Portal de Periódicos da CAPES

Chromogranin B (secretogranin I), a neuroendocrine-regulated secretory protein, is sorted to exocrine secretory granules in transgenic mice

1998; Springer Nature; Volume: 17; Issue: 12 Linguagem: Inglês

10.1093/emboj/17.12.3277

1460-2075

SHOICHI NATORI,

Lipid Membrane Structure and Behavior

Article15 June 1998free access Chromogranin B (secretogranin I), a neuroendocrine-regulated secretory protein, is sorted to exocrine secretory granules in transgenic mice Shoichi Natori Shoichi Natori Department of Neurobiology, University of Heidelberg, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany Search for more papers by this author Angus King Angus King Present address: Iizuka Hospital, 3-83 Yoshio-machi, Iizuka, 820 Japan Search for more papers by this author Andrea Hellwig Andrea Hellwig Present address: European Molecular Biology Laboratory, Postfach 10.2209, D-69012 Heidelberg, Germany Search for more papers by this author Ursula Weiß Ursula Weiß Department of Biochemistry, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 815, Departments of Pathology, Fukuoka, 812 Japan Search for more papers by this author Haruo Iguchi Haruo Iguchi Department of Biochemistry, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 815, Departments of Pathology, Fukuoka, 812 Japan Search for more papers by this author Benio Tsuchiya Benio Tsuchiya Kitasato University School of Allied Health, Fukuoka, 812 Japan Search for more papers by this author Toru Kameya Toru Kameya Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, 228 Japan Search for more papers by this author Ryoichi Takayanagi Ryoichi Takayanagi Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maedashi, Fukuoka, 812 Japan Search for more papers by this author Hajime Nawata Hajime Nawata Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maedashi, Fukuoka, 812 Japan Search for more papers by this author Wieland B. Huttner Corresponding Author Wieland B. Huttner Department of Neurobiology, University of Heidelberg, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany Search for more papers by this author Shoichi Natori Shoichi Natori Department of Neurobiology, University of Heidelberg, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany Search for more papers by this author Angus King Angus King Present address: Iizuka Hospital, 3-83 Yoshio-machi, Iizuka, 820 Japan Search for more papers by this author Andrea Hellwig Andrea Hellwig Present address: European Molecular Biology Laboratory, Postfach 10.2209, D-69012 Heidelberg, Germany Search for more papers by this author Ursula Weiß Ursula Weiß Department of Biochemistry, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 815, Departments of Pathology, Fukuoka, 812 Japan Search for more papers by this author Haruo Iguchi Haruo Iguchi Department of Biochemistry, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 815, Departments of Pathology, Fukuoka, 812 Japan Search for more papers by this author Benio Tsuchiya Benio Tsuchiya Kitasato University School of Allied Health, Fukuoka, 812 Japan Search for more papers by this author Toru Kameya Toru Kameya Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, 228 Japan Search for more papers by this author Ryoichi Takayanagi Ryoichi Takayanagi Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maedashi, Fukuoka, 812 Japan Search for more papers by this author Hajime Nawata Hajime Nawata Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maedashi, Fukuoka, 812 Japan Search for more papers by this author Wieland B. Huttner Corresponding Author Wieland B. Huttner Department of Neurobiology, University of Heidelberg, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany Search for more papers by this author Author Information Shoichi Natori1, Angus King2, Andrea Hellwig3, Ursula Weiß4, Haruo Iguchi4, Benio Tsuchiya5, Toru Kameya6, Ryoichi Takayanagi7, Hajime Nawata7 and Wieland B. Huttner 1 1Department of Neurobiology, University of Heidelberg, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany 2Present address: Iizuka Hospital, 3-83 Yoshio-machi, Iizuka, 820 Japan 3Present address: European Molecular Biology Laboratory, Postfach 10.2209, D-69012 