| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of * Neurobiology, Physiology, and Jules Stein Eye Research Institute, University of California at Los Angeles School of Medicine, Los Angeles, California; Department of Structural Chemistry, School of Pharmacy, University of Milan, Milan, Italy; and Department of ¶ Neurobiology and ** Center of Neuroengineering, Duke University Medical Center, Durham, North Carolina
Correspondence: Address reprint requests to G. A. Zampighi, Tel.: 310-206-2883; E-mail: gzampighi{at}mednet.ucla.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
15 nm in diameter (2% of the vesicle's surface) and contained a smaller 6 nm region where the proximal leaflets merged (hemi-fused). Hemi-fusion was only observed on the side of vesicles in contact with the active zone; at the side of contact between neighboring vesicles, the membranes were not hemi-fused. Approximately three-fourths of the docked vesicles contained hemi-fused regions. Vesicles fully fused to the active zone (exhibiting pores that appeared as interruptions of a single membrane) were less frequently observed (1 of 10 hemi-fused vesicles). In conclusion, our observations in cortical synapses strengthen the hypothesis that hemi-fusion is a stable intermediary that precedes full fusion and release. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
). Despite the success of the vesicular release model in explaining synaptic transmission, the series of steps leading to the formation of the fusion-pore and the mixing of the protein and lipid components of both membranes have remained topics of active investigation. The crux of the problem involves the fact that the phospholipid bilayer component of biological membranes represents formidable barriers to fusion (2–4). Consequently, synaptic vesicle fusion is viewed as a multistep, highly regulated process involving complexes formed between proteins in the membrane of the vesicle (synaptobrevin) with proteins in the membrane of the presynaptic terminal (syntaxin, SNAP-25) (5–8).
Once close membrane apposition has been established, two radically different models of fusion have been proposed (for review, see (9)). In one model, referred to as direct-fusion, small protein subunits form oligomers that transverse each of the membranes and link across the aqueous space to form a gap-junction-like pore (10,11). Such a fusion pore would first release the vesicle's contents and later facilitate the mixing of the protein and lipid components of vesicle and active zone. In the competing model, referred to as fusion-through-hemifusion, fusion proceeds via sequential merging of the proximal and distal leaflets comprising the membranes (12). First to fuse are the proximal leaflets that face each other across the aqueous space, and form the intermediary called hemi-fusion. During this hemi-fused state, the membranes can exchange lipids and fluorescent markers in the proximal leaflets of the membranes. The distal leaflets remain separate, a fact that hinders the release of the vesicular contents into the extracellular space. Hemi-fusion is thought to represent a stable intermediary in the fusion path of diverse systems, including protein-free bilayers with specific lipid composition (12–14), as well as fusion of viruses with the cell membrane (15). The subsequent merging of the distal leaflets forms the fusion-pore that mediates both release of the vesicular contents and intermixing of lipids and proteins in the distal leaflets of the membranes.
Distinguishing between the models involves solving the three-dimensional structure of the region of contact between vesicles and plasma membrane. In direct-fusion, the region of contact should be composed of two closely apposed membranes, while in the competing fusion-through-hemifusion model only a single hemi-fused membrane should be observed. Determining whether a given area of contact is comprised of a double or single membrane (i.e., 6 nm or 12 nm in overall thickness) requires imaging synapses using thin sectioning electron tomography, a method that eliminates the artifact induced by projecting the entire volume of the section onto a single plane (the projection-artifact) and with which it is possible to resolve individual membranes as trilayer structures (the unit-membrane pattern, (16,17)).
In this study, we have applied the general method of conical electron tomography (18,19) to determine the structure of the region of contact in an effort to test predictions from the models. The principal observation was that approximately three-fourths of the vesicles associated (docked) to the active zone exhibited small hemi-fused regions. These vesicles were also linked to the plasma membrane by densities shaped as fibrils that might correspond to trans-SNARE or core complexes (20). Fully fused vesicles were less frequent (1 per 10 hemi-fused) and were characterized by pores that appeared as interruptions in a single membrane, which most likely had been hemi-fused. Therefore, the demonstration of hemi-fusion in the region of contact between vesicles and active zone revealed here suggests that the fusion-through-hemifusion model is consistent with how neurotransmitters are released in chemical synapses of the neocortex.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
–23). We sacrificed three Sprague-Dawley adult rats 90–120 days of age and fixed by perfusion with 3% glutaraldehyde, and 4% paraformaldehyde in 0.2 M cacodylate buffer at pH 7.4. The brains were removed and sliced in a vibrotome in sections 100–150 µm. Tissue from the frontal association and motor neocortices was dissected from the sections and processed for thin sectioning electron microscopy. After OsO4 post-fixation and block staining, the tissue was embedded in EPON 812. Thin sections of thickness with gray-to-silver interference color (55–80 nm) were cut, collected on 200 mesh grids, and stained with solutions of uranyl and lead citrate. Gold particles, 10 nm in diameter, deposited on the surfaces of the thin sections were used for centering the conical series, estimation of the thin section thickness, and the calculation of preliminary three-dimensional reconstructions.
