Note to users. If you're seeing this message, it means that your browser cannot find this page's style/presentation instructions -- or possibly that you are using a browser that does not support current Web standards. Find out more about why this message is appearing, and what you can do to make your experience of our site the best it can be.



Abstract
Full Text
A Structural Model of Transcription Elongation
Nataliya Korzheva, Arkady Mustaev, Maxim Kozlov, Arun Malhotra, Vadim Nikiforov, Alex Goldfarb, and Seth A. Darst

Supplementary Material

RNA-protein Crosslinks in the Ternary Elongation Complex

For RNA crosslinking, we used two kinds of TECs. One type was generated using derivatized trinucleotide primers from the T7A1 promoter (Fig. 1a, b). Stalled elongation complexes were prepared by incubation of RNAP with the primer, T7A1 promoter DNA, and a limited set of substrates (ATP, GTP). The resulting TEC with an 11mer RNA transcript (TEC11) was extended by walking with limited sets of substrates to obtain TEC12-44, or shortened by limited pyrophosphorolysis to TEC7-9 (Fig. 1b). The second type of complex was assembled from E. coli core RNAP and synthetic RNA/DNA constructs ('minimal nucleic acid scaffold' or 'MS'; Fig. 1d).

Two kinds of reactive groups were used for RNA crosslinking (Fig. 1a), the fluorinated aromatic azide (I), which reacts non-selectively with any type of amino acid, and the aldehyde (II), which crosslinks exclusively to lysine residues. The RNAP b and/or b' subunits were the targets of crosslinking from all RNA and DNA positions (Fig. 1c, e).

Mapping RNA-protein crosslinks in the b subunit. To map crosslinks within the RNAP large subunits, we used the approach of Grachev et al. (9). The subunit carrying the radiolabeled, crosslinked adduct was subjected to single-hit chemical degradation, yielding two sets of nested fragments, N- and C-terminal. Crosslinks were localized between adjacent cleavage sites by determining the site closest to the N- or C-terminus of the subunit where cleavage resulted in removal of the radioactively labelled fragment.

SDS-PAGE analysis of the CNBr cleavage patterns of the b subunit crosslinked from TECs containing long RNAs (10, 11, 12, 13, 14, 16, 20, 34, or 44 nt) revealed characteristic C-terminal degradation patterns (9), indicating the labels resided near the C-terminus of the polypeptide?(Fig. 2a). In these cases, the radioactively labelled N-terminal cleavage products run near the top of the gel and do not complicate the analysis. For TEC14, 16, 20, 34, or 44 (which all gave identical degradation patterns), the crosslinking site was within the b subunit fragment 1304-1315, since the last radioactive product resulted from cleavage at Met1304 while the next product from cleavage at Met1315 was not labelled (Fig. 2a, lanes 1-3).

For TEC11, 12, or 13 (Fig. 2a, lanes 7-9), the last radioactive product resulted from cleavage at Met800/805 while the next product from cleavage at Met951 was not labelled, placing the crosslinking site within the fragment 805-951. This is clearly illustrated upon quantitation of the SDS gel (Fig. 2b). In this analysis, the bands from Fig. 2a, lane 1, were quantitated and the results shown as a histogram (Fig. 2b, 'lane 1'). This represents a reference C-terminal degradation pattern since the crosslink in this case resides very near the C-terminus of the b subunit. In the histograms underneath, the bands in Fig. 2a, lane 6 or lane 8 were quantitated and normalized with respect to the corresponding band in the reference pattern. In the histrogram for TEC11, 12, or 13 (Fig. 2b, 'lane 9'), the normalized intensity drops abruptly between Met800/805 and Met951, indicating the crosslink resides between those sites.

Quantitative analysis of the degradation pattern for TEC10 (Fig. 2b, 'lane 6') reveals crosslinking within the region 1243-1273. However, the histrogram reveals a two-fold drop in the normalized intensity for all of the cleavage products from Met951 to Met1243, indicating the presence of a second crosslinking site located between residues 805-951.

