Attenuated Vibrio cholerae vaccine strains specifically mutated in genes encoding cholera toxin (CT) are still capable of causing mild to moderate diarrhea. Culture supernatants of V. cholerae strains, both CT-positive and CT-negative, were examined in Ussing chambers, and a toxin was found that increases the permeability of the small intestinal mucosa by affecting the structure of the intercellular tight junction, or zonula occludens. The activity of this toxin is reversible, heat-labile, sensitive to protease digestion, and found in culture supernatant fractions containing molecules between 10 and 30 kDa in size. Production ofthis factor (named ZOT for zonula occludens toxin) correlates with diarrheagenicity of V. cholerae strains in volunteers and may represent another virulence factor of infectious diarrhea that must be eliminated to achieve a safe and effective live oral vaccine against cholera.Vibrio cholerae produces the copious diarrhea characteristic of cholera by means of a potent enterotoxin, cholera toxin (CT). The A subunit of CT, encoded by ctxA, stimulates adenylate cyclase in intestinal epithelial cells, which results in net secretion of fluid into the intestinal lumen (1). Initial recombinant V. cholerae vaccine strains, attenuated by removal of the ctxA gene, were greatly reduced in their ability to induce diarrhea in volunteers (2). However, despite the absence of CT, these strains were still capable of inducing an unacceptable amount of diarrhea. One such strain, V. cholerae CVD101, is a ctxA deletion mutant of V. cholerae strain 395 in which 94% of the sequence encoding the A1 peptide of CT has been removed (3). When evaluated in volunteer studies, CVD101 caused mild to moderate diarrhea (mean stool volume of 0.9 liter with a range of 0.3-to 2.1 liters) in 54% of subjects ingesting this organism (2). While greatly attenuated compared with the parent strain 395 [which induces a mean diarrheal stool volume of 5.5 liters with a range of 0.3-44 liters in >90% of volunteers (4)], the amount of diarrhea induced by CVD101 is still unacceptable for use of this strain as a vaccine.Given the magnitude of the diarrhea induced in the absence of CT, we hypothesized that V. cholerae produced a second toxin, which was still present in strains deleted of the ctxA sequence. To investigate this hypothesis, we examined V. cholerae strains with and without intact ctx genes by using rabbit intestinal tissue mounted in Ussing chambers, a classic technique for studying the process of transport across intestinal tissue (5, 6). The results indicate that V. cholerae produces a toxin that increases intestinal tissue conductance by altering the structure of intercellular tight junctions. Production of this toxin correlates with diarrheagenicity of V. cholerae strains in volunteers and may represent another virulence mechanism of infectious diarrhea. MATERIALS AND METHODSBacterial Strains and Growth Conditions. V. cholerae 395 is a classical Ogawa CT-positive strain that has been extensively studied in volunteer...

The number of large intramembrane particles associated with sites of synaptic vesicle release at the squid giant synapse was determined and compared to the average maximal presynaptic calcium current in order to derive an estimate of the conductance each particle would have ifit were a calcium channel. This value, 0.21 pS, compares favorably with conductances of calcium channels in other preparations, substantiating the idea that the large intramembrane particles, which are concentrated at "active zones," represent-calcium channels.The entry of calcium into a synaptic terminal through voltagedependent channels is known to be a key step in the release of neurotransmitter (1-3). Voltage-sensitive ionic channels are presumably proteins located in the plasma membrane (4). If calcium currents depend on structures analogous to sodium channels that span the thickness of the plasma membrane (5), they should, likewise, span the presynaptic membrane. Components spanning cell membranes can be visualized with the freeze-fracture technique, and indeed this technique has revealed a class oflarge intramembrane particles that characterize both neuromuscular and interneuronal synapses (6, 7). It has been proposed that these large intramembrane particles are necessary for transmitter release and that they may be the calcium channels (8).Transmitter release by exocytosis occurs in the vicinity of circumscribed patches oflarge intramembrane particles at neuromuscular junctions and at other synapses (6-9). The giant synapse of the squid Loligo pealei is no exception; here also patches of unusually large intramembrane particles define presynaptic "active zones" for synaptic vesicles exocytosis (10,11 (3), indicating a close proximity between the sites ofcalcium entry and the sites ofvesicular transmitter release (3). The present work uses data from freeze-fracture replicas and thin sections to estimate the total number of large particles at active zones in the squid giant synapse. Combining the number of presynaptic active zone particles with recent measurements ofthe calcium conductance through the presynaptic membrane of this synapse (3,(12)(13)(14), we calculate the conductance that a single particle ought to have if it were a calcium channel, and compare this value to that measured for single calcium channels in other systems.Stellate ganglia from small L. pealei (Marine Biological Laboratory, Woods Hole, MA) were excised and tested for the presence of normal synaptic transmission. They were fixed immediately (without further electrical stimulation). Next, using a tissue chopper, the ganglion was sectioned (125 jum) perpendicular to the long axis of the pallial nerve. Sections containing the giant synapse were sandwiched between a pair ofgold specimen carriers, frozen, and fractured, using the complementary replica holder. Three additional ganglia, two small and one large, were embedded in Araldite and sectioned perpendicular to the long axis ofthe pallial nerve. These synapses were serially sectioned by taking 20-30 thi...

