Effect of hyperthermia on the plasma membrane structure of Chinese hamster V79 fibroblasts: a quantitative freeze-fracture study

Hyperthermia in the range 41-45 degrees C can induce wide biochemical, physiological, and morphological changes in mammalian cells both in vivo and in vitro. In general, its effects on membranes, particularly on the plasma membrane, are still poorly understood. To investigate the effects of heat on this cell structure, Chinese hamster V79 fibroblasts were exposed to 43 degrees C hyperthermia for 1 h, immediately fixed with glutaraldehyde after treatment, and freeze-fractured for electron microscopic examination. Particular attention was given to the density and size of intramembranous particles (IMPs) on both protoplasmic (PF) and external (EF) fracture faces of the plasma membrane. The quantitative study performed by an interactive image analyzer on the IMPs, generally reported as plasma membrane proteins, showed in heat-treated cells a statistically significant increase in their density and size on both fracture faces. The differences observed demonstrate that in our experimental conditions, hyperthermia in plasma membranes produces structural changes whose biological significance has to be clarified. Moreover, our findings seem to support recent data indicating an involvement of membrane proteins in the cell response to hyperthermia.     

The ultrastructure of the cell surface and plasma membrane of exponential and stationary phase cells of Schizosaccharomyces pombe,grown in different media

In order to study the ultrastructure of the cell surface and plasma membrane of Schizosaccharomyces pombe as a function of growth conditions we investigated exponential and stationary phase cells grown in rich and minimal medium.Electron microscopic preparation techniques based on rapid cryofixation (without cryoprotectants) were used. The intramembraneous aspects of the plasma membrane were described by freeze fracturing. For the first time the dynamic surface structures could be directly analyzed by freeze drying in the scanning electron microscope and in thin section of freeze substituted samples. This preparation techniques reveal hair-like structures on the surface of yeast cells. The hairs of cells grown in the rich medium are longer than those grown in the minimal medium. A mutant defective in the structure of a cell surface galactomannoprotein (acid phosphatase) reveals (under conditions of maximal acid phosphatase expression) a cell surface structure that differs from the wild type. It is likely that the hairs represent the peripheral galactomannan layer or part of it.On the membrane fracture faces the number, shape, distribution and state of aggregation of the intramembraneous particles are different between membranes of growing and non-growing cells and between cells grown under different physiological conditions. In the minimal medium corresponding periodical structures on the plasmic and exoplasmic fracture faces were observed, which clearly differ between exponential and stationary phase cells. The number, length and depth of plasma membrane invaginations increase as the cells go from the exponential phase to the stationary phase. Short and flattened invaginations are filled with thin periodic structures.     

Freeze-fracture study of the receptor membranes in the olfactory organ of Alburnus alburnus (Teleostei)

Summary Olfactory receptor molecules are assumed to be integral membrane proteins which may be visualized on fracture faces of the membrane as intramembrane particles (IMPs). In the present study, the plasma membrane of the receptor dendrites and ciliated epithelial cells in the teleost fish Alburnus alburnus were studied by freeze-fracture electron microscopy. The IMP diameters on the membrane P-faces of both receptor dendrites and ciliated epithelial cells ranged from 5 nm to 11 nm. The average IMP densities on membrane fracture faces of the ciliated and microvillous sensory dendrites were 3130±780 for the cilia, 2070±550 for the microvilli, 2390±1190 on the knob regions and 3050±1130/m on the lateral dendrite membranes. The IMP densities on the P fracture faces of the cilia and knob regions were compared with the densities found on the lateral membranes of each individual dendrite. The ratios ranged from 0.5 to 0.96 in the case of the cilia/lateral membrane and from 0.5 to 0.90 in that of the knob/lateral membrane, indicating that, in contrast to the average densities, it is the lateral membrane which has the higher IMP densities and not the cilia. The great variations in the average IMP densities, as well as the considerable variety of the ratios, may be explained by the maturation and turnover of the olfactory sensory neurons.     

