Dhr1p, a putative DEAH-box RNA helicase, is associated with the box C+D snoRNP U3

Abstract

Putative RNA helicases are involved in most aspects of gene expression. All previously characterized members of the DEAH-box family of putative RNA helicases are involved in pre-mRNA splicing. Here we report the analysis of two novel DEAH-box RNA helicases, Dhr1p and Dhr2p, that were found to be predominantly nucleolar. Both genes are essential for viability, and MET-regulated alleles were therefore created. Depletion of Dhr1p or Dhr2p had no detectable effect on pre-mRNA splicing in vivo or in vitro. Both Dhr1p and Dhr2p were, however, required for 18S rRNA synthesis. Depletion of Dhr2p inhibited pre-rRNA cleavage at sites A(0), A(1), and A(2), while Dhr1p depletion inhibited cleavage at sites A(1) and A(2). No coprecipitation of snoRNAs was detected with ProtA-Dhr2p, but Dhr1p-ProtA was stably associated with the U3 snoRNA. Depletion of Dhr1p inhibited processing steps that require base pairing of U3 to the 5' end of the 18S rRNA. We speculate that Dhr1p is targeted to the preribosomal particles by the U3-18S rRNA interaction and is required for the structural reorganization of the rRNA during formation of the central pseudoknot.

FIG. 1

Pre-rRNA processing pathway in yeast. (A) Structure of the rRNA gene operon and location of the oligonucleotides used in this work. The mature 18S, 5.8S, and 25S rRNAs (bold lines) are released from the 35S primary transcript following cleavages in the 5′-ETS and 3′-ETS and ITS1 and ITS2. Cleavage sites are indicated by uppercase letters (A₀ to D). Oligonucleotides used for the Northern blot hybridization and primer extension experiments are indicated by lowercase letters (a to h). (B) Pre-rRNA processing pathway. The 35S pre-rRNA is successively cleaved in the 5′-ETS at site A₀ (generating the 33S RNA), at site A₁, the 5′ end of the mature 18S rRNA (generating the 32S RNA), and at site A₂ in ITS1 (generating the 20S and 27SA₂ pre-rRNAs). The 20S is dimethylated by Dim1p and cleaved at site D to generate the mature 18S rRNA. The 27SA₂ is matured to the 5.8S and 25S rRNAs following two alternative pathways. Eighty-five percent of the 27SA₂ population is cleaved at site A₃ in ITS1 by RNase MRP; this is rapidly followed by 5′-to-3′ trimming to site B_1S by Xrn1p and Rat1p. Fifteen percent of 27SA₂ is cleaved at site B_1L. Cleavage at site B₂ occurs concomitantly with cleavage at site B₁. The two forms of 27SB (27SB_S and 27SB_L) are matured following identical pathways involving cleavage at sites C₁ and C₂ and 3′-to-5′ exonucleolytic digestion to site E by the exosome. Both Dhr1p and Dhr2p are required for cleavages at sites A₁ and A₂; Dhr2p is also required for cleavage at site A₀. Base pairing between U3 and the pre-rRNA in the 5′-ETS and 18S regions is required for cleavage at sites A₁ and A₂; the U3–5′-ETS interaction is also required for cleavage at site A₀. (C) Structure of aberrant pre-rRNA processing intermediates detected on depletion of Dhr1p and Dhr2p. Premature cleavage of the 35S pre-rRNA at site A₃ in ITS1 generates the 23S RNA. The 22.5S RNA results from cleavage of pre-rRNA molecules at site D in the absence of cleavage at sites A₀ and A₁. The 22S RNA is accumulated when cleavages occur at sites A₀ and A₃ in the absence of cleavage at sites A₁ and A₂.

FIG. 2

Construction of MET-regulated alleles of DHR1 and DHR2. (A) Growth rates of MET::HA-dhr1 (closed diamonds), MET::HA-dhr2 (closed squares), and wild-type (open squares) strains following addition of methionine to the medium. Values are corrected for dilution. (B) Western blot analysis of strains described in panel A. Total protein was extracted from similar amounts of cells (according to OD₆₀₀), separated on SDS-polyacrylamide gels (8 and 15% PAGE for Dhr1p and Dhr2p, respectively), and transferred to nitrocellulose. Membranes were decorated with anti-HA antibodies. Two protein components of the signal recognition particle, Srp14p (16.4 kDa) and Srp68p (69.0 kDa), were decorated with specific antibodies as loading controls.