Obscurin (approximately 800 kDa) in striated muscle closely surrounds sarcomeres at the level of the M-band and Z-disk where, we hypothesize, it participates in the assembly of the contractile apparatus and membrane systems required for Ca2+ homeostasis. In this study, we used small inhibitory RNA (siRNA) technology to reduce the levels of obscurin in primary cultures of skeletal myotubes to study its role in myofibrillogenesis and the organization of the sarcoplasmic reticulum (SR). siRNA-treated myotubes showed a specific and dramatic reduction in the approximately 800 kDa form of obscurin by reverse transcription-polymerase chain reaction, immunoblotting, and immunofluorescence. M-bands and A-bands, but not Z-disks or I-bands, were disrupted when the synthesis of obscurin was inhibited. Small ankyrin 1, an integral protein of the network SR that binds to obscurin, also failed to align around developing sarcomeres in treated myotubes. Myosin and myomesin levels were significantly reduced in treated myotubes but alpha-actinin was not, suggesting that down-regulation of obscurin destabilizes proteins of the M-band and A-band but not of the Z-disk. Our findings suggest that obscurin is required for the assembly of the M-band and A-band and for the regular alignment of the network SR around the contractile apparatus.

The G-protein subunit ␣-gustducin, which is similar to rod transducin, has been implicated in the transduction of both sweet-and bitter-tasting substances. In rodents, there are differences in sensitivity to sweet and bitter stimuli in different populations of taste buds. Rat fungiform taste buds are more responsive to salts than to sweet stimuli, whereas those on the palate respond predominantly to sweet substances. In contrast, hamster fungiform taste buds are more sensitive to sweet-tasting stimuli. Taste buds in the vallate and foliate papillae of both species are sensitive to bitter compounds. These differences in sensitivity should be reflected in the numbers of gustducin-containing cells in different taste bud populations. We examined taste buds in the rat and hamster for immunoreactivity to an antibody against ␣-gustducin. Immunofluorescence of labeled taste cells was examined by confocal microscopy, and the cells were counted. Gustducin-positive cells were seen in all taste bud regions; they were spindleshaped, with circular cross-sections and apical processes that extended to the taste pore. Cells with this characteristic shape in rat vallate taste buds are Type II (light) cells. In the rat, taste buds of the fungiform papillae had fewer gustducin-positive cells (3.1/taste bud) than those of other regions, including the posterior tongue and palate (Ͼ8.9/taste bud). Hamster fungiform taste buds contained twice as many gustducin-expressing cells (6.8/taste bud) as those of the rat. These data support the hypothesis that ␣-gustducin is involved in the transduction of both sweet-and bitter-tasting stimuli by mammalian taste receptor cells. Key words: taste receptors; gustation; fungiform papillae; vallate papilla; palate; epiglottis; ␣-gustducin; G-protein; taste budsThe transduction of both sweet-and bitter-tasting substances is thought to involve membrane-bound receptors coupled to secondmessenger systems (Kinnamon and Cummings, 1992). Gustducin is an ␣-subunit of a G-protein closely related to the transducins that is expressed in taste tissue (McLaughlin et al., 1992). Studies with gustducin knockout mice implicate gustducin in the detection of both sweet-and bitter-tasting compounds. In both behavioral and electrophysiological experiments, mice homozygous for the null gustducin allele were much less responsive than wild-type mice to bitter and sweet stimuli, but not to salts or acids (Wong et al., 1996).Taste buds are specialized epithelial structures containing gustatory receptor cells. Taste buds in the rat are grouped into several populations: in fungiform papillae on the anterior portion of the tongue, in vallate and foliate papillae on the posterior tongue, in the epithelia of the nasoincisor ducts (NIDs) and "geschmacksstreifen" (GS) of the palate, and on the laryngeal surface of the epiglottis (Miller et al., 1978;Travers and Nicklas, 1990;Smith et al., 1993). These populations differ in their gustatory sensitivities, as shown by electrophysiological studies of peripheral taste fibers. In rats, ...

Cells of mammalian taste buds have been classified into morphological types based on ultrastructural criteria, but investigators have disagreed as to whether these are distinct cell types or the extremes of a continuum. To address this issue, we examined taste buds from rat vallate papillae that had been sectioned transversely, rather than longitudinally, to their longest axis. In these transverse sections, dark (Type I) and light (Type II) cells were easily distinguished by their relative electron density, shape and topological relationships. Cells with electron-lucent cytoplasm (light cells) were circular or oval in outline, while those with electron-dense cytoplasm (dark cells) had an irregular outline with sheetlike cytoplasmic projections that separated adjacent light cells. A hierarchical cluster analysis of 314 cells across five morphological parameters (cell shape and area, and nuclear ellipticity, electron density and invagination) revealed two distinct groups of cells, which largely corresponded to the dark and light cells identified visually. These cells were not continuously distributed within a principal components factor solution. Differences in the means for dark and light cells were highly significant for each morphological parameter, but within either cell type, changes in one parameter correlated little with changes in any other. These analyses all failed to reveal cells with a consistent set of intermediate characteristics, suggesting that dark and light cells of rat vallate taste buds are distinct cell types rather than extremes of a continuum. Sections of taste buds were stained with antibodies to several carbohydrates, then observed by indirect immunofluorescence. Optical sections taken with a confocal laser-scanning microscope showed that the Lewis antigen was present only on spindle-shaped cells with circular or oval outlines and lacking transverse projections; these characteristic shapes matched those of light cells seen by electron microscopy. The H blood group antigen and the 2B8 epitope appeared at most cell-cell interfaces in the bud and are present on dark cells and possibly on some light cells. These findings relate molecular markers to morphological phenotypes and should facilitate future studies of taste cell turnover, development and regeneration.

