Created on 05 June 2013

Because  of  the  high  number  of  RUs  participating to this project, details on specific activities and description of collaborations with foreign partners are deferred to Models B. A planning of the tasks is described below.

AI.1.  RECRUITMENT  OF INDIVIDUALS AND ADMINISTRATION OF DIARIES (RU1, 4, 6, 8; 1-6 months). Vegetarian and vegan individuals will be recruited with  the  cooperation  of  the  Italian  Scientific  Society  of  Vegetarian  Nutrition ( Omnivore individuals will be recruited through advertisements published at the Universities. Exclusion criteria will be fixed by RU4. About 50 healthy volunteers will be recruited by each of the 4 RUs, including an approximately equal number of omnivores, vegetarians and vegans (age 30-50 years, male/female ratio ca. 1:1). Recruited volunteers will be asked to sign a consensus document,  to  record  their  dietary habits and to collect biological samples (saliva, feces and urine). RU4 will be responsible to draft the diaries, which shall permit to get  detailed  information  on consolidated dietary habits and on food characteristics, allowing an easy identification of the products on the market and an estimation of the  presumptive  microbial load ingested. RU4 will elaborate the diaries at the end of period foreseen for the project. The Ethics Committee of the Universities will be informed  before  the  project  will  start.  Based on the possible withdrawal of individuals, the total number of volunteers recruited by the 4 RUs should be ca. 150 (50 omnivores + 50 vegetarians + 50 vegans). The statistical significance of 50 volunteers per each type of diet has been previously validated through Power Analysis.

AI.2.  HANDLING,  PREPARATION  AND  STORAGE  OF  BIOLOGICAL  SAMPLES  (RU1, 4, 6, 8; 1-6 months). Each individual will supply samples of saliva, feces  and  urine  weekly,  for  a  time  span of three weeks. Triplicate samples will be pooled before analyses in order to limit the intra-individual variability. Handling will  be  carried  out  differently  depending  on  the  type of biological samples (saliva, feces, urine) and subsequent analyses (e.g., DNA, RNA, proteome). Details are shown in the Models B of RU1, 4, 6 and 8.

AI.3.  MICROBIOLOGICAL  ANALYSES OF FECAL SAMPLES (RU1, 4, 6, 8; 1-6 months). The viable counts will be performed by plating fresh fecal material on different selective culture media, to enumerate the most common fecal microbial groups (39, 40).

AI.4.  SET  UP OF TECHNIQUES/METHODS (RU1, 2, 3, 5, 6, 7, 8, 9, 10; 1-6 months or more). Simultaneously with the recruitment of individuals and collection of biological  samples,  most  of the RUs will be involved in setting up the techniques/methods. These activities will vary depending on the specific tasks (see Models B). In order to establish an independent and competitive platform for meta-omics analyses, RU3 will purchase equipment for high throughput sequencing, which is a main economic investment in this project.

Based  on  the  information  from dietary diaries, the most representative foods of the 3 diets will be split in 3 categories: (i) low (TBC, £ 10^3 cfu/g); (ii)  intermediate (TBC  10^3  - 10^6 cfu/g); and (iii) high (TBC ³ 10^6 cfu/g) microbial load. For foods of group (i) the microbial number will be estimated based on literature data. For foods  of groups (ii) and (iii) the microbial number will be determined based on literature data in the case of well known products with low market differentiation (e.g., Parmigiano  Reggiano  cheese),  whereas  appropriate analyses will be done in the case of products with large market variability (e.g., Mozzarella cheese). In the latter case,  the analyses will involve a large number of products available in the market. The microbial diversity of foods will be studied more in depth first by PCR-DGGE and when needed by deep sequencing on the basis of level of complexity and current knowledge of the microbiota of the specific foods.

AII.1  QUANTITATIVE  DETERMINATION  OF MICROORGANISMS IN FOODS (RU 4, 5; 7-18 months). The analysis of foods will be shared between RU4 and 5,  and  some samples will be analysed by both RU to reciprocally validate the results. The quantification of the microbial groups commonly occurring in foods will be completed within 18 months.
Subsequently,  RU4  will  characterize selected foods by LH-PCR analysis (41), and the results will be compared with those obtained by RU7 using PCR-DGGE. RU5 will  identify  a  large  collection of isolates from foods, aiming at finding possible relationships with a reduced anti-genotoxic activity of the fecal waters and presence of specific microorgansims in the ingested foods.

