Rhodococcus equi-Specific Cytotoxic T Lymphocytes in Immune Horses and Development in Asymptomatic Foals
http://www.100md.com
感染与免疫杂志 2005年第4期
Department of Veterinary Microbiology and Pathology
Department of Veterinary Clinical Sciences, Washington State University, Pullman, Washington
ABSTRACT
Rhodococcus equi is an important cause of pneumonia in young horses; however, adult horses are immune due to their ability to mount protective recall responses. In this study, the hypothesis that R. equi-specific cytotoxic T lymphocytes (CTL) are present in the lung of immune horses was tested. Bronchoalveolar lavage (BAL)-derived pulmonary T lymphocytes stimulated with R. equi lysed infected alveolar macrophages and peripheral blood adherent cells (PBAC). As with CTL obtained from the blood, killing of R. equi-infected targets by pulmonary effectors was not restricted by equine lymphocyte alloantigen-A (ELA-A; classical major histocompatibility complex class I), suggesting a novel or nonclassical method of antigen presentation. To determine whether or not CTL activity coincided with the age-associated susceptibility to rhodococcal pneumonia, CTL were evaluated in foals. R. equi-stimulated peripheral blood mononuclear cells (PBMC) from 3-week-old foals were unable to lyse either autologous perinatal or mismatched adult PBAC targets. The defect was not with the perinatal targets, as adult CTL effectors efficiently killed infected targets from 3-week-old foals. In contrast, significant CTL activity was present in three of five foals at 6 weeks of age, and significant specific lysis was induced by PBMC from all foals at 8 weeks of age. As with adults, lysis was ELA-A unrestricted. Two previously described monoclonal antibodies, BCD1b3 and CD1F2/1B12.1, were used to examine the expression of CD1, a nonclassical antigen-presenting molecule, on CTL targets. These antibodies cross-reacted with both foal and adult PBAC. However, neither antibody bound alveolar macrophages, suggesting that the R. equi-specific, major histocompatibility complex-unrestricted lysis is not restricted by a surface molecule identified by these antibodies.
INTRODUCTION
Rhodococcus equi is a facultative intracellular bacterium with properties similar to those of Mycobacterium tuberculosis, a closely related pathogen. Both R. equi and M. tuberculosis are nocardioform actinomycetes, survive within the phagosomes of macrophages, have characteristic cell walls containing unique lipids, lipoproteins, and glycolipids, and produce pyogranulomatous lesions in the lung (16). Whereas M. tuberculosis produces tuberculosis in humans, R. equi is a common cause of life-threatening pneumonia in young horses. R. equi is also an opportunistic pathogen in AIDS patients (29).
R. equi is ubiquitous in the equine environment, and virtually all horses are considered exposed, likely in the first few weeks of life. Horses have a distinct age-associated susceptibility, such that equine rhodococcal pneumonia is almost exclusively a disease of foals less than 4 to 5 months of age. Even on farms where infection is endemic, a situation where a significant percentage of foals develop pneumonia and the environmental load of virulent R. equi is high, older horses do not manifest with disease. Previous research has provided strong evidence that the resistance of adult horses to R. equi challenge reflects acquired immunity and their ability to mount effective recall responses (18, 19). Therefore, immune adult horses, like humans who have successfully contained M. tuberculosis infection, provide a valuable system for understanding the mechanisms by which infection can be controlled and pneumonia prevented. Knowledge of these mechanisms will likely be critical to efforts to design an effective vaccine.
Rhodococcal pneumonia is an immunopathologic disease. Adoptive transfer and neutralizing antibody studies have shown that mice that develop Th2 responses, characterized by production of interleukin-4 (IL-4) rather than gamma interferon (IFN-), develop characteristic pulmonary lesions (23). In contrast, mice responding to infection with a Th1 response and IFN- production clear a virulent challenge. As with M. tuberculosis, CD8+ T lymphocytes can also contribute to immune clearance of R. equi (31). Even transgenic mice that lack major histocompatibility complex (MHC) class II-restricted CD4+ T lymphocytes significantly reduced R. equi numbers following pulmonary challenge (24).
The immune responses most relevant to protective immunity likely occur in the lung. For M. tuberculosis, both the nature of the response and the repertoire of antigens recognized can be different in comparisons of pulmonary responses to peripheral blood responses. Likewise, the outcome of host cell infection may be different in comparisons of alveolar macrophages to blood-derived monocytes. For example, human alveolar macrophages infected with M. tuberculosis are more resistant to cytotoxic T-lymphocyte (CTL)-mediated lysis than infected monocytes (42). When pulmonary responses to R. equi are examined via bronchoalveolar lavage (BAL), immune adult horses clear virulent organisms in association with increased numbers of CD4+ and CD8+ lymphocytes at the site of challenge (18). Although pulmonary T lymphocytes from most adult horses initially have low proliferative responses to R. equi antigen, challenge induces rapid, strong responses typical of a secondary immune response. Importantly, this recall response includes production of IFN- and is likely responsible for the immunity seen in adult horses (19, 27). As described with other pulmonary pathogens, the pulmonary responses to R. equi challenge are relatively compartmentalized. Responses are more marked, and the sensitivity to antigen dose is increased at the site of infection. Blood, including peripheral blood mononuclear cells (PBMC), has been an insensitive indicator of local pulmonary responses to R. equi (18).
More recently, research has examined the role of CTL in the immune control of R. equi (32). Equine PBMC stimulated with antigen-presenting cells (APC) either infected with R. equi or exposed to soluble R. equi antigen (SRA) lysed R. equi-infected target cells but not uninfected targets. Killing of R. equi-infected peripheral blood adherent cells (PBAC) by effector cells was equally effective against autologous and equine lymphocyte alloantigen-A (ELA-A) (classical MHC class I) mismatched targets. Lysis was decreased to background by depletion of either CD2+ or CD3+ cells, indicating that the effector cell had a T-lymphocyte but not an NK-cell phenotype. Depletion of CD8+ T cells also resulted in significantly decreased lysis of infected targets. These data indicate that immunocompetent adult horses have R. equi-specific CD8+ CTL, which may play a role in immunity. The apparent lack of restriction via classical MHC class I molecules suggests a novel method of antigen processing and presentation, such as presentation by CD1 or other nonclassical MHC molecules.
The goal of the experiments reported here was to further examine the relevance of R. equi-specific CTL to protective immunity. We first asked whether the CTL demonstrated in the peripheral blood of immune adult horses also occur in the lung. Next, we tested the hypothesis that susceptible perinatal foals are deficient in CTL that lyse R. equi-infected cells.
MATERIALS AND METHODS
Horses. Two healthy adult horses, one Quarterhorse (H71) and one Hanoverian (H68), presently maintained in the Washington State University (WSU) Veterinary Teaching Hospital herd, were selected based on mismatched ELA-A haplotypes and previous demonstration of CTL activity against R. equi-infected targets (32). For the experiments involving foals, six Arabian foals were selected based on age. Using previously described serologic reagents, ELA-A were determined by microcytotoxicity assay (43) (Table 1). Venous blood was collected from the jugular vein of each horse by using evacuated containers (Baxter, Deerfield, Ill.) containing 16 ml of anticoagulant citrate dextrose (ACD)/100 ml of blood. PBMC were isolated from venous blood using a Ficoll-Hypaque technique (43). PBMC were then used either as effector cells or for isolation of PBAC targets (see below).
Bacteria. The virulent R. equi strain ATCC 33701 (PL+), which contains the 80.6-kb virulence-associated plasmid and expresses the virulence-associated protein VapA, was utilized in all experiments. Bacteria were stored at –20°C as frozen stabilates and were reconstituted prior to each experiment. Bacteria were initially grown on brain heart infusion (BHI) agar and then cultivated overnight to log phase in BHI broth at 37°C in a shaking incubator at 200 rpm. After being washed once with phosphate-buffered saline (PBS), the bacteria were adjusted to 107/ml in antibiotic-free complete medium, which consisted of RPMI 1640 supplemented with 2 mM L-glutamine, 0.05 μM 2 mercaptoethanol, 6.25 mM HEPES, and 10% normal horse serum (NHS; Gibco, Grand Island, N.Y.). The adjusted bacterial concentration was initially confirmed by plating serial dilutions on BHI plates and calculating the CFU per milliliter.
Bronchoalveolar lavage and cell preparation. Horses were sedated with xylazine at 0.6 mg/kg of body weight. A Bivona BAL tube was inserted through the right nostril into the trachea and wedged in place. To collect BAL cells from adult horses, three separate aliquots of 180 ml of 0.9% NaCl containing 0.84% sodium bicarbonate (BAL wash) were instilled into the lung. Following each aliquot, fluid was aspirated. Fluid was placed into 50-ml conical tubes and centrifuged for 8 min at 600 x g at 4°C. Pellets were consolidated and washed three times with complete medium containing decreasing amounts of antibiotic/antimycotic (wash 1, 1 μg of amphotericin/ml [Gibco]; wash 2, 2 U of penicillin/ml, 2 U of streptomycin/ml, and 0.1 μg of amphotericin/ml [Gibco]; wash 3, antibiotic/antimycotic free). Resultant cells were plated at 4 x 106 cells/ml on 150-cm2 collagen/serum-coated petri dishes and incubated for 1 h at 37°C with 5% CO2 as previously described (32). Following incubation, nonadherent cells were collected by vigorous pipetting of the supernatant 10 times and washing again with PBS via pipetting 10 ml of PBS over the plate 10 times. Nonadherent cells (64% ± 29.6% CD2+ T lymphocytes and 35% ± 19.5% immunoglobulin M [IgM] positive) were plated at 4 x 106 cells/ml in 75-cm2 vented flasks for stimulation and used as effector cells. Adherent cells were used as targets following harvesting as described below.
Effector cell preparation and stimulation. PBMC or BAL cells in 25 ml of antibiotic-free complete medium were plated in 75-cm2 flasks at 4 x 106 cells/ml. The cells were stimulated with 1 multiplicity of infection (MOI) of live virulent R. equi 33701 PL+. Following a 1-h incubation at 37°C with 5% CO2, 1 μg of gentamicin/ml was added to kill extracellular bacteria. Culture flasks were then incubated for 5 days at 37°C with 5% CO2 and then rested without antigenic stimulation for 2 days prior to further use. Resting consisted of washing the cells twice with antibiotic-free complete medium, counting, and plating in fresh medium containing 1 μg of gentamicin/ml. Following stimulation and resting, cells were washed once, counted, and then characterized for surface marker expression by flow cytometry. Alternatively, the cells were used as effector cells in the CTL assay described below.
Target cell isolation. Autologous and ELA-A-mismatched PBAC were harvested from PBMC for use as target cells as previously described (33). Briefly, PBMC were plated on the gelatin/serum-coated dishes at a density of 1.5 x 108 cells/dish in 20 ml of antibiotic-free complete medium. Cells were incubated for 15 to 19 h at 37°C with 5% CO2. Following incubation, nonadherent cells were resuspended in the culture medium by repeated pipetting (10x) and then removed by washing each plate with 10 ml of PBS and repeated pipetting (10x). Adherent cells were eluted with 20 ml of antibiotic-free complete medium containing 20% (vol/vol) NHS mixed 1:1 with 10 mM EDTA for 35 min at 37°C with 5% CO2. The cells were then collected by vigorous pipetting of the eluting fluid 10 times followed by washing each dish twice with 10 ml of PBS per wash. Eluted cells were combined, washed once with antibiotic-free complete medium, counted, and added to 96-well plates at 105 cells/well in 100 μl of medium. Immediately prior to adding cells, 20 μl of commercial NHS was added to each well.
