Dobutamine stress cine‐MRI of cardiac function in the hearts of adult cardiomyocyte‐specific VEGF knockout mice

Abstract

A mouse model of non-necrotic vascular deficiency in the adult heart was studied using cine-magnetic resonance imaging (MRI) and other techniques. The mice lacked cardiomyocyte-derived vascular endothelial growth factor (VEGF) following a targeted knockout in the ventricular cardiomyocytes. Quantitative endothelial labeling showed that the capillary density was significantly reduced in the hearts of knockout mice. Gene expression patterns suggested that they were hypoxic. Semiautomated MR image analysis was employed to obtain both global and regional measurements of left ventricular function at 10 or more time points through the cardiac cycle. MRI measurements showed a marked reduction in ejection fraction both at rest and under low-and high-dose dobutamine stress. Regional wall thickness, thickening, and displacement were all attenuated in the knockout mice. A prolonged high-dose dobutamine challenge was monitored by MRI. A maximal response was sustained for 90 minutes, suggesting that it did not depend on endogenous glycogen stores. J. Index terms: VEGF; cine; dobutamine; mouse; heart THE RECENT PROLIFERATION of murine models of human heart disease has been remarkable (1) and is likely to continue. Although MRI has been elegantly applied by several groups to studies of the mouse heart (e.g., Refs. 2-5), it remains concentrated in relatively few centers. Cardiac MRI of the mouse, unlike of the rat (6), is not widespread in the pharmaceutical research setting, but that seems likely to change as the advantages of murine models increasingly outweigh their disadvantages. There are important biological differences between the hearts of different mammals, such as in calcium handling (7) and vascular anatomy (8), yet many of the perceived disadvantages of using mice as opposed to larger animal species are essentially technical in nature. This has spurred great efforts in adapting radiological, surgical, and physiological techniques to the temporal and spatial scale of the mouse heart (9,10).This paper describes MRI of the left ventricle (LV) under resting and stressed conditions in a mouse model of a non-necrotic vascular deficiency in the adult heart. This model may be of some value in designing and evaluating angiogenic therapies (11), and in understanding the pathophysiology of chronic ischemic conditions associated with coronary artery disease (12). The generation and preliminary phenotypic characterization of this mouse model has recently been described (13). Briefly, the Cre/lox system (14) was employed to generate a targeted knockout of vascular endothelial growth factor A (VEGF A) specific to the ventricular cardiomyocytes. VEGF A is a potent cytokine and mitogen involved in the formation and maintenance of blood vessels (15). Attempts to generate VEGF A knockouts by the more common germ-line strategies were confounded by the highly unusual embryonic lethality seen in VEGF A heterozygotes (16).The effect of the gene knockout was monitored directly (changes in VEGF A mRNA) and indirectly. Anticipating microvasc...

Optimization of cardiac cine in the rat on a clinical 1.5-T MR system

“…In addition to clinical use, there is intense research into imaging and assessing the heart function of small animals to better understand cardiovascular pathologies. Some investigators have demonstrated the feasibility of imaging and quantifying the heart function and dysfunction using 2-T [6,7], 4.7-T [8][9][10][11], or higher-field MR systems [12][13][14], dedicated exclusively to the small animal. However, such high-field systems are not widely available in contrast to 1.5-T MR systems used for clinical and research applications.…”

How to Perform an Accurate Assessment of Cardiac Function in Mice using High-Resolution Magnetic Resonance Imaging

“…The cine technique also allows the study of developmental changes in cardiac function and mass from neonatal to adult mice (26) or longitudinally in genetically or surgically manipulated mice (16). Applications in transgenic models have, for example, been shown in mice with cardiac hypertrophy (1), in mice with myocardial overexpression of tumor necrosis factor-α (3,27), and in adult cardiomyocyte-specific VEGF knockout mice (28). The orientation of the magnet bore (horizontal vs. vertical) also merits consideration.…”

Comparison of the Detection of Subtle Changes in Myocardial Regional Systolic Function Using Qualitative and Semi-Quantitative Techniques

“…1). A reference sector was placed at the inferior border where the right ventricle meets the left ventricle in both the ED and ES images to compensate for cardiac twist (2,9). The percent wall thickening between these contours in twelve different sectors was calculated and averaged to obtain wall-thickening values for six regions corresponding to the six regions used in the qualitative analysis.…”

“…The examiner has a distinct influence on the accuracy of contours (2) and in the appropriate hands high reproducibility and reliability of manual contour tracings can be obtained (15). Alternatively, reliable automated or semi-automated image segmentation software has shown the ability to overcome the limitations associated with the manual processing of the images (9,16). Thompson et al (17) found that two different semi-automatic cardiac software analysis packages (MASS and ARGUS) provided similar results for regional percent wall thickening, with only 6.1% variability between the two packages.…”

“…Qualitative wall motion assessment (regional endocardial surface motion) is a relatively quick and robust measurement but can be subject to bias and may be less accessible to researchers, whereas quantitative assessment of heart wall thickening (% change in wall thickness from end-diastole [ED] to end-systole [ES]) is more time-consuming (8) but less biased and most likely to yield reproducible results (9). It has been demonstrated previously that the use of quantitative analysis of cine MR images accurately determines degree and extent of regional LV dysfunction in the infarcted heart in vivo (10), where large regions of dysfunction are present.…”