ABSTRACT
A phosphatase secreted by Staphylococcus aureus strain 154 has previously been characterized and classified as a new member of the bacterial class C family of nonspecific acid phosphatases. As the acid phosphatase activity can be easily detected with a cost-effective plate screen assay, quantitatively measured by a simple enzyme assay, and detected by zymography, its potential use as a reporter system was investigated. The S. aureus acid phosphatase (sapS) gene has been cloned and expressed from its own regulatory sequences in Escherichia coli, Bacillus subtilis, and Bacillus halodurans. Transcriptional and translational fusions of the sapS gene with selected heterologous promoters and signal sequences were constructed and expressed in all three of the host strains. From the range of promoters evaluated, the strongest promoter for heterologous protein production in each of the host strains was identified, i.e., the E. coli lacZ promoter in E. coli, the B. halodurans alkaline protease promoter in B. subtilis, and the B. halodurans σD promoter in B. halodurans. This is the first report on the development of a class C acid phosphatase gene as a reporter gene with the advantage of being able to function in both gram-positive and gram-negative host strains.
Methods for the direct measurement of gene expression include mRNA detection with polynucleotide probes (Northern blot assays) or reverse transcriptase PCR methods, as well as protein detection methods that use antibodies (Western blot assays) or biological activities (9, 59). However, these methods are, in many cases, time-consuming and costly. Reporter genes provide an alternative method of genetic analysis that is faster and more convenient. Typically, reporter genes encode proteins that have a unique enzymatic activity or that are otherwise easily distinguishable from the mixture of intra- and extracellular proteins (2, 44). They have frequently been used to identify regulatory sequences, to monitor gene expression and function, and to characterize promoter strength and regulation (24, 37, 47, 57).
The choice of a reporter system is determined by a number of important criteria. These include the absence of activities similar to those of the reporter protein in the host organism and the availability of simple, rapid, and sensitive methods for the qualitative and quantitative assay of reporter protein activity. These methods should preferably allow assaying of the reporter protein activity in the presence of cellular components, thus obviating the need for purification steps prior to assay (21, 32). The most widely used reporter systems use genes that encode β-galactosidase (lacZ) (41, 50), chloramphenicol acetyltransferase (cat) (4, 34), and different sugar hydrolases, e.g., β-glucuronidase (gus) (21, 22). Although these reporter systems are convenient tools for semiquantitative plate-based assessment of promoter activities, more accurate quantification of promoter strength usually requires enzymatic assays, which typically involve bacterial cell disruption and addition of a substrate to drive the enzymatic reaction, followed by measurement of optical density (2). Another group of reporter systems is based on the emission of light (48, 60). In addition to the wild-type green fluorescent protein (GFP) from Aequorea victoria, many derivatives of GFP have been produced and subsequently used to monitor promoter activity both in the laboratory and in natural environments (6, 47, 48). However, naturally occurring fluorescence can lead to high background levels during in vitro and in vivo measurements. Alternative strategies have thus involved the luciferase-encoding luxAB genes, typically derived from Vibrio fischeri, Vibrio harveyi, and Photorhabdus luminescens (23, 30), and more recently the synthetic luxCDABE operon which alleviates the requirement for the addition of an exogenous aldehyde substrate in the light emission reaction (1, 15).
Since each reporter system has its own advantages and disadvantages that may limit its usefulness in specific host organisms and in specific types of studies, no single reporter gene is universally applicable (32). It is therefore desirable to have a number of reporter systems available for the same organism (20, 38). Consequently, modification of widely used reporter systems (7, 18, 54), the optimization of methods used for measuring reporter activity (17, 27, 51), and the isolation and evaluation of new reporters are continuing. The relevance of the latter is exemplified by several reports regarding the evaluation of new reporters, among others, β-galactosidase from Bacillus megaterium (45) and Thermus thermophilus (35), lichenase from Clostridium thermocellum (39), and α-galactosidase from Saccharopolyspora erythraea (40).
