ABSTRACT
The cytotoxicities of various strains of Mycoplasma mycoides subsp. mycoides small colony type (SC), the agent of contagious bovine pleuropneumonia (CBPP), were measured in vitro using embryonic calf nasal epithelial (ECaNEp) cells. Strains isolated from acute cases of CBPP induced high cytotoxicity in the presence of glycerol, concomitant with the release of large amounts of toxic H2O2 that were found to be translocated into the cytoplasms of the host cells by close contact of the Mycoplasma strains with the host cells. Currently used vaccine strains also showed high cytotoxicity and high H2O2 release, indicating that they are attenuated in another virulence attribute. Strains isolated from recent European outbreaks of CBPP with mild clinical signs, which are characterized by a defect in the glycerol uptake system, released small amounts of H2O2 and showed low cytotoxicity to ECaNEp cells. M. mycoides subsp. mycoides SC strain PG1 released large amounts of H2O2 but was only slightly cytotoxic. PG1 was found to have a reduced capacity to bind to ECaNEp cells and was unable to translocate H2O2 into the bovine cells, in contrast to virulent strains that release large amounts of H2O2. Thus, an efficient translocation of H2O2 into host cells is a prerequisite for the cytotoxic effect and requires an intact adhesion mechanism to ensure a close contact between mycoplasmas and host cells.
Contagious bovine pleuropneumonia (CBPP) is a severe respiratory disease of cattle and buffaloes caused by Mycoplasma mycoides subsp. mycoides small colony type (SC). It represents a major threat to raising cattle, particularly in Africa, where it creates great economic losses in regions of endemicity (6, 15, 21). CBPP was eradicated in the middle of the 20th century in the industrialized continents (5), but it reemerged in a few European countries in 1980 to 1999 in a milder, more insidious form that appeared to be largely chronic and showed a low mortality rate (11). A major genetic difference between European and African strains involves a segment of 8.84 kb which includes the genes gtsABC, which encode an ATP-binding cassette transporting system for the active uptake of glycerol (17, 18). Virulent field strains of M. mycoides subsp. mycoides SC isolated in Africa from animals with CBPP possess the glycerol transporter genes gtsABC for the assimilation of glycerol, which, after phosphorylation, is metabolized further to dihydroxyacetone phosphate with the release of hydrogen peroxide (H2O2). Strains from recent European outbreaks that showed very low mortality rates lack part of this glycerol uptake operon (18). Consequently, European strains were found to produce significantly smaller amounts of H2O2 in the presence of physiological concentrations of glycerol than the virulent African strains.
The l-α-glycerophosphate oxidase (GlpO) of M. mycoides subsp. mycoides SC was identified as a membrane protein that plays a central role in the cytotoxicity of M. mycoides subsp. mycoides SC to embryonic calf nasal epithelial (ECaNEp) cells (14). GlpO catalyzes the oxidation of glycerol-3-phosphate, followed by the release of H2O2 (18). Blocking GlpO by monospecific anti-GlpO antibodies blocked the release of H2O2 and concomitantly inhibited the cytotoxicity of M. mycoides subsp. mycoides SC to ECaNEp cells (14). The H2O2-mediated cytotoxic effect on host cells was found to be triggered by the close contact between mycoplasmas and ECaNEp cells allowing H2O2 to be translocated to the cytosol of the host cells, while exogenous H2O2 in axenic medium was not sufficient to cause cytotoxicity (14). This was evidenced using 5 (and 6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate-acetyl ester (CM-H2DCFDA) to monitor the oxidative stress in ECaNEp cells upon addition of glycerol. A strong accumulation of H2O2 was detected only in the cytoplasms of those cells that were infected with M. mycoides subsp. mycoides SC, suggesting the occurrence of a direct translocation of H2O2 from the mycoplasma to the host cell cytoplasm, presumably through an intimate contact between GlpO of adhering M. mycoides subsp. mycoides SC and host cells (14).
