Published online before print November 11, 2004, doi: 10.1161/01.RES.0000150592.88464.ad
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2004 American Heart Association, Inc.
UltraRapid Communications |
Peptidyl-Glycine 
-Amidating Monooxygenase Targeting and Shaping of Atrial Secretory Vesicles
Inhibition by Mutated N-Terminal ProANP and PBA
From the Department of Neuroscience, Centre Médical Universitaire, University of Geneva, Switzerland.
Correspondence Dr Alex J. Baertschi, Neuroscience, CMU, 1 Rue Michel-Servet, 1211 Geneva 4, Switzerland. E-mail alex.baertschi{at}medecine.unige.ch
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
ANP (atrial natriuretic peptide) is widely recognized as an important vasorelaxant, diuretic, and cardioprotective hormone. Little is known, however, about how ANP-secretory vesicles form within the atrial myocytes. Secretory vesicles were visualized by fluorescence microscope imaging in live rat atrial myocytes expressing proANP–enhanced green fluorescent protein (EGFP), or N-terminal–mutated fusion proteins thought to suppress the calcium-dependent aggregation of proANP. Results showed the following: (1) aggregates of proANP and coexpressed proANP-EGFP recruited peptidylglycine
-amidating monooxygenase (PAM)-1, an abundant atrial integral vesicle membrane protein; (2) coexpressed N-terminal–mutated (Glu23,24
Gln23,24) and N-terminal–deleted proANP-EGFP inhibited recruitment of PAM-1 by up to 60%; (3) 4-phenyl-3-butenoic acid (PBA) (10 µmol/L), a pharmacological inhibitor of the lumenal peptidylglycine -hydroxylating monooxygenase domain of PAM proteins, inhibited recruitment of endogenous PAM-1 and of coexpressed pro-EGFP–PAM-1; (4) PBA had no effect on exocytosis of the potassium inward rectifier KIR2.1; (5) PBA induced a deformation of the secretory vesicles but did not inhibit docking. These findings suggest that recruitment of PAM-1 to secretory vesicles depends on intact N-terminal proANP and on the lumenal domain of PAM-1. Conversely, PAM-1 participates in shaping the proANP-secretory vesicles. The full text of this article is available online at http://circres.ahajournals.org.
Key Words: atrium • EGFP • in vivo imaging • PAM • proANP • secretory vesicles
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Atrial natriuretic peptide (ANP), originally discovered by de Bold,1 is an endogenous hypotensive hormone involved in the regulation of blood pressure and fluid homeostasis.2,3 ANP is secreted from the cardiac atria in response to cardiac stress such as oxygen deficiency and mechanical overload (see Jiao and Baertschi4 for citations). Appropriately, ANP relieves the heart from hypoxic vasoconstriction of the pulmonary circulation and also helps protect the heart from mechanical overload.2–4 ANP and brain natriuretic peptide have become important clinical markers of cardiac hypertrophy and failure. Accordingly, there are thousands of studies on ANP release and action, in both animal and man.
In contrast to this wealth of knowledge, the cell biology of ANP-containing secretory vesicles is far less understood. ANP is synthesized as an inactive precursor, preproANP. A 24-aa signal sequence directs the nascent amino acid chain into the lumen of the endoplasmic reticulum, wherefrom it is processed through the Golgi stacks to the trans-Golgi network (TGN). The prohormone proANP is not cleaved but remains intact during its journey from the TGN to the plasma membrane.3 Sorting of proANP to secretory vesicles may be accomplished by calcium-dependent aggregation of proANP5,6 within electron-dense granules in the TGN.7 By analogy with other secretory systems,8–12 integral membrane proteins of the TGN may recognize the surface layer of the proANP core and enwrap it with TGN membrane to form the secretory vesicle. Our previous study indicates that mutations of the N-terminal calcium-binding domain of proANP results in deformed vesicles that no longer dock at the plasma membrane.13 This raises the question about the identity of the proteins that are recruited to shape and dock the secretory vesicles.
Although proteins involved in secretory vesicle budding and fusion are now well characterized in yeast,14 those of the cardiac ANP-secretory pathway are still largely unknown. Those known to play important roles in other mammalian secretory systems, such as vesicle associated protein and SCAMP secretory carrier–associated protein,14,15 could not be detected, by our preliminary studies, in significant amounts on atrial proANP vesicles. We found that peptidylglycine -amidating monooxygenase (PAM), an abundant protein family in the cardiac atrium,16–19 is localized on atrial proANP–enhanced green fluorescent protein (EGFP)–expressing vesicles and could thus be a viable candidate. PAM posttranslationally activates approximately half of all mammalian neuropeptides by converting their COOH-terminal glycine into an essential -amide moiety (see Prigge et al16 for review). PAM catalyzes this reaction in two consecutive steps via peptidylglycine -hydroxylating monooxygenase (PHM) and peptidyl--hydroxyglycine -amidating lyase. Several forms of PAM mRNA result from alternative splicing of the single copy rat PAM gene. PAM-1, the longest form, is composed of the PHM catalytic domain, a noncatalytic domain (exon A), the peptidyl--hydroxyglycine -amidating lyase catalytic domain (all within the vesicle lumen), a transmembrane domain (TM), and a C-terminal domain (CD) that extends into the cytoplasm. Deletion of exon A yields PAM-2, and further deletion of TM gives rise to soluble PAM-3 within the vesicle lumen (Figure 1C). In the heart atrium, most of the PAM-1 does not seem to be cleaved into smaller protein products.17–20
|
N, deletion of calcium binding N-terminal of proANPSo far, there is no known substrate for PAM in atrial vesicles; thus other functions of PAM have been explored. Because the lumenal catalytic domain of PAM is pH sensitive, PAM might play a role in signaling lumenal conditions to the cytosolic proteins.21 Mutation of the PAM-CD at juxtamembrane sites eliminates the ability of PAM to bind to the PAM C-terminal interacting protein and to modify regulated secretion.22–24 The pH-dependent aggregation of PAM25 could play a role in its segregation to secretory granules. If PAM interacted with lumenal proteins of the TGN, such as proANP, PAM could be involved in forming the atrial secretory vesicles.
