BACKGROUND Ectopic expression of gastric intrinsic factor (IF) has been described in rodent models of chronic gastritis. AIMS The current study undertook to determine if ectopic IF was also present in chronic gastritis in humans and might identify the process of ectopic protein expression as part of the response to chronic injury. METHODS Archived biopsies from mid-body, angularis and prepylorus of 9 patients with and without chronic gastritis and food-cobalamin malabsorption were examined in a blinded fashion by immunocytochemistry as were biopsies from 5 normal subjects. Cells with ectopic IF were further examined with antibodies against pepsin or with Griffonia simplicifolia II (GSII) to identity cells in the mucous neck cell compartment. RESULTS Ectopic IF production in non-parietal cells was identified in cells that were H(+),K(+)-ATPase-negative but IF-positive in 7 of the 9 patients (6/9 in the angularis and/or prepylorus biopsies and 1/9 only in the mid-body). These included 5 of the 6 H. pylori-infected patients and all 5 patients with severe food-cobalamin malabsorption. No normal control subjects demonstrated ectopic IF. The cells with ectopic IF were pepsinogen-positive peptic cells and were not GSII-positive. Expression was most extensive in patients and gastric regions with inflammation. In all but one sample, ectopic IF was observed near anatomical mucosal junctions, such as antral/body and prepylorus/duodenum junctions. CONCLUSIONS These data in humans with and without gastritis are consistent with the hypothesis that local factors influence ectopic gastric IF expression, arising from either the anatomical location, the focal inflammation, or both.

Paramecium bursaria cells harbor several hundred symbiotic algae in their cytoplasm. Algae-free cells can be reinfected with algae isolated from algae-bearing cells or cultivated Chlorella species through the digestive vacuoles. To determine the relationship between the infectivity of various Chlorella species and the nature of their cell wall components, algae-free P. bursaria cells were mixed with 15 strains of cultivated Chlorella species and observed for the establishment of endosymbiosis at 1 h and 3 weeks after mixing. Only 2 free-living algal strains, C. sorokiniana C-212 and C. kessleri C-531, were maintained in the host cells, whereas free-living C. sorokiniana C-43, C. kessleri C-208, C. vulgaris C-27, C. ellipsoidea C-87 and C-542, C. saccharophila C-183 and C-169, C. fusca var. vacuolata C-104 and C-28, C. zofingiensis C-111, and C. protothecoides C-150 and C-206 and the cultivated symbiotic Chlorella sp. strain C-201 derived from Spongilla fluviatilis could not be maintained. These infection-incapable strains could escape from the host digestive vacuole but failed to localize beneath the host cell membrane and were eventually digested. Labeling of their cell walls with Alexa Fluor 488-conjugated wheat germ agglutinin, GS-II, or concanavalin A, with or without pretreatment with 0.4 N NaOH, showed no relationship between their infectivity and the stainability with these lectins. Our results indicate that the infectivity of Chlorella species for P. bursaria is not based on the sugar residues on their cell wall and on the alkali-insoluble part of the cell wall components, but on their ability to localize just beneath the host cell membrane after escaping from the host digestive vacuole.

Chronic inflammation of the gastric epithelium is believed to induce mucosal changes that can eventually develop into gastric cancer. In gastrin-deficient (G-/-) mice exhibiting chronic inflammation in the hypochlorhydric stomach, we documented a prominent fundic mucous cell lineage sharing morphological similarity with preneoplastic changes reported in Helicobacter-infected mice. To study the identity and origin of this cell lineage, we screened for different gastric mucosal cell markers. The clusters of large, foamy cells stained for trefoil factor 2 (TFF2/SP), MUC6 and the lectin Griffonia Simplicifolia II (GSII), but not for the intestine-specific transcription factor Cdx2, suggested that they arise from gastric mucous neck cells. Ki67-labeled GSII-positive neck cells in Helicobacter felis-infected, but not G-/- stomachs, suggested that mucous neck cell proliferation accounted for expansion of this compartment in the H. felis model of gastritis, but not the G-/- model. Using RNase protection assays and quantitative PCR, we found that interferon gamma (IFNgamma) was the most abundant proinflammatory cytokine in the G-/- stomach. We also found that this Th1 cytokine can increase the abundance of mucous neck cells, since its infusion into mice recapitulated the appearance of these cells as observed in both G-/- and H. felis-infected mice. Using the human gastric cell line NCI-N87, we showed that IFNgamma induces the secretion of mucus and expression of MUC6, TFF2 and pepsinogen II, but not of pepsinogen I and intrinsic factor. In conclusion, our results demonstrate that inflammation, specifically the proinflammatory cytokine IFNgamma, induced expansion of the fundic mucous neck cell compartment, which likely represents both increased mucus production and cell number.