Adherent BAL cells were removed by covering the plates with 10 ml of Accutase (Sigma, St. Louis, Mo.) and incubating them for 10 min at 37°C with 5% CO2. The cells were then collected by repeated pipetting of the suspension over the plate 10 times followed by washing with 10 ml of PBS 10 times. Cells were consolidated via centrifugation at 600 x g at 4°C for 15 min, washed once with complete medium via the same centrifugation, and then counted and plated in 100 μl of complete medium on a serum-coated 96-well plate at 105 cells/well. The cells were incubated for 1 h prior to use as lung-derived target cells in the CTL assay (see below).
Cytotoxicity assay. After 1 h, each well of target cells was labeled with 50 μl of antibiotic-free complete medium containing 100 μCi of 51Cr per ml. Following 13 h of incubation, the targets were infected with 5 MOI of live R. equi bacteria by adding 50 μl of antibiotic-free complete medium containing 107 live R. equi cells per ml. One hour postinfection, 20 μl of complete medium containing 1 μg of gentamicin sulfate/ml was added to each culture well to kill any extracellular bacteria. Following an additional 7.5 h of incubation, the cells were washed four times with complete medium containing 1 μg of gentamicin sulfate/ml to remove extracellular 51Cr. Effector cells were then added to target cells and incubated for an additional 4.5 h at 37°C with 5% CO2. Following incubation, the plates were centrifuged at 500 x g for 5 min to pellet cells. One-hundred microliters of supernatant was removed, and the amount of 51Cr was measured using a MicroBeta plate reader. The formula for the percent specific lysis is [(E – S)/(M – S)] x 100, where E is the mean of three test wells, S is the mean spontaneous release from three target cell wells without effector cells, and M is the mean maximal release from three target cell wells with 3% Triton X-100 (48). As previously reported, significant lysis of infected cells was defined as 3 standard errors above the uninfected control target cells (43).
R. equi CTL responses in foals. To characterize the R. equi-specific immune response of foals in the first 2 months of life, PBMC were collected from six foals at 3, 6, and 8 weeks of age and used as effectors in CTL assays. PBMC from foals A2197, A2198, A2201, and A2203 were collected at 3 weeks of age. To ensure that a CTL response was induced, foals A2196 and A2197 were infected intrabronchially at 6 weeks of age with a small dose (104) of virulent R. equi. Following sedation (see the BAL protocol described above), bacteria in 1 ml of BAL wash was instilled into the BAL tube, followed by 10 ml of the BAL wash and 20 ml of air. PBMC were collected just prior to infection (at the 6-week-old time point) and then again 2 weeks postinfection (at 8 weeks of age). The remaining foals were housed with their dams and presumably were naturally exposed to R. equi in their environment (41). PBMC were also collected from the naturally exposed foals at 6 and 8 weeks of age. PBAC target cells from all foals were obtained 5 days after each effector cell collection date. Clinical parameters, including temperature, heart rate, and respiratory rate, of infected foals were monitored twice daily for 3 weeks. Lungs were simultaneously ausculted. Complete blood counts (CBC) plus fibrinogen, performed twice weekly, remained within normal limits for 3 weeks postinfection. Inoculated foals showed no signs of disease, except for mild pyrexia (less than 102.4°C) of 3 days' duration at 2 weeks postinfection.
Flow cytometry. Before and after R. equi stimulation, effector cells were labeled with a 15-μg/ml concentration of either anti-equine CD8 monoclonal antibody (MAb) HT14A, anti-equine CD4 MAb HB61A, anti-equine CD2 MAb HB88A, anti-equine IgG MAb1.9/3.2 (B-cell marker), anti-equine CD172a MAb DH59B (5) (granulocyte/monocyte marker), or isotype control antibody (COLIS69A) (control for nonspecific binding) (VMRD, Inc., Pullman, Wash.). Target cells were labeled with a 15-μg/ml concentration of anti-human CD1b MAb BCD1b3 subclone 1.3.1.6 (BCD1b3), 1:100 dilution of anti-guinea pig CD1 MAb CD1F2/1B12.1 (CD1F2; both CD1 antibodies graciously provided by S. Porcelli, Bronx, N.Y.), anti-equine CD172a MAb DH59B, or isotype control antibody (COLIS69A). Additional anti-CD1 antibodies, anti-guinea pig CD1 MAb CD1F2/5E3 (11, 20) (S. Porcelli), anti-guinea pig CD1 MAb CD1F2/6B5 (11, 20) (S. Porcelli, Bronx, NY), anti-bovine CD1b MAb CC-20 (28, 36) (C. J. Howard, Institute of Animal Health, Compton, Newbury, Berkshire, United Kingdom), anti-bovine CD1b TH97A (VMRD, Inc.), anti-ovine CD1b MAb 20-27 (11, 20, 28) (University of Melbourne, Center of Animal Biotechnology, Victoria, Australia), and anti-human CD1a MAb HI149 (17) (Biolegend, San Diego, Calif.) were also tested at previously reported concentrations. The mixtures were incubated on ice for 15 min at 4°C, washed three times, and pelleted at 4°C for 3 min at 500 x g. Cells were then suspended in 50 μl of (5 μg/ml) fluorescein-conjugated anti-mouse immunoglobulin serum absorbed with normal human and horse sera. Cells were next incubated for 15 min at 4°C in the dark, washed two times, and suspended in 200 μl of 2% formaldehyde in PBS (12). Labeled cells were analyzed with a FACScan equipped with a Macintosh computer and Cell Quest software (Becton Dickinson Immunocytometry Systems, San Jose, Calif.). The effector cell population was analyzed before stimulation and immediately prior to testing in the CTL assay. The values were compared using a paired t test, with P < 0.05 considered significant.
RESULTS
Pulmonary CTL responses. In previous studies, PBMC were shown to lyse R. equi-infected PBAC in a classical MHC class I-unrestricted manner (32). To determine whether a similar pattern of lysis would occur using cells from the lung (the target organ of R. equi infection), macrophages were collected from the lungs of immune adult horses. In three replicate trials using two horses, H71 and H68, PBMC effector cells stimulated with live virulent R. equi significantly lysed autologous and mismatched targets derived from the blood (Fig. 1A) and the lung (Fig. 1B) compared to that of uninfected targets. Thus, significant lysis of infected targets occurred regardless of tissue origin or MHC haplotype of the targets (see columns with asterisks in Fig. 1). At the optimal E:T (effector:target) ratios, the degree of lysis appeared similar for both BAL and PBAC mismatched and autologous targets, although a statistical comparison was not done.
In addition, pulmonary effectors were compared to blood-derived effectors (Fig. 2). As previously reported, lymphocyte numbers in BAL fluid of unstimulated immune horses were low (18), thus limiting the number of effectors that were available for evaluation (3.8 x 107 ± 2.1 x 107 lymphocytes before stimulation). At an E:T ratio of 1:1, both R. equi-stimulated BAL effectors and R. equi-stimulated PBMC effectors caused significant lysis of infected autologous and mismatched PBAC targets (Fig. 2A). Similarly, both R. equi-stimulated effectors caused significant lysis of infected autologous and mismatched BAL targets derived from the lung (Fig. 2B). Therefore, effectors from both the blood and lung significantly lysed mismatched targets from the blood and lung. The degree of lysis of the two target cell populations (lung versus blood derived) appeared similar. PBMC cultured for 7 days with no antigen (PBMC-NA) did not induce significant lysis of infected PBAC or BAL targets, thus, R. equi stimulation was required to induce PBMC effectors to lyse infected targets, regardless of tissue origin.
R. equi-specific CTL responses in perinatal foals. To test the hypothesis that perinatal foals are deficient in R. equi-specific CTL responses, PBMC isolated from four 3-week-old foals, A2197, A2198, A2201, and A2203, were stimulated for 7 days with 1 MOI of live virulent R. equi. In the initial cytotoxicity trial using A2197, a maximum E:T ratio of 64:1 was achieved. In all subsequent trials, where poststimulation lymphocyte numbers were lower, an E:T ratio of 27:1 was used. In all foals, increasing the E:T ratio from 1:1 to either 27:1 or 64:1 did not result in a significant percentage of specific lysis mediated by R. equi-stimulated foal PBMC (Fig. 3B). In all of the foals tested, the percentage of specific lysis was not significantly greater than that of the uninfected controls (Fig. 3B). This was in sharp contrast to the results obtained for simultaneously in vitro-stimulated adult H71 PBMC controls (Fig. 3A), which mediated lysis of 13 to 35% at an E:T ratio of either 27:1 or 64:1. Additionally, the percentage of CD8+ T lymphocytes did not significantly increase after stimulation of foal PBMC (prestimulation, 11% ± 2.7%; poststimulation, 16% ± 5.4%; n = 4; P = 0.1585; see Fig. 5), whereas those in the adult horse control had a significant increase after stimulation (prestimulation, 12% ± 2.9%; poststimulation, 31% ± 13.4%; 15 independent assays using H71; P < 0.0001).
These results suggested that there was a difference in either the foal effectors or in antigen presentation by foal PBAC targets. To assess the latter possibility, CTL assays using adult effectors against foal targets were performed. Interestingly, foal-derived infected PBAC targets were efficiently lysed (Fig. 3A). The lysis was similar or greater than the lysis of autologous adult infected PBAC.
Development of R. equi-specific CTL responses in foals. To further assess the development of R. equi-specific CTL activity, PBMC were collected from five foals at 6 weeks of age and four foals at 8 weeks of age. Foals A2197 and A2198 were evaluated at 3 weeks of age as well as at 6 and 8 weeks. Foal A2201 was tested at 3 and 6 weeks of age, and foal A2203 was only available for testing at 3 weeks. To ensure that CTL were stimulated in vivo, foals A2196 and A2197 were intrabronchially infected when they were 6 weeks old with a small dose of virulent R. equi. At 3 weeks of age, none of the foals had lymphocytes that significantly lysed infected autologous or mismatched targets following stimulation (Fig. 4A). At 6 weeks of age, stimulated lymphocytes from three of five foals significantly lysed both autologous and mismatched infected targets (Fig. 4B). At 8 weeks of age, lymphocytes from all foals stimulated with R. equi caused significant lysis of infected autologous and mismatched targets regardless of experimental or natural exposure (Fig. 4C). The percentage of specific lysis caused by lymphocytes from foals obtained at 8 weeks was increased over those obtained at 6 weeks in three of four foals (compare Fig. 4B and C), and lymphocytes from A2198 had percentages similar to those of the adult control (Fig. 4C). Lysis by R. equi-stimulated lymphocytes from both foals and adult horse controls was similar to that of previously observed classical MHC class I-unrestricted lysis. This type of lysis was shown to be mediated by R. equi-specific CD2+, CD3+, CD8+, CD4– T lymphocytes in the adult immune horse (32).
Fluorescence-activated cell sorter analysis of in vitro prestimulation and poststimulation of CD8+ T lymphocyte populations (Fig. 5) revealed that PBMC from 6-week-old foals (prestimulation, 10% ± 0.9%; poststimulation, 24% ± 3.6%; n = 5; P = 0.0007) as well as 8-week-old foals (prestimulation, 12% ± 1.3%; poststimulation, 33% ± 4.3%; n = 4; P = 0.0014) respond to antigenic stimulation by R. equi in a manner similar to that of adult horses (prestimulation, 12% ± 2.9%; poststimulation, 31% ± 13.4%; 15 independent assays using H71; P < 0.0001).
