INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Staphylococcus aureus is a prominent pathogen in bovine mastitis (24). This organism is frequently isolated from milk (2, 16, 40) and from cows with intramammary infection (IMI) (17). IMI caused by S. aureus tends to become chronic and may resist antibiotic therapy (49). It has been postulated that persistent infection with S. aureus is associated with an impairment of the immune response, mediated by factors produced by S. aureus (34). Thus far, however, no single factor has been clearly implicated.

Bovine isolates of S. aureus frequently produce one or more pyrogenic toxins (PTs), especially types C and D staphylococcal enterotoxins (SEs) and toxic shock syndrome toxin (24). The staphylococcal PTs are prototype microbial superantigens (SAgs), characterized by the ability to bind to major histocompatibility complex class II (MHCII) molecules and to specific V segments of T-cell receptor (TCR) outside the binding groove associated with MHC-restricted immune system recognition of processed peptides. By bypassing antigenic specificity, SAgs stimulate abnormally large numbers of T cells and are able, at nanomolar concentrations, to induce T-cell proliferation (33). Few studies have been performed to investigate the effects of SAgs on the bovine immune system (53); most studies have involved other animals. In several species, SAgs exert wide-ranging and deleterious effects, including induction of shock (4), T-cell unresponsiveness and deletion (23), differential stimulation of CD4⁺ and CD8⁺ T-cell subsets (47), and B-cell differentiation (46). Thus, although largely unconfirmed, there is a clear potential for SEs, and other SAgs, to modulate immune responses and contribute to the virulence and persistence of S. aureus in cattle.

The T-cell population consists of cells expressing either the TCR (TCR2) or the TCR (TCR1). While the roles of T cells in immune responses of many species have been well characterized, the function of T cells is less well understood (22). This is especially true in ruminants. Recent investigations have shown that the ruminant T cells comprise two disparate subpopulations, characterized by constitutive expression of cell surface molecules. One subpopulation, similar in composition and tissue distribution to T cells from other species, consists of cells that express CD2, CD5, and CD6 and are positive or negative for CD8 (10). These cells are present in low concentrations (3 to 5%) in peripheral blood and in high concentrations (35 to 40%) in spleen, gut epithelium, and mammary gland secretions (41). The second subpopulation, negative for CD2, CD6, and CD8, is unique and has been identified in only one other member of the Artiodactyla, swine (3, 28). This subpopulation is positive for CD5 and two lineage-restricted molecules, workshop cluster 1 (WC1) (32, 36, 50) and GD3.5 (21). The concentration of WC1⁺ GD3.5⁺ CD2 CD6 T cells is high (30 to 50%) in the peripheral blood in young ruminants, decreasing with age, and is low (3 to 8%) in secondary lymphoid tissues and mammary gland secretions (41, 52). Definitive data on the function of either of these major subpopulations of T cells have not been obtained. However, previous studies have suggested that they may be involved in regulating the proliferative response of CD4⁺ T cells to antigens (7).

Park et al. (42) identified a subset of T cells, positive for CD2 and CD8, in mammary gland secretions of cows infected with S. aureus. These cells had the ability to inhibit the proliferative response of bovine CD4⁺ cells to staphylococcal antigens (42). Although the mechanisms by which these cells were induced and mediated their effect were not determined, they clearly have the potential to contribute to the pathogenesis of staphylococcal IMI. Although not all bovine staphylococcal isolates produce known SAgs, it is important to determine whether SAg production could induce these or other immunosuppressive subpopulations in cows and promote the development of some infections such as IMI. The objective of this study was to extend these initial observations. We examined the effect of a representative SAg (SEC1) on the major subpopulations of bovine and T cells.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

SEC1 purification. SEC1 was purified from S. aureus (pMIN121), a recombinant harboring the sec_MNDON structural gene cloned into a nontoxigenic background (strain RN4220) (4, 19). Cultures were grown with aeration at 37°C in pyrogen-free dialyzable beef heart medium containing erythromycin (50 µg/ml). SEC1 was purified to homogeneity by standard preparative flat-bed isoelectric focusing techniques (44) with broad- and narrow-pH-range ampholytes in succession. Fractions containing purified toxin were identified by immunodiffusion, and the degree of purity was assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (26). Ampholytes were removed from the toxin solution by exhaustive dialysis against pyrogen-free water.

