Thymine at −5 Is Crucial for cpc Promoter Activity of Synechocystis sp. Strain PCC 6714

Abstract

The levels of transcripts of the cpc operon were highly reduced in a PD-1 mutant of cyanobacterium Synechocystis sp. strain PCC 6714. This was due to a substitution of C for T that occurred at 5 bp upstream of the transcription initiation site of the cpc operon. Any substitution for T at the −5 position drastically reduced both in vivo and in vitro promoter activity in cyanobacterium Synechococcus sp. strain PCC 7942 but not the in vivo activity in Escherichia coli. This suggests that the requirement of −5T appears to be specific for a cyanobacterial RNA polymerase-promoter combination.

Cyanobacteria are photosynthetic prokaryotes that are considered to be the ancestors of chloroplasts. Cyanobacteria have been investigated intensively, as a model of oxygen-evolving photosynthetic organisms, in various aspects such as photosynthesis and gene expression.

We have found that a single base substitution has occurred in the cpc operon in a phycocyanin-deficient mutant of the cyanobacterium Synechocystis sp. strain PCC 6714, PD-1 (13). The cpc operon comprises cpcBAC1C2D, which encodes phycocyanin and linker polypeptides located in peripheral rods in principal cyanobacterial light-harvesting antennae, i.e., phycobilisomes. In this study, we found that the substitution site that drastically reduces the transcription level is the −5 position, and we demonstrated that T at the −5 position is crucial for the promoter activities determined with in vivo and in vitro systems of Synechococcus sp. strain PCC 7942.

Organisms and growth conditions.

The wild type and PD-1 mutant of Synechocystis sp. strain PCC 6714 were grown as described previously (13, 14) for use for RNA extraction. The wild type and transformants of Synechococcus sp. strain PCC 7942 were grown at 30°C in BG11 medium (17) with aeration with ordinary air under continuous illumination at 30 microeinsteins m⁻² s⁻¹. Escherichia coli JM109, as the host for plasmid propagation, was grown in Luria-Bertani medium at 37°C.

Determination of the transcription initiation site of the cpc operon.

We previously reported that substitution of C for T at 259 bp upstream of the cpcB initiation codon of Synechocystis sp. strain PCC 6714 decreased the levels of transcripts drastically (13). To determine whether or not the site of the substitution is upstream of the transcription initiation site, transcripts of the cpc operon from Synechocystis sp. strain PCC 6714, the wild type and the PD-1 mutant, were analyzed by means of primer extension (Fig. 1A). Total RNA (10 to 20 μg) of the strains, which was prepared by the hot-phenol method described previously (13), was annealed with about 2 pmol of an end-labeled oligonucleotide, PE1 (5′-ATGGCTGCTCTCCATAAAAC-3′) (18) and then extended with a Moloney murine leukemia virus reverse transcriptase, ReverTra Ace (Toyobo, Osaka, Japan), for 30 min at 50°C. The reaction was stopped with formamide loading buffer. The products were electrophoresed on a 6% polyacrylamide gel along with a sequencing ladder. The 5′ end of the cpc mRNA was found to be located 254 bp upstream of the cpcB initiation codon in both the wild type and PD-1. Figure 1B shows that a substitution occurred at 5 bp upstream of the transcription initiation site (−5). The transcription initiation sites of the cpc operons in seven species of cyanobacteria have been determined so far. The −5 position of the promoter in Synechocystis sp. strain PCC 6714 is located in the nonconserved region between the −10 element and +1. However, T at 4 bp downstream of the 3′ end of a putative −10 element, GTATAA, seems to be conserved in unicellular cyanobacteria except for Synechocystis sp. strain PCC 9413.

FIG. 1.

The promoter region of the cpc operon. (A) Primer extension analysis of cpc operon transcripts. Lanes 1 and 2, reactions with mRNA isolated from Synechocystis sp. strain PCC 6714, the wild type and PD-1 mutant, respectively. The position of the 5′ end of the transcripts, indicated by a gray box, was determined by comparing the migration of a DNA sequence ladder extended with the same primer as that used for primer extension, i.e., PE1. (B) Sequence alignment of the promoter region of the cpc operon. The sequences of the cpc operon promoter regions of Synechocystis sp. strain PCC 6714 (this work), Synechocystis sp. strain PCC 6701 (1), Synechocystis sp. strain PCC 9413 (15), Synechococcus sp. strain PCC 7942 (10), Synechococcus sp. strain PCC 7002 (6), Anabaena sp. strain PCC 7120 (3), and Calothrix sp. strain PCC 7601 (4) were aligned with CLUSTAL W (23). The transcriptional start sites (+1) are indicated by boxes. Perfectly conserved nucleotides are indicated by shading. The substitution site of the PD-1 mutant at −5 is indicated by an arrow.