The purpose of this study was to determine whether PAM proteins were implicated in shaping and docking the atrial proANP vesicles. We answered the following questions. Are PAM proteins recruited to proANP-EGFP–expressing secretory vesicles? Do mutations of the N-terminal calcium-binding sequence of proANP suppress recruitment of PAM? Does a pharmacological inhibitor of PAM inhibit recruitment of PAM, deform the secretory vesicles, or suppress docking? The imaging of single atrial vesicles13 in live atrial myocytes was crucial for answering some of these questions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Primary Culture of Atrial Myocytes
Atrial appendage cells from neonatal rats (2- to 4-days old; Sprague-Dawley) were cultured as described previously13,26 on 22-mm glass slides in F10 medium. Late pregnant rats were obtained from Charles River Laboratories (L’Arbresle, France) or from the University of Geneva. Experiments were conducted according to the Guide for Care and Use of Laboratory Animals (with approval of the University Ethics Committee and the State of Geneva Veterinary Office).
Plasmid Constructs
Three pcDNA3-EGFP expression vectors were used for encoding preproANP-10aa-EGFP27 (a gift of Edwin Levitan, University of Pittsburgh, Pa), preproANP (Glu23,24Gln23,24)-10aa-EGFP, and preproANP(45–127)-10aa-EGFP13 (Figure 1A). The 10-aa sequence forms the spacer. Electroporation of these plasmids was successfully performed with a Gene Pulser II (Bio-Rad).13 The amount of fusion protein expressed in atrial secretory vesicles appears sufficiently high to explain their effect on vesicle shape and docking.13 From calibrations with EGFP protein, and the number of prohormone molecules contained in a dense core secretory vesicle,28 the mutated fusion proteins are estimated to be coexpressed in a ratio of at least 1:2 with endogenous proANP.
A new construct, encoding EGFP-PAM-1 preceded by the PAM-1 prosequence (Figure 1B), was made in several steps from the pCIneo plasmid (Promega Corporation, Madison, Wis) containing PAM-1 cDNA (a gift of R. Mains, University of Connecticut, Farmington). First, an ApaI restriction site was inserted by PCR between the prosequence and the PHM domain of full-length PAM-1 cDNA. Two primer pairs: BglII sense (tcgacagatcttcaat), prosequence antisense (atatgggcccaaacctcttaaagacaga); and PHM sense (atatgggcccagtttt-aaagaaactacc), EcoRV antisense (aagcactgatatcgcc; Microsynth, Balgach, CH) were designed to amplify two fragments, BglII/ApaI, including the signal peptide and the prosequence, and ApaI/EcoRV, containing part of the PHM domain. These two fragments were ligated using the engineered ApaI restriction site (shown in italics). The resulting fragment was subcloned in the pCIneoPAM-1 plasmid, replacing the BglII/EcoRV cassette (pCIneoPAM-1/ApaI). Second, to create a 10-aa spacer upstream of the PHM domain, the mutagenesis primer ApaI 10-aa sense (tatagggcccagtatcaaccctccagtagcatttaaagaaact) was designed to overlap the ApaI restriction site on the pCIneoPAM-1/ApaI plasmid and insert 21 new nucleotides overlapping the 5'-PHM domain. Together with the EcoRV primer, a new ApaI-10aa/EcoRV fragment was amplified by mutagenesis PCR. This fragment was subcloned in the pCIneoPAM-1/ApaI plasmid to generate the pCIneoPAM-1/ApaI10aa plasmid. Third, EGFP cDNA was amplified from the pEGFP-N1 plasmid sequence (Clontech, Palo Alto, Calif) by PCR with a primer pair, allowing to create ApaI restriction sites at the 5' and 3' ends of the cDNA and to mutate the ATG and TAA codons. This fragment was subcloned in the pCIneoPAM-1/ApaI and pCIneoPAM-1/ApaI10aa plasmids. The two plasmids, pCIneoPAM-1/ApaI EGFP and pCIneoPAM-1ApaI 10aa EGFP, were electroporated in atrial myocytes and HL-1 cells. Only the construct containing the 10-aa spacer gave rise to EGFP expression; the green fluorescence being localized to ANP-secretory vesicles (Figure 2C).
|
Treatment of Atrial Myocytes With 4-Phenyl-3-Butenoic Acid
Vehicle or 4-phenyl-3-butenoic acid (PBA) (10 µmol/L), the most potent irreversible, mechanism-based PHM inactivator known,29 was added to new culture medium each day. As a control experiment, the effects of PBA were tested on another regulated secretory system, the inwardly rectifying potassium channel KIR2.1,30,31 unrelated to proANP and without interactions with PAM-1.32 The exocytosis of EGFP-KIR2.1 was compared in 27 atrial myocytes treated with PBA or vehicle, as described in the legend to Figure 6.
