Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Mar 4, 2008; 105(9): 3315–3320.
Published online Feb 25, 2008. doi:  10.1073/pnas.0712074105
PMCID: PMC2265150

Monitoring RNA transcription in real time by using surface plasmon resonance


The decision to elongate or terminate the RNA chain at specific DNA template positions during transcription is kinetically regulated, but the methods used to measure the rates of these processes have not been sufficiently quantitative to permit detailed mechanistic analysis of the steps involved. Here, we use surface plasmon resonance (SPR) technology to monitor RNA transcription by Escherichia coli RNA polymerase (RNAP) in solution and in real time. We show that binding of RNAP to immobilized DNA templates to form active initiation or elongation complexes can be resolved and monitored by this method, and that changes during transcription that involve the gain or loss of bound mass, including the release of the sigma factor during the initiation–elongation transition, the synthesis of the RNA transcript, and the release of core RNAP and nascent RNA at intrinsic terminators, can all be observed. The SPR method also permits the discrimination of released termination products from paused and other intermediate complexes at terminators. We have used this approach to show that the rate constant for transcript release at intrinsic terminators tR2 and tR′ is ≈2–3 s−1 and that the extent of release at these terminators is consistent with known termination efficiencies. Simulation techniques have been used to fit the measured parameters to a simple kinetic model of transcription and the implications of these results for transcriptional regulation are discussed.

Keywords: intrinsic termination, kinetic modeling, RNA polymerase binding, elongation complex dissociation, sigma release

RNA transcription is a central event in gene expression and is tightly controlled during all of the phases of synthesis of the full-length transcript, including initiation, elongation, and termination. Misregulation during any of these processes can result in aberrant gene expression, compromise survival of single-celled creatures, and lead to disease in higher organisms. Although termination is required to stop transcript elongation at the ends of genes, this process also plays important regulatory roles at intermediate template positions during transcription elongation (14). The transcription of many essential genes across all phyla, including humans, is controlled in the postinitiation phase by termination events that are mediated by signals in the nascent RNA or DNA and are often further modulated by protein cofactors (5, 6). The fundamental structure of the nucleic acid scaffold that constitutes the core of the transcription elongation complex (TEC) is conserved in all organisms. Hence, the well described Escherichia coli TEC serves as an excellent model system to study basic aspects of transcription termination and its regulation.

Pathways that can potentially compete with elongation at each template position during transcription include pyrophosphorolysis (the chemical reverse of the NTP addition process), entrance into arrested or editing states, and termination (4, 710). Sequence-specific pausing of the TEC can also occur, often as a prelude to entering into other potential reaction pathways. The probability of a TEC venturing down any particular competing pathway at a given template position depends on the rate of that process relative to the rates of the other available pathways at that position (7, 1012). These rates can be formulated in terms of transition state barriers that control the entrance of the TEC to each pathway, with the relative heights of these competing barriers being regulated by elements of local nucleic acid sequence and protein transcription factors (4, 7).

To understand the mechanistic details and control of these alternative pathways, it is necessary to determine their relative rates under various regulatory conditions (4). Previous studies of transcription in vivo and in vitro, using both solution and single-molecule assays, have estimated elongation rates of 10–40 nt·s−1 (1, 8, 1320). However, all of these techniques have significant limitations for studying termination. The surface plasmon resonance (SPR) method we use here has permitted us to make real-time and comparative observations of RNA polymerase (RNAP) dissociation and transcript release rates at elongation and termination positions.

The stability and processivity of TECs is controlled by interactions of the RNAP with the nucleic acid scaffold as the complexes move along the template DNA. The scaffold includes the double-stranded DNA that brackets RNAP upstream and downstream of the TEC, the unwound template and nontemplate DNA of the open transcription bubble, and the 8- to 9-bp RNA-DNA hybrid that forms between the template DNA and the nascent RNA transcript within the bubble (supporting information (SI) Fig. 6) (21). Pausing and termination have been studied most for the TECs of E. coli, where termination is either intrinsic or Rho-dependent (4). Both types of termination are subject to modulation by additional protein factors that control the actual termination efficiency (TE) at any given intrinsic or Rho-dependent terminator.

In this study we focus on intrinsic termination, which typically depends on the appearance, within the moving transcription bubble, of a particularly weak RNA-DNA hybrid that consists primarily of rU-dA base pairs and is located immediately downstream of a palindromic DNA sequence that codes for a RNA hairpin with a G-C-rich (stable) stem. Intrinsic termination occurs when transcription has proceeded far enough along the template to synthesize the RNA of the terminator hairpin, followed by ≈8 rU residues. In this process the relatively weak rU-dA hybrid is further destabilized from the upstream end by melting or distortion induced by the terminator hairpin (4, 2224), driving release of the core RNAP and the nascent RNA chain and thus site-specific termination (4).

