Abstract
Using antibodies specific for pro-opiomelanocortin (POMC), amidated joining peptide (JP), and the prohormone convertase PC1, we showed immunocytochemically that PC1 in a corticotrophic tumor cell line, AtT-20, was co-localized either with POMC or with amidated JP in secretory granules, and also confirmed that POMC was cleaved mainly in secretory granules. Analysis using DAMP (3- [2,4-dinitroanilino]-3'-amino-N-methyldipropylamine) as the pH probe suggested a correlation between POMC processing and acidic pH in the secretory granules. Bafilomycin A1, a specific inhibitor of vacuolar-type H+-AT-Pase, completely inhibited POMC processing and caused constitutive secretion of the unprocessed precursor. By contrast, chloroquine, a weak base that is known to neutralize acidic organelles, was unable to inhibit POMC processing. Electron microscopic analysis revealed that, in AtT-20 cells treated with bafilomycin A1, the trans-Golgi cisternae were dilated and few secretory granules were present in the cytoplasm. These observations suggest that acidic pH provides a favorable environment for proteolytic processing of POMC by PC1 but is not required, and that integrity of the trans-Golgi network and sorting of POMC into secretory granules are important for POMC processing.
Keywords
I
Previous studies have shown that acidification of the TGN or secretory granules plays an important role in the proteolytic processing of prohormones, including POMC, and in the secretion of prohormone-derived peptides (Anderson and Orci 1988; Orci et al. 1986, 1987a,b; Moore et al. 1983), compatible with the fact that the PC1 and PC2 enzymes show the highest activity at relatively acidic pH. However, there have been other studies showing that weak bases, such as chloroquine and ammonium chloride, dissipate pH gradients but have no discernible effect on prohormone processing (Nagahama et al. 1991; Mains and May 1988).
To address the above issues, we first examined the subcellular localization of POMC, a processed product (amidated JP), and PC1 in a corticotrophic tumor cell line, AtT-20, using antibodies specific for a POMC cleavage site, amidated JP, and PC1. Second, we examined the relationship between acidification and POMC processing in the secretory granules using a basic congener of dinitrophenol (DNP), DAMP (3-[2,4-dinitro-anilino]-3'-amino-N-methyldipropylamine), as the pH probe. Finally, we examined the effects of bafilomycin A1, a specific inhibitor of vacuolar-type H+-ATPase (V-ATPase) (Bowman et al. 1988) on the processing of POMC and the morphology of intracellular compartments.
Materials and Methods
Cell Culture
AtT-20/D-16 cells were grown in Dulbecco's modified Eagle's medium supplemented with glucose, Hepes (4.76 g/liter), 10% fetal calf serum, penicillin (100K U/liter), and streptomycin (100 mg/liter) at 37C in 5% CO2. For analysis of immunoprecipitation and electron microscopy, cells were seeded at 1 × 106 cells per 3-cm tissue culture dish. For immunofluorescence, cells were seeded at 2 × 103 − 104 cells/ml on poly-
Antibodies
Antibodies used in this study have been prepared in rabbits and characterized as described previously: anti-human ACTH[1-39] serum (Tanaka and Kurosumi 1992); anti-POMC (ST-1), which was raised against the nonapeptide corresponding to the cleavage site between murine ACTH and β-LPH (Tanaka et al. 1991); anti-amidated JP (ST-3) against the C-terminal seven amino acids of murine amidated JP (Tanaka and Kurosumi 1992); anti-PC1 serum (ST-28) and anti-PC2 serum (ST-29) against synthetic peptides covering amino acids 442–459 and 613–629 of the respective mouse proteins (Tanaka et al. 1996).
Immunofluorescence
AtT-20 cells were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, and free aldehyde was quenched by incubating the fixed cells with PBS containing 50 mM NH4Cl. To obtain intense immunostaining for PC1 or PC2, the cells were then treated with Gomori's oxidation mixture according to Marcinkiewicz et al. (1993). After permeabilization with PBS containing 0.1% saponin and 0.5% BSA, the cells were sequentially incubated with 20% normal goat serum, the first antibody (anti-POMC, anti-amidated JP, anti-PC1, or anti-PC2; 1:500), and a mixture of rhodamine-labeled affinity-purified goat anti-rabbit IgG (Jackson Immunore-search; West Grove, PA) and DAPI (4',6-diamidino-2-phenylindole dihydrochloride) (Boehringer; Mannheim, Germany) as described previously (Kayo et al. 1996). The cells were examined with an Olympus BX50 microscope equipped with a BX-fluorescence attachment (Olympus Optical; Tokyo, Japan).
