Hyaluronan enters keratinocytes by a novel endocytic route for catabolism.

Hyaluronan synthesized in the epidermis has an exceptionally short half-life, indicative of its catabolism by epidermal keratinocytes. An intracellular pool of endogenously synthesized hyaluronan, from 1 to 20 fg/cell, inversely related to cell density, was observed in cultured rat epidermal keratinocytes. More than 80% of the intracellular hyaluronan was small (<90 kDa). Approximately 25% of newly synthesized hyaluronan was endocytosed by the keratinocytes and had a half-life of 2-3 h. A biotinylated aggrecan G(1) domain/link protein probe demonstrated hyaluronan in small vesicles of approximately 100 nm diameter close to the plasma membrane, and in large vesicles and multivesicular bodies up to 1300 nm diameter around the nucleus. Hyaluronan did not co-localize with markers of lysosomes. However, inhibition of lysosomal acidification with NH(4)Cl or chloroquine, or treating the cells with the hyaluronidase inhibitor apigenin increased intracellular hyaluronan staining, suggesting that it resided in prelysosomal endosomes. Competitive displacement of hyaluronan from surface receptors using hyaluronan decasaccharides, resulted in a rapid disappearance of this endosomal hyaluronan (t(12) approximately 5 min), indicating its transitory nature. The ultrastructure of the hyaluronan-containing vesicles, co-localization with marker proteins for different vesicle types, and application of specific uptake inhibitors demonstrated that the formation of hyaluronan-containing vesicles did not involve clathrin-coated pits or caveolae. Treatment of rat epidermal keratinocytes with the OX50 monoclonal antibody against the hyaluronan receptor CD44 increased endosomal hyaluronan. However, no CD44-hyaluronan co-localization was observed intracellularly unless endosomal trafficking was retarded by monensin, or cultivation at 20 degrees C, suggesting CD44 recycling. Rat epidermal keratinocytes thus internalize a large proportion of their newly synthesized hyaluronan into non-clathrin-coated endosomes in a receptor mediated way, and rapidly transport it to slower degradation in the endosomal/lysosomal system.     

Adhesion-induced receptor segregation and adhesion plaque formation: A model membrane study.

A model system to study the control of cell adhesion by receptor-mediated specific forces, universal interactions, and membrane elasticity is established. The plasma membrane is mimicked by reconstitution of homophilic receptor proteins into solid supported membranes and, together with lipopolymers, into giant vesicles with the polymers forming an artificial glycocalix. The homophilic cell adhesion molecule contact site A, a lipid-anchored glycoprotein from cells of the slime mold Dictyostelium discoideum, is used as receptor. The success of the reconstitution, the structure and the dynamics of the model membranes are studied by various techniques including film balance techniques, micro fluorescence, fluorescence recovery after photobleaching, electron microscopy, and phase contrast microscopy. The interaction of the functionalized giant vesicles with the supported bilayer is studied by reflection interference contrast microscopy, and the adhesion strength is evaluated quantitatively by a recently developed technique. At low receptor concentrations adhesion-induced receptor segregation in the membranes leads to decomposition of the contact zone between membranes into domains of strong (receptor-mediated) adhesion and regions of weak adhesion while continuous zones of strong adhesion form at high receptor densities. The adhesion strengths (measured in terms of the spreading pressure S) of the various states of adhesion are obtained locally by analysis of the vesicle contour near the contact line in terms of elastic boundary conditions of adhesion: the balance of tensions and moments. The spreading pressure of the weak adhesion zones is S approximately 10(-9) J/m(2) and is determined by the interplay of gravitation and undulation forces whereas the spreading pressure of the tight adhesion domains is of the order S approximately 10(-6) J/m(2).     

Polymer-cushioned bilayers. II. An investigation of interaction forces and fusion using the surface forces apparatus.

