Abstract
Interactions between the cAMP receptor protein (CRP) and the carboxy-terminal regulatory domain (CTD) of Escherichia coli RNA polymerase α subunit were analyzed at promoters carrying tandem DNA sites for CRP binding using a chemical nuclease covalently attached to α. Each CRP dimer was found to direct the positioning of one of the two α subunit CTDs. Thus, the function of RNA polymerase may be subject to regulation through protein–protein interactions between the two α subunits and two different species of transcription factors.
Keywords: protein–protein contact, protein–DNA contact, chemical nuclease, transcription regulation
The RNA polymerase holoenzyme of Escherichia coli is composed of core enzyme with subunit structure α2ββ′, responsible for RNA polymerization, and one of multiple species of σ subunit, responsible for promoter recognition. Promoter selectivity of the holoenzyme is modulated by direct or indirect interaction with many transcription factors, resulting in switching of the global pattern of gene transcription according to the environment. The best-characterized target on the RNA polymerase involved in molecular communication with transcription factors is the α subunit carboxy-terminal domain (CTD) that contains the contact sites for class I transcription factors. The α subunit, consisting of 329 amino acid residues, is composed of two structural domains, each responsible for distinct functions (1–3) and each forming independent structural domains connected by a protease-sensitive flexible linker (4–6). The amino (N)-terminal domain from residues 20 to 235 plays a key role in RNA polymerase assembly by providing the contact surface for α dimerization and binding of β and β′ subunits (7–10), whereas the CTD from residues 235 to 329 plays a regulatory role by providing the contact surfaces for trans-acting protein factors and cis-acting DNA elements (11–14).
Whereas the regulation of many E. coli promoters involves a single factor, some promoters are regulated by two or more transcription factors, and such coregulation systems involving multiple species of transcription factors can couple gene expression to diverse environmental conditions. Knowledge of the molecular mechanism of prokaryotic transcription regulation involving more than two factors would contribute much to understanding of the events carried out in eukaryotes, because the regulation of gene transcription in eukaryotes generally involves the action of multiple transcription factors. To gain insight into this problem, we analyzed interactions between RNA polymerase and cAMP receptor protein (CRP) dimers on promoters carrying tandem CRP-binding sites at various positions relative to the transcription start site. A set of promoters was constructed carrying one DNA site for CRP centered at position −41.5 upstream from the transcription start point and a second DNA site for CRP located further upstream (refs. 15 and 16; T.A.B., V. A. Rhodius, C. L. Webster & S.J.W.B., unpublished work). Experiments described herein suggest that the CTDs of the two α subunits can interact with different subunits of both bound CRP dimers, and this results in complexes of different architectures depending on the positions of the upstream CRP dimer. To investigate the effects of CRP in positioning of the two α subunits, we have applied an affinity DNA cleavage method by protein-conjugated EDTA⋅Fe (17–20), which provides information on the location of each of the two α subunits.
MATERIALS AND METHODS
Preparation of Mutant α Subunits.
Plasmids pGEMAX185, pGEMA(45A)-H6(C), pGEMA269C, and pGEMA(45A)269C encode wild-type α, [45A]α having a C-terminal His6-tag, [269C]α containing only one cysteine residue (Cys-269) with the three other cysteine residues substituted for alanine and [45A269C]α, respectively (20). For production of α derivatives with a His6-tag at the C-terminus, a sequence including six histidine codons was inserted between codon 329 and the stop codon. Wild-type α and all α-derivatives were expressed in E. coli BL21(λDE3) and purified by previously described procedures (21).
Conjugation of (p-bromoacetamidobenzyl)-EDTA⋅Fe (Fe⋅BABE).
