J.A. Hey, P. Kocis, S. Abushakra, J. Yu, A. Power, K. Blennow, M. Tolar
Abstract

Background
Beta amyloid (Aβ) oligomers play a critical role in the pathogenesis of Alzheimer’s disease (AD), and represent a promising target for drug development. Tramiprosate is a small-molecule Aß anti-aggregation agent that was previously evaluated in Phase 3 clinical trials for AD but did not meet the primary efficacy endpoints; however, a pre-specified subgroup analysis revealed robust, clinically meaningful cognitive and functional effects in APOE4/4 homozygous patients (Abushakra 2016, 2017). To advance this important efficacy attribute and to further improve its pharmaceutical properties, we have developed ALZ-801 a clinical stage prodrug of tramiprosate with substantially improved pharmaceutical properties for treatment of Alzheimer’s disease (AD). Here we characterize a multi-ligand enveloping effect of tramiprosate on Aβ42, which stabilizes Aβ monomers, and underpins anti-aggregation activity (Kocis et al. 2017). To select the ALZ-801 dose for the pivotal clinical trials of ALZ-801, we performed an integrated analysis that links tramiprosate exposures in the brain and its clinical effects in prior trials, to the anti-Aβ oligomer effect. Bridging human pharmacokinetic studies between tramiprosate and ALZ-801 are presented.

Methods:
Ion mobility mass spectrometry
To address the high conformational flexibility of Aβ42 and characterize its interaction with tramiprosate, we used ion mobility mass spectrometry (IMS MS), using Waters Synapt G2-quadrupole time of flight mass spectrometer (Q-TOF MS) with traveling wave ion mobility. IMS MS is a technique capable of separating molecular ions based on their size and conformation and can characterize the stoichiometry of ligand-protein complexes.

Nuclear magnetic resonance
To determine how tramiprosate binds to the target we used two-dimensional heteronuclear multiple quantum correlation nuclear magnetic resonance spectroscopy (2D 1H-15N HMQC NMR) of uniformly 15N-labeled Aβ42 peptide (in 90% H2O/10% D2O sodium phosphate buffer, pH 7.4 at 37°C). The NMR experiments were conducted at 800 MHz on a Bruker AVANCE II spectrometer using a 5mm HCN cryogenic probe. 2D1H-15N SOFAST-HMQC software with 3919 Watergate was used.

Molecular dynamics
To characterize the structure of Aβ42 alone and with different levels of excess tramiprosate, we conducted a series of all atom molecular dynamics simulations. All molecular modeling was performed using the Schrödinger suite. Molecular dynamics simulations were run using Desmond on GeForce GTX Titan Black GPU cards. To describe the large conformational changes observed in these simulations, we performed a principal component (PC) analysis of the free energy surface.

Human pharmacokinetic analyses, CSF Aβ42 measures, and pharmacokinetic-pharmacodynamic (PK-PD) translation
Plasma and CSF concentrations of tramiprosate at 78 weeks in the previous Phase 3 studies were determined in frozen samples using a validated LC-MS/MS method. The steady-state drug level in the brain was projected based on the brain/plasma drug exposure relationship from animals to humans. The CSF Aβ42 concentrations in AD patients in the tramiprosate Phase 2 trial were measured by ELISA, and was used in the PK-PD analyses.

Results:
We observed a multi-ligand interaction of tramiprosate with monomeric Aβ42, which differs from traditional 1:1 binding. This results in the stabilization of Aβ42 monomers and inhibition of oligomer formation and elongation, as demonstrated by IMS MS and molecular dynamics. Using NMR spectroscopy and molecular dynamics, we also showed that tramiprosate bound to Lys16, Lys28 and Asp23, the key amino acid side chains of Aβ42 that are responsible for both conformational seed formation and neurotoxicity. The projected molar excess of tramiprosate versus Aβ42 in humans using the dose effective in APOE4/4 homozygotes aligns with the molecular stoichiometry of the interaction providing a clear clinical translation of the mechanism of action.

Conclusions:
Aβ oligomers are considered a key driver of synaptic dysfunction and toxicity in AD. In summary, we have identified the molecular mechanism that may account for the observed clinical efficacy of tramiprosate in APOE4/4 homozygous AD patients. In addition, the integrated application of the molecular methodologies (i.e., IMS MS, NMR, and thermodynamics analysis) shows that it is possible to modulate and control the Aβ42 conformational dynamics landscape by a small molecule, and cause the Aβ42 peptide to alter its conformation, leading to a clinically relevant anti-aggregation action.

The 265mg dose of ALZ-801 determined from the preclinical to clinical translational PK/PD analyses is projected to provide tramiprosate plasma exposure that is bioequivalent to the clinically effective dose in completed Phase 3 studies, and to fully inhibit Aβ42 oligomer formation in the brain. This ALZ-801 dose will now be evaluated in an upcoming confirmatory pivotal program in APOE4/4 homozygous AD patients. Furthermore, the novel enveloping mechanism of action of tramiprosate described herein has potential utility in the development of disease-modifying therapies for AD and related neurodegenerative disorders caused by misfolded, prion-like proteins.