Published online before print October 29, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Biomedizinische ForschungsgmbH and
Immunology Outpatient Clinic, Vienna, Austria
2 Correspondence: Biomedizinische ForschungsgmbH, Lazarettgasse 19/2, A-1090 Vienna, Austria. E-mail: martha.eibl{at}meduniwien.ac.at
ABSTRACT
Staphylococcal toxic shock syndrome toxin 1 (TSST-1) is the major cause of toxic shock syndrome and is important in the pathophysiology of staphylococcal septic shock. Our study about the biological effects of TSST-1 in the rabbit 3 and 6 h and 7 days postinjection provides evidence that TSST-1 induces leukopenia, lymphopenia, and monocytopenia as a result of extravasation of cells in a Vß-unrestricted manner. Cells in the circulation, reduced significantly in numbers, show the same phenotypic distribution as before TSST-1 injection. Three hours post-in vivo TSST-1 injection, we demonstrated compartmentalization of the response. By quantitative RT-PCR, the induction of mRNA expression of TH1 and inflammatory cytokines in the spleen and lung and a complete lack of induction in PBMC could be shown. Proliferation assays revealed that 3 h after TSST-1, PBMC were neither activated nor responsive to in vitro restimulation, even when IL-2 was added. In contrast, 7 days later, PBMC and spleen cells were anergic: showing no response to TSST-1 but a vigorous response upon addition of IL-2. The results presented extend our understanding of the pathophysiology of toxic and septic shock as a result of superantigen toxin-producing Staphylococcus aureus. Demonstration of compartmentalization of the response proves that erroneous conclusions could be drawn by the exclusive analysis of PBMCs. The results reveal further that in nonresponsiveness to the antigen, different immunological mechanisms may be operational. Measurements of the induction of cytokine gene activation provide important complementary information to that of serum cytokine levels.
Key Words: rabbit • inflammation • superantigens • cell proliferation • anergy • extravasation
INTRODUCTION
Staphylococcus aureus represents a multifaceted bacterium that persists as a commensal bacterium in 10–30% of the human population. However, it can cause severe infections, including skin abscesses, wound infections, and life-threatening diseases such as osteomyelitis, endocarditis, necrotizing pneumonia, sepsis, and toxic shock syndrome (TSS); it may also be a cause of food poisoning. The staphylococcal superantigens (sAg) TSS toxin 1 (TSST-1) is the major cause of TSS. sAg bind to a nonpolymorphic region of MHC class II and unlike conventional antigens, do not need prior processing for presentation to the T cell [1
2
3
]. They activate up to 10–30% of the T cell population by interacting with the Vβ region of the TCR. The interactions among sAg, TCR, and MHC class II result in a massive proliferation of T cells and an uncontrolled release of cytokines. IFN-
, IL-1β, IL-6, and TNF-
are important mediators of the inflammatory response and are known to play a pivotal role in sAg-induced immune activation. These proinflammatory cytokines and IL-2 seem to contribute considerably to the life-threatening complications of TSS [4
, 5
].
Recently, it was shown that two staphylococcal sAg, staphylococcal enterotoxin A (SEA) and SEG, had similar sAg properties but different proinflammatory responses, and only the latter correlates with toxicity [6 ]. Upon SEA challenge, human monocytes secreted high levels of proinflammatory chemokines, whereas SEG was not able to induce inflammation. TSST-1 may have lethal toxicity independent of its superantigenicity. As a result of the key role of inflammatory cytokines in the pathogenesis of TSS, the clinical events in this and other immune disorders might be governed by the expression pattern of these cytokines.
For the study of tissue-specific cytokine expression, animal models are mandatory. It is well documented that humans and rabbits are much more susceptible to the effects of sAg as compared with mice [7 8 9 ].
The pathological effects of TSST-1 in rabbits and humans are highly similar [9 10 11 12 ]. In humans, staphylococcal TSS is an acute-onset illness characterized by high fever and hypotension that can lead to multiple organ failure. Three or more organ systems can be involved according to the Centers for Disease Control and Prevention (CDC) guidelines for diagnosis [13 , 14 ]. Rabbits challenged with TSST-1 develop fever, conjunctival hyperemia, anorexia, cachexia, and lethargy. Serum calcium declined, whereas blood urea nitrogen, serum creatinine, and serum glutamic pyruvic transaminase were highly elevated after application of TSST-1. At necropsy, most rabbits had mottled, congested livers, dark red cervical, thoracic, and mesenteric lymph nodes, dark, congested thymuses and spleens, and mottled, congested lungs [9 ]. The histological changes seen in tissues of rabbits infected with TSST-1-negative S. aureus, engineered by phage-transfer to express the TSST-1 gene and thus, produce the toxin, were remarkably similar to those described in tissues from human patients with TSS [11 , 13 ]. As a result of the lack of immunological reagents in the rabbit model, detailed analyses of pathophysiological changes in organs are still missing.
The effects of staphylococcal sAg have been studied intensely in mice. sAg are known to modify host immunity profoundly. They activate a large Vβ-positive subset of T cells leading to expansion and deletion of the responsive T cells [15 ]. More recently, a number of reports concentrated on anergy following in vivo sAg application leading to tolerance [16 17 18 19 ].
Application of milligram quantities of sAg to mice does not lead to signs of TSS, whereas life-threatening TSS in humans is already caused by 2–5 µg sAg [7 , 20 ]. To overcome the high resistance against sAg in mice, hepatotoxic agents, such as D-galactosamine, are coadministered, resulting in liver damage [21 , 22 ], whereas in human TSS, multiple organs are affected. The lower sensitivity of mice to sAg has been attributed to a lower affinity of the toxins for murine as compared with human MHC class II molecules [23 ].
