Drp35 has been identified as a protein that is induced in Staphylococcus aureus in response to exposure to certain antibiotics. Here we demonstrate that Drp35 is a lactonase that does not contribute directly to the resistance to the inducer antibiotics except for bacitracin. The detailed analysis on the expression of Drp35 revealed that in addition to a broad range of antibiotics, agents such as detergents that perturb the membrane integrity could induce its expression. The significance of this characteristic expression is discussed in relation to its activity similarity to the eukaryotic counterparts, paraoxonase family proteins.
Staphylococcus aureus is a Gram-positive bacterium that widely inhabits the skin surface and nasal cavity of warm-blooded animals. However, it is also a major pathogen that causes serious hospital infections because of the strains that have acquired antibiotic resistance. We previously identified a protein, Drp35, that was induced in S. aureus in response to treatment with antibiotics which affect the cell wall , but its actual physiological function is still unclear.
The cell wall affecting antibiotics have a variety of modes of action. β-Lactams inhibit the catalytic activity of penicillin binding proteins (PBPs) by the acylation of the active-site serine in the transpeptidase domain . Glycopeptide agents bind to the peptidoglycan precursors, UDP-N-acetylmuramyl-pentapeptides, and inhibit transglycosylation reactions . Fosfomycin inhibits UDP-N-acetylglucosamin enolpyruvoyl transferase (MurA) that is required for the first step in bacterial cell wall biosynthesis . Bacitracin binds to certain lipid carrier to block the supply of the cell wall components . These various modes of action result in cell autolysis by disrupting the balance between the rate of the cell wall synthesis and degradation by the autolytic enzymes. All the cell wall targeting antibiotics can induce Drp35 expression . This suggests the possibility that Drp35 might have a generic role in the action of all the cell wall affecting antibiotics.
In this report, we explored further the possible roles for Drp35 in the stress response and demonstrate that Drp35 is a functional protein with lactonase enzymatic activity. The detailed analysis of its expression profile revealed that Drp35 responds to a broad range of stresses affecting the membrane integrity. The significance of its characteristic activity and expression is discussed in relation to the eukaryotic counterparts, paraoxonase family proteins (PONs).
2Materials and methods
S. aureus strain N315, which is a pre-meticillin-resistant strain , was used to isolate the drp35 knockout mutant, NM6, and the Drp35 over expressing strain, NMDR, (described below). E. coli DH5α was used as the host for plasmid DNA preparation.
2.2Construction of drp35 disruptant, NM6
The drp35 disruptant, NM6, was generated by the method previously described . The drp35 targeting vector was constructed with pKILts plasmid and the 770-bp drp35 fragment amplified by PCR with a set of primers: 5′-TGAAAGGATTCATCATGAT-3′ and 5′-CACTCAAGCTTGACCATCAT-3′. The correct homologous recombination in NM6 was confirmed by the PCR and Southern blot analysis (data not shown). Western blot analysis also confirmed the absence of the Drp35 protein in NM6 (data not shown).
2.3Construction of drp35 overexpressing strain, NMDR
The drp35 gene containing the entire coding region was amplified by PCR from N315 with a set of primers: 5′-ATAAAATTGAAGCTTAAAACGAA-3′ and 5′-AAGACGCCTTGAATTCAGGTAGG-3′. The amplified DNA fragment was digested with Hin dIII and Eco RI, and cloned into pRIT5H  to generate pRIDR, which over expresses drp35 under the control of a proteinA promoter. The pRIDR plasmid was introduced into NM6 to generate a strain, NMDR. To generate a control strain NM6v, pRIT5H was introduced into NM6.
2.4Western blot analysis
Whole cell protein was extracted as described previously , and was centrifuged at 100,000g for 60 min at 4°C. The supernatant and the precipitate were collected as the soluble fraction and the membrane fraction, respectively. The protein concentrations were measured with DC Protein Assay (Bio-Rad Laboratories). Western blot analyses were performed as described previously . Protein bands were visualized using ECL western blotting analysis system (Amersham Biosciences).
