Secreted Wnt antagonists in scrub typhus

Abstract

Background: The mechanisms that control local and systemic inflammation in scrub typhus have only been partially elucidated. The wingless (Wnt) signaling pathways are emerging as important regulators of inflammation and infection, but have not been investigated in scrub typhus.

Methodology/principal findings: Plasma levels of secreted Wnt antagonists (i.e. DKK-1, sFRP-3, WIF-1 and SOST) were analyzed in patients with scrub typhus (n = 129), patients with similar febrile illness without O. tsutsugamushi infection (n = 31), febrile infectious disease controls, and in healthy controls (n = 31) from the same area of South India, and were correlated to markers of inflammation, immune and endothelial cell activation as well as for their association with organ specific dysfunction and mortality in these patients. We found i) Levels of SOST and in particular sFRP-3 and WIF-1 were markedly increased and DKK-1 decreased in scrub typhus patients at admission to the hospital compared to healthy controls. ii) In recovering scrub typhus patients, SOST, sFRP-3 and WIF-1 decreased and DKK-1 increased. iii) SOST was positively correlated with markers of monocyte/macrophage and endothelial/vascular activation as well as with renal dysfunction and poor outcome iv) Finally, regulation of Wnt pathways by O. tsutsugamushi in vitro in monocytes and ex vivo in mononuclear cells isolated from patients with scrub typhus, as evaluated by gene expression studies available in public repositories, revealed markedly attenuated canonical Wnt signaling.

Conclusions/significance: Our findings suggest that scrub typhus is characterized by attenuated Wnt signaling possibly involving dysregulated levels of several secreted pathway antagonists. The secreted Wnt antagonist SOST was strongly associated with renal dysfunction and poor prognosis in these patients.

Fig 1. Plasma levels of secreted Wnt antagonists in scrub typhus patients and healthy controls.

Levels of Dickopff-1 (DKK-1), secreted frizzled related protein-3 (sFRP-3), sclerostin (SOST) and Wnt inhibitory factor (WIF-1) in scrub typhus patients (n = 129) on admission (A) and at recovery (R) as well as comparative levels in healthy controls (HC, n = 31) and infectious disease controls (ID, n = 31). Data are given as medians and 25th, 75th percentiles. *p<0.05, **p<0.01 and ***p<0.001 versus healthy controls; ^††p < 0.01 and ^†††p < 0.001 versus infectious disease controls. Comparisons between levels at admission and recovery are also shown with p-values.

Fig 2. Plasma sclerostin (SOST) and associations with renal function, inflammation, and adverse outcome.

(A) Correlation between SOST and creatinine in patients with scrub typhus. The small graph shows SOST levels according to renal dysfunction (i.e. eGFR<60). (B) Tissue expression data from human protein atlas (proteinatlas.org/). (C) Association between SOST and inflammatory markers according to kidney function (eGFR<60, n = 53; eGFR>60, n = 76) and severity (Mild Sev = 0, n = 51; Moderate, Sev = 1, n = 37; Severe, Sev = 2/3, n = 41). (D) Correlation (Spearman) between plasma SOST and MIF and YKL-40 according to kidney function (eGFR<60). (E) Receiver operating characteristic analysis showing associations between mortality and SOST levels in scrub typhus patient on admission. (F) Logistic regression showing odds ratio (OR) for SOST in relation to all-cause death (UNI). The impact of each inflammatory marker and eGFR on this association is shown. SOST is presented as log/SD.

Fig 3. Regulation of Wnt pathways by O. tsutsugamushi in mononuclear cells and PBMC, data from public repositories.

Differences in Wnt pathway gene expression in PBMC between patients with scrub typhus and healthy controls (HC). Significant results (FDR adjusted p-values) are indicated in red/blue (for increased/decreased mRNA expression). The figure summarizes relevant genes for the (A) canonical,(B) non-canonical PCP pathway and (C) non-canonical Wnt/Ca2+ pathway. The figure is based on the Wnt signaling pathway in the KEGG database. (D) Heatmap over the regulated mRNA in the canonical pathway (red text), non-canonical PCP pathway (green) and non-canonical Wnt/Ca²⁺ pathway (blue) in PBMC between patients with scrub typhus (n = 4) and healthy controls (n = 2). Red indicates a higher expression. To the right is a table over regulated mRNA between patients with scrub typhus and healthy controls (PBMC), in monocytes (Mono) infected with O. tsutsugamushi compared to non-infected cells (n = 4 in each group) and in patients with scrub typhus compared to murine typhus (MT, Scrub n = 4, typhoid n = 7). *p<0.05, **p<0.01. (E) Correlation between fold change in PBMC (scrub patients vs. controls) and O. tsutsugamushi infected mononuclear cells. (F) Regulation of the canonical Wnt pathways by O. tsutsugamushi in mononuclear cells.