Activation of Bacteriophage Mu mom Transcription by C Protein Does Not Require Specific Interaction with the Carboxyl-Terminal Region of the α or ς⁷⁰ Subunit of Escherichia coli RNA Polymerase

Abstract

Late in its growth cycle, transcription of the phage Mu mom promoter (P_mom) is activated by the phage gene product, C, a site-specific DNA binding protein. In vitro transcription analyses showed that this activation does not require specific contacts between C and the carboxyl-terminal region of the α or ς⁷⁰ subunit of Escherichia coli RNA polymerase. Unexpectedly, these results are in contrast to those known for another Mu-encoded transcriptional activator, Mor, which has a high degree of sequence identity with C and appears to interact with the carboxyl termini of both α and ς⁷⁰.

In addition to the host Escherichia coli RNA polymerase (RNAP-ς⁷⁰) (22), transcription of the four Mu late operons (including mom) requires the phage-encoded C protein (14, 16, 21, 24, 25, 34, 38, 39). In vitro studies have shown that purified C alone is capable of activating transcription of the mom promoter, P_mom (8, 12). It is not surprising that an accessory protein is required for transcription because the mom promoter sequence has a poor match to the consensus E. coli promoter hexamers, TTGACA at −35 and TATAAT at −10, and the spacing between them is a suboptimal 19 bp (13, 15, 32). Although the Mu site has identities of 3 of 6 and 4 of 6 with the consensus −35 and −10 sequences, respectively, the three matches in the −35 hexamer are not at so-called invariant positions (28); this has led to the suggestion that the mom promoter lacks a recognizable −35 sequence (5, 23).

C is a site-specific DNA-binding protein (5, 27) that binds 5′ of and adjacent to a poor −35 site in P_mom and P_lys (5, 23). A bipartite consensus C-recognition site containing partial dyad symmetry, TTAT_X_5,6_ATAACC, has been demonstrated (37). The location of the C-binding site in P_mom, upstream of and overlapping a poor −35 hexamer, is analogous to that of several other promoters that are dependent on accessory proteins for activation, such as the O_R2 binding site for λ cI at the λ P_RM promoter (29) and the binding sites for AraC (20), OmpR (10), and MalT (30). A computer-assisted study of 107 E. coli RNAP-ς⁷⁰ promoters revealed that 47 of 48 activatable promoters have protein binding sites in the region from −65 to +20, which overlap the RNAP binding site (9), and 30 of those sites contact the −40 position, as does C. It has been suggested that the activator proteins for promoters with poor −35 hexamers functionally substitute for the −35 element in contacting RNAP (9). Busby and Ebright (7) have noted that many (but not all) activators that lack an interaction with the COOH-terminal domain (CTD) of the RNAP α subunit bind to sites overlapping the −35 element. However, it has been shown that the phage Mu Mor protein, which bears a striking homology to C and binds its target promoter from −33 to −56, requires contacts with both the α CTD and the ς⁷⁰ CTD (ς CTD) (2). In this report, we present evidence that C activation of P_mom transcription does not involve such an interaction(s).

(This work was submitted in partial fulfillment of the requirements for a Ph.D. by W.S.)

Activation of in vitro transcription from the mom promoter region: lack of interaction between C and the CTD of the RNAP α or ς⁷⁰ subunit.

The sequence of the mom promoter region is shown in Fig. 1. The C protein binds to P_mom at a site that partially overlaps the RNAP site around the −35 region. To investigate possible interaction between C and RNAP, we analyzed C transcriptional activation of enzyme reconstituted (11) with a truncated α CTD or ς CTD. These proteins were purified and characterized previously (11), and their identity was confirmed by sodium dodecyl sulfate-polyacrylamide gel analysis following shipment to the laboratory of S.H. (data not shown). For the in vitro studies, we used a single-round runoff transcription assay (2) and included DNA containing the modified P_RE^# promoter of phage λ as an internal control; P_RE^# is a so-called “extended −10” promoter (6, 19) whose −35 region bears little resemblance to the canonical −35 hexamer. However, due to interaction of its extended −10 site with region 2.5 of the RNAP-ς⁷⁰ subunit, the P_RE^# promoter can be transcribed by RNAP containing a ς CTD or α CTD truncation (4, 6, 19). In addition, the P_RE^# promoter lacks an UP element, an AT-rich sequence located just upstream of the −35 hexamer in some promoters, which interacts with the α CTD (33). Preliminary control experiments with the various RNAP forms alone were carried out to determine the amounts necessary to produce P_RE^# transcripts at levels comparable with those transcribed by the wild-type (wt) RNAP (data not shown). Thus, the P_RE^# promoter served as a reference for normalizing transcriptional activity at the P_mom promoter. It should be noted that in the absence of C, wt RNAP produced some leftward transcripts, but these were no longer observed when C was present and activated rightward transcription (36a).

FIG. 1.

