Soluble Adenylyl Cyclase Reveals the Significance of cAMP Compartmentation on Pulmonary Microvascular Endothelial Cell Barrier
http://www.100md.com
《循环研究杂志》
the Center for Lung Biology (S.S., M.A., T.S.) and Departments of Pharmacology (S.S., T.S.), Cell Biology and Neuroscience (M.A.), University of South Alabama College of Medicine, Mobile
Integrative Biology and Pharmacology (C.W.D.), University of Texas, Houston.
Abstract
Subtle elevations in cAMP localized to the plasma membrane intensely strengthen endothelial barrier function. Paradoxically, pathogenic bacteria insert adenylyl cyclases (ACs) into eukaryotic cells generating a time-dependent cytosolic cAMP-increase that disrupts rather than strengthens the endothelial barrier. These findings bring into question whether membrane versus cytosolic AC activity dominates in control of cell adhesion. To address this problem, a mammalian forskolin-sensitive soluble AC (sACI/II) was expressed in pulmonary microvascular endothelial cells. Forskolin stimulated this sACI/II construct generating a small cytosolic cAMP-pool that was not regulated by phosphodiesterases or Gs. Whereas forskolin simultaneously activated the sACI/II construct and endogenous transmembrane ACs, the modest sACI/II activity overwhelmed the barrier protective effects of plasma membrane activity to induce endothelial gap formation. Retargeting sACI/II to the plasma membrane retained AC activity but protected the endothelial cell barrier. These findings demonstrate for the first time that the intracellular location of cAMP synthesis critically determines its physiological outcome.
Key Words: phosphodiesterase signal transduction ExoY second messenger Pseudomonas aeruginosa
Introduction
The discovery of an adenine ribonucleotide (cAMP) as an intracellular signaling messenger, and the recognition of its highly hydrophilic nature, led to the initial idea that this molecule rapidly and equivalently accesses all cellular compartments. However, as studies advanced, it became unclear how different external stimuli, which similarly elevate cAMP, selectively control different physiological processes.1,2 These and related studies have given rise to the evolving hypothesis that cAMP signaling occurs within spatially and temporally confined microdomains.2eC5
Whole cell cAMP concentration is established by the activities of ACs that synthesize cAMP and phosphodiesterases that hydrolyze it. AC1 to -9 are all transmembrane proteins,6,7 and therefore cAMP synthesis emanates from the plasma membrane. Phosphodiesterases are distributed discretely throughout the cell where they inactivate cAMP before it escapes into physiologically inappropriate domains. Indeed, the coordinated activities of ACs and phosphodiesterases are responsible for generating cAMP gradients within a single cell, where high concentrations arise at the plasma membrane and lower concentrations exist in the bulk cytosol.8eC12 Yet it is unclear how cAMP is targeted to its physiologically appropriate effectors. Hall and colleagues have proposed that molecules of the signaling cascade may be orientated within a molecular complex such that cAMP is steered directly to its relevant effectors.13 Thus, the prevailing theory is that signaling fidelity is derived from spatial and temporal cAMP transitions generated within functional subcellular compartments where downstream targets are similarly located.2,3,14eC16
These paradigms presume that cAMP synthesis occurs exclusively at the plasma membrane. However, recent evidence suggests human soluble AC (hsAC) exists in a wide range of cells and tissues,17eC20 where it is tethered to mitochondria, nucleus, microtubules, and the microtubule organizing center,17 generating localized cAMP in response to bicarbonate fluctuations.20 This finding brings into question the role hsAC plays in the activation of downstream cytosolic cAMP targets. Although many cAMP effectors (eg, protein kinase A [PKA] and Epac) localize to nonplasma membrane cellular compartments,21,22 the current paradigm is that these targets are activated by cAMP that is generated at the plasma membrane, paradoxical to emerging evidence that free cAMP diffusion is quite limited. Currently, the physiological significance of cytosolic cAMP synthesis is not known.
Opposing actions of membrane versus cytosolic cAMP synthesis have been previously described in endothelial cells,23 providing evidence for physiologically relevant cAMP compartments. Pulmonary microvascular endothelial cells (PMVECs) predominantly express the calcium inhibited AC6.24,25 Calcium inhibition of AC6 cannot easily be resolved by standard whole cell cAMP measurements in PMVECs, yet 90% of its membrane activity is inhibited by submicromolar calcium concentrations.26 Heterologous expression of calcium stimulated AC8 reverses the calcium inhibition of AC6 but only modestly increases global cAMP concentrations. Despite such subtle shifts in the calcium regulation of cAMP, reversing the calcium inhibition to a stimulation of AC prevents thrombin from inducing interendothelial cells gaps.27 Thus, a membrane delimited cAMP pool that is not reflected by global cAMP concentrations has powerful barrier enhancing properties. In stark contrast to this membrane delimited cAMP pool, P aeruginosa transfers a soluble AC (eg, ExoY) into the cytosol of PMVECs where it exhibits a time-dependent increase in AC activity that is restricted exclusively to the cytosol.23 This cytosolic cAMP induces intercellular gaps in cultured PMVEC monolayers and increases the filtration coefficient of the isolated perfused lung.23 These studies suggest that the disparate compartmentation of cAMP synthesis, eg, plasma membrane versus cytosol, exerts exact opposite effects on endothelial barrier strength, although the ExoY-dependent cAMP rise occurs progressively over 4 hours without a concomitant increase in membrane cAMP. Thus, the goal of this study was to determine whether simultaneous rises in plasma membrane versus cytosolic cAMP dominates in the regulation of endothelial barrier strength. Our findings indicate that when cAMP is simultaneously generated at the plasma membrane and within the cytosol, cAMP synthesis in the cytosolic compartment overwhelms the barrier protective properties of membrane-restricted cAMP to induce endothelial cell gaps, demonstrating for the first time that the subcellular localization of signal generation is an important determinant of its physiological outcome.
Materials and Methods
Isolation and Culture of PMVECs
PMVECs were isolated, cultured, and passaged in EC media (DMEM+10% FBS+1% PBS) as described.25
Adenovirus Constructs and Infection
Adenovirus encoding soluble ACI/II28 and green fluorescence protein (GFP) (Ad 88929) have been described elsewhere. Adenovirus was added to 70% confluent PMVECs in EC media at multiplicity of infection (moi) 10:1 for 40 to 48 hours.
Retrovirus Constructs and Infection
The TM-AC8 fusion described previously,30 was inserted between HindIII and XhoI sites of a hygromycin-encoding retroviral vector pBABA-Hygro derivative. The sACI/II terminal codon was removed and BamHI sites attached. This was inserted between the N-terminal transmembrane domains of TM-AC8 to generate sACI/II-TM-AC8, and subcloned into the pBABE-Hygro derivative. Retrovirus media was added to 40% confluent PMVECs in 5.9 e/L polybrene, centrifuged for 1 hour at 750g at room temperature. Retrovirus media remained on the cells for 15 hours and was then replaced with EC media. Retroviral cell lines were passaged and plated for each experiment into hygromycin selection media (50 e/mL).
Whole Cell cAMP Measurements
cAMP concentration was assessed by standard radioimmunoassay (Biomedical Technologies, Stoughton, Mass). Forskolin (100 eol/L) or isoproterenol (1 eol/L) with or without rolipram (10 eol/L) was added as indicated.
