Proinflammatory Mediators of Toxic Shock and Their Correlation to Lethality

Bacterial exotoxins and endotoxins both stimulate proinflammatory mediators but the contribution of each individual toxin in the release of mediators causing lethal shock is incompletely understood. This study examines the cytokine response and lethality of mice exposed to varying doses of staphylococcal enterotoxin B (SEB) or lipopolysaccharide (LPS) and their combinations. In vivo, SEB alone induced moderate levels of IL-2 and MCP-1 and all mice survived even with a high dose of SEB (100 μg/mouse). LPS (80 μg/mouse) caused 48% lethality and induced high levels of IL-6 and MCP-1. SEB induced low levels of TNFα, IL-1, IFNγ, MIP-2, and LPS synergized with SEB in the expression of these cytokines and that of IL-6 and MCP-1. Importantly, the synergistic action of SEB and LPS resulted in lethal shock and hypothermia. ANOVA of cytokine levels by survival status of SEB-plus-LPS groups revealed significantly higher levels of TNFα, IL-6, MIP-2, and MCP-1 in nonsurvivors measured at 8 hours. Significantly higher levels of IFNγ and IL-2 were observed at 21 hours in nonsurvivors of toxic shock compared to those in survivors. Overall, synergistic action of SEB and LPS resulted in higher and prolonged levels of these key cytokines leading to toxic shock.


Introduction
Bacterial exotoxins and endotoxins are among the most common etiological agents that cause septic shock [1][2][3]. Although similar cytokines are released from host cells stimulated with these structurally distinct bacterial products, the stimulants act through distinct cell surface receptors on host cells. Staphylococcal enterotoxin B (SEB) and structurally related exotoxins are bacterial superantigens that potently activate antigen-presenting cells by binding directly to major histocompatibility complex (MHC) class II molecules [1,4,5]. These exotoxins also interact with specific Vβ regions of the T cell antigen receptors resulting in polyclonal T cell activation [6]. Interactions of superantigens with antigenpresenting cells and T cells lead to massive proinflammatory cytokine and chemokine release, causing clinical symptoms that include fever, hypotension, and shock [1,2,4,7,8].
In contrast, lipopolysaccharide (LPS) from gramnegative bacteria binds to a different receptor on monocytes/macrophages. An LPS-binding protein in serum first binds to LPS and facilitates its binding to cell surface protein CD14 on monocytes/macrophages and other cells [9,10]. The subsequent interaction of LPS/CD14 complex with Tolllike receptor 4 on these cells initiates recruitment of intracellular adaptors and downstream signaling pathways activating NFκB and results in hyperproduction of proinflammatory cytokines and chemokines [10,11]. High levels of these mediators induce systemic inflammatory response, vascular collapse, and shock [12].
In humans, either SEB or LPS alone can induce shock as humans are very sensitive to these bacterial products [2,12,13]. Current in vivo investigations on SEB-induced pathogenesis have relied heavily on murine models of lethal shock. However, mice are less susceptible (compared to humans) to SEB due to the decreased affinity of SEB to mouse MHC class II molecules [14,15] and enhancing agents are used in addition to SEB [16][17][18][19][20][21]. Potentiating agents such as LPS, D-galactosamine, or actinomycin D were used to amplify the toxic effects of SEB in vivo [18][19][20][21]. In these various animal models, there is a strong correlation 2 Mediators of Inflammation between toxicity and increased serum levels of inflammatory mediators, TNFα, IL-1, IL-6, and IFNγ [16][17][18][19]. However, the contribution of each individual cytokine or chemokine to toxicity has not been completely delineated. This study was undertaken to investigate the effects of different doses of SEB, LPS, and SEB plus LPS on the cytokine response and survival of mice exposed to these bacterial products.

Mouse Model of Lethal Shock.
Male Balb/c mice, weighing ∼20 g each (7-10 weeks old), were obtained from NCI (Frederick, MD). Mice were housed in conventional microisolator cages with food and water freely available at all times. SEB and LPS were administered intraperitoneally (i.p.; 200 μL) with a tuberculin syringe (26 G-3/8 inch needle). When SEB and LPS were used together, LPS was injected 4 hours after SEB as this was the optimal time previously determined to cause septic shock [19,[22][23][24]. All injections (0.2 mL/mouse) were given i.p. and all dilutions were made in saline. Doses of SEB used in survival experiments range from 1 to 100 μg/mouse as described in figure legends. A single high dose of LPS (80 μg/mouse) was used (n = 27). The effect of SEB alone, either 30 μg/mouse (n = 20) or 100 μg/mouse (n = 10) was also examined. Survival experiments were conducted with the different doses of SEB alone, LPS alone, and varying doses of SEB and LPS in groups of 10 to 15 mice per dose. Experiments were repeated multiple times as indicated in figures. The following combinations of SEB plus LPS were used: SEB 1 μg/mouse plus LPS 60 μg/mouse (n = 20), SEB 1 μg/mouse plus LPS 80 μg/mouse (n = 63), SEB 10 μg/mouse plus LPS 10 μg/mouse (n = 30), SEB 30 μg/mouse plus LPS 10 μg/mouse (n = 40), SEB 60 μg/mouse plus LPS 10 μg/mouse (n = 40), SEB 100 μg/mouse plus LPS 10 μg/mouse (n = 10), SEB 30 μg/mouse plus LPS 1 μg/mouse (n = 25), and SEB 100 μg/mouse plus LPS 1 μg/mouse (n = 10). Mice exposed to both SEB and LPS or high dose of LPS alone succumbed to death between 20 to 141 hours with the majority (97%) of death occurring between 21 and 70 hours after initial toxin dose. Lethal end points were monitored twice per day for 2 weeks. Temperature-based experiments were performed with each animal subcutaneously implanted (6-10 days before toxin exposure) with a Biomedic Data Systems (Seaford, DE) transponder located on the dorsum between the shoulder blades [24].
Animal research was conducted in compliance with the Animal Welfare Act as well as other federal statutes and regulations. The facility where this research was conducted is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. All efforts adhered to principles stated in the Guide for Care and Use of Laboratory Animals (National Research Council).

