Phospholipase Cε hydrolyzes perinuclear phosphatidylinositol 4-phosphate to regulate cardiac hypertrophy

Abstract

Phospholipase Cε (PLCε) is a multifunctional enzyme implicated in cardiovascular, pancreatic, and inflammatory functions. Here we show that conditional deletion of PLCε in mouse cardiac myocytes protects from stress-induced pathological hypertrophy. PLCε small interfering RNA (siRNA) in ventricular myocytes decreases endothelin-1 (ET-1)-dependent elevation of nuclear calcium and activation of nuclear protein kinase D (PKD). PLCε scaffolded to muscle-specific A kinase-anchoring protein (mAKAP), along with PKCε and PKD, localizes these components at or near the nuclear envelope, and this complex is required for nuclear PKD activation. Phosphatidylinositol 4-phosphate (PI4P) is identified as a perinuclear substrate in the Golgi apparatus for mAKAP-scaffolded PLCε. We conclude that perinuclear PLCε, scaffolded to mAKAP in cardiac myocytes, responds to hypertrophic stimuli to generate diacylglycerol (DAG) from PI4P in the Golgi apparatus, in close proximity to the nuclear envelope, to regulate activation of nuclear PKD and hypertrophic signaling pathways.

Figure 1

Conditional Deletion of PLCε in Cardiac Myocytes Prevents Development of Cardiac Hypertrophy. (A) Domain Structure of PLCε and Strategy for Conditional Deletion of PLCε. Exon 6 encodes the first common exon of two PLCε splice variants at the amino terminus of the CDC25 domain. Exon 6 was flanked by two LoxP sites as shown. Small arrows indicate location of primers for genotyping, small bars indicate the location of Southern blot probes. (B) 1 month old PLCε^fl/flCre⁺ and PLCε^fl/flCre⁻ mice were injected with 40 mg/kg of tamoxifen once/day for three consecutive days. PLCε mRNA was measured by real time quantitative PCR and normalized to GAPDH levels. (C) PLCε protein was immunoprecipitated and analyzed by western blotting. GAPDH from the lysates was immunoblotted as a loading control. KO controls are from globally deleted PLCε^−/− mice shown for comparison. (D) Anatomical view, histological HE stained sections, and trichome stained (for fibrosis, blue) sections from hearts from tamoxifen treated PLCε^fl/fl mice with or without 4 weeks of transaortic constriction. (E) Quantitation of heart weight to tibia length (HW/TL), atrial natriuretic factor (ANF)/GAPDH mRNA levels by RT-PCR, Ejection Fraction (by echocardiography), and Fractional Shortening (by echocardiography) (+/− SEM) . Analyzed by One way ANOVA, 7 mice Cre⁻, 11 mice Cre⁺, *p<0.05, **p<0.01, ***p<0.005. (F) Western blots of PKD, phosphoPKD, CamKII, phosphoCamKII, HDAC, PhosphoHDAC and GAPDH from tamoxifen injected mice with and without TAC. See also Figure S1.

Figure 2

Gα_q-stimulated hypertrophy in NRVMs is blocked by PLCε siRNA or disruption of nuclear envelope scaffolding via mAKAP. (A) NRVMs were transduced with 50 MOI of adenovirus expressing wild type Gα_q or LacZ. Cells were co-transduced with either PLCε siRNA or control scrambled siRNA (Ctl) adenovirus. After 48 hours cells were fixed, permeabilized and stained for α-actinin. Cell area was calculated from 200 cells each treatment with NIH Image J and pooled from 3 separate experiments in the right panel (+/− SEM). (B) Same as A except ANF/GAPDH ratio was determined by quantitative real time PCR (+/− SEM). (C) NRVMs were cotransduced with adenoviruses expressing Gα_q and mAKAP-SR domain. ANF/GAPDH ratio was measured after 48 h (+/− SEM). All experiments were repeated 3 times and were analyzed by One way ANOVA, *p<0.05; ***p<0.005.

