Fig 1.
Expression of the IL-33 and IL-33R/ST2L genes during Orientia infection.
C57BL/6J mice (4-5/group) were inoculated i.v. with O. tsutsugamushi Karp strain (4.5 x 106 FFU, gray bars) or with PBS (open bars). At 0, 2, 6 and 10 days post-infection (dpi), total RNA was extracted from indicated tissues, and mRNA levels of IL-33 and ST2L in the kidneys (A), livers (B) and lungs (C) were analyzed by qRT-PCR. Data are presented as “qPCR fold” (after normalization to the house-keeping genes), and are shown as mean ± SEM in each group. Representative results are shown from two independent studies with similar trends. **, p < 0.01; NS, no significance.
Fig 2.
Body weight and tissue bacterial loads of Orientia-infected mice.
Wild-type (WT) and IL-33-/- mice were lethally challenged with O. tsutsugamushi Karp strain, as in Fig 1, and monitored daily for whole body weight. Data are presented as total weight loss/mouse (A). Weight loss in WT mice began at 2–3 dpi and continued until they were moribund (9 dpi). IL-33-/- mice had significantly less weight-loss throughout the course of disease and were not moribund at the time of termination of the experiments (9 dpi). Bacterial loads in the kidneys (B), livers (C), and lungs (D) were determined by qPCR. **, p < 0.01; ***, p < 0.001; NS, no significance.
Fig 3.
Selective activation of cytokines in the kidneys of infected IL-33-/- mice.
WT and IL-33-/- mice were lethally challenged with O. tsutsugamushi Karp strain, as in Fig 2. (A) Total RNA was extracted from the kidneys at 9 dpi for measuring indicated transcripts by using qRT-PCR. Data are presented as “qPCR fold” (after normalization to the house-keeping genes), and are shown as mean ± SEM in each group. (B) Kidney tissue homogenates were prepared at 9 dpi for measuring protein levels via Bioplex. Representative results are shown from two independent studies with similar trends. *, p < 0.05; **, p < 0.01; NS, no significance.
Fig 4.
Renal histopathology of Orientia-infected mice.
WT and IL-33-/- mice were lethally challenged with O. tsutsugamushi Karp strain, as in Fig 1, and kidney tissues were collected at 9 dpi for analyses. (A) H&E stained sections. Glomerular cellular infiltration and condensed nuclei (in box). (B) Apoptotic cell staining in brown color. Bar = 50 μm. (C) Location of apoptotic cells in the endothelium of WT mice. Bar = 20 μm. (D) The numbers of apoptotic cells were counted for each section. (E) BCL-2 mRNA levels in the kidneys at 9 dpi. *, p < 0.05; ***, p < 0.001.
Fig 5.
Endothelia cell (EC) activation-associated gene expression in the kidneys of infected mice.
WT mice (4-5/group) were inoculated with O. tsutsugamushi Karp strain as in Fig 1. (A) At 0, 2, 6, and 10 dpi, kidney tissues were analyzed for angiopoietin (Ang) 1 and Ang2 expression by qRT-PCR. Data are presented as “qPCR fold” (after normalization to the housekeeping genes), and are shown as mean ± SEM in each group. (B) The Ang2/Ang1 ratios of individual samples were calculated based on the qRT-PCR data and compared with mock controls (0 dpi). WT and IL-33-/- mice (4-5/group) were infected and euthanatized at 9 dpi. Kidney tissues were analyzed for the expression of Ang1, Ang2 (C), Ang2/Ang1 ratio (D), endothelial nitric oxide synthase (eNOS), Endothlin-1 (ET-1) (E), and eNOS/ET-1 ratio (F), respectively. *, p < 0.05; **, p < 0.01; NS, no significance.
Fig 6.
Exacerbated disease and mortality in the presence of exogenous IL-33 during sub-lethal infection.
WT mice were infected with sub-lethal dose of O. tsutsugamushi Karp strain (OtK) and then injected with rIL-33 or PBS every other day. Signs of illness were monitored daily. (A) Body weight was monitored daily for 13 days. Shown are representative results from one of the three independent studies with similar trends. (B) Mouse survival was monitored as above. Shown are survival data pooled from three independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p <0.0001.
Fig 7.
Increased endothelial dysregulation in the kidney of Orientia-infected, rIL-33 injected mice.
WT mice were treated as in Fig 6. The kidney samples were collected at 9 dpi and analyzed by qRT-PCR for the expression of Ang1, Ang2, Ang2/1 ratios (A), eNOS, Endothlin-1, and their ratios (B), as well as for BCL-2, CXCL1 and CXCL2 (C). Data are presented as “qPCR fold” (after normalization to the house-keeping genes), and are shown as mean ± SEM in each group. Representative results are shown from three independent studies with similar trends. *, p < 0.05; ***, p < 0.001; NS, no significance.
Fig 8.
Induced IL-33 and ST2 expression in Orientia-infected human umbilical vein endothelial cells (HUVECs).
Confluent HUVEC monolayers in 24-well plates were left untreated (Medium only, M) or infected with bacteria at the MOI of 3 or 10. Total RNAs were extracted at 24 h (A) and 48 h (B) of infection for qRT-PCR analyses. Data are presented as “qPCR fold” (after normalization to the house-keeping genes), and are shown as mean ± SEM in each group. For each sample, mRNA levels of IL-33, ST2 receptor isoforms (sST2 and ST2L), Ang1 and Ang2 were measured. The Ang2/1 ratios were calculated, as in Fig 4. *, p < 0.05; **, p < 0.01; ***, p < 0.001; NS, no significance.
Fig 9.
A schematic illustration of EC responses in Orientia-infected WT versus IL-33-/- mice.
In WT mice, O. tsutsugamushi infection increases IL-33 and ST2L expression in ECs, and triggers IL-33 release from the nucleus of stressed or activated ECs. The binding of IL-33 to its membrane-bound receptor ST2L can promote EC activation and apoptosis, as judged by altered expression of Ang proteins. These events collectively contributed to increased ET-1 expression, decreased EC integrity, and increased vascular damage. In the absence of IL-33/ST2L signaling, EC stress and apoptosis are attenuated, partially due to relatively higher Ang1, eNOS, and BCL2 levels, but relatively lower ET-1 levels. The preserved EC integrity, together with the production of other immune cytokines facilitates tissue healing and host survival.