SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The association of the essential Escherichia coli protein NusA with RNA polymerase increases pausing and the efficiency of termination at intrinsic terminators. NusA is also part of the phage λ N protein-modified antitermination complex that functions to prevent transcriptional termination. We have investigated the structure of NusA using various deletion fragments of NusA in a variety of in vitro assays. Sequence and structural alignments have suggested that NusA has both S1 and KH homology regions that are thought to bind RNA. We show here that the portion of NusA containing the S1 and KH homology regions is important for NusA to enhance both termination and antitermination. There are two RNA polymerase-binding regions in NusA, one in the amino-terminal 137 amino acids and the other in the carboxy-terminal 264 amino acids; only the amino-terminal RNA polymerase-binding region provides a functional contact that enhances termination at an intrinsic terminator or antitermination by N. The carboxy-terminal region of NusA is also required for interaction with N and is important for the formation of an N–NusA–nut site or N–NusA–RNA polymerase–nut site complex; the instability of complexes lacking this carboxy-terminal region of NusA that binds N and RNA polymerase can be compensated for by the presence of the additional E. coli elongation factors, NusB, NusG and ribosomal protein S10.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

NusA is an essential E. coli protein that modulates elongation by associating with the core component of RNA polymerase (RNAP) after promoter escape and release of the σ70 subunit required for initiation at most promoters (Greenblatt and Li, 1981a). NusA influences elongation by increasing the dwell time for RNAP at certain pause sites (Greenblatt et al., 1981; Kassavetis and Chamberlin, 1981; Kingston et al., 1981; Farnham et al., 1982; Lau et al., 1983), possibly by interacting with and stabilizing the RNA hairpin structure often associated with pause sites (Landick and Yanofsky, 1987; Chan and Landick, 1993). It can also affect the overall rate of elongation in vitro by increasing the Ks for nucleoside triphosphates (Schmidt and Chamberlin, 1984). These effects on the RNAP elongation rate can influence gene regulation and may be necessary to couple transcription and translation and prevent premature termination caused by the Rho factor (Ruteshouser and Richardson, 1989; Zheng and Friedman, 1994). As well, NusA functions to enhance the efficiency of termination at many simple terminators (Greenblatt et al., 1981; Grayhack et al., 1985; Schmidt and Chamberlin, 1987; Whalen et al., 1988).

NusA and the additional host proteins NusB, NusG and ribosomal protein S10 are important for the N protein of bacteriophage λ to modify RNAP into a termination-resistant state (Das and Wolska, 1984; Goda and Greenblatt, 1985; Horwitz et al., 1987; Schauer et al., 1987). This modification requires a cis-acting element, called the nut site, that is composed of two elements, boxA and boxB (Salstrom and Szybalski, 1978; de Crombrugghe et al., 1979; Olson et al., 1982). N, the host proteins and nut site RNA assemble into a highly stable complex that associates with elongating RNAP (Barik et al., 1987; Horwitz et al., 1987; Mason et al., 1992a). Within this complex, N binds boxB RNA (Chattopadhyay et al., 1995; Mogridge et al., 1995; Tan and Frankel, 1995; Legault et al., 1998), and NusA binds to N (Greenblatt and Li, 1981b). NusB and ribosomal protein S10 form a heterodimer that binds the boxA portion of the nut site RNA (Mason et al., 1992b; Nodwell and Greenblatt, 1993; Mogridge et al., 1998a). NusG and S10, as well as NusA, bind RNAP (Greenblatt and Li, 1981a; Mason and Greenblatt, 1991; Li et al., 1992). Even in the absence of a DNA template, a stable complex can be assembled on the nut site RNA that contains N, RNAP and all the host cofactors (Mogridge et al., 1995).

While the presence of all of the host antitermination factors allows for the formation of a stable and highly processive antitermination complex (Barik et al., 1987; Horwitz et al., 1987; Mason et al., 1992a), high concentrations of N alone can cause nut site-independent antitermination in vitro (Rees et al., 1996). As this effect is enhanced by the presence of NusA in the reaction (Whalen et al., 1988; DeVito and Das, 1994), it was suggested that NusA functions to stabilize the N–NusA–RNAP–nut site complex (Rees et al., 1996).

Genetic studies on antitermination by the λ N protein have suggested that NusA may interact with the boxA portion of the nut site (Olson et al., 1982; 1984; Friedman and Olson, 1983), whereas gel mobility shift experiments have shown that mutations in both boxA and boxB specifically affect the interaction of NusA with an N–nut site complex (Mogridge et al., 1995; T.-F. Mah and J. Greenblatt, unpublished data). Protein–RNA cross-linking data have shown that NusA also interacts directly with the nascent single-stranded RNA in a transcription complex (Liu and Hanna, 1995a,b). In this context, however, boxA is not required for the NusA–RNA interaction.

Consistent with all the evidence that NusA may interact with RNA, sequence and structural alignments have indicated that NusA has S1 and KH homology regions, both of which are thought to interact with RNA (Gibson et al., 1993; Bycroft et al., 1997). S1 domains were originally identified in ribosomal protein S1, which has six S1 domains (Subramanian, 1983). Because this type of domain is found in other RNA-interacting proteins involved in the initiation of translation and turnover of mRNA (Bycroft et al., 1997), it must have some non-specific RNA-binding capacity. However, the S1 protein was also able to select specific RNA ligands from a random pool of RNAs and bind boxA-containing RNA with some selectivity, suggesting that S1 domains are capable of sequence-specific binding (Ringquist et al., 1995; Mogridge and Greenblatt, 1998). The KH domain was first identified in the human heterogeneous nuclear ribonucleoprotein (hnRNP) K protein (Siomi et al., 1993). Whereas some KH domain-containing proteins interact non-specifically with single-stranded RNA (Gibson et al., 1993), vigilin is a protein made up entirely of 14 KH domains that has been shown to bind directly to a specific mRNA (Dodson and Shapiro, 1997).

Relatively little is known about the S1 and KH homology regions in NusA, but some mutations in the S1 homology region of NusA prevent λ growth without affecting the viability of E. coli (T.-F. Mah, Y. Zhou, N. Yu, J. Mogridge et al., unpublished data; Friedman, 1971). These mutations often affect the ability of NusA to associate with an N–nut site RNA complex in gel mobility shift experiments, but do not directly affect the interaction between NusA and N (T. Mah, Y. Zhou, N. Yu, J. Mogridge et al., unpublished data; Mogridge et al., 1995).

Other studies have focused on the interaction of NusA with RNAP. The core RNAP enzyme consists of essential α, β and β′ subunits and one non-essential subunit, ω (Burgess et al., 1969; Gentry et al., 1991). Genetic experiments have suggested there may be interactions of NusA with β, β′ and α (Jin et al., 1988; Ito et al., 1991; Ito and Nakamura, 1993; 1996; Schauer et al., 1996). More recently, it has been shown that NusA can bind directly to the carboxy-terminal domain (CTD) of α, and possibly to β and β′, but not to the amino-terminal domain (NTD) of α (Liu et al., 1996). It has also been suggested that a direct interaction between the α-CTD and NusA is important for NusA's effects on pausing and termination, but not on N-mediated antitermination (Liu et al., 1996).