Animals. Purebred adult healthy Holstein-Frisian cattle were screened as prospective donors and for determination of the relative amounts of and T cells in their systemic circulation. Most experiments were performed with cells from two selected donors: one with a low concentration of T cells (13%) and the other with high concentration (82%) (designated donors A and F, respectively). Donor A (low ) was a 5-year-old dairy cow from the University of Idaho dairy farm, Moscow, Idaho. Donor F (high ) was a 2.5-year-old steer, housed at Washington State University, Pullman, Wash. The animals were maintained according to Association for the Assessment and Accreditation of Laboratory Animal Care, International, guidelines and regulations established by the Animal Care and Use Committees at Washington State University and the University of Idaho.

Blood collection and cell culture. Collection, processing, and conditions for culturing of peripheral blood mononuclear cells (PBMC) were described previously (42). PBMC suspensions were adjusted to 2.0 × 10⁶ cells per ml and incubated in plastic petri dishes for 4, 24, or 96 h at 37°C in 5% CO₂. The final concentrations of stimulants in cultures (SEC1, 0.1 µg/ml; concanavalin A ConA, 5.0 µg/ml) constituted the doses found to induce optimal T-cell proliferation in dose response assays (results not shown). Concanavalin A (ConA) was purchased from Sigma Chemical Co., St. Louis, Mo.

Flow cytometry. Before and after stimulation in culture, the cells were processed for single or multiple color flow cytometric analysis by established techniques (11). The monoclonal antibodies used in this study to detect cell surface molecules are listed in Table 1. Polyclonal and isotype-specific anti-mouse immunoglobulins (prepared in goats) conjugated to fluorescein isothiocyanate, phycoerythrin, or Tri-color (Caltag Laboratories, Burlingame, Calif.) were used as second-step reagents. Flow cytometric data were acquired with a FACSort apparatus equipped with a Macintosh computer and CellQuest software (Becton-Dickinson Immunocytometry Systems, San Jose, Calif.). The forward and side scatter gates for bovine leukocytes were set to exclude cell debris and dead cells. Typically, 5,000 events were acquired per sample. When necessary, gates were set on small subpopulations to collect 2,000 events for analysis. Analyses were performed with the CellQuest, PAINT A GATE, or ATTRACTORS analytical program. Specific subpopulations of cells were quantified by using fixed attractors, with a cutoff line at 1.1 log unit. Expression of interleukin-2 receptor  chain (IL-2R) was separated into high- and low-intensity categories, i.e., above and below 2.0 log units. Large cells, considered to be blasts, were readily identifiable on plots of linear forward/log right-angle light scatter. A cutoff value of linear forward light scatter, usually >612 channels, was set to distinguish small cells (comprising cells at a low or partial level of activation) from large (blast) cells. The data for these experiments are presented as an index of blastogenesis, which refers to the percentage of large cells in a given subpopulation.

Statistical analysis. Data on the expression of activation molecules are presented as arithmetic mean ± standard error of the mean (SEM) of triplicate measurements. To ascertain statistical significance of changes in cell numbers, a two-tailed t test (54) was performed on the counts after square root transformation, with P < 0.01 indicating significance.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Stimulation of bovine PBMC by SEC1 is unique and appears to be influenced by T cells. The percentage of T cells expressing WC1 and GD3.5 (referred to here as CD2 T cells) varies in bovine blood from over 80% in young animals to <10% in adults (52). Since this subpopulation of T cells has been reported to modulate T-cell responses (5, 7, 8, 52), the composition and general proliferative responses of PBMC from several prospective donors were assessed. As shown in Fig. 1, cumulative data from six animals revealed a marked difference in the proliferative response of CD4⁺ T cells to SEC1 in the presence of different concentrations of T cells, consistent with previous observations that indicated that T cells regulate activation of CD4⁺ T cells. Two initial observations pointed to a potential inhibitory effect mediated by T cells on SEC1-induced stimulation of CD4⁺ T cells. First, the relative percent of CD4⁺ T cells in cultures from all six donors declined during 96 h in the presence of SEC1. This decline was most prominent in cultures containing a high concentration of T cells. Second, most of the CD4⁺ T cells in cultures with a high concentration of T cells did not increase to blast size. A reduction in the percentage of T cells expressing CD8 was also observed with some donors. However, this reduction was smaller than that noted for CD4⁺ T cells and was evident only in cultures with high concentrations of T cells (donors D, E, and F). In contrast to CD4⁺, most CD8⁺ T cells were blast size after 96 h of culture, even in cultures derived from animals with high concentrations of T cells in peripheral blood.

