An ortholog of the Ro autoantigen functions in 23S rRNA maturation in D. radiodurans

Abstract

In both animal cells and the eubacterium Deinococcus radiodurans, the Ro autoantigen, a ring-shaped RNA-binding protein, associates with small RNAs called Y RNAs. In vertebrates, Ro also binds the 3' ends of misfolded RNAs and is proposed to function in quality control. However, little is known about the function of Ro and the Y RNAs in vivo. Here, we report that the D. radiodurans ortholog Rsr (Ro sixty related) functions with exoribonucleases in 23S rRNA maturation. During normal growth, 23S rRNA maturation is inefficient, resulting in accumulation of precursors containing 5' and 3' extensions. During growth at elevated temperature, maturation is efficient and requires Rsr and the exoribonucleases RNase PH and RNase II. Consistent with the hypothesis that Y RNAs inhibit Ro activity, maturation is efficient at all temperatures in cells lacking the Y RNA. In the absence of Rsr, 23S rRNA maturation halts at positions of potential secondary structure. As Rsr exhibits genetic and biochemical interactions with the exoribonuclease polynucleotide phosphorylase, Rsr likely functions in an additional process with this nuclease. We propose that Rsr functions as a processivity factor to assist RNA maturation by exoribonucleases. This is the first demonstration of a role for Ro and a Y RNA in vivo.

Figure 1.

Rsr is required for efficient 23S rRNA maturation. (A,B) Wild-type and the indicated mutant strains were grown to OD₆₀₀ = 0.2 at 30°C and shifted to 37°C. Cells were collected at 30°C (A) and after growth for 6 h at 37°C (B). Total RNA from the strains was fractionated in formaldehyde–agarose gels and subjected to Northern analysis. After methylene blue staining (top panels), the filter was probed with oligonucleotides complementary to 23S rRNA internal sequences (second panel), the 5′ leader (third panel), and the 3′ trailer (bottom panel). (C) Wild-type and Δrsr strains were grown at 30°C and shifted to 37°C at time 0. At intervals, RNA was extracted and analyzed by Northern blotting. The filters were stained with methylene blue (top panel) and probed to detect mature 23S rRNA (second panel), the 5′ leader (third panel), and the 3′ trailer (bottom panel). (D) Cells grown at 30°C (lanes 1–10), or for 4 h at 37°C (lanes 11–20) were labeled with ³²P_i for 5 min. Following addition of media containing excess unlabeled phosphate, aliquots were removed at intervals and RNA was extracted.

Figure 2.

The levels of both Rsr and the Y RNA increase at 37°C. (A) At intervals after the shift to 37°C, cells were harvested and subjected to Western blotting to detect Rsr. (Bottom panel) As a loading control, the blot was reprobed to detect the single-stranded DNA-binding protein SSB. (B,top) At intervals after the switch to 37°C, total RNA was subjected to Northern blotting to detect the Y RNA. (Bottom) As a control, the blot was reprobed to detect . (C) After growth at 30°C or after 4 h at 37°C, lysates were subjected to immunoprecipitation with anti-Rsr antibodies. RNAs in immunoprecipitates (lanes 4,5), supernatants (lanes 6,7), and an equivalent amount of the lysates were subjected to Northern analysis to detect the Y RNA. During immunoprecipitation, much of the Y RNA is nicked in an internal loop, resulting in a ∼90-nt fragment that is detected by the oligonucleotide probe (asterisk) and a ∼30-nt fragment that is detected when the full-length antisense RNA is used as a probe (data not shown). (D) Lysates of wild-type cells containing the vector pRAD1-SPC or pRAD1-SPC-expressing Rsr or Rsr-H189S under control of the katA promoter were subjected to Western blotting to detect Rsr. (Bottom) The blot was reprobed to detect SSB. (E) Total RNA extracted from wild-type cells containing pRAD1-SPC or pRAD1-SPC-expressing Rsr or Rsr-H189S was subjected to Northern blotting to detect the Y RNA (top) or (bottom). (F) Total RNA from the strains in D was fractionated in formaldehyde–agarose gels and subjected to Northern analysis. The filter was stained with methylene blue (top panel), then probed with oligonucleotides complementary to 23S rRNA internal sequences (second panel), the 5′ leader (third panel), and the 3′ trailer (bottom panel).

Figure 3.

Effects of deleting 3′-to-5′ exoribonucleases on 23S rRNA maturation. (A,B) Wild-type and the indicated mutant strains were grown to OD₆₀₀ = 0.2 at 30°C and shifted to 37°C. Cells were collected at 30°C (A) and after 6 h of growth at 37°C (B). Total RNA was extracted and subjected to Northern analysis. After methylene blue staining (top panels), the filters were probed to detect 23S rRNA internal sequences (second panel), the 5′ leader (third panel), and the 3′ trailer (bottom panel). (C) Wild-type, Δyrn, ΔyrnΔrph, and ΔyrnΔrnb strains were grown to OD₆₀₀ = 0.2 at 30°C. Total RNA was extracted and subjected to Northern analysis to detect pre-23S rRNAs. (D,E) The indicated strains were grown to OD₆₀₀ = 0.2 at 30°C (D) and shifted for 6 h to 37°C (E). Total RNA was subjected to Northern blotting to detect 23S rRNAs. (F,G, left panels) Lysates from wild-type and the indicated mutant strains were subjected to Western blotting to detect Rsr. (Bottom panels, left) The blots were reprobed to detect SSB. RNA extracted from the cells was subjected to Northern blotting to detect the Y RNA (right panels, top) and (bottom panels).