Imaging
Protocols
Between 16 and 50 hours after electroporation, the cultures were placed in HEPES-containing buffer (mmol/L: KCl, 5; CaCl2, 1; MgCl2, 1; NaCl, 118; glucose, 2.5; HEPES, 10; NaOH to pH 7.3; sucrose to290 milliosmol/kgH2O) on a temperature-controlled (30°C to 35°C) stage of a wide-field Nikon-DV200 Diaphot microscope.13 After imaging, the cells were fixed for 15 minutes in 2% paraformaldehyde and stored in PBS and darkness at 4°C until immunolabeling.
Acquisition
Image acquisition was performed with Metamorph software (Universal Imaging) at a monochromatic excitation of 480 nm (EGFP) or 575 nm (Texas Red) (±6-nm bandwidth),13 with illumination times of usually 500 ms (range 300 to 1000 ms). Forty to 80 optical sections were obtained at 150-nm vertical steps. The z-scan image stacks were deconvoluted using Huygens software (Bitplane) by the iterative constrained Tikhonov–Miller procedure and superposed by Metamorph software.13 Confocal optical sections were also acquired directly using laser excitations at 488 nm and 568 nm, a Nikon Diaphot 300 inverted microscope with a NA 1.3 x100 oil objective, a QLC100 spinning wheel confocal head (VisiTech International), and a Coolsnap HQ cooled charge-coupled device (CCD) camera (Visitron Systems). The pixel size (at 2x2 binning) in the x-y plane was 129 nm.
Measurement of Shape and Docking
The shape of live, peripherally located vesicles in time lapses or z scans was analyzed by zooming first at 150% and then at 800% (using edit-duplicate-stack-with-zoom function; MetaMorph). For each vesicle, gamma and contrast were adjusted from 1 to 5.43 and from 50 to 70, respectively, before automatic contour fitting. Each vesicle was coded by a number and classified twice by two observers as round when more than 50% of all contours were round. In controls, between 50% and 60% of vesicles were classified as round. Vesicles were classified as docked when the vesicles remained immobile in the periphery of the cell and over the total recording period; otherwise, they were classified as moving. A peripheral location was chosen so that secretory vesicles would not be confused with vesicles bulging from the TGN; the vesicles were close to the cell membrane, thus increasing the likelihood that they were docked. It was unavoidable that some vesicles that were close to the membrane but not docked could not be distinguished from truly docked vesicles.
Immunostaining
To study colocalization of green fluorescence with ANP, PAM-1, or CD-PAM, the fixed cultures were pretreated with 1% normal serum and one of the following primary antibodies was applied (4 days at 4°C): rabbit anti-rat ANP antiserum (IHC9103; Peninsula Europe) (dilution 1:1000); rabbit polyclonal antiserum to exon A of PAM-1 (No. 629; 1:5000)33; or C-terminal domain of PAMs (No. 571; 1:1500).33 The primary antibodies were visualized with Texas Red-X–labeled or Alexa594-labeled goat anti-rabbit 
-globulin (Santa Cruz or Molecular Probes; 1:200 to 1:400). Separation of green and red fluorescence was achieved using Nikon filters as described previously.13 Controls included application of nonimmune serum. Nuclei were stained with 4',6-diamidino-2-phenylindole dihydrochloride (DAPI) (D-1306; Molecular Probes). For example, an atrial myocyte expressing proANP-EGFP and stained for PAM-1 is shown in Figure 2A; another expressing proANP-EGFP and stained for ANP and DAPI, in Figure 2B; and a third expressing pro-EGFP–PAM-1 and stained for ANP and DAPI, in Figure 2C.
Western Blots
Western blots were performed similarly as described previously,13 using anti–PAM-1 antiserum 629 (1:2000 to 1:4000), anti–CD-PAM antiserum 571 (1:1000), and anti-PHM antiserum 475 (1:1000) (Figure 1C). For comparison purposes, each lane was loaded with the same amount of total protein (20 µg). All Western blots shown are from originally stained, nonstripped membranes.