Immunogold Electron Microscopy
AtT-20 cells were fixed with a mixture of 0.5% glutaraldehyde, 2% paraformaldehyde, and 1% picric acid in 0.1 M cacodylate buffer, pH 7.4, for 2 hr at 4C. The cells embedded in agarose were dehydrated in ethanol, embedded in LR White (London Resin; Basingstoke, UK), and then sectioned on a Reichert Ultracut-E (Reichert–Jung; Vienna, Austria). For immunolabeling of PC1 or PC2, ultrathin sections were subjected to Gomori's oxidation as described above. Double immunogold labeling of the sections with different antibodies was performed as described previously (Tanaka et al. 1996). Briefly, two faces of the grids were incubated with different antibodies (anti-POMC, 1:500; anti-amidated JP, 1:10,000; anti-PC1, 1:500; anti-PC2, 1:500) and then with anti-rabbit IgG conjugated with different sizes of gold particles (10-nm and 5-nm) (BioCell; Cardiff, UK). The immuno-labeled sections were fixed with 1% osmium tetroxide, and then stained with a mixture of uranyl acetate and methyl cellulose according to a published protocol (Roth et al. 1990). They were examined with a JEM 1200 EX electron microscope at 80 kV.
Measurement of pH
The pH of the interior organelles in AtT-20 cells was measured by the DAMP method by Orci et al. (1986), with a slight modification. Briefly, AtT-20 cells were incubated at 37C for 1 hr in the presence or absence of 10 nM bafilomycin A1. Then, DAMP (Dojin; Kumamoto, Japan) was added at a final concentration of 60 (μM and the cells were further incubated for 1 hr. After three washings with PBS, the cells were fixed with a mixture of 0.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4, for 1 hr, dehydrated in ethanol, and then embedded in the LR White resin. One face of each ultrathin section was treated with anti-DNP (1:300; BioMakor, Rehovot, Israel), followed by 5-nm anti-rabbit IgG-gold complexes. The other face was treated with either anti-POMC or anti-amidated JP, followed by 10-nm anti-rabbit IgG–gold complexes. The pH of a given organelle was estimated using the formula of Orci et al. (1986),
where N is the density of DAMP-specific gold particles in a pH 7.0 compartment (the nucleus) and L is the density of DAMP-specific gold particles in an organelle. Specimens containing AtT-20 cells from three different culture dishes were used for quantitative analysis. The number of gold particles per organelle was directly calculated from micrographs printed at a final magnification of × 100,000. The area of the organelle was measured with NIH Image software (version 1.47) on an Apple Macintosh computer. Statistical analysis was performed using the Cochran-Cox test. The statistical significance of the correlation between the accumulation of DAMP and either POMC or amidated JP labeling was evaluated using Spearman's rank correlation. p>0.05 was considered significant.
Immunoprecipitation
Immunoprecipitation was performed as described previously (Tanaka and Kurosumi 1992; Tanaka et al. 1991). Briefly, AtT-20 cells preincubated for 30 min in methionine-free medium were labeled for 30 min with [35S]-methionine (50 μCi) (TRAN 35S-LABEL; ICN Radiochemicals, Irvine, CA) and chased for 1.5 hr in normal medium. When indicated, bafilomycin A1 (Wako Pure Chemicals; Tokyo, Japan) or chloroquine (Sigma Chemical; St Louis, MO) was included through the preincubation, labeling, and chase periods. The cells and media were collected separately. The cell lysates and media were immunoprecipitated with anti-human ACTH[1–39] serum and protein A-Sepharose (Pharmacia LKB; Uppsala, Sweden). The precipitates were then analyzed using 10–20% SDS-polyacrylamide gel electrophoresis, followed by fluorography.
