Time machine

Author's Note

I’ve decided it’s prudent to take a break from the debate about the quality of reviewers on National Institutes of Health study sections. The American Society for Biochemistry and Molecular Biology governing council met in mid-November with Richard Nakamura, director of the NIH’s Center for Scientific Review. The discussion was enlightening, and the data presented will inform my future columns on this topic.

In the meantime, I encourage you to take a quick poll below.


This is the story of a spectacular triumph of biochemistry. The story builds upon the foundation of more than four decades of research on circadian rhythm.

Centuries ago, the French botanist Jean-Jacques d’Ortous de Mairan considered how it might be that the leaves of acacia trees zipper up at night and then open at dawn for maximal capture of sunlight. The most sensible interpretation, at least for a simple-thinking person like me, would have been that the earliest rays of sunlight might trigger the leaves to open up to optimize photosynthesis.

No, in the complete absence of sunlight, the leaves of the acacia tree opened right at dawn, giving evidence that the plant has a built-in clock. This internal timing device allows the plant to anticipate when the sun should be coming up. In hindsight, the advantage of anticipation is obvious.

Insight into the nature of the regulatory system controlling circadian rhythm got a huge boost from fruit fly genetics a bit more than 40 years ago. In an unusually inspired series of experiments, Ron Konopka and the late Seymour Benzer found mutations that cause fruit flies to have a longer than normal period of circadian rhythm, a shorter than normal period, or no circadian rhythm at all. Remarkably, all of these mutations mapped to the exact same gene – the fruit fly period gene.

Beautiful, beautiful genetics – yes. But advances toward mechanism had to wait decades until the period gene could be cloned. This was achieved independently by Mike Rosbash and Jeff Hall and Mike Young. How cool it was when these scientists showed that expression of the period gene was itself rhythmic over a 24-hour period.

Two decades of research in the lab of Joe Takahashi enhanced the understanding of how things fit together. Takahashi took exactly the same approach as Benzer and Konopka – forward genetics. Instead of fruit flies, Takahashi was sufficiently bold to use chemical mutagenesis and forward genetics with mice as his experimental species. Like Benzer and Konopka, Takahashi found his gene – Clock – wherein a specific ENU-induced mutation resulted in a lengthened circadian period. Like Young, Rosbash and Hall, Takahashi painstakingly chased down his Clock gene by positional cloning (See Antoch M.P. et al 1997 and King D.P. et al 1997)

Unlike his predecessors, however, Takahashi got lucky. Upon sequencing the Clock gene, he quickly recognized that it encoded a transcription factor – it had a distinct bHLH DNA binding domain. That discovery represented a critical, missing piece of the puzzle. In short order, Takahashi and others established that the genes controlling circadian rhythm specify the parts list for a negative transcription feedback cycle such that the pathway could be understood in clear and simple terms.

Perhaps the coolest piece of the circadian rhythm puzzle came from studies of the single-celled marine microbe Synechococcus elongates – a cyanobacterium. Susan Golden, Carl Johnson, Takao Kondo and others used forward genetics to discover cyanobacterial variants with altered circadian periods. In a particularly beautiful series of experiments, they showed that short or long period variants were at a fitness disadvantage relative to wild-type strains when grown under a 12-hour-to-12-hour light-to-dark cycle. Amazingly, variants with a long period – such as 28 hours – out-competed the wild-type strain when grown under a 15-hour-to-15-hour light-to-dark cycle.

These experiments demonstrated the importance of circadian timekeeping for biological fitness of cyanobacteria. Among the many mutants discovered to affect timekeeping in cyanobacteria, the most prevalent class fell into the kaiA, kaiB and kaiC genes. The products of these three genes organize into a complex centered with a KaiC hexamer having the approximate ratio of KaiA:KaiB:KaiC of 1:1:4. The KaiC hexamer has both autophosphorylation and autodephosphorylation activities, with KaiA enhancing autophosphorylation and KaiB attenuating the stimulatory effect of KaiA.

In a spectacular, two-page paper published in 2005, Kondo and colleagues showed that this Kai complex undergoes rhythmic changes in phosphorylation with a period of 24 hours. In other words, a complex reassembled from recombinant proteins has all of the properties of a biochemical clock – a virtual time machine. Complexes composed of mutated Kai variants known to have a short period in living cells displayed a short oscillatory period in the test tube, and complexes composed of long-period variant polypeptides displayed a long oscillatory period.

That the reconstituted, recombinant Kai complex has all of the working parts necessary to create an isolated and fully accurate biochemical clock represents a breathtaking discovery. This, my friends, is the beauty of biochemistry. The field needed genetics to open the door, but it was hardcore biochemistry that showed us the precise workings of the Kai time machine. What a triumph of science!

Steven McKnightSteven McKnight (steven.mcknight@
utsouthwestern.edu) is president of the American Society for Biochemistry and Molecular Biology and chairman of the biochemistry department at the University of Texas-Southwestern Medical Center at Dallas.