Localization of peripherin/rds in the disk membranes of cone and rod photoreceptors: relationship to disk membrane morphogenesis and retinal degeneration

The outer segments of vertebrate rod photoreceptor cells consist of an ordered stack of membrane disks, which, except for a few nascent disks at the base of the outer segment, is surrounded by a separate plasma membrane. Previous studies indicate that the protein, peripherin or peripherin/rds, is localized along the rim of mature disks of rod outer segments. A mutation in the gene for this protein has been reported to be responsible for retinal degeneration in the rds mouse. In the present study, we have shown by immunogold labeling of rat and ground squirrel retinas that peripherin/rds is present in the disk rims of cone outer segments as well as rod outer segments. Additionally, in the basal regions of rod and cone outer segments, where disk morphogenesis occurs, we have found that the distribution of peripherin/rds is restricted to a region that is adjacent to the cilium. Extension of its distribution from the cilium coincides with the formation of the disk rim. These results support the model of disk membrane morphogenesis that predicts rim formation to be a second stage of growth, after the first stage in which the ciliary plasma membrane evaginates to form open nascent disks. The results also indicate how the proteins of the outer segment plasma membrane and the disk membranes are sorted into their separate domains: different sets of proteins may be incorporated into membrane outgrowths during different growth stages of disk morphogenesis. Finally, the presence of peripherin/rds protein in both cone and rod outer segment disks, together with the phenotype of the rds mouse, which is characterized by the failure of both rod and cone outer segment formation, suggest that the same rds gene is expressed in both types of photoreceptor cells.     

Freeze-fracture studies of photoreceptor membranes: new observations bearing upon the distribution of cholesterol

We performed electron microscopy of replicas from freeze-fractured retinas exposed during or after fixation to the cholesterol-binding antibiotic, filipin. We observed characteristic filipin-induced perturbations throughout the disk and plasma membranes of retinal rod outer segments of various species. It is evident that a prolonged exposure to filipin in fixative enhances rather than reduces presumptive cholesterol detection in the vertebrate photoreceptor cell. In agreement with the pattern seen in our previous study (Andrews, L.D., and A. I. Cohen, 1979, J. Cell Biol., 81:215-228), filipin-binding in membranes exhibiting particle-free patches seemed largely confined to these patches. Favorably fractured photoreceptors exhibited marked filipin-binding in apical inner segment plasma membrane topologically confluent with and proximate to the outer segment plasma membrane, which was comparatively free of filipin binding. A possible boundary between these differing membrane domains was suggested in a number of replicas exhibiting lower filipin binding to the apical plasma membrane of the inner segment in the area surrounding the cilium. This area contains a structure (Andrews, L. D., 1982, Freeze-fracture studies of vertebrate photoreceptors, In Structure of the Eye, J. G. Hollyfield and E. Acosta Vidrio, editors, Elsevier/North-Holland, New York, 11-23) that resembles the active zones of the nerve terminals for the frog neuromuscular junction. These observations lead us to hypothesize that these structures may function to direct vesicle fusion to occur near them, in a domain of membrane more closely resembling outer than inner segment plasma membrane. The above evidence supports the views that (a) all disk membranes contain cholesterol, but the particle-free patches present in some disks trap cholesterol from contiguous particulate membrane regions; (b) contiguous inner and outer segment membranes may greatly differ in cholesterol content; and (c) the suggested higher cholesterol in the inner segment than in the outer segment plasma membrane may help direct newly inserted photopigment molecules to the outer segment.     

Proteomic analysis of the retinal rod outer segment disks

The initial events of vision at low light take place in vertebrate retinal rods. The rod outer segment consists of a stack of flattened disks surrounded by the plasma membrane. A list of the proteins that reside in disks has not been achieved yet. We present the first comprehensive proteomic analysis of purified rod disks, obtained by combining the results of two-dimensional gel electrophoresis separation of disk proteins to MALDI-TOF or nLC-ESI-MS/MS mass spectrometry techniques. Intact disks were isolated from bovine retinal rod outer segments by a method that minimizes contamination from inner segment. Out of a total of 187 excised spots, 148 proteins were unambiguously identified. An additional set of 61 proteins (partially overlapping with the previous ones) was generated by one-dimensional (1D) gel nLC-ESI-MS/MS method. Proteins involved in vision as well as in aerobic metabolism were found, among which are the five complexes of oxidative phosphorylation. Results from biochemical, Western blot, and confocal laser scanning microscopy immunochemistry experiments suggest that F 1F o-ATP synthase is located and catalytically active in ROS disk membranes. This study represents a step toward a global physiological characterization of the disk proteome and provides information necessary for future studies on energy supply for phototransduction.     

