Green fluorescent protein (GFP; Figure 1) has been a wonderful addition to the tools we use to study synapses. GFP is relatively small protein that fluoresces when irradiated by light.

Figure 1. A GFP crystal structure showing it forms a ß-can structure (left) consisting of 11 anti-parallel ß-strands which form a cylinder. Loops (white) and a little bit of alpha helical structure (pink) connect the 11 ß-strands. The chromophore, p-hydroxybenzylidene-imidazolidone (in green), is formed via a post-translational oxygen-requiring reaction with contributions from the Ser65-Tyr66-Gly67 tripeptide. GFP is a very stable molecule and the chromophore is well protected and in a very insulated local environment in the middle of the cylinder. Surprisingly, deletion of more than the N-terminal Met or more than seven amino acids from the C-terminus leads to a complete loss of fluorescence. Nevertheless, both N- and C- terminal fusions are fluorescent consistent with the N- and C-terminus of GFP being flexible enough not to disrupt the general structure of the molecule. Structure:1EMB from PDB displayed with RasMol 2.6

We have created several different GFP fusions to synaptic proteins that permit us to examine synapses. These include synaptobrevin-GFP, synaptogyrin-GFP and GFP-RAB-3. The most widely used of these is synaptobrevin-GFP which is used in cell culture, worms, and flies. The brightest of these fusions in our hands is GFP-RAB-3. In this case, GFP is added the N- instead of C-terminus because RAB-3 is post-translationally modified with the addition of a geranylgeranyl hydrophobic moiety to C-terminal cysteines which act as its anchor to the synaptic vesicle membrane. Because GFP:-RAB-3 is so much brighter we can now use the dissecting microscope to see certain sets of synapses in C. elegans . By expressing the GFP tag in different subsets of neurons using appropriate cell specific promoters we can examine different subsets of synapses. We mostly work on two different set of neurons (figure 2).

Figure 2 GFP-RAB-3 tagging the mechanosensory synapses An image of a live jsIs821 animal expressing GFP-RAB-3 in six mechanosensory neurons under the control of the mec-7 promoter. RAB3 associates with synaptic vesicles that accumulate at synapses. In this animal, a cluster of synapses in the nerve ring and a pair of synaptic puncta in the ventral nerve cord are easily visible. Dimmer autofluorescence derived from the intestine can also be seening along most of the length of the animals. These types of transgenic animals permit us to examine select synapses in live animals even under a dissecting microscope. We utilize synaptic-GFP tags extensively in the laboratory to study synapse formation. This image was taken using a 20X 0.5 na objective on an Olympus compound microscope. Photo M.Nonet.

We are also attempting to develop tags for other compartments of the presynapse. Other groups working in C. elegans and on other organisms have developed GFP tagged receptors and scaffolding proteins as markers for the postsynapse. These postsynaptic tags include the glutamate receptor subunit GLR-1, the acetylcholine receptor subunit UNC-29, the GABA receptor UNC-49 and scaffolding proteins like LIN-10.

Using spectral variants of GFP (Figure 3), one can create double labeled strains that tag both the presynapse and the postsynapse.

Excitation and emission of GFP variants. Graphs of spectra from Clontech.

There are now literally hundreds of different varients of GFP and dsRed and other fluorescent proteins. Here is a link to a list of some of the better ones.