Previous Article | Next Article ![]()
Infection and Immunity, June 2006, p. 3507-3512, Vol. 74, No. 6
0019-9567/06/$08.00+0 doi:10.1128/IAI.01726-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Animal Health, National Veterinary Institute, Oslo, Norway,1 Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, New Yok 148532
Received 21 October 2005/ Returned for modification 19 January 2006/ Accepted 29 March 2006
|
|
|---|
|
|
|---|
S. aureus produces a family of related superantigens (SAgs) that includes several staphylococcal enterotoxins (SE) and toxic-shock-syndrome toxin (TSST) variants (6). Staphylococcal SAgs are prototypical microbial superantigens, characterized by their ability to bind to major histocompatibility complex class II molecules and specific Vß segments of
ß T-cell receptors (32). SAgs bypass the antigenic specificities of T-cell receptors and stimulate abnormally large numbers of T cells. At extremely low concentrations, these molecules can induce profound disturbances in the homeostasis of the immune system (17, 44). These toxins play a critical role in human toxic shock syndrome and food poisoning, but their possible role in the onsets or maintenance of other diseases is not well understood (41).
Geographical differences exist in the occurrence of SAg-producing strains causing mastitis (22). Kenny et al. (19) found that S. aureus strains producing enterotoxin D (SED) alone or in combination with S. aureus enterotoxin C (SEC) and TSST-1 accounted for 22% of the isolates from New York State. In Norway, a previous study showed that 58% of S. aureus isolates expressed SAgs and that the production of SEC and TSST-1 in combination predominated (40). Some reports have suggested that S. aureus strains that express SEC and TSST-1 cause severe clinical mastitis unresponsive to therapy (12, 27), whereas other investigations have failed to find a significant correlation between SAg production and clinical manifestations of mastitis (23, 40).
Although the in vitro effect of some staphylococcal SAgs on bovine cells has been studied in detail (5, 8, 9, 45), evidence of in vivo production and the effect of these toxins on clinical disease is scarce. Niskanen et al. detected SEC, but not S. aureus enterotoxin A (SEA), in milk samples from experimentally infected cows and showed that the infusion of SEA caused inflammatory reactions in the udder (31). In a recent study, Kuroishi et al. measured antibodies to SEC and TSST-1 in mammary gland secretions and observed the inflammatory response after the intramammary infusion of these toxins (21). They found that SEC, but not TSST-1, had an impact on the severity of mastitis.
Studies on the effect of SED on bovine lymphocytes are lacking, as is information on the ability of specific bovine antibodies to modulate the effect of SED. The recruitment of neutrophils from blood to milk and their ability to take up and kill bacteria are important factors in the outcome of intramammary infections. An inhibitory effect of SEA on bovine neutrophils in an in vitro bactericidal assay has been reported (29), but there are few other reports on the effect of staphylococcal SAgs on neutrophil function.
The aim of the present study was to investigate the secretion of SED in experimental bovine S. aureus mastitis and to observe whether a measurable humoral immune response against this enterotoxin was generated during the course of infection. Experiments were performed to ascertain whether purified SED exerted a mitogenic effect on bovine lymphocytes or influenced neutrophil function in vitro.
|
|
|---|
Animals. Seventeen American Holstein Friesian cows in mid-lactation were used. The cows were in their first to third lactation. None of the animals had a history of clinical mastitis due to S. aureus, and all mammary quarters tested negative for S. aureus. One mammary gland from each cow was selected for infection on the basis of a somatic cell count (SCC) of less than 200,000 cells/ml and the absence of pathogens. After the cows were milked, the teat of the selected quarter was disinfected, and 1 ml of a bacterial suspension containing between 50 and 250 CFU of bacteria was infused using a teat cannula. For the in vitro assays, blood samples from clinically healthy Norwegian Red cows were used.
Sample collection and analysis. For the first 21 days postinoculation and thereafter at weekly intervals for 9 weeks, milk samples were collected for SCC analysis and bacteriological examination. Somatic cell counts were determined using a Fossomatic cell counter (Foss Electric, Hillerød, Denmark), while bacterial counts were determined by plating milk samples in triplicate on tryptic soy agar containing 5% bovine blood. Milk and blood samples for antibody analysis were collected at weeks 3, 6, 9, and 12. The milk samples were centrifuged, the fat was aspirated, and the cell pellet was discarded before both milk samples and sera were stored at 20°C.