Although many of the rearrangements of the transcription complex involved in transcript initiation, elongation, and termination have been well studied, estimates of termination rates have largely been indirect. In earlier studies measurements of TEC dissociation rates at terminators has been limited by the time resolution of the methods involved, and the difficulty of discriminating paused from released complexes (16, 22, 23, 2527). Here, we describe the application of SPR, which has been used to monitor some aspects of transcription elongation (28), to the quantitative analysis of the events of initiation, elongation, and intrinsic termination. We have used a simple kinetic competition model, together with simulation methods, to begin to test the “fit” of measured kinetic parameters to the proposed model and to set the stage for the determination of additional kinetic parameters that define the control of the elongation–termination switch by regulatory transcription factors.


RNA Polymerase Binding Interactions with the DNA Template Can Be Monitored by SPR.

RNA transcription begins with promoter binding by the holoenzyme (+ sigma) and formation of the initiation complex. We have used the Biacore X SPR system to monitor the time dependence of the interactions of E. coli RNAP with template DNA molecules containing different promoters and terminators.

DNA templates −tR (T7A1 promoter, no terminator), tR2 (T7A1 promoter), or tR′ (λ pR promoter) (Fig. 1) were immobilized on the Biacore chip surface and the binding of RNAP to these DNA templates was analyzed at high (50 μl/min) flow rates to reduce mass transport effects in the Biacore cell (see Materials and Methods; SI Text). A large increase in the surface index of refraction (SIR) signal [≈500–3,000 resonance units (RU)] was observed as holo-RNAP (≈450 kDa) was added to the DNA bound to the dextran-coated surface of the flow cell (see SI Fig. 8). The observed binding of the RNAP to DNA was multiphasic, with an initial fast rate followed by slower binding events (28). This suggests that more than one type of binding is being observed, consistent with the involvement of the nonspecific binding of E. coli RNAP to nonpromoter DNA sequences (28, 29). Similar signals were also observed for dissociation events, with ≈90% of the RNAP remaining bound to the DNA and ≈10% dissociating slowly (likely from nonspecific binding sites) for a period of 45 min under constant flow (SI Fig. 8A). Because SPR is a nonequilibrium measurement, these slow-dissociation events never quite reach equilibrium and the observed dissociation curves continue to drift downward slowly with time (SI Fig. 8B). Dissociation of RNAP is clearly visible in the SPR sensograms as a drop-in signal during either the dissociation phase of the experiment or during the regeneration of the DNA-bound surface with 1 M NaCl. These results suggest that the SPR method as used here has significant potential for quantitative monitoring of the dissociation of transcription complexes at terminator sites.

Fig. 1.
Transcription templates and intrinsic terminators. (A) The transcription templates used, indicating promoter type, the positions of the stalled complex (gray oval) and termination sequences (gray box), and the initiation, termination, and runoff positions ...

The Initiation to Elongation Transition and the Formation of Stable (Stalled) Elongation Complexes.

After binding RNAP to the promoter, transcription was initiated with three NTPs (I-mix) and continued to the stall site (the first template position at which UTP is required) to form stable or “stalled” TECs (Fig. 1A). The SIR signal decreases by ≈120 RU during the transition from initiation to elongation (Fig. 2A; SI Fig. 8A, +I-mix), representing a loss of ≈20% of the total mass of the final stable elongation complex, with the remaining mass (≈500 RU, Fig. 2B; SI Fig. 8A, +250 mM NaCl) corresponding to the stalled TECs after inactive initiation complexes and nonspecifically bound RNAP had been removed by salt challenge. This change in mass is consistent with the dissociation of the sigma subunit from the initiation complex during transition to the elongation phase (30), whereas the remaining ≈500 RU correspond to the mass of the core subunits of the stable TEC. Further analysis to conclusively identify the dissociating subunit was made difficult by its low concentration in the eluted sample. However, preliminary mass spectrometry analysis of the sample eluted during stalled TEC formation revealed a peak of ≈90 kDa, consistent with the mass of sigma70. This peak was not detected in control experiments that monitored the slow release of nonspecifically bound holo-RNAP at 4,000 s after injection; in these samples, sigma should remain bound to the core RNAP.

Fig. 2.
Typical sensograms showing transcript initiation, elongation, and termination. (A) Dissociation events during the initiation to elongation transition in the presence of initiation mix to form stalled complexes. (B) Comparison of the dissociation rates ...

