β₁- and β₂-adrenergic receptors (βARs) are known to differentially regulate cardiomyocyte contraction and growth. We tested the hypothesis that these differences are attributable to spatial compartmentation of the second messenger cAMP. Using a fluorescent resonance energy transfer (FRET)-based approach, we directly monitored the spatial and temporal distribution of cAMP in adult cardiomyocytes. We developed a new cAMP-FRET sensor (termed HCN2-camps) based on a single cAMP binding domain of the hyperpolarization activated cyclic nucleotide-gated potassium channel 2 (HCN2). Its cytosolic distribution, high dynamic range, and sensitivity make HCN2-camps particularly well suited to monitor subcellular localization of cardiomyocyte cAMP. We generated HCN2-camps transgenic mice and performed single-cell FRET imaging on freshly isolated cardiomyocytes. Whole-cell superfusion with isoproterenol showed a moderate elevation of cAMP. Application of various phosphodiesterase (PDE) inhibitors revealed stringent control of cAMP through PDE4>PDE2>PDE3. The β₁AR-mediated cAMP signals were entirely dependent on PDE4 activity, whereas β₂AR-mediated cAMP was under control of multiple PDE isoforms. β₁AR subtype–specific stimulation yielded ≈2-fold greater cAMP responses compared with selective β₂-subtype stimulation, even on treatment with the nonselective PDE inhibitor 3-isobutyl-1-methylxanthine (IBMX) (ΔFRET, 17.3±1.3% [β₁AR] versus 8.8±0.4% [β₂AR]). Treatment with pertussis toxin to inactivate G_i did not affect cAMP production. Localized β₁AR stimulation generated a cAMP gradient propagating throughout the cell, whereas local β₂AR stimulation did not elicit marked cAMP diffusion. Our data reveal that in adult cardiac myocytes, β₁ARs induce far-reaching cAMP signals, whereas β₂AR-induced cAMP remains locally confined.