Long-Term Staphylococcal Enterotoxin C1 Exposure Induces Soluble Factor-Mediated Immunosuppression by Bovine CD4⁺ and CD8⁺ T Cells^▿

Purification of SEC1.

SEC1 was purified from cultures of S. aureus RN4220(pMIN121), harboring the recombinant sec structural gene cloned from S. aureus MNDON (19). Cultures were grown in medium containing dialyzable beef heart broth extract and erythromycin (5 μg/ml). SEC1 was purified by ethanol precipitation of the bacterial cultures in the cold, followed by resolubilization and preparative isoelectric focusing with broad isoelectric point ( 3 to 10) and narrow isoelectric point (6 to 8) ranges of ampholytes in succession as described previously (6).

Preparation of heat-killed fixed S. aureus.

Overnight cultures of S. aureus Novel strain (29) were harvested, washed in phosphate-buffered saline (PBS), incubated at 80°C for 15 min, fixed in 2% paraformaldehyde, and then resuspended in culture medium as previously described (22).

Animals and purification of PBMC.

Blood was obtained from purebred adult healthy Holstein-Frisian steers (10 to 18 months old) via venipuncture of the jugular vein. 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 Committee at the University of Idaho. PBMC were isolated from blood by density gradient centrifugation using Ficoll-Hypaque (Pharmacia Biotechnology) according to standard techniques (6). The RPMI 1640 culture medium used in the present study contained 10% heat-inactivated fetal bovine serum (HyClone, Logan, UT), 2 μM l-glutamine, 100 U of penicillin-streptomycin/ml, and 5 × 10⁻⁵ M 2-mercaptoethanol.

Cell stimulation.

To determine the level of lymphocyte proliferation induced by SEC1, PBMC were processed as described above, adjusted to a concentration of 10⁶ cells/ml in RPMI 1640 culture medium with various concentrations of SEC1 (0.05 ng/ml to 10 μg/ml) added to PBMC cultures (200 μl of PBMC suspension per well in 96-well culture plates), and incubated at 37°C in 5% CO₂ for 72 h. Incorporation of [³H]thymidine into cellular DNA in a standard 4-day assay (34) was used to measure the level of lymphocyte proliferation. Cultures were pulsed (18 h) with 1 μCi of [³H]thymidine. Cellular DNA was harvested on glass fiber filters, and [³H]thymidine incorporation was quantified by liquid scintillation counting.

For phenotypic and gene expression analyses, 12 ml of the PBMC suspension (concentration of 10⁶ cells per ml in RPMI 1640 culture medium) was incubated in tissue culture dish (Falcon, Lincoln Park, NJ) for 2, 4, 6, 8, or 10 days at 37°C in 5% CO₂. SEC1 was added to some cultures to obtain a concentration of 5 ng/ml. In some experiments, PBMC were stained with 5- (and 6-)carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes, Eugene, OR) at 2 μM for 15 min at 37°C as described previously (36). CFSE-stained PBMC were stimulated with SEC1 for up to 10 days and immunolabeled by using antibodies specific for a variety of cell surface markers as described below.

MAbs.

Monoclonal antibodies (MAbs) specific for bovine CD4, CD8, CD62 ligand (CD62L) (BAQ92A, immunoglobulin G1 [IgG1]), CD45R (GS5A, IgG1), CD25 (CACT116A, IgG1), CD26 (CACT114A, IgG2b), and CD45R0 (GC44A, IgG3) were purchased from the Washington State University Monoclonal Antibody Center. Anti-IL-10 and IgG1 isotype control MAbs were purchased from Serotec Immunological (Oxford, United Kingdom); anti-TGF-β MAb was purchased from R&D Systems (Minneapolis, MN).

FC analysis of bovine PBMC.

Some properties of SEC1-stimulated bovine PBMC were analyzed by indirect immunofluorescence staining and flow cytometry (FC). After stimulation with SEC1 in culture, PBMC were harvested and incubated with the appropriate MAb at 4°C for 30 min. After being washed with PBS, the cells were incubated at 4°C for 30 min in the dark with isotype-specific anti-mouse immunoglobulins conjugated to fluorescein isothiocyanate or phycoerythrin (PE; Caltag Laboratories, Burlingame, CA). After further washing, PBMC were analyzed with FACSAria or FACScalibur flow cytometers equipped with FACSdiva and CellQuest software, respectively (Becton Dickinson Immunocytometry Systems, San Jose, CA).

