Properties of pancreatic zymogen granules studied by quasi-elastic light scattering

Physico-chemical properties of isolated zymogen granules of the mouse pancreas were studied by means of quasi-elastic light scattering. The average diameter of the granules in 0.3 M sucrose was found to be 1.1 ± 0.1 μm from the correlation time of intensity fluctuation of the scattered light. The average diameter altered depending on the osmolality of the medium in a manner that the alteration was smaller than that expected from the van't Hoff relation. Aggregation of the granules induced by the increase of Ca²⁺ concentration or the decrease of pH in the medium was also detected. The aggregation started at a critical level of 1 mM CaCl₂ or at pH 5.4.     

High-contrast en bloc staining of neuronal tissue for field emission scanning electron microscopy

Conventional heavy metal poststaining methods on thin sections lend contrast but often cause contamination. To avoid this problem, we tested several en bloc staining techniques to contrast tissue in serial sections mounted on solid substrates for examination by field emission scanning electron microscopy (FESEM). Because FESEM section imaging requires that specimens have higher contrast and greater electrical conductivity than transmission electron microscopy (TEM) samples, our technique uses osmium impregnation (OTO) to make the samples conductive while heavily staining membranes for segmentation studies. Combining this step with other classic heavy metal en bloc stains, including uranyl acetate (UA), lead aspartate, copper sulfate and lead citrate, produced clean, highly contrasted TEM and scanning electron microscopy (SEM) samples of insect, fish and mammalian nervous systems. This protocol takes 7-15 d to prepare resin-embedded tissue, cut sections and produce serial section images.     

Stabilization of exocytosis by dynamic F-actin coating of zymogen granules in pancreatic acini

Reorganization of F-actin in the apical region of mouse pancreatic acinar cells during Ca(2+)-dependent exocytosis of zymogen granules was investigated by two-photon excitation microscopy with intact acini. Granules were rapidly coated with F-actin in response to either agonist stimulation or photolysis of a caged-Ca(2+) compound. Such F-actin coating occurred exclusively at the surface of granules undergoing exocytosis and was prevented either by latrunculin-A, which inhibits actin polymerization, or by Clostridium botulinum exoenzyme C3, which inhibits the small GTPase Rho. Latrunculin-A or exoenzyme C3 also triggered the formation of vacuoles in acinar cells, a characteristic of acute pancreatitis. Stimulation of acini with high concentrations of cholecystokinin, which cause acute pancreatitis in mice, also impaired the F-actin coating of granules and induced vacuole formation. Latrunculin-A reduced the latency to exocytosis but did not affect the total number of exocytic events, suggesting that F-actin slows and further stabilizes exocytosis by facilitating F-actin coating. Rho-dependent F-actin coating of granule membranes thus stabilizes exocytic structures and is necessary for physiological progression of sequetial compound exocytosis in the exocrine pancreas and for prevention of acute pancreatitis.     

Dual modulation of chloride conductance by nucleotides in pancreatic and parotid zymogen granules.

The regulation of Cl- conductance by cytoplasmic nucleotides was investigated in pancreatic and parotid zymogen granules. Cl- conductance was assayed by measuring the rate of cation-ionophore-induced osmotic lysis of granules suspended in iso-osmotic salt solutions. Both inhibition and stimulation were observed, depending on the type and concentration of nucleotide. Under optimal conditions, the average inhibition measured in different preparations was 1.6-fold, whereas the average stimulation was 4.4-fold. ATP was inhibitory at 1-10 microM but stimulated Cl- conductance above 50 microM. Stimulation by ATP was more pronounced in granules with low endogenous Cl- conductance. The potency of nucleotides in terms of inhibition was ATP greater than adenosine 5'-[gamma-thio]triphosphate (ATP[S]) greater than UTP much greater than or equal to CTP much greater than or equal to GTP much greater than or equal to guanosine 5'-[gamma-thio]triphosphate (GTP[S]) much greater than or equal to ITP. The potency with respect to stimulation had the following order: adenosine 5'-[beta gamma-methylene]triphosphate (App[CH2]p) greater than ATP greater than guanosine 5'-[beta-thio]diphosphate (GDP[S]). Adenosine 5'-[beta gamma-imido]triphosphate (App[NH]p) was also stimulatory, and was more potent than ATP in the parotid granules, but less potent in the pancreatic granules. Aluminium fluoride stimulated Cl- conductance maximally at 15-30 microM-Al3+ and 10-15 mM-F. F was less effective at higher concentrations. Protein phosphorylation by kinases was apparently not involved, since the nucleotide effects (1) could be mimicked by non-hydrolysable analogues of ATP and GTP, (2) showed reversibility, and (3) were not abolished by the protein kinase inhibitors 1-(5-isoquinolinesulphonyl)-2-methylpiperazine (H-7) or staurosporine. The data suggest the presence of at least two binding sites for nucleotides, whereby occupancy of one induces inhibition and occupancy of the other induces stimulation.     

