SEARCH

SEARCH BY CITATION

Abstract

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

Transcription from the bop promoter in the haloarchaeon Halobacterium NRC-1, is highly induced under oxygen-limiting conditions. A DNA gyrase inhibitor, novobiocin, was previously shown to block bop gene induction and suggested that DNA supercoiling mediates transcriptional induction. A region of non-B structure was found 3′ to the TATA box within an 11 bp alternating purine–pyrimidine sequence (RY box), which correlated to both increased DNA supercoiling and transcriptional induction. Here, saturation mutagenesis of the RY box region has been used to show that single-base substitutions of A(r)G either 23 or 19 bp 5′ to the transcription start site temper the effect of DNA supercoiling based on novobiocin insensitivity of transcription. Mutagenesis of the region 5′ to the TATA box showed its involvement in DNA supercoiling modulation of transcription, defined the 3′ end of the upstream activator sequence (UAS) regulatory element, and ruled out the requirement for a TFB (TFIIB) Recognition Element. Spacing between the TATA box and UAS was found to be critical for promoter activity because insertion of partial or whole helical turns between the two elements completely inhibited transcription indicating that the UAS element does not function as a transcriptional enhancer. The results are discussed in the context of DNA melting and flexibility around the TATA box region and the involvement of multiple regulatory and transcription factors in bop promoter activity.


Introduction

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

Halobacterium species induce synthesis of bacteriorhodopsin, a protein–retinal complex, under conditions of low oxygen concentration and high light intensity (Yang and DasSarma, 1990; Shand and Betlach, 1991). The protein component of bacteriorhodopsin, bacterio-opsin, is encoded by the bop gene, one of the first archaeal genes to be cloned (Chang et al., 1981; Dunn et al., 1981). Two genes upstream to bop, brp and bat, were found to be involved in bop gene expression because insertions in these genes abolished or greatly lowered bop transcript levels (Betlach et al., 1984; Leong et al., 1988; Yang and DasSarma, 1990). The homology between the bat gene product and the flavin adenine dinucleotide-binding region of the bacterial transcriptional regulator, NifL, suggested a putative redox sensing function for Bat (Yang and DasSarma, 1990; Dixon, 1998). In contrast, the function of Brp, a putative hydrophobic protein, is unknown.

Transcription of bop is initiated two nucleotides upstream to the start codon (DasSarma et al., 1984) and the fully functional minimal promoter is contained within 53 bp upstream to the start site (Gropp et al., 1995; Yang et al., 1996). A region of the bop promoter, the upstream activator sequence (UAS; positioned −52 to −39) conserved among several co-ordinately regulated genes (Gropp et al., 1995) was analysed by saturation mutagenesis and found to be required for bop gene transcription (Baliga and DasSarma, 1999). Saturation mutagenesis of the TATA box region (positioned approximately −31 to −25) also showed its requirement, although its sequence was more mutable and shifted several nucleotides closer to the transcription start point. A consensus sequence was derived that was more AT rich and similar to other archaeal consensus sequences than to the wild-type promoter sequence, suggesting possible use of an alternative TATA-binding protein (TBP) in promoter recognition (Baliga and DasSarma, 1999; Baliga et al., 2000). The existence of multiple TBP genes has been shown in the Halobacterium NRC-1 genome project with four genes in a 191 kb mini chromosome (Ng et al., 1998).

An interesting region of the bop promoter, an 11 bp alternating purine–pyrimidine sequence overlapping the TATA box by 4 bp (RY box, Fig. 1B), was shown to adopt a non-B-DNA structure under conditions of high DNA supercoiling (Yang et al., 1996). Because lowered oxygen availability results in increased DNA supercoiling in facultative anaerobic organisms like Halobacterium (Pruss and Drlica, 1989; Yang et al., 1996), supercoiling is believed to be a mediator for bop gene regulation, in response to changes in dissolved oxygen concentration. The non-B-DNA structure, identified in the centre of the RY box element (Fig. 1B), results in a short region (2–3 bases) susceptible to a single-strand-specific nuclease immediately downstream to the TATA box. Addition of novobiocin, a DNA gyrase inhibitor, relaxes DNA supercoiling and also reduces transcription from the bop promoter allowing for experimental analysis of the DNA supercoiling effect (Yang and DasSarma, 1990; Yang et al., 1996). Mutagenic analysis of the UAS and TATA box regions showed that they are not involved in the observed DNA supercoiling effects (Baliga and DasSarma, 1999).

image

Figure 1. Map of the bop gene cluster, the minimal bop promoter and promoter mutagenesis.

A. The relative sizes and arrangement of the four genes, brp, bat, blp and bop, in a gene cluster on the chromosome is shown in boxes with position and orientation of the promoters (indicated by arrows above).