The cytotoxicity of strain L2, a strain representative of the European 1980-to-1999 outbreaks, was found to be significantly lower than that of the virulent African strain Afadé (14) and reflected the lack of an active glycerol uptake system and the low levels of H2O2 production of strain L2. Furthermore, these in vitro virulence assays were in agreement with in vivo assays of a comparative experimental infection of cattle, where strain L2 was significantly less virulent and induced a very low and late seroconversion compared to strain Afadé (1, 12). This showed that glycerol metabolism-dependent H2O2 production represents one of the major virulence attributes of M. mycoides subsp. mycoides SC (13, 14).
In the present study, we determined the cytotoxicity to ECaNEp cells of a broad collection of strains of M. mycoides subsp. mycoides SC of various origins, including vaccine strains. This set of strains was analyzed for the presence or absence of the glycerol transporter genes gtsABC; the gene glpO, encoding the l-α-glycerophosphate oxidase; and the production of H2O2 in the presence of glycerol.
MATERIALS AND METHODS
Strains, growth conditions, and DNA extraction.The strains of M. mycoides subsp. mycoides SC used in this study and their origins are listed in Table 1. Cultures of M. mycoides subsp. mycoides SC were grown in a standard mycoplasma medium (Axcell Biotechnologies) at 37°C to a density of approximately 5 × 108 cells per ml.
Mycoplasma strains used in this study
For DNA extraction, the cells were harvested by centrifugation at 8,000 × g for 10 min, washed three times in TES buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, 140 mM NaCl), and then resuspended in 0.1 volumes of TE buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA). A sample of 100 μl of resuspended cells was then lysed by addition of 500 μl of GES buffer (5 M guanidium thiocyanate, 100 mM EDTA, 0.5% N-lauroylsarcosine) for 10 min at room temperature, cooled on ice, and then mixed with 250 μl of 7.5 M ammonium acetate, pH 7.7. The lysate was then extracted three times with 500 μl of phenol-chloroform-isoamylalcohol (49.5:49.5:1) (Fluka), followed by one extraction with 500 μl of chloroform-isoamylalcohol (49:1) (Fluka), and the DNA was precipitated by the addition of 0.7 volumes of isopropanol and centrifugation at 17,500 × g for 15 min at 4°C. The pellet was washed three times with 80% ethanol, dried, and resuspended in 100 μl of H2O.
PCRs.PCR was performed using a Gene Amp 9600 DNA thermal cycler (Applied Biosystems) with a 50-μl reaction mixture [50 mM Tris-HCl, pH 9.2, 1.75 mM MgCl2, 16 mM (NH4)2SO4, and 350 μM of each deoxynucleoside triphosphate] containing 300 nM of each primer (Table 2), 1.75 U of Taq DNA polymerase (Roche Diagnostics), and approximately 50 ng of genomic DNA. The mixtures were subjected to 2 min of denaturation at 94°C, followed by 35 cycles of amplification with the following parameters: 30 s at 94°C, 30 s at 48°C, and 1.5 min of extension at 72°C for fragments of 1.2 kb or 2.5 min of extension at 72°C for fragments of 2.5 kb.
Oligonucleotide primers
For amplification of long DNA fragments (>3 kb), PCR was carried out using an Expand long-template PCR system kit (Roche Diagnostics) with a 50-μl reaction mixture including buffer 3, 500 μM of each deoxynucleoside triphosphate, 300 nM of each primer (Table 2), 3.75 U of a mixture of Taq and Pwo polymerases, and approximately 50 ng of genomic DNA as the template. The DNAs were amplified for 35 cycles (30 s of denaturation at 92°C, 30 s at 48°C, and elongation at 68°C for 10 min). An initial step of 2 min at 92°C was added to ensure denaturation.
The PCR amplification products were analyzed by electrophoresis through 0.7% agarose gels and visualized after staining with ethidium bromide on a UV transilluminator.