We have created phospholipid bilayers supported on soft polymer "cushions" which act as deformable substrates (see accompanying paper, Wong, J. Y., J. Majewski, M. Seitz, C. K. Park, J. N. Israelachvili, and G. S. Smith. 1999. Biophys. J. 77:1445-1457). In contrast to "solid-supported" membranes, such "soft-supported" membranes can exhibit more natural (higher) fluidity. Our bilayer system was constructed by adsorption of small unilamellar dimyristoylphosphatidylcholine (DMPC) vesicles onto polyethylenimine (PEI)-supported Langmuir-Blodgett lipid monolayers on mica. We used the surface forces apparatus (SFA) to investigate the long-range forces, adhesion, and fusion of two DMPC bilayers both above and below their main transition temperature (T(m) approximately 24 degrees C). Above T(m), hemi-fusion activation pressures of apposing bilayers were considerably smaller than for solid-supported bilayers, e.g., directly supported on mica. After separation, the bilayers naturally re-formed after short healing times. Also, for the first time, complete fusion of two fluid (liquid crystalline) phospholipid bilayers was observed in the SFA. Below T(m) (gel state), very high pressures were needed for hemi-fusion and the healing process became very slow. The presence of the polymer cushion significantly alters the interaction potential, e.g., long-range forces as well as fusion pressures, when compared to solid-supported systems. These fluid model membranes should allow the future study of integral membrane proteins under more physiological conditions.     

Avidin-biotin interactions at vesicle surfaces: adsorption and binding, cross-bridge formation, and lateral interactions.

Densely packed domains of membrane proteins are important structures in cellular processes that involve ligand-receptor binding, receptor-mediated adhesion, and macromolecule aggregation. We have used the biotin-avidin interaction at lipid vesicle surfaces to mimic these processes, including the influence of a surface grafted polymer, polyethyleneglycol (PEG). Single vesicles were manipulated by micropipette in solutions of fluorescently labeled avidin to measure the rate and give an estimate of the amount of avidin binding to a biotinylated vesicle as a function of surface biotin concentration and surface-grafted PEG as PEG-lipid. The rate of avidin adsorption was found to be four times less with 2 mol% PEG750 than for the unmodified surface, and 10 mol% PEG completely inhibited binding of avidin to biotin for a 2-min incubation. Using two micropipettes, an avidin-coated vesicle was presented to a biotinylated vesicle. In this vesicle-vesicle adhesion test, the accumulation of avidin in the contact zone was observed, again by using fluorescent avidin. More importantly, by controlling the vesicle membrane tension, this adhesion test provided a direct measure of the spreading pressure of the biotin-avidin-biotin cross-bridges confined in the contact zone. Assuming ideality, this spreading pressure gives the concentration of avidin cross-bridges in the contact zone. The rate of cross-bridge accumulation was consistent with the diffusion of the lipid-linked "receptors" into the contact zone. Once adherent, the membranes failed in tension before they could be peeled apart. PEG750 did not influence the mechanical equilibrium because it was not compressed in the contact zone, but it did perform an important function by eliminating all nonspecific adhesion. This vesicle-vesicle adhesion experiment, with a lower tension limit of 0.01 dyn/cm, now provides a new and useful method with which to measure the spreading pressures and therefore colligative properties of a range of membrane-bound macromolecules.     

Kinetics of membrane adhesion mediated by ligand-receptor interaction studied with a biomimetic system

We report the first measurement of the kinetics of adhesion of a single giant vesicle controlled by the competition between membrane-substrate interaction mediated by ligand-receptor interaction, gravitation, and Helfrich repulsion. To model the cell-tissue interaction, we doped the vesicles with lipid-coupled polymers (mimicking the glycocalix) and the reconstituted ligands selectively recognized by alpha(IIb)beta(3) integrin-mediating specific attraction forces. The integrin was grafted on glass substrates to act as a target cell. The adhesion of the vesicle membrane to the integrin-covered surface starts with the spontaneous formation of a small (approximately 200 nm) domain of tight adhesion, which then gradually grows until the whole adhesion area is in the state of tight adhesion. The time of adhesion varies from few tens of seconds to about one hour depending on the ligand and lipopolymer concentration. At small ligand concentrations, we observed the displacement xi of the front of tight adhesion following the square root law xi approximately t(1/2), whereas, at high concentrations, we found a linear law xi approximately t. We show both experimentally and theoretically that the t(1/2)-regime is dominated by diffusion of ligands, and the xi approximately t-regime by the kinetics of ligands-receptors association.     