TECs initially “synchronized” at the stall position could be further elongated (chased) to a termination site or to the end of the dsDNA template by adding all 4 NTPs. These site-specifically positioned, stalled TECs were extremely stable, with a half-life of ≈3 h at 25°C (unpublished data), and were resistant to challenge with 250 mM salt (9, 29, 31, 32). The dissociation rates for the stalled complexes were clearly slower than those for the initiation complexes, as evidenced by the dissociation curve triggered by a “salt pulse” of 1M NaCl (Fig. 2B). These stalled TECs also showed a higher final baseline (−3,200 RU) than did the initiation complexes; the latter dissociated almost immediately to a final baseline level of −3,700 RU (Fig. 2B). This shows that dissociation of stalled complexes was incomplete over the observation time of Fig. 2A.

In addition to TECs bound at the stall site, the SPR cell also likely contains some holo-RNAP molecules bound nonspecifically to the DNA (29). These nonspecific complexes also dissociated “instantly” (Fig. 2B) when pulsed with 1 M NaCl. Control experiments demonstrated that the stalled TECs, as expected, were much more stable than RNAPs bound either at promoters or at nonspecific DNA sites. These results show that the SPR system can be used to monitor the formation of initiation complexes at promoters on immobilized DNA templates, to follow the release of sigma during the initiation–elongation transition, and to observe the formation of functional stalled TECs at defined template sites.

Transcript Elongation Can Be Monitored by SPR.

A DNA template containing the T7A1 promoter and no terminator or significant pause site (150 μM −tR; Fig. 1 A and B) was used to show that the transcript elongation process itself can be monitored by SPR. An increase of 50 to 100 RU was observed when comparing the injection of chase mix (10 μl/min) over the preformed stalled TECs with the effect of a comparable injection containing 20 mM EDTA to prevent RNA synthesis (SI Fig. 9). This signal likely represents the increase in mass due to RNA synthesis and was confirmed by SDS/PAGE analysis (SI Fig. 11A).

Analysis of these data revealed an apparent rate constant for elongation (kFapp) of ≈0.04 s−1, corresponding to an apparent synthesis rate of RNA of ≈4 nt·s−1 per elongation complex (see SI Text). This rate is slower than rates reported for traditional solution or single-molecule reactions (1, 8, 1315, 1720), suggesting that the apparent rate measured here may be reduced by mass transport effects (33). These effects were reduced as described (SI Text) yielding synthesis of ≈70 nt in 7 s (data not shown). This corresponds to a lower limit for kFapp of ≈10 nts, comparable to transcription rates reported in gel assays and in single-molecule experiments (15, 17, 19, 20). Experiments at such low bound DNA concentrations were difficult, because of problems in regenerating the chip surface (see SI Text). Nevertheless, we can conclude that template-directed RNA transcription over ≈70 bp can be measured directly by this technique (SI Fig. 9).

SPR Can Be Used to Monitor Intrinsic Termination.

Stalled TECs chased with 4 NTPs and 500 μg/ml heparin were used (SI Fig. 8B) to track dissociation of active TECs (RNAP and nascent RNA; see SI Text) from templates containing two intrinsic terminators, tR2 and tR′ (Fig. 1). We used salt displacement techniques (see SI Text) to minimize the SIR contributions of nonspecifically bound RNAP molecules in these experiments; the absence of such selective washing procedures may have been responsible for the difficulties experienced by Pemberton and coworkers in measuring RNA chain extension in an earlier study of transcription by SPR (28, 30). The SIR signals for both templates decreased significantly on resumption of transcription from stalled TECs (+25 and +16) with the chase mix. The signal decrease averaged 40 RU for tR2 and 70 RU for tR′, showing that a fraction of the TECs had dissociated from both templates (Fig. 3). Because TECs at other positions were very stable and remained bound to the DNA under these conditions (see SI Text), this decrease in signal must have arisen solely from TEC dissociation at terminators. Because the molecular weight of RNAP is much greater than the nascent RNA, >80% of this signal must be caused by release of RNAP.

Fig. 3.
SPR analysis of termination at tR2 and tR′. Sensograms of transcription termination from SA chips with 10 μM DNA template bound at 50 μl/min monitoring dissociation from tR2 (solid line) and tR′ (dashed line) after injection ...

Consistent with the observed TE values of 0.4 and 0.7 measured for the tR2 and tR′ terminators in bulk-solution assays (8, 34), the amount of complex released from the tR2 template was always less than that released from tR′. Assuming formation of comparable amounts of elongation complex on the two templates, the ratio of the signal lost at the two terminators is very close to 1:2 (see SI Text), as expected from the relative TE values of the two terminators (0.4 and 0.7). The dissociation from the tR′ template was slightly delayed compared with tR2 and is concordant with the longer sequence between the stall and terminator positions for tR′ (see Fig. 1; SI Text). This template also contained some defined pauses between the stall site and the terminator, which should also contribute to the apparent delay in reaching the tR′ terminator (Fig. 3). Nascent RNA transcribed from these templates were collected and analyzed as described (SI Fig. 11B). These experiments demonstrate that it is possible to observe and quantify the dissociation of TECs at termination positions in real time.

