Autophagy-mediated post-transcriptional surveillance of meiotic translation in Saccharomyces Cerevisiae

ABSTRACT

In Saccharomyces cerevisiae, macroautophagy/autophagy plays a pivotal role and is indispensable for multiple meiotic processes. In this study, we demonstrate that Rim4, a meiosis-specific RNA-binding protein (RBP) that holds back the translation of a specific subset of meiotic transcripts until its programmed degradation by autophagy during meiotic divisions, forms a heterotrimeric complex in vivo with the yeast YWHA/14-3-3 proteins Bmh1 and Bmh2, which effectively expels mRNAs from Rim4’s binding grip. We pinpoint four distinct Bmh1 and Bhm2 binding sites (BBSs) in the Rim4 structure, with two of them nestled within the RNA recognition motifs (RRMs). The phosphorylation states at these BBSs controlled by counteracting PKA and Cdc14 phosphatase activities determine whether Rim4 interacts with Bmh1, Bmh2 or the mRNAs, thereby regulating Rim4’s subcellular distribution, function, and stability for autophagy. Remarkably, we found that Rim4 is an Atg11-dependent selective autophagy substrate and activates Atg1 during meiotic divisions, only after its sequential dissociation from mRNAs and Bmh1 or Bmh2 assisted by PKA and cytosolic Cdc14, respectively. These findings reveal an intricate mechanism that underpins the autophagy-mediated surveillance of Rim4-mRNA interactions, orchestrated by meiotic PKA and Cdc14 activities, to ensure stage-specific translation of key meiotic transcripts.

Main text

In budding yeast, meiotic progression beyond the S-phase depends on autophagy. Our previous research has underscored the pivotal role of autophagy in meiotic chromosome segregation, meiosis exit, and sporulation, partially achieved through degradation of Rim4, an essential meiosis-specific RBP that primarily suppresses the translation of specific mid-late mRNAs. However, the precise regulatory framework governing autophagy-mediated Rim4 degradation within the context of Rim4 ribonucleoprotein (mRNP) complex assembly and meiotic translation remains an enigma.

In the recent two back-to-back studies [1,2], we deviate from the previously reported amyloid-like form of Rim4 and have instead revealed a soluble heterotrimeric complex of Rim4 with Bmh1 and Bmh2. We revealed that, using a nuclear localization signal, Rim4 enters the nucleus to load its mRNA substrates. The Rim4 mRNP complex assembly in the nucleus is facilitated by Cdc14-mediated Rim4 dephosphorylation that dissociates Bmh1 and Bmh2 from Rim4. Importantly, only after mRNA loading can the translation-inhibitory Rim4 mRNP complex be exported efficiently from the nucleus into the cytosol (Figure 1, left). Such coupling of nuclear export with mRNA binding has been observed in other RBPs that primarily reside in the cytoplasm, such as Pab1 that binds to the poly(A) tail of an mRNA. Intriguingly, our data reveal that Rim4 forms complexes with Pab1 in an mRNA-dependent manner; since Pab1 regulates mRNA deadenylation and translation initiation, this finding suggests possible multifunctionality of Rim4 beyond passively sequestering mRNAs.

Figure 1.

Schematic model describing programed meiotic post-transcriptional translation suppression by Rim4 (left, in green shadow) and Rim4 degradation by selective autophagy after Rim4 releases its bound mRnas during meiotic divisions (right, in pink shadow). Left, Rim4 in complex with Bmh1-Bmh2 enters the nucleus using a nuclear localization signal; Rim4-mRNA mRNP assembly occurs in the nucleus, facilitated by Cdc14-mediated Rim4 dephosphorylation at the BBSs, which releases Bmh1-Bmh2 to stimulate mRNA loading; next, mRNA export drives Rim4-mRNA mRnps into the cytosol. Right, during meiotic divisions, PKA-mediated Rim4 phosphorylation at the BBSs facilitates mRNA release from Rim4 before anaphase; during anaphase, Cdc14 relocated from the nucleus to the cytosol disassembles the resulting Rim4-Bmh1-Bmh2 complex from mRNA release, thereby triggering selective autophagy-mediated Rim4 degradation in the cytosol.

In budding yeast, autophagy is active along meiosis progression. Remarkably, we have observed that autophagy activity spares the Rim4-mRNA complex in the cytosol through early meiotic stages. However, as meiotic divisions approach, with Rim4-bound mRNAs poised for release and translation, cytosolic Rim4 undergoes a sequential modification process at the Bmh1 and Bmh2 binding sites. Initially, phosphorylation of these BBSs by PKA stimulates Rim4-Bmh1-Bmh2 assembly, facilitating mRNA release. Subsequently, cytosolic Cdc14, during anaphase, engages in Rim4 dephosphorylation to remove Bmh1-Bmh2 interaction, rendering Rim4 susceptible to Atg11-dependent selective autophagy (Figure 1, right). Thus, autophagy mediates surveillance of Rim4-mRNA assembly during meiosis, thereby regulating meiotic translation.