Expression of CD1 on equine target cells. One possible mechanism of antigen presentation resulting in classical MHC class I-unrestricted lysis is through CD1 (10). Members of the CD1 family are highly conserved across species. Notably, antibodies produced against CD1 proteins of humans, sheep, cats, cattle, and guinea pigs have all been shown to react with APC in other species (11, 20). Thus, potential differences in surface expression of CD1 by foal and adult PBAC targets and by adult BAL cells were assessed. Following evaluation of eight CD1 MAb, two were shown to consistently bind to equine PBAC. For PBAC, a granulocyte/monocyte-specific MAb to CD172 was used to specifically gate on the monocytes in PBAC. A similar percentage of adult PBAC (65% ± 17.8%; n = 4) (Fig. 6A) and 6-week-old foal PBAC (80% ± 15.6%; n = 6) (Fig. 6A) expressed CD1 (P = 0.17), as assessed with MAb BCD1b3. Similarly, the percentage of cells with CD1 expression as assessed with MAb CD1F2 was the same in adult (39% ± 19.3%; n = 14) (Fig. 6A) and 6-week-old foal PBAC (52% ± 28.5%; n = 10; P = 0.35) (Fig. 6A). Data from a representative foal, A2196 at 6 weeks of age, is presented in Fig. 6B. This is the first reported evidence of CD1 expression on ex vivo equine APC. In contrast to the PBAC, however, pulmonary macrophages expressed minimal CD1 as measured by these two MAbs. BAL macrophages inconsistently expressed CD172a, so gating was performed via cellular size and granularity. Only 5% ± 1.6% of BAL macrophages reacted with MAb BCD1b3 (Fig. 6A) (n = 2), and 3% ± 1.4% of BAL cells reacted with MAb CD1F2 (Fig. 6A) (n = 6). Because adherent BAL cells effectively served as targets in the CTL assay (Fig. 1 and Fig. 2), it seems unlikely that CTL recognized R. equi-infected cells via a CD1 molecule identified by one of these two antibodies.
DISCUSSION
The concept that CTL play a role in the control of a phagosome-bound, intracellular bacterium is relatively new. Although CD8+ T lymphocytes were previously considered a potential source of IFN- during M. tuberculosis infection (35), there is now abundant evidence that the recognition and lysis of infected cells by CTL is important in limiting the disease (8, 38). Notably, lysis of M. tuberculosis-infected cells in humans has been inversely correlated with the severity of tuberculosis (13). In chronic M. tuberculosis infection of humans, the M. tuberculosis-specific immune response is often compartmentalized to the lungs. Higher frequency of M. tuberculosis-secreted protein Ag85-specific IFN- production was shown to occur in bronchoalveolar lavage-derived cells from healthy household contacts of patients with M. tuberculosis infection than in autologous PBMC (34). Likewise, although alveolar and blood-derived lymphocytes stimulated with M. tuberculosis purified protein derivative (PPD) and IL-2 showed equivalent lysis of blood-derived antigen-pulsed monocytes, antigen-pulsed alveolar macrophages were more resistant to lysis (42).
The immune response of R. equi-challenged immune adult horses has shown a similar compartmentalization. Whereas CD4+ and CD8+ pulmonary T lymphocytes expanded with challenge, the numbers of these cells in peripheral blood were unaltered. Likewise, in contrast to pulmonary cells, which proliferated in response to a small dose (2 μg/ml) of R. equi antigen, stimulation of PBMC from intrabronchially challenged adult horses proliferated only when exposed to a large dose (10 μg/ml) of antigen (18). Immune clearance of virulent R. equi in adult horses also correlated with increased numbers of IFN--positive CD4+ and CD8+ T lymphocytes at the site of pulmonary challenge (19). These data support a model by which immune animals clear R. equi due to an effective pulmonary recall response in which antigen-specific memory T lymphocytes expand locally in response to infection and/or migrate into the lung from the draining lymph nodes.
We recently showed that R. equi-specific CTL are present in the peripheral blood of immune horses (32). These CTL could be identified in all adult horses examined and, interestingly, lysed R. equi-infected targets (PBAC) in a classical MHC class I-unrestricted fashion. Uninfected targets were not killed, and significant lysis required antigen stimulation. To determine whether or not CTL activity also occurred in the lung, three separate trials using two ELA-A-mismatched, immune horses were performed. Pulmonary T lymphocytes effectively lysed R. equi-infected targets, and as with blood-derived effectors, lysis was not ELA-A restricted. Moreover, alveolar macrophage targets obtained by bronchoalveolar lavage were effectively killed by both PBMC and pulmonary effectors. These data are consistent with a role for CTL in protective immunity to R. equi. The data also support evaluation of peripheral blood CTL as a method to further study immune responses relevant to protection.
An important issue is the development of R. equi-specific CTL activity in early life, including its correlation with susceptibility and age-associated resistance. Foals that develop rhodococcal pneumonia are likely infected in the first few days to weeks of life, and all foals are presumably exposed (22). The understanding of the neonatal immune system is rudimentary at best, regardless of the species. In general, though, neonatal cell-mediated responses are thought to be Th2 biased (1, 3, 4). Given that rhodococcal pneumonia occurs when a Th2 immune response is induced, the Th2 bias of neonates is clearly a potential reason that foals are uniquely susceptible. However, many studies have revealed contradictory responses of neonates to antigen exposure, varying from limited to adult-like. The reaction is largely immunogen dependent, and recent evidence indicates that neonatal T lymphocytes can be induced to produce Th1 cytokines. Specifically, with increased costimulation, adult levels of Th1 cytokines are induced (2). Thus, an intrinsic inability of neonatal T lymphocytes to respond to antigen in an adult-like manner may not be present. Nevertheless, natural exposure to intracellular bacterial pathogens typically induces a Th2 response in neonates. These Th2-type responses are invariably associated with poor CTL activity (2). However, similar to Th1 reactions, activation of adult-like CTL responses can be achieved in neonates by methods such as DNA vaccination, reducing the antigen dose, and immunizing with complete Freund's adjuvant (2).
The present study is the first to examine CTL in immunocompetent foals and is notably the first to associate acquisition of R. equi-specific CTL activity with the well-known, age-associated development of resistance to rhodococcal pneumonia. As predicted, CTL activity was deficient in the youngest foals examined (3 weeks of age). This observation coincides with the period of time during which foals are first exposed and presumably infected due to being born and raised in a contaminated environment. At 6 weeks of age, some foals had developed R. equi-specific CTL activity, although the specific lysis was always less than that of the adult horse control that was run simultaneously (data not shown). In addition, significant ex vivo expansions of CD8+ cells occurred in R. equi-stimulated PBMC derived from 6-week-old foals. As with adults, killing of R. equi-infected targets was not ELA-A restricted. These finding are compatible with the acquisition of CTL in association with development of protective immune responses due to natural exposure and immunologic maturation.
At 8 weeks of age, all foals in this study had significant R. equi-specific CTL activity. There was no detectable difference between the foals that had been intrabronchially challenged with virulent R. equi versus foals that had not been experimentally challenged. This result may be due to small sample size (only two control and two challenged foals were examined) or to the low dosage of the challenge. The dose was intended to induce immune responses and not produce disease, but it could have been insufficient for either. More importantly, all foals may have already been responding effectively to natural exposure, as three out of five foals showed significant, albeit modest, CTL activity at 6 weeks of age. Importantly, none of the foals examined in these experiments developed rhodococcal pneumonia. It would be interesting to study a larger number of foals on a farm where infection is endemic to determine if a failure or delay in development of CTL is predictive of foals that develop clinical disease versus immune clearance.
Several important issues remain regarding R. equi-specific CTL activity in horses. Among these are the mechanisms by which R. equi-infected cells are recognized, the specific antigens recognized, and the method of antigen presentation. Based on cell surface markers, our previous study indicated that the effector cells are primarily CD8+ T lymphocytes (CD2+, CD3+) and not natural killer cells. There may also be a subpopulation of CD8– CD4– (double-negative) CTL (32). One possible explanation for classical MHC class I-unrestricted cytolysis due to CD8+ or double-negative T lymphocytes is recognition of lipid or glycolipid antigen presented via CD1 molecules. CD1 is a minimally polymorphic family of MHC class I-like glycoproteins expressed on the surface of mammalian APC in association with 2 microglobulin (6, 9). CD1 molecules, which have a variety of isoforms in different species, have been shown to be important in the presentation of unique lipid or lipid-modified antigens of M. tuberculosis, including the M. tuberculosis cell wall antigens lipoarabinomannan and mycolic acid (7, 37). Related cell wall antigens have been characterized in R. equi, including at least one that cross-reacts with M. tuberculosis (16, 30, 40).
Based on the reported conservation of CD1 molecules across species, we examined a panel of previously described anti-CD1 antibodies for cross-reactivity with equine APC using flow cytometry. The expression of CD1 was also compared with susceptibility to lysis by R. equi-specific CTL. The two CD1 MAb found to cross-react with surface molecules in the horse had been previously shown to bind several CD1b isoforms (CD1b2, CD1b3, and CD1b4) and the CD1c isoform CD1c3 in the guinea pig, suggesting that these or related isoforms also occur in the horse (20). CD1b was the isoform most commonly associated with presentation of M. tuberculosis lipid antigens (7, 37). In contrast to recent reports in which human neonatal cord cells lacked CD1 expression and had decreased CD1 expression poststimulation (26), perinatal and adult horse PBAC expressed similar levels of CD1. Therefore, a relative lack of expression, at least for these molecules, does not appear to play a role in either the perinatal susceptibility to rhodococcal pneumonia or the absence of R. equi-specific CTL at 3 weeks of age. More importantly, equine alveolar macrophages, which are as susceptible to lysis by R. equi-specific CTL as PBAC, expressed minimal levels of the CD1 molecules recognized by the two antibodies. This suggested that R. equi-specific CTL recognition of equine alveolar macrophages was not mediated by a CD1 molecule bound by the two MAbs used.
One explanation for the lack of association between cytolysis and CD1 expression is that important lipid and/or glycolipid antigens of R. equi may be presented via equine CD1 molecules not recognized by one of the two cross-reactive antibodies. Likewise, expression of CD1 and presentation of antigen may be altered in infected APC compared to that of uninfected cells (39). Further studies, including examination of additional antibodies, R. equi lipids, and equine CD1 genes, are likely required. Alternatively, however, other nonclassical presentation systems, such as glycolipid CD1d-restricted NK T cells or peptide nonclassical MHC class I-associated CTL, may be involved in R. equi-specific CTL activity in horses (14, 15, 25). The MHC class I molecules of the horse are incompletely characterized; however, four putative nonclassical MHC class I molecules have been identified, suggesting a potential for antigen presentation and CTL induction (21).
In conclusion, data from this study further extend the observations regarding R. equi-specific CTL in the natural host. Specifically, the isolation of CTL from the lung of immune horses and the early development of cytotoxic activity in young foals suggests a role for CTL in protection against rhodococcal pneumonia. Further dissection of the role of ELA-A-unrestricted CTL and improved understanding of the mechanism by which these cells recognize R. equi-infected cells will provide much-needed insight into how to induce protective immunity by vaccination. Importantly, because of the ubiquitous nature of R. equi, the goal of such a vaccine may be to drive the immune response of naturally exposed foals toward the protective phenotype (e.g., induce CTL at an earlier age).
ACKNOWLEDGMENTS
This work was supported by grant 5 K08 A1049391-03 from the National Institute of Allergy and Infectious Diseases and grant D0IEQ-42 from the Morris Animal Foundation.
We thank Wendy C. Brown and are grateful to Matt Littke, Deb Alperin, Linda Norton, Ashley Alger, and Zack Joseph for their excellent technical assistance.