Fractionation of Cellular Lysates
Confluent cells were rinsed in ice-cold buffer A (137 mmol/L NaCl, 2.7 mmol/L KCl, 10.14 mmol/L Na2HPO4, 10.76 mmol/L KH2PO4) followed by 10-minute incubation in ice-cold buffer B (40 mmol/L Tris-HCl, 1 mmol/L EDTA, 150 mmol/L NaCl). Cells were scraped using 10 mL ice-cold buffer C (20 mmol/L HEPES, pH 7.5, 5 mmol/L EDTA, 1 mmol/L EGTA, 2 mmol/L dithiothreitol, 200 mmol/L sucrose, protease inhibitor cocktail) and centrifuged for 10 minutes at 250g at 4°C. Supernatant 1 (media) was removed and the pellet resuspended in 2 mL buffer C using 20 strokes each of course and fine pestle with Dounce homogenizer. This suspension was centrifuged at 100 000g for 20 minutes at 4°C. Supernatant 2 (cytosolic fraction) was removed and the pellet (plasma membrane fraction) resuspended in 1.5 mL buffer C using 15 strokes of Duall homogenization. Fractions were stored at eC80°C. Protein concentration was determined using the modified Lowry method (Pierce).
Adenylyl Cyclase Activity Assay
Assays were performed using a 1-mL solution of 0.1 mmol/L ATP, 1 eol/L cAMP, 2.5 U/mL creatine phosphokinase, 12 mmol/L phosphocreatine, 1 U/mL adenine deaminase, 0.04 mmol/L GTP, 2 mmol/L MgCl2, and 25 mmol/L HEPES at pH 7.3 with physiological potassium solution (103.1 mmol/L KOH, 10 mmol/L KHCO3, and 26.9 mmol/L K2HPO4) and 100 eol/L calcium (determined using Fabiato free ion calculator).31 Reactions contained [-32P]ATP (1x106 cpm) with or without 100 eol/L forskolin and initiated by addition of 0.15 mg of protein, proceeded for 30 minutes at 30°C and terminated using 200 e蘈 30% trichloroacetic acid. Recovery was monitored using internal standard [2,8-3H]cAMP (15 000 cpm). Production of radio-labeled cAMP from substrate [-32P]ATP was determined following isolation of [32P]cAMP by Dowex cation exchange then alumina oxide chromatography.32
Time-Lapse Microscopy
Images were acquired at 1-minute intervals for 2 hours as described previously.23 Forskolin (100 eol/L) was added after 60 minutes.
Statistical Analysis
Values represent means±SD. One- or 2-way ANOVA followed by Bonferroni post test or t test was used for comparisons. Significance was accepted as P<0.05.
Results
sACI/II Elevates cAMP in PMVECs
Previous studies demonstrate that plasma membrane AC activity is endothelial barrier protective, whereas cytosolic activity is barrier disruptive.23 To determine which of these two sites of synthesis dominates in the regulation of endothelial permeability, we used a soluble forskolin-sensitive AC system that was generated by fusing the amino-terminal portion of the catalytic loops of mammalian transmembrane AC type I (C1a) and II (C2a) with a 14 amino acid linker33 (sACI/II, Figure 1a). Loss of the C1b domain of ACI renders the chimera insensitive to calcium.28 Dessauer and Gilman generated an adenovirus for expression of this soluble enzyme in mammalian cells.28 When expressed in PMVECs, activation by forskolin generated a time-dependent increase in intracellular cAMP (Figure 1b). Interestingly, no increase in basal cAMP was observed before the addition of forskolin, which was attributable to the lack of constitutive enzymatic activity. To confirm the sACI/II adenoviral specificity of elevated cAMP, the forskolin response was compared with cells infected with a control soluble GFP adenovirus. Forskolin activates endogenous transmembrane ACs increasing intracellular cAMP in uninfected and GFP infected control cells (Figure 1b). Expression of sACI/II increased the forskolin-stimulated intracellular cAMP above control groups, which was attributed to simultaneous activation of both endogenous transmembrane ACs and the sACI/II chimera (Figure 1b).
Previous studies have shown P aeruginosa competent for the type three secretion system transfers the AC ExoY into PMVECs, generating a cytosolic cAMP pool that is sufficient to increase endothelial permeability.23 At a sACI/II adenoviral dose moi 10:1, the maximal forskolin-stimulated cAMP response mimicked concentrations achieved by ExoY expressing P aeruginosa (Figure 1c).
Functional Compartments of Forskolin-Induced sACI/II-cAMP
Although these studies demonstrate sACI/II has activity in PMVECs, they provide no indication of the intracellular compartment where this activity resides. Phosphodiesterase activity is important to tightly restrict cAMP diffusion, thereby maintaining isolated cAMP pools8,10,34 and downstream signaling specificity. PMVECs express the rolipram-sensitive phosphodiesterase isoform 4, which controls the plasmalemmal cAMP pool.25 We sought to compare phosphodiesterase 4 regulation of the sACI/II-cAMP versus endogenously generated plasmalemmal cAMP. The forskolin-sensitive cAMP pool of uninfected control PMVECs is augmented by rolipram (Figure 2a). The substantial increase in cAMP generated by simultaneous forskolin and rolipram treatment was not sufficient to deplete total cellular ATP pools (data not shown). However, despite the fact that sACI/II expression increased the amount of cAMP produced in response to forskolin, rolipram did not significantly augment the forskolin-stimulated sACI/II pool (Figure 2b). Whereas rolipram induced a 110% increase in the forskolin-stimulated cAMP pool of uninfected and GFP infected controls, rolipram induced only a 40% rise in the forskolin-stimulated cAMP pool of sACI/II expressing cells (Figure 2c). This relative insensitivity of the sACI/II-dependent cAMP pool to hydrolysis by phosphodiesterase 4 suggests sACI/II and endogenous AC activity reside in different compartments.
To further determine compartmentation of sACI/II, we examined its integration into endogenous signaling pathways. Rolipram [10 eol/L, EC10025], used throughout this study to amplify global cAMP levels for detection purposes, doubles the intracellular cAMP concentrations of all treatment groups (data not shown). Isoproterenol activation of -adrenergic receptors is coupled via the trimeric G protein, GS, to endogenous transmembrane ACs. In PMVECs, isoproterenol (1 eol/L, EC10025) increases intracellular cAMP above rolipram alone (Figure 3). Interestingly, although in vitro kinetic analyses have demonstrated that GS can stimulate sACI/II,28 activation of GS in PMVECs expressing the sACI/II does not increase cAMP above control groups (Figure 3). Collectively, these data suggest that sACI/II resides in a compartment not readily accessible by GS.
Subcellular Localization of the sAC Activity
To determine the subcellular localization of sACI/II activity in PMVECs, cells were lysed into plasma membrane and cytosolic fractions and the AC activity measured. Similar basal- and forskolin-stimulated AC activity was detected in the plasma membrane fraction of all treatment groups (uninfected control, sACI/II and GFP infected cells) representative of the endogenous transmembrane ACs (Figure 4a). The plasma membrane fraction of sACI/II expressing cells did not possess elevated forskolin-sensitive AC activity above controls, demonstrating the chimera did not have activity at the plasmalemma. As the predominant endogenous AC in PMVECs is calcium sensitive isoform 6,26 calcium inhibition of forskolin activity further supports the plasma membrane composition of this fraction (Figure 4b). Although this is not the 90% calcium inhibition of AC activity in the lipid- and caveolin-enriched fraction,26 it demonstrates the presence of a calcium regulated AC in this bulk plasma membrane fraction. Within the cytosolic fraction, no basal AC activity was detected in any treatment group (Figure 4c); however, forskolin revealed 6 times more AC activity in the cytosolic fraction of sACI/II expressing cells compared with uninfected or GFP infected cells. These data demonstrate the unique cytosolic compartmentalized AC activity of sACI/II infected PMVECs. Interestingly, the modest level of forskolin-sensitive cytosolic sACI/II activity was comparable to the "basal" plasma membrane activity, but significantly less than the forskolin-stimulated membrane activity (Figure 4d). Therefore, when sACI/II expressing cells are stimulated by forskolin, there is 4.5 times more AC activity at the plasma membrane than in the cytosol, but 6 times more activity in the cytosolic compartment compared with basal conditions.
