Introduction

Sepsis and sequential multiple-organ failure/dysfunction syndrome (MOF/MODS) correlated with poor outcome and septic shock is the most common cause of death in intensive care units [1]. The immunological cascade of sepsis response leads to increased T-cell apoptosis, lymphopenia, and altered T-lymphocyte subpopulations [2]. The particular type of immune response is determined by the differentiation of precursor T helper (Th0) cells into Th1 or Th2 cells. Th1 cells produce proinflammatory cytokines, including interleukin (IL)-2, interferon (IFN)-γ and tumor necrosis factor (TNF)-α, and favor cell-mediated immune responses and IgG2 synthesis. Th2 cells secrete IL-4, IL-5, IL-13, IgG1, and IgE, and they play a pivotal role in the inflammatory response of various diseases [3]. Because Th1 and Th2 subsets demonstrate differing patterns of cytokine production, the ratio of Th1 to Th2 cells may be important for the outcome of sepsis. In addition, the balance of cytokines produced in severe sepsis is likely to have a marked effect on Th subset predominance in vivo. Moreover, Th2 antibody-mediated immune responses appear to predominate in sepsis [4, 5]. Other nosological entities, such as asthma and autoimmune diseases, have been reported to have a similar Th2 cell-type immune response [3].

The ST2 protein, also designated T1, Fit-1, and DER4, is a novel Th2 specific product. Three distinct type of ST2 gene products, a soluble ST2 (ST2), a transmembrane receptor form (ST2L), and a variant form (ST2 V) have been cloned [6, 7, 8]. The ST2 gene was originally described as primary response genes in mouse fibroblast cell lines and HA-ras oncogene-responsive gene [9, 10]. ST2 encodes an orphan receptor belonging to the Ig superfamily, and the ST2 gene is tightly linked to the genes encoding the IL-1 receptor type I and type II on human chromosome 2 [10, 11, 12]. According to the Human Gene Nomenclature Database the interleukin-1 receptor-like-1 (IL1R1) is the designated symbol for the soluble ST2, IL1RL1-a is the soluble ST2 receptor, and IL1Rl-b is ST2L [13]. The ST2 genes are preferentially expressed on activated Th2 cells but not Th1 cells, and play an essential role in Th2 effector functions [14, 15], characterized by elevated expression of the cytokines IL-4, IL-5, and IL-13 [16]. The ST2 protein is also expressed on cell membranes of mast cells [17], and serum protein levels are increased in patients with asthma and allergic airways inflammation [18, 19]. Recently, elevated serum concentrations of soluble ST2 protein were found in idiopathic pulmonary fibrosis [20], and various autoimmune diseases [21].

Since sepsis was shown to be related to increased production of anti-inflammatory Th2 cytokine [4, 5], we investigated the presence of soluble ST2, Th1-Th2 cytokine, and Ig content in critically ill septic and non-septic patients. Patients undergoing abdominal surgery and healthy controls served as controls.

Materials and methods

Patients and clinical features

The institutional review board of the Clinical Research Ethics Committee of University of Vienna Medical School, General Hospital, approved the study protocol. All study and control subjects or their legal executive signed a written informed consent. Patients were classified as septic based on the international criteria of the ACCP/SCCM consensus conference (Crit Care Med 1992; 20:864–874). Patients who underwent hemofiltration or radiation therapy, allograft and ventricular assist device recipients were excluded from the study. Moreover, none of our patients suffered from a chronic viral infection or received immunosuppressive therapy.

Immunological investigations were carried out in 15 patients with sepsis, in 13 trauma patients, in 11 patients who underwent abdominal surgery, and in 15 healthy controls. For demographic features see Table 1. All tests were performed immediately after blood sampling. Blood samples were obtained via peripheral venipuncture at 8:00 a.m. (routine blood sampling time) and were acquired within 48 h after the ICU staff diagnosed sepsis, within 24 h of admission to the trauma ICU, and 24 h after abdominal surgery.

Table 1 Profiles of patients and controls. ICU intensive care unit, MVA motor vehicle accident
Full size table

Quantification of serum soluble ST2 levels

Serum samples were aliquoted and deep frozen (−80°C). Soluble ST2 levels were measured with a commercial enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, Minn.). In brief, serum samples and standards were incubated in 96 well microplates precoated with anti-human ST2 antibody. After washing, peroxidase-conjugated anti-human ST2 antibody was added into the microwells and incubated. After another washing the substrate was added, and after adding the stop solution, the optical density was determined at 450 nm. The amount of protein in each sample was calculated according to a standard curve of optical density values constructed for known levels of ST2. The sensitivity of the ELISA kit is 25 pg/ml.