AII.2  MICROBIAL  DIVERSITY  IN FOODS (RU 2, 3 and 7; 7-30 months). First, this activity will consider poorly characterized foods that are specifically eaten by veregetarians  and  vegans  based on records from diaries. Both micro- and mico-biota will be investigated. PCR-DGGE analyses for yeasts and filamentous fungi will be  based  on  the  amplification of parts of the D1/D2 domain encoding for the subunit LSU or 26S of the rRNA, and the ITS regions. The V3 region of the 16S rRNA gene  will  be  targeted  to  detect bacteria. After DGGE analysis, the resulting gels will be digitalised and analysed by the software Bionumerics. Dendrograms will be subjected  to  cluster  analysis  to  exclude from the further sequencing those samples having a coefficient of similarity ³ 85%; such selection will save costs of the next generation sequencing, and will assure to get a manageable quantity of data to be analyzed via bio-informatics. RU3  will  perform  the  deep  sequencing  using  libraries  of  amplicons of variable genes of taxonomic interest and using specific procedures according to the type of sequencing platform that will be purchased. The sequencing results will be elaborated on a bioinformatic basis by the RU2.

AII.3  FOOD  METABOLOME  (RU  10;  7-9  months).  Based  on  the  information on dietary habits, the fermented foods most represented in the three diets, will be analysed  to  find  a  possible  relationship  between  chemical  compounds  ingested  (e.g.,  salycilic  acid)  and  presumptively  recovered  in  biological  samples. The metabolome analyses will be carried out by GC-MS/SPME and FTIR spectroscopy.

Preliminarily,  all  saliva  and  fecal  samples  from  all  the  ca. 150 volunteers will be subjected to PCR-DGGE analyses to get an overview of the microbial diversity. RT-PCR-DGGE will be also considered to estimate viable populations in feces. Based on these analyses, representative samples of each dietary habit will be subjected to  next generation sequencing. The output of all these results should allow, step by step, the selection of 4/5 fecal samples for each of the three diets to be subjected to meta-omics analyses.

AIII.1.1. PCR-DGGE PROFILING (RU 2, 3, 8; 7-24 months). PCR-DGGE analyses for yeasts and filamentous fungi will target 26S rRNA and ITS regions, while V3 and/or  V6-V8  regions  of  the  16S  rRNA  will  be  studied  for  bacteria.  Both  DNA  and  RNA (after RT-PCR) will be used as target to investigate total and viable populations,  respectively.  Under  the  same  experimental  conditions, RU3 will characterize the bacterial diversity of saliva samples. After DGGE, image and cluster analyses  from  DNA and RNA samples it will be possible to define a difference between the global and viable population. The dendrograms of similarity will be used to  exclude from the further sequencing activity those samples having a coefficient of similarity ³ 85%; such selection will save costs of the next generation sequencing analyses, and will assure to get a manageable quantity of data to be analyzed via bio-informatics.

AIII.1.2.  NEXT  GENERATION  SEQUENCING  (RU  3,  8;  10-30  months).  Deep  sequencing  on selected saliva and fecal samples will be done using libraries of amplicons  of  variable  genes  of  taxonomic  interest  as  above  described. Results for selection of fecal samples for -omics analyeses will be completed within 12-24 months,  but  deep  sequencing  will  continue  further  to  optimise  the  results  from the screening. All sequencing results obtained both for bacteria and fungi will be elaborated on a bioinformatic basis by the RU2.

AIII.1.3.  ANTIBIOTIC  RESISTANCE  (RU  7;  13-30  months). DNA extracted directly from saliva and fecals samples will be screened for the occurrence of genes encoding  for  resistances  to  several  antibiotics  using  PCR  assays.  Samples,  resulted  to  be  positive  for  antibiotic  resistance  (AR) genes, will be used to isolate antibiotic-resistant  lactobacilli,  lactococci  and  enterococci.  After identification, isolates showing values of minimum inhibitory concentration (MIC) higher than the corresponding  breakpoints  will  be  confirmed  by  PCR  amplification  of  the corresponding AR genes. The possibility of transferring AR genes localized on genetic mobile elements from isolates to bacteria of clinical interest will be evaluated by conjugation trials.

According  to  the  cascade  approach,  the  results  from  the study of the microbial diversity of fecal samples will select 4/5 individuals representative of each of the 3 types of diet (4/5 x 3 = 12/15) to be subjected to meta-omics analyses. The meta-omics approaches will provide an in house database, consisting of the DNA and RNA sequences from the fecal microbiomes of the 12/15 individuals, which will allow a complete view of the synthesized proteins.

AIII.2.1.  META-GENOMICS  (RU3,  8;  13-30 months). Fecal DNA will be quantified and a shotgun sequencing protocol will be applied for meta-genomic analysis. This  task  should  be performed by the RU3, depending on performance and availability of the sequencing platform, which will be engaged full time in the analysis of DNA and cDNA amplicons of feces and saliva. Alternatively, specialized companies will carry out this task as a service.

