Identification and functional distribution of intracellular ca channels in mouse lacrimal gland acinar cells.

We have determined the presence and cellular distribution of intracellular calcium channels, inositol 1, 4, 5-trisphosphate receptors (IP3Rs) and ryanodine receptors (RyRs) in adult and postnatal (P10) lacrimal gland acinar cells. Western blot analysis of both P10 cultures and adult tissue identified the presence of each IP(3)R and RyR isotypes. The immunocytochemistry analysis showed a differential cellular distribution of these calcium channels where the nuclear envelope, endoplasmic reticulum (ER) and Golgi apparatus membranes represent areas with highest levels of channel expression. This IP(3)R and RyR isotype distribution is confirmed by the immuno-EM results. The findings described in this study are in agreement with published pharmacological data that shows the participation of these channels in the secretion process of the lacrimal gland acinar cells. Furthermore, the differential subcellular distribution between the isoforms could indicate a potential role of these intracellular Ca(2+ )channels on the regulation of specific cellular functions.

Rat retinal progenitor cells and a retinal pigment epithelial factor.

Retinal progenitor cells were isolated from explants of neonatal rat retinas and characterized by transmission electron microscopy and reverse transcriptase-polymerase chain reaction and by their response to an isolated retinal pigment epithelial cell cell factor. The isolated progenitor cells demonstrated nuclei with abundant euchromatin typical of progenitor cells and showed the presence of nestin and opsin message. A protein ( approximately 67 kDa) isolated from conditioned media of cultured rat RPE cells promoted the survival of isolated retinal progenitor cells.

Positive regulation of motility and flhDC expression by the RNA-binding protein CsrA of Escherichia coli.

Many species of bacteria devote considerable metabolic resources and genetic information to the ability to sense the environment and move towards or away from specific stimuli using flagella. In Escherichia coli and related species, motility is regulated by several global regulatory circuits, which converge to modulate the overall expression of the master operon for flagellum biosynthesis, flhDC. We now show that the global regulator CsrA of E. coli K-12 is necessary for motility under a variety of growth conditions, as a result of its role as an activator of flhDC expression. A chromosomally encoded flhDC'-'lacZ translational fusion was expressed at three- to fourfold higher levels in csrA wild-type strains than in isogenic csrA mutants. Purified recombinant CsrA protein stimulated the coupled transcription-translation of flhDC'-' lacZ in S-30 extracts and bound to the 5' segment of flhDC mRNA in RNA mobility shift assays. The steady-state level of flhDC mRNA was higher and its half-life was approximately threefold greater in a csrA wild-type versus a csrA mutant strain. Thus, CsrA stimulates flhDC gene expression by a post-transcriptional mechanism reminiscent of its function in the repression of glycogen biosynthesis.

Immunolocalization of CD44 in the dystrophic rat retina.

The distribution of the cell surface adhesion/receptor molecule CD44 was studied in retinas of the Royal College of Surgeons (RCS) rat which exhibits an inherited retinal dystrophy. In this animal model, the retinal pigment epithelium fails to phagocytize shed photoreceptor outer segment material, a membranous debris layer accumulates in the subretinal space and the photoreceptor cells degenerate. Using immunoperoxidase and immunogold labeling, CD44 was localized to Müller cell apical microvilli in normal rat retinas, as noted in other species. For the RCS rat, immunoperoxidase labeling of 18 day and 1 month retinas showed the typical microvillar labeling pattern. At 2 months postnatal, following degeneration of most of the photoreceptors, a more condensed band of microvillar label was observed. At 3 months, when photoreceptor degeneration was virtually complete, only distinct regions of dense label remained between the neural retina and debris zone. Upon ultrastructural and immunogold analysis, these regions were found to contain closely packed Müller cell microvilli. At all ages studied, labeling for CD44 in the inner retina did not increase, as it does in other forms of retinal degeneration which lack a debris zone. However, by 3 months the debris zone was labeled for CD44 indicating that CD44 molecules remain on Müller cell microvilli and processes which have extended into and become part of the debris zone. This may be caused by an altered distribution of still undetermined ligands for CD44 which are present within the interphotoreceptor matrix of the RCS rat retina.

Immunocytochemical localization of CD44 in the mouse retina.