2.5Antibiotic susceptibility tests
Minimal inhibitory concentration (MIC) of cell wall affecting antibiotics were determined according to National Committee for Clinical Laboratory Standards (NCCLS) standard method with Dry Plate Eiken (Eiken, Japan) using cation-supplemented Mueller-Hinton broth (Eiken, Japan).
2.6Preparation of the recombinant Drp35 protein
The drp35 gene was amplified by PCR with primers: 5′-TGAAATGAAAGGATCCATCATGATGT-3′ and 5′-CCCTAAATTTAAGTCGACTTTCATTC-3′. The amplified DNA fragment was digested with Bam HI and Sal I, and ligated into pGEX-4T–1 (Amersham Biosciences) to obtain the pGX4TDR. A fusion protein of glutathione-S-transferase and Drp35 was expressed in E. coli strain BL21, and bound to the glutathione-sepharose beads according to the manufacture's instruction. The Drp35 protein was excised out from the beads by thrombin treatment. The eluate containing Drp35 and thrombin was passed through benzamidine-sepharose 6B (Amersham Biosciences) column to remove the thrombin. The final eluate containing the recombinant Drp35 was dialyzed against a buffer containing 50 mM Tris–HCl (pH 8.0), 20 % (w/v) glycerol, and 1 mM CaCl2, and stored at 4°C until use. Throughout the purification procedure, 1 mM Ca2+ was included. The protein concentration was measured with DC Protein Assay (Bio-Rad Laboratories). The protein was analyzed by SDS–PAGE and visualized by CBB-staining to check the purity, which was more than 95% in every preparation.
Lactonase activity was tested as previously described . The reaction mixture (1 ml) constituted 50 mM Tris–HCl (pH7.5), 1 mM CaCl2, 1 mM dihydrocoumarin, and 0.6 μg ml−1 Drp35 protein. The generation of the reaction product, 3-(2-hydroxyphenyl)propionic acid, was measured as the increase of the absorbance at 270 nm in a U-3000 spectrophotometer (Hitachi). The lactonase activity for 2-coumaranone was measured as the increase of the absorbance at 274 nm. The molar extinction coefficient of 1295 and 876 M−1 cm−1 was used for the calculation of dihydrocoumarin and 2-coumaranone, respectively.
The N-acylated homoserine lactones (N-Butylyl-dl-homoserine lactone, N-Octanoyl-dl-homoserine lactone, N-Dodecanoyl-dl-homoserine lactone) were purchased from Fluka, and used to test the acyl-homoserine lactonase activity of Drp35 by colorimetric assays using cresol purple .
3.1Is Drp35 required for the antibiotic resistance?
The drp35 disruptant strain (NM6) and overexpressing strain (NMDR) were generated as described in Section 2. No significant change in MIC values of antibiotics including β-lactams, glycopeptide and fosfomycin was detected for all the strains tested except for a slight but significant change that was observed for the resistance to bacitracin: the drp35 disruption (NM6) rendered cells susceptible to bacitracin and the overexpression of drp35 (NMDR) reversed the effect (Table 1). These data suggested that the drp35 gene is not essential for cells to maintain the levels of resistance to most of the cell wall affecting antibiotics although they can effectively induce its expression.
Table 1. MIC values for S. aureus strains: N315, NM6, NMDR, and NM6v
Note that the MIC values of bacitracin was precisely determined by using a series of concentrations at 10 μg/ml intervals.
3.2Drp35 is functional protein with calcium-dependent lactonase activity
Drp35 displays similarity to eukaryotic PONs that can hydrolyze the ester bond in lactones and other promiscuous substrates [9–11] (Fig. 1). The values of identity were 13.8%, 14.2% and 16.3% for M. musculus PON1, PON2 and PON3, respectively. The structure and 29 important residues of PON1  are also shown below the alignment. Nine of the important residues (31%) were shared between Drp35 and one of the PON family proteins in this alignment. This suggested the possibility that Drp35 has an enzymatic activity similar to the eukaryotic PONs.