B. The sequence of nucleotides within the minimal bop promoter and a few surrounding nucleotides with the TATA box, UAS and RY box, boxed. The start codon is underlined and the transcription start site (indicated by an arrow) is numbered +1. The three arrows within the RY box indicate the sites of S1 nuclease sensitivity. The 7 bp region between UAS and TATA box (UB sequence) is shown with a horizontal bracket on top. Identity and position of mutations introduced (−36 CG −35 to −36 TC −35) for engineering an XbaI site (−36 thru −31) are shown below the nucleotide sequence.

C. E. coliHalobacterium shuttle plasmid (pNB series) with the cloned bop gene (black arrow) and promoter (white box). The plasmid contains origins of replication for E. coli (ori ColE1) and Halobacterium (ori pNRC100, repH) in addition to selectable markers (bla for Ampr in E. coli, and mev for Mevr in Halobacterium[shaded arrows]).

Download figure to PowerPoint

In order to study the bop promoter further, we have conducted extensive mutagenesis of the regions flanking the TATA box region. The results show that nucleotides in the RY box are indeed responsible for modulating bop transcription in response to DNA supercoiling. We have also tested mutations in the region upstream to the TATA box and found the absence of a Transcription Factor B (TFB) Recognition Element (BRE), which has been shown to orient the transcription machinery correctly in other archaeal (and eukaryotic) promoters (Qureshi and Jackson, 1998). Finally, the spatial arrangement of the UAS and TATA box in the bop promoter has been studied by insertion of DNA between the two elements, which rules out an enhancer activity for the UAS.

Results

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

RY box mutagenesis

Saturation mutagenesis of the 7 bp of the RY box located 3′ to the TATA box (−24 to −18; Fig. 1B) by PCR yielded nearly 100% purple colonies in transformants of Halobacterium SD23. Because the phenotypic screen does not easily discriminate between fully induced and partially induced bop gene expression, we analysed 11 RY box mutants further by primer extension analysis. This method of estimation of transcript levels by primer extension was previously shown to be quantitative by mixing experiments (Yang and DasSarma, 1990; see Experimental procedures). Although sequencing of RY box mutants showed no statistically significant bias in the nucleotide distribution, when analysed for transcriptional regulation mediated by changes in DNA supercoiling, the promoters were all relatively less susceptible to inhibition by novobiocin (Fig. 2). This effect was consistently observed at both the transcript and protein levels. Therefore, the RY box was likely to function in transcriptional regulation, although specific nucleotides could not be identified as a result of multiple mutations in all cases. Because the bop message levels in all the RY box mutants remained consistently below the fully induced wild-type levels even in the absence of novobiocin, this provided further evidence that increased DNA supercoiling boosts transcription from the bop promoter above basal levels in conditions of limiting oxygen. Although transcription was deregulated in the RY box mutants, it was initiated at the same nucleotide as in the wild type (Fig. 2C).

image

Figure 2. Randomization of the RY box region. Strain designations, promoter sequences, phenotypes, bop transcript levels and/or BR content of RY box randomized mutants (A) and tally of nucleotides at all nucleotide positions in active promoters (B) are indicated. In panel A, the strain designations are shown in the first column, and sequence of nucleotides −24 to −18 (identity to the wild type base is denoted by a dot), the Pum phenotype (–, orange; +, light purple and ++, dark purple), and relative bop transcript ([bop]) and/or BR levels ([Bop]) in cultures grown with or without novobiocin (–/+novo) are shown to the right. *, not carried out. For panel B, distribution of individual nucleotides at each position, in active promoters, was determined. Panel C shows the results of primer extension analysis of the bop messenger.

Download figure to PowerPoint

To determine specific nucleotides functioning in DNA supercoiling-mediated regulation, we mutated the seven bases of the RY box individually using a PCR-based site-directed strategy. All 21 point mutations in this set resulted in purple [Pum+ (light purple) or Pum++ (dark purple)] colonies (Fig. 3) and primer extension analysis showed that the transcription start site remained unchanged (Fig. 4A). Analysis of transcript levels ± novobiocin showed the negative effect of the drug on transcription was drastically reduced with A(r)G transitions at positions −23 and −19 (Fig. 4). Mutagenesis of all other positions except −21 also resulted in moderation of the novobiocin effect.

image

Figure 3. Site-directed mutagenesis of the RY box region. Strain designations, promoter sequences, phenotypes and bop transcript content ([bop]) of RY box site-directed mutants are indicated. The strain designations are shown in the first column and sequence of nucleotides −24 to −18 (identity to the wild-type base is denoted by a dot), the Pum phenotype (–, orange; +, light purple and ++, dark purple), and relative bop transcript levels ([bop]) in cultures grown with or without novobiocin (–/+novo) are shown to the right. *, not carried out.