A Model for Transcription Regulation.

The elongation and termination pathways available to TECs at terminator positions are in kinetic competition with one another. It is likely that pause signals are involved in regulating these competing pathways (79, 16), with a decrease in pause efficiency and duration favoring elongation and an increase in these variables favoring termination. Fig. 4 shows a minimal kinetic model that is based on these concepts and can be used to simulate the reaction and to investigate the effects of altering the rate constants of either pathway on TE. Such simulation can also be used to determine how accurately the model fits this, and previous, data.

Fig. 4.
Minimal kinetic model for transcript elongation and intrinsic termination. A TEC with the next required NTP bound (A; TECn−1) undergoes nucleotide addition (rate constant kF) extending the nascent RNA by 1 nt and releasing PPi (B; TECn). The reverse ...

The elongation–termination switch at each template position is controlled by the probability that the sequence of steps leading to termination is either more or less likely than those leading to elongation (see below). It is reasonable to assume that termination requires the TEC to enter an initial paused state (Fig. 4D), followed by dissociation into its individual components (Fig. 4E). The transcription complex is thought to cycle rapidly between pretranslocated and translocated states at the active site when it is in its fully “elongation-competent” form (Fig. 4 A–C). Binding of the next templated NTP can only occur when the transcription complex is in the translocated state in which the 3′ end of the nascent RNA is in the product binding subsite of the RNAP, thus opening the substrate subsite to the next required NTP (Fig. 4B).

Nucleotide addition extends the transcript by 1 nt and allows translocation of the TEC downstream by 1 bp (Fig. 4C) (4). The rapid interconversion between the translocated and pretranslocated states is inhibited in the paused state at either pausing or terminator sites (24, 27, 35, 36), thus increasing the probability that the transcription complex will enter other reaction pathways (7, 8, 16, 34). A TEC poised at either a pause or a terminator site (Fig. 4B) may either pause (Fig. 4D) or undergo elongation to the next NTP position (Fig. 4C), with the relative probability of either event being determined by the relative rates for entering the paused state from the elongation state (kpause; Fig. 4 B–D) and for elongation to the next NTP position (kF; Fig. 4 B and C), and by the concentration of the next required (n + 1) NTP. Similarly, the relative probabilities of the paused TEC returning to the elongation-competent state (Fig. 4B) or progressing to the terminated state (Fig. 4E) are determined by the relative rates of escape from the paused state to the elongation-competent (kPE; Fig. 4 DB), or to the terminated state (krelease; Fig. 4 D and E; SI Text).

Simulations and Data Analysis.

Simulations of the termination process at the tR2 and the tR′ terminators were performed by using the Berkeley Madonna program and the simple model (Fig. 4 B–E; SI Text). The parameters used include the rate constants for each nucleotide addition step (kF; Fig. 4 A–C), entry into the paused state (kpause), escape to the elongation mode from the pause at the terminator (kPE), and termination (krelease). The simulated reactions contained excess concentrations of NTPs, stalled complex concentrations were normalized to unity, and templates were taken as being without pause sites between the stall position and the terminators (see Fig. 1A). Under these conditions an average transcription rate of 20 nt·s−1 was assumed at all template positions (except the terminator). This is the upper limit of the elongation rate determined here (Fig. 3; see below) and is the most commonly reported value for bulk-solution experiments (see SI Text) (1, 8, 13, 14).

By using these inputs we have analyzed our dissociation data and attempted to fit them to the simple quantitative model for the elongation-termination switch shown in Fig. 4. As pointed out earlier, the dissociation curves for TEC release at both terminators are multiphasic (Fig. 3), presumably reflecting the spreading of the statistically distributed population of TECs along the template before arrival at the terminator. Such statistical spreading for TECs transcribing along the tR′-containing template has been demonstrated (8). The observed kpause for the tR′ template (TE = 0.7) was twofold larger than for the tR2 template (TE = 0.4; SI Table 1; SI Fig. 7), and kPE was 10-fold smaller. Because the tR′ template is 170 bp longer between the stall position and the terminator (Fig. 1), the simulated rate profile for the terminated products of tR′ lags behind that for tR2 (SI Fig. 7), as also seen experimentally in the SPR data (Fig. 3; SI Text).

