Open Access
Issue
Vet. Res.
Volume 40, Number 4, July-August 2009
Number of page(s) 12
DOI http://dx.doi.org/10.1051/vetres/2009018
Published online 15 April 2009
How to cite this article Vet. Res. (2009) 40:35

© INRA, EDP Sciences, 2009

This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits unrestricted use, distribution, and reproduction in any noncommercial medium, provided the original work is properly cited.

1. INTRODUCTION

Glycerol metabolism is described to play a central role in virulence of Mycoplasma mycoides subsp. mycoides SC, the etiological agent of contagious bovine pleuropneumonia (CBPP) [4, 13, 14, 20], as well as in Mycoplasma pneumoniae [9]. Glycerol at physiological concentrations is incorporated and phosphorylated to obtain α-glycerol-3-phosphate (G3P) by M. mycoides subsp. mycoides SC via the highly active ATP-dependent glycerol transporter system GtsABC and then further metabolised to dihydroxyacetone phosphate (DHAP) in the presence of oxygen by the L-α-glycerophosphate oxidase (GlpO, also named α-glycerol-3-phosphate oxidase) with the release of the highly toxic compound hydrogen peroxide H2O2 [20]. Scattered electron microscopy and Triton X-114 partitioning revealed GlpO to be membrane-located in M. mycoides subsp. mycoides SC with a significant part of the protein being surface-exposed [13], which explains the fact that the mycoplasma is able to release significant amounts of H2O2 into the growth medium.

Using a cell culture infection model for M. mycoides subsp. mycoides SC it was shown that H2O2 is not only released by the mycoplasma but directly translocated into the eukaryotic host cells causing cytotoxicity and cell death [13]. M. mycoides subsp. mycoides SC strains generally possess the glycerol ABC transporter GtsABC for the assimilation of glycerol with the exception of strains from recent European outbreaks (period 1980–1999) producing significantly lower amounts of H2O2 [8, 20] and showing a very low cytotoxicity level towards eukaryotic host cells in the presence of glycerol compared to most other strains [4, 13]. Blocking the L-α-glycerophosphate oxidase GlpO by monospecific anti-GlpO antibodies hampers the production and release of H2O2 and concomitantly inhibits cytotoxicity of M. mycoides subsp. mycoides SC towards embryonic calf nasal epithelial (ECaNEp) cells [13]. These results indicate that GlpO may act as a useful antigen target to induce protective immunity against CBPP.

Furthermore, since GlpO-promoted H2O2 production is a central virulence attribute of M. mycoides subsp. mycoides SC, the glpO gene represents an ideal target for mutations in order to better attenuate vaccine strains, whose residual virulence is of concern. In this respect, it has to be mentioned that the live vaccine strains T1/44 and T1/Sm are currently used to vaccinate against CBPP [18]. However, both vaccines have been shown to contain a fully active glycerol transport system GtsABC and an enzymatically active L-α-glycerophosphate oxidase GlpO, which results in the release of significant amounts of H2O2 in the presence of physiological concentrations of glycerol [4, 21] and cytotoxicity against ECaNEp cells [4]. This feature might be the cause of unwanted side reactions encountered with these vaccines [17] and of CBPP elicitation in cattle upon inoculation of T1-derived vaccines via the endobronchial route [11].

In the current work, we studied the basic, structural and functional features of GlpO of M. mycoides subsp. mycoides SC with the aim of identifying mutation sites in the glpO gene for future live vaccine development. In this view, the mutations should result in (i) loss of H2O2-producing enzymatic activity; (ii) attenuation of currently used vaccines; and (iii) preservation of the antigenic properties of GlpO required to elicit neutralising antibodies like those induced by native GlpO [13]. GlpO (EC 1.1.3.21) is an enzyme that belongs to the family of oxidoreductases, specifically those acting on the CH-OH group of the donor with oxygen as the acceptor. It employs one cofactor, flavin adenine dinucleotide (FAD). In biochemistry, FAD is a redox cofactor involved in several important metabolic reactions. FAD can exist in two different redox states and its biochemical role usually involves alternating between these two states. FAD can be reduced to FADH2, whereby it accepts two hydrogen atoms. FADH2 is an energy-carrying molecule, and the reduced coenzyme can be utilised as a substrate for oxidative phosphorylation. FADH2 is re-oxidised to FAD, which makes it possible to produce two moles of the universal energy carrier ATP. The ADP portion of FAD interacts mostly via residues from the first β-strand–α-helix–β-strand (βαβ) fold in the N-terminal portion of GlpO, which includes the Gly-X-Gly-X-X-Gly fingerprint that interacts with the diphosphate moiety of FAD [7].