The sAg function of TSST-1 is considered as an important determinant of its lethal effect in humans and experimental animals [24 ]. Interestingly, spleen cells from various mouse strains show different proliferative capacity to sAg stimulation [25 ].
Strain/specific variation may be one of the reasons of contradictory findings in different studies relating to the induction of cytokine expression and release in mice [26 ]. Furthermore, as most of the results in the mouse system were obtained in experiments with spleen cells, and the studies in humans are done with PBMC, it appeared to us that this may also be a reason for discrepancies between findings. We therefore wanted to perform our studies on PBMC and spleen cells in parallel to have a true comparison under identical conditions and thus chose the rabbit model as most suitable.
In this paper, we describe the IFN--, IL-1β-, IL-2-, IL-6-, and TNF--mRNA expression profiles at two time-points in the lungs and in the spleen of rabbits after TSST-1 application. By applying quantitative real-time PCR (qRT-PCR), we show that significant mRNA expression for inflammatory cytokines is present in the organs but not in PBMC, already 3 h post-sAg administration. So far, it has been asserted that mainly sAg-reactive Vβ cells extravasate. Here, we show that 3 h after sAg application, more than 70% of lymphocytes, including CD4 and CD8 subsets of T cells as well as B cells, had left the circulation, proving by the numbers alone that extravasation cannot be Vβ-restricted. Interestingly, although all numbers decreased considerably, phenotypic distribution is comparable before and after extravasation, but the remaining cells were unresponsive to stimulation by the respective sAg.
MATERIALS AND METHODS
Animals
New Zealand White female rabbits, 1.5–2 kg, were purchased from Charles River Laboratories (Sulzfeld, Germany). Animals were kept in standard care facilities according to the guidelines of the Austrian Ministry for Science and Research and had free access to food (Altromin 2120 standard diet pellets, Marek Futtermittelwerke, Vienna, Austria) and water. The animal experiments had been approved and controlled by the Veterinary Department of the City of Vienna (Austria).
Antibodies for flow cytometry analyses
The murine-unlabeled or -conjugated mAb recognizing the rabbit cell-surface antigens CD4 (clone KEN-4), CD5 (clone KEN-5), CD8 (clone 12C7), CD45 (clone L12/201), IgM (clone NRBM), and CD14 (clone Tsk4) were purchased from Serotec (Oxford, UK). The corresponding IgG1- and IgG2a-negative control mAb were obtained from BD PharMingen (Becton Dickinson, San Jose, CA, USA).
Substances
Native TSST-1 (nTSST-1), recombinant TSST-1 (rTSST-1), as well as SEB were produced in our laboratory. The expression and purification of TSST-1 are described extensively by Gampfer et al. [27
, 28
]. All substances were tested for their biological activities in our laboratory and were proved to be endotoxin-free.
Injection of TSST-1, blood drawings, and extraction of organs from rabbits
The rabbits were challenged with 100 µg nTSST-1 or rTSST-1. The substances were dissolved in 1 ml PBS and sterile-filtered and were injected s.c. at two different sites (0.5 ml per site) or i.v. into the ear vein. In all experiments, results obtained with nTSST-1 and rTSST-1 were comparable and were given as a single mean value in Results. As a negative control, PBS was s.c. (2x0.5 ml)- or i.v. (1x1 ml)-administered.
For determination of white blood cells (WBC) and lymphocytes, as well as to perform flow cytometry analyses, venous EDTA blood was drawn prior and 1, 2, 6, 24, and 48 h after injection. WBC counts were determined by the use of a Coulter counter machine (Beckman Coulter, Fullerton, CA, USA) and in parallel, with total lymphocyte counts on a CELL-DYN® 3500 hematology analyzer (Abbott Laboratories, Abbott Park, IL, USA). In addition, the WBC were verified microscopically by counting cells with the help of a counting chamber (Brand, Wertheim, Germany).
For investigations of the cytokine gene expression in the organs (lung and spleen), each rabbit studied was i.v.-injected with 100 µg rTSST-1 or remained untreated (control rabbits). The organs were extracted from rabbits 3 h or 7 days after TSST-1 administration or from untreated animals described as follows. Preliminary experiments showed that the maximum of cytokine gene expression occurred 2–6 h after TSST-1 injection. Anesthesia was introduced with of 25 mg/kg ketamine hydrochloride (Ketalar®, Parke-Davis, Morris Plains, NJ, USA) and 2 mg/kg xylazine hydrochloride (Rompun®, Bayer AG, Leverkusen, Germany) administered intramuscularly. Before endotracheal intubation, anesthesia was deepened with i.v. application of ketamine hydrochloride and xylazine hydrochloride (100 mg ketamine hydrochloride, 4 mg xylazine hydrochloride, and 5 ml sodium chloride) to effect, followed by intubation and volume-controlled ventilation with 2% isoflurane (Forane®, Abbott Laboratories). After a midline laparotomy, a venflon was inserted into the abdominal aorta, and the whole blood was collected in heparinized tubes. Before organ extraction, 100 ml isotonic saline was infused via the abdominal aorta to remove traces of circulating blood cells from the organs studied. The organs used for cytokine gene expression studies were extracted at the time-points mentioned above, minced to small pieces, transferred immediately to liquid nitrogen, and stored at –80°C. For the proliferation assays of spleen cells, a good portion of the extracted spleens was transferred for transport into complete medium consisting of RPMI-1640 medium (Invitrogen, Paisley, UK), 10% FCS (HyClone, Logan, UT, USA), 2 mM L-glutamine (Invitrogen), and 100 U/ml penicillin and 100 µg/ml streptomycin (Invitrogen).