Download figure to PowerPoint

image

Figure 4. Analyses of bop transcripts and promoter strength in RY box point mutants by primer extension.

A. bop transcription start sites in the RY box point mutants were mapped by fractionating primer extension products adjacent to lane C of a sequencing ladder generated with the same primer, bop2, and plasmid, pMS1, containing the cloned bop gene. The identities of mutations are indicated above the gel picture.

B. The strength of the mutated promoters and the effect of novobiocin on their regulation were determined by quantifying the primer extension products by phosphoimager analysis. The relative transcript levels in all strains with respect to 100E levels are plotted on the Y-axis against the nature of mutations and culturing conditions (–/+novobiocin) on the X-axis.

C. Promoter strengths in the presence of novobiocin were calculated as a percentage of transcript levels in cultures grown without novobiocin for, e.g. the RY23-G promoter functions at 79% of its total capacity in the presence of novobiocin.

Download figure to PowerPoint

Mutagenesis between the UAS and TATA box

The −38 to −32 region, referred to as the ‘UB’ region, which includes the 7 bp between the UAS and the TATA box was randomized by PCR. Approximately 26% of SD23 transformants were of purple phenotype, suggesting one highly conserved nucleotide within the seven mutagenized bases. The G at −38, within the right extremity of the predicted UAS, was conserved in 15 out of 15 active promoters (Fig. 5B). Conversely, all inactive promoters (e.g. UB2–, UB3– and UB4–) did not have a G at −38. The transcription start site was confirmed by primer extension analysis to be the same irrespective of the nature of the mutation (data not shown). Of 11 mutants, the effect of novobiocin was significantly altered in two mutants viz. UB1 and UB10 (Fig. 5). UB1 and UB10 both contained a −36C(r)T transition, although neither was a single mutant. The mutations in the UB region had no effect on direction of transcription because primer extensions, from the opposite end using the universal T7 primer, did not reveal any transcripts in the reverse orientation (data not shown).

image

Figure 5. Randomization of the UB region. Strain designations, promoter sequences, phenotypes, bop transcript levels ([bop]) of UB sequence randomized mutants (A) and a tally of individual nucleotides at each position in active promoters (B) are indicated. In panel A the strain designations are shown in the first column and sequence of nucleotides −38 to −32 (identity to the wild-type base is denoted by a dot), the Pum phenotype (–, orange; +, light purple and ++, dark purple), and relative bop transcript levels ([bop]) in cultures grown with or without novobiocin (–/+novobiocin) are shown to the right. *, not carried out. In panel B, the consensus sequence for nucleotide positions −38 to −32 was determined by tallying the individual nucleotides at each position.

Download figure to PowerPoint

Spacing analysis: is the UAS an enhancer?

To test the UAS for enhancer-like functions we increased the distance to the TATA box by insertion of helical half- or full-turns. First, the CG dinucleotide between the TATA box and UAS, −36 to −35, was mutated to TC to introduce an XbaI site between the two elements (Fig. 1B). Insertions of varying lengths (4 bp, 14 bp, multimers of 10 bp and 11 bp HindIII linkers with XbaI overhangs, and a 733 bp HindIII fragment) were introduced into this site. Using this strategy, the distance between the UAS and TATA box elements was increased by half helical turns up to two turns (4, 10, 11, 14, 20 and 22 bp). Additional multimers of 10 and 11 bp were also introduced increasing the distance up to 99 bp (30, 33, 40, 44, 50, 55 and 99 bp). Finally, a large fragment [the 733 bp HindIII M fragment from pNRC100 (Ng et al., 1998)] was inserted into the HindIII site of a construct containing a single 10 bp monomer to increase the distance by 747 bp.

The effect of the insertions on promoter function was analysed phenotypically in Halobacterium SD23 and by primer extension analysis. All of the mutated promoters, except the engineered XbaI containing bop promoter, produced Pum phenotypes. Mutagenesis of the two bases, for construction of the XbaI site (bop-X in Fig. 6), resulted in 60% reduction in bop transcript levels compared with the wild type (100E) and insertion of only 4 bp (U4B in Fig. 6) into this site reduced transcription to undetectable levels (95% reduction of 100E levels). No transcripts could be detected from any of the other mutated promoters [e.g. those with insertions of 10 bp (U10B) and 11 bp (U11B) in Fig. 6 and data not shown].

image

Figure 6. Primer extension analysis of bop transcripts in Halobacterium SD23 transformed with pNB100E (100E), bop promoter plasmid with an engineered XbaI site (pNBbop-XbaI; bop-X), and mutants with increased spacing between UAS and TATA box (4 bp, U4B; 10 bp, U10B; 11 bp, U11B), respectively, by primer extension analysis.