REFERENCES
1. Adkins, B. 2000. Development of neonatal Th1/Th2 function. Int. Rev. Immunol. 19:157-171.
2. Adkins, B. 1999. T-cell function in newborn mice and humans. Immunol. Today 20:330-335.
3. Adkins, B., Y. Bu, V. Vincek, and P. Guevara. 2003. The primary responses of murine neonatal lymph node CD4+ cells are Th2-skewed and are sufficient for the development of Th2-biased memory. Clin. Dev. Immunol. 10:43-51.
4. Adkins, B., and R. Q. Du. 1998. Newborn mice develop balanced Th1/Th2 primary effector responses in vivo but are biased to Th2 secondary responses. J. Immunol. 160:4217-4224.
5. Alvarez, B., C. Sanchez, R. Bullido, A. Marina, J. Lunney, F. Alonso, A. Ezquerra, and J. Dominguez. 2000. A porcine cell surface receptor identified by monoclonal antibodies to SWC3 is a member of the signal regulatory protein family and associates with protein-tyrosine phosphatase SHP-1. Tissue Antigens 55:342-351.
6. Batuwangala, T., D. Shepherd, S. D. Gadola, K. J. Gibson, N. R. Zaccai, A. R. Fersht, G. S. Besra, V. Cerundolo, and E. Y. Jones. 2004. The crystal structure of human CD1b with a bound bacterial glycolipid. J. Immunol. 172:2382-2388.
7. Beckman, E. M., S. A. Porcelli, C. T. Morita, S. M. Behar, S. T. Furlong, and M. B. Brenner. 1994. Recognition of a lipid antigen by CD1-restricted alpha beta+ T cells. Nature 372:691-694.
8. Boom, W. H., D. H. Canaday, S. A. Fulton, A. J. Gehring, R. E. Rojas, and M. Torres. 2003. Human immunity to M. tuberculosis: T cell subsets and antigen processing. Tuberculosis (Edinburgh) 83:98-106.
9. Brigl, M., and M. B. Brenner. 2004. CD1: antigen presentation and T cell function. Annu. Rev. Immunol. 22:817-890.
10. Dascher, C. C., and M. B. Brenner. 2003. CD1 antigen presentation and infectious disease. Contrib. Microbiol. 10:164-182.
11. Dascher, C. C., K. Hiromatsu, J. W. Naylor, P. P. Brauer, K. A. Brown, J. R. Storey, S. M. Behar, E. S. Kawasaki, S. A. Porcelli, M. B. Brenner, and K. P. LeClair. 1999. Conservation of a CD1 multigene family in the guinea pig. J. Immunol. 163:5478-5488.
12. Davis, W. C., J. E. Davis, and M. Hamilton. 1995. Use of monoclonal antibodies and flow cytometry to cluster and analyse leukocyte differentiation molecules, p. 149. In W. C. Davis (ed.), Monoclonal antibody protocols. Humana Press, Inc., Totowa, N.J.
13. De La Barrera, S. S., M. Finiasz, A. Frias, M. Aleman, P. Barrionuevo, S. Fink, M. C. Franco, E. Abbate, and S. M. del Sasiain 2003. Specific lytic activity against mycobacterial antigens is inversely correlated with the severity of tuberculosis. Clin. Exp. Immunol. 132:450-461.
14. Dieli, F., M. Taniguchi, M. Kronenberg, S. Sidobre, J. Ivanyi, L. Fattorini, E. Iona, G. Orefici, G. De Leo, D. Russo, N. Caccamo, G. Sireci, C. Di Sano, and A. Salerno. 2003. An anti-inflammatory role for V alpha 14 NK T cells in Mycobacterium bovis bacillus Calmette-Guerin-infected mice. J. Immunol. 171:1961-1968.
15. Gansert, J. L., V. Kiessler, M. Engele, F. Wittke, M. Rollinghoff, A. M. Krensky, S. A. Porcelli, R. L. Modlin, and S. Stenger. 2003. Human NKT cells express granulysin and exhibit antimycobacterial activity. J. Immunol. 170:3154-3161.
16. Garton, N. J., M. Gilleron, T. Brando, H. H. Dan, S. Giguere, G. Puzo, J. F. Prescott, and I. C. Sutcliffe. 2002. A novel lipoarabinomannan from the equine pathogen Rhodococcus equi. Structure and effect on macrophage cytokine production. J. Biol. Chem. 277:31722-31733.
17. Hanau, D., D. A. Schmitt, T. Bieber, D. Schmitt, and J. P. Cazenave. 1990. Possible mechanism of action of CD1a antigens. J. Investig. Dermatol. 95:503-505.
18. Hines, M. T., K. M. Paasch, D. C. Alperin, G. H. Palmer, N. C. Westhoff, and S. A. Hines. 2001. Immunity to Rhodococcus equi: antigen-specific recall responses in the lungs of adult horses. Vet. Immunol. Immunopathol. 79:101-114.
19. Hines, S. A., D. M. Stone, M. T. Hines, D. C. Alperin, D. P. Knowles, L. K. Norton, M. J. Hamilton, W. C. Davis, and T. C. McGuire. 2003. Clearance of virulent but not avirulent Rhodococcus equi from the lungs of adult horses is associated with intracytoplasmic gamma interferon production by CD4(+) and CD8(+) T lymphocytes. Clin. Diagn. Lab. Immunol. 10:208-215.
20. Hiromatsu, K., C. C. Dascher, M. Sugita, C. Gingrich-Baker, S. M. Behar, K. P. LeClair, M. B. Brenner, and S. A. Porcelli. 2002. Characterization of guinea-pig group 1 CD1 proteins. Immunology 106:159-172.
21. Holmes, E. C., and S. A. Ellis. 1999. Evolutionary history of MHC class I genes in the mammalian order Perissodactyla. J. Mol. Evol. 49:316-324.
22. Horowitz, M. L., N. D. Cohen, S. Takai, T. Becu, M. K. Chaffin, K. K. Chu, K. G. Magdesian, and R. J. Martens. 2001. Application of Sartwell's model (lognormal distribution of incubation periods) to age at onset and age at death of foals with Rhodococcus equi pneumonia as evidence of perinatal infection. J. Vet. Intern. Med. 15:171-175.
23. Kanaly, S. T., S. A. Hines, and G. H. Palmer. 1995. Cytokine modulation alters pulmonary clearance of Rhodococcus equi and development of granulomatous pneumonia. Infect. Immun. 63:3037-3041.
24. Kanaly, S. T., S. A. Hines, and G. H. Palmer. 1993. Failure of pulmonary clearance of Rhodococcus equi infection in CD4+ T-lymphocyte-deficient transgenic mice. Infect. Immun. 61:4929-4932.
25. Lau, P., C. Amadou, H. Brun, V. Rouillon, F. McLaren, A. F. Le Rolle, M. Graham, G. W. Butcher, and E. Joly. 2003. Characterisation of RT1-E2, a multigenic family of highly conserved rat non-classical MHC class I molecules initially identified in cells from immunoprivileged sites. BMC Immunol. 4:1-19.
26. Liu, E., W. Tu, H. K. Law, and Y. L. Lau. 2001. Decreased yield, phenotypic expression and function of immature monocyte-derived dendritic cells in cord blood. Br. J. Haematol. 113:240-246.
27. Lopez, A. M., M. T. Hines, G. H. Palmer, D. C. Alperin, and S. A. Hines. 2002. Identification of pulmonary T-lymphocyte and serum antibody isotype responses associated with protection against Rhodococcus equi. Clin. Diagn. Lab. Immunol. 9:1270-1276.
28. MacHugh, N. D., A. Bensaid, W. C. Davis, C. J. Howard, K. R. Parsons, B. Jones, and A. Kaushal. 1988. Characterization of a bovine thymic differentiation antigen analogous to CD1 in the human. Scand. J. Immunol. 27:541-547.
29. Macias, J., J. A. Pineda, F. Borderas, J. A. Gallardo, J. Delgado, M. Leal, A. Sanchez-Quijano, and E. Lissen. 1997. Rhodococcus or mycobacterium An example of misdiagnosis in HIV infection. AIDS 11:253-254.
30. Moody, D. B., V. Briken, T. Y. Cheng, C. Roura-Mir, M. R. Guy, D. H. Geho, M. L. Tykocinski, G. S. Besra, and S. A. Porcelli. 2002. Lipid length controls antigen entry into endosomal and nonendosomal pathways for CD1b presentation. Nat. Immunol. 3:435-442.
31. Nordmann, P., E. Ronco, and C. Nauciel. 1992. Role of T-lymphocyte subsets in Rhodococcus equi infection. Infect. Immun. 60:2748-2752.
32. Patton, K. M., T. C. McGuire, D. G. Fraser, and S. A. Hines. 2004. Rhodococcus equi-infected macrophages are recognized and killed by CD8+ T lymphocytes in a MHC class I-unrestricted fashion. Infect. Immun. 72:7073-7083.
33. Raabe, M. R., C. J. Issel, and R. C. Montelaro. 1998. Equine monocyte-derived macrophage cultures and their applications for infectivity and neutralization studies of equine infectious anemia virus. J. Virol. Methods 71:87-104.
34. Schwander, S. K., M. Torres, C. C. Carranza, D. Escobedo, M. Tary-Lehmann, P. Anderson, Z. Toossi, J. J. Ellner, E. A. Rich, and E. Sada. 2000. Pulmonary mononuclear cell responses to antigens of Mycobacterium tuberculosis in healthy household contacts of patients with active tuberculosis and healthy controls from the community. J. Immunol. 165:1479-1485.
35. Shams, H., B. Wizel, S. E. Weis, B. Samten, and P. F. Barnes. 2001. Contribution of CD8(+) T cells to gamma interferon production in human tuberculosis. Infect. Immun. 69:3497-3501.
36. Siedek, E., S. Little, S. Mayall, N. Edington, and A. Hamblin. 1997. Isolation and characterisation of equine dendritic cells. Vet. Immunol. Immunopathol. 60:15-31.
37. Sieling, P. A., D. Chatterjee, S. A. Porcelli, T. I. Prigozy, R. J. Mazzaccaro, T. Soriano, B. R. Bloom, M. B. Brenner, M. Kronenberg, P. J. Brennan, et al. 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269:227-230.
38. Smith, S. M., and H. M. Dockrell. 2000. Role of CD8 T cells in mycobacterial infections. Immunol. Cell Biol. 78:325-333.
39. Stenger, S., K. R. Niazi, and R. L. Modlin. 1998. Down-regulation of CD1 on antigen-presenting cells by infection with Mycobacterium tuberculosis. J. Immunol. 161:3582-3588.
40. Sutcliffe, I. C. 1997. Macroamphiphilic cell envelope components of Rhodococcus equi and closely related bacteria. Vet. Microbiol. 56:287-299.
41. Takai, S. 1997. Epidemiology of Rhodococcus equi infections: a review. Vet. Microbiol. 56:167-176.
42. Tan, J. S., D. H. Canaday, W. H. Boom, K. N. Balaji, S. K. Schwander, and E. A. Rich. 1997. Human alveolar T lymphocyte responses to Mycobacterium tuberculosis antigens: role for CD4+ and CD8+ cytotoxic T cells and relative resistance of alveolar macrophages to lysis. J. Immunol. 159:290-297.
