Computational Competitive and Negative Design To Derive a Specific cJun Antagonist

Abstract

Basic leucine zipper (bZIP) proteins reside at the end of cell-signaling cascades and function to modulate transcription of specific gene targets. bZIPs are recognized as important regulators of cellular processes such as cell growth, apoptosis, and cell differentiation. One such validated transcriptional regulator, activator protein-1, is typically comprised of heterodimers of Jun and Fos family members and is key in the progression and development of a number of different diseases. The best described component, cJun, is upregulated in a variety of diseases such as cancer, osteoporosis, and psoriasis. Toward our goal of inhibiting bZIP proteins implicated in disease pathways, we here describe the first use of a novel in silico peptide library screening platform that facilitates the derivation of sequences exhibiting a high affinity for cJun while disfavoring homodimer formation or formation of heterodimers with other closely related Fos sequences. In particular, using Fos as a template, we have computationally screened a peptide library of more than 60 million members and ranked hypothetical on/off target complexes according to predicted stability. This resulted in the identification of a sequence that bound cJun but displayed little homomeric stability or preference for cFos. The computationally selected sequence maintains an interaction stability similar to that of a previous experimentally derived cJun antagonist while providing much improved specificity. Our study provides new insight into the use of tandem in silico screening/ in vitro validation and the ability to create a peptide that is capable of satisfying conflicting design requirements.