Interestingly, to date, no strain with more than one kaiA gene has been identified. On the other hand, there are strains lacking some or even all Kai components. For example, MED4 is lacking KaiA; UCYN-A

possesses neither KaiA nor KaiB; and the Gloeobacter genome does not encode any kai gene at all. This terrestrial strain shows strong phylogenetic distance to the other Cyanobacteria. The Gloeobacter lineage diverged early within the radiation of Cyanobacteria and chloroplasts, making it possible that Kai-based circadian timing systems arose in other Cyanobacteria later during evolution, even though a loss of the corresponding genes in Gloeobacter AZD2014 order also seems possible ( Nakamura et al., 2003). The symbiotic cyanobacterial strain UCYN-A is among the most abundant oceanic nitrogen-fixing microorganisms whose nitrogen-fixation rates are Neratinib molecular weight equal to or even greater than those of Trichodesmium ( Church et al., 2005, Langlois et al., 2008, Moisander et al., 2010, Montoya et al., 2004 and Zehr et al., 2001). Interestingly, because UCYN-A is lacking an oxygen-evolving photosystem II, nitrogen fixation can be continued during the light period, making a timed regulation of this specific process unnecessary. With a genome size of only 1.44 Mbp, UCYN-A shows a high degree of genome streamlining, with components of photosystem II, ribulose-1,5-bisphosphate carboxylase and the tricarboxylic

acid cycle being completely absent ( Thompson et al., 2012, Tripp et al., 2010 and Zehr et al., 2008). Therefore an obligate symbiosis is suggested for this species. And indeed, a unicellular eukaryotic alga was shown to be its symbiotic partner ( Thompson et al., 2012). To date, it is not clear which role a circadian clock might play for UCYN-A in this

relationship and if any timing mechanism is present. Interestingly, Nodularia harbors a KaiA protein (101 aa) that is shorter than that of the other organisms (approximately 300 aa). This shortened KaiA protein, also present in other group IV-strains, is equivalent to the C-terminal domain of the S. elongatus-KaiA. However, it lacks the N-terminal pseudoreceiver domain, which is thought to be important for direct interaction with the light-responsive redox Montelukast Sodium sensor CikA ( Williams et al., 2002). In this respect, it appears consistent that species like MED4, which lack the KaiA protein entirely, also lack CikA. However, UCYN-A possesses a CikA protein without harboring a KaiA homolog. In this case, the role of CikA might be restricted to its output function (described below). Contrarily, some strains harbor a full-length KaiA homolog, but no CikA (e.g. S. WH 7803). Accordingly, their timing input machinery seems to function differently, possibly relying on other external stimuli than light-responsive redox potentials ( Williams, 2009).