Statistics
The analysis of colocalization and of vesicle shape was performed on a total of 2992 vesicles in 335 atrial myocytes. The analysis of vesicle docking was performed on a total of 4813 vesicles in an overlapping population of 170 myocytes. The analysis was divided into two parts: (1) colocalization of proANP-EGFP or pro-EGFP–PAM-1 fusion proteins with PAM or ANP immunostaining; and (2) tests with the PHM inhibitor PBA. Results were displayed as histograms of weighted averages. Differences were analyzed for statistically significant differences by ANOVA tests on ranked data using SAS software from the SAS Institute (Carey, NC), the nonparametric Mann–Whitney U test, or 
2 test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Characterization of Experimental System
We used a cDNA construct encoding a fusion protein of proANP-EGFP27 that was localized by immunoelectron microscopy in dense core vesicles of atrial myocytes.13 A potential protease site at the C terminus of the proANP was replaced with a 10-aa spacer (GPGINPPVAT) to prevent cleavage of proANP-EGFP.27 As control, we electroporated myocytes with the proANP-EGFP expression vector and labeled the cells with an antibody directed against ANP. The great majority of cells, 88.8±2.3%, were labeled by EGFP as well as by anti-ANP antibody. An example is shown in Figure 2B, where the superposition of EGFP with ANP (red) yields yellow-orange vesicles and TGN surrounding the DAPI-stained (blue) nucleus. To further address the question of cleavage, the proANP-EGFP was expressed in HL-1 atrial myocytes. Western blots for EGFP showed a band corresponding to the intact proANP-EGFP fusion protein, confirming our previous study13 (not shown). Analogous results were obtained with the pro-EGFP–PAM-1 expression vector (eg, Figure 2C; Western blot not shown). Furthermore, Western blots for endogenous PAM-1 for atrial but not ventricular culture extracts showed a major band
125 kDa (Figure 2D1). Blots for CD-PAM displayed two major bands for atrial but not ventricular culture extracts 125 kDa and 110 kDa (Figure 2D2). A nonspecific band of lower molecular mass appeared for all extracts. Blots for PHM showed, in four atrial extracts, a weak band 125 to 130 kDa, a strong band 110 kDa, and a very faint band 90 kDa (Figure 2D3). Previous reports19,33 indicate that these bands correspond to PAM-1 (Figure 2D1), PAM-1 and PAM-2 (Figure 2D2), and PAM-1, PAM-2, and PAM-3 (Figure 2D3). Small amounts of PAM-3 in atrium (Figure 2D3) have been reported previously.19 Thus neither proANP-EGFP, nor pro-EGFP–PAM-1, nor endogenous PAM proteins are cleaved to a major degree in our culture system, as anticipated from previous reports.3,13,18,20
Effects of ProANP Mutations on Colocalization With PAM Proteins
We tested the hypothesis that point mutations or total deletion of the calcium-binding sequence of proANP would disrupt the aggregation of proANP and thus inhibit recruitment of PAM proteins. In 47 myocytes (435 vesicles), we examined whether these mutations suppressed the colocalization with PAM-1 (Figure 3A) and in 46 myocytes (469 vesicles), the colocalization with CD-PAM (Figure 3B). A wild-type proANP-EGFP–expressing myocyte is shown in Figure 2A, indicating a high degree of colocalization with PAM-1. Overall, there was a majority of vesicles (83.9%) where proANP-EGFP was colocalized with PAM-1 and a minority (16.1%) where proANP-EGFP was expressed alone (Figure 3A1). Point mutations (Glu23,24Gln23,24) or deletion of the calcium-binding N-terminal 44 aa reduced the proportion of PAM-1 colabeled vesicles to 58.7% (2=10.3; DF=1; P<0.01) and 33.2% (2=38.1; DF=1; P<0.001), respectively (Figure 3A2 and 3A3). Examples for such vesicles are shown in the top panels of Figure 3A. A slight decrease in anti–CD-PAM labeling was observed with the proANP mutations; a majority of vesicles being double-labeled: 92.2% for intact proANP; 89% for Glu23,24-mutated proANP (Figure 3B2); and 84.7% for N-terminal–deleted proANP (Figure 3B3; 2=4.3; DF=1; P<0.05). Because the CD-PAM antiserum labels PAM-1 and PAM-2, an estimate was made whether this decrease in colocalization could also have been attributable to PAM-2. In Western blots, PAM-1 accounts for 27.5±9.6% (mean±SEM, n=3) of PAM-1 plus PAM-2. If recruitment of PAM-1 alone was affected by deletion of N-terminal proANP, it is calculated that CD-PAM should colabel 76.9% of the green fluorescent vesicles, close to the 84.7% observed. The difference is within the margin of error of the measurements.
|
Effect of PHM Inhibitor PBA on Recruitment of PAM-1
We next tested whether PBA29,34 would modify recruitment of PAM-1 to secretory vesicles, in a total of 26 atrial myocytes (834 vesicles) expressing proANP-EGFP and treated for 72 hours with 10 µmol/L PBA or vehicle. Figure 4A1 shows that PBA strongly increased the mean percentage of PAM-1 negative green fluorescent vesicles per cell (2=12.3; DF=1; P<0.01). Figure 4A2 illustrates for individual myocytes that PBA produced a major shift toward PAM-1 negative vesicles, indicating a loss of recruitment of PAM-1. Microphotographs of Figure 4A3 illustrate that in a vehicle-treated myocytes (control), all EGFP-positive vesicles were labeled by anti–PAM-1 antiserum. In the PBA-treated myocyte, a few EGFP-positive vesicles were also labeled by anti–PAM-1 antiserum, but even strongly EGFP-positive vesicles were not stained at all. As a control, Figure 4B shows that PBA did not decrease the maximal immunofluorescence intensity for PAM-1 in TGN and vesicles; thus the difference in PAM-1 immunostaining of the vesicles in Figure 4A was not attributable to major changes in staining procedures. In another control, Western blots show that PBA did not produce a major change in the content of PAM-1 (Figure 4C) nor PHM (Figure 2D3) in atrial culture extracts. PBA had little effect on CD-PAM staining of secretory vesicles. Following PBA treatment, 83.2±7.3% of proANP-EGFP–expressing vesicles per cell were colabeled by CD-PAM antiserum (n=7 myocytes, 276 vesicles), similar to the percentage shown for vesicles expressing N-terminal truncated proANP-EGFP (Figure 3B3).
|
Effect of PBA on Recruitment of Pro-EGFP–PAM-1
In another experimental approach, we tested whether recruitment of pro-EGFP–PAM-1 to proANP vesicles was also affected by PBA. Pro-EGFP–PAM-1 was expressed in atrial myocytes, and these were treated for 3 days with PBA or vehicle (Figure 5). Most, if not all, green fluorescent vesicles were double-labeled for ANP, in both the controls and PBA-treated myocytes, presumably because of the extremely strong signal for ANP immunostaining. An example for a control is shown in Figure 2C; and for a PBA treated myocyte, in Figure 5A. However, PBA significantly changed the distribution of 180 secretory vesicles analyzed in 18 atrial myocytes (Figure 5B1), where the fluorescence intensity of EGFP was plotted as a function of ANP immunostaining. PBA thus reduced by approximately half the green fluorescence intensity relative to ANP (Figure 5B2).