Quantification of H2O2.To measure H2O2 production, strains of M. mycoides subsp. mycoides SC were grown in mycoplasma culture medium for 3 days to a density of approximately 5 × 108 cells/ml. The culture was centrifuged at 8,000 × g for 10 min at 4°C, washed once in incubation buffer (67.7 mM HEPES, pH 7.3, 140 mM NaCl, 7 mM MgCl2), resuspended in prewarmed incubation buffer at 37°C at a density of 109 cells/ml, portioned in aliquots of 1 ml, and incubated at 37°C for 1 h. To induce H2O2 production, glycerol was added to the mycoplasma suspensions at a final concentration of 100 μM, representing the physiological concentration of glycerol in bovine serum. The production of H2O2 was measured with a peroxide test (Merck KgaA) as described previously (18) at time zero and 1, 2, 5, 10, 20, and 40 min after the addition of glycerol. The assay was repeated three times for each strain, and mean values were calculated.
Time-dependent glycerol uptake assay.Measurements of time-dependent glycerol incorporation into cellular material of M. mycoides subsp. mycoides SC strains Afadé and KH3J were carried out with radioactive substrate, following a modified version of a protocol previously described (18). Briefly, 14-ml mycoplasmal cultures of both Afadé and KH3J, grown in mycoplasma culture medium to a density of approximately 5 × 108 cells/ml, were centrifuged at 12,000 × g for 10 min, and the cell pellets were washed with isotonic HEPES buffer (67.7 mM HEPES, pH 7.3, 140 mM NaCl) and then resuspended in 6 ml incubation buffer supplemented with 100 μM glycerol. Aliquots of 1 ml of cell suspensions were then adjusted to approximately 109 cells/ml. After incubation for 1 h at 37°C, the incubation buffer was adjusted to 280 nM [U-14C]glycerol (from a stock of 149 mCi/mmol; GE Healthcare). Aliquots of 200 μl were vacuum filtered at time intervals (30 s, 1 min 45 s, 3 min, 4 min 15 s, and 5 min 30 s), and filters (pore size, 0.22 μm; Millipore) were washed immediately with isotonic HEPES buffer and finally counted with a scintillation counter. It has to be noted that a minor amount of incorporated glycerol is not measured in this assay, since it is catabolized by the bacteria.
Assessment of cytotoxic activity.ECaNEp cells were prepared as already described (14). ECaNEp cells taken after seven passages in minimal essential medium (MEM)-Earle medium were grown to a nonconfluent state in 24-well plates to reach approximately 105 cells per well. Prior to the assay, the medium was removed and replaced by 200 μl of MEM-Earle medium without supplements or by MEM-Earle medium supplemented with 100 μM glycerol. The ECaNEp cells were then infected at a multiplicity of infection (MOI) of 500 mycoplasmas per cell and incubated for 3 h at 37°C in a humidified 5% CO2 atmosphere. Morphologically intact ECaNEp cells were counted after fixation and staining with 0.75% crystal violet, 0.25% NaCl, 1.75% formaldehyde, and 50% ethanol and photographed by phase-contrast microscopy as described previously (14).
Detection of oxidative stress caused by H2O2 in ECaNEp cells.To monitor intracellular H2O2 and reactive oxygen species in ECaNEp cells, the formation of dichlorofluorescein derivatives after cleavage of the ester groups of nonfluorescent CM-H2DCFDA (Molecular Probes) by H2O2-dependent oxidation was measured as already described (14). The same conditions used for the assessment of cytotoxic activity were employed, with the following modifications. The ECaNEp cells were incubated for 1 h with 10 μM CM-H2DCFDA and washed once with MEM-Earle medium without supplements. Cells were then infected at an MOI of 500 mycoplasmas per cell in the presence or absence of glycerol. Intracellular H2O2 was monitored 20 min after infection with mycoplasmas by fluorescence microscopy using a Nikon Eclipse TE 300 microscope. Note that all steps involving CM-H2DCFDA, including handling of this chemical, were performed in the dark.
