Figure 1. Principle of the glutathione-S-transferase (GST) system. After transformation of the expression plasmids to Escherichia coli, propagation of transformed cells and induction of expression by isopropyl--d-thiogalactopyranoside (IPTG) extracts are prepared. (a) Protein extracts containing GST (-fusion) proteins are subsequently incubated in the presence of gluthathione beads, which are then spun down and washed. (b) Purified fusion proteins attach to the beads and can be eluted from the beads by incubation with a buffer containing glutathione, after which they are ready for further use.

Figure 2. GST pull-down. GST (-fusion) proteins are incubated with in vitro translated proteins labeled with, for example, [³⁵S]methionine. After isolation of the fusion proteins plus associating proteins by centrifugation, the isolated proteins are subsequently detached from the beads by boiling in a denaturing buffer, followed by SDS-PAGE and detection. Lane 1, molecular size marker. Lane 2, input of labeled in vitro translated protein. Lane 3, results of a control pull-down experiment of GST proteins and the in vitro translated proteins. Lane 4, pull-down of GST-fusion proteins with the in vitro translated proteins. The translated protein binds to the GST-fusion protein (lane 4), but not to the GST control (lane 3).

Figure 3. Far-Western analysis. Protein extracts or (partially) purified proteins are loaded on an SDS-PAGE gel. The proteins are subsequently blotted to a membrane. Optionally, the proteins on the filter can be completely denatured, followed by renaturation. The blot is subsequently incubated with a purified and labeled protein probe, followed by washing of the membrane and detection. Lane 1, molecular size marker; lane 2, whole-cell extract (WCE) containing protein X; lane 3, protein fraction lacking proteins X; lane 4, purified protein X. The protein probe can bind directly to protein X. In the WCE, one smaller protein is present which also can directly bind to the protein probe.

Figure 4. The classic yeast two-hybrid assay exploits the modular structure of transcriptional activators: a protein–protein interaction is identified by reconstitution of the activities of a transcriptional activator. (a) One yeast expression plasmid encodes the DNA binding domain of the transcription factor GAL4 (GAL4-D) fused to the protein of interest (B, ‘bait’). This fusion protein binds to the promoter region, but, in the absence of a transcriptional activation domain, this protein will not give rise to significant transcriptional activation of the -galactosidase reporter (LacZ) gene. (b) The other yeast expression plasmid encodes the transcriptional activation domain of GAL4 fused to the protein of interest or proteins encoded by a cDNA library (P, ‘prey’). In the absence of a DNA-binding domain or promoter-targeting domain, this fusion protein will not give rise to transcriptional activation. (c) Two-hybrid experiment. The bait plasmid and the prey plasmid are cotransformed. The binding of the prey to the bait gives rise to the presence of a strong transcriptional activation domain in the promoter region, resulting in (strongly) increased transcriptional activation. (d, e) Two additional control experiments are necessary to verify that the interaction does not involve binding to GAL4-A or GAL4-D. (d) The bait together with the GAL activation domain alone should not give rise to transcriptional activation. (e) The GAL4 DNA-binding domain alone with the prey should not result in transcriptional activation.