DMR1 (CCM1/YGR150C) of Saccharomyces cerevisiae encodes an RNA-binding protein from the pentatricopeptide repeat family required for the maintenance of the mitochondrial 15S ribosomal RNA

Abstract

Pentatricopeptide repeat (PPR) proteins form the largest known RNA-binding protein family and are found in all eukaryotes, being particularly abundant in higher plants. PPR proteins localize mostly in mitochondria and chloroplasts, where they modulate organellar genome expression on the post-transcriptional level. The Saccharomyces cerevisiae DMR1 (CCM1, YGR150C) encodes a PPR protein that localizes to mitochondria. Deletion of DMR1 results in a complete and irreversible loss of respiratory capacity and loss of wild-type mtDNA by conversion to rho(-)/rho(0) petites, regardless of the presence of introns in mtDNA. The phenotype of the dmr1Delta mitochondria is characterized by fragmentation of the small subunit mitochondrial rRNA (15S rRNA), that can be reversed by wild-type Dmr1p. Other mitochondrial transcripts, including the large subunit mitochondrial rRNA (21S rRNA), are not affected by the lack of Dmr1p. The purified Dmr1 protein specifically binds to different regions of 15S rRNA in vitro, consistent with the deletion phenotype. Dmr1p is therefore the first yeast PPR protein, which has an rRNA target and is probably involved in the biogenesis of mitochondrial ribosomes and translation.

Figure 1.—

Phenotype of the DMR1 deletants. (A) Deletant strains are respiratory deficient and fail to grow on glycerol media, regardless of the presence (strain DPPR1) or absence (strain DPPR2) of mitochondrial introns. Diploids constructed from these dmr1Δ strains with a wild-type ρ⁰ tester (KL14-4A/60) are respiratory deficient as well, suggesting that functional mtDNA is irreversibly lost. Four 10-fold serial dilutions, beginning with the 10⁻¹ dilution of the saturated liquid YPD preculture are arranged vertically in each panel. Plates were grown for 3 days. (B) Residual mtDNA is still present in some dmr1Δ cells. Hoechst staining shows extranuclear signal in DPPR1 (dmr1Δ) cells, like in the ρ⁺ control CW04, but unlike in the ρ⁰ negative control (KL14-4A/60), leading to the conclusion that the dmr1Δ cells are ρ⁻. Bar, 10 μm.

Figure 2.—

Specific degradation of the 15S rRNA transcript in dmr1Δ mitochondria. (A) Hybridization with an oligonucleotide probe recognizing the 3′ fragment of the 15S rRNA shows that in ρ⁻ dmr1Δ cells (DPPR2/15S-7) expressing this gene the transcript is degraded into a series of defined fragments. Neither wild-type ρ⁺ strain (CW252), nor a strain with the same ρ⁻ genome introduced into wild-type nuclear background (CW15S), shows any signs of this degradation. (B) Independent dmr1Δ ρ⁻ clones (DPPR2/15S-2, DPPR2/15S-7, DPPR2/15S-24) expressing the 15S rRNA show the same degradation pattern. (C) Low-copy (pDMR1-CEN) or high-copy (pDMR1-2μ) plasmid vectors expressing the wild-type DMR1 gene, but not empty vector controls, rescue the phenotype of 15S rRNA degradation in dmr1Δ ρ⁻ mitochondria of the DPPR2/15S-7 strain. RNA prepared from mitochondria purified by differential centrifugation was used in all the hybridizations shown.

Figure 3.—

Other mitochondrial transcripts are not affected by the deletion of DMR1. Mitochondrial RNA preparations from dmr1Δ ρ⁻ clones expressing the 21S rRNA (DPPR2/21S-1 and DPPR2/21S-2) (A), or the COX2 mRNA (DPPR2/COX2) (B), together with positive (CW252, wild-type ρ⁺) and negative (KL14-4A/60, ρ⁰) controls were hybridized with appropriate plasmid probes (described in materials and methods).