The present study investigated the function of recombinant GlpO from M. mycoides subsp. mycoides SC, determining the FAD-binding site and the potential of mutant GlpOΔFAD, i.e., the recombinant GlpO without the FAD-binding site Gly12-Gly13-Gly14-Ile15-Ile16-Gly17, to induce antibodies neutralising cytotoxicity of M. mycoides subsp. mycoides SC.

2. MATERIALS AND METHODS

2.1. Strains, growth conditions and DNA extraction

M. mycoides subsp. mycoides SC strain Afadé, a highly virulent field strain isolated in 1968 in Cameroon, was used for the virulence studies unless otherwise stated. Mycoplasmal cultures were grown in a standard mycoplasma medium (Axcell Biotechnologies, St. Genis-L’Argentière, France) at 37 °C to a density of approximately 5 × 108 cells per mL. For genetic manipulation and subcloning, Escherichia coli strains DH5α and BL21(DE3) were used. The pETHIS-1 expression vector [15] was used for expression of recombinant N- and C-terminal polyhistidine fusion proteins.

2.2. ECaNEp cells

ECaNEp cells were prepared from foetuses obtained from a local abattoir and were maintained in minimal essential medium (MEM)-Earle medium supplemented with 7% foetal calf serum and penicillin (100 IU/mL) in 24-well microtiter plates and used at a confluent density of 2 × 105 cells per well, at 37 °C in a humidified 5% CO2 atmosphere [13]. Foetal calf serum and cell culture media were purchased from Seromed (Biochrom, Munich, Germany). The cells were routinely screened for contamination by mycoplasmas using PCR or for contamination by bovine viral diarrhoea virus using immunostaining [16].

2.3. Bioinformatics

Comparison of the GlpO protein sequence of strain Afadé with the EMBL/GenBank database was performed using the BLAST program blastp [1]. The PSIPRED Protein Structure Prediction Server v2.6 [12] was used to predict the secondary structure of GlpO of strain Afadé.

Since no 3D structures are annotated in the 3D databases for GlpO of M. mycoides subsp. mycoides SC, a search for 3D domains for bacterial GlpO sequences within available sequences from the NCBI site was performed. Among the 14 GlpO structures found, L-α-glycerophosphate oxidase 2RGH:A from Streptococcus [7] was selected for comparison with the GlpO of M. mycoides subsp. mycoides SC. The 3D-Mol Viewer software from the Vector NTI Advance 10.3.0 package (Invitrogen AG, Basel, Switzerland) was used to display the 3D protein structure model of 2RGH:A with its FAD molecule. The SWISS-MODEL software [2], accessible via the ExPASy web server, was used to display GlpO of M. mycoides subsp. mycoides SC taking 2RGH:A from Streptococcus as the template.

2.4. Cloning and expression of recombinant proteins

The glpO gene of M. mycoides subsp. mycoides strain Afadé was first amplified with the primers Afadé_EcoRI and MmmSC_GlpO_BamHI (Tab. I) containing the restriction sites for EcoRI and BamHI, respectively. As a template for PCR amplification, plasmid pETHIS-1::glpO containing mutagenised TGGTrp codons in the glpO gene [13] was used. The pETHIS-1 vector was amplified with the primers pETHIS-1_EcoRI and pETHIS-1_BamHI (Tab. I) to give the linearised vector pETHIS-1m harbouring EcoRI and BamHI restriction sites. This procedure was used to create GlpO fusion with a polyhistidine tail at the C-terminus only. Purified recombinant GlpO carrying the 10xHis-tagged C-terminus was shown to be enzymatically more active under the in vitro conditions used in this study compared to GlpO with polyhistidine tags at both N-terminal and C-terminal ends.

Table I.

Oligonucleotide primers.

