Isolation and expression of the catA gene encoding the major vegetative catalase in Streptomyces coelicolor Müller

We isolated the catA gene for the major vegetative catalase from Streptomyces coelicolor Müller. It encodes a polypeptide of 488 residues (55,440 Da) that is highly homologous to typical monofunctional catalases. We investigated catA expression by analyzing both catA mRNA and catalase activity. catA expression was increased by H2O2 treatment but did not increase during stationary phase. A putative catalase (CatB) cross-reactive with anti-CatA antibody appeared during stationary phase and in the aerial mycelium.

Catalase is an oxidoreductase which transfers two electrons to H 2 O 2 , decomposing it into O 2 and H 2 O. Most of the catalases characterized so far can be classified into one of two types based on their enzymological properties (reference 15 and references therein): monofunctional (typical) catalases and bifunctional catalase-peroxidases. Multiple catalases have been found in organisms such as Escherichia coli (5,6), Bacillus subtilis (16), Streptomyces coelicolor (13,14), Saccharomyces cerevisiae (19,20), and Aspergillus nidulans (17). The role of each enzyme in different stages of the growth of these organisms is not well understood.
In S. coelicolor Müller, multiple catalases exist in late phases of growth in liquid culture or after the onset of differentiation on surface culture (4,14). The major catalase (Cat4) is a typical catalase which consists of four identical subunits of 56 kDa and is expressed at a rather constant level throughout the different growth phases (14). A similar catalase has been reported in S. coelicolor A3(2) by Walker et al. (23). The involvement of H 2 O 2 or reactive oxygen species in signaling for cell development has been suggested from studies in many systems (3,10,18). In this respect, the role of each catalase in growth and differentiation deserves systematic investigation. As an initiating effort toward this goal, this paper describes the isolation and characterization of the catA gene for the major vegetative catalase Cat4, whose expression is induced by H 2 O 2 but not during the stationary phase of growth, unlike other major catalases in E. coli and B. subtilis (2,12).
Cloning and sequence analysis of the catA gene. The compilation and alignment of amino acid sequences from typical monofunctional catalases enabled us to design PCR primer pairs with which to isolate a catalase gene from S. coelicolor Müller (ATCC 10147) (see Fig. 1). A single PCR product of 276 bp was produced whose deduced amino acid sequence closely matched the conserved amino acid residues of typical catalases. Genomic Southern hybridization with this PCR fragment revealed a single prominent band from each restriction digest tested. There was also a minor hybridizing band which might bear a related gene. We screened a phage library of S. coelicolor Müller DNA to isolate the clone carrying the template gene (catA) for the PCR product. A 4.5-kb insert from one phage clone containing the catA gene was further charac-terized, and the nucleotide sequence of the internal 2.0-kb SalI-AvaI fragment was determined (Fig. 1). The sequence reveals a single open reading frame (ORF) encoding a polypeptide of 488 residues with a deduced molecular mass of 55,440 Da. A putative ribosome binding site ((GAGGAG) was located 6 nucleotides upstream from a putative initiation codon, UUG. The deduced gene product is almost the same size as the subunit of the major catalase Cat4 purified from S. coelicolor Müller, which is about 56 kDa as judged by its electrophoretic mobility (14). The catA ORF contained all the partial peptide sequences from the Cat4-like catalase purified from S. coelicolor A3(2) by Walker et al. (23) with a few differences. The comparison of amino acid sequences revealed significant homology between the deduced catA gene product and large regions of numerous typical catalases. Nearly all the residues important for catalytic activity and heme binding are conserved (8).
Expression of the catA gene in E. coli and Streptomyces lividans TK24. In order to verify the catA gene product, we expressed this gene on multicopy plasmids in both E. coli and S. lividans. E. coli cells containing the catA gene on pUC18 exhibited much higher levels of catalase activity (20-fold) than cells containing the parental vector ( Fig. 2A, lanes 1 and 2). The overproduced CatA catalase of E. coli comigrated with the S. coelicolor Cat4 enzyme on native gels (data not shown). When the catA gene was introduced on pIJ702, S. lividans TK24 cells produced two times more catalase activity, which again comigrated with Cat4 (data not shown). The identity of CatA with the major catalase Cat4 was confirmed by Western blot analysis with anti-Cat4 antibody ( Fig. 2A, lanes 3 and 4). A single polypeptide of about 56 kDa was detected in extracts of E. coli cells harboring the catA gene.
Determination of the transcription start site of the catA gene. We determined the transcription start site for catA by S1 nuclease mapping of RNAs from cells grown to midexponential phase (Fig. 2B). The probe used generated a single species of protected band and assigned the 5Ј end of the catA mRNA to the G residue 27 nucleotides upstream from the initiation codon. RNAs isolated from E. coli cells containing the catA gene also generated the same protected band, implying that the same promoter was utilized in both E. coli and S. coelicolor. Putative promoter elements were identified which closely match the consensus S. coelicolor E hrdB promoters (TTGCCT-N 16 -TAGGGT versus TTGACN-N 16-18-TAGAPuT) (Fig. 1) (22). Since this promoter resembles E. coli E rpoD promoters, it is most likely that the E rpoD holoenzyme recognized this promoter in E. coli.