Rhodopsin mutant P23H destabilizes rod photoreceptor disk membranes

Mutations in rhodopsin cause retinitis pigmentosa in humans and retinal degeneration in a multitude of other animals. We utilized high-resolution live imaging of the large rod photoreceptors from transgenic frogs (Xenopus) to compare the properties of fluorescently tagged rhodopsin, Rho-EGFP, and Rho(P23H)-EGFP. The mutant was abnormally distributed both in the inner and outer segments (OS), accumulating in the OS to a concentration of ～0.1% compared to endogenous opsin. Rho(P23H)-EGFP formed dense fluorescent foci, with concentrations of mutant protein up to ten times higher than other regions. Wild-type transgenic Rho-EGFP did not concentrate in OS foci when co-expressed in the same rod with Rho(P23H)-EGFP. Outer segment regions containing fluorescent foci were refractory to fluorescence recovery after photobleaching, while foci in the inner segment exhibited recovery kinetics similar to OS regions without foci and Rho-EGFP. The Rho(P23H)-EGFP foci were often in older, more distal OS disks. Electron micrographs of OS revealed abnormal disk membranes, with the regular disk bilayers broken into vesiculotubular structures. Furthermore, we observed similar OS disturbances in transgenic mice expressing Rho(P23H), suggesting such structures are a general consequence of mutant expression. Together these results show that mutant opsin disrupts OS disks, destabilizing the outer segment possibly via the formation of aggregates. This may render rods susceptible to mechanical injury or compromise OS function, contributing to photoreceptor loss.     

Sodium and Potassium Concentration Gradient in the Retinal Rod Outer Segment

THE Na⁺ and K⁺ concentration gradient across the cone disk membrane is the basis for the generation of the late receptor potential^1,2. The cone disks are invaginations of the outer plasmic membrane that separates the potassium-rich cytoplasm from the sodium-rich extracellular media. The permeability of a disk membrane is decreased when light is absorbed and this leads to hyperpolarization of the cell³. Rods are also capable of generating an electrical response to light. This response seems to be due to a decrease in permeability of the outer segment membrane⁴. The cone disk membranes are, none the less, the primary absorbers of the light quanta. The disks of rods are not continuous with the outer plasmic membrane and it is questionable whether there is any difference between interdisk and intradisk content of Na⁺ and K⁺-ions concentration.     

Some quantitative characteristics of the frog retinal rod outer segments]

Retinal rod outer segments in frogs were studied by means of light microscopy, refractometry, microspectrophotometry, and electron microscopy. Analysis of the data obtained shows that an unidentified substance, which makes up about 50% of outer segment dry weight, is lost during routine biochemical investigations. The protein parts of the rhodopsin molecules make up 85% of the outer segments proteins and 25% of outer segment dry weight. Rhodopsin molecules can be arranged in a square array with a unit cell side of about 7 nm on one side of each disk membrane. Lipids in a single membrane occupy only 2 nm, and disk membranes are strongly hydrated.     

Molecular mechanisms of photoreception. V. Isolation of cytoplasmic membranes of rod outer segments from the frog retina

Cytoplasmic membranes of rod outer segments from frog retina intact rods in retina were stained with fluorescent dye fluoresceinmonomercur acetate. The dye is covalently bound to proteins of cytoplasmic membrane and doesn't penetrate into the cells. Upon isolation of the purified outer segments with the labeled cytoplasmic membranes the cells were disrupted and fractionated in density sucrose gradient. Cytoplasmic membranes possess floating densities different from those of disk membranes and thus providing a mean for separating them from the latter. The main peptides of cytoplasmic membranes are 56, 53, 45, 30 and 28 kDa proteins.     

Evidence for melanogenesis in the retinal pigment epithelium of adult cattle and golden hamster.