|
Control Experiment Testing the Effect of PBA on Another Secretory System: Exocytosis of KIR2.1
Would PBA indiscriminately affect other secretory systems? As is the case for many transporters and channels, the exocytosis of the potassium inward rectifiers KIR2.x can be regulated (see Zitron et al30 and Malinowska et al31 for citations). The choice of KIR2.1 has the advantage that none of its numerous interacting partners includes PAM, as recently shown.32 Constructs encoding fusion proteins of KIR2.1 with EGFP were electroporated into atrial myocytes (Figure 6). After 3 days, green fluorescence was equally detectable in the plasma membrane of vehicle- or PBA-treated myocytes (examples in Figure 6A1 and 6A2). Quantification with line scans (Figure 6B1 and 6B2) showed there was no significant difference in membrane relative to intracellular green fluorescence intensity between 13 controls and 14 PBA-treated myocytes (Figure 6C1 and 6C2).
|
Effect of PBA on Shape and Docking of ProANP Vesicles in Live Myocytes
We tested in 136 myocytes, on 798 proANP-EGFP vesicles, whether 10 µmol/L PBA would change vesicle shape or docking. The proportion of deformed vesicles in vehicle-treated cells remained stable: 42.2% at 48 hours; 51.4% at 72 hours; and 52.2% at 96 hours (Figure 7A; examples in top panels). After 3 days of treatment, PBA produced a major change in vesicle shape, in a majority of spherical vesicles, to irregular (deformed) vesicles. The proportion of deformed vesicles increased from 52.1% at 48 hours to 82.4% at 72 hours (2=25.5; DF=1; P<0.001) and 73.7% at 96 hours (2=9.1; DF=1; P<0.01). We then tracked the movements of 4813 secretory vesicles in a total of 170 live myocytes (Figure 7B). There was no difference in the proportion of mobile vesicles of controls versus PBA-treated cells during the experiment: 79.9% versus 82.5% at 48 hours; 81.9% versus 84.8% at 72 hours; 85.2% versus 83.5% at 96 hours.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
ProANP, PAM-1, and PAM-2 are highly abundant proteins of the cardiac atrium,3,17–19 contained in the lumen and membrane, respectively, of the same secretory vesicles.35 The reasons for this intriguing, intimate relationship between proANP and PAM have remained unclear. The aims of this work were to test in living myocytes whether the calcium-dependent aggregation of proANP5,6 and the lumenal domain of PAM were required to recruit PAM-1 and PAM-2 to the vesicle membrane and, conversely, whether PAM proteins contributed to the genesis and docking of proANP vesicles. We show that recruitment of PAM-1, but not PAM-2, depends on an intact calcium-binding N-terminal sequence of proANP. We go on to show that PBA, a pharmacological inhibitor of PAM, inhibits recruitment of PAM-1 but not PAM-2, induces a deformation of the secretory vesicles, and has no effect on docking. These novel findings, summarized in Figure 8, were obtained on neonate atrial myocytes in culture, a favorable model for studying the cardiac secretory pathway.13,17–19,36
|
Recruitment of PAM-1 to Secretory Vesicles Is Inhibited by Mutations of the Calcium-Binding Domain of ProANP and by PBA
How do PAM-1 and PAM-2 target the atrial secretory vesicle membrane? At least two possibilities exist, or may coexist. First, the protein kinase PAM C-terminal interacting protein24 can bind tightly to the C-terminal cytosolic juxtamembrane domain of PAM-1 and PAM-2, and phosphorylate PAM-1 on Ser949.37 This may facilitate the aggregation of PAM-1 (see below) and recruitment of PAM-1 to secretory vesicles in AtT-20 cells. Ser949 phosphorylation also favors retrieval of PAM-1 from the plasma membrane onto endocytotic vesicles and/or retention of PAM-1 in the TGN. Second, it is reported that PAM-3, the soluble form of PAM-1 and PAM-2, precipitates either by itself or with lumenal cargo proteins at ionic conditions prevailing in the TGN,25 suggesting that the lumenal domain of PAM-1 could do likewise. From our data and these previous studies,24,25,37 we propose that PAM-1 is recruited to the proANP aggregate, either by coprecipitating with proANP on the surface of the aggregate or by recognizing surface motifs of the aggregate (Figure 8A). We base this suggestion on the following findings. When aggregation of proANP is perturbed by deletion or mutation of its calcium-binding domain,5,6 recruitment of PAM-1 to secretory vesicles is strongly diminished (Figure 3A), as illustrated in Figure 8B. The recruitment of PAM-1 is also diminished by PBA, as shown by two different experimental approaches. First, PBA reduces the colocalization of proANP-EGFP and PAM-1 on secretory vesicles (Figure 4A); and second, PBA reduces the colocalization of pro-EGFP–PAM-1 and ANP on secretory vesicles (Figure 5B). Interestingly, PBA has no effect on the exocytosis of KIR2.1 (Figure 6), suggesting that the actions of PBA may be confined to PAM. The mechanism of action of PBA in our experiments is not yet known. It is not established whether PAM-1 is functionally active within the TGN or the secretory vesicles, the more so because a substrate has not yet been identified. Changes in enzymatic activity of PHM in extracts were not measured here, because we would not have been able to distinguish between any changes occurring in TGN or secretory vesicles; the purification of secretory vesicles from atrial cell cultures has remained elusive so far. As an alternative possibility, the action of PBA on PHM may alter the conformation of PAM-1 (illustrated in Figure 8C) and thus inhibit the interaction between PHM and N-terminal proANP.
