ABSTRACT G-protein-coupled receptors (GPCRs) regulate several physiological and pathological processes and represent the target of approximately 30% of FDA-approved drugs. GPCR-mediated signaling was thought to occur exclusively at the plasma membrane. However, recent studies have unveiled their presence and function at subcellular membrane compartments. There is a growing interest in studying compartmentalized signaling of GPCRs. This requires development of novel tools to separate GPCRs signaling at the plasma membrane from the ones initiated at intracellular compartments. We took advantage of the structural and pharmacological information available for β1-adrenergic receptor (β1AR), an exemplary GPCR that functions at subcellular compartments, and rationally designed spatially restricted antagonists. We generated a cell impermeable β1AR antagonist by conjugating a suitable pharmacophore to a sulfonate-containing fluorophore. This cell-impermeable antagonist only inhibited β1AR on the plasma membrane. In contrast, a cell permeable β1AR agonist containing a non-sulfonated fluorophore, efficiently inhibited both the plasma membrane and Golgi pools of β1ARs. Furthermore, the cell impermeable antagonist selectively inhibited the phosphorylation of downstream effectors of PKA proximal to the plasma membrane in adult cardiomyocytes while β1AR intracellular pool remained active. Our tools offer promising avenues for investigating compartmentalized β1AR signaling in various context, potentially advancing our understanding of β1AR-mediated cellular responses in health and disease. They also offer a general strategy to study compartmentalized signaling for other GPCRs in various biological systems.
GPCRs are increasingly recognized to initiate signaling via heterotrimeric G proteins as they move through the endocytic network, but little is known about how relevant G protein effectors are localized. Here we report selective trafficking of adenylyl cyclase type 9 (AC9) from the plasma membrane to endosomes while adenylyl cyclase type 1 (AC1) remains in the plasma membrane, and stimulation of AC9 trafficking by ligand-induced activation of Gs-coupled GPCRs. AC9 transits a similar, dynamin-dependent early endocytic pathway as ligand-activated GPCRs. However, unlike GPCR traffic control which requires β-arrestin but not Gs, AC9 traffic control requires Gs but not β-arrestin. We also show that AC9, but not AC1, mediates cAMP production stimulated by endogenous receptor activation in endosomes. These results reveal dynamic and isoform-specific trafficking of adenylyl cyclase in the endocytic network, and a discrete role of a heterotrimeric G protein in regulating the subcellular distribution of a relevant effector.
Phospholipase C (PLC)-mediated PI4P hydrolysis pathway at the Golgi is important for regulating cardiac hypertrophy. cAMP stimulates this pathway via Epac-mediated activation of Rap1 which directly binds to and activates PLCε. Here, we demonstrate that a membrane permeant β-adrenergic agonist, dobutamine (dob), and the endogenous β-adrenergic agonist, norepinephrine (NE), induce Golgi PI4P hydrolysis in neonatal rat ventricular myocytes (NRVMs) and adult ventricular myocytes. However, a membrane impermeant β-adrenergic agonist, isoproterenol (iso), does not. In addition, a membrane permeant βAR antagonist, metoprolol, but not a membrane impermeant antagonist, sotalol, fully reverses dob-mediated PI4P hydrolysis. Taken together, this suggests that internal β-ARs are required for inducing PI4P hydrolysis. Subsequently,we used YFP-tagged mini Gs proteins that recruit to Gs-coupled receptors upon their activation, to monitor the location of activated receptors. Dob and NE induced robust and rapid recruitment of mini Gs to the plasma membrane and the Golgi, however, iso only induces recruitment to the plasma membrane. Additionally, inhibition of PLCεat the Golgi with either siRNA or the RA1 domain of PLCε, or the use of an Epac inhibitor also inhibits dob-mediated PI4P hydrolysis. This suggests that PI4P hydrolysis by dob requires Golgi-localized PLCεand Epac.Inhibition of the βAR with Golgi-targeted Nb80 prevented PI4P hydrolysis by both dob and NE confirming that Golgi-localized βARs are required for PI4P hydrolysis.Also, Oct3 transporter inhibitors prevent NE-induced PI4P hydrolysis but dyngo, an inhibitor of receptor internalization, had no effect. This indicates that agonist transport and not receptor internalization is required for NE-mediated Golgi βAR activation and PI4P hydrolysis. Furthermore, dob stimulates an increase in cell size and ANF expression in NRVMs that is significantly inhibited by metoprolol but less effectively by sotalol. Taken together, these data suggest that Golgi βARs are involved in mediating cardiomyocyte hypertrophy and may serve as a novel target for treating heart failure.