In vivo promoter activity in Synechococcus sp. strain PCC 7942.

To examine the effect of a nucleotide substitution at −5 on the transcription level, cpc promoter regions with A and G at −5, respectively, were prepared. In addition, promoter regions with T (wild type) and C (PD-1 mutant), which were constructed previously (13), and the promoter-luxAB fusions, were used for reporter assaying in a cyanobacterium. It has been shown that the mutational effect of the cpc promoter region of Synechocystis sp. strain PCC 6714 is also reflected in Synechococcus sp. strain PCC 7942, regardless of their somewhat different genetic backgrounds (13). Therefore, we performed reporter assays on Synechococcus sp. strain PCC 7942 by using an available plasmid, pAM1414 (2). The promoter regions (763 bp, −511 to +252 [when the transcription initiation site is +1]) of the wild type and PD-1 were inserted in a region upstream of promoterless luxAB genes in pAM1414 (2), and then −5T of the wild-type promoter was changed to A and G, respectively, by site-directed mutagenesis. The primers used were MF1 (5′-ACTAAGCTGATCCGGTGGAT-3′), MR1 (5′-GTGGCTGATAAGTGAGAAGG-3′), MUT (5′-CGAGTGATCCATTAATCTCC-3′), R1A (5′-GTAAACTGTGGGATTGCAAA-3′), and R1G (5′-GTGAACTGTGGGATTGCAAA-3′), according to the method of Ito et al. (8). Synechococcus sp. strain PCC 7942 was transformed with the four kinds of plasmids carrying the cpc promoters of Synechocystis sp. strain PCC 6714, with T, C, A, and G at −5, respectively. The promoter activities of the transformants were determined from the in vivo bioluminescence of luxAB gene products according to the method of Maeda et al. (11). One hundred microliters of a cell culture at the exponential-growth phase, which was diluted to about 1 μg of chlorophyll/ml, was transferred to a test tube and then mixed with 2 μl of a 0.1% n-decanal emulsion. The bioluminescence of the cell suspension was determined as described previously (13). As shown in Fig. 2, any substitution for T at the −5 position reduced the intrinsic promoter activity.

FIG. 2.

In vivo promoter activity in Synechococcus sp. strain PCC 7942. The activity was measured as the expression of luxAB reporter genes. The nucleotide indicated at the bottom of each column refers to the nucleotide located at −5 in each promoter. The data are expressed as means plus the standard deviations of the values for three independent experiments.

In vitro promoter activity with Synechococcus sp. strain PCC 7942 RNAP.

To determine whether or not the effect of the nucleotide at −5 on the transcription level is mediated by a transcription factor, we performed in vitro transcription with purified Synechococcus sp. strain PCC 7942 RNA polymerase (RNAP) (Fig. 3). The holoenzyme was reconstituted by mixing the core RNAP with recombinant sigma factor RpoD1. The RNAP core enzyme and principal sigma factor (RpoD1) were purified according to the previously described methods of Goto-Seki et al. (7). The core enzyme was mixed with a threefold molar excess of purified sigma protein (RpoD1), which was followed by incubation for 30 min at 30°C to allow formation of the holoenzyme. Single-round transcription reactions were performed under standard conditions for the E. coli RNAP with modifications (16). A transcription reaction mixture (35 μl) comprising 0.1 pmol of template DNA and 3 pmol of RNAP in T buffer (22) was incubated for 20 min at 30°C, after which RNA synthesis was initiated by the addition of 15 μl of a prewarmed substrate mixture containing 160 μM concentrations each of ATP, GTP, and CTP, as well as 50 μM UTP and 2 μCi of [α-³²P]UTP (Amersham Biosciences Co., Uppsala, Sweden) in T buffer. After incubation for 5 min at 30°C, the reaction was terminated by the addition of 50 μl of an ice-cold stop solution containing 40 mM EDTA and E. coli tRNA (300 μg/ml), and then the nucleic acids were precipitated with ethanol. The transcripts were electrophoresed through a 5% polyacrylamide gel containing 8 M urea and then examined with a BAS1000 image analyzer (Fuji Photo Film Co., Ltd., Tokyo, Japan). The lengths of transcripts were estimated with reference to the lengths of known transcripts of E. coli RNAP. The experiments were performed at least twice to confirm the reproducibility. The template for each reaction was prepared as follows: a 296-bp fragment comprising −205 to +91 of the cpc promoter with T, C, A, or G was PCR amplified with primers, i.e., ITB (5′-GTTCCCATTGAACATCAAGG-3′) and ITC (5′-CAACCCAAGGGAAAGTTACA-3′). A 326-bp fragment comprising −205 to +121 of the cpc promoter of pANY1 (13), which was used as an internal control for each reaction, was PCR amplified with primers, i.e., ITB and ITD (5′-AAGGGAATTTATGAGAGGCG-3′). As a result, 91- and 121-nt transcripts were produced through the respective transcription reactions, and so the expected promoter recognition could be detected. The in vitro promoter activity with the four kinds of promoters showed a similar tendency to the in vivo promoter activity in Synechococcus sp. strain PCC 7942 (Fig. 2 and 3).