As expected, the purified Drp35 protein exhibited the lactonase activity when dihydrocoumarin or 2-coumaranone was used as the substrate (Fig. 2(a) and (b)), while we could not detect apparent hydrolytic activity for p-nitrophenylacetate (data not shown). The N-acylhomoserine lactonase activity, which is related to quorum sensing , was not detected in Drp35. The lactonase activity of Drp35 strictly depended on the calcium ions as is reported for paraoxonases ; the addition of EGTA or EDTA in the reaction mixture reduced its activity (data not shown). These results clearly demonstrate that Drp35 is a functional enzyme that can hydrolyze certain substrates, although the endogeneous substrate is still unknown.
3.3Expression profile of Drp35
We explored the signals that can induce Drp35 to gain insight into its physiological role. Cell wall-affecting antibiotics could induce the expression of Drp35 as previously reported (Fig. 3(a)). The drug responsive expression of Drp35 was not the feature specific to MRSA, but was common to other S. aureus strains (data not shown).
Other stresses such as high temperature (46°C), heavy metal (1 mM CuSO4), and osmotic stress (2 M NaCl) had no inductive effect. In contrast, interestingly, detergents including Nonidet P40, TritonX-100, SDS, and CHAPS effectively induced the expression of Drp35 (Fig. 3(b)). Hence Drp35 could be induced by the broad range of the agents such as antibiotics and detergents that affect the membrane integrity.
We also examined the cellular distribution of Drp35 to determine whether it is a membrane-bound protein. Immunoblot analysis showed that this was not the case, in fact, it was detected in the soluble fraction and never in the membrane fraction (Fig. 3(c)).
4.1Drp35 contributes to the bacitracin resistance
The broad range of cell wall affecting antibiotics had the similar inhibitory effects on strains expressing different amount of Drp35 (Table 1). In contrast, bacitracin uniquely changed its minimal inhibitory concentration in a Drp35 mutant. These results suggest that Drp35 has a function that can target the action of the bacitracin molecule but not of other antibiotics. Bacitracin is known to block the recycling of a lipid carrier (undecaprenol) that transfers the peptidoglycan precursors toward outside of the cells . A gene designated as bacA encoding an enzyme catalyzing a step in the undecaprenol synthesis pathway is known to contribute to the bacitracin resistance . Drp35 may also have certain role in the undecaprenol metabolism, or on the proper homeostasis of the plasma membrane where the lipid cycle exists (see below).
4.2Drp35 acts in cells with perturbed membrane integrity
The actual physiological role of Drp35 is still unclear, but we can speculate the situation where Drp35 has its function. One important finding was that any detergent could induce the Drp35 expression. The primary effect of detergents on cells would be the perturbation of the membrane integrity, and it can be speculated that the cell wall affecting antibiotics also have the same effect on the plasma membrane because the rigid cell wall is the important structural layer supporting it. Namely, Drp35 seems to function when the plasma membrane is disturbed rather than the cell wall itself, suggesting that its possible physiological role is related to the membrane homeostasis. This assumption well explains why Drp35 cannot contribute to the increase of the resistance to the antibiotics targeting on cell walls.
4.3Drp35 is the bacterial counterpart of eukaryotic paraoxonase family proteins
The homology found between staphylococcal Drp35 and eukaryotic PONs (Fig. 1) suggested that they share common functional characteristics. In fact, they act as lactonases [9,10]. In addition, similar to PONs, the calcium ion is essential for the lactonase activity of Drp35. Therefore, Drp35 is probably the bacterial functional counterpart of the eukaryotic PONs, and possibly shares its evolutionary origin with them. The Drp35/PON homologues are broadly conserved, suggesting that they have certain important roles in a broad range of living organisms. The PONs are divided into three groups: PON1, PON2, and PON3 . The physiological roles of them have not been established, but their substrate specificities have been examined in detail. In the very recent report, Draganov et al. reported that undecanolactone is the most efficient substrate for PON3 among tested lactones with a series of tail length . PONs are also interesting enzymes with regard to their ‘evolutionary’ plasticity of enzymatic characteristics [17,18]. The detailed substrate specificity of the bacterial Drp35 will be examined in the extensive future studies to explore the endogeneous substrate among the undecaprenol derivatives or other membrane components.
We thank Mr. Yu Ohki for his technical assistance.