Glutaraldehyde cross-linking and the Criegee reaction
It is well known that fixatives used in the preparation of thin sectioning electron microscopy react with both the protein and lipid components of tissues. At the concentration used in this study, glutaraldehyde cross-links 
-amino groups as well as less reactive groups, such as tyrosinyl (24), guanidyl, imidazolyl, and sulfhydryl groups (25). This means that it will react with most proteins as well as the headgroups of phospholipids, including PE, PS, and PC. In these cross-linking reactions, however, studies using x-ray diffraction methods have shown that the principal features of the three-dimensional structure of proteins remain intact (26). It thus seems unlikely that these reactions could be responsible for changing the overall thickness of the region of contact from 12 nm to 6 nm, as would be necessary for a double membrane to appear hemi-fused.
OsO4 reacts initially with double bonds of unsaturated lipids in the hydrophobic core of bilayers to form monoesters, the Criegee reaction (27,28). These monoesters are further reduced, and liberate oxides of the type Os(IV)O4·nH2O, which then migrate to the hydrophilic regions of the phospholipid bilayer. A result of OsO4 fixation is that all biological membranes, independent of their chemical composition, exhibit a trilayer pattern comprised of two dense bands (the stained surfaces) flanking an electron lucent core (the hydrophobic center), together referred to as the unit-membrane pattern (16,17). Clearly, such a property of the Criegee reaction would enhance, not detract, from the ability to determine whether the region of contact between vesicles and the active zone is comprised of single- or double-unit membranes.
Collection of the conical series
We used the Gatan 650 Single Tilt Rotating Holder (Gatan, Pleasanton, CA) in a FEI Tecnai 12 electron microscope (FEI, Hillsboro, OR) operated at 120 KV to collect 19 conical series. The holder allowed 55° tilts and controlled 5° (72 projections) rotations in the azimuth. The images were collected in a 2k x 2k CCD Gatan camera at 26,000x final magnification (pixel size 0.409 nm). Imaging was carried out using a minimum-dose method: searching was done at 2700x magnification with minimum illumination and the regions of interest imaged by focusing 2 µm away. The total electron dose was 800 e/A2, which we previously found decreased the overall thickness of the thin section by 15% (19).
Centering, alignment, and preliminary three-dimensional reconstructions
To bring the conical series into a common reference system, we selected a specific gold particle as the common center to all images in the series and aligned the entire conical series (72 projections) around this center. After centering, the coordinates of other gold particles (3–8) were recorded for every projection of the series. These values were used to find all orientation parameters (Euler angles and origin position) of the micrographs of the conical series (19). These data were used to calculate the preliminary three-dimensional maps with the weighted back-projection algorithm (29).
Refinement by projection matching
The alignment of the series, based on the manual tracking of the gold particles, was further refined by a strategy of projection matching where the initial projections were iteratively cross-correlated with reprojections of an updating reconstruction. During the computation of the reprojections, we tested incremental adjustments in the three projection parameters, , ß, and 
, and monitored the results by statistical indices based on the variance of the volume. The overall process allowed the determination of the best values for all the shift and orientation parameters and consequently improved both alignment and resolution of the final maps (18,19).
