Changes in cardiomyocyte structure and cAMP/cGMP signalling during heart failure

Abstract

The contractile function of the heart depends on efficient β adrenergic receptor (βAR) signalling which involves cycling nucleotides as second messengers. Correct secondary messenger signalling is only possible in healthy, well structured cardiac myocytes. Of the three βAR subtypes present in human cardiomyocytes β1AR and β2AR classically signal via 3'-5' cyclic adenosine monophosphate (cAMP) to regulate contraction after catecholamine administration, whereby the second isoform may also be cardioprotective. The far less characterised β3AR has been controversially associated to both increasing contraction through cAMP and protecting the heart through 3'-5' cyclic guanosine monophosphate (cGMP) signalling. During the progression of heart failure following myocardial infarction (MI) both the normal cell structure and the regulation of cAMP and cGMP signalling are changed. This happens in part due to changes in catecholaminergic stimulation of the βARs and in mechanical load, as well as due to a progressive development of hypertrophy. Some of the alterations initially appear to be of a compensatory nature but escalate into HF by worsening cardiomyocyte function and cell survival. The work presented here (1) investigates the structural integrity of healthy, isolated, single cardiomyocytes by looking at the surface topography via Scanning Ion Conductance Microscopy (SICM) imaging and by examining the internal Transverse Axial Tubule (TAT) network via confocal imaging; (2) elucidates the cyclic nucleotide response to catecholamine stimulation following either global (in the solution) or local (in the SICM pipette) stimulation of either β2ARs or β3ARs and measuring either cAMP or cGMP levels via Förster Resonance Energy Transfer (FRET) sensors in a combined FRET/SICM imaging setup; (3) determines how both the structure and β2AR and β3AR dependent second messenger signalling change in a progressive rat model of HF 4, 8 and 16 weeks after the induction of chronic MI. The major findings of the presented work are as follows: In control cardiomyocytes the structure is highly intricate with regular Z-grooves and crest areas. In MI cells the normal suface topography progressively deteriorates, with the eventual disappearance of Z-grooves by week 16, which correlates with the disorganisation of the cardiomyocyte’s internal transverse axial tubule (TAT) network of T-tubules emanating from the cell surface and traversing into the cell centre. This is accompanied by the gradual redistribution of β2ARs from their normal position inside the T-tubules to the unstructured areas on the cardiomyocyte membrane. The regularity and density of the TAT network is already severely compromised at 4 weeks post MI; at the same time a significant drop in the expression of the structural protein Junctophilin 2 (JPH2) occurs. At 4 and 8 weeks post MI a potentially compensatory increase in the number of longitudinal elements takes place which was no longer detectable at 16 weeks. The production of cAMP following local stimulation of β2ARs in the T-tubule openings was already suppressed at 4 weeks post MI and a β2AR response becomes detectable after local stimulation at the cell crests (areas between Z-grooves) at 8 weeks post MI. At 16 weeks post MI the β2AR-dependent cAMP level following both global and local stimulations was reduced due to an overall decrease in the adenylate cyclase (AC) activity. The production of the second cyclic nucleotide, cGMP, following β3AR stimulation is evident in control cells and to a significantly lesser extent in myocytes isolated from hearts at the end stage of HF. These β3AR-cGMP levels were degraded mainly by phosphodiesterases (PDE) 2 and 5. Local stimulation through the SICM pipette reveals that functional β3ARs are primarily localized inside T-tubules in control cells but redistribute equally in between T-tubules and crests in cells isolated from failing hearts. To improve the accuracy and reliability of local application of agonists via the SICM nanopipette voltage was applied to the pipette, as opposed to previously employed displacement of the liquid in the pipette via air pressure. Mathematical modelling served to determine the correct settings for this voltage driven application. It shows that the SICM nanopipette can reliably and precisely unload the βAR agonist ISO onto the nanoscale structure of cardiomyocytes via voltage.