Ultrastructure of the Squid Axon Membrane as Revealed by Freeze-Fracture Electron Microscopy

The structure of the axolemma of the squid giant axon was studied by freeze-fracture electron microscopy. Three types of preparations were examined: intact axons, axons with their Schwann cell sheaths stripped off prior to freezing, and axons with their Schwann cell sheaths chemically detached but not mechanically removed. Because of a problem of cross-fracturing, the first two types of preparations revealed very few membrane faces of the axolemma. This cross-fracturing problem, however, was eliminated when we used a complementary replication method to fracture the third type of preparation. We found that the E-face of the axon membrane was smooth relative to the P-face, which showed many prominent intramembrane particles (IMP). The diameters of the typical IMP range from 6 to 15 nm. The P-face of the adjacent Schwann cells also showed many large IMP. The sizes and heights of the Schwann-cell IMP, however, appear to be more homogeneous than the P-face axolemma.     

Organization of the cell membrane inEuglena

Summary The cell membrane ofEuglena gracilis has been investigated with the freeze-fracture technique. When split, this membrane produces two fracture faces which are striking in their non-complementarity. The P fracture face is covered with a high density of 110 Å (average diameter) particles, while the E face is made up of a complex series of striations occurring at regular angles to the pellicle ridges which encircle the organism. Under certain conditions, however, the structure of the P fracture face assumes a more ordered configuration, and striations are visible on this fracture face which are precisely complementary to those observed on the E face. These observations suggest that the cortical cell membrane ofEuglena may be organized along the lines of a two dimensional crystal. However, this pattern of organization is restricted to the cortical region of the cell membrane; as the membrane invaginates near the anterior end of the cell the fracture faces change abruptly, and organization more typical of other cell membranes is observed. This invagination forms an extensive reservoir in the anterior of the cell, and the membrane bounding it is distinctly fluid in structure, with clear examples of endo- and exocytosis observable. These differences suggest that the cell membrane inEuglena is divided into two distinct but contiguous regions, each specialized with regard to structure and function.     

Heliobacterium chlorum: cell organization and structure

The basic cellular organization of Heliobacterium chlorum is described using the freeze-etching technique. Internal cell membranes have not been observed in most cells, leading to the conclusion that the photosynthetic apparatus of these organisms must be localized in the cell membrane of the bacterium. The two fracture faces of the cell membrane are markedly different. The cytoplasmic (PF) face is covered with densely packed particles averaging 8 nm in diameter, while the exoplasmic (EF) face contains far fewer particles, averaging approximately 10 nm in diameter. Although a few differentiated regions were noted within these fracture faces, the overall appearance of the cell membrane was remarkably uniform. The Heliobacterium chlorum cell wall is a strikingly regular structure, composed of repeating subunits arranged in a rectangular pattern at a spacing of 11 nm in either direction. We have isolated cell wall fragments by brief sonication in distilled water, and visualized the cell wall structure by negative staining as well as deep-etching.Abbreviations PF protoplasmic fracture face - EF exoplasmic fracture face     

Freeze-fracturing and surface labelling of embryonic neural retina cells

Freeze-fracturing of dissociated and aggregating neural retina cells from 7-day chick embryos revealed on the inner faces (PF) of the cell membrane numerous particles 6–20 nm in size. In contrast, the PF faces of blebs and some of the lobopodia that project from the cell surface were practically devoid of such particles. However, the elongated filopodia that abound on these cells showed numerous particles on their PF faces. These regional differences in the distribution of particles on PF faces of these cells are interpreted as reflecting membrane activity that leads to the formation of blebs and lobopodia. The frequent presence of “pits” at the basis of blebs and lobopodia is described. It is suggested that the “pits” are associated with the formation of these membrane projections; they may represent anchoring sites for microfilaments and for microtubules involved in the dynamic structure of the cell surface. ConA-binding sites on these cells were studied by scanning electron microscopy, using labeling with hemocyanin. The distribution of these sites on different regions of the cell surface coincided with the regional differences in the distribution of the inner membrane particles.     

Site-specific membrane particle arrays in magnesium-depleted Escherichia coli

The ultrastructure and polypeptide composition of a novel membrane junction in magnesium-starved Escherichia coli are described in this report. Freeze-fracture replicas reveal the junction as a site-specific membrane particle array with four fracture faces. Each junction consists of a cell membrane, a midline zone and a coupled membrane. Membrane particles associated with the junction extend from the hydrophobic region of the cell membrane across the hydrophilic midline zone and into the hydrophobic region of the coupled membrane. After negative staining or after rotary shadowing of freeze-fractured specimens, these particles were seen to consist of two similar but slightly offset bracket-shaped subunits separated by a small space. Optical analysis confirms this structure. Since the apposing membranes are bracketed or linked by their component particles, the name "bracket junction" is proposed for the complex. Methods are described for isolating a membrane fraction enriched in these junctional complexes; the fraction contains a prominent glycoprotein (mol wt 90,000) as well as a number of other components. The bracket junction is compared with the vertebrate gap junction in terms of both structure and possible roles in facilitating the permeation of the cell by small molecules.     