The recruitment of PAM-2 apparently does not depend on the aggregation of proANP. This is based on the finding that the staining of secretory vesicles with CD-PAM antiserum decreases only slightly with deletion or mutation of the N-terminal sequence of proANP (Figure 3B). The CD-PAM antiserum recognizes both PAM-1 and PAM-2. This report shows that PAM-1 accounts for 27.5% of the total amount of PAM-1 and PAM-2 in atrial myocytes, confirming previous studies.19,35,38 The slight decrease in staining can thus be attributed entirely to the loss of recruitment of PAM-1. The cytosolic routing motifs of PAM-2 (citations in24) may be sufficient for targeting PAM-2 to secretory vesicles, irrespective of potential interactions with proANP. The difference between PAM-1 and PAM-2 is the exon A sequence in PAM-1, located downstream of the PHM domain. It is speculated that the presence of exon A facilitates the interaction between PAM-1 and proANP. A potential influence of exon A on the conformation of PHM would explain why PBA inhibits recruitment of PAM-1 but not PAM-2 to secretory vesicles, even though PBA has the same site of action (PHM) in PAM-1 and PAM-2. Alternatively, exon A might interact with proANP; the effect of PBA would then be explained by a PBA-induced change in conformation of exon A. Further experiments are needed to clearly define the protein-protein interactions between proANP and PAM.
Role of PAM in Shaping Atrial Secretory Vesicles
Mutation or deletion of the calcium-binding N-terminal proANP results in deformation of atrial secretory vesicles.13 One explanation is that coexpressed N-terminal–mutated or N-terminal–deleted proANP would be unable to precipitate with endogenous proANP, thus strongly diminishing the interaction between proANP and a hypothetical integral vesicle membrane protein.13 As a consequence, packaging of the vesicle core by TGN membrane would be deranged, and the vesicle would be deformed. The new results support this hypothesis. Figure 7A shows that a 3-day treatment with PBA causes a striking increase in deformed secretory vesicles, suggesting that PBA, by perturbing the interaction between PAM-1 and proANP (see above) and diminishing the recruitment of PAM-1 (Figure 4A), deranges the ordered packaging of the vesicle core (illustrated in Figure 8C). Alternatively, PBA may inhibit the conversion of an unknown substrate to a product that participates in shaping the secretory vesicle. Recently, proteins have been identified, endowed with a so-called BAR domain, that impose curvature on biological membranes and thus shape the nascent vesicle.39 A tyrosine phosphatase, protein tyrosine phosphatase-MEG2, has been found whose overexpression causes a striking enlargement of secretory vesicles in rat basophilic mast cells and Jurkat T leukemia cells.40 To what degree integral vesicle transmembrane proteins may interact with PHM products, BAR (Bin/Amphiphysin/Rvs) domain proteins, or tyrosine phosphatases to shape the secretory vesicles remains open to investigation.
Is PAM Involved in Vesicle Docking?
Our previous study13 showed that deletion or mutation of N-terminal proANP not only deformed the secretory vesicles but also strongly diminished docking (illustrated in Figure 8B). Vesicle docking was thought to result from an interaction of docking receptors with plasma membrane docking proteins, and the lack of docking was attributed to suppression of recruitment of docking receptors. To test whether PAM could represent this docking receptor, vesicles were analyzed after 2 to 4 days of PBA treatment. This clearly reduced recruitment of PAM-1 and deformed a majority of secretory vesicles but had no effect on vesicle docking (Figure 7B), suggesting that PAM-1 is not likely to be involved in vesicle docking. So far, PAM-2 also is an unlikely candidate, because recruitment of PAM-2 is not affected by mutations of N-terminal proANP that are associated with decreased docking.13 A recent study on atrial secretory vesicle associated proteins38 should open the way to exploring several new candidates.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
PAM proteins have long been suspected to play a role in the secretory pathway of the heart atrium and to be recruited to secretory vesicles via their C-terminal domain19,23,35 (also see the introduction). In this study, we show that PAM-1 is recruited to the atrial vesicles via an additional mechanism, most likely via interactions of its lumenal domain with the calcium-binding domain of proANP. PAM-1 is involved in shaping the ANP-secretory vesicles, because the lack of recruitment of PAM-1 is associated with a high proportion of deformed ANP-secretory vesicles. The results further suggest that proteins distinct from PAM are involved in docking.
|
Acknowledgments |
|---|
This study was supported by Swiss National Science Foundation grant 31-066838.01, the Swiss University Conference project "Heart Remodeling in Health and Disease," the Carlos and Elsie de Reuter Foundation, the Geneva Academy of Sciences, the Novartis Research Foundation, the Swiss Heart Foundation, and the Gustave Prevot Foundation. Dr Robert Mains (University of Connecticut, Farmington) is gratefully acknowledged as the source of anti-PAM antisera and of the PAM-1 cDNA used in this study. We thank the Institute of Molecular Cardiobiology, Johns Hopkins Medical School (Baltimore, Md), for the EGFP-KIR2.1 construct and Dr Alfred L. George (Vanderbilt University, Nashville, Tenn) for the pCMV-Script-KCNJ2-R67W construct.
|
Footnotes |
|---|
Original received March 22, 2004; resubmission received July 2, 2004; revised resubmission received October 21, 2004; accepted November 2, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
1. de Bold AJ, Flynn TG. Cardionatrin I—a novel heart peptide with potent diuretic and natriuretic properties. Life Sci. 1983; 33: 297–302.[CrossRef][Medline] [Order article via Infotrieve]
2. Wilkins MR, Redondo J, and Brown LA. The natriuretic-peptide family. Lancet. 1997; 349: 1307–1310.[CrossRef][Medline] [Order article via Infotrieve]
3. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med. 1998; 339: 321–328.
