IL-33/ST2 correlates with severity of haemorrhagic fever with renal syndrome and regulates the inflammatory response in Hantaan virus-infected endothelial cells

Abstract

Background: Hantaan virus (HTNV) causes a severe lethal haemorrhagic fever with renal syndrome (HFRS) in humans. Despite a limited understanding of the pathogenesis of HFRS, the importance of the abundant production of pro-inflammatory cytokines has been widely recognized. Interleukin 33 (IL-33) has been demonstrated to play an important role in physiological and pathological immune responses. After binding to its receptor ST2L, IL-33 stimulates the Th2-type immune response and promotes cytokine production. Depending on the disease model, IL-33 either protects against infection or exacerbates inflammatory disease, but it is unknown how the IL-33/ST2 axis regulates the immune response during HTNV infection.

Methodology/principal findings: Blood samples were collected from 23 hospitalized patients and 28 healthy controls. The levels of IL-33 and soluble ST2 (sST2) in plasma were quantified by ELISA, and the relationship between IL-33, sST2 and the disease severity was analyzed. The role of IL-33/sST2 axis in the production of pro-inflammatory cytokines was studied on HTNV-infected endothelial cells. The results showed that the plasma IL-33 and sST2 were significantly higher in patients than in healthy controls. Spearman analysis showed that elevated IL-33 and sST2 levels were positively correlated with white blood cell count and viral load, while negatively correlated with platelet count. Furthermore, we found that IL-33 enhanced the production of pro-inflammatory cytokines in HTNV-infected endothelial cells through NF-κB pathway and that this process was inhibited by the recombinant sST2.

Conclusion/significance: Our results indicate that the IL-33 acts as an initiator of the "cytokine storm" during HTNV infection, while sST2 can inhibit this process. Our findings could provide a promising immunotherapeutic target for the disease control.

Fig 1. Increased IL-33 and sST2 levels in HFRS patients’ plasma.

Scatter diagram displaying the protein levels of IL-33 and sST2 in the plasma of HFRS patients. Comparison of plasma IL-33 (A) and sST2 (B) contents between HFRS patients and healthy donors (NC). Contents of IL-33 (C) or sST2 (D) in the acute phase of HFRS (including febrile, hypotensive, or oliguric stage), the convalescent phase of HFRS (including diuretic or convalescent stage), and healthy donors (NC). Data are the means±SE (HFRS, n = 59; NC, n = 28), ***p < 0.001, HFRS patients versus NC or acute phase versus convalescent phase and NC. Kinetic trends of IL-33 (E) and sST2 (F) levels in each HFRS patient. Kinetic analysis of plasma IL-33 (G) and sST2 (H) levels. Spearman correlation test showing that the plasma contents of sST2 and IL-33 in the entire group of subjects are positively correlated (I). The r and p values are indicated in the graphs. The 1–10 days group had higher ratios of sST2 to IL-33 than the > 10 days group (***p < 0.001) (J).

Fig 2. Increased levels of IL-33 and sST2 in HFRS patients’ plasma are positively correlated with HFRS severity.

Spearman correlation test representing that the plasma IL-33 contents in the entire group of subjects are positively correlated with the white blood cell (WBC) count (A) and viral load (B) but negatively correlated with the platelet count (PLT) (C). The plasma sST2 contents in the entire group of subjects are also positively correlated with the white blood cell (WBC) count (D) and viral load (E) but negatively correlated with the platelet count (PLT) (F). The r and p values are indicated in the graphs.

Fig 3. IL-33 induces the expression of pro-inflammatory cytokines in HTNV-infected HUVECs.

HUVECs were infected with HTNV/mock virus (MOI = 1) for 48 h or stimulated with IL-33 (20 ng/ml) for 6 h or first infected with virus for 48 h and then treated with IL-33 (20 ng/ml) for another 6 h. The mRNA expression of IL-1β, IL-6, IL-8, CCL2, CCL20, CXCL1, CXCL2, and CX3CL1 was determined by real-time PCR. Untreated HUVECs were set as the normal control (NC). Data are shown as the mean ± SD of triplicate samples and are representative of experiments with three independent HUVEC donors, *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 4. IL-33 mediates inflammatory responses via the ST2 receptor in HTNV-infected HUVECs.