43. Zhang, W., S. M. Lonning, and T. C. McGuire. 1998. Gag protein epitopes recognized by ELA-A-restricted cytotoxic T lymphocytes from horses with long-term equine infectious anemia virus infection. J. Virol. 72:9612-9620.(Kristin M. Patton, Travis)
Department of Veterinary Clinical Sciences, Washington State University, Pullman, Washington
ABSTRACT
Rhodococcus equi is an important cause of pneumonia in young horses; however, adult horses are immune due to their ability to mount protective recall responses. In this study, the hypothesis that R. equi-specific cytotoxic T lymphocytes (CTL) are present in the lung of immune horses was tested. Bronchoalveolar lavage (BAL)-derived pulmonary T lymphocytes stimulated with R. equi lysed infected alveolar macrophages and peripheral blood adherent cells (PBAC). As with CTL obtained from the blood, killing of R. equi-infected targets by pulmonary effectors was not restricted by equine lymphocyte alloantigen-A (ELA-A; classical major histocompatibility complex class I), suggesting a novel or nonclassical method of antigen presentation. To determine whether or not CTL activity coincided with the age-associated susceptibility to rhodococcal pneumonia, CTL were evaluated in foals. R. equi-stimulated peripheral blood mononuclear cells (PBMC) from 3-week-old foals were unable to lyse either autologous perinatal or mismatched adult PBAC targets. The defect was not with the perinatal targets, as adult CTL effectors efficiently killed infected targets from 3-week-old foals. In contrast, significant CTL activity was present in three of five foals at 6 weeks of age, and significant specific lysis was induced by PBMC from all foals at 8 weeks of age. As with adults, lysis was ELA-A unrestricted. Two previously described monoclonal antibodies, BCD1b3 and CD1F2/1B12.1, were used to examine the expression of CD1, a nonclassical antigen-presenting molecule, on CTL targets. These antibodies cross-reacted with both foal and adult PBAC. However, neither antibody bound alveolar macrophages, suggesting that the R. equi-specific, major histocompatibility complex-unrestricted lysis is not restricted by a surface molecule identified by these antibodies.
INTRODUCTION
Rhodococcus equi is a facultative intracellular bacterium with properties similar to those of Mycobacterium tuberculosis, a closely related pathogen. Both R. equi and M. tuberculosis are nocardioform actinomycetes, survive within the phagosomes of macrophages, have characteristic cell walls containing unique lipids, lipoproteins, and glycolipids, and produce pyogranulomatous lesions in the lung (16). Whereas M. tuberculosis produces tuberculosis in humans, R. equi is a common cause of life-threatening pneumonia in young horses. R. equi is also an opportunistic pathogen in AIDS patients (29).
R. equi is ubiquitous in the equine environment, and virtually all horses are considered exposed, likely in the first few weeks of life. Horses have a distinct age-associated susceptibility, such that equine rhodococcal pneumonia is almost exclusively a disease of foals less than 4 to 5 months of age. Even on farms where infection is endemic, a situation where a significant percentage of foals develop pneumonia and the environmental load of virulent R. equi is high, older horses do not manifest with disease. Previous research has provided strong evidence that the resistance of adult horses to R. equi challenge reflects acquired immunity and their ability to mount effective recall responses (18, 19). Therefore, immune adult horses, like humans who have successfully contained M. tuberculosis infection, provide a valuable system for understanding the mechanisms by which infection can be controlled and pneumonia prevented. Knowledge of these mechanisms will likely be critical to efforts to design an effective vaccine.
Rhodococcal pneumonia is an immunopathologic disease. Adoptive transfer and neutralizing antibody studies have shown that mice that develop Th2 responses, characterized by production of interleukin-4 (IL-4) rather than gamma interferon (IFN-), develop characteristic pulmonary lesions (23). In contrast, mice responding to infection with a Th1 response and IFN- production clear a virulent challenge. As with M. tuberculosis, CD8+ T lymphocytes can also contribute to immune clearance of R. equi (31). Even transgenic mice that lack major histocompatibility complex (MHC) class II-restricted CD4+ T lymphocytes significantly reduced R. equi numbers following pulmonary challenge (24).
The immune responses most relevant to protective immunity likely occur in the lung. For M. tuberculosis, both the nature of the response and the repertoire of antigens recognized can be different in comparisons of pulmonary responses to peripheral blood responses. Likewise, the outcome of host cell infection may be different in comparisons of alveolar macrophages to blood-derived monocytes. For example, human alveolar macrophages infected with M. tuberculosis are more resistant to cytotoxic T-lymphocyte (CTL)-mediated lysis than infected monocytes (42). When pulmonary responses to R. equi are examined via bronchoalveolar lavage (BAL), immune adult horses clear virulent organisms in association with increased numbers of CD4+ and CD8+ lymphocytes at the site of challenge (18). Although pulmonary T lymphocytes from most adult horses initially have low proliferative responses to R. equi antigen, challenge induces rapid, strong responses typical of a secondary immune response. Importantly, this recall response includes production of IFN- and is likely responsible for the immunity seen in adult horses (19, 27). As described with other pulmonary pathogens, the pulmonary responses to R. equi challenge are relatively compartmentalized. Responses are more marked, and the sensitivity to antigen dose is increased at the site of infection. Blood, including peripheral blood mononuclear cells (PBMC), has been an insensitive indicator of local pulmonary responses to R. equi (18).
More recently, research has examined the role of CTL in the immune control of R. equi (32). Equine PBMC stimulated with antigen-presenting cells (APC) either infected with R. equi or exposed to soluble R. equi antigen (SRA) lysed R. equi-infected target cells but not uninfected targets. Killing of R. equi-infected peripheral blood adherent cells (PBAC) by effector cells was equally effective against autologous and equine lymphocyte alloantigen-A (ELA-A) (classical MHC class I) mismatched targets. Lysis was decreased to background by depletion of either CD2+ or CD3+ cells, indicating that the effector cell had a T-lymphocyte but not an NK-cell phenotype. Depletion of CD8+ T cells also resulted in significantly decreased lysis of infected targets. These data indicate that immunocompetent adult horses have R. equi-specific CD8+ CTL, which may play a role in immunity. The apparent lack of restriction via classical MHC class I molecules suggests a novel method of antigen processing and presentation, such as presentation by CD1 or other nonclassical MHC molecules.
The goal of the experiments reported here was to further examine the relevance of R. equi-specific CTL to protective immunity. We first asked whether the CTL demonstrated in the peripheral blood of immune adult horses also occur in the lung. Next, we tested the hypothesis that susceptible perinatal foals are deficient in CTL that lyse R. equi-infected cells.
MATERIALS AND METHODS
Horses. Two healthy adult horses, one Quarterhorse (H71) and one Hanoverian (H68), presently maintained in the Washington State University (WSU) Veterinary Teaching Hospital herd, were selected based on mismatched ELA-A haplotypes and previous demonstration of CTL activity against R. equi-infected targets (32). For the experiments involving foals, six Arabian foals were selected based on age. Using previously described serologic reagents, ELA-A were determined by microcytotoxicity assay (43) (Table 1). Venous blood was collected from the jugular vein of each horse by using evacuated containers (Baxter, Deerfield, Ill.) containing 16 ml of anticoagulant citrate dextrose (ACD)/100 ml of blood. PBMC were isolated from venous blood using a Ficoll-Hypaque technique (43). PBMC were then used either as effector cells or for isolation of PBAC targets (see below).
Bacteria. The virulent R. equi strain ATCC 33701 (PL+), which contains the 80.6-kb virulence-associated plasmid and expresses the virulence-associated protein VapA, was utilized in all experiments. Bacteria were stored at –20°C as frozen stabilates and were reconstituted prior to each experiment. Bacteria were initially grown on brain heart infusion (BHI) agar and then cultivated overnight to log phase in BHI broth at 37°C in a shaking incubator at 200 rpm. After being washed once with phosphate-buffered saline (PBS), the bacteria were adjusted to 107/ml in antibiotic-free complete medium, which consisted of RPMI 1640 supplemented with 2 mM L-glutamine, 0.05 μM 2 mercaptoethanol, 6.25 mM HEPES, and 10% normal horse serum (NHS; Gibco, Grand Island, N.Y.). The adjusted bacterial concentration was initially confirmed by plating serial dilutions on BHI plates and calculating the CFU per milliliter.
Bronchoalveolar lavage and cell preparation. Horses were sedated with xylazine at 0.6 mg/kg of body weight. A Bivona BAL tube was inserted through the right nostril into the trachea and wedged in place. To collect BAL cells from adult horses, three separate aliquots of 180 ml of 0.9% NaCl containing 0.84% sodium bicarbonate (BAL wash) were instilled into the lung. Following each aliquot, fluid was aspirated. Fluid was placed into 50-ml conical tubes and centrifuged for 8 min at 600 x g at 4°C. Pellets were consolidated and washed three times with complete medium containing decreasing amounts of antibiotic/antimycotic (wash 1, 1 μg of amphotericin/ml [Gibco]; wash 2, 2 U of penicillin/ml, 2 U of streptomycin/ml, and 0.1 μg of amphotericin/ml [Gibco]; wash 3, antibiotic/antimycotic free). Resultant cells were plated at 4 x 106 cells/ml on 150-cm2 collagen/serum-coated petri dishes and incubated for 1 h at 37°C with 5% CO2 as previously described (32). Following incubation, nonadherent cells were collected by vigorous pipetting of the supernatant 10 times and washing again with PBS via pipetting 10 ml of PBS over the plate 10 times. Nonadherent cells (64% ± 29.6% CD2+ T lymphocytes and 35% ± 19.5% immunoglobulin M [IgM] positive) were plated at 4 x 106 cells/ml in 75-cm2 vented flasks for stimulation and used as effector cells. Adherent cells were used as targets following harvesting as described below.
Effector cell preparation and stimulation. PBMC or BAL cells in 25 ml of antibiotic-free complete medium were plated in 75-cm2 flasks at 4 x 106 cells/ml. The cells were stimulated with 1 multiplicity of infection (MOI) of live virulent R. equi 33701 PL+. Following a 1-h incubation at 37°C with 5% CO2, 1 μg of gentamicin/ml was added to kill extracellular bacteria. Culture flasks were then incubated for 5 days at 37°C with 5% CO2 and then rested without antigenic stimulation for 2 days prior to further use. Resting consisted of washing the cells twice with antibiotic-free complete medium, counting, and plating in fresh medium containing 1 μg of gentamicin/ml. Following stimulation and resting, cells were washed once, counted, and then characterized for surface marker expression by flow cytometry. Alternatively, the cells were used as effector cells in the CTL assay described below.
Target cell isolation. Autologous and ELA-A-mismatched PBAC were harvested from PBMC for use as target cells as previously described (33). Briefly, PBMC were plated on the gelatin/serum-coated dishes at a density of 1.5 x 108 cells/dish in 20 ml of antibiotic-free complete medium. Cells were incubated for 15 to 19 h at 37°C with 5% CO2. Following incubation, nonadherent cells were resuspended in the culture medium by repeated pipetting (10x) and then removed by washing each plate with 10 ml of PBS and repeated pipetting (10x). Adherent cells were eluted with 20 ml of antibiotic-free complete medium containing 20% (vol/vol) NHS mixed 1:1 with 10 mM EDTA for 35 min at 37°C with 5% CO2. The cells were then collected by vigorous pipetting of the eluting fluid 10 times followed by washing each dish twice with 10 ml of PBS per wash. Eluted cells were combined, washed once with antibiotic-free complete medium, counted, and added to 96-well plates at 105 cells/well in 100 μl of medium. Immediately prior to adding cells, 20 μl of commercial NHS was added to each well.