Forskolin-Stimulated sACI/II Activity Induces PMVEC Gap Formation
We next sought to determine whether plasma membrane or cytosolic AC activity would dominate in the regulation of the endothelial barrier. Forskolin was not sufficient to induce PMVEC gaps with the adenoviral GFP control. Whereas unstimulated sACI/II expressing PMVECs did not reveal gap formation, gaps rapidly formed with the addition of forskolin, as seen by the retraction of cell-cell borders (Figure 5; Movies 1 and 2 in the online data supplement available at http://circres.ahajournals.org). Thus, activation of the sACI/II chimera, although representing only a modest rise in cytosolic versus membrane activity, is sufficient to overwhelm the barrier protective effects of endogenous plasma membrane restricted AC activity to form endothelial cell gaps.
Relocation of sACI/II to the Plasma Membrane
To further test the hypothesis that the subcellular localization of AC activity governs its barrier protective or disruptive properties, the sACI/II chimera was relocalized to the plasma membrane. Gu et al30 have demonstrated that YFP sandwiched between the 2 transmembrane domains of AC8 (TM-AC8) (Figure 6a, i and ii) is sufficient for plasma membrane targeting. Plasma membrane localization of TM-AC8 was also detected in PMVECs (Figure 6b). Insertion of sACI/II between the transmembrane domains of AC8 generated sACI/II-TM-AC8 (Figure 6a, iii). Retroviral expression of sACI/II-TM-AC8 also localized to the membrane in PMVEC (Figure 6c). Whereas plasma membrane fractions of PMVECs expressing sACI/II-TM-AC8 exhibited elevated forskolin-dependent AC activity above controls (Figure 6d), no AC activity was detected in the cytosolic fraction (data not shown), illustrating its exclusive plasmalemma targeting. Forskolin stimulation of sACI/II-TM-AC8eCexpressing PMVECs did not lead to endothelial gap formation (Figure 6e and online data supplement, Movie 3), further demonstrating that AC activity exclusively restricted to the plasma membrane compartment with corresponding lack of activity in the cytosol does not disrupt the endothelial barrier.
Discussion
Elevated cAMP is widely recognized to be endothelial barrier protective.35 More specifically, only small cAMP elevations critically localized to the plasma membrane are needed to protect the PMVEC barrier.27 However, nature provides a paradox. Pathogenic bacteria have developed a mechanism to synthesize cAMP in the cytosol of eukaryotic cells, resulting in decreased endothelial barrier function. For example, edema factor of Bacillus anthracis elevates intracellular cAMP in CHO cells yet induces edema when injected into the skin of rabbits and guinea pigs.36 In addition, P aeruginosa inserts the adenylyl cyclase, ExoY, into the cytosol of PMVECs generating a barrier disruptive cytosolic cAMP pool.23 As theories of cAMP compartmentation have evolved, it appears that within many cell types not all cAMP elevations produce equivalent physiological outcomes. However, the discrete role of AC activity within different cellular microdomains has never been directly tested owing to the lack of a suitable physiological outcome measurement. Using novel techniques to manipulate compartments of AC activity, our present findings demonstrate that cytosolic AC activity in PMVECs is sufficient to overwhelm the barrier protective effects of plasma membrane activity to promote endothelial gap formation.
The mammalian sACI/II chimera is a fusion of C1a-ACI with C2a-ACII that contains an AC catalytic site and forskolin binding-pocket, thereby generating a conditionally active, soluble enzyme. Our findings that sACI/II has forskolin-dependent cytosolic AC activity in PMVECs that is not activated by Gs signaling in vivo support previous work of Dessauer and Gilman28; indeed sACI/II could not be activated by Gs signaling in vivo unless the enzyme was stimulated with forskolin.28 In PMVECs, sACI/II-cAMP was insensitive to phosphodiesterase 4 activity, collectively suggesting AC activity resides in a unique cytosolic compartment. Although forskolin simultaneously activates both the cytosolic sACI/II and endogenous transmembrane AC6, the magnitude of sACI/II activity only contributes a small fraction of the global cAMP pool. Despite the small contribution of the cytosolic cAMP to this global cAMP-pool, it was sufficient to induce PMVEC gaps. Currently it is unclear whether a change in the plasma membrane-to-cytosolic cAMP gradient, or a small rise in cytosolic AC activity, is responsible for overwhelming the barrier protective effects of plasma membrane AC activity resulting in endothelial gap formation. However, we have recently shown that disrupting membrane phosphodiesterase 4 activity reveals forskolin-stimulated PMVEC gap formation suggesting that cAMP must remain within the membrane domain for it to confer barrier protection (T.S., unpublished data, 2006).
When sACI/II is sandwiched between the two transmembrane domains of AC8, this plasma membrane targeted sACI/II retained its forskolin-stimulated AC activity, indicating "free" amino and carboxy termini are not required for enzymatic activity. Indeed, this is the first evidence that the entire AC catalytic pocket can reside in the C1 loop and still retain activity. When stimulated by forskolin, this chimeric protein generated small increases in cAMP similar to the level of activity achieved by sACI/II free in the cytosol. However, restriction of this sACI/II-dependent cAMP to the plasma membrane did not induce PMVEC gaps, definitively illustrating the site of cAMP synthesis critically mediates the physiological outcome.
The downstream cytosolic effectors that determine such diverse physiological responses of membrane and cytosolic cAMP transitions are unknown. Interestingly, mammalian, bicarbonate-stimulated, soluble ACs localize to mitochondria, nucleus, microtubules, and centrosomes where evidence is emerging that complete signalsomes exist. cAMP targets, such as PKA and Epac, also reside within the cytosol, some in association with complete signal transduction cascades.37 For example, PKA compartmentalized through its interactions with A kinase anchoring proteins (AKAPs), is involved in phosphorylation events that lead to rearrangement of the cytoskeleton.38 PKA phosphorylates microtubule associating proteins such as MAP2 and tau, which in turn reduces their ability to stabilize microtubules.39,40 Within the endothelium, microtubule disassembly increases permeability.41 An unregulated cytosolic cAMP pool could destabilize such cytoskeleton structures necessary to induce gap formation. Indeed, cAMP-signaling cascades are already in place within the cytosol, such that atypical cytosolic AC activity could lead to their activation. Future studies are necessary to pursue the identification of these cytosolic targets.
Thus, the present data demonstrate that novel targeting of minimal AC activity to the cytosol of PMVECs dominates over substantial plasmalemmal AC activity to induce endothelial gap formation. Relocalizing this same AC construct to the plasma membrane substantiates endothelial barrier protective effects of plasma membrane AC activity. Our data presented here contribute to the idea that compartmentalized microdomains of cAMP signaling play a key physiological and pathophysiological role on the regulation of pulmonary endothelial barrier.
Acknowledgments
This work was supported by NIH grants HL-60024 and HL-66299 (to T.S.) and by American Heart Association Southeast Consortium Predoctoral Fellowship 0315134B (to S.L.S.). We thank Dermot Cooper for providing the TM-AC8 plasmid construct and Judy Creighton for performing ATP studies.
References
Hayes JS, Brunton LL, Brown JH, Reese JB, Mayer SE. Hormonally specific expression of cardiac protein kinase activity. Proc Natl Acad Sci U S A. 1979; 76: 1570eC1574.
Steinberg SF, Brunton LL. Compartmentation of G protein-coupled signaling pathways in cardiac myocytes. Annu Rev Pharmacol Toxicol. 2001; 41: 751eC773.
Brunton LL, Hayes JS, Mayer SE. Functional compartmentation of cyclic AMP and protein kinase in heart. Adv Cyclic Nucleotide Res. 1981; 14: 391eC397.
Beavo JA, Brunton LL. Cyclic nucleotide research-still expanding after half a century. Nat Rev Mol Cell Biol. 2002; 3: 710eC718.
Bundey RA, Insel PA. Discrete intracellular signaling domains of soluble adenylyl cyclase: camps of cAMP Sci STKE. 2004; 2004: pe19.
Ludwig MG, Seuwen K. Characterization of the human adenylyl cyclase gene family: cDNA, gene structure, and tissue distribution of the nine isoforms. J Recept Signal Transduct Res. 2002; 22: 79eC110.