We have taken a systematic approach in an attempt to assign functions to the various regions of NusA. By making deletions at putative domain boundaries and assaying for effects on function, we were able to identify two RNAP-binding regions, one in the amino-terminal region and another in the carboxy-terminal region of NusA. We also found that the binding of NusA to N requires amino acids in the carboxy-terminal region of NusA. Interestingly, loss of the carboxy-terminal RNAP-binding region does not have an effect on termination or antitermination by N. Similarly, loss of the N-binding region does not have an effect on antitermination. However, loss of the S1 and KH homology regions or the amino-terminal RNAP-binding region abolishes NusA's ability to influence either of these processes. Surprisingly, the loss of the amino-terminal RNAP-binding region of NusA does not impair NusA's ability to assemble into a stable complex containing N, NusB, NusG, S10, RNAP and the nut site RNA, indicating that interaction of this region of NusA with RNAP may have a direct role in both termination and antitermination.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Production and characterization of NusA fragments

NusA binds to the core component of RNAP in order to modulate transcription (Greenblatt and Li, 1981a). A schematic diagram illustrating some of the putative domains of NusA is shown in Fig. 1A. These regions were assigned on the basis of comparisons with other proteins containing similar domains or on the basis of sequence alignments with NusA proteins from other organisms. Because of this, structural analysis must still be done to confirm the exact domain boundaries of NusA. The S1 and KH homology regions are putative RNA-binding domains, and the regions designated AR1 and AR2 are 50-amino-acid repeat sequences that are 39% identical and 54% similar and contain a high percentage of acidic residues (Gibson et al., 1993; Craven et al., 1994; Bycroft et al., 1997). In order to define structure–function relationships within NusA, we used these putative domain boundaries to make various polyhistidine-tagged deletion constructs. The resulting fragments of NusA were then overproduced in E. coli and purified to near homogeneity by chromatography on a nickel chelate resin (see Experimental procedures). All the NusA fragments were highly soluble in native conditions.

image

Figure 1. . Circular dichroism analysis of NusA deletion constructs. A. Predicted domain architecture of NusA. B. CD analysis of NusA deletion constructs. Wavelength scans were performed on selected NusA deletion constructs from 300 nm to 200 nm. C. Comparison of predicted percentage helix with actual percentage helix. Predicted secondary structure was generated by the PHDsec program. The values in the table for percentage helix were calculated using the following equation: ([θ222] + 2340)/30 3000 (Chen et al., 1972).

Download figure to PowerPoint

All the purified NusA fragments were analysed by circular dichroism (CD) spectroscopy, and some of the results are shown in Fig. 1B. With the exception of NusA (132–240), representing the isolated S1 homology region, there were obvious minima at 222 nm and 208 nm, indicating the presence of α-helices. In Fig. 1C are listed the percentage α-helix content of each NusA fragment and, for comparison, the percentage α-helix predicted by the phdsec secondary structure prediction program (Rost and Sander, 1993; 1994).

On the basis of these results, we think it is unlikely that the isolated S1 homology region in NusA (132–240) is properly folded. Also, because the 15% α-helical content of NusA (1–240) was much lower than the 31% predicted by the phdsec program and much lower than the 30% that would be predicted by comparison with the actual α-helical contents of NusA (1–348), NusA (1–137) and NusA (232–348), it is probable that the S1 homology region in NusA (1–240) is also not properly folded. This conclusion was confirmed by nuclear magnetic resonance (NMR) analysis, as shown in Fig. 2A. For this experiment, NusA (1–240) was labelled with 15N, and a two-dimensional 1H–15N quantum correlation (HSQC) spectrum was generated (Kay et al., 1992; Zhang et al., 1994). The lack of dispersion of the resonances in the spectrum suggested strongly that NusA (1–240) is improperly folded and highly aggregated. The probable lack of proper folding was taken into account when interpreting the negative results described below with NusA (1–240) and NusA (132–240).

image

Figure 2. . NMR analysis of fragments of NusA. (1H, 15N) HSQC spectra are shown for (A) NusA (1–240), (B) NusA (1–137) and (C) NusA (132–348). The sample concentrations were ≈ 0.6 mM, 1 mM and 0.1 mM respectively.

Download figure to PowerPoint

In contrast, CD analysis showed that the amino-terminal fragment, NusA (1–137), has a high α-helical content, similar to what is predicted. Also, this fragment was able to bind RNAP (see below). In CD experiments, NusA (1–137) showed co-operative and reversible melting with a TM of ≈ 50°C (data not shown), providing strong evidence for an independently folded domain. This conclusion was confirmed by analysing the HSQC spectrum of 15N-labelled NusA (1–137), as shown in Fig. 2B. In this case, the generally good dispersion of the resonances was indicative of a highly ordered, folded domain.

By comparing the 33% α-helical content of the active fragment NusA (1–348) (see below) and the 39% helical content of NusA (1–137), it can be predicted that NusA (132–348) would be about 29% α-helix, which is similar to the actual value of 28% (Fig. 1C). This suggested that the S1 and KH homology regions in NusA (132–348) should be folded for the most part. Again, this conclusion was verified by NMR (Fig. 2C). In this case, the good dispersion of the resonances in the HSQC spectrum suggested that NusA (132–348) also has a highly ordered structure, but the variation in signal intensity was suggestive of conformational heterogeneity. The likely presence in this NusA fragment of three domains, one S1 homology region and two KH homology regions (see Fig. 1A) could generate such conformational heterogeneity if one or more of the domains has freedom of motion with respect to the others. The longer active fragments, NusA (1–416) and NusA (1–495), and the multidomain inactive fragments, NusA (132–416) and NusA (132–495), which can interact with both N and with N–nut site complexes (see below), are also likely to be folded. We have made the assumption that these fragments of NusA, as well as NusA (1–137) and NusA (132–348), are properly folded when interpreting the results of some of the experiments that are described below.

Two RNAP-binding regions in NusA

To determine which part of NusA is required for the NusA–RNAP interaction, carboxy-terminal deletion mutants of NusA with His6 tags at the amino-terminus were used as column ligands in affinity chromatography experiments (Fig. 3A). E. coli extract containing additional RNAP core component (lane 2) was passed over the columns. The core polymerase subunits, β, β′ and α, bound specifically to all the deletion constructs that were tested (lanes 4–8), over and above the background binding to the Ni–agarose column matrix (lane 3). None of the NusA fragments had acquired an ability to bind proteins non-specifically, because few, if any, of the other proteins in the E. coli extract bound exclusively to any of the NusA columns (compare lane 3 with lanes 4–8). The smallest NusA fragment that bound RNAP, NusA (1–137) (lane 4), corresponds to a moderately α-helical amino-terminal domain of NusA that folds independently (see Figs 1B and 2B) and has not yet been assigned a function.

image

Figure 3. . Two RNAP binding-regions in NusA. A. NusA (1–137) binds RNAP. E. coli extract containing additional RNAP core enzyme (lane 2; RNAP is shown in lane 1) was passed over columns containing Ni–agarose (lane 3), 1.3 mg ml−1 His6-tagged NusA (lane 8) or various His6-tagged NusA deletion mutants (lanes 4–7). The concentration of each NusA deletion mutant on the column was adjusted so that each had the same molar concentration as the full-length NusA. Bound proteins were eluted with buffer containing 1 M NaCl, subjected to SDS–PAGE and stained with silver. B. NusA (232–495) binds RNAP. E. coli extract containing additional RNAP core enzyme (lane 1) was passed over columns containing Ni–agarose (lane 2), 1.3 mg ml−1 His6-tagged NusA (lane 6) or equivalent molar concentrations of various His6-tagged NusA deletion mutants (lanes 3–5). Bound proteins were eluted with buffer containing 1 M NaCl, subjected to SDS–PAGE and stained with silver.