Download figure to PowerPoint

Discussion

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

The bop promoter has been extensively studied by genetic approaches. The 53 bp minimal promoter was defined by deletion analysis and shown to be responsive to oxygen tension and DNA supercoiling (Gropp et al., 1995; Yang et al., 1996). Saturation mutagenesis was used to show the importance of a TATA box element located 30–25 bp upstream and a UAS element located 52–39 upstream from the transcription start point (Baliga and DasSarma, 1999). In the present study, we have conducted saturation mutagenesis on regions flanking the TATA box, including the 7 bp alternating Pu–Py (RY-box) sequence 3′ to the TATA box. In total, 35 bp out of the 53 bp within the minimal promoter have been randomized. In the course of this analysis, we have shown that both the TATA box and UAS are required for basal expression, with the 5′ and 3′ ends of the UAS being highly immutable. The TATA box consensus, derived by saturation mutagenesis, although very similar to other archaeal TATA box sequences, is significantly different from the wild type, unmutated, promoter sequence, suggesting that an alternative form of TBP may be used for bop gene transcription.

bop gene transcription was shown to be sensitive to novobiocin and therefore to DNA supercoiling (Yang and DasSarma, 1990). In subsequent mutagenic analysis, neither the TATA box nor the UAS was found to be responsible for the observed effects of DNA supercoiling on bop gene transcription. However, several mutations in the RY box, which was found to harbour a supercoiling-dependent S1 nuclease and OsO4-sensitive site (Yang et al., 1996) showed reduced sensitivity to novobiocin. The RY box region has now been systematically mutated to give a total of 21 single base changes corresponding to all mutations at seven nucleotide positions. Interestingly, an A[RIGHTWARDS ARROW]G transition at −23 (from the transcription start point), a major site for cleavage by S1 nuclease and OsO4, showed 79% transcription level in the presence of novobiocin and an A[RIGHTWARDS ARROW]G transition at −19 showed 94% transcription with novobiocin. Four of the randomized mutants with relatively novobiocin-insensitive bop expression shared these same base changes (ZB8 and ZD11 for −23 and ZD1 and ZD2 for −19). We speculate that replacement of the −23 A–T bp with a G–C bp, and possibly an A[RIGHTWARDS ARROW]G transition at −19, might be perturbing the propensity to form non-B-DNA structure and/or melting of the helix at this position. It has recently been demonstrated that modulating DNA flexibility and/or for opening the helix, in the vicinity of or downstream to the core promoter, can increase transcription by enhancing both TATA box-TBP interactions and open complex formation (Grove et al., 1998; Li and McClure, 1998). Hence, the change in DNA structure in the RY box may be assisting in binding of transcription factors and/or open complex formation by nucleating the opening of the helix in the bop promoter.

Several randomized mutants in the region between the UAS and TATA box elements, −38 to −32 to the bop transcription start point (UB1, UB9 and UB10) had a significantly lowered sensitivity to novobiocin (up to 80% reduction) on bop gene transcription. Reduction of sensitivity, in these cases, may be explained by the requirement for critical spacing between the two elements for promoter activity (below). This property may be a function of supercoiling. For example, the relative geometry of factors bound at the UAS and TATA box may be fine tuned by modulation of DNA supercoiling and or twisting. Hence, mutagenesis of the intervening region may be detrimental to this regulatory mechanism.

Based on the observation that 26% of the UB library transformants were purple, we could predict that one out of seven nucleotides in the UB sequence was highly conserved. This was confirmed upon sequence analysis, which showed that nucleotide −38G, a part of the predicted UAS element, is absolutely required for a functional promoter. All the active promoters (15/15) had a ‘G’ at this position and the likelihood of occurrence of such an event by chance was calculated to be one in 4.3 × 109. This finding extends the UAS consensus sequence to −52(5′-ACCcnactagTTnGG-3′)−38 (Baliga and DasSarma, 1999).

We tested the UAS for enhancer-like activity by physically moving it away from the TATA box by half or full helical turns. The strategy used for making these constructs allowed for the UAS to be in phase or out of phase with the TATA box over relatively short distances. We also inserted a larger DNA fragment to allow the UAS to potentially loop back. However, no promoter activity could be detected in any of the constructs with increased spacing between the UAS and TATA box. This leads us to conclude that the bop gene UAS is not an enhancer element. This is somewhat surprising given that the Halobacterium transcription machinery resembles the eukaryotic RNA polymerase II system in many ways (Reiter et al., 1990; Thomm, 1996). The UAS may recruit a transcriptional regulator, which interacts with the transcription factors and/or the RNA polymerase bound to or around the TATA box in a manner that is topologically constrained.