1. The ultrastructure of the retinal pigment epithelium (RPE) of adult Syrian golden hamsters and cattle was examined with respect to pigment granules and phagosomes involved in degradation of disk membranes from rod outer segments. 2. In the RPE of cattle, phagosomes were found that contained an electron-dense melanin-like material that was not autofluorescent and therefore not lipofuscin. 3. Disk membranes of rods are about 4 nm thick and become enlarged (7-20 nm) and electron-dense during degradation in the RPE. 4. Additionally electron-dense vesiculo-globular bodies (10-100 nm) were found in phagosomes during disk membrane degradation and in mature melanin granules. 5. In the RPE of adult hamsters that had been exposed to intense light, premelanosomes containing unmelanised filaments with a striated periodicity were found in the cytoplasm or in association with mature melanin granules. Early and late stage melanosomes were also present. Phagosomes in the RPE contained degraded disk membranes, melanin-like material and melanofilaments. 6. Dopa oxidase was detected ultrastructurally within shed disk membranes that were in close contact with the microvilli of the RPE. 7. The possibility of melanogenesis within phagosomes during disk membrane degradation is discussed.     

Localization of Usher 1 proteins to the photoreceptor calyceal processes,which are absent from mice

The mechanisms underlying retinal dystrophy in Usher syndrome type I (USH1) remain unknown because mutant mice lacking any of the USH1 proteins—myosin VIIa, harmonin, cadherin-23, protocadherin-15, sans—do not display retinal degeneration. We found here that, in macaque photoreceptor cells, all USH1 proteins colocalized at membrane interfaces (i) between the inner and outer segments in rods and (ii) between the microvillus-like calyceal processes and the outer segment basolateral region in rods and cones. This pattern, conserved in humans and frogs, was mediated by the formation of an USH1 protein network, which was associated with the calyceal processes from the early embryonic stages of outer segment growth onwards. By contrast, mouse photoreceptors lacked calyceal processes and had no USH1 proteins at the inner–outer segment interface. We suggest that USH1 proteins form an adhesion belt around the basolateral region of the photoreceptor outer segment in humans, and that defects in this structure cause the retinal degeneration in USH1 patients.     

Differential membrane protein phosphorylation in bovine retinal rod outer segment disk membranes as a function of disk age

The outer segment portion of photoreceptor rod cells is composed of a stacked array of disk membranes. Newly formed disks are found at the base of the rod outer segment (ROS) and are relatively high in membrane cholesterol. Older disks are found at the apical tip of the ROS and are low in membrane cholesterol. Disk membranes were separated based on their membrane cholesterol content and the extent of membrane protein phosphorylation determined. Light induced phosphorylation of ROS disk membrane proteins was investigated using magic angle spinning³¹P NMR. When intact rod outer segment preparations were stimulated by light, in the presence of endogenously available kinases, membrane proteins located in disks at the base of the ROS were more heavily phosphorylated than those at the tip. SDS-gel electrophoresis of the phosphorylated disk membranes subpopulations identified a phosphoprotein species with a molecular weight of approximately 68–72 kDa that was more heavily phosphorylated in newly formed disks than in old disks. The identity of this phosphoprotein is presently under investigation. When the phosphorylation reaction was carried out in isolated disk membrane preparations with exogenously added co-factors and kinases, there was no preferential protein phosphorylation. Taken collectively, these results suggest that within the ROS there is a protein phosphorylation gradient that maybe indicative of co-factor or kinase heterogeneity.     

Rhodopsin lateral diffusion as a function of rod outer segment disk membrane axial position

Rhodopsin lateral diffusion was measured at three points along the axis of frog outer segments using the method of absorbance recovery after photobleaching. Mean recovery times were slightly longer in distal disk membranes than in proximal disks. A small reduction of pigment mobility with disk age may reflect subtle changes in membrane composition.     

Fusion between disk membranes and plasma membrane of bovine photoreceptor cells is calcium dependent.

Disk membranes and plasma membrane vesicles were prepared from bovine retinal rod outer segments (ROS). The plasma membrane vesicles were labeled with the fluorescent probe octadecylrhodamine B chloride (R18) to a level at which the R18 fluorescence was self-quenched. At pH 7.4 and 37 degrees C and in the presence of micromolar calcium, an increase in R18 fluorescence with time was observed when R18-labeled plasma membrane vesicles were introduced to a suspension of disks. This result was interpreted as fusion between the disk membranes and the plasma membranes, the fluorescence dequenching resulting from dilution of the R18 into the unlabeled membranes as a result of lipid mixing during membrane fusion. While the disk membranes exposed exclusively their cytoplasmic surface, plasma membrane vesicles were found with both possible orientations. These vesicles were fractionated into subpopulations with homogeneous orientation. Plasma membrane vesicles that were oriented with the cytoplasmic surface exposed were able to fuse with the disk membranes in a Ca(2+)-dependent manner. Fusion was not detected between disk membranes and plasma membrane vesicles oriented such that the cytoplasmic surface was on the interior of the vesicles. ROS plasma membrane-disk membrane fusion was stimulated by calcium, inhibited by EGTA, and unaffected by magnesium. Rod photoreceptor cells of vertebrate retinas undergo diurnal shedding of disk membranes containing the photopigment rhodopsin. Membrane fusion is required for the shedding process.     

