Major Conformational Changes Occur during the Transition from an Initiation Complex to an Elongation Complex by T7 RNA Polymerase
2002; Elsevier BV; Volume: 277; Issue: 45 Linguagem: Inglês
10.1074/jbc.m206658200
ISSN1083-351X
AutoresKaiyu Ma, Dmitri Temiakov, Manli Jiang, Michael Anikin, William T. McAllister,
Tópico(s)DNA Repair Mechanisms
ResumoTo examine changes that occur during the transition from an initiation complex (IC) to an elongation complex (EC) in T7 RNA polymerase (RNAP), we used nucleic acid-protein cross-linking methods to probe interactions of the RNAP with RNA and DNA in a halted EC. As the RNA is displaced from the RNA-DNA hybrid ∼9 bp upstream from the active site (at −9) it interacts with a region within the specificity loop (residues 744–750) and is directed toward a positively charged surface that surrounds residues Lys-302 and Lys-303. Surprisingly, the template and non-template strands of the DNA at the upstream edge of the hybrid (near the site where the RNA is displaced) interact with a region in the N-terminal domain of the RNAP (residues 172–191) that is far away from the specificity loop before isomerization (in the IC). To bring these two regions of the RNAP into proximity, major conformational changes must occur during the transition from an IC to an EC. The observed nucleic acid-protein interactions help to explain the behavior of a number of mutant RNAPs that are affected at various stages in the initiation process and in termination. To examine changes that occur during the transition from an initiation complex (IC) to an elongation complex (EC) in T7 RNA polymerase (RNAP), we used nucleic acid-protein cross-linking methods to probe interactions of the RNAP with RNA and DNA in a halted EC. As the RNA is displaced from the RNA-DNA hybrid ∼9 bp upstream from the active site (at −9) it interacts with a region within the specificity loop (residues 744–750) and is directed toward a positively charged surface that surrounds residues Lys-302 and Lys-303. Surprisingly, the template and non-template strands of the DNA at the upstream edge of the hybrid (near the site where the RNA is displaced) interact with a region in the N-terminal domain of the RNAP (residues 172–191) that is far away from the specificity loop before isomerization (in the IC). To bring these two regions of the RNAP into proximity, major conformational changes must occur during the transition from an IC to an EC. The observed nucleic acid-protein interactions help to explain the behavior of a number of mutant RNAPs that are affected at various stages in the initiation process and in termination. RNA polymerase DNA polymerase initiation complex elongation complex nucleotides wild type hydroxylamine 2-nitro-5-thiocyano-benzoic acid N-chlorosuccinimide 4-thio-UMP template non-template 4-morpholinepropanesulfonic acid 4-morpholineethanesulfonic acid iron (S)-1-(p-bromoacetamidobenzyl)ethylenediamine-tetraacetate endoproteinase GluC (protease V8) E. coli outer membrane protease OmpT 4-thio-dTMP Like all RNA polymerases (RNAPs),1 T7 RNAP forms an unstable initiation complex (IC) that synthesizes and releases short abortive products before clearing the promoter and forming a stable elongation complex (EC) (Ref. 1Martin C.T. Muller D.K. Coleman J.E. Biochemistry. 1988; 27: 3966-3974Crossref PubMed Scopus (235) Google Scholar; for review, see Ref. 2McAllister W.T. Nucleic Acids Mol. Biol. 1997; 11: 15-25Crossref Google Scholar). The transition is accompanied by release of promoter contacts, changes in the size of the footprint of the polymerase on the DNA, and changes in accessibility to cleavage by a variety of proteases (3Ikeda R.A. Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 3614-3618Crossref PubMed Scopus (147) Google Scholar, 4Place C. Oddos J. Buc H. McAllister W.T. Buckle M. Biochemistry. 