Effect of polylysine on transformations and permeability of negative vesicular membranes

Small (40-60 nm in diameter) and large (300-350 nm) negative vesicles were complexed with a cationic polypeptide, poly-L-lysine (PL). Laser microelectrophoresis experiments showed that in small vesicles rendered anionic with the addition of cardiolipin (CL(2-)), only the CL(2-) in the outer leaflet is involved in the complexation with PL. Calorimetric and other data demonstrate that the binding of PL to the membrane surface causes domains ("rafts") of CL(2-) to form in the outer leaflet, and it is these domains that electrostatically bind the polymer. The kinetics of transmembrane permeation of doxorubicin (Dox, a fluorescent anti-tumor drug) was monitored with and without PL binding to the outer surface of the vesicles. It was found that PL mediates the permeation of Dox into the vesicle interior. In the absence of PL, the Dox molecule (possessing an amino group of pK(a)=8.6) binds to the anionic vesicles in the protonated form and, consequently, suffers an impaired mobility through the membrane. On the other hand, when the PL covers the vesicle surface, Dox passes though the membrane with greater ease. The effects of salt and polyanion on the stability of PL-vesicle complexes and the PL-mediated Dox permeation are also discussed.     

Interaction between DMPC liposomes and HM-PNIPAM polymer

The interactions of hydrophobically-modified poly-(N-isopropylacrylamides) (HM-PNIPAM) and dimyristoylphosphatidylcholine (DMPC) vesicles were investigated by the effect of the polymer on the binding of a fluorescent dye, oxonol VI, to DMPC vesicles, and on its diffusion across the membrane. On mixing with the vesicles, the dye exhibits an increase in fluorescence, which occurs in a two-stage process. The process was monitored by stopped-flow fluorescence spectrophotometry. According to the dependence of the reciprocal relaxation time on vesicle concentration, the rapid stage seems to be due to the second-order binding of the dye to the lipid membrane, a process that is almost diffusion-controlled, whereas the slow process is attributed to movement of the dye within the membrane phase. The polymer did not significantly affect the rate constant of the binding step, but it slowed down slightly the dissociation process of the dye from the membrane. However, the polymer affected the second stage, causing an increase in the reciprocal of its relaxation time, which suggests that the polymer makes the vesicle membrane more fluid.     

Comparison, by freeze-fracture electron microscopy, of chromatophores, spheroplast-derived membrane vesicles, and whole cells of Rhodopseudomonas sphaeroides.

By using freeze-fracture electron microscopy, chromatophores and spheroplast-derived membrane vesicles from photosynthetically grown Rhodopseudomonas sphaeroides were compared with cytoplasmic membrane and intracellular vesicles of whole cells. In whole cells, the extracellular fracture faces of both cytoplasmic membrane and vesicles contained particles of 11-nm diameter at a density of about 5 particles per 10(4) nm2. The protoplasmic fracture faces contained particles of 11 to 12-nm diameter at a density of 14.6 particles per 10(4) nm2 on the cytoplasmic membrane and a density of 31.3 particles per 10(4) nm2 on the vesicle membranes. The spheroplast-derived membrane fraction consisted of large vesicles of irregular shape and varied size, often enclosing other vesicles. Sixty-six percent of the spheroplast-derived vesicles were oriented in the opposite way from the intracellular vesicle membranes of whole cells. Eighty percent of the total vesicle surface area that was exposed to the external medium (unenclosed vesicles) showed this opposite orientation. The chromatophore fractions contained spherical vesicles of uniform size approximately equal to the size of the vesicles in whole cells. The majority (79%) of the chromatophores purified on sucrose gradients were oriented in the same way as vesicles in whole cells, whereas after agarose filtration almost all (97%) were oriented in this way. Thus, on the basis of morphological criteria, most spheroplast-derived vesicles were oriented oppositely from most chromatophores.     

