Post-transcriptional mechanisms underlying ATAXIN-2 dependent gene expression in Drosophila

Abstract

ATAXIN-2 (ATX2) is a RNA-binding protein that regulates gene expression at post-transcriptional levels. However, it is largely unknown what other factors contribute to ATX2-dependent gene regulation and how ATX2 controls the translation of its associating mRNAs. Here, I found two ATX2-interacting factors, LSM12 and ME31B, which play their distinct roles in post-transcriptional regulatory mechanisms in Drosophila circadian pacemaker neurons. LSM12 acts as an adaptor of the ATX2-associating protein complex to recruit TYF. The activator complex, ATX2-LSM12-TYF, associates with 5’ cap-binding translation initiation factors in an ATX2-dependent manner, thereby supporting TYF-dependent translational activation. On the other hand, a translational repressor/decapping activator ME31B/DDX6 facilitates the association of ATX2 with NOT1, a scaffold protein of CCR4-NOT deadenylase complex. The repressor complex, ATX2-ME31B-NOT1, contributes to NOT1-dependent gene silencing on selective mRNAs with short poly(A)-tails. These two opposing post-transcriptional regulator complexes govern circadian periodicity and rhythms amplitude, respectively, to sustain robust, 24-hour locomotor rhythms. To obtain additional insights for ATX2-dependent gene expression, I examined the post-transcriptional activity of RNA-tethered ATX2 on a series of RNA reporters. ATX2 tethering to 3’ end of reporter transcripts activates the reporter expression in a manner dependent on poly(A)-binding protein (PABP)-interacting motif (PAM2). The translational activation by ATX2 is most evident in poly(A) tail-deficient and non-circularized reporter transcripts. Inclusion of a poly(A)-track in the reporter or RNA interference (RNAi)-mediated depletion of PABP weakens the translational activation by ATX2. By contrast, ATX2 tethering to the 5' end of reporter transcripts represses its translation in a PAM2-independent manner. Finally, ATX2 has no effects on internal ribosomal entry sites (IRES)-mediated translation, suggesting that ATX2 specifically activates cap-dependent translation. Taken together, these data demonstrate that 1) ATX2 employs a specific factor to exhibit its post-transcriptional effects and sustain circadian locomotor rhythms in Drosophila; and 2) ATX2-PABP interaction might support mRNA circularization particularly in poly(A)-deficient transcripts to stimulate translational initiation by ribosome recycling given that PABP directly binds to the 5’ cap-binding translation initiation factor, eIF4G.