Conventional Electron Microscopy
AtT-20 cells incubated for either 30 min or 2.5 hr at 37C in the presence or absence of 10 nM bafilomycin A1 or 200 μM chloroquine were fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4, for 1 hr at 4C and postfixed for 1 hr at 4C in 1% osmium tetroxide, reduced with 1.5% potassium ferrocyanide, in the same buffer. After further washes with the above buffer, the cells were collected and embedded in agarose. The agarose blocks were dehydrated in ethanol, embedded in an Epon/Araldite mixture, and then the ultrathin sections were stained with uranyl acetate and lead citrate.
Results
Conversion of POMC to Mature Peptides Occurs Mainly in the Secretory Granules
To determine the intracellular sites at which POMC processing occurs in AtT-20 cells, we first examined the subcellular distribution of POMC and its processed peptides at the light microscopic level with indirect immunofluorescence. To do this, two antibodies that we had previously prepared were used: anti-POMC (ST-1), which was raised against a synthetic non-apeptide corresponding to the murine ACTH-β-LPH cleavage site and is able to recognize POMC but not its processed peptides (Tanaka and Kurosumi 1992; Tanaka et al. 1991); and anti-amidated JP (ST-3), which was against the seven C-terminal amino acids of amidated JP and is able to recognize amidated JP and 16K amino terminal peptide (Tanaka and Kurosumi 1992). As shown in Figure 1 a, ST-1 stained exclusively the perinuclear region. In contrast, ST-3 stained the entire cytoplasm, with the most intense staining in a punctate fashion at the tips of cell processes (Figure 1 b). The immunofluorescence obtained with the antibodies for POMC and amidated JP was completely eliminated by the addition of 1 μg/ml of the respective antigen peptides (data not shown). For more precise determination of the subcellular localization of POMC and amidated JP, we then performed immunogold electron microscopic analysis. Typical data of the double immunolabeling for POMC (large particles) and amidated JP (small particles) are shown in Figure 3. POMC was localized in secretory granules and on the rim of the Golgi cisternae, where condensing granules were formed (Figure 3 a). Immunolabeling for amidated JP was more commonly seen in secretory granules at the tips of cell processes (Figure 3 b). It is noteworthy that many granules contained both POMC and amidated JP, clearly demonstrating that processing of POMC takes place in the secretory granules. The relative labeling densities of POMC and amidated JP in the same secretory granules appeared to vary from granule to granule (Figure 4). However, one trend was apparent: Granules with POMC but without amidated JP were predominant around the Golgi area, whereas those containing only amidated JP were predominant at the process tips (compare Figures 3a and 3b). Taken together, these observation indicate that processing of POMC occurs coordinately with the maturation of secretory granules.
Co-localization of the Prohormone Convertase PC1 with POMC and Amidated JP in the Same Secretory Granules
Although immunocytochemical studies thus far have revealed the localization of prohormone convertases at the cellular level (Tanaka et al. 1996; Rouille et al. 1994; Marcinkiewicz et al. 1993, 1994; Hornby et al. 1993; Smeekens et al. 1992), only three electron microscopic studies, including ours, have researched their subcellular localization (Tanaka et al. 1996; Malide et al. 1995; Smeekens et al. 1992). Furthermore, these studies have been limited to endocrine cells in pancreatic islets, and there has been no report showing the co-localization of the convertases and prohormones in POMC-containing cells at the subcellular level. We therefore addressed this issue.
Immunofluorescence analysis showed the localization of PC1 in the perinuclear region and at the tips of cellular processes in AtT-20 cells (Figure 2 a). Pre-absorption of the anti-PC1 serum with 1 μg/ml of the corresponding antigen peptide eliminated the staining (Figure 2 b). No positive immunofluorescence was seen in any cellular compartments when AtT-20 cells were stained with anti-PC2. This was consistent with previous reports showing that PC1, but not PC2, is expressed in this cell line (Tanaka et al. 1996; Zhou and Mains 1994; Bloomquist et al. 1991; Seidah et al. 1990). As shown in Figure 5, immunoelectron microscopic analysis clearly demonstrated that PC1 (labeled with small gold particles) was co-localized with POMC (large particles, Figure 5 a) and amidated JP (large particles, Figure 5 b) in the same secretory granules. These observations support previous biochemical data showing that PC1 is involved in the processing of POMC in AtT-20 cells (Zhou et al. 1993; Bloomquist et al. 1991).
