Interactions of a Proteolytically Nicked RNA Polymerase of Bacteriophage T7 with Its Promoter*

The association of nicked RNA polymerase of bacteriophage T7 (Ikeda, R. A., and Richardson, C. C. (1987) J. Biol. Chem. 262, 3790-3799) with the T7 (610 promoter has been examined by DNA cleavage protection. The (610 promoter consists of a 23-base pair consensus sequence that extends from -17 to +6 with respect to the site of the initiation of transcription (+ 1). Nicked T7 RNA polymerase alone protects 20 bases from -21 to -2 (fl) base at each border. Initiation and synthesis of the trinucleotide r(GGG) expands and shifts the sequence protected by nicked T7 RNA polymerase. Twenty-five bases are protected from -17 to +8 (21). The polymerization of three additional ribonucle-otides, synthesis of the hexamer r(GGGAGA), further expands the protected sequence. Twenty-seven bases are protected from -17-+10 (2 1). Finally, the thesis of a pentadecaribonucleotide r(GGG-AGACCACGG), 22 bases from -2-+20 (2 1). In to sequences protected by T7 RNA zyme end are in small fragment of T7 in se-

protected by the T7 RNA polymerase change (Fig. 1). These changes in the sequences protected by the T7 RNA polymerase are reminiscent of the changes observed in the sequences protected by Escherichia coli RNA polymerase during the promoter-dependent initiation of transcription (Carpousis and Gralla, 1985;Hofer et al., 1985;Straney and Crothers, 1985). Consequently, a preliminary model for the initiation of transcription has been adapted from the sequential, multistep mechanism proposed for the initiation of E. coli RNA polymerase (Buc and McClure, 1985).
Free promoter (P)' and free RNA polymerase (R) associate and form a promoter-specific closed complex (RP,). The strands of the promoter are separated (RP,), and RNA synthesis is initiated (RP,). After polymerization of a number of ribonucleotides, the RPi isomerizes to a transcriptionally competent complex (RP,).
R+P+RP,+RP,@RP,-RP, It might be expected that the physical character of the T7 RNA polymerase-promoter complexes would mirror the enzymatic processes that must occur during initiation and elongation. In the preceding paper (Ikeda and Richardson, 1987) we have described the isolation of a T7 RNA polymerase that is proteolytically cleaved between amino acids 172 (lysine) and 173 (arginine) (Tabor and Richardson, 1985) and have shown that this nicked T7 RNA Polymerase is inefficient at initiating transcription from a T7 promoter and is likely to terminate transcription prematurely. To determine whether the changes observed in the enzymatic activities of nicked T7 RNA polymerase are reflected in the interaction of nicked RNA polymerase-promoter complexes, we have examined the binding of the nicked enzyme to the $10 promoter by DNA footprinting in the presence of MPE.Fe(II) (Van Dyke et al., 1982;Van Dyke and Dervan, 1983). We find that nicked T7 RNA polymerase can be readily distinguished from T7 RNA polymerase by the patterns of sequence protection exhibited during the initiation of transcription and by the stability of the polymerase-promoter complexes.  Schematic illustration of the sequences protected by T7 RNA polymerase. The 23-bp consensus sequence of the T7 610 promoter is boxed, and the numbering scheme begins (+I) at the first nucleotide in the RNA transcript. The brackets outline the sequences protected by T7 RNA polymerase ( k l base at each border) in the presence of no nucleoside 5"triphosphates (RP,), GTP only (RPil), GTP and ATP (RPi,), and GTP, ATP, and CTP (RP,). The lengths of the protected sequences are given a t the right. This figure has been reproduced from Ikeda and Richardson (1986) with permission from the National Academy of Science, U. S. A., for clarification of the "Introduction."

