Regular structures in membranes. I. Membranes in the endocytic complex of ileal epithelial cells.

An "apical endocytic complex" in the ileal lining cells of suckling rats is described. The complex consists of a continuous network of membrane-limited tubules which originate as invaginations of the apical plasma membrane at the base of the microvilli, some associated vesicles, and a giant vacuole. The lumenal surface of this tubular network of membranes and associated vesicles is covered with a regular repeating particulate structure. The repeating unit is an approximately 7.5-nm diameter particle which has a distinct subunit structure composed of possibly nine smaller particles each approximately 3 nm in diameter. The approximately 7.5-nm diameter particles are joined together with a center-to-center separation of approximately 15 nm to form long rows. These linear aggregates, when arranged laterally, give rise to several square and oblique two-dimensional lattice arrangements of the particles which cover the surface of the membrane. Whether a square or oblique lattice is generated depends on the center-to-center separation of the rows and on the relative displacement of the particles in adjacent rows. Four membrane faces are revealed by fracturing frozen membranes of the apical tubules and vesicles: two complementary inner membrane faces exposed by the fracturing process and the lumenal and cytoplasmic membrane surfaces revealed by etching. The outer membrane face reveals a distinct array of membrane particles. This array also sometimes can be seen on the outer (B) fracture face and is sometimes faintly visible on the inner (A) fracture face. Combined data from sectioned, negatively stained, and freeze-etched preparations indicate that this regular particulate structure is a specialization that is primarily localized in the outer half of the membrane mainly in the outer leaflet.

Role of microvilli in surface changes of synchronized P815Y mastocytoma cells.

The surface morphology of synchronized P815Y mastocytoma cells has been examined by scanning electron microscopy. Early G1 cells are comparatively smooth or light villated, whereas at later stages the surface becomes progressively more villated. In G1 cell most microvilli have a uniform diameter, whereas in S and G2 cells, many microvilli show branching and often originate from much larger surface protuberances. Small "blebs" are seen on the surface of many cells but these structures do not appear to be a characteristic feature of cells at any one stage of the cycle. The presence of microvilli increases the total surface of the cell to such an extent that the ratio of volume to surface area remains constant throughout the cell cycle. The mechanism of cytokinesis is thus a physical one, involving the unfolding of previously accumulated microvilli.

Regular structures in unit membranes. III. Further observations on the particulate component of the suckling rat ileum endocytic membrane complex.

Further morphological observations on the particulate components decorating the lumenal surfaces of membranes of the endocytic complex of the epithelial cells of the suckling rat ileum are presented. The particles each measure approximately 7.5 nm across and give the appearance of the capital letter H in frontal view. They consist of the enzyme n-acetyl-beta-glucosaminidase (NAG). They are arranged in rows called "decorated strips" with the symmetrical lateral bars in register and spaced approximately 14.5 nm apart. Decorated strips lie side-by-side in the external (lumenal) surface of the membrane. They are parallel and sometimes spaced approximately 14.5 nm apart making an orthogonal lattice. The lateral spacing between the decorated strips under certain conditions is reduced and sometimes there is shear between the adjacent ones. Occasionally, shear is present within the decorated strips themselves, with slight displacement of the two sides of each H-shaped particle. A purified preparation of these membranes has been studied by electron microscopy using thin sectioning, negative stain, Markham translation and optical diffraction computer image reconstruction methods. The individual particles comprising the array can be seen in the membrane surface in profile view when dried in a pool of negative stain. They appear either triangular or diamond-shaped in such views. If triangular, they appear to consist of three domains at the corners of an equilateral triangle. One side of each triangular figure is parallel to the membrane surface but separated from it by a dense band of negative stain approximately 2 nm thick that runs along the surface of the membrane. Sometimes a fourth symmetrical domain is visible within this dense band, giving a diamond-shaped figure. This fourth domain connects the particle to the membrane. Thus, each H-shaped particle is a double structure, with each half in profile view appearing as a diamond figure of four symmetrical domains. Each H-shaped particle is believed to consist of either two or four molecules of NAG.

Intimin and the host cell--is it bound to end in Tir(s)?