Increased adrenergic tone resulting from cardiovascular stress leads to development of heart failure, in part, through chronic stimulation of β1 adrenergic receptors (βARs) on cardiac myocytes. Blocking these receptors is part of the basis for β-blocker therapy for heart failure. Recent data demonstrate that G protein-coupled receptors (GPCRs), including βARs, are activated intracellularly, although the biological significance is unclear. Here we investigated the functional role of Golgi βARs in rat cardiac myocytes and found they activate Golgi localized, prohypertrophic, phosphoinositide hydrolysis, that is not accessed by cell surface βAR stimulation. This pathway is accessed by the physiological neurotransmitter norepinephrine (NE) via an Oct3 organic cation transporter. Blockade of Oct3 or specific blockade of Golgi resident β1ARs prevents NE dependent cardiac myocyte hypertrophy. This clearly defines a pathway activated by internal GPCRs in a biologically relevant cell type and has implications for development of more efficacious β-blocker therapies.
ABSTRACT G protein-coupled receptors (GPCRs) activate different signaling pathways through coupling to heterotrimeric G proteins consisting of four types of G alpha (Gα) subunits and the beta gamma (ýy) subunits. The C-terminal regions of Gα proteins is thought to determine coupling selectivity. However, there are several reports indicating that some receptors are promiscuous and can couple to multiple Gα proteins. The precise mechanism promoting promiscuous coupling is not fully understood. Here we show that anionic phospholipids such as phosphoinositide 4,5 bisphosphate (PI(4,5)P2) promote electrostatic interactions between anionic lipid head groups and positively charged amino acids in the transmembrane 4 (TM4) region of beta adrenergic receptors (βARs) to prime receptor coupling to the cognate Gαs and non-cognate Gαi proteins. The promiscuous coupling can only occur at the plasma membrane, a membrane compartment enriched in PI(4,5)P2, while receptors only couple to cognate Gαs proteins in compartments that are enriched in less negatively charged phosphoinositides such PI4P at the Golgi membranes. We took advantage of the rapamycin dimerization system to rapidly and inducibly deplete PI(4,5)P2 by recruiting a phosphatase to the plasma membrane and found that in the absence of anionic phospholipids, βARs couple only to their cognate Gαs protein. Finally, we found that mutating βARs PI(4,5)P2 binding motifs or depleting PI(4,5)P2 results in enhanced βARs-mediated cAMP response. Together, these findings reveal a role for anionic phosphoinositides in regulating GPCR activity at different subcellular compartments.
Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Increased adrenergic tone resulting from cardiovascular stress leads to development of heart failure, in part, through chronic stimulation of β1 adrenergic receptors (βARs) on cardiac myocytes. Blocking these receptors is part of the basis for β-blocker therapy for heart failure. Recent data demonstrate that G protein-coupled receptors (GPCRs), including βARs, are activated intracellularly, although the biological significance is unclear. Here we investigated the functional role of Golgi βARs in rat cardiac myocytes and found they activate Golgi localized, prohypertrophic, phosphoinositide hydrolysis, that is not accessed by cell surface βAR stimulation. This pathway is accessed by the physiological neurotransmitter norepinephrine (NE) via an Oct3 organic cation transporter. Blockade of Oct3 or specific blockade of Golgi resident β1ARs prevents NE dependent cardiac myocyte hypertrophy. This clearly defines a pathway activated by internal GPCRs in a biologically relevant cell type and has implications for development of more efficacious β-blocker therapies. https://doi.org/10.7554/eLife.48167.001 Introduction Cardiovascular diseases, including heart failure, are the leading cause of disease and death in the developed world. In response to long term pathological stress, such as hypertension or myocardial infarction, the levels of neurohumoral factors such as epinephrine, endothelin and angiotensin II increase. These hormones directly stimulate G protein-coupled receptors (GPCRs), including β-adrenergic receptors in cardiac myocytes (D'Angelo et al., 1997; Adams et al., 1998; Rockman et al., 2002). Chronic stimulation of these receptors, including stimulation of β-adrenergic receptors (βARs) by catecholamines, drives cardiac hypertrophy and ventricular remodeling, ultimately leading to heart failure. The efficacy of β-blocker therapy for treatment of heart failure results, at