The codons for the putative FAD-binding site Gly12-Gly13-Gly14-Ile15-Ile16-Gly17 on glpO were removed using the overlap extension-PCR method [5] to obtain the glpOΔFAD gene. The codons for the putative FAD-binding site located at nucleotides (nt) 34–51 downstream of the ATG start codon of the glpO sequence of strain Afadé with EMBL/GenBank accession number AJ581566 were removed subsequently with (i) primer pair MmmSC_glpO_FAD1F and MmmSC_glpO_FAD1R (Tab. I) to remove bases 34–39 coding for Gly12-Gly13, followed by (ii) primer pair MmmSC_glpO_FAD2F and MmmSC_glpO_FAD2R (Tab. I) to remove bases 40–45 coding for Gly14-Ile15, and (iii) primer pair MmmSC_glpO_FAD3F and MmmSC_glpO_FAD3R (Tab. I) to remove bases 46–51 coding for Ile16-Gly17. The products were cloned into pETHIS-1m by flanking EcoRI and BamHI cleavage sites. The constructs were analysed by DNA sequencing and the final clone was introduced into E. coli BL21(DE3) for expression. The expression of the cloned glpOΔFAD gene was induced by the addition of 1 mM IPTG at mid-exponential phase and further incubation for 4 h at 30 °C. Following induction, the polyhistidine-tailed fusion protein GlpOΔFAD was purified via Ni2+ chelation chromatography on a Ni-NTA column under denaturing conditions as described previously for other proteins [15, 19]. The fractions were dialysed and analysed on SDS-10% polyacrylamide gels.

Immunoblot analysis with mouse anti-His antibody (GE Healthcare, Otelfingen, Switzerland) (at a dilution of 1:3000) was carried out by standard methods [3], using SDS-10% polyacrylamide gels for separation of the proteins, which were blotted onto nitrocellulose membranes. Phosphatase-labelled conjugate goat anti-mouse immunoglobulin G (heavy plus light chains) (Kirkegaard and Perry Laboratories, Gaithersburg, MD, USA) at a dilution of 1:2000 was used for the visualisation of recombinant, polyhistidine-tailed proteins on blots.

2.5. In vitro activity assay

Both recombinant, polyhistidine-tailed proteins GlpO and GlpOΔFAD were also purified under native conditions in order to maintain their biological activity. For this purpose, E. coli BL21(DE3) cells expressing the recombinant protein GlpO were suspended in buffer A (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8.0) followed by sonification in a Branson Sonifier and binding for 1 h at 4 °C to Ni-NTA beads. The beads were then washed with buffer C (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, pH 8.0) and eluted with buffer D (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, pH 8.0). Recombinant GlpOΔFAD was purified by suspending the corresponding IPTG-induced E. coli cells in buffer B (50 mM NaH2PO4, 300 mM NaCl, 1 mM imidazole, pH 6.5) followed by sonification in a Branson Sonifier and binding for 1 h at 4 °C to Ni-NTA beads. The beads were then washed with buffer B and eluted with buffer D. The fractions were analysed on SDS-10% polyacrylamide gels.

The activities of purified recombinant proteins GlpO and GlpOΔFAD were measured by their ability to produce H2O2 in the presence of glycerol. The activity assay was performed in the presence of filter-sterilised incubation buffer (67.7 mM HEPES, 140 mM NaCl, 7 mM MgCl2, pH 7.3) supplemented with 9 mM ATP, 65 nM FAD and 6 mM α-glycerol-3-phosphate (G3P), to which recombinant GlpO or recombinant GlpOΔFAD was added at a final 20 μg/mL concentration. The production of H2O2 was measured with the peroxide test (Merck KgaA, Darmstadt, Germany) as described previously [20].

2.6. Fluorescence measurements

Fluorescence excitation and emission spectra of the proteins was measured with a Perkin-Elmer LS50 dual wavelength fluorescence spectrophotometer. The emission spectrum (370–600 nm) at an excitation wavelength of 350 nm and the excitation spectrum (250–410 nm) at an emission wavelength of 430 nm were recorded as described previously [6].

2.7. Sera, polyclonal antibodies and immunoglobulin purification

Immune sera directed against recombinant GlpO or GlpOΔFAD were obtained by subcutaneous immunisation of rabbits with 80 μg of purified recombinant polyhistidine-tailed proteins in 500 μL of phosphate-buffered saline (PBS) buffer (50 mM Na2HPO4/NaH2PO4, 140 mM NaCl, pH 8.0), mixed with 500 μL of Adjuvant 10 (Gerbu Biotechnik GmBH, Gaiberg, Germany). Primary immunisation was followed by booster injections with 80 μg of the respective protein 2 and 4 weeks later. The rabbits were bled 21 days after the last booster immunisation. Antisera were prepared from the blood samples and stored at −20 °C.