Cytokine and Chemokine Assays.
Previous results indicated that the optimal timing for serum cytokine collection after mice were given toxins is 8 hours after SEB [19,22,23]. Serum from an individual mouse was collected at 8 hours and 21 hours after SEB, from anesthetized mice retroorbitally or by cardiac puncture. Most experimental groups (n = 10) were repeated multiple times over a 1.5-year period. Sera were stored at −70 • C until assay in duplicates by ELISA. TNFα, IL-1, IL-2, IL-6, IFNγ, MIP-2, and MCP-1 were measured by ELISA using cytokine-specific antibodies and cytokine standards from R&D Systems (Minneapolis, MN) according to instructions of the manufacturers [19,22,23]. A standard curve was generated for each plate using five dilutions of cytokine standards. The detection limit for cytokine ELISA was 10 pg/mL. Background levels of each cytokine/chemokine, all found to be negligible, were derived from a prebleed of the same mice performed 2-4 days before each experiment.

Statistical Analysis.
Data were analyzed with the use of SAS software, version 9.2 (Cary, NC). Differences in cytokine/chemokine levels between treatment groups were assessed by obtaining geometric means with 95% confidence intervals. Statistical comparisons of survival and cytokine/chemokine data were performed using two-tailed Fisher's exact tests and analysis of variance (ANOVA), respectively. All reported P-values are two sided, and a P value of less than .05 was considered to indicate statistical significance.

Hypothermia in Nonsurvivors of Toxic
Shock. In addition to lethality as an endpoint, body temperature was also used as a marker of systemic shock as previous studies showed hypothermia precedes death in SEB-induced shock [24]. Figure 2 shows that mice given SEB alone maintained normal body temperature similar to temperature of control mice exposed to bovine serum albumin. However, mice exposed to 1 μg/mouse SEB and 80 μg/mouse LPS experienced hypothermia as early as 8 hours after SEB and temperature continued to drop dramatically. Interestingly, mice exposed to LPS alone (80 μg/mouse) had a slight temperature drop of 2 • C at 8 hours and moderate hypothermia was recorded at 21 hours. As 50% of mice survived in this group, re-examination of temperatures of nonsurvivors and survivors in the LPStreated group separately revealed that hypothermia was experienced only in nonsurvivors ( Figure 2). Thus, there is a good correlation between lethality and hypothermia regardless of the toxin and body temperature is an accurate indicator of systemic shock.

