T7 Lysozyme Represses T7 RNA Polymerase Transcription by Destabilizing the Open Complex during Initiation

Bacteriophage T7 lysozyme binds to T7 RNA polymerase and inhibits transcription initiation and the transition from initiation to elongation. We have investigated each step of transcription initiation to determine where T7 lysozyme has the most effect. Stopped flow and equilibrium DNA binding studies indicate that T7 lysozyme does not inhibit the formation of the preinitiation open complex (open complex in the absence of initiating nucleotide). T7 lysozyme, however, does prevent the formation of a fully open initiation complex (open complex in the presence of the initiating nucleotide). This is consistent with the results that in the presence of T7 lysozyme the rate of G ladder RNA synthesis is about 5-fold slower and the GTP Kd is about 2-fold higher, but T7 lysozyme does not inhibit the initial rate of RNA synthesis with a premelted bulge-6 promoter (bubble from -4 to +2). Neither the RNA synthesis rate nor the extent of promoter opening is restored by increasing the initiating nucleotide concentration, indicating that T7 lysozyme represses transcription by interfering with the formation of a stable and a fully open initiation bubble or by altering the structure of the DNA in the initiation complex. As a consequence of the unstable initiation bubble and/or the inhibition of the conformational changes in the N-terminal domain of T7 RNAP, T7 lysozyme causes an increased production of abortive products from 2- to 5-mer that delays the transition from the initiation to the elongation phase.

. This is consistent with the results that in the presence of T7 lysozyme the rate of G-ladder RNA synthesis is about 5-fold slower and the GTP K d is about 2-fold higher, but T7 lysozyme does not inhibit the initial rate of RNA synthesis with a premelted bulge-6 promoter (bubble from -4 to +2). Neither the RNA synthesis rate nor the extent of promoter opening is restored by increasing the initiating nucleotide concentration indicating that T7 lysozyme represses transcription by interfering with the formation of a stable and a fully open initiation bubble or by altering the structure of the DNA in the initiation complex. As a consequence of the unstable initiation bubble and/or the inhibition of the conformational changes in the N-terminal domain of T7 RNAP, T7 lysozyme causes an increased production of abortive products from 2-mer to 5-mer that delays the transition from the initiation to the elongation phase.

INTRODUCTION
The regulation of transcription in bacteriophage T7 RNA polymerase (RNAP) as in all RNA polymerases is dependent on the efficiency of each step of transcription, such as binding promoter DNA, binding nucleotides, overcoming abortive synthesis, RNA synthesis, and termination. Each of these steps is therefore a potential target for regulation. T7 transcription regulation begins with the gradual entry of the phage DNA into a host Escherichia coli cell (1) followed by the ability of the T7 RNAP, without the aid of auxiliary factors, to recognize and transcribe the phage promoters (2;3). The phage then makes T7 lysozyme, which represses transcription by T7 RNA polymerase. Unlike the LacI or the LexA repressors, which sterically block promoters from their polymerases, T7 lysozyme binds the T7 RNAP, not the DNA, forming a tertiary complex with the polymerase and DNA (4)(5)(6)(7). T7 lysozyme binds to the T7 RNAP at a site distal to the polymerase active site as confirmed by both biochemical data and a crystal structure of the T7 RNAP-T7 lysozyme complex (8)(9)(10).
Previous studies have shown that T7 lysozyme inhibits transcription initiation and promoter clearance, but not elongation (6;7). However, the exact mechanism by which these processes are inhibited is not known. We have therefore investigated T7 lysozyme inhibition under pre-steady state conditions enabling us to more fully dissect the steps that are altered during transcription. Transcription initiation occurs with a minimum of three steps (Reaction 1). In the first step, RNAP (E) binds to the promoter DNA (D) to form a closed complex, EDc, which isomerizes to form the pre-initiation open complex, open promoter DNA or a promoter DNA with even a single mismatch in the initiation region. The destabilization or the alteration of the structure of the initiation complex by T7 lysozyme is responsible for the inhibition of initial RNA synthesis and could also be the cause for the formation of more abortive products, which in turn delays the transition from initiation to elongation.

Synthetic DNA and other Materials
The oligodeoxynucleotides (unmodified and 2-AP-modified) were synthesized by Integrated DNA Technologies (Coralville, IA) and supplied as desalted samples. As described previously (11), the oligodeoxynucleotides were further purified by polyacrylamide gel electrophoresis, electroelution, and ethanol precipitation. The 3'-dGTP was purchased from TriLink Biotechnologies (San Diego, CA).

Protein
T7 RNAP was over-expressed in E. coli BL21/pAR1219 (12). The enzyme was purified as described previously (13)(14)(15) with the exception that the CM-Sephadex separation step was eliminated. The purified enzyme was stored at -80 °C in 20 mM sodium phosphate, pH 7.7, 1 mM trisodium EDTA, 1mM dithiothreitol, 100 mM sodium chloride and 50% (v/v) glycerol. The enzyme concentration was calculated from its absorbance at 280 nm and molar extinction coefficient of 1.4x10 5 M -1 cm -1 (16). T7 lysozyme was purified according to a reported procedure (10 and comb). The gels were exposed to a phosphor screen, scanned on a Typhoon instrument (Molecular Dynamics), and the RNA products quantified using the ImageQuaNT program.