AIII.2.2.   META-TRANSCRIPTOMICS   (RU   6;   13-30   months).   The   overall  microbial  gene  expression  profile  will  be  characterized  using  an  innovative Illumina-based  Metatranscriptomic  approach  and  the  protocol set up during first months of the project. cDNA will be sinthetised from RNA according to optimized procedures.  Random  amplification  of  the  cDNA  will  be  performed  and  the  cDNA  will  be  sequenced  by  using  the  Illumina  HiSeq  platform. The sequenced meta-transcriptomes  will  be  assembled  together  using  Velvet/MetaVelvet  or  SOAPdenovo  packages  to  achieve  the  largest  possible  consensus  sequence. The transcripts  sequences  will  be  annotated  using  well-established pipelines and transcripts will be searched using RPS-BLAST against KEGG, COG and the Genbank databases.  Bacteria-like  reads  identified  by  nr  BLASTX  will  be further searched against the COG database. The functional roles of the sequences will be assigned based on the KEGG and COG searches.

AIII.2.3.  META-PROTEOMICS (RU1; 13-30 months). Based on the workflow designed within the first months of activity, proteins will be extracted from each fecal sample  and  analysed  by  gel-free and/or gel-based proteomics. The identification of peptides will be performed using the mass spectrometer Finnigan LCQ Deca  XP MAX  (42).  Differential  proteomic  analysis  will  be  carried out. For peptide identification, the Open Mass Spectra Search Algorithm will be used to search MS/MS spectra  against  the  available  databases.  To retrieve further functional information, the proteins based on COG classification will be annotated. The identified  COGs will  be  mapped  on  KEGG  metabolic  pathways  database  and  visualized  by  the  online  application  of iPath. After FDR correction, all identified peptides will be searched  against UniProtKB and then mapped. To address the highest functional level of all the identified proteins, the identified COGs onto KEGG pathways will be mapped. Besides the common microbial core, inter-individual and inter-diet differences will be determined in terms of metabolic function.

The characterization  of  some  functional  features  of  fecal  samples  and of microbial isolates from fecal samples will strengthen the link between diet and intestinal microbiota.

AIV.1 FECAL GENOTOXIC AND ANTI-GENOTOXIC ACTIVITIES (RU 5; 7-30 months). Samples of fecal water (FW) will be prepared (43) and the genotoxicity will be determined by Comet assay (44). The HT29 enterocytes (10^6 cells/assay) co-exposed to FW and incorporated into LMA slide will be subjected to Single-Cell Gel-Electrophoresis  (SCGE)  and analysed by epifluorescence. Several microbial isolates from foods will be investigated for the inhibitory activity towards genotoxic and  mutagenic  compounds,  which  are  potentially  present  in  the  intestine.  The  effect  of the microorganism-genotoxin co-incubation will be assayed through the evaluation  of  the  residual  genotoxic activity of the compounds using the SOS-Chromotest (target Escherichia coli PQ37 sfiA:lacZ) and, in parallel, the Comet assay (target enterocytes HT29) (43).

AIV.2. FECAL MICROORGANISMS AND MODULATION OF THE IMMUNE RESPONSE (RU 9; 7-30 months). The conditions for growth and differentiation of dendritic cells  (DC) and Caco-2 cells will be set up within the first months of activity. Lactobacilli and bifidobacteria will be isolated from all fecal samples and used as  irradiated  bacteria  to  detect  the expression of immune mediators in enterocytes. Transcript and cognate protein secretion levels of IL-8 and tolerogenic TGF-beta and  TSLP  will  be  determined.  Irradiated  bacteria  will  be also used to stimulate surface markers in DC cells. Total RNA will be extracted from the Caco-2 and DC cells,  and  cDNA  will  be  prepared  to  look  at  the  relative  gene  expression  of  IL-8,  TGF-beta,  TSLP,  TNF-alpha, IL-12p40 and IL-10; the concentration of the corresponding gene products will be determined by ELISA.

To  complete  the  “omics”  approach,  the  metabolome  of  samples  of feces, urine and saliva will be analysed with three techniques (GC-MS/SPME, FTIR/ATR and NMR)  according  to the protocols previously set up. Chemical compounds will be identified using mass spectra databases as well as mass spectra data in the literature and/or  data  from  pure  chemical  compounds.  Upon  thawing  of  urine  and  saliva  samples,  the  NMR analysis will be performed within 2 h. Fecal samples will be prepared  giving  the  maximum  importance  to  the  complete  recovery  of  water-soluble  molecules.  The  NMR  spectra  will  be prepared for statistical analysis by correcting the little peaks misalignments through iCoshift algorithm (45).

During  project  development,  RU2  will  take  care  of  the maintenance of the website in collaboration with the partner CBS (Centraalbureau voor Schimmelcultures, Utrecht,  NL),  the  site  will  collect  all  the  results  from  the  different research activities. This database will be used for analysis in "R mode" to assess the impact of various  descriptors  in  determining  the  diversity  of the three diets, and in "Q mode" to determine difference between the three diets. During the same time, the most relevant results will be disseminated through scientific publications into peer-reviewed journals.Research activities.