Intimate bacterial adhesion to the intestinal epithelium is a pathogenic mechanism shared by several human and animal enteric pathogens, including enteropathogenic and enterohaemorrhagic Escherichia coli. Two bacterial protein partners involved in this intimate association have been identified, intimin and Tir. Some key remaining questions include whether intimin specifically interacts with one or more host-cell-encoded molecules and whether these contacts are a prerequisite for the subsequent intimate intimin-Tir association. Recent data support the hypothesis that the formation of a stable intimin-Tir relationship is the consequence of intimin protein interactions involving both host and bacterial components.

Purification of a 20 kDa phosphoprotein from epithelial cells and identification as a myosin light chain. Phosphorylation induced by enteropathogenic Escherichia coli and phorbol ester.

Previous studies on the mechanism of enteropathogenic Escherichia coli (EPEC) infection have revealed an increase in the phosphorylation state of a number of proteins in human laryngeal HEp-2 cells. The most prominent was an acidic phosphoprotein(s) of Mr 20-21 kDa. The present study reports: (a) a simple method for purification of phosphorylated 20 kDa protein; (b) identification of the 20 kDa phosphoprotein as myosin light chain; and (c) that the phorbol ester, TPA, also increased the phosphorylation of the 20 kDa myosin light chain. In contrast to the effects of EPEC, TPA stimulation resulted in the dissociation of myosin from the cytoskeleton to the cytosol.

Virulence properties of Enterobacteriaceae isolated from the small intestine of children with diarrhea.

Enterobacteriaceae isolated from the duodena of Peruvian children with persistent diarrhea (PD) have been examined for virulence factors and compared with Enterobacteriaceae isolated from children with acute diarrhea, those convalescent from PD and diarrhea-free controls. Escherichia coli were isolated from 42 of 186 (23%) of the aspirates. All 11 children with PD in whom multiple E. coli colonies were examined were colonized by a single serotype. DNA probes identified enterotoxigenic E. coli in 2 of 89 (2.2%) PD aspirates and 2 of 38 (5.3%) acute diarrhea aspirates and enteroaggregative E. coli in one PD and one control aspirate. Strains positive with the enteropathogenic E. coli adherence factor probe were identified from 2 of 89 (2.2%) patients with PD and 1 of 34 (2.9%) controls. A subset of 12 E. coli strains failed to show adhesion to human duodenal enterocytes although 5 of 9 showed sparse but polar attachment to ileal cells from a child with short bowel syndrome and PD. Three of 10 Enterobacteriaceae (two E. coli, one Klebsiella species) caused diarrhea in the reversible ileal tie adult rabbit model. Colonization with virulent Enterobactericeae did not explain the majority of episodes of PD. Examination of these duodenal bacteria in the rabbit model revealed some that caused diarrhea but were not recognized pathogens.

Carbon dioxide regulated secretion of the EaeB protein of enteropathogenic Escherichia coli.

An eaeA mutant (intimin deficient) of enteropathogenic Escherichia coli stimulated phosphorylation of several host cell proteins, showing that intimate adherence is not required to activate signal transduction pathways in enteropathogenic E. coli-infected cells. Growth of enteropathogenic E. coli in tissue culture medium in 5% CO2, in the presence or absence of cultured cells, resulted in the secretion of several bacterial proteins. Two of these, 36 kDa and 20 kDa in size, were expressed at significantly lower levels in air. N-terminal sequencing and analysis of secreted proteins of an eaeB mutant indicated that the 36 kDa secreted protein was EaeB, previously implicated in the stimulation of signalling pathways in enteropathogenic E. coli-infected cells.

An enteroaggregative Escherichia coli strain of serotype O111:H12 damages and invades cultured T84 cells and human colonic mucosa.