2000; 38: 4948-4957Crossref Scopus (35) Google Scholar, 5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar, 6Sousa R. Patra D. Lafer E.M. J. Mol. Biol. 1992; 224: 319-334Crossref PubMed Scopus (91) Google Scholar). Taken together, these changes indicate that significant alterations in the organization of the complex (and possibly in the structure of the RNAP) occur during the transition.Kinetic and biochemical analysis, the behavior of certain mutant RNAPs, and the effects of changes in template topology on abortive transcript synthesis suggest that there may be as many as three phases during the transition, one involving complexes that have extended ∼3–5 nt, another from ∼6 to 8 nt, and a final phase from ∼9 to 14 nt (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar,7Guajardo R. Sousa R. J. Mol. Biol. 1997; 265: 8-19Crossref PubMed Scopus (167) Google Scholar, 8Jia Y. Patel S.S. Biochemistry. 1997; 36: 4223-4232Crossref PubMed Scopus (95) Google Scholar, 9Imburgio D. Anikin M. McAllister W.T. J. Mol. Biol. 2002; 319: 37-51Crossref PubMed Scopus (19) Google Scholar, 10Mentesana P.E. Chin-Bow S.T. Sousa R. McAllister W.T. J. Mol. Biol. 2000; 302: 1049-1062Crossref PubMed Scopus (46) Google Scholar, 11Jiang M. Rong M. Martin C.T. McAllister W.T. J. Mol. Biol. 2000; 310: 509-522Crossref Scopus (29) Google Scholar, 12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). During the early stages (up to 8 nt) contacts between the upstream binding region of the promoter and the RNAP are maintained, whereas the active site moves downstream (3–5,13). Promoter clearance occurs when the RNA-DNA hybrid achieves a length of 8–9 bp (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar, 12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar,14Huang J. Sousa R. J. Mol. Biol. 2000; 303: 347-358Crossref PubMed Scopus (68) Google Scholar), but the length of the hybrid continues to increase to 10 bp before the transcription bubble collapses to yield a hybrid length of ∼8 bp, as is observed in the EC (12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). UV laser cross-linking experiments indicate that there may be multiple stages during promoter release, because contacts at −5 and −8 are lost before the contact at −17 (4Place C. Oddos J. Buc H. McAllister W.T. Buckle M. Biochemistry. 2000; 38: 4948-4957Crossref Scopus (35) Google Scholar). In the EC, the RNA-DNA hybrid is ∼8 bp and is enclosed in a transcription bubble of ∼9 bp (14Huang J. Sousa R. J. Mol. Biol. 2000; 303: 347-358Crossref PubMed Scopus (68) Google Scholar, 15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar, 16Liu C. Martin C.T. J. Mol. Biol. 2001; 308: 465-475Crossref PubMed Scopus (145) Google Scholar). The upstream border of the bubble is very close (within 1 bp) to the point at which the RNA is displaced from the template, and the downstream border is very close (within 1 bp) to the 3′ end of the RNA (16Liu C. Martin C.T. J. Mol. Biol. 2001; 308: 465-475Crossref PubMed Scopus (145) Google Scholar). As the RNA is displaced from the hybrid at −9 it becomes associated with a region of the RNAP (the specificity loop) that is involved in promoter recognition. An additional 4–6 nt at the 5′ end of the RNA remains protected from ribonuclease cleavage, presumably because of interactions with the RNAP (14Huang J. Sousa R. J. Mol. Biol. 2000; 303: 347-358Crossref PubMed Scopus (68) Google Scholar, 15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar).Crystal structures have now been solved for free RNAP, for RNAP complexed with a specific inhibitor of transcription (T7 lysozyme), for a binary promoter-RNAP complex, and for an IC that has transcribed the first three bases in the template strand (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar, 17Sousa R. Chung Y.J. Rose J.P. Wang B.C. Nature. 1993; 364: 593-599Crossref PubMed Scopus (338) Google Scholar, 18Cheetham G. Jeruzalmi D. Steitz T.A. Nature. 