INTRODUCTION

Ischemic heart failure is a leading cause of morbidity and mortality.¹ Multiple imaging techniques have been evaluated for detection of ischemic myocardium (IM) in various models including the stenosed coronary artery model.² Conventional methods such as PET, dobutamine stress echo or late gadolinium enhancement do not directly detect the functional state of the key organelle in oxidative metabolism, the mitochondrion. Hyperpolarized (HP) ¹³C-metabolic imaging has been used to assess mitochondrial function in preclinical models³ and is a potential modality for evaluating ischemic myocardium noninvasively. The major goal of this study is to evaluate myocardial response to dobutamine stimulation in ischemic rat hearts using HP-[1-¹³C]pyruvate.

METHODS

Study groups and animals: Animal studies were approved by IACUC. Metabolism of HP [1-¹³C]pyruvate was investigated in male SD rats in four groups: 1. normal myocardium (NM), 2. ischemic myocardium (IM), 3. NM+dobutamine, 4. IM+dobutamine. To induce myocardial ischemia, rats were anesthetized by intraperitoneal injection of ketamine/xylazine (50/2.5 mg/kg). A left parasternal thoracotomy was performed in the 4^th–5^th intercostal space following incubation and ventilation with 1% isoflurane mixed with oxygen. The left atrium was elevated to expose the left coronary artery and snare was positioned around the vessel in contact with a 300 μm diameter probe 1–2 mm from its origin. The vessel was allowed to expand upon probe removal, resulting in a partial occlusion of the coronary artery, a model described previously.⁴ For dobutamine stress HP-MRS, an intra-peritoneal injection of 1.5 mg/kg of dobutamine was administered before positioning the animal in the MRI scanner. ¹³C MRS and ¹³C chemical shift imaging (¹³C-CSI) of these animals were carried out 7 days after the surgery.

Polarization and injection of [1-¹³C]pyruvate: Neat [1-¹³C]pyruvic acid was mixed with 15 mM trityl radical OX63, and Gadoteridol ([Gd³⁺]=2 mM) was polarized at ~1.4 K, 3.35 T for 1.5 h in a HyperSense® polarizer. The polarized substrate was quickly dissolved in 4 mL hot PBS buffer (pH 7.4) and neutralized with NaOH. Hyperpolarized pyruvate solution (80 mM) was injected via a cannulated tail vein catheter over 12-15 seconds.

In-vivo ¹³C-MRS and ¹³C-CSI: ¹³C data were acquired in a 4.7 T MRI scanner (Agilent, Palo Alto, CA) using a Varian 60 mm dual tuned ¹H/¹³C linear volume coil. Anatomical ¹H images (GEMS; matrix size: 128x128; FOV: 60x60 mm; slice thickness: 2 mm) were acquired for slice planning and positioning. ¹³C intensity maps were generated and displayed as overlays on grayscale ¹H anatomical images. In separate experiments, in vivo HP-¹³C-MRS data acquisition (flip angle: 10 deg, TR: 2s, sw: 15000 Hz) was initiated immediately after the pyruvate injection was completed. HP-¹³C-MRS data were analyzed by measuring the area-under-the-curve (AUC) for the lactate, alanine and bicarbonate signals normalized to the total ¹³C signal from lactate, alanine and bicarbonate. Plots of bicarb/lactate, bicarb/alanine and lactate/alanine were generated from these data. HP-¹³C-CSI (8x8, TE:1.4 ms, TR:36.7 ms, sw:7485 Hz, data points: 512, thickness:12mm) data was also acquired to evaluate the spatial distribution of the metabolites in the myocardium. Statistical significance was evaluated by unpaired Student’s t-test using GraphPad Prism (v.8).

RESULTS AND DISCUSSION

Representative sequential ¹³C-MR spectra (Fig. 1A) acquired from an 18-mm axial slice of the heart show the evolution of ¹³C-metabolites after injection of HP-[1-¹³C]pyruvate. The corresponding summed ¹³C spectra (Fig. 1B) show well-resolved signals from [1-¹³C]lactate, [1-¹³C]alanine, and ¹³C-bicarbonate. In hearts with partial coronary ligation, the spectroscopy data were acquired from a mixture of ischemic and nonischemic myocardium. A 27% decrease in bicarbonate production was observed in hearts with ischemic myocardium compared to control hearts (Fig. 2A). After stimulation of NM with dobutamine, less bicarbonate was produced from HP-pyruvate likely due to an increased contribution of other endogenous substrates (circulating fatty acids/glucose/ketones) to acetyl-CoA. Production of bicarbonate was unchanged in IM hearts following adrenergic stimulation. Production of HP-lactate (Fig. 2B) from HP-pyruvate trended toward higher levels in both the IM and after dobutamine stimulation but these differences did not reach statistical significance. Alanine production (Fig. 2C) was comparable in both groups, with or without dobutamine, with high variability in signal intensity (Fig. 1). The average bicarbonate to lactate ratio (Fig. 2D) in IM hearts is 37% lower than the ratio in NM hearts. A 44% decrease in the bicarbonate-to-lactate ratio was observed as a result of dobutamine stimulation of NM hearts. Dobutamine did not change this ratio in IM hearts. A similar trend was observed when comparing bicarbonate-to-alanine ratios across all four groups of hearts (Fig. 2E). The lactate-to-alanine ratios (Fig. 2F) tended to be higher in both NM and IM hearts after dobutamine stimulation but, again, these differences did not reach statistical significance. Figure 3 show representative intensity maps of ¹³C-bicarbonate, ¹³C-lactate, as well as the injected substrate ¹³C-pyruvate overlayed onto anatomical ¹H images. These images confirm that pyruvate metabolism as detected by the in vivo ¹³C-spectroscopy largely detects the myocardium. The trend of the metabolic intensity maps agrees well with the quantitative analyses of ¹³C-pyruvate metabolism shown in Fig. 2.

CONCLUSION

Taken together, our results demonstrate that decreased oxidative metabolism of pyruvate is observed as a result of myocardial ischemia and dobutamine stimulation. No change in bicarbonate appearance was observed ischemic hearts even after adrenergic stimulation.

Acknowledgements

Funding support: AHA 18POST34050049, NIH 5R37-HL034557, and NIH P41-EB015908.