In Vivo Definition of an Archaeal Promoter - Journal of Bacteriology

JOURNAL OF BACTERIOLOGY, Apr. 1995, p. 1844–1849 0021-9193/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 177, No. 7

In Vivo Definition of an Archaeal Promoter JOHN R. PALMER

AND

CHARLES J. DANIELS*

Department of Microbiology, The Ohio State University, Columbus, Ohio 43210 Received 12 October 1994/Accepted 23 January 1995

We have used a plasmid-based transcriptional reporter system to examine the transcriptional effects of 33 single point mutations in the box A region (TATA-like sequence) of the Haloferax volcanii tRNALys promoter. The most pronounced effects on transcriptional efficiency were found when the nucleotides corresponding to the TATA-like region were altered. Promoters with wild-type or higher levels of transcriptional activity conformed to the general archaeal box A consensus, 5*-T/CTTAT/AA-3*. The preference for a pyrimidine residue in the 5* position of this region and the exclusion of guanine and cytosine in the next four positions in the 3* direction are defining characteristics shared by all efficient archaeal promoters. We have also observed that replacement of a 10-nucleotide purine-rich sequence, located 5* of the H. volcanii tRNALys box A element, completely abolished transcription from this promoter. These data show that the H. volcanii tRNALys promoter is dependent on two separate, and essential, sequence elements. The possible functions of these sequences, in view of the recent descriptions of eucaryal-like transcription factors for Archaea, are discussed. TFIIS (24, 28) and the observation that Pyrococcus woesei TBP binds specifically to TATA-containing DNA fragments (43) suggest that archaea use eucaryal-like DNA-binding proteins in the initiation of transcription. Before the relationship between these two systems can be fully defined, a better understanding of the archaeal transcription process is needed. In this study we used a plasmid-based transcriptional reporter system to examine the transcriptional effects of single point mutations in the box A region of the Haloferax volcanii tRNALys gene. We provide support for the hypothesis that all archaea share a common promoter structure and confirm that the TATA-like sequence is important in determining promoter efficiency in vivo. We also show that sequences other than the box A element can play an essential role in determining transcription efficiency in vivo.

The transcriptional apparatus of archaea (formerly archaebacteria [49]) bears a striking resemblance to the transcriptional systems found in eucarya (eucaryotes) (for a review, see reference 51). Unlike the relatively simple bacterial RNA polymerases, the archaeal and the eucaryal RNA polymerases display complex subunit compositions. Archaeal RNA polymerases typically contain 12 or more subunits, and these proteins share immunological cross-reactivity and sequence similarity with eucaryal RNA polymerase subunits (51). The relatedness of these two systems can also be seen in the structures of their promoter elements. A comparison of sequences upstream of archaeal genes has revealed two conserved sequence elements, box A and box B (17, 42, 48, 51). The box A element is located approximately 25 bp upstream of the transcription start site and has the general consensus 59-T/CTTAT/ AA-39. The core TATA sequence of this element is similar in location and sequence to the TATA promoter element recognized by eucaryal RNA polymerase II (5, 41, 47). The box B element contains the transcription start site and has a weak consensus 59-T/CG/A-39, with transcription beginning with the purine residue (17). Functional studies using in vitro transcription systems from Methanococcus vannielii, Methanococcus thermolithotrophicus, and Methanobacterium thermoautotrophicum (13, 14a, 18, 19, 25a) and Sulfolobus shibatae (17, 21a) have confirmed that the core box A TATA-like sequence is important in determining the efficiency of transcription initiation. The archaeal and eucaryal transcription systems may also share common mechanisms for promoter recognition and transcription initiation. In its most general form, initiation of transcription by eucaryal RNA polymerases is dependent on the interaction of the core RNA polymerase with proteins, transcription factors, and regulatory proteins already bound to the DNA. This is in contrast to bacterial RNA polymerases, which require the association of a sigma factor with the core enzyme before binding to the DNA. The recent descriptions of archaeal genes encoding the eucaryal-like transcription factors TFIID (31, 43) (TATA-binding protein [TBP]), TFIIB (7), and