Ultrastructure of the Bacteroides nodosus cell envelope layers and surface.

The surface structure and cell envelope layers of various virulent Bacteroides nodosus strains were examined by light microscopy and by electron microscopy by using negative staining, thin-section, and freeze-fracture-etch techniques. Three surface structures were described: pili and a diffuse material, both of which emerged from one or both poles of the bacteria (depending on the stage of growth and division), and large rodlike structures (usually 30 to 40 nm in diameter) associated with a small proportion of the bacterial population. No capsule was detected. The cell envelope consisted of four layers: a plasma membrane, a peptidoglycan layer, an outer membrane, and an outermost additional layer. The additional layer was composed of subunits, generally hexagonally packed with center-to-center spacing of 6 to 7 nm. The outer membrane and plasma membrane freeze-fractured through their hydrophobic regions revealing four fracture faces with features similar to those of other gram-negative bacteria. However, some unusual features were seen on the fracture faces of the outer membrane: large raised ring structure (11 to 12 nm in diameter) on cw 3 at the poles of the bacteria; complementary pits or ring-shaped depressions on cw 2; and small raised ring structures (7 to 8 nm in diameter) all over cw 2.     

THE INTRAMEMBRANOUS PARTICLE PROFILE OF THE PARAMYLON MEMBRANE DURING PARAMYLON SYNTHESIS IN EUGLENA (EUGLENOPHYCEAE)1

Paramylon is the β-1, 3-glucan storage product in euglenoid algae. It is a fibrous crystal that occurs as membrane-bound granules in the cytosol. The role of the surrounding membrane in paramylon synthesis was investigated by the use of freeze-etch electron microscopy. When Euglena gracilis Klebs strain Z (Pringsheim) cells were frozen in supercooled liquid nitrogen, the fracture plane primarily was throuh the paramylon membrane. A large intramembranous particle (IMP, mean diam range 5.6-6.5 nm) and a small IMP (mean diam range 9.6-10.3 nm) were predominant in both PF (protoplasmic fracture) and EF (exoplasmic fracture) faces of the paramylon membrane. During paramylon synthesis induction, the ratio of small to large IMPs increased in both fracture faces. The IMP density decreased in both fracture faces concomitant to paramylon synthesis increase. These changes in IMP profile and density suggest that the paramylon membrane is involved in the synthesis of paramylon.     

Restriction of glycolipids to the outer half of a plasma membrane: Concanavalin A labeling of membrane halves in acanthamoeba castellanii

Fracture-label, a method that permits the cytochemical characterization of faces produced by freezefracture, was used to determine the partition and distribution of a glycolipid on membrane fracture faces of Acanthamoeba castellanii cells. After treatment with concanavalin A (Con A), the glycolipid (a lipophosphonoglycan, LPG) was labeled with colloidal gold coated with horseradish peroxidase. The label was abundant over exoplasmic fracture faces (face E) of plasma membranes, but absent from protoplasmic fracture faces (face P). We conclude that, in A. castellanii, glycolipid molecules are restricted to the outer half of the plasma membrane. This conclusion is confirmed by experiments with cells disrupted by freezing and thawing, where access of label to the cell interior did not result in labeling of the inner surface. Our results establish the exclusive localization of a glycolipid to the outer half of a plasma membrane. Fracture-label is proposed as a new technique to investigate the distribution and partition of glycolipids in plasma and intracellular membrane halves.     

Analysis of the structure of intramembrane particles of the mammalian urinary bladder