The structural glycoprotein gene gp41 homologue of Spodoptera litura nucleopolyhedrosis virus (SpltNPV-I *) was identified in the 4.0 kb EcoRI-L fragment of the viral genome. The nucleotide sequence of 2063 bp of this fragment revealed an open reading frame of 1014 nucleotides to encode a polypeptide of 337 amino acids. Analysis of nucleotide and deduced amino acid sequences of the putative ORF indicated its identity with gp41 protein of other baculoviruses sharing maximum homology with that of Spodoptera frugiperda nucleopolyhedrosis virus (SfNPV). The coding sequence was preceded by an AT-rich region containing the consensus baculoviral late promoter motif RTAAG. The putative SpltNPV gp41 ORF was abundantly expressed as a 37 kDa apoprotein in E. coli and as a 50 kDa glycoprotein in Sf9 cells. The recombinant protein expressed in insect cells was glycosylated (20%) and has GlcNAc as the terminal sugar. The gene is conserved among baculoviruses and places SpltNPV-I close to Spodoptera frugiperda and Spodoptera exigua NPVs in phylogenetic tree.

The in vivo binding pattern of the lectin Griffonia simplicifolia II (GSLII) was evaluated in sections of adult cat optic nerve following reports that it is a marker for oligodendrocytes in adult rodent CNS and that it may also be an oligodendroglial lineage marker. Following as closely as possible the immunocytochemical methodology employed in these reports, staining for GSLII was incorporated into sets of consecutive one micron thick sections comprising known cell-type specific reference markers backed up by electron microscopy. With this correlative protocol both lectin positive and lectin negative cells could be reliably identified. The material examined included normal control tissue and tissue containing previously studied demyelinating lesions of various ages in which oligodendrocyte progenitors and precursors have been characterized. GSLII was found to stain not only mature oligodendrocytes in adult cat optic nerve but also activated microglia, macrophages, polymorphonuclear leucocytes and other haematogenous cells. Lectin positivity was not found in oligodendroglial precursors, endothelial cells, astrocytes or ramified microglia. This study emphasises that care needs to be taken before assigning lineage marker status to individual lectins or antibodies.

We recently described the cloning of putative human CMP-sialic acid transporter (hCST) cDNA [Ishida, N. et al.(1996) J. Biochem. 120, 1074-1078]. The hCST cDNA coded for a hydrophobic protein with an amino acid sequence showing a high degree of similarity (92% identity) to that of murine CMP-sialic acid transporter. In this report, we demonstrate that hCST corrects the CMP-sialic acid transporter-deficient phenotype of CHO-derived Lec2 cells, as judged from the recovery of WGA-sensitivity by transformants, and the recovery of CMP-sialic acid transporting ability by microsomal vesicles prepared from them. A peptide antibody against the C-terminus of the hCST protein detected the cDNA products expressed in the microsomes of the transformants. The subcellular localization of the hCST protein in the Golgi membrane was demonstrated by immunofluorescence microscopy, using the hCST-specific antibody. These results clearly indicate that hCST cDNA encodes the human CMP-sialic acid transporter protein. Plasma membrane-selective permeabilization combined with immunofluorescence microscopy provided strong evidence that the C-terminus of the human CMP-Sia transporter is exposed to the cytosol on the outer surface of the Golgi membrane.

Borna disease virus is a nonsegmented negative-strand RNA virus that causes neurologic disease in a wide variety of animal hosts. Here we describe identification and characterization of the first glycoprotein in this viral system. The 18-kDa glycoprotein, gp18, has been purified from infected rat brain. Isolation and microsequencing of this protein allowed identification of a 16.2-kDa open reading frame in the viral antigenome. Lectin binding and endoglycosidase sensitivity assays indicate that gp18 is an unusual N-linked glycoprotein.

Bandeiraea simplicifolia plant seeds contain a family of five alpha-D-galactopyranosyl-binding isolectins (BS I-A4, A3B, A2B2, AB3, B4) and N-acetyl-D-glucosamine-binding lectin (BS II). After Pi/NaCl extraction and ammonium sulfate fractionation BS II is specifically adsorbed onto p-aminobenzyl-1-thio-N-acetyl-beta-D-glucosaminide-succinylaminohexylaminyl--Sepharose-4B. The BS I isolectins pass through this column and BS II is selectively eluted by Pi/NaCl containing 2 mM N-acetyl-D-glucosamine or by 0.1 M sodium acetate buffer pH 3.6. The material not bound to the column is loaded onto p-aminophenyl-beta-D-galactopyranosyl-succinylaminohexylaminyl--Sepharose-4B. BS I-A4 is specifically eluted in a sharp peak with Pi/NaCl containing 1 mM N-acetyl-D-galactosamine. Then, BS I-A3B, A2B2, AB3 and B4 are selectively eluted, in a single peak for each isolectin. with Pi/NaCl containing 2 mM, 8 mM, 15 mM and 50 mM methyl alpha-D-galactopyranoside, respectively.