Figure 6. Increased adenylation activity is detrimental to mitochondrial gene expression.

(A) Western blotting for steady-state levels of OxPhos components and mtPAP (top). Mitochondrial translation products were labelled with [³⁵S]-Met in PDE12+/+ and PDE12−/− cells with either endogenous or overexpressed levels of mtPAP (bottom). (B) Radioactive MPAT assay for ND1 and COII extracted from PDE12+/+ and PDE12−/− cells with either endogenous or overexpressed levels of mtPAP. MPAT assay gel profiles were determined using ImageQuant. (C) Box-plot representation of percentage of mt-tRNAs molecules extended beyond the 3' CCA addition for all 22 mt-tRNAs, ascertained by MPAT-Seq, for PDE12+/+ and PDE12−/− cells with either endogenous or overexpressed levels of mtPAP. **p<0.01, ***p<0.001 with Student t-test relative to PDE12+/+. (p values: [+/+] vs. [+/+ + mtPAP] p=0.053, [+/+] vs [−/−] p=0.0089, [+/+] vs [−/− + mtPAP] p=5.52×10⁻¹¹). (D and E) Radioactive MPAT assay for 16S (D) and 12S (E) mt-rRNAs extracted from PDE12+/+ and PDE12−/− cells with either endogenous or overexpressed levels of mtPAP. MPAT assay gel profiles were determined using ImageQuant. (F) Model of PDE12 function. Polycistronic precursor RNA is endonucleolytically processed, liberating the individual mt-tRNA transcripts, which undergo post-transcriptional maturation (nucleotide modifications, CCA addition) before being aminoacylated with a cognate amino acid for use in mitochondrial translation (green). In the case of some mt-tRNAs a spurious poly(A) tail is added by mtPAP, preventing aminoacylation, leading to mitoribosome stalling during mitochondrial translation and to OxPhos defect as a consequence (red). Deadenylation by PDE12 removes spurious polyadenylation on mt-tRNAs restoring the properly matured pool of mt-tRNAs available for aminoacylation (green). Note: The role of PDE12 in the processing of mt-rRNA not included.

DOI: http://dx.doi.org/10.7554/eLife.27596.015

Figure 6—figure supplement 1. The role of PDE12 in maturation of mt-tRNA^Tyr.

(A) Adjacent organization of mt-tRNA^Tyr and mt-tRNA^Cys genes in human mtDNA indicating overlapping ‘A’ residue (red) required for both transcripts. This residue is required as the discriminator base (DB) in mt-tRNA^Tyr. (B) Schematic based on model proposed by Fiedler et al. (2015) for repair of mt-tRNA^Tyr-1 to provide discriminator base. In vitro data suggested that mtPAP polyadenylation of mt-tRNA^Tyr is required to add the DB and PDE12 is implicated in the exonucleolytic resolution of polyadenylated mt-tRNA^Tyr-1(A)_n intermediates to the repaired mt-tRNA^Tyr-1A carrying DB. (C–D) In order to test whether PDE12 participates in the biogenesis of mt-tRNA^Tyr in living cells, we have analysed the 3’ ends of this tRNA using the radioactive MPAT and MPAT-Seq assays in six conditions: control HEK293 cells, PDE12−/− cells, PDE12−/− cells complemented with wild-type PDE12 cDNA, PDE12−/− cells complemented with the catalytic mutant E351A, control HEK293 cells overexpressing mtPAP and PDE12−/− cells overexpressing mtPAP. Radioactive MPAT revealed that in control HEK293 cells a large proportion of mt-tRNA^Tyr molecules resolve at −1 nt compared to the 3’ end processing site, the direct product released from the 5’ cleavage of mt-tRNA^Cys by RNase P, consistent with the lack of both the DB and CCA in mt-tRNA^Tyr ((C) - ‘mt-tRNA^Tyr−1’). This confirms that cleavage of mt-tRNA^Tyr- mt-tRNA^Cys is likely to occur such that it leaves the overlapping nucleotide on mt-tRNA^Cys in a large proportion of the molecules.

The ratio of mt-tRNA^Tyr lacking DB compared to mature mt-tRNA^Tyr carrying CCA ((C) - ‘mt-tRNA^Tyr−1ACCA’) is not changed upon ablation of PDE12 (C). The analysis of the MPAT-Seq 3’ data revealed a drop in read count corresponding to DB in each condition tested, further supporting the presence of a substantial pool of mt-tRNA^Tyr-1, which lacks the DB (D). However, the proportion of mature mt-tRNA^Tyr-1ACCA compared to that lacking the DB (mt-tRNA^Tyr−1) appeared to be increased when mtPAP is overexpressed. This is consistent with the notion that mtPAP is involved in the process of adding the DB, as suggested by Fiedler et al. (2015). In all conditions, there is no evidence to suggest accumulation of mt-tRNA^Tyr species with just the DB added ((C) and (D), ‘mt-tRNA^Tyr-1A’), suggesting that when the single ‘A’ DB is incorporated, the CCA extension is rapidly synthetized.

The proportion of ‘A’ incorporated at the ‘CC’ positions of the CCA extension in MPAT-Seq would be indicative of the proportion of mt-tRNA^Tyr-1 molecules which carry oligoadenylate extensions (‘mt-tRNA^Tyr−1 A(n)”). However, no increase in the proportion of ‘A’ at these positions in the absence of PDE12 was observed, suggesting that mt-tRNA^Tyr-1 A(n) is not accumulating in PDE12−/− cells (D). In the absence of PDE12, mt-tRNA^Tyr is polyadenylated, but only beyond the CCA according to MPAT-Seq (D), suggesting that while ablation of PDE12 does not appear to participate in DB incorporation, it does lead to accumulation of spurious adenylation beyond the CCA extensions as seen with other mt-tRNAs (Figure 4). Together, our findings would support the role of mtPAP in providing the DB to the mt- tRNA^Tyr-1 repair substrate, as proposed by (Fiedler et al., 2015), but that PDE12 is not absolutely required for the trimming of the poly(A) tail to provide the single A residue for the DB. It is possible therefore that either mtPAP only incorporates a single A residue before the CCA adding enzyme acts to add the CCA extension to complete maturation of mt-tRNA^Tyr, or that other factors, for example ELAC2 proposed by (Fiedler et al., 2015), act to remove the poly(A) tail added by mtPAP to the mt-tRNA^Tyr-1 precursor.

DOI: http://dx.doi.org/10.7554/eLife.27596.016