INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Staphylococcus hyicus is known to be a causative agent of exudative epidermitis (EE) in pigs (28). EE is a generalized infection of the skin characterized by greasy exudation, exfoliation, and vesicle formation (7, 13).

Amtsberg (1) has shown that the culture filtrate of S. hyicus contains an exotoxin that causes exfoliation in piglets. He has also suggested that the exfoliative activity of this exotoxin is similar to the exfoliative toxin (ET) produced by Staphylococcus aureus. We isolated this exotoxin from the culture supernatant of S. hyicus P-1 and designated it SHET (26). In 1993, we described some of the characteristics of purified SHET from the culture supernatant of S. hyicus (29). The molecular mass of SHET was estimated to be 27 kDa, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and it was determined that SHET differs from ETA and ETB in antigenicity and animal susceptibility (29).

We have recently reported that the SHET-producing strain of S. hyicus causes EE in pigs while the non-SHET-producing strain does not and that SHET can be divided into more than two serotypes (30; H. Sato, T. Tanabe, T. Watanabe, K. Teruya, A. Ohtake, H. Saito, and N. Maehara, Proc. 14th IPVS Cong. Italy, p. 339, 1996). Andresen et al. (2, 3) have also reported that SHET has many serotypes. In this study, we detected the large plasmid in several strains of S. hyicus. We then confirmed the antigenic differences between SHETs from the plasmid-carrying strain and the plasmidless strain and between the SHET-producing ability of the plasmid-carrying strain and its plasmid-cured substrains.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains. Ninety-two S. hyicus strains, including P-1 (27), P-10, and P-23 (30; Sato et al., 14th IPVS Cong.) were used in this study. These bacteria were lyophilized and stored at 4°C. S. aureus strains ZM and J-sETB-8 were kindly supplied by S. Sakurai, Jikeikai University School of Medicine. All strains were lyophilized and stored at 4°C. The lyophilized bacteria were suspended in a heart infusion (HI) broth (Difco Laboratories, Inc., Detroit, Mich.) and cultured on HI agar (Difco) at 37°C for 18 h. The bacteria were then suspended in 20% glycerol and stored at 80°C.

SHET production test. S. hyicus strains cultured on HI agar at 37°C for 18 h were suspended in Dulbecco's phosphate-buffered saline (PBS; pH 7.2) without MgCl₂ and CaCl₂ at a concentration of 10⁹ CFU/ml. Then, 0.3 ml of this suspension was inoculated into 30 ml of TY broth (8). This culture was incubated at 37°C for 18 h in a Bio-shaker BR-160 LF (Taitec Inc., Tokyo, Japan) operated at 75 oscillations per min. The culture was centrifuged at 10,000 × g for 20 min at 4°C, and the supernatant was passed through a 0.45-µm-pore-size membrane filter (Toyo Roshi, Inc., Tokyo, Japan). Then, 0.5 ml of the culture filtrate was used as the undiluted culture filtrate (UCF) for inoculation of 1-day-old chickens. The remaining culture filtrate was concentrated to 3 ml with a UP-20 ultra-filter (Toyo Roshi). This solution was used as a 10-fold-concentrated culture filtrate (CCF) for the 1-day-old chicken inoculation test and the Western blotting.

