Playing RNase P evolution: swapping the RNA catalyst for a protein reveals functional uniformity of highly divergent enzyme forms

Abstract

The RNase P family is a diverse group of endonucleases responsible for the removal of 5' extensions from tRNA precursors. The diversity of enzyme forms finds its extremes in the eukaryal nucleus where RNA-based catalysis by complex ribonucleoproteins in some organisms contrasts with single-polypeptide enzymes in others. Such structural contrast suggests associated functional differences, and the complexity of the ribonucleoprotein was indeed proposed to broaden the enzyme's functionality beyond tRNA processing. To explore functional overlap and differences between most divergent forms of RNase P, we replaced the nuclear RNase P of Saccharomyces cerevisiae, a 10-subunit ribonucleoprotein, with Arabidopsis thaliana PRORP3, a single monomeric protein. Surprisingly, the RNase P-swapped yeast strains were viable, displayed essentially unimpaired growth under a wide variety of conditions, and, in a certain genetic background, their fitness even slightly exceeded that of the wild type. The molecular analysis of the RNase P-swapped strains showed a minor disturbance in tRNA metabolism, but did not point to any RNase P substrates or functions beyond that. Altogether, these results indicate the full functional exchangeability of the highly dissimilar enzymes. Our study thereby establishes the RNase P family, with its combination of structural diversity and functional uniformity, as an extreme case of convergent evolution. It moreover suggests that the apparently gratuitous complexity of some RNase P forms is the result of constructive neutral evolution rather than reflecting increased functional versatility.

Figure 1. Gene replacement and deletion strategy: substitution of RPR1 by PRORP3 and deletion of RPR2.

(A) The chromosomal RPR1 gene was replaced with the PRORP3 expression cassette (A. thaliana PRORP3 driven by the S. cerevisiae ADH1 promoter and terminator elements, and linked to a selectable marker (kanMX4)) by homologous recombination. (B) The chromosomal RPR2 gene was replaced with the HIS3MX marker by homologous recombination. In both cases, the selectable markers (flanked by loxP sites) were subsequently removed by transient expression of Cre recombinase. RPR1-coding sequences are indicated in blue, RPR2 in red, PRORP3 in magenta, markers used for selection in yellow, flanking loci in green, relevant promoter and terminator elements in grey. Auxiliary lines indicate insertion/deletion points.

Figure 2. tRNAs in RNase P-swapped yeast strains.

RNA was prepared from three independent clonal isolates of BY4743 and its RNase P-swapped derivatives, and analyzed by Northern blotting and RT-PCR. (A,B) Two blots were sequentially probed with oligonucleotides complementary to nucleus-encoded tRNAs and 5S rRNA. Blots were cropped to include the complete range of possible tRNA precursors (size estimates based on 5S rRNA hybridization signals). The relevant haploid genotypes of the homozygous diploid strains are indicated at the top. The RNA examined is specified to the left of each blot panel and presumed precursors indicated by asterisks. (C) The same samples were analyzed for the transcripts of the different RNase P genes by RT-PCR. (D) Quantitative analysis of 6 different tRNAs in RNase P-swapped yeast strains. Bands corresponding to mature tRNA were quantitated from the Northern blots (A,B) and normalized to 5S rRNA. Quantities are expressed relative to the mean of the parental BY4743 wild type strain. The mean and SD of the three clonal replicates are shown (see Table S2 for statistical analysis).

Figure 3. Levels of putative non-tRNA RNase P-substrates in RNase P-swapped yeast strains.

RNA was prepared from three independent clonal isolates of BY4743 and its RNase P-swapped derivatives. Precursor RNAs that were previously reported to accumulate in a conditional RNase P-deficiency model , were analyzed by quantitative RT-PCR. The RNase P deficiency strain JLY1 rpr1::HIS3 [rpr1-ts] and its wild type counterpart JLY1 rpr1::HIS3 [RPR1] were grown for two hours under restrictive conditions (37°C) and their RNA analyzed in parallel for comparison. Precursor RNA levels were normalized to the levels of ACT1 and CYC1 mRNA, and U6 snRNA. Quantities are expressed relative to (the mean of) the respective parental wild type strain(s) grown in parallel under identical conditions. The mean and SD of the three clonal replicates of the BY4743 wild type and of its RNase P-swapped derivatives, and the mean of technical duplicates of JLY1 rpr1::HIS3 [rpr1-ts] are shown (see Table S2 for statistical analysis). Note that the y-axis is split into two segments of different scale to accommodate the entire range of variation.

