Unphosphorylated SR-like protein Npl3 stimulates RNA polymerase II elongation

Abstract

The production of a functional mRNA is regulated at every step of transcription. An area not well-understood is the transition of RNA polymerase II from elongation to termination. The S. cerevisiae SR-like protein Npl3 functions to negatively regulate transcription termination by antagonizing the binding of polyA/termination proteins to the mRNA. In this study, Npl3 is shown to interact with the CTD and have a direct stimulatory effect on the elongation activity of the polymerase. The interaction is inhibited by phosphorylation of Npl3. In addition, Casein Kinase 2 was found to be required for the phosphorylation of Npl3 and affect its ability to compete against Rna15 (Cleavage Factor I) for binding to polyA signals. Our results suggest that phosphorylation of Npl3 promotes its dissociation from the mRNA/RNAP II, and contributes to the association of the polyA/termination factor Rna15. This work defines a novel role for Npl3 in elongation and its regulation by phosphorylation.

Figure 1. Npl3 stimulates RNAP II elongation.

Oligo(dC)-tail templates were used in transcription reactions with or without 78 nM Npl3, as indicated. Reactions were pulsed by the addition of ATP, GTP, and α ³²P-labeled CTP for 30 minutes as described in Materials & Methods. Transcripts were chased by the addition of excess CTP and limiting UTP for the indicated times. A schematic of the experimental scheme is shown on top. The oligo(dC)-template is represented next to the gel and the positions of the pause sites at stretches of Ts are indicated.

Figure 2. Npl3 interaction with RNAP II is affected by S411 phosphorylation.

(A) Unphosphorylated peptide RGG/RS1 (lane 1), RGG/RS2 (lane 2), RGG/RS3 (lane 3) and phosphorylated RGG/RS3 (lane 4) were used in pull-down assays with purified RNAP II, as indicated, and one fifth of the input is shown in lane 5. The precipitated sample (P) and supernatant (S) were analyzed by immunoblotting. (B) Full-length Npl3 or S411 mutants were used in immunoprecipitation assays with purified RNAP II (100 ng), as indicated. A control lane with RNAP II (10 ng) is shown in lane 13. (C) Oligo(dC)-tail transcription reactions were performed as described in Figure 1 for 30 minutes with RNAP II only (lane 1) or with equivalent concentrations of Npl3 (lane 2); Npl3-S411A (lane 3); -S411D (lane 4); -S411E (lane 5); Npl3-truncated (RRMs only, aa 121–280) (lane 6); or Npl3-120 (lane 7). Quantification of the transcription reactions was performed by calculating the ratio of the 450 nt to the 250 nt bands and normalized using the RNAP II only lane, as shown in the graph. The experiment was done twice, and a representative gel was chosen. (D) Unphosphorylated (lane 1 and 6), or phosphorylated Ser 5 (lane 2 and 7), Ser 2 (lane 3 and 8), and Ser 2/Ser 5 (lane 4 and 9) CTD peptides were used in pull-down assays with full-length Npl3, as indicated, and input is shown in lane 11. The IP and S shown were analyzed as described for (C) using antibodies specific for Npl3. Control Npl3 with no peptide is shown in lanes 5 and 10.

Figure 3. Cka1, the alpha catalytic subunit of CK2, phosphorylates Npl3.

(A) Recombinant His-Cka1 was incubated with His-Npl3 in the presence of radiolabeled ATP. Phosphorylation of Npl3 is reversed by the addition of increasing concentrations of λ-Phosphatase. (B) Using the in vitro kinase assay with mutants of Npl3-S411, this residue was uncovered as an additional phosphorylation site. Wild-type His-Npl3 or S411 point mutants, His-Npl3-S411A, -S411D or -S411E were incubated with His-Cka1, as described for (A) in the presence of radiolabeled-ATP. Coomasie stain representing the concentration of recombinant Npl3 proteins used is shown below. (C) CKA1 mutations show reduced phosphorylation of Npl3. Whole-cell extracts were prepared for cells grown for one hour at the permissive (25°C) or non-permissive (37°C) temperature for wild-type CKA1, cka1-12 or cka1-45, and immunoblot analysis was performed using antibodies specific for phosphorylated or non-phosphorylated Npl3, as indicated. Detection of Rpb3 with specific antibodies is shown as a loading control.

