Rhodococcus equi secreted antigens are immunogenic and stimulate a type 1 recall response in the lungs of horses immune to R. equi infection

Abstract

Rhodococcus equi is an opportunistic pathogen in immunocompromised humans and an important primary pathogen in young horses. Although R. equi infection can produce life-threatening pyogranulomatous pneumonia, most foals develop a protective immune response that lasts throughout life. The antigen targets of this protective response are currently unknown; however, Mycobacterium tuberculosis is a closely related intracellular pathogen and provides a model system. Based on previous studies of M. tuberculosis protective antigens released into culture filtrate supernatant (CFS), a bacterial growth system was developed for obtaining R. equi CFS antigens. Potential immunogens for prevention of equine rhodococcal pneumonia were identified by using immunoblots. The 48-h CFS contained five virulence-associated protein bands that migrated between 12 and 24 kDa and were recognized by sera from R. equi-infected foals and immune adult horses. Notably, the CFS contained the previously characterized proteins VapC, VapD, and VapE, which are encoded by genes on the R. equi virulence plasmid. R. equi CFS was also examined for the ability to stimulate a type 1-like memory response in immune horses. Three adult horses were challenged with virulent R. equi, and cells from the bronchoalveolar lavage fluid were recovered before and 1 week after challenge. In vitro stimulation of pulmonary T-lymphocytes with R. equi CFS resulted in significant proliferation and a significant increase in gamma interferon mRNA expression 1 week after challenge. These results were consistent with a memory effector response in immune adult horses and provide evidence that R. equi CFS proteins are antigen targets in the immunoprotective response against R. equi infection.

FIG. 1.

Initial characterization of R. equi CFS. R. equi ATCC 33701 was grown in DMEM-BHI for 4 to 144 h. Bacterial growth was determined by measuring the absorbance at 600 nm with a spectrophotometer. The accumulation of protein in CFS was measured by a BCA protein assay. The background protein content at time zero was subtracted from the CFS protein content at each time point.

FIG. 2.

Detection of immunogenic proteins in the CFS at three time points. Immunoblots of CFS and bacterial pellets from R. equi ATCC 33701 grown in DMEM-BHI for 4, 48, and 144 h were probed with serum from a foal with active rhodococcal pneumonia (1:800 dilution). Lanes were loaded with virulent R. equi ATCC 33701, unless indicated otherwise, as follows: lane 1, SRA (positive control); lane 2, SRA of plasmid-cured strain R. equi ATCC 33701 PC; lane 3, DMEM-BHI (medium control); lane 4, 4-h CFS; lane 5, 4-h pellet; lane 6, 48-h CFS; lane 7, 48-h pellet; lane 8, 144-h CFS; lane 9, 144-h pellet; lane 10, 144-h pellet diluted 1:10. All lanes containing CFS were loaded with 153 μg of protein. Lanes containing pellets were loaded by volume as follows: 15 μl for 4- and 48-h pellet antigens and 5 μl for 144-h pellet antigen (undiluted). The arrow on the right indicates the position of the characteristic doublet attributable to VapA. The numbers on the left indicate molecular masses (in kilodaltons).

FIG. 3.

Identification of proteins in 48-h CFS by using sera from R. equi-infected foals and immune adult horses. Immunoblots of CFS and bacterial pellets of R. equi ATCC 33701 grown in DMEM-BHI for 48 h were probed with sera from three foals infected with R. equi (A) or sera from three adult horses collected 2 weeks after intrabronchial challenge with R. equi ATCC 33701 (B). Lanes were loaded with 195 μg of a 48-h CFS or 3 μl of a 48-h pellet (P), as indicated. The antibody dilutions were as follows: foal 136 (#1), 1:800; foal CA11 (#2), 1:500; foal 151 (#3), 1:800; horse N124 (#4), 1:1,000; horse 20 (#5), 1:800; and horse 02 (#6), 1:800. The numbers on the left indicate molecular masses (in kilodaltons).

FIG. 4.

Immunogenic CFS proteins in R. equi lacking the virulence plasmid: immunoblot of R. equi ATCC 33701 CFS (+) and plasmid-cured R. equi ATCC 33701 PC CFS (PC) grown in DMEM-BHI for 48 h and probed with sera from three immune adult horses (as described in the legend to Fig. 3). Each lane was loaded with 195 μg of protein. The numbers on the left indicate molecular masses (in kilodaltons).

FIG. 5.

Identification of plasmid-encoded Vap proteins in R. equi CFS: immunoblots of R. equi CFS grown in DMEM-BHI for 48 h. (A) The anti-VapA monoclonal antibody reacted with VapA in SRA and the pellet antigen and six protein bands in the CFS antigen. The membrane was probed with 10G5, a monoclonal antibody against VapA. Lanes were loaded as follows: lane 1, SRA from strain ATCC 33701 (positive control); lane 2, SRA of plasmid-cured R. equi ATCC 33701 PC; lane 3, DMEM-BHI (medium control); lane 4, 48-h ATCC 33701 CFS; lane 5, 48-h ATCC 33701 bacterial pellet. The arrow indicates the position of the characteristic VapA doublet in the pellet. (B) Antibodies in sera from an infected foal recognize proteins that comigrate with VapC, VapE (arrows), and possibly VapD. CFS antigen was probed with monospecific antisera against VapC (1:100 dilution), VapD (1:250 dilution), and VapE (1:800 dilution) or with serum from an R. equi-infected foal (1:800 dilution). The nitrocellulose membrane was cut into five sections; four sections are shown, and the fifth section was probed with normal rabbit serum and did not react with any bands (data not shown). The numbers on the left indicate molecular masses (in kilodaltons).

FIG. 6.

Lymphoproliferation of BALF cells in response to R. equi CFS 1 week after challenge with R. equi ATCC 33701. BALF cells were cultured for 5 days before and after challenge and either not stimulated (n/s) or stimulated with 10 μg of DMEM-BHI per ml, C. pseudotuberculosis CFS (Cpstb CFS), R. equi CFS (Requi CFS), or SRA (Requi SRA). The data are the means for BALF cell cultures from all three horses. The error bars indicate the standard errors of the means. (A) Day 0; (B) day 7 after challenge. An asterisk indicates a significant difference (P < 0.05) compared to the values for the controls (not stimulated, DMEM-BHI, and C. pseudotuberculosis CFS).

FIG. 7.

R. equi CFS-specific IFN-γ expression by BALF cells 1 week after challenge with R. equi ATCC 33701. The data are mean normalized levels of IFN-γ expression from each of three horses: horse 71 (A), horse 44 (B), and horse 02 (C). BALF cells were not stimulated (n/s) or were stimulated for 24 h with 10 μg of DMEM-BHI per ml, C. pseudotuberculosis CFS (Cpstb CFS), R. equi CFS (Requi CFS), or SRA (Requi SRA) before (day 0) and after (day 7) intrabronchial challenge. An asterisk indicates a significant difference (P < 0.05) compared to the values for the controls (not stimulated, DMEM-BHI, and C. pseudotuberculosis CFS) at the same time.