MATERIALS AND METHODS Culture conditions and materials. H. volcanii WFD11 (6) was grown aerobically at 378C in complex medium (8), and when necessary to ensure maintenance of pWL-based expression plasmids, this medium was supplemented with 20 mM mevinolin (a gift from Merck and Co., Inc.). Escherichia coli DH5a-F9 and JM110 were cultured in Luria broth (LB) medium or LB medium supplemented with 100 mg of ampicillin per ml when cells carried pUC- or pWL-based plasmids. T4 polynucleotide kinase, T4 DNA ligase, and all restriction enzymes were purchased from Bethesda Research Laboratories; Sequenase, X-Gal (5-bromo4-chloro-3-indolyl-b-D-galactopyranoside), and IPTG (isopropyl-b-D-thiogalactopyranoside) were obtained from United States Biochemicals; AmpliTaq DNA polymerase and GeneAmp core reagents were purchased from Perkin Elmer; and Zeta-Probe nylon membranes were obtained from Bio-Rad Laboratories. All oligonucleotides used in this study were synthesized by The Ohio State University Biochemical Instrument Center. Amplification of DNA sequences. DNA sequences were amplified by PCR in a Perkin-Elmer Cetus DNA Thermal Cycler with AmpliTaq DNA polymerase and GeneAmp core reagents. Reactions were carried out in 100-ml volumes containing AmpliTaq DNA polymerase (5 U), 100 ng of template DNA, 0.2 mM primers, 200 mM deoxynucleotide triphosphates, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and 0.001% gelatin. Each reaction mixture was subjected to 30 cycles of amplification (1 cycle: strand separation at 948C, 1 min; template-primer annealing at 508C, 1 min; and DNA strand extension at 728C, 2 min). Generation of point mutations in H. volcanii tRNALys promoter box A element and replacement of purine-rich sequence. We have previously constructed a transcription reporter plasmid that carries the H. volcanii tRNALys promoter and a modified form of the Saccharomyces cerevisiae tRNAPro(UGG) gene as the reporter (37). In this construct the H. volcanii tRNALys promoter-containing fragment has been reduced to a 48-bp region that contains the box A and box B elements and a purine-rich element located 59 of the box A element (Fig. 1A). Transcription from this promoter begins within the expected box B element and

* Corresponding author. Mailing address: Department of Microbiology, The Ohio State University, 484 West 12th Ave., Columbus, OH 43210. Phone: (614) 292-4599. Fax: (614) 292-8120. Electronic mail address: [email protected]. 1844

VOL. 177, 1995

IN VIVO DEFINITION OF ARCHAEAL PROMOTER

1845

FIG. 2. Role of box A proximal purine-rich sequence in transcription from H. volcanii tRNALys promoter. (A) Sequence of wild-type tRNALys promoter and replacement mutant. The purine-rich sequence and its G1C-rich replacement sequence are presented in bold. (B) Northern hybridization patterns for RNAs isolated from cells carrying wild-type or replacement mutation promoter. Hybridization signals corresponding to the reporter tRNAProM RNA and the internal control, tRNALeu RNA, are indicated. The oligonucleotides TRNAPRO and LEU3E were used as the probes. FIG. 1. Construction of H. volcanii tRNALys box A mutants. (A) Schematic representation of the promoter-reporter module. The promoter sequence region encompassing the purine-rich box A, box B, and the transcription start site are presented. The region of mutagenesis, where single point mutations were introduced, is also shown. (B) Representative Northern assay for promoter activity. RNAs from cells carrying the wild-type and the three individual point mutations at position 225 were analyzed. The oligonucleotides TRNAPRO and LEU3E were used as probes, and each sample was tested in duplicate. Hybridization signals corresponding to the reporter, tRNAProM RNA, and an internal standard, tRNALeu, are indicated. The relative expression of each transcript, normalized to that of the wild-type box A (set at 100%), is given.

results in the production of a single stable transcript that can be detected by Northern (RNA) analysis (37, 39). A series of 11 oligonucleotides (termed MUTP) complementary to segments of the shortened tRNALys promoter-containing module were used to introduce point mutations into the box A element of the tRNALys promoter. Each oligonucleotide was degenerate for the 3 non-wild-type bases at a single site, beginning at position 229 and ending at position 219, with respect to the transcription initiation site (Fig. 1A). The sequences 59 to this site on each oligonucleotide were designed to be complementary to the promoter module sequence, up to position 241. Six noncomplementary bases were added to the 59 terminus of the oligonucleotides to generate a HindIII restriction site, and 10 complementary bases were added to the 39 site of the mutation to ensure efficient amplification. For example, the oligonucleotide MUTP-2 (59-CGTGCAAAGCTTGAAAG GAAAGTC[G/T/C]TTTTACCCAC-39) was used to introduce mutations at position 229. The plasmid pUC302, which contains the shortened H. volcanii tRNALys promoter immediately upstream of the yeast reporter gene (39), was used as the template DNA for PCR-based mutagenesis. The DNA between the regions of pUC302 complementary to each MUTP oligonucleotide and the universal pUC/M13 240 primer was amplified by PCR. Each of the amplified DNAs was digested with the restriction enzymes HindIII and EcoRI, and these DNAs were ligated into the equivalent sites of pUC19. The ligation products were then used to transform E. coli DH5a-F9, and the plasmid DNAs isolated from the resulting transformants were screened by DNA sequence analysis for each of the 33 possible point mutations in the tRNALys promoter box A element. Finally, each HindIII-EcoRI expression cassette was subcloned into the corresponding sites of pWL201 (35) to produce a series of H. volcanii-E. coli shuttle expression vectors constituting a complete set of tRNALys promoter box A point mutations between positions 229 and 219 (Fig. 1A). PCR-based mutagenesis was also used to construct a mutant tRNALys promoter in which the purine-rich region, residues 241 to 232 (Fig. 1A and 2A), was replaced with a pyrimidine-rich sequence. The oligonucleotide 5DELPUR (59-CGTGCAAAGCTTCCCTTGCGGCTCATTTTACCCACCGGCAGT-39) and the universal pUC/M13 240 primer were used to amplify sequences of the promoter reporter region of plasmid pUC302T. This plasmid differs from the plasmid pUC302 (38) by having a single point mutation at position 22, changing a cytosine to a thymine. This change had no effect on the transcriptional efficiency or the transcriptional start selection (data not shown). In addition to replacing the purine-rich sequence, amplification with the 5DELPUR oligonu-

