Daniel Yohannes Infinity Pharmaceuticals, Inc. 780 Memorial Drive, Cambridge, MA 02139
Introduction - The association of proteins with other proteins is one of the most common interactions in biology. Such interactions play a central role in the regulation of numerous cellular functions. While many of these processes are mediated by enzymes including kinases, proteases, or glycosylases, the pathways that regulate these cellular functions are initiated or inhibited via specific proteinprotein interactions. The precise regulation of these pathways often involves the assembly of multiple proteins. The resultant oligomeric protein complexes comprise many enzymes, viral proteins, and receptor-ligand partners (1). The assembly of protein complexes is critical for allosteric control, formation and conformational maintenance of active sites of oligomeric enzymes, regulatory processes such as signal transduction, cell-cell contacts, electron transport systems, and antigen-antibody interactions (2-12). Specific intervention of these partners may lead to an increased understanding of which components to target in pathological situations.
The wealth of information regarding protein-protein interactions as well as many descriptions of aberrant protein-protein interactions in disease has made the disruption of protein-protein interactions a focus of recent research activity in the pharmaceutical and academic scientific communities. The quest for biochemical probes and therapeutic agents which intervene at protein-protein interfaces provided the impetus for the discovery of antibodies, dominant-negative proteins, or short peptides which inhibit protein assembly in the literature. Modulation of proteinprotein interactions by small molecules has proved to be very challenging and there have been a paucity of successes reported in the literature. However, this chapter will cover recent advances in the identification and detection of protein-protein interactions, as well as the identification of protein complexes which have become recent targets of pharmaceutical intervention. The associated molecular tools and potential therapeutics which have emerged in the past year will also be described.
Identification and Detection of Protein-Protein Interactions - Before the advent of genomics, standard methods for the detection of protein-protein interactions were crosslinking, co-fractionation by chromatography and co-immunoprecipitation. In 1989, Fields and Song reported the "Yeast Two-Hybrid Assay" (13). This breakthrough genetic system studied protein-protein interactions by taking advantage of the properties of the GAL4 protein of yeast saccharomyces cerevisiae. GAL4 is a transcription factor made up of an N-terminal DNA binding domain and a C-terminal activation domain. A 'bait' protein was fused to the DNA binding domain and used to identify protein partners from a library of c-DNAs cloned into a vector encoding the GAL4 transactivation domain. Interaction of bait partners enhances the transcription of the GAL4 promoter. The binding partners were then sequenced. This unbiased process allows identification of direct protein-protein interactions.
The advancement of proteomics has allowed protein microarrays to complement gene expression profiling. Initially, the technology involved the arraying of cDNA expression libraries on PVDF membranes (14). Early methods in which proteins were arrayed at low density have been superceded by well defined high density
ANNUAL REPORTS IN MEDICINAL CHEMISTRY—38 ISSN: 0065-7743
protein arrays on glass surfaces in the past several years. In a seminal paper, the identification of protein-protein interactions has been disclosed by using fluorophore tagged proteins to probe proteins arrayed on aldehyde slides (15). In this system, the authors were able to elucidate the protein-protein interactions of FKBP12 using commercially available instrumentation. The current state of protein array methodologies has been recently reviewed (16). Further recent advances include a report describing a protein-domain chip used to identify novel protein-protein interactions (17). In this work, GST fusion proteins were arrayed onto nitrocellulose-coated slides to generate a protein domain chip on which the proteins maintained their binding integrity. These immobilized proteins interacted with proteins from a cell lysate and the interaction was detected with a specific antibody to provide an intracellular interaction map for a cytosolic protein of interest. From this work, the domain-binding profiles were determined for Sam68 (Src-associated during mitosis 68) and a core small nuclear ribonucleoprotein called SmB'. The use of red and green fluorescent proteins (RFP and GFP) has gained popularity in protein microarrays. This technique permits the investigation of protein-protein interactions without the need for additional labeling steps (fluorescent dye labelling and purification) of probe proteins. Such an approach was reported whereby recombinant proteins labeled with RFP and GFP were used in protein microarrays as tags to investigate antigen-antibody interactions and other protein-protein interactions (18). Although membrane protein-protein interactions are not commonly found by protein array technology, a recent report documents the immobilization of a membrane which provides for lateral fluidity of proteins, and which houses functional GPCRs as determined by their binding profiles (19).
While protein microarrays provide, in principle, potential for full access of the proteome on a solid phase, in a recent review on protein microarray technology, it has been suggested that it is fundamentally more challenging to work with proteins than with nucleic acids (20). Thus a newer technique has emerged: the immobilization of small molecules on a surface which permits the use of tagged proteins in the mobile phase for detection. The potential for an ultra-high-throughput full-proteome analysis with this approach is very high. An early report of this approach coupled this novel array methodology with large numbers of molecules derived from diversity-oriented synthesis (DOS) (21). By probing a high density microarray of DOS small molecules with fluorescently labeled yeast protein Ure2p (which suppresses transcription factors Gln3p and Nihp), a highly specific compound called uretupamine (!) was identified which activates a glucose-sensitive transcriptional pathway downstream of Ure2p (22). Cell based mechanistic assays with the identified small molecules allows this process to become a powerful tool, and it is currently being industrialized in a drug discovery setting. An alternative approach to small molecule microarrays has been taken wherein small libraries of designed (mechanism-based) covalently-modifying inhibitors conjugated to sequence-encoded peptide nucleic acids were able to identify activity-based profiles
Was this article helpful?