The conjugation of Fe⋅BABE onto the α derivatives, [269C]α and [45A269C]α, and the reconstitution of RNA polymerase carrying Fe⋅BABE on both of two α subunits, α(Fe)β⋅α(Fe)β′, or only the β′-associated α, αβ⋅α(Fe)β′, were carried out as described by Murakami et al. (20). The efficiency of Fe⋅BABE conjugation was determined by measuring free cysteine side chains with the fluorescent reagent N-[4-[7-(diethylamino)-4-methylcoumarin-3-yl]phenyl]maleimide (19). The hybrid RNA polymerase carrying Fe⋅BABE on the β-associated α, α(Fe)β⋅αβ′, was reconstituted by mixing Fe⋅BABE-labeled [269C], [45A]α-His6(C), β and β′ in a molar ratio of 1:10:1:1. All of the core enzymes thus assembled were separated from unassembled subunits by DEAE-column chromatography, and thereafter the hybrid RNA polymerases were isolated by Ni2+-affinity column chromatography as described in a previous report (20). Holoenzymes were prepared by mixing the reconstituted core enzymes with 4-fold molar excess of purified σ70 and incubated for 20 min at 30°C immediately before use.
DNA Cleavage by Fe⋅BABE.
DNA cleavage of CRP-dependent promoter fragments by Fe⋅BABE-labeled RNA polymerase was carried out as described by Murakami et al. (20) with minor modifications. Briefly, mixtures of 32P-end labeled DNA fragment and 20 nM Fe⋅BABE-conjugated RNA polymerase were incubated at 37°C for 15 min in 50 μl of the cleavage buffer [10 mM Tris⋅HCl (pH 7.8 at 37°C)/3 mM magnesium acetate/1 mM EDTA/50 mM NaCl/25 μg/ml BSA/10% glycerol]. Sonicated salmon sperm DNA (final 80 μg/ml) was added 1 min before the cleavage reaction to remove nonspecifically bound RNA polymerase. The reaction was initiated by adding sodium ascorbate (final concentration, 2 mM), carried out for 20 min at 37°C and terminated by adding 15 μl of a stop solution (0.1 M thiourea and 20 mM EDTA) and 65 μl of phenol/chloroform. The extracted DNA was precipitated with ethanol and analyzed by electrophoresis on a 6% polyacrylamide gel containing 8 M urea. Gels were visualized, and cleavage patterns were scanned with a BioImage Analyzer BAS2000 (FUJIX, Tokyo). Promoter fragments 5′-labeled on the upper strand were prepared by PCR using 32P-labeled forward primer and nonlabeled reverse primer. The primer sequences were: forward primer, 5′-TAAGAAACCATTATTATCAT-3′; reverse primer, 5′-GGGTAGCCAAATGCGTTGGC-3′.
RESULTS AND DISCUSSION
DNA Cleavage by Fe⋅BABE Bound to the RNA Polymerase α Subunit.
The affinity DNA cleavage method involves three steps: (i) site-specific conjugation of the DNA cleavage reagent, Fe⋅BABE, at Cys-269 of the α subunit; (ii) reconstitution of RNA polymerase containing the Fe⋅BABE specifically conjugated to one of the two α subunits, hereafter designated as α(Fe)β⋅αβ′ and αβ⋅α(Fe)β′, or to both α subunits, designated as α(Fe)β⋅α(Fe)β′; and (iii) DNA cleavage by the Fe⋅BABE-labeled RNA polymerases using a set of synthetic CRP-dependent promoters (15, 16), each carrying two CRP-binding sites. The promoters used contain a downstream binding site centered at position −41.5 and an upstream site at various positions ranging from −71.5 to −113.5 (ML−71.5, ML−74.5, ML−82.5, ML−90.5, ML−102.5, and ML−113.5). The level of transcription activation varies depending on the location of upstream bound CRP. Compared with the activation of CCD4 (a control promoter with a single site for CRP at −41.5), ML−90.5 was 4-fold more active, ML−102.5 was 3.5-fold more active, and ML−74.5, ML−82.5 and ML−113.5 all were 2-fold more active. In contrast, the activity of ML−71.5 is lower than that of CCD4, and the upstream CRP exerts no activation effect (15, 16).
Fe⋅BABE was conjugated at Cys-269 because this residue is not so important for contact with CRP (14) but is exposed on the surface of helix 1 (6) so as to allow monitoring of the location of α C-terminal domain (αCTD) by cleavage of DNA strands. The efficiencies of Fe⋅BABE conjugation were 83.6 and 81.6% for [269C]α and [45A269C]α, respectively, as determined by measuring free cysteine side chains (19). When the Fe⋅BABE-conjugated RNA polymerases were tested for their activities in CRP-directed transcription using the test promoters, the modified RNA polymerases were as active as the unmodified enzymes (data not shown), indicating that the RNA polymerases retain the CRP contact activity even after Fe⋅BABE conjugation.
