Protein interaction cloning

[Cheung C. Yue]

Quite some time back, I had responded to a post by stating a number
of new techniques have been described to clone genes for proteins 
which interact with other proteins.  A subsequet post requested references.
My desk has been a mess and it's taken me a while to find them, so here
they are (at least a partial list).


Interaction cloning: identification of a helix-loop-helix zipper protein 
that interacts with c-fos.  Science 256:1014; '92.  
     A fusion protein is constructed with the protein of interest linked to 3 
modules:  a) a FLAG sequence for easy purification, b) an enterokinase 
peptidase cleavage site, and c) a HMK (heart muscle kinase) recognition 
site for 32-P labeling.  The fusion protein is then used for screening an ex
pression library.
     
The two-hybrid system: a method to identify and clone genes for proteins 
that interact with a protein of interest.  PNAS 88:9578; '91.
     DNA binding domain of GAL-4 is fused to protein of interest.  A library 
of sequences is then fused to the activation domain of GAL-4.  Functional 
reconstitution of GAL-4 is read out by a reporter gene with GAL-4 binding sites.
     
A contingent replication assay for the detection of protein-protein 
interactions inanimal cells.  PNAS 88:1086; 91.
     Similar to the two-hybrid system.  Has an amplification loop using 
SV40 T antigen cloned behind GAL-4 binding site, and the fusion protein 
cloned behind SV40 ori.  Newly amplified plasmid sequences are DpnI-
resistant due to lack of methylation, and are selected for by DpnI digestion.
     
Direct selectionfor sequences encoding protease of known specificity.  PNAS 
88:5159;'91.
     Still using GAL-4, but now introduces a specific protease target site 
between the DNA binding domain and the activation domain.  If specific 
protease is co-expressed in the same cell, GAL-4 function would be lost, 
which can then be selected for.

ccy at po.cwru.edu

--