Toll-like receptor 2 impairs host defense in gram-negative sepsis caused by Burkholderia pseudomallei (Melioidosis)

Abstract

Background: Toll-like receptors (TLRs) are essential in host defense against pathogens by virtue of their capacity to detect microbes and initiate the immune response. TLR2 is seen as the most important receptor for gram-positive bacteria, while TLR4 is regarded as the gram-negative TLR. Melioidosis is a severe infection caused by the gram-negative bacterium, Burkholderia pseudomallei, that is endemic in Southeast Asia. We aimed to characterize the expression and function of TLRs in septic melioidosis.

Methods and findings: Patient studies: 34 patients with melioidosis demonstrated increased expression of CD14, TLR1, TLR2, and TLR4 on the cell surfaces of monocytes and granulocytes, and increased CD14, TLR1, TLR2, TLR4, LY96 (also known as MD-2), TLR5, and TLR10 mRNA levels in purified monocytes and granulocytes when compared with healthy controls. In vitro experiments: Whole-blood and alveolar macrophages obtained from TLR2 and TLR4 knockout (KO) mice were less responsive to B. pseudomallei in vitro, whereas in the reverse experiment, transfection of HEK293 cells with either TLR2 or TLR4 rendered these cells responsive to this bacterium. In addition, the lipopolysaccharide (LPS) of B. pseudomallei signals through TLR2 and not through TLR4. Mouse studies: Surprisingly, TLR4 KO mice were indistinguishable from wild-type mice with respect to bacterial outgrowth and survival in experimentally induced melioidosis. In contrast, TLR2 KO mice displayed a markedly improved host defenses as reflected by a strong survival advantage together with decreased bacterial loads, reduced lung inflammation, and less distant-organ injury.

Conclusions: Patients with melioidosis displayed an up-regulation of multiple TLRs in peripheral blood monocytes and granulocytes. Although both TLR2 and TLR4 contribute to cellular responsiveness to B. pseudomallei in vitro, TLR2 detects the LPS of B. pseudomallei, and only TLR2 impacts on the immune response of the intact host in vivo. Inhibition of TLR2 may be a novel treatment strategy in melioidosis.

Figure 1. Increased Expression of CD14, TLR1, TLR2, and TLR4 on Peripheral Blood Monocytes and Granulocytes of Patients with Melioidosis

(A) Increased expression of CD14, TLR1, TLR2, and TLR4 on peripheral blood monocytes and granulocytes of patients with severe melioidosis (triangles; n = 34) and healthy controls (squares; n = 32). Horizontal lines indicate median MFI. (B). Patients (n = 5) who survived after 2 wk of intensive treatment showed normalization of CD14, TLR2, and TLR4, but not of TLR1 expression on the cell surface of monocytes at discharge. ns, not significant. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2. TLR2 and TLR4 Contribute to Cellular Responsiveness to B. pseudomallei In Vitro

(A and B) Freshly isolated alveolar macrophages (A) and whole blood (B) of WT, TLR2 KO, and TLR4 KO mice (n = 8 per group) were incubated with RPMI 1640 (control) or heat-killed B. pseudomallei (equivalent 10⁵ CFU; target:effector ratio 1:10) for 16 h before TNFα production was measured. (C) HEK293 cells stably transfected with CD14, CD14/TLR2, or CD14/TLR4 were incubated for 6 h with medium, LPS of B. pseudomallei 1026, or heat-killed B. pseudomallei before measurement of IL8 in the supernatant. Data are mean ± SEM (n = 4). ***p < 0.001.

Figure 3. Survival of TLR2 KO Mice, but Not of TLR4 KO Mice, Is Enhanced during Experimental Melioidosis

(A) Survival after intranasal inoculation with B. pseudomallei in WT (open circles), TLR2 KO mice (closed squares) and (B) TLR4 KO mice (closed squares). Mortality was assessed twice daily for 6 wk. n = 12 per group. p-Value indicates the difference between TLR2 KO and WT mice. ns, not significant.