Adherence assay.For adherence assays, M. mycoides subsp. mycoides SC was grown in 10 ml of mycoplasma culture medium for 3 days at 37°C to a density of approximately 5 × 108 cells/ml. The mycoplasmas were harvested by centrifugation at 8,000 × g for 10 min at 4°C, washed twice in buffer A (50 mM Tris-HCl, pH 7.2, 100 mM NaCl, 1 mM CaCl2), and resuspended in 10 ml of buffer A.
To determine the optimal concentration of mycoplasmas for the adhesion assay, dilutions (1:1 to 1:512) of the 5 × 108-cell/ml culture were prepared in buffer A. After this, 200 μl of each dilution was transferred onto each ECaNEp monolayer (containing approximately 105 cells) and incubated for 2 h at 37°C. After removal of excess liquid, the ECaNEp cells were washed twice with 500 μl of buffer A to remove nonadherent mycoplasmas and overlaid with 500 μl of buffer A. ECaNEp cells bearing adherent mycoplasmas were then scratched out of the wells and analyzed by real-time PCR.
Quantification of adherent mycoplasmas.To quantify M. mycoides subsp. mycoides SC in adhesion assays, glpO-based TaqMan real-time PCR was used as described previously (2). To quantify ECaNEp cells, the TaqMan assay reagent for the 18S rRNA kit was used in TaqMan PCRs according to the manufacturer's protocol (Applied Biosystems). Reactions were performed by using 2.5 μl from each adherence assay, a 900 nM concentration of each TaqMan primer, 300 nM TaqMan probe, and TaqMan No AmpErase UNG universal PCR master mix (Applied Biosystems) in a 25-μl volume. PCRs were run on an ABI 7500 instrument (Applied Biosystems), using the following cycling parameters: after one initial step at 50°C for 2 min and at 95°C for 10 min, 40 cycles of denaturation at 95°C for 15 s and extension at 60°C for 1 min were performed. Real-time fluorescence measurements were taken for each sample by using Sequence Detector 1.3.1 software (Applied Biosystems), and the PCR cycle number at which the fluorescent signal crossed the cycle threshold limit (set at 0.1 for glpO and 0.06 for 18S rRNA) for each sample (2.5 μl) was recorded as the cycle threshold value. Each sample was assayed in duplicate.
TaqMan standard curves were produced by analyzing serial dilutions with known numbers of M. mycoides subsp. mycoides SC cells (glpO based) and of ECaNEp cells (18S rRNA based). The efficiencies of both TaqMan reactions were good, as illustrated by R2 values very close to 1.0 and by 10−1/slope values very close to 2.0. The estimated quantities of mycoplasmal and ECaNEp cells in 2.5 μl from the adherence assay and the relative adherence efficiencies were derived from the linear regressions of the standard curves.
Nucleotide sequence accession numbers.The KH3J DNA sequences of the gtsABC operon and the gene glpO, including its promoter region, have been deposited in the EMBL/GenBank database under accession numbers AM422015 and AM422016, respectively.
RESULTS
Presence of the glpO and the gtsABC transporter genes.Type strain PG1; 1 old European, 6 recent European, 7 African, and 3 Australian field strains; and 3 vaccine strains of M. mycoides subsp. mycoides SC were investigated for the presence of glpO and the glycerol transporter genes gtsABC (Table 1). The presence of the glpO gene was evidenced by PCR amplification using oligonucleotide primers Afadé_EcoRI and Afadé_BamHI (Table 2). This PCR amplified a fragment of about 1.2 kb from all strains analyzed, which indicates that the gene glpO, encoding l-α-glycerophosphate oxidase, is ubiquitous in M. mycoides subsp. mycoides SC (Table 1).
PCR amplification of a fragment containing the gtsABC operon, lppB, open reading frame 5 (ORF5), and ORF6 by use of primers 3480bp-L and 3480bp-R (Table 2) resulted, as expected, in a fragment of 11.3 kb that showed the presence of the entire gtsABC operon for type strain PG1 and all strains of the African and Australian continents, including the vaccine strains, except for strain 91130, which gave a 9.0-kb fragment (Table 1). PCR analysis with the same primers by use of DNA from European strains showed two patterns. Strains that were isolated from the recent CBPP outbreak in Europe at the end of the 20th century resulted in the amplification of a 2.5-kb fragment that indicated their clonality and the absence of the 8.84-kb portion of this genome segment that contains the gtsABC operon. The old European strain PO 1967, which was isolated in France in 1967 and is epidemiologically unrelated to the CBPP outbreaks at the end of the 20th century (9), showed an 11.3 kb-amplicon that indicates the presence of the gtsABC operon.