The first portions of the SPR dissociation curves obtained for each terminator, representing the leading edge of the statistically spread TECs in each case, were subjected to pseudo-first-order analysis. This resulted in apparent values of kdissapp of ≈2–3 s−1 for release from both terminators. Because dissociation of the TECs was apparent in the SPR experiments with both templates within 10 s after the start of the injection, it is reasonable to deduce an upper limit for the kFapp of at least 10–20 nt·s−1 from these data. This kFapp is comparable to rates observed directly at low DNA density on the chip surface for the −tR template in this study, and in previous single-molecule (15, 1720) and bulk-solution studies (1, 8, 13, 14). The simple model described above was altered to include a pause signal between the stall and termination sites and used to fit the normalized SPR data from the tR2 template (see Materials and Methods; SI Text; Fig. 5). This fit was satisfactory (rms deviation of 0.0436 with small random residuals) and yielded kpause, kPE, and krelease values of 17.5, 0.02, and 0.26 s−1, respectively (Fig. 5). The value for kpause is equivalent to that determined previously, but both krelease and kPE were changed 10-fold compared with the values reported in earlier studies (16, 27, 35).

Fig. 5.
Fitting the model to the tR2 SPR data. The Berkeley Madonna Program was used to fit (solid curve) the measured dissociation curves (circles, only data points at 1- to 5-s intervals are displayed) obtained with the tR2 template (Fig. 3) to the minimal ...

These findings permit us to conclude that the SPR method as developed here can be used to monitor the dissociation of TECs from DNA templates in real time, that the measured rates of transcription and dissociation are consistent with the template sizes and TE values used in the experiments, and that the model we have used contains the minimal number of parameters and states needed to fit the data. Finally, we note that the simulated kinetic model described above can be easily adapted to fit the SPR dissociation data directly, and that these analyses can be further refined by using a modification of the two-compartment model (33) to permit mass transport issues in the SPR experiments to be handled computationally.


SPR has been used to provide real-time observations of the kinetics of TEC dissociation at intrinsic terminators under bulk-solution conditions. This experimental approach can contribute significantly to our mechanistic understanding of the regulation of transcription. The binding of RNAP to template DNA at various loci (promoters, nonspecific binding sites, stall positions, and terminators), and its dissociation, is shown to be clearly detectable and measurable by using SPR technology (Fig. 3 and SI Fig. 8). We have followed single rounds of template-directed RNA synthesis by using immobilized DNA templates under flow (Fig. 3 and SI Fig. 9), have observed the release of sigma factor from the initiation complex during the initiation–elongation transition (Fig. 2A), and have measured the actual dissociation rates of TECs under salt-challenge conditions and at intrinsic terminators (Figs. 2B and and33).

Although single-molecule and SPR based techniques have been successfully used to detect nascent RNA transcription in real time, both have limitations when applied to termination (16, 28). The single-molecule data obtained by Yin and coworkers were fitted to a simple model that yielded a krelease of 0.016 s−1 (16), slower than the kdissapp and krelease values of 2–3 and 0.3 s−1 estimated for these parameters here (see SI Text; Figs. 3 and and5).5). This discrepancy may be due to immobilization of RNAP to the surface used in ref. 16 or to differences in the terminators used. The slower kPE of 0.02 s−1 observed for tR2 here, compared with previous work on the tR2 polyU sequence (0.25 s−1) (27, 35) suggest that the hairpin and weak-hybrid portions of the terminator act synergistically to pause the TEC at this position as suggested (24, 34, 37, 38).

The local concentration of reactants in SPR, particularly at high DNA template concentrations, is known to be affected by mass transport (SI Fig. 10) (33). Therefore, the lower limit for kF is probably higher than the 4–10 nt·s−1 observed in this study (SI Fig. 8). The observed statistical spreading of the TEC population over the template between the stall site and the terminator or runoff positions may also contribute to this underestimate for −tR template (see SI Text) (8). Further calibration of the system will include determination of the limit of resolution for the detection of RNA synthesis, and these data will be fitted to models that include mass transport corrections directly (see SI Text).

Intrinsic terminators constitute an RNA hairpin adjacent to a weak RNA-DNA hybrid, and termination is thought to involve pausing of the TEC in concert with other nucleic acid scaffold rearrangements (hairpin folding, hybrid melting, and bubble collapse) to form a preterminated complex (Fig. 4 F–H) (24, 27; reviewed in ref. 4). The rate at which these sequential events occur is likely to depend on the nucleic acid sequence of the hybrid, the downstream DNA, and the adjacent upstream RNA hairpin. Thus, a terminator may be characterized in terms of the thermodynamic stability changes induced by these hairpin and hybrid components (16, 21, 24, 34).