Immunofluorescence analyses and calculation of absolute cell numbers of peripheral blood (PB) subpopulations in rabbits
Whole blood (100 µl) was incubated for 30 min at room temperature with 10 µl unlabeled mAb specific for CD4, CD8, CD45, or IgM or 10 µl conjugated mAb directed against CD5 or CD14. As a negative control, an irrelevant, isotype-matched mAb was used. After incubation, cells were washed once with 2 ml PBS. Binding of the primary mAb was visualized using rabbit F(ab')2 anti-mouse Ig-FITC (STAR9B; Serotec, UK) in a working dilution of 1/100. Furthermore, cells were lysed by addition of 2 ml FACS lysing solution (Becton Dickinson) for 10 min. Finally, cells were washed with PBS, and the membrane fluorescence was analyzed on a FACSCalibur flow cytometer, supported by CellQuest software (Becton Dickinson).
Venous EDTA-whole blood samples drawn before and 1, 2, 6, 24, and 48 h after injection were analyzed for their leukocyte counts (i.e., absolute leukocyte numbers) with the help of a Coulter counter machine (Beckman Coulter), followed by flow cytometry analyses of these samples. FACS data allowed determining the percentage of granulocytes and mononuclear cells (MNC) within these whole blood samples by gating cells according to their size and granularity. Consequently, absolute granulocyte numbers were calculated. The knowledge of the calculated, absolute MNC numbers and the percentage of CD4+, CD5+, CD8+, CD45+, IgM+, and CD14+ cells within the MNC population allowed determination of absolute cell numbers of these subpopulations within the whole blood samples.
Cell isolation and cell stimulation
PBMC were isolated from heparinized blood of rabbits by density gradient centrifugation with LymphoprepTM (Axis-Shield PoC, Oslo, Norway). Subsequently, cells were washed twice in PBS and then resuspended in culture medium [RPMI supplemented with 10% FCS (HyClone)]. The cell number was adjusted to 6 x 106/ml, and cells were then stimulated with 30 ng/ml TSST-1 for 2.5 h or were grown in medium alone for the same period.
Splenic cells (SMC) were isolated from organs as follows: Spleens from treated or untreated animals were cut into small pieces, homogenized, and then filtered through a screen mesh of a 70-µm cell strainer (Becton Dickinson). Subsequently, the cell number was adjusted to 5 x 106/ml, and spleen cells were isolated using a density gradient with Lympholyte® as separation medium (Cedarlane, Ontario, Canada). Cells were washed twice in culture medium and then resuspended in 10 ml culture medium. The cell number was adjusted to 6 x 106/ml in culture medium. The SMC were then stimulated with 30 ng/ml TSST-1 for 2.5 h or grown in culture medium alone for the same period. Previous in vitro experiments indicated that the time period of 2.5 h was ideal for maximum gene expression of diverse cytokines.
Lymphocyte proliferation assay
Proliferation assays were performed with PBMC and SMC as follows: Cells/well (1x106) were cultured in 96-well round-bottom tissue-culture plates (Sarstedt, Newton, NC, USA) in complete medium consisting of RPMI-1640 medium (Invitrogen), 10% FCS (HyClone), 2 mM L-glutamine (Invitrogen), and 100 U/ml penicillin and 100 µg/ml streptomycin (Invitrogen) in the presence of different dilutions of rTSST-1 or nSEB (final concentrations: 100 ng/ml, 10 ng/ml, 1 ng/ml, 0.1 ng/ml) and the combinations of rTSST-1 or SEB plus human rIL-2 (Serotec) used in a final concentration of 100 U/ml. IL-2 has been described by Perkins et al. [29
] as being highly conserved between rabbits and human (amino acid identities of 85.2%), indicating that IL-2 may be used in bioassays across several species. PMA and ionomycin (both Sigma-Aldrich, St. Louis, MO, USA) were used in the respective final concentrations of 50 ng/ml and 250 nM.
All cells used were cultured in a humidified atmosphere at 37°C and 5% CO2 for 5 days. On Day 4, 0.5 µCi/well 3H-thymidine (ICN, Irvine, CA, USA) was added, and 18 h later, cells were frozen in and stored at –20°C until harvesting onto glass fiber filters. Incorporated radioactivity was measured with a MicroBeta Trilux 1450 scintillation counter (Wallac, Turku, Finland). In each experiment, the rabbit cells were tested in triplicates with all concentrations of the used stimuli. The proliferative response was expressed as the stimulation index (SI), calculated as the ratio of the proliferation of stimulated cells (in dpm) and the proliferation of cells without stimuli (in dpm).