We have previously isolated and characterized a novel class C nonspecific acid phosphatase secreted by Staphylococcus aureus strain 154 (11). The enzyme, designated SapS and encoded by sapS, is a stable monomeric protein of moderate size (296 amino acids, 30 kDa) that undergoes proteolytic cleavage of the N-terminal 31-amino-acid signal peptide to yield the mature protein. On the basis of its moderate size and the ease with which enzymatic activity tests may be performed (11), the present study focused on the development and evaluation of the SapS acid phosphatase as a reporter for the characterization of promoters and signal sequences in a gram-negative host (Escherichia coli), as well as in mesophilic and moderately thermophilic gram-positive (B. subtilis and B. halodurans, respectively) hosts. E. coli and members of the species Bacillus are the most frequently used prokaryotes for the production of heterologous proteins (26, 58) and were therefore included in this study to evaluate the sapS reporter system. The use of the reporter gene was evaluated in B. halodurans since it harbors the lac operon (19), and the commonly used lacZ reporter system can therefore not be used in this host organism. Furthermore, B. halodurans is currently being evaluated as a surface display expression system (8).
MATERIALS AND METHODS
Bacterial strains and growth conditions. E. coli DH10B (F−mcrA Δ(mrr-hsdRMS-mcrBC) (φ80dlacZΔM15) ΔlacX74 endA1 recA1 deoR Δ(ara-leu)7697 araD139 galU galK nupG rpsL λ−), obtained from Invitrogen, was used as an intermediary cloning host. Expression studies were done with E. coli CU1867, a BL21(DE3) strain with the chromosomal acid phosphatase appA gene disrupted (33), B. subtilis 154 (Δapr Δnpr amy spo) (42), and B. halodurans BhFC04 (ΔwprA Δhag) (12). E. coli and B. subtilis were cultured at 37°C in Luria-Bertani (LB) medium (0.5% [wt/vol] yeast extract, 1% [wt/vol] tryptone, 1% [wt/vol] NaCl, pH 7). When appropriate, E. coli growth media were supplemented with ampicillin (100 μg/ml), chloramphenicol (20 μg/ml), or erythromycin (300 μg/ml) and B. subtilis growth media were supplemented with chloramphenicol (5 μg/ml) or erythromycin (10 μg/ml). B. halodurans was grown at 37°C in LB medium (pH 8.5), and chloramphenicol (5 μg/ml) was added when appropriate.
Recombinant DNA techniques.Plasmid DNA was extracted with a Plasmid Midiprep Kit (QIAGEN, Hilden, Germany) and a Perfectprep Plasmid Mini Kit (Eppendorf, Hamburg, Germany). DNA fragments were purified from agarose gels by use of a DNA Extraction Kit (Fermentas, St. Leon-Rot, Germany). Procedures for DNA manipulations were carried out as described by Sambrook and Russell (43). Enzymes were obtained from Fermentas (St. Leon-Rot, Germany) and Roche Diagnostics (Mannheim, Germany) and were used according to the manufacturers’ protocols. Unless otherwise indicated, plasmid constructions were first established in E. coli DH10B and then transferred to E. coli CU1867, B. subtilis 154, and B. halodurans BhFC04. Transformation of bacteria was performed by electroporation for E. coli (10) and by protoplasting according to published procedures for B. subtilis (5) and B. halodurans (8). PCR was performed with Biotaq DNA polymerase (Bioline USA Inc., Randolph, MA) and a Progene thermocycler (Techne, Burlington, NJ). The oligonucleotides used in this study were obtained from Inqaba Biotechnical Industries, Pretoria, South Africa. Chromosomal DNA was extracted from S. aureus 154 and B. halodurans BhFC04 according to the method of Lovett and Keggins (29), except that lysozyme was added to a final concentration of 10 mg/ml. Nucleotide sequencing of all PCR products was performed with the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit v3.0 (Applied Biosystems, Foster City, CA), followed by resolution on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems), in accordance with the manufacturer's instructions. All plasmid constructions were verified by restriction endonuclease digestion, followed by agarose gel electrophoresis.
