ARTICLE Protein Interaction Affinity Determination by Quantitative FRET Technology Yang Song, 1 V.G.J. Rodgers, 1 Jerome S. Schultz, 1 Jiayu Liao 1,2 1 Department of Bioengineering, Bourns College of Engineering, University of California at Riverside, 900 University Avenue, Riverside, California 92521 2 Institute for Integrative Genome Biology, University of California at Riverside, 900 University Avenue, Riverside, California 92521; telephone: 951-827-6240; fax: 951-827-6416; e-mail: jiayu.liao@ucr.edu ABSTRACT: The dissociation constant, K d , is an important parameter for characterizing protein–protein interaction affinities. SUMOylation is one of the important protein post-translational modifications and it involves a multi- step enzymatic cascade reaction, resulting in peptide activa- tion and substrate conjugation. Multiple covalent and non- covalent protein–protein interactions are involved in this cascade. Techniques involving Fo ¨rster resonance energy transfer (FRET) have been widely used in biological studies in vitro and in vivo, and they are very powerful tools for elucidating protein interactions in many regulatory cas- cades. In our previous studies, we reported the attempt to develop a new method for the determination of the K d by FRET assay using the interaction of SUMO1 and its E2 ligase, Ubc9 as a test system. However, the generality and specifications of this new method have not been fully determined. Here we report a systematic approach for determining the dissociation constant (K d ) in the SUMOy- lation cascade and for further sensitivity and accuracy testing by the FRET technology. From a FRET donor to acceptor concentration ratio range of 4–40, the K d s of SUMO1 and Ubc9 consistently agree well with values from surface plas- mon resonance and isothermal titration calorimetry. These results demonstrate the high sensitivity and accuracy of the FRET-based K d determination approach. This technology, therefore, can be used in general for protein–protein inter- action dissociation constant determination. Biotechnol. Bioeng. 2012;109: 2875–2883. ß 2012 Wiley Periodicals, Inc. KEYWORDS: fluorescent proteins; quantitative FRET; Kd determination; SUMOylation Introduction Protein–protein interactions have pivotal roles in most physiology processes. In the post-genomic era, genome- wide studies of protein–protein interactions and structure have effectively identified physically interactive players and potential drug targets for various human diseases (Wells and McClendon, 2007). One of most important parameters for describing the protein–protein interactions affinities is the dissociation constant K d (Gilson and Zhou, 2007). K d has been determined by many techniques, including surface plasmon resonance (SPR) (Boozer et al., 2006; Homola, 2008), isothermal titration calorimetry (ITC) (Freire et al., 1990; Perozzo et al., 2004), and radio-ligand binding assay (Motulski and Christopoulos, 2003). Post-translational modifications by ubiquitin (Ub) and ubiquitin-like (Ubl) proteins, such as SUMO, are important mechanisms that regulate protein activities and half-lives in eukaryotes (Johnson, 2004; Melchior, 2000; Muller et al., 2001). The conjugation of Ub, SUMO, and other Ubls to target substrates involves an evolutionarily conserved but distinct enzyme cascade of actions by the E1, E2, and E3 ligases (Melchior et al., 2003). E1s activate Ub/Ubls in two major steps: adenylation of the Ub/Ubls at their C-termini by ATP and covalent thioester bond formation between the catalytic cysteine of E1 and the C-termini of Ub/Ubls by replacing the adenylate group. The activated Ubls are then transferred by a transthiolation reaction to E2, resulting in formation of a thioester bond between E2 catalytic cysteine and Ub/Ubls. A variety of E3s, which are generally considered to determine substrate specificity in vivo, finally conjugate the Ub/Ubls to lysine residues in the substrates. We are interested in understanding the mechanisms of sumoylation/ubiquitination cascade because multiple pro- tein–protein interactions are involved in the sequential transfer cascade (Knipscheer and Sixma, 2007). Among these interactions, the SUMO peptide has an intrinsic non- covalent affinity with its E2 ligase, Ubc9, and this interaction occurs without the activation cascade (Knipscheer and Sixma, 2007). It is important to understand the molecular The authors declare that they have no competing financial interests. Correspondence to: J. Liao Contract grant sponsor: National Institutes of Health Contract grant number: 5 R01 AI076504 Additional supporting information may be found in the online version of this article. Received 22 March 2012; Revision revised 27 April 2012; Accepted 15 May 2012 Article first published online 18 June 2012 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1002/bit.24564/abstract) DOI 10.1002/bit.24564 ß 2012 Wiley Periodicals, Inc. Biotechnology and Bioengineering, Vol. 109, No. 11, November, 2012 2875