Long-lived organisms tend to be resistant to many types of stress, whereas short-lived organisms tend to be stress sensitive. This happy coincidence allows us to screen for longevity mutants by looking for stress resistance rather than long life (advantage: it takes a lot less time to do the primary screen).

The same logic ought to apply to small-molecule drugs: Any compound that increases stress resistance has an improved change of extending lifespan. That hypothesis has been operationally tested by the Lithgow lab, who performed a small-scale screen of antioxidant compounds and looked for molecules that increased thermotolerance in the worm C. elegans. Several of these drugs also increased lifespan. From Benedetti et al.:

Compounds that confer thermal stress resistance and extended lifespan

The observation that long-lived and relatively healthy animals can be obtained by simple genetic manipulation prompts the search for chemical compounds that have similar effects. Since aging is the most important risk factor for many socially and economically important diseases, the discovery of a wide range of chemical modulators of aging in model organisms could prompt new strategies for attacking age-related disease such as diabetes, cancer and neurodegenerative disorders … . Resistance to multiple types of stress is a common trait in long-lived genetic variants of a number of species; therefore, we have tested compounds that act as stress response mimetics. We have focused on compounds with antioxidant properties and identified those that confer thermal stress resistance in the nematode Caenorhabditis elegans. Some of these compounds (lipoic acid, propyl gallate, trolox and taxifolin) also extend the normal lifespan of this simple invertebrate, consistent with the general model that enhanced stress resistance slows aging.

Note that the authors tested resistance to thermal, rather than oxidative stress — given their choice to screen only antioxidant compounds, to do the latter would have been a bit circular. Still, given the history of antioxidant compounds as candidate anti-aging compounds, and the widespread belief that reactive oxygen species per se are a causative force in aging, the decision to screen only antioxidants does raise the possibility that the lifespan extension is due to the antioxidant activity of these compounds and that the stress resistance is merely an epiphenomenon.

Then again, it’s quite impressive that so many different antioxidants of so many different types can confer thermotolerance and increased longevity, and suggests that perhaps the association between antioxidants and longevity may have never had much to do with oxidation as such, but rather with some as-yet-uncovered connection between antioxidants and the activation of stress response pathways.


Most microarray studies of aging animals try to associate gene expression with chronological age: they look for groups of genes that are upregulated or downregulated as we get older. But chronological age is often an imperfect proxy for the quantity we are really interested in – physiological age, or bodily health, which is notoriously difficult to quantify.

The search is on for informative correlates of physiological age (like telomere length) that can be used to assess current health and predict remaining lifespan. One seemingly relevant quantity that is simple to measure is liveliness: we might expect a sprightly 90 year old woman to look forward to more years of healthy life than a lethargic 80 year old.

In the latest issue of Aging Cell, Golden et al. argue that this kind of behavior is a useful proxy for physiological age, and then show that gene expression can be used to predict behavior:

Age-related behaviors have distinct transcriptional profiles in C.elegans

There has been a great deal of interest in identifying potential biomarkers of aging (Butler et al. 2004). Biomarkers of aging would be useful to predict potential vulnerabilities in an individual that may arise well before they are chronologically expected, due to idiosyncratic aging rates that occur between individuals. Prior attempts to identify biomarkers of aging have often relied on the comparisons of long-lived animals to a wild-type control (Dhahbi et al. 2004). However, the effect of interventions in model systems that prolong lifespan (such as single gene mutations, or caloric restriction) can sometimes be difficult to interpret due to the manipulation itself having multiple unforeseen consequences on physiology, unrelated to aging itself (Gems et al. 2002; Partridge and Gems 2006). The search for predictive biomarkers of aging therefore is problematic, and the identification of metrics that can be used to predict either physiological or chronological age would be of great value (Butler et al. 2004). One methodology which has been used to identify biomarkers for numerous pathologies is gene expression profiling. Here, we report whole-genome expression profiles of individual wild-type Caenorhabditis elegans covering the entire wild-type nematode life span. Individual nematodes were scored for either age-related behavioral phenotypes, or survival, and then subsequently associated with their respective gene expression profiles. This facilitated the identification of transcriptional profiles that were highly associated with either physiological or chronological age. Overall, our approach serves as a paradigm for identifying potential biomarkers of aging in higher organisms that can be repeatedly sampled throughout their lifespan.

In the study, worms are grouped into 3 broad categories based on their behavior: category A if they show “symmetric, spontaneous, and smooth movement, ” C if they “only move their nose or tail when prodded,” and B for anything in between. Using data from a previous work, Golden et al. show that behavior is a useful proxy for physiological age: C worms have a significantly shorter remaining lifespan than A worms (after controlling for chronological age). In other words, in our search for biomarkers of physiological aging, we should include genes whose expression levels predict behavior – not just those that predict chronological age.

Golden et al. then get on to their main experiment: they measure mRNA expression levels in worms of 7 different ages, and grade the behavior of each worm. Using a machine learning approach, they show that 71% of the time, gene expression levels accurately predict behavior class. Several of the genes that are more highly expressed in inert C worms (versus in active A worms) are involved in amino acid metabolism and/or related to actin – the authors speculate that physiologically old worms might upregulate these genes to try to repair their damaged cytoskeletons.

Their full results are available online in the GEO database (GSE12290) – this should prove a wonderful resource for bioinformaticians, especially as it is only the second full-genome microarray study of aging in the worm.