Stopped-Flow Kinetics
The Where F is the fluorescence intensity at time t; n is the number of exponential terms; A n and k obs,n are the amplitude and the observed rate constant of the nth term, respectively; and C is the fluorescence intensity at t=0. The observed rate constant (k obs ) was plotted as a function of GTP and the dependency was fit by non-linear regression analysis to the hyperbolic equation 2 (17) using SigmaPlot.
[GTP] K Where k conf is the rate of a conformational change upon GTP binding and K d is the equilibrium dissociation constant of GTP.

Promoter DNAs
Several synthetic promoter DNAs were used in these studies ( Table 1). The dsDNA is a fully duplex promoter that contains the T7 φ10 promoter consensus sequence from -21 to +19 and initiates RNA synthesis at +1 with the sequence GGG. The p-dsDNA and bulge-6 DNAs are considered mimics of an opened promoter. To monitor open complex formation and nucleotide binding the fluorescent adenine analog 2-AP was incorporated at position -4 in the template or at position +4 in the non-template strand relative to the transcription start site. Altered promoters with the initiation sequence (GAC) were used to distinguish binding of the +1 and +2 initiating nucleotides.

T7 lysozyme does not affect promoter binding or the rate of pre-initiation open complex formation
Previous studies have shown that T7 lysozyme inhibits both transcription initiation and the transition from initiation to elongation (7). T7 lysozyme does not reduce transcription by preventing the promoter DNA from binding to the T7 RNAP or by decreasing the affinity of the promoter for the RNAP (6;7;18). As this was determined using indirect method, we sought to verify the observation with a more direct assay. The kinetics of promoter binding and the formation of the pre-initiation open complex were measured in real time using a promoter DNA that was modified with the fluorescent adenine analog 2-AP (Table 1). During open complex formation, the -4 to +2/+3 region of the promoter is converted from a duplex to a single stranded region. If 2-AP is substituted for the adenines in the melted region, open complex formation is accompanied by an increase in 2-AP fluorescence, which can be easily monitored. Any adenine in the melting region may be substituted with 2-AP. The greatest increase in fluorescence upon binding to T7 RNAP is observed when the 2-AP is positioned at the -4 position of the template strand, t(-4) (19). This is because the adenine at t(-4) undergoes a large structural change becoming both unpaired and unstacked from its neighboring guanine (20), and it is the latter process that gives the large fluorescence increase(21).
We have verified that the insertion of 2-AP does not affect promoter binding or transcription (13;14).
The steps of DNA binding and the formation of the pre-initiation open complex were measured with the dsDNA promoter containing 2-AP at t(-4), as described previously (19). T7 RNAP in the presence or the absence of T7 lysozyme was mixed with the promoter DNA in a stopped-flow apparatus, and the increasing fluorescence intensity was monitored as a function of time. We used 12 µM of T7 lysozyme, which is above the K d of T7 lysozyme (7) GTP. The reactions were quenched after millisecond time intervals with the aid of a rapid chemical quenched flow apparatus and the RNA products quantitated. In the presence of GTP alone, we see the production of pppGpG and pppGpGpG RNA products and also some pppGpGpGpG at longer times (Figure 1a). The G-ladder production increased linearly with time up to 0.25 s and the slope provided the initial rate. The initial rate was plotted as a function of [GTP] and the results indicated that the rate at maximal [GTP] is reduced approximately 5-fold (from 7 s -1 to 1.5 s -1 ) when T7 lysozyme is present (Figure 1b). Due to the low signal, we were unable to get an accurate value of the GTP K d from this radiometric assay in the presence of T7 lysozyme. A 20-fold reduction in catalytic efficiency was calculated from the initial slope of rate versus [GTP] dependence (11.6 ± 2.5 s -1 mM -1 without T7 lysozyme and 0.56 ± 0.1 s -1 mM -1 in the presence of T7 lysozyme).
To determine the effect of T7 lysozyme on the K d of the initiating GTP, a more sensitive fluorescence assay was used. In this assay, GTP binding and a subsequent In the presence of T7 lysozyme, the pre-steady state rate of RNA synthesis is 3fold slower ( Table 2) that is evident from the decrease in the initial burst phase with T7 lysozyme shown in Figure 3b. The abortive products from 2-mer to 5-mer on the other hand are produced in greater amounts with T7 lysozyme (Figure 3c). Hence, the steady state rate of RNA synthesis is actually higher in the presence of T7 lysozyme ( Table 2).
The 19-mer runoff product is produced with a longer delay in the presence of T7 lysozyme (Figure 3d).
Similar measurement of the pre-steady state kinetics of RNA synthesis on a "premelted" DNA was carried out to determine whether T7 lysozyme inhibits the chemistry step. Bulge-6 is considered a mimic of a melted promoter as it contains 6 noncomplementary bases to the template strand in the initiation region (-4 to +2). This substrate was chosen over the p-dsDNA as the bulge-6 contains the non-template strand, which is capable of interacting with T7 RNAP. The initial rate of total RNA synthesis with the bulge-6 promoter is unaffected by the presence of T7 lysozyme (Table 2), as evident from the similar burst phase with and without T7 lysozyme in Figure 4a. The bulge-1 DNA shows a similar behavior as bulge-6 ( Figure 4d). This indicates that T7 lysozyme does not alter the chemical step of RNA synthesis as long as the DNA is premelted or easily melted. Even though T7 lysozyme had little effect on the RNA synthesis rate, the production of abortive products from 3-mer to 5-mer with the bulge-6 and 2-mer to 5-mer with the bulge-1 was increased in the presence of T7 lysozyme (Figure 4b and 4e). This results in a higher steady state rate with T7 lysozyme ( Table 2).

