|
|
||||||||
Published online before print August 2, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

,1
* Université Claude Bernard, INSERM E230, IFR-62, Faculté de Médecine Laennec, Lyon, France; and
Hospices Civils de Lyon, CH Lyon-Sud, and
Hôpital Neurologique, Laboratoire dImmunologie, Lyon, France
1 Correspondence: Laboratoire dImmunologie, Hôpital Neurologique, 59 bd Pinel, 69677 Bron Cedex, France. E-mail: guillaume.monneret{at}chu-lyon.fr
ABSTRACT
The severity of Staphylococcus aureus sepsis is positively associated with staphylococcal enterotoxin A (SEA) and negatively associated with the enterotoxin gene cluster (egc), which encodes five staphylococcal enterotoxins [1
]. We postulated that the variable, clinical severity of S. aureus sepsis might be a result of differences in the inflammatory properties of staphylococcal superantigens. We therefore compared the inflammatory properties of SEA with those of staphylococcal entérotoxin G (SEG), a member of the five egc superantigens. We found that SEA and SEG had similar superantigenic properties, as they induced CD69 expression on T lymphocytes and selective expansion of Vß subpopulations. Contrary to SEG, however, SEA induced a strong proinflammatory/Th1 response, including TNF-
and MIP-1
production. These results suggest that the association of SEA with the severity of S. aureus septic shock, characterized by a deleterious, inflammatory cascade, may be explained partly by the specific proinflammatory properties of this superantigen.
Key Words: superantigens bacterial cytokines chemokines Staphylococcus aureus
INTRODUCTION
Staphylococcus aureus is a versatile and potent human pathogen [2 , 3 ]. S. aureus infections can be subdivided schematically into suppurative diseases such as soft-tissue infections, abscesses and sepsis, and toxin-mediated syndromes such as the toxic shock syndrome (TSS) [2 ]. TSS toxin 1 (TSST-1) has been linked to menstrual TSS and in conjunction with staphylococcal enterotoxin B (SEB), SEC, to nonmenstrual TSS [4 , 5 ]. All of these staphylococcal toxins act as superantigens (SAGs), which directly cross-link certain TCR Vß domains with conserved structures in MHC Class II molecules [6 ]. These interactions trigger a strong Th1 proinflammatory response and activate large subpopulations of T lymphocytes. Clinical manifestations of TSS include fever, arterial hypotension, cutaneous signs (rash and then desquamation), and multiorgan failure [4 , 7 ].
SAG genes are not restricted to S. aureus strains associated with toxin-mediated diseases. In a recent study, Peacock et al. [8 ] found that the sea gene was significantly more prevalent in invasive S. aureus isolates recovered by blood culture than in noninvasive isolates. Likewise, we have previously reported that sea is associated with more severe S. aureus infections and especially, septic shock. In constrast to sea, the presence of other superantigenic enterotoxin genes (seg, sei, selm, seln, and selo) forming the enterotoxin gene cluster (egc) correlates positively with colonization and negatively with the gravity of infection [1 , 9 , 10 ]. Several studies have shown that the egc locus is frequent in S. aureus colonization isolates [11 12 13 ]. For instance, 56% of nasal isolates analyzed by Becker et al. [12 ] harbored the seg and sei genes. These data suggest that differences in the proinflammatory potential of SAG enterotoxins might explain the variable severity of S. aureus sepsis. For example, contrary to SEG and SEI, SEA might trigger a proinflammatory/Th1 cascade responsible for shock. Septic shock, which is mainly caused by S. aureus, profoundly disturbs immune homeostasis by inducing a potent initial systemic inflammatory response, which is rapidly followed by an anti-inflammatory process [14 ]. The pro-/anti-inflammatory balance and the Th1/Th2 ratio are now used to investigate host immunoinflammatory status during septic shock [15 , 16 ].
The aim of this study was to compare the effects of SEA and SEG on inflammatory mediator release by human mononuclear cells.
