Distinct submembrane localisation compartmentalises cardiac NPR1 and NPR2 signalling to cGMP

Figures

• 

  Fig. 1. Spatial distribution of NPR1 and NPR2 induced cGMP signals in cardiomyocytes. a, b Representative SICM image and corresponding FRET ratio traces recorded from whole single mouse ventricular cardiomyocytes expressing the cytosolic cGMP biosensor red cGES-DE5 after local NPR2 stimulation on the crest of the cell and in the T-tubule. The cell was superfused with buffer A, and NPR2 was locally stimulated from the scanning nanopipette filled with CNP (100 µM) by applying pressure (276 kPa). Estimated peptide concentration at the membrane was 700 nM (see Supplementary Fig. 3). c Quantification of the cGMP-FRET responses to CNP. d Spatio-temporal pattern of FRET response measured in different regions of the cell after CNP application to a single T-tubule, showing far-reaching cGMP signals. Scale bar, 10 µm. e, f Representative SICM image and corresponding FRET ratio traces after local NPR1 stimulation from the scanning nanopipette filled with 100 µM ANP. Quantification is in g. All data plotted as means ± s.e.m. Number of cells/mice is above the bars. Difference in c is not significant; in g, P < 0.01 by mixed ANOVA followed by Wald χ2-test. h Spatio-temporal pattern of FRET signals…

• 

  Fig. 2. Control SICM/FRET experiments. Recordings were performed in presence of 100 µM IBMX (unselective PDE inhibitor, preincubated for 3–5 min in the bath solution) (a, b) or in NPR2 knockout myocytes (c, d) as described in Fig. 1. Representative scans and data quantifications are shown. These data rule out the possibility that local PDEs might affect the specific NPR1 localisation measured by SICM/FRET in Fig. 1e–g and confirm the specificity of the applied stimulation protocol for the studied receptors/ligands. Data represent mean ± s.e.m. values of the indicated number of cells/mice. *P < 0.01 (Kruskal–Wallis test)

• 

  Fig. 3. Concentration-response dependences of CNP and ANP signals. Responses to different ANP and CNP concentrations were recorded either using SICM/FRET (a, b) as described in Fig. 1, or by simple FRET imaging with global ligand application into the bath solution (c, d). In a, b, calculated peptide concentrations at the membrane are shown. 700, 210, and 70 nM were obtained by filling SICM pipetted with 100 µM, 30 µM, and 10 µM of CNP or ANP, respectively. Data for 700 nM are reproduced from Fig. 1 to facilitate the comparison. Data represent mean ± s.e.m. In a, b, numbers of cells/mice are shown above the bars. In d, 9–12 cells from three mice each were analysed for each data point. In c, data are reproduced from our previous study19 for comparison

• 

  Fig. 4. Fluorescence recovery after photobleaching (FRAP) analysis for various NPR1-EYFP and NPR2-EYFP constructs. HEK293A cells were transfected with receptor constructs described in Supplementary Figure 7 and imaged 24–30 h later with and without MβCD treatment (1 mM for 1 h at 37 °C). Fluorescence was measured and bleached in a rectangular cell membrane region of interest using 488 nm laser and confocal microscope as described in Methods. Representative fluorescence recovery curves (a, b, d, e) and their time constant analysis after monoexponential fit (c, f) are shown. a NPR1-EYFP receptor mobility is sensitive to MβCD treatment. b This effect is abolished when TM domain of NPR1 is replaced by that of NPR2 in the NPR1/2 construct. d NPR2-EYFP receptor mobility is not sensitive to MβCD treatment. e However, it is changed by MβCD in the NPR2/1 construct which contained TM domain derived from NPR1. The opposite direction of MβCD effect (as compared to wild-type NPR1, compare e and a) might indicate that TM domain is crucial but not the only factor which affects NPR1 protein mobility. Data represent mean ± s.e.m. n = 10 cells each. * p < 0.05; *** p < 0.001; n.s., not significant by…

• 

  Fig. 5. T-tubular ANP/NPR1/cGMP responses are under tight local control by PDE2. a Schematic representation of a genetic construct used to generate pmDE5 transgenic mice expressing this targeted biosensor in adult VMs under the control of the α-myosin heavy chain (α-MHC) promoter. b Live-cell confocal imaging of isolated pmDE5 myocytes stained with di-8-ANEPPS (di-8) confirms the proper localisation of the sensor to caveolin-rich membrane domains. Representative images, n = 10. Scale bar, 10 µm. c, d, f Quantification of FRET responses recorded upon application of the selective PDE inhibitors BAY 60-7550 (PDE2, 100 nM), cilostamide (PDE3, 10 µM) and tadalafil (PDE5, 100 nM) either alone at basal state (c) or after 100 nM CNP (d) and 100 nM ANP (f) prestimulation. Data are means ± s.e.m. Number of cells/mice is above the bars. *P < 0.01, **P < 0.01 by mixed ANOVA followed by Wald χ2-test. In c, d and f, Sapp/RFP FRET ratio before addition of PDE inhibitors (at basal state of after NP stimulation) was normalised on 1 to calculate % changes in FRET ratio and to facilitate the comparison between various conditions. e Representative FRET trace from pmDE5 myocyte subsequently stimulated with…

• 

  Fig. 6. Differential subcellular compartmentation of NPR1 and NPR2/cGMP responses. a Spatio-temporal pattern of FRET signals after ANP application shows far-reaching cGMP response after pre-incubation with 100 nM of the PDE2 inhibitor BAY 60-7550. Scale bar, 10 µm. Quantification is in b. b includes also the quantification of experiments shown in Fig. 1d, h. Shown are means ± s.e.m five cells from three mice per condition were analysed. Differences between all three conditions were significant at p = 0.05 for each individual distance value using mixed ANOVA followed by Wald χ2-test. c Immunoblot analysis of phospholamban (PLN) phosphorylation on Ser16 in working hearts perfused with 100 nM ANP with or without 100 nM BAY 60-7550. * P < 0.05, ** P < 0.01, and n.s. not significant by one-way ANOVA. d Schematic diagram illustrating the major findings of the study. NPR2 is widely distributed across the membrane, generating far-reaching cGMP responses. This cGMP pool is under mild control by PDE3 and causes phosphorylation of PLN and troponin I (TnI), leading to negative inotropic and positive lusitropic effects. NPR1 is exclusively found in T-tubules where it generates locally confined cGMP…