Compartmentalization of secretory proteins in pancreatic zymogen granules as revealed by immunolabeling on cryo-fixed and molecular distillation processed tissue*

Summary— Immunogold labeling of amylase obtained over zymogen granules of rat pancreatic acinar cells processed through cryofixation, molecular distillation drying and embedding in resins was found to be of high intensity and displayed a particular pattern. Indeed, it was concentrated in certain areas of the granules leaving others devoid of gold particles. This pattern of labeling reflects a strong compartmentalization of the secretory proteins within each granule. In order to assess this phenomenon, we have compared the intensities and the pattern of distribution of the labelings in tissues processed through: chemical fixation with embedding in various resins, cryo-ultramicrotomy and cryo-fixation followed by molecular distillation drying. Serials sections and double labeling experiments were performed for further evaluation of the results and for assessing artefactual displacement of proteins during tissue preparation. The results obtained indicate that the secretory proteins are indeed segregated within the granule which appears thus as a well organized structure. Cryo-fixation combined with molecular distillation appears thus to be superior in terms of preservation of protein antigenicity and retention of cellular components close to their living state.     

Lysophospholipids in pig pancreatic zymogen granules in relation to exocytosis.

A hypothesis to explain the stimulatory role of cyclic AMP (adenosine 3':5'-monophosphate) in pancreatic enzyme secretion. has been tested. In this hypothesis cyclic AMP would activate a phospholipase activity, which would lead to a locally increased lysophospholipid formation, resulting in a fusion between the zymogen granule membrane and the apical plasma membrane. Cyclic AMP added to isolated pig pancreatic zymogen granules leads to an increased lysis of these granules, but the slowness of this effect makes its physiological significance dubious. In pancreatic homogenates or zymogen granules no stimulating effect of cyclic AMP on lipase of phospholipase activity could be demonstrated. Isolated zymogen granules have a high lysophospholipid content (27% of total phospholipids), consisting of the 1-acyl and 2-acyl forms of lysophosphatidylcholine and lysophosphatidylethanolamine. Experiments with radioactive phosphatidylcholine indicate that the lysophospholipids are due to the action of endogenous (phospho)lipases during the isolation procedure. It is concluded that these experiments do not lend support to the above hypothesis for the mechanism of action of cyclic AMP in pancreatic enzyme secretion     

Ultrastructural immunocytochemical localization of l-glutamate decarboxylase and GABA in rat pancreatic zymogen granules

Summary The ultrastructural immunohistochemical localization of gamma aminobutyric acid (GABA) and its regulating enzymes, l-glutamate decarboxylase (GAD) and gamma aminobutyrate--ketoglutarate transaminase, was determined utilizing an immunogold post-embedding protocol in pancreatic exocrine tissue. Within the acinar cell, GABA and its biosynthetic enzyme, GAD, were localized in zymogen granules. Quantitative analysis of the GABA immunoreactivity in the acinar cell revealed 1.7±0.5 gold particles/m² over the cytoplasm, 36.6±14.1 gold particles/ m² over the zymogen granules, and 2.9±2.1 gold particles /m² over the mitochondria. Quantitative analysis of the distribution of colloidal gold particles, representing glutamate decarboxylase immunoreactivity in the acinar cells, revealed 38.4±2.5 gold particles/m² over the zymogen granules, 4.7±1.1 gold particles/m² over the mitochondria and 6.3±0.5 gold particles/m² over the remainder of the cytoplasm. Substitution of normal sheep serum for the sheep anti-glutamate decarboxylase serum revealed a significant (p< 0.001) decrease of the colloidal gold particle distribution over the zymogen granules and cytoplasmic compartments of the acini. Gamma aminobutyrate --ketoglutarate transaminase, the catabolic enzyme for GABA, was not detected in the mitochondria, zymogen granules, and cytoplasm of the acinar cell, suggesting that GABA is not catabolized within the acinar cell. Preabsorption and substitution controls resulted in an absence of labeling. These results suggest that GABA may act extracellularly and/or have a role within the zymogen granule in the exocrine pancreas.     