Rhodopsin is spatially heterogeneously distributed in rod outer segment disk membranes

The visual photoreception takes place in the retina, where specialized rod and cone photoreceptor cells are located. The rod outer segments contain a stack of 500-2,000 sealed membrane disks. Rhodopsin is the visual pigment located in rod outer segment disks, it is a member of the G-protein-coupled receptor (GPCR) superfamily, an important group of membrane proteins responsible for the majority of physiological responses to stimuli such as light, hormones, peptides, etc. Alongside rhodopsin, peripherin/Rom proteins located in the disk rims are thought to be responsible for disk morphology. Here we describe the supramolecular structure of rod outer segment disk membranes and the spatial organization of rhodopsin and peripherin/Rom molecules. Using atomic force microscopy operated in physiological buffer solution, we found that rhodopsin is loosely packed in the central region of the disks, in average about 26?000 molecules covering approximately one third of the disk surface. Peripherin/Rom proteins form dense assemblies in the rim region. A protein-free lipid bilayer girdle separates the rhodopsin and peripherin/Rom domains. The described supramolecular assembly of rhodospin, peripherin/Rom and lipids in native rod outer segment disks is consistent with the functional requirements of photoreception.     

Disk formation in retinal cones of Tupaia belangeri (Scandentia)

Existing hypotheses on the mode of disk formation in the photoreceptor cells of mammals appear to be incompatible: (1) plasma membranes of adjacent evaginations form a disk which, subsequently, is internalized by a disk rim; (2) pinocytotic vesicles are pinched off from the plasma membrane and fuse into a larger vesicle, which flattens and forms a disk. We have studied the development of the cone outer segment and the disk formation in Tupaia belangeri by transmission electron microscopy. During the first two postnatal weeks, the distal part of the single cilium, which is inserted apically on the inner segment, becomes balloon-shaped. Apical to the axoneme, it contains tubular and vesicular material, which, most probably, has been detached from the axonemal microtubules. These tubules and vesicles do not contribute to disks. The balloon-shaped expansion, later retained as the ciliary backbone, establishes the contact with the pigment epithelium. Formation of disks, from the 12-day-old Tupaia onwards, occurs between adjacent evaginations at the outer segment base. The initial disk rims are “hooked” to the ciliary axonemal microtubules. The axonemal microtubules are involved in the initiation and in the alignment of the disks. Disk rim formation and, thus, internalization of disks proceeds from the base to the apex of the outer segment, that is, from the younger to the older disks. In the adult Tupaia, an uneven progression of disk rim formation on both sides of the axoneme is found among consecutive disks. The seemingly incompatible hypotheses on the mode of disk formation reflect a heterochrony of the internalization of membranes and of the disk formation among different mammals and, possibly, between cones and rods. Received: 24 July 1997 / Accepted: 10 September 1997     

Differences in the protein composition of bovine retinal rod outer segment disk and plasma membranes isolated by a ricin-gold-dextran density perturbation method

The plasma membrane and disk membranes of bovine retinal rod outer segments (ROS) have been purified by a novel density-gradient perturbation method for analysis of their protein compositions. Purified ROS were treated with neuraminidase to expose galactose residues on plasma membrane-specific glycoproteins and labeled with ricin-gold-dextran particles. After the ROS were lysed in hypotonic buffer, the plasma membrane was dissociated from the disks by either mild trypsin digestion or prolonged exposure to low ionic strength buffer. The dense ricin-gold-dextran-labeled plasma membrane was separated from disks by sucrose gradient centrifugation. Electron microscopy was used to follow this fractionation procedure. The dense red pellet primarily consisted of inverted plasma membrane vesicles containing gold particles; the membrane fraction of density 1.13 g/cc consisted of unlabeled intact disks and vesicles. Ricin-binding studies indicated that the plasma membrane from trypsin-treated ROS was purified between 10-15-fold. The protein composition of plasma membranes and disks was significantly different as analyzed by SDS gels and Western blots labeled with lectins and monoclonal antibodies. ROS plasma membrane exhibited three major proteins of 36 (rhodopsin), 38, and 52 kD, three ricin-binding glycoproteins of 230, 160, and 110 kD, and numerous minor proteins in the range of 14-270 kD. In disk membranes rhodopsin appeared as the only major protein. A 220-kD concanavalin A-binding glycoprotein and peripherin, a rim-specific protein, were also present along with minor proteins of 43 and 57-63 kD. Radioimmune assays indicated that the ROS plasma membrane contained about half as much rhodopsin as disk membranes.     