Figure 4.—

Mapping the ends of the degradation products. RNA from ρ⁻ dmr1Δ cells expressing the 15S rRNA of the DPPR2/15S-7 strain and the CW252 wild-type control was circularized by ligation and amplified in two successive rounds of RT–PCR and PCR. (A) General design of the experiment, showing the location of primers. (B) Results of the first round of amplification (PCR using primers L1 and RT) of the DPPR2/15S-7 strain and the CW252 wild-type control RNA preparations. M is the size marker (GeneRuler DNA ladder mix; Fermentas). (C) Locations of the 5′ ends of the degradation products in the mature 15S rRNA sequence. Fragment lengths shown in brackets, breakpoint positions in the sequence indicated by slashes. Line width is proportional to the number of times the particular fragment was identified in the sequenced pool. (D) Regions containing the breakpoints in the putative secondary structure model of 15S rRNA from the Comparative RNA Project (http://www.rna.ccbb.utexas.edu/).

Figure 5.—

Dmr1p is an integral mitochondrial protein. Mitochondria were purified from a strain expressing the Dmr1p–TAP fusion. (A) The TAP tag fusion does not affect the Dmr1p function. Total RNA from the isogenic wild-type strain (WT) and the Dmr1p–TAP strain was hybridized with an oligonucleotide probe recognizing the 3′ fragment of the 15S rRNA. (B) The Dmr1p–TAP fusion is detected in the mitochondrial fraction. Spheroplasts from the Dmr1–TAP strain were lysed and debris was removed by centrifugation at 1500 × g. Mitochondria (M) were then pelleted by centrifugation at 13,000 × g, and the postmitochondrial supernatant (S) corresponded mostly to the soluble cytoplasmic fraction. Proteins were analyzed by SDS–PAGE (12%) and immunoblotting with antibodies recognizing the TAP tag (PAP), the cytosolic protein Pgk1p, and the integral inner mitochondrial membrane protein Cox2p. (C) Purified mitochondria from the Dmr1p–TAP strain were converted to mitoplasts by hypotonic swelling. Intact mitochondria and mitoplasts were optionally treated with 20 μg/ml of proteinase K. Proteins from each fraction were then resolved on SDS–PAGE (15%) followed by Western blot analysis with antibodies recognizing the TAP tag (PAP), and the markers for the mitochondrial outer membrane (Tim70p and Fis1p), the intermembrane space (Cyc3p, Tim23p, and Tim13p), and the integral mitochondrial membrane protein Cox2p.

Figure 6.—

Results of the gel shift (EMSA) assays performed with the purified heterologous Dmr1 protein and labeled fragments of 15S rRNA (A) and 21S rRNA (B). The protein concentration was decreased from left to right in a series of twofold dilutions, with 1× corresponding to 0.6 μg of Dmr1p per reaction. The 1× concentration was not included due to the formation of aggregates that failed to enter the gel. The lengths and locations of the RNA fragments are described in the text.

Figure 7.—

15S rRNA transcripts in single and double pet127 and dmr1 deletants. Hybridization with an oligonucleotide probe recognizing the 3′ fragment of the 15S rRNA shows that in the context of a ρ⁺ mtDNA (intronless) the deletion of PET127 (pet127Δ strain DPET2) the 15S rRNA transcript is only slightly longer than in the isogenic and isomitochondrial control (CW252), but no other defects are apparent. In a ρ⁻ clone expressing the 15S rRNA transcript (all ρ⁻ clones were isomitochondrial to the DPPR2/15S-7 strain) with the wild-type alleles of PET127 and DMR1 (CW15S) only a slight overaccumulation of precursors is observed. In the context of the same ρ⁻ mtDNA, the single dmr1 deletant (DPPR2/15S-7) shows the typical 15S rRNA fragmetation pattern (cf. Figure 2), while both the single pet127Δ (DPETCW15S) and the double pet127Δ dmr1Δ (DDP/15S-7) mutant strains show a similar pattern of significant precursor overaccumulation with a decrease of the mature 15S rRNA level, but no detectable degradation products. RNA prepared from mitochondria purified by differential centrifugation was used in all the samples shown.