Identification of a secreted superoxide dismutase in Mycobacterium avium ssp. paratuberculosis


  • Xiaofei Liu,

    1. Department of Veterinary and Biomedical Sciences, 211 VBS, Fair and East Campus Loop, University of Nebraska, Lincoln, NE 68583-0905, USA
    Search for more papers by this author
  • Zhengyu Feng,

    1. Department of Veterinary and Biomedical Sciences, 211 VBS, Fair and East Campus Loop, University of Nebraska, Lincoln, NE 68583-0905, USA
    Search for more papers by this author
  • N.Beth Harris,

    1. Department of Veterinary and Biomedical Sciences, 211 VBS, Fair and East Campus Loop, University of Nebraska, Lincoln, NE 68583-0905, USA
    Search for more papers by this author
  • Jeffrey D. Cirillo,

    1. Department of Veterinary and Biomedical Sciences, 211 VBS, Fair and East Campus Loop, University of Nebraska, Lincoln, NE 68583-0905, USA
    Search for more papers by this author
  • Herve Bercovier,

    1. Department of Clinical Microbiology, Hebrew University-Hadassah Medical School, P.O. Box 12272, Jerusalem 91120, Israel
    Search for more papers by this author
  • Raúl G. Barletta

    Corresponding author
    1. Department of Veterinary and Biomedical Sciences, 211 VBS, Fair and East Campus Loop, University of Nebraska, Lincoln, NE 68583-0905, USA
    Search for more papers by this author

*Corresponding author. Tel.: +1 (402) 472 8543; Fax: +1 (402) 472 9690, E-mail:


Mycobacterium avium ssp. paratuberculosis (M. paratuberculosis), the causative agent of Johne's disease, is an important animal pathogen that has also been implicated in human disease. The major proteins expressed by M. paratuberculosis were analyzed by two-dimensional gel electrophoresis, and a superoxide dismutase (Sod) was identified from this protein profile. The M. paratuberculosis Sod has a molecular mass of 23 kDa and an isoelectric point of 6.1. Sequence analysis of the corresponding sodA gene from M. paratuberculosis indicates that this protein is a manganese-dependent enzyme. We show that the M. paratuberculosis Sod is actively secreted, suggesting that it may elicit a protective cellular immune response in the host during infection.


Mycobacterium avium ssp. paratuberculosis (M. paratuberculosis) is the etiologic agent of paratuberculosis, or Johne's disease, a granulomatous enteritis affecting ruminants worldwide [1]. M. paratuberculosis is also suspected to be the etiologic agent of Crohn's disease, a chronic enteritis in humans [2]. Unlike other mycobacteria, little is known on the mechanisms of immunity, pathogenesis, and the molecular genetics of M. paratuberculosis[2,3]. To date, only a few immunogenic proteins have been identified and characterized [1]. Of particular interest are the secreted proteins, which have been implicated as protective antigens in M. tuberculosis infections [4]. Secreted proteins have not been studied extensively in M. paratuberculosis because of the high concentration of bovine serum albumin required to sustain bacterial growth in vitro, which hinders the identification and separation of these proteins. Thus, only proteins with particular properties, such as the ability to bind to concavalin A, have been thoroughly characterized [5].

Superoxide dismutases (Sods; EC catalyze the dismutation of superoxide radicals into hydrogen peroxide and molecular oxygen, and may protect M. paratuberculosis against oxidative stress. A secreted Sod has been identified as a predominant protein in culture supernatant fluids of Mycobacterium tuberculosis[4]. Furthermore, Sod is secreted in significantly higher proportions by pathogenic M. tuberculosis than non-pathogenic Mycobacterium smegmatis[6]. Thus, the objectives of this study were to (i) determine levels of Sod in broth cultures; (ii) characterize the M. paratuberculosis sodA gene; (iii) determine the localization of Sod in M. paratuberculosis subcellular fractions.