Visualization of the maps
We used the Amira software package (Mercury Computer Systems, San Diego, CA) to visualize entire maps as well as those restricted to docked and fully fused vesicles contained in the synapses. The individual planes comprising the three-dimensional maps were always presented in inverse contrast (Figs. 2, D–F; 3, D–F; and 5). The reason for this inversion involves the fact that, during visualization, the white pixels represent signal and the black pixels, emptiness. Volume-rendering techniques, where voxels were assigned colors and intensities of light based on a number of user-defined parameters, were used to represent the maps. This type of representation was improved further by segmentation whereby voxels corresponding to vesicles that were hemi-fused or fully fused with the active zone were extracted from each plane. Simply stacking the demarcated regions generated three-dimensional models of individual vesicles as well as neighboring structures, such as the plasma membrane. The maps resulting from these segmentations proved to be exceptional assets in the analysis of the relationships between vesicles that were hemi-fused and fully fused with the active zone.
|
|
|
)), which requires two independent three-dimensional reconstructions. To satisfy this requirement, we split the series into subsets comprised of odd and even numbered images. The subsets were used to reconstruct two independent three-dimensional maps. FSC was computed using either the entire reconstructed volume (1024 x 1024 x 128 voxel) or small cubic regions (128 x 128 x 128 voxels) for a local measure of resolution (19). For computation we have used the EMAN (31) and BSOFT (32) software packages, as well as software developed by one of the members of our research team (S.L.). The resolution was assessed by cutting the FSC curves with thresholds corresponding to five-times the correlation noise (33).
The second method was to measure the thickness of the unit-membrane pattern (18). We used individual planes selected from three refined three-dimensional maps where the pattern was visible throughout the volume. For each map, we selected five planes and in each plane we performed 10–15 measurements using the measuring editor of the Amira package. Since there was significant variability in the distribution of the densities comprising the pattern, we measured regions of the plasma and vesicular membrane where the dense layers appeared equivalent. In these regions, we measured the distance between the centers of the layers (18). This method eliminated the bias in selecting the boundaries of the dense layers. By comparing the resolution between both methods, we have estimated that the resolution of the conical tomographic reconstructions was 3–4 nm (18).
Quantifications
Using the Amira or ImageJ software packages, we measured the length of the active zones in all reconstructions in which they were located (13 out of 19) by measuring the length of a curve line corresponding to the length of thick density in the postsynaptic terminal (densities colored green, Fig. 1 B). The area of the active zone was calculated by multiplying this length by its thickness estimated from the number of planes in the reconstruction (each plane was .82 nm and the average thickness of the thin sections was 47 ± 15 nm). Only vesicles the centers of which were contained in the reconstructed volume were included in the count. The total number of vesicles was obtained by examining individual planes comprising the reconstruction to insure that all vesicles were included and none counted twice. The number of docked vesicles was designated as the subset of the total number of vesicles that were in direct contact or located at <15 nm from the membrane of the active zone. The number of hemi-fused vesicles corresponds to the subset of docked vesicles where the area of contact with the active zone was comprised of a single unit membrane and measured 6 nm in overall thickness.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
,19), we reconstructed chemical synapses from the rat frontal association and motor neocortex prepared for thin sectioning electron microscopy. Key advantages of conical tomography include: 1), elimination of artifacts induced by projecting the entire thickness of the thin section onto a single plane (the projection-artifact); 2), isotropic in-plane resolution (3–4 nm)—approximately two orders-of-magnitude higher than that achieved by optical methods; 3), high throughput; and 4), elimination of the need to impose symmetry or use averaging methods to increase signal/noise ratios.
Of the 19 synapses reconstructed, 13 maps contained active zones. These three-dimensional maps were studied by rendering their entire volume (0.1 µm3) (Fig. 1 A), by visualizing their individual planes (Figs. 2; 3, D–F; 4, D–F; and 6, A–C), and by segmenting the volumes of plasma membranes and hemi-fused as well as fully fused vesicles (Figs. 1 B; 3, A–C; 4, A–C; and 5). These visualization methods showed that the presynaptic terminal contained the characteristic vesicles described in previous studies of chemical synapses using conventional thin section electron microscopy (34–36). These synaptic vesicles were clustered at the active zone (37), the region of the presynaptic membrane facing the thick layer of densities associated to the postsynaptic membrane (brown, Fig. 1 B). Other organelles such as coated pits, coated vesicles, and large membranous sacs, called endosomes, were also present in the terminal, but these were located outside the region defined in this study as the active zone (yellow, Fig. 1 B).
