Trauma hemorrhagic shock-induced lung injury involves a gut-lymph-induced TLR4 pathway in mice

Abstract

Background: Injurious non-microbial factors released from the stressed gut during shocked states contribute to the development of acute lung injury (ALI) and multiple organ dysfunction syndrome (MODS). Since Toll-like receptors (TLR) act as sensors of tissue injury as well as microbial invasion and TLR4 signaling occurs in both sepsis and noninfectious models of ischemia/reperfusion (I/R) injury, we hypothesized that factors in the intestinal mesenteric lymph after trauma hemorrhagic shock (T/HS) mediate gut-induced lung injury via TLR4 activation.

Methods/principal findings: The concept that factors in T/HS lymph exiting the gut recreates ALI is evidenced by our findings that the infusion of porcine lymph, collected from animals subjected to global T/HS injury, into naïve wildtype (WT) mice induced lung injury. Using C3H/HeJ mice that harbor a TLR4 mutation, we found that TLR4 activation was necessary for the development of T/HS porcine lymph-induced lung injury as determined by Evan's blue dye (EBD) lung permeability and myeloperoxidase (MPO) levels as well as the induction of the injurious pulmonary iNOS response. TRIF and Myd88 deficiency fully and partially attenuated T/HS lymph-induced increases in lung permeability respectively. Additional studies in TLR2 deficient mice showed that TLR2 activation was not involved in the pathology of T/HS lymph-induced lung injury. Lastly, the lymph samples were devoid of bacteria, endotoxin and bacterial DNA and passage of lymph through an endotoxin removal column did not abrogate the ability of T/HS lymph to cause lung injury in naïve mice.

Conclusions/significance: Our findings suggest that non-microbial factors in the intestinal mesenteric lymph after T/HS are capable of recreating T/HS-induced lung injury via TLR4 activation.

Figure 1. Pig T/HS lymph infusion in naïve CD1 mice recreates T/HS-induced lung injury.

Lung permeability to Evans Blue dye (EBD) was measured in CD-1 mice subjected to actual T/HS or T/SS for 60 min and 3 hr reperfusion as well as CD1 mice infused with porcine T/SS or T/HS lymph for 3 hr. Data expressed as mean ± SE (n = 5–6 mice/ group).

Figure 2. TLR4 deficiency attenuates T/HS lymph infused lung injury.

WT and TLR4^mut mice were infused with porcine T/SS and T/HS lymph for 3 hr. A) Lung permeability to EBD was performed. Data expressed as mean ± SE (n = 3–12 mice/ group). B) MPO levels (U/g) were measured in lung homogenates. Data expressed as mean ± SE (n = 4–7 mice/group).

Figure 3. TRIF and Myd88 deficiency confer full and partial protection against T/HS lymph induced microvascular permeability.

A) WT and Myd88^−/− and B) WT and TRIF^mut mice were infused with porcine T/SS and T/HS lymph for 3 hr and lung permeability to EBD was performed. Data expressed as mean ± SE (n = 5–8 mice/ group).

Figure 4. TLR2 does not mediate T/HS lymph-induced lung injury.

WT and TLR2^−/− mice infused with porcine T/HS and T/SS lymph for 3 hr. A) Lung permeability to EBD was performed. Data expressed as mean ± SE (n = 4–6 mice/ group). B) MPO levels (U/g) were measured in lung homogenates. Data expressed as mean ± SE (n = 5–6 mice/group).

Figure 5. TLR4 deficiency reduces T/HS lymph induced pulmonary iNOS protein levels.

A) and B) Western blot of iNOS in lung WCEs of WT and TLR4^mut mice infused with porcine T/SS and T/HS lymph for 3 hr. B) Densitometry was performed to quantify iNOS and total p42/p44 MAPK expression. Data expressed as mean ± SE (n = 4–7 mice/group).

Figure 6. Endotoxin- and bacterial DNA-independent factors in T/HS lymph mediate lung injury.

A) Pre- and post-polymyxin-immobilized column eluted T/SS and T/HS lymph samples were infused in naïve C57BL6 mice for 3 hours. Lung permeability to EBD. Data expressed as mean ± SE (n = 3–9 mice/ group). B) DNA isolated from lymph pooled from 3 pigs subjected to either T/HS or T/SS lymph was tested for bacterial DNA contaminants by PCR using 16S rDNA primers. DNA isolated from E.Coli served as our positive control was depicted as + and the negative control containing no DNA template was depicted as -. M denotes a 100 bp size marker.