In order to verify the integrity of the gtsABC operon, PCR analysis was made using the primer pair ORF0(BG7)-1N and DIG-5L (Table 2), matching the ends of the gtsABC operon. This analysis confirmed the presence of the entire gtsABC operon in all strains except in those from the new European outbreaks (Table 1).
Production of H2O2 in the presence of glycerol.H2O2 concentrations in culture supernatants of M. mycoides subsp. mycoides SC reached saturation 20 min after the addition of 100 μM glycerol. All strains isolated from the recent European CBPP outbreaks showed low levels of H2O2 production, in the range of 0.2 to 0.8 mg/liter. The old European strain PO 1967, in contrast, revealed high levels of H2O2 production (5.0 mg/liter), which reflects the presence of a functional gtsABC operon and a functional glpO gene (Table 1). Type strain PG1, the African field strains, and the African vaccine strain T1/44 showed high values for H2O2 production, ranging from 4.0 to 5.0 mg/liter, which also reflects the presence of the entire gtsABC operon and the glpO gene. The formerly used vaccine strain KH3J, in contrast, showed a reduced level of H2O2 production (2.0 mg/liter), despite the presence of the entire gtsABC operon and the glpO gene. The Australian field strains and vaccine strain V5 all showed high values for H2O2 production, although the value for field strain CF1 was slightly below the average of 4.6 mg/liter for this group of strains. Control experiments without addition of glycerol to the growth medium revealed no H2O2 production in any of the strains. The addition of catalase to the culture of M. mycoides subsp. mycoides SC after supplementation with glycerol reduced the H2O2 level in the medium to a concentration below 1 μM, as found previously (14).
In order to confirm the integrity of the gtsABC operon and the gene glpO, including its promoter region, in strain KH3J, the genomic segments containing these genes have been amplified by PCR using primer pairs ORF0(BG7)-1N/DIG-5L and MmmSC_glpO_Prom/Afadé_BamHI (Table 2), respectively, and their DNA sequences have been determined using the same oligonucleotide primers used for PCR amplification and additional internal primers. No differences in these genes, except two silent mutations, were found in comparison to those from the virulent strain Afadé or type strain PG1.
Analysis of glycerol uptake.Due to the reduced H2O2 production in the presence of 100 μM glycerol of vaccine strain KH3J in comparison to what was found for the other African strains (including vaccine strain T1/44), KH3J was subjected to time-dependent measurements of glycerol uptake, and the results were compared to those for strain Afadé, chosen as representative for African field strains. Uptake kinetics showed that KH3J has a reduced glycerol uptake capacity, approximately 75% of that of strain Afadé (Fig. 1).
Glycerol uptake. Numbers of disintegrations per minute (dpm) of 14C incorporated by mycoplasmas are shown as a measure for glycerol incorporation. Triangles, strain Afadé; squares, strain KH3J. Error bars indicate standard deviations.
Cytotoxicity of M. mycoides subsp. mycoides SC to bovine epithelial cells.When ECaNEp cells were inoculated with M. mycoides subsp. mycoides SC in MEM-Earle medium without glycerol, no cell death was detectable up to 3 h postinoculation, regardless of which strain was used, showing that M. mycoides subsp. mycoides SC displayed no particular cytotoxicity to ECaNEp cells under these conditions. In contrast, in the presence of 100 μM glycerol, which represents the physiological concentration in bovine serum, all African and Australian field strains caused cell mortalities of 95 to 98% in ECaNEp cells 3 h after infection (Table 1). The same cytotoxic effect in the presence of glycerol was observed with the vaccine strains T1/44 and V5. Vaccine strain KH3J, which was differentiated from the other vaccine strains tested by a relatively low level of H2O2 production, showed a very low cytotoxicity level, with only 5% cell death. European strains isolated from the outbreaks at the end of the 20th century showed low cytotoxicities, with 0 to 10% cell mortality, as expected from their low levels of H2O2 production. In contrast, the old European strain PO 1967 showed a high level of cytotoxicity, similar to that of the African field strains, with 94% cell death, correlating very well with the large amounts of H2O2 released when these strains are grown in the presence of 100 μM glycerol (Table 1).
