In this work we present a method for the differential analysis of gene co-expression networks and apply this method to look for large-scale transcriptional changes in aging. We derived synonymous gene co-expression networks from AGEMAP expression data for 16-month-old and 24-month-old mice. We identified a number of functional gene groups that change co-expression with age. Among these changing groups we found a trend towards declining correlation with age. In particular, we identified a modular (as opposed to uniform) decline in general correlation with age. We identified potential transcriptional mechanisms that may aid in modular correlation decline. We found that computationally identified targets of the NF-ΚB transcription factor decrease expression correlation with age. Finally, we found that genes that are prone to declining co-expression tend to be co-located on the chromosome. Our results conclude that there is a modular decline in co-expression with age in mice. They also indicate that factors relating to both chromosome domains and specific transcription factors may contribute to the decline. There is mounting evidence that mammalian aging is marked by increased gene transcriptional variation. This trend was shown not only by studying gene expression in single cells (Bahar et al. 2006), but at the coarse tissue resolution as well (Somel et al. 2006; Li et al. 2009). These led us to believe that looking at absolute changes in expression level alone may not tell the whole story of transcriptional changes in age. Instead the story may be in the more subtle changes in the coordination of expression among multiple genes. For this reason, we decided to look at changes in co-expression relationships with age. To this end, we developed a methodology for differential co-expression network analysis for the comparison gene co-expression on a global scale. We applied this methodology to compare co-expression between young (16-month) and old (24-month) mice. This allowed us to find both gene groups whose coordination appear to be affected by age and to propose potential mechanisms for the change. We believe our work is of broad importance because it represents a different paradigm for looking not only at aging but also at any complex condition or disease—away from changes in individual genes towards changes in gene relationships.
Low bone mineral density (BMD) leads to osteoporosis, and is a risk factor for bone fractures, including stress fractures. Using data from UK Biobank, a genome-wide association study identified 1,362 independent SNPs that clustered into 899 loci of which 613 are new. These data were used to train a genetic algorithm using 22,886 SNPs as predictors and showing a correlation with heel bone mineral density of 0.415. Combining this genetic algorithm with height, weight, age and sex resulted in a correlation with heel bone mineral density of 0.496. Individuals with low scores (2.2% of total) showed a change in BMD of -1.16 T-score units, an increase in risk for osteoporosis of 17.4 fold and an increase in risk for fracture of 1.87 fold. Genetic predictors could assist in the identification of individuals at risk for osteoporosis or fractures.
We developed a new statistical framework to find genetic variants associated with extreme longevity. The method, informed GWAS (iGWAS), takes advantage of knowledge from large studies of age-related disease in order to narrow the search for SNPs associated with longevity. To gain support for our approach, we first show there is an overlap between loci involved in disease and loci associated with extreme longevity. These results indicate that several disease variants may be depleted in centenarians versus the general population. Next, we used iGWAS to harness information from 14 meta-analyses of disease and trait GWAS to identify longevity loci in two studies of long-lived humans. In a standard GWAS analysis, only one locus in these studies is significant (APOE/TOMM40) when controlling the false discovery rate (FDR) at 10%. With iGWAS, we identify eight genetic loci to associate significantly with exceptional human longevity at FDR < 10%. We followed up the eight lead SNPs in independent cohorts, and found replication evidence of four loci and suggestive evidence for one more with exceptional longevity. The loci that replicated (FDR < 5%) included APOE/TOMM40 (associated with Alzheimer’s disease), CDKN2B/ANRIL (implicated in the regulation of cellular senescence), ABO (tags the O blood group), and SH2B3/ATXN2 (a signaling gene that extends lifespan in Drosophila and a gene involved in neurological disease). Our results implicate new loci in longevity and reveal a genetic overlap between longevity and age-related diseases and traits, including coronary artery disease and Alzheimer’s disease. iGWAS provides a new analytical strategy for uncovering SNPs that influence extreme longevity, and can be applied more broadly to boost power in other studies of complex phenotypes. Longevity is a complex phenotype, and few genetic variants that affect lifespan have been identified. However, aging and disease are closely related, and a great deal is known about the genetic basis of disease risk. Here, we show using genome-wide association studies (GWAS) of longevity and disease that there is an overlap between loci involved in longevity and loci involved in several diseases, such as Alzheimer’s disease and coronary artery disease. We then develop a new statistical framework to find genetic variants associated with extreme longevity. The method, informed GWAS (iGWAS), takes advantage of knowledge from 14 large studies of disease and disease-related traits in order to narrow the search for SNPs associated with longevity. Using iGWAS, we found eight SNPs that are significant in our discovery cohorts, and we were able to validate four of these in replication studies of long-lived subjects. Our results implicate new loci in longevity and reveal a genetic overlap between longevity and age-related diseases and traits. Beyond the study of human longevity, iGWAS can be applied to boost statistical power in any GWAS of a target phenotype by using larger GWAS of genetically-related conditions.