FIG. 3.

In vitro transcription experiments on reconstituted RNAP of Synechococcus sp. strain PCC 7942. (A) DNA templates carrying the cpc promoters with different nucleotides at −5 were examined by single-round transcription assaying in vitro. The nucleotide indicated at the bottom of each lane refers to the nucleotide located at −5 in each template. The target 91-nt and control 121-nt transcripts are indicated by arrows. (B) The relative band intensity for each promoter was quantified. The quantification was performed by dividing the band intensity of each 91-nt transcript with that of each 121-nt control transcript given in panel A. Each quantity represents the average of two independent experiments.

In vivo promoter activity in E. coli.

Cyanobacterial RNAPs are known to have characteristic subunit compositions, which are not found for other eubacterial RNAPs (19, 20). To determine whether the effect of −5T in the cpc promoter of Synechocystis sp. strain PCC 6714 is cyanobacterial RNAP specific or not, we used E. coli as a host for the in vivo promoter assay. The promoters (458 bp, −206 to +252) with T, C, A, and G at −5, respectively, were inserted into the BamHI site upstream of the lacZ gene in the promoterless lac operon fusion vector pRS415 (21). The E. coli JM109 transformants with the plasmids were grown at 37°C in M9 medium (12) containing 0.5% Casamino Acids and thiamine (100 μg/ml) until an optical density at 600 nm of approximately 0.8 was achieved. β-galactosidase activity was measured as described previously (12). We confirmed that transcription from the plasmid backbone accounted for less than 3% of all promoter activity. As shown in Fig. 4, the tendency of the four types of promoter activity in E. coli was different from that in Synechococcus sp. strain PCC 7942 (Fig. 2), i.e., the promoter activities in E. coli exhibited no −5 nucleotide specificity.

FIG. 4.

In vivo promoter activity of E. coli. The activity was measured as the expression of the lacZ reporter gene. The nucleotide indicated at the bottom of each column refers to the nucleotide located at −5 in each promoter. The rightmost column, designated as −P, shows the activity with a promoterless vector used as a control. The data are expressed as means plus the standard deviations of the values for three independent experiments.

Comparison of the RNAP structure between cyanobacteria and E. coli.

On the assumption that a potential factor is not copurified with RNAP, the results of in vitro transcription in Synechococcus sp. strain PCC 7942 indicate that the −5T recognition is not mediated by a transcription factor but may be directly performed by RNAP. In E. coli, however, an effect of substitution of C for T at −5 on the promoter activity could not be observed in vivo. This suggests that the requirement for T at the −5 position appears to be specific for a cyanobacterial RNAP-promoter combination.

Cyanobacterial RNAP shows two remarkable structural differences in the β′ subunit from that of E. coli (Fig. 5): One is a split separating β′ from γ (19, 20), and the other is a large insertion, which possibly interacts with DNA, as in the jaw module of the Rpb1 subunit of yeast RNAP II (5). It is conceivable that the effect of T at position −5 is somehow related to the presence of the split separating β′ from γ and/or the large insertion in the C-terminal region of β′ in cyanobacterial RNAP.

FIG. 5.

Schematic sequence comparison of the RNAP β′ subunits of E. coli (top) and cyanobacterium Synechocystis sp. strain PCC 6803 (bottom). The boxes designated A through H indicate evolutionally conserved regions (9). Striped boxes indicate the hypervariable regions (24); in the case of cyanobacteria, this region represents the insertion domain. The split site of the cyanobacterial subunit is indicated by an arrow. The N-terminal portion comprising regions A through D is the γ subunit, and the C-terminal portion comprising regions E through H is the β′ subunit. The gradations on the scale bar at the bottom each indicate 100 amino acids. The primary sequences of β′γ cyanobacterial subunits, which were available in genomic databases, were very similar to each other (data not shown). Therefore, the sequences of Synechocystis sp. strain PCC 6803, the species most related to both Synechocystis sp. strain PCC 6714 and Synechococcus sp. strain PCC 7942, were utilized as cyanobacterial representatives. The amino acid sequences of E. coli and Synechocystis sp. strain PCC 6803 were obtained through GenBank (http://www.ncbi.nlm.nih.gov/Entrez/) and CyanoBase (http://www.kazusa.or.jp/cyano/), respectively.