A region upstream of the TATA box has been shown to affect promoter strength by determining the orientation of transcription (Lagrange et al., 1998; Qureshi and Jackson, 1998). This region, referred to as the TFIIB(TFB) Recognition Element (BRE), interacts with TFIIB (TFB) and renders asymmetry to the otherwise symmetrical TATA box resulting in the preinitiation complex to be formed in a way, which allows transcription to continue in only one direction (Lagrange et al., 1998). By determining the orientation of transcription, this element also controls for strength of the promoter (Qureshi and Jackson, 1998). In the case of archaea, BREs have been identified experimentally in in vitro systems and also by sequence analysis; and the preferred sequence for TFB interactions with the promoter was predicted to be −36(5′-gNAA-3′)−33 or more generally −36(5′-PuNPu-3′)−34 (Qureshi and Jackson, 1998; Soppa, 1999). Surprisingly, we did not find any sequence bias in active promoters analysed from the randomized UB mutant library that was in agreement with the predicted BRE consensus sequences. In fact, although one mutant, UB6, has an exact match to the predicted consensus, the promoter strength is about one-third that of the wild-type promoter. Also, we did not detect transcripts in the opposite orientation in any of the UB mutants carrying mutations in the BRE region. However, the average level of transcription was significantly lower than the wild-type levels (Fig. 6) (for example, > 10-fold reduction in transcription was seen in the case of UB14) (Fig. 2). This observation suggests a role for the sequence in this region and may reflect a higher degree of complexity for transcription from the bop promoter.

Results thus far suggest that transcription from the bop promoter requires interplay between DNA supercoiling-mediated structural changes, TBP and possibly TFB factors, one or two regulators (Bat and perhaps Brp) and RNA polymerase. The finding of four TBP genes (Ng et al., 1998) and multiple TFB genes in the Halobacterium NRC-1 genome (Baliga et al., 2000), as well as multiple TBP and TFB factors in a related halophile (Thompson and Daniels, 1998), suggests that the system has potential for utilization of what are usually considered general transcription factors for gene regulation (Buratowski, 1997). The availability of mutants in the most critical residues of the bop promoter generated in this study will be extremely helpful for further advancing the understanding of the complex regulation of this archaeal promoter by a combination of biochemical and genetics experiments.

Experimental procedures

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

Halobacterium strains and culturing

Halobacterium S9, a purple membrane (Pum) constitutive strain and SD23, a Pum derivative of S9 with an ISH1 insertion at the 5′ end of bop, have been previously described (Yang et al., 1996). Culturing of these Halobacterium strains was carried out at 37°C in complete medium (CM) containing 4.5 M NaCl and trace metals as previously described. Culturing for studying effects of treatment with novobiocin (Sigma) was conducted according to Yang et al. (1996). Briefly, novobiocin was added to cultures, grown to an OD600 of 0.2 (early log phase) and to a final concentration of 0.05 µg ml−1; the cultures were allowed to grow further, in the presence of the drug, to stationary phase (OD600 > 1.8) at which point the cells were harvested for RNA preparations (see below).

Saturation mutagenesis

Saturation mutagenesis of the promoter was accomplished by PCR amplification of the cloned bop gene in pMS1 (Dunn et al., 1981) with mutagenic primers having degeneracies in seven positions corresponding to the region to be mutated [primer UINTR for UB sequence (UB is the region −38 to −32 to the transcription start site) and primer NB7 for the RY box (−24 to −18 to the transcription start site); see Table 1 for sequences of oligonucleotides] and a non-mutagenic primer, NB3, located downstream of the bop gene. All synthetic oligonucleotides were purchased from Genosys Biotechnologies. The PCR amplifications were carried out using Taq polymerase and standard conditions on the PTC-100 programmable thermal controller (MJ Research).

Table 1. Sequences of oligonucleotides used in this study.
NameSequence (5′-3′)
NB3G G G A A T T C T A C A A G A C C G A G T G G
NB7T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A N N N N N N N C C T C G T T A G G T A C T G
RY-24T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A N A C A T A T C C T C G T T A G G T A
RY-23T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C N C A T A T C C T C G T T A G G T A
RT-22T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C A N A T A T C C T C G T T A G G T A
RT-21T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C A C N T A T C C T C G T T A G G T A C T
RY-20T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C A C A N A T C C T C G T T A G G T A C T
RY-19T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C A C A T N T C C T C G T T A G G T A C T
RY-18T C G C G G A T C C G C G T A C C A T A C T G A T T G G G T C G T A G A G T T A C A C A C A T A N C C T C G T T A G G T A C T G T T G
UINTRT C G C G G A T C C G C G T A C C A T A C T G A T T G G N N N N N N N A G T T A C A C A C A T A T C C T C
bop2C C T G C G A T A C C C C C T
NB-XbaIT C G C G G A T C C G C G T A C C A T A C T G A T T G G G T T C T A G A G T T A C A C A C A T A T C C T C G T T A G G T A
U6BC T A G A A G C T T
U7B-1C T A G A A A G C T T
U7B-2C T A G A A G C T T T
M13 forwardC G C C A G G G T T T T C C C A G T C A C G A C
T7A A T A C G A C T C A C T A T A G