Serological methods. The total immunoglobulin G (IgG) anti-SED antibody response in serum was measured using an indirect enzyme-linked immunosorbent assay. A solution of 1.0 µg/ml SED (Toxin Technology, Sarasota, Fla.) was prepared in 50 mM carbonate-bicarbonate buffer (pH 9.6), and 0.1 ml was added to wells of MaxiSorp microtiter plates (Nunc, Roskilde, Denmark). Serum was diluted 1:500 in phosphate-buffered saline supplemented with 0.05% Tween 20 (PBST), and 0.1 ml was added to each well. Protein G conjugated to alkaline phosphatase (Sigma Chemical Co., St. Louis, Mo.), diluted 1:300 in PBST, was used as a conjugate. Both test sera and conjugate were incubated at room temperature for 2 h. P-nitrophenyl phosphate tablets (Sigma) dissolved (1 mg/ml) in 1 M diethanolamine buffer (pH 9.8), containing 0.5 mM MgCl2 and 0.02% sodium azide, were used as a substrate. The absorbance at 405 nm was read, and the reaction was stopped when the optical density (OD) of a pooled positive control serum sample was 1.0. The total IgG anti-SED antibody response in milk was measured as for serum, except that 0.1 ml of milk diluted 1:10 in PBST was added to each well.
An indirect enzyme-linked immunosorbent assay using monoclonal antibodies against bovine IgG1, IgG2, or IgM (18) was used to evaluate the isotype composition of anti-SED antibodies. Sera diluted 1:500 were added to microtiter plates coated with SED as described above, and mouse monoclonal antibodies specific for either IgG1, IgG2, or IgM were added at a 1:100, 1:200, or 1:200 dilution, respectively. Rabbit anti-mouse IgG conjugated to alkaline phosphatase (Sigma), diluted 1:500 in PBST, was used as a conjugate. All incubations were performed at 37°C for 2 h. The development and reading of the plates was performed as described above.
Detection of SED. Milk samples from 10 cows at day 0 and day 2 after inoculation were tested for the presence of SED. SED was detected using Ridascreen Set A, B, C, D, E (R-Biopharm, Darmstadt, Germany) according to the manufacturer's recommendations.
Lymphocyte proliferation assay. The lymphocyte proliferation assay was performed according to the method of Larsen (24). Three repeated experiments were performed, using blood samples from four cows. Briefly, 200 µl of heparinized whole blood diluted 1:5 in growth medium (RPMI 1640; Life Technologies, Rockville, Mass.) containing L-glutamine and gentamicin was added to 96-well tissue culture plates (Costar, Corning, Inc., Corning, Calif.) containing either 10 µg/ml concanavalin A (Sigma), growth medium, or purified SED (Toxin Technology). The plates were incubated for 3 days at 37°C in a 5% CO2 atmosphere before methyl-[3H]thymidine (Amersham Pharmacia Biotech, Buckinghamshire, England) was added. The cells were harvested 24 h later, and the incorporation of methyl-[3H]thymidine was measured in an automatic liquid scintillation counter (United Technologies Packard, Downers Grove, Ill.). The proliferative response was converted to median counts per minute (CPM) of triplicate samples. In order to evaluate the inhibitory effect of anti-SED antibodies on lymphocyte proliferation, 10 µl of a pooled serum sample taken from three cows prior to and 6 weeks after inoculation or 10 µl of SED antiserum from an immunized sheep was added to the wells. Fetal calf serum was used as a negative control.
Sheep anti-SED antiserum. An adult female sheep was immunized subcutaneously twice weekly with 1 to 5 µg of SED in saline for a period of 4 weeks. In weeks 5 to 8, the sheep received 25 to 50 µg of SED emulsified in incomplete Freund's adjuvant, and after a 2-week rest period, the sheep was bled.
Measurement of bovine gamma interferon (IFN-
).
Heparinized blood samples from five clinically healthy cows were incubated with SED concentrations ranging from 10 pg/ml to 500 ng/ml in a cell culture plate at 37°C in an atmosphere containing 5% CO2. After incubation, the plate was centrifuged, and the amount of IFN-
in the supernatant was determined using a BioSource Bovine IFN-
kit (BioSource, Nivelles, Belgium). The level of IFN-
was found by relating the OD values to those of a standard curve made using known amounts of recombinant bovine IFN-
.