View larger version (26K): [in this window] [in a new window]   FIG. 1.   Influence of T cells on CD4+ and CD8+ T cells in PBMC cultures following a 96-h incubation with SEC1. Data are percentages of T cells obtained in a representative experiment. Donors A through F were healthy Holstein-Frisian cattle differing in / T-cell ratios as indicated in the figure. Each of the six donors was tested in the following number of replicate experiments: donor A, 7; donor B, 2; donor C, 3; donor D, 4; donor E, 3; and donor F, 6. They consistently yielded similar results.

View larger version (17K): [in this window] [in a new window]   FIG. 2.   Proliferative responses of bovine PBMC in cultures stimulated with SEC1. Data are means of triplicate measurements ± SEM. (A) Incorporation of [3H]thymidine in a standard 4-day proliferation assay. (B) Changes in cell numbers within bovine T-cell subpopulations following a 96-h culture of PBMC with SEC1 or ConA. Statistically significant increases (P < 0.01) are indicated (*).

Lack of correlation between SEC1-induced blastogenesis and proliferation. As shown in Fig. 2B, the net increase in cell number in cultures derived from donor A cells and stimulated with SEC1 was attributed to the expansion of CD8⁺ cells. Interestingly, though, subsequent experiments suggested that SEC1 partially activates other subpopulations of T cells, presumably through signaling pathways that do not lead to cell division. Analysis of forward and side light scatter patterns of cells from cultures stimulated with SEC1 revealed that many cells had increased in cell size, consistent with the early events of cell activation and blastogenesis. Stimulation with SEC1 caused a significant increase in the number of blast-sized cells within several T-cell subpopulations, especially in cultures with a low concentration of T cells. The majority (76.9%) of CD8⁺ and a lower but significant (53.9%) percentage of CD4⁺ cells reached blast size in cultures derived from donor A (Table 2), even though there was no net increase in the number of CD4⁺ cells at 96 h.

Activation of CD4⁺ cells by SEC1. T-cell activation is typically associated with sequential upregulation of several IL receptors and other cell-surface molecules. For example, following activation, IL-2R expression may be upregulated within hours, followed by upregulation of MHCII molecules on T cells. To further analyze factors affecting CD4⁺ T-cell activation, we measured the expression of IL-2R and MHCII to monitor the progression of T cells through early and late stages of activation. Since the expression of IL-2R is affected by both constitutive and inducible factors, the density of this receptor on the surface varies depending on the level of stimulation (1). High-density expression requires IL-2R gene transcription as well as mRNA stabilization (6). Flow cytometric dot plots, of SEC1-stimulated cells labeled with anti-IL-2R and one additional antibody (for phenotyping the cell population under study), were used to distinguish between low- and high-density expression. Cells with low-density expression were defined as cells with fluorescence staining intensities less than 2.0 log units; cells considered to have high density expression had intensities above 2.0 log units.

View larger version (33K): [in this window] [in a new window]   FIG. 3.   Expression of IL-2R, MHCII, and ACT3 by subpopulations of bovine T cells in PBMC cultures stimulated with SEC1 or ConA. Data are numbers of cells expressing a given marker (mean ± SEM of three measurements obtained in a representative experiment). The cells expressing IL-2R are separated into nonoverlapping populations characterized by a small or large amount of the receptor on the cell surface. (A) CD4+ cells; (B) CD8+ cells. Data represented by the open and solid circles are means ± SEM of the total number of cells expressing CD4 or CD8 at each time point.