1999; 399: 80-83Crossref PubMed Scopus (271) Google Scholar, 19Jeruzalmi D. Steitz T.A. EMBO J. 1998; 17: 4101-4113Crossref PubMed Scopus (150) Google Scholar). However, no structure has yet been published for an EC or for intermediate complexes that may form during the transition from an IC to an EC. Here, we have probed the organization of the T7 RNAP EC through the use of nucleic acid-protein cross-linking methods. The results are not easily reconciled with previously known structures of the enzyme and indicate that major structural rearrangements involving the N-terminal domain occur during the transition to an EC.DISCUSSIONAlthough a number of structures of T7 RNAP have been solved, no crystallographic information has yet been published regarding the structure of a T7 RNAP EC. The results described in this work, in which the organization of the EC was probed by nucleic acid-polymerase cross-linking methods, are not easily reconciled with the structure of a T7 RNAP IC and suggest that major conformational changes must occur during the transition from an IC to an EC.In earlier studies, Cheetham and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar) characterized the structure of a T7 RNAP IC in which the first three bases of the template strand had been transcribed (see Fig.6). In this structure, the upstream promoter contacts that were observed in the binary T7 RNAP-promoter complex (in the absence of transcription) are maintained, whereas the active site moves downstream along the T strand, resulting in packing ("scrunching") of the intervening portion of the T strand into a hydrophobic pocket. It was proposed that the strain associated with this packing contributes to the release of abortive products and promoter clearance during the initial phase of transcription (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar,11Jiang M. Rong M. Martin C.T. McAllister W.T. J. Mol. Biol. 2000; 310: 509-522Crossref Scopus (29) Google Scholar, 13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar).In considering the structure of the EC, Cheetham and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar) conclude that extension of the 3-bp RNA-DNA hybrid would result in steric clash with the N-terminal domain of the protein, leading these authors to propose that the length of the hybrid could not exceed 3 bp and to suggest an exit pathway for the displaced RNA as shown by thered arrow in Fig. 6 A. However, subsequent studies demonstrated that the RNA-DNA hybrid in a T7 EC is 7–8 bp (14Huang J. Sousa R. J. Mol. Biol. 2000; 303: 347-358Crossref PubMed Scopus (68) Google Scholar, 15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar, 16Liu C. Martin C.T. J. Mol. Biol. 2001; 308: 465-475Crossref PubMed Scopus (145) Google Scholar) and that as the RNA is displaced from the hybrid it becomes associated with the specificity loop and is directed toward a distal surface of the enzyme (Fig. 6 A, yellow arrow) (15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar). In agreement with the latter findings, the region of the RNAP that becomes cross-linked to the RNA nucleotide at −14 in the EC (the R-14 cross-link) lies along this path (see Figs. 4 and 5). The proposed exit pathway for the RNA is also consistent with the results of recent experiments involving RNA cleavage by Fe-BABE conjugated to amino acid residues in the N-terminal domain of the RNAP (31Mukherjee S. Brieba L.G. Sousa R. Cell. 2002; 110: 1-20Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar).The R-9 cross-link defines the position of the RNA just after it has been displaced from the upstream boundary of the RNA-DNA hybrid at −8 (15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar). It should therefore be in proximity to the T and NT strands of the DNA at the upstream edge of the transcription bubble. However, DNA nucleotides in the T and NT strands at −9 (and nearby) form cross-links to a segment of the protein (residues 172–191; Fig.6 B, yellow) that is quite distant (>40 Å) from the R-9 site, and which is separated by considerable intervening structure from the N-terminal domain (dark blue). Significant rearrangements in the organization of the RNAP must occur during the transition from an IC to an EC to bring these regions into proximity. Pending the availability of structural information we do not know the nature of these rearrangements, but they are likely to be considerable.In considering the possible changes in the IC that must occur during the transition, we extended the 3-bp RNA-DNA hybrid observed in the IC by superimposing the RNA-DNA hybrid observed in a yeast polymerase II transcription complex (25Gnatt A.L. Cramer P., Fu, J. Bushnell D.A. Kornberg R.D. Science. 