Mechanism of Transcription Termination by RNA Polymerase III Utilizes a Non-template Strand Sequence-Specific Signal Element

Abstract

Understanding the mechanism of transcription termination by a eukaryotic RNA polymerase (RNAP) has been limited by lack of a characterizable intermediate that reflects transition from an elongation complex to a true termination event. While other multisubunit RNAPs require multipartite cis-signals and/or ancillary factors to mediate pausing and release of the nascent transcript from the clutches of these enzymes, RNAP III does so with precision and efficiency on a simple oligo(dT) tract, independent of other cis-elements or trans-factors. We report an RNAP III pre-termination complex that reveals termination mechanisms controlled by sequence-specific elements in the non-template strand. Furthermore, the TFIIF-like RNAP III subunit C37 is required for this function of the non-template strand signal. The results reveal the RNAP III terminator as an information-rich control element. While the template strand promotes destabilization via a weak oligo(rU:dA) hybrid, the non-template strand provides distinct sequence-specific destabilizing information through interactions with the C37 subunit.

Figure 1. Pol III pauses at T1–T4 of a terminator before transcript release

A) Schematic of the experimental system. 10 nt RNA primer and 81 bp DNA are shown; oligo(dT) on the non-template strand is indicated as a T9 stretch. B) Time course assay for transcript release by RNAP III. R and B above lanes indicate released and bound transcripts respectively. Quantification of termination efficiency (TE) (Methods) reflect percentage of transcripts released at the terminator, below the lanes. RT and T to the sides of gel images represent read-through and terminated transcripts respectively. C) Lane tracings of B. D) Time course of RNAP III transcription in seconds; the 0.5 point is approximate. E) Transcript release assay for a 4T terminator with TE below; FL = full-length. F) Lane trace profiles comparing total transcripts at early time points on a template with a 4T stretch; positions of 1U–4U transcripts are indicated above. Numbers below show the % of total signal at the 4T stretch (see text). See also supplemental figures S1, and S2.

Figure 2. PTCs are metastable, active for release or nucleotide addition

A) Isolated PTCs were treated with different reagents as indicated above the lanes (SM = starting material; Buffer alone; 4NTP = 0.5 mM each ATP, CTP, GTP, UTP & 5 mM MgCl₂; UTP = 0.5 mM UTP & 5 mM MgCl₂; 3NTP = 0.5 mM ATP, GTP, CTP & 5 mM MgCl₂. T, R and B above lanes indicate total, released and bound. RT indicates read-through. The -1C with arrow points to a band representing the C nucleotide immediately preceding the terminator. Transcript bands representing the proximal and distal parts of the terminator are indicated by brackets to the left. Vertical line between lanes 10 & 11 indicate that lanes between them in the gel were removed. B) Lane tracings of the starting material (SM), and the released and bound transcripts from panel A.

Figure 3. C53/C37 and C11 are required for PTC formation

A) Time course transcription by RNAP III-holo and RNAP II-core, the latter with and without recombinant C53/C37 and C11 as indicated above the lanes. Labeling conventions are same as in Fig 1. Vertical lines between lanes indicate non-contiguous sections; lanes 4–6 and 16–18 were from a different gel than the others. B) Lane tracings of the 5 second time-points in A. C) Release and bound assay for the components indicted above the lanes. Transcripts from the proximal and distal parts of the terminator are indicated to the left; quantification of termination efficiency in the proximal and distal parts of the terminator are shown under the lanes as TE_P and TE_D respectively (Experimental procedures). D) Lane tracings of ‘released’ lanes in C; vertical arrows below help asses ratios of 8U to 9U transcripts (text). See also supplemental figure S3.

Figure 4. Non-template (NT) strand bases T3, T4 and T5 are required for PTC formation and transcript release

A) Time course of transcription by RNAP III-holo with full-length NT strands containing the terminator NT sequences above the lanes. No NT strand was added to reaction lanes 1–3. B) Transcript release assay for RNAP III-holo and RNAP III-core with terminator NT strand sequences as indicated above the lanes; I = deoxyinosine (dI) and U = deoxyuridine (dU). No NT strand was added to reaction lanes 3–4 and 7–8. The vertical lines between some lanes indicate non-contiguous sections. C) Transcript release assay for single base substitutions in the terminator NT strand. T5 and T4 indicate T->A substitution at NT strand positions 5 and 4 respectively. Labeling conventions are as in Fig 1. Proximal and distal parts of the terminator are indicated by brackets. Transcripts from the proximal and distal parts of the terminator are indicated to the left; termination efficiency in the proximal and distal parts of the terminator are shown under the lanes as TE_P and TE_D respectively. D) A model for the RNAP III holoenzyme mechanism of transcription termination. 1) The C53/C37/C11 subcomplex transforms the RNAP III EC to a 2) PTC upon addition of 4 Us. T3 and T4 of the NT strand constitute a signal for PTC formation. 3) C37 amino acids 226–230 and T5 of NT strand contribute to switching the PTC to 4) transcript release mode upon nucleotide addition. See also supplemental figure S4.