Abstract

The aging population in developed countries has prompted experts to warn of the impending “silver tsunami” of older adults poised to overload already strained healthcare systems. Advanced age is the primary risk factor for most chronic diseases, and the vast majority of the elderly population is diagnosed with at least one such condition, if not several. The field of geroscience has emerged to study the processes responsible for age-related decline and devise methods to reduce this burden. A major branch of geroscience is the genetics of aging, including examining gene expression changes over time as well as exploring how genetic modifications can reduce senescence. Hundreds of studies over decades have compared gene expression in samples from the young against samples from the old, and these data have been analyzed, compiled into databases, and even re-analyzed in complex meta-analyses. Though the findings are largely logical and congruent, it is very difficult to distinguish correlation from causation, and causation from compensatory responses, when the data are entirely observational. As for interventional genetics, the nematode C. elegans is a popular model organism for geroscience studies, owing both to natural features and the wealth of accessible, well-established experimental protocols. Like humans, these small worms experience post-reproductive senescence with diverse features ranging from cognitive decline to sarcopenia, but these events occur on a compressed timescale of only a few short weeks. Genetic interventions, especially knockdown via the RNA interference feeding method, are simple and inexpensive. The major shortcoming is that findings in microscopic worms are challenging to defend as relevant to human physiology and medicine. Here I aimed to synthesize these two approaches in geroscience to produce a single unified workflow that harnesses the strengths and mitigates the weaknesses of each method alone, yet remains accessible and scalable. First, I designed a meta-analysis strategy based on the value-counting method to identify genes consistently differentially expressed with age across a variety of mammalian species and tissues. I hypothesized that a subset of these identified genes would be drivers of age-related decline that could be modulated in vivo to produce therapeutic results. Then, in order to identify members of this subset, orthologs of the differentially expressed mammalian genes were evaluated as targets for anti-aging, life-extending interventions in C. elegans using a series of post-developmental RNA interference lifespan screens. For this meta-analysis, I used a combination of R and python programming languages to apply a modified value-counting method to 25 distinct gene expression datasets hosted on the National Biotechnology Information Gene Expression Omnibus. Each dataset compared samples from younger adults against older adults and adhered to specific inclusion and exclusion criteria. In order from highest to lowest number of datasets, samples were derived from muscle, brain, adipose, immune, heart, liver, and other miscellaneous tissues from mice, humans, rats, and dogs. After ranking genes according to the frequency and consistency of differential expression with age, cut-off ranks were set to allow 130 genes to proceed to pathway analysis and 45 genes to proceed to in vivo testing in C. elegans. The results of gene ontology analysis were consistent with previous studies, highlighting trends like age-downregulated mitochondrial pathways and age-upregulated adaptive immune processes, suggesting the meta-analysis was designed appropriately. Out of the 45 highest ranking mammalian genes, comprising 16 age-downregulated and 29 age-upregulated genes, 27 (60%) genes could be matched to orthologs in C. elegans. Of these 27 orthologs, 19 corresponded to verified bacterial clones for RNA interference and were able to be tested in worms. Each clone was tested for significant and reproducible lifespan-extension in at least one screening experiment of roughly 80-100 worms per clone and subsequently a more tightly controlled validation experiment with roughly 100-150 worms per clone. Ultimately, post-developmental knockdown of 6 of the tested 19 (32%) orthologs of mammalian genes differentially expressed with age reliably extended lifespan in C. elegans. In order of greatest mean lifespan extension, these six genes were: fzy-1 (ortholog of CDC20), ost-1 (SPARC), spch-2 (RSRC1), C42C1.8 (DIRC2/SLC49A4), csp-3 (CASP1), and cah-3 (CA4). Though slightly more age-upregulated (10) than age-downregulated (9) genes were tested, only two of the six lifespan extending clones were derived from age-upregulated genes: spch-2 and csp-3. Here, I successfully identified six evolutionarily conserved drivers of aging using accessible, inexpensive methods. Interestingly, although many studies have proceeded under the assumption that genes upregulated with age should be targeted for downregulation, the results of my study do not support this assumption. Ultimately, I hope my dissertation serves as a blueprint for a scalable workflow that others will use and improve to translate the massive public repositories of gene expression data into actionable therapeutic targets for combating age-related decline.

Committee Chair

Roberto Civitelli

Committee Members

Erica Scheller, Elizabeth Pollina; John Edwards; Kerry Kornfeld

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

4-21-2026

Language

English (en)

Download

Share

COinS