Adherent BAL cells were removed by covering the plates with 10 ml of Accutase (Sigma, St. Louis, Mo.) and incubating them for 10 min at 37°C with 5% CO2. The cells were then collected by repeated pipetting of the suspension over the plate 10 times followed by washing with 10 ml of PBS 10 times. Cells were consolidated via centrifugation at 600 x g at 4°C for 15 min, washed once with complete medium via the same centrifugation, and then counted and plated in 100 μl of complete medium on a serum-coated 96-well plate at 105 cells/well. The cells were incubated for 1 h prior to use as lung-derived target cells in the CTL assay (see below).
Cytotoxicity assay. After 1 h, each well of target cells was labeled with 50 μl of antibiotic-free complete medium containing 100 μCi of 51Cr per ml. Following 13 h of incubation, the targets were infected with 5 MOI of live R. equi bacteria by adding 50 μl of antibiotic-free complete medium containing 107 live R. equi cells per ml. One hour postinfection, 20 μl of complete medium containing 1 μg of gentamicin sulfate/ml was added to each culture well to kill any extracellular bacteria. Following an additional 7.5 h of incubation, the cells were washed four times with complete medium containing 1 μg of gentamicin sulfate/ml to remove extracellular 51Cr. Effector cells were then added to target cells and incubated for an additional 4.5 h at 37°C with 5% CO2. Following incubation, the plates were centrifuged at 500 x g for 5 min to pellet cells. One-hundred microliters of supernatant was removed, and the amount of 51Cr was measured using a MicroBeta plate reader. The formula for the percent specific lysis is [(E – S)/(M – S)] x 100, where E is the mean of three test wells, S is the mean spontaneous release from three target cell wells without effector cells, and M is the mean maximal release from three target cell wells with 3% Triton X-100 (48). As previously reported, significant lysis of infected cells was defined as 3 standard errors above the uninfected control target cells (43).
R. equi CTL responses in foals. To characterize the R. equi-specific immune response of foals in the first 2 months of life, PBMC were collected from six foals at 3, 6, and 8 weeks of age and used as effectors in CTL assays. PBMC from foals A2197, A2198, A2201, and A2203 were collected at 3 weeks of age. To ensure that a CTL response was induced, foals A2196 and A2197 were infected intrabronchially at 6 weeks of age with a small dose (104) of virulent R. equi. Following sedation (see the BAL protocol described above), bacteria in 1 ml of BAL wash was instilled into the BAL tube, followed by 10 ml of the BAL wash and 20 ml of air. PBMC were collected just prior to infection (at the 6-week-old time point) and then again 2 weeks postinfection (at 8 weeks of age). The remaining foals were housed with their dams and presumably were naturally exposed to R. equi in their environment (41). PBMC were also collected from the naturally exposed foals at 6 and 8 weeks of age. PBAC target cells from all foals were obtained 5 days after each effector cell collection date. Clinical parameters, including temperature, heart rate, and respiratory rate, of infected foals were monitored twice daily for 3 weeks. Lungs were simultaneously ausculted. Complete blood counts (CBC) plus fibrinogen, performed twice weekly, remained within normal limits for 3 weeks postinfection. Inoculated foals showed no signs of disease, except for mild pyrexia (less than 102.4°C) of 3 days' duration at 2 weeks postinfection.
Flow cytometry. Before and after R. equi stimulation, effector cells were labeled with a 15-μg/ml concentration of either anti-equine CD8 monoclonal antibody (MAb) HT14A, anti-equine CD4 MAb HB61A, anti-equine CD2 MAb HB88A, anti-equine IgG MAb1.9/3.2 (B-cell marker), anti-equine CD172a MAb DH59B (5) (granulocyte/monocyte marker), or isotype control antibody (COLIS69A) (control for nonspecific binding) (VMRD, Inc., Pullman, Wash.). Target cells were labeled with a 15-μg/ml concentration of anti-human CD1b MAb BCD1b3 subclone 1.3.1.6 (BCD1b3), 1:100 dilution of anti-guinea pig CD1 MAb CD1F2/1B12.1 (CD1F2; both CD1 antibodies graciously provided by S. Porcelli, Bronx, N.Y.), anti-equine CD172a MAb DH59B, or isotype control antibody (COLIS69A). Additional anti-CD1 antibodies, anti-guinea pig CD1 MAb CD1F2/5E3 (11, 20) (S. Porcelli), anti-guinea pig CD1 MAb CD1F2/6B5 (11, 20) (S. Porcelli, Bronx, NY), anti-bovine CD1b MAb CC-20 (28, 36) (C. J. Howard, Institute of Animal Health, Compton, Newbury, Berkshire, United Kingdom), anti-bovine CD1b TH97A (VMRD, Inc.), anti-ovine CD1b MAb 20-27 (11, 20, 28) (University of Melbourne, Center of Animal Biotechnology, Victoria, Australia), and anti-human CD1a MAb HI149 (17) (Biolegend, San Diego, Calif.) were also tested at previously reported concentrations. The mixtures were incubated on ice for 15 min at 4°C, washed three times, and pelleted at 4°C for 3 min at 500 x g. Cells were then suspended in 50 μl of (5 μg/ml) fluorescein-conjugated anti-mouse immunoglobulin serum absorbed with normal human and horse sera. Cells were next incubated for 15 min at 4°C in the dark, washed two times, and suspended in 200 μl of 2% formaldehyde in PBS (12). Labeled cells were analyzed with a FACScan equipped with a Macintosh computer and Cell Quest software (Becton Dickinson Immunocytometry Systems, San Jose, Calif.). The effector cell population was analyzed before stimulation and immediately prior to testing in the CTL assay. The values were compared using a paired t test, with P < 0.05 considered significant.
RESULTS
Pulmonary CTL responses. In previous studies, PBMC were shown to lyse R. equi-infected PBAC in a classical MHC class I-unrestricted manner (32). To determine whether a similar pattern of lysis would occur using cells from the lung (the target organ of R. equi infection), macrophages were collected from the lungs of immune adult horses. In three replicate trials using two horses, H71 and H68, PBMC effector cells stimulated with live virulent R. equi significantly lysed autologous and mismatched targets derived from the blood (Fig. 1A) and the lung (Fig. 1B) compared to that of uninfected targets. Thus, significant lysis of infected targets occurred regardless of tissue origin or MHC haplotype of the targets (see columns with asterisks in Fig. 1). At the optimal E:T (effector:target) ratios, the degree of lysis appeared similar for both BAL and PBAC mismatched and autologous targets, although a statistical comparison was not done.
In addition, pulmonary effectors were compared to blood-derived effectors (Fig. 2). As previously reported, lymphocyte numbers in BAL fluid of unstimulated immune horses were low (18), thus limiting the number of effectors that were available for evaluation (3.8 x 107 ± 2.1 x 107 lymphocytes before stimulation). At an E:T ratio of 1:1, both R. equi-stimulated BAL effectors and R. equi-stimulated PBMC effectors caused significant lysis of infected autologous and mismatched PBAC targets (Fig. 2A). Similarly, both R. equi-stimulated effectors caused significant lysis of infected autologous and mismatched BAL targets derived from the lung (Fig. 2B). Therefore, effectors from both the blood and lung significantly lysed mismatched targets from the blood and lung. The degree of lysis of the two target cell populations (lung versus blood derived) appeared similar. PBMC cultured for 7 days with no antigen (PBMC-NA) did not induce significant lysis of infected PBAC or BAL targets, thus, R. equi stimulation was required to induce PBMC effectors to lyse infected targets, regardless of tissue origin.
R. equi-specific CTL responses in perinatal foals. To test the hypothesis that perinatal foals are deficient in R. equi-specific CTL responses, PBMC isolated from four 3-week-old foals, A2197, A2198, A2201, and A2203, were stimulated for 7 days with 1 MOI of live virulent R. equi. In the initial cytotoxicity trial using A2197, a maximum E:T ratio of 64:1 was achieved. In all subsequent trials, where poststimulation lymphocyte numbers were lower, an E:T ratio of 27:1 was used. In all foals, increasing the E:T ratio from 1:1 to either 27:1 or 64:1 did not result in a significant percentage of specific lysis mediated by R. equi-stimulated foal PBMC (Fig. 3B). In all of the foals tested, the percentage of specific lysis was not significantly greater than that of the uninfected controls (Fig. 3B). This was in sharp contrast to the results obtained for simultaneously in vitro-stimulated adult H71 PBMC controls (Fig. 3A), which mediated lysis of 13 to 35% at an E:T ratio of either 27:1 or 64:1. Additionally, the percentage of CD8+ T lymphocytes did not significantly increase after stimulation of foal PBMC (prestimulation, 11% ± 2.7%; poststimulation, 16% ± 5.4%; n = 4; P = 0.1585; see Fig. 5), whereas those in the adult horse control had a significant increase after stimulation (prestimulation, 12% ± 2.9%; poststimulation, 31% ± 13.4%; 15 independent assays using H71; P < 0.0001).
These results suggested that there was a difference in either the foal effectors or in antigen presentation by foal PBAC targets. To assess the latter possibility, CTL assays using adult effectors against foal targets were performed. Interestingly, foal-derived infected PBAC targets were efficiently lysed (Fig. 3A). The lysis was similar or greater than the lysis of autologous adult infected PBAC.
Development of R. equi-specific CTL responses in foals. To further assess the development of R. equi-specific CTL activity, PBMC were collected from five foals at 6 weeks of age and four foals at 8 weeks of age. Foals A2197 and A2198 were evaluated at 3 weeks of age as well as at 6 and 8 weeks. Foal A2201 was tested at 3 and 6 weeks of age, and foal A2203 was only available for testing at 3 weeks. To ensure that CTL were stimulated in vivo, foals A2196 and A2197 were intrabronchially infected when they were 6 weeks old with a small dose of virulent R. equi. At 3 weeks of age, none of the foals had lymphocytes that significantly lysed infected autologous or mismatched targets following stimulation (Fig. 4A). At 6 weeks of age, stimulated lymphocytes from three of five foals significantly lysed both autologous and mismatched infected targets (Fig. 4B). At 8 weeks of age, lymphocytes from all foals stimulated with R. equi caused significant lysis of infected autologous and mismatched targets regardless of experimental or natural exposure (Fig. 4C). The percentage of specific lysis caused by lymphocytes from foals obtained at 8 weeks was increased over those obtained at 6 weeks in three of four foals (compare Fig. 4B and C), and lymphocytes from A2198 had percentages similar to those of the adult control (Fig. 4C). Lysis by R. equi-stimulated lymphocytes from both foals and adult horse controls was similar to that of previously observed classical MHC class I-unrestricted lysis. This type of lysis was shown to be mediated by R. equi-specific CD2+, CD3+, CD8+, CD4– T lymphocytes in the adult immune horse (32).
Fluorescence-activated cell sorter analysis of in vitro prestimulation and poststimulation of CD8+ T lymphocyte populations (Fig. 5) revealed that PBMC from 6-week-old foals (prestimulation, 10% ± 0.9%; poststimulation, 24% ± 3.6%; n = 5; P = 0.0007) as well as 8-week-old foals (prestimulation, 12% ± 1.3%; poststimulation, 33% ± 4.3%; n = 4; P = 0.0014) respond to antigenic stimulation by R. equi in a manner similar to that of adult horses (prestimulation, 12% ± 2.9%; poststimulation, 31% ± 13.4%; 15 independent assays using H71; P < 0.0001).