Plasmids.Plasmids pNW33N, an E. coli-Bacillus-Geobacillus shuttle vector obtained from the Bacillus Genetic Stock Center (Columbus, OH), and pMG36e, a Lactococcus expression vector (52) which also replicates in E. coli and B. subtilis, were used to express the S. aureus acid phosphatase gene (sapS) in the gram-positive and gram-negative host strains. In all instances, except for pMG36e-SapS, pNW33N served as the genetic backbone into which different transcriptional and translational fusions were inserted to evaluate the sapS gene as a reporter (Fig. 1). The SapS enzyme was processed differently in S. aureus and E. coli compared to the Bacillus spp. (31 and 43 N-terminal amino acids were deleted, respectively, to produce the mature SapS protein). Consequently, translational fusions were made with both the mature sapS gene determined for E. coli and S. aureus and the Bacillus sp. truncated mature sapS gene (Δ12). The oligonucleotides used in this study are listed in Table 1. S. aureus strain 154 chromosomal DNA was used for amplification of the sapS gene and its derivatives. B. halodurans BhFC04 chromosomal DNA was used as the template for PCR amplification of the β-glucanase and alkaline protease promoter and signal sequences and the σD promoter. The SPO2 promoter was obtained from plasmid pPL608 as a 300-bp EcoRI DNA fragment (46). The following vectors were constructed.
Schematic representation of the vector constructs harboring the S. aureus sapS gene for expression in E. coli CU1867, B. subtilis 154, and B. halodurans BhFC04. Abbreviations: Pgluc, B. halodurans β-glucanase promoter; Papr, B. halodurans alkaline protease promoter; PSPO2, Bacillus temperate phage SPO2 promoter; $$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$, B. halodurans hag gene (flagellin protein) promoter; PlacZ, E. coli lacZ promoter; P32, strong lactococcal promoter; PsapS, S. aureus sapS promoter. Δ12 serves to indicate that the N-terminal sequence of the S. aureus SapS protein expressed in B. halodurans was 12 amino acids shorter than that determined for the native S. aureus SapS protein (11).
Oligonucleotides used in this study
pNW33-SapS.The 1.140-kb full-length sapS gene was PCR amplified with primers Sap-F and Sap-R1.
pNW33N1.The 227-bp β-glucanase promoter (Pgluc) fragment was PCR amplified with primers Glu-F and Glu-R, and the 888-bp sapS gene fragment, lacking the ATG initiation codon (ΔATGsapS), was PCR amplified with primers ΔATGSap-F1 and Sap-R2.
pNW33N2.The 315-bp β-glucanase promoter and signal peptide (Pgluc+sp) was PCR amplified with primers Glu-F and GluS-R, and the 798-bp sapS DNA fragment, lacking the N-terminal 31-amino-acid signal peptide (Δ31sapS), was PCR amplified with primers Δ31Sap-F and Sap-R3.
pNW33N3.A 762-bp sapS fragment, lacking the N-terminal 43 amino-acid signal peptide as determined for B. halodurans BhFC04 (Δ12sapS), was PCR amplified with primers Δ43Sap-F and Sap-R3 and ligated to the 315-bp β-glucanase promoter and signal peptide (Pgluc+sp) from pNW33N2.
pNW33N4.The 280-bp alkaline protease promoter (Papr) was PCR amplified with primers Apr-F1 and Apr-R, and the 888-bp sapS DNA fragment, lacking the ATG initiation codon (ΔATGsapS), was PCR amplified with primers ΔATGSap-F2 and Sap-R2.
pNW33N5.The 327-bp alkaline protease promoter and signal peptide (Papr+sp) was PCR amplified with primers Apr-F2 and AprS-R and ligated to the 798-bp Δ31sapS fragment from pNW33N2.
pNW33N6.The 327-bp alkaline protease promoter and signal peptide (Papr+sp) from pNW33N5 was ligated to the 762-bp Δ12sapS DNA fragment from pNW33N3.
pNW33N7.The SPO2 promoter was cloned into pNW33N, and the plasmid was designated pNWSpo. The 891-bp sapS gene, starting from the ATG initiation codon, was amplified by PCR with primers ATGSap-F and Sap-R2 and ligated to pNWSpo to generate pNW33N7.
pNW33N8.The 230-bp sigma D promoter ($$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$) fragment was PCR amplified with primers Sig-F and Sig-R. The 888-bp sapS gene fragment, lacking the ATG initiation codon (ΔATGsapS), was generated by PCR with primers ΔATGSap-F3 and Sap-R4.
