Phosphoinositide 3-kinase gamma/delta inhibition limits infarct size after myocardial ischemia/reperfusion injury

Abstract

Although phosphoinositide 3-kinases (PI3Ks) play beneficial pro-cell survival roles during tissue ischemia, some isoforms (gamma and delta) paradoxically contribute to the inflammation that damages these same tissues upon reperfusion. We therefore considered the possibility that selectively inhibiting proinflammatory PI3K isoforms during the reperfusion phase could ultimately limit overall tissue damage seen in ischemia/reperfusion injuries such as myocardial infarction. Panreactive and isoform-restricted PI3K inhibitors were identified by screening a novel chemical family; molecular modeling studies attributed isoform specificity based on rotational freedom of substituent groups. One compound (TG100-115) identified as a selective PI3K gamma/delta inhibitor potently inhibited edema and inflammation in response to multiple mediators known to participate in myocardial infarction, including vascular endothelial growth factor and platelet-activating factor; by contrast, endothelial cell mitogenesis, a repair process important to tissue survival after ischemic damage, was not disrupted. In rigorous animal MI models, TG100-115 provided potent cardioprotection, reducing infarct development and preserving myocardial function. Importantly, this was achieved when dosing well after myocardial reperfusion (up to 3 h after), the same time period when patients are most accessible for therapeutic intervention. In conclusion, by targeting pathologic events occurring relatively late in myocardial damage, we have identified a potential means of addressing an elusive clinical goal: meaningful cardioprotection in the postreperfusion time period.

Fig. 1.

Models of PI3K isoform-kinase inhibitor interactions. (a) Superposition of TG100-115, TG100713, and TG101110 in PI3Kγ showing preferred rotatable angles between ring A and the pteridine core. (b) Superposition of the same three compounds in PI3Kα. (c) Model structure of human PI3Kγ kinase (ribbon) with TG100-115 located in the catalytic domain.

Fig. 2.

Inhibition of cell proliferation and signaling. (a) EC were cultured in the presence of vehicle (DMSO) or PI3K inhibitors (all at 10 μM); cell proliferation was assessed 24 (open bar), 48 (gray bar), or 72 h (black bar) later. Data are presented as OD at 450 nm (mean ± SEM, n = 6; at all timepoints, vehicle and TG100-115 groups differ from all others but not each other by P < 0.001). (b) EC were cultured in serum-free medium (nonstimulated), medium with added VEGF, or medium with VEGF plus 10 μM TG100-115. Cell lysates were then processed for Western blot analyses to detect phosphorylated VE-cadherin or ERK1/2, or total ERK2 (as a loading control).

Fig. 3.

Inhibition of edema and inflammation. (a and b) Rats were injected i.v. with Evans blue dye and then intradermally with saline, VEGF, or histamine. Pretreatment with TG100-115 (1 mg/kg) (b) reduced edema formation relative to vehicle-treated animals (a). (c–f) Rat hindpaws were injected with either PAF (c and d) or dextran (e and f) and processed 3 h later as H&E-stained paraffin sections. Pretreatment with TG100-115 (d and f; 5 mg/kg) blocked both the edema and leukocytic infiltrate induced by these two inflammatory mediators relative to animals dosed with vehicle alone (c and e). Images were taken of paws representing the mean group value for volume as presented in Results. (Original magnification ×200.)

Fig. 4.

Reduction of infarct development in a rodent MI model. (a) Rats were subjected to 60 min of LAD occlusion followed by vehicle or TG100-115 delivery (at the indicated dose) 60 min after reperfusion. Both ischemic area [area at risk (AAR)] and infarct area were then determined 24 h after study initiation. Data are shown as infarct area as a percentage of the AAR (means ± SEM, n = 6; ∗, 0.5 and 5 mg/kg TG100-115 dose groups differ from vehicle control by P < 0.05 but not from one another). (b) Animals were treated as in a, except that a single TG100-115 dose (0.1 mg/kg) was delivered at 0-3 h after reperfusion; in one group, animals were dosed at both 0 and 3 h. Data shown are as in a (n = 5–9; ∗, all TG100-115 groups differ from vehicle control by P < 0.001 but not from one another).

Fig. 5.

Reduction of infarct development in a porcine MI model. Pigs were subjected to 90 min of LAD occlusion followed by vehicle or TG100-115 (0.5 mg/kg) delivery 30 min after reperfusion. At 24 h after study initiation, total ischemic area (AAR), viable AAR, and infarcted AAR were determined, for the entire LV as well as the free wall alone. Data are shown as infarct as a percentage of AAR or viable AAR as a percentage of total AAR (means ± SEM, n = 12–13; ∗, vehicle and TG100-115 groups differ for all measures by P ≤ 0.05).