Top strand sequence of the mom promoter. The −10 and −35 hexamers are underlined, and the DNase I footprints of C and RNAP are indicated by horizontal bars; the upstream boundary protected by RNAP is deduced from the P1 site in the tin7 mutant mom promoter (3, 36).

The results of in vitro transcription assays with both C and reconstituted forms of RNAP are shown in Fig. 2A. Note that in these experiments higher concentrations of the mutant RNAPs were necessary to produce P_RE^# transcripts at levels comparable with those transcribed by the wt RNAP (Fig. 2A; compare lanes 3 to 5 with lanes 6 to 8 and 9 to 12), suggesting lower efficiency of holoenzyme assembly with the COOH-terminally truncated α or ς⁷⁰ subunits, or intrinsically lower activity of the mutant enzymes. The autoradiographs were computer scanned, and the relative transcriptional activities of P_mom and P_RE^# are presented in Fig. 2B (with the densities of the wt RNAP bands set to 1). Under the experimental conditions, RNAP with an α CTD truncation from amino acid 235 still retained 62% activity, and 79 to 55% activities were seen for the ς CTD deletions (starting at residues 556 to 529, respectively). These results demonstrate clearly that activation of P_mom transcription does not require specific interaction between C and the COOH-terminal region of either the α or the ς⁷⁰ subunit of RNAP.

FIG. 2.

Gel analysis of single-round in vitro transcription by reconstituted wt RNAP and mutant RNAPs with COOH-terminally truncated α or ς⁷⁰ subunits. (A) A mix of 5-fmol P_RE^# and 3-fmol P_mom fragments (except for lanes 1 and 2, which contained P_RE^# and P_mom alone, respectively) was incubated with a saturating concentration of C (80 nM) (in a solution containing 25 mM Tris · HCl (pH 7.9), 50 mM KCl, 5 mM MgCl₂, 1 mM dithiothreitol, 3% glycerol, 25 μg of bovine serum albumin/ml, and 0.05% Nonidet P-40), and then wt or mutant RNAP was added. Transcription was initiated by the addition of radioactive [α-³²P]UTP (40 μM; 8 Ci/mmol), unlabeled ATP, GTP, and CTP (160 μM each), and heparin (200 μg/ml). After 15 min of incubation at 37°C, the reactions were terminated by addition of an equal volume (12 μl) of loading dye. Ten microliters of each sample was electrophoresed on 5% sequencing gels. Amounts of RNAP used in lanes 1 to 12 were 113, 113, 113, 75, 50, 211, 141, 94, 50, 480, 370, and 800 fmol, respectively. Marker runoff transcripts (lane M) were generated from the T7 promoter of pBluescript II SK(−) (Stratagene) and its derivative (constructed by replacing the HindIII/SalI fragment with the corresponding polylinker sequence from pBend2 [18]). The shorter P_RE^# transcript (marked with an asterisk) is presumably due to pausing or premature termination. (B) Relative transcriptional activities at P_mom and P_RE^# for wt and mutant RNAPs. The autoradiographs were computer scanned, and the densities of the wt RNAP bands were set to 1. Note that the value for Δα-235 RNAP was taken from panel A, lanes 5 and 8.

Our observations are in contrast to those for the phage Mu transcriptional activator, Mor, which appears to require interactions with the COOH termini of both the α and ς⁷⁰ subunits to activate transcription of the middle promoter (P_m) (2). This was quite surprising because Mor bears a high degree of homology to C, and both recognize imperfect hyphenated AT-rich inverted-repeat sequences (12, 26, 37). However, there are differences in how these proteins interact with their respective promoters and with RNAP in vitro. For instance, C binding to P_mom introduces a DNA deformation (bending and/or untwisting) and a new DNase I hypersensitive site (31, 36, 37), while Mor has no such effects on P_m (1, 17). Furthermore, addition of RNAP was shown to introduce an extension of DNase I protection 5′ to the Mor protection boundary in P_m (2), which was attributed to additional contacts made by the α CTD. In contrast, no such upstream extension of DNase I protection was observed with C and RNAP (12, 36). Thus, whether and how a transcriptional activator protein interacts with RNAP may be influenced by the location of the activator binding site and the sequence of the target promoter DNA, as well as the biochemical properties and structure of the activator protein itself. It will be interesting to examine whether there is any requirement for C-RNAP interaction in transcriptional activation of the other late Mu promoters, P_lys, P_P, and P_I. Although we cannot rule out the possibility that C interacts with some other region of α or ς⁷⁰, or with the β or β′ subunit of RNAP, we currently favor the notion that C activation of P_mom transcription is mediated through a deformation in DNA structure, which is known to be induced by C binding in this region (31, 36, 37). In this regard, transcriptional activation by the Hg(II)-dependent MerR factor appears to be mediated by unwinding of merOP DNA (reviewed in reference 35), although it is still not clear whether specific MerR-RNAP interaction is also required. Thus, it remains open whether transcriptional activation can be mediated purely through factor-induced alteration in DNA topology.