Heidelberg, Germany 4Department of Biochemistry, National Kyushu Cancer Center, 3-1-1 Notame, Fukuoka 815, Departments of Pathology, Fukuoka, 812 Japan 5Kitasato University School of Allied Health, Fukuoka, 812 Japan 6Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, 228 Japan 7Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maedashi, Fukuoka, 812 Japan The EMBO Journal (1998)17:3277-3289https://doi.org/10.1093/emboj/17.12.3277 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Chromogranin B (CgB, secretogranin I) is a secretory granule matrix protein expressed in a wide variety of endocrine cells and neurons. Here we generated transgenic mice expressing CgB under the control of the human cytomegalovirus promoter. Northern and immunoblot analyses, in situ hybridization and immunocytochemistry revealed that the exocrine pancreas was the tissue with the highest level of ectopic CgB expression. Upon subcellular fractionation of the exocrine pancreas, the distribution of CgB in the various fractions was indistinguishable from that of amylase, an endogenous constituent of zymogen granules. Immunogold electron microscopy of pancreatic acinar cells showed co-localization of CgB with zymogens in Golgi cisternae, condensing vacuoles/immature granules and mature zymogen granules; the ratio of immunoreactivity of CgB to zymogens being highest in condensing vacuoles/immature granules. CgB isolated from zymogen granules of the pancreas of the transgenic mice aggregated in a mildly acidic (pH 5.5) milieu in vitro, suggesting that low pH-induced aggregation contributed to the observed concentration of CgB in condensing vacuoles. Our results show that a neuroendocrine-regulated secretory protein can be sorted to exocrine secretory granules in vivo, and imply that a key feature of CgB sorting in the trans-Golgi network of neuroendocrine cells, i.e. its aggregation-mediated concentration in the course of immature secretory granule formation, also occurs in exocrine cells although secretory protein sorting in these cells is thought to occur largely in the course of secretory granule maturation. Introduction Sorting of secretory proteins to the regulated pathway has been studied extensively (Burgess and Kelly, 1987; Arvan and Castle, 1992; Tooze et al., 1993; Halban and Irminger, 1994; Thiele et al., 1997). Two sorting compartments can be distinguished, the trans-Golgi network (TGN) and the immature secretory granule, a vesicular intermediate in secretory granule biogenesis. In the TGN, secretory proteins destined to secretory granules (regulated secretory proteins, RSPs) and other lumenal cargo proteins are segregated from each other into immature secretory granules and constitutive secretory vesicles, respectively. The RSPs and other lumenal cargo proteins are segregated further from each other in the course of the fission of the immature secretory granule into the mature secretory granule and an AP1 adaptor/clathrin-coated vesicle (Dittié et al., 1996), respectively, with the latter mediating constitutive-like secretion presumably via early endosomes (von Zastrow and Castle, 1987; Arvan and Castle, 1992; Kuliawat and Arvan, 1992; Huttner et al., 1995; Kuliawat et al., 1997). In both sorting compartments, the segregation of the RSPs from other lumenal cargo proteins involves their progressive self-interaction, i.e. their aggregation during the formation of the immature secretory granule from the TGN and their further condensation during the conversion of the immature to the mature secretory granule (Gerdes et al., 1989; Tooze et al., 1989, 1993; Arvan and Castle, 1992; Thiele et al., 1997). Though the mechanism of segregation of the RSPs from other lumenal cargo proteins utilizes a common principle in both compartments, the extent to which sorting occurs at the level of the TGN versus the immature secretory granule may vary between cell types, depending on the ratio of volume flow from the TGN into immature secretory granules versus constitutive secretory vesicles, and from immature secretory granules into mature secretory granules versus the AP1 adaptor/clathrin-coated vesicles (Thiele et al., 1997). Cells containing the regulated pathway of