Critical role of extracellular heat shock cognate protein 70 in the myocardial inflammatory response and cardiac dysfunction after global ischemia-reperfusion

Abstract

Previous studies showed that Toll-like receptor 4 (TLR4) modulates the myocardial inflammatory response to ischemia-reperfusion injury, and we recently found that cytokines link TLR4 to postischemic cardiac dysfunction. Although TLR4 can be activated in cultured cells by endogenous agents including heat shock protein 70, how it is activated during myocardial ischemia-reperfusion is unknown. In the present study, we examined 1) whether heat shock cognate protein 70 (HSC70), which is constitutively expressed in the myocardium, is released during ischemia-reperfusion; 2) whether extracellular HSC70 induces the myocardial inflammatory response and modulates cardiac function; and 3) whether HSC70 exerts these effects via TLR4. We subjected isolated mouse hearts to global ischemia-reperfusion via the Langendorff technique. Immunoblotting and immunostaining detected the release of HSC70 from the myocardium during reperfusion. Treatment with an antibody specific to HSC70 suppressed myocardial cytokine expression and improved cardiac functional recovery after ischemia-reperfusion. Recombinant HSC70 induced NF-kappaB activation and cytokine expression and depressed myocardial contractility in a TLR4-dependent manner. These effects required the substrate-binding domain of HSC70. Fluorescence resonance energy transfer analysis of isolated macrophages demonstrated that extracellular HSC70 interacts with TLR4. Therefore, this study demonstrates for the first time that 1) the myocardium releases HSC70 during ischemia-reperfusion, 2) extracellular HSC70 contributes to the postischemic myocardial inflammatory response and to cardiac dysfunction, 3) HSC70 exerts these effects through a TLR4-dependent mechanism, and 4) the substrate-binding domain of HSC70 is required to induce these effects. Thus extracellular HSC70 plays a critical role in regulating the myocardial innate immune response and cardiac function after ischemia-reperfusion.

Fig. 1

Heat shock cognate protein 70 (HSC70) released from the myocardium during ischemia-reperfusion (I/R) contributes to cardiac dysfunction and myocardial cytokine expression. Hearts isolated from C3H/HeN (wild-type) mice were subjected to global ischemia-reperfusion or perfusion only (control). A: coronary effluent was collected before 20-min ischemia (basal) and during the first 5 min (R5) and the last 5 min (R60) of reperfusion. Although immunoblotting did not detect HSC70 in the coronary effluent before ischemia, HSC70 was present in both samples collected during reperfusion. Two HSC70-immunoreactive bands with lower molecular size were also detected in the R5 sample. Myocardial tissue sections were immunofluorescently stained to localize HSC70 after ischemia-reperfusion (20-min ischemia/5-min reperfusion). Top: HSC70 (red) and nuclei (blue); bottom: cell surfaces (green). In control hearts, HSC70 is intracellular and associated with myocyte striations. After ischemia-reperfusion, the striation pattern is lost, and HSC70 is localized in the extracellular space (arrows). B: hearts were subjected to global ischemia-reperfusion (20-min ischemia/60-min reperfusion) with or without treatment with polyclonal anti-HSC70 (HSC Ab, 0.2 µg/ml) or control IgG (IgG, 0.2 µg/ml) for 10 min before ischemia and for 30 min after initiation of reperfusion. Control hearts were perfused with perfusion buffer without being subjected to ischemia. Anti-HSC70 improved the recovery of left ventricular developed pressure (LVDP) and dP/dt max while control IgG had no effect. C: anti-HSC70 reduced myocardial levels of cytokine (TNF-α, IL-1β, and IL-6) mRNA and peptides after ischemia-reperfusion while control IgG had no effect. Data are means ± SE. rHSC, recombinant HSC70; M, myocyte (n = 6 in each group). *P < 0.05 vs. I/R; **P < 0.05 vs. control.

Fig. 2

HSC70 induces the activation of myocardial p38 MAPK and NF-κB through Toll-like receptor 4 (TLR4). Hearts isolated from C3H/HeN (TLR4-competent) and C3H/HeJ (TLR4-defective) mice were perfused with recombinant HSC70 (0.5 µg/ml) for 30 min followed by 60 min washout. Control hearts were perfused without exposure to HSC70. A and B: representative blot and mean densitometry data (means ± SE) of 2 separate experiments show that HSC70 induced phosphorylation of p38 MAPK and ERK1/2 in TLR4-competent hearts. In TLR4-defective hearts, ERK 1/2 phosphorylation was minimally changed, but the effect of HSC70 on p38 MAPK phosphorylation was abrogated. C: HSC70 increased NF-κB DNA-binding activity in TLR4-competent hearts, an effect that was greatly attenuated in TLR4-defective hearts. NF-κB DNA-binding data are means ± SE (n = 6 in each group). *P < 0.05 vs. HeN HSC. p-p38, phosphorylated-p38; p-ERK1/2, phosphorylated-ERK1/2.

Fig. 3

HSC70 induces expression of cardiodepressant cytokines through TLR4. Hearts isolated from C3H/HeN (TLR4-competent) and C3H/HeJ (TLR4-defective) mice were perfused with recombinant HSC70 (0.5 µg/ml) for 30 min followed by 60-min washout. Control hearts were perfused with perfusion buffer without HSC70. A: TLR4-competent hearts expressed cytokine (TNF-α, IL-1β, and IL-6) mRNA in response to HSC70 treatment; TLR4-defective hearts did not. B: cytokine (TNF-α, IL-1β, and IL-6) peptide levels increased in TLR4-competent hearts treated with HSC70 but not in TLR4-defective hearts. Data are means ± SE (n = 6 in each group). *P < 0.05 vs. HeN HSC; **P < 0.05 vs. HeN control.

Fig. 4

HSC70 induces cardiac contractile depression through TLR4. Hearts isolated from C3H/HeN (TLR4-competent) and C3H/HeJ (TLR4-defective) mice were perfused with recombinant HSC70 (0.5 µg/ml) for 30 min followed by 60-min washout. Perfusion with HSC70 depressed LVDP (A) and dP/dt_max (B) in TLR4-competent hearts but not in TLR4-defective hearts. Data are expressed as means ± SE (n = 6 in each group). *P < 0.05 vs. HeJ.

Fig. 5

The effect of HSC70 on cardiac contractility is abrogated in the absence of the substrate-binding domain. Hearts isolated from C3H/HeN (TLR4-competent) mice were perfused with either recombinant full-length HSC70 (0.5 µg/ml) or recombinant HSC70 fragment (without the substrate-binding domain, 0.5 µg/ml) for 30 min followed by 60-min washout. A and B: without the substrate-binding domain, the ability of HSC70 to depress cardiac contractility was lost. Data are means ± SE (n = 6 in each group). *P < 0.05 vs. full-length HSC70.

FIG. 6

HSC70 interacts with TLR4 in macrophages. Macrophages from C3H/HeN (TLR4-competent) mice were incubated with recombinant HSC70 (0.5 µg/ml) for 15 min or left untreated (control). Double immunofluorescent staining labeled HSC70 red and TLR4 green (left). Nuclei were counterstained blue. In untreated cells (top), HSC70 was located in the cytoplasm and no fluorescence resonance energy transfer (FRET) signal between HSC70 and TLR4 was detected after photo-bleaching. In cells incubated with HSC70 (bottom), increased cell-surface localization of HSC70 and colocalization of HSC70 and TLR4 (yellow areas, arrows) were observed. In addition, FRET signals (bottom right, arrows) were detected in the 2 photobleached cells at bottom right but not in the nonphotobleached (control) cell at top left.