Isolation of plasmid DNA. Cleared lysates were prepared from 30-ml TY cultures that were shaken overnight by an adaptation of the method described by Novick and Bouanchaud (17, 18). The cells were harvested and washed once with TNE (0.1 M Tris [pH 7.5], 0.1 M EDTA [pH 7.5], 0.15 M NaCl) buffer. The washed cells were resuspended in 2.0 ml of lysis buffer (10 mM Tris [pH 7.0], 10 mM MgCl₂, 2.5 M NaCl); 100 µl of lysostaphin (1 mg/ml; Sigma Chemical Inc., St. Louis, Mo.) in 50 mM Tris-HCl (pH 7.5) was then added, and the cells were incubated at 37°C for 20 min. A 3.0-ml portion of the lysis mixture (0.36 M EDTA [pH 8.0], 10% Brij 58, 0.4% sodium deoxycholate; 8:1:1, by volume) was then added. The lysate was cleared by centrifugation at 18,000 × g for 30 min. The cleared lysate was diluted with a 0.5 volume of TE (10 mM Tris-HCl [pH 8.0], 1 mM EDTA [pH 8.0]) buffer and then deproteinized by extraction three times with phenol equilibrated in 10 mM Tris-HCl (pH 8.0). Traces of phenol were removed by chloroform extraction. Sodium acetate was added to a concentration of 0.3 M, and the nucleic acids were precipitated by adding 2 volumes of absolute ethanol, followed by incubation at 20°C for 18 h. The DNA was pelleted by centrifugation at 18,000 × g for 10 min. The DNA pellet was dried under vacuum and dissolved in 100 µl of TE buffer. The plasmid DNA content was examined by electrophoresis in 0.8% (wt/vol) agarose gels as follows.

Isolation of chromosomal DNA. Chromosomal DNA was prepared from 30-ml TY cultures that were shaken overnight by an adaptation of the method described by Lindberg et al. (15). The cells were harvested and washed once with TNE buffer. The washed cells were resuspended in 5.0 ml of TNE buffer. A 300-µl portion of lysostaphin (1 mg/ml) in 50 mM Tris-HCl (pH 7.5) was then added, and the cells were incubated at 37°C for 30 min, followed by treatment with pronase (final concentration, 2 mg/ml; Sigma) at 37°C for an additional 60 min. A 0.4-ml portion of SDS (5% in 45% ethanol) was added, and the mixture was shaken by hand for 30 min at room temperature. The lysed suspension was then mixed with an equal volume of phenol equilibrated in 10 mM Tris-HCl (pH 8.0), and the mixture was shaken by hand for 15 min at room temperature. The resulting emulsion was broken by slow-speed centrifugation. The aqueous phase was collected, and the phenol treatment was repeated three times. The aqueous phase was extracted with chloroform several times to remove the phenol. Finally, sodium acetate was added to the aqueous phase to a final concentration of 0.3 M. The nucleic acids were precipitated by gently mixing the solution with 2 volumes of cold ethanol. After incubation at 20°C for 1 h, a thread-like precipitate was collected and dissolved in 5 ml of TE buffer. To remove the RNA, ribonuclease A (Sigma) was added to a final concentration of 50 µg/ml, and the mixture was incubated at 37°C for 30 min. The DNA was precipitated by the addition of 2 volumes of cold ethanol and then collected and dissolved in 200 µl of TE buffer.

Restriction endonuclease analysis. To determine the size of the large plasmid, the restriction endonucleases EcoRI, BamHI, and HindIII (Nippon Gene, Inc., Tokyo, Japan) were used according to the manufacturer's instructions, except that the incubation periods were extended to 2 h. The patterns of the DNA fragments were examined by electrophoresis on an 0.8% agarose gel. Lambda phage DNA cleaved with HindIII (Nippon Gene) was used as a DNA size marker.

Agarose gel electrophoresis. Plasmids were analyzed by agarose gel electrophoresis of the cleared lysate of each strain by the method described by Meyers et al. (16, 18). The standard electrophoresis buffer consisted of final concentrations of 45 mM Tris, 2.5 mM EDTA, and 8.9 mM boric acid (0.5× TBE, pH 8.0). Three milliliters of TBE and 2 ml of dye solution (0.02% BPB, 0.1% SDS, 25% sucrose in 0.5 M Tris-HCl, pH 7.6) were added to 3 ml of cleared lysate. A 10-µl portion of this solution was then dispensed into the wells of a 0.8% agarose (Seakem GTG agarose; FMC BioProducts, Inc., Rockland, Maine) gel, and the gel was then electrophoresed for 90 min at 100 V in an AE-6100 submerged electrophoresis apparatus (Atto, Inc., Tokyo, Japan). The gel was removed and stained for 15 min with 1 mg of ethidium bromide per ml of TBE. The gel was then washed with pure water and visualized with a DT-20MP transilluminator (Atto).

