Production and glycan binding characterization of human properdin and structural elucidation of c-Jun N-terminal kinase 3 inhibitors

Abstract

Properdin, the only known positive regulator of the alternative complement pathway (AP), has been a focus of controversy since its discovery by Louis Pillemer in 1954 [1], [2]. In the last years evidence accumulated that this glycoprotein acts as a pattern recognition molecule and initiates the AP upon recognition of specific glycan markers, apart from its established role in the stabilization of AP convertases [3]. In this work, human full-length properdin was produced in mammalian cells and the purified protein was found to be active and to form the previously reported dimeric, trimeric and tetrameric cyclic structures. STD-NMR experiments performed with a set of glycans suggested that negative charge is required for binding to properdin and that glycosaminoglycans (GAGs) are potential pathogen-associated molecular patterns (PAMPs) for properdin and may mediate direct AP activation by properdin. The results also confirmed reports that both positive and negative regulators of the AP, namely properdin and complement Factor H, bind to different epitopes on identical glycans. The structural complexity that contributes to the diversity of GAGs and glycans represents a significant challenge for their isolation for functional and structural studies. A common approach involves the use of GAG-depolymerizing enzymes, such as heparinase I from Pedobacter heparinus, followed by fractionation of the obtained oligosaccharides [4], [5]. However, since the most biological relevant oligosaccharides have a higher degree of polymerization than those usually obtained from heparinase, a structure-based engineering approach to rationally design this enzyme could alter the product distribution. Although recombinant heparinase I could be produced in E. coli and expression constructs with varying N- and C-terminal sequences were tested, an intrinsic low thermal and conformational stability was observed and no protein crystals were obtained for structure determination by X-ray crystallography and subsequent enzyme engineering. Over the past decades, inhibition of the c-Jun N-terminal kinase 3 (JNK3), a mitogen-activated protein kinase (MAPK) involved in the regulation of cellular responses to extracellular stimuli has become a promising strategy for treatment of neurodegenerative disorders such as Alzheimer’s and Parkinson’s diseases. However, up to date, no inhibitors targeting JNK3 have been approved by the FDA [6], [7]. Altering the substitution pattern of a pyridinylimidazole scaffold from a dual p38α/JNK3 MAPK inhibitor proved to be effective in shifting the selectivity towards JNK3 [8]. A similar binding mode of the two most potent inhibitors with an IC50 value <1 μM in the ATP binding pocket was confirmed by X-ray crystallography. While selectivity was achieved by addressing the hydrophobic region I of JNK3 with a small methyl group, addition of a S-methyl group contributed to the stability of the G-rich loop of JNK3, thus increasing the inhibitory potency. Future strategies to increase the inhibitory potency while preserving selectivity were devised from the determined JNK3 crystal structures.