protein secretion include exocrine and neuroendocrine cells. However, our understanding as to the extent of conservation of RSP sorting between exocrine and neuroendocrine cells is incomplete. On the one hand, certain RSPs normally produced by exocrine cells such as trypsinogen (Burgess et al., 1985), proline-rich protein (Castle et al., 1992) and amylase (Castle et al., 1997) have been found qualitatively to be sorted to secretory granules upon expression in neuroendocrine cells, although quantitative differences in the sorting efficiency compared with the endogenous neuroendocrine RSPs have been noted (Castle et al., 1997). On the other hand, an anchorless, secretory version of GP2, a lumenal protein of pancreatic zymogen granules normally bound to the membrane via a glycerophosphatidylinositol (GPI) anchor, is sorted to secretory granules in AR42J cells (Colomer et al., 1994), a pancreatic cell line exhibiting some exocrine features (Rosewicz et al., 1992), but not in the neuroendocrine cell line AtT-20 (Colomer et al., 1994). These observations indicate that an exocrine RSP may, but will not necessarily be sorted to neuroendocrine secretory granules. Whether or not the converse, the sorting of neuroendocrine RSPs to exocrine secretory granules, may also occur, is unknown. Two neuroendocrine RSPs, pancreatic polypeptide and gastrin, whose biogenesis in neuroendocrine cells involves proteolytic processing of larger precursors and C-terminal amidation in secretory granules, have been recovered in fully processed form from cell extracts upon expression in AR42J cells (Dickinson et al., 1993). However, the significance of these observations with regard to a possible sorting of the two prohormones to exocrine secretory granules is unclear because AR42J cells exhibit not only exocrine, but also neuroendocrine, features (Rosewicz et al., 1992) and a possible storage of the two hormones in exocrine secretory granules was not investigated (Dickinson et al., 1993). An additional problem with both approaches, the expression of exocrine RSPs in neuroendocrine cells and of neuroendocrine RSPs in exocrine cells, is that cultured cell lines rather than cells in tissues were used as hosts for the foreign RSPs. Hence, the significance of the observed, at least partial conservation of exocrine versus neuroendocrine RSP sorting for the in vivo situation remains to be established. A neuroendocrine RSP, growth hormone, has been expressed in pancreatic acinar cells and found to be secreted into the lumen of the pancreatic duct (Ornitz et al., 1985). While the pathway of secretion (i.e. constitutive, constitutive-like or regulated) was not investigated, this study nonetheless documents that the exocrine pancreas of a transgenic mouse can serve as a suitable model to address the conservation of exocrine versus neuroendocrine RSP sorting in vivo. In the present study, we have generated transgenic mice expressing chromogranin B (CgB, also referred to as secretogranin I), an RSP found in a wide variety of neuroendocrine cells (Rosa et al., 1985; Huttner et al., 1991; Winkler and Fischer-Colbrie, 1992). Upon placing the CgB gene (Pohl et al., 1990) under the control of the human cytomegalovirus (CMV) promoter, the highest level of expression was detected in the exocrine pancreas. We exploited this observation to investigate the question of whether or not this neuroendocrine RSP is sorted to exocrine secretory granules in vivo. Two considerations in addition to the ones delineated above made this investigation worthwhile. First, the membrane-bound form of carboxypeptidase E (mCPE), a processing enzyme (Fricker, 1988) found in neuroendocrine, but not exocrine cells, has been reported (Cool et al., 1997) to bind to the N-terminal disulfide-bonded loop of pro-opiomelanocortin and to various other neuroendocrine RSPs including chromogranin A (CgA), a protein related to CgB (Benedum et al., 1987; Huttner et al., 1991; Winkler and Fischer-Colbrie, 1992), and has been proposed to have an essential role in their sorting to secretory granules (Cool et al., 1997; Shen and Loh, 1997). Since both CgA and CgB also contain N-terminal disulfide-bonded loops which are highly homologous