RNA isolation from rabbit organs, PBMC, or SMC and RT
RNA was extracted from rabbit organs using the High Pure RNA tissue kit from Roche Diagnostics (Mannheim, Germany). In short, 
25 mg lung or spleen of rabbits, treated as mentioned above, was homogenized in 0.7 ml lysis/binding buffer provided in the kit using an Ultra-Turrax T8 homogenizator (IKA, Staufen, Germany). Samples were then centrifuged at 13,000 g for at least 30 s through a filter tube, where the RNA binds selectively to a glass fiber fleece. Residual DNA was then digested by DNase I. After a series of wash and spin-steps, the RNA was eluted from the glass fiber fleece with 30 µl elution buffer (sterile, double-distilled water) and by centrifuging for 2 min at 8000 g. The concentration of the eluted RNA was then measured spectrophotometrically at 260 nm. RNA from PBMC and SMC was extracted using the High Pure RNA isolation kit following the same principle as above. The concentration of the eluted RNA was adjusted to 60 ng/µl for RT. RNA was reversely transcribed using the SuperScript III Platinum two-step qRT-PCR kit (Invitrogen). RNA (900 ng) was mixed with 20 µl 2x RT reaction mix and 4 µl RT enzyme mix to a final volume of 40 µl. The 2x RT reaction mix contains a mixture of oligo(dT)20 and random hexamers for priming, dNTPs, and a buffer optimized for RT. The samples were incubated at room temperature for 10 min. RT followed at 42°C for 50 min. The reaction was terminated at 85°C for 5 min, followed by a subsequent chill on ice. Aliquots of cDNA were stored long-term at –20°C and stored short-term at +4°C.
Primers
Gene-specific oligonucleotide primers were designed by primer design Primer Express® v2.0 software from Applied Biosystems (Foster City, CA, USA). Forward and reverse primers were designed to span different exons. As a result of the lack of genomic sequence information about rabbit consensus, sequences of the cytokine genes from human and mouse served as a basis for primer design. Primer pairs (Table 1
) were synthesized at MWG Biotech (Heidelberg, Germany) and at Invitrogen.
|
View this table: [in a new window] |
Table 1. Primers Used for qRT-PCRs
|
|
View this table: [in a new window] |
Table 2. Primers Used for Cloning of cDNA
|
The RT reaction (2 µl) was amplified in duplicates in 12.5 µl Platinum SYBR Green qPCR SuperMix- uracil DNA glycosylase, 0.5 µl 6-carboxyl-X-rhodamine reference dye, and 1 µl primer mix (10 µMol each). The reaction was filled up with DEPC-treated water to give a final volume of 25 µl. qRT-PCRs were performed in 0.2 ml thin-wall eight-tube strips locked with optical thin-wall eight-cap strips (Abgene, Epsom, UK) under the following conditions: After a hold for 2 min at 50°C, an initial denaturation at 94°C for 2 min was performed followed by 40 consecutive cycles at 95°C for 15 s (denaturation), 57°C for 30 s (annealing), and 72°C for 30 s (extension phase). Fluorescence data were collected during extension phase. As an add-on application, a melting-curve analysis was performed. Samples were first melted at 95°C for 15 s and then equilibrated at 60°C for 20 s before being reheated slowly (dissociated-melted) back to 95°C. The ramping time at Stage 3 (from 60°C to 95°C) varied from default (usually ramping is as fast as possible) to 19 min and 59 s. Fluorescence data were collected during the ramping from 60°C to 95°C. The threshold was positioned at a fluorescence level that was 10 times higher than the background signal. Target cDNA copy numbers in the cellular samples were then calculated based on a standard curve that was created by plotting the cycles at a comparative threshold against the logarithmic values of the cDNA standard copy number. Data were normalized by reference to GAPDH as an internal standard.
RESULTS
TSST-1 induces leukopenia, lymphopenia, and monocytopenia in rabbits
Although exposure to sAg has been described as resulting in hematological changes in several species including monkeys, mice, and rabbits [9
, 12
, 30
31
32
33
34
35
36
], data about the effects on the different lymphocyte subsets in the circulation are inhomogeneous and incomplete. Reports addressing this question are rare, controversial, and especially focused on the extravasation of Vβ-expressing cells and their interaction with endothelial cells [33
, 35
36
37
]. Furthermore, the kinetics of extravasation of individual cell populations has not been investigated.
In this study, rabbits were administered with TSST-1 or PBS, and the numbers of circulating leukocytes were determined at different time-points. Prior to injection, WBC of all rabbits tested (n=8) were 6293 ± 1824 cells/µl (m±SD; Fig. 1A
). Six hours after injection of TSST-1 (n=4), the number of circulating leukocytes decreased to 1913 ± 688 cells/µl, which was about one-third of baseline levels. The number of circulating PB-leukocytes had started to decline 1 h after TSST-1 application, reached its minimum level 6 h after injection, and remained nearly unaltered until 24 h after TSST-1 injection. Subsequently, 48 h after TSST-1 administration, PB-leukocyte counts returned to baseline levels (Fig. 1A)
. In parallel, as shown in Figure 1B
, total lymphocyte counts in the PB of rabbits decreased considerably 6 h after TSST-1 administration. Prior to injection, the numbers of lymphocytes were found to be 5960 ± 1937 (n=8). TSST-1 injection caused a strong decline of circulating lymphocytes leading to 490 ± 379 cells/µl (n=4), which was 10% of baseline levels (Fig. 1B)
.
