The ubiquitous second messenger 3′,5′-cyclic guanosine monophosphate (cGMP) regulates multiple physiologic processes in the cardiovascular system. Its intracellular effects are mediated by stringently controlled subcellular microdomains. In this review, we will illustrate the current techniques available for real-time cGMP measurements with a specific focus on live cell imaging methods. We will also discuss currently accepted and emerging mechanisms of cGMP compartmentation in the cardiovascular system.

The ubiquitous second messenger 3′,5′-cyclic adenosine monophosphate (cAMP) regulates cardiac excitation-contraction coupling (ECC) by signaling in discrete subcellular microdomains. Phosphodiesterase subfamilies 4B and 4D are critically involved in the regulation of cAMP signaling in mammalian cardiomyocytes. Alterations of PDE4 activity in human hearts has been shown to result in arrhythmias and heart failure. Here, we sought to systematically investigate specific roles of PDE4B and PDE4D in the regulation of cAMP dynamics in three distinct subcellular microdomains, one of them located at the caveolin-rich plasma membrane which harbors the L-type calcium channels (LTCCs), as well as at two sarco/endoplasmic reticulum (SR) microdomains centered around SR Ca²⁺-ATPase (SERCA2a) and cardiac ryanodine receptor type 2 (RyR2). Transgenic mice expressing Förster Resonance Energy Transfer (FRET)-based cAMP-specific biosensors targeted to caveolin-rich plasma membrane, SERCA2a and RyR2 microdomains were crossed to PDE4B-KO and PDE4D-KO mice. Direct analysis of the specific effects of both PDE4 subfamilies on local cAMP dynamics was performed using FRET imaging. Our data demonstrate that all three microdomains are differentially regulated by these PDE4 subfamilies. Whereas both are involved in cAMP regulation at the caveolin-rich plasma membrane, there are clearly two distinct cAMP microdomains at the SR formed around RyR2 and SERCA2a, which are preferentially controlled by PDE4B and PDE4D, respectively. This correlates with local cAMP-dependent protein kinase (PKA) substrate phosphorylation and arrhythmia susceptibility. Immunoprecipitation assays confirmed that PDE4B is associated with RyR2 along with PDE4D. Stimulated Emission Depletion (STED) microscopy of immunostained cardiomyocytes suggested possible co-localization of PDE4B with both sarcolemmal and RyR2 microdomains. In conclusion, our functional approach could show that both PDE4B and PDE4D can differentially regulate cardiac cAMP microdomains associated with calcium homeostasis. PDE4B controls cAMP dynamics in both caveolin-rich plasma membrane and RyR2 vicinity. Interestingly, PDE4B is the major regulator of the RyR2 microdomain, as opposed to SERCA2a vicinity, which is predominantly under PDE4D control, suggesting a more complex regulatory pattern than previously thought, with multiple PDEs acting at the same location.

In mouse cardiomyocytes, the expression of two subfamilies of the calcium/calmodulin-regulated cyclic nucleotide phosphodiesterase 1 (PDE1)—PDE1A and PDE1C—has been reported. PDE1C was found to be the major subfamily in the human heart. It is a dual substrate PDE and can hydrolyze both 3′,5′-cyclic adenosine monophosphate (cAMP) and 3′,5′-cyclic guanosine monophosphate (cGMP). Previously, it has been reported that the PDE1 inhibitor ITI-214 shows positive inotropic effects in heart failure patients which were largely attributed to the cAMP-dependent protein kinase (PKA) signaling. However, the role of PDE1 in the regulation of cardiac cGMP has not been directly addressed. Here, we studied the effect of PDE1 inhibition on cGMP levels in adult mouse ventricular cardiomyocytes using a highly sensitive fluorescent biosensor based on Förster resonance energy transfer (FRET). Live-cell imaging in paced and resting cardiomyocytes showed an increase in cGMP after PDE1 inhibition with ITI-214. Furthermore, PDE1 inhibition and PDE1A knockdown amplified the cGMP-FRET responses to the nitric oxide (NO)-donor sodium nitroprusside (SNP) but not to the C-type natriuretic peptide (CNP), indicating a specific role of PDE1 in the regulation of the NO-sensitive guanylyl cyclase (NO-GC)-regulated cGMP microdomain. ITI-214, in combination with CNP or SNP, showed a positive lusitropic effect, improving the relaxation of isolated myocytes. Immunoblot analysis revealed increased phospholamban (PLN) phosphorylation at Ser-16 in cells treated with a combination of SNP and PDE1 inhibitor but not with SNP alone. Our findings reveal a previously unreported role of PDE1 in the regulation of the NO-GC/cGMP microdomain and mouse ventricular myocyte contractility. Since PDE1 serves as a cGMP degrading PDE in cardiomyocytes and has the highest hydrolytic activities, it can be expected that PDE1 inhibition might be beneficial in combination with cGMP-elevating drugs for the treatment of cardiac diseases.