SEC1 induces an unique phenotype of activated CD8⁺ cells. Activation molecule 3 (ACT3) is a 120-kDa membrane molecule with no identified human or mouse ortholog (13, 42, 45). In ConA-stimulated cultures of PBMC, ACT3 was expressed on at least 50% of CD4⁺ cells after 96 h (Fig. 3A). Levels of ACT3 expression on CD4⁺ cells in donor A PBMC cultures, stimulated with either SEC1 or ConA, were similar to the levels of expression of IL-2R^high and MHCII (Fig. 3A). An increase in the expression of ACT3 was clearly evident only in ConA-stimulated cultures of PBMC from donor F. In contrast to CD4⁺ cells, a much smaller proportion of CD8⁺ cells expressed ACT3 in ConA-stimulated cultures, but SEC1 induced ACT3 expression on more than 50% of CD8⁺ cells in culture from donor A and on more than 40% of CD8⁺ cells in culture from donor F (Fig. 3B). Most of the CD8⁺ blasts (60 to 80%) were positive for ACT3 in SEC1-stimulated cultures, whereas fewer than 15% were positive in ConA-stimulated cultures (data not shown). Combined, these data suggest that SEC1-induced expression of ACT3 on bovine CD8⁺ T cells coincides with a highly activated state.

Effect of SEC1 on T cells. The responses of the CD2 and CD2⁺ CD8⁺ subpopulations of T cells differed. CD2⁺ CD8⁺ cells failed to expand in cultures stimulated with either ConA or SEC1 (Fig. 2B). The absence of proliferation correlated with minimal activation, evidenced by low blastogenesis and moderate expression of IL-2R (Table 2; Fig. 4A). In contrast to other T-cell subpopulations, expression of MHCII was higher than expression of IL-2R at 96 h.

View larger version (24K): [in this window] [in a new window]   FIG. 4.   Expression of IL-2R and MHCII on subpopulations of bovine T cells in PBMC cultures stimulated with SEC1 or ConA. Data are numbers of cells positive for a given marker (mean ± SEM of three measurements obtained in a representative experiment). The cells positive for IL-2R are separated into nonoverlapping populations characterized by small or large amounts of the receptor on the cell surface. (A) CD2+ CD8+ cells; (B) CD2 cells. Notice the different scale in panels A and B.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

SAg-induced oligoclonal activation of T cells results from the binding of SAgs to MHCII molecules and TCRs that bear specific V gene products. For PT SAgs such as SEs and toxic shock syndrome toxin type 1, structural data suggest this interaction does not bring the peptide binding groove of the MHCII molecule in juxtaposition with the TCR peptide binding pocket (15, 20, 25). Thus, cross-linking is sufficient to transduce signals of activation. How these signals differ from those that occur following MHC-restricted immune system recognition has not been fully elucidated. However, it is clear that SAg-mediated signaling is unique and that altered signaling leads to a cascade of events associated with toxic shock syndrome and aberrations in the function of lymphocytes (43).

The present study suggests that SAgs have the potential to affect the bovine immune system in a manner that could promote staphylococcal persistence and infections such as IMI. Comparing the response of PBMC to ConA and SEC1 revealed differences in cell activation, proliferation, and expression of activation molecules. Stimulation with SEC1 led to partial activation of CD4⁺ T cells and T cells as evidenced by the expression of MHCII and IL-2R. A significant number of CD4⁺ T cells also expressed ACT3 and increased in size. However, proliferation was limited or absent. Few CD2 T cells increased in size in 4-day cultures, and there was no increase in the numbers of CD2⁺ CD8⁺ T cells in spite of some increase in cell size. In contrast, stimulation with SEC1 led to activation and proliferation of CD8⁺ T cells and the unique expression of the activation molecule, ACT3. Of interest, the level of activation of CD4⁺ and CD8⁺ T cells was affected by the proportion of T cells present in the culture, suggesting that these cells may play a modulatory role in the activation of T cells. The evolution of the immune system in swine and ruminants is unique since the T cell concentration is highly variable in these animals (3, 5, 10, 21, 22, 32, 36, 41, 50, 51, 52). Although T-cell levels are influenced by age, other factors, such as prior disease or exposure, have not been shown to influence the / T cell ratio. Whether animals with high levels of T cells are more susceptible to infection, especially by SAg-producing organisms, is one possibility currently under investigation.