 View larger version (38K): [in a new window] |
Figure 1. Significant decrease of CD4+, CD8+, IgM+, and CD14+ cells in PB of rabbits treated with TSST-1. Each rabbit was injected with 100 µg/ml TSST-1 (n=4) [ ] or 1 ml PBS (n=4) [x]. Prior to application as well as 1, 2, 6, 24, and 48 h after injection, the numbers of WBC (A) and total lymphocytes (B) were determined from venous blood of rabbits using a Coulter counter machine or a hematology analyzer. The absolute numbers of the lymphocyte subsets CD4+ (C), CD8+ (D), IgM+ (E), and CD14+ (F) were calculated from FACS analyses at the time-points indicated, as described in Materials and Methods. Figures show mean values ± SD of the numbers of the indicated cell populations of four TSST-1-treated and four control rabbits.
|
Changes in granulocyte counts in the first hours after TSST-1 injection were not significant. Forty-eight hours after TSST-1 exposure, granulocyte levels were increased significantly within the circulation (data not shown).
Diminished absolute lymphocyte numbers but unchanged phenotypic distribution of lymphocytes in the circulation after TSST-1 application
Few studies focus on the evaluation of cell migration during the early phase after sAg exposure [33
, 36
]. It is, for example, unclear whether the extravasation, upon sAg administration, is limited to a certain subset. Therefore, we addressed the question of whether the observed, strong leukopenia and monocytopenia influenced the distribution of the lymphocyte subsets in the PB after TSST-1 injection.
Figure 2A shows that the absolute numbers of CD45+ lymphocytes, CD5+, CD4+, and CD8+ T cells and IgM+ B cells declined significantly in the whole blood 3 h after administration of TSST-1. The reduction of the different subsets was found to be between 68.8% and 76.8%. However, the distribution by percentage of these lymphocyte subsets in the PB did not change significantly 3 h after TSST-1 injection compared with untreated rabbits (Fig. 2B) .
 View larger version (28K): [in a new window] |
Figure 2. Three hours after TSST-1 injection, total numbers of lymphocytes in all subsets are decreased substantially, but phenotypic composition remains unchanged. Rabbits were injected with 100 µg TSST-1. Prior to administration (open bars) and 3 h after TSST-1 application (solid bars), venous blood was analyzed by flow cytometry. The absolute cell numbers (A) of the indicated lymphocyte subsets were calculated, as described in Materials and Methods. The distribution of the lymphocyte subsets within the MNC gate (B) was assessed before and after treatment. The figure shows mean values ± SD of four rabbits prior to and 3 h after TSST-1 application.
|
Measuring the cytokine mRNA expression in PBMC, we found no significant induction of IL-2 and the inflammatory cytokines IL-1β, IL-6, IFN-, and TNF- 3 h after TSST-1 injection (Fig. 3A
). In contrast to the circulation, there was a strong increase in the expression of the cytokine genes in the spleen (Fig. 3B
, upper). IL-6 was induced most strongly in the spleen (255-fold) and approximately tenfold in the lung (Fig. 3B
, lower). IL-1β, IL-2, TNF-, and IFN- were also induced in the spleen (three- to 21-fold), although all were weaker than IL-6. Fold-induction levels of the cytokine mRNA were lower in the lung than in the spleen.
 View larger version (11K): [in a new window] |
Figure 3. Discordance of cytokine gene expression in PBMC and in the organs 3 h after TSST-1 treatment. Assessment of mRNA expression of IFN-, IL-1β, IL-2, IL-6, and TNF- via qRT-PCR in PBMC (A) and in spleen and lung (B) of rabbits 3 h after TSST-1 injection. Fold induction was determined from values normalized for expression of the housekeeping gene GAPDH and then normalized to the mean values derived from PBMC or corresponding organs of three unchallenged rabbits, as described in Materials and Methods. Data represent mean values from three TSST-1-treated animals.
|
cytokine gene expression and proliferation
Three hours after TSST-1 administration, the PBMC were isolated and in vitro-restimulated by TSST-1 for 2.5 h. As shown in Figure 4 A and B
, neither IL-2 nor IFN- gene expression was induced after restimulation by TSST-1 (middle bars). However, in vitro stimulation with PMA and ionomycin (right bars) led to strong elevation of IL-2 and IFN- mRNA in PBMC isolated 3 h after in vivo TSST-1 treatment. As expected, both cytokine transcripts were enhanced strongly in PBMC taken from untreated rabbits and stimulated in vitro by TSST-1 for 2.5 h. As shown in Figure 4 A and B
(left bars), IL-2 mRNA was induced 260-fold and IFN- mRNA, 22-fold.
 View larger version (27K): [in a new window] |
Figure 4. Three hours after in vivo TSST-1 administration, PBMC are unresponsive to in vitro stimulation by TSST-1, as determined by cytokine gene expression and proliferation. IL-2 (A) and IFN- (B) gene expression of in vitro-stimulated PBMC assessed by qRT-PCR. PBMC isolated from rabbits 3 h after in vivo TSST-1 treatment (A and B, middle and right bars) or from unchallenged rabbits (A and B, left bars) were in vitro-stimulated by 30 ng/ml TSST-1 for 2.5 h. Fold induction was determined from values normalized for expression of the housekeeping gene GAPDH and then normalized to the values derived from PBMC of three untreated rabbits, as described in Materials and Methods. Data represent mean values from five TSST-1-treated and three untreated animals. As a positive control, cytokine gene expression in PBMC of TSST-1-treated rabbits was determined after stimulation with PMA and ionomycin (A and B, right bars). For proliferation (C), PBMC of rabbits 3 h after TSST-1 injection (sold bars) or of untreated rabbits (hatched bars) were cultured in triplicates in the presence of the indicated concentrations of TSST-1, TSST-1 plus IL-2, SEB, and PMA plus ionomycin for 5 days. For the calculation of the SI, the mean response (in dpm) of the individual stimuli tested was divided per the mean of the medium controls (in dpm) of the respective experiments. (C) SI of the mean proliferative response to TSST-1, TSST-1 plus IL-2, and SEB of three independent experiments of three untreated rabbits and of two to six experiments of six treated rabbits.