cleotide introduced a HindIII site at the promoter end of the DNA fragment. The amplified DNA was digested with HindIII and EcoRI, and this DNA was cloned into the corresponding sites of pUC19 to give the plasmid pUC5DELPUR. The mutation was verified by DNA sequence analysis, and the HindIII-EcoRI fragment of pUC5DELPUR was subcloned into the HindIIIEcoRI sites of the H. volcanii-E. coli shuttle vector pWL201 to give the plasmid pWL5DELPUR. Transformation of H. volcanii. H. volcanii cells used in the transformation were grown to a culture density corresponding to an A550 of 0.5, pelleted by centrifugation (3,500 3 g for 5 min), and gently resuspended in 1/20 volume of spheroplasting solution (35). Transformation was carried out as described previously (35), except that the plasmid DNAs were first introduced into E. coli JM110. Previous results have demonstrated that plasmid DNA with dam-methylated A residues (propagated in E. coli DH5a-F9) transform H. volcanii at only 2% of the efficiency observed by using unmethylated DNA (propagated in E. coli JM110) (21, 36). Therefore, immediately prior to all H. volcanii transformations, plasmid DNA was passed through E. coli JM110. Quantitation of H. volcanii tRNALys promoter strength. The transcriptional strength of each tRNALys promoter was calculated by using Northern analysis to determine in vivo levels of the tRNAProM transcript (37). Cultures of each H. volcanii transformant (100 ml) were grown to a cell density corresponding to an A550 of 1.0. Cells from each culture were harvested by centrifugation and gently resuspended in 4.5 ml of lysis buffer (10 mM Tris-HCl [pH 8.0], 10 mM NaCl, 1 mM trisodium citrate, 1.5% sodium dodecyl sulfate [SDS]) and 126 ml of diethyl pyrocarbonate (DEPC). The resulting lysate was incubated at 378C for 10 min and then transferred to an ice bath. After a 10-min incubation, 2.25 ml of NaCl-saturated water was gently mixed into the lysate, and the incubation was continued for an additional 15 min. The tube contents were then transferred to a 30-ml Corex centrifuge tube, and the SDS-protein-DNA precipitate was pelleted by centrifugation (12,000 3 g for 10 min at 48C). The supernatant, in 6-ml aliquots, was pipetted into 30-ml Corex tubes, and the RNA was precipitated by the addition of 2.5 volumes of ethanol. The RNA was isolated by centrifugation (12,000 3 g for 10 min at 48C), washed once in 70% ethanol, and dissolved in 1.2 ml of DEPC-treated water. For transcript analysis, two 0.4-ml aliquots from each RNA preparation were transferred to separate 1.5-ml microcentrifuge tubes, and the RNA was again precipitated. Finally, the RNA pellets were air dried and dissolved in 60 ml of 13 RNA loading buffer (20 mM Tris-HCl [pH 7.4], 1 mM EDTA [pH 8.0], 8 M urea, 0.05% bromophenol blue, 0.05% xylene cyanol). The RNAs from each duplicate preparation were separated by electrophoresis through a denaturing (8.3 M urea) 6% polyacrylamide gel and transferred to a Zeta-Probe nylon membrane with an electrophoretic blotter (Idea Scientific). RNA was fixed to the membrane by baking under vacuum (808C, 30 min), and each filter was prehybridized at 508C for 30 min in 53 SSC (13 SSC contains 150 mM NaCl plus 15 mM trisodium citrate)–20 mM Na2HPO4 (pH 7.2)–7% SDS– 0.5% nonfat powdered milk–100 mg of denatured salmon sperm DNA per ml. The exon 2-specific oligonucleotide TRNAPRO (59-CGAGCTGGGAATT GAACCCAGG-39) was used to detect levels of tRNAProM transcript. The tRNAProM hybridization signals were standardized by reference to their companion hybridization signals derived from the oligonucleotide LEU3E (59-GGG