We relied on the unit membrane pattern (16,17), representing the phospholipid bilayer structure to estimate the resolution achieved in the maps. To perform this task, we sectioned the reconstructed volume along orthogonal directions (x,y; x,z; and y,z planes). We measured the pattern from the center of the stained layers in the x,y (Fig. 2 A), y,z (Fig. 2 B), and y,z planes (Fig. 2 C), and determined an overall thickness of 6.1 ± 0.25 nm. The resolution of the maps was thus 3 nm, and was unaffected by either the distance of features from the center of the map or their orientation with respect to the tilt axis.
Having determined that the resolution of the maps was 3–4 nm in all directions, we proceeded to analyze the presynaptic terminals by counting the total number of vesicles and by measuring their diameter and the area of active zone to which they were associated. The diameter of these vesicles had a bimodal distribution: In four reconstructions, vesicle diameters measured 50 ± 7 nm (mean ± SD, n = 16), while in the remaining 15 it was 62 ± 5 nm (n = 18). Whether or not this bimodal distribution reflects functional differences among the synapses in the region (inhibitory/exhibitory) will require reconstructions of synapses that have been characterized functionally previous to chemical fixation.
We counted a total of 778 vesicles, 53 of which (or 7% of the total) were docked to 0.4 µm2 of active zone (Table 1). A vesicle was considered docked when its membrane was within 15 nm of the active zone. To distinguish between the direct-fusion and the fusion-through-hemifusion models, we then determined whether double or single membranes comprised the region of contact that docked vesicles established with the active zone. We found that regions comprised of a single membrane (hemi-fusion) could be seen in 42 of these 53 vesicles (or 74% of docked vesicles). In the remaining 11 docked vesicles, two closely apposed unit membranes comprised the regions of contact. These observations indicate that the overwhelming majority of docked vesicles were hemi-fused to the membrane of the active zone.
|
15 nm in total depth and exhibited a complex structure. Two-thirds (10 nm) of the area of contact was comprised of two closely apposed unit membranes (Fig. 3, D and G), while the remaining region (5 nm) was hemi-fused and comprised of a single unit membrane (vertical arrows, Fig. 3, E, F, H, and I). Using 50 nm for the diameter of an average synaptic vesicle, we estimated that the entire region of contact occupied 2% and that the smaller hemi-fused region occupied only 0.2% of the vesicle's area. Therefore, the region of contact between docked vesicles and the active zone includes sections of close membrane apposition (Fig. 3, D and G), and hemi-fusion (Fig. 3, E, F, H, and I).
Three additional features characterized the region of contact, all of which have been proposed as properties leading to the opening of the fusion pore in the fusion-through-hemifusion model. The first was a pronounced curvature that, in the plane, gave the docked vesicle a triangular appearance (Fig. 3, D–I) and, in three dimensions, a pear shape. The second was a shallower curvature in the plasma membrane toward the vesicle; since the two curvatures were of the opposite sign, the region of contact appeared as a stalk (Fig. 3, D and G). The third was the existence of densities linking the hemi-fused side with the active zone (arrowheads, Fig. 3, D and F). These densities appeared either as fibrils or as rings of 15 nm in diameter (arrowhead, Fig. 3 F). The fibrils were oriented perpendicularly, with the longer fibril (30 nm) attached to the plasma membrane (arrowhead, Fig. 3 D) and the shorter fibril (8.5 nm) attached to the vesicle. The finding of these types of densities is consistent with the idea that protein complexes bring vesicles into close proximity with the active zone.
Our study also identified four vesicles that were fully fused and appeared as small dimples in the membrane of the active zone (Figs. 4 and 5). The principal attribute of fully fused vesicles was a prominent pore that appeared as an interruption, (3–8 nm) in width, of the membrane at the region of contact. The pore was patent when examining individual planes of the region (arrow, Fig. 4, E and F) as well as when the volume of the fully fused vesicles was segmented and visualized by surface triangulation (arrow, Fig. 5 D). Using this visualization method, the pore was found to open toward the perimeter of the hemi-fused region rather than at its center (arrow, Fig. 4 E). We estimated a ratio of approximately one fully fused vesicle to every 10 hemi-fused vesicles (42:4). Despite this expectedly low frequency (see discussion), their observation in four independent reconstructions indicates that they appear to be a normal feature of synapses that have not been stimulated previous to fixation. Once again, conventional thin sectioning will make visualization of a pore of such dimensions impossible.