Immunoglobulin G (IgG) fractions from decomplemented rabbit anti-GlpO and anti-GlpOΔFAD sera were purified with the HiTrap Protein G kit (Amersham Pharmacia Biotech, Uppsala, Sweden) as directed by the manufacturer. Protein concentrations were determined using a NanoDrop ND-1000 spectrophotometer (Witec AG, Littau, Switzerland).

2.8. Quantification of H2O2 and inhibition assay

To measure H2O2 production, strains of M. mycoides subsp. mycoides SC were grown in mycoplasma culture medium for 3 days at 37 °C 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, resuspended in pre-warmed 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 the peroxide test (Merck KgaA) as described previously [20] at time zero and 1, 2, 5, 10, 20 and 40 min after the addition of glycerol. In order to measure the inhibitory potential of anti-GlpO or anti-GlpOΔFAD antibodies, mycoplasmas were pre-treated with IgG preparations of either anti-GlpO or anti-GlpOΔFAD antiserum at concentrations of 10 μg/mL for 30 min at 37 °C, followed by two washes with PBS buffer.

2.9. Detection of translocation of H2O2 from infecting mycoplasmas into ECaNEp cells

To monitor intracellular H2O2 and accompanying reactive oxygen species (ROS) in ECaNEp cells, the formation of dichlorofluorescein derivatives after cleavage of the ester groups of nonfluorescent 5 (and 6)-chloromethyl-2’,7’-dichlorodihydrofluorescein diacetate-acetyl ester (CM-H2DCFDA; Molecular Probes, Eugene, OR, USA) by H2O2-dependent oxidation was measured as already described [13]. The ECaNEp cells were incubated for 1 h with 10 μM CM-H2DCFDA and washed once with MEM-Earle medium. Cells were then infected with M. mycoides subsp. mycoides SC strain Afadé at a multiplicity of infection (MOI) of 500 mycoplasmas per cell in the presence or absence of 100 μM glycerol. To assess blocking of the GlpO activity, M. mycoides subsp. mycoides SC cells were pre-treated with anti-GlpO or anti-GlpOΔFAD IgG at 8 to 28 μg/mL. 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.

3. RESULTS

3.1. FAD-binding site Gly-X-Gly-X-X-Gly of GlpO

The results obtained with the BLAST program blastp showed that the 387-aa GlpO of M. mycoides subsp. mycoides SC strain Afadé (EMBL/GenBank accession number CAE46342) has strong homologies to several bacterial glycerol-3-phosphate oxidases and glycerol-3-phosphate dehydrogenases. In some cases, these similar proteins are larger than 450 aa in length and the GlpO-homologous portions reside in their N-termini (not shown).

The secondary structure prediction of GlpO of strain Afadé indicates the presence of a βαβ fold in its N-terminal portion, which includes the Gly-Gly-Gly-Ile-Ile-Gly fingerprint that is assumed to interact with the diphosphate moiety of FAD (Fig. 1). A comparison with protein 2RGH:A, the L-α-glycerophosphate oxidase from Streptococcus [7], demonstrated the presence of two putative G3P-binding sites also in GlpO from M. mycoides subsp. mycoides SC. They are located in the region spanning aa residues 310–350, close to the C-terminus (Fig. 1).

thumbnail Figure 1.

Secondary structure of GlpO of M. mycoides subsp. mycoides SC strain Afadé. Boxes represent FAD-binding (red) and putative G3P-binding (blue) segments, as defined previously [7]. Yellow arrows and green cylinders represent β-sheets and α-helices, respectively, as obtained by the protein structure prediction program PSIPRED. (For a colour version of this figure, please consult www.vetres.org..)

Protein 2RGH:A, whose 3D structure is known, was chosen as a model to envisage the possible 3D structure of GlpO from M. mycoides subsp. mycoides SC. 2RGH:A showed the three Gly residues in the Gly-X-Gly-X-X-Gly fingerprint (in this case Gly39-Gly40-Gly41-Ile42-Thr43-Gly44) in a βαβ fold in the vicinity of the diphosphate moiety of FAD (Fig. 2A), which corresponds to the putative FAD-binding site Gly12-Gly13-Gly14-Ile15-Ile16-Gly17 in the GlpO of M. mycoides subsp. mycoides SC strain Afadé (Fig. 2B).

thumbnail Figure 2.