Expression of CD1 on equine target cells. One possible mechanism of antigen presentation resulting in classical MHC class I-unrestricted lysis is through CD1 (10). Members of the CD1 family are highly conserved across species. Notably, antibodies produced against CD1 proteins of humans, sheep, cats, cattle, and guinea pigs have all been shown to react with APC in other species (11, 20). Thus, potential differences in surface expression of CD1 by foal and adult PBAC targets and by adult BAL cells were assessed. Following evaluation of eight CD1 MAb, two were shown to consistently bind to equine PBAC. For PBAC, a granulocyte/monocyte-specific MAb to CD172 was used to specifically gate on the monocytes in PBAC. A similar percentage of adult PBAC (65% ± 17.8%; n = 4) (Fig. 6A) and 6-week-old foal PBAC (80% ± 15.6%; n = 6) (Fig. 6A) expressed CD1 (P = 0.17), as assessed with MAb BCD1b3. Similarly, the percentage of cells with CD1 expression as assessed with MAb CD1F2 was the same in adult (39% ± 19.3%; n = 14) (Fig. 6A) and 6-week-old foal PBAC (52% ± 28.5%; n = 10; P = 0.35) (Fig. 6A). Data from a representative foal, A2196 at 6 weeks of age, is presented in Fig. 6B. This is the first reported evidence of CD1 expression on ex vivo equine APC. In contrast to the PBAC, however, pulmonary macrophages expressed minimal CD1 as measured by these two MAbs. BAL macrophages inconsistently expressed CD172a, so gating was performed via cellular size and granularity. Only 5% ± 1.6% of BAL macrophages reacted with MAb BCD1b3 (Fig. 6A) (n = 2), and 3% ± 1.4% of BAL cells reacted with MAb CD1F2 (Fig. 6A) (n = 6). Because adherent BAL cells effectively served as targets in the CTL assay (Fig. 1 and Fig. 2), it seems unlikely that CTL recognized R. equi-infected cells via a CD1 molecule identified by one of these two antibodies.
DISCUSSION
The concept that CTL play a role in the control of a phagosome-bound, intracellular bacterium is relatively new. Although CD8+ T lymphocytes were previously considered a potential source of IFN- during M. tuberculosis infection (35), there is now abundant evidence that the recognition and lysis of infected cells by CTL is important in limiting the disease (8, 38). Notably, lysis of M. tuberculosis-infected cells in humans has been inversely correlated with the severity of tuberculosis (13). In chronic M. tuberculosis infection of humans, the M. tuberculosis-specific immune response is often compartmentalized to the lungs. Higher frequency of M. tuberculosis-secreted protein Ag85-specific IFN- production was shown to occur in bronchoalveolar lavage-derived cells from healthy household contacts of patients with M. tuberculosis infection than in autologous PBMC (34). Likewise, although alveolar and blood-derived lymphocytes stimulated with M. tuberculosis purified protein derivative (PPD) and IL-2 showed equivalent lysis of blood-derived antigen-pulsed monocytes, antigen-pulsed alveolar macrophages were more resistant to lysis (42).
The immune response of R. equi-challenged immune adult horses has shown a similar compartmentalization. Whereas CD4+ and CD8+ pulmonary T lymphocytes expanded with challenge, the numbers of these cells in peripheral blood were unaltered. Likewise, in contrast to pulmonary cells, which proliferated in response to a small dose (2 μg/ml) of R. equi antigen, stimulation of PBMC from intrabronchially challenged adult horses proliferated only when exposed to a large dose (10 μg/ml) of antigen (18). Immune clearance of virulent R. equi in adult horses also correlated with increased numbers of IFN--positive CD4+ and CD8+ T lymphocytes at the site of pulmonary challenge (19). These data support a model by which immune animals clear R. equi due to an effective pulmonary recall response in which antigen-specific memory T lymphocytes expand locally in response to infection and/or migrate into the lung from the draining lymph nodes.
We recently showed that R. equi-specific CTL are present in the peripheral blood of immune horses (32). These CTL could be identified in all adult horses examined and, interestingly, lysed R. equi-infected targets (PBAC) in a classical MHC class I-unrestricted fashion. Uninfected targets were not killed, and significant lysis required antigen stimulation. To determine whether or not CTL activity also occurred in the lung, three separate trials using two ELA-A-mismatched, immune horses were performed. Pulmonary T lymphocytes effectively lysed R. equi-infected targets, and as with blood-derived effectors, lysis was not ELA-A restricted. Moreover, alveolar macrophage targets obtained by bronchoalveolar lavage were effectively killed by both PBMC and pulmonary effectors. These data are consistent with a role for CTL in protective immunity to R. equi. The data also support evaluation of peripheral blood CTL as a method to further study immune responses relevant to protection.
An important issue is the development of R. equi-specific CTL activity in early life, including its correlation with susceptibility and age-associated resistance. Foals that develop rhodococcal pneumonia are likely infected in the first few days to weeks of life, and all foals are presumably exposed (22). The understanding of the neonatal immune system is rudimentary at best, regardless of the species. In general, though, neonatal cell-mediated responses are thought to be Th2 biased (1, 3, 4). Given that rhodococcal pneumonia occurs when a Th2 immune response is induced, the Th2 bias of neonates is clearly a potential reason that foals are uniquely susceptible. However, many studies have revealed contradictory responses of neonates to antigen exposure, varying from limited to adult-like. The reaction is largely immunogen dependent, and recent evidence indicates that neonatal T lymphocytes can be induced to produce Th1 cytokines. Specifically, with increased costimulation, adult levels of Th1 cytokines are induced (2). Thus, an intrinsic inability of neonatal T lymphocytes to respond to antigen in an adult-like manner may not be present. Nevertheless, natural exposure to intracellular bacterial pathogens typically induces a Th2 response in neonates. These Th2-type responses are invariably associated with poor CTL activity (2). However, similar to Th1 reactions, activation of adult-like CTL responses can be achieved in neonates by methods such as DNA vaccination, reducing the antigen dose, and immunizing with complete Freund's adjuvant (2).
The present study is the first to examine CTL in immunocompetent foals and is notably the first to associate acquisition of R. equi-specific CTL activity with the well-known, age-associated development of resistance to rhodococcal pneumonia. As predicted, CTL activity was deficient in the youngest foals examined (3 weeks of age). This observation coincides with the period of time during which foals are first exposed and presumably infected due to being born and raised in a contaminated environment. At 6 weeks of age, some foals had developed R. equi-specific CTL activity, although the specific lysis was always less than that of the adult horse control that was run simultaneously (data not shown). In addition, significant ex vivo expansions of CD8+ cells occurred in R. equi-stimulated PBMC derived from 6-week-old foals. As with adults, killing of R. equi-infected targets was not ELA-A restricted. These finding are compatible with the acquisition of CTL in association with development of protective immune responses due to natural exposure and immunologic maturation.
At 8 weeks of age, all foals in this study had significant R. equi-specific CTL activity. There was no detectable difference between the foals that had been intrabronchially challenged with virulent R. equi versus foals that had not been experimentally challenged. This result may be due to small sample size (only two control and two challenged foals were examined) or to the low dosage of the challenge. The dose was intended to induce immune responses and not produce disease, but it could have been insufficient for either. More importantly, all foals may have already been responding effectively to natural exposure, as three out of five foals showed significant, albeit modest, CTL activity at 6 weeks of age. Importantly, none of the foals examined in these experiments developed rhodococcal pneumonia. It would be interesting to study a larger number of foals on a farm where infection is endemic to determine if a failure or delay in development of CTL is predictive of foals that develop clinical disease versus immune clearance.
Several important issues remain regarding R. equi-specific CTL activity in horses. Among these are the mechanisms by which R. equi-infected cells are recognized, the specific antigens recognized, and the method of antigen presentation. Based on cell surface markers, our previous study indicated that the effector cells are primarily CD8+ T lymphocytes (CD2+, CD3+) and not natural killer cells. There may also be a subpopulation of CD8– CD4– (double-negative) CTL (32). One possible explanation for classical MHC class I-unrestricted cytolysis due to CD8+ or double-negative T lymphocytes is recognition of lipid or glycolipid antigen presented via CD1 molecules. CD1 is a minimally polymorphic family of MHC class I-like glycoproteins expressed on the surface of mammalian APC in association with 2 microglobulin (6, 9). CD1 molecules, which have a variety of isoforms in different species, have been shown to be important in the presentation of unique lipid or lipid-modified antigens of M. tuberculosis, including the M. tuberculosis cell wall antigens lipoarabinomannan and mycolic acid (7, 37). Related cell wall antigens have been characterized in R. equi, including at least one that cross-reacts with M. tuberculosis (16, 30, 40).
Based on the reported conservation of CD1 molecules across species, we examined a panel of previously described anti-CD1 antibodies for cross-reactivity with equine APC using flow cytometry. The expression of CD1 was also compared with susceptibility to lysis by R. equi-specific CTL. The two CD1 MAb found to cross-react with surface molecules in the horse had been previously shown to bind several CD1b isoforms (CD1b2, CD1b3, and CD1b4) and the CD1c isoform CD1c3 in the guinea pig, suggesting that these or related isoforms also occur in the horse (20). CD1b was the isoform most commonly associated with presentation of M. tuberculosis lipid antigens (7, 37). In contrast to recent reports in which human neonatal cord cells lacked CD1 expression and had decreased CD1 expression poststimulation (26), perinatal and adult horse PBAC expressed similar levels of CD1. Therefore, a relative lack of expression, at least for these molecules, does not appear to play a role in either the perinatal susceptibility to rhodococcal pneumonia or the absence of R. equi-specific CTL at 3 weeks of age. More importantly, equine alveolar macrophages, which are as susceptible to lysis by R. equi-specific CTL as PBAC, expressed minimal levels of the CD1 molecules recognized by the two antibodies. This suggested that R. equi-specific CTL recognition of equine alveolar macrophages was not mediated by a CD1 molecule bound by the two MAbs used.
One explanation for the lack of association between cytolysis and CD1 expression is that important lipid and/or glycolipid antigens of R. equi may be presented via equine CD1 molecules not recognized by one of the two cross-reactive antibodies. Likewise, expression of CD1 and presentation of antigen may be altered in infected APC compared to that of uninfected cells (39). Further studies, including examination of additional antibodies, R. equi lipids, and equine CD1 genes, are likely required. Alternatively, however, other nonclassical presentation systems, such as glycolipid CD1d-restricted NK T cells or peptide nonclassical MHC class I-associated CTL, may be involved in R. equi-specific CTL activity in horses (14, 15, 25). The MHC class I molecules of the horse are incompletely characterized; however, four putative nonclassical MHC class I molecules have been identified, suggesting a potential for antigen presentation and CTL induction (21).
In conclusion, data from this study further extend the observations regarding R. equi-specific CTL in the natural host. Specifically, the isolation of CTL from the lung of immune horses and the early development of cytotoxic activity in young foals suggests a role for CTL in protection against rhodococcal pneumonia. Further dissection of the role of ELA-A-unrestricted CTL and improved understanding of the mechanism by which these cells recognize R. equi-infected cells will provide much-needed insight into how to induce protective immunity by vaccination. Importantly, because of the ubiquitous nature of R. equi, the goal of such a vaccine may be to drive the immune response of naturally exposed foals toward the protective phenotype (e.g., induce CTL at an earlier age).
ACKNOWLEDGMENTS
This work was supported by grant 5 K08 A1049391-03 from the National Institute of Allergy and Infectious Diseases and grant D0IEQ-42 from the Morris Animal Foundation.
We thank Wendy C. Brown and are grateful to Matt Littke, Deb Alperin, Linda Norton, Ashley Alger, and Zack Joseph for their excellent technical assistance.
