Mechanisms for Controlling Gene Expression in Bacteria and Bacteriophage

Robert Weisberg, PhD, Head, Section on Microbial Genetics
Natalia Komissarova, PhD, Staff Scientist
Erik Read, PhD, Postdoctoral Fellow
Tatyana Velikodvorskaya, PhD, Postdoctoral Fellow
Sieghild Sloan, MS, Microbiologist

Temperate bacteriophages can establish lysogeny, a long-term and mutually beneficial association between phage and bacterial host. Lysogeny is a major mechanism of horizontal gene transfer between bacterial families, and temperate bacteriophages frequently contain genes that allow bacteria both to survive environmental challenges and to cause disease in mammalian hosts. To maintain lysogeny, temperate phages control the expression of their own genes so that host viability is unimpaired. Many phages control gene expression by regulating the elongation step of transcription, which is catalyzed by DNA-dependent RNA polymerase (RNAP). After initiating polymerization on a DNA template, the transcript elongates until RNAP reaches a terminator, where the enzyme, template, and transcript (the elongation complex or EC) dissociate from each other. Anti-terminators can control the efficiency of termination and hence the expression of downstream genes. The molecular mechanisms of termination and anti-termination are poorly understood. We aim to understand these mechanisms by studying transcription directed by an Escherichia coli phage, HK022, which uses a robust yet relatively simple anti-termination pathway to control gene expression. We recently expanded our studies to transcriptional regulatory mechanisms in a different temperate phage, B40-8, which parasitizes Bacteroides fragilis, a very distantly related bacterium.

Inhibition of a transcriptional pause by RNA anchoring to RNA polymerase

Komissarova, Velikodvorskaya, Sen, King

HK022 uses a novel, RNA-based anti-termination mechanism to express many essential genes. A nascent transcript of a viral sequence called put binds to the enzyme that catalyzed put’s synthesis and remains associated with the EC as the transcript continues to elongate. put RNA modifies the EC to which it is bound so that transcription no longer stops at terminators. No other protein factor is absolutely required, other ECs in the same cell are unaffected, and no obvious terminator specificity is involved. The apparent simplicity of this anti-termination pathway makes it an attractive target for deeper analysis. Modification of the EC by put RNA also suppresses transcriptional pausing at a U-rich sequence located very close to the put site. EC s have a high probability of pausing for variable amounts of time at particular template positions, and such pauses can have important physiological functions. Moreover, pausing is thought to precede termination, and intrinsic terminators contain U tracts at the termination point. Therefore, we decided to investigate the mechanism of put-mediated anti-pausing at this site.

We showed that the U-rich sequence promotes “backtracking” of the EC. Backtracking is a retrograde movement during which the nucleotides at the 3′ end of the transcript are melted from the template DNA strand and extruded from RNAP while compensating amounts of upstream DNA and nascent RNA re-enter the EC. The length of the RNA:DNA hybrid remains constant. Backtracking is favored by the replacement of thermodynamically weak with thermodynamically strong RNA:DNA hybrid (e.g., rU:dA with rC:dG). In vitro transcription and footprinting assays revealed that put RNA suppresses the U-rich pause by restricting re-entry of the nascent transcript into the RNA exit channel, thus limiting backtracking. The restriction is local and relaxes as the transcript elongates. The number of nucleotides over which the restriction operates allows us to use RNA as a “molecular tape measure” to locate the put binding site. Our results indicate that put RNA binds to the surface of polymerase within 10 to 28 Å of the end of the RNA exit channel, a region that includes amino acid residues known to be important for anti-termination and RNA binding. Even though binding is essential for anti-pausing and anti-termination, these two activities of put differ; anti-pausing is limited to the immediate vicinity of the put site, but anti-termination is not. We know that put RNA remains stably bound as the EC translocates and presumably ensures the persistence of anti-termination. While the mechanism of anti-termination remains unknown, our current work argues against the possibility that put acts by strengthening thermodynamically weak RNA:DNA hybrid.

Transcriptional pausing is an important component of many regulatory networks, and our knowledge of the networks is growing. For example, in a large number of Drosophila embryo genes, ECs stall near the transcription start site, possibly waiting for activation at later developmental stages. Backtracking is a mechanism that accounts for many known pauses, and RNA anchoring to RNAP is potentially a general mechanism that regulates backtracking-associated pauses. Moreover, given that the effect of anchoring is local, individual pauses can be targeted. Examination of the properties of several well-studied pause sites leads us to propose that RNA anchoring to RNAP is a widespread mechanism of pause regulation.

Komissarova N, Velikodvorskaya T, Sen R, King RA, Banik-Maiti S, Weisberg RA. Inhibition of a transcriptional pause by RNA anchoring to RNA polymerase. Mol Cell 2008;31:683-694.
Sloan S, Rutkai E, King RA, Velikodvorskaya T, Weisberg RA. Protection of antiterminator RNA by the transcript elongation complex. Mol Microbiol 2007;63:1197-1208.

