The SardiNIA study population cohort comprises over 7,000 subjects, starting at ages from 14-102, from a cluster of four towns in Sardinia. The study has been measuring >600 quantitative traits (endophenotypes or quantitative risk-related genetic or environmental factors) that can be scored on a continuous scale, and is designed as a longitudinal studies, with 4 visits over the first 15 years of its tenure. Traits of special interest include a range of cardiovascular risk factors, anthropometric measurements, blood test values, and facets of personality. Fourth visits have been completed for the study cohort to permit more incisive assessment of longitudinal trends and outcomes, as well as the assessment of additional phenotypes related to bone density and frailty as a function of age. For example, 24 hour blood pressure measurements and ECHOcardiography are extending the analysis of cardiovascular traits; hearing tests and retinograms are extending studies to sensory deficits in aging; and the cohort has been specifically extended to over 250 individuals over 92 years of age to analyze effects of extreme age. With this cohort, a variety of epidemiological and genetic factors have been identified, including recent epidemiological studies have been done of personality traits associated with white coat or masked hypertension, with circadian blood pressure patterns, or with uric acid levels; of the prevalence of chronic kidney disease and unknown thyroid disorders; and of arterial stiffness and influences of the metabolic syndrome. Full-genome scans with batteries of single-nucleotide markers were conducted, and were supplemented in the lastrecent years with full genome DNA sequencing and genotyping with specialized chips (metabochip, immunochip, and exome chip, and a chip designed to give equal coverage across the entire genome). The whole cohort has also been analyzed to yield genetic variation data down to allelic frequencies of 0.1% 23,107,086 single nucleotide polymorphisms (SNP) and 3,053,323 insertions/deletions (indels) available for the analysis of genetic factors in traits and autoimmune diseases. GWAS pointed to genes/variants that determine a significant portion of the genetic contribution to variance for each trait and disease. In conjunction with consortium efforts on other population cohorts, including the Baltimore Longitudinal Study of Aging and the InCHIANTI study supported by the NIA, an increasing number of publications have resulted that identify genes associated with obesity, cardiovascular traits, and levels of lipids and blood components. GWAS consortium studies have also advanced technical approaches, including the usefulness of averaging of quantitative blood pressure traits; quality control; integrative annotation of variants; and mining the human phenome using allelic scores. As for GWAS of genetic factors involved in traits and diseases, Consortium efforts in the last year have discovered variants associated with the QT interval; thyroid peroxidase antibodies and clinical thyroid disease; fibrinogen levels; and BMI in adolescents and young adults. Supporting the range and depth of results are, for example, a publication in which human demographic history was reconstructed from DNA sequences of 1,204 Y chromosomes in the cohort; and a second that identified a genetic locus that is associated with part of hereditary capacity for educational attainment. In a new initiative, we are deepening the analysis of sensory capacity, with completed analyses of hearing tests and ongoing studies of microvasculature in retinal photographs. Those have detailed age-related changes that often correlate with disease phenotypes. In recent years, GWAS findings have been reported for several traits in three Nature Genetics papers: Height -- Two new gene variants were found to have a significant effect on stature in Sardinians. Together, they appear to reduce height by 6 centimeters. Researchers found additional height-decreasing gene variants consistently across the Sardinian population that reflect an observed island effect in which large, island-dwelling mammals decrease in size over time. Lipids and Inflammation -- We identified 14 novel genetic variants associated with serum lipid levels, as well as 19 associated with inflammatory markers in the blood. Both lipid levels and inflammatory markers influence risk of heart disease. Hemoglobin -- For the first time, researchers concurrently analyzed gene regulation of A1, A2 and fetal hemoglobin levels, to see if there was any coordination in their regulation. Twenty-three associations were seen at 10 loci, including five new gene candidates. Half the variants showed associations with more than one type of hemoglobin. Hemoglobin deficiency can cause life-threatening diseases like beta-thalassemia and sickle cell anemia. In particular, genes associated with HbF levels as a modulator of thalassemia/sickle cell disease severity have been identified. In another approach, cohorts of Sardinian patients and controls were assembled and genotyped for GWAS for each of several diseases, Type 1 diabetes, multiple sclerosis (MS), and breast cancer. This has permitted our report in the New England Journal of Medicine ofIn another ongoing initiative we FACS characterized cell counts, cell ratios and expression of cell surface markers -- assessed as median fluorescence intensity (MFI) -- of helper T cell and cytotoxic T cells (250 traits), regulatory T cells (207 traits), B cells (189 traits), monocyte cells (88 traits), and dendritic cells (78 traits), resulting in 812 quantitative immune cell traits in 4,000 SardiNIA individuals. Correlating the quantitative immune cell traits mentioned above with our ultra-dense genetic map yielded a total of 294 associated loci correlated with autoimmunity in 40 coincident associations (manuscript in preparation). We have extended the studies to range of related measures, including published data on C-reactive protein monocyte chemotactic protein-1 (MCP-1), interleukin-6 (IL-6), adiponectin, and soluble BAFF. Furthermore, we have measured the circulating levels of the main immunoglobulin types (IgM, IgG and IgA) using both turbidimetric and Luminex assays in individuals from the same cohort, and saw 18 genome-wide associations.
Showing the most recent 10 out of 110 publications
