Overview of Protein–Protein Interaction Analysis

See all protein interaction analysis products

Proteins are the workhorses that facilitate most biological processes in a cell, including gene expression, cell growth, proliferation, nutrient uptake, morphology, motility, intercellular communication and apoptosis. But cells respond to a myriad of stimuli, and therefore protein expression is a dynamic process; the proteins that are used to complete specific tasks may not always be expressed or activated. Additionally, all cells are not equal, and many proteins are expressed in a cell type–dependent manner. These basic characteristics of proteins suggest a complexity that can be difficult to investigate, especially when trying to understand protein function in the proper biological context.

Critical aspects required to understand the function of a protein include:

• Protein sequence and structure—used to discover motifs that predict protein function
• Evolutionary history and conserved sequences—identifies key regulatory residues
• Expression profile—reveals cell-type specificity and how expression is regulated
• Post-translational modifications—phosphorylation, acylation, glycosylation and ubiquitination suggest localization, activation and/or function
• Interactions with other proteins—function may be extrapolated by knowing the function of binding partners
• Intracellular localization—may allude to the function of the protein

Until the late 1990s, protein function analyses mainly focused on single proteins. However, because the majority of proteins interact with other proteins for proper function, they should be studied in the context of their interacting partners to fully understand their function. With the publication of the human genome and the development of the field of proteomics, understanding how proteins interact with each other and identifying biological networks has become vital to understanding how proteins function within the cell.

Protein interactions are fundamentally characterized as stable or transient, and both types of interactions can be either strong or weak. Stable interactions are those associated with proteins that are purified as multi-subunit complexes, and the subunits of these complexes can be identical or different. Hemoglobin and core RNA polymerase are examples of multi-subunit interactions that form stable complexes.

Transient interactions are expected to control the majority of cellular processes. As the name implies, transient interactions are temporary in nature and typically require a set of conditions that promote the interaction, such as phosphorylation, conformational changes or localization to discrete areas of the cell. Transient interactions can be strong or weak, and fast or slow. While in contact with their binding partners, transiently interacting proteins are involved in a wide range of cellular processes, including protein modification, transport, folding, signaling, apoptosis and cell cycling. The following example provides an illustration of protein interactions that regulate apoptotic and anti-apoptotic processes. 

Heavy BAD protein–protein interaction. Panel A: Coomassie-stained SDS-PAGE gel of recombinant light and heavy BAD-GST-HA-6xHIS purified from HeLa IVT lysates (L), using glutathione resin (E1) and cobalt resin (E2) tandem affinity. The flow-through (FT) from each column is indicated. Panel B: Schematic of BAD phosphorylation and protein interactions during cell survival and cell death (i.e., apoptosis). Panel C: BAD protein sequence coverage showing identified Akt consensus phosphorylation sites (red box). Panel D: MS spectra of stable isotope-labeled BAD peptide HSSYPAGTEDDEGmGEEPSPFr.

Proteins bind to each other through a combination of hydrophobic bonding, van der Waals forces, and salt bridges at specific binding domains on each protein. These domains can be small binding clefts or large surfaces and can be just a few peptides long or span hundreds of amino acids. The strength of the binding is influenced by the size of the binding domain. One example of a common surface domain that facilitates stable protein–protein interactions is the leucine zipper, which consists of α-helices on each protein that bind to each other in a parallel fashion through the hydrophobic bonding of regularly-spaced leucine residues on each α-helix that project between the adjacent helical peptide chains. Because of the tight molecular packing, leucine zippers provide stable binding for multi-protein complexes, although all leucine zippers do not bind identically due to non-leucine amino acids in the α-helix that can reduce the molecular packing and therefore the strength of the interaction.

Two Src homology (SH) domains, SH2 and SH3, are examples of common transient binding domains that bind short peptide sequences and are commonly found in signaling proteins. The SH2 domain recognizes peptide sequences with phosphorylated tyrosine residues, which are often indicative of protein activation. SH2 domains play a key role in growth factor receptor signaling, during which ligand-mediated receptor phosphorylation at tyrosine residues recruits downstream effectors that recognize these residues via their SH2 domains. The SH3 domain usually recognizes proline-rich peptide sequences and is commonly used by kinases, phospholipases and GTPases to identify target proteins. Although both SH2 and SH3 domains generally bind to these motifs, specificity for distinct protein interactions is dictated by neighboring amino acid residues in the respective motif.

The result of two or more proteins that interact with a specific functional objective can be demonstrated in several different ways. The measurable effects of protein interactions have been outlined as follows:

• Alter the kinetic properties of enzymes, which may be the result of subtle changes in substrate binding or allosteric effects
• Allow for substrate channeling by moving a substrate between domains or subunits, resulting ultimately in an intended end product
• Create a new binding site, typically for small effector molecules
• Inactivate or destroy a protein
• Change the specificity of a protein for its substrate through the interaction with different binding partners, e.g., demonstrate a new function that neither protein can exhibit alone
• Serve a regulatory role in either an upstream or a downstream event

Usually a combination of techniques is necessary to validate, characterize and confirm protein interactions. Previously unknown proteins may be discovered by their association with one or more proteins that are known. Protein interaction analysis may also uncover unique, unforeseen functional roles for well-known proteins. The discovery or verification of an interaction is the first step on the road to understanding where, how and under what conditions these proteins interact in vivo and the functional implications of these interactions.

