A systems biology network analysis of nutri(epi)genomic changes in endothelial cells exposed to epicatechin metabolites

Although vasculo-protective effects of flavan-3-ols are widely accepted today, their impact on endothelial cell functions and molecular mechanisms of action involved is not completely understood. The aim of this study was to characterize the potential endothelium-protective effects of circulating epicatechin metabolites and to define underlying mechanisms of action by an integrated systems biology approach. Reduced leukocyte rolling over vascular endothelium was observed following epicatechin supplementation in a mouse model of inflammation. Integrative pathway analysis of transcriptome, miRNome and epigenome profiles of endothelial cells exposed to epicatechin metabolites revealed that by acting at these different levels of regulation, metabolites affect cellular pathways involved in endothelial permeability and interaction with immune cells. In-vitro experiments on endothelial cells confirmed that epicatechin metabolites reduce monocyte adhesion and their transendothelial migration. Altogether, our in-vivo and in-vitro results support the outcome of a systems biology based network analysis which suggests that epicatechin metabolites mediate their vasculoprotective effects through dynamic regulation of endothelial cell monocyte adhesion and permeability. This study illustrates complex and multimodal mechanisms of action by which epicatechin modulate endothelial cell integrity.


Microcirculation mouse model -Intravital fluorescence microcopy
Mice were anesthetized by an intraperitoneal (i.p.) administration of ketamine (60 µg/g body weight) and xylazine (10 µg/g body weight). Upon surgical anesthesia, the entire back of the mice was shaven and a chemical hair remover was used to remove any remaining hair. The technique of implanting an Aluminium access chamber in the dorsal skin flap of rats, originally described by Papenfuss et al. (Papenfuss et al, 1979), was modified and adapted to meet carefully established criteria of stability (thin, translucent skin) in mice. Therefore, spacers made of stainless steel were used yielding in a frame-to-frame distance of 400-450 µm while the chamber was affixed in a position perpendicular to the animal's back. The thickness of the preparation and thus the distance between both frames were chosen on the basis of optical translucency needed for proper visualization of microscopic blood channels without compressing larger blood vessels. This chamber was tolerated by all animals.
The surgical procedure was done by placing the animal on a surgical stage. A midpoint line was drawn along the back and a dorsal skin fold was affixed to the surgical stage by means of silk sutures (5-0) to stabilize the preparation. One part of the chamber frame consisting of the frame, two lower bolts and spacers was slightly pushed under the skin-fold causing the skin to protrude.
Two holes were cut carefully through the skin, this part of the chamber introduced through the holes and fixed using two baby mosquito hemostats. A template, equivalent to the outer diameter of the chamber's collar served to mark a circle outlining the subsequent incision. A crisscross cut resulting in four skin flaps was made, and each of the four flaps was removed with a fine, curved scissor tracing precisely the perimeter of the outline with an effort to follow the hypodermis. The area exposed was freed from overlaying fascia utilizing an operating microscope leaving one layer of s.c. tissue at the opposite side of the epidermis intact. During this surgical procedure, the area was kept moist by allowing drops of warmed normal saline to irrigate the preparation that was limited to one side of the skin-fold only. After trimming the area under observation, the matching side of the chamber was inserted. Mechanical connection of the two chamber frames was accomplished using metal nuts for top and lower bolts. At seven points, the 5-0 sutures were fixed around the chamber to support the position of the skin flap. Animals were allowed for recovery for 72h before first intravital fluorescence microscopy was performed. Pain medication (buphrenorphine 0.05mg/kg body weight) was administered subcutaneously immediately after surgery and every 8 hours thereafter.
For the in vivo microscopy analysis, mice were immobilized and the skin fold preparation was attached to the microscope stage. After intravenous injection of 0.05 ml of 5% Fluorescein isothiocyanate-dextran (Sigma-Aldrich, Germany) which stains blood plasma, and in vivo leukocyte staining by 0.05 ml 1% Rhodamine 6G chloride (Invitrogen, Germany) injection in a tail vein or the retro-bulbar venous plexus, in vivo microscopy was performed using a Leica DM 4000M microscope. The observations were recorded by video camera and transferred to a video system for off-line evaluation. The microscope images were recorded on DVD and analyzed during playback by using CapImage (Dr. Zeintl, Germany). Microscopy was performed at baseline (72 h after application of dorsal skin fold chamber) and 12h after CLP induction.

Microcirculatory analysis
Stained plasma allowed the measurement of vessel diameters and red blood cell velocity (RBCV). RBCV was assessed in the centerline of the respective vessel by frame-to-frame analysis. Stained rolling leukocytes were defined as cells moving along the endothelial lining at a velocity markedly slower than that of the surrounding red cell and are given as percentage of the total number of calculated leukocytes passing through the observed vessel segment in 1 minute.

Sepsis induction
Mice were rendered septic by cecum ligation and puncture (CLP) (Merx et al, 2005;Merx et al, 2004) 6 h after baseline intravital fluorescence microscopy was performed. Anesthesia was induced by intraperitoneal (i.p.) administration of ketamine (60 µg/g body weight) and xylazine (10 µg/g body weight). Through a 1-cm abdominal midline incision, the cecum was ligated below the ileo-cecal valve with careful attention to avoid obstruction of the ileum or colon. The cecum was then subjected to a single "through and through" perforation with a 20-gauge needle. After repositioning of the bowel, the abdominal incision was closed in layers with standard silk surgical suture 4-0 (Ethicon, Somerville, New Jersey). Sham operated mice underwent the same procedure, except for ligation and perforation of the cecum. Pain medication (buphrenorphine 0.05mg/kg body weight) and volume support (NaCl 0.9%, 0.05 ml/g BW) were administered subcutaneously immediately after sepsis induction and every 8 hours thereafter. All mice had unlimited access to food and water. Twelve hours after sepsis induction, intravital microscopy was performed again to analyze the effects of microbial sepsis in microcirculation.   Supplemental information: western-blot gels for p38 and Phospho-p38. Gels were analyzed using Odyssey Li-Cor (Lincoln, Nebraska, USA) detection system and density of the signals was quantified using Odyssey software. TNF: tumor necrosis factor; MIX: mixture of studied flavanol metabolites; NR: samples not related to this work Supplemental Table S1: Site map property values and Dscore ranks for the p38 MAPK (Pdb id: 4F9Y)