Plasmid elimination. The method described by Rogolsky et al. (24) was used for the elimination of the large plasmid. To eliminate the large plasmid at high temperatures, volumes of 10⁷ CFU of bacteria per ml were inoculated into TY broth, incubated for 24 h in a water bath set at 44°C, plated onto HI agar, and then incubated at 37°C for 24 h. In the curing experiments, cell suspensions were exposed to some oscillation during serial dilutions to break up cell clusters which might have contained a mixture of P⁺ (plasmid-carrying bacteria) and P (plasmid-cured bacteria) types. The above procedure was repeated five times, and 25 colonies in final culture were randomly selected and tested for the existence of P bacteria.

Synthesis of DNA probe. The DNA probe for the detection of the SHET gene was chosen on the basis of the conservative nucleotide sequences among genes coding for ETA and ETB (4, 14, 23). This probe was a 21-mer coding for amino acids 209 to 215 from the signal sequences of the ETB molecule, and it had the following nucleotide sequence: 5'-TCTGGATCAGGTATATTTAAT-3'. Biotinylated DNA probe was synthesized by an automated DNA synthesizer (Model 391DNA synthesizer; Perkin-Elmer, Inc., Applied Biosystems Division, Foster City, Calif.) and was designated the ET probe.

Dot blot hybridization. A 50-µl portion of cleared lysate from each strain was mixed with 40 ml of 1 mM EDTA in 10 mM Tris-HCl (pH 8.0) and 10 ml of 2 N NaOH and denatured for 10 min. After denaturation, 100 ml of 2 M ammonium acetate was added to the above solution and used as a DNA sample. The nitrocellulose filter (BA 85, 0.45-µm pore size; Schleicher & Schuell, Inc., Dassel, Germany) was soaked in 2× SSC (300 mM NaCl and 30 mM sodium citrate, pH 7.0) for 1 h and fixed on a Bio-Dot apparatus (Bio-Rad Laboratories, Inc., Hercules, Calif.). A 50-µl DNA sample was dispensed into each well of the Bio-Dot apparatus, adsorbed on a nitrocellulose sheet by vacuum aspiration for 20 min, and then air dried. The nitrocellulose sheet was washed with 2× SSC solution, and DNA was cross-linked for 3 min by UV lights by using a UVC508 ultraviolet crosslinker (Ultrarüm, Inc., Carson, Calif.). Prehybridization was carried out for 30 min at 65°C in 6.25× SSC buffer containing 0.5% (wt/vol) bovine serum albumin, 0.5% (wt/vol) polyvinylpyrrolidone, and 1% (wt/vol) SDS. The filters were transferred into the hybridization solution (6× SSC buffer containing 1% bovine serum albumin, 1% polyvinylpyrrolidone, and 1 mM EDTA) containing 0.15 pmol of ET probe per ml. Hybridization was performed for 30 min at 65°C, followed by two washes (5 min each) at 65°C with 1× SSC buffer containing 1% SDS. The sheet was then washed three times with PBS, and the avidin-biotin complex (ABC) reagent (Vectastain ABC kit; Vector Laboratories, Inc., Burlingame, Calif.) was applied on a sheet and incubated at room temperature for 1 h. After incubation, the sheet was washed with PBS for 10 min. The substrate solution (0.5 mg of 3,3'-diaminobenzidine per ml and 0.01% H₂O₂ in 100 mM Tris-HCl [pH 7.2]) was then applied on a sheet and incubated at room temperature for 5 to 10 min. The sheet was then washed with tap water, which immediately stopped the color reaction.