Differential Rhodopsin Regeneration in Photoreceptor Membranes is Correlated with Variations in Membrane Properties

Rhodopsin, the major transmembrane protein in both the plasma membrane and the disk membranes of photoreceptor rod outer segments (ROS) forms the apo-protein opsin upon the absorption of light. In vivo the regeneration of rhodopsin is necessary for subsequent receptor activation and for adaptation, in vitro this regeneration can be followed after the addition of 11-cis retinal. In this study we investigated the ability of bleached rhodopsin to regenerate in the compositionally different membrane environments found in photoreceptor rod cells. When 11-cis retinal was added to bleached ROS plasma membrane preparations, rhodopsin did not regenerate within the same time course or to the same extent as bleached rhodopsin in disk membranes. Over 80% of the rhodopsin in newly formed disks regenerated within 90 minutes while only 40% regenerated in older disks. Since disk membrane cholesterol content increases as disks are displaced from the base to the apical tip of the outer segment, we looked at the affect of membrane cholesterol content on the regeneration process. Enrichment or depletion of disk membrane cholesterol did not alter the % rhodopsin that regenerated. Bulk membrane properties measured with a sterol analog, cholestatrienol and a fatty acid analog, cis parinaric acid, showed a more ordered, less fluid, lipid environment within plasma membrane relative to the disks. Collectively these results show that the same membrane receptor, rhodopsin, functions differently as monitored by regeneration in the different lipid environments within photoreceptor rod cells. These differences may be due to the bulk properties of the various membranes.     

Photoreceptor number and outer segment disk membrane surface area in the retina of the rat: stereological data for whole organ and average photoreceptor cell

A random sampling scheme is employed to obtain stereological estimates of disk membrane surface area in the entire retina and in the average photoreceptor cell. The scheme involves the use of vertical sections with combined light and electron microscopy at several magnification levels. Left and right retinas from six albino animals were analysed. There were no significant lateral differences. On average, the retina had a volume of 16 mm3, thickness of 200 μm and surface area of 80 mm2 (representing about 56% of the external surface of the eyeball). Photoreceptor disk membranes within outer segments amplified total retinal surface by almost 1000-fold (final surface 770 cm2 per retina). The retina contained 3×107 photoreceptors (packing density 374 000 mm-2) with an average disk membrane surface area of 2600 μm2. Mean nuclear volume in photoreceptor cells was 59 μm3 and the coefficient of variation for the distribution of nuclear volumes was 57%. The data are consistent with an average of 700 disks per photoreceptor cell, a membrane area of 4 μm2 per disk and a convergence ratio of ～260 photoreceptors per optic nerve fibre. The basic scheme could be modified for other species and for direct cell counts conducted on rods and cones separately.     