The pathogenic mechanisms of enteroaggregative Escherichia coli (EAEC) are not well defined. We investigated the interaction of EAEC strain 236 (serotype O111:H12) with polarised Caco-2 and T84 human intestinal epithelial cells lines, and with human jejunal and colonic mucosa. Strain 236 adhered to both polarised cell lines and to both intestinal tissue types, but caused severe damage and was invasive only in T84 cells and colonic mucosa. In contrast, prototype EAEC strain 042, which also adhered to the cultured intestinal cell lines, did not adhere to or invade jejunal or colonic tissue. These observations suggest a heterogeneity of virulence properties within the EAEC category of diarrhoea-causing E. coli.

Molecular and ultrastructural characterisation of EspA from different enteropathogenic Escherichia coli serotypes.

Enteropathogenic Escherichia coli (EPEC) encode a type III secretion system located on a pathogenicity island known as the locus for enterocyte effacement. Four proteins are known to be exported by this type III secretion system--EspA, EspB and EspD required for subversion of host cell signal transduction pathways and a translocated intimin receptor protein (Tir) required for intimin-mediated intimate attachment and attaching and effacing lesion formation. The espA gene is located within the locus for enterocyte effacement and the EspA polypeptide from the prototype EPEC strain E2348/69 (O127:H6) has recently been shown to be a component of a filamentous structure involved in bacteria-host cell interaction and locus for enterocyte effacement-encoded protein translocation involved in attaching and effacing lesion formation. In this study we have extended our investigation of EspA to strains belonging to other classical EPEC serotypes. DNA sequencing demonstrated that the espA gene from the different EPEC strains share at least 65% DNA identity. In addition, we detected morphologically and antigenically similar EspA filaments in all but one of the bacterial strains examined including recombinant, non-pathogenic E. coli expressing espA from a cloned locus for enterocyte effacement region (HB101(pCVD462)).

Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli.

A collection of 44 Campylobacter isolates (37 C. jejuni and seven C. coli) from children with colitis (21 strains) or watery diarrhoea (23 strains) was analysed for toxin production, association with HeLa cells, and invasion of differentiated Caco-2 cell cultures. There was no obvious association of clinical symptoms with species, biotype or enterotoxin production. All colitis strains and most of the isolates from watery diarrhoea were cytotoxic for Chinese hamster ovary cells. Measurements of bacterial association indices with HeLa cells varied with time, and were considered to be unreliable for discriminating between isolates from the two diagnostic groups. Statistically significant differences were observed between the two groups (all colitis strains and 65% of strains from non-inflammatory diarrhoea) with respect to invasion of both HeLa and Caco-2 cell monolayers. However, among the strains from non-inflammatory diarrhoea that did invade, numbers of internalised bacteria were similar to the range observed for colitis strains. Of the colitis strains, 86% were able to transcytose through polarised Caco-2 monolayers grown on filters, compared with 48% of isolates from non-inflammatory disease. We propose the use of Caco-2 cells as a model for studying invasion of intestinal epithelia by C. jejuni and C. coli.

Pathological changes in the rabbit ileal loop model caused by Campylobacter jejuni from human colitis.

Four strains of Campylobacter jejuni isolated from children with inflammatory diarrhoea were assayed in the rabbit ileal loop model of infectious diarrhoea. All caused inflammatory reactions with severe macroscopic and microscopic damage in infected rabbit ileal tissue similar to that observed in the patients by endoscopy and histological analysis of colonic biopsies. Haemoglobin and other proteins were observed in loop fluids, consistent with leakage of serum from damaged mucosa. Loop fluids also contained significant bicarbonate concentrations, indicative of an active secretory component similar to that in control loops inoculated with cholera toxin. However, although three of the four clinical strains produced small amounts of a protein immunologically related to cholera toxin in vitro, none such was detected in either tissues or fluids of infected ileal loops. We propose instead that host-derived mediators of secretion may be important in pathogenesis. A mutant strain of C. jejuni with impaired motility, obtained from the National Collection of Type Cultures, did not induce tissue damage or fluid secretion in rabbit ileal loops.

Cellular responses to enteropathogenic Escherichia coli infection.