Cell purification.

Bovine CD4⁺, CD4⁻, CD8⁺, and CD8⁻ T-cell populations were purified from PBMC prior to or after stimulation with SEC1 by positive and negative selection. Briefly, cell suspensions were labeled with an anti-CD4 or anti-CD8 MAb in PBS for 15 min at 4°C. The cells were washed and then incubated with PE-conjugated isotype-specific antibodies in PBS for 15 min at 4°C. After washing, the cells were incubated with anti-PE-coated microbeads (Miltenyi Biotech, Auburn, CA) for 15 min at 4°C. Cell separation was performed with an LS column (Miltenyi Biotech) according to the manufacturer's instructions. In some experiments, CD4⁺ and CD8⁺ T-cell populations were further separated into CD25⁺ and CD25⁻ subpopulations. Each cell population was incubated with the anti-CD25 MAb at 4°C for 30 min. After a washing step, the cells were incubated with isotype-specific anti-mouse immunoglobulins conjugated with fluorescein isothiocyanate (Caltag Laboratories). The cells were washed again, and the various subpopulations were sorted by using the FACSAria cytometer. The purities of sorted subpopulations were determined to be 98.5 ± 0.5% in three separate experiments, as assessed by FC (data not shown).

Real-time PCR.

RNA was extracted from approximately 5 × 10⁶ SEC1-stimulated PBMC or purified T cells by using TRIzol reagent (Life Technologies, Gaithersburg, MD). First-strand cDNA was generated from 1 μg of RNA by using Superscript II reverse transcriptase and oligo(dT) primers (both from Life Technologies). The reverse transcription reaction was performed in a 20-μl volume according to the manufacturer's specifications.

Primers for PCR amplification were designed by Primer Express (version 2.0; PE Applied Biosystems, Foster City, CA) using GenBank sequences (Table 1) . The real-time PCRs were performed by using the SYBR green I dye master mix and ABI Prism 7500 real-time PCR system (PE Applied Biosystems) according to the manufacturer's instructions. After 40 cycles of amplification, a melting curve was generated by slowly increasing the temperature of the reaction mixture at a rate of 0.1°C/s from 60 to 95°C. During this time, the fluorescence was measured continuously to verify the amplification specificity.

Real-time PCR data were analyzed by using sequence detector systems software (version 1.2.2; PE Applied Biosystems). Briefly, the threshold cycle (C_T) was calculated by the sequence detection software as the cycle number at which the ΔRn crossed the baseline. The data were normalized by calculating the ΔC_T (C_T of target − C_T of the internal control [β-actin gene]). Normalized ΔC_T data were used for the comparative C_T method (ΔΔC_T) for relative quantification. Normalized ΔC_T data from each time point of stimulation were compared to data from unstimulated cells (day 0).

Identification of the bovine foxp3 gene.

A partial sequence of the bovine foxp3 gene was obtained from the bovine genome database (24) based on its high similarity to human foxp3 and used to generate the following PCR primers: forward, 5′-CACTGGTTTACACGCATGTTTG-3′, and reverse, 5′-TGACTGAGGCAGGCTGTGTGT-3′. PCRs, using these primers with cDNA produced from SEC1-stimulated PBMC, generated the expected size amplicons (330 bp). Based on the sequence of this product, a reverse primer (5′-GTGCACACCTTACTTCTTGGT-3′) was designed and used with a RACE (rapid amplification of cDNA ends) adaptor primer to obtain the complete cDNA sequence of bovine foxp3, using the FirstChoice RLM-RACE kit (Ambion, Austin, TX) according to the manufacturer's instructions.

Coculture experiments.

