Role of the mammalian RNA polymerase II C-terminal domain (CTD) nonconsensus repeats in CTD stability and cell proliferation

Abstract

The C-terminal domain (CTD) of mammalian RNA polymerase II (Pol II) consists of 52 repeats of the consensus heptapeptide YSPTSPS and links transcription to the processing of pre-mRNA. The length of the CTD and the number of repeats diverging from the consensus sequence have increased through evolution, but their functional importance remains unknown. Here, we show that the deletion of repeats 1 to 3 or 52 leads to cleavage and degradation of the CTD from Pol II in vivo. Including these repeats, however, allowed the construction of stable, synthetic CTDs. To our surprise, polymerases consisting of just consensus repeats could support normal growth and viability of cells. We conclude that all other nonconsensus CTD repeats are dispensable for the transcription and pre-mRNA processing of genes essential for proliferation.

FIG. 1.

The mammalian Pol II CTD is composed of 52 repeats of the consensus heptapeptide YSPTSPS (left-hand panel). Deviations from the consensus sequence, largely in the distal half of the CTD, are marked red. Repeats 1 to 3 and 52 (CTD52), the focus of the present study, are boxed in red. A genomic clone of Pol II LS*wt was manipulated to introduce a variety of deletions and substitutions; an N-terminal HA tag allows identification after Western transfer (upper panel). The schematics highlight the essential features of each mutant: repeats conforming to the consensus YSPTSPS are turquoise, nonconsensus repeats are black or white, and CTD52 is red. PAGE reveals the phosphorylated (IIo), nonphosphorylated (IIa), and CTD-less (IIb) forms of Pol II LS*. Mutant 1 was produced by replacing repeats 51 and 52 with a sequence encoding part of repeat 52 (YSLTSPAISPTG; yellow). Mutants 2 to 6 were produced by cloning the given acidic and basic polypeptides or domains into 52′ of mutant 1. Mutants 7 to 11 display a variety of deletions and substitutions, as published previously. Stable Raji cell lines were produced for each mutant; after expression, cells were cultivated 48 h in the presence of α-amanitin before cells were harvested, and the different forms produced by each mutant separated by polyacrylamide gel electrophoresis, followed by Western analysis. The effect of these mutations on degradation to the Pol IIb form is visible in each blot and summarized as “Stability” in the right-hand column. A wt clone, and a deletion mutant (Δ50) known to degrade to IIb were included as controls.

FIG. 2.

Replacement of repeats 1 to 3 by consensus repeats induces degradation to Pol IIb. Deletion and substitution mutants of repeats 1 to 3 were created both in the wt and Δ50 backgrounds. Mutants 2 and 6 contain a substitution of repeats 1 to 3 for a motif from Abl1 (CTD-ID), known to interact with CTD52 (sequences shown below). Similarly, mutants 3 and 7 contain a random linker of the same length. In mutants 4 and 8, repeats 1 to 3 were replaced by repeat 52 flanked at each side by a consensus repeat. Stable Raji cell lines were produced for each mutant; after expression, cells were cultivated 48 h in the presence of α-amanitin before being harvested, and the different forms produced by each mutant were separated by polyacrylamide gel electrophoresis, followed by Western analysis. The Western blot reveals the different forms produced by each mutant; stability is summarized in the right-hand column.

FIG. 3.

Fusions with EGFP, but not acidic polypeptides, prevent degradation in the absence of repeats 1 to 3 and 52. (A) EGFP was cloned at the C terminus of mutants containing deletions of repeats 1 to 3, CTD52, and in combination. (B) Repeats 1 to 3 were deleted in the background of truncated CTD52 and fusions thereof with acidic or basic hexapeptides. The sequence used in mutants 11 and 12 corresponds to the C-terminal amino acids of D. melanogaster CTD. Stable Raji cell lines were produced after expression, cells were cultivated 48 h in the presence of α-amanitin before being harvested, and the different forms produced by each mutant were separated by polyacrylamide gel electrophoresis, followed by Western analysis; stability is summarized in the right-hand column.

FIG. 4.

Deletion of repeats 1 to 3 and CTD52 has no apparent effect on RNA processing, cell growth, and viability. (A) Deletion mutants of repeats 1 to 3 and CTD52 fused with EGFP were recloned into the expression vector pSTC-TK under the control of the constitutive thymidine-kinase promoter. HeLa cells transfected with these mutants were cultured in the presence of α-amanitin for several weeks, after which total RNA was extracted. As a control (K), nontransfected cells were cultured in α-amanitin for 48 h before RNA extraction. (B) The abundance of five different RNAs examined (upper panel), and the 18S and 28S rRNAs (lower panel) are shown.

FIG. 5.

Repeats 4 to 51 can be replaced by consensus repeats and still support growth and viability. (A) Artificial CTDs were produced containing 20 (mutant 1), 35 (mutant 2), and 55 (mutant 3) consensus repeats in place of repeats 4 to 51. The ability of these mutants to replace the endogenous Pol II LS was tested by cultivation of cells in the presence of 2 μg of α-amanitin/ml. The cumulative growth (cell number × 10⁴; B) and viability (C) of these cells were monitored over a period of weeks. A mutant known to be viable (wt) and two nonviable mutants (Δ50 and mutant 4) were included as controls. The number of living (N₁) and dead cells (N_d) was determined by trypan blue staining. The percentage of viable cells (V) was calculated by using the formula V = 100 × N₁(N₁ + N_d). Cumulative growth was calculated by multiplying the number of cells present by the factor by which the culture was split over the course of the experiment.