|
). As stated above, the hemi-fused region occupies a small fraction of the total thickness of the section (1/8). To test the possibility that projection artifacts might be responsible for the failure, we compared individual planes gathered from hemi-fused vesicles (Fig. 6, A–C) with projections of the entire volume (64 consecutive planes or 50 nm in thickness) of these same vesicles (Fig. 6, D–F). Analysis of the individual planes showed the two defining properties of hemi-fused vesicles at the region of contact: the merging of the proximal leaflets and the deformation of the membranes (red arrows, Fig. 6, A–C). When examining projections of these regions of contact, however, the images were quite different (Fig. 6, D–F). In two projected images, diffuse densities linking the vesicle to the plasma membrane replaced the hemi-fused region (yellow arrows, Fig. 6, D and F). In another projection, the docked vesicle was barely visible (Fig. 6 E). Since projection images mimic imaging by conventional electron microscopy, this simulation indicates that features smaller than the thickness of the section (50 nm), such as an 5–6 nm diameter hemi-fused region, cannot be resolved with this method.
|
Based on these considerations, we thus suggest that the hemi-fused state is a key structural feature of synaptic vesicles that were docked with the active zone in the rat neocortex.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
)). This model gives rise to the prediction that the region of contact should be comprised of two intimately apposed unit-membranes and measure 12 nm in overall thickness. In the fusion-through-hemifusion model, it is proposed that fusion proceeds via the sequential merging of first the proximal and then the distal leaflets of the apposing membranes. The state created by the hemi-fusion of the proximal leaflets hinders release of the vesicular contents into the extracellular space. Only after the distal leaflets have merged, can the pore open and the vesicular contents be released, completing the mixing of lipids and proteins (for review, see (9)). This model gives rise to the prediction that the region of contact should partially consist of a single unit membrane, measuring 6 nm in overall thickness, and that the fusion pore should traverse a single unit membrane (Fig. 7). The three-dimensional maps gathered in this study allowed us to test the predictions derived from these two models. Analysis of the individual planes of the region of contact of hemi-fused vesicles indicated that a single unit membrane was present in the region of contact (Fig. 3, D–F). Moreover, similar analysis performed at the region of contact of vesicles that were fully fused with the active zone showed that their fusion-pores span only this single membrane (Fig. 4, D–F). We thus suggest, based on our reconstructions, that the fusion-through-hemifusion or stalk-model reflects the structure of docked and fully fused vesicles more accurately than the direct-fusion model.
The hemi-fused region occupied only a small part of the region of contact between docked vesicles and the active zone (Fig. 3, E and F). Using the figure of 50 nm for the diameter of an average synaptic vesicle, we estimated that the smaller hemi-fused region occupied only 0.2% of the vesicle's area. The small dimension of the hemi-fused region with respect to the section's thickness (40–50 nm) additionally helps to explain why studies using conventional thin sectioning electron microscopy have failed to resolve this critical structural characteristic of docked vesicles.
It is useful to express the number of docked and hemi-fused vesicles as a function of the area of the active zone (Table 1). Since 53 vesicles (7% of the total) were docked to 0.4 µm2 of active zone, we estimated that the density of docked vesicles should be 130/µm2. To further understand the significance of our measurements, we can express this density as an average synapse from the 13 maps (0.6 µm in length and, assuming circular geometry, 0.28 µm2 of active zone area). From the proportionality between vesicles and the active zone established experimentally, we can further conclude that such a synapse would have 37 docked and 28 hemi-fused vesicles (Table 1). This prediction marks a first estimate of docked vesicles expressed per unit area of active zone.
The three-dimensional structure of the vesicles docked to the active zone contains features pertaining to the hypothesis that populations (i.e., pools) of vesicles with distinct functional properties exist in the presynaptic terminal (39,40). Conventional electron microscopy studies have hitherto identified no significant structural feature beyond that of distance from the active zone, which can be used to distinguish among vesicles in the terminal. Yet numerous functional studies indicate that synaptic vesicles can be separated into pools: those that are immediately available on stimulation, and those that release only upon intense stimulation (for review, see (41)). The finding of hemi-fusion marks the first instance of an additional characteristic directly relevant to this hypothesis: It follows simply that vesicles hemi-fused to the active zone may comprise the immediately-releasable pool (39,41).
