Experimental Controls in Lipid Resuscitation Therapy

We read with interest the recently published work by Umar et al.¹  The authors assert that antagonism of free-fatty acid receptor G-protein-coupled Receptor 40 (GPR40) with a G-protein-coupled receptor small molecule inhibitor (GW1100) blocks all cardioprotective effects of lipid emulsion in animal models of ischemia-reperfusion and bupivacaine-induced cardiotoxicity. The hypothesis is appealing given that a fat-based antidote could logically work by interacting with fatty-acid receptors. We believe, however, that the study lacks appropriate controls and relies on untested assumptions to justify its conclusions.

First, the role of GPR40 in cardiac tissue is unclear. Itoh et al. identified GPR40 as a channel in the gastrointestinal-tract that modulates insulin secretion from the pancreas in response to stimulation with free-fatty acids.²  The current work is the first to identify GPR40 in cardiac tissue but fails to identify its physiologic role in normal cardiac function. Given that insulin signaling modulates bupivacaine toxicity,^3–5  and that GPR40 inhibition will abolish insulin-release from the pancreas, appropriate controls for insulin and glucose levels would strengthen the argument for a cardiac (instead of pancreatic) effect. Alternatively, the authors could use a cardiac-specific knockout to confirm a cardiac specific effect.

Second, the authors use GW1100, a small molecule antagonist of GPR40 without a clear understanding of the cardiac side effects of inhibiting it. Other, off-target effects of GW1100 (e.g., pancreatic G-protein coupled receptor 120) are known and without characterizing its pharmacokinetics at the doses used, it is possible that it perturbed pancreatic calcium homeostasis and insulin release.⁶  Experiments with other GPR40 antagonists (e.g., DC260126⁷ ) could confirm a GPR40-based effect, but as presented the results are not specific for GPR40.

The paper lacks positive controls to test whether the combination of GW1100 and ischemia (or bupivacaine) is an unrecoverable insult. The authors used a predosing control of GW1100, which showed physiologic effects of GW1100. Other physiologically relevant drugs (including insulin and the protein kinase B inhibitor Wortmannin) exert combinatorial toxicity.^2 ,4 Given that such an interaction between GW1100 and ischemia-reperfusion injury and/or bupivacaine toxicity could explain their results, the lack of positive controls (e.g., a titratable and recoverable insult) is puzzling.

The authors’ data are impressive, but the results are not specific for cardiac effects mediated by GPR40 or lipid emulsion. For example, replacing GW1100 with concentrated potassium chloride would produce a near-identical set of data. Evidence from several laboratories indicates that multiple properties (e.g., volume expansion, accelerated redistribution, positive inotropy, and attenuation of ischemia-reperfusion injury) drive the benefit lipid resuscitation therapy.⁸  The authors do not address how their results would comport with these other theories or use controls to obviate the other effects (e.g., How would GW1100 alter volume or redistribution effects?). Given these concerns, the simplest explanation of the findings by Umar et al. is that the combination of GW1100 and ischemia or bupivacaine causes cardiotoxicity that has no bearing on the function of lipid emulsion.

Dr. Weinberg is an officer, shareholder, and paid consultant of ResQ Pharma, Inc., Chicago, Illinois.