MATERIALS AND METHODS
Reagents
SEA was purchased from Sigma-Aldrich (St. Louis, MO). SEG was produced by molecular techniques. Briefly, the seg DNA sequence was introduced into S. aureus plasmid pLUG345 with his-Tag (pLUG274::agr1-sa nt18191475; P3 promotor to start codon)::StuI::BamHI cloning sites::linked to nt 1095751 rnaIII (transcription terminator) [17
]. The resulting plasmid was transformed in S. aureus RN6390, a strain normally producing no superantigens. Recombinant SEG was purified with NiNTA columns. The purity of SEG and SEA preparation was assessed by SDS-PAGE (Fig. 1
). The Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD) showed that the endotoxin content of the SEA and SEG preparations was, respectively, <0.05 and <0.5 endotoxin units/ml.
|
Flow cytometry
CD69 assay
As described previously, we assessed T cell activation upon SAG challenge by measuring surface CD69 expression [19
]. Briefly, whole blood was incubated with SEG or SEA (100 ng/mL) in Eagles minimum essential medium containing 5% heat-inactivated FCS (Gibco Invitrogen, Paisley, UK) for 24 h at 37°C in humidified air with 5% CO2. Culture medium and PHA (20 µg/ml) were used as negative and positive controls, respectively. After ammonium chloride erythrocyte lysis, leukocytes were incubated with a mixture of anti-CD3 conjugated to cyanin-5-PE (Dako, Glostrup, Denmark) and anti-CD69 conjugated to PE (Beckman Coulter, Miami, FL). The cells were then analyzed with a FACScan® flow cytometer (BD Biosciences, San Jose, CA), and the results were expressed as the percentage of CD3+ lymphocytes expressing CD69.
Analysis of T cell Vß repertoires
PBMC were cultured in RPMI-1640 medium with 5% heat-inactivated FCS, 20 mM HEPES buffer, 2 mM L-glutamine (Sigma-Aldricht), 100 IU/ml penicillin G, and 100 µg/ml streptomycin (Sigma-Aldricht) at a density of 2 x 105 cells per mL. Cells were stimulated with SEA (250 ng/mL), SEG (250 ng/mL), or RPMI 1640 (negative control) and were then cultured for 9 days with increasing concentrations of recombinant human IL-2 (up to 100 U/ml, Eurobio, Courtaboeuf, France). To determine specific TCR Vß profiles, we used the IO Test Beta Mark® (Beckman Coulter). Briefly, the kit comprises eight vials, marked AH, each containing three Vß families tagged to PE, FITC, or both. The kit was used as recommended by the manufacturer. Vß profiles were assessed by flow cytometry.
Cytokine and chemokine assays
Whole blood was collected from healthy volunteer laboratory staff, who denied recent anti-inflammatory drug intake. PBMC were extracted by differential gradient centrifugation over endotoxin-free Ficoll-Paque Plus® (Amersham Biosciences, UK) and washed three times in Ca- and Mg-free PBS. Cell viability was measured with the trypan blue exclusion test. Cells were used at a density of 5 x 105 PBMC/ml.
TNF-
, IL-10, MIP-1
, thymus and activation-regulated chemokine (TARC), and IFN-
levels were measured in culture supernatants with ELISA kits (Biosource, Nivelles, Belgium, for cytokine assays and R&D Systems, Minneapolis, MN, for chemokine assays). The detection limits were 4 pg/ml for TNF-
, 0.2 IU/ml for IFN-
, 5 pg/ml for IL-10, 10 pg/ml for MIP-1
, and 7 pg/ml for TARC. The ELISA plates were read at 450 nm with a Dynatech (Alexandria, VA) plate-reader.
Statistical analysis
Statistical analyses were done with Stat View® software Version 5 for Windows® (SAS Institute, Carry, NC). Cytokine and chemokine values were expressed as means ± SEM and were subsequently analyzed for significance (P<0.05) with the nonparametric Mann-Whitney U-test. The correlation between TNF-
and MIP-1
levels was tested with Spearmans test.
