There are several critical requirements for successful identification (or exclusion) of specific alterations in expression profiles in brain samples from individuals with neuropsychiatric disorders. These requirements include antemortem neuropsychiatric assessment, short post-mortem intervals, preservation of RNA and protein integrity, and, because of likely heterogeneity, large numbers of cases and controls. The Department of Psychiatry of the Mount Sinai School of Medicine has been collecting brain samples from antemortem assessed subjects since 1985. The original focus was on Alzheimer disease (AD), but was expanded a decade ago to include psychiatrically ill patients, especially those with schizophrenia. By collaborating with—and providing clinical services for—long-term care facilities, the Brain Bank has established a system whereby residents of the facilities are followed for long periods and tested annually or semi-annually with an extensive array of neuropsychiatric instruments. Post-mortem intervals (PMI) are short (one-half less than seven hours and a one-sixth under three hours), and tissue from half the brain is immediately snap frozen for RNA and other studies. The remaining tissue is paraformaldehyde fixed and examined for a large battery of neuropathological changes in a blinded manner. Recently, RNA profiling, first using differential display and now using microarrays, has been undertaken in both the AD and schizophrenia cohort.
In AD we have been studying a large cohort (n=79) of cases chosen to particularly represent early cognitive decline (clinical dementia rating of 0–2), with the expectation that early markers of dementia will be identified. The sample cohort was selected from a group of 278 consecutively, rapidly autopsied subjects (between 1986 and 1997) who had been residents of the Jewish Home and Hospital in Manhattan and the Bronx, New York. A multi-step approach was applied to the assignment of clinical dementia rating (CDR) score based on cognitive and functional status during the last six months of life. Note that all patients with neuropathological lesions other than those of AD were excluded from further studies. Using this sample base, we are identifying molecular changes that occur very early in cognitive decline, before a diagnosis of definite AD. We are studying molecular changes in five different brain areas, including the middle frontal gyrus, superior temporal gyrus, the entorhinal cortex, inferior parietal lobule and the primary visual cortex. As part of these studies, we have begun to make use of cDNA arrays for RNA profiling in brain tissue to identify changes in gene expression that occur with the earliest changes in cognition. Our initial focus for RNA profiling is the middle frontal gyrus, an area free of neuropathology in early dementia but where we have identified changes in total amyloid-β protein (Aβ) levels. The Aβ we measure represents soluble and perhaps preaggregated Aβ and apparently increases before any neurofibrillary changes or amyloid plaque formation. Our first studies are being carried out in ten controls (CDR 0), ten individuals with questionable dementia (CDR 0.5) and ten individuals with mild dementia (CDR 1.0). Note that the controls are age and gender matched, are derived from the same population as the cases and have undergone the same antemortem and post-mortem assessments.
This is a preview of subscription content, access via your institution