The mammalian circadian clock strongly influences when and to what extent certain genes are transcribed in each tissue throughout the day. A new study, published today (February 2) in Science, shows that this daily schedule of gene expression depends partially on a person’s sex and age. Specifically, the team found that women have more genes with rhythmicity compared to men, and that these rhythms break down as people age.
To study circadian gene expression in human tissues, the team used postmortem RNA quantification and sequencing data from 914 donors, stored in the Genotype-Tissue Expression (GTEx) catalog. They developed an algorithm to calculate the reference circadian phase for each donor, that is, the internal clock state in the skeletal muscle following death. They used to their advantage the fact that they had data from typically 10 to 20 tissues per donor, and that the internal times of these tissues were correlated since they came from the same person. This resulted in a more accurate estimate of each donor’s internal phase, which is not necessarily equivalent to the “external” time of death, explains lead study author and Swiss Federal Institute of Technology Lausanne researcher Felix Naef, as the former might be advanced or delayed depending on other factors, such as the person’s chronotype or their geographic location within a time zone.
Naef and his colleagues used their algorithm to first assign the internal muscle time reference for each donor and then, shifted the internal time of the other tissues with respect to that reference, which increased the robustness of the timestamp. Using this approach to analyze the data of roughly 16,000 samples, they were able to characterize the rhythmicity of gene expression in 46 types of human tissue.
Scientists have previously looked at postmortem tissues to study circadian rhythms. But what makes this study unique is the algorithm developed by the team that allows them to look “at such scale, both in terms of the number of samples [and] number of different tissues” says Perelman School of Medicine University of Pennsylvania physician-scientist Garret FitzGerald. “I think it’s been a very valuable contribution,” adds FitzGerald, who didn’t contribute to the study, but he does advise the aging research-focused company Calico Laboratories.
Once the internal circadian phases of each donor were defined, Naef and colleagues assessed whether there were differences between the rhythms of men and women’s gene expression. They found that while the expression of core circadian clock genes—those essential for the generation of circadian rhythms—was conserved in both groups, there were significant differences in the physiological output in some tissues. Overall, women had twice as many genes showing rhythmicity, especially those that are active in the liver and the adrenal gland. This may have pharmacological implications, says Naef, since many of these highly rhythmic genes in the female liver influence how the body metabolizes drugs.
Next, to test the role of age, the team compared two groups of donors: those under 50 and those older than 60 years old at the time of death. As in the sex comparison, Naef and colleagues found no significant differences in the core circadian clock. However, the analyses revealed that the rhythm of expression weakened in most tissues in the older group. In this case, one of the largest reductions in rhythmicity was found in the coronary arteries, for example, in programs regulating cholesterol and fatty acid metabolism. This physiological change could be somehow related to the high “incidence rate of cardiovascular diseases in the elderly,” says Naef, but this is only a speculation.
The algorithm developed by Naef’s team is now publicly available and he hopes that it can open the door to analyses of other types of human samples, such as biopsies. He adds that moving in this direction may hopefully help the field go beyond the study of correlations and slowly begin to understand causal relationships between the circadian clock and human disease states, for instance, in cancer or metabolic disorders.