Neutrophil function assays. Neutrophils from peripheral blood samples were isolated as described previously (4). Isolated neutrophils and heparinized whole-blood samples from 8 to 13 cows were preincubated with various concentrations of SED for 60 min at 37°C and 38.5°C, respectively. Thereafter, a bactericidal assay using 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) was performed essentially as described by Stevens et al. (16, 37). Pretreated neutrophils were incubated with opsonized, live bacteria of S. aureus strain M60 at a bacteria-to-cell ratio of 50:1, and the absorbance at 550 nm was read. The proportion of bacteria killed was determined by comparing the OD with that of a standard curve, and the results are reported as percentages of bacteria killed. A Phagotest kit (Orpegen Pharma, Heidelberg, Germany), containing unopsonized fluorescein isothiocyanate-labeled Escherichia coli isolates, was used according to the manufacturer's instructions, except that the temperature was kept at 38.5°C. The results are reported as percentages of phagocytosing granulocytes, measured by flow cytometry (14). A whole-blood flow-cytometric respiratory burst assay was performed as described by Kampen et al. (15), except that S. aureus strain M60 was used. The results are reported as increases in the geometric means of the fluorescence intensity of the granulocytes.
Statistical analysis. The results are presented as means with standard errors of the means (SEMs). Student's t test was used for comparison of antibody levels. The effects of different concentrations of SED on lymphocyte proliferation and neutrophil function in vitro were assessed using the Tukey-Kramer honestly significant difference test. All tests were performed with a significance level of 5%.
|
|
|---|
![]() View larger version (8K): [in a new window] |
FIG. 1. Mean and individual values for bacterial counts (A) and somatic cell counts (SCC) (B) in milk samples from 17 cows infected with S. aureus strain M60.
|
![]() View larger version (8K): [in a new window] |
FIG. 2. Mean and individual values for the total levels of IgG against SED in sera (A) and milk samples (B) from 17 and 13 cows, respectively, infected with S. aureus strain M60.
|
![]() View larger version (8K): [in a new window] |
FIG. 3. Mean serum levels of IgG1, IgG2, and IgM against SED in 13 cows infected with S. aureus strain M60. Bars indicate SEMs.
|
Mitogenic activity of SED.
The lymphocyte proliferation assay showed that SED caused a significant, dose-dependent proliferation of bovine lymphocytes after 72 h of incubation starting at doses of 0.1 pg/ml and peaking at a count of 52,218 at a concentration of 100 ng/ml (Fig. 4). The addition of 10 µl of sheep anti-SED serum per well neutralized the lymphoproliferative effect of SED for concentrations of SED up to 100 ng/ml (Fig. 5). The same volume of sera, taken from cows prior to intramammary inoculation, had an inhibitory effect on proliferation for SED concentrations up to 10 ng/ml. Sera taken from animals after infection could reduce the effects of SED concentrations of 50 ng/ml (Fig. 5). Stimulation by SED caused a dose-dependent production of IFN-
in whole blood when evaluated after 24 h of incubation (Fig. 6).
![]() View larger version (5K): [in a new window] |
FIG. 4. Proliferative responses of bovine lymphocytes stimulated with SED in a whole-blood assay. Values are means from three repeated experiments with four cows, expressed as CPM of incorporated [3H]thymidine after 72 h. Bars represent SEMs.
|
![]() View larger version (19K): [in a new window] |
FIG. 5. Proliferative responses of bovine lymphocytes stimulated with SED in a whole-blood assay. Fetal calf serum was used as a control, and pooled sera taken from cows prior to intramammary infection with the SED-producing S. aureus strain M60, pooled sera taken from cows after the intramammary infection, or anti-SED antiserum from a vaccinated sheep was added to the wells in order to investigate their inhibitory effect on SED-induced lymphocyte proliferation. Values are medians of three repeated experiments, expressed as CPM of incorporated [3H]thymidine after 72 h.
|
![]() View larger version (11K): [in a new window] |
FIG. 6. Mean IFN- production levels of bovine lymphocytes in whole-blood samples taken from five cows and stimulated with SED in vitro. Bars indicate SEMs.
|
|
View this table: [in a new window] |
TABLE 1. Results for neutrophil function assays used in this study
|
|
|
|---|
Many studies have reported the secretion of staphylococcal enterotoxins by bovine mammary isolates of S. aureus when the bacteria are cultured under laboratory conditions. The detection of an antibody response to SED in serum and milk samples following intramammary infection with S. aureus strain M60 and the finding of toxin in selected milk samples provide evidence that SED is secreted in the bovine mammary gland during infection. The relative importance of clones expressing SED as the cause of mastitis varies geographically and over time. Strains expressing SED were prevalent in New York State in the 1990s (19), and strains carrying the sed gene are still prevalent in the United States (22). More recently, clones expressing both SEC and TSST-1, most likely harboring the pathogenicity island SaPIbov (10), seem to predominate among bovine isolates in many areas (13, 36). However, the relative contributions, if any, of the different SAgs in the pathogenesis of mastitis are poorly understood.
