Vascular Dysfunction in Paediatric IBD

Plan of Investigation:

Patients: 130 children and adolescents (8-21y) with an expected ratio of 60% (n=78) Crohn's disease (CD), 35% (n=45) Ulcerative colitis (UC) 5% (n=6) Indeterminate colitis (IC). 78 age and sex-matched controls will be investigated. Sample size calculations are based on Circulating Endothelial Cells (CECs) as the primary end-point, as suggested by an on-going study by one of the co-applicants, on healthy children and children with Kawasaki disease (Brogan P et al, ref published).

40 subjects/ group are required to detect a doubling average of CECs in CD and UC vs. control with 90% power, significance 0.05 and this should be achievable by our initial recruitment. Non-normality and the need to use non-parametric or transform prior to analysis, increases the number to 47; hence we will aim to recruit 50 to the UC group. This will provide adequate power, and is feasible based on the clinical cohort available to us.

This patient cohort is an appropriate candidate group for the present investigation as

1. Evidence indicates that vascular changes that occur between 10-20 years of age are a critical determinant of future vasculature health;

2. Initiation into smoking habits is most likely in teenage years and

3. Other established traditional risk factors for atherosclerosis (such as hypertension, type II diabetes, or even fully established atherosclerotic disease) that would act as confounding variables, are not yet present in adolescence.

Data will be collated onto a de-identified excel sheet and will include age, sex, age at diagnosis, coronary artery status at presentation, growth, body mass index, blood pressure, family history of CVD, and smoking status.

Aim 1: Do Children with IBD have evidence of a MP mediated prothrombotic tendency? MPs will be identified by flow cytometry as previously described by our group19. Briefly, platelet-poor plasma (PPP) will be obtained from blood and stored at -80 °C for future batch testing. 200 μL of PPP will allow sedimentation of MPs which will be resuspended in Annexin V (AnV) binding buffer prior to staining with FITC- or phycoerythrin -AnV (BD PharMingen). Endothelial, platelet and neutrophil-derived MPs (EMPs/PMPs/NMPs) and Tissue-Factor (TF) will be enumerated by detection with anti-human (CD62E, CD41 and CD11b activation epitope which binds to activated neutrophils respectively, plus relevant isotype controls). Latex beads (1.1 μm) are used to gate MPs < 1.1 μm. The thrombotic potential of MPs will be quantified by suspending MPs in control microparticle-free plasma (MPFP) containing trypsin inhibitor [inhibits contact activation] followed by exposure to calcium-fluorogenic substrate (Z-G-G-R-AMC). Kinetics of thrombin generation will be recorded up to 90min post-stimulation. Lag time min, peak thrombin nM, velocity index nM/min and endogenous thrombin potential (ETP) nM × min will be calculated. To investigate the relative contribution of PS and TF to MP-mediated thrombin generation, MPs will be pre-incubated with increasing concentrations of recombinant AnV protein or a blocking anti-TF (or isotype control) prior to thrombin analysis.

Aim 2. Do Children with IBD have evidence of Endothelial Injury? In addition to investigating arterial health (below), we will quantify vascular injury by measuring CECs. CECs will be isolated by immunomagnetic bead extraction (based on an international consensus protocol) and counted using a Nageotte chamber/fluorescence microscopy. CEC enumeration is defined as Ulex-europaeus-lectin bright cells of >10 μm in size, with 5 magnetic beads attached19. Data will be analysed in the context of EMP data (Aim 1) as the combination will give an insight into the degree of endothelial injury in IBD versus control.

Aim 3. Do Children with IBD have evidence of Structural Arterial Disease? Pulse Wave Velocity (PWV) will be an indicator of arterial structural health. Pressure waveforms will be recorded simultaneously at two sites (carotid-femoral) using the VICORDER analysis software (Skidmore Medical Limited);

Aim 4. What is the relationship between indices of inflammation and established mediators of vascular Injury? Levels of hs-CRP, serum amyloid A (SAA), TNF-α, IL-1α, IL-1β, IL-6, MCP-1, VEGF, fasting lipids and angiopoietin 1/2 will be correlated with validated clinical assessment of disease activity [Paediatric UC Activity Index20 (PUCAI) & Paediatric CD Activity Index (PCDAI)] in addition to conventional markers (CRP, ESR, D-Dimers and platelets in active and inactive disease). In many cases, conventional circulating markers do not correlate with endoscopic findings in active disease; the non-conventional markers may show a higher sensitivity in detecting those with on-going active inflammation.

Methodology:

1. Clinical data collection: Subject clinical data will be collated onto a de-identified excel spread sheet from a paper data collection proforma as used in the pilot study. Clinical data collected will include age, sex, age at IBD diagnosis, coronary artery status at presentation and currently, growth and body mass index, blood pressure, family history of cardiovascular disease, and smoking status.

2. Chronic inflammation will be assessed using Meso Scale Discovery (MSD, Maryland, USA) multi-array electrochemiluminescence detection of plasma: hs-CRP, SAA, TNF-α, IL-6, 8 and 10, MCP-1, VEGF, angiopoietin 1 and 2 (the latter mediating endothelial detachment in systemic vasculitis (66). This technique allows many parameters to be assayed from very small amounts of blood and was used (as per manufacturer instructions) with great success in the pilot study; d. Endothelial activation/injury will be assessed by: CEC count quantified by immunomagnetic bead (IB) isolation from whole blood using anti-CD146 coated beads, then stained with FITC-conjugated Ulex and enumerated using fluorescent microscopy and a Nageotte counting chamber as previously described by our group and others (46;67); EMPs expressing CD105, E-selectin, ICAM-1, VCAM-1 CD144, CD31, but negative for the platelet markers CD42a, or CD62P will be quantified from platelet poor plasma using a BD FACSArray flow cytometer as previously described by our group (44); e. Endothelial repair potential will be assessed in a total of 45 subjects (15 from each group: KD CAA+, CAA-, and healthy controls) by assessing (i) EPC colony forming unit (CFU) capacity (68), and (ii) EPC potential to incorporate into HUVEC vascular structures in matrigel (69), and now routinely performed by our group (see below). EPCs are prepared from PBMCs isolated from 5ml of blood are plated into culture dishes coated with human fibronectin and maintained in EGM-2 supplemented with 20% fetal calf serum and 40ng/ml VEGF. After 4 days in culture, non-adherent cells are removed by washing with phosphate buffered saline and the adherent cells maintained in culture until day 7. Early EPC colony forming units are then counted. Amplified EPCs harvested as described above are then labelled with Di -I-LDL and replated with HUVEC on top of a solidified matrigel layer and incubated at 37 °C for 24 h. Fluorescent microscopy is used to assess incorporation of EPCs into the HUVEC capillary lattice, and HUVEC tubule formation (structure exhibiting length four times its width) will be compared using EPC from KD (+/- CAA) to healthy controls. Five independent fields are assessed for each well, and the mean number of tubules/×100 field are determined. f. Vascular stiffness and carotid IMT will be assessed by: (i) carotid-femoral and carotid-radial pulse wave velocity (PWV) using the Vicorder device (Skidmore Medical Limited); All these techniques are routinely performed by our group in accordance with AHA recommendations (70), as illustrated by our pilot data and our previously published studies (71-73).