Based on his studies of echocardiography and molecular and cellular biology, Kitabatake shared his views on the recent progress in research of heart failure and the prospect for its treatment.

Traditionally heart failure has been defined as a clinical syndrome resulting from the inability of the heart to pump sufficient blood to supply the metabolic needs of the body. Left ventricular (LV) systolic dysfunction has been considered to play a major role in heart failure. However, epidemiological studies suggest that 20-40% of patients with heart failure have preserved systolic function. Diastolic function was of little concern twenty years ago because of its difficult evaluation. Work by Kitabatake in the 1970s showed that physiological properties of the papillary muscle in situ might be affected by coronary perfusion pressure. That is, the systemic environment affects the mechanical properties of the left ventricle. Further work in the animal model showed that the ejection timing, rather than peak LV pressure, was the primary determinant of the LV relaxation rate.

Work by Kitabatake and colleagues showed that measurement of the transmitral flow pattern (E/A ratio) is feasible but pseudo-normalized, and that the flow propagation velocity (FPV) is much more useful to non-invasively evaluate LV diastolic function in humans and is not pseudo-normalized, even in patients with severe heart failure. FPV closely correlates with Tau, exercise capacity and prognosis in patients with impaired systolic function.

A new ultrasound system combining pulse Doppler technique and two-dimensional echocardiography to non-invasively evaluate LV diastolic properties in humans was developed by Kitabatake and colleagues. The direction of the pulse Doppler flow meter with an electronic beam sector scanning echocardiograph enabled the precise location of the sample volume of the Doppler beam on 2-D echocardiogram. Regional blood flow evaluation is possible anywhere in the cardiovascular system with this system, as is non-invasive assessment of pulmonary arterial pressure. Accurate measurement of transmitral blood flow velocity, shown to reflect the diastolic behavior of the left ventricle in health and disease, is also possible.

Throughout the diastolic period, the area of the mitral valve orifice remains nearly unchanged, so the transmitral velocity flow pattern reflects LV volume changes essentially equivalent to the dv/dt curve in diastole. Thus, Doppler measurement may allow the estimation of sequential LV volume changes without invasive scintigraphy.

By the mid-1980s, a consensus was achieved that LV diastolic function might decrease E wave amplitude because of incompetent early relaxation of the ventricle and increased A wave amplitude because of compensatory atrial contraction, resulting in a decrease in the E/A ratio. Pseudo-normalization of the transmitral flow pattern in patients with dilated cardiomyopathy (DCM), in which in the absence of heart failure the E wave amplitude and the E/A ratio is decreased to less than 1.0, while in the presence of severe heart failure the E wave amplitude and E/A ratio increased, was identified by Kitabatake and colleagues. The peak systolic to diastolic flow velocity ratio (S/D) in primary venous flow allows differentiation of pseudo-normalized from normal transmitral flow pattern. However, this method was somewhat complicated.

Color M-mode Doppler echocardiography was then studied to improve clinical evaluation of LV diastolic function. The flow propagation velocity (FPV) as a new diastolic index was defined as the ratio between the maximal velocity around the mitral orifice (Lmax) in early diastole and the decrease to 70% of the maximal velocity (L₇₀). FPV is not pseudo-normalized, making it a useful index for evaluating LV relaxation, even in patients with severe heart failure. Hence, evaluation of LV diastolic function, regardless of heart failure severity, was made possible.

Diastolic function is a major determinant of exercise capacity, as shown by the significant correlation between peak VO₂ and FPV on color M-mode Doppler echocardiography. Peak VO₂ as an index of exercise capacity did not correlate with ejection fraction as determined by 2-D echocardiogram. Cumulative event-free survival was significantly better in the mild diastolic dysfunction than in the severe dysfunction group. This finding strongly indicated that the degree of diastolic dysfunction of the ventricle is relevant to the prognosis in patients with impaired systolic function.

Neurohormonal activation worsens the natural history of myocardial dysfunction and remodeling. In essence, multiple neurohormonal signaling pathways such as alpha-1 and angiotensin II type-1 receptors are activated in the failing heart and promote maladaptive remodeling and myocardial dysfunction. Improved understanding of the molecular biology and mechanisms of the signal transduction system would improve the understanding of cardiac performance.

The BIO TO2 strain of the cardiomyopathic hamster is an animal model of DCM and heart failure. At 20 weeks in this model, LV diastolic dimension was significantly higher and the percent fractional shortening (%FS) was significantly lower. At 26 weeks, the ventricular wall was thinner and the left ventricle more dilated than in the control animal. Interstitial fibrosis could also be seen in the BIO TO2 animals, with increased interstitial space. The decreased nuclear density in the BIO TO2 animal was closely correlated with %FS, demonstrating that cardiac dysfunction is accompanied by a loss of cardiomyocytes. Total collagen content increased with age in the BIO TO2 animals while it remained the same in the control animals, and the difference was significant at 10-20 weeks of age. As total collagen content increased, the %FS decreased more. These findings indicate that an increase in collagen content might be due to LV dysfunction.

