Is There a Classical Nonsense-Mediated Decay Pathway in Trypanosomes?

In many eukaryotes, messenger RNAs with premature termination codons are destroyed by a process called "nonsense-mediated decay", which requires the RNA helicase Upf1 and also, usually, an interacting factor, Upf2. Recognition of premature termination codons may rely on their distance from either a splice site or the polyadenylation site, and long 3'-untranslated regions can trigger mRNA decay. The protist Trypanosoma brucei relies heavily on mRNA degradation to determine mRNA levels, and 3'-untranslated regions play a major role in control of mRNA decay. We show here that trypanosomes have a homologue of Upf1, TbUPF1, which interacts with TbUPF2 and (in an RNA-dependent fashion) with poly(A) binding protein 1, PABP1. Introduction of a premature termination codon in either an endogenous gene or a reporter gene decreased mRNA abundance, as expected for nonsense-mediated decay, but a dependence of this effect on TbUPF1 could not be demonstrated, and depletion of TbUPF1 by over 95% had no effect on parasite growth or the mRNA transcriptome. Further investigations of the reporter mRNA revealed that increases in open reading frame length tended to increase mRNA abundance. In contrast, inhibition of translation, either using 5'-secondary structures or by lengthening the 5'-untranslated region, usually decreased reporter mRNA abundance. Meanwhile, changing the length of the 3'-untranslated region had no consistent effect on mRNA abundance. We suggest that in trypanosomes, translation per se may inhibit mRNA decay, and interactions with multiple RNA-binding proteins preclude degradation based on 3'-untranslated region length alone.