3' FOO T W I N TS
the T7 410 promoter have been visualized by the protection of specific sequences from cleavage by methidiumpropyl-EDTA. Fe(I1) (MPE. Fe(I1)). A 312-bp EcoRVINurI restriction fragment containing the T 7 RNA polymerase promoter 410 was uniquely labeled a t either the 5' terminus or the 3' terminus of the NurI restriction site (Maxam and Gilbert, 1980;Maniatis et ul., 1982). The labeled restriction fragment was equilibrated with nicked T7 RNA polymerase. MPE. Fe(I1) (Hertzberg and Dervan, 1984) was added to the reaction, and nicking of the DNA was initiated with dithiothreitol. The resulting fragments were separated on an 8% polyacrylamide denaturing gel, and the gel was visualized by autoradiography. The sequences that are protected by nicked T7 RNA polymerase appear on the autoradiographs as a series of bands of reduced intensity. Nicked T 7 RNA polymerase like intact T7 RNA polymerase interacts weakly with the strongT7 410 promoter. In standard transcription buffer (Chamberlin et ul., 1970), nicked T 7 RNA polymerase does not protect the $10 promoter in the absence of RNA synthesis. The specific complex of nicked T 7 RNA polymerase bound to the 410 promoter can only be observed under conditions that favor electrostatic interactions, and although these conditions tend to favor nonspecific interactions, the accompanying increase in the stability of the closed complex makes protection of the binding site detectable. Specific protection of the 410 promoter by the nicked enzyme is apparent after the concentration of Mg2+ present in the reactions is reduced from 20 to 2 mM (Figs. 2 and 3). Nicked T7 RNA polymerase protects 19 bases on the 3' labeled (antisense) strand and 20 bases on the 5' labeled (sense) strand defining a sequence from -21 to -2 ( f l ) base a t each border, and the efficiency of the protection increases with increasing amounts of added RNA polymerase.
The initiation of RNA synthesis stabilizes the promoterspecific binding of nicked T 7 RNA polymerase. In the presence of GTP, ATP, and CTP both nicked T7 RNA polymerase and T7 RNA polymerase efficiently bind linear templates that contain a single 410 promoter. These transcriptionally stabilized complexes are retained (data not shown) by nitrocellulose filters (Matson and Richardson, 1985). Analogous experiments by Smeekens and Romano (1986) showed that complexes of T7 RNA polymerase and linear T 7 DNA templates are not retained by nitrocellulose when the complexes are formed in the absence of nucleoside 5"triphosphates.
These observations parallel the protection of the $10 promoter. In standard transcription buffer containing 20 mM MgC12, nicked T 7 RNA polymerase protects the 410 promoter only after the initiation of transcription (Fig. 4). In the absence of initiation little protection of the promoter is apparent.
Two different initiation complexes and a transcription complex can be observed by limiting the nucleoside 5"triphos-

3802
Interactions of Nicked Ti' RNA Polymerase with Its Promoter FIG. 3. Schematic illustration of the sequences protected by nicked T7 RNA polymerase. The 23-bp consensus sequence of the T7 $10 promoter is boxed, and the numbering scheme begins (+1) at the first nucleotide in the RNA transcript. The brackets outline the sequences protected by the nicked T 7 RNA polymerase (el base a t each border) in the presence of no nucleoside 5'triphosphates (RP,), G T P only (RPi,), GTP and ATP (RPiz), and GTP, ATP, and C T P (RP,). The height of each bracket schematically gives the relative efficiency of protection at that point in the sequence. The heights of the brackets in Fig. 1 also give the relative efficiency of protection along each protected sequence, but with T7 RNA polymerase each of the complexes uniformly protects the binding site. The lengths of the protected sequences are given at the right.  show the protection of the 3' strand and the 5' strand of the T7 $10 promoter by nicked T 7 RNA polymerase during the initiation of transcription. The brackets outline the protected sequences that have been identified by densitometry, and the labels G, GA, and GAC indicate the ribonucleoside 5'-triphosphates that were present in the reaction. The numbering scheme is as described in the legend to Fig. 2. The samples in lanes 1-5 all contained 50 mM Tris. HC1, p H 8, 20 mM MgCI2, 0.5 pg of tRNA, and 30 ng of the 3' ( A ) or 5' ( B ) labeled 312-bp EcoRVINarI restriction fragment. The samples in lanes 1 contained 400 p~ GTP and no nicked T 7 RNA polymerase; the samples in lanes 2 contained 400 p~ GTP and 13.1 pg of nicked T 7 RNA polymerase; the samples in lanes 3 contained 400 p~ GTP, 400 p~ ATP, and 13.1 pg of nicked T 7 RNA polymerase; the samples in lanes 4 contained 400 p~ GTP, 400 p~ ATP, 400 p~ CTP, and 13.1 pg of nicked T 7 RNA polymerase, and the samples in lanes 5 contained no nucleoside 5"triphosphates and 13.1 pg of nicked T 7 RNA polymerase. The lanes marked G and C>T are Maxam-Gilbert sequencing samples of the labeled restriction fragments.