Crosslinking experiments with the Lys-specific aldehyde probe (reagent II, Fig. 1a) allowed us to track the Lys residues along the trajectory of the exiting, upstream RNA. In this case, identical CNBr degradation patterns were obtained from TEC14, 16, 20, 34, or 44 (Fig. 2a, lanes 1-3) and TEC11, 12, or 13 (Fig. 2a, lanes 7-9) with reagent II as with reagent I. There is only one Lys residue within the interval 1304-1315 (Lys1306), identifying it as the crosslinking target from TEC14-44. There are 8 lysines in the segment 805-951. To refine the Lys-specific mapping in this case, we employed limited proteolysis. In the TEC, trypsin cleaves the bsubunit predominantly at Arg903, generating a 115 kDa N-terminal fragment and a 45 kDa C-terminal fragment (9, 34). For the trypsin cleavage experiments (Fig. 3), we used crosslinked TEC11 complexes with wild type and Xho19 mutant RNAP [R. Landick, J. Stewart, D. N. Lee, Genes Dev. 4, 1623 (1990)] containing a 3 amino acid deletion that eliminates Lys886 (Fig. 3a, bottom). Both of the trypsin cleavage products from the wild-type RNAP were labeled (Fig. 3b), indicating at least one crosslinking site between residues 805-903 and another between 904-951. There are 5 lysines in the former region and 3 in the latter. The mutant RNAP also gave two radioactive tryptic fragments, excluding Lys886 as a crosslinking target (Fig. 3b). To refine the mapping, the smaller tryptic fragment was subjected to single-hit hydroxylamine cleavage (31) at Asn-Gly residues. There are 3 potential cleavage sites in the fragment, between residues 922-923, 1089-1090, and 1299-1300 (Fig. 3a). Three hydroxylamine degradation products were detected; the shortest of which had the mobility expected for the cleavage product at Asn922 (Fig. 3c, lane 2). Hence, the crosslinking target in this case resided within the interval 904-922, which contains two lysine residues, Lys909 and Lys914 (Fig. 3a).

RNA-protein crosslinks from transcription complexes containing short RNAs were generated using the minimal scaffold (MS) system (Fig. 1d) and mapped (Fig. 4). SDS-PAGE analysis of the CNBr cleavage patterns of the b subunit crosslinked from MS complexes containing RNAs of 4, 5, or 6 nt revealed characteristic N-terminal degradation patterns (Fig. 4a, lanes 4-6, quantitated in Fig. 4b, 'lane 6'), as judged by comparison with a reference N-terminal degradation pattern (Fig. 4a, lanes 7-9, quantitated in Fig. 4b, 'lane 7'). The quantitative analysis (Fig. 4b) clearly shows the loss of radioactivity in the cleavage product 492-515.

For crosslinks from TEC7, 8, or 9, the cleavage pattern represents a combination of N- and C-terminal sets of fragments (Fig. 4a, lanes 2, 3). Comparison of the histogram in this case (Fig. 4b, 'lane 2') with reference N-terminal (Fig. 4a, lane 7, quantitated in Fig. 4b) and C-terminal (Fig. 4a, lane 1, quantitated in Fig. 4b) patterns indicate two crosslinking sites. The major crosslink gives rise to an N-terminal pattern, with the crosslink localized in the fragment 492-515. A minor crosslink gives rise to a C-terminal degradation pattern, with the crosslink localized in the fragment 1230-1243.

The segment 492-515 contains a short evolutionarily conserved segment (Fig. 4c, boxed). According to the TEC model, this segment directly contacts RNA within the RNA:DNA hybrid.

Mapping RNA-protein crosslinks in the b' subunit. CNBr degradation of the b' subunit crosslinked from different elongation complexes always resulted in N-terminal degradation patterns (Fig. 5). A typical CNBr degradation pattern for the b' subunit crosslinked from TEC10, 11, 12, 13, 14, 16, 20, 34, or 44 is shown in Fig. 5a (lanes 12-14). The pattern of radioactive products in this case was explained by the location of the label between residues 1-29 for both the non-specific and Lys-specific crosslinking reagents, indicating that lysines from this region contact RNA exiting the TEC. Intensities of the cleavage products from Fig. 5a, lane 14, were quantitated and used as an N-terminal reference pattern for analysis of the other crosslinking sites (Fig. 5b, 'lane 14').

For crosslinking from TEC7-9 (reagent I) and MS6, 7, or 8 (reagents I and II), the degradation patterns placed the crosslink site within the fragment 298-330 (Fig. 5a, lanes 4-8), which was confirmed by analysis of the quantitative histograms of lane 8 (Fig. 5b, 'lane 8').