The luminal and discoid vacuole membranes of the superficial cell layer of the transitional epithelium of the mammalian urinary bladder have been studied by thin-sectioning and freeze-fracture-etch (FFE) electron microscope methods. For the FFE studies membranes were deposited on a cationized glass surface, covered by a thin copper disc, and fractured under liquid N2. Specimens were etched at -100 degrees C and replicated at -190 degrees C. A model of the lattice membrane derived from thin sections was used to predict the heights of the fracture faces above the glass surface. A hexagonal pattern of globular intramembrane particles spaced 160 A apart was seen in the external fracture (EF) face plaques as previously described and regarded as the dominant structure. However, very extensive areas of another pattern, seen before in only limited areas, have beeen found in the EF faces. The pattern consists of a smooth hexagonal lattice with the same space constant as the globular one but a different structure. By image analysis it consists of overlapping domains bordered by shared but incomplete metal rims. Each domain has a central spot of metal encircled by a shadow. The surface of the smooth lattice is partly complementary to the corresponding protoplasmic fracture (PF) face which shows a similar hexagonal lattice with the same space constant. The height of the smooth EF lattice above the glass substrate is the same as the plane of the center of the lipid bilayer predicted by the model. The mean heights of the particles of the globular EF lattice are greater than the total thickness of the membrane as predicted by the model and confirmed by measurements. The globular EF lattice is not complementary and it is concluded that the globular particles do not exist in the native membrane but arise artifactually during the preparatory procedures.     

Evidence for a vertical displacement of intramembranous particles on the plasma membrane of Tetrahymena cells by a local anesthetic

We examined the effect of a local anesthetic, dibucaine, on the plasma membrane of Tetrahymena pyriformis strain NT-1 using freeze-fracture electron microscopy. Intramembranous particles (IMPs) were distributed homogeneously on the plasma membrane of untreated cells. But, when Tetrahymena cells had been treated with 1.3 mM dibucaine for 5 min at growth temperature, freeze-fracture micrographs of the plasma membrane showed marked alterations. Although IMPs showed an almost homogeneous distribution, their density was elevated markedly on the protoplasmic fracture (PF) face but greatly reduce on the exoplasmic fracture (EF) face. Areas around deciliated portions had a reverse IMP density distribution for the PF and EF faces. These results suggest that dibucaine induced vertical displacement of the IMPs in the plasma membrane.     

Leishmania mexicana: distribution of intramembranous particles and filipin sterol complexes in amastigotes and promastigotes

The density and distribution of intramembranous particles was analyzed in freeze fracture replicas of the plasma membrane of amastigotes, and infective as well as noninfective promastigotes of Leishmania mexicana amazonensis. The density of intramembranous particles on both protoplasmic and extracellular faces was higher in infective than in noninfective promastigotes and it was lower in amastigotes than in promastigotes. Amastigotes purified immediately after tissue homogenization were surrounded by a membrane which corresponded to the membrane which lined the endocytic vacuoles where the parasites were located within the tissue macrophages. Aggregation of the particles was seen in the flagellar membrane at the point of emergence of the flagellum from the flagellar pocket. Differences in the organization of the particles were seen in the membrane which lined the flagellar pocket of amastigotes and promastigotes. The polyene antibiotic, filipin, was used as a probe for the detection of sterols in the plasma membrane of L. m. amazonensis. The effect of filipin in the parasite's structure was analyzed by scanning electron microscopy and by transmission electron microscopy of thin sections and freeze fracture replicas. Filipin sterol complexes were distributed throughout the membrane which lined the cell body, the flagellar pocket, and the flagellum. No filipin sterol complexes were seen in the cell body-flagellar adhesion zone. The density of filipin sterol complexes was lower in the membrane lining the flagellum than in that lining the cell body of promastigotes.     

Membrane differentiations at sites specialized for cell fusion

Fusion of plasma membranes between Chlamydomonas reinhardtii gametes has been studied by freeze-fracture electron microscopy of unfixed cells. The putative site of cell fusion developes during gametic differentiation and is recognized in thin sections of unmated gametes as a plaque of dense material subjacent to a sector of the anterior plasma membrane (Goodenough, U.W., and R.L. Weiss. 1975.J. Cell Biol. 67:623-637). The overlying membrane proves to be readily recognized in replicas of unmated gametes as a circular region roughly 500 nm in diameter which is relatively free of "regular" plasma membrane particles on both the P and E fracture faces. The morphology of this region is different for mating-type plus (mt+) and mt- gametes: the few particles present in the center of the mt+ region are distributed asymmetrically and restricted to the P face, while the few particles present in the center of the mt- region are distributed symmetrically in the E face. Each gamete type can be activated for cell fusion by presenting to it isolated flagella of opposite mt. The activated mt+ gamete generates large expanses of particle-cleared membrane as it forms a long fertilization tubule from the mating structure region. In the activated mt- gamete, the E face of the mating structure region is transformed into a central dome of densely clustered particles surrounded by a particle-cleared zone. When mt+ and mt- gametes are mixed together, flagellar agglutination triggeeeds to fuse with an activated mt- region. The fusion lip is seen to develop within the particle-dense central dome. We conclude that these mt- particles play an active role in membrane fusion.     

