3.3 Fusion of vesicles with the target membrane
In this section, we shall look at how vesicles fuse with the appropriate target membrane. The targeting of different classes of transport vesicles to their distinct membrane destinations is essential in maintaining the distinct characteristics of the various eukaryotic organelles. Because coat proteins, such as clathrin, are found in different trafficking pathways, it follows that other proteins in the coat must specify the direction of transport of a particular vesicle and its ultimate destination.
What controls the specificity of vesicle targeting and docking? The interaction and fusion of the membranes of the vesicle and the target is a multistage event. The first stage is membrane recognition. The next step is a loose interaction called tethering. The subsequent interaction brings the opposing membranes much closer to each other and is called docking. Docking leads to membrane fusion.
Thus, targeting specificity might be thought of as a series of events that includes:
specification of the vesicle delivery site;
the recruitment of components capable of initiating vesicle ‘capture’;
the formation of a bridge between the vesicle and the target membrane;
conformational change that allows the vesicle and target membrane proteins to come close enough to interact;
dissociation of the tethering proteins, to free them for another round of transport.
Proteins involved in the tethering processes are called TRAPPs (transport protein particles). The TRAPP complex (Mr ~ 1,100,000) is made of 10 subunits and is essential for vesicle trafficking. TRAPP proteins are highly conserved and integral to the membrane. The discovery of different tethering complexes involved in the process of vesicle trafficking suggests that tethering is more complex than a simple cross-bridging of two membranes. It is more likely to be a series of events that involves pairing of specific membrane proteins (Figure 23).
Two families of integral membrane proteins ‘tag’ the membrane of the target organelle and the vesicle. Members of one family are known as vesicle SNAREs (v-SNAREs) and members of the other family are known as target SNAREs (t-SNAREs). (SNAREs may be categorised according to their amino acid sequence as well as by the v/t nomenclature. Components involved in the fusion between the donor membrane of the vesicle and the target membrane were identified because of their interactions with a soluble ATPase called NSF, identified by its sensitivity to NEM (N -ethylmaleimide). Thus SNARE is an acronym for Soluble NEM-sensitive factor Attachment protein REceptors.)
What are SNAREs and how is SNARE function studied? Most of the discoveries on SNARE function and their interactions in membrane trafficking have been made through studies of the proteins that attach to the membrane of synaptic vesicles (Box 2). We shall use this example to illustrate how SNAREs function.
Box 2 Synaptic vesicles
Synaptic vesicles are secretory vesicles containing neurotransmitters that are found clustered near the plasma membrane at the terminals of a nerve axon. These vesicles fuse with specialised areas of the plasma membrane called ‘active zones’, releasing the neurotransmitter into the extracellular space (the synaptic cleft). This process is the basis of signal transmission between nerves and it depends on the v/t SNARE complex.
Relatively pure synaptic vesicles from the nervous system may be prepared by subcellular fractionation. Brain tissue is homogenised in a medium that allows subcellular organelles, such as mitochondria and synaptic vesicles, to remain intact. Different organelles can be separated according to their buoyant density by sedimentation in sucrose density gradients. In this way it is possible to obtain relatively pure preparations of vesicles from neuronal as well as other secretory tissues. The availability of pure vesicles has enabled some of the key molecules in the process of exocytosis to be identified and the relationship between these molecules to be unravelled.
The synaptic SNARE complex, which comprises a v-SNARE (synaptobrevin) and two t-SNAREs (syntaxin and SNAP-25), is made of three proteins: synaptobrevin, syntaxin and SNAP-25 (named for SyNaptosomal Associated Protein of Mr 25,000).
The v-SNARE synaptobrevin is a transmembrane synaptic vesicle protein with a short C-terminal region inside the vesicle and the bulk of the protein in the cytoplasm. The t-SNARE syntaxin has a very similar structure but is located in the plasma membrane of the synapse, with the bulk of the protein also in the cytoplasm. In contrast, the t-SNARE SNAP-25 is firmly anchored to the plasma membrane by palmitoyl chains (see Section 3.3.3). These three proteins together form a complex that links the synaptic vesicle to the plasma membrane and whose structure has been determined by X-ray crystallography. They assemble with a 1 : 1 : 1 stoichiometry into a tight ternary complex called a fusion or SNARE complex (Figure 24). Much of our knowledge of these processes has been gained from the study of neurotoxins (Box 3).
Box 3 Neurotoxins and SNARE research
Important insights into the function of SNAREs have been gained through the use of toxins that poison the nervous system (neurotoxins). In the case of neurotransmitter release, evidence that SNAP-25 and synaptobrevin are essential for membrane docking was obtained in studies using tetanus toxin and botulinum toxin, which are bacterial neurotoxins, which cause paralysis by blocking neurotransmitter release. They are produced by bacteria as single proteins that are then cleaved into two different subunits, known as the light chain and the heavy chain. The heavy chain binds to the external membrane of neurons and facilitates the entry of the light chain into the cytoplasm of the nerve cell at the synaptic terminal. Once the light chains enter the cytoplasm they act as proteases that rapidly destroy specific target proteins. For example, botulinum B and tetanus toxin destroy only synaptobrevin, whereas botulinum A selectively cleaves SNAP-25. Another toxin, botulinum C1, destroys syntaxin. Because each of these toxins is very specific in their action they have provided direct evidence that SNARE proteins are critical components of the process of neurotransmitter release.
Note that tetanus toxin produces spastic paralysis by preventing transmitter release in the CNS, whereas botulinum B toxin causes flaccid paralysis by preventing acetylcholine release at the neuromuscular junction.
All SNARE proteins share a characteristic motif that consists of a stretch of approximately 60 amino acids called the SNARE motif. The SNARE motif is the principal protein–protein interaction region; it is where synaptobrevin binds tightly to corresponding SNARE motifs in syntaxin and SNAP-25 to form an exceptionally stable complex.
After docking of the synaptic vesicle, SNARE proteins undergo a priming step that probably involves the transition of syntaxin to an open conformation required for SNARE complex formation. Another protein, called Unc (named after uncoordinated phenotype of C. elegans ), interacts with the syntaxin in its open state to stabilise the conformation, which enables the SNARE complex to form. The stable assembly of the SNARE complex is believed to drive membrane fusion, which involves a number of other proteins shown in Figure 24, although the precise interactions are not yet known.
For fusion and cargo release to proceed normally, the SNARE complex must become disassembled. In nerve terminals, this process occurs at some point after fusion, and is a step required for recovery and recycling of the membrane. Disassembly is carried out by proteins called NSF and three different SNAPs (unrelated to SNAP-25), which prise apart the SNARE complex in a process that requires ATP hydrolysis (see Figure 26).
v/t SNARE pairing presents an attractive lock-and-key mechanism, which may underlie the specificity of vesicle targeting and docking. In other words, the v/t SNARE recognition process could provide a means by which a cell controls the specificity of a vesicle–target membrane fusion. Direct evidence for the involvement of SNAREs in bilayer fusion was lacking until recently, when studies in which SNAREs were reconstituted into liposomes and the liposomes placed inside permeabilised living cells provided strong evidence that SNARE complex formation is associated with the physiological fusion step.
However, several lines of evidence indicate that the same v-SNARE can reside both on vesicles moving towards the plasma membrane and on those being recovered from it. Also it appears that one single v-SNARE can interact with several t-SNAREs. Hence the v–t interaction is not sufficient to ensure the necessary targeting/docking/fusion specificity. This suggests that SNAREs cannot be the sole determinants of vesicle docking that occurs prior to fusion.