Coculture experiments were conducted in Transwell (Corning Life Sciences, Nagog Park Acton, MA) culture plates to determine the ability of soluble factors from SEC1-stimulated PBMC or purified subpopulations to suppress naive PBMC proliferation in response to heat-killed fixed S. aureus. Naive PBMC (10⁵ or 10⁶) were placed in 24-well culture plates (Corning Life Sciences) in the presence or absence of heat-killed fixed S. aureus. Various numbers of SEC1-stimulated PBMC (from 0 to 10⁶) or purified SEC1-stimulated CD4⁺ or CD8⁺ T cells (10⁵) were seeded into the upper chamber of Transwell plates, which were then inserted into 24-well culture plates. After 72 h of culture, [³H]thymidine (1 μCi/well) was added. The PBMC from 24-well plates were harvested 18 h later, and the [³H]thymidine incorporation was quantified as described above. In some experiments, anti-IL-10 and/or anti-TGF-β MAb (1 or 10 μg/ml) or isotype control MAbs were added to 24-well plates to determine whether the suppression was mediated by IL-10 and TGF-β.

Statistical analysis.

Statistical significance was analyzed with the Student t test using OriginPro software (OriginLab, Northampton, MA).

Differential activation of bovine T-cell populations by SEC1.

Bovine PBMC were stimulated for 4 days with various concentrations of SEC1 (0.5 to 10 μg/ml). [³H]thymidine incorporation was measured as an indicator of cell proliferation (Fig. 1A). A near-linear dose response was observed between 50 and 500 ng/ml. Subsequent experiments were conducted using the 5-ng/ml concentration of SEC1, which was near the midpoint of the dose-response curve and generated a response that was ca. 50% of the maximum observed.

The phenotypes of SEC1-stimulated PBMC were analyzed by FC. Before stimulation, the CD4/CD8 ratio, normally 1.18 to 1.82 (32), was initially 1.35 but decreased to 0.91 by day 10 (Fig. 1B). The proportion of γδ⁺ T cells present in the culture dramatically decreased from ∼39 to 7% between days 4 and 10 of exposure to SEC1 (Fig. 1B).

The percentages of CD4⁺ and CD8⁺ T cells expressing naive T-cell markers (CD62L and CD45R) dramatically decreased after 4 days of exposure to SEC1 in culture. This observation was accompanied by a corresponding, but more gradual, increase in cells expressing T-cell activation markers (CD25 and CD26) and CD45R0 (Fig. 1C and D). These data indicated that both CD4⁺ and CD8⁺ T cells were activated by SEC1.

To determine whether the decrease in the percentages of γδ⁺ T cells was attributable to a lack of stimulation by SEC1 or to a difference in the relative rate of proliferation compared to that of CD4⁺ and CD8⁺ T cells, PBMC were labeled with CFSE prior to incubation to monitor the rate of proliferation of each cell population. As expected, unstimulated cells were represented as a single peak of fluorescence (Fig. 2A). Upon stimulation, additional peaks representing PBMC fractions with progressively lower concentrations of CFSE appeared as a result of CFSE dilution during successive rounds of cell division at 4 and 10 days. The rates of cell division by the CD4⁺ and CD8⁺ populations were similar at 4 days based on the distribution of the peaks. At 10 days, very few cells remained among the CD4⁺ and CD8⁺ populations represented by the high fluorescence levels in the initial peaks, indicating that most of CD4⁺ and CD8⁺ T cells had proliferated. In contrast, the majority of γδ T cells contained maximal CSFE levels throughout the entire 10-day period of culture, indicating that few of these cells had proliferated.

The distribution of CD4⁺, CD8⁺, and γδ⁺ T cells in the CFSE-stained fraction representing the most highly divided (lowest CSFE content) cells after 10 days was determined (Fig. 2B). CD4⁺, CD8⁺, and γδ⁺ T cells represented 25, 43.8, and 9.2% of the PBMC, respectively (Fig. 2B). The results indicated that CD4⁺ and CD8⁺ T cells proliferated in response to stimulation with SEC1, with the proliferation of CD8⁺ T cells being more extensive than that of CD4⁺ T cells. This observation could explain, at least in part, the reversal of the CD4/CD8 ratio after 10 days of culture and also indicates that the relative decrease in γδ⁺ T cells after 10 days was due to their failure to proliferate.

Functional characterization of bovine T-cell subpopulations induced by SEC1.