Download figure to PowerPoint

As direct binding experiments (Liu et al., 1996) and NusA–RNAP cross-linking experiments (J. Greenblatt and J. Li, unpublished data) have shown that NusA may be able to bind to more than one subunit of RNAP, we reasoned that there could be another RNAP binding site on NusA. Therefore, His6-tagged constructs that lacked either the first 136 amino acids of NusA (data not shown) or its first 232 amino acids (Fig. 3B) were coupled to Ni–agarose and used as affinity chromatography ligands. In this case, the β, β′ and α subunits of RNAP were specifically retained on a column containing full-length NusA (lane 6), as well as on a column containing NusA (232–495) (lane 5), but deletion of the carboxy-terminal 79 amino acids in NusA (232–416) resulted in the loss of RNAP binding (lane 4). Similarly, a NusA protein beginning at amino acid 132 would not bind RNAP when the carboxy-terminal 79 amino acids of NusA were deleted (data not shown). Thus, there are two RNAP-binding regions in NusA, one in the amino-terminal 1–137 amino acids and another in the carboxy-terminal 264 amino acids. These conclusions were confirmed by two additional experiments showing that both highly purified RNAP core enzyme and the endogenous RNAP in a crude extract bind specifically to columns containing immobilized NusA (1–137), NusA (232–495) and NusA (1–495) (data not shown). Therefore, NusA (1–137) and NusA (232–495) can bind directly and independently to RNAP.

Only the amino-terminal RNAP-binding region of NusA is necessary for enhancement of termination at an intrinsic terminator

NusA increases the termination efficiency of RNAP at many intrinsic terminators (Greenblatt et al., 1981; Ward and Gottesman, 1981; Farnham et al., 1982; Schmidt and Chamberlin, 1987). It was therefore of interest to determine the impact of the loss of one or other of the RNAP-binding regions on NusA's function in termination. To this end, in vitro transcription was performed on a template containing a promoter, a wild-type nut site and the simple λ terminator, tR′ (Fig. 4A). When no NusA was added, RNAP alone gave 43% readthrough of the tR′ terminator. NusA (1–495) substantially decreased the amount of readthrough to 18%. In contrast to this effect, the addition of NusA (1–137) or NusA (1–240) had no effect on the efficiency of termination. Surprisingly, even though NusA (1–348) and NusA (1–416) had lost the ability to bind RNAP via their carboxy-terminal RNAP-binding regions, both proteins enhanced termination almost as well as the full-length NusA protein. Thus, the presence of the carboxy-terminal RNAP-binding region of NusA is not required for the enhancement of termination. The ability of NusA (1–137) to bind RNAP and the evidence from CD spectroscopy that this fragment of NusA is folded (see above), combined with the inability of NusA (1–137) and NusA (1–240) to enhance termination, indicated that the region containing the S1 and KH homology regions (amino acids 132–348) is required for the enhancement of termination. As the S1 homology region in NusA (1–240) is unlikely to be folded (see above), it is still unclear which of the S1 and KH homology regions is individually required for NusA to enhance termination.

image

Figure 4. . Regions of NusA important for enhancement of termination in vitro. A. The carboxy-terminal RNAP-binding region of NusA is not necessary for termination enhancement. B. The amino-terminal RNAP-binding region of NusA is necessary for termination enhancement. Transcription reactions containing 25 nM RNAP and 50 nM NusA, or various NusA deletion mutants (as indicated), were electrophoresed on 6 M urea–4% polyacrylamide gels, dried and exposed to film. Positions of the terminated and run-off transcripts are indicated.

Download figure to PowerPoint

The fact that NusA constructs containing only the amino-terminal RNAP-binding region and the S1 and KH homology regions were able to enhance termination suggested that the amino-terminal RNAP-binding region might be very important for this process. To test this idea, in vitro transcription was carried out using four NusA constructs of various lengths that lacked this region (Fig. 4B), and none was able to enhance termination. As CD and NMR experiments on NusA (132–348), which cannot enhance termination, had indicated that this portion of NusA is folded on its own, these results showed that the amino-terminal RNAP-binding region is necessary for NusA to enhance termination and cannot be compensated for by the carboxy-terminal RNAP-binding region that is present in NusA (132–495).

A carboxy-terminal region of NusA is necessary for the binding of N

NusA also binds the bacteriophage λ N protein (Greenblatt and Li, 1981b), and these proteins together can associate with RNAP to prevent termination of transcription when a nut site is present (Whalen et al., 1988; DeVito and Das, 1994; Mogridge et al., 1995; Rees et al., 1996). In order to determine which part of NusA is required for a direct interaction with N, GST-N was used as an affinity ligand in binding experiments (Fig. 5). A mixture of His6-tagged full-length NusA (1–495) and four carboxy-terminally deleted proteins, NusA (1–137), NusA (1–240), NusA (1–348) and NusA (1–416), as shown in lane 1, was passed over a GST control column and a GST-N column with sufficient capacity to bind all the NusA fragments. All of the proteins flowed through the GST column (lane 2) and were absent from the salt eluate of this column (lane 4). In contrast, the flowthrough from the GST-N column contained only the three smallest proteins, NusA (1–137), NusA (1–240) and NusA (1–348) (lane 3), whereas the salt eluate from this column contained significant amounts of only full-length NusA and the deletion construct that had the smallest carboxy-terminal truncation, NusA (1–416) (lane 5). As NusA (1–348) is active in termination and antitermination assays (see above and below), it must be properly folded. Therefore, these results indicated that the N-binding ability of NusA requires a carboxy-terminal portion of NusA, but the extreme carboxy-terminal 80 amino acids are not required for this interaction.

image

Figure 5. . N binds a carboxy-terminal region of NusA. A mixture of His6-tagged NusA and His6-tagged NusA deletion mutants (as indicated, lane 1) was passed over affinity columns containing 2 mg ml−1 GST (lanes 2 and 4) or 0.5 mg ml−1 GST-N (lanes 3 and 5). The flowthrough (lanes 2 and 3) and the 1 M NaCl eluate (lanes 4 and 5) fractions were subjected to SDS–PAGE and stained with silver.

Download figure to PowerPoint

The amino-terminal RNAP-binding region of NusA is essential for enhancing antitermination by N

NusA enhances an intrinsic ability of N to antiterminate transcription (Whalen et al., 1988; DeVito and Das, 1994; Rees et al., 1996). Therefore, in vitro transcription was performed in order to assess the effects of progressive carboxy-terminal deletions in NusA on the ability of NusA to influence antitermination (Fig. 6). The addition of N protein alone to a reaction increased the amount of readthrough from 46% to 68% (Fig. 6B). When NusA was added to a reaction containing N and RNAP, the readthrough was increased further to 86%. The addition of NusA (1–137) or NusA (1–240) to the reaction had no effect on the ability of N to antiterminate (Fig. 6A). Surprisingly, however, NusA (1–348), which had lost its ability to bind directly to N, enhanced the effect of N on readthrough of the tR′ terminator as well as the two N-binding constructs, NusA (1–416) and NusA (1–495) (Fig. 6A). Combined with CD and NMR evidence that NusA (1–137) is folded and other evidence that it binds RNAP, the inability of NusA (1–240) and NusA (1–137) to enhance antitermination by N indicated that the region containing the KH and S1 homology regions (amino acids 132–348) must be critical for this process.

image

Figure 6. . Regions of NusA necessary for enhancement of antitermination by N. A. The carboxy-terminal N-binding region of NusA is not necessary for enhancement for antitermination by N. B. The amino-terminal RNAP-binding region of NusA is essential for enhancement of antitermination by N. Transcription reactions containing 25 nM RNAP, 100 nM N and 50 nM NusA or various NusA deletion mutants (as indicated) were electrophoresed on 6 M urea 4%–polyacrylamide gels, dried and exposed to film. Positions of the terminated and run-off transcripts are indicated.