pNW33N9.The 905-bp sapS gene, including the putative ribosome binding site (RBS), was obtained by PCR with primers RBS-Sap-F and Sap-R3 and ligated to pNW33N. The resulting plasmid, pNW33N9, harbored the vector-borne lacZ promoter (PlacZ) translationally fused to the sapS reporter gene.
pMG36e-SapS.The 888-bp sapS gene fragment, lacking the ATG initiation codon (ΔATGsapS), was generated by PCR with primers ΔATGSap-F4 and Sap-R5. The sapS gene fragment was ligated into pMG36e, an expression vector that harbors the strong lactococcal P32 promoter and an ATG initiation codon, thus placing the reporter gene fragment in phase with the initiation codon.
Protein sample preparation and protein concentration determination.Bacterial strains harboring the plasmid constructs were inoculated into LB medium with the appropriate antibiotics and incubated at 37°C for 24 h on a rotary shaker (175 rpm). Protein samples from the cultures were prepared as described by Van der Vaart et al. (53), with the following modifications. For cell fractionation, 40 ml of the respective cultures was harvested after a 15-min centrifugation at 12,000 × g at 4°C. The cell-free supernatants (extracellular fraction) were retained, and the proteins were precipitated with ice-cold acetone prior to being suspended in 0.1 M sodium acetate buffer (pH 5). The cell pellets were washed once with sterile distilled water, resuspended in 5 ml of 0.1 M sodium acetate buffer (pH 5), and sonicated on ice for 20 min with a model HD2070 Sonoplus Ultrasonic Homogenizer (Bandelin Electronic, Berlin, Germany). The cell lysate was clarified by centrifugation at 12,000 × g for 15 min, and the supernatant, considered the intracellular fraction, was recovered. The pellet (cell wall fraction) was washed once with sterile distilled water and resuspended in 5 ml of 0.1 M sodium acetate buffer (pH 5). Whole cell protein samples were prepared by harvesting the cells from 5 ml of the respective cultures by centrifugation as described above. The cell pellets were washed with sterile distilled water and resuspended in 5 ml of 0.1 M sodium acetate buffer (pH 5). The protein concentration of samples was determined by the method of Bradford (3), with the Bio-Rad protein assay kit (Bio-Rad, Hercules, CA) and with bovine serum albumin as the standard.
Qualitative, quantitative, and zymographic detection of acid phosphatase activity.Colonies were grown on LB agar supplemented with the appropriate antibiotic and screened for acid phosphatase activity by flooding the surface with 0.1 M sodium acetate (pH 5) containing 0.1% α-naphthyl phosphate (Roche Diagnostics) and 0.2% Fast Garnett GBC salt (Sigma-Aldrich, Aston Manor, South Africa). Acid phosphatase-positive colonies produce a black precipitate.
Acid phosphatase activity was quantified according to the method of Golovan et al. (14), with the following modifications. The assays were performed by incubating 200 μl of enzyme preparation with 200 μl of pNPP substrate (Roche Diagnostics), at a final concentration of 25 mM, in 0.1 M sodium acetate (pH 5). Following incubation at 37°C for 30 min, the reaction was terminated by the addition of 1 ml of 1 M NaOH and the liberated p-nitrophenol (pNP) was measured at 405 nm. The extinction coefficient of p-nitrophenyl was taken to be 18.5 cm2 · μmol−1 (56), and 1 U of enzyme activity was defined as the amount of enzyme able to release 1 μmol of pNP per min under the assay conditions used. All assays were performed in triplicate, and the results are expressed as means ± standard deviations.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed with 12% polyacrylamide gels by the method of Laemmli (25) after the samples had been heated at 37°C for 30 min. Molecular weight markers (Bio-Rad) were included in each gel. Following electrophoresis, the gels were either stained with Coomassie brilliant blue R-250 to visualize the protein bands or incubated for 16 h at room temperature in several changes of renaturation buffer for zymographic analysis (16). After renaturation treatment, gels were equilibrated for 1 h at 37°C in 0.1 M sodium acetate buffer (pH 5) and incubated at 37°C for 15 min to 1 h in 0.1 M sodium acetate (pH 5) containing 0.1% (wt/vol) α-naphthyl phosphate and 0.2% (wt/vol) Fast Garnet GBG salt (13). Phosphatase activity was indicated by the presence of black-stained bands.
