| dc.description.abstract | Up to half of the eukaryotic genome is controlled by the endogenous circadian clock at the level of rhythmic transcript abundance. The clock also controls post-transcriptional events, including clock regulation of the levels and phosphorylation state of translation factors that are also thought to promote rhythmic protein synthesis. However, if, and how, the clock controls translation initiation is unknown. I discovered that phosphorylation of eIF2(alpha), a conserved translation initiation factor, is clock-controlled, peaking during the subjective day. The rhythm in phosphorylated eIF2(alpha) (P-eIF2(apha) requires rhythmic activation of the eIF2(alpha) kinase CPC-3, the homolog of yeast and mammalian GCN2. Binding of uncharged tRNA to GCN2, such as what occurs during histidine starvation, is required to activate the kinase. Consistent with rhythmic activation of N. crassa CPC-3 by binding of uncharged tRNA, starvation of wild type cells for histidine, or a CPC-3 mutation that constitutively activates CPC-3 in the absence of bound uncharged tRNA, abolished rhythmic P-eIF2(alpha) levels. Further, translational activator GCN1 cycles and is required for eIF(alpha) hosphorylation by CPC-3. The rhythm in P-eIF2α accumulation led to reduced translation during the day in an in vitro cell-free translation system and is required for cycling levels of the mannosyl transferase protein, but not the core clock protein FRQ, in vivo, suggesting clock regulation of translation of specific mRNAs. To identify the mechanism of translational regulation of specific mRNAs through rhythmic eIF(alpha) phosphorylation, ribosome profiling in parallel with RNA-seq was carried out. 913 genes showed rhythmic changes in translational efficiency (TE) in WT cells, of which 554 were dependent on CPC-3.
Of those 554 genes, 426 have rhythmic TE but arrhythmic mRNA, suggesting these genes are translationally regulated through rhythmic eIF2(alpha) phosphorylation. Of these 426 genes, 113 day-peaking genes contain T-rich motifs, 189 night-peaking genes contain A-rich motifs and 19% have putative uORFs in the 5’ UTR. These elements suggest possible mechanisms for how specific mRNAs are controlled by cycling P-eIF2(alpha) levels. Finally, loss of rhythmic control of P-eIF2(alpha) levels led to reduced growth rates, supporting the idea that partitioning translation to the night, when energy resources are high, provides a growth advantage to the organism. Together, these data reveal a fundamental mechanism by which the clock regulates rhythmic protein production | en |