Three-dimensional model showing the FAD-binding site of GlpO. The “ball and stick” views were obtained with the 3D-Mol Viewer program to display: (A) the predicted structure model of the FAD-binding site of 2RGH:A of Streptococcus [7] and its bound FAD molecule (“stick” view); and (B) the SWISS-MODEL-simulated structure of the FAD-binding site of GlpO of M. mycoides subsp. mycoides SC strain Afadé obtained by taking 2RGH:A from Streptococcus as the template (FAD is not displayed by SWISS-MODEL-generated structure simulation). The protein chains (excluding the FAD-binding sites Gly-X-Gly-X-X-Gly) are illustrated as blue wires. The displayed FAD molecule is shown in red and the amino acids X of the FAD-binding sites are shown in blue while their three amino acids Gly are displayed in yellow. (For a colour version of this figure, please consult www.vetres.org..)

3.2. Structural properties and enzymatic activity of recombinant GlpO and GlpOΔFAD

Immunoblot analysis of recombinant proteins GlpO and GlpOΔFAD, both carrying a 10xHis-tagged C-terminus, with anti-His antibodies revealed two protein bands of 45 and 90 kDa for GlpO and one single protein band of 44 kDa for GlpOΔFAD (Fig. 3A). Both polyclonal monospecific anti-recombinant GlpO and GlpOΔFAD sera strongly reacted with a 43-kDa protein and poorly reacted with 85- and 125-kDa proteins from total antigen of M. mycoides subsp. mycoides SC strain Afadé (Fig. 3B).

thumbnail Figure 3.

Immunoblot detection of GlpO and GlpOΔFAD. (A) One-microgram samples of purified recombinant GlpO (lane 1) and GlpOΔFAD (lane 2) were run by SDS-10% polyacrylamide gels, transferred onto a nitrocellulose membrane and probed with anti-His antibody. (B) Total antigen (approximately 10 μg per lane) of M. mycoides subsp. mycoides SC strain Afadé was separated on SDS-10% polyacrylamide gels and immunoblotted using either anti-GlpO (lane 1) or anti-GlpOΔFAD antiserum (lane 2) at a dilution of 1:1000. Std, molecular mass standards.

To test the activity of both enzymes, recombinant proteins GlpO and GlpOΔFAD were purified under native conditions and tested for their functional and enzymatic activities. A final concentration of 20 μg/mL for both recombinant proteins was used in the assays. Under in vitro conditions, recombinant GlpO of M. mycoides subsp. mycoides SC oxidised G3P with release of 2 μg/mL H2O2 into the assay solution after 10 min incubation and 10 μg/mL H2O2 after 40 min incubation. In contrast, recombinant GlpOΔFAD showed no release of H2O2 into the assay solution.

3.3. Spectroscopic analysis of GlpO and GlpOΔFAD

It was observed that cell lysates of GlpO-producing E. coli had a yellowish appearance, whereas cell lysates with the protein GlpOΔFAD had a whitish appearance, like E. coli BL21(DE3) harbouring vector pETHIS-1 without insert. This indicates that cell lysates with recombinant GlpO contained protein-bound flavin (i.e. bound FAD), whereas GlpOΔFAD did not bind FAD.

In order to confirm this, fluorescence spectra of the two recombinant proteins GlpO and GlpOΔFAD were measured and compared with the spectra of an FAD solution, the latter displaying an emission maximum at 525 nm. The fluorescence spectra of recombinant GlpO-samples displayed the typical protein-derived excitation maximum at 280 nm and also showed an emission maximum at 525 nm (Fig. 4), indicating that recombinant GlpO contained bound FAD that must have been acquired during overexpression in E. coli. In contrast, fluorescent spectroscopic analysis of recombinant GlpOΔFAD displayed solely an excitation maximum at 280 nm and no peak visible at 525 nm in the emission spectrum, showing that this recombinant protein did not contain bound FAD (Fig. 4). These results indicate that recombinant GlpOΔFAD lacking the putative FAD-binding site Gly12-Gly13-Gly14-Ile15-Ile16-Gly17 was unable to acquire and bind the cofactor FAD when expressed in E. coli, hence explaining why it is enzymatically inactive.

thumbnail Figure 4.

Fluorescence excitation and emission spectra of GlpO, GlpOΔFAD and FAD. Spectra of GlpO (4.6 μM), GlpOΔFAD (1.2 μM) and FAD (7.7 μM) in 20 mM PBS were recorded. Excitation wavelength was 350 nm, and emission wavelength was 430 nm. Excitation spectra were measured at 250–410 nm; emission spectra were measured at 370–600 nm. The bold line represents the recombinant protein GlpO, the dashed line represents GlpOΔFAD and the dotted line represents free FAD.