Animals. Twenty-eight 1-day-old chickens (White Leghorn; Kanto Shokkei, Inc., Tokyo, Japan) were used for the 1-day-old chicken inoculation test. Eight Landrace piglets (3 weeks old) bred on a Kitasato University farm were used for experimental infection with S. hyicus.

One-day-old chicken inoculation test. UCF and CCF (0.4 ml each) were inoculated subcutaneously into 1-day-old chickens. After inoculation, these chickens were observed for 3 h for the Nikolsky sign (peeling off the skin surface easily caused by slight rubbing with the fingertip). The Nikolsky sign was graded as follows: , no reaction; +, localized exfoliation by CCF; ++, exfoliation of a wide area by CCF; +++, exfoliation of a wide area by UCF.

Experimental infection. A SHETA-producing strain (P-1), a SHETB-producing strain (P-23), and plasmid-cured substrains (P-23C1 and P-23 C2) of strain P-23 were used for this experiment. Bacteria grown at 37°C for 18 h on HI agar were suspended in PBS at a concentration of 10⁹ CFU/ml. One milliliter of each strain was inoculated into the external ear of two piglets. After inoculation, the piglets were observed for 1 week for skin lesions.

Antisera. Antisera against SHET produced by S. hyicus strains P-1 (SHETA) and P-23 (SHETB) were prepared by the method described by Tanabe et al. (29). The antibody titers of these two antisera were determined by enzyme-linked immunosorbent assay (29). The enzyme-linked immunosorbent assay titers of anti-SHETA and anti-SHETB sera were 1:25,600 and 1:12,800, respectively.

SDS-PAGE. SDS-PAGE was performed at room temperature by the method described by Laemmli (12). A mixture of 0.05 ml of 500 mM Tris-HCl buffer (pH 6.8), 0.08 ml of 10% SDS, 0.02 ml of 2-mercaptoethanol (Bio-Rad Laboratories), and 0.05 ml of 0.02% bromophenol blue in 80% glycerol was added to 0.2 ml of CCF, and the mixture was allowed to stand overnight at room temperature. This sample solution was layered on SDS-12.5% polyacrylamide gel slabs and run at 30 mA per gel. The proteins in the slabs were transferred to polyvinylidene difluoride membranes (Atto Inc.), stained with 0.25% Coomassie brilliant blue R-250 (CBB; Merck, KGaA Inc., Darmstadt, Germany), and decolorized with 7% acetic acid by the method described by Fairbanks et al. (5).

Western blotting. Western blotting was carried out by the method of Towbin et al. (31). The antigens (CCF from each strain) were prepared by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The first antisera were anti-SHETA or anti-SHETB sera at a 1:2,000 dilution in 10% skim milk (Difco). The second antibody was peroxidase-conjugated anti-mouse immunoglobulin G (lot F44989; Seikagaku Kogyo Inc., Tokyo, Japan) diluted 1:2,000 with 10% skim milk. The color reaction of the substrate was developed with 50 mM Tris-HCl buffer (pH 7.7) containing 0.05% (wt/vol) 3,3'-diaminobenzidine and 0.01% H₂O₂.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Plasmid profiles of S. hyicus strains. Large plasmids were found in the cleared lysates from several strains of S. hyicus. The mobility of these plasmids on agarose gel was almost the same as that of the 42-kb plasmid from the ETB-producing strain of S. aureus. Of the 92 S. hyicus strains, 22 SHET-producing strains (23.9%) possessed large plasmids. However, none of the non-SHET-producing strains possessed such plasmids (Table 1). The large plasmid of strain P-23 was digested by several restriction enzymes, and their restriction patterns were analyzed. EcoRI digests of the large plasmids resulted in seven fragments (11.9, 7.9, 7.6, 5.1, 4.8, 3.6, and 1.1 kb), and the total size of these fragments was 42 kb. BamHI digests resulted in a 25.4-kb fragment. HindIII digests resulted in seven fragments (12.4, 9.0, 7.3, 4.8, 4.2, 2.3, and 1.7 kb), and the total size of these fragments was 41.7 kb. From these results, the size of the large plasmid was estimated to be 42 kb and was designated pKUH-1.