In this study, one crosslink from the RNA 3'-terminus was mapped. The crosslinkable analog 4-thio-U (4SU) was incorporated into the RNA 3'-terminus of TEC21. A crosslink was then generated and mapped. The loss of radioactivity in the cleavage product 400-466 (Fig. 5a, lanes 1-3, quantitated in Fig. 5b, 'lane 3') places the crosslink in that segment.

Visual inspection of the cleavage pattern for TEC7-9 in the case of the Lys-specific reagent II (Fig. 5a, lanes 9-11) suggested two crosslinking sites. This was confirmed by comparison of the corresponding histogram with the reference (Fig. 5b, 'lane 11'). The first crosslinking site (localized to the fragment 298-330) was predominant (about 80%), while the second one (localized to the fragment 1-29) was minor (about 20%). Interestingly, crosslinks from MS6, 7, or 8 only mapped to one of these regions (298-330). Our interpretation is that the minor site observed from TEC7-9 (1-29) was a crosslink generated from TEC11, an intermediate used to form TEC7-9 by pyrophosphorolysis (Fig. 1b). In the region 298-330, there are two potential targets of Lys-specific modification, Lys321 and Lys325. Lys325 is evolutionarily conserved (Fig. 5c), an indication of its functional significance.

DNA-Protein Crosslinks in the Ternary Elongation Complex To map DNA-protein contacts, we used synthetic constructs (Fig. 6b) assembled from minimal (4) or extended (12) nucleic acid scaffolds. The crosslinking probe (Fig. 6a) was attached to the DNA via synthetically-incorporated phosphorothioate linkages (13) at defined positions (Fig. 6b). Crosslinks were observed to b and/or b' (Fig. 6c) and strong crosslinks were mapped.

Mapping DNA-protein crosslinks in the b' subunit. CNBr degradation of the b' subunit crosslinked from different DNA positions yielded N-terminal degradation patterns (Fig. 7a, lanes 1-6), as judged by comparison with a reference N-terminal pattern (Fig. 7a, lane 7), and C-terminal patterns (Fig. 7a, lanes 8-13). For crosslinks from the upstream DNA duplex (positions -12 of the template strand and -15 of the nontemplate strand), the last radioactive product was due to cleavage at Met102 (Fig. 7a, lanes 4-6, quantitated in Fig. 7b, 'lane 6'), placing the crosslink site within the fragment 29-102. With the probe incorporated at position +6 on the DNA template strand, (lanes 1-3), the degradation pattern indicated that the label was located within the fragment 102-130 (Fig. 7a, lanes 1-3, quantitated in Fig. 7b, 'lane 3'). The significant (about two fold) drop of normalized intensity for the fragment 130-151 compared with 102-130 (Fig. 7b, 'lane 3') indicates the presence of two crosslinking sites, one in the region 102-130 (50%), the other within 130-151 (50%).

For crosslinks from positions +9 and +11 of the template strand, the degradation patterns placed the crosslink within the fragment 1189-1260 (Fig. 7a, lanes 8-10). For positions +5 and +15 of the nontemplate strand, the degradation patterns placed the crosslink within the fragment 1095-1189 (Fig. 7a, lanes 11-13).

Mapping DNA-protein crosslinks in the b subunit. Crosslinks to the bsubunit were mapped from positions +8 and +9 of the template strand, and -5, -3, -1, and +5 of the nontemplate strand. Partial CNBr cleavage in these cases yielded N-terminal patterns, placing all the crosslinks within the interval 130-239 (Fig. 8, lanes 7-9). The mapping was refined using NBS, which cleaves preferentially at Trp residues. For positions -1, -3 and -5 of the nontemplate strand, the shortest peptide was the N-terminal cleavage product at Trp183, pointing to the segment 130-189 as the crosslinking region (Fig. 8, lanes 1-3). For the position +8 of the template strand, both the N-terminal and C-terminal fragments resulting from cleavage at Trp183 were labelled, (Fig. 8, lanes 4-6), indicating two crosslinking sites: 130-183 and 184-239.

The correspondence between the TEC structural model and the crosslinking data is illustrated in Fig. 9 (see legend for details).