3.4. Neutralising activity of anti-GlpOΔFAD antibodies

With respect to future vaccine design, it appears highly desirable to use enzymatically inactive GlpO eliciting antibodies in immunised animals able, like those directed against native GlpO [13], to neutralise GlpO and thereby obliterating the cytotoxic effect of M. mycoides subsp. mycoides SC. The neutralising activity of rabbit antibodies that were induced by immunisation with recombinant GlpOΔFAD was therefore assessed.

As observed with rabbit anti-GlpO [13], decomplemented rabbit anti-GlpOΔFAD serum and protein G-purified IgG prepared thereof were shown to specifically block the release of H2O2 by M. mycoides subsp. mycoides SC strain Afadé grown in liquid medium supplemented with 100 μM glycerol. Pre-immune serum or IgG of the unvaccinated rabbit had no inhibitory effect. Consequently, the GlpOΔFAD protein lacking the FAD-binding site and the native GlpO have similar capacities to induce GlpO-neutralising antibodies.

In order to evaluate whether anti-GlpOΔFAD antibodies are also able to neutralise translocation of H2O2 and the accompanying oxygen radicals from M. mycoides subsp. mycoides SC into ECaNEp cells, and the resulting cytotoxic effects, ECaNEp cells were pre-treated with CM-H2DCFDA to detect intracellular oxidation of this compound by fluorescence microscopy. 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, an event followed by cell death. No fluorescence was detected in the control experiment where ECaNEp cells were infected with the Afadé strain without the addition of glycerol (Fig. 5). Intracellular oxidation of CM-H2DCFDA in ECaNEp cells as visualised by fluorescence was inhibited when infecting M. mycoides subsp. mycoides SC were pre-treated with antibodies directed either against GlpO (Fig. 5C) or against GlpOΔFAD (Fig. 5D). This showed that GlpOΔFAD is able to induce antibodies that prevent H2O2 translocation from infecting M. mycoides subsp. mycoides SC to the host cells.

thumbnail Figure 5.

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. ECaNEp cells 20 min after infection with strain Afadé in the absence (A) and in the presence of glycerol (B), or with strain Afadé pre-treated with anti-GlpO IgG (C) or with anti-GlpOΔFAD IgG (D) in the presence of glycerol.

4. DISCUSSION

With the aim of designing new live vaccines against CBPP using specific, targeted, attenuating mutations in the genome of M. mycoides subsp. mycoides SC, we investigated the role of the FAD-binding site of the L-α-glycerophosphate oxidase GlpO, a central enzyme for the virulence of this pathogen [4, 13, 14].

Spectroscopic analysis revealed that the modified recombinant GlpOΔFAD – the GlpO protein-mutant whose FAD-binding site was knocked out by recombinant gene technology – was devoid of bound FAD while the intact recombinant GlpO did contain bound FAD. It is interesting to note at this point that purified recombinant GlpO, in contrast to GlpOΔFAD, showed two protein bands on SDS-polyacrylamide gels and immunoblots, the larger band being attributed to the formation of a dimeric aggregate of GlpO, in accordance with the finding that this oxidase is a dimeric protein in other organisms [6, 7]. The observation that GlpOΔFAD is present uniquely as a monomer on immunoblots suggests that the lack of the FAD-binding site (or the failure of binding FAD) also changed structural properties of this protein. Furthermore, the FAD-binding site of GlpO was demonstrated to play a central role in the enzymatic activity of the oxidase, as shown by the inability of GlpOΔFAD to oxidise α-glycerol-3-phosphate (G3P) and to produce toxic H2O2.

Antibodies that were raised in rabbits against purified recombinant GlpOΔFAD, like anti-GlpO antibodies, were able to (i) recognise three bands (representing monomer, dimer and a possible trimer on immunoblots) of GlpO from total proteins of M. mycoides subsp. mycoides SC; (ii) block the H2O2 production of M. mycoides subsp. mycoides SC metabolising glycerol; and (iii) neutralise the resulting cytotoxicity towards host cells in an in vitro assay.

Taken together, these results suggest that the removal of the FAD-binding site from the glpO gene of M. mycoides subsp. mycoides SC provides an ideal means to attenuate cytotoxicity of future live vaccines, with preserved capacity to elicit efficient GlpO-neutralising antibodies in the host. In order to replace the glpO wild type gene in the chromosome of M. mycoides subsp. mycoides SC with the glpOΔFAD allele for creation of a targeted deletion mutant, novel, more powerful methods for allelic replacement still need to be developed since this organism does not seem to be particularly apt to homologous recombination [10].