Figure 3

PLCε is in a multicomponent signaling complex with mAKAP, Epac, PKCε, PKD and Ryr2 in the heart. (A) perinuclear localization of PLCε in NRVMs. mCherry tagged PLCε was expressed in NRVMs (left panel) and costained with DAPI (right panel). The boundaries of the cell are outlined with dashed lines. (B) Epac co-immunoprecipitates with PLCε and mAKAP from heart lysates. (C) Ryr2 immunoprecipitates with PLCε from heart lysates. (D) PKCε immunoprecipitates with PLCε and mAKAP from heart lysates. (E) PKD immunoprecipitates with PLCε and mAKAP from heart lysates. PLCε^−/− mice were tested as a specificity control in B,C and D. See also Figure S2.

Figure 4

PLCε regulates nuclear PKD activation. (A) NRVMs were infected with PLCε or control (Ctl) siRNA, followed by treatment with 100 nM ET-1 or 10 μM Norepinephrine (NE) for 1h and ET-1 for 24h, and assessment of PKD phosphorylation by Western blotting each repeated 3–4 times. Statistically different from Ctl siRNA **p<0.01, ***p<0.005 (B) NRVMs were infected with either control YFP, mAKAP-SR or PLCε-RA expressing adenoviruses to disrupt PLCε-mAKAP scaffolding, followed by treatment with 100 nM ET-1 for 1h or 24h and assessment of total PKD activation. Statistically different from Ctl siRNA ***p<0.005 (C) NRVMs were transduced with adenovirus expressing nuclear targeted D kinase activation reporter (nDKAR). The YFP to CFP ratio is shown as a pseudocolor image to emphasize the high level of FRET in the nucleus. D) a region in the nucleus from NRVMs expressing showing nDKAR FRET was selected and the YFP/CFP ratio was followed for the indicated times after 100 nM ET-1 addition in PLCε siRNA or ctl siRNA adenovirus treated NRVMs. E) Pooled data for the YFP/CFP ratio of nDKAR analyzed 20 min after addition of ET-1 (+) or vehicle (−) from 5 independent experiments in PLCε siRNA or Ctl siRNA expressing NRVMs (50–100 cells each condition) ***p<0.005. F) NRVMs were cotransfected with plasmids expressing nDKAR and mAKAP-SR1 or control LacZ. ET1 (100nM) was added at the indicated time. Data are pooled from 4 independent cells for each treatment from 2 separate NRVM preparations. All quantitative data is (+/− SEM). See also Figure S3.

Figure 5

PLCε is involved in regulation of nuclear Ca²⁺ elevation. (A) NRVMs were loaded with Fura2 and the Fura2 ratio in the nucleus was measured with time after 200 nM ET-1 addition in the presence of a 10 μM nifedepine/ 2 μM mibefradil. Left panel is a representative trace from two individual cells. Middle left panel is the combined data from multiple cells comparing Ctl siRNA and PLCε siRNA treated NRVMs, ***p<0.005. Middle right panel is the combined data from cells expressing either YFP or the YFP-RA domain to disrupt mAKAP scaffolding. Both sets of data are from 4 independent experiments from cells isolated from 4 different NRVM preparations with >10 cells for each coverslip and 3 coverslip averages for each for each NRVM preparation, *p<0.05 calculated based on an average of 12 coverslip averages for each condition. Right panel shows that the ET-1-dependent nuclear Ca²⁺ response is not blocked by Ryanodine but the Caffeine response is (This experiment was repeated with 2 separate sets of NRVMs, data is representative of one experiment with 30 cells each condition). All data are +/− SEM. (B) 2-photon microscopy was used to measure nuclear Ca²⁺ levels in fluo-4 loaded NRVMs. Cells were perfused with imaging buffer containing 10 μM nifedepine and 1.8 μM mibefradil to block Ca²⁺ transients associated with voltage-dependent Ca²⁺ release followed by addition of 100 nM ET-1. Individual panels show the level of Fluo4 fluorescence at the indicated times (See Supplemental Movie 1). The boxes in the first panel correspond to the areas measured shown in the traces shown on the right.