Unexpectedly, type strain PG1 revealed a relatively low level of cytotoxicity, with 13% cell death at an MOI of 500 mycoplasmas per ECaNEp cell, in spite of the presence of the full genes for the glycerol transporter and the l-α-glycerophosphate oxidase and its capacity to release high levels of H2O2. Increment of the MOI for PG1 to 1,000 and 2,000 mycoplasmas per cell raised the cytotoxicity rates to 67% and 93%, respectively (Table 3).
Correlation between adherence and cytotoxicity
Detection of oxidative stress in host cells.We have monitored the oxidative stress in ECaNEp cells caused by intracellular H2O2 after infection with M. mycoides subsp. mycoides SC in the presence or absence of glycerol, using ECaNEp cells pretreated with CM-H2DCFDA to detect intracellular oxidation of this compound by fluorescence microscopy (Fig. 2). As shown in Fig. 2B, infection with the African M. mycoides subsp. mycoides SC field strain Afadé produced a strong induction of fluorescence in ECaNEp cells 20 min after the addition of glycerol, reflecting the presence of intracellular H2O2. In contrast, no fluorescence was detected in ECaNEp cells infected with strain Afadé without the addition of glycerol (Fig. 2A). Addition of glycerol to ECaNEp cells in the absence of mycoplasmas did not result in fluorescence of CM-H2DCFDA-treated ECaNEp cells, indicating that H2O2 inside the ECaNEp cells that were infected with M. mycoides subsp. mycoides SC was not produced endogenously.
Fluorescence micrographs showing detection of intracellular H2O2 in ECaNEp cells infected with M. mycoides subsp. mycoides SC at an MOI of 500 mycoplasmas per cell. Results are shown for ECaNEp cells 20 min after infection with strain Afadé in the absence (A) and in the presence (B) of glycerol, with strain PG1 in the absence (C) and in the presence (D) of glycerol, and with strain L2 in the absence (E) and in the presence (F) of glycerol.
Intracellular oxidation of CM-H2DCFDA did not occur when ECaNEp cells were infected at an MOI of 500 mycoplasmas per cell with type strain PG1 (Fig. 2C and D) or with the European strain L2 (Fig. 2E and F), either in the presence or in the absence of glycerol, showing a good correlation between cytotoxicity and presence of intracellular H2O2 in infected ECaNEp cells. When the MOI for PG1 was increased to 1,000 or 2,000 mycoplasmas per cell, oxidation of CM-H2DCFDA in infected ECaNEp cells was detected, in parallel with the observed increment in cytotoxicity (Table 3).
Adhesion to host cells.Considering the significantly low cytotoxicity of type strain PG1, despite its high level of H2O2 production upon incubation with glycerol and its low capacity to translocate H2O2 into the host's cytoplasm, we compared its process of adhesion to ECaNEp cells with that for strain Afadé. At different ratios of MOI, the relative adherence of type strain PG1 to ECaNEp was 30 to 65% lower than that of strain Afadé (Fig. 3). Determination of the multiplicity of attachment (MOA) indicated, however, that about 300 mycoplasmal cells adhering to a single ECaNEp cell were necessary for both strains Afadé (requiring an MOI of 500) and PG1 (requiring a higher MOI of 2,000) to achieve a cytotoxicity rate of approximately 90% (Table 3).