Measuring the elasticity of clathrin-coated vesicles via atomic force microscopy

Using a new scheme based on atomic force microscopy (AFM), we investigate mechanical properties of clathrin-coated vesicles (CCVs). CCVs are multicomponent protein and lipid complexes of approximately 100 nm diameter that are implicated in many essential cell-trafficking processes. Our AFM imaging resolves clathrin lattice polygons and provides height deformation in quantitative response to AFM-substrate compression force. We model CCVs as multilayered elastic spherical shells and, from AFM measurements, estimate their bending rigidity to be 285 +/- 30 k(B)T, i.e., approximately 20 times that of either the outer clathrin cage or inner vesicle membrane. Further analysis reveals a flexible coupling between the clathrin coat and the membrane, a structural property whose modulation may affect vesicle biogenesis and cellular function.     

Acrosome differentiation in the ascidians Clavelina lepadiformis and Ciona intestinalis

The spermatozoa of both Clavelina lepadiformis and Ciona intestinalis have architectural features characteristic of ascidian spermatozoa that have been previously described. They have an elongated head (6 microm and 3 microm long, respectively) and a single mitochondrion that is closely applied laterally to the nucleus; they lack a midpiece. The acrosome of Clavelina lepadiformis spermatozoa is a moderately electron-dense, pear-shaped flattened vesicle, approx. 300 nm x 200 nm x 40 nm in length, width, and height, respectively. The acrosome of Ciona intestinalis spermatozoa is a moderately electron-dense, round flattened vesicle with an electron-dense plate in its central region. It is approx. 200 nm x 200 nm x 50 nm in length, width, and height, respectively. During spermiogenesis in both ascidians, several proacrosomal vesicles (50-70 nm in diameter) appear in a blister at the future apex of the spermatids. These vesicles appear to be associated with the inner surface of the plasma membrane enclosing the blister. They come into contact with each other along the inner surface of the plasma membrane and fuse to form a horseshoe-shaped acrosomal vesicle, which becomes a round, flattened vesicle during further differentiation. Some speculations about the mechanism of acrosome differentiation, the possible role of the acrosome during fertilization, and in the speciation of ascidians are presented.     

Adhesion-induced domain formation by interplay of long-range repulsion and short-range attraction force: a model membrane study.

We study the role of the interplay of specific and universal forces for the adhesion of giant vesicles on solid supported membranes. To model the situation of cell adhesion, we incorporated lipopolymers (phospholipids with polyethyleneoxide headgroups) as artificial glycocalix, whereas attractive lock-and-key forces are mimicked by incorporating biotinylated lipids into both membranes and by mediating the strong coupling through streptavidin. Adhesion is studied by quantitative reflection interference contrast microscopy (RICM), which enables visualization of the contact zone and reconstruction of the height profile of the membrane beyond the contact line (outside the contact zone) up to a height of 1 micron. We demonstrate that adhesion is accompanied by lateral phase separation, leading to the formation of domains of tight adhesion (adhesion plaques) separated by areas of weak adhesion exhibiting pronounced flickering. By analyzing the height profile S(x) near the contact line in terms of the tension equilibrium (Young equation) and the moment equilibrium, respectively, the adhesion energy and membrane tension can be approximately measured locally. We show that the adhesion energy is about three orders of magnitude larger for the adhesion plaques than for the weekly adhering regions. The adhesion is studied as a function of the excess area of the vesicle generated by temperature variation. A very remarkable finding is that increased excess area is not always stored in the contact area, but leads to the formation of microbuds (diameter approximately 2 microns).     

