Tel.+49 30 838 - 518 34 Fax.+49 30 838 451 834
Scrub typhus, an infectious disease caused by O. tsutsugamushi, is widely distributed in Asia and the Pacific region including Northern Australia, with more than one billion people at risk of contracting the disease and probably more than one million new cases every year. O. tsutsugamushi, which was classified as a separate genus in the Rickettsiaceae family, is an obligate intracellular bacterium that is transmitted via the skin by mite bites. Little is known about the early immunological events of O. tsutsugamushi infection. Particularly, the role of immediate innate immune responses induced by pattern recognition receptors such as toll-like receptors (TLRs) and NOD-like receptors in response to O. tsutsugamushi remains poorly understood.
The goal of this study was to identify an innate receptor responsible for recognition of bacterial surface structures of O. tsutsugamushi and to investigate its role in a mouse infection model. First, an in vitro overexpression system for innate receptors was used in order to test reactivity with O. tsutsugamushi. HEK293 cells transiently transfected with human TLR2, TLR4, NOD1 or NOD2, were stimulated or infected with O. tsutsugamushi and the production of IL-8 was measured in cell culture supernatants. It was thus shown that TLR2, but not TLR4, NOD1 or NOD2, has a role in recognition of both inactivated and live O. tsutsugamushi. The TLR2 ligand of O. tsutsugamushi was heat-stable and showed complete sensitivity to treatment with hydrogen peroxide and sodium hydroxide, and partial sensitivity to proteinase K, suggesting the presence of two different TLR2 ligands in O. tsutsugamushi, possibly a lipopeptide and a protein.
The requirement of TLR2 for the induction of pro-inflammatory responses was investigated in the murine macrophage cell lines 232 (C57BL/6 wild-type) and 261 (TLR-deficient). It was found that O. tsutsugamushi activated the NF-B signal transduction pathway depending on the presence of TLR2. TLR2 deficiency increased the intracellular replication of O. tsutsugamushi in macrophage cell lines in vitro. This increase in replication in TLR2-deficient macrophages was correlated with significantly lowered levels of TNF- production, indicating that decreased TNF- production by TLR2-deficient macrophages could be in part responsible for enhanced growth of O. tsutsugamushi. In consistence with these results, treatment with recombinant TNF- reduced the growth of infectious O. tsutsugamushi in a concentration-dependent manner to a similar extent in both 232 and 261 macrophage cell lines.
Finally, the influence of TLR2-dependent signaling on the kinetics of O. tsutsugamushi dissemination was investigated in vivo. Infection was initiated by subcutaneous inoculation, thus mimicking the natural transmission of O. tsutsugamushi via the skin. TLR2-deficient mice and C57BL/6 wild types showed no significant differences in the recruitment of inflammatory monocytes and neutrophils to the regional lymph node during the first week of infection. On a cellular level, at day 5 post infection, a significantly lower bacterial load was found in TLR2-deficient inflammatory monocytes compared to wild type cells, while neutrophils had very low bacterial loads in both mouse strains. Although TLR2-deficient macrophages supported more efficient bacterial growth in vitro, it is possible that the higher bacterial load in wild type monocytes in vivo reflects enhanced TLR2-mediated bacterial uptake rather than replication. The suppressive effect of TLR2 on bacterial replication may thus be masked in vivo by enhanced uptake.
The O. tsutsugamushi loads in the lymph node and other target organs during the 3-week course of infection did not differ between TLR2-deficient and wild type mice. All animals survived the infection. Thus, TLR2-mediated pathogen recognition influenced neither the recruitment of monocytes and neutrophils nor the bacterial dissemination in vivo or overall survival following s.c. infection. Possibly, TLR2-mediated danger signaling is redundant in vivo during O. tsutsugamushi infection and can be compensated by other innate signaling pathways.
Upon intraperitoneal inoculation of O. tsutsugamushi, which is more severe and potentially lethal infection, surprisingly TLR2 deficiency did not predispose for overwhelming infection, but even ameliorated the severity of symptoms and protected mice from lethal outcomes. This unexpected finding suggests the development of a TLR2-mediated immunopathology in severe, highly replicative infections with O. tsutsugamushi.
This study shows for the first time that TLR2 serves as an innate receptor for O. tsutsugamushi. Instead of contributing significantly to antibacterial immunity in vivo, as it is known from infections with other pathogens, TLR2-dependent signals rather deteriorate the course of infection.