Figure 4.

D. radiodurans lacking Rsr accumulate longer and shorter forms of 23S rRNA. (A) Following growth of the indicated strains at 30°C (lanes 1–8) or 37°C (lanes 9–16), RNA was extracted and subjected to site-directed cleavage using RNase H and a 2′-O-methyl RNA–DNA chimeric oligonucleotide that directs cleavage 122 nt from the mature 23S rRNA 3′ end. Following Northern blotting, 3′ precursors were detected with an oligonucleotide complementary to sequences 3′ of the cleavage site. RNAs in lanes 1–8 and 9–16 were fractionated in separate gels. As a loading control, the blot was reprobed to detect . (B) RNA was subjected to cleavage as in A, except that the oligonucleotide used directs cleavage 67 nt from the mature 5′ end. Following Northern blotting, 5′ precursors were detected using an oligonucleotide complementary to sequences 5′ of the cleavage site. The blot was reprobed to detect . (C) A secondary structure for pre-23S rRNA predicted by Mfold. The 5′ extension is green and the 3′ extension is red. The mature 5′ and 3′ ends are indicated by green and red arrows, respectively. The mapped 5′ end is 3 nt longer than predicted from comparison with E. coli. For Δrsr, Δrph, and Δrnb strains, the 5′ and 3′ ends of pre-23S rRNAs are indicated by green and red solid arrowheads, respectively, while for ΔrsrΔpnp cells these ends are indicated by open arrowheads. The 5′ end of the pre-23S rRNA may be the transcription start, as it is preceded by a sequence resembling D. radiodurans promoters (Meima et al. 2001). Mfold predicts five possible structures for the RNA, all of which contain extensive base-pairing beween the 5′ and 3′ extensions. (D) Organization of the 23S rRNA transcription unit in D. radiodurans.

Figure 5.

Extended and truncated forms of 23S rRNA are present in polyribosomes. Following growth at 30°C, lysates from wild-type (A) and ΔrsrΔpnp (B) cells were fractionated in sucrose gradients. RNA extracted from each fraction was subjected to site-directed cleavage and Northern blotting to visualize 23S rRNA 3′ ends. Positions of 30S and 50S subunits, 70S ribosomes, and polysomes were determined by monitoring OD₂₆₀. The fractions were analyzed in two gels that are joined at the line. To examine whether pre-23S rRNAs were stable, RNA extracted from lysates (lane 1) was compared with RNA prepared by direct phenol extraction (lane 2). The prominent precursors with 71 and 79 extra 3′ nucleotides are indicated, along with a minor species containing 36 additional nucleotides. (Asterisk) A pre-23S rRNA degradation product.

Figure 6.

Rsr interacts with PNPase. (A) Serial fivefold dilutions of the indicated mutant strains were spotted on TGY agar and grown at 16°C, 25°C, 30°C, and 37°C. (B) Lysates from Flag₃-rsr (lanes 1,3,4) or untagged (lanes 2,5,6) strains were incubated with anti-PNPase antibody (lanes 3,5) or preimmune sera (lanes 4,6). Proteins in immunoprecipitates were subjected to Western blotting with an anti-Flag antibody. (Asterisk) A degradation fragment of Flag₃-Rsr. Although the cells for this experiment were grown at 30°C, similar results were obtained from cells grown at 37°C. (C) Serial fivefold dilutions of the strains were spotted on TGY agar containing either 0 or 0.4 mM hydrogen peroxide and grown at 25°C. (D) Wild-type (solid squares), Δpnp (solid circles), Δrsr (open triangles), and ΔrsrΔpnp (open circles) cells were irradiated with the indicated doses of UV. After irradiation, aliquots were removed and plated on TGY agar, and colonies were counted to determine the fraction of surviving cells.

Figure 7.

Possible model for the role of Rsr in 23S rRNA maturation. (A) In wild-type cells at 30°C, maturation of 23S rRNA by PNPase or another ribonuclease is inefficient, resulting in precursor accumulation. (B) In wild-type cells at 37°C, maturation is efficient and requires Rsr, RNase PH, and RNase II. Although only Rsr and the ribonucleases are shown, maturation may involve additional proteins such as helicases. (C) In Δrsr cells, maturation is always inefficient, resulting in pre-23S rRNA accumulation. (D) Cells lacking the Y RNA contain excess free Rsr at both temperatures, resulting in efficient maturation. (E) In Δpnp cells, another ribonuclease, together with Rsr, carries out efficient maturation at both temperatures. (F) In ΔrsrΔpnp cells, neither the inefficient pathway involving PNPase nor the other efficient pathway involving Rsr are operational. In these cells, truncated 23S rRNAs (open arrowheads) accumulate, presumably through the action of other nuclease(s).