The ß-adrenergic receptor system plays a key role in cardiac function in normal and diseased states. In end stage human heart failure, the ß-1 AR mRNA and protein are selectively downregulated, while the ß-2 AR are not changed for the most part (Fig. 1). Both the ß-1 and ß-2 receptors are uncoupled, mediated by the beta receptor kinase ßARK, a member of the family of G protein coupled receptor kinases. ßARK mRNA protein inactivity is increased about 3-fold in human heart failure, which can lead to the enhanced phosphorylation of ß receptors and the functional uncoupling seen in cardiac disease.

ß-2 AR activation causes coupling to the heterochimeric G protein. Gas can activate cyclase. The beta-gamma dimer effector molecules can lead to activation of downstream target events 1) ion channel activation, 2) activation of certain forms of adenyl cyclase, 3) importantly, specifically binding residues on the C terminus of ßARK (ß ARKct), bringing ßARK into more contact with the membrane where it can phosphorylate agonist-occupied receptors. So, peptides from this region of ßARK have been shown to be in vivo and in vitro ßARK inhibitors.

Mouse models developed in Koch's laboratory showed that inhibiting the activity or lowering the expression of ßARK allows direct regulation of cardiac contractility. Two transgenic mice models of hypercontractile function were generated using the alpha myosin heavy chain promoter: selective cardiac overexpression of the human ß-2 AR and inhibition of ßARK using the inhibitor ßARKct. Overexpression of ß-2 AR in a dose-dependent manner led to enhanced cardiac function and to DCM. At more modest levels of overexpression no pathology was seen.

Their work is the first in vivo demonstration that ßARK could be present and desensitize beta receptors. The ßARKct peptide inhibitor had enhanced basal dP/dtmax and supersensitivity to isoproterenol. A reciprocal regulation of cardiac contractility was found when they overexpressed the ßARK-1 enzyme or the ßARKct peptide inhibitor. ß ARK overexpressing mice had attenuated function. Heterozygote ßARK knockout animals with 50% less ßARK had increased basal dP/dtmax and supersensitivity to isoproterenol, much like the ßARKct animals.

To determine whether inhibiting ßARK or overexpressing receptors had any therapeutic value, the adenoviral approach in larger animals was used in Koch's laboratory. No mouse models of heart failure were available at that time. The development of DCM was prevented by the expression of the ßARKct peptide when the MLP knockout mouse (with DCM) and the ß ARKct animals were bred. The catalytic domain of the ßARK-1 protein is very similar to protein kinase A (PKA) and C (PKC). The N terminal domain has some interesting functions that Koch's laboratory is now delineating. The beta-gamma binding domain is in the carboxy terminus, the region used for the ßARKct.

Prior work has shown that the wild type mice have normal systolic and diastolic function on echo, whereas the MLP knockout mice have extreme dilatation, thin walls, and very limited fractional shortening. In the MLP-ßARKct hybrid animal, systolic and diastolic function is normalized, with no dilatation.

Interestingly, work by other investigators has shown positive effects with ß-2 AR overexpression and variable effects with ßARKct. In the Gaq transgenic mouse in Dorn's laboratory, modest ß-2 adrenergic overexpression led to some functional and hypertrophic rescue. In Leiden's model of idiopathic DCM dominant negative Kreb mice, the ßARKct had some positive inotropic effects that did not seem to rescue the mortality in these animals; a variable effect. Leinwand's mutant myosin heavy chain animal with a cardiomyopathic phenotype was rescued by ßARKct.

Targeted CSQ overexpression led to altered SR function and calcium handling in the CSQ overexpressing model generated by Jones. These mice have early hypertrophy followed by progressive DCM. The CSQ transgenic mice at 7 weeks have concentric hypertrophy, with normalized function that rapidly deteriorates to a state of dilatation and low systolic function at 14 weeks. All the mice are dead by 16 weeks; severe early mortality in the CSQ transgenic mice in contrast to MLP knockout mice. At 7 and 14 weeks of age there was no beta receptor responsiveness in relation to normalized function as seen by echo. Biochemically, decreases in beta receptor density and a 2-fold increase in ßARK protein and activity was seen. These beta receptor abnormalities were present at 7 weeks, before the onset of dilatation.

The CSQ transgenic have large left ventricles as shown by the LV end diastolic dimension and in the CSQ-ßARKct animals there was substantial rescue. The fractional shortening that was significantly impaired in the CSQ mice, was significantly elevated in the hybrid animals from about 15% to about 40% in experiments in Koch's laboratory.

