Pausing by bacterial RNA polymerase is mediated by mechanistically distinct classes of signals - PubMed (original) (raw)
Pausing by bacterial RNA polymerase is mediated by mechanistically distinct classes of signals
I Artsimovitch et al. Proc Natl Acad Sci U S A. 2000.
Abstract
Transcript elongation by RNA polymerase is discontinuous and interrupted by pauses that play key regulatory roles. We show here that two different classes of pause signals punctuate elongation. Class I pauses, discovered in enteric bacteria, depend on interaction of a nascent RNA structure with RNA polymerase to displace the 3' OH away from the catalytic center. Class II pauses, which may predominate in eukaryotes, cause RNA polymerase to slide backwards along DNA and RNA and to occlude the active site with nascent RNA. These pauses differ in their responses to antisense oligonucleotides, pyrophosphate, GreA, and general elongation factors NusA and NusG. In contrast, substitutions in RNA polymerase that increase or decrease the rate of RNA synthesis affect both pause classes similarly. We propose that both pause classes, as well as arrest and termination, arise from a common intermediate that itself binds NTP substrate weakly.
Figures
Figure 1
Properties of two classes of pause signals. (Top) Possible positions of the RNA 3′ nt during active elongation or pausing. Structures of class I and class II paused TECs are depicted with DNA (black) entering RNAP (blue) from downstream at right and separating near the active site (white circles). RNA (red) pairs with template DNA in an eight-bp hybrid (vertical red lines), then exits under the β flap domain (37). In the examples, the RNA nt in the hybrid (in the active state) are underlined. Nucleotide addition occurs when the RNA 3′ OH and template-specified NTP simultaneously occupy the left and right halves of RNAP's bipartite active site (i and i+1, respectively). At a class I pause, interaction of the pause hairpin with the β flap domain displaces the RNA 3′ OH away from the catalytic center. At a class II pause, RNAP enters pretranslocated or backtracked conformations (dashed red line). Backtracked RNA may enter the secondary channel through which NTPs are thought to enter (gray dotted outline; see refs. , , and 47). These structures are consistent with the tabulated differences in sensitivity to pyrophosphorolysis and transcript cleavage, effect of changes in hybrid stability on pausing, 5′ limit of positions at which antisense oligos can reduce pausing, effects of NusA and NusG, and effects of fast and slow RNAP mutants.
Figure 2
Pausing at the his and ops pause sites. (A) Preformed [α-32P]CMP-labeled A29 complexes were incubated with 10 μM GTP, 150 μM ATP, CTP, and UTP on the his (Left) or ops (Right) template. Samples were taken at the times indicated in seconds above each lane. Prominent transcripts (and lengths for his and ops templates, respectively) are P, pause RNA transcript (71 and 62 nt); T, terminated transcript (150 and 138 nt); and RO, run-off RNA transcript (381 and 369 nt). The fractions of the pause RNA (closed circles) were plotted against time; the pause half-life and efficiency were determined as described previously (24). Fractions of pause RNAs from preformed complexes (open squares; gels not shown) were determined after paused TECs were halted by NTP deprivation for ≈5 min.
Figure 3
Oligos and pyrophosphate affect the his and ops pause complexes differently. (A) The pause half-lives are plotted (as a fraction of a “no-oligo” control) by the 3′-most nucleotide of the nascent RNA that remains outside the RNA:oligo duplex (equivalent to the 5′-terminal base of the 22-nt oligonucleotides). Each value is an average of at least two independent measurements. (B) Immobilized TECs were halted along the templates encoding the his (Left) or ops (Right) pause signals at the positions indicated below each panel and treated with increasing concentrations of PPi (0, 0.001, 0.01, 0.1, and 1 mM) as described previously (16).
Figure 4
Elongation factors NusA and NusG preferentially target his and ops pause complexes, respectively. [α-32P]CTP-labeled A29 complexes (40 nM) were preformed on his or ops pause templates. Transcription was allowed to resume in the absence of additional factors (Left) or in the presence of 50 nM NusA (Center) or NusG (Right). The pause half-life and efficiency (see Materials and Methods) are indicated below each panel.
Figure 5
Fast and slow RNAP mutants recognize his and ops pause sites similarly. [α 32P]-CTP-labeled A29 complexes were formed on his (Left) or ops (Right) pause templates with wild-type (WT), RpoB5101 and RpoB8 RNAPs and then incubated with NTPs (see Materials and Methods).
Figure 6
Slow intermediate mechanism. RNAP is depicted as in Fig. 1. At most template positions, fast translocation from the pretranslocated state to the active state allows tight NTP binding (assisted by a properly positioned 3′OH) and rapid transcription (top horizontal pathway)]. When RNAP encounters a pause, arrest, or termination site, it isomerizes to a slow intermediate in which the RNA 3′ end frays away from the DNA. A slight conformational opening of RNAP may precede and accelerate this change or may accompany it. Further rearrangement of the slow intermediate produces the different classes of paused, arrested, or terminating complexes. Escape of the slow intermediate back to the elongation pathway occurs by weak NTP binding and recapture of the 3′ OH in the active site. Amino acid substitutions in RNAP favor or disfavor the slow intermediate, whereas elongation factors NusA and NusG stabilize hairpin-RNAP interaction or inhibit backtracking, respectively, at later steps in the pathway. Whether termination sometimes involves hairpin-RNAP interaction (dotted line) and whether it occurs via hairpin-induced bubble collapse or RNA pull-out (18, 36) remains to be determined.
References
- Richardson J, Greenblatt J. In: Escherichia coli and Salmonella: Cellular and Molecular Biology. 2nd Ed. Neidhardt F, et al., editors. Washington, DC: ASM Press; 1996. pp. 822–848.
- Uptain S, Kane C, Chamberlin M. Annu Rev Biochem. 1997;66:117–172. - PubMed
- Ingham C, Dennis J, Furneaux P. Mol Microbiol. 1999;31:651–663. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases