The SMC Loader Scc2 Promotes ncRNA Biogenesis and Translational Fidelity

Abstract

The Scc2-Scc4 complex is essential for loading the cohesin complex onto DNA. Cohesin has important roles in chromosome segregation, DSB repair, and chromosome condensation. Here we report that Scc2 is important for gene expression in budding yeast. Scc2 and the transcriptional regulator Paf1 collaborate to promote the production of Box H/ACA snoRNAs which guide pseudouridylation of RNAs including ribosomal RNA. Mutation of SCC2 was associated with defects in the production of ribosomal RNA, ribosome assembly, and splicing. While the scc2 mutant does not have a general defect in protein synthesis, it shows increased frameshifting and reduced cap-independent translation. These findings suggest Scc2 normally promotes a gene expression program that supports translational fidelity. We hypothesize that translational dysfunction may contribute to the human disorder Cornelia de Lange syndrome, which is caused by mutations in NIPBL, the human ortholog of SCC2.

Fig 1. scc2-4 is a partial loss of function mutation.

(A) Scc2 has conserved HEAT repeats in the C terminal domain. (B) The amino acid mutated in scc2-4 is conserved from yeast to human (E534). (C) A growth curve of the WT and scc2-4 mutant strains in YPD at 30°C is shown (left). Growth curves were measured using a TECAN machine. The scc2-4 mutant has a slower maximum growth rate at 30°C compared to the WT, p = 0.0001 (right). (D) The levels of Scc2-Myc were measured by Western blotting (left) and quantified (right, in triplicate with error bars). Pgk1 serves as a loading control. The E534K mutation does not significantly reduce the amount of Scc2 protein at permissive temperature (30°C).

Fig 2. The scc2-4 mutation compromises the association with genic regions at 30°C.

WT and scc2-4 mutant strains were cultured to mid-log phase (~OD₆₀₀ = 0.5–0.8) in YPD medium. Strains were crosslinked and chromatin extracted for ChIP. Metagene analysis was carried out for Scc2-Myc and Scc2^E534K-Myc for ChIP seq data. Two biological replicates for each are shown. (A) The scc2-4 mutation does not affect the association of Scc2 with centromere regions. (B) The mutation reduces the association with ribosomal protein genes (132) (C) rDNA (D) snoDNAs (77) and (E) tDNAs (275).

Fig 3. Hundreds of genes are differentially expressed in the scc2-4 mutant compared to WT at 30°C in YPD.

(A) Gene expression values (mean of triplicate samples) are shown in MA plot. Differentially expressed genes (adjusted p-value <0.05) were colored in orange, and then colored red or green after a minimum fold change cut-off of 1.5 was applied. There are 2644 genes differentially expressed genes in the scc2-4 mutant, with 1285 up-regulated and 1359 down-regulated. Applying a fold-change cutoff of 1.5 (corresponding to an absolute log₂ value of 0.6) returns a more reasonable number of genes for GO analysis and general comparison. (B) GO term analysis for the up-regulated genes shows enrichment for genes important for RNA processing/metabolism and ribosome biogenesis. (C) GO term analysis for down-regulated genes shows enrichment for genes important for biological processes such as oxidative phosphorylation, electron transport chain, and carbohydrate metabolic processes.

Fig 4. Ribosomal RNA production is compromised in the scc2-4 mutant.

(A) The RNA synthesis rate was examined using ³H-uridine incorporation in WT and scc2-4 mutant strains. Strains were grown in triplicate in SD-ura medium with minimal uracil to an approximate OD₆₀₀ of 0.3. ³H-uridine was added for 5 minutes and incorporation was quantified, averaged, and expressed relative to WT. Standard error bars are indicated for n = 4. The difference between WT and mutant was significant at p<0.0001. (B). WT and scc2-4 mutant strains were grown in SD-met at 30°C into mid-log phase. Equal numbers of cells were pulse-labeled with ³H-methylmethionine for 5 min, followed by RNA extraction and electrophoresis of the RNA on a formaldehyde gel. The gel was photographed following ethidium bromide staining (bottom). RNA was transferred to a membrane and detected by autoradiography (top). 25S and 18S rRNA were cut from the membrane, quantified with a scintillation counter, and the average is expressed relative to WT (p<0.05). p-values were calculated using a student two-tailed t-test. Standard error is indicated for n = 4.