The results of DNA cleavage are shown in Fig. 1, and scans of cleavage patterns are shown in Fig. 2. The Fe⋅BABE-conjugated RNA polymerases did not cleave any of the test promoter DNAs in the absence of CRP. Thus, all the cleavage was CRP dependent. The control CCD4 promoter has only one CRP-binding site centered at position −41.5 bp upstream from the transcription start site (this is the prototype of a class II CRP-dependent promoter). The ternary complex of doubly modified RNA polymerase, α(Fe)β⋅α(Fe)β′, and CRP at CCD4 showed three separate cleavage regions (−83 ≈ −78, −74 ≈ −66, and −58 ≈ −59) (Fig. 1A, lane 4, and Fig. 2A, blue line). The hybrid-modified RNA polymerase, α(Fe)β⋅αβ′, cleaved the promoter DNA at positions corresponding to the two downstream regions, whereas the second hybrid enzyme, αβ⋅α(Fe)β′, resulted in cleavage at positions corresponding to the two upstream regions (Fig. 1A, lanes 2 and 3, and Fig. 2A, red and green lines). This observation indicates that the two αCTDs of RNA polymerase are located at different positions upstream of the bound CRP dimer and is consistent with the previous DNase I and hydroxyl radical footprinting experiments (22).
Figure 1.
Cleavage of synthetic promoters by RNA polymerase-conjugated with Fe⋅BABE. The top strand of promoter DNA fragments was 5′ end-labeled with 32P, and used for cleavage by RNA polymerase containing α subunit-associated Fe⋅BABE. The promoters used were: (A) CCD4, (B) ML−74.5, (C) ML−82.5, (D) ML−90.5, (E) ML−102.5, and (F) ML−113.5. The Fe⋅BABE-conjugated RNA polymerases used are indicated on each gel lane. These include α(Fe)β⋅α(Fe)β′, α(Fe)β⋅αβ′, and αβ⋅α(Fe)β′. Lane A+G shows the A+G specific Maxam-Gilbert sequence marker.
Figure 2.
Gel scan of the DNA cleavage patterns. Gel patterns A–F shown in Fig. 1 were scanned using a BioImage Analyzer BAS2000. Blue lines represent the cleavage by the doubly labeled RNA polymerase [α(Fe)β⋅α(Fe)β′], and the cleavage patterns by the singly labeled enzymes are shown by green [αβ⋅α(Fe)β′] or red [α(Fe)β⋅αβ′] lines.
Repositioning of the CTD of α Subunit by a Second CRP.
Using the Fe⋅BABE-labeled RNA polymerase, we then examined the effect of a second upstream CRP molecule on the positioning of the CTD of each α subunit. For each promoter the cleavage pattern was determined for RNA polymerase with both α subunits labeled (blue scans in Fig. 2 B–F), with only the β-associated α labeled (red scans) and with only β′-associated α labeled (green scans). When the second upstream CRP is located at −71.5 (ML−71.5), the modified RNA polymerases did not cleave DNA efficiently (data not shown), presumably because insufficient space on the DNA between the two CRP dimers does not allow the association of α subunits. When the second upstream CRP is located at −74.5 (ML−74.5), the double-modified enzyme cleaved four separates regions, i.e., two regions between the two bound CRP dimers (−71 ≈ −66 and −60 ≈ −58), and two regions upstream of the upstream CRP dimer (−103 ≈ −99 and −92 ≈ −89) (Fig. 1B, lane 3, and Fig. 2B, blue line). The results of DNA with the hybrid enzymes clearly show that the β-associated α contacts DNA between two CRP-binding sites and that the β′-associated α binds upstream of the second CRP site (Fig. 1B, lanes 1 and 2, and Fig. 2B, red and green lines).