Sayner SL, Stevens T. Adenylyl cyclase and cAMP regulation of the endothelial barrier. In: Patterson CE, ed. Perspectives on Lung Endothelial Barrier Function. 1st ed. Amsterdam: Elsevier; 2005: 105eC138.
Rich TC, Fagan KA, Nakata H, Schaack J, Cooper DM, Karpen JW. Cyclic nucleotide-gated channels colocalize with adenylyl cyclase in regions of restricted cAMP diffusion. J Gen Physiol. 2000; 116: 147eC161.
Jurevicius J, Skeberdis VA, Fischmeister R. Role of cyclic nucleotide phosphodiesterase isoforms in cAMP compartmentation following beta2-adrenergic stimulation of ICa,L in frog ventricular myocytes. J Physiol. 2003; 551: 239eC252.
Rich TC, Fagan KA, Tse TE, Schaack J, Cooper DM, Karpen JW. A uniform extracellular stimulus triggers distinct cAMP signals in different compartments of a simple cell. Proc Natl Acad Sci U S A. 2001; 98: 13049eC13054.
Brunton LL. PDE4: arrested at the border. Sci STKE. 2003; 2003: PE44.
Bacskai BJ, Hochner B, Mahaut-Smith M, Adams SR, Kaang BK, Kandel ER, Tsien RY. Spatially resolved dynamics of cAMP and protein kinase A subunits in Aplysia sensory neurons. Science. 1993; 260: 222eC226.
Karpen JW, Rich TC. The fourth dimension in cellular signaling. Science. 2001; 293: 2204eC2205.
Jurevicius J, Fischmeister R. cAMP compartmentation is responsible for a local activation of cardiac Ca2+ channels by beta-adrenergic agonists. Proc Natl Acad Sci U S A. 1996; 93: 295eC299.
Davare MA, Avdonin V, Hall DD, Peden EM, Burette A, Weinberg RJ, Horne MC, Hoshi T, Hell JW. A beta2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science. 2001; 293: 98eC101.
Schwartz JH. The many dimensions of cAMP signaling. Proc Natl Acad Sci U S A. 2001; 98: 13482eC13484.
Zippin JH, Chen Y, Nahirney P, Kamenetsky M, Wuttke MS, Fischman DA, Levin LR, Buck J. Compartmentalization of bicarbonate-sensitive adenylyl cyclase in distinct signaling microdomains. FASEB J. 2003; 17: 82eC84.
Pastor-Soler N, Beaulieu V, Litvin TN, Da Silva N, Chen Y, Brown D, Buck J, Levin LR, Breton S. Bicarbonate-regulated adenylyl cyclase (sAC) is a sensor that regulates pH-dependent V-ATPase recycling. J Biol Chem. 2003; 278: 49523eC49529.
Buck J, Sinclair ML, Schapal L, Cann MJ, Levin LR. Cytosolic adenylyl cyclase defines a unique signaling molecule in mammals. Proc Natl Acad Sci U S A. 1999; 96: 79eC84.
Chen Y, Cann MJ, Litvin TN, Iourgenko V, Sinclair ML, Levin LR, Buck J. Soluble adenylyl cyclase as an evolutionarily conserved bicarbonate sensor. Science. 2000; 289: 625eC628.
Qiao J, Mei FC, Popov VL, Vergara LA, Cheng X. Cell cycle-dependent subcellular localization of exchange factor directly activated by cAMP. J Biol Chem. 2002; 277: 26581eC26586.
Scott JD, Stofko RE, McDonald JR, Comer JD, Vitalis EA, Mangili JA. Type II regulatory subunit dimerization determines the subcellular localization of the cAMP-dependent protein kinase. J Biol Chem. 1990; 265: 21561eC21566.
Sayner SL, Frank DW, King J, Chen H, VandeWaa J, Stevens T. Paradoxical cAMP-induced lung endothelial hyperpermeability revealed by Pseudomonas aeruginosa ExoY. Circ Res. 2004; 95: 196eC203.
Stevens T, Nakahashi Y, Cornfield DN, McMurtry IF, Cooper DM, Rodman DM. Ca(2+)-inhibitable adenylyl cyclase modulates pulmonary artery endothelial cell cAMP content and barrier function. Proc Natl Acad Sci U S A. 1995; 92: 2696eC2700.
Stevens T, Creighton J, Thompson WJ. Control of cAMP in lung endothelial cell phenotypes. Implications for control of barrier function. Am J Physiol. 1999; 277: L119eCL126.
Creighton JR, Masada N, Cooper DM, Stevens T. Coordinate regulation of membrane cAMP by Ca2+-inhibited adenylyl cyclase and phosphodiesterase activities. Am J Physiol Lung Cell Mol Physiol. 2003; 284: L100eCL107.
Cioffi DL, Moore TM, Schaack J, Creighton JR, Cooper DM, Stevens T. Dominant regulation of interendothelial cell gap formation by calcium-inhibited type 6 adenylyl cyclase. J Cell Biol. 2002; 157: 1267eC1278.
Dessauer CW, Gilman AG. Purification and characterization of a soluble form of mammalian adenylyl cyclase. J Biol Chem. 1996; 271: 16967eC16974.
Ruchko M, Gorodnya O, LeDoux SP, Alexeyev MF, Al-Mehdi AB, Gillespie MN. Mitochondrial DNA damage triggers mitochondrial dysfunction and apoptosis in oxidant-challenged lung endothelial cells. Am J Physiol Lung Cell Mol Physiol. 2005; 288: L530eCL535.
Gu C, Cali JJ, Cooper DM. Dimerization of mammalian adenylate cyclases. Eur J Biochem. 2002; 269: 413eC421.
Ahlijanian MK, Cooper DM. Calmodulin may play a pivotal role in neurotransmitter-mediated inhibition and stimulation of rat cerebellar adenylate cyclase. Mol Pharmacol. 1987; 32: 127eC132.
Salomon Y, Londos C, Rodbell M. A highly sensitive adenylate cyclase assay. Anal Biochem. 1974; 58: 541eC548.
Tang WJ, Gilman AG. Construction of a soluble adenylyl cyclase activated by Gs alpha and forskolin. Science. 1995; 268: 1769eC1772.
Zaccolo M, Pozzan T. Discrete microdomains with high concentration of cAMP in stimulated rat neonatal cardiac myocytes. Science. 2002; 295: 1711eC1715.
Moore TM, Chetham PM, Kelly JJ, Stevens T. Signal transduction and regulation of lung endothelial cell permeability. Interaction between calcium and cAMP. Am J Physiol. 1998; 275: L203eCL222.
Leppla SH. Anthrax toxin edema factor: a bacterial adenylate cyclase that increases cyclic AMP concentrations of eukaryotic cells. Proc Natl Acad Sci U S A. 1982; 79: 3162eC3166.
Wong W, Scott JD. AKAP signalling complexes: focal points in space and time. Nat Rev Mol Cell Biol. 2004; 5: 959eC970.
Diviani D, Scott JD. AKAP signaling complexes at the cytoskeleton. J Cell Sci. 2001; 114: 1431eC1437.
Itoh TJ, Hisanaga S, Hosoi T, Kishimoto T, Hotani H. Phosphorylation states of microtubule-associated protein 2 (MAP2) determine the regulatory role of MAP2 in microtubule dynamics. Biochemistry. 1997; 36: 12574eC12582.
Litersky JM, Johnson GV, Jakes R, Goedert M, Lee M, Seubert P. Tau protein is phosphorylated by cyclic AMP-dependent protein kinase and calcium/calmodulin-dependent protein kinase II within its microtubule-binding domains at Ser-262 and Ser-356. Biochem J. 1996; 316 (pt 2): 655eC60.