The products obtained from the PCR amplification reactions with mutagenic primers were fractionated on a 0.8% agarose gel, gel purified using the Qiaex II kit (Qiagen) and digested with fivefold excess of each EcoRI and BamHI (New England Biolabs). The digested inserts were repurified after fractionation on a second 0.8% agarose gel and used for cloning. The vector, a Halobacterium–Escherichia coli shuttle plasmid, pNB148, a pNG168 derivative (See Fig. 1C), was prepared by removing a 1.2 kb EcoRI–BamHI stuffer fragment and subsequently gel purified with the Qiaex II kit. The insert DNAs (0.02 µg) were ligated to this vector (0.05 µg) and electroporated into Electromax E. coli DH10B cells (Gibco-BRL). A small fraction, 0.4%, of the transformation was plated out on (Luria–Bertani) LB agar plates containing ampicillin (100 µg ml−1) to determine the efficiency of cloning and the remainder was amplified in a 1-L culture. The libraries were amplified in E. coli DH10B and plasmid DNA was prepared on a large scale from a 1 litre culture by the alkaline lysis method. Plasmid DNA was purified on a cesium–chloride (CsCl) density gradient and quantified spectrophotometrically. All standard recombinant DNA procedures used have been previously described (Sambrook et al., 1989).

DNA from the amplified libraries was transformed into the PumHalobacterium strain SD23 using the PEG–EDTA transformation method (Cline and Doolittle, 1987). Halobacterium transformants were selected on CM plates containing 16 µg ml−1 mevinolin. A total of 25 000–30 000 colonies was obtained in each case and was considered a good representation for the library (each library can have a maximum of 47 or 16 384 different sequences in the randomized region). The purple colony (Pum+, Pum++) phenotype was used as a screen to select candidates for further analysis. The plasmids from mevinolin resistant Halobacterium SD23 colonies were recovered by electroporation of E. coli DH5α with halobacterial total DNA. Plasmid DNA extracted from single isolated E. coli colonies was retransformed into Halobacterium SD23 to confirm the observed ‘Pum’ phenotype.

Site-directed mutagenesis

The seven nucleotides in the RY box were mutated one at a time using seven different mutagenic oligonucleotides (Table 1) in conjunction with NB3 (above) in separate PCR reactions. The PCR products were cloned as described above. Each mutagenesis experiment yielded three mutated and one wild-type promoter giving a total of 21 point mutations in the RY box. Plasmids containing point mutations, purified from E. coli DH5α, were used for transformations into Halobacterium SD23 to determine the phenotype.

Insertion mutagenesis

In order to insert a unique XbaI site between the UAS and TATA box, the bop gene was amplified with NB3 and a mutagenic primer NB-XbaI (Table 1). The amplified bop gene had two point mutations at positions −36 and −35, which resulted in an XbaI recognition site (5′-TCTAGA-3′). pNBbop-XbaI was constructed by ligating the mutated, EcoRI and BamHI digested bop gene PCR product to pNB148ΔXbaI digested with the same enzyme. pNB148ΔXbaI is a derivative of pNB148 where an XbaI site in the multiple cloning site (MCS) was mutated by digestion with XbaI followed by a Klenow mediated fill-in reaction and religation.

Synthetic 10 bp and 11 bp HindIII linkers were prepared by annealing a partially complementary 6 bp oligonucleotide (U6B) (Table 1) to itself and two 7 bp oligonucleotides (U7B1 and U7B2, also partially complementary to each other) (Table 1) respectively. The T4 kinase (New England Biolabs) phosphorylated linkers with XbaI overhangs were ligated in 100-fold molar excess to dephosphorylated, XbaI digested pNBbop-XbaI to get multiple inserts. After E. coli electroporation into DH5α, plasmids were prepared by the alkaline lysis method and screened for the presence of single or multiple inserts by sequencing.

Cloning of a 733 bp HindIII fragment between the UAS and TATA box was achieved by ligating HindIII digest of total Halobacterium NRC-1 genomic DNA to pNBU10B (pNBbop-XbaI containing a single 10 bp HindIII linker) digested with the same enzyme. After transformation into E. coli DH5α and screening by sequencing plasmid DNA prepared by the alkaline lysis method the identity of the insert as a 733 bp pNRC100 HindIII fragment was determined by sequence analysis.