Figure 4. Bacterial Loads in Lungs, Spleen, and Blood

TLR2 KO mice demonstrate strongly reduced bacterial loads at 72 h after infection in lungs (A), spleen (B), and blood (C). TLR4 KO mice showed a modestly enhanced outgrowth at 24 h after infection in spleen (E) and blood (F) but not in lungs (D). Data are mean ± SEM (n = 8 per group) at 24, 48, and 72 h after inoculation with B. pseudomallei in WT mice (open bars), TLR2 KO mice (black bars), and TLR4 KO mice (crosshatched bars). *p < 0.05; ***p < 0.0001. ns, not significant.

Figure 5. Reduced Lung Inflammation in TLR2 KO Mice 72 Hours after Infection

Representative lung histology of WT (A, C, and E) and TLR2 KO mice (B, D, and F) at 24 h (A and B), 48 h (C and D), and 72 h (E and F) after inoculation with 5 × 10² CFU B. pseudomallei. TLR2 KO showed significantly less inflammation, pleuritis, peribronchial inflammation, edema, endothelialitis, and necrosis 72 h after infection in the mice (F and G) compared to WT (E–G). Pathology scores (mean ± SEM) (G) were calculated as described in the Methods section; open bars represent WT mice, and solid bars represent the TLR2 KO mice. The insets (A–F) are representative images of immunostaining for granulocytes, showing dense granulocytic infiltrations and confirming reduced inflammation and granulocyte influx in the TLR2 KO mice at 72 h after inoculation with B. pseudomallei. TLR2 KO mice that survived the experiment shown in Figure 3A were examined 6 wk after inoculation; as expected, they showed decreased organ inflammation as compared to the 72 h time point The abscesses that were still present were encapsulated (H). Magnification, 20×. All data are from eight mice per group at each time point, except for (H) (n = 5).

Figure 6. TLR2 KO Mice, but Not TLR4 KO Mice, Show Reduced Pulmonary Cytokine Levels during Experimental Melioidosis

WT mice (open symbols) (A–F), TLR2 KO mice (closed symbols) (A, C, and E), and TLR4 KO mice (closed symbols) (B, D, and F) were intranasally infected with 5 × 10² CFU of B. pseudomallei. 24, 48, and 72 h after inoculation mice were sacrificed, lungs were removed, and TNFα (A and B), IL6 (C and D), and IL10 (E and F) were measured in lung homogenates. Data are means ± SEM of eight mice per group per time point. *p < 0.05; **p < 0.001; ***p < 0.0001.

Figure 7. No Difference in TLR4 Expression in TLR2 KO Mice Compared to WT Mice during Infection with B. pseudomallei

TLR4 cell surface expression is depicted over time (0, 24, 48, and 72 h after intranasal inoculation with B. pseudomallei), on the lung macrophages (A), blood monocytes (B), and blood granulocytes (C). Data are expressed as mean ± SEM and n = 4 mice per group at each time point. MFI, mean fluorescence intensity.

Figure 8. TLR2 KO Mice Demonstrate Reduced Distant Organ Injury

Mice were intranasally inoculated with 5 × 10² CFU of B. pseudomallei. After 72 h, representative spleen histology of WT (A) and TLR2 KO (B) mice demonstrated decreased inflammation and less thrombus formation in the TLR2 KO spleens. Magnification, 12.5×. Arrows indicate increased thrombus formation in the WT mice compared to TLR2 KO mice. The reduced inflammation and injury in the spleen at 72 h after infection (C) was confirmed by the semiquantitative pathology score described in the Methods. At this time point TLR2 KO mice also showed less hepatic injury, as reflected by the plasma concentrations of ASAT (D) and ALAT (E), and less renal failure, as reflected by plasma creatinine (F) and blood urea nitrogen (BUN) (G). Data are expressed as mean ± SEM of eight WT mice (open bars) and eight TLR2 KO mice (black bars). *p < 0.05; **p < 0.01 versus WT control.