Immunofluorescent detection of POMC (a) and amidated JP (b) in AtT-20 cells. Note that the intense staining for POMC (red) is found exclusively in the Golgi region, and the punctate staining for amidated JP (red) is distributed at the tips of cell processes, as revealed by the rhodamine labeling method (arrowheads). Blue color indicates nuclear DNA stained with DAPI. Bar = 20 μm.
Immunofluorescent localization of PC1 in AtT-20 cells. (a) PC1 (red) is localized at the tips of the cell processes and in the Golgi region (arrowheads). Bar = 20 μm. (b) No positive reaction was detected when anti-PC1 was pre-absorbed with the corresponding antigen peptide. Blue color indicates nuclear DNA stained with DAPI. Bar = 20 μm.
Electron micrographs of AtT-20 cells double immunolabeled for POMC (large gold particles) and amidated JP (small gold particles). (a) Only the secretory granules near the Golgi apparatus (G) and the rim of the Golgi cisternae were immunolabeled with the POMC antibody. Bar = 0.25 μm. (b) At the tips of cell processes, the labeling density of POMC and amidated JP varied from granule to granule. Arrowheads, granules bearing only POMC; double arrowheads, granules bearing both POMC and amidated JP; arrows, granules bearing only amidated JP; M, mitochondria. Bar = 0.25 μm.
The pH of Secretory Granules is Lower than That of the Golgi Cisternae
To elucidate the relationship between the processing of POMC and the pH of secretory granules in AtT-20 cells, we examined the acidity of granules with the DAMP method. AtT-20 cells were incubated with 60 μM DAMP, fixed with aldehyde fixatives, and embedded in LR White resin. The LR White resin sections were immunolabeled with anti-DNP antibody followed by immunogold reagent. Many gold particles for DAMP were located in the secretory granules, but few were present in the Golgi apparatus, cell nucleus, and mitochondria (Figures 6a and 6b; small gold particles). In sections singly labeled for DAMP, the density of gold particles in the nucleus, the Golgi apparatus, and the mitochondria was 70.0 ± 8.2 (n = 20), 74.5 ± 5.6 (n = 20), and 68.3 ± 10.2 (n = 20) particles/μm2, respectively. No significant difference in the density was observed among these compartments. In contrast, a variable density of DAMP labeling was found in the secretory granules. As shown in Figure 7, the pH values of secretory granules, estimated according to the formula of Orci et al. (1986), ranged from 5.2 to 7.0.
The labeling density of DAMP was lower in POMC-positive secretory granules near the Golgi apparatus than in amidated JP-positive granules at the tips of the cell processes (Figures 6a and 6b). To define more precisely the relationship between the DAMP labeling density and the processing of POMC to amidated JP in secretory granules, we performed quantitative analysis by double labeling for DAMP and either POMC or amidated JP. As shown in Figure 8, the accumulation of DAMP and the degree of POMC processing appeared to be related, with a tendency towards a negative correlation between the labeling density for DAMP and POMC (r = −0.443; p>0.001) and a positive correlation between DAMP and amidated JP (r = 0.463; p>0.001).
Distribution of the frequency of POMC and amidated JP immunolabeling in secretory granules of AtT-20 cells. To quantitatively evaluate the maturation of POMC, the number of gold particles in each secretory granule was counted. In 15 randomly selected cells, 134 granules were examined, and the values were expressed as a percentage of small particles (amidated JP) per total number of large (POMC) and small (amidated JP) particles in each granule.