Membrane attach complex of complement (MAC): three-dimensional analysis of MAC-phospholipid vesicle recombinants

The three-dimensional structure of recombinants of the isolated membrane attack complex (MAC) of complement with single bilayer dioleoyllecithin (DOL) vesicles and with dimyristoyllecithin (DML) vesicles was determined. A total of four MAC-vesicle complexes were analyzed by imaging negatively stained specimens at various defined tilting angles under minimal dose conditions in the electron microscope and by computer-aided three-dimensional reconstruction. The information on electron micrographs obtained at 6 degrees angular increments from +60 degrees to -60 degrees was digitized by densitometric scanning, Fourier-transformed, corrected for imaging errors, cross-correlated, and synthesized to the three-dimensional image. All four MAC-vesicle recombinants showed stain penetration into the interior of the vesicle, indicating increased permeability of the bilayer to negative stain. The MAC appeared as a hollow structure of 16-nm height, 2.0-nm wall thickness, and a 3.0-nm torus at the free end with an outer and inner diameter of 20.0 nm and 10.0 nm. In MAC-DOL vesicles the hollow core of the MAC terminated at the membrane-binding site, and only small pores of up to 2.0-nm in diameter penetrated the bilayer. In one MAC-DML vesicle lipid discontinuities on the outer circumference of the MAC binding site mediated stain penetration. The second MAC-DML vesicle showed a channel of approximately 4.0 nm connecting the hollow core of the MAC across the bilayer with the vesicle interior. The results suggest the MAC may mediate increased membrane permeability by protein channel formation in addition to lipid reorientation.     

Location of the 100 kd-50 kd accessory proteins in clathrin coats.

We present a three-dimensional map of the clathrin coat of coated vesicles, generated from tilt series of electron micrographs of unstained specimens embedded in vitreous ice. We have examined native placental coated vesicles and coats reassembled from their purified constituents, namely clathrin triskelions and accessory proteins of approximate mol. wts 100 kd and 50 kd. Our results show that the accessory proteins contribute a further shell of density within the double shell of the clathrin cage, extending from the terminal domains of the clathrin to the membrane of the vesicle. The thickness of the complete coat is approximately 22 nm.     

Adhesion of lymphoid cells to CD44-specific substrata: the consequences of attachment depend on the ligand

Interactions between cell-surface adhesion receptors and immobilized specific substrata can exert profound effects on cell morphology. Using phase-contrast microscopy, we show that CD44-expressing mouse lymphoid cells display a spread morphology when adhering to CD44-specific monoclonal antibody (mAb) immobilized on plastic. This spread morphology is different from that of these same cells when adhering to immobilized hyaluronan, the natural ligand of CD44. Morphometric measurements, in combination with intracellular actin staining and fluorescence microscopy, revealed that the adhesion of lymphoid cells to hyaluronan required essentially no cytoskeletal reorganization and resulted in no fundamental change in morphology. On the other hand, cells adhering to immobilized CD44-specific mAb rearranged their actin structure and established multiple membrane contact sites (spread). Cell spreading on antibody, but not attachment to hyaluronan, was inhibited by cytoskeleton-disrupting agents. Transfection of CD44-negative lymphoid cells with full-length and tailless CD44 enabled these cells to bind to both immobilized hyaluronan and mAb. However, the transfectant lacking the cytoplasmic tail of CD44 spread only transiently on the antibody-coated surface. Our results suggest that CD44 may mediate lymphocyte attachment to its carbohydrate ligand hyaluronan by mechanisms broadly similar to those used by selectins. When immobilized CD44-specific antibody is the ligand, however, CD44 may regulate the activity of the cytoskeleton by mechanisms broadly similar to those used by integrins. In the latter case, the cytoplasmic domain of CD44 contributes to cell spreading.     