Effect of SEB, LPS, and Varying Doses of SEB Plus LPS on Serum Cytokines and Chemokines.
Proinflammatory mediators have critical pathophysiologic effects in vivo and many of the manifestations of septic shock have been correlated to the exaggerated release of these cytokines upon interaction of host cells with SEB and/or LPS [1,2,7,8,12].
We and others showed that serum TNFα, IL-1, IL-6, and IFNγ were critical in inducing lethality in murine models of SEB-induced shock [16,19,23]. Here we also examined the serum levels of these cytokines and two chemokines, MIP-2 and MCP-1, from mice exposed to varying doses of SEB plus LPS as all mice survived, even at 100 μg of SEB. Two time points, 8 and 21 hours were chosen as some cytokines such as TNFα and IL-1 are induced relatively early [19] and the later time at 21 hours might reveal the cytokines induced later by syngergistic effects. The 21 hours time point is also the time near the lethal end point. We first investigated serum cytokines levels of mice treated with either SEB or LPS alone. SEB (100 μg/mouse, n = 10) induced high levels of IL-2 and MCP-1, 24174 pg/mL and 3103 pg/mL, respectively, measured at 8 hours after SEB (Figure 3(a)). Both of these cytokines dropped dramatically to 272 pg/mL and 6 pg/mL for IL-2 and MCP-1, respectively at 21 hours. In contrast, LPS (80 μg/mouse, n = 16) induced significantly higher levels of IL-6, MIP-2, and MCP-1 (4584 pg/mL, 179 pg/mL, 29721 pg/mL) but negligible IL-2 (2 pg/mL) when compared to the SEB-treated group at 8 hours. This is not surprising as IL-2 is a T cell growth factor mostly induced by T-cell mitogens or stimulants. At this early time point, a 6000-fold higher IL-6 and 10-fold higher MCP-1 were seen in the LPS-treated group when compared to the SEB-treated group. The low level of TNFα induced by LPS seen here was likely due to the sampling time after LPS exposure (4 hours) as TNFα peaked 90 minutes to 120 minutes after LPS and disappeared within 6 to 8 hours [24,25]. The sampling time of 8 hours after SEB (4 hours after LPS) was a compromise time chosen to accommodate the other cytokines/chemokines that appeared later after TNFα. At 21 hours after toxin exposure, all cytokines decreased to lower levels (>100 pg/mL) except that of IL-6 (983 pg/mL) and MCP-1 (3341 pg/mL) in LPStreated mice and 272 pg/mL of IL-2 in SEB-treated mice (Figure 3(b)). Pairwise comparison of mice treated singularly with SEB or LPS indicated the levels of IL-6, IL-2, and MCP-1 were significantly different between the two groups at 8 hours and IL-6 and MCP-1 remained significantly different at 21 hours.
We also analyzed and compared the cytokine levels between two groups of mice treated with a constant SEB dose of 30 μg but different low LPS doses of 1 μg (n = 24) or 10 μg (n = 14). Figure 4(a) shows significantly higher TNFα, IL-6, IL-2, and MIP-2 were found in the 30 μg SEB plus 10 μg LPS group at 8 hours. At 21 hours, significantly higher TNFα, IL-1, IL-6, IL-2, and MCP-1 were found in the 30 μg SEB plus 10 μg LPS group compared to the 30 μg SEB plus 1 μg LPS group (Figure 4(b)). Thus there was a substantial potentiation of cytokines and chemokines with the higher LPS dose when SEB was kept constant. Pairwise comparison of these two groups showed prolonged higher levels of TNFα, IL-6, and MCP-1 levels in the higher LPS dose (10 μg) group. It appears that the higher cytokine response, especially at the later time point, was also influenced mostly by the LPS dose.
The cytokine response using a combination of varying doses of SEB with LPS, SEB 30 μg/mouse plus LPS 1 μg/mouse (n = 24), SEB 100 μg/mouse plus LPS 1 μg/mouse  Figure 3: Serum levels of TNFα, IL-1, IL-6, IFNγ, IL-2, MIP-2, and MCP-1 at (a) 8 hours after SEB or LPS administration (b) 21 hours after SEB or LPS administration. Points represent the geometric mean ± standard deviation (SD) for each group. SEB group consisted of mice treated with 100 μg SEB and LPS group was treated with 80 μg LPS. The " * " indicates P < .05 between groups treated with SEB or LPS.

Discussion
The exaggerated systemic response to SEB is similar to LPS where excessive proinflammatory cytokine release causes increase in vascular permeability, cell adhesion, and coagulation [2,12]. Either SEB or LPS alone induce these proinflammatory mediators and it is not clear whether there is a threshold level of these mediators above which either singularly or in combination they can trigger shock. LPS naturally synergizes with superantigens to induce the proinflammatory cascade culminating in shock. Moreover, the cytokines, TNFα, IL-1, and IFNγ induced by either SEB or LPS also synergizes with each other to promote inflammation [26]. In this study we investigated the contribution of the cytokines, TNFα, IL-1, IL-6, IFNγ, and IL-2, and chemokines, MIP-2 and MCP-1, to lethal shock in mice exposed to SEB or LPS and their combination. Our survival analysis indicated that increasing LPS dose had more effect in influencing mortality than SEB dose with SEB plus LPSinduced shock. The cytokine data also showed that a higher LPS dose with a constant SEB dose was responsible for significantly higher serum levels of TNFα, IL-1, IL-6, IL-2, and MCP-1. Clearly, the mortality data paralleled the cytokine data indicating synergistic action of SEB and LPS, with LPS as the dominant factor in this mouse model of toxic shock.
Cytokines are intracellular signaling proteins released from virtually all nucleated cells that regulate cell differentiation, proliferation, and inflammation [26]. Dysregulation of cytokine production has been associated with a variety of diseases, including autoimmune disorders, infectious diseases, cardiovascular diseases, asthma, and allergy. We examined the cytokines TNFα, IL-1, IL-6, IFNγ, and IL-2 as they are present in high levels in various animal models of shock [16][17][18][19]. The chemokines are chemoattractants produced by many cell types and are potent molecules involved in host defense as they direct leukocyte migration to sites of infection and injury [26]. We chose to examine two prototypical chemokines, MIP-2 and MCP-1, in this study as they influence leukocyte migration in various animal models of infectious diseases and inflammation [27].