2Materials and methods

2.1Bacterial strains, plasmids and growth conditions

The M. paratuberculosis strains used in this study were ATCC 19698, and the bovine clinical isolates K-10, K-39, K-48, S-8, and Lorraine (provided by D.L. Whipple, National Animal Disease Center, Ames, IA, USA). These strains were grown in Middlebrook 7H9 broth (7H9 broth) supplemented with oleic acid, albumin, dextrose, cycloheximide and mycobactin J as previously described [7]. M. avium strain MAC104 (provided by L.E. Bermudez, Kuzell Institute) was grown in 7H9 broth containing albumin, dextrose and cycloheximide [7].

2.2Fractionation of M. paratuberculosis proteins

To minimize cell autolysis, M. paratuberculosis strain K-10 cells were harvested in early exponential phase (OD600 nm ca. 0.3) by centrifugation at 5000×g at 4°C for 15 min. The culture supernatant fluid was removed and filtered through a 0.2-μm filter and stored at −20°C. The cell pellet was washed with 10 ml of phosphate-buffered saline (PBS) and centrifuged as described above. Cells were disrupted by sonication and fractionated as described [8]. Protein concentration of the fractions was determined using the DC assay kit (Bio-Rad Laboratories).

2.3Two-dimensional (2D) gel electrophoresis analysis and N-terminal sequencing

Proteins from the cytosolic and cell membrane/cell wall extracts were resolved by 2D gel electrophoresis according to the method of O'Farrell [9] (Kendrick Labs). Isoelectric focusing was carried out in glass tubes of 2.0-mm inner diameter, using 2% resolytes pH 4–8 ampholines for 9600 Volt-hour. After equilibration for 10 min in buffer (50 mM dithiothreitol, 2.3% SDS, 0.0625 M Tris–HCl, pH 6.8), the tube gels were sealed to the top of a stacking gel. Proteins were resolved on a 10% polyacrylamide slab gel (0.75 mm thick) at 12.5 mA for 4 h. Gels were stained with Coomassie blue, and patterns of polypeptide spots were compared by visual inspection. Images from the gels were digitally captured and processed only to enhance brightness and contrast.

For determination of the N-terminal amino acid sequence, the gel was placed in transfer buffer (12.5 mM Tris–HCl, pH 8.8, 86 mM glycine, 10% methanol), following 2D electrophoresis, and transblotted overnight onto a PVDF membrane at 100 mA and 100 V. The membrane was stained with Coomassie blue, and the protein spot of interest was eluted and subjected to N-terminal amino acid sequencing (University of Nebraska-Lincoln Protein Core Facility).

2.4Construction of M. paratuberculosis recombinant genomic library

Genomic DNA from M. paratuberculosis strain K-10 was isolated and partially digested with Sau3AI. DNA fragments in the 2–3-kb size range were purified from 1% agarose gels using glassmilk and ligated to the Escherichia coli–Mycobacterium shuttle vector pMV262 [10], which had been linearized with BamHI and dephosphorylated. The ligation mixture was transformed into E. coli STBL2 (Life Technologies) and approximately 29 600 recombinants were obtained for a theoretical representation of P>99.9% of the M. paratuberculosis genome. This library was screened with a 0.7-kb radiolabeled fragment of M. avium MAC101 sodA gene by colony hybridization [11].

2.5Southern hybridization

Genomic DNA from strain K-10 was digested separately with BamHI, EcoRI or PstI, separated on a 0.8% agarose gel, transferred onto a nitrocellulose membrane, and hybridized with the 0.7-kb probe containing the M. avium sodA gene. The membrane was washed under stringent conditions as previously described [11], and positive hybridization bands were visualized by autoradiography. These images were digitally captured and processed for enhancement of brightness and contrast.

2.6PCR amplification

The oligonucleotide primers used for PCR amplification of sodA genes were MuSODL (5′-ATCTTCCTGAACGAAAAGAACCT-3′) and MuSODR (5′-CCAGTTGACCACGTTCCAGA-3′). PCR amplifications were performed in a 50-μl reaction volume containing 0.25 u Taq DNA polymerase (Fisher Scientific), 10 mM Tris–HCl (pH 8.3), 2.5 mM MgCl2, 0.2 mM deoxynucleoside triphosphate, 0.1 mg of gelatin ml−1, 0.2 mM spermidine and 10% dimethyl sulfoxide. Thermocycling was performed using a Perkin–Elmer Gene Amp 9600 thermocycler (Roche Molecular Systems) and using the following conditions: 95°C for 4 min, then denaturation at 95°C for 30 s, annealing at 55°C for 30 s, and elongation at 72°C for 45 s (30 cycles), and a final elongation step of 72°C for 7 min. The amplified products were purified using QIAquick PCR purification kit (Qiagen) and subjected to DNA sequencing procedure.