1846

PALMER AND DANIELS

GACGAGATTCGAACTCGCGAACCCCTACG-39), which is complementary to the H. volcanii chromosomally encoded tRNALeu RNA sequence. Each oligonucleotide was 59 end labeled with [g-32P]ATP and T4 polynucleotide kinase, and each oligonucleotide (2.0 3 106 cpm) was added to the hybridization solution. Hybridization was allowed to take place at 508C for 24 h. The membranes were then washed three times at room temperature, 10 min for each wash, in 250 ml of 23 SSC–0.5% SDS. Quantitation of hybridizing RNA as counts per minute was determined by using a Betagen Betascope 603 blot analyzer. The values of TRNAPRO-derived signals were then normalized by dividing them by the value of their companion LEU3E-derived signal. The relative transcriptional strength of each promoter was then calculated by expressing this ratio as a percentage of the ratio derived for the wild-type tRNALys promoter, which was arbitrarily set at 100. Each value presented is the result of two independent analyses. The reproducibility of the assay was determined by using the wild-type tRNALys promoter. In 17 independent determinations, the level of tRNAProM hybridization relative to that of the tRNALeu internal control was found to be 33% 6 6% (37). Therefore, only those promoter mutations that gave transcriptional levels 25% above or below the wild-type promoter activity were considered significantly different. It was interesting that the chromosomally encoded tRNALeu RNA had a higher signal level in Northern analysis than the plasmid-encoded tRNAProM RNA (H. volcanii tRNALys promoter), even though the tRNAProM gene was present in multiple copies. Unfortunately, we cannot directly compare expression from these two promoters since we have not characterized the tRNALeu gene, or the turnover of its transcript. Shuttle expression vector copy number determination and transcription start site identification. The copy numbers of four shuttle expression vectors (wild type, 227 adenine, 225 adenine, and 224 cytosine) were determined to test whether the introduction of point mutations into the tRNALys promoter influenced the plasmid copy number. Total DNA was isolated from each of the appropriate H. volcanii transformants (35), and approximately 10 mg of this DNA was digested to completion with EcoRI and MluI in the presence of RNase A (80 mg/ml). After electrophoretic separation through a 1% agarose gel, the DNA was transferred by capillary action to a Zeta-Probe nylon membrane. The membrane was then probed with the 32P-59-end-labeled oligonucleotide 3-hydroxy-3-methylglutaryl-coenzyme A (HMGCoA) (59-GCTGTTGATGACCGAGCAACC-39). The HMGCoA reductase gene, present as a single copy on the plasmid and the chromosome, contains a sequence complementary to this oligonucleotide. Following hybridization for 24 h at 458C, the membrane was washed three times, 1 h each, in 250 ml of 23 SSC–0.5% SDS. The hybridization signals from the chromosomal and plasmid HMGCoA reductase genes were measured in counts per minute with a Betagen Betascope 603 blot analyzer. The plasmid copy number was determined by dividing the plasmid-derived (counts per minute) value by the chromosome-derived (counts per minute) value. Each determination was made in triplicate, and the average value was calculated. The copy numbers for plasmids carrying the wild-type promoter and the 227 adenine, 225 adenine, and 224 cytosine mutations were 16, 12, 17, and 16, respectively. Primer extension analysis was used to map the transcription initiation start sites of representative promoter mutants. By using the 32P-59-end-labeled oligonucleotide TRNAPRO as the primer, the transcription start site was determined for the wild-type promoter and the mutant promoters 227 adenine, 225 adenine, and 222 adenine. In each case, transcripts were initiated, as expected, at the guanine residue of the box B element, 59-CTGC-39 (data not shown).

RESULTS We have previously shown that a 48-bp DNA fragment originating from the 59 region of the H. volcanii tRNALys gene is capable of directing efficient and precise transcription when introduced into these cells on plasmid pWL201 (39). Using the archaeal box A consensus (41, 47, 48, 51) as a guide, combined with the observation that box A elements are generally located approximately 25 bp 59 of the transcription initiation site, we identified the sequence 59-ATTTTACCCAC-39 as a potential box A element within the tRNALys DNA fragment. To evaluate the role of individual nucleotides within this proposed box A element, we constructed single point mutations that changed each nucleotide in this sequence element to each of the other three possible nucleotides (Fig. 1A). Each mutant promoter was then placed upstream of the yeast tRNAProM reporter gene on the H. volcanii-E. coli pWL201 shuttle vector (35) and introduced into H. volcanii. In examining the transcriptional activities of the 33 mutant promoters, we observed mutations that decreased promoter activity and others that increased transcription to levels higher than that of the wild-type promoter. The most pronounced effects on transcription occurred when changes were intro-

J. BACTERIOL. TABLE 1. Relative transcriptional strength of box A mutants Transcription level of a: Position

229 228 227 226 225 224 223 222 221 220 219

A

G

C

T

100 3 205 49 249 100 160 75 76 100 66

27 2 11 4 105 55 90 88 74 9 186

134 60 8 4 16 8 100 100 100 36 100

271 100 100 100 100 28 69 93 90 110 84

a Experimental values are expressed as a percentage of the wild-type level (arbitrarily set at 100%).