Efforts to elucidate the mechanisms of SAg-driven T-cell responses in other animals have shown that aberrant signaling induced by SAgs leads to the production of multiple regulatory cytokines in vivo (29). This is followed by deletion of cells expressing specific V segments (30, 31) and anergy or hyporesponsiveness to stimuli in the remaining cells (23, 35). Such hyporesponsiveness may be associated with downregulation of the expression of IL-2R and inhibition of JAK-3 kinase, demonstrated for human CD4⁺ T cells (38). The present study of bovine cells is in agreement with studies performed in other systems, showing that CD4⁺ T-cell function may be inhibited more than that of CD8⁺ T cells by prolonged exposure to SAgs (9, 23). Long-term studies in vivo showed that CD8⁺ T cells predominate among SAg-reactive T cells (i.e., those bearing specific V segments) surviving treatment with SAg (18, 31). The results obtained in the present study are also in line with results of experiments with mice, whereby treatment with SAgs gave rise to CD8⁺ T cells with an altered phenotype and functional activity. Specifically, CD8⁺ cells isolated from mice treated with SAg were shown to lack cytotoxic activity, although they were responsive to cytokines and retained the ability to proliferate (48). Further studies are required to determine whether such cells could modulate the activation and function of CD4⁺ T cells through the production of cytokines that downregulate their capacity to respond to stimuli.

Previous studies with T cells from cows have shown that a subpopulation of activated CD8⁺ T cells downregulates the response of CD4⁺ cells to staphylococcal antigens presented by antigen-presenting cells (42). Elevated levels of these CD8⁺ T cells were demonstrated in mammary secretions from glands infected with S. aureus, indicating that they might play a role in pathogenesis (42). Although the potential role of microbial products in inducing these cells was not established, the present study indicates that SAg-activated CD8⁺ T cells may play a significant role in the pathogenesis of mastitis caused by SAg-producing S. aureus.

The activation of bovine CD8⁺ T cells by SEC1 occurs through a mechanism that is associated with unique expression of ACT3, a 120-kDa molecule with no identified ortholog in any other species. Antibodies that recognize this molecule were assigned to workshop cluster 10 in the Second International Workshop on Ruminant Leukocyte Differentiation Antigens (45). High-level expression of ACT3 on activated CD8⁺ T cells has not been previously observed and could be associated with the expression of cytokines that modulate the capacity of CD4⁺ T cells to respond to mitogenic stimuli. Previous studies have shown ACT3 is expressed predominantly by CD4⁺ T cells in ConA-stimulated cultures (13). More recently, investigations have shown that ACT3 is highly expressed on CD4⁺ and T cells in long-term cultures derived from animals stimulated with Babesia bovis (5). Further studies are now needed to determine specifically whether SEC1-activated CD8⁺ T cells affect the proliferative and functional activity of CD4⁺ T cells.

This study also suggests that activation of CD4⁺ and CD8⁺ T cells by SEC1 and, to a lesser degree, by ConA is significantly influenced by the proportion of T cells present in cultures. Since the content of T cells is a highly variable characteristic of ruminants, the general susceptibility of animals to SAg-mediated immunosuppression may vary greatly. The mechanism of a putative regulatory influence of T cells remains to be determined. It is not clear whether T cells in ruminants can be activated directly through the binding of SAgs to the TCR. However, expression of activation markers on their surface, following exposure to SEC1, indicates that either direct or indirect activation of T cells does occur.

In conclusion, evidence has been obtained which shows that SAgs could induce immunosuppression in dairy animals and contribute to the pathogenesis of staphylococcal mastitis. The staphylococcal SAg SEC1 induces a unique and aberrant activation of T-cell subpopulations. Further studies are warranted on this basis alone. However, the findings also suggest that the bovine system might serve as a useful model for the investigation of the mechanisms by which SAgs modulate immune system function and cause disease in other animals.