FIG. 3

Dhr1p and Dhr2p localize to the nucleolus. Indirect immunofluorescence on cells expressing an HA epitope fusion of either Dhr1p or Dhr2p. The Dhr fusions were detected with a rat anti-HA (α-HA) monoclonal antibody, followed by FITC staining. For Nop1p detection, a mouse anti-Nop1p (α-Nop1p) monoclonal antibody was used in combination with Cy3 staining. Both HA fusions showed an intranuclear staining (green channel) which colocalized with Nop1p (red channel) and revealed a classical nucleolar crescent-like shape structure (yellow in merged images). DNA was stained with DAPI (blue in merged images). W.T., wild type.

FIG. 4

Dhr1p and Dhr2p are required for 18S rRNA synthesis. Total RNA was extracted from MET::dhr1, MET::dhr2, and otherwise isogenic wild-type (W.T.) strains grown in selective minimal medium lacking methionine (0-h time points) and following addition of methionine for the times indicated. RNA was resolved in a 1.2% agarose-formaldehyde gel, transferred to nylon membrane and hybridized with the oligonucleotides indicated. Oligonucleotide (oligo) b does not distinguish between the 33S and 32S pre-rRNAs, while oligonucleotides d and e do not distinguish between the 27SA₂ and 27SA₃ pre-rRNAs. However, the 33S and 27SA₃ pre-rRNAs are very low in abundance and the signals observed are predominantly due to the 32S and 27SA₂ pre-rRNAs.

FIG. 5

Dhr1p and Dhr2p are not involved in the maturation of the 7S pre-rRNA. Total RNA was extracted from MET::dhr1, MET::dhr2, and isogenic wild-type (W.T.) strains grown in selective minimal medium lacking methionine (0-h time points) and following addition of methionine for the times indicated. RNA was resolved in an 8% polyacrylamide gel, transferred to nylon membrane, and hybridized with oligonucleotide (oligo) e (I) or h (II).

FIG. 6

Dhr1p and Dhr2p are required for cleavages in 5′-ETS and ITS1. Primer extension analysis through 5′-ETS and ITS1 was performed from oligonucleotides a and e (diagram on left). Total RNA was extracted from MET::dhr1, MET::dhr2, and isogenic wild-type (W.T.) strains grown in selective minimal medium lacking methionine (0-h time points) and following addition of methionine for the times indicated. On Dhr1p depletion, the primer extension stop at site A₀ is substantially elevated and the corresponding panel (lanes 1 to 6) is from a sixfold shorter exposure than lanes 7 to 14.

FIG. 7

Dhr1p-ProtA is specifically associated with the U3 snoRNA. Coprecipitation experiments with IgG-agarose beads were performed with lysates from cells expressing ProtA epitope-tagged versions of Dhr1p and Dhr2p. For Dhr1p-ProtA, two independently isolated strains were used. (A and B) Northern blot analyses. RNA was extracted from equivalent amounts of the total (T), supernatant (S), and pellet (P) fractions and loaded in a 1:1:1 ratio. Membranes were hybridized with oligonucleotides specific for the box C+D snoRNAs (U3, U14, and snR190) and the H+ACA snoRNAs (snR10 and snR30).

FIG. 8

Dhr1p-ProtA is specifically associated with Mpp10p. Coprecipitation experiments with IgG-agarose beads were performed with two independently isolated strains expressing a Dhr1p-ProtA fusion. Proteins from the total (T), supernatant (S), and pellet (P) fractions (in a 1:1:35 ratio; see Materials and Methods) were resolved by SDS-PAGE and transferred to nitrocellulose. Membranes were decorated with PAP or anti-Mpp10p antibodies.

FIG. 9

Dhr1p and Dhr2p are not required for accumulation of snoRNAs. Total RNA was extracted from MET::dhr1, MET::dhr2, and otherwise isogenic wild-type (W.T.) strains grown in selective minimal medium lacking methionine (0-h time points) and following addition of methionine for the times indicated. Total RNA was resolved in an 8% polyacrylamide gel, transferred to a nylon membrane, and hybridized with oligonucleotides specific for the box C+D snoRNAs (U3 and U14) and the H+ACA snoRNAs (snR10 and snR30).