to each other (Benedum et al., 1987) and essential for sorting (Chanat et al., 1993; Krömer et al., 1998), an answer to the above question would provide information as to a possible role of mCPE in the sorting of the chromogranins to secretory granules. Second, in certain endocrine cells, CgB has been found to be enriched in a subpopulation of secretory granules containing a distinct set of hormones (Bassetti et al., 1990). Moreover, from the analysis of endocrinologically silent pituitary adenomas, it has been concluded that the granins alone are sufficient to form a secretory granule matrix (Rosa et al., 1992). If CgB should turn out to be sorted to secretory granules in pancreatic acinar cells, an ultrastructural analysis would provide insight as to whether or not this reflects the formation of a neuroendocrine-like subpopulation of secretory granules in an exocrine cell or the co-packaging with zymogens into true exocrine secretory granules. Here, we report biochemical and morphological evidence that CgB expressed in the exocrine pancreas of transgenic mice is sorted to zymogen granules, and discuss the implications of our ultrastructural analysis for current concepts of RSP sorting. Results Generation of CgB-transgenic mice We generated transgenic mice expressing CgB under the control of a constitutive promoter. For this purpose, a CgB expression vector was constructed in which the mouse CgB gene without its endogenous promoter was placed under the control of the human CMV promoter (Figure 1A). This CgB expression construct proved to be functional because NIH 3T3 cells stably transfected with this vector, but not wild-type NIH 3T3 cells, produced CgB mRNA and protein; the latter showed a punctate perinuclear localization in immunofluorescence, indicative of targeting of the expressed CgB to the secretory pathway (data not shown). The CMVmCgB insert of the expression vector was microinjected into fertilized mouse oocytes. Nineteen pups were born, of which three showed integration of the CgB transgene into the genome as revealed by Southern blot analysis (Figure 1B). Two of these mice could transmit the CgB transgene to their offsprings in a Mendelian fashion and were used to found CgB-transgenic mouse lines, referred to as 380.6 and 382.6. Both lines showed the massive expression of the CgB transgene in the exocrine pancreas which is documented below and further characterized for the 380.6 line. Figure 1.Transgenic mice expressing CgB under the control of the human CMV promoter. (A) The pCMVmCgB expression vector containing the mouse CgB gene under the control of the human CMV promoter (open box). Filled boxes E1–E5, exons 1–5; hatched box, 1.5 kb SacI–XbaI fragment used as a probe in Southern blot analysis (B); dotted box, 1.6 kb EcoRI fragment used as probe in Northern blot analysis (Figure 2A) and as template for the cRNA probes for in situ hybridization (Figure 3). The asterisk indicates the HindIII site generated 5′ to the SacI site to distinguish, in Southern blot analysis of HindIII–EcoRI-digested genomic DNA, the CgB transgene (T, 3.6 kb fragment) from the endogenous CgB gene (E, 6.3 kb fragment). (B) Southern blot analysis of HindIII–EcoRI-digested genomic DNA from control (C) and CgB-transgenic (T) mice showing the endogenous CgB gene (arrow) and the CgB transgene (arrowhead). Download figure Download PowerPoint Ectopic expression of the CgB transgene in the exocrine pancreas Northern blot analysis showed that in contrast to control mice, in which the CgB mRNA is known to be present specifically in neuroendocrine tissues (Rosa and Gerdes, 1994) such as brain (Figure 2A, bottom), CgB-transgenic mice contained detectable levels of CgB mRNA also in various non-neuroendocrine tissues (Figure 2A, top). By far the highest CgB mRNA level in CgB-transgenic mice was observed in the pancreas. In CgB-transgenic mice, this tissue also contained the highest level of CgB protein as shown by immunoblot analysis (Figure 2B, top). This was in contrast to control mice (Figure 2B, bottom), in which CgB, though readily observed in immunoblots of tissues rich in neuroendocrine cells such as adrenal, pituitary and brain, is not easily detected in the