|
As expected, PBMC isolated from untreated rabbits showed marked proliferation in the presence of TSST-1, TSST-1 plus IL-2, and SEB (Fig. 4C , hatched bars). The stimulation indices ranged from 110.8 to 136.6 using TSST-1 concentrations between 0.1 ng/ml and 100 ng/ml. Stimulation by TSST-1 plus IL-2 led to proliferation with SI of 146.4–177.3. Cells exposed to 100 ng/ml or 10 ng/ml SEB reached nearly equal stimulation indices as with TSST-1 (SI: 83.9 –113.9). However, lower SEB concentrations resulted in a negligible-proliferative response (data not shown).
Three hours after in vivo TSST-1 administration, spontaneous activation of SMC and lack of further response to restimulation by TSST-1 assessed by cytokine expression and proliferation
As a next step, we addressed the question of whether the cytokine induction and the proliferative capacity of SMC of TSST-1-injected rabbits are affected in response to TSST-1 restimulation.
As shown in Figure 5 A and B
, the fold-induction levels of IL-2 and IFN- gene expression were negligible in SMC taken from rabbits 3 h after in vivo TSST-1 injection, measured after restimulation by TSST-1 (middle bars) or medium alone (data not shown). However, in vitro stimulation of SMC by PMA plus ionomycin isolated 3 h after TSST-1 treatment resulted in strong cytokine gene induction (Fig. 5 A and B
, right bars). As expected, IL-2 expression was clearly induced (50-fold) in SMC isolated from untreated animals when in vitro-stimulated by TSST-1 (Fig. 5A
, left bar), whereas the transcripts of IFN- were scarce (Fig. 5B
, left bar). The proliferative capacity of SMC goes in line with the results observed in cytokine gene expression studies. SMC isolated 3 h after TSST-1 injection were clearly activated, indicated by their spontaneous proliferation (28,311±24,742 dpm) and by their enhanced response to IL-2 (110,697±27,899 dpm; SI: 3.91; Table 3
). However, SMC could not be stimulated further by TSST-1 (SI: 1.3–3.1), TSST-1 plus IL-2 (SI: 3.8–5.4), or SEB (SI: 1.6–3.7; Fig. 5C
, solid bars). As expected, SMC of untreated rabbits responded with marked proliferation to TSST-1 (SI: 46.3–64.5) enhanced by IL-2 (SI: 74.2–103.5; Fig. 5C
, hatched bars). SEB, in concentrations of 100 ng/ml and 10 ng/ml, also induced proliferation in SMC (SI: 26.6–51.7).
 View larger version (27K): [in a new window] |
Figure 5. No response to in vitro stimulation by TSST-1 as determined by cytokine gene expression and proliferation of SMC 3 h after TSST-1 administration. qRT-PCR analyses of IL-2 (A) and IFN- (B) gene expression of SMC isolated 3 h after TSST-1 injection (A and B, middle and right bars) or from control rabbits (A and B, left bars) after in vitro stimulation by 30 ng/ml TSST-1 for 2.5 h. Fold induction was determined from values normalized for expression of the housekeeping gene GAPDH and then normalized to the values derived from SMC of three untreated rabbits as described in Materials and Methods. Data represent mean values from three TSST-1-treated and three unchallenged animals. For positive control, cytokine gene expression in SMC of two TSST-1-treated rabbits after in vitro stimulation by PMA and ionomycin was determined (A and B, right bars). For proliferation (C), SMC of rabbits 3 h after TSST-1 injection (solid bars) or of untreated rabbits (hatched bars) were cultured in triplicates in the presence of the indicated concentrations of TSST-1, TSST-1 plus IL-2, SEB, and PMA plus ionomycin for 5 days. For the calculation of the SI, the mean response (in dpm) of the individual stimuli tested was divided per the mean of the medium controls (in dpm) of the respective experiments. (C) SI of the mean proliferative response to TSST-1, TSST-1 plus IL-2, and SEB of two to three experiments of three TSST-1-treated and three control rabbits.
|
|
View this table: [in a new window] |
Table 3. Proliferation (in dpm) of Indicated Cells to Medium, IL-2 Alone, and PMA Plus Ionomycin
|
mRNA was induced significantly in either of the cell types (data not shown). Thus, compartmentalization could not be detected at that time-point. PBMC, isolated 7 days after TSST-1 injection, responded weakly to in vitro restimulation by TSST-1 with SI ranging from five- to 44-fold induction of medium levels. This proliferation could be enhanced strongly by the addition of IL-2 leading to SI between 449 and 751 for all tested concentrations, indicating an anergic state in PBMC. IL-2 alone was able to activate PBMC moderately. SEB stimulation resulted in marked proliferation with SI of 166–230 at the concentrations 100 ng/ml and 10 ng/ml (Fig. 6A solid bars). When studying SMC (Fig. 6B) at that time-point, we found similar behavior to the cells in the circulation. TSST-1 alone could only induce a minor proliferative response (SI: 3–17). TSST-1 plus IL-2 resulted in strong proliferation with SI between 302 and 318 (Fig. 6B , solid bars). The anergic state was only detectable for TSST-1; the addition of SEB in the concentration of 100 ng/ml and 10 ng/ml led to a proliferative response (SI: 49–96). Seven days after TSST-1 administration, we did not find cells in the circulation nor in the spleen, which proliferated spontaneously, as indicated by the low baseline levels in medium of 204 ± 74 dpm or 645 ± 295 dpm, respectively (Table 3) . However, there seemed to be still high numbers of IL-2R-expressing cells in the PB and spleen, resulting in high-proliferation counts in response to IL-2 alone (Table 3) .