least in part, by ameliorating chronic βAR stimulation in the heart (Bristow, 2000; Wang et al., 2018). Previous studies by our group, and others, have implicated specific phospholipase C (PLC) enzymes downstream of Gq and Gs-coupled GPCRs in the regulation of cardiac hypertrophy. PLCβ isoforms stimulated by Gq have been implicated in cardiac hypertrophy driven by cell surface α-adrenergic receptors (αAR) (Filtz et al., 2009; Grubb et al., 2011). Our laboratory identified a critical role of PLCε in cardiac hypertrophy driven downstream of the endothelin (ET-1A) receptor and cAMP signaling by βARs (Zhang et al., 2011; Zhang et al., 2013). PLCε is poised as a nexus for multiple receptor signaling systems due to its diversity of upstream regulators including heterotrimeric G protein βγ subunits and small GTPases, including Rap, Rho and Ras, and cAMP via the Rap guanine nucleotide exchange factor (GEF), Epac (Kelley et al., 2004; Malik et al., 2015; Kelley et al., 2001; Harden and Sondek, 2006; Smrcka et al., 2012). PLCε and Epac are scaffolded together at the nuclear envelope in cardiac myocytes by the hypertrophic organizer, muscle-specific A kinase anchoring protein (mAKAPβ), in close proximity to the Golgi apparatus (Dodge-Kafka et al., 2005). Activation of PLCε in cardiac myocytes induces hydrolysis of phosphatidylinositol-4-phosphate (PI4P) at the Golgi apparatus leading to release of diacylglycerol (DAG) and inactive IP2 in the vicinity of the nucleus to facilitate nuclear protein kinase D (PKD) and subsequent downstream hypertrophic pathways (Zhang et al., 2013; Vega et al., 2004). In previous work we found that either the Epac-selective cAMP analog, cpTOME, or forskolin stimulation of adenylate cyclase, activate PLCε-dependent Golgi PI4P hydrolysis in cardiac myocytes. Surprisingly, the βAR agonist isoproterenol (Iso), does not stimulate Golgi PI4P hydrolysis despite strongly stimulating cAMP production (Nash et al., 2018). As an explanation for this apparent paradox we demonstrated PLCε-dependent PI4P hydrolysis can be controlled by two distinct pools of cAMP delimited by distinct PDE isoforms. One pool, limited by PDE3 acts through Epac and Rap1 to activate PLCε, while a second pool controlled by PDE2 and/or PDE9A inhibits PLCε activity through activation of PKA (Nash et al., 2018). cAMP generated downstream of Iso cannot access the Epac/mAKAPβ/PLCε complex unless PDE3 is specifically inhibited, partially explaining the lack of activation of Golgi PI4P hydrolysis by Iso. A new paradigm for generating localized cAMP signals in cells has been established by a number of recent studies indicating that GPCRs in different intracellular compartments have discrete signaling outputs (Vilardaga et al., 2014; Irannejad et al., 2017; Irannejad et al., 2013; Irannejad and von Zastrow, 2014; Stoeber et al., 2018; Calebiro et al., 2009). An emergent idea is that βARs internalized into endosomes continue to stimulate cAMP accumulation resulting in cellular outcomes distinct from those generated by βARs at the cell surface. Irannejad et al. (2017) established that β1ARs expressed at the Golgi apparatus in HeLa cells can induce G protein activation in the Golgi when stimulated by either a cell permeant agonist, dobutamine, through passive diffusion, or the physiological hormone epinephrine, when transported across the cell membrane by the organic cation transporter, Oct3. These ideas led us to hypothesize that in cardiac myocytes, βARs resident at the Golgi apparatus could generate cAMP that has privileged access to the Epac/mAKAPβ/PLCε scaffold at the nuclear envelope/Golgi interface, thereby yielding a set of signals divergent from those generated by cell surface βARs. We demonstrate that endogenous β1ARs at the Golgi apparatus in cardiac myocytes are required to stimulate hypertrophic Epac/PLCε-dependent PI4P hydrolysis at the Golgi, and that these intracellular βARs can be accessed by physiological neurotransmitters, and synthetic β-blockers and agonists. These data present a potential new paradigm for drug development for treating heart failure through deliberate targeting of internal βARs in cardiac myocytes. Results Dobutamine induces β1-AR activation and PI4P hydrolysis at the Golgi In previous studies we showed that cAMP produced upon stimulation with the βAR agonist, Iso, does not induce PI4P hydrolysis at the Golgi in NRVMs without addition of a PDE3 inhibitor (Nash et al., 2018). We hypothesized that intracellular βARs may be required to generate a specific pool of cAMP with privileged access to the Epac/mAKAPβ/PLCε complex, and that the cell impermeant agonist, Iso, is unable to access these internal receptors. To test this hypothesis, we utilized the membrane permeant β1AR-agonist, dobutamine, either