Amino-terminal amino acid sequencing.Amino-terminal amino acid sequencing was determined as described previously (11).
RESULTS
Expression of the S. aureus sapS gene in E. coli, B. subtilis, and B. halodurans.Plasmid pNW33-SapS, harboring the promoter and coding region of the S. aureus SapS preprotein (signal sequence and mature protein) was transformed into E. coli CU1867, B. subtilis 154, and B. halodurans BhFC04. The acid phosphatase enzyme was successfully expressed in all three host strains as determined by in vitro enzyme assays (Fig. 2) and zymography (Fig. 3). No acid phosphatase activity was detected with the in vitro enzyme assays of the host strains harboring the pNW33N vector. In contrast to Bacillus spp., where the highest acid phosphatase activity was obtained for the whole-cell fractions, in E. coli the highest activity was obtained for the intracellular fraction (Fig. 2).
Extracellular, whole-cell, and intracellular in vitro acid phosphatase activity (act) results of the host strains harboring pNW33-SapS. A, E. coli CU1867; B, B. subtilis 154; C, B. halodurans BhFC04.
Zymographic analysis of S. aureus SapS acid phosphatase activity in E. coli CU1867(pNW33-SapS), Bacillus 154(pNW33-SapS), and S. aureus (wild type). Lane 1, low-range protein molecular weight marker (Bio-Rad); lane 2, S. aureus supernatant fraction; lane 3, E. coli CU1867(pNW33-SapS) cell wall fraction; lane 4, B. subtilis 154(pNW33-SapS) cell wall fraction; lane 5, B. halodurans BhFC04(pNW33-SapS) cell wall fraction.
Zymographic analysis of the cell wall fractions of the three host strains harboring pNW33-SapS was performed following sodium dodecyl sulfate-polyacrylamide gel electrophoresis and protein renaturation (Fig. 3). The cell wall fractions were chosen as they gave rise to high activity levels. For E. coli, the SapS activity band was found at the molecular mass position of the mature S. aureus 154 acid phosphatase protein band, indicating that the processing had occurred at or close to the cleavage site determined previously for the S. aureus 154 SapS protein (11). The molecular masses of the acid phosphatase activity bands obtained for the B. subtilis and B. halodurans cell wall fractions (Fig. 3, lanes 4 and 5) were lower than that obtained for E. coli (Fig. 3, lane 3). In order to determine if the protein was processed differently in the gram-positive Bacillus spp., N-terminal sequencing of the enzyme was performed. The N-terminal sequence of the S. aureus SapS protein expressed in gram-positive B. halodurans was determined to be NH2-SIPASQKANL, which is 12 amino acids shorter than the native S. aureus SapS protein N-terminal sequence (11). Consequently, the coding regions of both the S. aureus 154 and B. halodurans BhFC04 mature SapS proteins were included in the vector construction.
Evaluation of heterologous promoters and signal sequences with sapS as a reporter gene in E. coli CU1867.To ascertain the feasibility of using the sapS gene as a reporter gene in E. coli CU1867, the acid phosphatase activity of the host strain harboring the reporter gene constructs (Fig. 1) was determined qualitatively with the plate screen assay (Fig. 4A) and quantitatively with in vitro enzyme assays (Fig. 4B).
(A) Plate screen assay showing acid phosphatase activities of 24-h cultures of E. coli CU1867 harboring the various constructs. Control (C), pNW33N; 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNw33N8; 9, pNW33N9; 10, pNW33-SapS. (B) Whole-cell in vitro acid phosphatase activity (act) results of 24-h cultures of E. coli CU1867 harboring different promoter constructs. 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNW33N8; 9, pNW33N9; 10, pMG36e-SapS.
E. coli CU1867 harboring the pNW33N vector showed no acid phosphatase activity after activity staining (Fig. 4A). The host strain harboring pNW33-SapS stained pitch black, indicating high levels of enzyme activity. Various levels of brown to black color development were detected for the host strain harboring the heterologous promoter-reporter gene constructs pNW33N1 through pNW33N9, indicating differences in promoter strength. No black color development was detected for E. coli CU1867(pNW33N8), indicating that the B. halodurans$$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$ gene ligated to the reporter gene was not expressed.