 View larger version (27K): [in a new window] |
Figure 6. Seven days after in vivo TSST-1 treatment, PBMC and SMC are unresponsive as a result of anergy; responsiveness can be restored by addition of IL-2. PBMC (A) and SMC (B), isolated from untreated rabbits (hatched bars) or from rabbits that were injected with TSST-1 7 days before (solid bars), were cultured in triplicates in the presence of the indicated concentrations of TSST-1, TSST-1 plus IL-2, SEB, and PMA plus ionomycin for 5 days. Figure shows the SI of three independent experiments of three TSST-1-treated and three control rabbits.
|
Our study gives new and important insights into the biology of TSS. We provide evidence that extravasation after TSST-1 injection is not restricted to TSST-1-reactive T cells. Our detailed study of the initial phase of the response to TSST-1 in the rabbit model showed that a significant proportion of leukocytes, mostly lymphocytes, disappears from the circulation within 6 h after TSST-1 injection (Fig. 1 A and B)
. TSST-1 is known to react specifically with T cells bearing Vβ2 in humans, thereby engaging 10% of all β+ T cells [2
]. Although we could not determine the exact percentage of the TSST-1-reactive Vβ subset in rabbits as a result of the lack of specific reagents, it is certain that the observed, highly significant loss of lymphocytes occurred in a Vβ-unrestricted manner, as 84–94% of all T and B cells had disappeared from the PB 6 h after TSST-1 application, and CD4+ and CD8+ T cells as well as B cells extravasated with comparable kinetics (Fig. 1 C and E)
. Surprisingly, when total lymphocyte counts in the circulation were 20–30% of preinjection counts, the phenotypic distribution of lymphocytes (T cells, B cells, CD4 and CD8 subsets) were similar to the initial distribution (Fig. 2B)
. Furthermore, we provide evidence that monocytes extravasate with different kinetics than lymphocytes after sAg application (Fig. 1 C-F)
. The majority (87%) of monocytes was absent from the circulation already 1 h and up to 24 h after TSST-1 administration.
Reports addressing the analysis of subpopulations in the circulation of sAg-treated animals are rare and controversial. In accordance with our results, leukopenia and lymphopenia followed by leukocytosis have been observed after TSST-1 or SEB exposure in rabbits, mice, or monkeys [9 , 12 , 30 , 31 , 34 ]. Lymphopenia has also been described as a frequent finding, especially in rapidly progressive cases of toxic shock as well as in patients with septic shock caused by S. aureus [38 ]. A striking, Vβ-specific decrease of lymphocytes in the circulation upon SEB challenge has been described in macaques. The SEB-reactive Vβ3 and Vβ19 population disappeared from the circulation 2 h–2 days after SEB injection. This coincided with a decrease in total lymphocyte counts (CD3, CD4, and CD8 T lymphocytes) and the relative increase in CD20+ B lymphocytes [24 ]. In mice injected with SEB, 90% of blood Vβ8+ cells disappeared within 2 h, whereas cells expressing Vβ6+, which do not react with SEB, were also depleted from blood but much less than Vβ8+ T cells [36 ].
Looking carefully at the results described in the paper of Vasseur et al. [36 ], as well as at our results demonstrating that the cells remaining in the circulation are completely nonresponsive to TSST-1 stimulation, even if IL-2 is added (Fig. 4 , Table 3 ), it appears likely that TSST-1-responsive lymphocytes are activated and leave the circulation and home in lymphoid organs (spleen) soon after activation. In addition, as it has also been observed by Vasseur and coworkers [36 ], bystander cells, i.e., B cells and T cells, become activated and leave the circulation, leading to the lymphopenia observed. The interaction of activated endothelial cells with the activated lymphocytes, which has also been reported by Brogan et al. [37 ], could be an additional reason, explaining the finding that granulocytes do not leave the circulation.
Analyzing cytokine gene expression and lymphocyte-proliferative responses to TSST-1 in parallel in PBMC and in the spleen, the compartmentalization of the response became evident. Three hours post-in vivo TSST-1 injection, PBMC did not show any sign of activation by the induction of cytokine genes (Fig. 3A) , and restimulation with TSST-1 had no effect (Fig. 4 A and B ; although respective genes could be induced normally, bypassing receptor-mediated signaling by PMA/ionomycin [18 ]). Three hours post-TSST-1 application, activated cells could not be detected in the circulation. Proliferation assays revealed that PBMC were unresponsive to TSST-1, and adding IL-2 had no effect. The nonresponsiveness was specific for TSST-1, as a response to SEB, albeit reduced, could be observed, and the response to PMA/ionomycin was unimpaired (Fig. 4C) .
Unlike PBMC, the expression of inflammatory cytokines, especially IL-6 but also IL-1β, and IFN- was highly induced in the spleen and could also be detected, albeit at a low level, in the lung 3 h after in vivo TSST-1 treatment. When activated SMC were restimulated with TSST-1 in vitro, no additional cytokine gene activation could be seen (Fig. 5 A and B)
. One of the reasons could be a down-regulation of the TCR complex [39
]. Cell activation in SMC could also be detected by their spontaneous-proliferative response (in medium) 3 h post-in vivo TSST-1 (Table 3)
. These activated spleen cells could not be activated further in culture by TSST-1, and even the addition of IL-2 had no effect (Fig. 5C)
.
