Evidence that myocardial Na/I symporter gene imaging does not perturb cardiac function

Abstract

Adenoviral Na/l symporter (NIS) gene transfer has emerged as a promising method for myocardial gene imaging but concern over possible perturbation of cardiac function persists. In this study, we addressed this issue with cultured cardiac cells and serial echocardiography, creatine kinase (CK) measurements, and histologic examination of rats intramyocardially injected with an adenovirus that expresses both NIS and enhanced green fluorescent protein (EGFP) (Ad.EGFP.NIS) or a control virus (Ad.EGFP). Methods: H9C2 cardiac myoblasts differentiated into cardiomyocytes were evaluated for the effect of Ad.EGFP.NIS and Ad.EGFP infection on viable cell number and 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays. Rats injected intramyocardially with 3 x 10(8) plaque-forming units of Ad.EGFP.NIS (n = 9) or Ad.EGFP (n = 8) underwent serial echocardiographic measurements of heart rate, left ventricular (LV) dimensions, ejection fraction (EF), and fractional shortening (FS) on the day before and on days 4 and 9 after gene transfer. Five Ad.EGFP.NIS rats also underwent repeated I-123 imaging from which ratios of cardiac to mediastinal counts (C/ M ratios) were obtained. Separate rats underwent serial measurements of serum CK, myocardial myeloperoxidase assays, and microscopic assessment of inflammation. Results: Cultured cardiac cells showed no change in cell viability or proliferation at 4 and 9 d after Ad.EGFP.NIS or Ad.EGFP infection compared with controls. I-123 scintigraphy demonstrated high cardiac radio-uptake at Ad.EGFP.NIS injection sites by days 2 and 4 (C/M ratios, 5.0 +/- 0.6 and 5.1 +/- 1.0, respectively), followed by a complete loss of uptake by day 9 (C/M ratio, 1.4 +/- 0.0). Serial echocardiography revealed no difference in heart rate, LV dimensions, or functional parameters between Ad.EGFP.NIS and Ad.EGFP groups at any given time. Mild reductions in LVEF and LVFS by day 9 compared with baseline were similar for both Ad.EGFP (88.2% +/- 6.4% vs. 79.6% +/- 5.0% for LVEF and 0.55 +/- 0.10 vs. 0.44 +/- 0.05 for LVFS) and Ad.EGFP.NIS groups (88.0% +/- 5.4% vs. 78.7% +/- 4.6% for LVEF and 0.54 +/- 0.09 vs. 0.42 +/- 0.05 for LVFS). Serial serum CK and myocardial myeloperoxiclase activities were not elevated in either group, in contrast to substantial increases found after ischemia-reperfusion injury. Histology revealed similar mild inflammatory cell infiltration restricted to the injection site for both groups. Conclusion: The results of this study demonstrate that myocardial NIS gene imaging does not cause significant myocardial injury or perturbed cardiac function, other than mild effects likely due to adenoviral vector-associated host response. Thus, this practical and convenient reporter gene strategy can be used safely for noninvasive myocardial gene imaging in living subjects.