Prokaryotes: not so “simple” after all

One of my research challenges for 2008 is to marry my previous work (and my “true interest”), microbial genomics, with my current project which focuses on protein kinase substrate prediction. It should prove to be fascinating and fun, provided that publications such as this one keep appearing:

Schneiker, S. et al. (2007)
Complete genome sequence of the myxobacterium Sorangium cellulosum.
Nature Biotech. 25: 1281-1289
doi:10.1038/nbt1354 | Abstract | Full Text (subscription) | PubMed

It’s years since I read a textbook but I imagine that many of them will tell you that cell signalling via serine/threonine/tyrosine phosphorylation is “eukaryotic”, whereas Bacteria and Archaea use two-component sensor/histidine kinase systems. Some far-sighted individuals have been educating us otherwise for years, but only with the advent of microbial genomics has it become apparent that eukaryotic-like protein kinases (ELKs) are widespread in prokaryotes.

Back to Sorangium. This bacterium has the largest-known bacterial genome to date (~ 13 Mbp), encodes ~ 50% more proteins than the eukaryote S. cerevisiae and – get this – has 317 putative ELKs. That’s a greater length of DNA encoding for ELKs than some entire bacterial genomes and the highest percentage of the genome devoted to ELKs of any organism (bear in mind that the human/rat/mouse kinome comprises ~ 600 kinases). On top of that Sorangium synthesises a heap of interesting secondary metabolites, including an anti-cancer compound.

Let’s hear it for those prokaryotes.

8 thoughts on “Prokaryotes: not so “simple” after all

  1. I’ve had a draft post lying around for months listing features found in prokaryotes that are normally thought of as eukaryotic: splicing/introns, internal membranes (including membrane-bound nucleoids), EKLs, ncRNAs, multiple chromosomes, multi-subunit transcriptional complexes…the list goes on. I must write it up sometime.

  2. Neil, a very interesting topic! The concept that prokaryotes do not have ‘proper eukaryotic’ Ser/Thr phosphorylation did actually cause some trouble for me back in 1995. Here is the short version of the sorry tale:

    In this dark (pre-genomic) age, there was no solid example of a bacterial S/T kinase around, at least not to my knowledge. In 1994/95, I worked on an interesting sequence motif (nowadays called FHA domain), which to me looked a lot like a phospho-Ser/Thr recognition motif. If there only hadn’t been the ugly finding that there were bacterial proteins harbouring this domain (at that time mostly in Anabaena) . I talked to some people who managed to convince me that this would rule out a general role of the FHA domain in phospho-Ser/Thr recogntion, as the latter does not occur in bacteria. How stupid!

    Nowadays, we all know that there are plenty of bacterial FHA domains and plenty of bacterial Ser/Thr kinases. The two domains often coexist in the same polypeptide. It was really a stupid mistake of mine that I did not advocate the true function of FHA domains more forcefully.

  3. Very interesting, so am I right in saying that the only true difference to teach between Eu and Pro karyote is the very meaning of the respective words, with or without nuclear membrane respectively?

  4. Pingback: Distractions » Blog Archive » Prokaryotes: not so “simple” after all

Comments are closed.