Download figure to PowerPoint

Experiments were also carried out to assess the effect of deleting the amino-terminal RNAP-binding region of NusA on antitermination by N (Fig. 6B). Loss of the amino-terminal RNAP-binding region of NusA in NusA (132–348), NusA (132–416) and NusA (132–495) greatly affected the ability of NusA to enhance antitermination by N. The levels of readthrough observed with these amino-terminally truncated NusA constructs were not increased beyond the level that N could promote on its own. In fact, there was actually less readthrough, suggesting that these mutant NusA proteins may be interfering with antitermination by N via an unknown mechanism. Thus, even the amino-terminally deleted N-binding constructs, NusA (132–416), which can interact with an N–nut site complex (see below), and NusA (132–495), which can even assemble into complexes containing RNAP (see below), were unable to influence antitermination, suggesting that the amino-terminal RNAP-binding region of NusA is critical for this process.

Effects of NusA deletions on the formation of N–NusA–nut site complexes

Gel mobility shift experiments can be used to assess the interaction of N, the E. coli Nus factors and RNAP with 32P-labelled nut site-containing RNA (Mogridge et al., 1995). In order to assess the importance of the various NusA domains on the binding of NusA to N–nut site complexes, gel shift experiments were performed with N and various NusA deletion constructs (Fig. 7). N protein alone is sufficient to bind and retard the mobility of RNA containing a wild-type nut site (lanes 1 and 2), whereas full-length NusA cannot shift the RNA on its own (Mogridge et al., 1995). When full-length NusA was added to the reaction containing N, the RNA was supershifted from the N–nut site complex (lanes 7, 12 and 16). NusA (1–416), which can also bind N directly (see Fig. 5), was also capable of supershifting the N–RNA complex (lane 6). In this case, the supershifted complex had lower mobility, perhaps because deleting the last 80 amino acids of NusA alters its conformation or causes it to dimerize. As expected, the deleted, non-functional NusA proteins that were not able to bind N, namely NusA (1–137) and NusA (1–240), were not able to supershift the N–RNA complex (lanes 3 and 4), and even NusA (1–348), which supports antitermination but cannot bind N, interacted more weakly with the complex (lane 5). None of the NusA fragments except NusA (1–416) can bind the RNA directly in the absence of N (T.-F. Mah and J. Greenblatt, manuscript in preparation). Thus, only NusA constructs that can bind N can interact with an N–nut site complex strongly enough to create a supershift in a non-denaturing gel. Because the carboxy-terminal N-binding region of NusA is important for the supershift, we could not determine whether the KH and S1 homology regions of NusA are also necessary for complex formation.

image

Figure 7. . The effects of NusA deletions on the formation of N–NusA–nut site complexes. Reactions containing 32P-labelled nut site RNA, 1.5 μM N and 1.0 μM NusA or various NusA deletion mutants (as indicated) were electrophoresed on 7.5% non-denaturing gels, dried and exposed to film. The results from three separate gels were combined, but control lanes 1 and 2, originally present in all the gels, are shown only once.

Download figure to PowerPoint

To assess the effect of progressive amino-terminal deletions on the ability of NusA to supershift an N–nut site complex, gel mobility shift experiments were performed with NusA constructs truncated at either amino acid 132 or amino acid 232, and thus missing either the amino-terminal RNAP-binding region (amino acids 1–137) or this region plus the S1 homology region (amino acids 136–233). NusA (132–416) and NusA (132–495) were able to support the formation of an N–NusA–nut site complex (lanes 10 and 11). These proteins retained the S1 and KH homology regions and the N-binding region of NusA. Therefore, the amino-terminal RNAP-binding region is not necessary for binding to an N–nut site complex, even though it is necessary for NusA to enhance antitermination. As before, further carboxy-terminal deletions to amino acids 240 or 348, which eliminated the N-binding site of NusA, reduced or eliminated the ability of NusA to join the N–nut site complex (lanes 8 and 9). None of the constructs lacking the S1 homology region, as well as the amino-terminal RNAP-binding region, namely NusA (232–348), NusA (232–416) and NusA (232–495), could interact with the N–nut site complex (lanes 13–15). As CD experiments indicated that the KH homology regions in these fragments were likely to be folded, and as constructs containing the S1 homology region and extending to amino acids 416 or 495 were able to support complex formation (lanes 10 and 11), this result suggested that the S1 homology region is required for interaction with the N–nut site complex. This conclusion would be consistent with our earlier observation that the nusA1 mutation in the S1 homology region prevents NusA from binding to the N–nut complex (Mogridge et al., 1995). Thus, at least the S1 homology region of NusA and an N-binding region between amino acids 348 and 416 are required in order for NusA to associate strongly with an N–nut site complex.

Effects of NusA deletions on the formation of complete complexes containing N, the Nus factors and RNAP

A NusA mutant truncated at amino acid 343, which lacks the N binding site of NusA, is partially functional for antitermination in vivo and somewhat temperature sensitive for E. coli growth (Tsugawa et al., 1988), suggesting that most of the wild-type function is retained in this protein. To investigate whether the presence of the remaining Nus factors, NusB, NusG and ribosomal protein S10, as well as RNAP, could somehow stabilize the association of such a truncated NusA molecule with the N–nut site complex, we performed gel mobility shift experiments in the presence of all the Nus factors, RNAP and RNA containing a wild-type nut site (Fig. 8A; Mogridge et al., 1995). Wild-type NusA supports the formation of a complete, lower mobility complex containing N, all of the Nus factors and RNAP (compare lane 17 with lanes 15 and 16; Mogridge et al., 1995). The low-mobility complex was still formed with NusA (1–416), which lacks the carboxy-terminal RNAP-binding region but still retains the carboxy-terminal N-binding region and the amino-terminal RNAP-binding region (lane 14). Interestingly, although NusA (1–348) seemed unable to support the formation of the unstable N–NusA–RNAP–nut site complex (lane 10) that was observed with NusA (1–416) (lane 13) or full-length NusA (lane 16), it was able to support complete complex formation (lane 11), although not as efficiently as NusA (1–495) or NusA (1–416) (compare lane 11 with lanes 14 and 17). This result correlated well with the in vivo data (Tsugawa et al., 1988) and suggests that the additional stability provided by NusB, NusG and S10 can partially compensate for instability caused by the loss of the N-binding and RNAP-binding regions in the carboxy-terminal region of NusA. The shorter NusA deletion constructs, NusA (1–137) and NusA (1–240), were not able to support complete complex formation (lanes 5 and 8), indicating that the portion of NusA containing the KH and S1 homology regions is important for complete complex formation.

image

Figure 8. . The effects of NusA deletions on the formation of complete complexes containing N, the Nus factors and RNAP. A. The portion of NusA containing the KH and S1 homology regions is important for the formation of the complete complex. B. Either RNAP-binding region is sufficient to support complete complex formation. Reactions containing 32P-labelled nut site RNA and various combinations of 500 nM N, 50 nM NusB, 50 nM NusG, 50 nM S10, 25 nM RNAP and 100 nM NusA or NusA deletion mutants (as indicated) were electrophoresed on 5% non-denaturing gels, dried and exposed to film.