~* P ---! ' ---C T T C T G A~A G A C T T C G A~A T~A A T A C G~C T C A C T A T A~G G A G A~C A~A A C G G T T T C~C T C T A G A C G~A T C
phates available for incorporation into RNA. As deduced from the sequence of the DNA fragment that contains the 410 promoter, the first three nucleotides of the RNA transcript are all guanylates. In the presence of GTP, the trinucleotide r(GGG) is synthesized. With both GTP and ATP present, the hexanucleotide r(GGGAGA) is synthesized, and with GTP, ATP, and CTP present the pentadecanucleotide r(GGGAG-ACCACAACGG) is synthesized. Each of the three initiated complexes protects the 410 promoter from cleavage by MPE. Fe(II), and each complex protects different sequences within and surrounding the consensus sequence of the 410 promoter.
In the presence of GTP, the r(GGG)-initiated complex protects 25 bases on the 3' labeled (antisense) strand and 24 bases on the 5' labeled (sense) strand, a sequence from -17 to +8 ( f l ) base at each border. In the presence of GTP and ATP (hexanucleotide synthesis) the 5' end of the sequence protected by nicked T 7 RNA polymerase is identical to the 5' end of the sequence protected by the r(GGG)-initiated complex. At the 3' end of the protected sequence, the r(GGGAGA)-initiated complex extends 2 bases farther downstream. Twenty-five bases are protected on the antisense strand, and 27 bases are protected on the sense strand covering a sequence from -17 to +10 (fl). Finally, in the presence of GTP, ATP, and CTP (pentadecanucleotide synthesis) the protected sequence is translocated downstream. Nineteen bases are protected on the antisense strand, and 24 bases are protected on the sense strand, a sequence from -2 to +20 ( f l ) (Figs. 3 and 4).
The closed complex observed in the presence of no nucleoside 5"triphosphates and the transcription complex observed in the presence of GTP, ATP, and CTP (pentadecanucleotide synthesis) protect their binding sites uniformly. In contrast, the two initiated complexes formed by trinucleotide synthesis (GTP, r(GGG), and hexanucleotide synthesis (GTP and ATP, r(GGGAGA)) efficiently protect only 5 bases at the 5' end of the protected sequences of both complexes and 14 bases at the 3' end of the r(GGG)-initiated complex and 16 bases at the 3' end of the r(GGGAGA)-initiated complex. The protected sequences are broken into two blocks by the short sequence of relatively unprotected bases (Figs. 4 and 5). Five bases on the 3' labeled strand and 6 bases on the 5' labeled strand from approximately -12 to -7 are inefficiently protected by both complexes. With T7 RNA polymerase the binding sites are uniformly protected by all four complexes (Fig. 1). The sequences protected by the two initiation complexes are not broken into two segments as is observed with the nicked T7 RNA polymerase (Fig. 5 ) .
Protection of the $10 Promoter Is Proportional to the Concentration of RNA Polymerase-The efficiency of protection of the $10 promoter increases with rising concentrations of T7 RNA polymerase and nicked T7 RNA polymerase. The ratio of the percentage of bound promoter to the percentage of free promoter can be determined from densitometer tracings of the autoradiographs of the denaturing polyacrylamide gels that were used to separate the fragmented DNA from the protection reactions (Hawley et al., 1985;Ikeda and Richardson, 1986). The ratio of bound promoter to free promoter is a direct measure of the efficiency of protection of the $10 promoter and is a function of the added RNA polymerase.