Antigenic difference of SHETs produced by plasmidless and plasmid-carrying strains of S. hyicus. The Western blot analysis of SHETs from the plasmidless (P-1) and plasmid-carrying strains (P-10 and P-23) is shown in Fig. 1 and Table 2. The SHET from strain P-1 reacted with antiserum to SHET from strain P-1 alone. The SHET from strain P-23 also reacted with antiserum to SHET from strain P-23 alone. However, SHET from strain P-10 reacted with both antisera. These results suggest that strains P-1 and P-23 produce different serotypes of SHET and that strain P-10 produces both serotypes of SHET. We therefore designated the SHETs from strains P-1 and P-23 as SHETA and SHETB, respectively.

View larger version (90K): [in this window] [in a new window]   FIG. 1.   Western blot analysis of the CCF from three different strains of S. hyicus. Lane M, marker proteins (CBB staining); lane 1, purified SHETA (CBB staining); lanes 2 and 3, CCF from strain P-1; lanes 4 and 5, CCF from strain P-10; lanes 6 and 7, CCF from strain P-23; lanes 2, 4, and 6, SHETA antibody; lanes 3, 5, and 7, SHETB antibody.

View this table: [in this window] [in a new window]   TABLE 2.   Antigenicity of SHETs from P+ and P strainsa

Detection of SHET genes. Detection of the SHET genes of the plasmid-carrying (P-23, P-10) and plasmidless strains (P-1) of S. hyicus is shown in Fig. 2 and Table 3. The ET probe hybridized with chromosomal DNA of S. aureus ZM (plasmidless strain, ETA producer) and plasmid DNA of S. aureus J-sETB-8 (plasmid-carrying strain, ETB producer). These results suggest that the ET probe specifically hybridized with eta and etb genes. The ET probe also hybridized with chromosomal DNA of S. hyicus strain P-1 (plasmidless strain, SHETA producer) and plasmid DNA of S. hyicus P-23 (plasmid-carrying strain, SHETB producer). Moreover, the ET probe hybridized with both chromosomal and plasmid DNA of S. hyicus P-10 (plasmid-carrying strain, SHETA and SHETB producer). These results suggest that the genes coding for SHETA and SHETB are located in the chromosomal and plasmid DNAs, respectively.

View larger version (71K): [in this window] [in a new window]   FIG. 2.   Dot blots hybridized to ET probe. Lanes: 1, S. aureus ZM (ETA producer); 2, S. aureus J-sETB-8 (ETB producer); 3, S. hyicus P-1 (SHETA producer); 4, S. hyicus P-23 (SHETB producer); 5, S. hyicus P-10 (SHETA and SHETB producer).

Toxic activities of parental strain and plasmid-cured substrains. The toxic activities of the parental strains and the plasmid-cured substrains are shown in Table 4. The extended exfoliation was observed in the chickens inoculated with UCF from parental strains (P-23 and P-10). The extended exfoliation was observed in the chickens inoculated with CCF from plasmid-cured substrains (P-10C1 and P-10C2) of P-10, while no exfoliation was observed for the chickens inoculated with CCF from plasmid-cured substrains (P23C1 and P-23C2) of P-23. These results suggest that the plasmid-cured substrains of strain P-10 possessed the SHETA gene in their chromosomal DNA, whereas the plasmid-cured substrains of P-23 did not possess the SHETA and SHETB genes.