When the second CRP site was shifted to −82.5 (ML−82.5), the doubly modified RNA polymerase showed three separate cleavage regions (Fig. 1C, lane 4, and Fig. 2C, blue line). The sites of DNA cleavage by α(Fe)β⋅αβ′ were essentially the same as in the case of CCD4 (Fig. 2A) and ML−74.5 (Fig. 2B) (Fig. 1C, lane 2 and Fig. 2C, red line), although interestingly, DNA cleavage by αβ⋅α(Fe)β′ occurred between the two CRP dimers and adjacent to the regions cleaved by β-associated α (Fig. 1C, lane 3, and Fig. 2C, green line). Thus, in contrast to the situation with CRP dimers centered at −41.5 and −74.5 (ML−74.5), sufficient space appears to be available to accommodate the two αCTDs on the DNA between two CRP dimers centered at −41.5 and −82.5 (ML−82.5).
At the promoters with the second upstream CRP centered further upstream, the doubly modified RNA polymerase generated cleavage only between the two CRP-binding sites (ML−90.5, ML−102.5, and ML−113.5; Figs. 1 D–F and 2 D–F). At these promoters, α(Fe)β⋅αβ′ and αβ⋅α(Fe)β′ cleaved sites on the DNA corresponding to the downstream and upstream regions, respectively, that are cleaved by the double-labeled RNA polymerase. The DNA cleavage patterns generated by the β-associated α are essentially the same as those observed with CCD4 without the second upstream CRP, supporting the conclusion that the β-associated α remains in contact with the downstream CRP dimer. On the other hand, both the location and level of DNA cleavage by the β′-associated α varied greatly depending on the location of the second upstream CRP. Comparison with the case of a single CRP site (CCD4, Figs. 1A and 2A) showed that the level of DNA cleavage by αβ⋅α(Fe)β′ increased when the second CRP was centered at −90.5 (Fig. 2D) and −102.5 (Fig. 2E). In the latter case (ML−102.5), the positions of DNA cleavage shifted (Fig. 2E). These changes were, however, not observed when the CRP-binding site was shifted even further upstream to position −113.5 (Fig. 2F). These results indicate that the β′-associated α interacts with the upstream CRP dimer. The upstream limit for the position of the upstream CRP dimer such that it can interact with the β′-associated α appears to be around position −100 bp from the transcription start site.
Involvement of Protein–Protein Contacts in the Positioning of RNA Polymerase α Subunits.
The results presented here argue that two CRP dimers can interact independently with the two α subunits of RNA polymerase and direct their positioning along the promoter DNA. One possibility is that the altered binding of α is due to CRP-induced changes in DNA conformation rather than to direct CRP-α contacts. To investigate this point, DNA cleavage experiments were carried out using the [H159L]CRP mutant, which carries the His-to-Leu substitution at residue 159. This substitution results in a positive control (PC) mutant of CRP that can bind its target site on DNA and induce DNA bending, but is defective in transcription activation at both class I and class II CRP-dependent promoters (23, 24).
The DNA cleavage pattern of the control CCD4 promoter by the modified RNA polymerase was qualitatively similar with both wild-type CRP and [H159L]CRP: with the mutant CRP the amount of DNA cleavage is reduced, particularly at positions −58 and −59 (Fig. 3A), consistent with decreased interaction between the β-associated α and CRP. When a second upstream CRP dimer was bound at −74.5 (Fig. 3B) and at −90.5 (Fig. 3C), the cleavage patterns observed by the modified RNA polymerase was markedly different between the [H159L]mutant and wild-type CRP. The upstream cleavage regions due to the β′-associated α completely disappeared. This indicates that the β′-associated αCTD interacts directly with the upstream site-bound CRP and that His-159 of CRP is required for this interaction.
Figure 3.
Effect of H159L positive control substitution in CRP on DNA cleavage by Fe⋅BABE-conjugated RNA polymerase. The DNA cleavage reaction was carried out using the wild-type CRP or [H159L]CRP mutant, doubly labeled RNA polymerase [α(Fe)β⋅α(Fe)β′], and the promoters (A) CCD4, (B) ML−74.5, and (C) ML−90.5 with 32P-end-labeled top strand. The gel patterns were scanned with a BioImage Analyzer BAS2000.
Flexibility of the α Subunit CTD.