Regulation of catA gene expression by H 2 O 2 and growth phase. The total catalase activity in S. coelicolor in the presence or absence of H 2 O 2 was measured at different growth phases. As shown in Fig. 3A, compared with the early exponential phase culture, the total catalase activity increased about sevenfold when cells were cultured until stationary phase in YEME liquid medium (11). In rapidly growing cells, 100 M H 2 O 2 induced catalase activity twofold. The inducibility by H 2 O 2 decreased as cells entered stationary phase. We resolved different catalases in cell extracts by native polyacrylamide gel electrophoresis and examined the expression of each catalase under the same conditions (Fig. 3B). We observed that Cat4 (CatA) is the major catalase in rapidly growing cells, is induced by H 2 O 2 , and increases during exponential culture for up to 20 h. At late exponential phase (40-h culture), Cat4 decreased slightly but a new activity (Cat5) appeared, contributing to the increase in total catalase activity (Fig. 3B, lane 5). During stationary phase, we observed an emergence of high-molecular-weight catalase activity (Cat1) as well as an increase in the levels of activities comigrating with Cat4 and Cat5 (Fig. 3B,  lane 7). The emergence of at least two catalases other than CatA contributed to the increase in total catalase activity at later phases of growth. However, the quantitative assessment of each individual catalase cannot be accurately achieved from this type of activity staining analysis.
We next monitored the changes in catA expression by measuring both its RNA and its protein product (Fig. 4). Northern  (Fig. 2B) is indicated with a bent-end arrow, and the putative Shine-Dalgarno sequence is presented in boldface. Potential promoter sequences for E hrdB recognition are in boldface and underlined. The coding region starts at the TTG codon at nucleotide 392 and ends at the TGA codon at nucleotide 1855, thus encoding a polypeptide of 488 amino acids. The amino acid residues in boldfaced italics are those that are identical to previously reported peptide sequences (23). The two arrows indicate the regions that bind PCR primers (5Ј-TTCGACCGSGAGCGSATCCCSGAC-3Ј and 5Ј-GATGAAGAA-GAYSGGSGTGTTGTT-3Ј, where S is G or C and Y is C or T).  (11). S1 nuclease mapping analysis was done with a 629-nucleotide SalI-BglII fragment as a probe as described by Smith and Chater (21). The initiating G nucleotide is indicated with an arrow. TϩC, Maxam-Gilbert sequencing ladder.
analysis revealed a single mRNA species of 1.7 kb (Fig. 4A). This corresponds to the size of the catA ORF and implies that the message is monocistronic. The level of catA mRNA increased up to fourfold following H 2 O 2 treatment during exponential phase (Fig. 4A). No full-length catA mRNAs were detected at stationary phase. However, it was extremely difficult to obtain intact RNAs, from cells at this growth phase, probably because of increased level of RNases. We therefore tried to detect a portion of catA mRNA near the 5Ј end by S1 nuclease protection with the same probe used for the experiment shown in Fig. 2B (Fig. 4B). The results confirm the observation made after by Northern analysis that catA mRNA is induced by H 2 O 2 until the late exponential (40-h) phase. Both Northern and S1 analyses demonstrated that the level of catA mRNA in the absence of H 2 O 2 increases slightly between 10 and 20 h of culture but decreases as cells enter stationary phase. The level of CatA polypeptide was determined under the same conditions by using anti-Cat4 antiserum. The results, presented in Fig. 4C, demonstrate that the level of CatA polypeptide increased slightly from 10 to 20 h, consistent with the increase in RNA level, but remained almost constant from 20 to 100 h. The level of the CatA polypeptide in rapidly growing cultures also increased about twofold in the presence of H 2 O 2 .
In Western blot analysis, we observed the emergence of a ϳ87-kDa polypeptide (CatB) that cross-reacted with anti-Cat4 antibody during the stationary phase of growth. This polypeptide is likely to correspond to one of the catalases enhanced in 100-h cultures, either Cat1 or the increased catalase comigrating with Cat4 (Fig. 3B). Another cross-reacting band (CatX) that is weaker than CatB was also observed. The significance of these bands in relation to total catalase activity is not clear at this point. Cells were also harvested from solid medium at stages of substrate, aerial, and sporulated mycelia and analyzed for CatA protein by Western blotting. The level of CatA did not change significantly, consistent with the behavior of catA gene expression observed in liquid culture (data not shown). The presence of the CatB polypeptide was observed in the later stages of growth when aerial mycelium and spores were present, consistent with the result in the liquid phase.
The characteristics of the S. coelicolor catA gene product and its mode of regulation suggest that S. coelicolor CatA is different from vegetative catalases of E. coli (HPI) and B. subtilis (catalase 1) in gene regulation and/or biochemical aspects. E. coli HPI is an rpoS-controlled catalase-peroxidase whose level increases during stationary phase (5,12). B. subtilis catalase 1 is controlled by spo0A and increases during stationary phase (2). Since the level of CatA gradually increases and is inducible by H 2 O 2 during the exponential growth phase, it is likely that the transcriptional level of the catA gene reflects the intracellular concentration of H 2 O 2 and/or the specific growth rate. As suggested for HPI of E. coli, it may play a major role in reducing the concentration of H 2 O 2 inside the cell (9).
Nucleotide sequence accession number. The nucleotide sequence of the catA gene has been deposited in the EMBL/ GenBank/DDBJ database (accession no. X96981).  Fig. 3. The level of catA transcript was assessed by Northern blot analysis (A) and S1 nuclease mapping (B). The amount of CatA polypeptide was estimated by Western blot analysis (C). Each sample contained 30 g of RNA (A), 50 g of RNA (B), or 50 g of total protein (C). Northern analysis revealed a transcript of 1.7 kb, consistent with the predicted size of monocistronic catA mRNA. In Western analysis, two weak cross-reacting bands were detected beside the major 56-kDa CatA protein; they are indicated as CatX (92 kDa) and CatB (87 kDa).