pancreas because of the low amount of islet tissue relative to the vast amount of exocrine tissue which physiologically does not express CgB (Rosa and Gerdes, 1994). No significant difference in the pancreatic CgB level was noted between female and male CgB-transgenic mice (data not shown). Figure 2.Tissue distribution of CgB mRNA and protein in control and CgB-transgenic mice. (A) Northern blot of total RNA (10 μg) from salivary gland (Sa), brain (Br), muscle (Mu), kidney (Ki), testis (Te), lung (Lu), heart (H), thymus (Thy), liver (Li), pancreas (Pa) and spleen (Spl) of CgB-transgenic (upper panel) and control (lower panel) mice. To document expression of the CgB transgene in the various tissues, an autoradiogram with an overexposed signal of the CgB mRNA in the transgenic pancreas is shown; this accentuates the CgB mRNA fragments resulting from degradation by pancreatic RNase. (B) Immunoblot of the heat-stable proteins prepared from adrenal gland (Adr, 24–30 μg), pituitary gland (Pi, 17–24 μg), salivary gland (Sa), brain (Br), muscle (Mu), kidney (Ki), lung (Lu), heart (H), thymus (Thy), liver (Li), pancreas (Pa) and spleen (Spl) (100 μg each) of CgB-transgenic (upper panel) and control (lower panel) mice. To document the presence of endogenous CgB in adrenal gland, pituitary gland and brain, an autoradiogram with an overexposed signal of CgB in the transgenic pancreas is shown; arrow, full-length CgB; arrowhead, N-terminally truncated degradation product of CgB. Download figure Download PowerPoint The high amounts of CgB mRNA and protein in the pancreas of CgB-transgenic mice suggested that the transgene was ectopically expressed in the exocrine cells of this tissue. In situ hybridization showed that this was indeed the case (Figure 3). Staining for the CgB mRNA was obtained with the antisense (Figure 3C–E), but not the sense (Figure 3A and B) probe and was observed in acinar cells of CgB-transgenic (Figure 3C), but not control (Figure 3D) mice. In these cells, staining for the CgB mRNA was concentrated in the basal, perinuclear region (Figure 3E), consistent with the subcellular localization of the rough endoplasmic reticulum. Staining for the CgB mRNA in the pancreatic islets showed a very similar pattern for both CgB-transgenic (Figure 3C) and control (Figure 3D) mice, with a low expression level in many cells and high expression in a few scattered cells. Figure 3.In situ hybridization showing expression of CgB mRNA in the exocrine pancreas of CgB-transgenic mice. In the control pancreas (B and D), islet cells, but not acinar cells are stained with the CgB antisense probe (D), but not the sense probe (B, note unstained islet in the lower left corner). In the CgB-transgenic pancreas (A, C and E), in addition to the staining of some islet cells (C, lower left corner), there is massive staining of acinar cells with the CgB antisense probe (C and E), but not with the sense probe (A). Staining of acinar cells is concentrated in the basal area (see higher magnification in E), consistent with the intracellular distribution of the rough endoplasmic reticulum. (A–D) are the same magnification, the bar in (A) corresponds to 25 μm; (E) is a higher magnification, the bar corresponds to 10 μm. Download figure Download PowerPoint Consistent with the results of in situ hybridization, immunocytochemistry showed CgB immunoreactivity in acinar cells of CgB-transgenic (Figure 4B and C), but not control (Figure 4A) mice. CgB immunoreactivity was concentrated in the apical region of the cells and had a punctate pattern (Figure 4C), suggesting a localization of the ectopically expressed CgB in secretory vesicles. CgB immunoreactivity in the pancreatic islets showed a very similar pattern for both CgB-transgenic (Figure 4B) and control (Figure 4A) mice, with significant staining in many cells and high staining in a few scattered cells, consistent with previous observations on non-transgenic animals (Winkler and Fischer-Colbrie, 1992; Rosa and Gerdes, 1994). Figure 4.Immunocytochemistry showing the presence of CgB in pancreatic acinar cells of CgB-transgenic mice. In the control pancreas (A), CgB immunoreactivity is observed in islet cells, but not in acinar