Affiliations: [a] Department of Pharmaceutical Sciences, University of Southern California, Los Angeles, CA 90033, USA | [b] Research Center for Alcoholic Liver and Pancreatic Diseases, University of Southern California, Los Angeles, CA 90033, USA | [c] Department of Computational and Molecular Biology, University of Southern California, Los Angeles, CA 90089, USA | [d] Center for Dementia Research, Nathan Kline Institute, Orangeburg, NY 12053, USA | [e] GeneGo, Inc., 500 Renaissance Drive, Suite 106, St. Joseph, MI 49085, USA
Correspondence:
[**]
Corresponding auhto: Austin Yang, Ph.D., University of Southern California, School of Pharmacy, Department of Pharmaceutical Sciences, Los Angeles, CA 90089, USA. Tel.: +323 442 4118; E-mail: austiny@usc.edu.
Note: [*] These authors contributed equally to this work.
Abstract: Increasing evidence suggests that oxidative injury is involved in the pathogenesis of many age-related neurodegenerative disorders, including Alzheimer's disease (AD). Identifying the protein targets of oxidative stress is critical to determine which proteins may be responsible for the neuronal impairments and subsequent cell death that occurs in AD. In this study, we have applied a high-throughput shotgun proteomic approach to identify the targets of protein carbonylation in both aged and PS1+AβPP transgenic mice. However, because of the inherent difficulties associated with proteomic database searching algorithms, several newly developed bioinformatic tools were implemented to ascertain a probability-based discernment between correct protein assignments and false identifications to improve the accuracy of protein identification. Assigning a probability to each identified peptide/protein allows one to objectively monitor the expression and relative abundance of particular proteins from diverse samples, including tissue from transgenic mice of mixed genetic backgrounds. This robust bioinformatic approach also permits the comparison of proteomic data generated by different laboratories since it is instrument- and database-independent. Applying these statistical models to our initial studies, we detected a total of 117 oxidatively modified (carbonylated) proteins, 59 of which were specifically associated with PS1+AβPP mice. Pathways and network component analyses suggest that there are three major protein networks that could be potentially altered in PS1+AβPP mice as a result of oxidative modifications. These pathways are 1) iNOS-integrin signaling pathway, 2) CRE/CBP transcription regulation and 3) rab-lyst vesicular trafficking. We believe the results of these studies will help establish an initial AD database of oxidatively modified proteins and provide a foundation for the design of future hypothesis driven research in the areas of aging and neurodegeneration.