The transmembrane glycoprotein CD44 is a cell adhesion molecule which has recently been localized in a variety of cell types. It mediates cell attachment to extracellular matrix components and also binds to the actin cytoskeleton within the cell. In this study, the presence and distribution of CD44 in the mouse retina was investigated in order to determine whether this molecule might be important for retinal cell adhesion. With immunoperoxidase techniques, positive labeling for CD44 was found at the level of the outer limiting membrane and in the region just sclerad to it. However, from these light microscope results it was not clear if the label was in the photoreceptor inner segments, the Müller cell apical microvilli, or both. In order to answer this question, cryoultramicrotomy and immunogold labeling were used to demonstrate that CD44 is specifically localized to the Müller cell microvilli which appose the interphotoreceptor matrix. Western blotting of a retina homogenate showed that the anti-CD44 antibody is specific for a protein of approximately 90 kDa which is the correct molecular weight for the most abundant form of CD44. These results, along with the known binding characteristics of CD44 for the extracellular matrix, suggest that CD44 could play a role in mediating the attachment of the neural retina to components of the interphotoreceptor matrix.

Identification and molecular characterization of csrA, a pleiotropic gene from Escherichia coli that affects glycogen biosynthesis, gluconeogenesis, cell size, and surface properties.

Current evidence suggests that a few global regulatory factors mediate many of the extensive changes in gene expression that occur as Escherichia coli enters the stationary phase. One of the metabolic pathways that is transcriptionally activated in the stationary phase is the pathway for biosynthesis of glycogen. To identify factors that regulate glycogen biosynthesis in trans, a collection of transposon mutants was generated and screened for mutations which independently increase or decrease glycogen levels and the expression of a plasmid-encoded glgC'-lacZ fusion. The glycogen excess mutation TR1-5 was found to be pleiotropic. It led to increased expression of the genes glgC (ADPglucose pyrophosphorylase) and glgB (glycogen branching enzyme), which are representative of two glycogen synthesis operons, and the gluconeogenic gene pckA (phosphoenolpyruvate carboxykinase), and it exhibited effects on cell size and surface (adherence) properties. The mutated gene was designated csrA for carbon storage regulator. Its effect on glycogen biosynthesis was mediated independently of cyclic AMP (cAMP), the cAMP receptor protein, and guanosine 3'-bisphosphate 5'-bisphosphate (ppGpp), which are positive regulators of glgC expression. A plasmid clone of the native csrA gene strongly inhibited glycogen accumulation and affected the ability of cells to utilize certain carbon sources for growth. Nucleotide sequence analysis, complementation experiments, and in vitro expression studies indicated that csrA encodes a 61-amino-acid polypeptide that inhibits glycogen biosynthesis. Computer-assisted data base searches failed to identify genes or proteins that are homologous with csrA or its gene product.

Amplification of the Golgi complex in MDCK cells secreting human growth hormone.

MDCK cells were transfected with pXGH5, a plasmid containing the human growth hormone (hGH) gene, and permanently expressing cell lines were selected. Clone 3A cells, which secrete quantities of hGH through both apical and basolateral surfaces, were examined in detail. Immunofluorescence analysis using anti-hGH antibody revealed bright perinuclear staining coinciding with the area delineated by anti-p52 kDa protein (a resident Golgi protein) antibody. There appeared to be less Golgi-specific fluorescence in untransfected cells. This difference correlated with an increased amount of 52 kDa in the clone 3A cells. Morphometric analysis was performed on electron micrographs of clone 3A and untransfected cells using the fractionator to estimate average number of Golgi stacks per cell, and values were statistically analyzed. It was found that clone 3A cells contained 3.3 and untransfected cells 1.6 stacks (P < or = 0.005), respectively. When clone 3A cells were placed into defined medium, the synthesis and secretion of hGH declined 4-fold, and the number of Golgi stacks also decreased to the untransfected level within seven days. The number of Golgi stacks per untransfected cell was not affected by the presence of exogenous hGH, indicating that Golgi amplification was directly related to secretory demand. Generation times and cell volumes were identical for both cell types under all growth conditions. In addition, the kinetics of protein secretion from radiolabelled cells demonstrated that clone 3A cells generally secrete lower amounts of endogenously synthesized apical proteins than do untransfected cells, while basolateral secretion remains the same. In both cases hGH comprised only about 10% of total secretory proteins, so that the increase in total protein secretion did not seem to warrant the two-fold elaboration of Golgi by 3A cells. But there might be significant amounts of hGH which traverse the Golgi to end up in lysosomes, rather than being secreted, leading to Golgi amplification.