The ratio of bound promoter to free promoter was plotted against the concentration of added nicked T7 RNA polymerase. The data obtained for the r(GGG)-initiated complex formed in the presence of 50 mM Tris. HCl, pH 8.0, 20 mM MgC12, and 400 PM G T P yields an apparent Keq of 2 X lo5 M" for protection of the $10 promoter (Fig. 6). For both T7 RNA polymerase and nicked T 7 RNA polymerase, protection of the $10 promoter in reactions containing 20 mM MgC1, is not apparent in the absence of nucleoside 5"triphosphates; however, protection of $10 promoter is detectable in 50 mM The ratio of bound promoter to free promoter was determined as described under "Experimental Procedures" and is plotted against the concentration of added nicked T 7 RNA polymerase (POL). The protection of the T7 $10 promoter in the presence of GTP increases with increasing concentrations of nicked T 7 RNA polymerase. Assuming a linear function, the dashed lines represent the possible error for the fit of the bold line to the data. The apparent Keq determined from the slope of the bold line is 2 X lo5 M-'. The samples represented on the graph contained 50 mM Tris.HC1, pH 8.0, 20 mM MgCl,, 400 p~ GTP, 0.5 pg of tRNA, 20 ng of the 3' labeled 312-bp EcoRV/NarI restriction fragment, and 2. 62, 3.92, 5.23, 6.54, 7.85,9.16, 10.46, 11.77, or 14.39 pg of nicked T 7 RNA polymerase. The control lane contained no nicked T 7 RNA polymerase.
The protection of the $10 promoter by T7 RNA polymerase alone is not as simple as the protection of the $10 promoter by nicked T7 RNA polymerase alone. Nonspecific binding of T7 RNA polymerase interferes with the identification and characterization of the closed complex of T7 RNA polymerase. At low concentrations (<3 FM) of T7 RNA polymerase faint protection of the $10 promoter appears, but at higher  concentrations of T 7 RNA polymerase protection of the 410 promoter does not increase (Fig. 7). Protection of the 410 promoter by T 7 RNA polymerase alone is difficult to detect visually; however, densitometer scans of the autoradiograph of the polymerase titration reveal protection of the promoter. The protection appears to be proportional to the concentration of the added enzyme over the lowest concentrations of T7 RNA polymerase (Fig. 8).
Protection of the 410 Promoter Is Sensitive to the Concentration of M$+-The appearance of protection of the 610 promoter in reactions containing 2 mM M$+ and the observation of no protection of the 410 promoter in reactions containing 20 mM Mg2+ suggest that an electrostatic component contributes to the binding of the RNA polymerases to the 610 promoter. The binding of a positively charged molecule to a negatively charged polyelectrolyte has been theoretically described by Manning (1978). Both the polyelectrolyte theory and the application of the theory have been modified by Record and co-workers (de Haseth et al., 1977;Record et al., 1976Record et al., , 1978 to describe the effect of ions on the binding equilibria of proteins and nucleic acids. The theory predicts that the association of a positively charged protein with negatively charged DNA should be inversely related to the concentration of cations present in the solution. The ratio of bound promoter to free promoter was determined as a function of the concentration of Mg2+ for the protection of the 610 promoter by the r(GGG)-initiated complex of nicked T7 RNA polymerase (Figs. 9 and lo), the r(GGG)-initiated complex of nicked T 7 RNA polymerase (Figs. 9 and lo), and the closed complex of nicked T 7 RNA polymerase (Fig. 11). A log-log plot of the apparent Keq versus the concentration of Mg2+ yields the functions for the Mg2+ titrations of the three different complexes. The slopes of the functions are 1.20 f 0.5 for the simple complex of nicked T 7 RNA polymerase (Fig. l l ) , 2.2 f 0.6 for the r(GGG)-initiated complex of nicked T 7 RNA polymerase (Fig. lo), and 3.7 f 1.0 for the r(GGG)initiated complex of T7 RNA polymerase (Fig. 10). These slopes represent an estimate of the number of Mgz+ ions that are released during the formation of the polymerase-promoter complexes. It appears that the r(GGG)-initiated complex of nicked T7 RNA polymerase displaces fewer, Mg2+ ions than the r(GGG)-initiated complex of T 7 RNA polymerase. Unfortunately, the same data cannot be compared for the closed complexes of the two enzymes. Data for the closed complex of T 7 RNA polymerase could not be obtained. Nonspecific binding by the intact enzyme obscures the results. If the trend observed with the r(GGG)-initiated complexes applies to the closed complexes of T 7 RNA polymerase and nicked T 7 RNA polymerase one might expect that the closed complex of the intact enzyme would displace more Mg2+ ions than the closed complex of the nicked enzyme.