Figure 6

PI4P localizes to perinuclear Golgi surrounding the nuclear envelope in cardiac myocytes. (A) Detection of PI4,5P₂ localization in NRVMs. Left panel: HEK-293 cells transfected with Tubby GFP and stained with DAPI, middle: NRVMs transfected with Tubby-GFP and right: NRVMs transfected with PLCδ-PH-GFP and analyzed by confocal microscopy. (B) Detection of PI4P at the nuclear envelope. The indicated cell types were transfected with either OSBP-PH-GFP or FAPP-PH-GFP and analyzed by confocal microscopy. (C) Inhibition of ARF eliminates perinuclear staining with FAPP-PH-GFP. NRVMs transfected with FAPP-PH-GFP were treated with 100 ng/mL Brefeldin A and GFP fluorescence monitored with time. Cells treated with vehicle are shown in the bottom panels indicate a lack of photobleaching in these experiments. (D) Inhibition of PI4kinase with PAO depletes perinuclear fluorescence associated with FAPP-PH-GFP. NRVMs transfected with FAPP-PH-GFP were treated with 10 μM PAO and fluorescence analyzed by confocal microscopy. All experiments were repeated a minimum of 3 times. See also Figure S4.

Figure 7

Perinuclear PI4P is a substrate for mAKAP-scaffolded PLCε, stimulated by either Epac or ET-1 receptors. (A) NRVMs transfected with FAPP-PH-GFP were analyzed by live cell confocal microscopy before and after treatment with 10 μM cpTOME for 1 h. (See also Supplemental Movie 2) (B) Individual regions GFP fluorescence in NRVMs transfected with FAPP-PH-GFP were monitored with confocal microscopy and followed with time after treatment with vehicle, 10 μM cpTOME, 50 nM ET-1 or 50 nM ET-1+ 100 nM BQ-123 (top panels are representative traces). Data was pooled from 4 experiments at 0 and 50 min for quantitation and statistics (Bottom panels). (C) NRVMs were treated with Vehicle or 10 μM cpTOME for 50 min followed by extraction of PI4P and assay using a PI4P protein-lipid overlay assay according to the manufacturer's instructions. Data from three separate experiments are quantitated in the bottom panel and analyzed by a student's t-test. (D) NRVMs were cotransfected with FAPP-PH-GFP and either mAKAP-SR1 or control plasmid. Perinuclear GFP fluorescence was monitored as in B. ET-1 experiments in B and D were done in parallel so the ET-1 alone representative traces are the same in both panels. Top panel is a representative trace and bottom panel is pooled data from 3 experiments analyzed by one way ANOVA. (E) NRVMs were transfect with FAPP-PH-GFP and transduced with viruses expressing PLCε siRNA or random control siRNA and perinuclear GFP fluorescence was monitored as in B and D. Data are pooled from 5 independent experiments each. (F) NRVMs were transfected with YFP-C1b-Y123W to detect DAG localization. Left panel shows localization of YFP-C1b-Y123W by confocal microscopy. Right panel: NRVMs transfected with YFP-C1b-Y123W were stimulated with 10 μM cpTOME and perinuclear regions and cytoplasm were imaged over time. The ratio of perinuclear fluorescence (F_pn) to the cytoplasmic fluorescence (F_c) was calculated and normalized to the starting F_pn/F_c before cpTOME addition. Data are pooled from 4 independent experiments, F_pn/F_c at 9–12 min were individually compared with F_pn/F_c at 0 min with a one way ANOVA. G) Golgi specific depletion of PI4P blocks ET-1-dependent nuclear PKD activation. Left panel; cells were transduced with GFP-tagged, Golgi targeted Sac-1 and imaged by confocal microscopy. Right panel: Cells were cotransfected with plasmids expressing nDKAR and Golgi targeted Flag-tagged Sac-1 plasmids and nDKAR FRET was monitored as in Fig. 4 D and F. 50 nM ET-1 was added at the indicated time. Data pooled from 4 independent cells for each treatment from 2 separate NRVM preparations. All data are +/− SEM. See also Figures S6 and S7.