Supramaximal cholecystokinin displaces Munc18c from the pancreatic acinar basal surface, redirecting apical exocytosis to the basal membrane

Abstract

Exocytosis at the apical surface of pancreatic acinar cells occurs in the presence of physiological concentrations of cholecystokinin (CCK) but is inhibited at high concentrations. Here we show that Munc18c is localized predominantly to the basal membranes of acinar cells. Supramaximal but not submaximal CCK stimulation caused Munc18c to dissociate from the plasma membrane, and this displacement was blocked by protein kinase C (PKC) inhibitors. Conversely, whereas the CCK analog CCK-OPE alone failed to displace Munc18c from the membrane, this agent caused Munc18c displacement following minimal PKC activation. To determine the physiological significance of this displacement, we used the fluorescent dye FM1-43 to visualize individual exocytosis events in real-time from rat acinar cells in culture. We showed that supramaximal CCK inhibition of secretion resulted from impaired apical secretion and a redirection of exocytic events to restricted basal membrane sites. In contrast, CCK-OPE evoked apical exocytosis and could only induce basolateral exocytosis following activation of PKC. Infusion of supraphysiological concentrations of CCK in rats, a treatment that induced tissue changes reminiscent of mild acute pancreatitis, likewise resulted in rapid displacement of Munc18c from the basal membrane in vivo.

Figure 1

Localization of Munc18c, syntaxin-4, and SNAP-23 in dispersed rat pancreatic acinar cells. (a) Western blots of purified pancreatic acinar plasma membranes (PM) and ZG membranes (ZGM). Highly purified membranes (10 μg protein) were separated on a 15% SDS-PAGE and immunoblotted with anti-Munc18c, –syntaxin-4, –SNAP-23, and –VAMP-2 antibodies. (b) Immunoprecipitation of Munc18c from purified pancreatic acinar plasma membranes (0.5 mg), and coimmunoprecipitation of syntaxin-4 (lane 1). Immunoprecipitation of total lysates (10-cm dish) of 3T3-L1 adipocytes were used as a positive control (25, 26) (lane 2). (c–h) Laser confocal microscopy of dispersed acinar cells. Acinar cells were plated on coverslips and probed with (c) anti-Munc18c, (e) anti–syntaxin-4, or (g) anti–SNAP-23 antibodies, along with double-labeling with FITC-phalloidin in d, f, and h, respectively, to indicate the plasma membrane staining and concentration of actin in the apical portion of the acinar cells.

Figure 2

Effects of secretagogues on the cellular localization of Munc18c in dispersed rat pancreatic acini. The dispersed acini were stimulated by the following protocols and were then all labeled with anti-Munc18c antibody except for h and i, which were labeled with anti–syntaxin-4 and anti–SNAP-23 antibodies as indicated in the figure. The cellular locations of these proteins were then visualized by confocal microscopy as described in Methods. (a) Submaximal CCK-8 (30 pM, 1 hour); (b, h, and i) supramaximal CCK-8 (10 nM, 15–30 minutes); (c) supramaximal CCK-OPE (1 μM, 1 hour); (d) 1 nM TPA (30 minutes); (e) 1 μM TPA (30 minutes); (f) 1 nM TPA (15 minutes) + 1 μM CCK-OPE (30 minutes); (g) calphostin C (500 nM, 40 minutes) + 10 nM CCK-8 (30 minutes). (j) Acini were stimulated with 10 nM CCK-8 at the indicated times and then fractionated into plasma membrane (PM) and cytosol (Cyt) fractions (see Methods). Thirty micrograms protein of the plasma membrane fractions and 40 μg of the cytosol fractions were separated on SDS-PAGE and immunoblotted with anti-Munc18c and –syntaxin-4 antibodies. This is a representative of three independent experiments.