alone, or in combination with the βAR antagonists, metoprolol and sotalol, which are membrane-permeant and -impermeant, respectively (Irannejad et al., 2017). NRVMs were transduced with an adenovirus expressing the PI4P biosensor, FAPP-PH-GFP and stimulated with dobutamine (100 nM) or PBS. In contrast to Iso, dobutamine stimulated rapid and sustained PI4P hydrolysis (Figure 1A, left). Addition of metoprolol, but not sotalol, blocked dobutamine-stimulated PI4P hydrolysis (Figure 1A middle and right panels). This indicates that stimulation of an intracellular population βARs is required for dobutamine-mediated PI4P hydrolysis. To visualize βAR activation at the Golgi apparatus, NRVMs were transfected with translocatable Venus-based sensor of Gs coupled receptor activation, NES-Venus-mini-Gs (Wan et al., 2018). When a Gs coupled receptor is activated, NES-Venus-mini-Gs binds to the activated receptor, which is then visualized as translocation from the cytoplasm to membranes where the activated receptor is located, providing information on receptor activation in subcellular compartments. In unstimulated cells NES-Venus-mini-Gs is distributed throughout the cytoplasm. We did not detect translocation of NES-Venus-mini-Gs to endogenously expressed receptors (data not shown), although, as has been previously reported, we were able to detect staining of endogenous β1ARs at the Golgi in cardiac myocytes (Figure 1—figure supplement 1), so NRVMs were cotransfected with FLAG-β1AR. Addition of dobutamine to cells co-expressing NES-Venus-mini-Gs and FLAG-β1AR, caused a rapid clearance of cytoplasmic fluorescence and accumulated fluorescence at the PM and punctate structures, possibly corresponding to microtubules. This was followed by slower translocation of Venus associated fluorescence to the perinuclear region of the cell colocalizing with the CFP-giantin, a marker for the Golgi apparatus (Figure 1B,D and E and Figure 1—video 1). Iso also caused rapid translocation to the PM but, in contrast to dobutamine, no accumulation at the perinuclear region was observed (Figure 1C,D and E and Figure 1—video 2). To further confirm association of NES-Venus-mini-Gs with the Golgi, cells were treated with dobutamine for 8 min to cause NES-Venus-mini-Gs translocation to the perinuclear region, followed by treatment with Brefeldin A. Brefeldin A treatment significantly reversed NES-Venus-mini-Gs association with the perinuclear region, confirming Golgi localization of NES-Venus-mini-Gs (Figure 1—figure supplement 2 and Figure 1—video 3). Taken together, these data support the idea that a population of βARs present at the Golgi can be activated by exogenous agonists and stimulate Gs, ultimately resulting in PLC activation and PI4P hydrolysis. Figure 1 with 5 supplements see all Download asset Open asset Dobutamine induces PI4P hydrolysis through the activation of internal βARs. (A) NRVMs were transduced with FAPP-PH-GFP and stimulated as indicated. Time lapse live cell microscopy was used to quantitate Golgi associated FAPP-PH-GFP fluorescence was quantitated as previously described (Zhang et al., 2013; Malik et al., 2015). NRVMs were stimulated with dobutamine alone (100 nM, left) (N = 9), dobutamine in the presence or absence of metoprolol (100 μM, center) (N = 4), or dobutamine in the presence or absence of sotalol (5 mM, right) (N = 8). Data are not significantly different between dobutamine and dobutamine + sotalol. Data are from at least 4 cells for each N. (B) NRVMs were transfected with β1-ARs and NES-Venus-mini-Gs, followed by viral transduction with CFP-Giantin. Representative confocal fluorescence images of dobutamine-mediated NES-Venus-mini-Gs recruitment (100 nM, B left), CFP-Giantin Golgi marker (B, center) and histogram of representative NES-Venus-mini-Gs recruitment (B, right). Red arrow = Perinuclear region, Black arrow = sarcolemma. The yellow line indicates where histogram data was captured. Scale bars are 10 μm. (C) Representative confocal fluorescence images of Iso-mediated NES-Venus-mini-Gs recruitment (100 nM, left), CFP-Giantin Golgi marker (B, center) and histogram of representative NES-Venus-mini-Gs recruitment. Yellow line indicates where histogram data was captured. Scale bars = 10 μm (D) Mean data of fluorescence intensity of NES-Venus-mini-Gs at the perinuclear region corresponding to the Golgi ± SEM from at least 5 cells. (E) Pearson's correlation coefficient for overlap of YFP-Mini-Gs and CFP-Giantin images. All symbols on time course graphs are presented as mean ± standard error from N = 4 or more independent preparations of myocytes. Agonists were added where indicated by the arrow. https://doi.org/10.7554/eLife.48167.002 Figure 1—source data 1 PI4P hydrolysis is stimulated