2001; 292: 1876-1882Crossref PubMed Scopus (741) Google Scholar). In agreement with the cross-linking results presented here and in previous work (15Temiakov D. Mentesana P.E., Ma, K. Mustaev A. Borukhov S. McAllister W.T. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14109-14114Crossref PubMed Scopus (74) Google Scholar), the trajectory of the modeled RNA-DNA hybrid is directed toward the specificity loop and the proposed exit pathway for the displaced RNA. However, as noted by Cheetham and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar), extension of the RNA-DNA hybrid in this manner would result in major steric clashes with the N-terminal region of the RNAP (Fig. 7). We suggest that the changes that occur during the transition to an EC involve significant rearrangements of the N-terminal domain and, at a minimum, require movement of the 172–191 region toward the R-9 contact in the specificity loop (Fig. 6 B, dashed arrow). An alteration in the position of the specificity loop may also occur during the transition.Figure 7Organization of the active site. Panel A, clash of the RNA-DNA hybrid with elements in the N-terminal domain. Extension of the heteroduplex to 4–5 bp would result in steric clash with a helix-turn-helix motif involving α helices F (dark blue) and G (green) and with helix N (magenta). The α carbons of Glu-148 (E148), Ser-393 (S393), and Arg-394 (R394) are indicated.Panel B, clash of Tyr-639 with the incoming nucleotide in the active site and comparison with the TaqDNAP open complex. The structure shown is that of the T7 RNAP IC (1QLN (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar)). The first two bp of the RNA-DNA hybrid observed in the IC are indicated inred (RNA) and white (DNA). The RNA portion of the heteroduplex from the yeast polymerase II transcription complex (yellow) has been superimposed onto this hybrid as described in Fig. 6; the positions of the bases in the I+1, I-1 and I-2 sites are indicated. α helices Y and Z of T7 RNAP are indicated in light blue, and the interval that has been identified in the R-1 cross-link is indicated in green, including the side chain of Tyr-639. The structures of TaqDNAP and T7 RNAP were aligned by superimposing key conserved residues in both enzymes (see "Experimental Procedures"). The 0 and 01 helices in theTaqDNAP open complex are in red, and the side chain of Tyr-671 is indicated. After the transition from an open to a closed complex, the 0 and 01 helices are in a different position (blue), and Tyr-671 is rotated out of the active site (dashed arrow). The slightly offset positions of Tyr-671 and of the 0 and 01 helixes in the TaqDNAP open complex as compared with that of Tyr-639 and the Y and Z helixes of T7 RNAP may be due to the different geometry of the RNA-DNA heteroduplexversus the DNA-DNA primer-template homoduplex.View Large Image Figure ViewerDownload (PPT)Previous studies of polymerase mutants that are affected in the 172–191 region or RNAPs that have been endoproteolytically cleaved at 172 or 179 by OmpT or by trypsin revealed a number of changes in enzyme behavior including: decreased ability to displace the transcript from the template, the formation of an extended RNA-DNA hybrid, decreased stability of halted elongation complexes, and failure to terminate at class II termination signals (10Mentesana P.E. Chin-Bow S.T. Sousa R. McAllister W.T. J. Mol. Biol. 2000; 302: 1049-1062Crossref PubMed Scopus (46) Google Scholar, 32Ikeda R.A. Richardson C.C. J. Biol. Chem. 1987; 262: 3790-3799Abstract Full Text PDF PubMed Google Scholar, 33Gopal V. Brieba L.G. Guajardo R. McAllister W.T. Sousa R. J. Mol. Biol. 1999; 290: 411-431Crossref PubMed Scopus (49) Google