Effect of Bafilomycin A1 on Processing of POMC and the Morphology of Organelles
To investigate further the correlation between the granular pH and the degree of POMC processing, we examined the effect of bafilomycin A1, a V-ATPase inhibitor known to dissipate the pH gradient across organellar membranes. AtT-20 cells were pulse-labeled with [35S]-methionine and chased in the continuous presence or absence of bafilomycin A1, and then the cell lysates and media were immunoprecipitated with anti-human ACTH[1–39] serum. As shown in Figure 9, the proteins present in the pulse-labeled cells were limited to two main bands of 32 and 29 kD, corresponding to the size of glycosylated POMC (Roberts et al. 1978) and to some intermediate forms of it (26, 23, and 20 kD), whereas the fully processed forms of POMC, ACTH[1–39] and glycosylated ACTH[1–39], were not observed (Figure 9, Lane 1C). No immunoreactive material was observed in the medium (Figure 9, Lane 1M). In the cells after the chase, POMC-like molecules were no longer observed, but a 13-kD band, which was presumed to be glycosylated ACTH[1–39] (Roberts et al. 1978), appeared (Figure 9, Lane 2C). Glycosylated POMC molecules of 34 and 32 kD, which are formed through further glycosylation (Roberts et al. 1978), and glycosylated ACTH[1–39] were secreted during the chase period (Figure 9, Lane 2M). On the other hand, in the cells chased in the presence of bafilomycin A1, a large amount of POMC was secreted and processed products were not observed (Figure 9, Lanes 3C and 3M). In addition, the secretion of POMC molecules from the drug-treated cells was not enhanced by 5 mM 8-bromo-cyclic AMP (data not shown), which is known to stimulate regulated secretion in AtT-20 cells (Moore et al. 1983). These data indicate that inhibition of V-ATPase prevents POMC processing and causes constitutive secretion of the unprocessed precursor.
We then examined the effect of bafilomycin A1 on the morphology of organelles. Untreated AtT-20 cells are typically characterized by a well-developed rough endoplasmic reticulum and a Golgi apparatus consisting of a stack of six or seven cisternae (Figure 10 a). A few secretory granules were scattered near the Golgi apparatus, and secretory granules were more numerous at the tips of the cell processes. Figure 10 b shows an electron micrograph of AtT-20 cells treated with 10 nM bafilomycin A1 for 2.5 hr. In these cells, many large vacuoles and dilated cisternae were found near the Golgi apparatus. Careful observation revealed that these vacuoles were partially coated with a clathrin-like substance and did not contain any dense core material (Figures 10b and 10c, arrows). Secretory granules were rarely seen in any cytoplasmic region. After as little as 30 min of treatment, the trans-cisternae of the Golgi apparatus were dilated and few secretory granules were observed in the cytoplasm (data not shown).
When the bafilomycin A1-treated AtT-20 cells were immunolabeled with anti-POMC or anti-amidated JP, labeling for POMC was found in the vacuoles coated with a clathrin-like substance and in the dilated cisternae (Figure 11), but no immunoreactivity for amidated JP was detected in any subcellular compartment (data not shown). These morphological observations substantiate the data from the pulse-chase analysis.
When the cells treated with bafilomycin A1 were processed for DAMP immunogold labeling, the accumulation of the pH probe in secretory granules was blocked at the background level (the density of gold particles was 93.4 ± 9.2 particles/μm2; n = 50).
Electron micrograph of an AtT-20 cell double immunolabeled for PC1 (small gold particles) and POMC (large gold particles) (a) or amidated JP (large gold particles) (b). Note that there are granules that are labeled with both sizes of particles (arrowheads). Bars = 0.25 μm.
Chloroquine Does Not Inhibit Processing of POMC
Previously, Mains and May (1988) have reported that chloroquine, which is also known to dissipate the pH gradient of organelles, but a different mechanism from that of bafilomycin A1, is unable to inhibit the processing of POMC. We confirmed their results: Synthesis, processing, or secretion of POMC was not significantly affected by treating AtT-20 cells with chloroquine, even at concentrations of up to 200 μM (data not shown). These data could imply that the acidic pH of organelles is not always crucial for processing of POMC. When the cells treated with 200 μM chloroquine were observed by electron microscopy, lysosome-like structures were detected in the cytoplasm, and no significant morphological change in the Golgi cisternae was observed (data not shown).
