POPDC1 scaffolds a complex of adenylyl cyclase 9 and the potassium channel TREK‐1 in heart

Figures

• 

  Figure 1. Deletion of Adcy9 results in bradycardia at rest and HR variability during the recovery period after ISO injection A Heart rate (HR) of WT and Adcy9 −/− mice averaged over 5 min of nonactive versus sustained activity on home cage running wheels. The Student's t‐test P‐values for indicated comparisons are shown. Circles represent data from individual animals. The boxplot central band is the median while whiskers represent the 10th and 90th percentiles.B, CHR and R‐R interval of WT (black bars) and Adcy9 −/− mice (red bars) before and after ISO (1 μg/g) injection; mean of 5 min intervals ± SE are shown.DRepresentative ECG recordings of 7 months WT and Adcy9 −/− mice, ~ 55 min post‐ISO injection.E, FHR variability was calculated before and after ISO injection using the root mean square of successive RR interval differences (RMSSD; panel (E)) and percentage of sequential R‐R intervals differing by >6 ms (pNN6; panel (F)) methods. Data information: Two‐way ANOVA repeated‐measures analysis (time and genotype) with the Holm–Sidak method used for pairwise comparisons for HR, RR, and HR variability with ISO; n = 9 WT and n = 10 Adcy9 −/− mice; *P < 0.05; **P < 0.01; ***P < 0.001. Source…

• 

  Figure 2. Cellular interaction of POPDC isoforms with AC9 A, BPLA assay performed in neonatal cardiomyocytes expressing GFP‐Flag or AC9‐Flag using antibodies against Flag and endogenous POPDC1, POPDC2, or Gβγ. Images (A); scale bar is 20 μm; nuclear staining by DAPI shown in blue) and quantitation (B) of PLA signal is shown. The unpaired two‐sided Student's t‐test (n = 4 independent experiments, **P = 0.0086) was performed for each antibody pair. Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.CCartoon of AC9‐POPDC BiFC complex.DQuantification of COS‐7 cells expressing BiFC constructs for AC9, POPDC 1–3. The expressed proteins tagged with VN (top line) and VC (bottom line) are shown. The Kruskal–Wallis one‐way ANOVA analysis on ranks was performed (n = 4 experiments) with multiple comparisons with control by the Dunnett's method (# P < 0.05). Individual comparisons were performed with a two‐tailed Student's t‐test (**P = 0.003, *P = 0.018) or Mann–Whitney rank‐sum test (^P = 0.004 versus control) Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are…

• 

  Figure EV1. PLA signal between GFP‐tagged AC9 and POPDC1‐Myc A, BImages (A) of PLA signal (red) and DAPI (blue) performed in HEK293 cells using antibodies against GFP and MYC tags. Scale bars are 20 μm. Mean cellular fluorescence intensity of PLA signal (B) was quantified by high content microscopy. The Kruskal–Wallis one‐way ANOVA analysis was performed (n = 7 experiments, P = 0.003 between groups) with multiple comparisons by the Bonferroni t‐test (*P < 0.05). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the 10th and 90th percentiles are represented by the whiskers.CCompetition of PLA signal between GFP‐tagged AC9 and POPDC1‐Myc with overexpression of nontagged EGFR (control), Flag‐AC9, or POPDC1‐Flag in HEK293 cells. Mean cellular fluorescence intensity was quantified by high content microscopy and normalized to GFP control. The Kruskal–Wallis one‐way ANOVA analysis was performed (n = 3 experiments, >1,000 cells per experiment) with multiple comparisons by the Student–Newman–Keuls method (*P < 0.05). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.

• 

  Figure 3. AC9 interacts with the POPDC1 transmembrane and Popeye domains HEK293 cells expressing Flag‐AC9 in the presence or absence of Myc‐tagged POPDC1 or − 2 were subjected to co‐immunoprecipitation (Co‐IP) with anti‐MYC and assayed for AC activity with 300 nM Gαs‐GTPγS. The Kruskal–Wallis one‐way ANOVA analysis on ranks was performed (n = 6 experiments, P = 0.003 between all groups) with multiple comparisons with AC9 control by the Dunn's method and Mann–Whitney rank‐sum test (**P = 0.002). Data are plotted as mean ± SEM.A portion of the lysates and Co‐IP from (A) were subjected to western blot (WB) analysis. Sf9 cells expressing Flag‐AC9 served as a positive WB control. Molecular weight markers are denoted as M. Quantitation of Flag‐AC9 WB by one‐way ANOVA, n = 3–4 experiments, with comparisons by the Tukey test, **P = 0.003, ***P = 0.001. Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.HEK293 cells expressing Flag‐AC9 +/− POPDC1‐Myc or POPDC2‐Myc were subjected to Co‐IP with anti‐FLAG and subjected to WB analysis with anti‐MYC and anti‐FLAG. Quantitation of anti‐MYC WB by one‐way ANOVA, n = 4 experiments,…