by dobutamine and inhibited by a cell permeable antagonist. https://doi.org/10.7554/eLife.48167.005 Download elife-48167-fig1-data1-v2.zip Figure 1—source data 2 Dobutamine but not Iso stimulates Mini Gs translocation. https://doi.org/10.7554/eLife.48167.006 Download elife-48167-fig1-data2-v2.zip PI4P hydrolysis by Dobutamine requires Golgi-localized PLCε and Epac Previous data from our laboratory demonstrated that the Epac/mAKAPβ/PLCε complex is responsible for cAMP-mediated PI4P hydrolysis at the Golgi in NRVMs (Zhang et al., 2013; Nash et al., 2018). To determine if this complex is responsible for dobutamine stimulated PI4P hydrolysis we utilized adenoviral shRNA to deplete PLCε, or adenoviral transduction of the RA1 domain of PLCε, which competes for the interaction between PLCε and mAKAPβ, disrupting the mAKAPβ-PLCε complex (Zhang et al., 2011). PLCε (PLCE1) shRNA depletion of PLCε completely inhibited PI4P hydrolysis stimulated by dobutamine, while negative control (NC) scrambled shRNA had no effect (Figure 2A). Viral expression of the PLCε RA1 domain also inhibited PI4P hydrolysis stimulated by dobutamine (Figure 2B). This demonstrates that mAKAPβ- scaffolded PLCε is required for dobutamine-mediated PI4P hydrolysis. The Epac-selective inhibitor HJC0726 inhibited (Zhu et al., 2015) PI4P hydrolysis stimulated by dobutamine (Figure 2C), however, the βγ inhibitor Gallein had no effect (Figure 2D). Taken together, these data indicate that Golgi PI4P hydrolysis stimulated by intracellular GPCRs requires the Epac/mAKAPβ/PLCε complex. Figure 2 Download asset Open asset Dobutamine-mediated PI4P hydrolysis requires Epac and mAKAPβ bound PLCε. NRVMs were transduced with FAPP-PH-GFP, stimulated with treatments as indicated and Golgi associate fluorescence was monitored with time. (A) PLCε knockdown prevents dobutamine-mediated PI4P hydrolysis. NRVMs were transduced for 48 hr with adenovirus expressing shRNA for PLCε or scrambled control shRNA before stimulation with 100 nM dobutamine (at arrow) (N = 3 independent experiments). (B) Disruption of PLCε-mAKAPβ interaction prevents PI4P hydrolysis stimulated by dobutamine. NRVMs were transduced with RA1 domain expressing adenovirus or control virus (NC-negative control) as previously described 24 hr before experimentation. Cells were stimulated with 100 nM dobutamine (at arrow) (N = 3 independent experiments). (C) Epac is required for dob-mediated PI4P hydrolysis. The Epac inhibitor HJC0726 (1 μM) was added to NRVMs 15 min before imaging and dobutamine (100 nM) was added at the arrow. (N = 3 independent experiments) (D) Gβγ is not required for dobutamine-mediated PI4P hydrolysis. The Gβγ inhibitor, Gallein (10 μM) was added 15 min prior to imaging and dobutamine added as indicated by the arrow (N = 3 independent experiments). Data are not significant between Dob and Dob + Gallein. Images for A, B,C and D PI4P hydrolysis were taken from at least 4 cells for each separate preparation of NRVMs. (E) Western blot of Adenoviral sh-RNA knockdown of PLCε. The two bands likely represent two splice variants of PLCε. https://doi.org/10.7554/eLife.48167.010 Figure 2—source data 1 Effects of inhibition of PLCε, Epac and Gβγ on dobutamine stimulated PI4P hydrolysis. https://doi.org/10.7554/eLife.48167.011 Download elife-48167-fig2-data1-v2.zip Inhibition of Golgi βARs prevents PI4P hydrolysis stimulated by dobutamine The data presented so far suggest that dobutamine acts through βARs at the Golgi apparatus to stimulate Golgi PI4P hydrolysis. To more directly demonstrate a requirement for endogenous βAR activation in the Golgi we targeted the βAR selective nanobody, Nb80, to the Golgi apparatus using the rapamycin inducible FRB-FKBP system. This approach has been previously used to inhibit βARs at the Golgi in Hela cells (Irannejad et al., 2017). NRVMs were transduced with virus containing CFP-Nb80-FRB and FKBP-mApple-GalT protein for Golgi targeting, along with a virus expressing GFP-FAPP-PH. Cells were selected that were positive for mApple, CFP and GFP fluorescence. Upon addition of rapamycin, translocation of CFP-Nb80-FRB to the Golgi apparatus was observed (Figure 3A). In cells pretreated with vehicle control, dobutamine stimulated PI4P hydrolysis. However, following addition of 1 μM Rapamycin, dobutamine-induced PI4P hydrolysis was significantly inhibited (Figure 3B). These data confirm that βARs expressed directly on the Golgi are required for dobutamine-mediated PI4P hydrolysis at the Golgi. Figure 3 Download asset Open asset β1ARs in the Golgi are required for PI4P hydrolysis stimulated by dobutamine. NRVMs were transduced with FRB-CFP-Nb80 and FKBP-GalT-mApple containing adenovirus, along with adenovirus containing FAPP-PH-GFP for 24 hr. (A) Confocal