Abstract

Myocardial infarction (MI) remains a major killer despite highly optimised systems for the delivery of primary percutaneous coronary intervention (PPCI), highlighting a need for novel therapeutics that could be administered in the days following the event. The healing myocardium undergoes a macrophage driven inflammatory response which is a compelling therapeutic target, though clinical exploration of this process has been limited because conventional imaging techniques cannot assess cellular inflammation in the heart.

We hypothesised that the huge increases in signal-to-noise ratio provided by hyperpolarized MRI could provide a solution to this problem. In rodent models, hyperpolarized [1–13C]pyruvate MRI using a custom designed metabolite mapping sequence in vivo demonstrated that experimental MI caused intense [1–13C]lactate signal in healing myocardial segments at both day 3 (paralleling the maximal ‘inflammatory’ phase of the macrophage response) and also at day 7 (‘reparative’ phase), compared to sham operated controls. Monocyte/macrophage depletion using clodronate liposomes normalised the [1–13C]lactate signal at both timepoints.

Gene expression analysis of monocytes/macrophages sorted from infarct tissue demonstrated regulation of key enzymes of glycolysis, suggesting that monocyte/macrophage recruitment and metabolic reprogramming of those cells underlies the high lactate signal detected. Hyperpolarized [1–13C]pyruvate MR spectroscopy in macrophage-like cell suspensions confirmed that cellular activation and polarisation almost doubles hyperpolarized lactate label flux rates in vitro; blockade of glycolysis with 2-deoxyglucose (2-DG) in activated macrophage-like cells normalised lactate label flux rates and also markedly inhibited production of key pro-inflammatory cytokines at both mRNA and protein level, without major cytotoxicity.

Systemic administration of 2-deoxyglucose following rodent MI normalised hyperpolarized [1–13C]lactate signal in healing myocardial segments and also caused dose dependent improvement in IL-1β expression in infarct tissue, providing proof-of-concept of ‘MR visible’ immunomodulation. Furthermore, cine MRI demonstrated improvements in myocardial remodelling and systolic function in 2-DG treated rats at 3 months. Finally, we present initial human experience of cardiac hyperpolarized [1–13C]pyruvate MR, demonstrating unprecedented improvements in signal-to-noise ratio and highlighting the potential for rapid clinical translation of these findings.

We conclude that hyperpolarized MRI provides a novel biomarker of cardiac inflammation and repair post-MI by detecting the induction of an immuno-metabolic pathway in cardiac macrophages which controls key inflammatory cytokine production and influences myocardial remodelling. In addition to a role in the development of novel therapeutics to improve remodelling post MI, hyperpolarized MRI may have broad applications in other inflammatory cardiovascular diseases.