Previous Article | Next Article ![]()
Clinical and Vaccine Immunology, December 2006, p. 1322-1327, Vol. 13, No. 12
1071-412X/06/$08.00+0 doi:10.1128/CVI.00205-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Enteric Diseases and Food Safety Research, National Disease Center, USDA, Agriculture Research Service, Ames, Iowa 50010,1 Institute for Hygiene and Infectious Diseases of Animals, Justus Liebig University, D-35392 Giessen, Germany,2 U.S. Meat Animal Research Center, USDA, Agriculture Research Service, Clay Center, Nebraska 68933,3 Cerner Corporation, Kansas City, Missouri 64117,4 Novartis Animal Health US, Greensboro, North Carolina 274085
Received 2 June 2006/ Returned for modification 1 August 2006/ Accepted 5 October 2006
|
|
|---|
|
|
|---|
Ruminants are important sources of EHEC O157:H7 strains because they frequently shed STEC in their feces (26). Single STEC O157:H7 clones can be repeatedly isolated from a herd (6, 14), implying that these clones persist within that herd. This could result from frequent transmission between and reinfection of different animals (2) or from a true persistence in single animals (3, 7, 8). Experimental infections of calves revealed that several bacterial factors involved in the "attaching and effacing" adherence of E. coli O157:H7 to epithelial cells (e.g., intimin [10], other products of the locus of enterocyte effacement [11], and efa-1 [32]) promote bacterial colonization of the intestinal mucosa. However, recent studies failed to link the induction of attaching and effacing to the establishment of persistent infections in animals (4, 18).
Several lines of evidence support the hypothesis that Stx can suppress the bovine immune response. Stx1 targets bovine peripheral blood (23, 25, 31) and intraepithelial lymphocytes (IEL) (22) in vitro. Stx1 binds to Stx receptor-expressing (Gb3-/CD77-positive) lymphocytes at early activation stages (31) and blocks the proliferation of distinct lymphocyte subpopulations (CD8
+ T cells, B cells) in vitro (25). A significant portion of bovine CD8
+ IEL expresses Stx-receptors in situ (22). Inoculation of ligated ileal loops in 2-week-old calves with Stx1-producing STEC reduces the number of CD8
+ IEL within 12 h (24). Direct evidence for a suppression of immune cell functions in the course of STEC infections in cattle, as observed during experimental STEC infections in pigs (5), has been lacking. Although bovine lymphocytes are sensitive to minute concentrations of Stx1 and Stx2 (1, 12, 13, 25), Stxs do not completely abolish the development of specific immune responses in naturally and experimentally infected cattle. Antibodies against O157 lipopolysaccharide (LPS), Stx1, and Stx2 are frequently detected in bovine sera and mucosal secretions (colostra) (17, 28). Since Stx1 suppresses bovine lymphocytes in vitro without inducing cell death (25), we hypothesized that in vivo, Stxs primarily reduce or delay the host's cellular immune response, thereby generating an opportunity for STEC bacterial colonization. The objectives of the present study were to monitor the levels and durations of fecal shedding of Stx-producing (Stx+) and Stx-negative (Stx) E. coli and to examine specific cellular and humoral immune responses in calves experimentally inoculated with either human pathogenic Stx2+ or Stx E. coli O157:H7 bacteria.
|
|
|---|
Inoculations. The strains used in this study were EHEC O157:H7 strain 86-24, a streptomycin-resistant mutant of an Stx2-producing strain isolated from an outbreak in Washington state (36); E. coli O157:H7 strain 87-23, a nalidixic-resistant mutant of a strain isolated from the same outbreak as 86-24 and lacking both Stx1 and Stx2 genes (19, 36); and, as a control strain, a nalidixic acid-resistant mutant of E. coli strain 123 (O43:H28), a porcine isolate which does not produce Stx and is not pathogenic in cattle (9). Escherichia coli cultures were prepared and stored at 80°C as previously described (9).