2.7DNA sequence analysis

DNA sequencing was performed by the fluorescent dideoxy chain terminator method. The sequence of the sodA gene and flanking regions reported herein is under the GenBank accession number AF180816. Theoretical calculation of the Sod isoelectric point was performed using the GCG Wisconsin Sequence Analysis program (version 10.1, Genetics Computer Group).

2.8Western immunoblotting

After separation of the samples by SDS–polyacrylamide gel electrophoresis (SDS–PAGE), the gel was transblotted to a nitrocellulose membrane that was blocked overnight with PBS containing 0.05% Tween 20 and 3% dry milk. Primary monoclonal antibodies IT-13 (anti-65-kDa heat shock protein, anti-HSP65, [12]) and IT-17 (anti-Sod, [13]) were applied at dilutions of 1:25 and 1:1000, respectively, to the membrane for 1 h at room temperature. After washing with PBS–0.05% Tween 20, the membrane was incubated with goat anti-mouse IgG at 1:200 (IT-13) or 1:500 (IT-17), and visualized using the ECL chemiluminescent system (Amersham-Pharmacia Biotech). Densitometric quantification of band intensities was performed by capturing the images from the autoradiogram. Images were again processed only to enhance brightness and contrast. The results of this analysis are reported as an average (±S.D.) of three independent measurements.

3Results and discussion

3.1Analysis of M. paratuberculosis proteins by 2D gel electrophoresis

To analyze the M. paratuberculosis protein profile, bacteria were grown to early exponential phase, harvested, and protein extracts were prepared by sonic disruption. Proteins in these extracts were separated by 2D gel electrophoresis. The cytosolic fraction was resolved into approximately 150 spots as determined by Coomassie blue staining, and 15 prevailing proteins were identified (Fig. 1). The M. tuberculosis Sod has been described as a protein with a molecular mass of 23 kDa and a pI of 6.0 [14]. In the gel presented in Fig. 1, we detected two fused spots located at this approximate position. This is consistent with the observation that the Sod from M. tuberculosis may yield multiple spots in 2D gels [15]. To perform N-terminal amino acid sequencing, another 2D gel was developed under identical conditions and transferred to a PVDF membrane. This membrane was stained with Coomassie blue and the region corresponding to the fused spots was excised and sequenced. The resulting sequence (AEYTLPDLDW) was identical to that from the M. avium Sod [16], confirming that this protein is the M. paratuberculosis Sod.

Figure 1.

2D gel electrophoresis of M. paratuberculosis cellular proteins. Approximately 250 μg of total proteins was resolved by 2D gel electrophoresis and stained with Coomassie blue. 1.0 μg of Tropomyosis was added to the sample as a standard, and is seen as two polypeptides of similar isoelectric points. The lower spot of 33 kDa and pI 5.2 is indicated by an arrow. Increasing isoelectric points are from left to right. Molecular mass size markers are shown on the left. The two fused spots corresponding to the Sod protein (23 kDa, pI=6.1) are bounded by a square.

3.2M. paratuberculosis genome possesses one copy of the sodA gene

To determine the copy number of the M. paratuberculosis sodA gene, genomic DNA from M. paratuberculosis K-10 was digested with either BamHI or EcoRI, which do not cut within the sodA gene. Southern hybridization using a 0.7-kb fragment of the M. avium sodA gene as a probe yielded single bands of ca. 20 and 4.0 kb, respectively (Fig. 2, lanes 1 and 2). Conversely, digestion of genomic DNA with PstI, which cuts three times within the sodA gene, resulted in hybridization bands of 1.8 and 4.5 kb, and two bands of less than 0.2 kb in size (Fig. 2, lane 3), confirming the presence of a single sodA gene in M. paratuberculosis.

