The nuclear exosome is active and important during budding yeast meiosis

Abstract

Nuclear RNA degradation pathways are highly conserved across eukaryotes and play important roles in RNA quality control. Key substrates for exosomal degradation include aberrant functional RNAs and cryptic unstable transcripts (CUTs). It has recently been reported that the nuclear exosome is inactivated during meiosis in budding yeast through degradation of the subunit Rrp6, leading to the stabilisation of a subset of meiotic unannotated transcripts (MUTs) of unknown function. We have analysed the activity of the nuclear exosome during meiosis by deletion of TRF4, which encodes a key component of the exosome targeting complex TRAMP. We find that TRAMP mutants produce high levels of CUTs during meiosis that are undetectable in wild-type cells, showing that the nuclear exosome remains functional for CUT degradation, and we further report that the meiotic exosome complex contains Rrp6. Indeed Rrp6 over-expression is insufficient to suppress MUT transcripts, showing that the reduced amount of Rrp6 in meiotic cells does not directly cause MUT accumulation. Lack of TRAMP activity stabilises ∼ 1600 CUTs in meiotic cells, which occupy 40% of the binding capacity of the nuclear cap binding complex (CBC). CBC mutants display defects in the formation of meiotic double strand breaks (DSBs), and we see similar defects in TRAMP mutants, suggesting that a key function of the nuclear exosome is to prevent saturation of the CBC complex by CUTs. Together, our results show that the nuclear exosome remains active in meiosis and has an important role in facilitating meiotic recombination.

Figure 1. TRAMP mutants undergo normal meiosis despite CUT stabilisation.

A: FACS analysis of DNA content in wild-type and trf4Δ mutants taken at 2 hour time-points during meiosis, showing that meiotic DNA replication is normal in trf4Δ cells. B: Northern analysis showing expression of meiosis-specific mRNAs SPO11 and DMC1 in wild-type and trf4Δ cells. Replicate northern blots of the same samples were prepared for each probe; a representative SCR1 control is shown from the SPO11 blot. Quantification for SPO11 is given in Figure S1C. C: Schematic diagram of NEL025 and adjacent CUTs at the HPA3, YEL067C and NPR2 loci. D: Northern blot analysis of NEL025 and adjacent CUTs in log-phase cells growing on YPA and at 2 hour time-points during meiosis. SCR1 is shown as a loading control. Samples were run on replicate blots and each probed for a different CUT to avoid signal from inefficient stripping, a representative loading control is shown.

Figure 2. Genome-wide distribution of CUT transcripts during meiosis.

A: Scatter plot of log-transformed read counts from Cbc2-associated RNA isolated from wild-type and trf4Δ cells after six hours of meiosis. Each point represents the read count from a single 500 bp genome region. B: Venn-diagram showing overlap between CUT transcripts identified in meiotic cells and published CUT datasets from mitotic cells , . Actual numbers are given in the table below, note that overlapping counts are different depending on the direction of analysis as multiple CUTs defined in one dataset can map to a single locus in another dataset. For the Venn diagram, the number of overlapping CUTs was taken as the average of the values from both directions. C: Distribution of sequencing reads in 100 bp segments around the NEL025 locus from meiotic Cbc2-associated RNA in wild-type and trf4Δ cells. This graph differs from Figure 2A in that actual read counts as opposed to log-transformed values are shown, the points in Figure 2A that correspond to this region are shown in Figure S2C. D: Distribution of sequencing reads across chromosome V, as in C.

Figure 3. Exosome characterisation in meiotic cells.

A: Western blot for Rrp6-TAP and Pgk1. Left-hand blot compares Rrp6 in log phase YPA and at the initiation of meiosis; right-hand blot, shown at different exposure, shows the gradual decline in Rrp6 levels across meiosis. B: Analysis of the effect of estradiol-induced GFP-Rrp6 over-expression on MUT stability. RT-PCR reactions for RRP6 and MUTs were performed on RNA from cells without and with estradiol, compared to an SCR1 loading control. MUT 100 is expressed throughout meiosis whereas MUT 523 is only expressed after 8 hours, neither is repressed by Rrp6 overexpression. C: Silver-stained protein gel showing Csl4-TAP immunoprecipitations from meiotic and mitotic cells, compared to purifications from untagged strains. Mitotic cells were grown on YPD, meiotic cells were harvested after six hours in SPO media. D: Plots of peptide score vs. molecular weight for proteins identified by mass spectrometry in two independent immunoprecipitation experiments. Non-yeast proteins and proteins also discovered in the untagged control sample were discarded, then the proteins were divided into exosomal and non-exosomal sets, both of which are displayed. Key exosome proteins including Rrp6 are highlighted.

Figure 4. Meiotic DSB formation is defective in trf4Δ cells.

A: PFGE analysis of transient DSBs in chromosome III during meiosis in wild-type and trf4Δ cells, truncated chromosome fragments caused by DSB formation are indicated by arrows. B: PFGE analysis showing the accumulation of cleaved chromosome III fragments during meiosis in sae2Δ and sae2Δ trf4Δ cells. C: Quantification of chromosome III DSBs in sae2Δ and sae2Δ trf4Δ cells. Error bars indicate ±1 s.e., * - p<0.05, *** - p<0.01 by student's t-test, n = 5. D: RT-PCR analysis of mRNA expression for selected meiotic recombination proteins in wild-type and trf4Δ cells.