Real-time imaging of cGMP signaling shows pronounced differences between glomerular endothelial cells and podocytes

Abstract

Recent clinical trials of drugs enhancing cyclic guanosine monophosphate (cGMP) signaling for cardiovascular diseases have renewed interest in cGMP biology within the kidney. However, the role of cGMP signaling in glomerular endothelial cells (GECs) and podocytes remains largely unexplored. Using acute kidney slices from mice expressing the FRET-based cGMP biosensor cGi500 in endothelial cells or podocytes enabled real-time visualization of cGMP. Stimulation with atrial natriuretic peptide (ANP) or SNAP (NO donor) and various phosphodiesterase (PDE) inhibitors elevated intracellular cGMP in both cell types. GECs showed a transient cGMP response upon particulate or soluble guanylyl cyclase activation, while the cGMP response in podocytes reached a plateau following ANP administration. Co-stimulation (ANP + SNAP) led to an additive response in GECs. The administration of PDE inhibitors revealed a broader basal PDE activity in GECs dominated by PDE2a. In podocytes, basal PDE activity was mainly restricted to PDE3 and PDE5 activity. Our data demonstrate the existence of both guanylyl cyclase pathways in GECs and podocytes with cell-specific differences in cGMP synthesis and degradation, potentially suggesting new therapeutic options for kidney diseases.

Keywords: FRET; GECs; Glomerular endothelial cell; Podocyte; cGMP signaling; cGi500; pGC; sGC.

Fig. 1

Podocytes respond to ANP/pGC stimulation with a long-lasting cGMP increase. Acute kidney slices obtained from (a) Tie2:Cre/cGi500 mice or (d) Pod: Cre/cGi500 mice showing cell-specific cGi500 biosensor expression with overlapping CFP/YFP fluorescence patterns in GECs and podocytes. (b, e) AKS were superfused (dotted lines) from 200 s to 570 s with 1 µM ANP (blue trace), 1 mM SNAP (orange trace), or the combination of both (black trace). Stimulation was followed by a washout phase with 1X KHB (37 °C). Time-lapse recordings for each cell type represent the mean of baseline-normalized FRET (CFP/YFP) ratios (R/R_o) of all analyzed glomeruli. Relative signal changes in ∆R/R_o (%) reflect changes in the intracellular cGMP concentration. (c, f) Scatter dot plots display the maximal ∆R/R_o response for each glomerulus derived from the indicated stimulation of (b) Tie2:Cre/cGi500 mice or (e) Pod:Cre/cGi500 mice during the entire measurement period of 1500 s. Data represent mean ± SEM of at least 9 glomeruli per condition obtained from slices of six Tie2:Cre/cGi500 mice and five Pod:Cre/cGi500 mice. One-way ANOVA, Tukey´s post hoc test, *P < 0.05. Scale bar 25 µM.

Fig. 2

Addition of the PDE inhibitor IBMX reveals higher PDE activity in GECs compared to podocytes. Acute kidney slices from (a) Tie2:Cre/cGi500 mice or (b) Pod:Cre/cGi500 mice were superfused (outer dotted lines) from 200 s to 860 s with 1 µM ANP (blue trace), 1 mM SNAP (orange trace) or the combination of both (black trace). In addition to the ongoing stimulation, 500 µM IBMX was administered continuously from 530 s to 860 s (gray area). Time-lapse recordings for each cell type represent the mean of baseline-normalized FRET (CFP/YFP) ratios (R/R_o) of all analyzed glomeruli. (c) Comparison of the additional cGMP/FRET response after IBMX administration between GECs and podocytes. Scattered bar plot displays the maximal ∆R/R_o response for each glomerulus in the respective measurement interval (w/o IBMX: 260–590 s; with IBMX: 600–1500 s). The delta (mean difference, shown in red) between the two time intervals was compared between GECs (E) and podocytes (P). Data represent mean ± SEM of at least 7 glomeruli per condition obtained from slices of three Tie2:Cre/cGi500 mice and four Pod:Cre/cGi500 mice. One-way ANOVA (Brown Forsythe (p < 0.0001), Welch test (p < 0.0001)), Dunnett T3 post hoc test, *P < 0.05.

Fig. 3

PDE inhibition reveals predominant PDE2a activity in GECs and PDE3 & PDE5 activity in podocytes. Acute kidney slices from (a) Tie2:Cre/cGi500 mice or (c) Pod:Cre/cGi500 mice were superfused for 370 s with 50 µM BAY 60-7550, 250 µM Cilostamide, 250 µM Roflumilast, 250 µM Avanafil or 100 µM PF-04449613. The primarily inhibited PDE isoform is stated in brackets. Baseline-normalized FRET (CFP/YFP) ratios (R/Ro) of all analyzed glomeruli were calculated. (b, d) Time-lapse recordings display a sustained cGMP response after BAY 60-7550 and Cilostamide administration in GECs (b), whereas Cilostamide and Avanafil lead to a transient cGMP response in podocytes (d). Scattered bar plots display the maximal ∆R/Ro response for each compound during the entire measurement period of 1500 s. Data represent mean ± SEM of at least 8 glomeruli per condition obtained from slices of nine Tie2:Cre/cGi500 mice and six Pod:Cre/cGi500 mice.

Fig. 4

PDE2a inhibition prevents cGMP degradation in GECs, whereas PDE3 & PDE5 inhibition augment the ANP/pGC/cGMP pathway in podocytes. Acute kidney slices from (a) Tie2:Cre/cGi500 mice or (c, e) Pod:Cre/cGi500 mice were superfused from 200 s to 800 s (outer dotted lines) with 1 µM ANP (blue trace), 1 mM SNAP (orange trace) or the combination of both (black trace). In addition to the ongoing stimulation, Tie2:Cre/cGi500 slices were superfused continuously with 50 µM BAY 60-7550 and Pod:Cre/cGi500 slices were superfused with 250 µM Cilostamide or 250 µM Avanafil in the time period from 500 s to 800 s (gray area). Time-lapse recordings represent the mean of baseline-normalized FRET (CFP/YFP) ratios (∆R/R_o) of all analyzed glomeruli. (b, d, f) Scattered bar plots display the maximal ∆R/R_o response for each glomerulus in the respective measurement interval (w/o PDE inhibitor: 200–570 s; with PDE inhibitor: 580–950 s) for (b) BAY 60-7550, (d) Cilostamide or (f) Avanafil. The delta (mean difference) between the two time periods is shown in red. Data represent mean ± SEM of at least 8 glomeruli per condition obtained from slices of eight Tie2:Cre/cGi500 mice after administration of BAY 60-7550, eight Pod:Cre/cGi500 mice for Cilostamide and nine Pod:Cre/cGi500 mice for Avanafil. Two-tailed paired t-test. *P < 0.05.