Figure 2.

Southern hybridization. Genomic DNA of M. paratuberculosis K-10 was digested with BamHI (lane 1), EcoRI (lane 2) or PstI (lane 3) and hybridized using a 0.7-kb fragment of the M. avium sodA as a probe. Molecular size standards in kb are indicated on the right.

3.3Cloning and sequence analysis of the M. paratuberculosis sodA gene

A M. paratuberculosis K-10 genomic library was screened by colony hybridization, yielding three positive clones. One of the recombinant plasmids (pBUN151) carrying a 3-kb insert was chosen for DNA sequencing. Sequence analysis of the insert revealed two ORFs in the same transcriptional direction: the first ORF encodes the Sod protein and the second ORF a protein homologous to the M. tuberculosis Rv3847 [17]. The genetic organization of these ORFs is similar to those of M. tuberculosis[17] and Mycobacterium leprae[18]. The putative start codon of the M. paratuberculosis sodA gene is GTG (Fig. 3) and it encodes a 23-kDa protein of 207 amino acids with a theoretical isoelectric point of 5.97. This matches the data obtained by 2D gel electrophoretic analysis. The derived N-terminal amino acid sequence was identical to the amino acid sequence after removal of the N-terminal methionine. The putative ATG start codon of the second ORF is located 158 bp downstream from the sodA stop codon. A putative transcription terminator is between the two ORFs, suggesting the second ORF corresponds to a separate transcriptional unit.

Figure 3.

DNA sequence and analysis of M. paratuberculosis sodA. The putative ribosome-binding site coding sequence is underlined, and the start (GTG) and stop (TGA, indicated with an asterisk) codons are in boldface type. Nucleotides in the M. paratuberculosis sequence differing from the M. avium sequence (indicated above the M. paratuberculosis sequence) are italicized and underlined. The PstI restriction sites are indicated in boldface type. The characteristic Val133 in the M. paratuberculosis Sod amino acid sequence is in boldface type. The inverted arrows show a putative transcriptional terminator.

Subsequently, sodA genes from five different M. paratuberculosis strains, K-39, K-48, S-8, Lorraine, ATCC 19698, and the M. avium strain MAC104, were PCR-amplified and sequenced. Analysis of sodA genes from six M. paratuberculosis, the M. avium strains MAC104 and TMC 724 (GenBank accession number U11550), revealed six nucleotide differences in the coding sequence between M. paratuberculosis and M. avium (Fig. 3). The sodA genes from all six M. paratuberculosis strains were identical to each other; likewise, the two M. avium sodA genes were identical. The nucleotide differences noted above determine a restriction site polymorphism in the sodA gene, in which a unique AgeI site is present in the M. avium sequence but is absent from M. paratuberculosis. In addition, all M. paratuberculosis Sod proteins display a valine residue at position 133, while leucine is present in the M. avium counterpart.

The amino acid sequence of Sod is highly conserved among mycobacteria [6,14,16]. The M. paratuberculosis Sod has 82% identity and 90% similarity to the Sod from M. tuberculosis, and a 90% identity and 94% similarity to the Sod from M. leprae[19]. This high degree of homology suggests the potential for cross-reactivity in antibody-based tests, thus precluding the use of this antigen diagnostically. However, a diagnostic assay could be developed based on the AgeI polymorphism described above. A similar PCR-based test has been described for the M. paratuberculosis hsp65 gene [20]. The M. paratuberculosis Sod amino acid sequence displays putative metal cofactor binding sites characterized by the following amino acids: His28, His76, Asp160, and His164[16]. Furthermore, the amino acid residues at positions 71–74, 78, 79, 82, 87, 146, 147, 166, and 177 are characteristic of a Mn-dependent Sod [16,21].