duced in the nucleotides at positions 229 to 224, the TATAlike region of the box A element (Fig. 1). The pattern of transcriptional activities of these mutants revealed preferences for specific nucleotides at some positions and an apparent preference for A-T base pairing at other positions. The effects of these mutations are presented in Table 1, and the salient features are summarized below. Replacement of the wild-type adenine at position 229 with thymine led to the highest level of transcription (270%) observed for all of the promoters tested. A modest increase in transcription was observed when a cytosine was present at this position; however, guanine at this position was inhibitory. Transcriptional properties of mutations at position 228 indicated that a pyrimidine is required at this position. Substitution of the wild-type thymine by cytosine at this position led to a small decrease in transcription, whereas the introduction of a purine at this position essentially abolished transcription. There was a marked preference for adenine or thymine at positions 227 and 226; the presence of guanine or cytosine at these positions strongly inhibited transcription. The introduction of an adenine at position 227 is also accompanied by a twofold increase in transcription. A similar stimulation in transcription was observed when the thymine at position 225 was changed to an adenine, again creating a TA dinucleotide. Among the remaining positions examined there was discrimination by the transcription machinery of the nucleotide at position 224, where adenine was the preferred nucleotide, and at position 219, where there was a slight preference for guanine. Changes in the nucleotides at positions 223 to 221 had little effect on transcriptional activity. The high level of transcription associated with the 229 adenine-to-thymine mutation was unexpected and prompted us to examine sequences surrounding the box A element as possible modulators of transcription. We had previously observed that a purine-rich sequence is often found upstream of halobacterial stable RNA genes (9). To determine if these nucleotides could function in transcriptional control, we constructed a mutant promoter module in which the purine-rich element, 59-GAAAGGAAAG-39, was replaced with the G1Crich sequence 59-CCCTTGCGGC-39 (Fig. 2A). Analysis of RNAs isolated from cells carrying the modified upstream region indicated that the replacement of the purine-rich sequence with a G1C-rich sequence abolished transcription from this promoter (Fig. 2B).

VOL. 177, 1995


1847

FIG. 3. Halobacterial promoter sequences. (A) Sequences of halobacterial promoter regions for genes for which the transcription start site has been mapped. Transcription start sites are aligned under the asterisk, and potential box A sequences are underlined. A consensus halobacterial box A element derived from these sequences is shown at the bottom. The following criteria were used to identify possible halobacterial box A elements: the element should be approximately 25 bp 59 of the transcription start site, the element should begin with a pyrimidine, and guanine should be excluded from the four adjacent 39 positions. Also shown are the archaeal consensus box A (17) and the preferred S. shibatae 16S rRNA box A, derived from in vitro transcription studies (41). The effects of single base mutations on the transcription from the H. volcanii tRNALys promoter are also shown. Mutations resulting in greater than a 70% increase or decrease, with respect to the wild-type promoter, are presented above and below the wild-type sequence, respectively. Hk Aa2.2, Haloferax sp.; Hcut, Halobacterium cutirubrum; HGRB, Halobacterium sp. strain GRB; Hhal, Halobacterium halobium; Hmar, Halobacterium marismortui; Hmed, Haloferax mediterranei; Hmor, Halococcus morrhuae; Hvol, Haloferax volcanii.

1848

PALMER AND DANIELS

DISCUSSION We have used an in vivo transcription reporter system to analyze the effects of single point mutations in the box A region of the H. volcanii tRNALys promoter. Within this group of mutant promoters, we observed promoters with unchanged, increased, or decreased transcriptional efficiency compared with the wild-type promoter (Table 1). The transcriptional response was most sensitive to changes in the 229 to 224 region of the box A element; this region corresponds to the TATA-like sequence of the archaeal box A consensus 59-T/ CTTAT/AA-39 (Fig. 3). The general pattern of sensitivity of the tRNALys box A to mutations in this region was strikingly similar to the pattern of sensitivity observed in an in vitro analysis of the S. shibatae 16S rRNA promoter, for which a similar analysis of point mutations in the box A region was performed (17). For the H. volcanii tRNALys and S. shibatae 16S rRNA promoters there is a requirement for a pyrimidine residue at the 59 end of the box A element (position 228 of the tRNALys box A) (Table 1 and Fig. 3). This region is followed by a 3- or 4-nucleotide region where there is a preference for A-T base pairs. Cytosine and guanine at these positions strongly inhibit transcriptional activity (17) (Table 1). Both box A elements also share a preference for an adenine nucleotide in the third and fourth positions 39 of the T/C nucleotide. This leads to the formation of a TA dinucleotide, a characteristic that is common among highly efficient archaeal promoters, and a feature that is in agreement with the general archaeal box A consensus (18, 41) (Fig. 3). The nucleotides located in positions 5 through 7 relative to the T/C nucleotide, are less important in determining transcriptional efficiency. A similar pattern of transcriptional inhibition was observed in in vitro studies with an M. vannielii RNA polymerase transcription system when guanine residues were introduced into the box A region of the M. vannielii tRNAVal promoter (18). An additional study involving the characterization of the H. volcanii HMGCoA reductase gene (27) provides further support for the proposal that TATA-like elements are important for in vivo gene expression in archaea. In this case a guanine-tothymine transversion, converting the box A sequence 59-GTT AGGG-39 to 59-TTTAGGG-39, leads to a dramatic increase in the level of HMGCoA reductase transcription in these cells (27). This is consistent with the results which we have observed for the H. volcanii tRNALys promoter (Table 1) and the in vitro transcription studies (18, 41) in which G-C base pairs in this region lead to lower transcriptional levels. It is not clear why G-C base pairs are excluded from archaeal TATA elements. This effect could reflect a specific recognition of the adenine or thymine nucleotides by the TBP, or it may reflect a general requirement for A-T base pairing in this region, possibly enhancing DNA strand separation. Alternatively, the presence of guanine or cytosine could interfere with the binding interaction of the TBP with the DNA. In Eucarya, it appears that the presence of guanine or cytosine in the TATA region prevents binding of DNA to TBP. Cocrystals of a yeast TBP carboxyl-terminal peptide with a TATA-containing DNA fragment indicated that the presence of guanine in this region would interfere with van der Waals interactions between a hydrophobic surface on the TBP and the minor groove of the TATA element, destabilizing DNA binding (25, 50). Since the amino acid residues involved in the binding of the TATA-element are conserved in P. woesei (43) and Thermococcus celer (31) TBPs, it is likely that these TBPs bind DNA in a manner similar to that of the eucaryal TBPs. The recent characterization of an H. volcanii TBP (38) which also exhibits a high degree of similarity to the eucaryal TBPs sug-