Our simulations have revealed that the TE values for tR′ (0.7) and tR2 (0.4) are sensitive to changes in the values used for kpause and kPE, whereas the apparent rates of release from both terminators are not (SI Table 1, SI Fig. 7, and SI Text). Because the weak-hybrid sequences are almost identical for the tR2 and tR′ terminators, whereas the hairpins differ significantly (Fig. 1 C and D), these simulations suggest that the longer, more thermodynamically stable hairpin of tR′ may contribute more to increasing the efficiency and half-life of the pause at the terminator than to increasing krelease itself. Given that other studies have shown that the hairpin and weak-hybrid elements of an intrinsic terminator each act to pause the TEC (24, 34, 37, 38), it is likely that these elements function synergistically to increase pause efficiency and pause duration at terminators. Conversely, krelease may be influenced by the thermodynamic stability of the hybrid sequence of the intrinsic terminators (7, 37). Finally, we note that antitermination complexes reduce terminator efficiency in response to environmental signals and are likely to function by decreasing kpause, increasing kPE, and/or by reducing krelease.

In conclusion, this study has shown that SPR technology is a powerful tool that can be used to explore the relationship between pause half-life (rate of pause escape), the rate of preterminated complex formation (induced by hairpin folding or Rho-helicase activity for intrinsic or Rho-dependent terminators, respectively), and the overall rate of TEC dissociation. In combination with simulations of transcription activity, SPR transcription termination data should be useable to quantitate changes in activation free-energy barrier heights at these regulatory positions and the effects on these changes of transcription cofactors, and thus to further elucidate the relative probabilities of alternative reaction pathways under different transcription conditions (Fig. 4) (4, 12).

Materials and Methods

Buffers and Reagents.

DNA oligonucleotides, including those with 5′-biotin modifications, were purchased from Integrated DNA Technologies and subjected to PCR amplification with Vent DNA polymerase (New England Biolabs) to create the DNA templates used in this study. Templates for the SPR transcription reactions were amplified by using 5′-biotinylated upstream primers, whereas those for the bulk-solution reactions were created by using unmodified primers. Heparin, dinucleotide ApU, and high-purity ribonucleotide sets for transcription were purchased from Calbiochem/EMD, Oligos Etc., and Amersham Biosciences, respectively, whereas E. coli RNAP holoenzyme was from Epicentre Biotechnologies. SPR reagents, such as the streptavidin (SA) SPR chips and P20 surfactant were from Biacore Lifesciences. Single-round SPR transcription experiments were conducted by using a Biacore X SPR instrument and running buffer] 20 mM Hepes, pH 7.8, 100 mM KAc, 5 mM Mg(Ac)2, and 0.05% surfactant P20] or pre-SDS wash buffer (20 mM Hepes, pH 7.8) to reduce the possibility of SDS-induced precipitation of salt in the integrated microfluidics cartridge (IFC) during regeneration cycles.

SPR Monitoring of Transcription Reactions.

DNA templates used were −tR (T7A1 promoter, no terminator), tR2 (T7A1 promoter, λ tR2 terminator), and tR′ (λ pR promoter, λ tR′ terminator) (Fig. 1). All SPR reactions were performed at 30°C. SA chips were prepared and injections were conducted as suggested in the Biacore manual. Biotinylated DNA was bound to the surface of the flow cell #1 (at a flow rate of running buffer of 5 μl/min) to a density, unless otherwise stated, of 10 μM [assuming that 1,000 RU represent 0.78 ng/μl of DNA (28)]. Unbound DNA was washed off with 1 M NaCl and signal quality was maintained by running weekly instrumental maintenance protocols and by washing the chip surfaces regularly with running buffer. Unless otherwise stated, flow rates were 50 μl/min and data were collected at “medium” (1 data point per s) or “high” (5 data points per s) rates.

RNAP holoenzyme, diluted into 80 μl of running buffer at a concentration of 50–100 nM, was injected across both the control (no DNA) and the sample cell at flow rates of 10 or 50 μl/min. Two minutes of dissociation were allowed before completion of the injection cycle. The RNAP-bound DNA surface was regenerated either by injecting 100 μl of 1 M NaCl or by use in a transcription experiment. Stalled TECs were prepared by injecting 80 μl of initiation mix (I-mix; 0.5 mM ApU and 0.1 mM each rATP, rCTP, and rGTP in running buffer) across both flow cells, followed by removal of nonspecifically bound RNAP and initiation complexes by injections of 250 mM NaCl. Transcription at stall positions was restarted by injection of “chase mix” (1 mM each rUTP, rATP, rCTP, and rGTP; 500 μg/ml heparin) in running buffer. The chips were regenerated by injection of 1 M NaCl or by washing with pre-SDS wash buffer, followed by a 10 s pulse of 0.05% SDS in wash buffer and further washing in pre-SDS wash buffer.