Verin AD, Birukova A, Wang P, Liu F, Becker P, Birukov K, Garcia JG. Microtubule disassembly increases endothelial cell barrier dysfunction: role of MLC phosphorylation. Am J Physiol Lung Cell Mol Physiol. 2001; 281: L565eCL574(Sarah L. Sayner, Mikhail )
Integrative Biology and Pharmacology (C.W.D.), University of Texas, Houston.
Abstract
Subtle elevations in cAMP localized to the plasma membrane intensely strengthen endothelial barrier function. Paradoxically, pathogenic bacteria insert adenylyl cyclases (ACs) into eukaryotic cells generating a time-dependent cytosolic cAMP-increase that disrupts rather than strengthens the endothelial barrier. These findings bring into question whether membrane versus cytosolic AC activity dominates in control of cell adhesion. To address this problem, a mammalian forskolin-sensitive soluble AC (sACI/II) was expressed in pulmonary microvascular endothelial cells. Forskolin stimulated this sACI/II construct generating a small cytosolic cAMP-pool that was not regulated by phosphodiesterases or Gs. Whereas forskolin simultaneously activated the sACI/II construct and endogenous transmembrane ACs, the modest sACI/II activity overwhelmed the barrier protective effects of plasma membrane activity to induce endothelial gap formation. Retargeting sACI/II to the plasma membrane retained AC activity but protected the endothelial cell barrier. These findings demonstrate for the first time that the intracellular location of cAMP synthesis critically determines its physiological outcome.
Key Words: phosphodiesterase signal transduction ExoY second messenger Pseudomonas aeruginosa
Introduction
The discovery of an adenine ribonucleotide (cAMP) as an intracellular signaling messenger, and the recognition of its highly hydrophilic nature, led to the initial idea that this molecule rapidly and equivalently accesses all cellular compartments. However, as studies advanced, it became unclear how different external stimuli, which similarly elevate cAMP, selectively control different physiological processes.1,2 These and related studies have given rise to the evolving hypothesis that cAMP signaling occurs within spatially and temporally confined microdomains.2eC5
Whole cell cAMP concentration is established by the activities of ACs that synthesize cAMP and phosphodiesterases that hydrolyze it. AC1 to -9 are all transmembrane proteins,6,7 and therefore cAMP synthesis emanates from the plasma membrane. Phosphodiesterases are distributed discretely throughout the cell where they inactivate cAMP before it escapes into physiologically inappropriate domains. Indeed, the coordinated activities of ACs and phosphodiesterases are responsible for generating cAMP gradients within a single cell, where high concentrations arise at the plasma membrane and lower concentrations exist in the bulk cytosol.8eC12 Yet it is unclear how cAMP is targeted to its physiologically appropriate effectors. Hall and colleagues have proposed that molecules of the signaling cascade may be orientated within a molecular complex such that cAMP is steered directly to its relevant effectors.13 Thus, the prevailing theory is that signaling fidelity is derived from spatial and temporal cAMP transitions generated within functional subcellular compartments where downstream targets are similarly located.2,3,14eC16
These paradigms presume that cAMP synthesis occurs exclusively at the plasma membrane. However, recent evidence suggests human soluble AC (hsAC) exists in a wide range of cells and tissues,17eC20 where it is tethered to mitochondria, nucleus, microtubules, and the microtubule organizing center,17 generating localized cAMP in response to bicarbonate fluctuations.20 This finding brings into question the role hsAC plays in the activation of downstream cytosolic cAMP targets. Although many cAMP effectors (eg, protein kinase A [PKA] and Epac) localize to nonplasma membrane cellular compartments,21,22 the current paradigm is that these targets are activated by cAMP that is generated at the plasma membrane, paradoxical to emerging evidence that free cAMP diffusion is quite limited. Currently, the physiological significance of cytosolic cAMP synthesis is not known.
Opposing actions of membrane versus cytosolic cAMP synthesis have been previously described in endothelial cells,23 providing evidence for physiologically relevant cAMP compartments. Pulmonary microvascular endothelial cells (PMVECs) predominantly express the calcium inhibited AC6.24,25 Calcium inhibition of AC6 cannot easily be resolved by standard whole cell cAMP measurements in PMVECs, yet 90% of its membrane activity is inhibited by submicromolar calcium concentrations.26 Heterologous expression of calcium stimulated AC8 reverses the calcium inhibition of AC6 but only modestly increases global cAMP concentrations. Despite such subtle shifts in the calcium regulation of cAMP, reversing the calcium inhibition to a stimulation of AC prevents thrombin from inducing interendothelial cells gaps.27 Thus, a membrane delimited cAMP pool that is not reflected by global cAMP concentrations has powerful barrier enhancing properties. In stark contrast to this membrane delimited cAMP pool, P aeruginosa transfers a soluble AC (eg, ExoY) into the cytosol of PMVECs where it exhibits a time-dependent increase in AC activity that is restricted exclusively to the cytosol.23 This cytosolic cAMP induces intercellular gaps in cultured PMVEC monolayers and increases the filtration coefficient of the isolated perfused lung.23 These studies suggest that the disparate compartmentation of cAMP synthesis, eg, plasma membrane versus cytosol, exerts exact opposite effects on endothelial barrier strength, although the ExoY-dependent cAMP rise occurs progressively over 4 hours without a concomitant increase in membrane cAMP. Thus, the goal of this study was to determine whether simultaneous rises in plasma membrane versus cytosolic cAMP dominates in the regulation of endothelial barrier strength. Our findings indicate that when cAMP is simultaneously generated at the plasma membrane and within the cytosol, cAMP synthesis in the cytosolic compartment overwhelms the barrier protective properties of membrane-restricted cAMP to induce endothelial cell gaps, demonstrating for the first time that the subcellular localization of signal generation is an important determinant of its physiological outcome.
Materials and Methods
Isolation and Culture of PMVECs
PMVECs were isolated, cultured, and passaged in EC media (DMEM+10% FBS+1% PBS) as described.25
Adenovirus Constructs and Infection
Adenovirus encoding soluble ACI/II28 and green fluorescence protein (GFP) (Ad 88929) have been described elsewhere. Adenovirus was added to 70% confluent PMVECs in EC media at multiplicity of infection (moi) 10:1 for 40 to 48 hours.
Retrovirus Constructs and Infection
The TM-AC8 fusion described previously,30 was inserted between HindIII and XhoI sites of a hygromycin-encoding retroviral vector pBABA-Hygro derivative. The sACI/II terminal codon was removed and BamHI sites attached. This was inserted between the N-terminal transmembrane domains of TM-AC8 to generate sACI/II-TM-AC8, and subcloned into the pBABE-Hygro derivative. Retrovirus media was added to 40% confluent PMVECs in 5.9 e/L polybrene, centrifuged for 1 hour at 750g at room temperature. Retrovirus media remained on the cells for 15 hours and was then replaced with EC media. Retroviral cell lines were passaged and plated for each experiment into hygromycin selection media (50 e/mL).
Whole Cell cAMP Measurements
cAMP concentration was assessed by standard radioimmunoassay (Biomedical Technologies, Stoughton, Mass). Forskolin (100 eol/L) or isoproterenol (1 eol/L) with or without rolipram (10 eol/L) was added as indicated.
Fractionation of Cellular Lysates
Confluent cells were rinsed in ice-cold buffer A (137 mmol/L NaCl, 2.7 mmol/L KCl, 10.14 mmol/L Na2HPO4, 10.76 mmol/L KH2PO4) followed by 10-minute incubation in ice-cold buffer B (40 mmol/L Tris-HCl, 1 mmol/L EDTA, 150 mmol/L NaCl). Cells were scraped using 10 mL ice-cold buffer C (20 mmol/L HEPES, pH 7.5, 5 mmol/L EDTA, 1 mmol/L EGTA, 2 mmol/L dithiothreitol, 200 mmol/L sucrose, protease inhibitor cocktail) and centrifuged for 10 minutes at 250g at 4°C. Supernatant 1 (media) was removed and the pellet resuspended in 2 mL buffer C using 20 strokes each of course and fine pestle with Dounce homogenizer. This suspension was centrifuged at 100 000g for 20 minutes at 4°C. Supernatant 2 (cytosolic fraction) was removed and the pellet (plasma membrane fraction) resuspended in 1.5 mL buffer C using 15 strokes of Duall homogenization. Fractions were stored at eC80°C. Protein concentration was determined using the modified Lowry method (Pierce).
