The relevance of TLR2 and TLR4 for recognizing Chlamydia pneumoniae in vivo during pulmonary infection and to survive the infection was explored. We found that early immune responses triggered by C. pneumoniae partially depended on TLR2, but not on TLR4. The chemokines MIP-2 and MIP-1alpha were not induced, while IL-12p40 levels were higher in TLR2(-/-) mice compared to wild-type mice. Secretion of TNF, keratinocyte-derived chemokine and monocyte chemoattractant protein-1 was attenuated in TLR2(-/-) mice, while IFN-gamma was increased as in wild-type mice. The pulmonary cyto- and chemokine response of TLR2(-/-) x TLR4(d/d) was similar to TLR2(-/-) mice. TLR2(-/-) and TLR2(-/-) x TLR4(d/d) mice also attracted fewer polymorphonuclear neutrophils into the lung, while TLR4(d/d) mice recruited them. Attenuated recruitment of polymorphonuclear neutrophils correlated with reduced weight loss in TLR2(-/-) and TLR2(-/-) x TLR4(d/d) mice and a lower chlamydial burden 3 days post infection. At 9 days post infection, TLR2(-/-) and TLR2(-/-) x TLR4(d/d) mice produced cyto- and chemokines as efficiently as wild-type mice, indicating that the involvement of TLR in inflammation varies over time. All TLR2(-/-) x TLR4(d/d) mice succumbed to the infection, while about 50% of TLR2(-/-) mice died. Taken together, the function of TLR2 and TLR4 is required to survive pulmonary infection with C. pneumoniae.