Rim4 possesses three RNA recognition motifs separated by low complexity sequences that facilitate Rim4 self-assembly. Interestingly, we found that purified recombinant Rim4 and Rim4(∆C289) that lacks its C-terminal low complexity sequence, can activate Atg1 in cell lysates and in immunoprecipitated Atg1 complex within Rim4’s physiological concentration range. Moreover, activation of Atg1 by Rim4 can be inhibited by Bmh1, Bmh2 and mRNAs, which interact with Rim4’s RRMs. Thus, it is plausible that the intrinsic properties of exposed RRMs of Rim4 might trigger selective autophagy. This aligns with our observation that the rapid removal of cytosolic Rim4 by autophagy necessitates Rim4’s sequential phosphorylation by PKA for mRNA release and then dephosphorylation by Cdc14 for Bmh1-Bmh2 dissociation. As a result, we speculate that Rim4 conducts self-assembly, enabling the exposure of normally concealed Rim4 sites in the structured RRMs, thereby stimulating Atg1 activity and phagophore assembly site formation. Whether this process requires a selective autophagy receptor remains unclear. Nonetheless, this notion is supported by the colocalization of Atg8 with spatially concentrated cytosolic Rim4 (Rim4 puncta) during meiotic divisions, when Rim4 loses bound mRNAs and is degraded by autophagy. Moreover, recombinant Rim4 at high concentration intriguingly loses its activation effect on Atg1 in vitro, indicating that the sizes of Rim4’s self-assembly affected by Rim4 concentration might determine its effect on Atg1 activation.

Previous studies have shown that autophagy can degrade RNAs in various organisms. A key revelation from our study is that autophagy selectively spares the Rim4-mRNA complex. Moreover, the mRNAs liberated from Rim4’s hold, exemplified by the model substrate CLB3, become susceptible to autophagy. This suggests that binding to an RBP might constitute a general mechanism by which autophagy selectively spares certain mRNAs, a selection process likely guided by the RRMs. It is worth noting that recombinant RRM-containing Pab1 also stimulates Atg1 activity in vitro, albeit requiring higher concentration than Rim4. The RRMs are among the most common protein motifs in eukaryotes and are found in approximately 2% of all human genes. Given that, future studies should delve deeper into elucidating how the consensus sequence or structure of RRMs provides cues for autophagy-mediated degradation.

In 2022, we reported in JCB that Cdc14 upregulates autophagy by dephosphorylating Atg13, coupled with its programmed relocation from the nucleus to cytosol at the meiotic anaphases. Here, we reported a dual functionality of Cdc14 in Rim4 regulation, depending on its subcellular localization. Specifically, by releasing Bmh1-Bmh2 from Rim4, the nucleus-localized Cdc14 stimulates Rim4-mRNA complex formation; in contrast, Cdc14 stimulates autophagic Rim4 degradation in the cytosol. Intriguingly, upon translocation to the cytosol, we previously reported that cytosolic Cdc14 partially colocalizes with the Atg13-positive phagophore assembly site structure, suggesting that cytosolic Cdc14 might play a spatial role in connecting autophagy to its meiotic substrates, e.g., Rim4.

During mitosis, Cdc14 remains predominantly inactive in the nucleus, except for anaphase, a well-established principle that also extends to meiosis. However, we have unveiled a surprising revelation that Cdc14 exhibits at least partial activity before meiotic divisions. This is supported by various lines of evidence, including normal mitosis in temperature-sensitive cdc14–1 cells at 30°C; in sharp contrast, 30°C blocks sporulation, concomitant with enhanced phosphorylation of numerous proteins by meiotic prophase I, including Rim4 at the BBSs. This phenotype can be rescued by lowering the sporulation medium temperature to 23°C, thereby enabling more enzymatic activity from cdc14–1. These observations lead us to speculate that meiosis employs and regulates Cdc14 differently from mitosis.

Cdc14 primarily acts to counterbalance Cdc28/CDK1 activity in both mitosis and meiosis. Interestingly, our findings reveal that meiotic Cdc14 also antagonizes PKA activity at canonical PKA target sites, denoted as RRxS/T, which happen to be harbored by Rim4’s BBSs. The acute inhibition of PKA (tpk-as) by 1 NM-PP1 at meiosis entry leads to the complete abolishment of Rim4 phosphorylation at the BBSs and has reduced global phosphorylation of RRxS/T sites in many proteins by prophase I, accompanied by abnormal nuclear morphology and significantly reduced sporulation. Given the evolutionary conservation of PKA, Cdc14, and YWHA/14-3-3 proteins from yeast to humans, the mechanism uncovered in this study may have broader implications for gametogenesis in various species.

Funding Statement

The work was supported by the National Institute of General Medical Sciences [R01GM133899]; Welch Foundation [I-2019-20190330].

Disclosure statement

No potential conflict of interest was reported by the author(s).