17 November 2006
Johan Elf
Department of Chemistry and Chemical Biology
Harvard University
I will discuss a theoretical model for how the charging levels of different tRNA isoacceptors respond when supply of their amino acid becomes growth limiting [1]. The charged levels of some tRNA isoacceptors are predicted to go to very low values while the charged levels of others are predicted to remain at high values. This theory explains the codon usage in attenuation control leader sequences [2] as well as the biased codon usage in several amino acid biosynthetic operons. We have verified the first class of predictions from the theory, concerning selective tRNA charging in wild type bacteria, by micro arrays and Northern blots for three isoacceptor families (Leu, Thr and Arg) in E. coli [3] as well as by ribosomal by-passing experiments [4]. The second class of predictions, concerning selective tRNA charging when tRNA concentrations are perturbed, initially appeared to be in disagreement with experiments. The theoretical predictions could only be correct if the established codon recognition table for the leucine codons was incorrect. Further experiments showed that it was the codon table that was inconclusive [5].
I will also discuss the near-critical properties of unsaturated multi-substrate reactions [6,7]. In essence, one substrate pool can decrease while another increases, such that the overall flux remains unchanged. This results in high sensitivity in substrate concentrations to flow imbalances, but it also induces large zero order fluctuations. The near-critical properties of coupled flow processes are predicted to have important consequences for the flow of different amino acids into proteins synthesis [8]. In particular I will discuss the implication for transcriptional control of amino acid biosynthetic operons.