Figure 4. Quantitative phenotypic profiling of RNase P-swapped yeast strains.

Three independent clonal isolates of BY4743 (A) and CEN.PK (B) wild type (RPR1 RPR2) strains and their respective RNase P-swapped (rpr1Δ::PRORP3 rpr2Δ0) derivatives were grown in suspension micro culture under different conditions. Media were inoculated with defined numbers of stationary starved cells and growth monitored by continual automated optical density reading until all cultures had apparently reached the stationary phase. Maximal growth rates were derived from the logarithmically transformed data. Growth was analyzed (i) in different standard media (synthetic complete, complex rich (YPD), or minimal (containing only essential components) medium with 2% glucose) at standard (30°C) or stress temperatures (37 and 40°C) each; (ii) in synthetic complete medium with different mono-, di-, or trisaccharides at 2% as fermentable carbon source, or with 3% ethanol or glycerol as a non-fermentable carbon source; (iii) under different forms of osmotic and/or ionic stress (16% “high” glucose, 1.5 M sorbitol, 1.0/0.5 M NaCl, 1.5 M KCl); (iv) under either organic acid (40 mM acetic acid) or alkaline stress (“high” pH; adjusted to 7.8 with Tris base); (v) under conditions of either limiting or increased availability of critical metal ions (330 µM EDTA (general depletion of trace metal ions), 42 µM “low” (100-fold lower than normal) MgCl₂, 420 mM “high” (100-fold higher) MgCl₂); and (vi) under different forms of toxic stress (333 µM CuSO₄, 240/20 mM LiCl, 6 mM MnCl₂, 8/4 mM ZnCl₂, 600 µM paraquat, 100/20 mM hydroxyurea, 400/200 µM hygromycin B, 1.8/2 µM amphotericin) (in cases the conditions differed between the two strain backgrounds they are specified in the order BY4743/CEN.PK; unless otherwise specified, the basal medium was a synthetic complete medium containing 2% glucose). For a given condition all strains were analyzed in parallel. The mean and SD of the three clonal replicates are shown (*, P<0.05; **, P<0.01; ***, P<0.001; see Table S2 for P value listing); (A) BY4743-based strains; (B) CEN.PK-based strains.

Figure 5. Competitive growth of yeast strains with ribonucleoprotein RNase P versus strains with protein-only RNase P.

Wild type (RPR1 RPR2) BY4743 (A) and CEN.PK (B) were grown in co-culture with their respective diploid RNase P-swapped counterpart (rpr1Δ::PRORP3 rpr2Δ0) in glucose-containing synthetic complete medium. To distinguish the co-cultured strains they were mutually labeled with green fluorescent protein (GFP) by replacing one allele of the leu2Δ0 and the leu2-3_112 locus, respectively, with a TEF1 promoter-driven, yeast-enhanced GFP (yeGFP-labeled wild type strains, open squares and broken line; yeGFP-labeled RNase P-swapped strains, filled squares and continuous line). Logarithmic starter cultures were mixed in equal proportion and a sample withdrawn for analysis. Co-cultures were grown to stationary phase, samples taken for analysis, and fresh medium was inoculated at 1:1000 with the stationary cells to reinitiate growth. Subsequently, seven full cycles of growth (each corresponding to ∼10 cell generations) and dilution were carried out with samples always taken at stationary phase. The fraction of GFP-positive cells in the samples was determined by flow cytometry. The mean and the 95% confidence interval of three experiments (from independent starting cultures) are shown; (A) BY4743-based strains; (B) CEN.PK-based strains.