Bacteriorhodopsin assays

Halobacterium cultures grown to an OD600 of ≈ 1.8 were used for preparation of purple membrane by the method described by Oesterhelt (1995). Cells from 50 ml of each culture were harvested by centrifugation at 7000 r.p.m. for 20 min. The cell paste was resuspended in 4 ml of basal salts solution containing 20 µg ml−1 DNase I. The cells were gradually lysed by overnight dialysis (Spectrum Labs) against distilled water and the lysate was then analysed spectrophotometrically at 568 nm to quantify BR content. The absorption values were normalized to an OD280 value of 1.5 and corrected by subtracting out the background (normalized reading for host Halobacterium sp. strain SD23).

Primer extension analysis

Crude RNA was prepared from 1.5 ml of Halobacterium cultures using the RNeasy kit (Qiagen). Primer extension was performed on approximately 10 µg of RNA, using end-labelled bop2 primer (or the universal T7 primer for the opposite strand) and the Primer extension kit (Promega) according to the manufacturer's instructions. End-labelling was carried out using [γ-32P]-ATP (Amersham Life Science or NEN Life Science Products) in a T4 polynucleotide kinase (New England Biolabs) catalysed reaction. The cDNA product was analysed by electrophoresis in denaturing conditions on a 6% polyacrylamide/8.3 M urea gel on a Genomyx LR sequencer (Beckman Coulter). Band intensities were quantified by phosphoimager analysis of the gel using a Cyclone Storage Phosphor Imaging system and the Optiquant imaging software (Packard).

Some batch-to-batch variation was observed for novobiocin sensitivity (compare results for wild-type promoter 100E in Figs 2A, 3 and 5). However, it is notable to mention that all experiments were carried out double blind in duplicate or triplicate with freshly prepared RNA. Results were consistent and reproducible for a given batch of novobiocin.

Sequence analysis

Plasmids were prepared by the alkaline lysis method from cultures of E. coli DH5α transformants for sequencing. The universal forward and bop2 oligonucleotides were used to sequence both the strands by the dideoxy cycle-sequencing method (Sanger et al., 1977; Reeve and Fuller, 1995) using either the radio-isotope-based chemistry (for primer extension experiments) (Sequenase version 2.0 Sequencing kit, Amersham Life Science) on a Genomyx LR sequencer or the fluorescent dye-terminator chemistry (Thermosequenase dye terminator cycle sequencing premix kit, Amersham Life Science) on an ABI373A sequencer (Perkin Elmer). Sequences were analysed using the gcg software package (Genetics Computer Group) running on an SGI O2 workstation (Silicon Graphics).

Acknowledgements

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

We thank R. Niang for assistance with mutagenesis experiments.

