Evidence supporting sex steroids as being essential for normal cognitive health includes: 1) 12 studies demonstrating improvements in cognitive performance of cognitively healthy postmenopausal women taking 17?-estradiol (E2), the major female estrogen, 2) improvements in cognition in subjects treated with physiologically relevant sex steroids: women treated with E2 (3 controlled/1 uncontrolled studies) and men treated with testosterone (T;2 controlled studies), 3) the abrupt cognitive deficits observed in premenopausal women following 'chemical castration', deficits that are reversible with simultaneous administration of E2, 4) the increased risk of dementia in premenopausal women who have had a bilateral oophorectomy, 5) the increased prevalence of cognitive disease in women, which correlates with the abrupt earlier loss of gonadal function, 6) the elevated concentration of E2 in the brain compared to the circulation, and 7) the negative correlation between serum sex steroids, but positive correlation between serum gonadotropins, in women and men with AD. The above findings should NOT be confused with the widely reported Women's Health Initiative study which found a negative consequence of using UNPHYSIOLOGICAL conjugated equine estrogens. The above data indicate that the decline in circulating physiological sex steroids at menopause and during andropause would greatly impact brain functioning, and that those individuals with a higher synthetic capacity post-menopause and during andropause would be more likely to be spared the senescent changes induced by low sex steroid concentrations. We therefore hypothesize that those individuals with a lower neurosteroid synthetic capacity are more likely to develop AD, while those with a higher neurosteroid synthetic capacity will be spared longer the neurodegeneration induced by the age-related loss of peripheral sex steroids. Importantly, APOE ?4, the major risk factor for AD, and ABCA7, recently associated with AD from GWAS, alter the neuronal transport of cholesterol, the precursor from which all sex steroids are synthesized. Our preliminary studies have assessed sex steroid synthetic capacity by examining single nucleotide polymorphisms (SNPs) in genes of steroidogenic pathway family members (SPFMs) that regulate sex steroid synthesis. Identification of those genetic variations in members of the SPFM that cumulatively lower normal pathway function, i.e. that result in lower sex steroid synthesis are expected to associate with cognitive deficits. In this respect, we have found that gene-gene interactions between any 2 of APOE ?4, lhcgr2 and fsh1 together with female gender significantly increases the risk of developing AD to >90%. This data is very promising, but to date we have analyzed only 240 case-control samples. This grant therefore aims to perform genotype and hormone analyses in a larger sample size to gain greater confidence that these interactions are important predictors of AD. To this end, in Specific Aim 1 we will genotype a further 2760 age- and sex- matched case-control samples and use analytic tools (logistic regression, multi-dimensionality reduction and recursive partitioning) to identify well supported gene-gene interactions that predict AD.
In Specific Aim 2, we will measure 1) the concentrations of luteinizing hormone, follicle-stimulating hormone, E2, pregnenolone (P5), progesterone (P4) and T in matched plasma, CSF and hippocampus from 200 AD and age-matched control brains. This data will be analyzed to determine if plasma, CSF and brain concentrations of sex steroids are lower in AD, the relationship between plasma, CSF and brain sex steroids, and together with Specific Aim 1 genotyping data, if specific gene-gene interactions associated with AD are predictive of plasma, CSF and brain sex steroid concentrations. These experiments will determine that genetic variants in SPFMs regulate circulating and brain sex steroid synthesis and predict AD. Successful completion of these studies would enable the development of prognostic and diagnostic tests for AD, and provide insights into the neuroendocrine mechanisms of AD.

Public Health Relevance

Dementia is a major health problem for Veterans;however, the diagnosis of dementia is often complicated by confounding factors in the patient history (trauma for example) or cognitive test scores of the individual. Identification of gene-gene interactions that predict with a >90% certainty the diagnosis of dementia (e.g., Alzheimer's disease) would greatly compliment current diagnostic procedures (patient and family history, neuropsychological test results, physical exam and MRI and PET scans) in the diagnosis/prognosis of Alzheimer's disease. As therapeutic strategies reach the market, a cheap, prognostic genetic test would greatly help direct physicians in providing suitable treatment options. Such a genetic test would be invaluable in ruling out other neurological conditions.

National Institute of Health (NIH)
Veterans Affairs (VA)
Veterans Administration (I21)
Project #
Application #
Study Section
Special Emphasis - Research on Clinical Application of Genetics (SPLC)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Wm S. Middleton Memorial Veterans Hosp
United States
Zip Code