The transcriptional levels of several key activation marker and cytokine genes were analyzed by real-time PCR. The data from SEC1-stimulated and unstimulated PBMC were compared. The relative transcriptional level resulting from this comparison is expressed on a log₂ scale in Fig. 3 and a natural scale in the text. Consistent with the FC results, transcription of CD25 increased (∼10-fold at 10 days) rapidly and dramatically during exposure to the toxin (Fig. 3A). The transcription of CD152 increased (∼10-fold at 10 days), and the pattern of expression was similar to that for CD25. The transcription of CD28 also increased but only modestly (∼1.8-fold) compared to CD25 and CD152 (Fig. 3A). The transcription of IFN-γ and granulocyte-macrophage colony-stimulating factor (GM-CSF) also increased (∼60-fold at 10 days). After induction to maximal levels on 4 days, mRNA levels for both genes were sustained throughout the remainder of the culture period (Fig. 3B). Transcription of IL-2 increased early, peaked at 4 days (∼13-fold), and then gradually decreased to the baseline level by 10 days. Based on these data, one may conclude that CD4⁺ and CD8⁺ T-cell proliferation at later time points (after 4 days) was accompanied by increased transcription of CD25 but not of IL-2. The transcription of IL-12 initially declined (at 2 days) and then increased slightly at later time points before returning to baseline levels on 10 days. A small but persistent increase (∼1.6-fold) in the transcription of IL-4 was observed beginning in 4-day-old cultures. In contrast, the increase in the transcription of IL-5 and IL-13 after 4 days of culture was much more dramatic (ca. 20- to 160-fold) (Fig. 3C). The transcription of TGF-β gradually increased and peaked on 8 days (∼2.8-fold), whereas that of IL-6 and IL-10 decreased during the entire culture period (∼25-fold) (Fig. 3C and D).

The CD4⁺ CD25⁺ and CD8⁺ CD25⁺ T-cell subpopulations were isolated from 10-day-old SEC1-stimulated PBMC cultures (CD4⁺ CD25⁻ and CD8⁺ CD25⁻ T-cell subpopulations were isolated from unstimulated control cultures) and subjected to gene expression studies by using real-time PCR. Compared to unstimulated controls, SEC1 exposure increased the transcription of CD25 and CD152 in both CD4⁺ CD25⁺ and CD8⁺ CD25⁺ T-cell subpopulations (Fig. 4A). A smaller but consistent increase in CD28 transcription (∼1.4-fold) was measured in CD4⁺CD25⁺ T cells, whereas a corresponding relative decline in expression occurred in the CD8⁺ CD25⁺ subpopulation (∼1.2-fold) (Fig. 4A). The transcription of IL-2 increased ∼7-fold in CD8⁺ CD25⁺ T cells; a smaller but consistent increase was observed in CD4⁺ CD25⁺ T cells (∼2-fold) (Fig. 4B). Consistent with data acquired from analysis of SEC1-stimiulated PBMC cultures, the transcription of IFN-γ and GM-CSF was highly increased within both purified T-cell subpopulations (ca. 20- to 40-fold) (Fig. 4B). The transcription of IL-12 was increased in both purified T-cell subpopulations (approximately four- to eightfold) (Fig. 4B). This level of increase was not observed in PBMC exposed for 10 days, suggesting that IL-12 expression among other T-cell subpopulations was dramatically reduced. The transcription of IL-5 and IL-13 increased in both subpopulations, whereas IL-4 transcription decreased in CD4⁺ CD25⁺ T cells and slightly increased in CD8⁺ CD25⁺ T cells (Fig. 4C). The transcription of IL-6 was consistently decreased in both subpopulations (ca. 50- to 200-fold) (Fig. 4C). The levels of TGF-β transcription in either subpopulation stimulated with SEC1 did not change substantially compared to baseline levels in unstimulated controls (Fig. 4D). However, it is noteworthy that baseline TGF-β transcription was inherently high and was comparable to that of β-actin prior to SEC1 exposure (data not shown). These data suggest that, even though the transcription of TGF-β did not change, it remained at a high level of transcription during SEC1 stimulation. The transcription of IL-10 increased in CD4⁺ CD25⁺ T cells (∼10-fold) and decreased in CD8⁺ CD25⁺ subpopulation (∼30-fold) (Fig. 4D). This result is consistent with the overall decline in IL-10 expression in PBMC as demonstrated in Fig. 3D.

Bovine foxp3 identification and expression characterization.