A common property of aging in all animals is that chronologically and genetically identical individuals age at different rates. To unveil mechanisms that influence aging variability, we identified markers of remaining lifespan for Caenorhabditis elegans. In transgenic lines, we expressed fluorescent reporter constructs from promoters of C. elegans genes whose expression change with age. The expression levels of aging markers in individual worms from a young synchronous population correlated with their remaining lifespan. We identified eight aging markers, with the superoxide dismutase gene sod-3 expression being the best single predictor of remaining lifespan. Correlation with remaining lifespan became stronger if expression from two aging markers was monitored simultaneously, accounting for up to 49% of the variation in individual lifespan. Visualizing the physiological age of chronologically-identical individuals allowed us to show that a major source of lifespan variability is different pathogenicity from individual to individual and that the mechanism involves variable activation of the insulin-signaling pathway. One of the long-standing mysteries in aging is that some individuals die early whereas others die late. The age at which a specific individual will die is difficult or impossible to predict, and thus a fundamental aspect of aging in all animals is that it is stochastic. Aging stochasticity is particularly interesting in model organisms such as C. elegans because they are genetically inbred, can have the exact same chronological age, and can be grown under standard lab conditions. In this paper, we uncover a major mechanism underlying stochasticity in aging. To do this, we first developed a fluorescent aging marker (sod-3::GFP) whose expression declines with age and can be used to measure physiological age. In young animals, the level of expression of this fluorescent marker indicates which animals will live longer and which will die earlier. We used this fluorescent aging marker to show that variable pathogenicity from individual to individual is a major source of lifespan variability and that the mechanism involves variable activation of the insulin-signaling pathway.
The normal aging process is associated with stereotyped changes in gene expression, but the regulators responsible for these age-dependent changes are poorly understood. Using a novel genomics approach, we identified HOX co-factor unc-62 (Homothorax) as a developmental regulator that binds proximal to age-regulated genes and modulates lifespan. Although unc-62 is expressed in diverse tissues, its functions in the intestine play a particularly important role in modulating lifespan, as intestine-specific knockdown of unc-62 by RNAi increases lifespan. An alternatively-spliced, tissue-specific isoform of unc-62 is expressed exclusively in the intestine and declines with age. Through analysis of the downstream consequences of unc-62 knockdown, we identify multiple effects linked to aging. First, unc-62 RNAi decreases the expression of yolk proteins (vitellogenins) that aggregate in the body cavity in old age. Second, unc-62 RNAi results in a broad increase in expression of intestinal genes that typically decrease expression with age, suggesting that unc-62 activity balances intestinal resource allocation between yolk protein expression and fertility on the one hand and somatic functions on the other. Finally, in old age, the intestine shows increased expression of several aberrant genes; these UNC-62 targets are expressed predominantly in neuronal cells in developing animals, but surprisingly show increased expression in the intestine of old animals. Intestinal expression of some of these genes during aging is detrimental for longevity; notably, increased expression of insulin ins-7 limits lifespan by repressing activity of insulin pathway response factor DAF-16/FOXO in aged animals. These results illustrate how unc-62 regulation of intestinal gene expression is responsible for limiting lifespan during the normal aging process. The normal aging process in Caenorhabditis elegans involves coordinated changes in expression of about a thousand genes. In order to find upstream regulators responsible for these aging genes, we used a genomics approach to screen for transcription factors that bind to them. We focused on one such regulator, unc-62/Homothorax, that encodes a co-factor for a Hox transcription factor. Although essential for development, expression of unc-62 in adults limits lifespan. We find that unc-62 is a transcriptional activator of yolk protein genes, which are synthesized in the intestine and encode essential nutrients for progeny but accumulate to toxic levels with age. Additionally, analysis of unc-62 knockdown indicates that when vitellogenin transcription is decreased, transcription of intestinal genes with somatic functions increases. Thus, activation of yolk protein gene expression by unc-62 is both a burden on maternal health and a necessary resource for embryonic growth. Surprisingly, we also found that the intestine of old worms has novel expression of non-intestinal proteins that are toxic when expressed in the intestine and are activated by unc-62 in old age. Our work on transcriptional regulation of intrinsic gene expression differences during normal aging has revealed insights into the mechanisms that limit lifespan.