Discussion
With respect to the intracellular compartments in which proteolytic processing of POMC takes place in AtT-20 cells, there have been two immunocytochemical studies. Using an antibody that specifically recognizes a cleavage site of POMC, Tooze et al. (1987) have shown that some POMC is packaged into secretory granules before its cleavage. On the other hand, Schnabel et al. (1989) have concluded that processing of POMC begins in the trans-most cisterna of the Golgi apparatus, based on their detection of processed products in this compartment. However, neither study has examined whether or not both the precursor and mature peptides are present in the same compartments in AtT-20 cells. Addressing this point, the experiments presented here clearly demonstrate that by using antibodies specific for the cleavage site of POMC and for a mature peptide (amidated JP), both the precursor and amidated JP can be detected in the same secretory granule. POMC was detected predominantly in the secretory granules around the Golgi apparatus and in condensing secretory materials on the rims of the trans-Golgi cisternae, whereas amidated JP was predominant in the secretory granules at the tips of the cell processes. Furthermore, the present study shows that PC1, a convertase that is capable of cleaving POMC at paired basic amino acid sites, is co-localized with POMC and amidated JP in the same secretory granule. Considering these points together, we conclude that POMC is cleaved mainly in secretory granules coordinately with their maturation. This conclusion is compatible with the data of Tooze et al. (1987). However, we failed to obtain evidence showing that POMC processing begins in the trans-Golgi area, as reported by Schnabel et al. (1989). This discrepancy between their data and ours might be due either to the difference in the reactivity between the antibodies used for amidated JP or to the difference in the immunocytochemical sensitivity between the used sections, i.e., the ultrathin cryosections and the LR White sections.
Location of DAMP (small gold particles) and POMC (large gold particles) (a), and of DAMP (small gold particles) and ami-dated JP (large gold particles) (b) on the same ultrathin sections of AtT-20 cells. Cells were incubated in the presence of 60 μM DAMP for 1 hr at 37C. After fixation, LR White ultrathin sections were made and immunolabeled with gold particles of different sizes for DAMP and POMC or DAMP and amidated JP. The labeling density of DAMP was Lower in POMC-positive granules than on amidated JP-positive granules. Arrowheads, secretory granules; M, mitochondria; N, nucleus. Bar = 0.25 μm (for a and b).
Percentage showing classes of secretory granules in AtT-20 cells after arranging the intensity of DAMP-labeling in each granule in terms of pH. The data were analyzed in 400 secretory granules from 50 cells singly labeled for DAMP.
Enzymological studies have shown that the prohormone convertases PC1 and PC2 are most active at a relatively acidic pH (pH 5.0–6.5) (Lamango et al. 1996; Rufaut et al. 1993; Zhou and Lindberg 1993; Davidson et al. 1988), whereas furin, which is localized mainly in the TGN and is involved in the cleavage of precursors for constitutively secreted proteins and plasma membrane proteins rather than prohormones, has a neutral pH optimum (Molloy et al. 1994, 1992; Hatsuzawa et al. 1992). This indicates that there may be a close relationship between the optimal pH of the enzymes and their subcellular localization. It is therefore important to know whether or not the pH of secretory granules is adequate for the action of prohormone convertases. Previous studies by Orci et al. (1986, 1987b, 1994), using DAMP as the pH probe, have shown that in pancreatic β-cells conversion of proinsulin to insulin occurs coordinately with the acidification of maturing secretory granules. However, there has been no report about the pH of secretory granules in other cell types, including AtT-20 cells. We have shown, using the DAMP method, that in AtT-20 cells the granular pH varies from one granule to another over a range of 5.2 to 7.0. The granular pH values are in good agreement with those in pancreatic β-cells reported by Orci et al. (1986, 1987b, 1994). Furthermore, there was a tendency for the labeling density for DAMP to correlate negatively with that for POMC and positively with that for amidated JP. Taking these data together with the observations of the co-localization of POMC and amidated JP in the same granule noted above, we presume that POMC processing in AtT-20 cells occurs mainly in acidifying secretory granules.
Relationship between accumulation of DAMP and proteolytic cleavage of POMC. There is (a) a slightly negative correlation between DAMP and POMC labeling and (b) a slightly positive correlation between DAMP and amidated-JP labeling. Immunolabeling of individual secretory granules is expressed as the number of gold particles/μm2 of secretory granule area after double labeling with anti-POMC and anti-DNP antibodies or with anti-amidated JP and anti-DNP antibodies. Each granule is represented by a dot. Two hundred granules were evaluated for each of the two kinds of double labeling.
