Lessons From Cyclosporine A: Structural Determinants of Conformation-Switching and Passive Membrane Penetration

Abstract

The structural complexity of ‘beyond-rule-of-5’ compounds, such as peptide macrocycles, may facilitate access to additional biological target space beyond the enzymatic active site. Naturally occurring cyclic peptides, in particular, exhibit a wide variety of unusual and potent biological activities. Cyclosporine A (CsA), the canonical ‘rule-breaking’ molecule, is heavily N-methylated, and it is widely held that this contributes to its unexpectedly high oral bioavailability. This has been a major motivator of the use of N-methylation to enhance the ability of cyclic peptides to cross lipid bilayers. The evidentiary basis of this, stems largely from structural studies which predict that all the backbone amide bonds of CsA that are not engaged in transannular hydrogen bonds in the suspected ‘cell-permeable conformation,’ are otherwise N-methylated. However, to our knowledge, no direct, systematic analysis of the effect of N-methylation on cell permeability and target engagement of CsA has been reported. In order to gain a deeper understanding of this structurally-significant molecule, we have engineered a new tool for the study CypA ligands. A destabilized variant of CsA presenting protein, Cyclophilin A (CypA) that persists in the cell only when bound to a stabilizing ligand, was used to evaluate the target binding and cell permeability of selected CsA anlaogues. Subsequent structure-activity studies show that seemingly small structural changes, such as the loss of a single N-methyl group or the introduction of a less conformationally-constrained side chain, can result in large conformational changes that alter CsA’s ability to engage CypA. We also propose that this tool may be used to identify or evaluate novel CypA-binding scaffolds with altered target specificity.