T7 lysozyme affects the formation of the open complex in the presence of the initiating nucleotide
Our results indicate that the decrease in the rate of initiation cannot be attributed to a defect in the formation of the pre-initiation complex or a defect in chemistry or solely due to an increase in the GTP K d . Increased abortive synthesis in the presence of T7 Since, the 2-mer RNA synthesis rate was not affected by T7 lysozyme with an already or readily melted promoter, one would predict that T7 lysozyme would not affect the formation of the initiation complex with these promoters. This is indeed the case as shown in Figure 5c. The fluorescence of bulge-6 and bulge-1 in complex with T7 RNAP is similar to that of the p-dsDNA. Upon addition of GTP, the dsDNA complex showed the characteristic increase in fluorescence from 30% to 100% but no change was observed in the bulge DNAs. In the presence of T7 lysozyme (10 µM), a slight decrease in the fluorescence of bulge DNA pre-initiation complexes was observed, but this was overcome by the addition of nucleotide. Thus, the initiation complex of bulge-6, bulge-1 or p-dsDNA was unaffected by the presence of T7 lysozyme. The fluorescence of EL·dsDNA with GTP was greater than with 3′-dGTP (65% versus 40%) most likely due to the stabilizing effects of the newly synthesized RNA.

DISCUSSION
The experiments presented here were carried out to investigate the mechanism by which T7 lysozyme inhibits T7 RNAP transcription. It is already known that unlike many other transcriptional repressors, which sterically block promoters from their polymerases, T7 lysozyme does not bind DNA. Rather it directly binds to T7 RNAP to form a tertiary complex with the polymerase and DNA (7). More specifically, a crystal structure of the T7 lysozyme-T7 RNAP complex revealed that T7 lysozyme binds to a site distal to the polymerase active site and causes little change in the overall T7 RNAP structure with the exception of the extreme C-terminus. These last four residues (FAFA 883 ) are disordered in the complexed structure, where as in the T7 RNAP-DNA complex structure they are located below the polymerase active site (8). Furthermore, biochemical data revealed that mutations in any of these four C-terminal residues results in decreased T7 RNAP activity (24) and that the C-terminus is more sensitive to proteolysis in the presence of T7 lysozyme (25). It has been proposed based on this information and steady state transcription assays that showed an increase in the NTP apparent K m during initiation, that T7 lysozyme inhibits T7 RNAP by stabilizing the conformation of the T7 RNAP with an altered C-terminus. This model predicts that inhibition can be overcome by the addition of nucleotide (18;25). We have further investigated this model for T7 lysozyme inhibition of T7 RNAP transcription initiation using pre-steady state techniques.
During T7 RNAP transcription initiation there are several potential points for regulation by T7 lysozyme. The first potential regulatory site is promoter binding.
Stopped-flow experiments indicated that T7 lysozyme does not prevent T7 RNAP from binding the promoter. Similarly, the pre-initiation complex formation rate was affected only to a small extent in the presence of T7 lysozyme. We next explored the possibility that T7 lysozyme altered the binding of the initiating nucleotide to the T7 RNAPpromoter pre-initiation complex. A quenched-flow radiometric assay measuring the rate of RNA synthesis during initiation at increasing [GTP] showed that the rate of G-ladder synthesis decreased five-fold in the presence of T7 lysozyme. The rate could not be restored even by high concentrations of nucleotide (2 mM GTP). Since obtaining a reliable GTP K d value was impossible due to the difficulty in measuring the products synthesized in the presence of T7 lyszoyme, a fluorescent assay monitoring an increase in 2-AP fluorescence upon GTP binding was employed. The stopped-flow assay also indicated that the maximal rate of a conformational change occurring upon GTP binding (k conf ) was reduced approximately three-fold while the average GTP K d was increased at most two-fold in the presence of T7 lysozyme. These results are only partly in agreement with the proposed mechanism of Villemain and Sousa (18)  Nonetheless, due to the increased abortive synthesis the RNAP spends more time in the recycling mode, which causes a delay in the transition from initiation to elongation. The fact that abortive products from 3-mer to 5-mer increases with T7 lysozyme also indicates that the delay in the transition from the initiation to the elongation phase could be caused by the inhibition of some or all the conformational changes of the N-terminal domain (29;30) that are necessary to make RNA products longer than 3-mer. T7    Table 2. The pre-steady state and steady state rates are listed in Table 2.