In order to quantify the acid phosphatase activity results, the production levels of the sapS enzymatic reporter genes from the various heterologous promoters and signal sequences were monitored after 24 h by in vitro acid phosphatase activity assays. In E. coli CU1867, the highest enzyme activity (835 mU/mg of protein) was obtained for PlacZ ligated to the reporter gene (pNW33N9), followed by activities obtained for the Papr-reporter gene construct pNW33N4 (589 mU/mg) and the Papr+sp-reporter gene construct pNW33N5 (541 mU/mg). The activity measured for Papr+sp ligated to the truncated reporter gene (pNW33N6) was approximately fivefold less (81 mU/mg of protein). A lower level of phosphatase activity was obtained for the Pgluc-reporter gene construct pNW33N1 (189 mU/mg) than for Papr-reporter gene construct pNW33N4 (589 mU/mg). This result indicated that the B. halodurans BhFC04 alkaline protease promoter is a stronger promoter than the β-glucanase promoter. The enzyme activity determined for the lactococcal promoter P32 ligated to the reporter gene (pNW33N10) was 177 mU/mg of protein. The enzyme activity determined for the Bacillus temperate phage PSPO2 ligated to the reporter gene (pNW33N7) was 475 mU/mg of protein. In accordance with the plate screen assay, no activity was detected from the B. halodurans$$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$-reporter gene construct (pNW33N8) in E. coli CU1867. A low level of extracellular enzyme activity (52 mU/mg) was detected for E. coli harboring the PlacZ promoter (pNW33N9) after 24 h. The extracellular activity can probably be ascribed to leakage, as opposed to secretion, in the E. coli host strain. No extracellular activity was detected for any of the other constructs expressed in E. coli.
Evaluation of heterologous promoters and signal sequences with sapS as a reporter gene in B. subtilis 154.The acid phosphatase activity of B. subtilis 154 harboring the various constructs was determined qualitatively (Fig. 5A) and quantitatively (Fig. 5B).
(A) Plate screen assay showing acid phosphatase activities of 24-h cultures of B. subtilis strain 154 harboring the various constructs. Control (C), pNW33N; 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNW33N8; 9, pNW33N9; 10, pNW33-SapS. (B) Whole-cell in vitro acid phosphatase activity (act) results of 24-h cultures of B. subtilis strain 154 harboring the different promoter constructs. 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNW33N8; 9, pNW33N9; 10, pMG36e-SapS.
B. subtilis 154 harboring pNW33N showed no acid phosphatase activity in the plate screen assay (Fig. 5A). The host strain harboring pNW33-SapS stained pitch black, indicating high levels of enzyme activity. As for the E. coli CU1867 host strain, various levels of brown to black color development were observed for the heterologous promoter-reporter gene constructs pNW33N1 through pNW33N9. B. subtilis 154 harboring the Papr-reporter gene construct (pNW33N4), the Papr+sp-mature reporter gene construct (pNW33N5), and the Papr+sp-truncated mature reporter gene construct (pNW33N6) stained black, indicating high levels of reporter gene activity. No black color development was detected for B. subtilis(pNW33N8), indicating that $$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$ ligated to the reporter gene was not expressed in B. subtilis 154. Similarly, no activity was observed for B. subtilis(pNW33N9), showing that the E. coli PlacZ-reporter gene construct was not expressed in the gram-positive host strain.
