Circadian rhythms and noise

One source of noise in human gene expression data may be circadian. A new study (PDF) in Molecular Endocrinology by Lemos, Downs, and Urbanski examines gene expression with respect to time in adrenal tissue. The abstract:

The suprachiasmatic nucleus plays a key role in the circadian secretion of adrenocortical hormones. However, there is evidence from mouse studies that components of the circadian clock, are also expressed within the adrenal gland itself. In the present study we performed genome-wide expression profiling to determine whether the adrenal gland of rhesus monkeys shows temporal gene expression across a 24-h period. We identified 322 transcripts with rhythmic patterns of expression, and found that the phase distribution of cycling transcripts varied across the day, with more genes showing activation duirng the night. We classified the transcripts by their function, and clustered them according to their participation in common biochemical pathways: 1) catecholamine synthesis and reuptake; 2) cholesterol cleavage and dehydroepiandrosterone sulfate synthesis; 3) protein synthesis and turnover; 4) the circadian clock mechanism. In an additional experiment, we assessed the expression of various clock genes at two time points, 12-h apart. We found that expression of Bmal1 and Cry1 was higher at 13 h, or zeitgeber time 6 (ZT6), whereas expression of Per1 was higher at 1 h (ZT18). Expression levels of Rev-erb alpha er itself goes through several steps of discovering which genes are responsive to changes in light levels, from the large number that alter their expression over a 24-hour cycle. </p>

It is interesting to consider the kind of time control applied to human tissue libraries -- from which a lot of gene expression results have been derived. Could different exposures to light in the processes of preservation and study trigger circadian expression processes? Does it make a difference when in the day tissue extractions were done? And of course, what other processes might be disrupted when cells are taken ex vivo?

Science Blog has the press release that also discusses hormone treatment in humans and circadian effects on stress:

"Of course, different genes peaked in function at different times of the day," explained Dario Lemos, an OHSU graduate student in the Urbanski lab and first author of the study. "For instance, genes controlling catecholamine secretion were more active in the day with function greatly decreasing at night. Catecholamines are involved in many important body functions, such as stress and mood."
This research provides important new information regarding the complex, rhythmic, 24-hour functions of the body. The research may also impact current therapies for a variety of diseases. For instance, data gathered in this study and future studies may suggest that certain therapies be delivered at certain times to synchronize with normal body functions controlled by body clocks.
"One example is testosterone replacement, a common treatment for certain disorders in males such as sexual dysfunction and depression," explained Urbanski. "Patients receiving testosterone late in the day often complain of sleep loss. This is likely due to the fact that in healthy people, testosterone levels are lower in the afternoon and evening. As more data is gathered about body clock functions in our lab and others, we will likely learn of a specific window of time during the day where testosterone therapy is effective, but less disruptive for patients."

Wouldn't it be ironic if telling patients to take medication with a meal was exactly the wrong thing to do! On the other hand, it might be a nightmare to figure out all these effects: how many therapies will include "time of day" in their research trials?


Lemos DR, Downs JL, Urbanski HF. 2006. Twenty-four hour rhythmic gene expression in the rhesus macaque adrenal gland. Mol Endocrin (early access) DOI link