Supplemental Figure 1. Systems used to map the 5'-terminal RNA contacts in transcription complexes.

a) Photoreactive fluorinated (F) aromatic azide (I) and Lys-specific aldehyde (II) RNA primers used in crosslinking experiments were synthesized as described for reagent IV (9) starting from 4-azidotetrafluorobenzylamine (37) and 2-hydroxybenzaldehyde.

b) The sequence of RNA synthesized on T7A1 promoter and the route to defined elongation complexes containing the reactive groups. 'R' denotes the crosslinking probe attached to the 5'-terminus of the trinucleotide primer (either I or II from a). The binary complex was formed by the incubation of 5-10 pmol of T7A1 promoter fragment with 1.5 molar excess of His6-tagged RNAP immobilized on Ni2+-NTA agarose for 5 min at 37o in 20-50 ml of transcription buffer (TB) containing 20 mM Hepes (pH 8.0), 50 mM NaCl, and 10 mM MgCl2. RNA synthesis to register 11 (TEC11) was initiated by addition of reactive trinucleotide primer (10 mM), GTP (25 mM), and [a-32P]ATP (0.6 mM), and incubation for a further 10 min at 37°. TEC7-9 were obtained from TEC11 by treatment with 10 mM pyrophosphate at 37° for 5 min. TEC10 was obtained from TEC7-9 by incubation with 25 mM GTP for 2 min at 20°. The rest of the elongation complexes were obtained from TEC11 as described [E. Nudler, A. Goldfarb, M. Kashlev, Science 265, 793 (1994)]. Radioactive RNA products were analyzed on 23% sequencing gels.

c) Autoradiogram showing the reactive RNA products obtained according to the above scheme. Below is shown an autoradiogram of an SDS-PAGE gel of E. coli RNAP band b' subunits radioactively labeled by virtue of RNA crosslinking from the indicated TECs. To achieve the crosslinks, defined elongation complexes obtained as described above were either irradiated at 254 nm for 2 min at 0° (in the case of the photoreactive reagent I) or treated with 10 mM NaBH4 for 15 min at 20° (in the case of the aldehyde reagent II). Products were analyzed by SDS/PAGE as described (19) and visualized by autoradiography and/or quantitated by phosphorimagery (Molecular Dynamics).

d) The structure of minimal RNA:DNA scaffolds (MS) used for crosslinking experiments. 'R' denotes the crosslinking probe attached to the 5'-terminus of the RNA oligonucleotides (either I or II from a). RNA and DNA components of the scaffold (10-20 pmol of each) were mixed in 20-50 ml of TB at 50° and the mixture was allowed to cool to 20° over 10-30 min. A fraction of the solution containing 5-10 pmol of annealed scaffold was added to a 1.5 molar excess of His6-tagged core RNAP immobilized on Ni2+-NTA agarose in 20 ml of TB and kept at 20° for 10 min. This was followed by the addition of [a-32P]CTP (0.6 mM). After incubation for 20 min at 20°, the beads were washed with TB (2 x 1 ml) and the crosslinking performed as described above.

The synthetic complexes (containing 8 nt RNA) had a half-life of about 20 min in TB with 1 M KCl (compared with about 80 min under the same conditions for 'natural' TECs). At low salt conditions (40 mM KCl), the synthetic complexes were as stable as 'natural' TECs.

e) Autoradiogram of an SDS-PAGE gel of E. coli RNAP b and b' subunits crosslinked from MS complexes with 5'-terminal non-specific (I) or Lys-specific (II) reagents.


Medium version | Full size version


Supplemental Figure 2. Mapping of RNA-protein contacts in the C-terminal part of the b subunit.

Single-hit CNBr degradation was performed as described (9).

a) Left - CNBr degradation map of the E. coli RNAP b subunit. The top horizontal bar represents the 1342-amino acid b polypeptide. Met residues (CNBr cleavage sites) are indicated on the bar below. Terminal residues of cleavage products are indicated by lettered numbers, where the letter indicates whether the fragment is N- or C-terminal. Solid and open circles and rectangle indicate the position of the mapped crosslinking sites.