IgG2 was the principal immunoglobulin found against SED in serum after the experimental intramammary infection. The finding is in accordance with what has been reported for antibodies against staphylococcal alpha-toxin in cows (18). IgG2 is a major opsonin for the phagocytosis of S. aureus by bovine blood neutrophils (2) and has been shown to be cytophilic for bovine blood neutrophils and ovine mammary neutrophils (25, 43). The occurrence of low levels of antibodies in the cows at day 0 suggests prior exposure to SED. Although this did not appear to influence the clinical course of the disease or the total level of induced antibodies, the possibility that the predominance of IgG2 might be attributed to a secondary response to the antigen could not be excluded. Also, because the first blood samples were taken 3 weeks postinfection, an early rise in IgM antibodies against SED might have gone undetected.
SED was detected in mastitic milk samples from two cows at concentrations of 5 to 10 ng/ml, and these were the two animals that displayed the highest anti-SED titers in milk samples after infection. It is possible that in vivo, SED secreted by bacteria binds to a major histocompatibility complex class II expressed by cells found in mammary epithelium and connective tissue so that little becomes free in milk (11). Niskanen et al. (31) detected SEC in mastitic bovine milk samples containing 100 to 1,000 CFU of S. aureus per milliliter. They could not, however, detect SEA in infected mammary secretions when using an assay with a sensitivity of 1 µg/ml, although the in vitro production of this toxin was higher than that of SEC. This might reflect in vivo production levels of toxins or various absorption levels of secreted toxin in the gland. Differences in the abilities of SAgs to cross mucosal barriers have been described (35), but these aspects have not been investigated in the bovine udder.
Concentrations of SED in picograms per milliliter caused proliferation of bovine lymphocytes in a whole-blood assay. The potencies of staphylococcal SAgs differ with toxin type and the origin of the cells used in the proliferation assay, but generally concentrations as low as 0.1 to 100 ng/ml have mitogenic effects (3, 45). In addition to causing proliferation of T cells and an uncontrolled release of cytokines like tumor necrosis factor alpha, interleukin-2, and IFN-
(45), SAgs can induce populations of suppressor cells and reduce humoral immune responses (7, 34). In the present study, the rise in IFN-
production with increasing SED concentrations corresponded with the findings in the lymphocyte proliferation assay. Lymphocytes from the mammary glands of S. aureus-infected cows exhibit decreased responses to lectins and staphylococcal antigens compared to T cells from uninfected cows (33).
SED concentrations up to 300 ng/ml did not influence neutrophil function in vitro, while in the bactericidal assay the two highest concentrations tested appeared to inhibit bacterial killing. In a previous report (29), high SEA concentrations resulted in reduced bacterial killing in an MTT-based bactericidal assay. The SED concentrations required to exert an effect in the MTT bactericidal assay were more than a millionfold higher than the concentrations needed to induce proliferation of bovine lymphocytes. Studies reporting an influence of SAgs on human or bovine neutrophils postulate an indirect effect mediated through proinflammatory cytokine production by mononuclear cells (20, 28). As none of the SED concentrations tested seemed to have any effect in the whole-blood assays in the present study, the apparent effect of high concentrations of SED in the bactericidal assay remains unclear.
In the present study, serum antibodies were able to reduce the mitogenic effect of SED in vitro. The finding that antiserum from a sheep immunized with SED inhibited the proliferative effect of SED in vitro raises the question of a possible role of SAgs as antigens in vaccines against S. aureus diseases. Vaccines based on SE toxoids have been evaluated in models of SE-induced toxicosis (26, 38) and recently in murine models of S. aureus infection (30). Although the contribution of staphylococcal SAgs to the pathogenesis of mastitis remains ambiguous, the finding that SED is secreted in vivo and can be detected in mastitic secretions suggests that SAgs may be involved in the disease. Further research is needed to determine whether SAgs play an active role in the pathogenesis of S. aureus intramammary infections in cattle and merit inclusion as antigens in a future S. aureus mastitis vaccine.
Present address: Department of Agriculture and Food, Abbotstown, Dublin 15, Ireland. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»