 2003 (D-nr 167/03).
Consent to participation in the study ‘‘Ulcerative colitis
and Crohn’s disease in twins’’ and to treatment of
personal information
I have been informed in writing about this actual study and
have had time in peace and quiet to read through the information
and to ask questions by telephone. I have also been provided with
a copy of the written information and my written consent.
Through my signature I provide my consent to:
– participate in the study.
– that my personal information can be used as in the written
information.
– that my samples are treated as in the written information.
– that Jonas Halfvarson, gastroenterologist at USO¨ , can request
copies of my medical journal.
I am aware that participation is voluntary, and that at the same
time I may at any time and without excuse cancel my participation
without influencing my future care.
(Direct translation from Swedish).
Supporting Information
Supporting Information S1 Additional figures, tables, a
note regarding technical and twin reproducibility in the
metaproteomes and peptide-level false discovery rates.
(PDF)
Table S5 Normalized total spectra counts across all
subjects and 24 MS runs for the matched metagenome
(MM) database searches.
(XLS)
Table S6 Normalized total spectra counts across all
subjects and 24 MS runs for the human microbial isolate
reference genome database (HMRG) searches.
(XLS)
Table S7 Human microbial isolate reference genome
database (HMRG) database components. 51 bacterial
isolates were downloaded from the JGI IMG human microbiome
project (IMG-HMP) into a single FASTA-formatted protein
sequence database.
(XLSX)
Table S8 Distribution of all normalized ‘unique’ spectra
counts (worksheet 1) for a metaproteome genus-level
comparison of all 24 MS runs against the HMRG
database. Three comparisons (worksheet 2–4) between different
phenotypes (healthy, ICD, and CCD) were performed with Wilson
Integrated Omics of Crohn’s Disease
PLOS ONE | www.plosone.org 12 November 2012 | Volume 7 | Issue 11 | e49138
rank sum: Q value (adjusted P value) less than 0.05, difference
between medians of the two conditions greater than 5, and more
than 4 runs with greater than 5 spectral counts. Only the ‘healthy’
versus ‘ICD’ comparison have several genera that are significantly
changed.
(XLSX)
Table S9 Core and unique microbial protein clusters
identified in the metaproteomes. Common core microbial
prot

Di niente Roberto, su questo argomento per momentanea mancanza di tempo vado un pò a traino, avevo trovato questi studi da un pò ma non potendoci dare uno sguardo approfondito preferisco che intanto ne discutiate voi, comunque sono io che ringrazio te per ravvivare e rimpinguare questa discussone, di epigenetica molti ne parlano ma si comprende che rimane un aspetto poco chiarito anche per la difficoltà dell'argomento. Io stesso ho poco più che una infarinatura e solo un libro letto a tal proposito. :ok:

Epigenetica per combattere l'epilessia infantile
Identificato il difetto alla base dell'epilessia associata all'instabilita' da triplette del gene Arx: una rara forma che si manifesta nei primi mesi di vita dei bambini maschi
http://www.chimici.info/epigenetica-per-combattere-l-epilessia-infantile_news_x_13657.html
Scoperto il difetto epigenetico alla base dell’epilessia con deficit di apprendimento nei bambini malati di una forma genetica maligna legata al cromosoma X. A identificare il link funzionale nel complesso labirinto dei difetti molecolari associati a questa malattia lo studio realizzato da un team di ricercatori dell’Istituto di genetica e biofisica ‘Adriano Buzzati Traverso’ del Consiglio nazionale delle ricerche (Igb-Cnr) di Napoli, finanziato dall’associazione francese ‘Jerome Lejeune’ e pubblicato sull’American Journal of Human Genetics.

“L’epilessia associata all’instabilità da triplette del gene Arx, attivatore della trascrizione del Dna, è una rara ma devastante forma di epilessia maligna, che si manifesta nei primi mesi di vita dei bambini maschi – ha spiegato Maria Giuseppina Miano dell’Igb-Cnr. “Le alterazioni di Arx innescano alterazioni a carico del gene bersaglio Kdm5c, la cui proteina, un regolatore epigenetico, svolge un ruolo fondamentale nello stabilire quali geni devono essere esclusivamente espressi per garantire un corretto sviluppo del cervello embrionale”.

La novità di questa ricerca consiste nella scoperta che “mutazioni del gene Arx danneggiano gravemente l’attivazione di KDM5C e, di conseguenza, il controllo epigenetico a valle, confermando una relazione diretta tra danno genetico e gravità della malattia -continua la ricercatrice -. Per comprendere le dinamiche alla base della patologia è stata analizzata quindi l’interazione irregolare tra i due geni e l’effetto che questo difetto ha sul differenziamento dei neuroni. Ciò ha consentito di identificare un’alterazione globale che avviene durante la maturazione dei neuroni GABAergici, una classe di interneuroni inibitori gravemente danneggiata nei bambini malati, provocando uno sbilanciamento funzionale dei geni”.

Questi bambini sin da piccoli presentano una disorganizzazione generale e caotica delle scariche elettriche con danni a carico della maturazione del cervello che, nell’arco di pochi anni, manifesta difetti di apprendimento invalidanti. “Ne deriva – spiega Maria Giuseppina Miano - la necessità di somministrare loro tempestivamente farmaci antiepilettici che forniscano benefici efficaci, prima che si completi lo sviluppo cognitivo. Purtroppo, ad aggravare ulteriormente il quadro, spesso i neonati con tali deficit non rispondono alla terapia antiepilettica convenzionale”. “In quest’ottica lo studio getta le basi per una ricerca mirata a sviluppare nuovi metodi di cura delle epilessie refrattarie, in grado di limitare i danni che le convulsioni producono sullo sviluppo neurologico.

Acetilazione degli istoni -esercizio fisico indotta- gioca un ruolo con il genoma
http://jp.physoc.org/content/588/6/905.full.pdf
[...] Dal momento che l'esercizio di resistenza si consiglia di avere un potenziale terapeutico di obesità, diabete di tipo II, sindrome metabolica e dei relativi co-morbidità, è interessante per chiarire i meccanismi che si attivano al momento dell'esercizio. Oltre alla canonica attivazione delle chinasi di segnalazione e fattori di trascrizione, è ormai evidente che l'esercizio fisico può anche giocare tag con il nostro genoma con rimodellamento della cromatina, aprendo nuove vie di ricerca e la creazione di obiettivi per gli interventi terapeutici a base di esercizio.