The role of cholesterol in rod outer segment membranes

The photoreceptor rod outer segment (ROS) provides a unique system in which to investigate the role of cholesterol, an essential membrane constituent of most animal cells. The ROS is responsible for the initial events of vision at low light levels. It consists of a stack of disk membranes surrounded by the plasma membrane. Light capture occurs in the outer segment disk membranes that contain the photopigment, rhodopsin. These membranes originate from evaginations of the plasma membrane at the base of the outer segment. The new disks separate from the plasma membrane and progressively move up the length of the ROS over the course of several days. Thus the role of cholesterol can be evaluated in two distinct membranes. Furthermore, because the disk membranes vary in age it can also be investigated in a membrane as a function of the membrane age. The plasma membrane is enriched in cholesterol and in saturated fatty acids species relative to the disk membrane. The newly formed disk membranes have 6-fold more cholesterol than disks at the apical tip of the ROS. The partitioning of cholesterol out of disk membranes as they age and are apically displaced is consistent with the high PE content of disk membranes relative to the plasma membrane. The cholesterol composition of membranes has profound consequences on the major protein, rhodopsin. Biophysical studies in both model membranes and in native membranes have demonstrated that cholesterol can modulate the activity of rhodopsin by altering the membrane hydrocarbon environment. These studies suggest that mature disk membranes initiate the visual signal cascade more effectively than the newly synthesized, high cholesterol basal disks. Although rhodopsin is also the major protein of the plasma membrane, the high membrane cholesterol content inhibits rhodopsin participation in the visual transduction cascade. In addition to its effect on the hydrocarbon region, cholesterol may interact directly with rhodopsin. While high cholesterol inhibits rhodopsin activation, it also stabilizes the protein to denaturation. Therefore the disk membrane must perform a balancing act providing sufficient cholesterol to confer stability but without making the membrane too restrictive to receptor activation. Within a given disk membrane, it is likely that cholesterol exhibits an asymmetric distribution between the inner and outer bilayer leaflets. Furthermore, there is some evidence of cholesterol microdomains in the disk membranes. The availability of the disk protein, rom-1 may be sensitive to membrane cholesterol. The effects exerted by cholesterol on rhodopsin function have far-reaching implications for the study of G-protein coupled receptors as a whole. These studies show that the function of a membrane receptor can be modulated by modification of the lipid bilayer, particularly cholesterol. This provides a powerful means of fine-tuning the activity of a membrane protein without resorting to turnover of the protein or protein modification.     

Cholesterol heterogeneity in bovine rod outer segment disk membranes

Rod outer segment disk membranes have been used to study visual transduction events. Numerous studies have also focused on protein-lipid interactions in these membranes. The possible heterogeneity of the disk membrane composition has not been addressed in such studies. Freeze fracture studies (Andrews, L. D., and Cohn, A. I. (1979) J. Cell Biol. 81, 215-220; Caldwell, R., and McLaughlin, B. (1985) J. Comp. Neurol. 236, 523-537) suggest a difference in cholesterol content between newly formed and old disks. This potential heterogeneity in disk membrane composition was investigated using digitonin. Osmotically intact bovine rod outer segment disk membranes prepared by Ficoll flotation were separated based on the cholesterol content of the disks. The addition of digitonin to disk membrane suspensions in a one-to-one molar ratio with respect to cholesterol produced an increase in the density of the membranes in proportion to the amount of cholesterol present. The digitonin-treated disks were separated into subpopulations using a sucrose density gradient. Disks were shown to vary in cholesterol to phospholipid ratio from 0.30 to 0.05. The ratio of phospholipid to protein remained constant in all disk subpopulations at approximately 65 phospholipids per protein. No significant change in the fatty acid composition of the disks was observed as a function of change in cholesterol content. This work demonstrates compositional heterogeneity in disk membranes which may ultimately affect function.     

Cell envelope associations of Aquaspirillum serpens flagella.

Specific regions of the cell envelope associated with the flagellar basal complex of the gram-negative bacterium Aquaspirillum (Spirillum) serpens were identified by studying each of the envelope layers: outer membrane, mucopeptide, and plasma membrane. The outer membrane around the flagella insertion site was differentiated by concentric membrane rings and central perforations surrounded by a closely set collar. The perforations in both the outer membrane and the isolated mucopeptide layer were of a size accomodating the central rod of the basal complex but smaller than either the L or the P disks. The P disk of the complex may lie between the mucopeptide and the outer membrane. Electron microscopy of intact, spheroplasted, or autolyzed preparations did not adequately resolve the location of the inner pair of disks of the basal complex. Freeze-etching, however, revealed differentiation within the plasma membrane that appeared to be related to the basal complex. The convex fracture face showed depressions which are interpreted as impressions of a disk surrounded by a set of evenly spaced macromolecular studs and containing a central "plug" interpreted as the central rod. In thin sections, blebs, which appear to be associated with the flagellar apparatus, were seen on the cytoplasmic side of the plasma membrane. Superimposing the dimensions of the flagellar basal complex and the spacings of the cell envelope layers and using the position of the L disk within the outer membrane for reference, showed that the S disk might be within and the M disk beneath the plasma membrane. A tentative model was developed for comparison with that based on the structure of the Escherichia coli basal complex.