RNA transcription products were eluted from the chip surface, as described in the Biacore manual, during injection of the chase mix for templates tR2 and tR′, and during regeneration with SDS for the −tR template. Eluted RNA was dephosphorylated with Antarctic phosphatase (New England Biolabs), radiolabeled with [γ-32P]ATP by using T4 polynucleotide kinase (Amersham Biosciences), and analyzed by denaturing 5% polyacrylamide (1:19 acrylamide/bisacrylamide), 8 M urea 1× Tris-borate-EDTA gel electrophoresis. Transcription assays in bulk solution were performed as described in ref. 8.

Data Analysis.

Experimental rate data were normalized to the baseline signal by using the BIAnalysis software, followed by analysis assuming initial pseudo-first-order kinetics or by simulation with the Berkeley Madonna program (University of California, Berkeley; see SI Text) to fit the data to the minimal kinetic model described in Fig. 4. SPR data were normalized to unity for fitting purposes, assuming TEs equal to those previously determined in bulk-solution assays for each terminator used (8, 9). These data were then fit to the inverse rate of appearance of the terminated products (also normalized to unity; see SI Text).

Supplementary Material

Supporting Information:


We thank our laboratory colleagues, Diane Hawley and others, for helpful discussions. This work was supported in part by National Institutes of Health Grant GM-15792 (to P.H.v.H.). S.J.G. was a Postdoctoral Fellow (AHA-0425713Z) of the American Heart Association, J.P.G. was supported by National Institutes of Health Training Grant GM-07759, L.J.M. is a Murdock Foundation Partner in Science, and P.H.v.H. is an American Cancer Society Research Professor of Chemistry.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0712074105/DC1.