Routing of a secretory protein to the endocytic compartment in transfected Madin Darby canine kidney cells.

Transfected Madin Darby canine kidney (MDCK) cells (3A) expressing human growth hormone (hGH) contain twice as many Golgi stacks as untransfected cells. How MDCK cells, lacking a regulated pathway, deal with (over)expression of a protein hormone, or any exogenous protein, has not been examined in detail. Since hGH constituted 10% of total secreted proteins, it was not apparent why Golgi amplification was needed, unless some enters a nonsecretory compartment. Studies were undertaken to determine hGH fate. By using an inhibitor of protein synthesis, or by analyzing pulse labeled immunoprecipitated hGH, 20-30% of hGH was shown to remain intracellular even after 4 h. That portion might be localized in the endosome/lysosome compartment, because it is post-Golgi. Immunoelectron microscopy with antibodies against hGH, clathrin, and cathepsin D demonstrated clathrin and hGH colocalized, as did hGH and cathepsin D. The latter were found in large vesicles, but no hGH appeared in lysosomes, due to its degradation. Analysis of isolated lysosome/endosomes revealed vesicles containing both hGH and cathepsin D, but more containing only cathepsin D. Endocytosis studies suggested the 3A basolateral endosomal compartment may be more capacious than normal. Thus, 3A Golgi amplification resulted in an expanded endosome compartment to accommodate secretory protein (over)expression.

Immunohistochemical localization of basic fibroblast growth factor (bFGF) within growing and atretic mouse ovarian follicles.

Basic fibroblast growth factor is a biologically active peptide with a strong affinity for heparin. This growth factor has been previously shown to be mitogenic for a variety of mesoderm and neuroectoderm-derived cells. The immunohistochemical localization of basic FGF within mouse growing and atretic ovarian follicles is presented in the study. Ovarian tissue samples were obtained either (a) randomly from mice housed in a controlled light environment or (b) following the administration of exogenous gonadotropins to stimulate follicle development. Ovarian samples were fixed in Bouin's fluid for no longer than 18 h. Following fixation and paraffin embedding, sections were exposed to a primary antibody made in rabbits against either (a) human recombinant basic FGF or (b) the 1-24 synthetic fragment of bovine basic FGF. The primary antibody was followed by biotinylated goat anti-rabbit IgG and a biotin-avidin-peroxidase complex. There were no differences in the immunolocalization of basic FGF using either source of primary antibody or between randomly obtained ovarian samples and those obtained from mice given exogenous gonadotropins. Basic FGF was immunolocalized in follicle basal laminae and was also closely associated with individual follicle cells during all stages of ovarian follicle development. Basic FGF was absent in the theca interna, oocyte cytoplasm, zona pellucida and follicle fluid of normal growing follicles. Individual corpora luteal cells were surrounded by basic FGF but lacked cytoplasmic staining. Atretic follicles exhibited staining patterns similar to their respective stage of follicle development. However, when present, follicle fluid within atretic follicles was strongly positive for basic FGF. These results indicate that basic FGF may be an important factor involved in intraovarian control mechanisms.

Difference in the response to PIF/activin between animal caps excised from mid- or late blastula stages of Xenopus laevis.

Animal caps from Xenopus embryos at Stage 7/8 cultured in salt solution, were capable of elongating, and formed various embryonic tissues including axial mesoderm, nervous tissues and pigmented retina. In contrast, animal caps from Stage 9 only developed into permanent blastulae. Following exposure to PIF/activin, however, such animal caps displayed morphogenetic movements. They formed tube-shaped embryoids that frequently had one or more tails but no head. We conclude that animal caps from Stage 9 produce more reliable results than those from Stage 7/8 because the latter are most likely contaminated with mesodermal cells.

An ultrastructural study of in-vitro interaction of guinea-pig and mouse blastocysts with extracellular matrices.