Calves (five calves/group) were inoculated intrarumenally, on day 0 and day 21, with approximately 1010 CFU of strain 86-24 (Stx2+O157), 87-23 (StxO157), or 123 (control) bacteria in 100 ml of trypticase soy broth (Becton Dickinson, Cockeysville, MD) as previously described (11, 33). Three weeks after the second inoculation (day 42), all calves were similarly inoculated with Stx2+O157. This third inoculation of all calves with the same Stx+O157 strain is referred to as the "challenge" to clearly differentiate it from the two initial inoculations of calves in different groups with different strains.
Bacteriology. Freshly isolated fecal samples (approximately 2 g) were diluted 1:10 and serially diluted in sterile phosphate-buffered saline. Sorbitol-negative STEC O157:H7 bacteria were quantitated on sorbitol-MacConkey agar (Oxoid, Basingstoke, United Kingdom) supplemented with streptomycin (200 µg per ml) or nalidixic acid (20 µg per ml) to select for strains 86-24 or 87-23, respectively. Randomly selected antibiotic-resistant, sorbitol-negative colonies were tested for O157 LPS antigen by slide agglutination (Oxoid). E. coli strain 123 bacteria were quantitated on MacConkey agar (Becton Dickinson, Cockeysville, MD) containing 20 µg per ml of nalidixic acid. Enrichment cultures were performed by adding up to 10 g of feces to trypticase soy broth containing 0.01% bile salts 3 (Difco, Detroit, MI) to a final volume of 100 ml. These cultures were incubated overnight, without shaking, at 37°C. Samples (100 µl) from enrichment cultures of feces that were negative by direct plating were incubated on sorbitol-MacConkey agar with the appropriate antibiotic. Enrichment plates were scored as either positive or negative and were not quantitated. Positive enrichment cultures were arbitrarily assigned a value of 10 CFU g1, and those which were negative were given a value of 1 CFU/g. Student's t test was used to determine statistical significance.
Detection of plasma antibodies to O157 LPS and Stx2. Titers against O157 LPS were determined by a blocking enzyme-linked immunosorbent assay (20), using monoclonal antibody MARC 13B3 (38) and highly purified LPS from E. coli O157:H7 (strain CDC EDL 933). MARC 13B3 lacks cross-reactivity with other bacteria, such as Brucella spp. and Yersinia enterocolitica, a problem often found with polyclonal antisera against O157. Endpoints were defined using receiver-operator-characteristic analysis of 1:4 dilutions of known positive and negative serum samples (20).
Vero cell cytotoxicity neutralization assays were performed as previously described (15). Briefly, dilutions of plasma were added to samples containing Stx2. After incubation for 1 h at 37°C, the plasma-toxin mixtures were added to Vero cells and incubated at 37°C for 48 h. The cells were then stained for viability with trypan blue. Antisera obtained from four calves hyperimmunized by subcutaneous injections with 100 µg of a noncytotoxic amino acid substitution mutant of Stx2e (15) were pooled and used as a positive anti-Stx control serum. Antibodies against Stx2e cross-reacted with and neutralized Stx2 in our system.
Preparation of antigen for peripheral blood mononuclear cell (PBMC) stimulation. Heat-killed Stx2+ E. coli O157:H7 bacteria (HK-O157) were used for in vitro stimulation. One milliliter containing 1010 CFU of E. coli O157:H7 strain 86-24 (harvested from a fresh overnight culture) was boiled for 15 min. The cells were pelleted by centrifugation and resuspended in 1 ml of complete media (RPMI; 25 mM HEPES and L-glutamine [Gibco, Carlsbad, CA], 10% fetal calf serum, and 50 µg gentamicin per ml [Sigma, St. Louis, MO]). Aliquots of the cells were tested for sterility by plating them on trypticase soy agar plates and stored at 80°C until needed.
