Antibody Microarray Study of CSF Samples From Depressed Patients ? 1.0 Introduction and Specific Aim:? Our specific aim has been to identify prospective biomarkers for disease and to develop candidate targets for drug discovery.? 1.1. Rationale:? The medical rationale is based upon the understanding that a core proteome governs neuropsychiatric diseases and that an in depth knowledge of these proteins is fundamental to understanding their roles in health and disease. ? 1.2. Overview:? The mechanism by which only some Hepatitis C patients develop clinical depression after treatment with Interferon alpha (IFNa) is not known. However, we anticipated that depression-specific CSF protein biomarkers in this class of patients might be indicative of biomarkers for the general class of depressed patients. The biological fluids of patients with specific diseases often contain proteins that have been leaked from the specifically diseased cells. Examples are lipases and amylase in pancreatitis; SGOT/SGPT in hepatitis; troponin I in myocardial infarction; PSA from BPH prostate tissue, and many others. Based on the depression literature, we anticipated finding changes in certain serotonin receptors; protein kinase Cs; glutamate receptors; TNFa proteins, and other proinflammatory signaling proteins. The antibody microarray chosen for this pioneer analysis contained 507 features antibodies in duplicate, covering more than 30 signaling pathways, many transcription factors, and many chaperones.? Only recently has the antibody microarray technology become sufficiently mature for reproducible, robust detection of low abundance cellular proteins Fathman etal., 2005; Coombes etal., 2005; MacBeath etal., 2001; MacBeath, 2002; Srivastava etal., 2006. The advantage of this new technology is that it is an intrinsically robust and quantitative system, delivering high throughput and parallel detection. Antibody microarrays have yet to be applied to low abundance proteins in cerebrospinal fluid (CSF). ? 2.0 Experimental Plan: ? CSF samples from Hepatitis C patients all treated with Interferon a have been pooled and compared with equivalent pools of control CSF in order to identify patterns of protein expression unique to depression. Using highly specific and validated antisera, we are using an antibody microarray that distinguishes between CSF of depressed and non-depressed patients. ? 3.0 Materials and Methods:? 3.1. Patients: Based on the available set of CSF samples, we determined that the male patients could be used for this pathfinder study. All patients had hepatitis C (HC). The CSF from the 4 patients from each of the 3 groups indicated below was pooled. ? 3.1.1. Male HC Patients (=Controls, n=4) were non-treated.? 3.1.2. Male HC Patients treated with Interferon a, who did not become depressed (=IFN,n=4).? 3.1.3. Male HC Patients treated with Interferon a, who did become depressed (=IFN,n=4).? 3.2. CSF sample acquisition: CSF samples were obtained from the NIMH Mood and Anxiety Disorders Program (Dr. Husseini Manji).? 3.2.1. Antibody microarray technology? The antibody microarrays used in these experiments were obtained from BD Biosciences Clontech Antibody Microarray, and handled as recommended by the manufacturer. Each of the 3 classes of samples was pooled by volume. Each composite sample was then labeled with Cy3. A parallel sample of gold standard cell extract was labeled with Cy5. Both were mixed and incubated with antibody microarrays. Each feature was then measured for binding using the GENEPIX reader and proprietary MATLAB-based software for analysis. Following incubation and washing, the dried slides were imaged on a Typhoon 9410 Variable Mode Imager. The data were then exported offline for analysis using proprietary algorithms. ? 3.2.2. Bioinformatics and Statistics? Intensity values of spots on the microarrays were normalized to the median of all spots of the reespective color (Cy3 or Cy5) on the same array. The amount of each protein was calculated by averaging the normalized intensities for that protein. In essence, this approach gives the protein levels normalized to the median of the 507 proteins on the array in each given sample. In order to compare the absolute levels, we calculated the absolute protein level ratios using the manufacturers protocol (the square-root method). Rank-order non-parametric methods were employed for identifying the candidate CSF proteomic signature. ? 4.0 Preliminary Data:? In this work we show that an antibody microarray of 507 antibodies against low-abundance intracellular proteins can quantitatively discriminate between CSF from depressed and non-depressed interferon a treated patients. ? The data (see LIST) shows a selection of a few of the ca.30 significantly depression-specific candidate biomarkers derived from the complete analysis.? There are insignificant differences between control and IFNa-treated patients who don't have depression, but significant differences when comparing IFNa-treated depressed vs. not depressed patients.? ? B-cell CLL/lymphoma 2? Bcl-2-asscociated transcription factor? Tumor necrosis factor-alpha? heat shock transcription factor 4? discs, large homolog1? son of sevenless homolog1? ADP-ribosyltransferase? insulin-like growth factor binding protein 3? disabled homolog 2, mitogen-responsive phosphoprotein? DnaJ (HSP40) homolog, subfamilyA, member1? proteosome 26S subunit, ATPase, 5? nuclear factor of activated T-cells, calcineurin-dependent 2? symplekin? syntaxin binding protein? phosducin-like? interleukin 5? general transcription factor IIF, polypeptide 2? chromogranin B (secretogranin 1)? survival of motor neuron protein interacting protein 1? megakaryocyte-associated tyrosine kinase? flotilin 2? baculoviral IAP repeat-containing 6? branched chain aminotransferase 1, cystolic? ? The rationale for this paradigm shift is compelling evidence in the psychiatric field that a CSF/serum protein signature consisting of multiple proteins is a considerably more powerful predictor of clinical state than considering individual proteins one at a time.? Our approach is to identify and characterize patterns of protein expression in the CSF that are predictive of psychiatric diseases such as depression and bipolar disease. Our plan is to define and construct a clinical database and repository consisting of clinically annotated CSF patient samples.? A potential therapeutic benefit from the development of a CSF proteomic signature will be its use as a surrogate endpoint for the development of new therapeutic approaches to CSF.
