One popular method to elucidate protein-protein interactions involves the native co-purification of an affinity-tagged protein and its interacting partners, which are subsequently identified through mass spectrometry (MS) (1). Although straightforward, reproducible, and broadly used, this strategy is hampered by the efficacy of protein recoveries both in terms of sensitivity and specificity. This is especially pertinent to methodologies that use a single-step purification, in which suboptimal enrichment of the bait protein and its partners over background can lead to masking of their signals. Although improvements to MS instrumentation generally increase peptide detection sensitivities, the problem of specificity (i.e., distinguishing specific from nonspecific interacting proteins) remains. Thus ultimately, the limiting factor in the identification of specific interacting proteins lies with the purification itself.

An effort to resolve this specificity issue has been made with the introduction of the tandem affinity purification (TAP) tag. This construct consists of an immunoglobulin G (IgG) binding domain and calmodulin binding peptide domain separated by a tobacco etch virus (TEV) protease cleavage site (2). The TAP method was originally developed in yeast and has best demonstrated its utility in the systematic identification of numerous multiprotein complexes in the yeast proteome (3). Although modifications to the original TAP methodology have been successful in examining the protein networks of mammalian cells (3,4,5,6,7), the strategy offers a relatively low yield of bait and specific interacting proteins (8), and the success rate usually varies on a case-by-case basis. Additionally, problems remain that are inherent to any protein tagging strategy: (i) variable exposure of the affinity tag; (ii) disruption of the bait protein's ability to fold properly; (iii) steric exclusion of interacting partners; and (iv) ectopic overexpression of the fusion protein, which can lead to complications in both the purification and identification of true interactions.

We generated five novel dual-tag purification vectors, each with a different combination of affinity tags (two per construct, varying by composition, size, and terminal location) and including either a constitutive (CMV_p) or tetracycline-regulatable promoter (Tet_p) to allow for controlled expression of the tagged bait proteins (Figure 1A). We chose Strep-Tactin binding peptide (StrepII-tag®; IBA, St. Louis, MO, USA) in most of our dual-tag combinations due to both its high binding affinity and its small (8-mer) size compared with the original streptavidin binding peptide (strep tag: 38-mer). Other novel features include a second TEV protease recognition site to improve cleavage efficiency (data not shown) and a tetracysteine motif (CCPGCC) (9) (except for CHAtP) to easily monitor bait protein expression, purification, and localization. Moreover, all our dual-tags are constructed in Invitrogen Gateway®-compatible destination vectors, allowing for easy cloning through site-specific recombination (see the supplementary materials available online at www.BioTechniques.com).

Figure 1. Features and validation of the dual-tags. (Click to enlarge)

We selected human telomeric repeat binding factor 2 (TRF2) to evaluate our dual-tagging system. TRF2 is a key telomere binding protein that functions to stabilize the t-loop configuration, a structure that both protects the chromosome end from being recognized as damaged DNA and represses telomere elongation by telomerase (10). Several telomere-associated and DNA damage repair proteins are known to interact with TRF2 (10,11) and thus provides both an effective means to assess the efficacy of our tagging system and an opportunity to gather potentially novel insight regarding TRF2 function.

As shown in Figure 1, B and F, all TRF2 fusions produced proteins of anticipated size (or very nearly so). Moreover, the level of Tet_p-driven TRF2 fusion protein expression was tetracycline concentration-dependent (Figure 1C). This demonstrates the capability to modulate fusion protein expression, potentially overcoming problems encountered in overexpression systems. For example, not only can expression be adjusted to near physiological level, but also, bait protein that would otherwise impede cellular growth and/or viability can be repressed until specific experimental conditions are met. This feature greatly expands the tags’ applicability.

The CCPGCC motif featured in four out of five dual-tags allows for the visualization of the fusion protein in both live cells and cellular lysates using Lumio™ (Invitrogen, Carlsbad, CA, USA), a conditionally fluorescent, membrane-permeable compound based on the fluorescein arsenical hairpin (FIAsH) reagent (9,12). The expected co-localization of TRF2-C-StH with the telomere in fixed cells (Figure 1D) and its similar punctate staining pattern in live cells (Figure 1E) indicate that the tag does not interfere with TRF2's subcellular localization. This useful feature provides a means to (i) rapidly infer bait protein function following tagging based on proper localization; (ii) assess transfection efficiency; (iii) confirm putative interacting partners by co-localization; or (iv) monitor the purification progress directly by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Figure 1F). Interestingly, the tetracysteine motif can also function as an affinity tag when paired with F1AsH-conjugated agarose beads (12), providing yet another means by which the bait-complex could be purified.