PBMC stimulation. PBMCs were isolated from whole blood collected at weekly intervals. The blood was centrifuged for 25 min at 1,170 x g. Plasma samples were removed and stored at 80°C for further analysis. Buffy coat cells were subjected to two rounds of hypotonic red blood cell lysis. PBMCs were then isolated by Percoll (Pharmacia, Piscataway, NJ) density gradient (P = 1.08) centrifugation at 450 x g for 40 min. Cells were washed twice in complete media. Cells were then diluted to 6.6 x 106 cells per ml and aliquoted into 96-well plates at 150 µl of cells (106 cells) per well. Stimulants (HK-O157, equivalent to 5 x 104 CFU E. coli O157:H7 cells, or 1 µg per ml ConA), 50 µl per well, were added to each well. Plates were incubated for 72 h in 5% CO2 at 39°C. PBMCs were then pulsed with media containing 50 µCi of [3H]thymidine and incubated for an additional 18 h. Cells were harvested with a Harvester 96 (Tomtech, Orange, CN) and analyzed using a 1450 Microbeta Plus scintillation counter (Wallac, Gaithersberg, MD). The stimulation index (SI) is the mean for three wells of cells incubated with stimulant divided by the mean for three wells incubated with media alone. Cell preparations in which the SI for ConA was less than 10 times that for cells incubated with media alone were considered poor and were excluded from further analysis. Statistical analysis was performed using the SAS software package and the analysis of variance technique.
|
|
|---|
The inoculum strain was recovered in higher numbers and for a longer time from calves initially inoculated with Stx2+ O157:H7 strain 86-24 than from calves inoculated with either of the other two E. coli strains (Fig. 1). More than 102 CFU of Stx2+O157 bacteria/g of feces were recovered for 9 and 14 days after the first and second inoculations, respectively. Fecal shedding of StxO157 decreased at a rate similar to that of the control bacteria, with both strains at <102 CFU g1 of feces by the fifth day after each inoculation. The differences in fecal shedding between the Stx2+ and StxO157 strains were statistically significant (P < 0.05 by Student's t test) between days 2 and 13 after the first inoculation and days 22 and 27 after the second inoculation.
![]() View larger version (9K): [in a new window] |
FIG. 1. Fecal shedding of E. coli by calves inoculated with different strains. Numbers of inoculum bacteria recovered from feces of calves inoculated with Stx2+ E. coli O157:H7 strain 86-24 (Stx2+), Stx E. coli O157:H7 strain 87-23 (Stx), or control E. coli O43:H28 strain 123 (Control) are shown. Samples which were negative by direct culturing methods (<100 CFU g1) were tested by enrichment culturing. All values below the dashed line represent enrichment levels. Samples which were positive by enrichment analysis were given an arbitrary value of 10 CFU g1, and those which were negative were given a value of 1 CFU/g. Arrows indicate dates of inoculation. Data are presented as mean numbers (± standard errors of the means) of CFU g1 for the indicated groups.
|
![]() View larger version (15K): [in a new window] |
FIG. 2. Fecal shedding of the challenge strain by three groups of calves which were challenged with Stx2+ O157:H7 strain 86-24 following prior inoculations with different strains of E. coli. The arrow indicates the date of the challenge. Data are presented as mean numbers (± standard errors of the means) of CFU g1 of sorbitol-negative Strr O157:H7 bacteria in the feces of the indicated groups. Group designations are same as those in the legend to Fig. 1.
|
16) of anti-O157 LPS, and the highest titer in this group was seen at week 4 in the calf which had serum antibodies to LPS O157 prior to the first inoculation. Calves in all three groups developed antibodies to O157 LPS following challenge with Stx2+O157 (Fig. 3). None of the calves in any of the three groups developed antibodies against Stx2.
![]() View larger version (9K): [in a new window] |
FIG. 3. Development of antibody titers against O157 LPS in calves inoculated with different E. coli strains. Data are shown as the means (± standard errors of the means) of log2 titers of plasma antibodies. For the designations of calf groups, see the legend to Fig. 1. Arrows indicate the dates of the initial inoculations (small arrows) and challenge (large arrow).
|
![]() View larger version (26K): [in a new window] |
FIG. 4. Comparison over time of the lymphoproliferative responses to heat-killed Stx2+ E. coli O157:H7 bacteria (HK-O157) in calves inoculated with different E. coli strains. Calves were inoculated twice (on days 0 and 21) with Stx2+ E. coli O157:H7 strain 86-24 (Stx2+), non-Shiga-toxigenic E. coli O157:H7 strain 87-23 (Stx), or control E. coli O43:H28 strain 123 (Control) prior to a final challenge (at 7 weeks postinoculation) with Stx2+ E. coli O157:H7 strain 86-24. Results are shown as the means (± standard errors of the means) of the SIs obtained with PBMCs from five calves per group. The index for an individual animal was determined by dividing the mean counts per minute for three wells stimulated with HK-O157 by the mean counts per minute for three wells of nonstimulated cells. Lymphoproliferative responses to ConA stimulation are shown in the insert. Data obtained with blood samples during weeks 1, 2, and 3, during weeks 4 and 6, and postchallenge (during weeks 7, 8 and 10) were combined within those respective groups.