cells. In the CgB-transgenic pancreas (B and C), both islet cells (B, left) and acinar cells show CgB immunoreactivity. Staining of acinar cells is concentrated in the apical area and has a granular appearance (see higher magnification in C). No staining was observed with non-immune serum (not shown). (A) and (B) are the same magnification, the bar in (A) corresponds to 25 μm; (C) is a higher magnification, the bar corresponds to 10 μm. Download figure Download PowerPoint CgB ectopically expressed in pancreatic acinar cells is sorted to zymogen granules Given the ectopic expression of CgB in pancreatic acinar cells of CgB-transgenic mice, we investigated its intracellular localization by subcellular fractionation and immunoelectron microscopy. Upon purification of zymogen granules, the specific CgB immunoreactivity determined by immunoblot analysis (Figure 5A) increased in parallel with the specific activity of amylase (Figure 5B). When a crude zymogen granule fraction was analyzed by equilibrium and velocity sucrose gradient centrifugation, CgB immunoreactivity coincided with amylase activity in the dense fractions of the equilibrium gradient (Figure 6) and the pellet of the velocity gradient (Figure 7). This indicated that the CgB-containing vesicles were indistinguishable from zymogen granules in density and size, and suggested the co-localization of CgB with zymogens in the latter organelles. Figure 5.Parallel enrichment of CgB and amylase upon purification of zymogen granules from the pancreas of CgB-transgenic mice. (A) CgB immunoblot of the heat-stable protein fraction prepared from equal amounts (1 mg protein) of the initial homogenate (I), crude zymogen granule fraction (C) and purified zymogen granule fraction (P). For reference, the heat-stable protein fraction prepared from a homogenate (500 μg protein) of mouse adrenal gland (Adr) is shown. (B) Comparison of the CgB immunoreactivity (●) and amylase activity (○) per mg of protein in the homogenate and zymogen granule fractions. For CgB immunoreactivity, the bands of full-length CgB and the CgB degradation products shown in (A) were quantified by densitometric scanning and the sum of the absorbance values calculated. Download figure Download PowerPoint Figure 6.Co-fractionation of CgB and amylase upon equilibrium sucrose gradient centrifugation of zymogen granules from the pancreas of CgB-transgenic mice. A crude zymogen granule fraction prepared from the pancreas of CgB-transgenic mice was subjected to equilibrium centrifugation on a 0.5–2 M sucrose gradient and an aliquot of equal volume of each gradient fraction analyzed (fraction 12 = bottom). (A) CgB immunoblot. (B) Comparison of CgB immunoreactivity (●) and amylase activity (○). For CgB immunoreactivity, the bands of full-length CgB and the CgB degradation products shown in (A) were quantified by densitometric scanning and the sum of the absorbance values calculated. Download figure Download PowerPoint Figure 7.Co-sedimentation of CgB and amylase upon velocity sucrose gradient centrifugation of zymogen granules from the pancreas of CgB-transgenic mice. A crude zymogen granule fraction prepared from the pancreas of CgB-transgenic mice was subjected to a 15 min velocity centrifugation on a 0.3–1.2 M sucrose gradient and an aliquot of equal volume of each gradient fraction including the resuspended pellet (P) analyzed (2+3, pool of half an aliquot of fractions 2 and 3). (A) CgB immunoblot. (B) Amylase activity. Download figure Download PowerPoint Immunogold electron microscopy provided direct evidence for the presence of the ectopically expressed CgB in zymogen granules (Figures 8 and 9). Double labeling for CgB and zymogens (Figure 8) showed CgB immunoreactivity over the matrix of most, if not all, zymogen granules. We did not observe subpopulations of secretory granules containing either mostly zymogens or mostly CgB, nor did we detect secretory granules in which CgB and zymogens were obviously segregated from each other within the secretory granule matrix, as has been reported for distinct RSPs in neuroendocrine cells (Fumagalli and Zanini, 1985; Ehrhart et al., 1986; Hashimoto et al., 1987; Fisher et al., 1988; Bassetti et