Expression of mRNA of angiotensinogen, renin, and ACE was about 2-fold higher in the left ventricle of spontaneously hypertensive rats (SHR) than in Wistar Kyoto rats (WKR), suggesting that the tissue angiotensin II formation pathway was enhanced in the hypertrophied remodeled heart, as in the failing heart of the BIO TO2 animals.

In LV hypertrophy (LVH), extracellular matrix protein, collagen and cardiomyocytes are major components of the LV wall. The effects of angiotensin II on collagen synthesis in cardiofibroblasts isolated from SHR and WKR was studied. A 2-fold increase in collagen synthesis resulted from 2-hours of angiotensin II stimulation of cardiofibroblasts, which was completely inhibited by the angiotensin II type-1 antagonist MK954, but not by the type-2 antagonist PD123177. Thus, in the remodeled myocardium of SHR, tissue expression of the components of the renin angiotensin system (RAS) is increased. This cardiac RAS activation may cause myocardial fibrosis by stimulating collagen synthesis.

Phosphatidyl inositol (PI) metabolism during the cellular response to norepinephrine in the pressure overloaded hypertrophic rat heart was examined in SHR. The calcium influx stimulated by norepinephrine was higher in SHR and the release of DAG from cardiomyocytes was markedly increased. The accumulation of IP3 significantly increased, indicating that PI-specific PLC activity increased in SHR. PLC activity was also enhanced in SHR. Thus, in the remodeled myocardium, the cytosolic calcium concentration may increase in response to stimuli such as angiotensin II or endothelin, whose receptors are coupled to Gq proteins, like norepinephrine.

Enhanced cardiac RAS may increase IP3 and DAG production in cardiomyocytes in BIO TO2 animals. In fact, IP3 release was markedly increased by angiotensin II and the intracellular calcium concentration was higher in BIO animals from 5 to 20 weeks of age (Figure 1). These findings suggest that angiotensin formation via ACE was increased and the angiotensin II signaling and the PI turnover might produce a higher intracellular calcium level in the BIO TO2 animals. These changes in PI turnover are essentially similar in hypertrophied or failing myocardium.

PI signaling was altered in isolated cardiomyocytes from SHR and BIO TO2 animals. Activated PIP2-PLC resulted in increases in secondary messengers such as IP3 and DAG, which induced translocation of PKC from cytosol to the plasma membrane leading to activation of the MAP-kinase cascade. The PI turnover pathway may play an important role in the remodeling process. Figure 2 illustrates the PI signaling system in remodeled myocardium. Receptor-mediated activation of myocardial Gq signaling is a biochemical mechanism postulated to be responsible for inducing pressure overload hypertrophy.

Serial echocardiographic findings in Gq-transgenic mice and non-transgenic mice show that the LV is more dilated and wall motion is decreased after aortic banding in the mice overexpressing myocardial Gq. The LV mass of both species increased after aortic banding. The %FS progressively declined after aortic banding in the Gq-transgenic mice, while it was preserved in the non-transgenic mice. Gq-stimulated cardiac hypertrophy is functionally deleterious and compromises the ability of the heart to adapt to an increased mechanical load. The hypertrophic response to receptor-mediated signaling might differ from that in Gq-stimulated signaling, implying that physiological stimuli are regulated by integrated crosstalk with other signal transduction systems.

Heart failure is associated with a diminished contractile response to catecholamines. In contrast to the redundancy associated with the Gq-coupled receptor signaling, the number of beta-adrenergic receptors and the agonist-stimulated adenyl cyclase (AC) activity is reduced in the failing myocardium.

Forskolin-stimulated AC activity was markedly decreased at the age of 16 and 28 weeks. G-protein coupled receptor kinases (GRKS) are known for their agonist-dependent phosphorylation of beta-adrenergic receptors. Semi-quantitative RTPCR revealed that expression of GRK2 and GRK5 was significantly higher in the BIO TO2 than in the F1b animals. It has been demonstrated that gene targeting of GRKs directly affects the cardiac function. Taken together with the findings by Kitabatake, it is likely that GRKs might be one of the responsible factors for reducing cardiac contractility.

A new molecule, rap1GAPII, was found to be a Gi-associated isoform of rap1GAP. The association of Gi and rap1GAPII resulted in the activation of ERK MAP-kinase by inactivating rap1. An agonist dependent decrease in GTPrap1 occurred with co- transfection of M2 muscarinic-receptors. These results suggested that M2-dependent Gi activation may cause GTPrap1 hydrolysis attributed to the activation of rap1GAPII. Mice overexpressing rap1GAPII in the heart are now being developed by Kitabatake and colleagues.