Fusion of liposomes with the plasma membrane of epithelial cells: fate of incorporated lipids as followed by freeze fracture and autoradiography of plastic sections

The fusion of liposomes with the plasma membrane of influenza virus- infected monolayers of an epithelial cell line, Madin-Darby canine kidney cells (van Meer et al., 1985. Biochemistry. 24:3593-3602), has been analyzed by morphological techniques. The distribution of liposomal lipids over the apical and basolateral plasma membrane domains after fusion was assessed by autoradiography of liposomal [3H]dipalmitoylphosphatidylcholine after rapid freezing or chemical fixation and further processing by freeze substitution and low temperature embedding. Before fusion, radioactivity was solely detected on the apical cell surface, indicating the absence of redistribution artifacts and demonstrating the reliability of lipid autoradiography on both a light and electron microscopical level. After induction of fusion by a low pH treatment, the basolateral plasma membrane domain became progressively labeled, indicative of rapid lateral diffusion of [3H]dipalmitoylphosphatidylcholine in the plasma membrane. Analysis of individual fusion events by freeze fracture after rapid freezing confirmed the rapid diffusion of the liposomal lipids into the plasma membrane, as intramembrane particle-free lipid patches were never observed. After the induction of liposome-cell fusion, well-defined intramembrane particles were present on the otherwise smooth liposomal fracture faces and on the fracture faces of the plasma membrane. Morphological evidence thus was obtained in favor of a local point fusion mechanism with an intramembrane particle as a specific structural fusion intermediate.     

Membrane fusion without cytoplasmic fusion (hemi-fusion) in erythrocytes that are subjected to electrical breakdown

There are many reports of hemi-fusion in phospholipid vesicles but few published studies on hemi-fusion in cells. We report evidence from both fluorescence microscopy and freeze-fracture electron microscopy for hemi-fusion in the electrofusion of human erythrocytes. We have also characterised the conditions that favour hemi-fusion as opposed to complete fusion, and discuss the possibility that hemi-fusion might precede complete electrically-induced cell fusion. A membrane probe (DiIC16) and a cytoplasmic probe (6-carboxyfluorescein) were used to investigate the behaviour of doubly-labelled human erythrocytes which were aligned in chains by dielectrophoresis and then exposed to high voltage breakdown pulses. Some of the cells were fused by the pulses, as shown by diffusion of both membrane and cytoplasmic probes from labelled to unlabelled cells. With other cells, the membrane probe diffused into unlabelled cells after the breakdown pulses, without the cytoplasmic probe diffusing into unlabelled cells or leaking into the medium. Membrane fusion (hemi-fusion) thus occurred without cytoplasmic fusion in these erythrocytes. Such cells were irreversibly, but fragilely, attached to their neighbours by the breakdown pulses. There was an inverse relationship between conditions that permit complete fusion and those that favour hemi-fusion, with respect to breakdown pulse length, breakdown voltage and, in particular, osmolarity and temperature. The incidence of hemi-fusion in 250 mM erythritol was twice that in 150 mM erythritol, and hemi-fusion was 5-fold greater at 25 degrees C than at 20 degrees C. Hemi-fused erythrocytes occasionally fused completely on heating to 50 degrees C, demonstrating that hemi-fusion can proceed to complete cell fusion. Freeze fracture electron micrographs of preparations of hemi-fused cells revealed long-lived, complementary depressions and protrusions on the E- and P-fracture faces, respectively, of tightly apposed cells that may mediate hemi-fusion. The possibility that the fusion of closely adjacent human erythrocytes by electrical breakdown pulses may involve an intermediate, shared bilayer structure, which is stable in certain conditions but which can be ruptured by osmotic swelling of the permeabilised cells, is discussed.     

Label-fracture: a method for high resolution labeling of cell surfaces

We introduce here a technique, "label-fracture," that allows the observation of the distribution of a cytochemical label on a cell surface. Cell surfaces labeled with an electron-dense marker (colloidal gold) are freeze-fractured and the fracture faces are replicated by plantinum/carbon evaporation. The exoplasmic halves of the membrane, apparently stabilized by the deposition of the Pt/C replica, are washed in distilled water. The new method reveals the surface distribution of the label coincident with the Pt/C replica of the exoplasmic fracture face. Initial applications indicate high resolution (less than or equal to 15 nm) and exceedingly low background. "Label-fracture" provides extensive views of the distribution of the label on membrane surfaces while preserving cell shape and relating to the freeze-fracture morphology of exoplasmic fracture faces. The regionalization of wheat germ agglutinin receptors on the plasma membranes of boar sperm cells is illustrated. The method and the interpretation of its results are straightforward. Label-fracture is appropriate for routine use as a surface labeling technique.     