The linker connecting the N- and C-terminal domains of α is sufficiently flexible to allow the CTD to move freely at least 30 bp along the promoter DNA. This is supported by previous reports that a single CRP site can be shifted from −61.5 to around −90, with retention of transcription activation (25) and with retention of the same critical determinants of CRP (26). NMR studies also indicate the presence of a flexible linker of at least 13 residues between the N- and C-terminal domains (27). The DNA cleavage experiments in this study agree with the proposal that this linker is indeed sufficiently flexible enough to allow movement of the αCTD over a distance of about 30 bp on the DNA (Fig. 4), because: (i) β′-associated α is able to contact up to position −96 if the second CRP is located at −74.5 (Figs. 1B and 2B); and (ii) β′-associated α can interact with the upstream CRP bound at −102.5 (Figs. 1E and 2E). Each Fe⋅BABE-labeled α subunit generates two regions of DNA cleavage separated by about 10 bp (or approximately one helical turn of DNA). These two cleavage regions probably are due to stretching of a single α-conjugated Fe⋅BABE (approximate Fe⋅BABE chelate length is 12 Å) in either of two directions along the DNA, and the position of the αCTD on the DNA at different promoters is estimated, by assuming that the α subunit is located at the center between the two DNA cleavage peaks.
Figure 4.
Model for the mechanism of independent positioning of two α subunit CTDs by two CRP dimers. On the CCD4 promoter carrying a single DNA site for CRP, both of the two α subunit CTDs bind upstream of the CRP dimers (A). In the case of promoters carrying tandem DNA sites for CRP, each of the two α subunit CTDs interacts independently with each of two CRP dimers and the location of the CTD of β′-associated α shifts depending on the location of upstream CRP (B) ML−74.5 and (C) ML−82.5, ML−90.5, and ML−102.5. When the upstream DNA site for CRP is shifted further to position −113.5, the α subunit CTD is unable to make contact with the upstream CRP (D) ML−113.5.
The observations described here demonstrated that the two RNA polymerase α subunits are indeed able to contact two different factors independently (Fig. 4). If other transcription factors interact with the σ, β, or β′ subunits, RNA polymerase potentially could participate in regulatory complexes involving three or more different factors.
Acknowledgments
We thank Nobuyuki Fujita for many useful discussions about this work. This work was supported by Grants-in-Aid from the Ministry of Education, Science and Culture of Japan, the Proposed-Based Advanced Industrial Technology Research and Development Program from the New Energy and Industrial Technology Development Organization (NEDO) of Japan, and the Core Research for Evolutional Science and Technology (CREST) of the Japan Science and Technology Corporation (JST). The construction of the different promoters was funded by a project grant from the Wellcome Trust.
ABBREVIATIONS
- Fe⋅BABE
(p-bromoacetamidobenzyl)-EDTA⋅Fe
- CRP
cAMP receptor protein
- CTD
carboxy-terminal domain
References
- 1.Ishihama A. Mol Microbiol. 1992;6:3283–3288. doi: 10.1111/j.1365-2958.1992.tb02196.x. [DOI] [PubMed] [Google Scholar]
- 2.Ishihama A. J Bacteriol. 1993;175:2483–2489. doi: 10.1128/jb.175.9.2483-2489.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ebright R H, Busby S. Curr Opin Genet Dev. 1995;5:197–203. doi: 10.1016/0959-437x(95)80008-5. [DOI] [PubMed] [Google Scholar]