REFERENCES
1. Adkins, B. 2000. Development of neonatal Th1/Th2 function. Int. Rev. Immunol. 19:157-171.
2. Adkins, B. 1999. T-cell function in newborn mice and humans. Immunol. Today 20:330-335.
3. Adkins, B., Y. Bu, V. Vincek, and P. Guevara. 2003. The primary responses of murine neonatal lymph node CD4+ cells are Th2-skewed and are sufficient for the development of Th2-biased memory. Clin. Dev. Immunol. 10:43-51.
4. Adkins, B., and R. Q. Du. 1998. Newborn mice develop balanced Th1/Th2 primary effector responses in vivo but are biased to Th2 secondary responses. J. Immunol. 160:4217-4224.
5. Alvarez, B., C. Sanchez, R. Bullido, A. Marina, J. Lunney, F. Alonso, A. Ezquerra, and J. Dominguez. 2000. A porcine cell surface receptor identified by monoclonal antibodies to SWC3 is a member of the signal regulatory protein family and associates with protein-tyrosine phosphatase SHP-1. Tissue Antigens 55:342-351.
6. Batuwangala, T., D. Shepherd, S. D. Gadola, K. J. Gibson, N. R. Zaccai, A. R. Fersht, G. S. Besra, V. Cerundolo, and E. Y. Jones. 2004. The crystal structure of human CD1b with a bound bacterial glycolipid. J. Immunol. 172:2382-2388.
7. Beckman, E. M., S. A. Porcelli, C. T. Morita, S. M. Behar, S. T. Furlong, and M. B. Brenner. 1994. Recognition of a lipid antigen by CD1-restricted alpha beta+ T cells. Nature 372:691-694.
8. Boom, W. H., D. H. Canaday, S. A. Fulton, A. J. Gehring, R. E. Rojas, and M. Torres. 2003. Human immunity to M. tuberculosis: T cell subsets and antigen processing. Tuberculosis (Edinburgh) 83:98-106.
9. Brigl, M., and M. B. Brenner. 2004. CD1: antigen presentation and T cell function. Annu. Rev. Immunol. 22:817-890.
10. Dascher, C. C., and M. B. Brenner. 2003. CD1 antigen presentation and infectious disease. Contrib. Microbiol. 10:164-182.
11. Dascher, C. C., K. Hiromatsu, J. W. Naylor, P. P. Brauer, K. A. Brown, J. R. Storey, S. M. Behar, E. S. Kawasaki, S. A. Porcelli, M. B. Brenner, and K. P. LeClair. 1999. Conservation of a CD1 multigene family in the guinea pig. J. Immunol. 163:5478-5488.
12. Davis, W. C., J. E. Davis, and M. Hamilton. 1995. Use of monoclonal antibodies and flow cytometry to cluster and analyse leukocyte differentiation molecules, p. 149. In W. C. Davis (ed.), Monoclonal antibody protocols. Humana Press, Inc., Totowa, N.J.
13. De La Barrera, S. S., M. Finiasz, A. Frias, M. Aleman, P. Barrionuevo, S. Fink, M. C. Franco, E. Abbate, and S. M. del Sasiain 2003. Specific lytic activity against mycobacterial antigens is inversely correlated with the severity of tuberculosis. Clin. Exp. Immunol. 132:450-461.
14. Dieli, F., M. Taniguchi, M. Kronenberg, S. Sidobre, J. Ivanyi, L. Fattorini, E. Iona, G. Orefici, G. De Leo, D. Russo, N. Caccamo, G. Sireci, C. Di Sano, and A. Salerno. 2003. An anti-inflammatory role for V alpha 14 NK T cells in Mycobacterium bovis bacillus Calmette-Guerin-infected mice. J. Immunol. 171:1961-1968.
15. Gansert, J. L., V. Kiessler, M. Engele, F. Wittke, M. Rollinghoff, A. M. Krensky, S. A. Porcelli, R. L. Modlin, and S. Stenger. 2003. Human NKT cells express granulysin and exhibit antimycobacterial activity. J. Immunol. 170:3154-3161.
16. Garton, N. J., M. Gilleron, T. Brando, H. H. Dan, S. Giguere, G. Puzo, J. F. Prescott, and I. C. Sutcliffe. 2002. A novel lipoarabinomannan from the equine pathogen Rhodococcus equi. Structure and effect on macrophage cytokine production. J. Biol. Chem. 277:31722-31733.
17. Hanau, D., D. A. Schmitt, T. Bieber, D. Schmitt, and J. P. Cazenave. 1990. Possible mechanism of action of CD1a antigens. J. Investig. Dermatol. 95:503-505.
18. Hines, M. T., K. M. Paasch, D. C. Alperin, G. H. Palmer, N. C. Westhoff, and S. A. Hines. 2001. Immunity to Rhodococcus equi: antigen-specific recall responses in the lungs of adult horses. Vet. Immunol. Immunopathol. 79:101-114.
19. Hines, S. A., D. M. Stone, M. T. Hines, D. C. Alperin, D. P. Knowles, L. K. Norton, M. J. Hamilton, W. C. Davis, and T. C. McGuire. 2003. Clearance of virulent but not avirulent Rhodococcus equi from the lungs of adult horses is associated with intracytoplasmic gamma interferon production by CD4(+) and CD8(+) T lymphocytes. Clin. Diagn. Lab. Immunol. 10:208-215.
20. Hiromatsu, K., C. C. Dascher, M. Sugita, C. Gingrich-Baker, S. M. Behar, K. P. LeClair, M. B. Brenner, and S. A. Porcelli. 2002. Characterization of guinea-pig group 1 CD1 proteins. Immunology 106:159-172.
21. Holmes, E. C., and S. A. Ellis. 1999. Evolutionary history of MHC class I genes in the mammalian order Perissodactyla. J. Mol. Evol. 49:316-324.
22. Horowitz, M. L., N. D. Cohen, S. Takai, T. Becu, M. K. Chaffin, K. K. Chu, K. G. Magdesian, and R. J. Martens. 2001. Application of Sartwell's model (lognormal distribution of incubation periods) to age at onset and age at death of foals with Rhodococcus equi pneumonia as evidence of perinatal infection. J. Vet. Intern. Med. 15:171-175.
23. Kanaly, S. T., S. A. Hines, and G. H. Palmer. 1995. Cytokine modulation alters pulmonary clearance of Rhodococcus equi and development of granulomatous pneumonia. Infect. Immun. 63:3037-3041.
24. Kanaly, S. T., S. A. Hines, and G. H. Palmer. 1993. Failure of pulmonary clearance of Rhodococcus equi infection in CD4+ T-lymphocyte-deficient transgenic mice. Infect. Immun. 61:4929-4932.
25. Lau, P., C. Amadou, H. Brun, V. Rouillon, F. McLaren, A. F. Le Rolle, M. Graham, G. W. Butcher, and E. Joly. 2003. Characterisation of RT1-E2, a multigenic family of highly conserved rat non-classical MHC class I molecules initially identified in cells from immunoprivileged sites. BMC Immunol. 4:1-19.
26. Liu, E., W. Tu, H. K. Law, and Y. L. Lau. 2001. Decreased yield, phenotypic expression and function of immature monocyte-derived dendritic cells in cord blood. Br. J. Haematol. 113:240-246.
27. Lopez, A. M., M. T. Hines, G. H. Palmer, D. C. Alperin, and S. A. Hines. 2002. Identification of pulmonary T-lymphocyte and serum antibody isotype responses associated with protection against Rhodococcus equi. Clin. Diagn. Lab. Immunol. 9:1270-1276.
28. MacHugh, N. D., A. Bensaid, W. C. Davis, C. J. Howard, K. R. Parsons, B. Jones, and A. Kaushal. 1988. Characterization of a bovine thymic differentiation antigen analogous to CD1 in the human. Scand. J. Immunol. 27:541-547.
29. Macias, J., J. A. Pineda, F. Borderas, J. A. Gallardo, J. Delgado, M. Leal, A. Sanchez-Quijano, and E. Lissen. 1997. Rhodococcus or mycobacterium An example of misdiagnosis in HIV infection. AIDS 11:253-254.
30. Moody, D. B., V. Briken, T. Y. Cheng, C. Roura-Mir, M. R. Guy, D. H. Geho, M. L. Tykocinski, G. S. Besra, and S. A. Porcelli. 2002. Lipid length controls antigen entry into endosomal and nonendosomal pathways for CD1b presentation. Nat. Immunol. 3:435-442.
31. Nordmann, P., E. Ronco, and C. Nauciel. 1992. Role of T-lymphocyte subsets in Rhodococcus equi infection. Infect. Immun. 60:2748-2752.
32. Patton, K. M., T. C. McGuire, D. G. Fraser, and S. A. Hines. 2004. Rhodococcus equi-infected macrophages are recognized and killed by CD8+ T lymphocytes in a MHC class I-unrestricted fashion. Infect. Immun. 72:7073-7083.
33. Raabe, M. R., C. J. Issel, and R. C. Montelaro. 1998. Equine monocyte-derived macrophage cultures and their applications for infectivity and neutralization studies of equine infectious anemia virus. J. Virol. Methods 71:87-104.
34. Schwander, S. K., M. Torres, C. C. Carranza, D. Escobedo, M. Tary-Lehmann, P. Anderson, Z. Toossi, J. J. Ellner, E. A. Rich, and E. Sada. 2000. Pulmonary mononuclear cell responses to antigens of Mycobacterium tuberculosis in healthy household contacts of patients with active tuberculosis and healthy controls from the community. J. Immunol. 165:1479-1485.
35. Shams, H., B. Wizel, S. E. Weis, B. Samten, and P. F. Barnes. 2001. Contribution of CD8(+) T cells to gamma interferon production in human tuberculosis. Infect. Immun. 69:3497-3501.
36. Siedek, E., S. Little, S. Mayall, N. Edington, and A. Hamblin. 1997. Isolation and characterisation of equine dendritic cells. Vet. Immunol. Immunopathol. 60:15-31.
37. Sieling, P. A., D. Chatterjee, S. A. Porcelli, T. I. Prigozy, R. J. Mazzaccaro, T. Soriano, B. R. Bloom, M. B. Brenner, M. Kronenberg, P. J. Brennan, et al. 1995. CD1-restricted T cell recognition of microbial lipoglycan antigens. Science 269:227-230.
38. Smith, S. M., and H. M. Dockrell. 2000. Role of CD8 T cells in mycobacterial infections. Immunol. Cell Biol. 78:325-333.
39. Stenger, S., K. R. Niazi, and R. L. Modlin. 1998. Down-regulation of CD1 on antigen-presenting cells by infection with Mycobacterium tuberculosis. J. Immunol. 161:3582-3588.
40. Sutcliffe, I. C. 1997. Macroamphiphilic cell envelope components of Rhodococcus equi and closely related bacteria. Vet. Microbiol. 56:287-299.
41. Takai, S. 1997. Epidemiology of Rhodococcus equi infections: a review. Vet. Microbiol. 56:167-176.
42. Tan, J. S., D. H. Canaday, W. H. Boom, K. N. Balaji, S. K. Schwander, and E. A. Rich. 1997. Human alveolar T lymphocyte responses to Mycobacterium tuberculosis antigens: role for CD4+ and CD8+ cytotoxic T cells and relative resistance of alveolar macrophages to lysis. J. Immunol. 159:290-297.
43. Zhang, W., S. M. Lonning, and T. C. McGuire. 1998. Gag protein epitopes recognized by ELA-A-restricted cytotoxic T lymphocytes from horses with long-term equine infectious anemia virus infection. J. Virol. 72:9612-9620.(Kristin M. Patton, Travis)