Developing high-throughput tools for study of cyclic nucleotide signalling in aging hipsc-cm

Abstract

There is potential in the use of HiPSC-CMs (human induced pluripotent stem cell derived cardiomyocytes) in drug testing, genetic studies, and cell replacement therapy. However, current methods of differentiation produce cells that structurally and physiologically differ from adult human cardiomyocytes. Adult cardiomyocytes regulate the signalling molecules cAMP and cGMP that control contraction and relaxation into strict nano-domains through structural elements such as caveolae, t-tubules, and the SR, then further regulate them through a hydrolysing protein family of phosphodiesterases (PDEs). Of these signalling molecules, cGMP is relatively lacking in descriptive studies. In our study we first produce tools and methods needed for experiments studying cGMP signalling in HiPSC-CM. Then, using our novel tools and methods as well as existing ones, we examine the maturation of cGMP signalling between D30 and D90 HiPSC-CM by characterising pathways that produce cGMP, the PDEs that can hydrolyse cGMP pools produced through each of these pathways, and the effect of structural compartmentation through caveolae on these pools. MultiFRET is a novel and highly flexible high-throughput real-time FRET measurement and analysis tool, which was developed in Java for use with the Icy Bioimaging suite. With MultiFRET we increase the number of cells measured in an experiment 50 times what was possible before under the same circumstances. Though it is technically possible to obtain even more cells, provided the computer has enough RAM and the dish contains enough cells, this was not tested in our experiments. MultiFRET further exhibits functionality for enhanced real-time and post-experiment analysis, as well as for the simultaneous measurement of multiple FRET sensors. We test several cGMP-detecting FRET sensors for appropriate sensitivity and find that ScGi performs the best for our needs. We then lay the groundwork for the generation of a transgenic HiPSC-CM line that expresses ScGi and one that expresses a cAMP sensor with compatible fluorophores for multiplexed FRET measurements with ScGi. Experiments using the ScGi FRET sensor to measure the effects of stimulating three different cGMP producing pathways; the NO-pathway (nitric oxide), the NP-pathway (natriuretic peptide), and the β3-pathway. After this stimulation, we inhibit one of the following PDEs relevant to cGMP: PDE1, PDE3, PDE5, or PDE9. Our results show that the NO-pathway produces less cGMP in D90 versus D30, that PDE3 regulation of NO-cGMP decreases with this maturation, that NP-cGMP regulation by PDE3 and PDE5 decrease after maturation, and that PDE2 and PDE3 regulation of β3-cGMP decreases in D90. We then examine the effects of lipid depletion to remove caveolae on our previous experimental conditions and find that removal of caveolae results in a decrease in NO-cGMP response in D30 and D90 and an increased regulation by PDE2 and PDE9 in D30, that caveolae removal increases NP-cGMP production in both D30 and D90, and that caveolae removal enhances β3-cGMP production in D90. Supporting our FRET studies, we examine the change in expression of relevant genes and proteins and find increases in PDE5A and ADRB3 as well as a decrease in PDE9A gene Page 4 of 236 expression from D30 to D90. Protein expression however only showed a decrease in PDE1C and PDE2A as well as an increase in PDE4B. Finally, we compare contraction and calcium handling dynamics between the experimental conditions of some of our FRET experiments using a new CytoCypher device, but surprisingly find no significant effect of our drugs. In conclusion we provide a comprehensive characterisation of the regulation of cGMP and its change between D30 and D90 HiPSC-CM, as well as the tools and methods to dive deeper.