Guinea-pig (intrusive) and mouse (displacement) blastocysts display different cellular mechanisms of implantation. Blastocysts were placed in CMRL-1066 supplemented with either 10 or 20% fetal calf serum, 0.1M L-glutamine and antibiotics and then transferred to dishes previously coated with either Matrigel or type I collagen. After culture for 48 or 72 h, the dishes were processed for transmission electron microscopy. Blastocysts had attached to both extracellular matrices by 48 h. Matrigel elicited minimal trophoblast cell activity. Trophoblast cell projections were oriented parallel to the Matrigel and displayed little invasive activity, but trophoblast cells displayed active interaction with type I collagen. By 72 h, trophoblast cells exhibited slender, anastomosing projections which extended into the collagen matrix. Bundles of microfilaments running parallel with the long axis of the projections were observed. The morphology of type I collagen was altered in the immediate vicinity of the trophoblast projections. The projections interdigitated and desmosomes developed between processes. Projections appeared to meet, fuse and entrap matrix. These results suggest that trophoblast cells do not significantly interact with Matrigel, but penetrate into type I collagen.

A method for using low-temperature embedding media for electron microscopy of cells grown on microporous supports.

Attempts were made to embed Madin-Darby canine kidney (MDCK) cells grown on nitrocellulose microporous supports in Lowicryl, a medium designed for the retention of antigenicity for electron microscopic immunocytochemical studies. It was found that the membrane fragmented during the prescribed embedding procedure leading to loss of the cell sample. This necessitated the formulation of a new combination of fixative and dehydration agents which would allow: (i) preservation of membrane and cellular integrity; (ii) infiltration and embedding in Lowicryl; and (iii) detection of a specific antigen in thin sections. The cellular monolayer and organelle profiles were best preserved with glutaraldehyde fixation at room temperature followed by ethanol dehydration. Since the latter was carried out at temperatures attainable with an ice-salt bath, there was no need for a special ultralow-temperature apparatus. This procedure was applied to a MDCK cell clone that consisted of stable secretors of human growth hormone (hGH), as a result of being transfected with a plasmid containing the hGH gene. Thus, it was demonstrated that hGH could be detected by immunogold labeling of thin sections and localized to specific cellular structures. The procedure developed in this report is applicable to cells grown on two other supports and may be extended further.

Ultrastructure of oocyte migration through the mouse ovarian surface epithelium during neonatal development.

Previous studies using light microscopy have described the presence of oocytes within the ovarian surface epithelium (OSE) of the neonate ovary and their subsequent release into the periovarian space. The ultrastructural examination and quantitative estimate of oocyte migration through the OSE is described in this study. The surface of the mouse ovary is covered by a simple squamous to simple cuboidal epithelium resting on a distinct basal lamina. Healthy, non-atretic primordial follicles located in the periphery of the ovarian cortex interact with the OSE. Oocytes within the primordial follicles are large (50-70 microns), spherical cells surrounded by a single layer of squamous granulosa cells. Migratory oocytes initially display a ruffled border and extend pseudopodia-like cellular extensions towards the OSE. These processes subsequently compromise the basal lamina of the OSE and extend between the epithelial cells. Granulosa cells retract as junctional complexes between them and the oocyte are no longer observed. The oocyte migrates between the OSE cells and passes into the periovarian space without the loss of either OSE or granulosa cells. The rate of oocyte migration reaches a peak during the first week of neonatal development and then gradually diminishes until by day 28 of development no oocyte migration was observed.

Ultrastructural and immunocytochemical characterization of cultured cells from rat molar stellate reticulum.

By scanning electron microscopy, the cultured cells were stellate and had numerous filipodia--characteristics of stellate reticulum cells in vivo. By transmission electron microscopy, they contained bundles of intermediate filaments, numerous mitochondria, large electron-dense granules and desmosomes--all features of the stellate reticulum in vivo. Moreover, the stellate reticulum was the only region of the enamel organ in vivo that contained large, electron-dense granules. By immunocytochemistry, the cultured cells contained cytokeratins, confirming their epithelial nature, and stellate reticulum cells in vivo and in vitro did not have an EGF receptor. Thus, these combined ultrastructural and immunocytochemical findings suggest that the cell culture was of stellate reticulum.