Download figure to PowerPoint

To test directly whether the amino-terminal RNAP-binding region is important for forming a low-mobility complex containing RNAP, we used NusA constructs that lack this region in gel mobility shift experiments with N, RNAP and the rest of the Nus factors (Fig. 8B). The constructs that lacked the carboxy-terminal RNAP-binding region, as well as the amino-terminal RNAP-binding region, NusA (132–240), NusA (132–348) and NusA (132–416), were unable to support the formation of the complete complex (lanes 5, 8 and 11). Interestingly, the presence of the carboxy-terminal RNAP-binding region in NusA (132–495) compensated for the loss of the amino-terminal RNAP-binding region, such that a low-mobility complex was formed (compare lanes 14 and 17) whose stability was increased by NusB, NusG and S10 (compare lanes 13 and 14). Therefore, the presence of either RNAP-binding region is sufficient to support the formation of a low-mobility stable complex even though only the amino-terminal RNAP-binding region is needed for NusA to enhance antitermination (Fig. 6).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We used deletion constructs of NusA in various assays to identify functional regions of the protein (see Fig. 9). Two RNAP-binding regions were identified by affinity chromatography: one was localized to amino acids 1–137 of NusA and appears to be a folded domain; the other in amino acids 232–495 was eliminated by deleting amino acids 417–495. Although both of these regions bound about equally well to the RNAP core enzyme, in vitro transcription assays revealed that the amino-terminal RNAP-binding domain of NusA is very important for NusA's effects on termination at an intrinsic terminator and on antitermination by N, while the carboxy-terminal RNAP-binding region is dispensable for both activities.

image

Figure 9. . Summary of the results of the deletion analysis of NusA. See text for details.

Download figure to PowerPoint

In addition to the previously observed interaction between N and full-length NusA (Greenblatt and Li, 1981b), N protein also bound NusA (1–416). However, further truncation of NusA, to amino acid 348, resulted in the loss of N binding. There are two acidic repeats in the carboxy-terminal region of NusA (Craven et al., 1994). One of these repeats is lost when NusA is truncated to amino acid 416, while the second is lost when NusA is further truncated to amino acid 348. It may be that the basic N protein interacts with both of these regions equally well and that the loss of one of the two acidic repeats has only a minor effect on N binding.

Results from the in vitro antitermination assay, in which loss of N-binding ability did not affect NusA function, calls into question the importance of the N–NusA interaction. However, full-length NusA cannot interact with RNA on its own, and we speculate that the binding of N to the carboxy-terminal region of NusA facilitates the binding of NusA to RNA in the context of the antitermination complex. Indeed, interaction of NusA with RNA appears to be essential for antitermination, because the nusA1 point mutation in the S1 homology region of NusA (NusA R183A; Friedman, 1971) or deletion of the S1 and KH homology regions in NusA (1–137) can each impair NusA's ability to support antitermination by N. Thus, we think that N may transform wild-type NusA from a termination factor into an antitermination factor by promoting a NusA–nut site RNA interaction that would lead to antitermination rather than an alternative NusA–RNA interaction that would lead to termination or pausing.

Formation of complexes in gel mobility shift experiments gives an indication of the stability of a complex. Using the NusA deletion constructs with this assay, it was possible to determine which regions of NusA are important for interaction with the other components of the system. When gel shifts were performed with N, NusA and the nut site, loss of the N-binding region or the S1 homology region prevented the formation of the N–NusA–nut site complex, but neither RNAP-binding region was essential for the formation of this complex. When the reactions contained RNAP, as well as N, NusA and nut site RNA, the N-binding region and one or other of the RNAP-binding regions of NusA were critical for the formation of the characteristically unstable RNAP-containing complex that functions in antitermination only over a short distance (see Figs 8 and 9; Whalen et al., 1988; DeVito and Das, 1994; Mogridge et al., 1995; Rees et al., 1996). The ability of NusA (132–495) to form this complex, as well as a more stable complex in the presence of NusB, NusG and S10, even though this complex lacks the amino-terminal RNAP-binding region of NusA and is non-functional in antitermination, implies that N alters the amino-terminal termination-enhancing region of NusA, so that it actually interferes with termination by RNAP.

NusA (1–348) was unable to support the formation of observable amounts of the N–NusA–RNAP–nut site complex in a gel mobility shift assay. However, as more Nus factors were added to the reaction, so that there were additional protein–protein and protein–RNA interactions, NusA (1–348) was able to support the formation of a low-mobility, stable complex. Thus, NusB, NusG and S10 can restore the stability of the complex and compensate for loss of the carboxy-terminal RNAP-binding region and N-binding region in NusA (1–348). Interestingly, although the carboxy-terminal RNAP-binding region could not compensate for the loss of the amino-terminal RNAP-binding region in transcription assays, it could do so in the gel mobility shift experiment, suggesting that, whereas either protein–protein contact between NusA and RNAP is enough to stabilize the complex, only the amino–terminal interaction has very critical consequences for termination and antitermination. This is consistent with the observation that an E. coli strain containing NusA (1–343), which lacks the carboxy-terminal RNAP-binding region, is viable, although temperature-sensitive, and supports partial antitermination by N (Tsugawa et al., 1988). Whereas NusA (1–348) supports both termination and antitermination, neither NusA (1–137), which binds RNAP, nor NusA (132–348), which contains the S1 and KH homology regions, can function on its own, even though CD and NMR analysis indicate that these NusA fragments have highly ordered, folded structures. We have not determined whether a combination of NusA (1–137) and NusA (132–348) is functional.

The S1 and KH homology regions are predicted RNA-binding domains (Gibson et al., 1993; Bycroft et al., 1997). Gel mobility shift experiments with nut site RNA have shown that point mutations in the S1 homology region are detrimental to the formation of various RNA-bound complexes containing N just as they are detrimental for antitermination (Mogridge et al., 1995; T.-F. Mah, Y. Zhou, N. Yu, J. Mogridge et al., unpublished data). Our observations here that NusA (232–416) and NusA (232–495), two constructs that lack the S1 homology region and are likely to be folded, do not bind N–nut site complexes also suggest that the S1 homology region is important for complex formation. Although NusA (1–240), which lacks the KH homology regions, did not support termination, antitermination or the formation of RNA-bound complexes, CD and NMR experiments indicated that the S1 homology region in this construct is unlikely to be folded. Therefore, it is still unclear whether the KH homology regions of NusA participate in termination, antitermination or RNA binding.