While the various methods and approaches to studying protein–protein interactions are too numerous to describe here, the table below and the remainder of this section focuses on common methods to analyze protein–protein interactions and the types of interactions that can be studies using each method. In summary, stable protein–protein interactions are easiest to isolate by physical methods like co-immunoprecipitation and pull-down assays because the protein complex does not disassemble over time. Weak or transient interactions can be identified using these methods by first covalently crosslinking the proteins to freeze the interaction during the co-IP or pull-down. Alternatively, crosslinking, along with label transfer and far–western blot analysis, can be performed independent of other methods to identify protein–protein interactions.

Co-immunoprecipitation (co-IP) is a popular technique for protein interaction discovery. Co-IP is conducted in essentially the same manner as an immunoprecipitation (IP) of a single protein, except that the target protein precipitated by the antibody, also called the "bait", is used to co-precipitate a binding partner/protein complex, or "prey", from a lysate. Essentially, the interacting protein is bound to the target antigen, which is bound by the antibody that is immobilized to the support. Immunoprecipitated proteins and their binding partners are commonly detected by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and western blot analysis. The assumption that is usually made when associated proteins are co-precipitated is that these proteins are related to the function of the target antigen at the cellular level. This is only an assumption, however, that is subject to further verification.

Co-immunoprecipitation of cyclin B and Cdk1. The Thermo Scientific Pierce Protein A/G Magnetic Beads bind to Cdk1 antibody complexed with Cdk1. Cyclin B is bound to the Cdk1, and is captured along with its binding partner.

Pull-down assays are similar in methodology to co-immunoprecipitation because of the use of beaded support to purify interacting proteins. The difference between these two approaches, though, is that while co-IP uses antibodies to capture protein complexes, pull-down assays use a "bait" protein to purify any proteins in a lysate that bind to the bait. Pull-down assays are ideal for studying strong or stable interactions or those for which no antibody is available for co-immunoprecipitation.

General schematic of a pull-down assay. A pull-down assay is a small-scale affinity purification technique similar to immunoprecipitation, except that the antibody is replaced by some other affinity system. In this case, the affinity system consists of a glutathione S-transferase (GST)–, polyHis- or streptavidin-tagged protein or binding domain that is captured by glutathione-, metal chelate (cobalt or nickel) – or biotin-coated agarose beads, respectively. The immobilized fusion-tagged protein acts as the "bait" to capture a putative binding partner (i.e., the "prey"). In a typical pull-down assay, the immobilized bait protein is incubated with a cell lysate, and after the prescribed washing steps, the complexes are selectively eluted using competitive analytes or low pH or reducing buffers for in-gel or western blot analysis.

Most protein–protein interactions are transient, occurring only briefly as part of a single cascade or other metabolic function within cells. Crosslinking interacting proteins is an approach to stabilize or permanently adjoin the components of interaction complexes. Once the components of an interaction are covalently crosslinked, other steps (e.g., cell lysis, affinity purification, electrophoresis or mass spectrometry) can be used to analyze the protein–protein interaction while maintaining the original interacting complex.

Homobifunctional, amine-reactive crosslinkers can be added to cells to crosslink potentially interacting proteins together, which can then be analyzed after lysis by western blotting. Crosslinkers can be membrane permeable, such as DSS, for crosslinking intracellular proteins, or they can be non–membrane permeable, such as BS3, for crosslinking cell-surface proteins. Furthermore, some crosslinkers can be cleaved by reducing agents, such as DSP or DTSSP, to reverse the crosslinks.

Alternatively, heterobifunctional crosslinkers that contain a photoactivatable group, such as SDA product or Sulfo-SDA, can be used to capture transient interactions that may occur, such as after a particular stimulus. Photoactivation can also be also be after metabolic labeling with photoactivatable amino acids such as L-Photo-Leucine or L-Photo-Methionine.

Crosslinking sites between proteins can be mapped by high precision using mass spectrometry, especially if a MS-cleavable crosslinker such as DSSO or DSBU is used.  

Label transfer involves crosslinking interacting molecules (i.e., bait and prey proteins) with a labeled crosslinking agent and then cleaving the linkage between the bait and prey so that the label remains attached to the prey. This method is particularly valuable because of its ability to identify proteins that interact weakly or transiently with the protein of interest. New non-isotopic reagents and methods continue to make this technique more accessible and simple to perform by any researcher.

Experimental strategy for Sulfo-SBED biotin label transfer and analysis by western blotting.

Just as pull-down assays differ from co-IP in the detection of protein–protein interactions by using tagged proteins instead of antibodies, so is far–western blot analysis different from western blotanalysis, as protein–protein interactions are detected by incubating electrophoresed proteins with a purified, tagged bait protein instead of a target protein-specific antibody, respectively. The term "far" was adopted to emphasize this distinction.

Diagram of far–western blot to analyze protein–protein interactions. In this example, a tagged bait protein is used to probe either the transfer membrane or a gel for the prey protein. Once bound, enzyme (horseradish peroxidase; HRP)-conjugated antibody that targets the bait tag is used to label the interaction, which is then detected by enzymatic chemiluminescence. This general approach can be adjusted by using untagged bait protein that is detected by antibody, biotinylated bait protein that is detected by enzyme-conjugated streptavidin, or radiolabeled bait protein that is detected by exposure to film.

1. Golemis E (2002) Protein-protein interactions: A molecular cloning manual. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press. p ix, 682.
2. Phizicky EM, Fields S (1995) Protein-protein interactions: Methods for detection and analysis. Microbiol Rev 59:94–123.