In order to quantify the acid phosphatase activity results in B. subtilis 154, the production levels of the sapS enzymatic reporter gene from the various heterologous promoters and signal sequences were monitored after 24 h by in vitro acid phosphatase activity assays. The highest enzyme activity (1,923 mU/mg of protein) was obtained for the Papr+sp-mature reporter gene construct pNW33N5, followed by activities obtained for the Papr+sp-truncated mature reporter gene construct pNW33N6 (1,620 mU/mg) and the Papr-reporter gene construct pNW33N4 (968 mU/mg). A lower level of phosphatase activity was obtained for the Pgluc-reporter gene construct pNW33N1 (65 mU/mg) than for Papr-reporter gene construct pNW33N4 (968 mU/mg). The in vitro assay results correlated with the plate screen assay; i.e., the recombinant strains harboring the β-glucanase promoter-reporter gene constructs (pNW33N1, pNW33N2, and pNW33N3) stained lighter than the recombinant strains harboring the alkaline protease promoter-reporter gene constructs (pNW33N4, pNW33N5, and pNW33N6). As for E. coli, this result indicated that the B. halodurans BhFC04 alkaline protease promoter is a stronger promoter than the β-glucanase promoter. The P32 lactococcal promoter was also effectively recognized, since 166 mU/mg enzyme activity was measured with the pNW33N10 construct. No acid phosphatase activity was detected for the PlacZ-reporter gene construct (pNW33N9) in the host strain. This is not surprising since Bacillus is very stringent in its recognition of promoters (36). Extracellular acid phosphatase activity was detected for the Papr+sp-mature reporter gene construct pNW33N5 (60 mU/mg) and the Papr+sp-truncated mature reporter gene construct pNW33N6 (44 mU/mg). No extracellular activity was detected for any of the other constructs evaluated in this host strain (results not shown).
Evaluation of heterologous promoters and signal sequences with sapS as a reporter gene in B. halodurans BhFC04.The acid phosphatase activity of B. halodurans BhFC04 harboring the expression vector constructs was determined qualitatively (Fig. 6A) and quantitatively (Fig. 6B).
(A) Plate screen assay showing acid phosphatase activities of 24-h cultures of B. halodurans BhFC04 harboring the various constructs. Control (C), pNW33N; 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNw33N8; 9, pNW33N9; 10, pNW33-SapS. (B) Whole-cell in vitro acid phosphatase activity (act) results of 24-h cultures of B. halodurans BhFC04 harboring the different promoter constructs. 1, pNW33N1; 2, pNW33N2; 3, pNW33N3; 4, pNW33N4; 5, pNW33N5; 6, pNW33N6; 7, pNW33N7; 8, pNW33N8; 9, pNW33N9.
B. halodurans BhFC04 harboring pNW33N showed no acid phosphatase activity in the plate screen assay (Fig. 6A). The host strain harboring pNW33-SapS stained black, indicating high levels of expression of the sapS gene from its own promoter and signal sequence. E. coli CU1867 harboring the $$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$-reporter gene construct (pNW33N8) stained light brown in the plate screen assay. No color was detected for the host strain harboring any of the other promoter-reporter gene constructs (Fig. 6A).
Of the heterologous promoter-reporter gene constructs, the best activity (18 mU/mg of protein) was obtained with B. halodurans BhFC04 harboring the $$mathtex$$\(P_{{\sigma}}^{D}\)$$mathtex$$-reporter gene construct (pNW33N8). Since the acid phosphatase activity measured in this host strain was very low and no enzyme activity could be detected for the extracellular fractions of any of the B. halodurans transformants harboring the various constructs, the reporter gene could not be used for the evaluation of the efficacy of the isolated B. halodurans gluc and apr signal sequences for the extracellular production of heterologous proteins.
DISCUSSION
In this study, we evaluated the S. aureus acid phosphatase SapS enzyme as a reporter for promoter and signal sequence characterization in E. coli CU1867, B. subtilis 154, and B. halodurans BhFC04. Initial studies focused on determining whether the sapS gene was expressed in the three host strains from its own promoter and signal sequence. A zymogram of active acid phosphatase led to the determination of the molecular masses of the sapS gene expressed in the three different host strains and showed that the enzyme was processed differently in E. coli than in B. subtilis and B. halodurans. The precursors of secreted proteins from gram-positive bacteria generally have longer and more hydrophobic signal peptides than those of gram-negative bacteria (55). In S. aureus, alanine is at the −1 position and lysine is at the +1 position of the signal peptide of the sapS gene, as determined by N-terminal sequencing. In B. halodurans BhFC04, alanine was found to be at the −1 position and serine was at the +1 position in the signal peptide of the sapS gene. In B. subtilis, alanine is the predominant residue (>90%) at the −1 and −3 positions of the Bacillus signal peptides (31). Furthermore, for B. subtilis the SapS activity band was found at the molecular mass position of the mature B. halodurans BhFC04 acid phosphatase protein band, indicating that the processing had occurred at or close to the cleavage site determined for B. halodurans BhFC04.
