As you probably know, my specialty are birds, so writing this series on clocks in microorganisms was quite an eye-opener for me and I have learned a lot. The previous two posts cover the clocks in the cyanobacterium Synechococcus elongatus, the first bacterium in which circadian rhythms were discovered and, thus, the species most studied to date.
The work in Synechococcus has uncovered a cluster of three genes - kaiA, kaiB and kaiC - that are essential for circadian rhytmicity in this species. kaiA positively regulates the kaiBC promoter and overexpression of kaiC represses the kaiBC promoter. Deletion of any one of the three genes leads to the complete loss of rhythmicity.
Synechococcus is a unicellular cyanobacterium. It was thought that circadian clock evolved in it due to incompatibility between nitrogen fixation and photosynthesis. Thus, temporal separation of these two processes was needed, phosynthesis occuring only during the day, while nitrogen fixation was relagated to the night time. It is known that filamentous cyanobacteria, those that build chains of cell, utilize a different strategy, that of spatial separation, some cells being involved in nitrogen fixation and others in photosynthesis. The two cell types exchange the end-results of those processes. Thus, it was thought that filamentous cyanobacteria have no need for a circadian clock.
However, it appears that Synechococcus is not the only bacterium to have a clock. Laboratory of Eviatar Nevo in Israel has taken a look at another cyanobacterium, this time a filamentous, chain-forming species, Nostoc linckia, and the work that ensued suggests that a number of other bacteria may possess a circadian clock as well [1,2,3].
Cyanobacteria are some of the oldest organisms on Earth, at least 3.5 billion years old, appearing in the fossil record relatively soon after the split between Eubacteria and Archaea (3.8 billion years ago). For most of the evolutionary history of cyanobacteria, the environment was very harsh, and UV radiation was one of the major factors influencing the evolution of prokaryotes. For most of that evolutionary history, the environment has undergone large changes, not just in oxygen levels, but also in the levels of UV radiation.
Volodymir Dvornik, Eviatar Nevo and collaborators hypothesized that a circadian clock, involved in temporal processing of light (including UV light) may be an important adaptation in all cyanobacteria and have detected the kaiABC cluster in Nostoc. Moreover, they hypothesized that Nostoc living in harsh, exposed environments (on sun-bathed slopes of so-called Evolution Canyons in Israel) would show greater mutation rate and higher nuclotide polymorphisam in the kai genes than Nostoc living on less harsh slopes of the Canyons. This is exactly what they found .
Some of the data from that study was intiguing - suggesting gene duplications and horizontal gene transfer of kai genes. So, they followed this up with a study of kai genes in a number of species of cyanobacteria  and later in a number of species of Eubacteria and Archea . Here is the tree of kaiC (right) compared to the tree of 16S rRNA genes (left) - with quite amazing overlap:
Their analysis suggests that kaiC is the oldest element of the complex, while the kaiA is the youngest. kaiA occurs only in cyanobacteria, while kaiB, kaiC and the kaiBC complex occur in other types of bacteria and Archaea. There are also two types of kaiC: short and long. The long, double-domain kaiC (dd-kaiC) is found only in photosynthetic bacteria. Likewise, kaiBC cluster is found only in photosynthetic bacteria. Here is the tree of the kaiBC cluster:
Non-photosynthetis bacteria tend to have the short version of kaiC (sd-kaiC), as well as independent kaiB elsewhere in the genome (i.e., not in a cluster with kaiC). Analysis of the trees of kai gene evolution sugests many duplication events, as well as many occurences of gene loss and horizontal tranfer. Curiously, all the horizontal tranfers occured from cyanobacteria, as donors, to other types of bacteria and Archaea as recipients. Here is the proposed evolutionary history of the kai genes:
Thus, a number of bacteria and Archaea posses one, two or three kai genes, sometimes in multiple copies. Does that mean they have functioning circadian clocks?
Bacteria other than cyanobacteria do not have kaiA. Deletion of kaiA in Synechococcus abolishes rhythms. It is not inconceivable that a different gene (and several additional transcription factors besides kaiA are involved in the Synechococcus clock, so there is no lack of potential candidates) may fulfill that role in other microorganisms. Still, Synechococcus is the only prokaryote in which circadian rhythms have been measured and studied (OK, there is a recent exception - but you will have to wait for the next post to hear about it). Is it possible that kai genes in other bacteria have other functions and only in cynobacteria they got exapted for the circadian role? Time and new research will tell.
 Volodymyr Dvornyk, Oxana Vinogradova, and Eviatar Nevo, Long-term microclimatic stress causes rapid adaptive radiation of kaiABC clock gene family in a cyanobacterium, Nostoc linckia, from "Evolution Canyons" I and II, Israel, PNAS, February 19, 2002, vol. 99, no. 4, 2082-2087
 Volodymyr Dvornyk, Eviatar Nevo, Evidence for Multiple Lateral Transfers of the Circadian Clock Cluster in Filamentous Heterocystic Cyanobacteria Nostocaceae, JMol Evol (2004) 58:341-347
 Volodymyr Dvornyk, Oxana Vinogradova, and Eviatar Nevo, Origin and evolution of circadian clock genes in prokaryotes, PNAS, March 4, 2003, vol. 100, no. 5, 2495-2500