Right - SDS-PAGE and autoradioagraphy of degradation products of crosslinked b. Single-hit CNBr cleavage products of the b subunit are radioactively labeled by virtue of crosslinking from the 5'-end of RNA in TEC14, 16, 20, 34 or 44 (lanes 1-3), TEC10 (lanes 4-6), or TEC11, 12, or 13 (lanes 7-9). Solid lines connect corresponding cleavage products from each lane. The position of the radioactive label was localized between the CNBr cleavage sites giving rise to the shortest labelled cleavage product and the next shortest product which was not radioactive (indicated by dashed lines).

Quantitation of CNBr cleavage products. Top panel represents the histogram of the intensity for the cleavage products from lane 1. Panels below represent the intensity of the cleavage products from the indicated lanes, normalized with respect to lane 1.


Medium version | Full size version


Supplemental Figure 3. Detailed mapping of the RNA crosslink to the b subunit from TEC11, 12, or 13.

a) The mapping scheme. Horizontal bars represent the b polypeptide or its proteolytic fragments. Lettered boxes designate evolutionarily conserved regions (1). Hydroxylamine and trypsin cleavage sites are indicated by labelled arrows. Terminal residues of proteolytic fragments are numbered. At the bottom, the positions of lysine residues, which are the potential targets of crosslinking from reagent II, are indicated on the sequences of E. coli (Ecoli) and T. aquaticus (Theaq). The T. aquaticus sequence corresponding to the flap domain is indicated.

b) Partial trypsinolysis of the labeled wild-type enzyme (left) and Xho 19 deletion mutant (right). TEC11 containing Lys-specific crosslinking group (reagent II of Fig. 1a) on the RNA 5'-terminus was immobilized on Ni2+-NTA agarose. After treatment with NaBH4, the complexes were supplemented with the indicated amounts of trypsin (in TB) and incubated for 30 min at 20°. The mixture was supplemented with phenylmethylsulfonylfluoride (100 mg/ml) and incubated for another 10 min at room temperature. The cleavage products were analyzed as described (31). Large (Fr1) and small (Fr2) tryptic fragments are indicated.

c) Autoradiogram of 'single hit' hydroxylamine cleavage (31) products of the small tryptic fragment (Fr2) indicated in (b). Lanes 1, 2 - untreated and hydroxylamine-treated tryptic fragment (Fr2). Lane 3 - markers generated by limited CNBr degradation of the b' subunit in TEC11 labeled in the N-terminal region by the same crosslinking probe.


Medium version | Full size version


Supplemental Figure 4. Mapping of RNA-protein contacts in the N-terminal part of the b subunit.

a) Middle - CNBr degradation map of the E. coli RNAP b subunit. Met residues (CNBr cleavage sites) are marked as in Fig. 2a. Open circle and rectangle indicate the position of the mapped crosslinking sites.

Left - SDS-PAGE and autoradiography of degradation products of crosslinked b. Single-hit CNBr cleavage products for RNAP labeled from the 5'-end of RNA in TEC7-9 (lanes 2, 3) or MS4, 5, or 6 (lanes 4-6). Lane marked 'M' shows pattern of CNBr cleavage products of the b subunit labeled near the C-terminus at His1237 (9).

Right - Marker represents the N-terminal ladder of CNBr cleavage products of b labeled in the region Met130-239. Continuous and dashed lines are as described in Fig. 2a.

b) Quantitation of CNBr cleavage products. Top 2 panels represent the histograms of the intensity for the cleavage products from lane 1 (C-terminal reference) and lane 7 (N-terminal reference). Panel below represents the intensity of the cleavage products from lane 6, normalized with respect to lane 7. Bottom panel represents the intensity of the cleavage products from lane 2, normalized with respect to lanes 1 or 7 (whichever contains the appropriate reference band).

c) Sequence alignment of the segment of conserved region D of bsubunit where the crosslink in TEC7-9 and MS4, 5, 6 was mapped. Ecoli, Escherechia coli; Theaq, Thermus aquaticus; Bacsu, Bacillus subtilis; Tobac, Tobacco chloroplasts; Sulac, Sulfolobus acidocaldarius; Metth, Methanobacterium thermoautotrophicum; Sacce II, S. cerevisiae polymerase II; Human II, human polymerase II. The most conserved region of sequence is boxed. Rif I above the arrow indicates the beginning of cluster I of rifampicin resistant mutations (21).