fluorescence images of NRVMs expressing FRB-CFP-Nb80 in the CFP channel and FKBP-GalT-mApple in the red channel before and after addition of rapamycin (see Materials and methods for details) Pearson's correlation coefficient = 0.32 ± 0.03. Scale bars are 10 μm. (B) NRVMs were incubated with either rapamycin (1 μM) or DMSO control for 15 min prior to addition of dobutamine (100 nM) added at the arrow. Golgi associated FAPP-PH-GFP fluorescence was monitored by time lapse fluorescence video microscopy as in Figure 1A. (C) Cells were pretreated treated with the cell permeant β1AR selective antagonist, CGP-20712 (100 nM), or vehicle, followed by dobutamine addition and assessed as in B. Images were taken from at least n = 9 cells each from at least four separate preparations of NRVMs. Data were analyzed as means from N = 4 experiments. Agonists were added as indicated by the arrow. https://doi.org/10.7554/eLife.48167.012 Figure 3—source data 1 Effects of Golgi targeted NB80 and β1AR-specific antagonism on dobutamine stimulated PI4P hydrolysis. https://doi.org/10.7554/eLife.48167.013 Download elife-48167-fig3-data1-v2.zip Nb80 can bind to and block both β1 and β2ARs, and while dobutamine is relatively selective for β1ARs over β2ARs, it can still potentially stimulate β2ARs. To confirm that dobutamine stimulated PI4P hydrolysis is through β1ARs, cells were treated with the cell permeant, highly selective, β1AR antagonist CGP-20172 (100 nM) followed by stimulation with dobutamine. CGP-20172 completely blocked dobutamine-stimulated PI4P hydrolysis (Figure 3C), indicating that the dobutamine stimulation of Golgi localized PI4P hydrolysis is through Golgi localized β1ARs. Norepinephrine (NE) induces PI4P hydrolysis and βAR activation at internal membranes in cardiac myocytes Thus far we have shown that the synthetic β1AR selective agonist, dobutamine, can enter cells, activate intracellular βARs, and stimulate intracellular PLC activity which has interesting pharmacological implications but may not be relevant to physiological cardiac regulation. For this reason, we determined if the sympathetic neurotransmitter norepinephrine (NE) could stimulate PI4P hydrolysis and βAR activation at the Golgi in NRVMs. Cells transduced with adenovirus containing FAPP-PH-GFP were stimulated with NE (10 μM) at 37°C. NE caused delayed, but robust PI4P hydrolysis (Figure 4A, left). PI4P hydrolysis initiated by NE was blocked by metoprolol (100 μM, Figure 4A, center) but not sotalol (5 mM, Figure 4A, right) indicating that NE can stimulate internal βARs. To further determine if NE could enter cells and activate βARs at the Golgi, we monitored NES-Venus-mini-Gs recruitment to the Golgi in response to NE. NE initially stimulated NES-Venus-mini-Gs translocation to the plasma membrane as was observed for dobutamine, followed by slower, NES-Venus-mini-Gs translocation to the Golgi (Figure 4B and Figure 4—video 1). Since PM and Golgi recruitment are difficult to see in the same confocal plane for Figure 4—video 2 a separate cell was imaged in a confocal plane that emphasizes Golgi recruitment of Venus-mini-Gs. Thus, NE enters cardiac myocytes, accesses internal βARs, and stimulates PI4P hydrolysis. Figure 4 with 2 supplements see all Download asset Open asset The physiological neurotransmitter, norepinephrine, can induce PI4P hydrolysis through internal receptors and can activate βARs at the Golgi. (A) NRVMs were transduced with FAPP-PH-GFP and stimulated with norepinephrine (10 μM, left) in the presence of metoprolol (100 μM, center) or sotalol (5 mM, right) and analyzed as in Figure 1A. Data are not significant between norepinephrine and norepinephrine + sotalol. (B) NRVMs were transfected with β1AR and NES-Venus-mini-Gs, followed by viral transduction with CFP-Giantin. Representative images of norepinephrine-mediated NES-. Venus-mini-Gs recruitment (10 μM, left columns), CFP-Giantin Golgi marker (B, right). The Pearson's Coefficient for colocalization of Venus-mini-Gs and CFP-Giantin is R = 0.455 ± 0.06. Scale bars are 10 μm. All experiments were performed in humidified environmental chamber at 37°C. Images for PI4P hydrolysis were collected as in Figure 1A, and were from at least n = 7 cells each from three separate preparations of NRVMs. Data were analyzed as means from N = 3 experiments. Agonists were added where indicated by the arrow. https://doi.org/10.7554/eLife.48167.014 Figure 4—source data 1 NE stimulates PI4P hydrolysis. https://doi.org/10.7554/eLife.48167.015 Download elife-48167-fig4-data1-v2.zip Inhibition of the membrane cation transporter, OCT3, prevents norepinephrine induced PI4P hydrolysis Data from other laboratories has demonstrated that transport of norepinephrine or epinephrine into cells by organic cation