The suitability of the sapS gene as an expression reporter system was evaluated by the ligation of a range of heterologous promoters that included both well-known and newly isolated promoters and signal sequences. The expression and secretion abilities of the transcriptional and translational fusion products were estimated by acid phosphatase activity determination. Since Nagarajan et al. (31a) reported the successful use of the Bacillus amyloliquefaciens neutral and alkaline protease promoter and signal sequences for overexpression of heterologous proteins, the B. halodurans alkaline protease gene promoter and signal sequence was isolated and evaluated for the ability to express and secrete the reporter gene. The thermostable endo-(1,3-1,4)β-glucanase-encoding gene from B. halodurans has previously been expressed successfully in E. coli, B. subtilis (28), and Lactobacillus plantarum (unpublished results). Therefore, the β-glucanase gene promoter and signal sequence was isolated and evaluated for the ability to express and secrete the reporter gene. It has been reported that the Bacillus temperate phage SPO2 promoter functions well in B. subtilis (46) and was included in the range of promoters to be evaluated. The strong lactococcal promoter P32 was used to express genes from prokaryotic and eukaryotic origins in lactococci, B. subtilis, and E. coli (52). Thus, sapS gene expression from this promoter was also evaluated. The σD promoter region of the B. halodurans hag gene (flagellin protein) was included in the range of promoters evaluated as it was used in the development of a surface display system in B. halodurans Alk36 (8).
Enzyme studies performed with E. coli, B. subtilis, and B. halodurans harboring the various transcriptional and translational reporter gene constructs demonstrated that the sapS gene can be used as a reporter in all three of the host strains. The enzyme activity obtained for recombinant strains harboring the heterologous promoter-reporter gene constructs was less than the activity measured for sapS expressed from its own promoter and signal sequence. This could be due to the reduction in the quantity of fusion proteins produced depending on the differences in promoter strengths and not necessarily from misfolding. The decrease in enzymatic activity of fusion proteins containing heterologous promoters was also found for GFP, lacZ, and luciferase. GFP and luciferase reporters retain approximately 5% of their activity compared to nonfused controls (49). Piruzian et al. (39) reported a decrease in thermostable lichenase (LicB) and Gus activities for cells expressing the fusion constructs compared to the native proteins. SapS activity was obtained with translational fusions of isolated promoter and signal sequences to the native mature 798-bp sapS gene sequence, as well as the truncated 762-bp sapS gene sequence in all three of the host strains, showing the versatility of the sapS gene as an enzymatic reporter gene. The facts that very low levels of extracellular SapS activity were detected for the constructs in the three host strains and the activity was whole cell associated make it unsuitable for the isolation or evaluation of signal peptides for the extracellular production of heterologous proteins.
From the range of promoters evaluated with this system, the strongest promoter for the expression of heterologous proteins was easily identified in each of the three host strains. These include the E. coli lacZ promoter in E. coli, the B. halodurans alkaline protease promoter in B. subtilis, and the B. halodurans σD promoter in B. halodurans. Although the Bacillus temperate phage SPO2 promoter was reported to be a strong promoter for heterologous protein production in B. subtilis (46), similar activity levels were obtained in both E. coli CU1867 and B. subtilis 154. Although the lactococcal P32 promoter was used for heterologous protein production in E. coli and B. subtilis (52), we found in our study that the B. subtilis temperate phage SPO2 and B. halodurans BhFC04 alkaline protease promoters both gave rise to higher levels of enzyme activity in E. coli and B. subtilis than did the P32 promoter.
This is the first report on the development of a class C acid phosphatase gene as a reporter gene with the advantage of being able to function in both gram-positive and gram-negative host strains. Furthermore, the sapS enzymatic reporter gene has shown potential for use in the characterization and evaluation of a range of heterologous promoters that could find application in the development of expression vectors for improved production of industrially important proteins.
FOOTNOTES
- Received 9 May 2007.
- Accepted 19 September 2007.
- Copyright © 2007 American Society for Microbiology