We attached paraformaldehyde-fixed human erythrocyte ghosts to coated coverslips and sheared them to expose the cytoskeleton. Quick-freeze, deep-etch, rotary-replication, or tannic acid/osmium fixation and plastic embedding revealed the cytoskeleton as a dense network of intersecting straight filaments. Previous negative stain studies on spread skeletons found 5-6 spectrin tetramers intersecting at each actin oligomer, with an estimated 250 such intersections/microns 2 of membrane. In contrast, we found 3-4 filaments at each intersection and approximately 400 intersections/microns 2 of membrane. Immunogold labeling verified that the filaments were spectrin, but their lengths (29-37 nm) were approximately one-third that of extended spectrin dimers. The length and diameter of the filaments were sufficient to accommodate spectrin dimers, but not spectrin tetramers. Our results suggest that, in situ, spectrin dimers may associate as hexamers and octamers, rather than tetramers. We present several explanations that can reconcile our observations on intact cytoskeletons with previous reports on spread material. Extracting sheared ghosts with solutions of low ionic strength removed the cytoskeleton to reveal projections from the cytoplasmic surface of the membrane. These projections contained band 3, as shown by immunogold labeling, and they aggregated to a similar extent as intramembrane particles (IMP) when the cytoskeleton was removed, suggesting a direct relationship between these structures. Quantification indicated a stoichiometry of 2 IMP for each cytoplasmic projection. Cytoplasmic projections presumably contain other proteins besides band 3 since further treatment with high ionic strength solutions extracts peripheral proteins and reduces the diameter of projections by approximately 3 nm.

The clustering of acetylcholine receptors (AChR) in the postsynaptic membrane of newly innervated muscle fibers is one of the earliest events in the development of the vertebrate neuromuscular junction. Here, we describe two hypotheses that can account for AChR clustering in response to innervation. The "trophic factor" hypothesis proposes that the neuron releases a soluble factor that interacts with the muscle cell in a specific manner and that this interaction results in the local accumulation of AChR. The "contact and adhesion" hypothesis proposes that the binding of the nerve to the muscle cell surface is itself sufficient to induce AChR clustering, without the participation of soluble factors. We present a model for the molecular assembly of AChR clusters based on the contact and adhesion hypothesis. The model involves the sequential assembly of three distinct membrane domains. The first domain to form serves to attach microfilaments to the cytoplasmic surface of the muscle cell membrane at sites of muscle-nerve adhesion. The second domain to form is clathrin-coated membrane; it serves as a site of insertion of additional membrane elements, including AChR. Upon insertion of AChR into the cell surface, a membrane skeleton assembles by anchoring itself to the AChR. The skeleton, composed in part of actin and spectrin, binds and immobilizes significant numbers of AChR, thereby forming the third membrane domain of the AChR cluster. We make several predictions that should distinguish this model of AChR clustering from one that invokes soluble, trophic factors.

To evaluate the function and substrate specificity of human breast cancer resistance protein (BCRP, ABCG2) in the absence of cofactors or heterologous partner proteins, Xenopus laevis oocytes were injected with cRNA of wild-type or mutant (R482T) BCRP. High expression of BCRP was observed on the oocyte surface. Accumulation and efflux assays revealed that oocytes expressing R482T transported daunorubicin (DNR), mitoxantrone (MX), rhodamine 123, and flavopiridol (FLV), whereas wild-type BCRP transported only MX and FLV, in agreement with observations in mammalian and other systems. Transport activity was completely inhibited by fumitremorgin C, a known inhibitor of BCRP. Injection of oocytes with cRNA containing mutations of serine 187 in the ATP-binding cassette signature motif (S187T or S187A) resulted in strong expression of the mutant forms; however, these oocytes were devoid of transporter activity. When oocytes were coinjected with R482T and R482T/S187T, DNR transport was inhibited in a manner dependent on the amount of R482T/S187T cRNA added, consistent with the idea that the active form of BCRP is a homodimer or homomultimer. Substrate interaction studies found that no two substrates reciprocally inhibited the efflux of the other. Although FLV proved to be an effective inhibitor of both MX and DNR transport, and MX inhibited DNR transport, the other substrates tested had only weak or no inhibitory activity, indicating a complex nature of substrate interaction with the BCRP homodimer. We conclude that the X. laevis oocyte heterologous expression system is a valid and effective means of studying BCRP function and substrate specificity.