transporter family of proteins (OCT proteins) is required for activation of internal receptors in adult cardiac myocytes (Wright et al., 2008) and HELA cells (Irannejad et al., 2017). The non-selective cation transporter, OCT3 has been shown to be responsible for this transport, therefore we utilized three structurally distinct inhibitors of OCT3, corticosterone (100 μM, Figure 5A), abacavir (10 μM, Figure 5B) and lamotrigine (10 μM, Figure 5C) (Dickens et al., 2018) to determine if OCT3 is required for NE-stimulated intracellular PI4P hydrolysis. Preincubation of NRVMs with any of these inhibitors prevented NE from inducing PI4P hydrolysis, suggesting that OCT3-mediated transport is required for stimulation of PI4P hydrolysisby this catecholamine. In addition, we sought to determine if receptor internalization is required for NE-mediated PI4P hydrolysis and βAR activation at the Golgi. An inhibitor of dynamin-mediated receptor internalization, Dyngo (40 μM), had no effect on either PI4P hydrolysis (Figure 5D) or recruitment of NES-Venus-mini-Gs to the Golgi apparatus (Figure 5E and Figure 5—video 1) by NE. As a positive control, Dyngo does prevent internalization of FLAG-β1AR at the cell surface (Figure 5—figure supplement 1A). Additionally, corticosterone (100 μM) did not alter cAMP production from cell surface βARs (Figure 5—figure supplement 1B). Corticosterone did not block dobutamine-mediated PI4P hydrolysis but rather enhanced the rate of onset relative to dobutamine alone (Figure 5—figure supplement 1C). The reason for this enhancement is unclear but the differential effects of corticosterone on dobutamine vs NE are consistent with blockade of NE entry into the cell, but not dobutamine. Together, these data indicate that transport by OCT transporters, not receptor internalization, is required for NE-mediated Golgi βAR activation and PI4P hydrolysis. Figure 5 with 2 supplements see all Download asset Open asset Norepinephrine requires OCT transporters but not receptor internalization to stimulated PI4P hydrolysis. NRVMs were transduced with FAPP-PH-GFP and stimulated with norepinephrine (10 μM) in the presence of corticosterone (100 μM, (A), abacavir (10 μM, (B), lamotrigine (10 μM, (C) or Dyngo (40 μM, (D). Data are not significant between norepinephrine and norepinephrine + Dyngo. Images were collected as in Figure 1A for PI4P hydrolysis (N = 3) were from at least 4 cells each from separate preparations of NRVMs. Agonists were added where indicated by the arrow. (E) Dyngo has no effect on NES-Venus-mini-Gs Golgi recruitment by norepinephrine. Representative image of norepinephrine-mediated NES-Venus-mini-Gs recruitment in the presence of Dyngo (40 μM). (N = 3) Scale bars are 10 μm. (F) NRVMs were transduced with adenovirus containing FRB-CFP-Nb80 and FKBP-GalT-mApple along with adenovirus containing FAPP-PH-GFP for 24 hr prior to experimentation. NRVMs were incubated with either rapamycin (1 μM) or DMSO control for 15 min prior to addition of NE (10 μM) added at the arrow. Images were collected as in Figure 1A for PI4P hydrolysis and were from at least n = 10 cells each from three separate preparations of NRVMs. Data were analyzed as means from N = 3 experiments. https://doi.org/10.7554/eLife.48167.018 Figure 5—source data 1 NE requires membrane transport by OCT3 and Golgi resident βARs to stimulate PI4P hydrolysis. https://doi.org/10.7554/eLife.48167.020 Download elife-48167-fig5-data1-v2.zip Additionally, NRVMs were transduced with viruses containing FRB-CFP-Nb80 and FKBP-mApple-GalT, along with FAPP-PH-GFP to determine if Golgi β1AR were required for PI4P hydrolysis to NE. In cells pretreated with vehicle, NE induces PI4P hydrolysis. However, following addition of 1 μM Rapamycin, NE-induced PI4P hydrolysis was significantly inhibited (Figure 5F). To confirm that dobutamine and NE stimulated Golgi PI4P hydrolysis is not unique to neonatal myocytes we tested the ability of dobutamine and NE to stimulate PI4P hydrolysis in acutely isolated murine adult ventricular myocytes. Infection of AVMs for 24 hr with FAPP-PH-GFP labels the Golgi surrounding the nucleus as well as other structures within the myocyte as we have previously reported (Figure 6A) (Malik et al., 2015). Stimulation of AVMs with dobutamine (Figure 6B) or NE (Figure 6C) stimulated PI4P depletion in the Golgi. NE-stimulated PI4P hydrolysis was blocked by preincubation with the Oct3 inhibitor abacavir (Figure 6C) indicating that NE transport into AVMs is required for stimulation of PI4P hydrolysis, as was seen in NRVMs. These data indicate that internal receptors are required for stimulation of PI4P hydrolysis in AVMs as well as NRVMs. Figure 6 Download asset Open asset Dobutamine and NE stimulate PI4P hydrolysis in Adult