Scholar, 34He B. Kukarin A. Temiakov D. Chin-Bow S.T. Lyakhov D.L. Rong M. Durbin R.K. McAllister W.T. J. Biol. Chem. 1998; 273: 18802-18811Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar, 35Lyakhov D.L., He, B. Zhang X. Studier F.W. Dunn J.J. McAllister W.T. J. Mol. Biol. 1997; 269: 28-40Crossref PubMed Scopus (55) Google Scholar, 36Lyakhov D.L., He, B. Zhang X. Studier F.W. Dunn J.J. McAllister W.T. J. Mol. Biol. 1998; 280: 201-213Crossref PubMed Scopus (74) Google Scholar, 37Macdonald L.E. Durbin R.K. Dunn J.J. McAllister W.T. J. Mol. Biol. 1994; 238: 145-158Crossref PubMed Scopus (70) Google Scholar, 38Macdonald L.E. Zhou Y. McAllister W.T. J. Mol. Biol. 1993; 232: 1030-1047Crossref PubMed Scopus (108) Google Scholar). All of these effects are consistent with the notion that this region of the RNAP interacts with the upstream edge of the transcription bubble, perhaps helping to resolve or stabilize the bubble and/or aid in RNA displacement. For example, termination at class II signals requires proper resolution of the bubble and RNA displacement (34He B. Kukarin A. Temiakov D. Chin-Bow S.T. Lyakhov D.L. Rong M. Durbin R.K. McAllister W.T. J. Biol. Chem. 1998; 273: 18802-18811Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar, 39Hartvig L. Christiansen J. EMBO J. 1996; 15: 4767-4774Crossref PubMed Scopus (43) Google Scholar, 40Kwon Y.-S. Kang C. J. Biol. Chem. 1999; 41: 29149-29155Abstract Full Text Full Text PDF Scopus (18) Google Scholar); the observation that polymerases whose structure is altered in this region do not terminate at these signals is consistent with the proposed role of this element in resolving the bubble. Similarly, the observation that the mutant or proteolytically cleaved RNAPs are less stable when halted and are sensitive to changes in template topology that decrease RNA displacement, were taken to indicate that they are defective in resolving the transcription bubble and displacing the transcript (10Mentesana P.E. Chin-Bow S.T. Sousa R. McAllister W.T. J. Mol. Biol. 2000; 302: 1049-1062Crossref PubMed Scopus (46) Google Scholar,33Gopal V. Brieba L.G. Guajardo R. McAllister W.T. Sousa R. J. Mol. Biol. 1999; 290: 411-431Crossref PubMed Scopus (49) Google Scholar).The trajectory of the extended RNA-DNA hybrid proposed here is such that the first steric clashes (at 4–5 nt) would involve a helix-turn-helix motif that includes α helices F and G (which project into the binding cleft) and α helix N (which lies at the base of the thumb) (Fig. 7 A). Substitutions of residue Glu-148 (located at the base of helix G) results in a mutant RNAP that specifically aborts transcription at 5 nt (41He B. Rong M. Durbin R.K. McAllister W.T. J. Mol. Biol. 1997; 265: 275-288Crossref PubMed Scopus (44) Google Scholar), and it had previously been suggested that this region of the RNAP might be involved in the transition from an IC to an EC by sensing the presence of an RNA-DNA hybrid of suitable length (41He B. Rong M. Durbin R.K. McAllister W.T. J. Mol. Biol. 1997; 265: 275-288Crossref PubMed Scopus (44) Google Scholar). Substitutions of key residues in the N-helix (i.e. Ser-393 and Arg-394, which lie at the base of the thumb) also result in increased termination at 5 nt, presumably reflecting the importance of interactions with the thumb domain once the heteroduplex has achieved a length of 4–5 bp (42Brieba L.G. Gopal V. Sousa R. J. Biol. Chem. 2000; 276: 10306-10313Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar).As noted above, a variety of lines of evidence suggests that there may be three stages during the transition process, one occurring from 3 to 5 nt, another from 6 to 8 nt, and a final phase from 9 to 14 nt. In this regard, we note that further extension of the heteroduplex beyond the clash with the F-G helices would result in a clash with the Val-237 loop (at 6–7 nt). It is possible that events that occur in the transition process after 5 nt may involve contacts with the emerging heteroduplex or with the displaced RNA product. These effects may be superimposed