Medium version | Full size version


Supplemental Figure 5. Mapping of RNA-protein crosslinks in the b' subunit.

a) Right - CNBr degradation map of the 1407-amino acid E. coli b' subunit. The horizontal bar at the top represents the 1407-amino acid b' polypeptide. Positions of Met residues are marked. Terminal residues of N-terminal cleavage products are numbered. Solid and open circles and solid rectangle indicate the positions of mapped crosslinking sites.

Left - SDS-PAGE and autoradioagraphy of degradation products of crosslinked b'. Lanes 4-14, limited CNBr cleavage products of the b' subunit labeled from TEC or MS complexes (indicated) by the indicated crosslinking reagent attached to the RNA 5'-terminus. The RNA length (nt) in each TEC or MS is indicated; lanes 1-3, CNBr degradation pattern of b' crosslinked from TEC21 containing a 3' terminal 4-thiouridine residue (4SU). Continuous and dashed lines are as described in Fig. 2a.

b) Quantitation of CNBr cleavage products. Top panel represents the histogram of the intensity for the cleavage products from lane 14. Panels below represent the intensity of the cleavage products from the indicated lanes, normalized with respect to lane 14.

c) Sequence alignments for the segment of b' conserved region C where crosslinks from MS6, 7, 8 and TEC7-9 were mapped. Ecoli, Escherechia coli; Theaq, Thermus aquaticus; Anasp, Anabaena sp.; Pinth, Pinus thunbergii; Sulac, Sulfolobus acidocaldarius; Metth, Methanobacterium thermoautotrophicum; Human II, human polymerase II. Segment which comprises the rudder in the T aquaticus core RNAP structure (6) is indicated. Lysine residues, which are the potential targets of crosslinking, are numbered. The most conserved region of sequence is boxed.


Medium version | Full size version


Supplemental Figure 6. DNA-protein contacts in the ternary elongation complex.

a) Structure of the photoreactive probe, 1-(4-azidotetrafluorophenyl)-2-bromoethanone (37), attached to internucleotide thiophosphates of the DNA. Circled letters indicate adjacent nucleotide residues in the DNA sequence.

b) The structure of nucleic acid scaffolds used for DNA-protein crosslinking. T, NT-template and nontemplate DNA strands, respectively. Solid circles mark the positions where reactive groups were introduced. The derivatization reaction was performed as described [A. N. Mayer, F. Barany Gene 153, 1, (1995)]. Modified oligonucleotide (10-20 pmol) was 5'-terminally labeled using T4 polynucleotide kinase with [g-32P]ATP and assembled in a scaffold as described in legend to Fig. 1d. After incubation with a 1.5 molar excess of His6-RNAP immobilized on Ni2+-NTA agarose for 10 min at 20°, the beads were washed with 1 ml of TB with 1 M NaCl, then with 1 ml of TB. The beads were suspended in 30-40 ml of TB and irradiated at 308 nm for 2 min. After crosslinking, the mixture was supplemented with 1% SDS and 200 pmol of DNA oligo complementary to the strand of the scaffold which did not contain the crosslinking group. The mixture was incubated for 2 min at 60° and analyzed by 4% SDS-PAGE.

c) Autoradiogram of labeled RNAP subunits crosslinked from the indicated DNA positions. Lanes 1-4, 5-7, 9-14, 8, 9-14, and 15 show results obtained in different experiments.


Medium version | Full size version


Supplemental Figure 7. Mapping of DNA contacts with the b' subunit in ternary elongation complexes.

a) Middle - CNBr degradation map of the RNAP b' subunit. Met residues (CNBr cleavage sites) are marked as in Fig. 2a. Open and solid circles and rectangles indicate the position of the mapped crosslinking sites.

Left and right - Lanes 1-6, N-terminal CNBr degradation patterns; lanes 8-13, C-terminal patterns. Lane 7 - reference pattern (M) for the crosslink from RNA 5'-terminus in TEC20.

b) Quantitation of CNBr N-terminal cleavage products. Top panel represents the histogram of the intensity for the cleavage products from lane 7. Panels below represent the intensity of the cleavage products from the indicated lanes, normalized with respect to lane 7.


Medium version | Full size version


Supplemental Figure 8. Mapping of DNA contacts with the b subunit in ternary elongation complexes.