Adherence assay. Shown is a graph representing the relative levels for adherence of strain Afadé and type strain PG1 to ECaNEp cells in culture at different MOIs after 2 h of incubation at 37°C. Triangles, strain Afadé; squares, strain PG1. Lines are trend lines. Error bars represent standard deviations for three measurements.
DISCUSSION
In this study, we investigated the cytotoxicities of several strains of M. mycoides subsp. mycoides SC to bovine epithelial cells based on the previously reported ability of the l-α-glycerophosphate oxidase GlpO to produce cytotoxic H2O2 in the presence of glycerol (14). Glycerol at physiological concentrations is incorporated and phosphorylated by M. mycoides subsp. mycoides SC via the highly active ATP-binding cassette glycerol transporter composed of GtsA, GtsB, and GtsC and is finally converted to dihydroxyacetone phosphate and H2O2 by GlpO (13, 18). The gene glpO was found to be present and functional in all M. mycoides subsp. mycoides SC strains investigated, while the gtsABC operon was entire in all strains except those from the recent European outbreaks. Hydrogen peroxide was shown to be translocated from M. mycoides subsp. mycoides SC into host cells presumably through the close contact of M. mycoides subsp. mycoides SC with its host (14). It is worth noting at this point that H2O2 in the medium had no visible cytotoxic effect on host cells in the absence of mycoplasmas, and cytotoxicity was induced only at very high doses of exogenous H2O2 in axenic media, i.e., at concentrations at least 30 times higher than that measured in culture media of M. mycoides subsp. mycoides SC grown in the presence of glycerol (14). Inside the host cells, H2O2 acts as a powerful mediator of cell injury, which then leads to inflammatory processes.
Field strains from the African and Australian continents as well as a European field strain isolated in 1967 showed high cytotoxicity to ECaNEp cells in the presence of glycerol due to their high levels of H2O2 production (Table 1). The capacity to translocate H2O2 into the cells was demonstrated in detail for strain Afadé (Fig. 2). These results are consistent with the presence of the gtsABC and glpO genes in these strains. They also confirm that the cytotoxic effects of these strains, which are caused by highly efficient glycerol uptake and GlpO-entailed metabolism, as already reported (14), are a general virulence attribute of highly pathogenic field strains of M. mycoides subsp. mycoides SC. However, the vaccine strains T1/44 and V5 also showed production of large amounts of H2O2 and high cytotoxicity to ECaNEp cells (Table 1). The high cytotoxic capacities of these vaccine strains could be the origin of inflammatory reactions, known as Willems reactions, observed at the site of injection in some vaccinated animals (8, 16, 19). In spite of their full cytotoxicity, vaccine strains T1/44 and V5 are known to be attenuated (5). Hence, their attenuation must affect virulence genes other than those involved in oxidative glycerol metabolism. It has been reported previously that less virulent strains of M. mycoides subsp. mycoides SC produce significantly less capsular polysaccharide, thus rendering them less persistent and invasive during infection (3, 10). Therefore, it has been suggested that the attenuation mutation in several vaccine strains might be located in the genes encoding the biosynthetic enzymes of galactan or in the genes responsible for the regulation of such enzymes (for a review, see reference 13).
Significantly lower cytotoxicities were found in all strains of the recent European outbreaks at the end of the last century. The low cytotoxic values for these European strains reflect the low levels of H2O2 production as a consequence of the lack of glycerol uptake genes. Some cytotoxicity in these strains, however, can still be demonstrated upon extended incubation in the presence of glycerol. The low levels of cytotoxicity of these strains may be due to a residual production of H2O2 originating from the metabolism of glycerol that penetrates the mycoplasmas by passive diffusion or via the linear bypass uptake system conferred by the glycerol facilitator gene glpF (13, 14, 18). The low cytotoxicities of the strains isolated from the European outbreaks in 1980 to 1999 correlate with the lower virulence of this group of strains, as determined by experimental infections (1, 12), and reflect the more chronic aspects of CBPP with low mortality rates during these outbreaks (11).