J. BACTERIOL.

gests that the halobacterial TBPs may also share this mechanism for DNA binding. On the basis of the results observed for mutations in the H. volcanii tRNALys box A element, it appears that the TBP has a high degree of tolerance for binding to the TATA element, provided G-C base pairs do not occur within this 4-bp region. A relaxed recognition pattern is consistent with the observation that a TATA element of five consecutive thymine residues functions as an effective promoter (Table 1). This general preference for A-T base pairing in the box A region is also predicted for the consensus for halobacterial box A elements (Fig. 3). Following the argument that the archaeal transcriptional machinery is eucaryal-like, the archaeal TBP should recruit other protein factors to establish an initiation complex, and binding of TBP would be only one component in determining the frequency of transcription initiation. In the case of the H. volcanii tRNALys promoter, we observed that replacement of the purine-rich sequence, immediately 59 of the box A element, with a G1C-rich sequence completely abolished transcription from this promoter. At present we cannot distinguish whether the purine-rich sequence is essential for expression from this promoter or whether the replacement sequence is inhibitory. However, the fact that sequences upstream of halobacterial box A elements are not highly conserved (Fig. 3) argues that the replacement sequence is not inhibitory; it is more likely that the purine-rich sequence acts to enhance expression from the tRNALys promoter. The enhancement of expression could arise from alterations in the local DNA structure, increasing the efficiency of binding of the transcriptional machinery to the box A element or hastening open complex formation. Alternatively, these sequences could aid in the interaction of factors that are specific for the expression of tRNA genes in this organism. Both possibilities are consistent with our current understanding of the transcriptional apparatus of archaea. Further studies will be needed to determine the precise role of these sequences in transcription activation. ACKNOWLEDGMENTS This work was supported by a grant from the Department of Energy, DE-FG02-91ER20041. C.J.D. is an associate of the Canadian Institute for Advanced Research. REFERENCES 1. Arndt, E., W. Kro ¨mer, and T. Hatakeyama. 1990. Organization and nucleotide sequence of a gene cluster coding for eight ribosomal proteins in the archaebacterium Halobacterium marismortui. J. Biol. Chem. 265:3034–3039. 2. Benachenhou, N., and G. Baldacci. 1991. The gene for a halophilic glutamate dehydrogenase: sequence, transcription analysis and phylogenetic implications. Mol. Gen. Genet. 230:345–352. 3. Betlach, M., J. Friedman, H. W. Boyer, and F. Pfeifer. 1984. Characterization of a halobacterial gene affecting bacterio-opsin gene expression. Nucleic Acids Res. 12:7949–7959. 4. Blanck, A., and D. Oesterhelt. 1987. The halo-opsin gene. II. Sequence, primary structure of halorhodopsin and comparison with bacteriorhodopsin. EMBO J. 6:265–273. 5. Bucher, P., and E. N. Trifonov. 1986. Compilation and analysis of eukaryotic POL II promoter sequences. Nucleic Acids Res. 14:10009–10026. 6. Charlebois, R. L., W. L. Lam, S. W. Cline, and W. F. Doolittle. 1987. Characterization of pHV2 from Halobacterium volcanii and its use in demonstrating transformation of an archaebacterium. Proc. Natl. Acad. Sci. USA 84:8530–8534. 7. Creti, R., P. Londei, and P. Cammarano. 1993. Complete nucleotide sequence of an archaeal (Pyrococcus woesei) gene encoding a homolog of eukaryotic transcription factor IIB (TFIIB). Nucleic Acids Res. 21:2942. 8. Daniels, C. J., S. E. Douglas, A. H. Z. McKee, and W. F. Doolittle. 1986. Archaebacterial tRNA genes: structure and intron processing, p. 349–355. In L. Leive, P. F. Bonventre, J. A. Morello, S. D. Silver, and H. C. Wu (ed.), Microbiology—1986. American Society for Microbiology, Washington, D.C. 9. Daniels, C. J., R. Gupta, and W. F. Doolittle. 1985. Transcription and excision of a large intron in the tRNATrp gene of an archaebacterium, Halobac-

