Arabidopsis AtRRP44A is the functional homolog of Rrp44/Dis3, an exosome component, is essential for viability and is required for RNA processing and degradation

Abstract

The RNA exosome is a multi-subunit complex that is responsible for 3' to 5' degradation and processing of cellular RNA. Rrp44/Dis3 is the catalytic center of the exosome in yeast and humans. However, the role of Rrp44/Dis3 homologs in plants is still unidentified. Here, we show that Arabidopsis AtRRP44A is the functional homolog of Rrp44/Dis3, is essential for plant viability and is required for RNA processing and degradation. We characterized AtRRP44A and AtRRP44B/SOV, two predicted Arabidopsis Rrp44/Dis3 homologs. AtRRP44A could functionally replace S. cerevisiae Rrp44/Dis3, but AtRRP44B/SOV could not. rrp44a knock-down mutants showed typical phenotypes of exosome function deficiency, 5.8S rRNA 3' extension and rRNA maturation by-product over-accumulation, but rrp44b mutants did not. Conversely, AtRRP44B/SOV mutants showed elevated levels of a selected mRNA, on which rrp44a did not have detectable effects. Although T-DNA insertion mutants of AtRRP44B/SOV had no obvious phenotype, those of AtRRP44A showed defects in female gametophyte development and early embryogenesis. These results indicate that AtRRP44A and AtRRP44B/SOV have independent roles for RNA turnover in plants.

Figure 1. Arabidopsis Rrp44/Dis3 homologs.

(A) Schematics of the Rrp44/Dis3 homologs S. cerevisiae Rrp44 (ScRrp44), human RRP44/DIS3 (hRRP44), and A. thaliana AtRRP44A and AtRRP44B/SOV. Yellow and blue boxes represent the PIN and RNB domains, respectively, that are conserved among Rrp44/Dis3 homologs. aa represents amino acids. (B) AtRRP44A complements the S. cerevisiae rrp44 doxycycline (DOX) repressible mutant. Growth phenotypes resulting from the expression of plasmid-borne AtRRP44A, AtRRP44B/SOV, AtRRP44A and AtRRP44B/SOV, and ScRrp44 in S. cerevisiae BSY1883 strain, and negative control alleles were assessed in the presence (repressed chromosomal ScRrp44) or absence (expressed chromosomal ScRrp44) of DOX after incubation for 90 h at 30°C. –LEU-TRP, without leucine and tryptophan. (C) Diagram of the intron–exon structure of AtRRP44A and AtRRP44B/SOV. UTRs are indicated by grey boxes, exons by black boxes and introns by solid lines. T-DNA insertion sites for rrp44a-1 (SALK_037533), rrp44a-2 (SALK_141741), rrp44a-3 (SALK_051800), rrp44b-1 (SAIL_804_F05), rrp44b-2 (SALK_017934) and rrp44b-3 (SALK_010765) are shown in red arrowheads.

Figure 2. Establishment of rrp44a knock-down mutants by artificial microRNA (amiR).

(A) Schematics of the amiR precursor. A black circle represents cap structure. amiR and amiR* represent guide strand and passenger strand, respectively. amiR sequences targeting AtRRP44A (amiR_AtRRP44A-1 and amiR_AtRRP44A-2) and E. coli β-glucuronidase (GUS) (amiR_GUS-2; vector control with no target sites in the Arabidopsis genome) are shown. (B) AmiR sequences and target sites in the AtRRP44A (AT2G17510) mRNA. (C) Expression of amiR_RRP44A-1 and amiR_RRP44A-2 were detected by small RNA gel blot analysis. Small RNA gel blots were hybridized with an antisense oligonucleotide complementary to the amiRs. U6 RNA (U6) served as a loading control for small RNA. Total RNAs were isolated from 25 days post-germination (dpg) rosette leaves of T3 homozygous lines carrying a unique insertion in Col-0 background plants expressing amiR_RRP44A-1 (AtRRP44A Knocked Down-1; rrp44aKD-1#7-3-1), amiR_RRP44A-2 (rrp44aKD-2#6-2-1), or amiR_GUS-2 (gusKD-2#2-10-3: vector control (VC)). (D) The amounts of AtRRP44A mRNA in gusKD-2#2-10-3, rrp44aKD-1#7-3-1 and rrp44aKD-2#6-2-1 were analyzed by qRT-PCR. Total RNAs were isolated from 25 dpg rosette leaves. Error bars represent standard errors. Six biological replicates and two technical replicates were performed. * indicates significant difference (p < 0.01, Tukey’s test) between gusKD-2 (VC) and rrp44aKD-1 and -2.