Experimental infection of the plasmid-cured and parental strains. The macroscopic lesions of pigs inoculated with P⁺ and P strains are shown in Table 5. Pigs inoculated with the SHETA-producing (P-1) and SHETB-producing (P-23) strains exhibited the typical clinical signs of EE. However, the non-SHET-producing strains (N-10 and N-13) and P substrains of P-23 did not exhibit any clinical signs. These results suggest that the SHETB-producing strains as well as the SHETA-producing strain cause typical clinical signs of EE and that plasmid-cured substrains of the SHETB-producing strains lose their pathogenicity.

View this table: [in this window] [in a new window]   TABLE 5.   Macroscopic lesions of pigs inoculated with P+ and P strains

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

More than 70% of S. hyicus strains from healthy pigs and pigs with EE produce SHET. Of S. aureus strains, only 5% have been shown to produce ET (22). These findings suggest that the SHET-producing rate of S. hyicus is higher than that of S. aureus and that there are no obvious differences in the SHET-producing ability among the strains of S. hyicus from different sources.

The ETs produced by S. aureus were divided into two serotypes, ETA and ETB (9, 10, 11). ETA is a heat-stable toxin, while ETB is heat labile (9, 10). The production of ETA and ETB is genetically controlled by chromosomal DNA and a 42-kb plasmid, respectively (19, 20, 32). The genes coding for ETA and ETB have already been determined (6, 14, 21, 25), but the genes coding for SHET have not been determined. In this study, we located the large plasmids in several SHET-producing strains of S. hyicus, but these plasmids could not be detected in all the non-SHET-producing strains. The size of the large plasmid was determined to be 42 kb based on summing the sizes of its restriction fragments.

Our recent study (30; Sato et al., 14th IPVS Cong.) indicates that SHET has at least two serotypes. Andresen et al. (2, 3) have also reported that SHET can be divided into more than three serotypes; however, the location of the SHET gene based on these studies was still not clear. In the present study, CCF from all P strains reacted with antibody to SHET of the plasmidless strain (P-1) alone and CCF from all P⁺ strains reacted with antibody to SHET of the plasmid-carrying strain (P-23). Based on the above, we designated the SHETs of the plasmidless and plasmid-carrying strains as SHETA and SHETB, respectively. Strain P-23 produce SHETB, while plasmid-cured substrains of P-23 could not produce SHETB. In other plasmid-carrying strains, such as P-10, the level of toxic activity was decreased by curing the 42-kb plasmid, but toxic activity could still be detected. The CCF of strain P-10 reacted with both anti-SHETA and anti-SHETB antibodies in the Western blot analysis. These findings suggest that strain P-1 produces SHETA alone, strain P-23 produces SHETB alone, and strain P-10 produces both SHETs. Recently, we succeeded in cloning the gene coding for SHETB in Escherichia coli (data not shown). From these findings, it appears that the chromosomal DNA and the 42-kb plasmid DNA control the production of SHETA and SHETB, respectively.

The synthesized DNA probe based on the C-terminal active region of ET (4, 14, 23) hybridized with the chromosomal DNA of the ETA-producing strain (ZM) and the plasmid DNA of the ETB-producing strain (J-sETB-8). We then carried out dot blot hybridization using the above DNA probe (ET probe), since it was confirmed that this probe specifically hybridizes with genes coding for ETA and ETB. The ET probe also hybridized with the chromosomal DNA of the SHETA-producing strain (P-1) and plasmid DNA of the SHETB-producing strain (P-23). These findings suggest that the nucleotide sequences of the active region among ETs and SHETs are conserved and that the genes coding for SHETA and SHETB are located on the chromosomal and plasmid DNA, respectively.

The typical clinical signs (exudation, exfoliation, and crusting) of EE were observed in the piglets inoculated with the plasmid-carrying (SHETB-producing) strain as well as those inoculated with plasmidless (SHETA-producing) strains. However, plasmid-cured substrains of P-23 did not cause any clinical signs. The above results suggest that SHETB-producing strains as well as the SHETA-producing strain of S. hyicus cause EE in piglets.