Intestinal mucosal adherence and translocation of commensal bacteria at the early onset of type 2 diabetes: molecular mechanisms and probiotic treatment

A fat-enriched diet modifies intestinal microbiota and initiates a low-grade inflammation, insulin resistance and type-2 diabetes. Here, we demonstrate that before the onset of diabetes, after only one week of a high-fat diet (HFD), live commensal intestinal bacteria are present in large numbers in the adipose tissue and the blood where they can induce inflammation. This translocation is prevented in mice lacking the microbial pattern recognition receptors Nod1 or CD14, but overtly increased in Myd88 knockout and ob/ob mouse. This ‘metabolic bacteremia’ is characterized by an increased co-localization with dendritic cells from the intestinal lamina propria and by an augmented intestinal mucosal adherence of non-pathogenic Escherichia coli. The bacterial translocation process from intestine towards tissue can be reversed by six weeks of treatment with the probiotic strain Bifidobacterium animalis subsp. lactis 420, which improves the animals' overall inflammatory and metabolic status. Altogether, these data demonstrate that the early onset of HFD-induced hyperglycemia is characterized by an increased bacterial translocation from intestine towards tissues, fuelling a continuous metabolic bacteremia, which could represent new therapeutic targets.


Microscopic visualization and quantification of GFP-E. coli in ileal mucosa
Two or five hours after gavage with 10 9 GFP-E. coli, ileal mucosa was scrapped off, spread onto glass slides and observed by fluorescence microscopy for image acquisition (Microscope Zeiss Axio Observer Z1 with AxioVision 4.7.2 acquisition software, Carl Zeiss MicroImaging GmbH, Germany). GFP-E. coli were enumerated in 10 different fields per mouse.

Immunodetection of dendritic cells in ileum and mesenteric lymph nodes
Two hours after gavage with 10 9 GFP-E. coli, mesenteric lymph nodes (MLN) and a segment of ileum were formalin-fixed and frozen. Cryostat sections (8µm) were incubated with hamster anti-mouse CD11c as primary antibody (clone HL3; BD Biosciences Pharmingen, California, USA; 1/1,000, overnight, room temperature), biotin mouse anti-hamster IgG as secondary antibody (BD Biosciences Pharmingen, California, USA; 1/1,000, 1h 30 min, room temperature), then with Texas Red dye-conjugated streptavidin (Jackson ImmunoResearch Laboratories, Pennsylvania, USA, 1/1,000, 30 min in the dark at room temperature) and finally counterstained with 4,6-diamidino-2-phenylindole (200 ng/mL, Sigma-Aldrich, France). Slides were then observed by using a Zeiss Axio Observer Z1 fluorescence microscope (Carl Zeiss MicroImaging GmbH, Germany). Images were collected by using a Zeiss AxioCam MRm digital camera and AxioVision 4.7.2 acquisition software (Carl Zeiss MicroImaging GmbH, Germany) and merged with Adobe Photoshop CS3 Extended 10.0 (Adobe Systems Inc., California, USA). GFP-E. coli appear in green, cell nuclei in blue, and CD11c-positive cells in red. CD11c-positive cells co-localized with GFP-E. coli are yellow. CD11c-positive cells co-localized with GFP-E. coli in ileum and MLN, were enumerated in 10 different fields per mouse.

Quantification of GFP-E. coli in blood by RT-qPCR
Blood was collected from the inferior vena cava 2 hours after gavage with 10 9 GFP-E. coli. Blood cells were first mechanically disrupted by mini-beads vibrating at 30 hertz for 2 x 3 min on a Retsch Tissue Lyser II (Qiagen GmbH, Germany) with 200 mg of acid-washed glass beads (Ø < 106 µm; Sigma-Aldrich, France) in Tripure isolation reagent before total mRNAs were extracted according to manufacturer's procedure (Roche Applied Science, Germany). Contaminating DNA was eliminated by DNase I (RNase-free, Ambion Ltd, UK) and the DNase I was removed by a second phenol/chloroform extraction with Tripure, prior to reverse transcription using eAMV-RT (Sigma-Aldrich, France). Primers specific for GFP were designed for this study and were as follows: GFP forward, 5'-CTACCTGTTCCATGGCCAAC-3'; GFP reverse: 5'-AGGGTATCACCTTCAAACTTGACT-3'. The qPCR assay was performed from the cDNAs with a Stepone Plus Real-Time PCR System and Power SYBR Green PCR Master Mix (Applied Biosystems, California, USA). Data were analyzed by absolute quantification using a standard curve of cDNA synthesized from RNA obtained from a pure culture of GFP-E. coli, in order to express the results as cfu-equivalent/mL blood.