The equations describing the variation of Keg as a function of Mg2+ concentration can also be used to compare association constants obtained by different investigators under different Mg2+ concentrations. In a personal communication3 S. Gunderson and R. Burgess reported to us that they had obtained a binding constant of 10 f 1.7 x lo7 M" for the r(GGG)stabilized complex of T 7 RNA polymerase. In contrast, we have previously reported a binding constant of 5 X lo5 M" for the same complex. The two values are however determined under different conditions. Gunderson and Burgess3 obtained their data from reactions containing 10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 5 mM MgC12, and 400 p~ GTP, whereas our reactions typically contained 50 mM Tris HCl, pH 8.0, 20 mM MgC12, 400 p~ GTP, and 0.5 pg of tRNA. Extrapolation of our data to 5 mM Mg2+ via the equation log Keq = -3.7 log [Mg2+] -0.48 yields an estimated Keq of 1.0 X 10* f 0.2 X 10' M". This corresponds to the Keg obtained by Gunderson and Burgess.3 It suggests that the differences in the binding constants reported by the two groups are largely determined by the differences in Mg2+ concentrations and that nonspecific binding of T 7 RNA polymerase of tRNA must be negligible a t high Mg2+ concentrations (>lo mM Mg2+). Furthermore, it also suggests that small differences in pH, NaCl concentra-S. Gunderson and R. Burgess, personal communication.

Increasing Mg 2+
. tion, or Tris.HC1 concentration do not greatly affect the binding of T7 RNA polymerase.

DISCUSSION
As with T 7 RNA polymerase, the complexes observed during the initiation of transcription by nicked T7 RNA polymerase resemble the complexes observed during the initiation of transcription by E. coli RNA polymerase (Carpousis and Gralla, 1985;Hofer et al., 1985;Straney and Crothers, 1985). We again adopt the sequential, multistep mechanism proposed for the initiation of E. coli RNA polymerase (Buc and McClure, 1985) to provide a possible model for discussion of the steps in the initiation of transcription by nicked T7 RNA polymerase.
The sequences protected by nicked T 7 RNA polymerase as it changes conformation from RP, to RP, are similar to the sequences protected by the analogous T7 RNA polymerase complexes; however, four significant changes differentiate the protection of the 410 promoter by nicked T 7 RNA polymerase and T7 RNA polymerase. First, in all cases that are not complicated by nonspecific binding, a higher concentration of nicked T 7 RNA polymerase is necessary to protect the T 7 410 promoter efficiently. Second, the 5' border of the protected sequences translocates downstream much earlier with the nicked enzyme than with T7 RNA polymerase. Third, the sequences protected by the two observable initiation complexes and the single observable transcription complex of nicked T7 RNA polymerase are shorter than the corresponding sequences protected by T7 RNA polymerase (Figs. 1 and  5). Fourth, bases toward the middle of the protected sequences of the two initiated complexes of nicked T7 RNA polymerase are not as efficiently protected as the bases at the ends of the protected sequences.
It may be possible that the conformation of the r(GGG)and r(GGGAGA)-initiated complexes of nicked T7 RNA polymerase allow the DNA near the middle of the protected sequences to be exposed through the nick in the proteolytically cleaved enzyme. Cleavage of the exposed DNA by MPE.  12 mM (lanes 2), 14 (lanes 3), 16 (lanes 41, 18 (lanes 5)  Promoter recognition by T7 RNA polymerase has been attributed to amino acid residues between amino acid 220 and amino acid 530 and to amino acid residues in the carboxyl terminus of T7 RNA polymerase (Ryan and McConnell, 1982;Bailey et al., 1983;McGraw et al., 1986;King et al., 1986). In nicked T7 RNA polymerase, the amino acid residues implicated in promoter recognition reside in the large fragment of the enzyme.