Acknowledgments

This work has been supported by grant No. 075804 “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. We are grateful to Hartmut Porzig (Institute of Pharmacology, University of Bern, Switzerland) for expert help with fluorescence measurements and for most valuable critical comments and editorial help. We thank Yvonne Schlatter and Antoinette Golomingi (Institute of Veterinary Virology, University of Bern) for excellent technical assistance.

References

  1. Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J., Basic local alignment search tool, J. Mol. Biol. (1990) 215:403–410 [CrossRef] [PubMed].
  2. Arnold K., Bordoli L., Kopp J., Schwede T., The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling, Bioinformatics (2006) 22:195–201 [CrossRef] [PubMed].
  3. Ausubel F.M., Brent R., Kingston R.E., Moore D.D., Seidman J.G., Smith J.A., Struhl K., Current protocols in molecular biology, John Wiley & Sons, Inc., New York, 1999.
  4. Bischof D.F., Janis C., Vilei E.M., Bertoni G., Frey J., Cytotoxicity of Mycoplasma mycoides subsp. mycoides small colony type to bovine epithelial cells, Infect. Immun. (2008) 76:263–269 [CrossRef] [PubMed].
  5. Braman J., Papworth C., Greener A., Site-directed mutagenesis using double-stranded plasmid DNA templates, Methods Mol. Biol. (1996) 57:31–44 [PubMed].
  6. Claiborne A., Studies on the structure and mechanism of Streptococcus faecium L-alpha-glycerophosphate oxidase, J. Biol. Chem. (1986) 261:14398–14407 [PubMed].
  7. Colussi T., Parsonage D., Boles W., Matsuoka T., Mallett T.C., Karplus P.A., Claiborne A., Structure of $\alpha$-glycerophosphate oxidase from Streptococcus sp.: a template for the mitochondrial $\alpha$-glycerophosphate dehydrogenase, Biochemistry (2008) 47:965–977 [CrossRef] [PubMed].
  8. Djordjevic S.P., Vilei E.M., Frey J., Characterization of a chromosomal region of Mycoplasma sp. bovine group 7 strain PG50 encoding a glycerol transport locus (gtsABC), Microbiology (2003) 149:195–204 [CrossRef] [PubMed].
  9. Hames C., Halbedel S., Hoppert M., Frey J., Stülke J., Glycerol metabolism is important for cytotoxicity of Mycoplasma pneumoniae, J. Bacteriol. (2009) 191:747–753 [CrossRef] [PubMed].
  10. Janis C., Bischof D., Gourgues G., Frey J., Blanchard A., Sirand-Pugnet P., Unmarked insertional mutagenesis in the bovine pathogen Mycoplasma mycoides subsp. mycoides SC: characterization of a lppQ mutant, Microbiology (2008) 154:2427–2436 [CrossRef] [PubMed].
  11. Mbulu R.S., Tjipura-Zaire G., Lelli R., Frey J., Pilo P., Vilei E.M., et al., Contagious bovine pleuropneumonia (CBPP) caused by vaccine strain T1/44 of Mycoplasma mycoides subsp. mycoides SC, Vet. Microbiol. (2004) 98:229–234 [CrossRef] [PubMed].
  12. McGuffin L.J., Bryson K., Jones D.T., The PSIPRED protein structure prediction server, Bioinformatics (2000) 16:404–405 [CrossRef] [PubMed].
  13. Pilo P., Vilei E.M., Peterhans E., Bonvin-Klotz L., Stoffel M.H., Dobbelaere D., Frey J., A metabolic enzyme as a primary virulence factor of Mycoplasma mycoides subsp. mycoides Small Colony, J. Bacteriol. (2005) 187:6824–6831 [CrossRef] [PubMed].
  14. Pilo P., Frey J., Vilei E.M., Molecular mechanisms of pathogenicity of Mycoplasma mycoides subsp. mycoides SC, Vet. J. (2007) 174:513–521 [CrossRef] [PubMed].
  15. Schaller A., Kuhn R., Kuhnert P., Nicolet J., Anderson T.J., MacInnes J.I., et al., Characterization of apxIVA, a new RTX determinant of Actinobacillus pleuropneumoniae, Microbiology (1999) 145:2105–2116 [CrossRef] [PubMed].
  16. Schweizer M., Peterhans E., Oxidative stress in cells infected with bovine viral diarrhoea virus: a crucial step in the induction of apoptosis, J. Gen. Virol. (1999) 80:1147–1155 [PubMed].
  17. Thiaucourt F., Yaya A., Wesonga H., Huebschle O.J.B., Tulasne J.J., Provost A., Contagious bovine pleuropneumonia: a reassessment of the efficacy of vaccines used in Africa, Ann. N. Y. Acad. Sci. (2000) 916:71–80 [PubMed].
  18. Thiaucourt F., Aboubakar Y., Wesonga H., Manso-Silvan L., Blanchard A., Contagious bovine pleuropneumonia vaccines and control strategies: recent data, Dev. Biol. (Basel) (2004) 119:99–111 [PubMed].
  19. Vilei E.M., Abdo E.-M., Nicolet J., Botelho A., Gonçalves R., Frey J., Genomic and antigenic differences between the European and African/Australian clusters of Mycoplasma mycoides subsp. mycoides SC, Microbiology (2000) 146:477–486 [PubMed].
  20. Vilei E.M., Frey J., Genetic and biochemical characterization of glycerol uptake in Mycoplasma mycoides subsp. mycoides SC: its impact on H2O2 production and virulence, Clin. Diagn. Lab. Immunol. (2001) 8:85–92 [PubMed].
  21. Vilei E.M., Correia I., Ferronha M.H., Bischof D.F., Frey J., $\beta$-D-Glucoside utilization by Mycoplasma mycoides subsp. mycoides SC: possible involvement in the control of cytotoxicity towards bovine lung cells, BMC Microbiol. (2007) 7:31 [CrossRef] [PubMed].