4. Jiao JH, Baertschi AJ. Neural control of the endocrine rat heart. Proc Natl Acad Sci U S A. 1993; 90: 7799–7803.
5. Thibault G, Doubell AF. Binding and aggregation of pro-atrial natriuretic factor by calcium. Am J Physiol. 1992; 2: C907–C915.
6. Canaff L, Brechler V, Reudelhuber TL, Thibault G. Secretory granule targeting of atrial natriuretic peptide correlates with its calcium-mediated aggregation. Proc Natl Acad Sci U S A. 1996; 93: 9483–9487.
7. Jamieson JD, Palade GE. Specific granules in atrial muscle cells. J Cell Biol. 1964; 23: 151–172.
8. Gerdes H-H, Glombik MM. Signal-mediated sorting to the regulated pathway of protein secretion. Anat Anz. 1999; 181: 447–453.[Medline] [Order article via Infotrieve]
9. Arvan P, Castle D. Sorting and storage during secretory granule biogenesis: looking backward and looking forward. Biochem J. 1998; 332: 593–610.[Medline] [Order article via Infotrieve]
10. Thiele C, Huttner WB. Protein and lipid sorting from the trans-Golgi network to secretory granules: recent developments. Semin Cell Dev Biol. 1998; 9: 511–516.[CrossRef][Medline] [Order article via Infotrieve]
11. Tooze SA. Biogenesis of secretory granules in the trans-Golgi network of neuroendocrine and endocrine cells. Biochim Biophys Acta. 1998; 1404: 231–244.[Medline] [Order article via Infotrieve]
12. Kuiper RP, Martens GJM. Prohormone transport through the secretory pathway of neuroendocrine cells. Biochem Cell Biol. 2000; 78: 289–298.[CrossRef][Medline] [Order article via Infotrieve]
13. Baertschi AJ, Monnier D, Schmidt U, Levitan ES, Fakan S, Roatti A. Acid prohormone sequence determines size, shape, and docking of secretory vesicles in atrial myocytes. Circ Res. 2001; 89: E23–E29.[CrossRef][Medline] [Order article via Infotrieve]
14. Bonifacino JS, Glick BS. The mechanisms of vesicle budding and fusion. Cell. 2004; 116: 153–166.[CrossRef][Medline] [Order article via Infotrieve]
15. Bennett MK, Scheller RH. The molecular machinery for secretion is conserved from yeast to neurons. Proc Natl Acad Sci U S A. 1993; 90: 2559–2563.
16. Prigge ST, Mains RE, Eipper BA, Amzel LM. New insights into copper monooxygenases and peptide amidation: structure, mechanism and function. Cell Mol Life Sci. 2000; 57: 1236–1259.[CrossRef][Medline] [Order article via Infotrieve]
17. Eipper BA, May V, Braas KM. Membrane-associated peptidylglycine alpha-amidating monooxygenase in the heart. J Biol Chem. 1988; 263: 8371–8379.
18. Stoffers DA, Green CB, Eipper BA. Alternative mRNA splicing generates multiple forms of peptidyl-glycine alpha-amidating monooxygenase in rat atrium. Proc Natl Acad Sci U S A. 1989; 86: 735–739.
19. Maltese JY, Eipper BA. Developmental expression of peptidylglycine alpha-amidating monooxygenase (PAM) in primary cultures of neonatal rat cardiocytes: a model for studying regulation of PAM expression in the rat heart. Mol Endocrinol. 1992; 6: 1998–2008.
20. Braas KM, Stoffers DA, Eipper BA, May V. Tissue specific expression of rat peptidylglycine alpha-amidating monooxygenase activity and mRNA. Mol Endocrinol. 1989; 3: 1387–1398.
21. Bell-Parikh LC, Eipper BA, Mains RE. Response of an integral granule membrane protein to changes in pH. J Biol Chem. 2001; 276: 29854–29863.
22. Alam MR, Steveson TC, Johnson RC, Back N, Abraham B, Mains RE, Eipper BA. Signaling mediated by the cytosolic domain of peptidylglycine alpha-amidating monooxygenase. Mol Biol Cell. 2001; 12: 629–644.
23. Steveson TC, Keutmann HT, Mains RE, Eipper BA. Phosphorylation of cytosolic domain Ser(937) affects both biosynthetic and endocytic trafficking of peptidylglycine alpha-amidating monooxygenase. J Biol Chem. 1999; 274: 21128–21138.
24. Alam MR, Caldwell BD, Johnson RC, Darlington DN, Mains RE, Eipper BA. Novel proteins that interact with the COOH-terminal cytosolic routing determinants of an integral membrane peptide-processing enzyme. J Biol Chem. 1996; 271: 28636–28640.
25. Colomer V, Kicska GA, Rindler MJ. Secretory granule content proteins and the luminal domains of granule membrane proteins aggregate in vitro at mildly acidic pH. J Biol Chem. 1996; 271: 48–55.