VOL. 177, 1995 terium volcanii. J. Biol. Chem. 260:3132–3134. 10. DasSarma, S., U. L. RajBhandary, and H. G. Khorana. 1984. Bacterio-opsin mRNA in wild-type and bacterio-opsin-deficient Halobacterium halobium strains. Proc. Natl. Acad. Sci. USA 81:125–129. 11. Dennis, P. P. 1985. Multiple promoters for the transcription of the ribosomal RNA gene cluster in Halobacterium cutirubrum. J. Mol. Biol. 186:457–461. 12. Englert, C., M. Horne, and F. Pfeifer. 1990. Expression of the major gas vesicle protein gene in the halophilic archaebacterium Haloferax mediterranei is modulated by salt. Mol. Gen. Genet. 222:225–232. 13. Frey, G., M. Thomm, B. Bru ¨digam, H. P. Gohl, and W. Hausner. 1990. An archaebacterial cell-free transcription system: the expression of tRNA genes from Methanococcus vannielii is mediated by a transcription factor. Nucleic Acids Res. 18:1361–1367. 14. Gerl, L., and M. Sumper. 1988. Halobacterial flagellins are encoded by a multigene family: characterization of five flagellin genes. J. Biol. Chem. 263:13246–13251. 14a.Gohl, H. P., W. Hausner, and M. Thomm. 1992. Cell-free transcription of the nifH1 gene of Methanococcus thermolithotrophicus indicates that promoters of archaeal nif genes share basic features with the methanogen consensus promoter. Mol. Gen. Genet. 231:286–295. 15. Gropp, F., B. Grampp, P. Stolt, P. Palm, and W. Zillig. 1992. The immunityconferring plasmid pfHL from the Halobacterium salinarium phage fH: nucleotide sequence and transcription. Virology 190:45–54. 16. Gropp, F., P. Palm, and W. Zillig. 1989. Expression and regulation of Halobacterium halobium phage fH genes. Can. J. Microbiol. 35:182–188. 17. Hain, J., W. D. Reiter, U. Hu ¨depohl, and W. Zillig. 1992. Elements of an archaeal promoter defined by mutational analysis. Nucleic Acids Res. 20: 5423–5428. 18. Hausner, W., G. Frey, and M. Thomm. 1991. Control regions of an archaeal gene: a TATA box and an initiator element promote cell-free transcription of the tRNA(Val) gene of Methanococcus vannielii. J. Mol. Biol. 222:495– 508. 19. Hausner, W., and M. Thomm. 1993. Purification and characterization of a general transcription factor, aTFB, from the archaeon Methanococcus thermolithotrophicus. J. Biol. Chem. 268:24047–24052. 20. Holmes, M. L., and M. L. Dyall-Smith. 1991. Mutations in DNA gyrase result in novobiocin resistance in halophilic archaebacteria. J. Bacteriol. 173:642–648. 21. Holmes, M. L., S. D. Nuttall, and S. M. Dyall. 1991. Construction and use of halobacterial shuttle vectors and further studies on Haloferax DNA gyrase. J. Bacteriol. 173:3807–3813 21a.Hu ¨depohl, U., W. D. Reiter, and W. Zillig. 1990. In vitro transcription of two rRNA genes of the archaebacterium Sulfolobus sp. B12 indicates a factor requirement for specific initiation. Proc. Natl. Acad. Sci. USA 87:5851–5855. 22. Jones, J. G., N. R. Hackett, J. T. Halladay, D. J. Scothorn, C. F. Yang, W. L. Ng, and S. DasSarma. 1989. Analysis of insertion mutants reveals two new genes in the pNRC100 gas vesicle gene cluster of Halobacterium halobium. Nucleic Acids Res. 17:7785–7793. 23. Joshi, P., and P. P. Dennis. 1993. Characterization of paralogous and orthologous members of the superoxide dismutase gene family from genera of the halophilic archaebacteria. J. Bacteriol. 175:1561–1571. 24. Kaine, B. P., I. J. Mehr, and C. R. Woese. 1994. The sequence, and its evolutionary implications, of a Thermococcus celer protein associated with transcription. Proc. Natl. Acad. Sci. USA 91:3854–386. 25. Kim, J. L., D. B. Nikolov, and S. K. Burley. 1994. Co-crystal structure of TBP recognizing the minor groove of a TATA element. Nature (London) 365: 520–527. 25a.Knaub, S., and A. Klein. 1990. Specific transcription of cloned Methanobacterium thermautotrophicum transcription units by homologous RNA polymerase in vitro. Nucleic Acids Res. 18:1441–1446. 26. Kro ¨mer, W. J., and E. Arndt. 1991. Halobacterial S9 operon: three ribosomal protein genes are cotranscribed with genes encoding a tRNA(Leu), the enolase, and a putative membrane protein in the archaebacterium Haloarcula (Halobacterium) marismortui. J. Biol. Chem. 266:24573–24579. 27. Lam, W. L., and W. F. Doolittle. 1992. Mevinolin-resistant mutations identify a promoter and the gene for a eukaryote-like 3-hydroxy-3-methylglutarylcoenzyme A reductase in the archaebacterium Haloferax volcanii. J. Biol. Chem. 267:5829–5834. 28. Langer, D., and W. Zillig. 1993. Putative tfIIs gene of Sulfolobus acidocaldarius encoding an archaeal transcription elongation factor is situated directly downstream of the gene for a small subunit of DNA-dependent RNA