|
|
|
|---|
Our studies were not designed to examine the mechanism underlying the observed immunosuppression. However, the observation that the PBMCs obtained from the different inoculation groups did not significantly differ in their proliferative responses to ConA provides preliminary evidence that the effect of STEC infection on the bovine immune system was not generalized. Furthermore, the observed immunosuppressive effects of STEC infections were restricted to antigen-specific in vitro PBMC proliferative responses. This contrasts with a generalized suppression of T cells described for mice and piglets, which is characterized by a drop in the reactivities of lymphocytes to ConA following injection with purified Stx2 and inoculation with Stx1+ E. coli strain O111:NM, respectively (5, 34). Differences in levels of systemic Stx may be responsible for the observed differences in the effects of Stx on the immune system in the different studies. The mouse and piglet studies were more likely to involve high systemic concentrations of Stx. Mice were intravenously injected and piglets were orally inoculated with high doses of Stx+ bacteria at 24 h of age, when gut closure may not have been complete and the likelihood of toxin absorption from the intestine was increased. Consequently, the effects of Stx on the immune system in these models may have been high enough to affect "general" functions, like mitogenic responsiveness. The effects of Stx in calves infected at 6 to 8 weeks of age (notwithstanding the high inoculum doses) may better reflect the effects of "physiologic" infections that occur in the field.
Thus, we interpret the present findings as support for our hypothesis that Stxs are immunosuppressive factors which play a role in promoting STEC colonization in the bovine host primarily by reducing or delaying the host's antigen-specific cellular immune response against the bacteria.
As previously postulated (31), the immunosuppressive effects of STEC infections appeared to result from a prevention of the onset of an immune response rather than the downregulation of an established one. The development of an in vitro PBMC response was prevented when Stx2 was present at the time that the naïve O157:H7 E. coli-specific immune cells were first stimulated by the initial inoculations (i.e., in calves initially inoculated with Stx2+O157). In contrast, the Stx2+O157 strain did not exert an immunosuppressive effect after the immune response was efficiently induced by two inoculations with StxO157. Calves initially inoculated with StxO157 showed a continued proliferative response that, similar to an anamnestic response, was even more pronounced after they were challenged with Stx2+O157.
Remarkably, inoculation of calves with Stx2+O157, on the contrary, did not hinder the development of a humoral response against O157 LPS (17; this study). B-cell responses in calves apparently were less sensitive to STEC-induced suppression than in vitro PBMC proliferative responses (25), particularly when the B cells were directly activated in vivo by T-cell independent B-cell antigens, like LPS. The absence of antibodies against Stx2 in our calves, which is consistent with previous observations after experimental STEC O157:H7 inoculations in cattle (17), can therefore not be explained by a suppression of B cells.
Despite the different immune responses in the three groups of calves following the initial inoculations, the shedding patterns for Stx2+O157 bacteria were similar in all groups in the first weeks after the calves were challenged with the Stx2+O157 strain. Neither the induction of O157 antibodies in the Stx2+O157 group nor the induction of O157 antibodies and HK-O157-specific lymphocytes in the StxO157 group reduced the duration or magnitude of shedding after the challenge compared to what was found for the control group during the observation period. Further experiments will be necessary to determine how O157:H7 E. coli-specific cellular immune responses affect long-term and intermittent STEC shedding (3, 8).
The effects of STEC infections on the development of specific cellular immune responses must be addressed in designing vaccines aimed at reducing STEC O157:H7 infections in cattle, especially vaccines directed against Stx, which is the only virulence factor shared by all STEC strains.
Disclaimer: the use of names is necessary to report factually on available data; however, the USDA neither guarantees nor warrants the standard of any product, and the use of a name by the USDA implies no approval of the product to the exclusion of others that may also be available.
Published ahead of print on 18 October 2006. ![]()
|
|
|---|
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»