Multistep Modeling Strategy To Improve the Binding Affinity Prediction of PET Tracers to Aβ42: Case Study with Styrylbenzoxazole Derivatives

Artificial Intelligence and the Future of Diagnostic and Therapeutic Radiopharmaceutical Development:: In Silico Smart Molecular Design

Novel diagnostic and therapeutic radiopharmaceuticals are increasingly becoming a central part of personalized medicine. Continued innovation in the development of new radiopharmaceuticals is key to sustained growth and advancement of precision medicine. Artificial intelligence has been used in multiple fields of medicine to develop and validate better tools for patient diagnosis and therapy, including in radiopharmaceutical design. In this review, we first discuss common in silico approaches and focus on their usefulness and challenges in radiopharmaceutical development. Next, we discuss the practical applications of in silico modeling in design of radiopharmaceuticals in various diseases.

Multidisciplinary approaches for targeting the secretase protein family as a therapeutic route for Alzheimer's disease

The continual increase of the aging population worldwide renders Alzheimer's disease (AD) a global prime concern. Several attempts have been focused on understanding the intricate complexity of the disease's development along with the on- andgoing search for novel therapeutic strategies. Incapability of existing AD drugs to effectively modulate the pathogenesis or to delay the progression of the disease leads to a shift in the paradigm of AD drug discovery. Efforts aimed at identifying AD drugs have mostly focused on the development of disease-modifying agents in which effects are believed to be long lasting. Of particular note, the secretase enzymes, a group of proteases responsible for the metabolism of the β-amyloid precursor protein (βAPP) and β-amyloid (Aβ) peptides production, have been underlined for their promising therapeutic potential. This review article attempts to comprehensively cover aspects related to the identification and use of drugs targeting the secretase enzymes. Particularly, the roles of secretases in the pathogenesis of AD and their therapeutic modulation are provided herein. Moreover, an overview of the drug development process and the contribution of computational (in silico) approaches for facilitating successful drug discovery are also highlighted along with examples of relevant computational works. Promising chemical scaffolds, inhibitors, and modulators against each class of secretases are also summarized herein. Additionally, multitarget secretase modulators are also taken into consideration in light of the current growing interest in the polypharmacology of complex diseases. Finally, challenging issues and future outlook relevant to the discovery of drugs targeting secretases are also discussed.

Core Binding Site of a Thioflavin-T-Derived Imaging Probe on Amyloid β Fibrils Predicted by Computational Methods

Development of new diagnostic imaging probes for Alzheimer's disease, such as positron emission tomography (PET) and single photon emission computed tomography (SPECT) probes, has been strongly desired. In this study, we investigated the most accessible amyloid β (Aβ) binding site of [¹²³I]IMPY, a Thioflavin-T-derived SPECT probe, using experimental and computational methods. First, we performed a competitive inhibition assay with Orange-G, which recognizes the KLVFFA region in Aβ fibrils, suggesting that IMPY and Orange-G bind to different sites in Aβ fibrils. Next, we precisely predicted the IMPY binding site on a multiple-protofilament Aβ fibril model using computational approaches, consisting of molecular dynamics and docking simulations. We generated possible IMPY-binding structures using docking simulations to identify candidates for probe-binding sites. The binding free energy of IMPY with the Aβ fibril was calculated by a free energy simulation method, MP-CAFEE. These computational results suggest that IMPY preferentially binds to an interfacial pocket located between two protofilaments and is stabilized mainly through hydrophobic interactions. Finally, our computational approach was validated by comparing it with the experimental results. The present study demonstrates the possibility of computational approaches to screen new PET/SPECT probes for Aβ imaging.

Effect of Alzheimer Familial Chromosomal Mutations on the Amyloid Fibril Interaction with Different PET Tracers: Insight from Molecular Modeling Studies

Alzheimer's disease (AD) is the most common neurodegenerative disorder. Along with an increasing number of elderly worldwide, it poses a great challenge for the society and health care. Although sporadic AD is the common form of AD, 2-3% of the AD cases are expected to be due to mutations in the β region of the amyloid precursor protein, which is referred to as autosomal dominant AD (ADAD). These mutations may cause changes in the secondary structure of the amyloid β fibrils and may alter the fibrillization rate leading to changes in the disease development and could also affect the binding to tracers used in diagnosis. In particular, from some recent clinical studies using PET tracers for detection of fibrillar amyloids, it is evident that in ADAD patients with Arctic mutation no amyloid plaque binding can be detected with the ¹¹C-Pittsburgh Compound B (¹¹C-PIB). However, for in vitro conditions, significant binding of ³H-PIB has been reported for the amyloid fibrils carrying the Arctic mutation. The aim of the present study is to investigate if there is any mutation specific binding of commonly used amyloid tracers, namely, florbetaben, florbetapir, FPIB, AZD4694, and AZD2184, by means of molecular modeling techniques. Other than Arctic, ADAD mutations, such as the Dutch, Italian, Iowa, and Flemish mutations, are considered in this study. We report that all tracers except florbetapir show reduced binding affinity toward amyloid β fibrils with the Arctic mutation when compared to the native type. Moreover, florbetapir is the only tracer that binds to all mutants with increased affinity when compared to the native fibril. The results obtained from these studies could increase the understanding of the structural changes caused by mutation and concomitant changes in the interaction pattern of the PET tracers with the mutated variants, which in turn can be useful in selecting the appropriate tracers for the purpose of diagnosis as well as for designing new tracers with desirable properties.