Effect of bafilomycin A1 on POMC processing in AtT-20 cells. Cells were preincubated for 30 min, pulse-labeled with [35S]-methionine for 30 min (Lane 1), and chased for 1.5 hr in the absence (Lane 2) or presence (Lane 3) of 10 nM bafilomycin A1. Cell lysates (C) and media (M) were then prepared as described in Materials and Methods. Newly synthesized POMC and POMC-related peptides were immunoprecipitated with human ACTH[1–39] antibody and analyzed by SDS-PAGE, followed by fluorography.
In an attempt to support the above observations, we examined the effect of the V-ATPase inhibitor, bafilomycin A1, on POMC processing. Recently, it was reported that the internal acidic environment in the secretory pathways was generated by a V-ATPase (Nelson and Taiz 1989). V-ATPase has since been purified and characterized from the Golgi apparatus (Moriyama and Nelson 1989) and the secretory granules (Moriyama and Futai 1990). Bafilomycin A1 is a specific inhibitor of V-ATPase in chromaffin granules, lysosomes and the Golgi apparatus (Moriyama and Nelson 1989; Bowman et al. 1988). Therefore, the application of bafilomycin A1 is expected to define the role of V-ATPase-mediated acidification in the secretory pathways. The present pulse-chase experiments showed that bafilomycin A1 inhibits POMC processing, and electron microscopic analysis detected POMC, but not amidated JP, in bafilomycin A1-treated cells. Since the accumulation of DAMP in granules was blocked at background levels in the drug-treated cells, it appears that the V-ATPase inhibitor neutralized the granules, thereby inhibiting POMC processing. On the other hand, the pulse-chase experiments showed that bafilomycin A1 causes constitutive secretion of the unprocessed precursor. Furthermore, morphological analysis has revealed that the drug causes the dilation of the trans-Golgi cisternae and the appearance of many vacuoles containing POMC but not amidated JP. Similar morphological changes induced by the drug have also been observed in GH3 cells that produce growth hormone and prolactin (Henomatsu et al. 1993). It is therefore likely that bafilomycin A1 commonly induces such morphological changes. These observations lead to another possibility, that the inhibition of POMC processing in the bafilomycin A1-treated AtT-20 cells resulted from the morphological changes. The changes might have blocked the condensation and packaging of POMC into regulated secretory granules, since no dense core material was observed in the POMC-containing vacuoles. Supporting this possibility is the fact that chloroquine, which also dissipates intracellular acidic environments, but by a mechanism different from that of bafilomycin A1, does not inhibit POMC processing (Mains and May 1988), nor does it cause significant morphological changes around the Golgi region. The chloroquine data indicate that acidic environments are not essential for processing of POMC by PC1. This view is indirectly supported by enzymological studies showing that although PC1 has an acidic pH optimum, it still exhibits some activity around neutral pH (Rufaut et al. 1993; Zhou and Lindberg 1993; Davidson et al. 1988), and is in agreement with the data of Schnabel et al. (1989) that POMC processing begins in the TGN. Certainly, the data presented here clearly demonstrate that proteolytic processing of POMC in AtT-20 cells occurs coordinately with the maturation of secretory granules. It is therefore likely that the packaging of POMC into secretory granules is a crucial step for proper processing of the prohormone.
Electron micrographs of normal and bafilomycin A1-treated AtT-20 cells. (a) The Golgi apparatus in untreated AtT-20 cells consists of several cisternae. Small arrows, secretory granules. Bar = 0.25 μm. (b) In bafilomycin A1-treated cells, large vacuoles (arrows) and dilated cis-ternae (large arrowheads) are seen near the Golgi apparatus. Bar = 0.25 μm. (c) Vacuoles in the drug-treated cells were partially coated with a clathrin-like material (large arrows). G, Golgi cisternae. Bar = 0.25 μm.
Immunogold electron micrograph showing the presence of POMC (gold particles) in the vacuoles (arrows) near the Golgi apparatus (G) and in the Golgi cisternae (arrowheads) in bafilomycin A1-treated AtT-20 cells. Bar = 0.25 μm.
Footnotes
Acknowledgements
This work was supported by a research grant (No. 0467002) from the Japanese Ministry of Education, Science, and Culture (ST).