upon the effects of the strain introduced by DNA packing. Unlinking the binding and initiation regions of the promoter by disrupting the intervening portion of the T strand dramatically reduces the release of abortive products 6 nt (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar,11Jiang M. Rong M. Martin C.T. McAllister W.T. J. Mol. Biol. 2000; 310: 509-522Crossref Scopus (29) Google Scholar). This issue will require further study.As shown in Fig. 3, the base at the 3′ end of the RNA in a halted EC forms a cross-link (R-1) with the interval comprising residues 636–643, most likely with Tyr-639. In the structure of the T7 RNAP IC, Tyr-639 is close to the base at the 3′ end of the growing transcript (in the I-1 site), suggesting that the organization of this region of the active site is similar in both complexes (see Fig.7 B).The RNA-DNA hybrid depicted in Figs. 6 and 7 is from a yeast RNAP II complex that is in the pretranslocated state (i.e. formation of the phosphodiester bond has been completed, but the product has not yet been moved out of the active site). Modeling of this hybrid into the IC results in a clash between Tyr-639 and the nucleotide in the I+1 site (Fig. 7 B). A similar clash was observed by Cheetam and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar) when an incoming nucleotide (which would occupy the same position as the terminal nucleotide in the pretranslocated RNA) was modeled into the I+1 site based upon homology with a T7 DNAP complex. In studies of quaternary complexes of TaqDNAP, Li et al. (26Li Y. Korolev S. Waksman G. EMBO J. 1998; 17: 7514-7525Crossref PubMed Scopus (653) Google Scholar) observed movement of Tyr-671 (which is analogous to Tyr-639 in T7 RNAP) and the O and O1 helices (which are analogous to the Y and Z helices of T7 RNAP) during the transition from an open to a closed complex. In the open complex (in which the incoming dNTP is bound but has not yet been positioned in a catalytically productive configuration) Tyr-671 was observed to stack on the template base at the end of the primer-template duplex and to partially occupy the active site. In the transition to the closed complex, Tyr-671 vacates this position, allowing the incoming dNTP to pair correctly with the template base. As noted in Fig. 7 B, the disposition of Tyr-639 in the T7 RNAP IC is similar to that of Tyr-671 in theTaqDNAP open complex but not in the closed complex. These considerations lead us to suggest that the T7 RNAP IC complex solved by Cheetham and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar) represents a post-translocated complex that is in the open configuration (i.e. the I+1 site is not yet occupied by the incoming NTP) and further to suggest that movement of helices Y and Z (and of Tyr-639) may be important during the translocation event and in the transition from an open to closed state, as has been proposed for DNAPs (see also Ref. 43Sousa R. Uirusu. 2001; 1: 81-94Crossref Scopus (5) Google Scholar). Significantly, Cheetham and Steitz (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar) report that in another crystal structure of the IC in which the incoming nucleotide was visible, Tyr-639 was poorly ordered. A proposed role for Tyr-639 in the nucleotide binding step would be in agreement with biochemical and genetic studies in which it was shown that this residue is involved in discrimination against the incorporation of dNTPs (44Sousa R. Padilla R. EMBO J. 1995; 14: 4609-4621Crossref PubMed Scopus (243) Google Scholar) (a previous suggestion that His-784 plays a role in substrate discrimination, based upon modeling of the incoming nucleotide from a closed T7 DNAP complex into the IC (13Cheetham G. Steitz T.A. Science. 1999; 286: 2305-2309Crossref PubMed Scopus (288) Google Scholar), is not in agreement with subsequent genetic analysis (45Brieba L.G. Sousa R. Biochemistry. 