Effect of Ethylenediaminetetraacetate upon the Surface of Escherichia coli

The effect of ethylenediaminetetraacetate (EDTA) on the envelope of two strains of Escherichia coli (B and Cla) was studied with freeze-fracturing methods. Untreated cells showed the outer membrane's outer surface with a fine texture of randomly spaced depressions of about 4.5-nm diameter; small areas with symmetrical arrangements of structural surface elements were also observed. The outer membrane's fracture plane revealed a random distribution of particles on its "concave" plane, only occasionally interrupted by particle-free areas. The "convex" aspect of the outer membrane's fracture plane showed only a few scattered particles. The cleavage plane of the inner membrane was often interrupted by many localized elevated plateaus, at which the cleaving process had, for short distances, switched to the outer membrane. The effects of EDTA treatment were mainly seen in the structure of the freeze-etched outer membrane: (i) the pits as well as the symmetrical surface elements of the outer membrane's outer surface had disappeared; (ii) a number of plateaus (about 20 to 50/cell) were seen at which a cleavage plane within the inner membrane had switched to the hydrophobic portion of the outer membrane (outer membrane's fracture plane). These plateaus were also visible in untreated cells; however, EDTA treatment apparently caused an increased exposure of plateaus. Surface areas, exposed by freeze-etching, revealed the underlying plateaus as elevations in the surface contour of the cell, suggesting a slower etching rate in the zones of the plateaus relative to the rest of the outer membrane. Well-defined, particle-free patches in the outer membrane's fracture plane, concave, were more frequent and larger in size after EDTA treatment than in the controls. In the presence of glycerol, the cells often cleaved in the outer membrane's fracture plane, but isolated plateaus were rarely observed. After metabolic poisoning of cells for 15 to 25 min at 37 degrees C, the plateaus had widened. These data suggest that the material of the plateaus has a slow rate of lateral diffusion. Placement of EDTA-treated cells in fresh medium at 37 degrees C caused, after 3 to 5 min, the reoccurrence of the pitted surface structure. We propose that the plateaus represent localized zones, at which newly synthesized lipopolysaccharide has been inserted.     

Vectorial and nonvectorial transphosphorylation catalyzed by enzymes II of the bacterial phosphotransferase system

The plasmolytic response of Bacillus licheniformis 749/C cells to the increasing osmolarity of the surrounding medium was quantitated with stereological techniques. Plasmolysis was defined as the area (in square micrometers) of the inside surface of the bacterial wall not in association with bacterial membrane per unit volume (in cubic micrometers) of bacteria. This plasmolyzed surface area was zero when the cells were suspended in a concentration of sucrose solution lower than 0.5 M, but increased linearly when the sucrose molarity rose above 0.5 M, reaching a plateau value of 3.61 micrometers2/micrometers3 in 2 M sucrose. In contrast, when the bacterial cells were treated with lysozyme plasmolysis increased abruptly from 0.06 micrometers2/micrometers3 in 0.75 M sucrose to 4.09 micrometers2/micrometers3 in 1 M sucrose. When the time of exposure was prolonged, the degree of plasmolysis increased gradually for the duration of the experiment (30 min) after exposure to 1 M sucrose without lysozyme, whereas with lysozyme plasmolysis reached a maximum (4.09 micrograms2/micrometers3) in 2 to 5 min. The examination of ultrastructure showed that the protoplast bodies of lysozyme-treated cells in 1 M sucrose and untreated cells in 2 M sucrose are maximally retracted from the intact wall of the bacteria; hardly any retraction of protoplasts could be seen for untreated cells in 1 M sucrose. The data suggest that the B. licheniformis cells are isoosmotic to 800 to 1,100 mosM solutions, but are able to withstand much greater osmotic pressure with no signs of plasmolysis because the cell wall and the plasma membrane are held in close association, perhaps by a covalent bond. It is likely that lysozyme weakens this bond by degradation of the peptidoglycan layer. Cellular autolysis also weakens this wall-membrane association.