Figure 4. Cka1 disrupts the ability of Npl3 to effectively compete for binding to an Rna15-preferred sequence.

(A) Recombinant Npl3, Rna15 and Cka1 were incubated with a radiolabeled RNA oligo (N4), UV cross-linked, and resolved in denaturing 10% SDS-PAGE gels. This RNA oligo consists of an A-rich repeat (described in Materials and Methods), which is the preferred binding site for Rna15 and is commonly found upstream of polyA sites . Representative UV-cross-linking experiments are shown where increasing Cka1 is added to reactions containing Npl3 with Rna15. The graph below each gel shows quantification for the average of three experiments (Npl3, white bars; Rna15, black bars). Binding levels were calculated as fractions relative to a reaction containing the highest concentration of the individual RNA-bound protein (lane 1 for Npl3 and lane 2 for Rna15). Control UV-crosslinking experiment where Cka1 has been added with or without ATP is also shown. Values represent total PhosphorImager units (PIU). (B) Recombinant Npl3, Rna15 and Cka1 were incubated with a radiolabeled U/G/C-rich RNA oligo (N2), UV-cross-linked, and resolved in denaturing 10% SDS-PAGE gels and quantified as described for (A). (C) Similar reactions were performed as described for Figure 1 for 0 and 30 minutes without (lane 1 and 5) or with 78 nM Npl3 (lanes 2–3 for 0 min time-point; and 6–7 for the 30 min time-point). 200 nM Rna15 was added to the transcription reactions in the presence or absence of Npl3 (lanes 3–4 for 0 min time-point; and 7–8 for the 30 min time-point). The oligo(dC)-template is represented next to the gel and the positions of the pause sites at stretches of Ts are indicated. Quantification of the transcription reactions was performed as shown for Figure 2C.

Figure 5. Opposing effects in termination demonstrated using RNA-binding or phosphorylation defective npl3 alleles.

RNA was extracted from NPL3, npl3-120 and npl3-S411A and whole transcript sense strand cDNA was synthesized and used for hybridization to tiled arrays. The hybridization signals for (A) FBA1, MPE1 and (B) TDH3 for two independent experiments were used to calculate the ratio of mutant vs. wild-type. The top panel shows the ratio of rna15-2/RNA15; bottom panel shows the ratio of each npl3 mutant/NPL3; a solid line for npl3-120/NPL3, or dashed line for npl3-S411A/NPL3. The corresponding position and orientation (Watson (W+), dashed bars, or Crick (C-) strand, black solid bars) of the genes is also shown. (C–D) A 300 bp region at the beginning of each ORF (+50 to +350 relative to the start codon), was selected for comparison to the same size region at the 3′UTR (−100 to +200 relative to the stop codon) of the genes that showed increase readthrough for the npl3-S411A strain. The mean value of the log₂ ratio of npl3-S411A/NPL3 was calculated and graphed in a scatter plot. (D) The total number of genes corresponding to the Crick strand were used to calculate the mean value of the same region described for (C).

Figure 6. Two activities of Npl3 promote elongation.

Npl3 directly stimulates the elongation rate of RNAP II by physically interacting with phospho-Ser 2 of the CTD and the RNA. Phosphorylation regulates Npl3's interaction with RNAP II and RNA, and promotes binding of polyadenylation/termination factors. Binding of Npl3 to the nascent RNA may stabilize interactions between CTD-RNAP II and the RNA. Npl3, white circle; Rna15, small black diamond; CFI, large black diamond; RNAPII, grey shaded oval.