Detection of Ig by solid-phase ELISA

Ninety-six well microtiter plates were coated overnight with 100 µl/well of bovine serum albumin (BSA) as control or with anti-human monoclonal IgE, IgG1, IgG2, or IgM (Sigma, St. Louis, Mo.). Content of IgG1, IgG2, IgM, and IgE was determined by standard IgGs which were diluted/applied in serial dilutions (10–1000 ng/ml) and values compared hereto, respectively. Nonspecific binding was blocked with PBS+1% BSA for 1 h at room temperature. Before use, the serum samples were diluted 1:3 for IgE and 1:5 for IgG and IgM with PBS+1% BSA. Each plate was equipped with human IgE, IgG1, IgG2, and IgM as standard and with the diluted sera of 15 septic patients, 13 critically ill patients, and 15 healthy controls, and was then incubated for 2 h at room temperature. After another washing step with 0.05% Tween/PBS, Ig binding was quantified by incubating with peroxidase conjugated rabbit anti-human IgE, IgG or IgM (Sigma) at a dilution of 1:1000 for 1 h at room temperature followed by the substrate reaction with 1,2 phenylenediamine. Plates were read at 490 nm and specific binding was calculated according to a standard curve of optical density values constructed for known levels of Ig.

Quantification of serum cytokine levels

Circulating serum levels of IL-2, IL-10, and IFN-γ were measured by ELISA (Biosciences, San Diego, Calif.). The sensitivity of all ELISA is 2 pg/ml. The amount of protein in each sample was calculated according to a standard curve of optical density values constructed for known levels of IL-2, IL-10, and IFN-γ.

Statistical analysis

Results are presented as mean±SEM, if not otherwise mentioned. Due to the relatively small sample size, the Mann-Whitney U test was used to calculate significance. This is a descriptive study with no main a priori hypothesis; a p value of 0.05 was deemed to be significant.

Results

Serum T1/ST2 levels

Figure 1 demonstrates a significant increase in the concentration of ST2 in the sera from patients with sepsis (8420±2169 pg/ml) as compared with that of trauma patients (2936±826 pg/ml), abdominal surgery patients (1423±373 pg/ml), and healthy controls (316±72 pg/ml). These differences were significantly altered (sepsis vs trauma, p=0.04; sepsis vs surgery, p=0.014; sepsis vs healthy, p=0.001; trauma vs healthy, p=0.003; surgery vs healthy, p=0.003).

Fig. 1
figure 1

Concentration of soluble ST2 protein. The level of ST2 serum in the sera from 15 septic patients, 13 trauma patients, 11 abdominal surgery patients, and 15 healthy controls were measured by ELISA. The bars represent the mean values of ST2 (in pg/ml), the whiskers represent the SEM.* p<0.05, ** p<0.01, *** p<0.001

Full size image

Serum isotype Ig levels

The results depicted into Fig. 2 show significant differences in the Ig concentrations between sepsis and critically ill patients, for TH1 specific IgG1 (p=0.001) and IgG2 (p=0.001). Differences between the septic and the healthy cohort were significant for IgG1 (p=0.003) and IgG2 (p<0.001). The difference in IgG2 levels between critically ill patients and healthy controls was also significant (p=0.001).

Fig. 2
figure 2

Concentration of cytokine. The concentration of cytokines in the sera from 15 septic patients, 13 trauma patients, 11 abdominal surgery patients, and 15 healthy controls were measured by ELISA. The bars represent the mean values of IgE, IgG1, IgG2, and IgM in the sera (in pg/ml), the whiskers represent the SEM. * p<0.05, ** p<0.01, *** p<0.001

Full size image

Serum cytokine levels

As can seen from Table 2 septic patients demonstrated decreased IL-2 and IFN-γ values as compared with the other patient cohorts (p<0.05 for all). Moreover, IL-10 was significantly increased in septic and trauma patients as compared with the surgery group and healthy controls (p<0.01 for all).