Enteropathogenic Escherichia coli (EPEC), first described in the 1940's and 1950's, remain an important cause of severe infantile diarrhoea in many parts of the developing world. EPEC do not produce enterotoxins and are not invasive; instead their virulence depends upon exploitation of host cell signalling pathways and the host cell cytoskeleton both as a means of colonizing mucosal surfaces of the small intestine and causing diarrhoea. Following initial mucosal attachment, EPEC secrete 'signalling' proteins and express a surface adhesin, intimin, to produce 'attaching & effacing' lesions in the enterocyte brush border membrane characterised by localised destruction of brush border microvilli, intimate bacterial adhesion and cytoskeletal reorganisation and accretion beneath attached bacteria. The pathophysiology of EPEC diarrhoea is also complex and probably results from a combination of epithelial cell responses including both electrolyte secretion and structural damage.

Studies of membrane fusion. III. Fusion of erythrocytes with polyethylene glycol.

Freeze-fracture electron microscopy has been used to investigate the mechanism of polyethylene glycol-induced cell fusion. Interaction of cells with the high concentrations of polyethylene glycol required for cell fusion results in cell agglutination with large planar areas of very close contact between adjacent cell membranes. An aggregation of intramembrane particles into large patches at the sites of cell-cell contact accompanies cell agglutination. Fusion occurs following the removal of most of the PEG when cells only remain in close contact at small (approximately 0.1 micrometer diameter) plaques of smooth membrane resulting in cells connected by one (or more) small cytoplasmic connexions. Expansion to form spherical fused cells occurs by a process of cell swelling.

Studies of membrane fusion. IV. Fusion of HeLa cells with Sendai virus.

The Sendai virus-induced fusion of HeLa cells has been studied by freeze-fracture electron microscopy. Freeze-fracture observations confirm previous scanning electron-microscope studies (1977) and show that at 4 degrees C virus particles bind to the cell surface and that cell agglutination results from the crosslinking by virus particles of microvilli on adjacent cells. Incubation at 37 degrees C initiates a change in viral envelope structure and fusion of 'altered' virus particles with the cell plasma membrane. Fusion of a virus particle with two crosslinked cells is probably the membrane fusion event which initiates cell-cell fusion; fusion is completed as a result of virally induced cell swelling. Lateral diffusion of viral envelope components following virus-cell fusion and, in some instances, an aggregation of plasma membrane intramembrane particles occurs in swollen cells. These observations show that the mechanisms of viral envelope-cell and probably cell-cell fusion are the same as have been reported for erythrocytes. Although endocytosis of intact virus particles does occur, the specialized cell-mediated mechanism for fusion of the viral envelope with the cell plasma membrane suggests that this, and not viropexis, is the mechanism of Sendai virus infection.

Studies of membrane fusion. V. Fusion of erythrocytes with non-haemolytic Sendai virus.

The fusion of human erythrocytes with non-haemolytic '1-day' Sendai virus has been studied by electron microscopy. The mechanism of viral envelope-cell fusion is the same as that described previously for haemolytic '3-day' Sendai virus except that fusion is frequently arrested at an initial stage when 2 segments of smooth linear viral membrane fuse and become incorporated into the erythrocyte membrane. After longer periods of incubation at 37 degrees C, in addition to many partly fused virus particles, long (up to 4 micrometer) lengths of smooth linear viral membrane are seen within the erythrocyte membrane which arise by linear aggregation of shorter (approximately 0.25 micrometer long) segments of smooth linear membrane derived from individual fused viral envelopes. Cell-Cell fusion, as a result of the fusion of a viral envelope with 2 adjacent erythrocytes also occurs but, in the absence of cell swelling, fusion is arrested at this stage with cells joined by one (or more) small cytoplasmic bridges. Typical fused cells are produced if such cells are swollen with hypotonic buffer. These observations provide further evidence that membrane fusion and cell swelling are distinct events in cell fusion and that cell swelling is the driving force both for completing the incorporation of the viral envelope into the cell membrane and for expanding cells connected by small cytoplasmic bridges to form spherical fused cells. Little lateral diffusion of viral envelope components occurs in the absence of cell swelling; in fact, some aggregation of components occurs. Comparison with previous studies using haemolytic '3-day' Sendai virus suggests that virally induced cell swelling perturbs membrane structure so as to allow the rapid lateral diffusion of integrated viral envelope components.