Figure 3

Confocal microscopy of a large acinus showing a 15-minute time course of stimulation by 1 μM CCK-OPE. This is a projection of a series of Z-sections across the equatorial plane, which were digitally collapsed using Adobe Photoshop software (Adobe Systems Inc., San Jose, California, USA) to capture the vesicles of adjoining confocal planes. Arrow 1 indicates the acinar ductal lumen; arrow 2 points to its “neck” as it exits out of the acinus, which is marked by arrow 3. Note that the ductal zymogen protein contents are stained with FM1-43, which fills the ductal lumen. Here, 1 μM CCK-OPE stimulation caused a time-dependent (1–6 minutes) increase in fluorescence within the ductal lumen and an increase in the diameter of the lumen. Note tubular structures branching off from the ductal lumen that seem to extend into the apical surfaces of the acinar cells. At 6 minutes, the ductal lumen decreased in size, FM1-43 fluorescence intensity (compared with 4 minutes) decreased, and some of the tubular structures shrank or collapsed. At 15 minutes, the duct was “reduced” in size to resting levels. Note small FM1-43 fluorescent puncta inside the acinar cells at 15 minutes. Bar = 20 μm.

Figure 4

Fluorescence microscopy of 1 μM CCK-OPE–stimulated apical exocytosis. (a) A 3D-reconstructed image of 0.6-μm confocal cuts of the full thickness of a FM1-43–labeled triplet acinus stimulated by 1 μM CCK-OPE for 5 minutes (right). The phase contrast image is also shown (left). (b) Epifluorescence FM1-43 staining of another triplet acinus similarly stimulated with 1 μM CCK-OPE. Shown are the phase contrast image (left), basal fluorescent levels (middle), and the fluorescent image after 1 minute of stimulation (right). (c) The same acinus as in b where images were acquired at 1 frame per second. Regions of interest were drawn as indicated and graphed to show the kinetics of fluorescent changes in real time at these regions — the ZG pole, the apex, and a representative basal plasma membrane region. (d) A graphical analysis of these fluorescent changes (16 cells from doublets to four-cell acini from four experiments) obtained at peak levels. (e–j) The washout of FM1-43 fluorescence from an acinus stimulated with 1 μM CCK-OPE to show that the fusion pores of ZGs that have undergone exocytotic fusion remained open. (e) A phase contrast of the acinus. (f) Epifluorescent microscopy of FM1-43 fluorescence of this acinus stimulated by 1 μM CCK-OPE for 5 minutes. This acinus was then subjected to four washes (g–j) with FM1-43–free media. Note that with each wash, the FM1-43 fluorescence progressively diminished from the innermost portions of the ZG poles of the acinus toward the apical lumen (first [g] to third [i] washes), and finally the apical lumenal fluorescence was washed off (fourth wash [j]). Bar = 20 μm.

Figure 5

Supramaximal CCK (10 nM) stimulation redirects exocytosis from the apical to the basal plasma membrane. (a) A confocal section across the equatorial plane of a four-cell acinus stained with FM1-43, which was stimulated by 10 nM CCK-8 for 1, 5, and 15 minutes compared with the basal state. Note that most of the membrane hotspots were already present in the cells prior to stimulation (indicated by filled arrowheads) and then increased in intensity and size after stimulation. De novo hotspots not previously present also appeared with CCK stimulation (open arrowheads in the 1-minute image). Bar = 20 μm. (b and c) Epifluorescence microscopy of a small triplet-cell pancreatic acinus stimulated by 10 nM CCK-8. Real-time changes in fluorescence were obtained at 1 frame/sec. (b) The fluorescence images at 0 time, 40 seconds, 2 minutes, and 10 minutes of stimulation. (c) Regions of interest were drawn as indicated on this diagram of the acinus shown in b. The left fluorescence tracing shows two representative hotspots that were present at basal state (filled arrowheads). The right fluorescence tracing shows two de novo hotspots (open arrowheads, see 40-second image in b). (d and e) Graphical summaries of the peak fluorescence data from (d) epifluorescence studies (12 cells, four experiments as in c), including the two populations of basal plasma membrane (B-PM) hotspots (n = 26 each, indicated by filled and open arrowheads), and (e) confocal studies (13 cells, three experiments as in a) of 10 nM CCK-8–evoked basal membrane FM1-43 exocytosis (n = 18 hotspots) at 1, 5, and 15 minutes of stimulation.