Analysis of a bacteriophage that parasitizes a commensal anaerobe

Read

We have sequenced and annotated the genome of bacteriophage B40-8, whose host is Bacteroides fragilis, a human commensal anaerobe that is frequently pathogenic. B40-8, a member of the Siphoviridae, has a circular double-stranded DNA genome of 45,805 basepairs. The genome encodes 64 real or potential proteins, of which only 18 can be assigned functions based on homology, structural predictions, or N-terminal sequencing of capsid proteins. The predicted open reading frames are oriented in the same direction and generally conform to the temporal cassette organization of terminase-head-tail-lysis, except that the major tail gene is located after the terminase and before the head genes. The distribution of numerous predicted promoters, terminators, and conserved sequence motifs within the intergenic regions suggests a complex regulatory network of several overlapping transcripts similar to that of Streptomyces phage phiC31. A paucity of discernible Shine-Dalgarno sequences and the presence of a conserved YA_3-7Y element in their place suggest a different mechanism of translational initiation than that usually found in prokaryotes. The presence of submolar restriction fragments of the virus chromosome, alignment of other terminase largesubunit coding sequences with CDS2, and other evidence strongly support a processive headful DNA-packaging mechanism. Sequencing of the chromosomal ends places the first cut at basepair 12013 and the second cut between bases 17491 and 17753, generating a first headful of approximately 51,400 base pairs and 10.9 percent terminal redundancy. Thus, B40-8 packaging is distinct from that of other well-studied phages; the first cut site is not within but rather almost 10 kb downstream of the putative terminase genes. PCR analysis of purified phage demonstrates that host DNA is also packaged into capsids, and sequencing of genomic clones suggests that about 1 percent of the phage particles contain host DNA.

Action of a sequence-specific transcript termination factor

Sloan, Gottesman

The Nun protein of phage HK022 protects HK022 lysogens against superinfection by phage lambda. Nun binds to and terminates EC s that transcribe lambda DNA , and nascent transcripts of the lambda nutL and nutR sites promote binding. Nun recognizes these transcripts with high specificity, effectively limiting termination to the lambda transcripts. In vitro, Nun arrests transcript elongation but does not by itself dissociate the arrested ECs. However, in vivo, Nun dissociates ECs that have transcribed the nutR site, presumably with the help of cellular termination factors such as Mfd and/or Rho, which are known to act on arrested ECs.

Our laboratory is identifying the requirements for Nun-dependent dissociation of the EC in vivo by using Northern blots to measure the abundance and stability of Nun-terminated transcripts containing the nutL site and initiated at the lambda P_L promoter. We reasoned that dissociation of the EC should allow cellular 3′–5′ exonucleases, the major catalysts of mRNA degradation in E. coli, access to the 3′ end of the terminated transcript. By contrast, simple arrest leads to a transcript with a protected 3′ end (it is inside the EC), likely increasing transcript stability. In addition, the arrested EC will sterically block transcription by ECs that follow, which should back up on the template and eventually occlude the promoter (“constipation”). Therefore, the factors required for dissociation of arrested ECs are likely to reduce the stability of Nun-terminated transcripts while increasing their rate of synthesis. We are currently determining the effects of inactivating Mfd, Rho, and cellular 3′–5′ exonucleases on the abundance and stability of such transcripts. Given that other work has suggested that dissociation is favored with increased distance, we will determine the effect of changing the distance between the promoter and the nutL site.

Our initial experiments show that, in cells with the wild-type P_L–nutL distance (32 bp), Nun increases the half-life of P_L mRNA more than 5-fold while decreasing its abundance about 2-fold. As expected, transcripts produced in the presence of Nun have 3′ ends that are dispersed over about 100 nt downstream of nutL while transcripts produced in its absence are considerably longer. This result argues that undissociated, Nun-arrested EC s protect at least some transcripts from degradation and that “constipation” caused by Nun-mediated arrest reduces promoter activity. Inactivation of Mfd increases transcript stability about another 5-fold in the presence of Nun, but has no effect in its absence. Therefore, Mfd appears to promote dissociation of a fraction of the arrested ECs. We suggest that Mfd dissociates only those ECs that have lost the initiation factor sigma when arrest occurs. Work from other laboratories suggests that sigma release occurs stochastically after an EC leaves the promoter and that sigma-containing ECs are resistant to the action of Mfd.

Weisberg RA, Hinton DM, Adhya S. Transcriptional regulation in bacteriophage. In: Mahy BWJ, Van Regenmortel MHV, eds. Encyclopedia of Virology. Elsevier, 2008;174-186.

Collaborators

Max Gottesman, MD, PhD, Institute of Cancer Research, Columbia University, New York, NY
Rodney King, PhD, Western Kentucky University, Bowling Green, KY
Ranjan Sen, PhD, Center for DNA Fingerprinting and Diagnostics, Hyderabad, India

For further information, contact weisberr@mail.nih.gov or visit http://smg.nichd.nih.gov.