1849

polymerase. Nucleic Acids Res. 21:2251. 29. Lechner, J., and M. Sumper. 1987. The primary structure of a procaryotic glycoprotein: cloning and sequencing of the cell surface glycoprotein gene of halobacteria. J. Biol. Chem. 262:9724–9729. 30. Leffers, H., F. Gropp, F. Lottspeich, W. Zillig, and R. A. Garrett. 1989. Sequence, organization, transcription and evolution of RNA polymerase subunit genes from the archaebacterial extreme halophiles Halobacterium halobium and Halococcus morrhuae. J. Mol. Biol. 206:1–17. 31. Marsh, T. L., C. I. Reich, R. B. Whitelock, and G. J. Olsen. 1994. Transcription factor IID in the Archaea: sequences in the Thermococcus celer genome would encode a product closely related to the TATA-binding protein of eukaryotes. Proc. Natl. Acad. Sci. USA 91:4180–4184. 32. May, B. P., and P. P. Dennis. 1990. Unusual evolution of a superoxide dismutase-like gene from the extremely halophilic archaebacterium Halobacterium cutirubrum. J. Bacteriol. 172:3725–3729. 33. Mevarech, M., T. S. Hirsch, S. Goldman, E. Yakobson, H. Eisenberg, and P. P. Dennis. 1989. Isolation and characterization of the rRNA gene clusters of Halobacterium marismortui. J. Bacteriol. 171:3479–3485. 34. Mojica, F. J., G. Juez, and V. F. Rodriguez. 1993. Transcription at different salinities of Haloferax mediterranei sequences adjacent to partially modified PstI sites. Mol. Microbiol. 9:613–621. 35. Nieuwlandt, D. T., and C. J. Daniels. 1990. An expression vector for the archaebacterium Haloferax volcanii. J. Bacteriol. 172:7104–7110. 36. Nieuwlandt, D. T., E. S. Haas, and C. J. Daniels. 1991. The RNA component of RNase P from the archaebacterium Haloferax volcanii. J. Biol. Chem. 266:5689–5695. 37. Palmer, J. R., and C. J. Daniels. 1994. A transcriptional reporter for in vivo promoter analysis in the archaeon Haloferax volcanii. Appl. Environ. Microbiol. 60:3867–3869. 38. Palmer, J. R., and C. J. Daniels. Unpublished data. 39. Palmer, J. R., D. T. Nieuwlandt, and C. J. Daniels. 1994. Expression of a yeast intron-containing tRNA in the archaeon Haloferax volcanii. J. Bacteriol. 176:3820–3823. 40. Pfeifer, F., J. Griffig, and D. Oesterhelt. 1993. The fdx gene encoding the [2Fe--2S] ferredoxin of Halobacterium salinarium (H. halobium). Mol. Gen. Genet. 239:66–71. 41. Reiter, W. D., U. Hu ¨depohl, and W. Zillig. 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: 9509–9513. 42. Reiter, W. D., P. Palm, and W. Zillig. 1988. Analysis of transcription in the archaebacterium Sulfolobus indicates that archaebacterial promoters are homologous to eukaryotic pol II promoters. Nucleic Acids Res. 16:1–19. 43. Rowlands, T., P. Baumann, and S. P. Jackson. 1994. The TATA-binding protein: a general transcription factor in eukaryotes and archaebacteria. Science 264:1326–1329. 44. Salin, M. L., M. V. Duke, D. P. Ma, and J. A. Boyle. 1991. Halobacterium halobium Mn-SOD gene: archaebacterial and eubacterial features. Free Radical Res. Commun. 12-13:443–449. 45. Scholzen, T., and E. Arndt. 1992. The alpha-operon equivalent genome region in the extreme halophilic archaebacterium Haloarcula (Halobacterium) marismortui. J. Biol. Chem. 267:12123–12130. 46. Shimmin, L. C., and P. P. Dennis. 1989. Characterization of the L11, L1, L10 and L12 equivalent ribosomal protein gene cluster of the halophilic archaebacterium Halobacterium cutirubrum. EMBO J. 8:1225–1235. 47. Thomm, M., and W. Hausner. 1992. Genes for stable RNAs and their expression in Archaea, p. 36–53. In Madeleine Sebald (ed.), Genetics and molecular biology of anaerobic bacteria. Springer-Verlag, New York. 48. Thomm, M., and G. Wich. 1988. An archaebacterial promoter element for stable RNA genes with homology to the TATA box of higher eukaryotes. Nucleic Acids Res. 16:151–163. 49. Woese, C. R., O. Kandler, and M. L. Wheelis. 1990. Towards a natural system of organisms: proposal of the domains Archaea, Bacteria, and Eucarya. Proc. Natl. Acad. Sci. USA 87:4576–4579. 50. Youngchang, K., J. H. Geiger, S. Hahn, and P. B. Sigler. 1994. Crystal structure of a yeast TBP/TATA-box complex. Nature (London) 365:512–520. 51. Zillig, W., P. Palm, H.-P. Klenk, D. Langer, U. Hu ¨depohl, J. Hain, M. Landendorfer, and I. Holz. 1993. Transcription in the Archaea, p. 367–391. In M. Kates, D. Kushner, and A. Matheson (ed.), The Biochemistry of archaea (archaeabacteria), vol. 26. Elsevier, New York.