Figure 6

Supramaximal CCK-evoked basal membrane exocytosis by a PKC-mediated pathway. (a) A confocal section across the equatorial place of a four-cell acinus stained with FM1-43, which was preincubated with 1 nM TPA for 15 minutes 37°C and then combined with 1 μM CCK-OPE. Images were taken during the 1 nM TPA preincubation, and the fluorescence was identical to basal levels before exposure to TPA (not shown). Many of the basal membrane hotspots were already present prior to CCK-OPE stimulation and then increased in intensity and size after stimulation (filled arrowheads); only a few more de novo membrane hotspots appeared later (open arrowheads). Bar = 20 μm. (b) Epifluorescence microscopy (1 frame per second) performed on a doublet-cell acinus subjected to the same protocol of preincubation with 1 nM TPA, followed by addition of 1 μM CCK-OPE. A phase contrast image of the acinus and the regions of interest drawn on this acinus for analysis are shown. 0 time is when 1 μM CCK-OPE was added into the 1 nM TPA-containing media. The 15-minute TPA preincubation is not shown. Note the synchronous increase in the fluorescence intensity of the basal plasma membrane hotspots (B-PM1 and B-PM2); in contrast, there was no change in the FM1-43 fluorescence intensity in the apex or ZG poles of the acinus. (c) A graphical summary of the peak fluorescence data obtained from several confocal experiments performed as in a (n = 15 basal membrane hotspots from 9 cells; three separate experiments).

Figure 7

Supramaximal CCK (100 nM)-evoked basal membrane FM1-43 hotspots persist and immunostain precisely with exocytosed amylase at the basal plasma membrane. The acini were stimulated with 100 nM CCK for 20 minutes (37°C), followed by costaining with anti-amylase antibody (red) and FM1-43 (green) as described in Methods. The phase contrast image of the acinar cells is shown to demonstrate their health after the CCK stimulation. Superimposition of the anti-amylase and FM1-43 shows a precise colocalization (yellow spots) of most of the basal membrane FM1-43 hotspots and amylase. Bar = 20 μm.

Figure 8

Confocal microscopy localization of Munc18c and actin and histology of pancreatic tissues from rats treated with supramaximal cerulein concentrations. Rats were intravenously infused with saline control (a–c) or supramaximal cerulein (10 μg/kg/h) for 10 minutes (d–f), 1 hour (g–i), 2 hours (j–l), or 4 hours (m–o). The rats were immediately perfused with a fixative and the pancreata removed and prepared for confocal microscopy (see Methods) and histology (hematoxylin and eosin stain). For the confocal studies, the pancreatic tissue sections were labeled with anti-Munc18c (a, d, g, and j) and double-labeled with FITC-phalloidin (b, e, h, and k). Note the progressive displacement of Munc18c and its diminished levels in the pancreatic acinar plasma membranes. These changes of Munc18c correlate to the progressive actin disassembly and the formation of cytoplasmic vacuoles (arrows in f and i) in the acinar cells. The pancreatic acinar cells nonetheless remain completely intact (f, i, and l). (m–o) Pancreata from rats treated with supramaximal cerulein for 4 hours. Treatment did not affect syntaxin-4 (m) or SNAP-23 (o) plasma membrane locations but did cause actin disassembly (n).

Figure 9

Minimal PKC activation in vivo causes CCK-OPE to mimic cerulein in inducing pancreatitis in rats. Rats were intraperitoneally injected with 50–100 μl of saline control (a and b), low TPA (1 nmol/kg) (c and d), supramaximal CCK-OPE (10 μg/kg) (e and f), low TPA (1 nmol/kg) + CCK-OPE (10 μg/kg) (g and h), or supramaximal cerulein (10 μg/kg) concentrations (i and j). Four hours after the injections of these reagents, the rats were treated and the pancreata removed as in Figure 8 (and in Methods). A portion of each pancreas was prepared for confocal microscopy studies (a, c, e, g, and i) to determine Munc18c localization, and another portion was prepared for histologic studies (b, d, f, h, and j) (hematoxylin and eosin staining) to determine the early morphologic features of mild edematous pancreatitis. In the low TPA + CCK-OPE protocol, the rat was first injected with 1 nmol/kg TPA and, after 10 minutes, was injected with CCK-OPE. Note the similar changes in the TPA + CCK-OPE–treated (g and h) and the cerulein-treated (i and j) pancreatic tissues, including Munc18c disruption (arrows) and diminished plasma membrane levels (arrowheads) in g and i, and the formation of small and large cytoplasmic vacuoles (arrows) in h and j. These changes caused by supramaximal cerulein (i and j) were, however, more marked and generalized than the changes in the low TPA + CCK-OPE–treated tissues (g and h).