3.4Distribution of Sod in culture supernatant fluids and subcellular fractions of M. paratuberculosis

Sods from M. tuberculosis and M. avium are secreted [14,16]. To investigate whether the M. paratuberculosis Sod is also secreted, protein extracts representing cytosolic, cell wall/cell membrane and culture supernatant fractions were analyzed by Western immunoblotting using the broadly reactive monoclonal antibody IT-17 (Fig. 4). The subcellular fractions loaded on SDS–PAGE gels were generated from an equal number of cells to determine the relative amounts of Sod protein present in each fraction. Using densitometric analysis, we found that 60.7±2.5% of the M. paratuberculosis Sod is cytosolic (Fig. 4, panels 4.1 and 4.2, lane C), 9.5±5.2% is present in the membrane and cell wall extract (Fig. 4, panels 4.1 and 4.2, lane M), and approximately 29.7±3.2% is found in the culture supernatant fluids (Fig. 4, panels 4.1 and 4.2, lane S). To rule out autolysis as a source of the extracellular Sod, the HSP65, which is commonly used as a marker for autolysis [22], was also measured using the monoclonal antibody IT-13 [12]. No HSP65 protein was identified in the culture supernatant fluids using this antibody (Fig. 4, panel 4.3). Furthermore, culture supernatant fluids did not have any activity for the cytoplasmic enzyme lactate dehydrogenase (data not shown), similar to results for M. tuberculosis[6]. Thus, we concluded that the M. paratuberculosis Sod is secreted into the culture supernatant fluid.

Figure 4.

Western immunoblot analysis of M. paratuberculosis subcellular protein fractions. Proteins were resolved by 10% SDS–PAGE gels, blotted onto a nitrocellulose membrane and immunostained with antibody against Sod or HSP65. Amounts of subcellular fractions loaded were generated from equal numbers of cells. Panel 4.1: the entire membrane immunostained with anti-Sod antibody. Panel 4.2: (Sod) shows the relative levels of Sod in the different subcellular fractions. Panel 4.3: (HSP65) shows that the control protein is not found in the supernatant. Lane C, cytosolic extract; lane M, cell membrane/cell wall extract; lane S, culture supernatant fluid. Molecular mass standards in kDa are shown on the left.

All secreted Sods have no leader peptide sequences, suggesting that an internal signal sequence may direct the secretion process. Recombinant M. smegmatis carrying the Sod from either M. tuberculosis or M. avium secretes this protein [6,16]. In contrast, recombinant E. coli cannot secrete the Sod, suggesting that the export of Sod is dependent on the specific export machinery of mycobacteria [6]. Alternatively, the level of secretion may be directly correlated with its expression level. It appears that secretion of Sod can be correlated with mycobacterial pathogenesis. M. paratuberculosis produces approximately 0.6 fg of Sod per cell, based on a similar analysis of Sod from M. tuberculosis and M. smegmatis[6]. This amount is five-fold less than that produced by M. tuberculosis, but 20-fold more than the amount expressed by non-pathogenic M. smegmatis[6]. Considering both the expression and secretion levels, M. paratuberculosis secretes approximately 0.2 fg of Sod per cell, which is 30-fold more than the amount secreted by M. smegmatis and about 10% of the amount secreted by M. tuberculosis[6].

In summary, our results demonstrating that M. paratuberculosis secretes Sod in significant amounts support the hypothesis that only pathogenic mycobacteria exhibit this property. Furthermore, the M. tuberculosis-secreted Sod was shown to stimulate protective cell-mediated immunity in guinea pigs [23]. The secreted M. paratuberculosis Sod may also elicit a protective immune response in the host that may be exploited for future development of subunit vaccines against Johne's disease.


We would like to thank Dr. John T. Belisle (Colorado State University) for kindly providing the antibodies IT-13 and IT-17. We also thank Dr. L.E.M. Bermudez and Dr. F. Sangari (Kuzell Institute) for providing the recombinant plasmid carrying the M. avium sodA gene. We thank Dr. Gautam Sarath at the University of Nebraska–Lincoln Protein Core Facility for performing N-terminal amino acid sequencing. This study, filed as Journal No. 13168 of the Nebraska Agricultural Experiment Station, was supported by Grants BARD-USDA #IS-2564-95C, NRI CGP-USDA # 1999-02316, and NIH R01 AI43199 (J.D.C.). Z.F. is a recipient of a Maude Hammond Fling-Bukey Memorial Fund Fellowship from Graduate Studies, University of Nebraska-Lincoln.