A.K. Dunker. Center for Computational Biology and Bioinformatics, Indiana University,.
Día: 9 de julio de 2018, lunes
Hora: 12:00 h
Lugar: Salón de Actos de la EEAD
A. Keith Dunker and Jingwei Meng
Center for Computational Biology and Bioinformatics
Indiana University School of Medicine, Indianapolis, IN 46202
The overall structure of the genetic code, with a very uneven number of codons for the various amino acids and with the more complex amino acids having fewer codons, led Crick, Woese, and others to speculate that the present genetic code evolved from a simpler precursor with significantly fewer amino acids. Many years later Trifonov summarized the work on the evolution of the genetic code and developed a “consensus speculation” regarding the likely identity of the earliest amino acids and a possible order of addition for the ones that came later. According to this analysis, the codons for the aromatic amino acids were among the very last to be added.
We and our collaborators had developed a disorder propensity scale that ranked the amino acids from those that most strongly promote structure to those that most strongly promote disorder. We noticed that the earliest amino acids were comprised of the most disorder-promoting and the last-to-be added amino acids were comprised of the most structure-promoting, such as the aromatic amino acids, suggesting that protein structure formation could have provided the selective advantage for the particular genetic code that ultimately prevailed.
Next we searched for lists of the earliest proteins identified as those that are most widely distributed among the 3 domains of life. These early proteins contained mostly disordered proteins with functions such as RNA binding and only a few structured proteins including a small number of enzymes. Next we searched for an early enzyme that contained the fewest aromatic amino acids, which turned out to be dephospho-CoA kinase (DPC kinase), which transfers a phosphate from ATP to dephospho-CoA to make CoA. CoA is a co-factor for a large number of enzymes. We had the DNA coding sequences synthesized for the DPC kinases with the lowest number of aromatic amino acids from four organisms; also synthesized were four DPC kinases in which all of the codons for the aromatics were replaced with the codons for leucine. Cloning and expression of the four wild-type and four aromatic-minus mutants were carried out. All of the wild-type expressions yielded functional DPC Kinases with specific enzyme activities similar to those reported in the literature, whereas just two of the aromatic-minus mutants gave soluble proteins.
Far UV circular dichroism spectra of the two aromatic-minus mutants were very similar to the spectra of the wild-type proteins, with slight differences likely due mainly to the absence of aromatic side chains, but the aromatic-minus proteins were highly unstable, exhibited molten-globule features, and gave very low enzyme activity only slightly above background levels. Overall these data suggest that proteins evolved from disordered forms (including molten globules) ⇒ structured proteins, with this transformation occurring concomitantly with the evolution of the genetic code and enabled by the incorporation of the last-to-be added (e.g. aromatic) amino acids.