ATP sensitive K+ conductance in pancreatic zymogen granules: Block by glyburide and activation by diazoxide

Summary The properties of transporters (or channels) for monovalent cations in the membrane of isolated pancreatic zymogen granules were characterized with an assay measuring bulk cation influx driven by a proton diffusion potential. The proton diffusion potential was generated by suspending granules in an isotonic monovalent cation/acetate solution and increasing the proton conductance of the membrane with a protonophore. Monovalent cation conductance had the sequence Rb⁺ > K⁺ > Na⁺ > Cs⁺ > Li⁺ > N-methyl glucamine⁺. The conductance could be inhibited by Ca²⁺, Mg²⁺, Ba²⁺, and pharmacological agents such as quinine, quinidine, glyburide and tolbutamide, but not by 5 mm tetra-ethyl ammonium or 5mm 4-aminopyridine, when applied to the cytosolic surface of the granule membrane. Over 50% of K⁺ conductance could be inhibited by millimolar concentrations of ATP or MgATP. The inhibition by MgATP, but not by ATP itself, was reversed by the K⁺ channel opener diazoxide. The inhibitory effect is probably by a noncovalent interaction since it could be mimicked by nonhydrolyzable analogs of ATP and by ADP. The reversal of MgATP inhibition by diazoxide may be mediated by phosphorylation since it was not affected by dilution, and was blocked by the protein kinase inhibitor H7. The properties of the K⁺ conductance of pancreatic zymogen granule membranes are similar to those of ATP-sensitive K⁺ channels found in the plasma membrane of insulin-secreting islet cells, neurons, muscle, and renal cells.This research was supported by grants from the Cystic Fibrosis Foundation (ZO298) and NIH (DK-39658). F.T. is recipient of a Fellowship from the American Cystic Fibrosis Foundation. K.C.V. is a participant of a summer research program for undergraduate students from Knox College, Galesburg, IL.     

Rab27b localizes to zymogen granules and regulates pancreatic acinar exocytosis

To understand the function of pancreatic zymogen granules, we performed a proteomics analysis to identify ZG membrane components. Here we report the identification of Rab27b through this proteomics study and validate its role in granule function. MALDI-MS peptide mass fingerprint was matched to rat Rab27b with 43% sequence coverage, and the identification was also confirmed by tandem mass spectrometry. The localization of Rab27b on ZGs was confirmed by Western blotting and immunocytochemistry. To examine the function of Rab27b in acinar secretion, we overexpressed wild type and mutant Rab27b protein in pancreatic acini using recombinant adenoviruses. Wild type Rab27b had no effect on amylase secretion, while Rab27b Q78L enhanced, and Rab27b N133I inhibited, CCK-induced amylase release by 92+/-13% and 53+/-8%, respectively. This enhancement and inhibition occurred at all points on the CCK dose-response curve and over a 30min time course. These results demonstrate that Rab27b is present on ZGs and plays an important role in regulating acinar exocytosis.     

In vitro stability of pancreatic zymogen granules: roles of pH and calcium

Purified preparations of pancreatic zymogen granules have the peculiar property of lysing instantaneously at neutral pH, a property clearly irreconcilable with the cytoplasmic pH of the acinar cell. Two important factors known for regulating the stability of secretory granules are calcium and pH. Fluorescence microscopy of acinar cells in the presence of weak bases showed that zymogen granules have an acidic pH. In vivo, abolition of the delta pH by NH4Cl did not induce any lysis of the granules. In vitro, with purified granules, an acidic intragranular pH was measured. This delta pH was produced by a Donnan potential. The importance for granule stability of keeping the intragranular pH acidic has been confirmed in vitro by addition of K+ and nigericin to the suspension medium. These conditions produced alkalinization of the granule matrix and caused instantaneous solubilization of the granules. Concentrations of 15 mM total, and 10 mM free calcium were measured in purified granules. The importance of intragranular Ca2+ was evaluated by means of the ionophore A23187 which induced calcium efflux and granule lysis. The lysis induced by the calcium ionophore was in direct relation with the calcium efflux, since addition of Ca2+ to the medium, at concentrations corresponding to that measured in the granule, relieved the effect. The role of calcium-binding sites on the cytoplasmic surface of the granules was investigated with Ca2+, EGTA, and La3+. Calcium did not have any damaging effects; EGTA induced a slight lysis, while lanthanum yielded a strong and spontaneous lysis at micromolar concentrations. In addition to calcium-binding sites, La3+ would bind to specific sites on the granule that would be directly coupled to maintenance of its stability. These findings suggest that the intragranular acidic pH and calcium are both important for the in vitro stability of the zymogen granule and that purified granules have lost, in the course of purification, some cytoplasmic factors that in vivo, control the permeability of the membrane to protons, and chloride more particularly. Calcium-binding sites and other specific sites probed with La3+, presumably on proteins at the surface of the granule, are also believed to have key roles in preserving the integrity of the membrane and the resulting stability of the granule.     

Histochemical staining of zymogen granules of pancreatic acinar cells using a permanganate-HID-technique

Summary Using a permanganate-HID technique specific staining of cystine-containing zymogen granules of exocrine pancreas is possible. This method is based upon breaking and oxidation of disulphide bonds into sulphonic acids which subsequently are demonstrable with the HID-reaction.N,N-dimethyl-p-phenylendiamine-dihydrochloride was obtained from Fluka AG/ Basel; N,N-dimethyl-m-phenylendiamine-dihydrochloride from Schuchardt/Munich.     

Proteomic analysis of pancreatic zymogen granules: identification of new granule proteins

The composition of zymogen granules from rat pancreas was determined by LC-MS/MS. Enriched intragranular content, peripheral membrane, and integral membrane protein fractions were analyzed after one-dimensional SDS-PAGE and tryptic digestion of gel slices. A total of 371 proteins was identified with high confidence, including 84 previously identified granule proteins. The 287 remaining proteins included 37 GTP-binding proteins and effectors, 8 tetraspan membrane proteins, and 22 channels and transporters. Seven proteins, pantophysin, cyclic nucleotide phosphodiesterase, carboxypeptidase D, ecto-nucleotide phosphodiesterase 3, aminopeptidase N, ral, and the potassium channel TWIK-2, were confirmed by immunofluorescence microscopy or by immunoblotting to be new zymogen granule membrane proteins.     

A lysophospholipase specific for exocrine pancreatic cells is stored in zymogen granules and secreted into pancreatic juice

We separated by two-dimensional (2D) gel electrophoresis the content of isolated rat zymogen granules and from the gel excised a protein of apparent MW 77,500 and an isoelectric point of about 4.7. A rabbit antiserum against this previously uncharacterized rat zymogen granule protein recognized two cDNA clones in a rat pancreas expression library. The cDNA inserts of these two clones had sequences showing perfect homology to the published cDNA sequence of rat pancreatic lysophospholipase. The antiserum recognized only a single protein, lysophospholipase, on one and two-dimensional immunoblots of rat pancreas homogenates and isolated zymogen granules. The antiserum did not react with any protein in homogenates of rat liver, spleen, adrenal, parotid, and prostate tissue. The zymogen granule protein of the guinea pig, previously identified as Lipase 1, was recognized specifically by the antiserum against rat lysophospholipase. This guinea pig protein can now be regarded as lysophospholipase. The same protein was demonstrated in the transformed rat acinar cell line AR4-2J, where both the rate of total enzyme synthesized and the amount of mRNA increased following treatment with dexamethasone. Immunogold labeling established that pancreatic lysophospholipase is restricted exclusively to exocrine cells where it occurs only in compartments of the exocytotic pathway. It could also be detected in pancreatic juice in the ducts of the tissue. Finally, we have shown that lysophospholipase is not related to the zymogen granule membrane protein GP2. This work establishes that lysophospholipase is a normal member of the set of soluble enzymes and proenzymes that are stored in zymogen granules and secreted into pancreatic juice.     

Effects of a cytosolic protein on the interaction of rat pancreatic zymogen granules in vitro

Photon correlation spectroscopy has been used to study the kinetics of aggregation of isolated rat pancreatic zymogen granules in vitro by monitoring time-dependent changes in mean particle size derived from the photon count autocorrelation function, g2(tau). Isolated granules were stable in isotonic sucrose (pH 5.4-7.0). At pH 6.0 they maintained a mean diameter of 1225 +/- 18 nm with a polydispersity index of 0.199 +/- 0.007. The mean granule diameter showed a limited decrease (approx. 20%) with increasing pH within the range 5.4-7.0, but the polydispersity index was unaltered. At pH greater than 7.0 granule instability was indicated by a rapid reduction in total photon counts. In solutions of monovalent cations ([M+] greater than 10 mM) and divalent cations ([M2+] greater than 0.5 mM) zymogen granules aggregated at a rate dependent upon both ion and granule concentration. These effects were consistent with the bimolecular nature of the interaction mechanism and were clearly distinguishable from the limited size changes associated with osmolarity. At concentrations of Na+ or K+ salts greater than 50 mM granule aggregation was accompanied by anion-dependent solubilisation. A soluble protein fraction separated from the pancreatic acinar cell cytosol by gel filtration reduced the mean diameter and polydispersity index of zymogen granules suspended in isotonic sucrose, inhibited cation-induced aggregation and stabilised granules to solubilisation induced by raising pH greater than 7.0 or exposure to high ionic strength media. The inhibitory effects of this protein were apparent at concentrations less than or equal to 10 micrograms X ml-1 (i.e. at inhibitor: granule protein ratios less than 1:20) and could not be mimicked by bovine serum albumin, the Ca2+-binding proteins calmodulin and troponin C (less than or equal to 100 micrograms X ml-1), nor the highly negatively charged polymer polyglutamate (less than or equal to 10 micrograms X ml-1). Inhibitory activity was also absent from fractions of rat liver cytosol prepared identically to pancreatic acinar cytosol. These observations are consistent with the presence in pancreatic acinar cells of a specific cytosolic granule stabilisation factor (or factors) that normally restricts zymogen granule interaction and may therefore play an important role in the regulation of granule mobility and exocytosis.     

Selective fluorescence of zymogen granules of pancreatic acinar cells stained with hematoxylin and eosin

Following staining with hematoxylin and eosin Y, paraffin sections of mouse pancreas were examined by transmitted light, epifluorescence and confocal laser scanning microscopy. Light microscopy revealed that the nuclei of pancreatic acinar cells were located basally, while the apices of the cells appeared eosinophilic, although the secretory granules were difficult to visualize. Under violet-blue light excitation, the zymogen granules at the apices of the acinar cells showed strong yellowish fluorescence; the other part of the cytoplasm was only faintly fluorescent and the nuclei and the supporting tissues were nonfluorescent. Confocal laser scanning microscopy resulted in clear pictures of the zymogen granules and their distribution within the cell. The fluorescent emission of zymogen granules was certainly the result of eosin Y staining, because hematoxylin is not a fluorochrome and the zymogen granules are not autofluorescent. Staining with eosin Y alone, however, did not result in clear fluorescent images of zymogen granules or any other cellular structures. Our observation shows that the fluorescence emission of eosin Y allows easy and precise recognition of zymogen granules of pancreatic cells.     

Selective fluorescence of zymogen granules of pancreatic acinar cells stained with hematoxylin and eosin

Following staining with hematoxylin and eosin Y, paraffin sections of mouse pancreas were examined by transmitted light, epifluorescence and confocal laser scanning microscopy. Light microscopy revealed that the nuclei of pancreatic acinar cells were located basally, while the apices of the cells appeared eosinophilic, although the secretory granules were difficult to visualize. Under violet-blue light excitation, the zymogen granules at the apices of the acinar cells showed strong yellowish fluorescence; the other part of the cytoplasm was only faintly fluorescent and the nuclei and the supporting tissues were nonfluorescent. Confocal laser scanning microscopy resulted in clear pictures of the zymogen granules and their distribution within the cell. The fluorescent emission of zymogen granules was certainly the result of eosin Y staining, because hematoxylin is not a fluorochrome and the zymogen granules are not autofluorescent. Staining with eosin Y alone, however, did not result in clear fluorescent images of zymogen granules or any other cellular structures. Our observation shows that the fluorescence emission of eosin Y allows easy and precise recognition of zymogen granules of pancreatic cells.     

Selective fluorescence of zymogen granules of pancreatic acinar cells stained with hematoxylin and eosin.

Following staining with hematoxylin and eosin Y, paraffin sections of mouse pancreas were examined by transmitted light, epifluorescence and confocal laser scanning microscopy. Light microscopy revealed that the nuclei of pancreatic acinar cells were located basally, while the apices of the cells appeared eosinophilic, although the secretory granules were difficult to visualize. Under violet-blue light excitation, the zymogen granules at the apices of the acinar cells showed strong yellowish fluorescence; the other part of the cytoplasm was only faintly fluorescent and the nuclei and the supporting tissues were nonfluorescent. Confocal laser scanning microscopy resulted in clear pictures of the zymogen granules and their distribution within the cell. The fluorescent emission of zymogen granules was certainly the result of eosin Y staining, because hematoxylin is not a fluorochrome and the zymogen granules are not autofluorescent. Staining with eosin Y alone, however, did not result in clear fluorescent images of zymogen granules or any other cellular structures. Our observation shows that the fluorescence emission of eosin Y allows easy and precise recognition of zymogen granules of pancreatic cells.     

G(alpha)(i3) in pancreatic zymogen granules participates in vesicular fusion

Earlier studies indicated that a G(i)-like protein localized in pancreatic zymogen granule (ZG) membrane mediates vesicle swelling, and is a potentially important prerequisite for vesicle fusion at the cell plasma membrane (PM) [Jena et al. (1997) Proc. Natl. Acad. Sci. USA 94, 13317-13322]. In the present study, we demonstrate the presence of G(alpha)(i3) immunoreactivity in ZGs of rat exocrine pancreas using immunoblot assays, light and electron immunomicroscopy. Since GTP has been implicated in the fusion of isolated ZG with PM fractions [Nadin et al. (1989) J. Cell Biol. 109, 2801-2808], the potential role of ZG-associated G(alpha)(i3) was investigated. Immunoblot assays demonstrate an increase in G(alpha)(i3) protein in ZGs isolated from carbamylcholine stimulated pancreas. Thin layer chromatography shows an increase in GTP hydrolysis by GTPase in ZGs isolated from stimulated compared to resting pancreas. In vitro fusion assays demonstrate that ZGs isolated from carbamylcholine-stimulated pancreatic lobules fuse with the PM at a greater potency in the presence of GTP, mastoparan (G protein agonist) and its analogue mas7. Furthermore, G(alpha)(i3)-specific a recombinant GAIP (G alpha interacting protein), potentiates ZG-PM fusion in the presence of GTP but not in presence of the non-hydrolyzable GTP analogue Gpp(NH)p. Our immunoblot analysis demonstrates the recruitment of Galpha(i3) immunoreactivity to ZG from stimulated acinar cells, and these isolated ZGs are more potent and efficient in fusing with plasma membrane fractions, suggesting the possible involvement of G(alpha)(i3) in ZG-PM fusion. The participation of ZG-associated G(alpha)(i3) in ZG-PM fusion is further confirmed by the influence of the G(alpha)(i3)-specific GAIP, which is known to interact specifically with G(alpha)(i3), and not with G(alpha)(i2) or G(alpha)(q) [DeVries et al. (1995) Proc. Natl. Acad. Sci. USA 92, 11916-11920]. Additionally, our data suggest that GTP hydrolysis is a requirement for ZG-PM fusion since GAIP in the presence of Gpp(NH)p shows little or no effect on fusion, whereas GAIP in the presence of GTP significantly potentiates ZG-PM fusion. Our studies suggest a possible role for ZG-associated G(alpha)(i3) in ZG-PM fusion.     

The protein content and morphogenesis of zymogen granules

When zymogen granules, the secretion granules of pancreatic acinar cells, fill, secretory product is accumulated in immature granules, condensing vacuoles. Mature granules are formed when this product (protein) condenses into an osmotically inactive aggregate and, bulk water is expelled. This hypothesis for granule morphogenesis has two elements. The first is that immature granules are precursors to mature granules. The second is that a particular maturational event, condensation, which involves the aggregation of protein, takes place. These hypotheses lead to two straightforward predictions. One, that condensing vacuoles on average, should contain less protein than filled or mature granules. And two, that, due to condensation, mature granules should contain protein at a common concentration. In the current work, both of these predictions were tested using measurements of the protein content of individual granules acquired by X-ray microscopy. Neither prediction was affirmed by the experimental results. First, there was no distinguishable difference in the distribution of protein between immature and mature granules. Second, the protein concentration of mature granules varied widely between preparations, although granules from the same preparation had similar concentrations. From the data we conclude that: 1) mature granules and condensing vacuoles are different, though not necessarily unrelated, types of secretory vesicle, and not two forms of the same object; 2) as such, condensing vacuoles are not precursors to mature granules; 3) all granules do not contain protein at one particular concentration when full, or mature; 4) granule maturation does not involve a condensation step; 5) concentration is not determined by such physical limits as the space available for protein packing or condensation; and 6) the amount of protein contained is physiologically regulated.