2000; 39: 919-923Crossref PubMed Scopus (34) Google Scholar)).CONCLUSIONSThe results of these studies indicate that the transition from an IC to an EC by T7 RNAP is a complex process that may involve multiple steps and involves major reorganization of the transcription complex, particularly in the N-terminal domain. Although available structures have captured some complexes along this pathway, it is clear that multiple structures will need to be solved to gain a complete understanding of the process. Until such structures emerge, biochemical and genetic studies will continue to provide useful insights (and will be necessary to corroborate structural analyses). Because the single subunit phage RNAPs carry out all of the same steps in the transcription process as the more complex multisubunit RNAPs, the results of these studies will be of considerable interest in guiding and interpreting similar experiments in multisubunit systems. Like all RNA polymerases (RNAPs),1 T7 RNAP forms an unstable initiation complex (IC) that synthesizes and releases short abortive products before clearing the promoter and forming a stable elongation complex (EC) (Ref. 1Martin C.T. Muller D.K. Coleman J.E. Biochemistry. 1988; 27: 3966-3974Crossref PubMed Scopus (235) Google Scholar; for review, see Ref. 2McAllister W.T. Nucleic Acids Mol. Biol. 1997; 11: 15-25Crossref Google Scholar). The transition is accompanied by release of promoter contacts, changes in the size of the footprint of the polymerase on the DNA, and changes in accessibility to cleavage by a variety of proteases (3Ikeda R.A. Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 3614-3618Crossref PubMed Scopus (147) Google Scholar, 4Place C. Oddos J. Buc H. McAllister W.T. Buckle M. Biochemistry. 2000; 38: 4948-4957Crossref Scopus (35) Google Scholar, 5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar, 6Sousa R. Patra D. Lafer E.M. J. Mol. Biol. 1992; 224: 319-334Crossref PubMed Scopus (91) Google Scholar). Taken together, these changes indicate that significant alterations in the organization of the complex (and possibly in the structure of the RNAP) occur during the transition. Kinetic and biochemical analysis, the behavior of certain mutant RNAPs, and the effects of changes in template topology on abortive transcript synthesis suggest that there may be as many as three phases during the transition, one involving complexes that have extended ∼3–5 nt, another from ∼6 to 8 nt, and a final phase from ∼9 to 14 nt (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar,7Guajardo R. Sousa R. J. Mol. Biol. 1997; 265: 8-19Crossref PubMed Scopus (167) Google Scholar, 8Jia Y. Patel S.S. Biochemistry. 1997; 36: 4223-4232Crossref PubMed Scopus (95) Google Scholar, 9Imburgio D. Anikin M. McAllister W.T. J. Mol. Biol. 2002; 319: 37-51Crossref PubMed Scopus (19) Google Scholar, 10Mentesana P.E. Chin-Bow S.T. Sousa R. McAllister W.T. J. Mol. Biol. 2000; 302: 1049-1062Crossref PubMed Scopus (46) Google Scholar, 11Jiang M. Rong M. Martin C.T. McAllister W.T. J. Mol. Biol. 2000; 310: 509-522Crossref Scopus (29) Google Scholar, 12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). During the early stages (up to 8 nt) contacts between the upstream binding region of the promoter and the RNAP are maintained, whereas the active site moves downstream (3–5,13). Promoter clearance occurs when the RNA-DNA hybrid achieves a length of 8–9 bp (5Brieba L.G. Sousa R. EMBO J. 2001; 20: 6826-6835Crossref PubMed Scopus (45) Google Scholar, 12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar,14Huang J. Sousa R. J. Mol. Biol. 2000; 303: 347-358Crossref PubMed Scopus (68) Google Scholar), but the length of the hybrid continues to increase to 10 bp before the transcription bubble collapses to yield a hybrid length of ∼8 bp, as is observed in the EC (12Liu C. Martin C.T. J. Biol. Chem. 2002; 277: 2725-2731Abstract Full Text Full Text PDF PubMed Scopus (100) Google Scholar). UV laser cross-linking experiments indicate that there may be multiple stages during promoter release, because contacts at −5 and −8 are lost before the contact at −17 (4Place C. Oddos J. Buc H. McAllister W.T. Buckle M. Biochemistry. 2000; 38: 4948-4957Crossref
Referência(s)