Fig 5. Ribosome protein distribution is compromised in the scc2-4 mutant.

Protein components of the small (Rps2) and large (Rpl25) ribosomal subunits were tagged with GFP and imaged. Representative images are shown for WT and mutant strains (A and D). Using flow cytometry, the peak GFP intensity was quantified for independent biological replicates (B and E). At least 10000 cells were examined per replicate. A KS-distance was calculated from cumulative distribution frequency curves of the fluorescence which allows us to determine statistical significance (C and F) (p<0.0001).

Fig 6. Pseudouridylation of rRNA is reduced in the scc2-4 mutant.

(A) The Box H/ACA snoRNAs that guide sequence-specific pseudouridylation are down-regulated in the scc2-4 mutant. The error bars represent the standard deviation from triplicate samples and the asterisks indicate statistical significance at an adj p<0.05. (B) Reverse transcription with primers corresponding to residues Ψ 1003 and Ψ1123 for SNR5 and residues Ψ2258 and Ψ2260 for SNR191 was performed. Samples were treated with or without CMC, exposed to pH 10.4 for 4 hrs and reverse transcribed. The rectangles indicate the bands quantified to the right. Pseudouridylation assays were performed at least two times; a representative experiment is shown.

Fig 7. Scc2 and Paf1 recruitment at snoDNAs is co-dependent.

(A) WT, scc2-4, and paf1Δ mutant strains were cultured in YPD medium to mid-log phase (~ OD₆₀₀ = 0.5–0.8). Strains were crosslinked and chromatin extracted for ChIP. ChIP/qPCR analysis was carried out for Scc2-Myc and Scc2^E534K-Myc and Scc2-Myc paf1Δ. ChIP performed without the addition of primary antibody served as a negative control. ChIP experiments were performed at least three times. p-values were calculated using a student t-test. Standard error bars are indicated for n = 3. Values different from the WT are indicated by an asterisk (p<0.05). qPCR analysis shows reduced enrichment of indicated snoDNAs in mutant relative to WT in Paf1 ChIP (α-HA antibody, 12CA5, Roche). (B) Western blot analysis shows that the protein level of Paf1 is reduced in the scc2-4 mutant. Scc2 levels also appear reduced in the paf1Δ strain. Pgk1 served as the loading control. (C) Western blot analysis shows that the protein level of Ctr9 is reduced in the scc2-4 mutant. (D) qPCR analysis shows reduced enrichment of snoDNAs in the Scc2-Myc ChIP in the paf1Δ mutant strain relative to WT (α-Myc antibody, 9B11, Cell signaling). (E) A cartoon model for the co-recruitment of Scc2 and Paf1 at snoDNAs is shown. The Tbf1 transcription factor is shown in purple, RNA Pol II in yellow, and Scc2 and Paf1c in blue and green, respectively.

Fig 8. Translational fidelity is reduced in the scc2-4 mutant.

Dual luciferase reporters were used to measure translation. (A, top) Schematic diagram of CrPV IGR IRES containing reporter. Renilla luciferase is translated by a cap-dependent mechanism, and firefly luciferase synthesis requires cap-independent initiation mediated by the IRES. While cap-dependent translation is mildly impaired (p = 0.01), IRES-dependent translation is more strongly inhibited in the scc2-4 mutant (p = 0.001). (B, top) A dual luciferase reporter was used to monitor -1 frameshifting mediated by sequence derived from the yeast L-A virus, and +1 frameshifting promoted by the yeast Ty1 sequence. In-frame renilla luciferase translation serves as a normalization control, and efficiencies were determined as previously described (Landry et al., 2009) The mutant shows an increase in -1 frameshifting. (C). Readthrough efficiency for three different stop codons (UAA, UAG, UGA) was measured for the WT and scc2-4 strains. The translation of renilla luciferase serves as a normalization control. (D) The maximal growth rates of the WT and scc2-4 mutant strains were compared, with the WT growth rate set to 100%. Strains were grown in YPD or YPD with paromomycin. p-values were calculated using student two-tailed t-test (asterisk indicates p<0.05). Standard error is indicated for at least three independent measurements.