Right - CNBr and N-Bromosuccinimide (NBS) degradation maps of the E. coli b subunit. Lines with vertical bars represent the cleavage maps with NBS and CNBr. Solid squares mark the positions of cross-linking sites. Cleavage products used for localization of the crosslinking sites are numbered as described above.

Left - Autoradiogram of an SDS gel showing 'single hit' NBS (top, lanes 1-6) and CNBr (bottom, lanes 7-9) degradation products. Top regions in lanes 1 and 6 are magnified.


Medium version | Full size version


Supplemental Figure 9. Correspondence of TEC model with crosslinking data.

Three views of the TEC model showing the correspondence between the model and the crosslinking data mapped onto the RNAP crystal structure. The model consists of three components: 1) the Taq core RNAP crystal structure, shown as a molecular surface , 2) the DNA template (template strand, red; non-template strand, yellow), and 3) the RNA transcript (gold) and incoming nucleotide substrate (green). The nucleic acid backbones are shown as worms with CPK phosphate atoms. For base-paired sections of the nucleic acids, the bases are shown as sticks. For clarity, not all of the crosslinking results are mapped onto the structure. In these views, the RNAP structure is shown as a white molecular surface but with regions to which crosslinks were mapped color-coded (unless color-coded, protein surfaces that have been sliced away for clarity are colored grey). Phosphate atoms for which crosslinks were not tested are colored according to the standard scheme, otherwise the phosphate atoms are colored to match the protein fragment the crosslink was mapped to. For example, the phosphate at +15 on the non-template DNA strand (magenta in a), modified with the non-specific reagent, crosslinks to b'Ec1095-1189 (Ta1250-1305; also colored magenta in a). Residues for which specific crosslinks were mapped from the Lys-specific reagent are colored red.

a) View showing the downstream duplex DNA from about +3 to +17, with crosslinks mapped from specific phosphates to b' fragments color-coded and labeled. The reactive groups attached to the backbone of the DNA template strand at positions +6 and +9/+11 crosslink to b'Ec102-130 (Ta90-118; orange) and Ec1189-1260 (Ta1305-1376; pink), respectively. The reactive group attached to the backbone of the DNA nontemplate strand at positions +5 and +15 crosslinks to b'Ec1095-1189 (Ta1250-1305; magenta). These regions of b' form the side walls of the 'trough' that accommodate the downstream DNA. The phosphates at +8 and +9 (labeled) also crosslink to bEc184-239 (Ta172-207), which is directly between the DNA and the viewer but is sliced away to reveal the DNA and the b' trough accomodating it.

b) View showing the RNA/DNA hybrid from +1 to about -8 and corresponding crosslinks. A 3'-terminal thiouridine occupying the i+1 site (+1, colored green) crosslinks to b'DEc400-466 (Ta676-745), also colored green. This region includes the invariant motif -NADFDGD- responsible for chelating the active site Mg2+ ion (magenta sphere). The RNA phosphates at -3 and -4 (light green) crosslink to bRifr (Ec492-515 or Ta372-395). The RNA phosphate at -8 (fuschia), crosslinks to the b'C rudder. The RNA phosphates from -5 to -7 are colored light green and fuschia since they each crosslink to two separate protein fragments, bRifr and the b'C rudder. When modified by the Lys-specific reagent, positions -6 to -8 crosslink to b'EcK325 (TaR601, red).

c) View approximately along the axis of the upstrema duplex DNA, with obscuring parts of the upstream DNA removed, and the b'C rudder and its coiled-coil platform shown as a backbone worm. The RNA phosphate at -9 (medium blue) crosslinks to two protein regions, a segment of bI (also medium blue), and the bG flap (cyan). When modified with the Lys-specific reagent, RNA phosphates from -9 to -12 crosslink to two Lys residues on the underside of the bG flap (and therefore not visible), bEcK890 (TaK762; just visible in e) and bEcK914 (TaK786). The RNA phosphate at -13 (dark blue) crosslinks to the C-terminal b fragment also colored dark blue. When modified with the Lys-specific reagent, the crosslink maps to bEcK1306 (TaQ1068), shown in red.


Medium version | Full size version


To Advertise     Find Products


Science. ISSN 0036-8075 (print), 1095-9203 (online)