Cyclic nucleotide phosphodiesterases 2A (PDE2A) and PDE3A play an important role in the regulation of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP)-to-cAMP crosstalk. Each of these PDEs has up to three distinct isoforms. However, their specific contributions to cAMP dynamics are difficult to explore because it has been challenging to generate isoform-specific knock-out mice or cells using conventional methods. Here, we studied whether the CRISPR/Cas9 approach for precise genome editing can be used to knock out Pde2a and Pde3a genes and their distinct isoforms using adenoviral gene transfer in neonatal and adult rat cardiomyocytes. Cas9 and several specific gRNA constructs were cloned and introduced into adenoviral vectors. Primary adult and neonatal rat ventricular cardiomyocytes were transduced with different amounts of Cas9 adenovirus in combination with PDE2A or PDE3A gRNA constructs and cultured for up to 6 (adult) or 14 (neonatal) days to analyze PDE expression and live cell cAMP dynamics. A decline in mRNA expression for PDE2A (~80%) and PDE3A (~45%) was detected as soon as 3 days post transduction, with both PDEs being reduced at the protein level by >50–60% in neonatal cardiomyocytes (after 14 days) and >95% in adult cardiomyocytes (after 6 days). This correlated with the abrogated effects of selective PDE inhibitors in the live cell imaging experiments based on using cAMP biosensor measurements. Reverse transcription PCR analysis revealed that only the PDE2A2 isoform was expressed in neonatal myocytes, while adult cardiomyocytes expressed all three PDE2A isoforms (A1, A2, and A3) which contributed to the regulation of cAMP dynamics as detected by live cell imaging. In conclusion, CRISPR/Cas9 is an effective tool for the in vitro knock-out of PDEs and their specific isoforms in primary somatic cells. This novel approach suggests distinct regulation of live cell cAMP dynamics by various PDE2A and PDE3A isoforms in neonatal vs. adult cardiomyocytes.

The therapeutic benefit of stimulating the cGMP pathway as a form of treatment to combat heart failure, as well as other fibrotic pathologies, has become well established. However, the development and signal compartmentation of this crucial pathway has so far been overlooked. We studied how the three main cGMP pathways, namely, nitric oxide (NO)-cGMP, natriuretic peptide (NP)-cGMP, and β₃-adrenoreceptor (AR)-cGMP, mature over time in culture during cardiomyocyte differentiation from human pluripotent stem cells (hPSC-CMs). After introducing a cGMP sensor for Förster Resonance Energy Transfer (FRET) microscopy, we used selective phosphodiesterase (PDE) inhibition to reveal cGMP signal compartmentation in hPSC-CMs at various times of culture. Methyl-β-cyclodextrin was employed to remove cholesterol and thus to destroy caveolae in these cells, where physical cGMP signaling compartmentalization is known to occur in adult cardiomyocytes. We identified PDE3 as regulator of both the NO-cGMP and NP-cGMP pathway in the early stages of culture. At the late stage, the role of the NO-cGMP pathway diminished, and it was predominantly regulated by PDE1, PDE2, and PDE5. The NP-cGMP pathway shows unrestricted locally and unregulated cGMP signaling. Lastly, we observed that maturation of the β₃-AR-cGMP pathway in prolonged cultures of hPSC-CMs depends on the accumulation of caveolae. Overall, this study highlighted the importance of structural development for the necessary compartmentation of the cGMP pathway in maturing hPSC-CMs.

Adenine nucleotide (AN) 2nd messengers, such as 3′,5′-cyclic adenosine monophosphate (cAMP), are central elements of intracellular signaling, but many details of their underlying processes remain elusive. Like all nucleotides, cyclic nucleotide monophosphates (cNMPs) are net-negatively charged at physiologic pH which limits their applicability in cell-based settings. Thus, many cellular assays rely on sophisticated techniques like microinjection or electroporation. This setup is not feasible for medium- to high-throughput formats, and the mechanic stress that cells are exposed to raises the probability of interfering artefacts or false-positives. Here, we present a short and flexible chemical route yielding membrane-permeable, bio-reversibly masked cNMPs for which we employed the octanoyloxybenzyl (OB) group. We further show hydrolysis studies on chemical stability and enzymatic activation, and present results of real-time assays, where we used cAMP and Ca²⁺ live cell imaging to demonstrate high permeability and prompt intracellular conversion of some selected masked cNMPs. Based on these results, our novel OB-masked cNMPs constitute valuable precursor-tools for non-invasive studies on intracellular signaling.