HMGB1 and TLR4 mediate skeletal muscle recovery in a murine model of hindlimb ischemia.

BACKGROUND

We have previously shown that the danger signal high-mobility group box 1 (HMGB1) promotes angiogenesis when administered to ischemic muscle. HMGB1 signals through Toll-like receptor 4 (TLR4) as well as the receptor for advanced glycation end-products (RAGE). However, the actions of these receptors in ischemic injury and muscle recovery are not known. We hypothesize that TLR4 mediates tissue recovery and angiogenesis in response to ischemia.

METHODS

Femoral artery ligation was performed in control, TLR4 competent (C3H/HeOuJ) and incompetent (C3H/HeJ) mice, as well as RAGE knockout mice and their C57B6 control counterparts. In other experiments, control mice were pretreated with anti-HMGB1 neutralizing antibody before femoral artery ligation. After 2 weeks, limb perfusion was evaluated using laser Doppler perfusion imaging and reported as the ratio of blood flow in the ischemic to nonischemic limb. Muscle necrosis, fat replacement, and vascular density in the anterior tibialis muscle were quantified histologically. In vitro, TLR4 and RAGE expression was evaluated in human dermal microvascular endothelial cells in response to hypoxia. Human dermal microvascular endothelial cells treated with HMGB1 alone and in the presence of anti-TLR4 antibody were probed for phosphorylated extracellular signal-regulated kinase (ERK), a signaling molecule critical to endothelial cell (EC) angiogenic behavior.

RESULTS

Both anti-HMGB1 antibody as well as defective TLR4 signaling in HeJ mice resulted in prominent muscle necrosis 2 weeks after femoral artery ligation. Control HeOuJ mice had less necrosis than TLR4 incompetent HeJ mice, but a greater amount of fat replacement. In contrast to control C3H mice, control C57B6 mice demonstrated prominent muscle regeneration with very little necrosis. Muscle regeneration was not dependent on RAGE. While vascular density did not differ between strains, mice with intact RAGE and TLR4 signaling had less blood flow in ischemic limbs compared with mutant strains. In vitro, EC TLR4 expression increased in response to hypoxia while TLR4 antagonism decreased HMGB1-induced activation of extracellular signal-regulated kinase.

CONCLUSIONS

Both HMGB1 and TLR4 protect against muscle necrosis after hindlimb ischemia. However, muscle regeneration does not appear to be tied to vascular density. HMGB1 likely activates angiogenic behavior in ECs in vitro, and this activation may be modulated by TLR4. The improvement in blood flow seen in mice with absent TLR4 and RAGE signaling may suggest anti-angiogenic roles for both receptors, or vasoconstriction induced by TLR4 and RAGE mediated inflammatory pathways.