Finally, the structure of docked vesicles supports the hypothesis that complexes comprised of SNARE proteins link synaptic vesicles to the active zone. Our observations that densities appearing as filaments or rods connecting hemi-fused vesicles with the plasma membrane strongly support this prediction (arrowheads, Fig. 3, D and F). Our observations offer no inconsistencies with atomic models of the trans-SNARE complexes determined by x-ray diffraction methods (42); however, a precise identification of individual SNAREs will require reconstructions of synapses where one or several of these proteins have been manipulated or knocked out. Despite these limitations in identifying individual SNAREs, our observations nonetheless provide direct evidence that complexes that form in vitro by mixing liposomes and purified SNAREs (43) could also exist in a synaptic environment.
In conclusion, our observations add docked synaptic vesicles to a long list of systems, in which hemi-fusion has been observed as a normal constituent of membranes in close apposition. Along with protein-free bilayers (12–14), SNARE-containing liposomes (44–46), viral fusion (2), SNARES that are ectopically expressed in cells (47), and physiological homotypic fusion of yeast vacuoles (48), docked synaptic vesicles also exhibit regions of contact where sequential merging of proximal and distal leaflets of apposing membranes lead to a fusion pore traversing only a single unit membrane. As with its observation elsewhere, the finding of hemi-fusion in synapses has profound consequences on our understanding of the system as a whole, not the least of which are its impact on the model of vesicular fusion, and its implication on the organization of functional pools.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This work was supported by grant No. EY-04110 as well as funds from the Foundation to Prevent Blindness (to G.Z.); grant No. GM-27278, from Philip Morris USA and Philip Morris International (to S.S.); FIRST 2001 (to S.L.); and grant No. DK44602 (to E.W).
Submitted on March 10, 2006; accepted for publication May 31, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
2. Chernomordik, L. V., and M. M. Kozlov. 2003. Protein-lipid interplay in fusion and fission of biological membranes. Annu. Rev. Biochem. 72:175–207.[CrossRef][Medline]
3. Cohen, F. S., and G. B. Melikyan. 2004. The energetic of membrane fusion from binding, through hemifusion, pore formation, and pore enlargement. J. Membr. Biol. 199:1–14.[CrossRef][Medline]
4. Malinin, V. S., and B. R. Lentz. 2004. Energetics of vesicle fusion intermediaries: comparison of calculations with observed effects of osmotic and curvature stresses. Biophys. J. 86:2951–2964.
5. Sudhof, T. C. 1995. The synaptic vesicle cycle: a cascade of protein-protein interactions. Nature. 375:645–653.[CrossRef][Medline]
6. Jahn, R., and T. C. Sudhof. 1999. Membrane fusion and exocytosis. Annu. Rev. Biochem. 68:863–911.[CrossRef][Medline]
7. Jahn, R., T. Lang, and T. C. Sudhof. 2003. Membrane fusion. Cell. 112:519–533.[CrossRef][Medline]
8. Li, L., and L.-S. Chin. 2003. The molecular machinery of synaptic vesicle exocytosis. Cell. Mol. Life Sci. 60:942–960.[Medline]
9. Chernomordik, L. V., and M. M. Kozlov. 2005. Membrane hemifusion: crossing a chasm in two leaps. Neuron. 123:375–382.
10. Peters, C., M. J. Bayer, S. Buhler, J. S. Andersen, M. Mann, and A. Mayer. 2001. Trans-complex formation by proteolipid channels in the terminal phase of membrane fusion. Nature. 409:581–588.[CrossRef][Medline]
11. Han, X., C. T. Wang, J. Bai, E. R. Chapman, and M. B. Jackson. 2004. Transmembrane segments of syntaxin line the fusion pore if Ca2+-triggered exocytosis. Science. 304:289–292.
12. Chernomordik, L., A. Chanturiya, J. Green, and J. Zimmerberg. 1995. The hemifusion intermediary and its conversion to complete fusion: regulation by membrane composition. Biophys. J. 69:922–929.
13. Cohen, F. S., J. Zimmerberg, and A. Finkelstein. 1980. Fusion of phospholipids vesicles with planar phospholipids bilayer membranes. II. Incorporation of a vesicular membrane marker into the planar membrane. J. Gen. Physiol. 75:251–270.