Two of the strains analyzed, strain KH3J and type strain PG1, exhibited low cytotoxic activities that were unexpected due to the presence of intact gtsABC and glpO genes (Table 1). Strain KH3J revealed a reduced level of H2O2 production, which in part reflects its low cytotoxic activity. However, DNA sequence analysis of the gtsABC and glpO genes showed no differences compared to those of the virulent field strain Afadé, with the exception of two silent mutations. Hence, the finding that glycerol uptake and metabolism in KH3J are hampered must be due to defects in yet unknown functions that are required to ensure the functionality of GtsABC and GlpO. It has to be noted that strain KH3J is known to be the least virulent of the vaccine strains (5). Thus, the findings that KH3J not only produces much less capsular polysaccharide galactan (7) but also shows a reduced cytotoxicity may explain its very low virulence.
Most surprisingly, type strain PG1 showed a low cytotoxicity, in spite of the fact that it was able to produce the same large amounts of H2O2 from glycerol metabolism as the highly virulent field strains (Table 1). Furthermore, there were no differences in the glycerol uptake genes gtsABC or in the l-α-glycerophosphate oxidase gene glpO between PG1 and the virulent field strain Afadé. However, PG1 was unable to translocate H2O2 into ECaNEp cells upon infection of the latter, in contrast to strain Afadé, which was shown to efficiently translocate H2O2 into ECaNEp cells (Fig. 2). Since M. mycoides subsp. mycoides SC seems to secrete toxic substances that act only over short distances, such as H2O2, adhesion is particularly important for an intimate interaction with host cells, which is considered an early step of pathogenesis for most microorganisms (13). Therefore, we tested the adhesion properties of type strain PG1 compared to those of the virulent strain Afadé. Under in vitro conditions, type strain PG1 showed a significantly lower level of adherence to the eukaryotic cells than Afadé. This reduced difference in adherence might explain the difficulty of PG1 to translocate H2O2 into host cells and hence may account for the attenuated cytotoxic activity of PG1. This finding supports the general notion that type strain PG1 is virtually nonpathogenic (4). While strains that possess lower levels of adherence showed reduced cytotoxicity, some vaccine strains, also known to have reduced virulence, have been identified to contain a functional glycerol metabolic pathway, to adhere to host cells like the virulent strains, and to show high cytotoxicity. Hence, these vaccine strains are supposed to possess attenuations of their virulence in genes of other pathogenicity mechanisms, such as those responsible for persistence or invasive capacity (3, 10).
Determination of MOA indicated that a cytotoxicity rate in the reaction well of approximately 90% was achieved when about 300 cells of M. mycoides subsp. mycoides SC were adhering to a single ECaNEp cell. For strain PG1, it was necessary to increase the MOI to 2,000 mycoplasmas per ECaNEp cell to obtain 340 adhering mycoplasmas per single cell and a cytotoxicity rate of 93%, while similar results were obtained with virulent strains at an MOI of 500 mycoplasmas per cell. Thus, the current study showed that the cytotoxic effect of M. mycoides subsp. mycoides SC is clearly dependent on an intact mechanism for adhesion of the pathogen to its host cells and that adhesion is a common virulence attribute of virulent strains of M. mycoides subsp. mycoides SC. The cytotoxic effect is triggered by the particular metabolism of glycerol and by the capacity to translocate the toxic metabolite H2O2 into host cells, the latter process being strictly dependent on adhesion between pathogen and host cell.
ACKNOWLEDGMENTS
We are grateful to Yvonne Schlatter and Antoinette Golomingi for technical assistance.
This work was supported by grant no. 075804, designated “a genomics approach to understanding the immunopathology of contagious bovine pleuropneumonia (CBPP): improvement of current live vaccines and the development of next generation vaccines,” of the Wellcome Trust, London, United Kingdom. Carole Janis received an exchange scholarship from the Germaine de Staël grant designated “Programmes actions intégrées PAI.”
FOOTNOTES
- Received 10 July 2007.
- Returned for modification 7 September 2007.
- Accepted 29 October 2007.
- Copyright © 2008 American Society for Microbiology