Quantification of bacterial RNA in mesenteric adipose tissue by RT-qPCR
Mesenteric adipose tissue (MAT) samples were processed for mRNAs extraction as described for blood, with additional homogenization in Tripure isolation reagent with Ultra-Turrax before cells disrupting by mini-beads vibrating. Contaminating DNA was eliminated by DNase I (RNase-free, Ambion Ltd, UK) and DNase I was removed by a second phenol/chloroform extraction with Tripure, prior to reverse transcription using eAMV-RT (Sigma-Aldrich, France). Total bacterial RNA was quantified from synthesized cDNAs, using EUBAC 16S rRNA universal primers already described above for total bacterial DNA quantification. Data were analyzed by absolute quantification using a standard curve of cDNA synthesized from RNA obtained from a pure culture of GFP-E. coli.

Denaturing gradient gel electrophoresis (DGGE) of bacteria in caecum, blood and mesenteric adipose tissue
Genomic DNA was isolated from caecal contents, blood, or mesenteric adipose tissue (MAT), using Tripure isolation reagent (Roche Applied Science, Germany) as described above for bacterial DNA quantification by qPCR in the same tissues. The variable V3 region of the 16S rRNA gene was amplified by final point PCR with primers amplifying the conserved V3 surrounding region. The nucleotide sequences of the primers are as follows: HDA1-GC, 5'-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGGCCTACGGGAGGCAGCAG-3' (GC clamp in bold); HDA-2, 5'-ATTACCGCGGCTGCTGG-3'. PCR amplification was done with the thermal cycler VWR Unocycler (VWR International, Belgium). The samples were first incubated for 5 min at 94°C to denature the DNA template and subsequently cooled down to 80°C, at which point 0.35 U of Taq DNA polymerase (Sigma-Aldrich, Saint-Louis, USA) was added: this hot start technique minimizes non-specific annealing of primers to non-target DNA. Twenty cycles were then carried out as follows: 94°C for 30 sec, annealing for 45 sec and 72°C for 1 min. The annealing temperature started 10°C above the expected annealing temperature and was decreased by 1°C every second cycle until a touchdown at 55°C, this procedure reducing the formation of spurious by-products during the amplification process. Ten more cycles were then carried out at 55°C as annealing temperature and the final primer extension was carried out at 72°C for 30 min. Amplification products were first analyzed by electrophoresis in 6% acrylamide gels and EtBr staining to check for the presence of the 233bp expected product. DGGE was performed with a DGGE-2401 system (CBS Scientific Company, California, USA). Electrophoresis was carried out with 0.75mm-thick 8% polyacrylamide gels (ratio of acrylamide to bisacrylamide, 37.5:1) submerged in TAE buffer (40 mM Tris, 20 mM acetic acid, 1 mM EDTA; pH 7.4) at 60°C. The electrophoresis was run for 16 h at 60 V in a linear 35 to 55% denaturing gradient with 100% denaturant defined as 7 M urea and 40% deionized formamide. The gels were stained for 1 h in TAE buffer with Sybr Safe DNA gel stain (Invitrogen, Eugène, USA), visualized by using a Typhoon 9400 variable mode imager fluorescent scanner and analyzed with ImageQuant TL software (v2003; Amersham Biosciences, Little Chalfont, England).

Quantification of bacterial groups in mesenteric adipose tissue and ileal mucosa by qPCR
Bacterial DNA was extracted from mesenteric adipose tissue (MAT) and ileal mucosa by first disrupting tissues and cells by milling with ceramic (ø 1.4 mm) and glass beads (ø 0.1 mm) (6800 rpm, 3 × 30 s; Precellys 24, Bertin Technologies, Montigny, France), followed by extraction and purification of DNA with the QIAamp Stool kit (Qiagen, Leiden, The Netherlands) according to manufacturer's instructions. Quantitative real-time PCR (Applied Biosystems Fast 7500; Foster City, CA, USA) was used to quantify Bacteroides-Prevotella-Porphyromonas group (Rinttila et al, 2004), genus Enterococcus (Rinttila et al, 2004), Enterobacteriaceae group (Matsuda et al, 2007) and genus Lactobacillus (Makivuokko et al, 2005) from MAT and the genus Bifidobacterium (Makivuokko et al, 2005) from ileal mucus as previously described.

Fasted plasma insulin concentration
Plasma insulin concentrations were determined using an enzyme-linked immunosorbent assay (ELISA) kit (Mercodia, Upssala, Sweden) at completion of the fasting period (6 h) in 10 µL of plasma collected from tail blood.