How Our Biological Clock Could Dictate Medical Treatments


You might thinkthe time on the clock controls when and how you live your life. But ticking away inside each of us is a biological timekeeper that holds powerful sway over our bodies and behaviors. When we eat and when we sleep, our heart rates and our hormones—they're all regulated by our so-called circadian clocks.

Many researchers think medicine would be safer and more effective if it were practiced not according to a clock on a wall but the clocks inside patients. Proponents of the idea call it circadian medicine, and evidence suggests it could boost the effectiveness of present and future treatments, not only prescription medications but also surgery and radiation therapy.

The challenges facing this growing field are twofold. First, reading a person’s biological clock is expensive and time-consuming. (The gold-standard test, the dim light melatonin onset assay, or DLMO for short, involves closely monitoring a patient's melatonin levels by sitting them in a dimly lit room and collecting their spit or blood every 30 minutes for a day or more.) And second, it's not always easy or practical to tell which therapies would benefit most from optimally timed administration.

This week, researchers announced advances that confront both those issues: A simple blood test that researchers say could help infer a person's circadian rhythm, and a database of clock genes that encode targets for thousands of existing drugs. Together, they shift us toward a future in which we know not only what therapies to deliver at a set time but when to deliver them to specific people.

By the latest estimates, your biological clock controls the expression of half your genome on a roughly 24-hour schedule, operating independently of the clock on your phone's lock screen. Rosemary Braun, a computational biologist at Northwestern University, set out to see if it was possible to gauge a person’s internal time from the gene activity in a blood sample.

To do it, Braun and her colleagues analyzed 1,116 blood samples collected from 73 people and trained a machine learning algorithm to predict the time of day when the test subjects' blood had been drawn, based on the expression of just 41 genes. The resulting algorithm, which they call TimeSignature, can take two samples of blood and identify the three-hour window in which they were each drawn, as the researchers describe in this week's issue of Proceedings of the National Academy of Sciences.

Many genes peak in activity at different times of day. This set of 41 genes, each shown as a different color, shows a robust wave of circadian expression. By monitoring the level of each gene relative to the others, the TimeSignature algorithm learns to ‘read’ your body’s internal clock.
Rosemary Braun/Northwestern University

Here's where things get hair-splitty. Technically speaking, TimeSignature guesses when a subjects' blood was collected—not the subject's circadian state at the time of sampling. Because these samples came from people whose circadian rhythms align well with the clocks that go tick-tick, Braun expects that those 41 genes are also indicators of biological time. But that link still needs testing, as does whether the algorithm works equally well on people with different chronotypes–morning people, night owls, and so on. “That’s exactly the experiment we want to do next," she says. The goal is to emerge with a blood test that can dramatically simplify the process of figuring out your own chronotype.

Knowing your chronotype is useful, in theory, because it can affect how certain medicines interact with your body. That's where the second study comes in. In an investigation recounted in the latest issue of Science Translational Medicine, researchers analyzed thousands of tissues samples from more than 600 people in search of the genes that are both under circadian control and associated with known drugs—drugs that could, in theory, benefit from timed administration. “That’s my real concern," says genome biologist John Hogenesch, deputy director of the Center for Chronobiology at Cincinnati Children’s Hospital, who led the study. "Not enough people are working on coming up with the best time for therapies."

They winnowed down the list of genes they examined to produce a database of about 900 that appeared to code for proteins involved in the transport, metabolism, or activity of known drugs. (For instance: 136 of them encode drug targets for existing heart medications.) That kind of information could be a boon to anyone trying to determine the best time to deliver a drug.

"What this database could help answer is: What is the best time to deliver treatments?" says integrative physiologist Kenneth Wright, director of the Sleep and Chronobiology laboratory at the University of Colorado, Boulder, who was unaffiliated with both studies. "These are the kinds of results that companies and researchers can take and say, okay, based on the activity of this drug target at this time, I should test delivering the drug at this time or that time—rather than once in the morning or once at night."

Another compelling feature of Hogenesch's database: It catalogues patterns of gene expression in tissues throughout the body. In the future, researchers could use that data to deconstruct the body's circadian rhythms into what Wright calls satellite clocks. "There might come a time when we want to know what time it is in the liver, or the pancreas," he says. If different parts of our bodies operate on different time zones, as it were, timing drug delivery to the circadian state of each might one day bring about even more therapeutic precision.

  • How a domino master builds 15,000-piece creations
  • This hyper-real robot will cry and bleed on med students
  • Inside the haywire world of Beirut's electricity brokers
  • Tips to get the most out of Gmail’s new features
  • How NotPetya, a single piece of code, crashed the world
  • Looking for more? Sign up for our daily newsletter and never miss our latest and greatest stories