RESULTS
Superantigenic properties
Induction of CD 69 expression
SEA and SEG rapidly induced CD69 large expression on T cells (Fig. 2
). Levels were similar with the two toxins (23% and 22%, respectively) and also with a SEA- and SEB-containing, filtered culture supernatant of S. aureus strain ATCC 13566, used as positive control (28%).
|
|
Pro-/anti-inflammatory responses
We compared the pro-/anti-inflammatory properties of SEA and SEG by measuring the release of TNF-
(a proinflammatory cytokine) and IL-10 (an anti-inflammatory cytokine) by PBMC after 24, 48, and 72 h of incubation. At 24 h, cytokine release was similar to negative control values (data not shown). At 48 and 72 h, we observed concentration-dependent TNF-
production in response to SEA (from 1 to 100 ng/ml), whereas SEG induced only weak TNF-
release at 100 ng/ml (Table 1
). Neither toxin induced IL-10 production (Table 1)
. Consequently, SEA was associated with a significantly higher, proinflammatory behavior than SEG (Table 1)
.
|
(CCL-3, a Th1 chemokine) and TARC (CCL-17, a Th2 chemokine) in 48 h and 72 h PBMC culture supernatants. At 48 h, we observed concentration-dependent MIP-1
production in response to SEA, whereas SEG induced only weak MIP-1
release at 100 ng/ml (Table 2
). At 72 h, only SEA induced concentration-dependent MIP-1
production (Table 2)
. Neither toxin induced significant TARC production (Table 2)
. Consequently, SEA induced a significantly stronger Th1 response than SEG (Table 2)
. IFN-
was just detectable after 48 h of culture with each toxin (SEA and SEG at 100 ng/ml; data not shown).
|
and MIP-1
values obtained after 48 and 72 h of SEA challenge, we found a positive and highly significant correlation between levels of the two mediators (r=0.781, P<0.001), supporting concomitant induction of proinflammatory (TNF-
) and Th1 (MIP-1
) responses by SEA. DISCUSSION
The main finding of this study is that the S. aureus superantigen SEA induces stronger proinflammatory and Th1 responses than SEG in PBMC from health volunteers. We also observed selective expansion of T cells bearing particular TCR Vß regions, as reported previously [10
, 19
, 21
]. These results confirmed the functional properties of our toxin preparations [3
, 19
, 20
, 22
]. The inflammatory properties of SEA were in keeping with those observed by Krakauer and coworkers [7
, 23
24
25
] and Hale et al. [25
] with SEB and TSST-1, both of which induced proinflammatory responses. Contrary to SEG, however, we found that SEA induced marked Th1 chemokine production. As Th2 cytokine production by cultured human cells is barely detectable, we studied Th1/Th2 responses in terms of chemokine production [26
]. We studied Th1 chemokine responses by measuring MIP-1
/CCL-3 [27
28
29
] and Th2 chemokine responses by measuring TARC/CCL-17, a member of the CC family [30
31
32
]. We found that SEA induced Th1 chemokine release (abundant MIP-1
and no TARC), as reported previously by Tessier et al. [33
], Brandt et al. [34
], and Gehring et al., using TSST-1 [34
, 35
]. These findings indicate that most S. aureus superantigens induce proinflammatory/Th1 responses. In contrast, we showed, for the first time, that SEG did not induce a proinflammatory response.
Based on their amino acid sequences, SAGs can be divided into four distinct groups [4 ]. SEA and SEG belong to different structural families. SEA belongs to one group, SEG and SEB to a second group, and TSST-1 to a third group. This phylogenetic classification, based on amino acid sequence homology, does not reflect the different pro-/anti-inflammatory behavior of SEA and SEG observed here.
The respective roles of monocytes and T lymphocytes in the inflammatory response to SAGs are controversial. Dinges et al. [36
] showed that rabbit with profound T cell deficiency remained completely susceptible to TSST-1, indicating that lethal effects of superantigens can occur in the absence of T cells. SEB induced TNF-
production by isolated monocytes but not by nonmonocytic cells or purified T cells, suggesting that proinflammatory reactions do not require T cells [37
]. On monocytes, SAGs mainly interact with MHC Class II molecules. SEA, contrary to SEG, possesses a cystein domain involved in high-affinity MHC Class II binding [38
, 39
]. SEA binds to two sites on Class II molecules, one with high-affinity and the other with low-affinity [26
, 40
]. In contrast, SEG appears to bind only the
chain of MHC Class II molecules with low-affinity [22
]. These differences might in part account for the different inflammatory properties of S. aureus superantigens. Grossman et al. [41
] showed that monocyte TNF-
production, in response to SEA and TSST-1, could be blocked significantly by anti-MHC Class II antibodies. TNF-
production by monocytes generally correlates with toxin binding to MHC Class II molecules. This was confirmed in experiments with "mutated toxins" such as F47-SEA, which lacks one of the MHC Class II-binding sites [42
]. Wright and Chapes [43
] proposed another mechanism of SAG action on monocytes. They showed that SEA could also interact with MHC Class I molecules on monocytic cells lacking MHC Class II expression and could induce marked cytokine release (notably, TNF-
and IL-6). This cross-linking of SEA and MHC Class I molecules offers an alternative solution to explain the differential, proinflammatory behavior of SAGs [43
].