Adhesion of phospholipid vesicles to Chinese hamster fibroblasts: Role of cell surface proteins

The adhesion of artificially generated lipid membrane vesicles to Chinese hamster V79 fibroblasts in suspension was used as a model system for studying membrane interactions. Below their gel-liquid crystalline phase transition temperature, vesicles comprised of dipalmitoyl lecithin (DPL) or dimyristoyl lecithin (DML) absorbed to the surfaces of EDTA- dissociated cells. These adherent vesicles could not be removed by repeated washings of the treated cells but could be released into the medium by treatment with trypsin. EM autoradiographic studies of cells treated with[(3)H]DML or [(3)H]DPL vesicles showed that most of the radioactive lipids were confined to the cell periphery. Scanning electron microscopy and fluorescence microscopy further confirmed the presence of adherent vesicles at the cell surface. Adhesion of DML or DPL vesicles to EDTA-dissociated cells modified the lactoperoxidase-catalyzed iodination pattern of the cell surface proteins; the inhibition of labeling of two proteins with an approximately 60,000- dalton mol wt was particularly evident. Incubation of cells wit h (3)H-lipid vesicles followed by sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis showed that some of the (3)H-lipid migrated preferentially with these approximately 60,000-mol wt proteins. Studies of the temperature dependence of vesicle uptake and subsequent release by trypsin showed that DML or DPL vesicle adhesion to EDTA- dissociated cells increased with decreasing temperatures. In contrast, cells trypsinized before incubation with vesicles showed practically no temperature dependence of vesicle uptake. These results suggest two pathways for adhesion of lipid vesicles to the cell surface-a temperature-sensitive one involving cell surface proteins, and a temperature-independent one. These findings are discussed in terms of current models for cell-cell interactions.     

Recurrent dynamics of rupture transitions of giant lipid vesicles at solid surfaces

Single giant unilamellar vesicles (GUVs) rupture spontaneously from their salt-laden suspension onto solid surfaces. At hydrophobic surfaces, the GUVs rupture via a recurrent, bouncing ball rhythm. During each contact, the GUVs, rendered tense by the substrate interactions, porate, and spread a molecularly transformed motif of a monomolecular layer on the hydrophobic surface from the point of contact in a symmetric manner. The competition from pore closure, however, limits the spreading and produces a daughter vesicle, which re-engages with the substrate. At solid hydrophilic surfaces, by contrast, GUVs rupture via a distinctly different recurrent burst-heal dynamics; during burst, single pores nucleate at the contact boundary of the adhering vesicles, facilitating asymmetric spreading and producing a “heart”-shaped membrane patch. During the healing phase, the competing pore closure produces a daughter vesicle. In both cases, the pattern of burst-reseal events repeats multiple times, splashing and spreading the vesicular fragments as bilayer patches at the solid surface in a pulsatory manner. These remarkable recurrent dynamics arise, not because of the elastic properties of the solid surface, but because the competition between membrane spreading and pore healing, prompted by the surface-energy-dependent adhesion, determine the course of the topological transition.     

Electron cryomicroscopy of bacteriorhodopsin vesicles: mechanism of vesicle formation.