We have shown previously that there are nucleotides in both boxA and boxB that are important for NusA association with the N–nut site complex (Mogridge et al., 1995). We have also shown previously that ribosomal protein S1 specifically binds boxA RNA (Mogridge and Greenblatt, 1998), suggesting that the S1 homology region of NusA may interact with boxA. If the S1 homology region interacts with one portion of the nut site, it is possible that at least one of the KH homology regions interacts with the other. As there is evidence that individual KH domains have different sequence specificities (Dejgaard and Leffers, 1996), it is possible that one of the KH homology regions in NusA is required for nut site binding and the other is required to bind to nascent mRNA. An alternative possibility is provided by the observation that the hnRNP K protein has three KH domains, and mutation or deletion of any one of them severely affects the ability of the protein to interact with C-rich sequences (Siomi et al., 1994). This suggests that both KH homology regions in NusA may be required for binding one type of RNA sequence.

The amino-terminal RNAP-binding domain and the S1 and KH homology regions of NusA are the minimal regions required for wild-type function in our in vitro assays for termination and antitermination. Further work needs to be done to elucidate the RNA sequence preferences of the S1 and KH RNA-binding regions in NusA and to identify the precise functions of the critical interaction between the amino-terminal domain of NusA and RNAP.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Plasmids, enzymes and strains

RNAP, N, NusA, NusB and NusG were purified as described previously (Burgess and Jendrisak, 1975; Greenblatt et al., 1980; Greenblatt et al., 1981; Swindle et al., 1988; Li et al., 1992). Purified S10 was generously provided by Dr Volker Nowotny.

All plasmids were prepared in the bacterial strain DH5α (Life Technologies) and purified on Qiagen-tip 500 columns (Qiagen). The oligonucleotides used for cloning were purchased from ACGT Corp. (Toronto). RNAguard was bought from Pharmacia Biotech. Restriction enzymes and DNA ligase were purchased from New England Biolabs. T7 RNAP was obtained from Life Technologies.

Construction of His6-tagged NusA proteins

Polymerase chain reaction (PCR) primers were designed to amplify NusA or fragments of NusA from the plasmid pJL4. Forward and reverse primers contained NcoI and BamHI restriction sites, respectively, for subsequent cloning into the vector pET-11d (Novagen). The forward primers also contained the sequence for six histidines.

Purification of His6-tagged NusA proteins

The E. coli strain BL21 (pREP4) containing the His6-tagged NusA plasmids was grown in 1 l of LB medium to an A600 of 0.5 and induced for 3 h with 2 mM IPTG. Cells were harvested by centrifugation for 20 min at 4500 r.p.m. in a Sorvall H6000A swinging bucket rotor, resuspended in 10 ml of binding buffer (5 mM imidazole, 0.5 M NaCl, 20 mM Tris, pH 7.9, 10% glycerol, 5 mM β-mercaptoethanol), sonicated and centrifuged for 30 min at 15 000 r.p.m. in a Sorvall SS34 rotor. The extracts were mixed with 0.25 ml of His Bind beads (Qiagen) and incubated at 4°C for 2 h. The beads were added to 1 ml Bio-Spin columns (Bio-Rad) and washed sequentially with 2 ml of binding buffer, 4 ml of wash buffer (45 mM imidazole, 0.5 M NaCl, 20 mM Tris, pH 7.9, 10% glycerol, 5 mM β-mercaptoethanol) and 1 ml ACB (10 mM HEPES, pH 7.0, 10% glycerol, 0.1 mM EDTA, 5 mM β-mercaptoethanol containing 0.1 M NaCl). Columns were eluted with 1 ml of ACB containing 0.1 M NaCl, 0.4 M imidazole, pH 7.9, and 1 mM dithiothreitol (DTT).

Affinity chromatography

For experiments with GST-N, extract containing GST or GST-N (Mogridge et al., 1998b) was incubated with glutathione Sepharose 4B beads (Pharmacia Biotech) and rotated at room temperature for 15 min. The beads were washed sequentially with 1 M NaCl buffer A [20 mM Tris-HCl, pH 7.8, 0.2 mM EDTA, 1 mM DTT, 1 mM phenylmethylsulphonyl fluoride (PMSF)] and 0.1 M NaCl buffer A. Beads (20 μl) were added to 200 μl pipette tips that contained 10 μl of 212–300 micron glass beads (Sigma). The column bed was washed with 10 column volumes of 1 M NaCl ACB (10 mM HEPES, pH 7.0, 10% glycerol, 0.1 mM EDTA, 1 mM DTT) and then washed with 10 column volumes of 100 mM NaCl ACB. Columns were loaded with 100 μl of a mixture of His6-tagged NusA and His6-tagged NusA deletion mutants (3.0 μg each) in 100 mM NaCl ACB buffer supplemented with 0.2 mg ml−1 insulin. The columns were washed with 10 column volumes of 100 mM NaCl ACB and eluted with 1 M NaCl ACB.

For experiments with RNAP and the His6-tagged NusA and the His6-tagged NusA deletion mutants, purified His6-tagged protein was incubated with Ni–agarose (Qiagen) for 30 min. Beads (20 μl) were added to 200 μl pipette tips and treated as described above except that these columns were loaded with 100 μl of 1 mg ml−1E. coli extract containing 6 μg of additional RNAP core enzyme.

In vitro transcription

Transcription reactions were performed as described previously (Whalen and Das, 1990) using the template pJD12 (generous gift from A. Das). Concentrations of proteins are indicated in the figure legends. Levels of terminated and run-off transcripts were quantified using a phosphoimager.

Gel mobility shift experiments

Gel mobility shift assays were performed as described previously (Mogridge et al., 1995). Concentrations of proteins are indicated in the figure legends.