The demonstration that SEC1 exposure upregulates the transcription of CD25, CD152, and IL-10 and maintains constitutive transcription levels of TGF-β in CD4⁺ CD25⁺ T cells suggested that this subpopulation was analogous to T_regs in other species. In T cells from other species, Foxp3 (forkhead/winged helix transcription factor 3) is a key marker for T_regs (38). Since the sequence of the bovine foxp3 gene had not been determined at the time of the study, we cloned and sequenced the bovine orthologue of foxp3. The identified foxp3 sequence was deposited in GenBank (GenBank accession no. DQ322170). The translated protein sequence of bovine Foxp3 was compared to other known orthologues of Foxp3 by using the multisequence alignment program CLUSTALX (20) and showed a high similarity with other orthologues of this protein (human, 90%; monkey, 89%; mouse, 86%; rat, 85%) (4, 14).

Real-time PCR primers for bovine foxp3 were designed (Table 1) and used to determine the level of transcription in SEC1-stimulated PBMC cultures. A gradual increase in the transcription of foxp3 was clearly evident by 6 days of culture, and mRNA levels reached a peak by between 8 and 10 days (ca. four- to fivefold) (Fig. 5A). In other species, Foxp3 is a transcriptional repressor and activator of the IL-2 and CD152 genes, respectively (18). The transcription patterns of IL-2 and CD152 after SEC1 exposure (Fig. 3B and A, respectively) were consistent with the pattern of expression of foxp3 (Fig. 5A). After a rapid induction and peak on 4 days, IL-2 mRNA gradually decreased thereafter, when foxp3 transcription increased dramatically. In contrast, expression of CD152 closely paralleled that of foxp3. Real-time PCR analysis of various purified T-cell subpopulations after exposure to SEC1 indicated that the transcription of foxp3 occurred in CD4⁺ T cells. No transcription increase could be detected in CD8⁺ T cells, regardless of whether or not they expressed CD25 (Fig. 5B).

SEC1-stimulated PBMC inhibit the proliferation of naive PBMC in response to heat-killed fixed S. aureus via soluble factors.

In other species, some T_regs are immunosuppressive through a mechanism involving the secretion of IL-10 and TGF-β. To assess whether SAgs could induce this type of bovine T_reg cell, we conducted experiments to determine whether the exposure of bovine PBMC to SEC1 induces T cells with this property. Naive PBMC, with heat-killed and fixed S. aureus cells, were cocultured with SEC1-stimulated PBMC. Cell-to-cell contact between naive and SEC1-stimulated PBMC was prevented by using a Transwell apparatus. In control cultures without SEC1-stimulated PBMC, naive PBMC proliferated well in response to heat-killed fixed S. aureus (Fig. 6A). The addition of SEC1-stimulated PBMC to the cultures resulted in a dose-dependent suppression of proliferation to levels of ca. 22.51% compared to the control when both cell types were cocultured in a 1:1 ratio (Fig. 6A). IL-10-specific or TGF-β-specific MAbs added to the cultures resulted in partial restoration of the naive PBMC proliferative response to heat-killed S. aureus. Both MAbs used in combination resulted in an additive effect (Fig. 6B). However, the suppressive activity of SEC1-stimulated PBMC was not fully abrogated, suggesting the MAbs only partially neutralized their respective cytokines or the existence of additional mechanisms of suppression induced by SEC1.

To address which T-cell population was responsible for the suppression, CD4⁺ and CD8⁺ T cells were purified from SEC1-stimulated PBMC and used in coculture experiments. Either CD4⁺ or CD8⁺ T cells generated from SEC1 stimulation strongly suppressed naive PBMC proliferation in response to heat-killed S. aureus (Fig. 6C and D). The CD4⁺ suppressive effect was partially abrogated by adding anti-IL-10-specific MAb, anti-TGF-β-specific MAb, or a combination of both MAbs. The IL-10-specific MAb had a small effect on the CD8⁺ T-cell suppression. Used alone, only the higher concentration of this MAb induced a statistically significant reversal (Fig. 6D), which was much less dramatic than that observed with CD4⁺ T cells or with PBMC. The TGF-β-specific MAb did not have a significant effect at either concentration tested. These results indicate CD4⁺ and CD8⁺ T cells mediate their suppression by different mechanisms.