Adenylyl Cyclase Activity Assay
Assays were performed using a 1-mL solution of 0.1 mmol/L ATP, 1 eol/L cAMP, 2.5 U/mL creatine phosphokinase, 12 mmol/L phosphocreatine, 1 U/mL adenine deaminase, 0.04 mmol/L GTP, 2 mmol/L MgCl2, and 25 mmol/L HEPES at pH 7.3 with physiological potassium solution (103.1 mmol/L KOH, 10 mmol/L KHCO3, and 26.9 mmol/L K2HPO4) and 100 eol/L calcium (determined using Fabiato free ion calculator).31 Reactions contained [-32P]ATP (1x106 cpm) with or without 100 eol/L forskolin and initiated by addition of 0.15 mg of protein, proceeded for 30 minutes at 30°C and terminated using 200 e蘈 30% trichloroacetic acid. Recovery was monitored using internal standard [2,8-3H]cAMP (15 000 cpm). Production of radio-labeled cAMP from substrate [-32P]ATP was determined following isolation of [32P]cAMP by Dowex cation exchange then alumina oxide chromatography.32
Time-Lapse Microscopy
Images were acquired at 1-minute intervals for 2 hours as described previously.23 Forskolin (100 eol/L) was added after 60 minutes.
Statistical Analysis
Values represent means±SD. One- or 2-way ANOVA followed by Bonferroni post test or t test was used for comparisons. Significance was accepted as P<0.05.
Results
sACI/II Elevates cAMP in PMVECs
Previous studies demonstrate that plasma membrane AC activity is endothelial barrier protective, whereas cytosolic activity is barrier disruptive.23 To determine which of these two sites of synthesis dominates in the regulation of endothelial permeability, we used a soluble forskolin-sensitive AC system that was generated by fusing the amino-terminal portion of the catalytic loops of mammalian transmembrane AC type I (C1a) and II (C2a) with a 14 amino acid linker33 (sACI/II, Figure 1a). Loss of the C1b domain of ACI renders the chimera insensitive to calcium.28 Dessauer and Gilman generated an adenovirus for expression of this soluble enzyme in mammalian cells.28 When expressed in PMVECs, activation by forskolin generated a time-dependent increase in intracellular cAMP (Figure 1b). Interestingly, no increase in basal cAMP was observed before the addition of forskolin, which was attributable to the lack of constitutive enzymatic activity. To confirm the sACI/II adenoviral specificity of elevated cAMP, the forskolin response was compared with cells infected with a control soluble GFP adenovirus. Forskolin activates endogenous transmembrane ACs increasing intracellular cAMP in uninfected and GFP infected control cells (Figure 1b). Expression of sACI/II increased the forskolin-stimulated intracellular cAMP above control groups, which was attributed to simultaneous activation of both endogenous transmembrane ACs and the sACI/II chimera (Figure 1b).
Previous studies have shown P aeruginosa competent for the type three secretion system transfers the AC ExoY into PMVECs, generating a cytosolic cAMP pool that is sufficient to increase endothelial permeability.23 At a sACI/II adenoviral dose moi 10:1, the maximal forskolin-stimulated cAMP response mimicked concentrations achieved by ExoY expressing P aeruginosa (Figure 1c).
Functional Compartments of Forskolin-Induced sACI/II-cAMP
Although these studies demonstrate sACI/II has activity in PMVECs, they provide no indication of the intracellular compartment where this activity resides. Phosphodiesterase activity is important to tightly restrict cAMP diffusion, thereby maintaining isolated cAMP pools8,10,34 and downstream signaling specificity. PMVECs express the rolipram-sensitive phosphodiesterase isoform 4, which controls the plasmalemmal cAMP pool.25 We sought to compare phosphodiesterase 4 regulation of the sACI/II-cAMP versus endogenously generated plasmalemmal cAMP. The forskolin-sensitive cAMP pool of uninfected control PMVECs is augmented by rolipram (Figure 2a). The substantial increase in cAMP generated by simultaneous forskolin and rolipram treatment was not sufficient to deplete total cellular ATP pools (data not shown). However, despite the fact that sACI/II expression increased the amount of cAMP produced in response to forskolin, rolipram did not significantly augment the forskolin-stimulated sACI/II pool (Figure 2b). Whereas rolipram induced a 110% increase in the forskolin-stimulated cAMP pool of uninfected and GFP infected controls, rolipram induced only a 40% rise in the forskolin-stimulated cAMP pool of sACI/II expressing cells (Figure 2c). This relative insensitivity of the sACI/II-dependent cAMP pool to hydrolysis by phosphodiesterase 4 suggests sACI/II and endogenous AC activity reside in different compartments.
To further determine compartmentation of sACI/II, we examined its integration into endogenous signaling pathways. Rolipram [10 eol/L, EC10025], used throughout this study to amplify global cAMP levels for detection purposes, doubles the intracellular cAMP concentrations of all treatment groups (data not shown). Isoproterenol activation of -adrenergic receptors is coupled via the trimeric G protein, GS, to endogenous transmembrane ACs. In PMVECs, isoproterenol (1 eol/L, EC10025) increases intracellular cAMP above rolipram alone (Figure 3). Interestingly, although in vitro kinetic analyses have demonstrated that GS can stimulate sACI/II,28 activation of GS in PMVECs expressing the sACI/II does not increase cAMP above control groups (Figure 3). Collectively, these data suggest that sACI/II resides in a compartment not readily accessible by GS.
Subcellular Localization of the sAC Activity
To determine the subcellular localization of sACI/II activity in PMVECs, cells were lysed into plasma membrane and cytosolic fractions and the AC activity measured. Similar basal- and forskolin-stimulated AC activity was detected in the plasma membrane fraction of all treatment groups (uninfected control, sACI/II and GFP infected cells) representative of the endogenous transmembrane ACs (Figure 4a). The plasma membrane fraction of sACI/II expressing cells did not possess elevated forskolin-sensitive AC activity above controls, demonstrating the chimera did not have activity at the plasmalemma. As the predominant endogenous AC in PMVECs is calcium sensitive isoform 6,26 calcium inhibition of forskolin activity further supports the plasma membrane composition of this fraction (Figure 4b). Although this is not the 90% calcium inhibition of AC activity in the lipid- and caveolin-enriched fraction,26 it demonstrates the presence of a calcium regulated AC in this bulk plasma membrane fraction. Within the cytosolic fraction, no basal AC activity was detected in any treatment group (Figure 4c); however, forskolin revealed 6 times more AC activity in the cytosolic fraction of sACI/II expressing cells compared with uninfected or GFP infected cells. These data demonstrate the unique cytosolic compartmentalized AC activity of sACI/II infected PMVECs. Interestingly, the modest level of forskolin-sensitive cytosolic sACI/II activity was comparable to the "basal" plasma membrane activity, but significantly less than the forskolin-stimulated membrane activity (Figure 4d). Therefore, when sACI/II expressing cells are stimulated by forskolin, there is 4.5 times more AC activity at the plasma membrane than in the cytosol, but 6 times more activity in the cytosolic compartment compared with basal conditions.
Forskolin-Stimulated sACI/II Activity Induces PMVEC Gap Formation
We next sought to determine whether plasma membrane or cytosolic AC activity would dominate in the regulation of the endothelial barrier. Forskolin was not sufficient to induce PMVEC gaps with the adenoviral GFP control. Whereas unstimulated sACI/II expressing PMVECs did not reveal gap formation, gaps rapidly formed with the addition of forskolin, as seen by the retraction of cell-cell borders (Figure 5; Movies 1 and 2 in the online data supplement available at http://circres.ahajournals.org). Thus, activation of the sACI/II chimera, although representing only a modest rise in cytosolic versus membrane activity, is sufficient to overwhelm the barrier protective effects of endogenous plasma membrane restricted AC activity to form endothelial cell gaps.