CD and NMR spectroscopy

Circular dichroism (CD) was performed in an Aviv 62A DS circular dichroism spectrometer. All measurements were collected at 25°C in 10 mM Tris-HCl, 0.1 mM EDTA, 250 mM NaCl and 0.1 mM DTT in a 0.1 cm cuvette. Spectra were collected at a scanning speed of 1 nm s−1 from 300 nm to 200 nm. 15N-labelled fragments of NusA were prepared, and HSQC NMR spectra were generated as described previously for the λ N protein (Mogridge et al., 1998b).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The authors thank R. Muhandiram and L. Kay for the NMR experiments. This work was supported by the Medical Research Council of Canada (MRC). J.G. is an International Research Scholar of the Howard Hughes Medical Institute and an MRC Distinguished Scientist.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Barik, S., Ghosh, B., Whalen, W., Lazinski, D., Das, A. (1987) An antitermination protein engages the elongating transcription apparatus at a promoter-proximal recognition site. Cell 50: 885899.
  • 2
    Burgess, R.R. & Jendrisak, J.J. (1975) A procedure for the rapid, large-scale purification of Escherichia coli DNA-dependent RNA polymerase involving Polymin P precipitation and DNA-cellulose chromatography. Biochemistry 14: 46344638.
  • 3
    Burgess, R.R., Travers, A.A., Dunn, J.J., Bautz, E.K. (1969) Factor stimulating transcription by RNA polymerase. Nature 221: 4346.
  • 4
    Bycroft, M., Hubbard, T.J., Proctor, M., Freund, S.M., Murzin, A.G. (1997) The solution structure of the S1 RNA binding domain: a member of an ancient nucleic acid-binding fold. Cell 88: 235242.
  • 5
    Chan, C.L. & Landick, R. (1993) Dissection of the his leader pause site by base substitution reveals a multipartite signal that includes a pause RNA hairpin. J Mol Biol 233: 2542.
  • 6
    Chattopadhyay, S., Garcia-Mena, J., DeVito, J., Wolska, K., Das, A. (1995) Bipartite function of a small RNA hairpin in transcription antitermination in bacteriophage lambda. Proc Natl Acad Sci USA 92: 40614065.
  • 7
    Chen, Y.H., Yang, J.T., Martinez, H.M. (1972) Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion. Biochemistry 11: 41204131.
  • 8
    Craven, M.G., Granston, A.E., Schauer, A.T., Zheng, C., Gray, T.A., Friedman, D.I. (1994) Escherichia coli-Salmonella typhimurium hybrid nusA genes: identification of a short motif required for action of the lambda N transcription antitermination protein. J Bacteriol 176: 13941404.
  • 9
    De Crombrugghe, B., Mudryj, M., DiLauro, R., Gottesman, M. (1979) Specificity of the bacteriophage lambda N gene product (pN): nut sequences are necessary and sufficient for antitermination by pN. Cell 18: 11451151.
  • 10
    Das, A. & Wolska, K. (1984) Transcription antitermination in vitro by lambda N gene product: requirement for a phage nut site and the products of host nusA, nusB, and nusE genes. Cell 38: 165173.
  • 11
    Dejgaard, K. & Leffers, H. (1996) Characterisation of the nucleic-acid-binding activity of KH domains. Different properties of different domains. Eur J Biochem 241: 425431.
  • 12
    DeVito, J. & Das, A. (1994) Control of transcription processivity in phage lambda: Nus factors strengthen the termination-resistant state of RNA polymerase induced by N antiterminator. Proc Natl Acad Sci USA 91: 86608664.
  • 13
    Dodson, R.E. & Shapiro, D.J. (1997) Vigilin, a ubiquitous protein with 14 K homology domains, is the estrogen-inducible vitellogenin mRNA 3′-untranslated region-binding protein. J Biol Chem 272: 1224912252.
  • 14
    Farnham, P.J., Greenblatt, J., Platt, T. (1982) Effects of NusA protein on transcription termination in the tryptophan operon of Escherichia coli. Cell 29: 945951.
  • 15
    Friedman, D. (1971) A bacterial mutant affecting lambda development. In The Bacteriophage Lambda. Hershey, A.D. (ed.). Cold Spring Harbor, NY: Cold Spring Harbour Laboratory Press, pp. 733738.
  • 16
    Friedman, D.I. & Olson, E.R. (1983) Evidence that a nucleotide sequence, ‘boxA’, is involved in the action of the NusA protein. Cell 34: 143149.
  • 17
    Gentry, D., Xiao, H., Burgess, R., Cashel, M. (1991) The omega subunit of Escherichia coli K-12 RNA polymerase is not required for stringent RNA control in vivo. J Bacteriol 173: 39013903.
  • 18
    Gibson, T.J., Thompson, J.D., Heringa, J. (1993) The KH domain occurs in a diverse set of RNA-binding proteins that include the antiterminator NusA and is probably involved in binding to nucleic acid. FEBS Lett 324: 361366.
  • 19
    Goda, Y. & Greenblatt, J. (1985) Efficient modification of E. coli RNA polymerase in vitro by the N gene transcription antitermination protein of bacteriophage lambda. Nucleic Acids Res 13: 25692582.
  • 20
    Grayhack, E.J., Yang, X.J., Lau, L.F., Roberts, J.W. (1985) Phage lambda gene Q antiterminator recognizes RNA polymerase near the promoter and accelerates it through a pause site. Cell 42: 259269.
  • 21
    Greenblatt, J. & Li, J. (1981 a) Interaction of the sigma factor and the nusA gene protein of E. coli with RNA polymerase in the initiation–termination cycle of transcription. Cell 24: 421428.
  • 22
    Greenblatt, J. & Li, J. (1981 b) The nusA gene protein of Escherichia coli. Its identification and a demonstration that it interacts with the gene N transcription anti-termination protein of bacteriophage lambda. J Mol Biol 147: 1123.
  • 23
    Greenblatt, J., Malnoe, P., Li, J. (1980) Purification of the gene N transcription anti-termination protein of bacteriophage lambda. J Biol Chem 255: 14651470.
  • 24
    Greenblatt, J., McLimont, M., Hanly, S. (1981) Termination of transcription by nusA gene protein of Escherichia coli. Nature 292: 215220.
  • 25
    Horwitz, R.J., Li, J., Greenblatt, J. (1987) An elongation control particle containing the N gene transcriptional antitermination protein of bacteriophage lambda. Cell 51: 631641.
  • 26
    Ito, K. & Nakamura, Y. (1993) Pleiotropic effects of the rpoC10 mutation affecting the RNA polymerase beta′ subunit of Escherichia coli on factor-dependent transcription termination and antitermination. Mol Microbiol 9: 285293.
  • 27
    Ito, K. & Nakamura, Y. (1996) Localization of nusA-suppressing amino acid substitutions in the conserved regions of the beta′ subunit of Escherichia coli RNA polymerase. Mol Gen Genet 251: 699706.
  • 28
    Ito, K., Egawa, K., Nakamura, Y. (1991) Genetic interaction between the beta′ subunit of RNA polymerase and the arginine-rich domain of Escherichia coli nusA protein. J Bacteriol 173: 14921501.
  • 29
    Jin, D.J., Walter, W.A., Gross, C.A. (1988) Characterization of the termination phenotypes of rifampicin-resistant mutants. J Mol Biol 202: 245253.
  • 30
    Kassavetis, G.A. & Chamberlin, M.J. (1981) Pausing and termination of transcription within the early region of bacteriophage T7 DNA in vitro. J Biol Chem 256: 27772786.
  • 31
    Kay, L.E., Keifer, P., Saarinen, T. (1992) Pure absorption gradient enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity. J Am Chem Soc 114: 1066310665.
  • 32
    Kingston, R.E., Nierman, W.C., Chamberlin, M.J. (1981) A direct effect of guanosine tetraphosphate on pausing of Escherichia coli RNA polymerase during RNA chain elongation. J Biol Chem 256: 27872797.
  • 33
    Landick, R. & Yanofsky, C. (1987) Isolation and structural analysis of the Escherichia coli trp leader paused transcription complex. J Mol Biol 196: 363377.
  • 34