- 4.Blatter E, Ross W, Tang H, Gourse R L, Ebright R H. Cell. 1994;78:889–896. doi: 10.1016/s0092-8674(94)90682-3. [DOI] [PubMed] [Google Scholar]
- 5.Negishi T, Fujita N, Ishihama A. J Mol Biol. 1995;248:723–728. doi: 10.1006/jmbi.1995.0254. [DOI] [PubMed] [Google Scholar]
- 6.Jeon Y H, Negishi T, Shirakawa M, Yamazaki T, Fujita N, Ishihama A, Kyogoku Y. Science. 1995;270:1495–1497. doi: 10.1126/science.270.5241.1495. [DOI] [PubMed] [Google Scholar]
- 7.Kimura M, Fujita N, Ishihama A. J Mol Biol. 1994;242:107–115. doi: 10.1006/jmbi.1994.1562. [DOI] [PubMed] [Google Scholar]
- 8.Kimura M, Ishihama A. J Mol Biol. 1995;248:756–767. doi: 10.1006/jmbi.1995.0258. [DOI] [PubMed] [Google Scholar]
- 9.Kimura M, Ishihama A. J Mol Biol. 1995;254:342–349. doi: 10.1006/jmbi.1995.0621. [DOI] [PubMed] [Google Scholar]
- 10.Kimura M, Ishihama A. Genes Cells. 1996;1:517–528. doi: 10.1046/j.1365-2443.1996.d01-258.x. [DOI] [PubMed] [Google Scholar]
- 11.Igarashi K, Ishihama A. Cell. 1991;32:319–325. [Google Scholar]
- 12.Igarashi K, Hanamura A, Makino K, Aiba H, Aiba H, Mizuno T, Nakata A, Ishihama A. Proc Natl Acad Sci USA. 1991;88:8958–8962. doi: 10.1073/pnas.88.20.8958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gaal T, Ross W, Blatter E E, Tang H, Jia X, Krishnan V V, Assa-Munt N, Ebright R H, Gourse R L. Genes Dev. 1996;10:16–26. doi: 10.1101/gad.10.1.16. [DOI] [PubMed] [Google Scholar]
- 14.Murakami K, Fujita N, Ishihama A. EMBO J. 1996;15:4358–4367. [PMC free article] [PubMed] [Google Scholar]
- 15.Busby S, West D, Lawes M, Webster C, Ishihama A, Kolb A. J Mol Biol. 1994;241:341–352. doi: 10.1006/jmbi.1994.1511. [DOI] [PubMed] [Google Scholar]
- 16.Savery N, Rhodius V, Busby S. Philos Trans R Soc London B. 1996;351:543–550. doi: 10.1098/rstb.1996.0053. [DOI] [PubMed] [Google Scholar]
- 17.Dervan P B. Methods Enzymol. 1991;208:497–515. doi: 10.1016/0076-6879(91)08026-e. [DOI] [PubMed] [Google Scholar]
- 18.Ebright Y W, Chen Y, Pendergrast P S, Ebright R H. Biochemistry. 1992;31:10664–10670. doi: 10.1021/bi00159a004. [DOI] [PubMed] [Google Scholar]
- 19.Greiner D P, Miyake R, Moran J K, Jones A D, Negishi T, Ishihama A, Meares C F. Bioconjugate Chem. 1997;8:44–48. doi: 10.1021/bc9600731. [DOI] [PubMed] [Google Scholar]
- 20.Murakami K, Kimura M, Owans J T, Meares C F, Ishihama A. Proc Natl Acad Sci USA. 1997;94:1709–1714. doi: 10.1073/pnas.94.5.1709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fujita N, Ishihama A. In: Methods in Enzymology: RNA Polymerase and Associated Factors. Adhya S, editor; Adhya S, editor. San Diego: Academic; 1996. pp. 121–130. [Google Scholar]
- 22.Belyaeva T, Bown J, Fujita N, Ishihama A, Busby S. Nucleic Acids Res. 1996;24:2243–2251. doi: 10.1093/nar/24.12.2243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bell A, Gaston K, Williams R, Chapman K, Kolb A, Buc H, Minchin S, Williams J, Busby S. Nucleic Acids Res. 1990;18:7243–7250. doi: 10.1093/nar/18.24.7243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zhou Y, Zhang X, Ebright R H. Proc Natl Acad Sci USA. 1993;90:6081–6085. doi: 10.1073/pnas.90.13.6081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ushida C, Aiba H. Nucleic Acids Res. 1990;18:6325–6330. doi: 10.1093/nar/18.21.6325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhou Y, Merkel T J, Ebright R H. J Mol Biol. 1994;243:603–610. doi: 10.1016/0022-2836(94)90035-3. [DOI] [PubMed] [Google Scholar]
- 27.Jeon Y H, Yamazaki T, Otomo T, Ishihama A, Kyogoku Y. J Mol Biol. 1997;267:953–962. doi: 10.1006/jmbi.1997.0902. [DOI] [PubMed] [Google Scholar]