Ventricular Myocytes (AVMs). Freshly isolated mouse AVMs were infected with FAPP-PH-GFP (100 MOI) for 24 hr. (A) representative image of an AVM expressing FAPP-PH-GFP showing strong labeling surrounding the nucleus. (B) FAPP-PH-GFP expressing AVMs were treated with either dob or PBS control as indicated and were imaged as for NRVMs as in Figure 1A for PI4P hydrolysis. Fluorescence intensity in the region surrounding the nucleus was quantitated at 5 min intervals. C) AVMs were treated with PBS control, NE (10 μM) or NE plus abacavir and analyzed as in B. Data for B and C are from at least n = 3 cells each from three separate preparations of AVMs. Data were analyzed as means from N = 3 experiments. https://doi.org/10.7554/eLife.48167.022 Figure 6—source data 1 NE-Stimulated PI4P hydrolysis in adult cardiac myocytes requires OCT3. https://doi.org/10.7554/eLife.48167.023 Download elife-48167-fig6-data1-v2.zip Cardiac hypertrophy induced by dobutamine is more effectively inhibited by a cell permeant antagonist To determine if signaling by internal βARs is important for promotion of cellular hypertrophy, NRVMs were treated with dobutamine (100 nM) for 48 hr and two measures of hypertrophy were assessed, cell area and ANF expression. Dobutamine stimulated an increase cell area (Figure 7A), and ANF expression (Figure 7B), after 48 hr of stimulation. Co-incubation with the cell permeant antagonist, metoprolol, strongly inhibited these hypertrophic responses. The cell impermeant antagonist, sotalol, on the other hand was significantly less effective at blocking these hypertrophy measures than metoprolol, (Figure 7A and B). These data, taken together with PI4P hydrolysis data, suggest that internal βARs are involved in mediating cardiomyocyte hypertrophy. Figure 7 Download asset Open asset Dobutamine induced cardiomyocyte hypertrophy requires intracellular βARs. (A) Dobutamine induces internal-receptor dependent increases in cell area. NRVMs were transduced with YFP virus prior to stimulation for 48 hr with dobutamine in the presence of the indicated antagonists or vehicle control. Following fixation, cell area was measured using image J. Representative images are on the left with mean data (fold over basal, FOB) on the right (N = 4 separate preparations of NRVMs), analyzed with a paired two way ANOVA. ****p<0.0001 vs. Dob and Dob+sotalol. Scale bars are 10 μm. B) Dobutamine induces an increase in ANF expression via an internal receptor-dependent mechanism. NRVMs were stimulated with dobutamine in the presence or absence of the indicated antagonists or vehicle control for 48 hr before fixation and staining for ANF. Fluorescence of ANF rings was then captured by confocal microscopy, followed by fluorescence intensity analysis with Image J. Representative images are on the left with mean data (fold over basal, FOB) on the right (N = 3 separate preparations of NRVMs), analyzed with a paired two way ANOVA. **p=0.007 vs. Dob and Dob+sotalol. The total number of cells analyzed was greater than 200 cells for each. https://doi.org/10.
Acidic vesicles and organelles play fundamental roles in a broad range of cellular events such as endocytosis, lysosomal degradation, synaptic transmission, pathogen fate, and drug delivery. Fluorescent reporters will be invaluable for studying these complex and multifunctional systems with spatiotemporal resolution, yet common fluorescent proteins are generally nonfluorescent at acidic conditions due to the decrease of anionic chromophores upon protonation, but are fluorescent at physiological pH, creating interfering fluorescence from nonvesicle regions. Here we developed a novel acid-brightening fluorescent protein (abFP) that fluoresces strongly at acidic pH but is nonfluorescent at or above neutral pH, boasting a pH profile opposite to that of common fluorescent proteins. Through expansion of the genetic code, we incorporated a quinoline-containing amino acid Qui into the chromophore of EGFP to reverse the chromophore charge. Protonation of Qui rendered a cationic chromophore, which resulted in unique fluorescence increase only at acidic pH in vitro, in E. coli cells, and on the mammalian cell surface. We further demonstrated that abFP-tagged δ opioid receptors were fluorescently imaged in lysosome showing distinct features and without background fluorescence from other cellular regions, whereas EGFP-tagged receptors were invisible in lysosome. This Qui-rendered cationic chromophore strategy may be generally applied to other fluorescent proteins to generate a palette of colors for acidic imaging with minimal background, and these abFPs should facilitate the study of molecules in association with various acidic vesicles and organelles in different cells and model organisms.