14. Lee, J., and B. R. Lentz. 1997. Evolution of lipidic structures during model membrane fusion and the relation of this process to cell membrane fusion. Biochemistry. 36:6251–6259.[CrossRef][Medline]
15. Kemble, G. W., T. Danieli, and J. M. White. 1994. Lipid-anchored influenza hemagglutinin promotes hemifusion, not complete fusion. Cell. 76:383–391.[CrossRef][Medline]
16. Robertson, J. D. 1959. The ultrastructure of cell membranes and their derivatives. Biochem. Soc. Symp. 16:3–43.[Medline]
17. Robertson, J. D. 1987. The early days of electron microscopy of nerve tissue and membranes. Int. Rev. Cytol. 100:129–201.[Medline]
18. Lanzavecchia, S., F. Cantele, P. L. Bellon, L. Zampighi, M. Kreman, E. M. Wright, and G. A. Zampighi. 2005. Conical tomography of freeze-fracture replicas: a method for the study of integral membrane proteins inserted in phospholipid bilayers. J. Struct. Biol. 149:87–98.[CrossRef][Medline]
19. Zampighi, G. A., L. Zampighi, N. Fain, E. M. Wright, F. Cantele, and S. Lanzavecchia. 2005. Conical Tomography II: a method for the study of cellular organelles in thin sections. J. Struct. Biol. 151:263–274.[CrossRef][Medline]
20. Sollner, T., S. W. Whiteheart, M. Brunner, H. Erdjument-Bromage, S. Geromanos, P. Tempst, and J. E. Rothman. 1993. SNAP receptors implicated in vesicle targeting and fusion. Nature. 362:318–324.[CrossRef][Medline]
21. Zampighi, G. A., J. M. Corless, and J. D. Robertson. 1980. On gap junction structure. J. Cell Biol. 86:190–198.
22. Zampighi, G. A., J. E. Hall, G. R. Ehring, and S. A. Simon. 1989. The structural organization and protein composition of lens fiber junctions. J. Cell Biol. 108:2255–2275.
23. Zampighi, G. A., and R. S. Fisher. 1997. Polyhedral protein cages encase synaptic vesicles and mediate their docking to the pre-synaptic membrane. J. Struct. Biol. 119:347–359.[CrossRef][Medline]
24. Bowes, J. H., and C. W. Carter. 1968. The interaction of aldehydes with collagen. Biochim. Biophys. Acta. 168:322–341.
25. Habeeb, A. F., and R. Hiramoto. 1968. Reaction of proteins with glutaraldehyde. Arch. Biochem. Biophys. 126:1–16.[CrossRef][Medline]
26. Hopwood, D. 1972. Theoretical and practical aspects of glutaraldehyde fixation. Histochem. J. 4:99–127.
27. Criegee, R. 1936. Osmium saure-ester als mischerproducte beir oxidation. [Osmium acid-ester as byproduct of oxidation.]. Ann. Chem. 522:75–96.
28. Criegee, R., B. Marchand, and H. Wannowius. 1942. Zur kentniss der organischer osmium-verbinguen gen. [On organic osmium compounds.]. Ann. Chem. 550:99–133.
29. Radermacher, M. 1992. Weighted back-projection methods. In Electron Tomography. J. Frank, editor. Plenum Press, New York.
30. van Heel, M. 1987. Similarity measures between images. Ultramicroscopy. 21:95–100.[CrossRef]
31. Ludtke, S. J., P. R. Baldwin, and W. Chin. 1999. EMAN: semiautomated software for high-resolution single particle reconstructions. J. Struct. Biol. 128:82–97.[CrossRef][Medline]
32. Heyman, B. J. 2001. BSOFT: image and molecular processing in electron microscopy. J. Struct. Biol. 133:156–169.[CrossRef][Medline]
33. Radermacher, M. 1988. Three-dimensional reconstruction of single particles from random and non-random tilt series. J. Electron Microsc. Tech. 9:359–394.[CrossRef][Medline]
34. Sjostrand, F. S. 1953. Ultrastructure of retinal rod synapses of Guinea-pig eye. J. Appl. Phys. 24:1422–1423.
35. Palay, S. L., and G. E. Palade. 1955. The fine structure of neurons. J. Biophys. Biochem. Cytol. 1:68–88.
36. Gray, E. G. 1963. The granule cells, mossy synapses and Purkinje spine synapses of the cerebellum: light and electron microscope observations. J. Anat. (Lond.). 97:101–106.[Medline]
37. Couteaux, R. 1961. Principaux criteres morphologiques et cytochimiques utilisables aujourd'hui pour definir les divers types de synapses. [Principal morphological and cytochemical criteria to define the diverse types of synapses.]. Acta Neurophysiol. 3:143–173.