Table 2 Serum cytokine levels
Full size table

ST2 as a predictor for mortality

To determine the value of high ST2 levels as a predictor for mortality we compared ST2 serum levels of the patients who died with the levels of those who survived. In the sepsis group the difference did not reach statistical significance (p=0.535). Surprisingly, we found a difference in the trauma group; the mean ST2 level (pg/ml) in the cohort that survived was 4133±1008 (n=9), whereas that in the patients who died was 244±147 (n=4; p=0.029).

Discussion

Our study showed for the first time that serum ST2 and humoral immune response, as indicated by increased IgG1 isotype production, is significantly augmented in patients with sepsis and trauma as compared with control populations. Furthermore, a marked decrease in the concentrations of proinflammatory cytokines IL-2 and IFN-γ was found in the sera from septic patients but not with trauma or abdominal surgery, whereas IL-10 levels were significantly elevated in sepsis as well as in trauma patients. Because ST2 is known as a stable marker for Th2 cells [14, 15, 16], our data showing increase IL-10 and decrease IL-2 and IFN-γ production are in line with previous reports [22, 23], and demonstrate a Th2 shift in septic patients.

The biological function of the T1/ST2 receptor remains unclear, but its homology with the Toll-like receptor (TLR) [20] and other IL-1 family members [9, 22, 23] suggests that it may play a central role in innate and adaptive immune responses. Injection of anti-ST2 antibody in Lesishmmania major-infected mice increases the production of IFN-γ, decreases the synthesis of IL-4 and IL-5, and exacerbates collagen-induced arthritis [15]. In murine asthma models, the administration of recombinant ST2 fusion protein or soluble ST2 gene transfer attenuates eosinophilic inflammation of airway, and suppresses IL-4 and IL-5 production [12, 15]. Furthermore, in murine lipopolysaccharide (LPS)-induced shock models, the ST2 fusion protein enhances the survival rate and suppresses IL-6, IL-12, and TNF-α production by acting directly on macrophages via the ST2-TLR-4 route [24]. ST2 expression was shown to be induced by IL-1β, IL-1α, and TNF-α in fibroblasts, macrophages, muscle, and in spleen following LPS challenge that is comparable to the sepsis syndrome [22, 23]; thus, macrophages respond to LPS by producing Th2-biased cytokines that in turn can induce the expression of a putative soluble ST2 protein. Recently, enhanced concentrations of soluble ST2 receptor was found in sera from patients as well as from mice after myocardial infarction, and was considered to be a predictor of subsequent mortality [25].

The mechanism by which soluble ST2 protein suppresses the production of proinflammatory cytokines is presently not well understood. Nuclear factor (NF)-kβ is important transcription factor required for the expression of inflammatory cytokines. In mast cells, soluble ST2 is induced by GATA-1 transcription factor via activation of the distal exon-1a promoter region [26]. ST2 can activate c-Jun terminal kinase, p42/p44, and p38 mitogen-activated protein (MAP) kinases, and the transcription factor activator protein-1 (AP-1), but it apparently does not activate NF-kβ [27, 28]. These findings are consistent with a role of ST2 in Th2 cell function since NF-kβ activation is more associated with Th1 type responses. It has been further shown that in vitro production of Th2-type cytokines precedes the expression of ST2 in Th2 cell polarized [29]. Most importantly, the cross-linking of ST2 provided a costimulatory signal for Th2 but not Th1 cells, and directly induces cell proliferation and Th2 cytokine production [29]. Further evidence that ST2 molecule is a regulator of type-1 immune response, and that signaling through ST2 molecule is not necessary for Th2 differentiation, have been recently shown in a experimental model of nonhealing Leishmania major infection [30]; therefore, it can be speculated from these data that ST2 may function as an important mediator in this negative feedback loop for preventing uncontrolled inflammatory reactions.

Conclusion

In conclusion, our results demonstrate that soluble ST2, a marker for Th2 cytokine-producing cells, is increased during sepsis as well as in trauma patients. They provide further evidence pertaining to a dispute concerning the shift from Th1 to Th2 cytokine milieu in septic patients. To the best of our knowledge, we show for the first time an increased levels of soluble ST2 and IgG1 production, concomitant with a marked decrease in the Th1 cytokine profile in sera of critically ill patients. We suggest that these serum markers have the potential to determine the immune status and the progression of sepsis and trauma in general. Our study is, however, limited through the unknown variables such as circadian rhythm and the serum half time of ST2. But the relative significance, most notably trauma vs sepsis, supports the importance of our finding.