Radiolabeled lipophilic cationic compounds, such as 18 F-labeled phosphonium salt, accumulate in the mitochondria through a negative inner transmembrane potential. The purpose of this study was to develop and evaluate ( 18 F-fluoropentyl)triphenylphosphonium salt ( 18 F-FPTP) as a myocardial PET agent. Methods: A reference compound of 18 F-FPTP was synthesized via 3-step nucleophilic substitution reactions and was radiolabeled via 2-step nucleophilic substitution reactions of no-carrier-added 18 F-fluoride. Accumulations of 18 F-FPTP, 3 H-tetraphenylphosphonium, and 99m Tc-sestamibi were compared in a cultured embryonic cardiomyoblast cell line (H9c2). The biodistribution of 18 F-FPTP was assessed using BALB/c mice. The 18 F-FPTP small-animal PET study was performed in Sprague-Dawley rats with or without left coronary artery (LCA) ligation. Results: 18 F-FPTP was synthesized with a radiochemical yield of 15%-20% and radiochemical purity of greater than 98%. Specific activity was greater than 6.3 TBq/mmol. Cell uptake of 18 F-FPTP was more than 15-fold higher in H9c2 than in normal fibroblasts (human normal foreskin fibroblasts). Selective collapse of mitochondrial membrane potential substantially decreased cellular uptake for 18 F-FPTP and 3 H-tetraphenylphosphonium, compared with that for 99m Tc-sestamibi. The biodistribution data in mice (n 5 24) showed rapid blood clearance and high accumulation in the heart. Heart-to-blood ratios at 10 and 30 min were 54 and 133, respectively. Heart-to-lung and heart-to-liver ratios at 10, 30, and 60 min were 4, 4, and 7 and 4, 5, and 7, respectively. Dynamic small-animal PET for 60 min after injection of 18 F-FPTP showed an initial spike of radioactivity, followed by retention in the myocardium and rapid clearance from the background. 18 F-FPTP small-animal PET images in LCA-occluded rats demonstrated sharply defined myocardial defects in the corresponding area of the myocardium. The myocardial defect size measured by 18 F-FPTP small-animal PET correlated closely with the hypoperfused area measured by quantitative 2,3,5-triphenyltetrazolium chloride staining (r 2 5 0.92, P , 0.001). Conclusion: The excellent pharmacokinetics of 18 F-FPTP and its correlation with 2,3,5-triphenyltetrazolium chloride staining in normal and LCAoccluded rats suggest that this molecular probe may have a high potential as a mitochondrial voltage sensor for PET. This probe may also allow high throughput, with multiple daily studies and a wide distribution of PET myocardial imaging in the clinic.

Optimization of the specific affinity of cardiac delivery vector could significantly improve the efficiency of gene/protein delivery, yet no cardiac vectors to date have sufficient target specificity for myocardial infarction (MI). In this study, we explored bacterial tropism for infarcted myocardium based on our previous observations that certain bacteria are capable of targeting the hypoxic regions in solid tumors. Out of several Escherichia coli or Salmonella typhimurium strains, the S. typhimurium defective in the synthesis of ppGpp (ΔppGpp S. typhimurium) revealed accumulation and selective proliferation in the infarcted myocardium without spillover to noncardiac tissue. The Salmonellae that were engineered to express a variant of Renilla luciferase gene (RLuc8), under the control of the E. coli arabinose operon promoter (P(BAD)), selectively targeted and delivered RLuc8 in the infarcted myocardium only upon injection of L-arabinose. An examination of the infarct size before and after infection, and estimations of C-reactive protein (CRP) and procalcitonin indicated that intravenous injection of ΔppGpp S. typhimurium did not induce serious local or systemic immune reactions. This current proof-of-principle study demonstrates for the first time the capacity of Salmonellae to target infarcted myocardium and to serve as a vehicle for the selective delivery of therapeutic agents in MI.