Several reports point to a major role of T lymphocytes in inducing proinflammatory responses. Marrack et al. [44
] showed that pathological effects of SEB might result from massive T cell stimulation, and T cell-deficient mice being less susceptible to the toxic effects of SEB. Faulkner et al. [45
] recently demonstrated that
/ß TCR was critical for toxic shock lethality in HLA-DR1 transgenic mice. However, when we measured CD69 expression on CD3 cells, we observed no significant difference in T cell activation induced by SEA and SEG.
The difference in the pro-/anti-inflammatory properties of SEA and SEG observed here is in agreement with the results of several epidemiological studies. Ferry et al. [1 ] showed a higher prevalence of S. aureus strains harboring sea genes in septic patients with shock than in patients without shock. In contrast, the prevalence of egc-harboring strains, which produced SEG, was higher among strains causing suppurative infections than invasive diseases [1 ]. Ferry also observed a negative relationship between the egc cluster and clinical severity. Similarly, Peacock et al. [8 ], in an epidemiological study designed to identify links between S. aureus virulence factors and deep-seated infections, observed a high prevalence of sea and sej in invasive strains. Finally, Becker et al. [12 ] studied the prevalence of genes encoding pyrogenic superantigenic toxins in blood and nasal S. aureus isolates and found that 73% of all isolates harbored superantigen toxin genes and that over 50% of these isolates were positive for seg and sei. However, they found no significant differences between the two types of sample.
Together, these results suggest that the association of SEA with the severity of S. aureus septic shock, characterized by a deleterious, inflammatory cascade, may be explained partly by the specific proinflammatory properties of this superantigen. This hypothesis now deserves to be investigated in vivo in animal models. It also remains to be shown whether SEG properties apply to other egc toxins.
In conclusion, we show that SEA is more proinflammatory than SEG. This might be a result of differences in monocyte binding. Our findings are in keeping with epidemiological data showing a higher prevalence of SEA in S. aureus strains associated with severe clinical syndromes.
ACKNOWLEDGEMENTS
We thank Fabienne Venet for her excellent advice and Florence Couzon for technical assistance. We also thank Professor Jacques Bienvenu and the immunology team of Lyon Sud (especially Carmen Fernandez and Marie-Claude Gutowski) for technical assistance with cytokine assays. G. L. and G. M. contributed equally to this work.
Received March 31, 2006; revised June 6, 2006; accepted June 13, 2006.
REFERENCES
/CCL3 on the expression of hosts innate immunities against infectious complications J. Leukoc. Biol. 72,1190-1197
and MIP-1ß during experimental endotoxemia and human sepsis J. Infect. Dis. 179,136-141[CrossRef][Medline]
J. Immunol. 161,1204-1211
-chain J. Immunol. 158,3698-3704[Abstract]
secretion Cell. Immunol. 197,129-135[CrossRef][Medline]This article has been cited by other articles:
![]() |
D. Grumann, S. S. Scharf, S. Holtfreter, C. Kohler, L. Steil, S. Engelmann, M. Hecker, U. Volker, and B. M. Broker Immune Cell Activation by Enterotoxin Gene Cluster (egc)-Encoded and Non-egc Superantigens from Staphylococcus aureus J. Immunol., October 1, 2008; 181(7): 5054 - 5061. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Sola, H. A. Saka, the Cordoba MRSA Collaborative Study Group, A. Vindel, and J. L. Bocco Emergence and Dissemination of a Community-Associated Methicillin-Resistant Panton-Valentine Leucocidin-Positive Staphylococcus aureus Clone Sharing the Sequence Type 5 Lineage with the Most Prevalent Nosocomial Clone in the Same Region of Argentina J. Clin. Microbiol., May 1, 2008; 46(5): 1826 - 1831. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |