Fine structure of the secretory and nonsecretory ameloblasts in the frog. I. Fine structure of the secretory ameloblasts.

Amelogenesis in the tooth germs of the frog Rana pipiens was examined by electron microscopy at different stages of tooth development. Cellular changes in secretory ameloblasts during this process showed many basic similarities to those in mammalian amelogenesis. Amelogenesis can be divided into three stages based on histological criteria such as thickness of enamel and the relative position of the tooth germ within the continuous succession of teeth. These stages are early, transitional and late. The fine structure of the enamel-secreting cells reflects the functional role of these ameloblasts as primarily secretory in the early stage, possibly transporting in the late stage and reorganizing between the two functions in the transitional stage. In early amelogenesis the cell exhibits well-developed granular endoplasmic reticulum, Golgi complex, microtubules, dense granules, smooth and coated vesicles, lysosome-like bodies in supranuclear and distal portions of the cell and mitochondria initially concentrated in the basal part of the cell. Numerous autophagic vacuoles are observed concomitant with the loss of some cell organelles at the transitional stage. During late amelogenesis the ameloblasts exhibit numerous vesicles, granules, convoluted cell membranes, junctional complexes and widely distributed mitochondria. Toward the end of amelogenesis, cells become oriented parallel to the enamel surface and the number of organelles is reduced. Amelogenesis in the frog is an extracellular process and mineralization seems to occur simultaneously with matrix formation.     

Study of human secretory immunoglobulin A. I. Obtaining monospecific antiserum to human secretory immunoglobulin A]

A method of obtaining monospecific antiserum to the human secretory IgA is described. Immunochemically pure secretory IgA (isolated from human colostrum by fractionation with ammonium sulfate and gel-filtration on Sephadex G-200) was used for immunization of rabbits or sheep. Heterologous antibodies were removed by adsorption with commercial gamma globulin, normal serum, the serum of a patient suffering from A-myeloma with the IgA polymere and purified lactoferrin. Monospecific antiserum to the secretory IgA gave a reaction of complete immunological identity with the secretory IgA and a free secretory component.     

Disulfide bonding of secretory component to a single monomer subunit in human secretory IgA.

The arrangement of disulfide bonds joining secretory component (SC) to the alpha chains in secretory IgA was studied by determining the molecular size of the principal fragments resulting from CNBr digestion of secretory dimeric Fc fragments from IgA (Fc)2alpha fragments). In vitro complexes formed by incubating 125I-free SC and myeloma 131I-(Fc)2alpha fragments were isolated by gel filtration and subsequently digested with cyanogen bromide. The CNBr digests of SC-(Fc)2alpha fragments were analyzed by gel filtration in 5 M guanidine. Two principal fragments were obtained, one containing a monomeric Fc fragment from IgA (Fcalpha) associated with SC (m.w. congruent to 110,000) and a second containing the second Fcalpha monomer (m.w. congruent to 50,000) from the dimeric SC-(Fc)2alpha. Similar results were obtained when secretory (Fc)2alpha fragments isolated from native secretory IgA dimer were subjected to CNBr digestion. The data indicate that SC is disulfide bonded to a single monomer subunit in secretory IgA dimer.     

Distribution of covalently bound and non-covalently bound secretory component on sbuclasses of rabbit secretory IgA.

The distribution of non-covalently bound secretory component (SC) on the two subclasses, IgA-f and IgA-g of rabbit secretory IgA (sIgA) was determined; the two subclasses were separated from each other by the use of antibody-immunosorbent columns and were subjected to SDS polyacrylamide gel electrophoresis. No SC appeared to be dissociated from the IgA-f molecules from each of 11 different rabbits; the IgA-g molecules, however, did have SC which was dissociated by SDS. Thus, all of the noncovalently bound SC on rabbit sIgA resides on the IgA-g subclass molecules.     

Influence of secretory immunoglobulin A and purified secretory component on dextran-sucrase activity of Streptococcus mutans.

The net stimulation of dextransucrase EC 2.4.1.5) activity from Streptococcus mutans HS6 by dextran, secretory immunoglobulin A, or secretory component was investigated. Approximately equal stimulation resulted from treatment with these three components.     

Coincident localization of secretory and plasma membrane proteins in organelles of the yeast secretory pathway.

Immunoelectron microscopy of Saccharomyces cerevisiae cells embedded in Lowicryl K4M has been used to localize invertase and plasma membrane (PM) ATPase in secretory organelles. sec mutant cells incubated at 37 degrees C were prepared for electron microscopy, and thin sections were incubated with polyclonal antibodies, followed by decoration with protein A-gold. Specific labeling of invertase was seen in the lumen of the endoplasmic reticulum, Golgi apparatus, and secretory vesicles in mutant cells that exaggerate these organelles. PM ATPase accumulated within the same organelles. Double-immune labeling revealed that invertase and PM ATPase colocalized in secretory vesicles. These results strengthen the view that secretion and plasma membrane assembly are biosynthetically coupled in yeast.     

Proprotein-processing endopeptidases of the insulin secretory granule.

Enzymological studies have implicated two Ca2+ dependent endopeptidases in the conversion of proinsulin to insulin: a type 1 activity and a type 2 activity which cleave on the C-terminal side of R31R32 and K64R65 in proinsulin, respectively. These activities were further characterized and their relationship to the mammalian family of subtilisin-like proteases was investigated. PC2 was expressed in neuroendocrine tissues and in insulinoma secretory granule fractions predominantly as a 65kDa protein. On anion-exchange chromatography of solubilized granules, PC1/3 immunoreactivity comigrated with a peak of type 1 activity whereas PC2 immunoreactivity coeluted with the peak of type 2 endopeptidase activity. PC2 antiserum gave a specific immunoprecipitation of type 2 activity from insulin granule extracts. It was concluded that the PC2 gene-product has type 2 endopeptidase activity.     

Biogenesis of processing-competent secretory organelles in vitro.

Propeptide processing occurs in specific compartments of the secretory pathway, but how these processing-competent organelles are generated from their processing-incompetent precursor compartments is unknown. To dissect the process biochemically, we have developed a novel cell-free system reconstituting the production of processing-competent secretory granules in AtT-20 cells. Using donor membranes containing [(35)S]sulfate labeled pro-opiomelanocortin (POMC)(5) in the trans-Golgi, we can reconstitute cytosol- and ATP-dependent prohormone processing as well as incorporation of processed ACTH into immature secretory granules (ISGs). Under limiting cytosol conditions, both reactions are greatly stimulated by ADP-ribosylation factor 1 (ARF1) but not by the GDP-bound ARF1 T31N mutant. pH studies show that lumenal acidification, most likely due to ARF-mediated sorting of proton pumps and leaks during budding, confers processing competency to the resulting organelle. Surprisingly, comparison of onset of processing and ISG release reveals that they are distinct biochemical processes with different kinetics and separate cytosolic requirements. Moreover, ARF regulates the onset of prohormone processing but not ISG release. Our data suggest a two-step mechanism (onset of processing followed by ISG release) for the production of processing-competent organelles from the trans-Golgi and provide the first system with which these two steps may be individually dissected.     

Carbohydrate moieties in human secretory component.

Human secretory component has seven putative sites for N-linked glycosylation. From tryptic and Glu-C digests we have isolated peptides encompassing asparagines 65, 72, 117, 168, 403, 451 and 481. Analysis by on line HPLC-electrospray mass spectrometry indicated that these residues were fully glycosylated and that the major carbohydrate moieties were far less diversified in composition than expected. Fast atom bombardment mass spectrometry performed on oligosaccharides released by peptide-N-glycosidase F treatment of fractionated and unfractionated SC digests showed the following glycan compositions: Fuc(2)Hex(5)HexNAc(4), Fuc(3)Hex(5)HexNAc(4), NeuAcFucHex(5)HexNAc(4), NeuAcFuc(2)Hex(5)HexNAc(4), NeuAc(2)Hex(5)HexNAc4 and NeuAc(2)FucHex(5)HexNAc(4). Three of these oligosaccharides are the major carbohydrate moieties in human lactoferrin. A possible biological role of the secretory component glycans in the protection of mucosal surfaces is discussed.     

Characterization of epitopes of human secretory component on free secretory component, secretory IgA, and membrane-associated secretory component

We have compared the epitopes present in various forms of human secretory component by using a panel of hybridoma-derived antibodies elicited by immunizing mice with free secretory component (FSC) or secretory IgA (sIgA). Enzyme-linked immunosorbent binding assays (ELISA) were used to assess antibody binding to FSC- and SC-containing antigens, including sIgA isolated from milk, reduced and alkylated sIgA, and sIgA assembled in vitro by incubating dimeric IgA with FSC. Immunofluorescence assays were also used to assess binding to a human epithelial tumor cell line (HT29) that expresses secretory component as an integral protein of the plasma membrane. The results can be summarized as follows. 1) Most antibodies from fusions in which sIgA was the immunizing antigen bound preferentially to sIgA. 2) Most antibodies from fusions in which FSC was the immunizing antigen bound preferentially to FSC. 3) Antibodies that bound preferentially to sIgA invariably bound sIgA assembled in vitro; antibodies that bound preferentially to FSC invariably did not. 4) Antibodies that bound readily to both sIgA and FSC were rare in all fusions. 5) The monoclonal antibodies defined at least six classes of epitopes on SC, including epitopes that were a) FSC specific and reduction sensitive, b) FSC specific and reduction insensitive, c) sIgA specific and reduction-sensitive, d) sIgA specific and reduction insensitive, e) shared by FSC and sIgA and reduction-sensitive, and f) shared by FSC and sIgA and reduction-insensitive. 6) Antibodies that mediated intense immunofluorescent staining of secretory component on HT29 cell membranes were rare and constituted a distinct subset of those which recognized epitopes shared by FSC, reduced and alkylated sIgA, and some preparations of native sIgA. Results of these antibody-binding studies indicate that most SC epitopes are not shared by FSC and sIgA. Most SC-related epitopes on sIgA appear to be generated by the physical interaction of SC with dimeric IgA, whereas most epitopes on FSC are masked or altered by this interaction. Finally, epitopes that are shared by membrane SC and FSC and/or sIgA represent a minor and immunochemically distinct subset of epitopes on SC. The high proportion of unique epitopes on the different physical forms of SC suggest that the epitopes of this molecule are highly sensitive to its molecular environment. The monoclonal reagents described here will be useful in studying the structure and function of SC; quantitating FSC, sIgA, and membrane SC; purifying various molecular forms of SC by immunoaffinity chromatography; and localizing SC in human tissues and cultured cells by immunocytochemical techniques.     

Protein processing within the secretory pathway.

Endoproteolytic cleavage of hormone and neuropeptide precursors, as well as many complex proteins, such as coagulation factors and viral glycoproteins, is a key process in the generation of bioactive polypeptides. These cleavages typically occur at the dibasic amino acid residues Lys-Arg or Arg-Arg. The enzymes responsible for the processing belong to a newly discovered family of serine proteases related to the bacterial subtilisins. These include PACE (furin), PC1/PC3, PC2 and PACE4, which have all been characterized functionally and structurally.     

Comparison of the secretory processes in the parotid and sublingual glands of the mouse. 1. Regulation of the secretory processes.

1. Secretion from the mucous sublingual gland of the mouse has been investigated and compared with the serous parotid gland. The influence of acetylcholine, noradrenalin and adrenalin on the secretion of glycoproteins (e.g. mucins) and proteins (e.g. amylase) from these glands in vitro, and the involvement of cyclic AMP and Ca2+ has been studied. 2. Secretion from the parotid gland could be stimulated by both acetylcholine and the catecholamines. It appears that cyclic AMP plays an important role in the adrenergic secretory process, but not in the cholinergic-induced secretion. In the latter case, exogenous Ca2+ strongly increased the secretion. 3. Mucin secretion from the sublingual gland could be affected by acetylcholine in the presence of exogenous Ca2+. Noradrenalin and adrenalin induced only a slow mucin secretion and, for this secretory process, exogenous Ca2+ is also required. Though cyclic AMP is present in the sublingual gland, no influence on its level could be detected in this gland after stimulation of the adrenergic beta-receptor, whereas, in contrast to the parotid gland, dibutyryl cyclic AMP induced only a slow secretion. Because it was observed that the sublingual gland of the mouse is not innervated sympathetically, it seems reasonable to suppose that the catecholamines stimulate the mucin secretion from this gland via hormonal receptors and not via the adrenergic beta-receptor. 4. The protein secretion from the sublingual gland could be stimulated by both acetylcholine and the catecholamines. An involvement of cyclic AMP in this process was not observed. Addition of exogenous Ca2+ is less important, as was found for the mucin secretion. So it has been concluded that protein and mucin secretion from the sublingual gland are regulated via different pathways.     

Basketball exercise and secretory immunoglobulin A.

This study examined saliva levels of immunoglobulin A (IgA) before and after three games and three practice sessions during the basketball season. Saliva was collected from 27 prepubescent boys (10-12 years) in a small Fry league and 23 postpubescent boys (16-18 years) on a high school varsity team. Saliva samples were frozen for later assay using a standard enzyme-linked immunosorbent assay technique. IgA levels were significantly increased after games 1 and 3 in both age groups and after practice 3 in the high school athletes. Over the 2 months of saliva collections the pre-exercise IgA increased significantly with games 2 and 3 higher than game 1, and practice 3 higher than practices 1 and 2, in both age groups. These results indicate that basketball exercise can increase saliva IgA levels and that chronic exercise over the basketball season may increase the resting levels of IgA. These changes may give athletes more protection against respiratory infections both after exercise and in the resting state later in the season.