The principal research goal of the Microbial Genetics Section is to understand molecular mechanisms that underlie the control of gene expression. An organisms's ability to adjust to environmental stress and to choose between alternative pathways of development requires that organism to control its patterns of gene expression. Transcription, the first step in gene expression, is catalyzed by RNA polymerase (RNAP), a large, multisubunit enzyme that is a target of many genetic pathways and whose core is structurally and functionally conserved in all kingdoms of life. Transcription can be divided into three stages: initiation, elongation, and termination. We concentrate our efforts on understanding the mechanisms of elongation and termination in the bacterium E. coli. These two steps, like initiation, are the targets of regulatory pathways but are less well understood than initiation. Moreover, the conservation of structure and enzymatic properties among multisubunit RNAPs argues that insights obtained from studies of the bacterial enzyme will be applicable to other organisms. We are intensively studying an elongation control mechanism found in phage HK022, a virus that parasitizes E. coli and uses the host RNAP to express its genes.

After RNAP initiates transcription, it continues to elongate the transcript until it reaches a terminator site. At this point, the enzyme, template, and transcript have a high probability of disassociating from each other. Termination sites insulate contiguous groups of genes from each other and thus allow independent control of the expression of each group. In addition, the efficiency of termination and, hence, the expression of genes downstream of terminators can be controlled. The molecular mechanisms of termination are incompletely understood. In some cases, the transcript of the terminator site recruits a termination factor to the adjacent RNA polymerase while, in other cases, the transcript of the terminator site has a direct role in stopping elongation, with no additional proteins required. To understand the mechanism and control of termination, we study "antiterminators," elements that reduce the efficiency of termination.

We showed earlier that the phage HK022 chromosome contains antiterminator sites that differ from any previously described. The HK022 sites are called put, for polymerase utilization. A nascent Put transcript associates with the RNAP molecule that synthesized it and converts the enzyme to a terminator-resistant form. No other protein factor is required for this conversion, and no other RNAP molecule in the cell is affected. Put RNA suppresses both factor-dependent and intrinsic terminators and has no apparent terminator specificity.

Antitermination is critical for the phage because it increases the expression of genes that are required for autonomous virus growth. The elongation control pathway appears simpler than others found to date, and its simplicity makes it an attractive target for deeper analysis. We asked what is special about the antiterminator RNA and how it changes the properties of RNAP.

FIGURE 34

Computer simulations, enzymatic probing, and analysis of mutations in the antiterminator site have provided clues about the details of the RNA structure (Figure 34). Our evidence suggests that the newly synthesized transcript containing the antiterminator sequences folds into a structure whose details are important for its association with and modification of RNAP. Characterization of RNAP mutants that affect antitermination have provided insights into how the modification occurs. The mutations alter a very highly conserved domain of the protein, the Zn-finger motif, of the b' subunit. To study the interaction between the antiterminator RNA and RNA polymerase, we stalled the elongation complex downstream of put (Figure 35) and determined the sensitivity of the transcript to ribonuclease cleavage. Part of put RNA was protected from cleavage by wild-type polymerase, but not by a Zn-finger motif mutant with a defect in put-dependent antitermination. We also exposed the stalled complex to oligonucleotides complementary to put RNA, restarted transcription, and measured antitermination. Some but not all complementary oligonucleotides inhibited antitermination. Finally, cleavage of the RNA between put and the 3'-end released put RNA from the stalled complex and prevented antitermination. We conclude that nascent put RNA binds specifically to RNAP and must remain bound to suppress termination. The zinc-finger motif mutant prevents antitermination by preventing binding.

FIGURE 35

The Zn-finger motif contains two pairs of invariant cysteines flanking a slightly less well-conserved segment of 13 amino acids, which is rich in basic residues. To investigate the role of this region in enzyme function, we replaced each of the charged residues with alanine and determined the effects of the substitutions on cell phenotype and enzymatic activity. All the mutated genes complemented a temperature-sensitive RNAP b' mutant for cell growth at a nonpermissive temperature, and the tested mutant enzymes transcribed and terminated normally in vitro. Therefore, the mutant enzymes are competent to perform essential cellular functions. However, none of the mutants antiterminated transcription with normal efficiency in response to put sites. We found that the severity of the antitermination defect depended on the sequence of the upstream stem-loop of put RNA. Some but not all Zn-finger mutants could distinguish between put variants that differed in this region. Such mutant specificity suggests that put RNA interacts directly and specifically with the Zn-finger motif to effect antitermination. Our conclusion is consistent with the finding that one of the Zn-finger mutants fails to bind nascent put RNA during transcription in vitro. Substitution or deletion of the invariant cysteine residues differs from substitution of the charged residues in that the cysteine mutants are unable to complement a temperature-sensitive b' mutant for cell growth. The RNAP produced by the mutants is active in transcription but terminates less efficiently than the wild-type enzyme on templates that lack a put site. In addition, the mutant enzymes do not respond to put sites. It therefore appears that the Zn-finger motif has a role in termination as well as in binding put RNA. These two properties could well be related: binding of put RNA by wild-type polymerase could alter the structure of the Zn-finger motif in the same way as mutations of the invariant cysteines.