Relocation of sACI/II to the Plasma Membrane
To further test the hypothesis that the subcellular localization of AC activity governs its barrier protective or disruptive properties, the sACI/II chimera was relocalized to the plasma membrane. Gu et al30 have demonstrated that YFP sandwiched between the 2 transmembrane domains of AC8 (TM-AC8) (Figure 6a, i and ii) is sufficient for plasma membrane targeting. Plasma membrane localization of TM-AC8 was also detected in PMVECs (Figure 6b). Insertion of sACI/II between the transmembrane domains of AC8 generated sACI/II-TM-AC8 (Figure 6a, iii). Retroviral expression of sACI/II-TM-AC8 also localized to the membrane in PMVEC (Figure 6c). Whereas plasma membrane fractions of PMVECs expressing sACI/II-TM-AC8 exhibited elevated forskolin-dependent AC activity above controls (Figure 6d), no AC activity was detected in the cytosolic fraction (data not shown), illustrating its exclusive plasmalemma targeting. Forskolin stimulation of sACI/II-TM-AC8eCexpressing PMVECs did not lead to endothelial gap formation (Figure 6e and online data supplement, Movie 3), further demonstrating that AC activity exclusively restricted to the plasma membrane compartment with corresponding lack of activity in the cytosol does not disrupt the endothelial barrier.
Discussion
Elevated cAMP is widely recognized to be endothelial barrier protective.35 More specifically, only small cAMP elevations critically localized to the plasma membrane are needed to protect the PMVEC barrier.27 However, nature provides a paradox. Pathogenic bacteria have developed a mechanism to synthesize cAMP in the cytosol of eukaryotic cells, resulting in decreased endothelial barrier function. For example, edema factor of Bacillus anthracis elevates intracellular cAMP in CHO cells yet induces edema when injected into the skin of rabbits and guinea pigs.36 In addition, P aeruginosa inserts the adenylyl cyclase, ExoY, into the cytosol of PMVECs generating a barrier disruptive cytosolic cAMP pool.23 As theories of cAMP compartmentation have evolved, it appears that within many cell types not all cAMP elevations produce equivalent physiological outcomes. However, the discrete role of AC activity within different cellular microdomains has never been directly tested owing to the lack of a suitable physiological outcome measurement. Using novel techniques to manipulate compartments of AC activity, our present findings demonstrate that cytosolic AC activity in PMVECs is sufficient to overwhelm the barrier protective effects of plasma membrane activity to promote endothelial gap formation.
The mammalian sACI/II chimera is a fusion of C1a-ACI with C2a-ACII that contains an AC catalytic site and forskolin binding-pocket, thereby generating a conditionally active, soluble enzyme. Our findings that sACI/II has forskolin-dependent cytosolic AC activity in PMVECs that is not activated by Gs signaling in vivo support previous work of Dessauer and Gilman28; indeed sACI/II could not be activated by Gs signaling in vivo unless the enzyme was stimulated with forskolin.28 In PMVECs, sACI/II-cAMP was insensitive to phosphodiesterase 4 activity, collectively suggesting AC activity resides in a unique cytosolic compartment. Although forskolin simultaneously activates both the cytosolic sACI/II and endogenous transmembrane AC6, the magnitude of sACI/II activity only contributes a small fraction of the global cAMP pool. Despite the small contribution of the cytosolic cAMP to this global cAMP-pool, it was sufficient to induce PMVEC gaps. Currently it is unclear whether a change in the plasma membrane-to-cytosolic cAMP gradient, or a small rise in cytosolic AC activity, is responsible for overwhelming the barrier protective effects of plasma membrane AC activity resulting in endothelial gap formation. However, we have recently shown that disrupting membrane phosphodiesterase 4 activity reveals forskolin-stimulated PMVEC gap formation suggesting that cAMP must remain within the membrane domain for it to confer barrier protection (T.S., unpublished data, 2006).
When sACI/II is sandwiched between the two transmembrane domains of AC8, this plasma membrane targeted sACI/II retained its forskolin-stimulated AC activity, indicating "free" amino and carboxy termini are not required for enzymatic activity. Indeed, this is the first evidence that the entire AC catalytic pocket can reside in the C1 loop and still retain activity. When stimulated by forskolin, this chimeric protein generated small increases in cAMP similar to the level of activity achieved by sACI/II free in the cytosol. However, restriction of this sACI/II-dependent cAMP to the plasma membrane did not induce PMVEC gaps, definitively illustrating the site of cAMP synthesis critically mediates the physiological outcome.
The downstream cytosolic effectors that determine such diverse physiological responses of membrane and cytosolic cAMP transitions are unknown. Interestingly, mammalian, bicarbonate-stimulated, soluble ACs localize to mitochondria, nucleus, microtubules, and centrosomes where evidence is emerging that complete signalsomes exist. cAMP targets, such as PKA and Epac, also reside within the cytosol, some in association with complete signal transduction cascades.37 For example, PKA compartmentalized through its interactions with A kinase anchoring proteins (AKAPs), is involved in phosphorylation events that lead to rearrangement of the cytoskeleton.38 PKA phosphorylates microtubule associating proteins such as MAP2 and tau, which in turn reduces their ability to stabilize microtubules.39,40 Within the endothelium, microtubule disassembly increases permeability.41 An unregulated cytosolic cAMP pool could destabilize such cytoskeleton structures necessary to induce gap formation. Indeed, cAMP-signaling cascades are already in place within the cytosol, such that atypical cytosolic AC activity could lead to their activation. Future studies are necessary to pursue the identification of these cytosolic targets.
Thus, the present data demonstrate that novel targeting of minimal AC activity to the cytosol of PMVECs dominates over substantial plasmalemmal AC activity to induce endothelial gap formation. Relocalizing this same AC construct to the plasma membrane substantiates endothelial barrier protective effects of plasma membrane AC activity. Our data presented here contribute to the idea that compartmentalized microdomains of cAMP signaling play a key physiological and pathophysiological role on the regulation of pulmonary endothelial barrier.
Acknowledgments
This work was supported by NIH grants HL-60024 and HL-66299 (to T.S.) and by American Heart Association Southeast Consortium Predoctoral Fellowship 0315134B (to S.L.S.). We thank Dermot Cooper for providing the TM-AC8 plasmid construct and Judy Creighton for performing ATP studies.
References
Hayes JS, Brunton LL, Brown JH, Reese JB, Mayer SE. Hormonally specific expression of cardiac protein kinase activity. Proc Natl Acad Sci U S A. 1979; 76: 1570eC1574.
Steinberg SF, Brunton LL. Compartmentation of G protein-coupled signaling pathways in cardiac myocytes. Annu Rev Pharmacol Toxicol. 2001; 41: 751eC773.
Brunton LL, Hayes JS, Mayer SE. Functional compartmentation of cyclic AMP and protein kinase in heart. Adv Cyclic Nucleotide Res. 1981; 14: 391eC397.
Beavo JA, Brunton LL. Cyclic nucleotide research-still expanding after half a century. Nat Rev Mol Cell Biol. 2002; 3: 710eC718.
Bundey RA, Insel PA. Discrete intracellular signaling domains of soluble adenylyl cyclase: camps of cAMP Sci STKE. 2004; 2004: pe19.
Ludwig MG, Seuwen K. Characterization of the human adenylyl cyclase gene family: cDNA, gene structure, and tissue distribution of the nine isoforms. J Recept Signal Transduct Res. 2002; 22: 79eC110.