HUVECs were transfected with an siRNA specific to ST2L for 6 h and then infected with HTNV (MOI = 1) for 48 h. The cells were harvested being treated with 20 ng/ml IL-33 for another 6 hours. The mRNA (A) and protein levels (B) of ST2L were reduced when compared to the scramble control. *p < 0.05, siST2 versus nontargeting control siNC. (C) HUVECs were transfected with an siRNA specific to ST2L for 6 h or first prior incubated with soluble recombinant human ST2 protein (sST2, 100 ng/ml) for 2 h; siNC or isotype IgG (100 ng/ml) was set as the control, respectively. The treated cells were then stimulated with both HTNV and IL-33, as indicated above. The mRNA levels of IL-1β, IL-6, IL-8, CCL2, CCL20, CXCL1, CXCL2, and CX3CL1 were determined by real-time PCR. The results shown are the mean± SD of triplicate samples and are representative of experiments with three independent HUVEC donors, ***p < 0.001.

Fig 5. Expressions of sST2 and ST2L in HUVECs.

HUVECs were treated as previously indicated, and whole-cell lysates were harvested. The mRNA levels of sST2 (A) and ST2L (B) were determined by real-time PCR. The protein levels of sST2 and GAPDH were analysed by western blotting (C), and the level of ST2L expression was determined by flow cytometric analysis (D). Data are the means ± SE, *p < 0.05, **p < 0.01.

Fig 6. IL-33 and HTNV synergistically induce the activation of the NF-κB pathway.

(A) HUVECs were infected with HTNV/mock virus (MOI = 1) for 48 h or were stimulated with IL-33 (20 ng/ml) for 8 min or were pre-infected with HTNV for 48 h and treated with IL-33 (20 ng/ml) for another 8 min. Whole-cell lysates were harvested, and the amounts of phospho-IKK, total IKK, phospho-IκB, total IκB, and GAPDH were determined by western blotting. (B) A semiquantitative analysis of p-IKK, IKK, p-IκB, and IκB levels was performed using Image J software. The expression of GAPDH was used as the control. (C) The plasmid pNF-κB-luc was transfected into HUVECs. Four hours after transfection, the cells were infected with HTNV for 48 h or were treated with IL-33 (20 ng/ml) for 6 h or cells were first infected with HTNV for 48 h and then treated with IL-33 (20 ng/ml) for another 6 h. HUVECs treated only with growth medium were set as the normal control (NC). The relative luciferase activities were detected. NC is set as one, and the others are presented as fold values relative to NC. (D) The phospho-JNK, total JNK, phospho-ErK, total ErK, phospho-p38, total p38, and tubulin levels were determined by western blotting using the same cell lysates indicated above. (E) A semiquantitative analysis of the p-JNK, JNK, p-ErK, ErK, p-p38, and p38 levels was performed using Image J software. The expression of β-tubulin was set as the control. Data are the means ± SE, *p < 0.05, **p < 0.01, ***p < 0.001.

Fig 7. Divergent effects of sip65 and PDTC on IL-33-mediated inflammatory responses in HUVECs infected with HTNV.

(A) An siRNA specific to p65 was transfected into HUVECs for 6 h, and the cells were treated as indicated above. Half of the cell lysate was collected, and the expression of p65 and GAPDH was analysed by western blotting. (B) RNA was extracted from the other half, and the mRNA levels of pro-inflammatory cytokines were determined by real-time PCR. All the experiments were repeated in triplicate. The mRNA data were generated from three independent experiments using three independent HUVECs donors. (C) HUVECs were first exposed to PDTC (100 μM) for 2 h and infected with HTNV (MOI = 1) for 48 h and treated with IL-33 (20 ng/ml) for another 6 h. Cells in a medium containing DMSO were set as the reagent control. RNA was extracted from these cells, and the mRNA levels of pro-inflammatory cytokines were determined by real-time PCR. All the experiments were repeated in triplicate. The mRNA data were generated from three independent experiments using three independent HUVECs donors. Data are the means ±SE. *p < 0.05, **p < 0.01, ***p < 0.001, siRNA specific to p65 versus siNC-treated cells, PDTC 0 μM versus PDTC 100 μM, or PDTC 0 μM versus the DMSO control.