B. Nicked T 7 RNA Po~merase
Analogously, trypsin proteolysis of T3 RNA polymerase produces a large fragment similar to the large fragment of nicked T 7 RNA polymerase. The isolated large fragment of T 3 RNA polymerase retains T 3 RNA polymerase activity and accurately initiates transcription from a T3 promoter (Bautz, 1976). Since the initiation of transcription occurs from the downstream end of the T3 promoter, the large fragment of T 3 RNA polymerase must contact the downstream end of the promoter. Consequently, it is likely that T7 RNA polymerase binds the T7 promoter with the carboxyl half (large fragment) of the enzyme covering the downstream end of the promoter. Furthermore, if the division of the protected sequences of the two initiation complexes of nicked T7 RNA polymerase is due to the exposure of DNA through the cleavage site in the nicked enzyme, the small size of the block of sequence efficiently protected near the upstream end of the T7 promoter might suggest that the small fragment of the nicked enzyme binds near the upstream end of the promoter. This would indicate that T7 RNA polymerase binds its promoter with the amino half of the enzyme oriented near the upstream end of the promoter and with the carboxyl half of the enzyme oriented near the downstream end of the promoter.
The nicking of T 7 RNA polymerase not only alters the sequences protected by the polymerase-promoter complexes, but also destabilizes the complexes. The apparent pseudobimolecular equilibrium constant for the formation of the r(GGG)-initiated complex of nicked T7 RNA polymerase is 2 X IO5 M-', whereas the apparent pseudo-bimolecular equilibrium constant for the formation of the r(GGG)-initiated complex of T7 RNA polymerase is 5 X lo5 M-'. The hypothesis that the enzymatic deficiencies of nicked T7 RNA polymerase reflect a destabilization of the DNA.enzyme complexes appears to be confirmed by the reduced Keq for the formation of the initiated complex of nicked T7 RNA polymerase.
At low salt concentrations, T 7 RNA polymerase binds nucleic acid nonspecifically. The true Keq for the formation of the closed complex of T7 RNA polymerase must then be greater than the apparent Keq = 2 X lo5 M-' determined by protection of the 410 promoter from cleavage by MPE.Fe(I1) (see "Experimental Procedures"). Qualitatively, sequence protection by the closed complex of nicked T7 RNA polymerase shows much less interference from nonspecific binding than sequence protection by the closed complex of T7 RNA polymerase (Fig. 2). As a consequence, protection of the binding site is more noticeable with nicked T7 RNA polymerase than with T7 RNA polymerase. The apparent Keg determined for the formation of the closed complex of the nicked enzyme is 5 X lo5 M-*; however, nonspecific binding of nicked T7 RNA polymerase may not be negligible under our low Mg2+ conditions. As a consequence the apparent K,, determined for the closed complex of nicked T7 RNA polymerase is only an estimate of the lower limit of the true Keg for formation of complex. How good an estimate the apparent Keg of the closed complex of nicked T7 RNA polymerase is cannot be determined from this data. Alternative methods will need to be used to fully characterize these complexes.
The decrease in the apparent Keg for the formation of a polymerase-promoter complex with increasing M$+ concentrations is related to the number of M$+ ions that are released by the polymerase-promoter complex. The slope of a log-log plot of the apparent Keg versus the M$+ concentration is equivalent to the number of Mg2+ ions that are displaced from the DNA template during the binding of the RNA polymerase. Plots for the r(GGG)-initiated complex of T7 RNA polymerase and the r(GGG)-initiated complex of nicked T7 RNA polymerase suggest that the r(GGG)-initiated complex of T7 RNA polymerase releases four Mg2+ ions and the r(GGG)initiated complex of nicked T7 RNA polymerase two M$+ ions.
Assuming that for Mg'+ is equal to 0.47 (Record et al., 1976), the number of ion pairs formed by a polymerasepromoter complex is given by the slope of the log K,, versus -log [M$+] divided by 0.47. The r(GGG)-initiated complex of T7 RNA polymerase forms eight ion pairs, whereas the r(GGG)-initiated complex of nicked T7 RNA polymerase forms four ion pairs. Apparently, the proteolysis of T7 RNA polymerase disrupts four of the enzyme-phosphate pairs that are normally formed in the r(GGG)-initiated complex. The reduced stability of the nicked enzyme complex is probably related to this reduction in the number of ion pairs formed between the enzyme and the phosphates. If less than the normal number of ion pairs are formed in all of the complexes of the nicked T7 RNA polymerase, the accompanying reduction in the stability of all of the polymerase.DNA complexes may account for the reduced initiation efficiency and the increased termination frequency of the nicked T7 RNA polymerase. In addition, this disruption of ionic interactions would affect nonspecific binding of the nicked enzyme more dramatically than promoter specific binding since electrostatic interactions are more important for nonspecific binding than for specific binding.