• 

  Figure 4. AC9 is part of a POPDC1‐TREK‐1 complex Co‐localization of CFP‐TREK‐1 with the BiFC signal from AC9‐VN:AC9‐VC homodimer (top) and POPDC1‐VN:AC9‐VC complex (bottom) in HEK293 cells. Scale bar: 10 μm.The lifetime distribution of Cerulean‐tagged proteins +/− the indicated YFP‐tagged proteins expressed in HEK293 cells. The Mann–Whitney rank‐sum test was performed (***P < 0.001; n = cell number indicated on each bar). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.Co‐IP of Myc‐tagged POPDC1 (anti‐MYC) pulled down both Flag‐AC9 and YFP‐tagged TREK‐1 (n = 3 experiments). The one‐way ANOVA was performed with multiple comparisons by the Holm–Sidak method (# P = 0.007 compared with control). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.Co‐IP of TREK‐1 (anti‐GFP) pulls down Flag‐AC9 and endogenous POPDC1. AKAP79 is not pulled down in complex (n = 3 experiments). The one‐way ANOVA was performed with multiple comparisons by the Student–Newman–Keuls method (# P < 0.05 compared with control; *P = 0.013 and 0.026 comparing bars 2–3 and…

• 

  Figure 5. TREK‐1 pulls down AC9 activity in heart, which requires POPDC1 expression IP‐AC assay with IgG versus anti‐TREK‐1 in heart homogenates from WT and Adcy9 −/− mice; AC activity stimulated with 300 nM Gαs‐GTPγS. IP‐AC with anti‐AKAP150 is shown as a positive control. The two‐way ANOVA (overall P < 0.001, n = 3 mice per genotype) with multiple comparisons by the Holm–Sidak test; # P < 0.001 for indicated comparison or to respective IgG controls. IgG versus anti‐TREK‐1 in Adcy9 −/− was analyzed by the Student's t‐test (**P = 0.009, n = 3). A portion of the co‐IP was subjected to western blotting with anti‐TREK‐1 and anti‐AKAP150. Data are plotted as mean ± SEM.IP‐AC assay with IgG versus anti‐TREK in WT and Popdc1 −/− mouse heart homogenates; AC activity stimulated as in (A). Statistics as in (A), n = 4 mice per genotype, #P < 0.001. Paired t‐test within Popdc1 −/−, P = 0.123. Data are plotted as mean ± SD.Total AC activity in heart homogenates from WT or Popdc1 −/− (P1 −/−) mice stimulated with 300 nM Gαs‐GTPγS or 100 μM calcium and 300 nM calmodulin. Student's t‐test (n = 4, n.s.). Data are plotted as mean ± SD.IP‐AC assay with anti‐MYC from cell lysates expressing AC1 or AC8 in the…

• 

  Figure 6. Isoproterenol treatment reduces AC9 and POPDC1 interaction with TREK‐1 ATreatment of HEK293 cells for 10 min in the presence of 10 μM ISO reduced the amount of Flag‐AC9 and POPDC1‐Myc pulled down by TREK‐1 co‐IP.B, CQuantification (IP/input) of Flag‐AC9 (B) and POPDC1‐Myc (C) between ISO treatment groups by a two‐tailed paired Student's t‐test (*P = 0.028, **P = 0.007; n = 3 biological replicates). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range. Source data are available online for this figure.

• 

  Figure 7. ISO‐dependent loss of TREK‐1 from AC9‐POPDC1 complex requires an intact Popeye domain ABiFC signal between the indicated VN‐ and VC‐tagged proteins (POPDC1, P1). Kruskal–Wallis one‐way ANOVA on ranks (n = 5 experiments, P < 0.001 between groups) with multiple comparisons with VN and VC controls by Student–Newman–Keuls tests (*P < 0.05). Data are plotted as mean ± SEM.BPercent decrease of BiFC signal in (A) with ISO treatment (10 μM, 10 min at 37°C). Paired t‐test for each condition (vehicle versus ISO) using raw data prior to calculation of percent of vehicle control (n = 6; *P < 0.05; **P < 0.01). Boxplots show the median as the central band, the box size as the lower and upper quartiles, while the whiskers are the range.CISO dose–response curves for BiFC interactions between POPDC1:AC9, TREK‐1:AC9, and TREK‐1 with a catalytically inactive AC9 (TREK‐1:AC9D). Two‐way ANOVA with multiple comparisons by the Bonferroni method (n = 3–4; *P < 0.05 for comparisons between TREK:AC9 versus TREK:AC9D and POPDC1:AC9). Data are plotted as mean ± SEM.DTREK‐1:AC9 BiFC signal is absent in COS‐7 cells lacking detectable POPDC1. TREK‐1:TREK‐1 BiFC is a positive control. One‐way ANOVA with…

• 

  Figure 8. AC9 activity opposes the enhancement of TREK‐1 currents by POPDC1 ARepresentative two‐electrode voltage‐clamp measurements in Xenopus oocytes injected with TREK‐1 alone (orange) or TREK‐1 co‐injected with cRNAs encoding the indicated proteins. Voltage was ramped from −120 to +40 mV.BRelative TREK‐1 current amplitudes at +40 mV normalized to TREK‐1 injected alone. The number of oocytes is given within the bar graphs (n = 44–97). 3–5 batches of oocytes were used for each condition. Data are presented as mean ± SEM. Two‐tailed Student's t‐test between the indicated groups (***P = 0.00018; ****P < 1 × 10−5).CRelative currents normalized to TREK‐1 plus POPDC1 in the absence and presence of theophylline. Data are plotted as mean ± SEM. Two‐tailed Student's t‐test (**P = 0.0011; ***P = 0.0002; ****P < 1 × 10−5; n = 40–85 biological replicates).D, EOocytes were treated with 0.5 mM theophylline prior to voltage‐clamp measurements as in (A, B). Data are presented as mean ± SEM. Two‐tailed Student's t‐test between the indicated groups (**P = 0.0011; ***P = 4 × 10−5; ****P < 1 × 10−7). Source data are available online for this figure.