    Lau, L.F., Roberts, J.W., Wu, R. (1983) RNA polymerase pausing and transcript release at the lambda tR1 terminator in vitro. J Biol Chem 258: 93919397.
  • 35
    Legault, P., Li, J., Mogridge, J., Kay, L.E., Greenblatt, J. (1998) NMR structure of the bacteriophage lambda N peptide/boxB RNA complex: recognition of a GNRA fold by an arginine-rich motif. Cell 93: 289299.
  • 36
    Li, J., Horwitz, R., McCracken, S., Greenblatt, J. (1992) NusG, a new Escherichia coli elongation factor involved in transcriptional antitermination by the N protein of phage lambda. J Biol Chem 267: 60126019.
  • 37
    Liu, K. & Hanna, M.M. (1995 a) NusA contacts nascent RNA in Escherichia coli transcription complexes. J Mol Biol 247: 547558.
  • 38
    Liu, K. & Hanna, M.M. (1995 b) NusA interferes with interactions between the nascent RNA and the C- terminal domain of the alpha subunit of RNA polymerase in Escherichia coli transcription complexes. Proc Natl Acad Sci USA 92: 50125016.
  • 39
    Liu, K., Zhang, Y., Severinov, K., Das, A., Hanna, M.M. (1996) Role of Escherichia coli RNA polymerase alpha subunit in modulation of pausing, termination and anti-termination by the transcription elongation factor NusA. EMBO J 15: 150161.
  • 40
    Mason, S.W. & Greenblatt, J. (1991) Assembly of transcription elongation complexes containing the N protein of phage lambda and the Escherichia coli elongation factors NusA, NusB, NusG, and S10. Genes Dev 5: 15041512.
  • 41
    Mason, S.W., Li, J., Greenblatt, J. (1992 a) Host factor requirements for processive antitermination of transcription and suppression of pausing by the N protein of bacteriophage lambda. J Biol Chem 267: 1941819426.
  • 42
    Mason, S.W., Li, J., Greenblatt, J. (1992 b) Direct interaction between two Escherichia coli transcription antitermination factors, NusB and ribosomal protein S10. J Mol Biol 223: 5566.
  • 43
    Mogridge, J. & Greenblatt, J. (1998) Specific binding of Escherichia coli ribosomal protein S1 to boxA transcriptional antiterminator RNA. J Bacteriol 180: 22482252.
  • 44
    Mogridge, J., Mah, T.F., Greenblatt, J. (1995) A protein–RNA interaction network facilitates the template-independent cooperative assembly on RNA polymerase of a stable antitermination complex containing the lambda N protein. Genes Dev 9: 28312845.
  • 45
    Mogridge, J., Mah, T.F., Greenblatt, J. (1998 a) Involvement of boxA nucleotides in the formation of a stable ribonucleoprotein complex containing the bacteriophage lambda N protein. J Biol Chem 273: 41434148.
  • 46
    Mogridge, J., Legault, P., Li, J., Van Oene, M.D., Kay, L.E., Greenblatt, J. (1998 b) Independent ligand-induced folding of the RNA-binding domain and two functionally distinct antitermination regions in the phage lambda N protein. Mol Cell 1: 265275.
  • 47
    Nodwell, J.R. & Greenblatt, J. (1993) Recognition of boxA antiterminator RNA by the E. coli antitermination factors NusB and ribosomal protein S10. Cell 72: 261268.
  • 48
    Olson, E.R., Flamm, E.L., Friedman, D.I. (1982) Analysis of nutR: a region of phage lambda required for antitermination of transcription. Cell 31: 6170.
  • 49
    Olson, E.R., Tomich, C.S., Friedman, D.I. (1984) The nusA recognition site. Alteration in its sequence or position relative to upstream translation interferes with the action of the N antitermination function of phage lambda. J Mol Biol 180: 10531063.
  • 50
    Rees, W.A., Weitzel, S.E., Yager, T.D., Das, A., Von Hippel, P.H. (1996) Bacteriophage lambda N protein alone can induce transcription antitermination in vitro. Proc Natl Acad Sci USA 93: 342346.
  • 51
    Ringquist, S., Jones, T., Snyder, E.E., Gibson, T., Boni, I., Gold, L. (1995) High-affinity RNA ligands to Escherichia coli ribosomes and ribosomal protein S1: comparison of natural and unnatural binding sites. Biochemistry 34: 36403648.
  • 52
    Rost, B. & Sander, C. (1993) Prediction of protein secondary structure at better than 70% accuracy. J Mol Biol 232: 584599.
  • 53
    Rost, B. & Sander, C. (1994) Combining evolutionary information and neural networks to predict protein secondary structure. Proteins 19: 5577.
  • 54
    Ruteshouser, E.C. & Richardson, J.P. (1989) Identification and characterization of transcription termination sites in the Escherichia coli lacZ gene. J Mol Biol 208: 2343.
  • 55
    Salstrom, J.S. & Szybalski, W. (1978) Coliphage lambda nutL: a unique class of mutants defective in the site of gene N product utilization for antitermination of leftward transcription. J Mol Biol 124: 195221.
  • 56
    Schauer, A.T., Carver, D.L., Bigelow, B., Baron, L.S., Friedman, D.I. (1987) lambda N antitermination system: functional analysis of phage interactions with the host NusA protein. J Mol Biol 194: 679690.
  • 57
    Schauer, A.T., Cheng, S.W., Zheng, C., St Pierre, L., Alessi, D., Hidayetoglu, D.L., et al (1996) The alpha subunit of RNA polymerase and transcription antitermination. Mol Microbiol 21: 839851.
  • 58
    Schmidt, M.C. & Chamberlin, M.J. (1984) Amplification and isolation of Escherichia coli nusA protein and studies of its effects on in vitro RNA chain elongation. Biochemistry 23: 197203.
  • 59
    Schmidt, M.C. & Chamberlin, M.J. (1987) nusA protein of Escherichia coli is an efficient transcription termination factor for certain terminator sites. J Mol Biol 195: 809818.
  • 60
    Siomi, H., Matunis, M.J., Michael, W.M., Dreyfuss, G. (1993) The pre-mRNA binding K protein contains a novel evolutionarily conserved motif. Nucleic Acids Res 21: 11931198.
  • 61
    Siomi, H., Choi, M., Siomi, M.C., Nussbaum, R.L., Dreyfuss, G. (1994) Essential role for KH domains in RNA binding: impaired RNA binding by a mutation in the KH domain of FMR1 that causes fragile X syndrome. Cell 77: 3339.
  • 62
    Subramanian, A.R. (1983) Structure and functions of ribosomal protein S1. Prog Nucleic Acid Res Mol Biol 28: 101142.
  • 63
    Swindle, J., Zylicz, M., Georgopoulos, C., Li, J., Greenblatt, J. (1988) Purification and properties of the NusB protein of Escherichia coli. J Biol Chem 263: 1022910235.
  • 64
    Tan, R. & Frankel, A.D. (1995) Structural variety of arginine-rich RNA-binding peptides. Proc Natl Acad Sci USA 92: 52825286.
  • 65
    Tsugawa, A., Saito, M., Court, D.L., Nakamura, Y. (1988) nusA amber mutation that causes temperature-sensitive growth of Escherichia coli. J Bacteriol 170: 908915.
  • 66
    Ward, D.F. & Gottesman, M.E. (1981) The nus mutations affect transcription termination in Escherichia coli. Nature 292: 212215.
  • 67
    Whalen, W.A. & Das, A. (1990) Action of an RNA site at a distance: role of the nut genetic signal in transcription antitermination by phage-lambda N gene product. New Biol 2: 975991.
  • 68
    Whalen, W., Ghosh, B., Das, A. (1988) NusA protein is necessary and sufficient in vitro for phage lambda N gene product to suppress a rho-independent terminator placed downstream of nutL. Proc Natl Acad Sci USA 85: 24942498.
  • 69
    Zhang, O., Kay, L.E., Olivier, J.P., Forman-Kay, J.D. (1994) Backbone 1H and 15N resonance assignments of the N-terminal SH3 domain of drk in folded and unfolded states using enhanced-sensitivity pulsed field gradient NMR techniques. J Biomol NMR 4: 845858.
  • 70
    Zheng, C. & Friedman, D.I. (1994) Reduced Rho-dependent transcription termination permits NusA-independent growth of Escherichia coli. Proc Natl Acad Sci USA 91: 75437547.