In summary, nicked T7 RNA polymerase initiates transcription via the same sequential multistep mechanism observed for T7 RNA polymerase. Although the steps in the initiation of transcription are identical for the two enzymes, the complexes formed by nicked T7 RNA polymerase are different from the complexes formed by T7 RNA polymerase. It seems apparent that some of the molecular changes in the polymerase-promoter complexes of nicked T7 RNA polymerase are reflected in the enzymatic changes noted in the nicked enzyme, but a complete analysis of the relationship of the structure of T 7 RNA polymerase to its enzymatic activities will require the systematic examination of each of the domains of the protein. Chemicals--Inorganic s a l t s of reagent grade were obtained from purchased from Pharmacia. I T -32PlATP was obtained from ICN, and 1s -32PldATP and l a -32PldCTP were Obcalned from New England Nuclear.
Enn mes--T7 RNA polymerase (Tabor and Richardson, 19851 and n i c k e h A polymerase (Ikeda and Richardson. 19871 were pucifled as described. The T7 RNA polymerase ( 1 . 1 8 mg/mll has a specif~c actlvlty of 255.000 U/mg (21, and the nicked T7 RNA polymerase (0.857 mg/'oli has a specific activity of 12. 500 U/mg. RestrictLon enzymes, large fragment of Escherichia coll UNA polymerase I and T4 polynucleotide kinase were p u r c h e S e d f r a e w England 8iolabl. E. coil bacterial alkaline phosphatase was purchased from Bethesda iieieszh In the absence of T7 RNA polymerase 139 of the total radioactivity) was treated as background and vas subtracted from the tabulated "dl'Je*.
Promoter--The a R V / X I fragment of pRI10 that c o n t a~n s the (10 promoter was used i n the protection studies. The 32P label was introduced at the %I Lerrninl at either the 5"end ICY -32PlATP and TI PolynucleOLide klnasel Or the 3"end ( I o -32PldCTP and the large fragment of E. coli DNA pOlymeraIe 11 iPLaxdm and GllberC, 19801. After digesting% labeled DNA w i t h EcORV, the deslred 312 bp restriction fragment was purlfied by ~y a c r y l a m i d e g e l electcophoresls. The fragment was eluted from the gel makrix with 0.6M N H~O A C , and t h e e l u a n t was flltered through a 0.45 urn filter Precipitation ~n the pcesence of c a r r i e r tRNA IManlatis %dl., Inaxam and GliberC, 1980). The UNA was concentrated by ethanol 19821. The fragment was then redissolved I n water for use in the . .
HkIIIC mlcrodenritoneter. Protection of the (10 promoter by T7 RNA polymerase or the proteolytically nicked T7 RNA polymerase *as derived from samples contalnlng RNA polymerase with densitometec scans identified by comparative a n a l y s i s of the densitometer scans Of lanes of control lanes derived from samples l a c k i n g T7 RNA polymerase.
The auloradiograph~ of the gels were scanned on a Joyce-Loebl described above was modlfled to examine the behavior of the t w o Tl RNA polymerase during the binding of the (10 promoter and during the lnlriation of transcriptLon from the (LO promoter. The modifications included changes ~n the concentration of Tcls'HC1. of UgC12. and of As Indicated i n the Plgures and the -, the procedure RNA polymerase. In addLCLon, the nucleoside 5"tciphosphates. ATP, CTP, and GTP, were either added or omitted as needed. MPE'FelII) preferentially nicks double stranded nucleic acid efficienrlv and n~n~~e~i f~c a l l r : Conseauentlv. the fraamentatlon of Determination of the fractional Occupancy at the (10 Promoter" ONA is uniform. and degradation is propOCLlOna1 Lo the cnncentrdtlon of I I P E ' F e ( I 1 ) . When DNA seavences are protected from cleavage by MPE'FelIII. the protected sequences can be eaSlly dlstlngvlshed from nonprotecred sequences. . .