The demonstration of compartmentalization of the sAg/toxin response is of great significance. It is an important example that results obtained in PBMC, which are usually thought to mirror the immunological situation in the patient, may be totally misleading. Activated cells may leave the circulation and home to lymphoid organs, which remain unexplored in the clinical setting. Definite TSS is usually diagnosed according to the clinical criteria proposed by the CDC [40 ]. In cases in which the evaluation of clinical parameters is not sufficient, the analysis of the percentage of Vβ2+ T cells in PBMC was proposed for diagnosis [40 , 41 ]. Our data implicate that because of the compartmentalization of the response, an increase of the respective Vβ population may be absent. However, our findings reflect the biological effects of a sAg bolus. Further studies will have to clarify the effect of chronic sAg stimulation. It could well be that under those circumstances, an increase of the sAg-reactive Vβ population could be present.
IL-6 is an important and sensitive marker of inflammation. It may be used as an indicator for the presence of bacteremia, for the diagnosis of early-onset sepsis in neonates, as it is present earlier in the plasma than C-reactive protein [42 ]. IL-6 is a cytokine that not only correlates with the severity of inflammation but also with mortality in septic patients. Particularly high levels of IL-6 were found in nonsurviving septic patients, suggesting that circulating levels of IL-6 may serve as a predictive variable for survival in sepsis [43 , 44 ]. This early burst of cytokines and their subsequent release into the circulation might contribute considerably to lethality upon sAg exposure.
The absence of elevation of IL-6 mRNA expression in cells of the circulation 3 h (Fig. 3A) and 7 days (data not shown) after TSST-1 administration was first surprising, as levels of IL-6 protein are known to be elevated in plasma in sepsis [43 ]. One has to be aware that results obtained by the estimation of levels of (secreted) cytokines in the circulation and by the analysis of cytokine gene expression in the circulation and/or in organs provide important but different information, and these results do not necessarily correlate with each other. Direct comparison may lead to inhomogeneous and even contradictory results. Measurement of cytokine concentration in the circulation is a helpful diagnostic aid for the assessment of the inflammatory process [42 ]. It has to be taken into consideration, however, that cell-associated cytokines and the short half-life of the cytokines in the circulation might influence the results, as described by Munoz et al. [43 ]. Measurements of cytokine gene activation are well-suited for the detection of the sites and cells of production, but as in our case, compartmentalization of the response could lead to incorrect conclusions.
qRT-PCR offers the advantage to detect small amounts of cytokine messages of all proteins, and it allows detailed and sensitive analyses of the kinetics of transcription, although the effect of post-transcriptional events is not picked up. To our knowledge, apart from our work, there are no in vivo data available about qRT-PCR measuring cytokine gene expressions upon sAg injection.
Seven days after in vivo TSST-1 administration, cytokine gene expression for IL-2 and for proinflammatory cytokines was absent in the circulation and in the spleen and could not be induced by stimulation of MNC with TSST-1 (data not shown). At the same time, PBMC and SMC only proliferated weakly in response to TSST-1 stimulation (Fig. 6) . Both populations were strongly activated by IL-2 alone, and the TSST-1 response has been potentiated significantly by the addition of IL-2 (Fig. 6 A and B) . The most likely explanation is that as activation faded in cells in the spleen, these cells went into an anergic state and started to recirculate. Looking closer at the nonresponsiveness of the cell population studied at different time-points after TSST-1 injection, we realize that what is often described as anergy is obvious nonresponsiveness as a result of different underlying mechanisms. Three hours after in vivo administration, PBMC are not activated and show specific unresponsiveness to TSST-1 (Table 3 and Fig. 4 ). The response cannot be induced by IL-2. At the same time, SMC are highly activated but do not respond to further stimulation. Seven days after TSST-1 challenge, the nonresponsiveness appears similar at first sight, but at this time, PBMC and SMC are truly anergic. They do not proliferate when unstimulated (lack of IL-2 production), but they respond vividly to IL-2, and the TSST-1 response is potentiated by IL-2 (Table 3 and Fig. 6 A and B ).
The results presented extend our understanding of the multifactorial pathophysiology of toxic shock and of septic shock as a result of sAg/toxin-producing S. aureus. It clearly indicates that assessment of cytokine gene expression in the circulation is insufficient for understanding the sequence of events. Estimation of cytokine concentration in the circulation is essential but does not reveal the site of production. Further analysis will have to delineate the biological consequences of sAg/toxin in the different compartments and will help to bring understanding to results that look controversial at present.
ACKNOWLEDGEMENTS
This study was supported by the Center for Innovation and Technology (ZIT/LS04/16). The authors thank Prof. Dr. Udo Losert for his continuous support with the animal experiments at the facilities of the Institute of Biomedical Research at the General Hospital Vienna (Austria).
FOOTNOTES
1 These authors contributed equally to this work. 
Received January 31, 2008; revised August 18, 2008; accepted August 19, 2008.
REFERENCES
J. Exp. Med. 186,71-81 requires transcriptional arrest J. Immunol. 153,1778-1788[Abstract] interferon during induction of hypersensitivity to lipopolysaccharide by toxic shock syndrome toxin 1 in mice Infect. Immun. 69,1256-1264
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