Figure 3. Analysis of rRNA processing and degradation.

(A) Diagram illustrating the 5.8S rRNA processing intermediates and the rRNA maturation by-product generated from the 5ʹ ETS (P-Pʹ) compared with the 35S precursor [27]. Horizontal red arrows represent the positions of oligonucleotide probes used in this study. (B) The 5.8S rRNA 3ʹ extension is processed by AtRRP44A, AtRRP4 and AtRRP41, but not AtRRP44B/SOV. (C) The 5ʹ ETS is degraded by AtRRP44A, AtRRP4 and AtRRP41, but not AtRRP44B/SOV. RNA gel blots of 5.8S rRNA precursors (B) or the 5ʹ ETS (C). Total RNAs were isolated from 10 dpg rosette leaves of Col-0 (wild type: WT), gusKD-2 (VC), rrp44aKD-1, rrp44aKD-2, rrp44b-1, rrp44b-2 and mtr4-1 plants or from gusKD-2, rrp4KD-3, rrp41KD-1 and rrp44aKD-1 plants (B and C). mtr4-1 was used to determine the sequence of 5.8S processing intermediates [27]. Total RNAs were separated on 6% polyacrylamide gels. Methylene blue staining of 5S rRNA is shown as a loading control. Relative RNA levels estimated from band signals are indicated at the bottom of each lane as mean values ± SE with RNA levels in Col-0 plants set to 1.0.. Values for which P<0.05 (Tukey’s test) compared to corresponding wild type plants (gusKD-2 or Col-0) were shown in red. Two (B and C: Left panels) or three (B and C: Right panels) biological replicates were performed for all RNA gel blots.

Figure 4. Levels of MRP RNA and snoRNA31 in rrp44aKD-1, rrp44b-2 and the exosome core mutants.

(A and B) qRT-PCR revealed that accumulation of the MRP RNA and snoRNA31 was upregulated in rrp4KD-3, rrp41KD-1 and rrp44aKD-1, but not in rrp44b-1. AtRRP4 and AtRRP41 represent the Arabidopsis exosome core. Total RNAs were isolated from 10 dpg rosette leaves. EF1a mRNA was used as an endogenous control. Error bars represent standard errors. Three biological replicates and two technical replicates were performed. * indicates significant difference (p < 0.05, Tukey’s test) between mutant and wild type plants.

Figure 5. Levels of selected subsets of RNAs in rrp44aKD-1, rrp44b-2 and the exosome core mutants.

qRT-PCR analysis of total RNAs isolated from 7 dpg leaves for AtRRP4 and AtRRP41 (AT5G11090 3ʹ extension, AT5G27720-Intron) and AtRRP41L (NCED3) substrates [10,30] (A–C). UTRs are indicated by grey boxes, exons by black boxes, introns by solid lines and the 3ʹ extended region by a black broken line (A and B). Green lines show the coverage of amplicons used for qRT-PCR. Error bars represent standard errors. Three biological replicates and two technical replicates were performed. EF1a mRNA was used as an endogenous control. * indicates significant difference (p < 0.05, Tukey’s test) between mutants and gusKD-2 (VC) or Col-0 (WT) plants.

Figure 6. Characterization of AtRRP44A and AtRRP44B/SOV T-DNA insertion mutants.

(A) Semi-sterility phenotype of the rrp44a-1 mutant. (B) rrp44b-1, rrp44b-2 and rrp44b-3 have normal phenotypes. rrp44b-1 and rrp44b-2 mutants are RNA null. rrp44b-3 mutants showed reduced RNA levels. Reverse Transcriptional PCR analysis of AtRRP44B/SOV, using total RNA isolated from 24 dpg rrp44b-1, rrp44b-2 and rrp44b-3 homozygous mutants. All samples showed similar levels of EF1a mRNA (Lower), but AtRRP44B/SOV mRNAs were not detectable (in rrp44b-1 and rrp44b-2) or were decreased (in rrp44b-3). Primers used for this analysis are listed in Table S3.

Figure 7. Model for the roles of A. thaliana AtRRP44A and AtRRP44B/SOV in RNA processing and degradation.

AtRRP44A localizes to the nucleus and processes rRNAs with the exosome complex. However, AtRRP44B/SOV localizes to the cytoplasm and targets a select subset of mRNAs.