al., 1990). Co-localization of CgB and zymogen immunoreactivity was already observed at the level of the Golgi cisternae and single condensing vacuoles (Figure 8). Quantitation of the gold particles for CgB and zymogens revealed that the ratio of CgB to zymogen immunoreactivity over condensing vacuoles/immature granules (for definition see Materials and methods) was 4- to 5-fold greater than that over Golgi cisternae and mature zymogen granules (Figure 8, Table I). This was largely due to the fact that for CgB, the apparent immunoreactivity per unit area increased from the Golgi cisternae to the condensing vacuoles/immature granules and decreased from condensing vacuoles/immature granules to mature zymogen granules, whereas no such decrease was observed in the case of the zymogens. This was evident from both double labeling for CgB and zymogens (Figure 8) and single labeling for either CgB or zymogens (Figure 9). Figure 8.Double immunogold electron microscopy showing co-localization of CgB with zymogens in the Golgi complex, condensing vacuoles and zymogen granules of the pancreas of CgB-transgenic mice. Ultrathin cryosections prepared from the pancreas of CgB-transgenic mice were double-immunolabeled for CgB (arrows) and zymogens (arrowheads), and immunoreactivity was revealed by 9 and 14 nm protein A–gold. (A, D and E) CgB, 14 nm gold; zymogens, 9 nm gold. (B and C) CgB, 9 nm gold; zymogens, 14 nm gold. Note the co-localization of CgB and zymogen immunoreactivity in the cisternae of the Golgi complex (white G), in the condensing vacuoles (cv) seen in continuity with, or in proximity to, the Golgi cisternae, and in zymogen granules (zg). (B and C) The ratio of CgB to zymogen immunoreactivity is greater in the condensing vacuoles than in Golgi cisternae and in zymogen granules (see also Table I). All panels are the same magnification; the bar in (E) corresponds to 500 nm. Download figure Download PowerPoint Figure 9.Single immunogold electron microscopy of the pancreas of CgB-transgenic mice showing the decrease in CgB, but not zymogen, immunoreactivity in mature zymogen granules relative to the condensing vacuoles/immature granules. Ultrathin cryosections prepared from the pancreas of CgB-transgenic mice were immunolabeled for either CgB (A and C) or zymogens (B and D), and immunoreactivity was revealed by 9 nm protein A–gold. Note that the CgB immunoreactivity (A and C) is higher in the condensing vacuoles/immature granules (cv; for definition, see Materials and methods) than in the cisternae of the Golgi complex (white G) and in mature zymogen granules (zg, white arrowheads), whereas zymogen immunoreactivity (B and D) is similar in condensing vacuoles/immature granules and in mature zymogen granules. All panels are the same magnification; the bar in (D) corresponds to 500 nm. Download figure Download PowerPoint Table 1. CgB and zymogen immunoreactivity in Golgi cisternae, condensing vacuoles/immature granules and zymogen granules of pancreatic acinar cells of CgB-transgenic mice Field Golgi cisternae Condensing vacuoles/immature granules Zymogen granules CgB Zym. CgB/Zym. Ratio CgB Zym. CgB/Zym. Ratio CgB Zym. CgB/Zym. Ratio Field 1 (Figure 8A) 5 16 0.31 2.58 3 4 0.75 6.25 18 149 0.12 1 Field 2 (Figure 8B) 6 10 0.60 0.60 41 14 2.93 2.93 38 38 1.00 1 Field 3 (Figure 8C) 12 9 1.33 1.85 85 17 5.00 6.94 33 46 0.72 1 Field 4 (Figure 8D) 3 44 0.07 0.44 n.d. n.d. n.d. n.d. 69 442 0.16 1 Field 5 (Figure 8E) 3 34 0.09 0.75 n.d. n.d. n.d. n.d. 38 309 0.12 1 Field 6 n.d. n.d. n.d. n.d. 26 22 1.18 3.58 24 72 0.33 1 Mean ± SD 1.24 ± 0.93* 4.93 ± 1.97* 1 Gold particles indicative of either CgB or zymogen (Zym.) immunoreactivity over Golgi cisternae, condensing vacuoles/immature granules and zymogen granules were counted in the five fields, parts of which are shown in Figure 8A–E, and in another field (Field 6) of double immunogold-labeled ultrathin cryosections. The ratio of CgB immunoreactivity/zymogen immunoreactivity was calculated (CgB/zym.). For each field, the ratio obtained for zymogen granules was set to 1, and the ratio for Golgi cisternae and for condensing vacuoles/immature granules (for definition, see Materials and methods) was expressed rela