26. Jiao JH, Baumann P, Baron A, Roatti A, Pence RA, Baertschi AJ. Sulfonylurea receptor ligands modulate stretch-induced ANF secretion in rat atrial myocyte culture. Am J Physiol. 2000; 278: H2028–H2038.
27. Burke NV, Han W, Li D, Takimoto K, Watkins SC, Levitan ES. Neuronal peptide release is limited by secretory granule mobility. Neuron. 1997; 19: 1095–1102.[CrossRef][Medline] [Order article via Infotrieve]
28. Howell SL. The mechanism of insulin secretion. Diabetologia. 1984; 26: 319–327.[Medline] [Order article via Infotrieve]
29. Driscoll WJ, Konig S, Fales HM, Pannell LK, Eipper BA, Mueller GP. Peptidylglycine-alpha-hydroxylating monooxygenase generates two hydroxylated products from its mechanism-based suicide substrate, 4-phenyl-3-butenoic acid. Biochemistry. 2000; 39: 8007–8016.[CrossRef][Medline] [Order article via Infotrieve]
30. Zitron E, Kiesecker C, Lück S, Kathöfer S, Thomas D, Kreye VAW, Kiehn J, Katus HA, Schoels W, Karle CA. Human cardiac inwardly rectifying current IKIR2.2 is upregulated by activation of protein kinase A. Cardiovasc Res. 2004; 63: 520–527.
31. Malinowska DH, Sherry AM, Tewari KP, Cuppoletti J. Gastric parietal cell secretory membrane contains PKA- and acid-activated KIR2.1 K+ channels. Am J Physiol Cell Physiol. 2004; 286: C495–C506.
32. Leonoudakis D, Conti LR, Radeke CM, McGuire LM, Vandenberg CA. A multiprotein trafficking complex composed of SAP97, CASK, Veli, and Mint1 is associated with inward rectifier Kir2 potassium channels. J Biol Chem. 2004; 279: 19051–19063.
33. Maltese JY, Eipper BA. Maturation, internalization, and turnover of soluble and membrane proteins associated with atrial myocyte secretory granules. Endocrinology. 1993; 133: 2579–2587.
34. Mueller GP, Driscoll WJ, Eipper BA. In vivo inhibition of peptidylglycine-alpha-hydroxylating monooxygenase by 4-phenyl-3-butenoic acid. J Pharmacol Exp Ther. 1999; 290: 1331–1336.
35. O’Donnell PJ, Driscoll WJ, Back N, Muth E, Mueller GP. Peptidylglycine-alpha-amidating monooxygenase and pro-atrial natriuretic peptide constitute the major membrane-associated proteins of rat atrial secretory granules. J Mol Cell Cardiol. 2003; 35: 915–932.[CrossRef][Medline] [Order article via Infotrieve]
36. Slot JW, Garruti G, Martin S, Oorschot V, Posthuma G, Kraegen EW, Laybutt R, Thibault G, James DE. Glucose transporter (GLUT-4) is targeted to secretory granules in rat atrial cardiomyocytes. J Cell Biol. 1997; 137: 1243–1254.
37. Steveson TC, Zhao GC, Keutmann HT, Mains RE, Eipper BA. Access of a membrane protein to secretory granules is facilitated by phosphorylation. J Biol Chem. 2001; 276: 40326–40337.
38. Muth E, Driscoll WJ, Smalstig A, Goping G, Mueller GP. Proteomic analysis of rat atrial secretory granules: a platform for testable hypotheses. Biochim Biophys Acta. 2004; 1699: 263–275.[Medline] [Order article via Infotrieve]
39. Peter BJ, Kent HM, Mills IG, Vallis Y, Butler PJ, Evans PR, McMahon HT. BAR domains as sensors of membrane curvature: the amphiphysin BAR structure. Science. 2004; 303: 495–499.
40. Wang X, Huynh H, Gjorloff-Wingren A, Monosov E, Stridsberg M, Fukuda M, Mustelin T. Enlargement of secretory vesicles by protein tyrosine phosphatase PTP-MEG2 in rat basophilic leukemia mast cells and Jurkat T cells. J Immunol. 2002; 168: 4612–4619.
41. Andelfinger G, Tapper AR, Welch RC, Vanoye CG, George AL Jr, Benson DW. KCNJ2 mutation results in Andersen syndrome with sex-specific cardiac and skeletal muscle phenotypes. Am J Hum Genet. 2002; 71: 663–668.[CrossRef][Medline] [Order article via Infotrieve]
42. Holt JR, Johns DC, Wang S, Chen ZY, Dunn RJ, Marban E, Corey DP. Functional expression of exogenous proteins in mammalian sensory hair cells infected with adenoviral vectors. J Neurophysiol. 1999; 81: 1881–1888.
This article has been cited by other articles:
 |
|
 |
|
  P. Philip-Couderc, N. I. Tavares, A. Roatti, R. Lerch, C. Montessuit, and A. J. Baertschi Forkhead Transcription Factors Coordinate Expression of Myocardial KATP Channel Subunits and Energy Metabolism Circ. Res., February 1, 2008; 102(2): e20 - e35. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Beauchamp, K. A. Yamada, A. J. Baertschi, K. Green, E. M. Kanter, J. E. Saffitz, and A. G. Kleber Relative Contributions of Connexins 40 and 43 to Atrial Impulse Propagation in Synthetic Strands of Neonatal and Fetal Murine Cardiomyocytes Circ. Res., November 24, 2006; 99(11): 1216 - 1224. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||