38. Heuser, J. E., and T. S. Reese. 1981. Structural changes after transmitter release at the frog neuromuscular junction. J. Cell Biol. 88:564–580.
39. Delgado, R., C. Maureira, C. Oliva, Y. Kidokoro, and P. Labarca. 2000. Size of vesicular pools, rates of mobilization, and recycling and neuromuscular synapses of a Drosophila mutant, shibire. Neuron. 28:941–953.[CrossRef][Medline]
40. Schikorski, T., and C. F. Stevens. 2001. Molecular correlates of functionally defined synaptic vesicle populations. Nat. Neurosci. 4:391–395.[CrossRef][Medline]
41. Rizzoli, S. O., and W. J. Betz. 2005. Synaptic vesicle pools. Nat. Neurosci. Rev. 6:57–69.[CrossRef]
42. Sutton, R. B., D. Fasshauser, R. Jahn, and A. T. Brunger. 1998. Crystal structure of a SNARE complex involved in synaptic exocytosis at 2.4 Å resolution. Nature. 395:347–353.[CrossRef][Medline]
43. Weber, T., B. V. Zemelman, J. A. McNew, B. Westermann, M. Gmachl, F. Parlati, T. H. Sollner, and J. E. Rothman. 1998. SNAREpins: minimal machinery for membrane fusion. Cell. 92:759–772.[CrossRef][Medline]
44. Lu, X., F. Zhang, J. A. McNew, and Y.-K. Shin. 2005. Membrane fusion induced by neuronal SNAREs transits through hemifusion. J. Biol. Chem. 280:30538–30541.
45. Xu, Y., F. Zhang, Z. Su, J. A. McNew, and Y. K. Shin. 2005. Hemifusion in SNARE-mediated membrane fusion. Nat. Struct. Mol. Biol. 12:417–422.[CrossRef][Medline]
46. Dennison, M. S., M. E. Bowen, A. T. Brunger, and B. R. Lentz. 2006. Neuronal SNAREs do not trigger fusion between synthetic membranes but do promote PEG-mediated membrane fusion. Biophys. J. 90:1661–1675.
47. Giraudo, C. G., C. Hu, D. You, A. M. Slovic, E. V. Mosharov, D. Sulzer, T. J. Melia, and J. E. Rothman. 2005. SNAREs can promote complete fusion and hemifusion as alternative outcomes. J. Cell Biol. 170:259–260.
48. Reese, C., F. Heise, and A. Mayer. 2005. Trans-SNARE pairing can precede a hemifusion intermediary in intracellular membrane fusion. Nature. 436:410–414.[Medline]
49. Bloom, F. E., and G. K. Aghajanian. 1966. Cytochemistry of synapses: a selective staining method for electron microscopy. Science. 154:1575–1577.
50. Pfenninger, K., K. Akert, H. Moor, and C. Sandri. 1972. The fine structure of freeze-fractured presynaptic membranes. J. Neurocytol. 1:129–149.[CrossRef][Medline]
51. Phillips, G. R., J. K. Huang, Y. Wang, H. Tanaka, L. Shapiro, W. Zhang, W.-S. Shan, K. Arndt, M. Frank, M. A. Gawinowicz, and D. R. Colman. 2001. The presynaptic particle web: ultrastructure, composition, dissolution, and reconstitution. Neuron. 32:63–77.[CrossRef][Medline]
52. Kozlovsky, Y., L. V. Chernomordik, and M. M. Kozlov. 2002. Lipid intermediaries in membrane fusion: formation, structure, and decay of the hemifusion diaphragm. Biophys. J. 83:882–895.
This article has been cited by other articles:
 |
|
 |
|
  G. A. Zampighi, N. Fain, L. M. Zampighi, F. Cantele, S. Lanzavecchia, and E. M. Wright Conical Electron Tomography of a Chemical Synapse: Polyhedral Cages Dock Vesicles to the Active Zone J. Neurosci., April 16, 2008; 28(16): 4151 - 4160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Liu, T. Wang, E. R. Chapman, and J. C. Weisshaar Productive Hemifusion Intermediates in Fast Vesicle Fusion Driven by Neuronal SNAREs Biophys. J., February 15, 2008; 94(4): 1303 - 1314. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