Sayner SL, Stevens T. Adenylyl cyclase and cAMP regulation of the endothelial barrier. In: Patterson CE, ed. Perspectives on Lung Endothelial Barrier Function. 1st ed. Amsterdam: Elsevier; 2005: 105eC138.
Rich TC, Fagan KA, Nakata H, Schaack J, Cooper DM, Karpen JW. Cyclic nucleotide-gated channels colocalize with adenylyl cyclase in regions of restricted cAMP diffusion. J Gen Physiol. 2000; 116: 147eC161.
Jurevicius J, Skeberdis VA, Fischmeister R. Role of cyclic nucleotide phosphodiesterase isoforms in cAMP compartmentation following beta2-adrenergic stimulation of ICa,L in frog ventricular myocytes. J Physiol. 2003; 551: 239eC252.
Rich TC, Fagan KA, Tse TE, Schaack J, Cooper DM, Karpen JW. A uniform extracellular stimulus triggers distinct cAMP signals in different compartments of a simple cell. Proc Natl Acad Sci U S A. 2001; 98: 13049eC13054.
Brunton LL. PDE4: arrested at the border. Sci STKE. 2003; 2003: PE44.
Bacskai BJ, Hochner B, Mahaut-Smith M, Adams SR, Kaang BK, Kandel ER, Tsien RY. Spatially resolved dynamics of cAMP and protein kinase A subunits in Aplysia sensory neurons. Science. 1993; 260: 222eC226.
Karpen JW, Rich TC. The fourth dimension in cellular signaling. Science. 2001; 293: 2204eC2205.
Jurevicius J, Fischmeister R. cAMP compartmentation is responsible for a local activation of cardiac Ca2+ channels by beta-adrenergic agonists. Proc Natl Acad Sci U S A. 1996; 93: 295eC299.
Davare MA, Avdonin V, Hall DD, Peden EM, Burette A, Weinberg RJ, Horne MC, Hoshi T, Hell JW. A beta2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science. 2001; 293: 98eC101.
Schwartz JH. The many dimensions of cAMP signaling. Proc Natl Acad Sci U S A. 2001; 98: 13482eC13484.
Zippin JH, Chen Y, Nahirney P, Kamenetsky M, Wuttke MS, Fischman DA, Levin LR, Buck J. Compartmentalization of bicarbonate-sensitive adenylyl cyclase in distinct signaling microdomains. FASEB J. 2003; 17: 82eC84.
Pastor-Soler N, Beaulieu V, Litvin TN, Da Silva N, Chen Y, Brown D, Buck J, Levin LR, Breton S. Bicarbonate-regulated adenylyl cyclase (sAC) is a sensor that regulates pH-dependent V-ATPase recycling. J Biol Chem. 2003; 278: 49523eC49529.
Buck J, Sinclair ML, Schapal L, Cann MJ, Levin LR. Cytosolic adenylyl cyclase defines a unique signaling molecule in mammals. Proc Natl Acad Sci U S A. 1999; 96: 79eC84.
Chen Y, Cann MJ, Litvin TN, Iourgenko V, Sinclair ML, Levin LR, Buck J. Soluble adenylyl cyclase as an evolutionarily conserved bicarbonate sensor. Science. 2000; 289: 625eC628.
Qiao J, Mei FC, Popov VL, Vergara LA, Cheng X. Cell cycle-dependent subcellular localization of exchange factor directly activated by cAMP. J Biol Chem. 2002; 277: 26581eC26586.
Scott JD, Stofko RE, McDonald JR, Comer JD, Vitalis EA, Mangili JA. Type II regulatory subunit dimerization determines the subcellular localization of the cAMP-dependent protein kinase. J Biol Chem. 1990; 265: 21561eC21566.
Sayner SL, Frank DW, King J, Chen H, VandeWaa J, Stevens T. Paradoxical cAMP-induced lung endothelial hyperpermeability revealed by Pseudomonas aeruginosa ExoY. Circ Res. 2004; 95: 196eC203.
Stevens T, Nakahashi Y, Cornfield DN, McMurtry IF, Cooper DM, Rodman DM. Ca(2+)-inhibitable adenylyl cyclase modulates pulmonary artery endothelial cell cAMP content and barrier function. Proc Natl Acad Sci U S A. 1995; 92: 2696eC2700.
Stevens T, Creighton J, Thompson WJ. Control of cAMP in lung endothelial cell phenotypes. Implications for control of barrier function. Am J Physiol. 1999; 277: L119eCL126.
Creighton JR, Masada N, Cooper DM, Stevens T. Coordinate regulation of membrane cAMP by Ca2+-inhibited adenylyl cyclase and phosphodiesterase activities. Am J Physiol Lung Cell Mol Physiol. 2003; 284: L100eCL107.
Cioffi DL, Moore TM, Schaack J, Creighton JR, Cooper DM, Stevens T. Dominant regulation of interendothelial cell gap formation by calcium-inhibited type 6 adenylyl cyclase. J Cell Biol. 2002; 157: 1267eC1278.
Dessauer CW, Gilman AG. Purification and characterization of a soluble form of mammalian adenylyl cyclase. J Biol Chem. 1996; 271: 16967eC16974.
Ruchko M, Gorodnya O, LeDoux SP, Alexeyev MF, Al-Mehdi AB, Gillespie MN. Mitochondrial DNA damage triggers mitochondrial dysfunction and apoptosis in oxidant-challenged lung endothelial cells. Am J Physiol Lung Cell Mol Physiol. 2005; 288: L530eCL535.
Gu C, Cali JJ, Cooper DM. Dimerization of mammalian adenylate cyclases. Eur J Biochem. 2002; 269: 413eC421.
Ahlijanian MK, Cooper DM. Calmodulin may play a pivotal role in neurotransmitter-mediated inhibition and stimulation of rat cerebellar adenylate cyclase. Mol Pharmacol. 1987; 32: 127eC132.
Salomon Y, Londos C, Rodbell M. A highly sensitive adenylate cyclase assay. Anal Biochem. 1974; 58: 541eC548.
Tang WJ, Gilman AG. Construction of a soluble adenylyl cyclase activated by Gs alpha and forskolin. Science. 1995; 268: 1769eC1772.
Zaccolo M, Pozzan T. Discrete microdomains with high concentration of cAMP in stimulated rat neonatal cardiac myocytes. Science. 2002; 295: 1711eC1715.
Moore TM, Chetham PM, Kelly JJ, Stevens T. Signal transduction and regulation of lung endothelial cell permeability. Interaction between calcium and cAMP. Am J Physiol. 1998; 275: L203eCL222.
Leppla SH. Anthrax toxin edema factor: a bacterial adenylate cyclase that increases cyclic AMP concentrations of eukaryotic cells. Proc Natl Acad Sci U S A. 1982; 79: 3162eC3166.
Wong W, Scott JD. AKAP signalling complexes: focal points in space and time. Nat Rev Mol Cell Biol. 2004; 5: 959eC970.
Diviani D, Scott JD. AKAP signaling complexes at the cytoskeleton. J Cell Sci. 2001; 114: 1431eC1437.
Itoh TJ, Hisanaga S, Hosoi T, Kishimoto T, Hotani H. Phosphorylation states of microtubule-associated protein 2 (MAP2) determine the regulatory role of MAP2 in microtubule dynamics. Biochemistry. 1997; 36: 12574eC12582.
Litersky JM, Johnson GV, Jakes R, Goedert M, Lee M, Seubert P. Tau protein is phosphorylated by cyclic AMP-dependent protein kinase and calcium/calmodulin-dependent protein kinase II within its microtubule-binding domains at Ser-262 and Ser-356. Biochem J. 1996; 316 (pt 2): 655eC60.
Verin AD, Birukova A, Wang P, Liu F, Becker P, Birukov K, Garcia JG. Microtubule disassembly increases endothelial cell barrier dysfunction: role of MLC phosphorylation. Am J Physiol Lung Cell Mol Physiol. 2001; 281: L565eCL574(Sarah L. Sayner, Mikhail )