All Tables

Table I.

Oligonucleotide primers.

All Figures

thumbnail Figure 1.

Secondary structure of GlpO of M. mycoides subsp. mycoides SC strain Afadé. Boxes represent FAD-binding (red) and putative G3P-binding (blue) segments, as defined previously [7]. Yellow arrows and green cylinders represent β-sheets and α-helices, respectively, as obtained by the protein structure prediction program PSIPRED. (For a colour version of this figure, please consult www.vetres.org..)

In the text
thumbnail Figure 2.

Three-dimensional model showing the FAD-binding site of GlpO. The “ball and stick” views were obtained with the 3D-Mol Viewer program to display: (A) the predicted structure model of the FAD-binding site of 2RGH:A of Streptococcus [7] and its bound FAD molecule (“stick” view); and (B) the SWISS-MODEL-simulated structure of the FAD-binding site of GlpO of M. mycoides subsp. mycoides SC strain Afadé obtained by taking 2RGH:A from Streptococcus as the template (FAD is not displayed by SWISS-MODEL-generated structure simulation). The protein chains (excluding the FAD-binding sites Gly-X-Gly-X-X-Gly) are illustrated as blue wires. The displayed FAD molecule is shown in red and the amino acids X of the FAD-binding sites are shown in blue while their three amino acids Gly are displayed in yellow. (For a colour version of this figure, please consult www.vetres.org..)

In the text
thumbnail Figure 3.

Immunoblot detection of GlpO and GlpOΔFAD. (A) One-microgram samples of purified recombinant GlpO (lane 1) and GlpOΔFAD (lane 2) were run by SDS-10% polyacrylamide gels, transferred onto a nitrocellulose membrane and probed with anti-His antibody. (B) Total antigen (approximately 10 μg per lane) of M. mycoides subsp. mycoides SC strain Afadé was separated on SDS-10% polyacrylamide gels and immunoblotted using either anti-GlpO (lane 1) or anti-GlpOΔFAD antiserum (lane 2) at a dilution of 1:1000. Std, molecular mass standards.

In the text
thumbnail Figure 4.

Fluorescence excitation and emission spectra of GlpO, GlpOΔFAD and FAD. Spectra of GlpO (4.6 μM), GlpOΔFAD (1.2 μM) and FAD (7.7 μM) in 20 mM PBS were recorded. Excitation wavelength was 350 nm, and emission wavelength was 430 nm. Excitation spectra were measured at 250–410 nm; emission spectra were measured at 370–600 nm. The bold line represents the recombinant protein GlpO, the dashed line represents GlpOΔFAD and the dotted line represents free FAD.

In the text
thumbnail Figure 5.

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. ECaNEp cells 20 min after infection with strain Afadé in the absence (A) and in the presence of glycerol (B), or with strain Afadé pre-treated with anti-GlpO IgG (C) or with anti-GlpOΔFAD IgG (D) in the presence of glycerol.

In the text