Proteomic-Based Approach for Biomarker Discovery to Predict Silent Cerebral Infarct in Patients with Sickle Cell Disease.

Silent cerebral infarction (SCI) is part of a spectrum of cerebrovascular disease, occurs in approximately 22% of children with sickle cell disease (SCD) and is associated with decreased cognitive function. While several plasma biomarkers have been shown to be elevated in acute stroke, to our knowledge none have been evaluated in SCD or SCI. The aim of this study was to develop a reliable pipeline to identify low abundance plasma proteins that correlate with SCI in patients with SCD. We used a proteomic discovery approach involving three sequential separation steps in order to compare the plasma proteomes in a discovery set of 15 children with SCD (7 children with SCI and 8 children without SCI). Baseline steady state plasma samples obtained from the Silent Infarct Transfusion (SIT) Trial Biologic Repository were matched for age and WBC. In the first dimension, hemoglobin was depleted with nickel-nitrilotriacetic acid (Ni-NTA). Subsequently, second dimension separation and enrichment was achieved by immunoaffinity depletion of the 12 most abundant proteins (ProteomeLab IgY-12 LC10) followed by third dimension separation by reverse phase liquid chromatography fractionation (RPLC) using a C18 column and a linear acetonitrile gradient. Collected fractions were subjected to tryptic digestion and analyzed using label-free quantification on a LTQ-Orbitrap (Thermo) mass spectrometer. The MS/MS data were analyzed using the Proteomics Analyzer Software System (PASS) version 4.0.10 (Integrated Analysis Inc, Bethesda, MD) with X! Tandem searches (www.thegpm.org; version 2008.12.01) using the human IPI database. Identified proteins were compared to databases of brain specific and brain enriched proteins to identify candidate biomarker proteins for SCI. After hemoglobin depletion, 71% of total protein was removed. On average, protein recovery after the LC-12 column was 4% of the total Ni-NTA depleted protein. Of the 9800 proteins that were identified in the plasma proteome of children with SCD, 23 were brain specific proteins. Evaluation of the relative abundance by spectral counts (SC) revealed 3 brain-related proteins that were over-represented in patients with SCI: microtubule-associated protein tau (a neurofibrillary tangle protein implicated in Alzheimer disease, frontotemporal dementia and parkinsonism), neuroligin-3 (a neuronal cell surface protein proposed to be involved in cell-cell-interactions via binding to beta-neurexins and implicated in autism and Asperger syndrome), and nucleosome-remodeling factor subunit BPTF (a histone-binding component of nucleosome-remodeling factor that is abundantly expressed in the fetal brain and re-expressed in neurodegenerative diseases). Reticulon-4, a potent neurite growth inhibitor involved in the restriction of axonal regeneration after injury, was under-represented in patients with SCI. After depletion of hemoglobin and other high abundant proteins, we were able to develop a database of plasma proteins in children with SCD and to identify brain specific proteins as potential surrogate markers of brain injury. These markers may be implicated in the pathophysiology of SCI. Although validation studies are necessary to determine the relevance of these candidate biomarkers in SCI and SCD, our methodology appears to be a practical approach to proteomic discovery in patients with hemolytic anemia.

Casella:Boehringer Ingelheim: Honoraria, Research Funding, Travel funding.