We obtained vesicles from purple membrane of Halobacterium halobium at different suspension compositions (pH, electrolytes, buffers), following the procedure of Kouyama et al. (1994) (J. Mol. Biol. 236:990-994). The vesicles contained bacteriorhodopsin (bR) and halolipid, and spontaneously formed during incubation of purple membrane suspension in the presence of detergent octylthioglucoside (OTG) if the protein:OTG ratio was 2:1 by weight. The size distribution of the vesicles was precisely determined by electron cryomicroscopy and was found to be almost independent on the incubation conditions (mean radius 17.9-19 nm). The size distribution in a given sample was close to the normal one, with a standard deviation of approximately +/- 1 nm. During dialysis for removal of the detergent, the vesicles diminished their radius by 2-2.5 nm. The results allow us to conclude that the driving force for the formation of bR vesicles is the preferential incorporation of OTG molecules in the cytoplasmic side of the membrane (with possible preferential delipidation of the extracellular side), which creates spontaneous curvature of the purple membrane. From the size distribution of the vesicles, we calculated the elasticity bending constant, K(B) approximately 9 x 10(-20) J, of the vesicle wall. The results provide some insight into the possible formation mechanisms of spherical assembles in living organisms. The conditions for vesicle formation and the mechanical properties of the vesicles could also be of interest with respect to the potential technological application of the bR vesicles as light energy converters.     

Polymer-cushioned lipid bilayers in porous alumina

Using small-angle neutron scattering (SANS) and cyclic voltammetry (CV), we show that model biological membranes can be deposited on a polymer cushion confined in highly regular porous alumina. The thicknesses of the dilute polymer cushion chemically bound to the alumina and of the supported bilayer are obtained for two polyethylene glycol cushions (PEG₅₀₀₀ and PEG₂₀₀₀₀) and for a cushion made of chains bearing a lipid anchor at their free end (DSPE-PEG₃₄₀₀). The bilayers are studied well below and well above the chain melting temperature of the lipid mixture (DMPC/DMPE: 80/20), using a coenzyme (Ubiquinone, UQ₁₀) as a redox probe for the voltammetry experiments. Analysis of the SANS form factor of the bilayers shows that the bilayer thickness can be extracted in this particular geometry. Using PEG chains grafted at a low surface density (D<2R_g), the thickness of the complete molecular construction is obtained by CV, which shows (after subtracting the bilayer thickness) that the polymer cushion thickness can be varied from 50 to 150 Å. The values obtained with three different chain lengths, are in perfect agreement with the radius derived from the Flory theory.     

In vitro biosynthesis, core glycosylation and membrane integration of opsin

A membrane-integrated , core-glycosylated form of bovine opsin was synthesized in vitro when bovine retina mRNA was translated in a wheat germ cell-free system supplemented with dog pancreas microsomal vesicles; glycosylation and integration of opsin into membranes were coupled to translation. Proteolysis with themolysin was used to probe the orientation of opsin within the dog pancreas microsomal membrane, and to compare it with that of opsin in rod cell disk membranes isolated from bovine retina. Intact microsomal or disk vesicles were required for production of discrete, membrane-associated thermolysin fragments of opsin; no discrete opsin fragments were detected when membranes were incubated with thermolysin in the presence of the nonionic detergent, Triton X-100. The major opsin fragments produced by themosylin treatment of intact microsomal vesicles resembled those from disk vesicles in their size, oligosaccharide content, and order of appearance. In each case, the first cleavage of opsin took place at the COOH-terminus, generating a glycosylated fragment, O’, which was only slightly smaller than intact opsin. Both the microsomal and disk membrane forms of O’ were next cleaved internally; glycosylated fragments of similar sizes in both cases were detected which were derived from the NH(2)-terminal portion of O’. Several smaller NH(2)-terminal fragments of opsin were detected only in thermolysin-treated microsomal membranes, and not in disk membranes. The data suggest that the topology of opsin integrated into dog pancreas microsomal vesicles is similar to that in rod cell disk vesicles, although not identical. In each case, the glycosylated NH(2)-terminal region of opsin is located within the lumen of the vesicle, while discrete COOH-terminal and internal segments of opsin apparently emerge at the outer, cytoplasmic face of the membrane. Thus, opsin in the heterologous microsomal membrane, like its counterpart in the native disk membrane, may cross the bilayer at least three times. The internal domain of the polypeptide that emerges at the outer membrane surface is apparently more highly exposed in the case of opsin in microsomal membranes, evidenced by the additional internal thermolysin cleavage sites detected.