1. Uptain SM, Kane CM, Chamberlin MJ. Basic mechanisms of transcript elongation and its regulation. Annu Rev Biochem. 1997;66:117–172. [PubMed]
2. Conaway JW, Conaway RC. Transcription elongation and human disease. Annu Rev Biochem. 1999;68:301–319. [PubMed]
3. Condon C, Squires C, Squires CL. Control of rRNA transcription in Escherichia coli. Microbiol Rev. 1995;59:623–645. [PMC free article] [PubMed]
4. Greive SJ, von Hippel PH. Thinking quantitatively about transcriptional regulation. Nat Rev Mol Cell Biol. 2005;6:221–232. [PubMed]
5. Keene RG, Mueller A, Landick R, London L. Transcriptional pause, arrest and termination sites for RNA polymerase II in mammalian N- and c-myc genes. Nucleic Acids Res. 1999;27:3173–3182. [PMC free article] [PubMed]
6. Gromak N, West S, Proudfoot NJ. Pause sites promote transcriptional termination of mammalian RNA polymerase II. Mol Cell Biol. 2006;26:3986–3996. [PMC free article] [PubMed]
7. von Hippel PH, Yager TD. Transcript elongation and termination are competitive kinetic processes. Proc Natl Acad Sci USA. 1991;88:2307–2311. [PMC free article] [PubMed]
8. Rees W, Weitzel S, Das A, von Hippel P. Regulation of the elongation-termination decision at intrinsic terminators by antitermination protein N of phage (lambda). J Mol Biol. 1997;273:17. [PubMed]
9. Wilson K, von Hippel P. Stability of Escherichia coli transcription complexes near an intrinsic terminator. J Mol Biol. 1994;244:36–51. [PubMed]
10. von Hippel PH, Pasman Z. Reaction pathways in transcript elongation. Biophys Chem. 2002;101–102:401–423. [PubMed]
11. McDowell JC, Roberts JW, Jin DJ, Gross C. Determination of intrinsic transcription termination efficiency by RNA polymerase elongation rate. Science. 1994;266:822–825. [PubMed]
12. von Hippel PH, Rees WA, Rippe K, Wilson KS. Specificity mechanisms in the control of transcription. Biophys Chem. 1996;59:231–246. [PubMed]
13. Chamberlin MJ, Nierman WC, Wiggs J, Neff N. A quantitative assay for bacterial RNA polymerases. J Biol Chem. 1979;254:10061–10069. [PubMed]
14. Kassavetis GA, Chamberlin MJ. Pausing and termination of transcription within the early region of bacteriophage T7 DNA in vitro. J Biol Chem. 1981;256:2777–2786. [PubMed]
15. Schafer DA, Gelles J, Sheetz MP, Landick R. Transcription by single molecules of RNA polymerase observed by light microscopy. Nature. 1991;352:444–448. [PubMed]
16. Yin H, Artsimovitch I, Landick R, Gelles J. Nonequilibrium mechanism of transcription termination from observations of single RNA polymerase molecules. Proc Natl Acad Sci USA. 1999;96:13124–13129. [PMC free article] [PubMed]
17. Davenport RJ, Wuite GJ, Landick R, Bustamante C. Single-molecule study of transcriptional pausing and arrest by E. coli RNA polymerase. Science. 2000;287:2497–2500. [PubMed]
18. Epshtein V, Nudler E. Cooperation between RNA polymerase molecules in transcription elongation. Science. 2003;300:801–805. [PubMed]
19. Neuman KC, Abbondanzieri EA, Landick R, Gelles J, Block SM. Ubiquitous transcriptional pausing is independent of RNA polymerase backtracking. Cell. 2003;115:437–447. [PubMed]
20. Shaevitz JW, Abbondanzieri EA, Landick R, Block SM. Backtracking by single RNA polymerase molecules observed at near-base-pair resolution. Nature. 2003;426:684–687. [PMC free article] [PubMed]
21. Yager TD, von Hippel PH. A thermodynamic analysis of RNA transcript elongation and termination in Escherichia coli. Biochemistry. 1991;30:1097–1118. [PubMed]
22. Kashlev M, Komissarova N. Transcription termination: Primary intermediates and secondary adducts. J Biol Chem. 2002;277:14501–14508. [PubMed]
23. Komissarova N, Becker J, Solter S, Kireeva M, Kashlev M. Shortening of RNA:DNA hybrid in the elongation complex of RNA polymerase is a prerequisite for transcription termination. Mol Cell. 2002;10:1151–1162. [PubMed]
24. Datta K, von Hippel PH. Direct spectroscopic study of reconstituted transcription complexes reveals that intrinsic termination is driven primarily by thermodynamic destabilization of the nucleic acid framework. J Biol Chem. 2008;283:3537–3549. [PMC free article] [PubMed]
25. Arndt KM, Chamberlin MJ. Transcription termination in Escherichia coli. Measurement of the rate of enzyme release from Rho-independent terminators. J Mol Biol. 1988;202:271–285. [PubMed]
26. Jin DJ, Burgess RR, Richardson JP, Gross CA. Termination efficiency at rho-dependent terminators depends on kinetic coupling between RNA polymerase and rho. Proc Natl Acad Sci USA. 1992;89:1453–1457. [PMC free article] [PubMed]
27. Gusarov I, Nudler E. The mechanism of intrinsic transcription termination. Mol Cell. 1999;3:495–504. [PubMed]
28. Pemberton IK, Buckle M. Real time in vitro analysis of transcription by RNA polymerase on immobilized DNA fibres. J Mol Recognit. 1999;12:322–327. [PubMed]
29. deHaseth PL, Lohman TM, Burgess RR, Record MT., Jr Nonspecific interactions of Escherichia coli RNA polymerase with native and denatured DNA: Differences in the binding behavior of core and holoenzyme. Biochemistry. 1978;17:1612–1622. [PubMed]
30. Gill SC, Weitzel SE, von Hippel PH. Escherichia coli sigma 70 and NusA proteins. I. Binding interactions with core RNA polymerase in solution and within the transcription complex. J Mol Biol. 1991;220:307–324. [PubMed]
31. Richardson JP. The binding of RNA polymerase to DNA. J Mol Biol. 1966;21:83–114. [PubMed]
32. Nudler E, Avetissova E, Markovtsov V, Goldfarb A. Transcription processivity: protein-DNA interactions holding together the elongation complex. Science. 1996;273:211–217. [PubMed]
33. Myszka DG, He X, Dembo M, Morton TA, Goldstein B. Extending the range of rate constants available from BIACORE: Interpreting mass transport-influenced binding data. Biophys J. 1998;75:583–594. [PMC free article] [PubMed]
34. Wilson KS, von Hippel PH. Transcription termination at intrinsic terminators: The role of the RNA hairpin. Proc Natl Acad Sci USA. 1995;92:8793–8797. [PMC free article] [PubMed]
35. Gusarov I, Nudler E. Control of intrinsic transcription termination by N, NusA: The basic mechanisms. Cell. 2001;107:437–449. [PubMed]
36. Pasman Z, von Hippel PH. Active Escherichia coli transcription elongation complexes are functionally homogeneous. J Mol Biol. 2002;322:505–519. [PubMed]
37. Reynolds R, Chamberlin MJ. Parameters affecting transcription termination by Escherichia coli RNA. II. Construction and analysis of hybrid terminators. J Mol Biol. 1992;224:53–63. [PubMed]
38. Chan CL, Wang D, Landick R. Multiple interactions stabilize a single paused transcription intermediate in which hairpin to 3′ end spacing distinguishes pause and termination pathways. J Mol Biol. 1997;268:54–68. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...