This work was supported by NSF grant MCB-9604443 and MCB-9812330.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Baliga, N.S. & DasSarma, S. (1999) Saturation mutagenesis of the TATA box and the upstream activator sequence in the haloarchaeal bop gene promoter. J Bacteriol 181: 25132518.
  • 2
    Baliga, N.S., Goo, Y., Ng, W.V., Hood, L., Daniels, C.J., DasSarma, S.(2000) Is gene expression in Halobacerium NRC-1 regulated by multiple TBP and TFB transcription factors? Mol Microbiol 36: 11841185.
  • 3
    Betlach, M., Friedman, J., Boyer, H.W., Pfeifer, F. (1984) Characterization of a halobacterial gene affecting bacterio-opsin gene expression. Nucleic Acids Res 12: 79497959.
  • 4
    Buratowski, S. (1997) Multiple TATA-binding factors come back into style. Cell 91: 1315.
  • 5
    Chang, S.H., Majumdar, A., Dunn, R., Makabe, O., RajBhandary, U.L., Khorana, H.G., et al. (1981) Bacteriorhodopsin: partial sequence of mRNA provides amino acid sequence in the precursor region. Proc Natl Acad Sci USA 78: 33983402.
  • 6
    Cline, S. & Doolittle, W.F. (1987) Efficient transfection of the archaebacterium Halobacterium halobium. J Bacteriol 169: 13411344.
  • 7
    Danner, S. & Soppa, J. (1996) Characterization of the distal promoter element of halobacteria in vivo using saturation mutagenesis and selection. Mol Microbiol 19: 12651276.
  • 8
    DasSarma, S., RajBhandary, U.L., Khorana, H.G. (1984) Bacterio-opsin mRNA in wild-type and bacterio-opsin deficient Halobacterium halobium strains. Proc Natl Acad Sci USA 81: 125129.
  • 9
    Dixon, R. (1998) The oxygen-responsive NIFL-NIFA complex: a novel two-component regulatory system controlling nitrogenase synthesis in γ-proteobacteria. Arch Microbiol 169: 371380.
  • 10
    Dunn, R., McCoy, J., Simsek, M., Majumdar, A., Chang, S.H., Rajbhandary, U.L., et al. (1981) The bacteriorhodopsin gene. Proc Natl Acad Sci USA 78: 67446748.
  • 11
    Gropp, F., Gropp, R., Betlach, M.C. (1995) Effects of upstream deletions on light- and oxygen-regulated bacterio-opsin gene expression in Halobacterium halobium. Mol Microbiol 16: 357364.
  • 12
    Grove, A., Galeone, A., Yu, E., Mayol, L., Geiduschek, P. (1998) Affinity, stability and polarity of the TATA binding protein governed by flexure at the TATA box. J Mol Biol 282: 731739.
  • 13
    Lagrange, T., Kapanidis, A.N., Tang, H., Reinberg, D., Ebright, R.H. (1998) New core promoter element in RNA polymerase II-dependent transcription: sequence-specific DNA binding by transcription factor IIB. Genes Dev 12: 3444.
  • 14
    Leong, D., Boyer, H., Betlach, M. (1988) Transcription of genes involved in bacterio-opsin gene expression in mutants of a halophilic archaebacterium. J Bacteriol 170: 49104915.
  • 15
    Li, X.-Y. & McClure, W.R. (1998) Stimulation of open complex formation by nicks and apurinic sites suggests a role for nucleation of DNA melting in Escherichia coli promoter function. J Biol Chem 273: 2355823566.
  • 16
    Ng, W.-L., Yang, C.-F., Halladay, J.T., Arora, P., DasSarma, S. (1995) Isolation of genomic and plasmid DNAs from Halobacterium halobium. In Archaea: A Laboratory Manual. Halophiles. DasSarma, S., Robb, F.T., Place, A.R., Sowers, K.R., Schreier, H.J., and Fleischmann, E.M. (eds). Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, pp. 179184.
  • 17
    Ng, W.-L., Ciufo, S.A., Smith, T.M., Bumgardner, R.E., Baskin, D., Faust, J., et al. (1998) Snapshot of a dynamic replicon in a halophilic archaeon: megaplasmid or minichromosome? Genome Res 8: 11311141.
  • 18
    Oesterhelt, D. (1995) Isolation of purple membranes. In Archaea: A Laboratory Manual. Halophiles. DasSarma, S., Robb, F.T., Place, A.R., Sowers, K.R., Schreier, H.J., and Fleischmann, E.M. et al. (eds). Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press, pp. 5557.
  • 19
    Pruss, G.J. & Drlica, K. (1989) DNA supercoiling and prokaryotic transcription. Cell 56: 521523.
  • 20
    Qureshi, S.A. & Jackson, S.P. (1998) Sequence-specific DNA binding by the S. shibatae TFIIB homolog, TFB, and it's effect on promoter strength. Mol Cell 1: 389400.
  • 21
    Reeve, M.A. & Fuller, C.W. (1995) A novel thermostable polymerase for DNA sequencing. Nature 376: 796797.
  • 22
    Reiter, W.-D., Hüdepohl, U., Zillig, W. (1990) Mutational analysis of an archaebacterial promoter: Essential role of a TATA box for transcription efficiency and start site selection in vitro. Proc Natl Acad Sci USA 87: 95099513.
  • 23
    Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 24
    Sanger, F., Nicklen, S., Coulson, A.R. (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74: 54635467.
  • 25
    Shand, F.R. & Betlach, M.C. (1991) Expression of the bop gene cluster of Halobacterium halobium is induced by low oxygen tension and by light. J Bacteriol 173: 46924699.
  • 26
    Soppa, J. (1999) Transcription initiation in Archaea: facts, factors and future aspects. Mol Microbiol 31: 12951305.
  • 27
    Thomm, M. (1996) Archaeal transcription factors and their role in transcription intiation. FEMS Microbiol Rev 18: 159171.
  • 28
    Thompson, D.K. & Daniels, C.J. (1998) Heat shock inducibility of an archaeal TATA-like promoter is controlled by adjacent sequence elements. Mol Microbiol 27: 541551.
  • 29
    Yang, C.-F. & DasSarma, S. (1990) Transcriptional induction of purple membrane and gas vesicle synthesis in the archaebacterium Halobacterium halobium is blocked by a DNA gyrase inhibitor. J Bacteriol 172: 41184121.
  • 30
    Yang, C.-F., Kim, J.-M., Molinari, E., DasSarma, S. (1996) Genetic and topological analyses of the bop promoter of Halobacterium halobium: stimulation by DNA supercoiling and non-B-DNA structure. J Bacteriol 178: 840845.