Importantly, a survival study showed that ßARKct chronically expressed in the heart failure model lead to positive effects. The CSQ animals die very early at 15-16 weeks of age. In the ßARKct animals a statistically significant increase in survival to more than 20 weeks was found. However, once the animals begin to die the curves are nearly parallel, showing that despite the substantial survival effect, there is a threshold the ß ARKct can not overcome.

Koch's laboratory attempted to deliver the ß-2 receptor and ßARKct to the hearts of rabbits by intracoronary delivery. The objective was to inhibit ßARK-1 mediated desensitization by expressing the ßARKct to enhance cardiac contractility and restore normal beta receptor responsiveness, a limitation of the failing heart.

In a study using a percutaneous approach, a subselective catheterization was done of a rabbit coronary artery and 5 x 10¹¹ total viral particles of an adenovirus containing either the ß-2 receptor, ßARKct or beta-Gal injected. Ventricular-targeted expression of the transgene is possible as shown by the beta gal; going down the right coronary artery only transduces the right ventricle, the left circumflex is distributed through the left ventricular free wall.

Left ventricular overexpression, with no overexpression in the right ventricle, was obtained injecting the ß-2 receptor through the left circumflex artery (Fig. 2). Higher levels of expression are obtained in the right ventricle as it is easier to transduce, probably due to it being thinner with less pressure. ß receptor overexpression in the range of 6- to 10-fold led to enhanced contractility of the rabbit heart at 7-21 days after gene delivery. ßARKct was also deliverable using this method.

Heart failure study

Gene delivery of adenoviral ßARKct in a rabbit model of heart failure resulted in a significant increase in regional fractional shortening in the ßARKct animals, and 2-D echo showed a slight but significant increase in fractional shortening. Each animal served as its own control, and serial echos showed an increase in fractional shortening. The failing control animals that received beta-Gal worsened or had no change.

In this study, heart failure was induced by performing a left circumflex ligation in a marginal branch to produce a 30-40% infarct in the LV free wall. Hemodynamic failure developed within 3 to 4 weeks, and the ischemic model was allowed to progress to heart failure. Percutaneous left circumflex-mediated delivery of the adenoviral ßARKct as the ßARKct inhibitor was performed. At one and two weeks after gene delivery, in vivo assessment was performed regionally using sonomicrometry crystals placed on the plane of the LV and by 2-D and milar micromanameter catheterization.

Expression of ßARK in human heart failure is associated with severity. Transgenic mice with cardiac-targeted ßARKct expression have enhanced cardiac performance without pathology, and do not have a higher susceptibility to ischemic injury. The heterozygote ßARK knockout mice have increased cardiac function.

Heart failure has been prevented or attenuated by cardiac expression of the ßARKct in several mouse models of cardiomyopathy, and it increased survival in the CSQ model. In the rabbit model, adenoviral-mediated cardiac delivery of ßARKct increased systolic and diastolic function in normal hearts. Heart failure was prevented after myocardial infarction by ßARKct delivery, as shown in a study to be published soon. Cardiac function following heart failure development can be improved.

The ßARKct acts as an inhibitor of beta gamma activation of ßARK. It is known from previous studies that ß ARKct inhibits ßARK in the heart, beta receptor desensitization is attenuated, and beta receptor signaling is enhanced. However, since ßARK probably phosphorylates hundreds of receptors, including several in the heart, there could be other signaling systems responsible for the phenotype

Novel paradigm of receptor signaling

Koch's laboratory is now testing the interesting speculation that ßARK translocation is the mechanism triggering beta gamma-dependent MAP kinase activation. If ßARK translocation is the contributing mechanism, it is due to ßARK inhibition. Other labs have shown that this beta-gamma G protein coupled receptor activation of the RAS-MAP kinase pathway, progresses from the internalization of receptors by beta-Arrestin, which binds GRK phosphorylated receptors. This can not occur unless the receptor is phosphorylated by GRK or ß ARK. The ßARKct is likely inhibiting desensitization of the beta receptor in the heart. Classical signaling through cAMP through the calcium channels is enhanced, which could support the decreased contractility of the heart. Since it is decreasing desensitization, fewer internalized receptors to activate MAP kinase are available. In this novel paradigm of receptor signaling, the desensitized receptor, classically shown to be turned off, is actually the signaling molecule.