De novo Design of G Protein-Coupled Receptor 40 Peptide Agonists for Type 2 Diabetes Mellitus Based on Artificial Intelligence and Site-Directed Mutagenesis

Xu Chen,Shengbin Li,Daiyun Xu,Junqing Wang,Yongxiao Li,Yonghui Lyu,Zhidong Chen,Zhe Wang

doi:10.3389/fbioe.2021.694100

Abstract

G protein-coupled receptor 40 (GPR40), one of the G protein-coupled receptors that are available to sense glucose metabolism, is an attractive target for the treatment of type 2 diabetes mellitus (T2DM). Despite many efforts having been made to discover small-molecule agonists, there is limited research focus on developing peptides acting as GPR40 agonists to treat T2DM. Here, we propose a novel strategy for peptide design to generate and determine potential peptide agonists against GPR40 efficiently. A molecular fingerprint similarity (MFS) model combined with a deep neural network (DNN) and convolutional neural network was applied to predict the activity of peptides constructed by unnatural amino acids (UAAs). Site-directed mutagenesis (SDM) further optimized the peptides to form specific favorable interactions, and subsequent flexible docking showed the details of the binding mechanism between peptides and GPR40. Molecular dynamics (MD) simulations further verified the stability of the peptide–protein complex. The R-square of the machine learning model on the training set and the test set reached 0.87 and 0.75, respectively; and the three candidate peptides showed excellent performance. The strategy based on machine learning and SDM successfully searched for an optimal design with desirable activity comparable with the model agonist in phase III clinical trials.

Highlights

Type 2 diabetes mellitus (T2DM) is a degenerative disease caused by impairment in insulin action and pancreatic β-cell function, characterized by the inability to maintain glucose homeostasis
To obtain high-precision activity prediction results, convolution neural network (CNN) and deep neural network (DNN) were constructed to compare their fitness with the task
Both neural networks were trained based on all molecules in the data set

Summary

Introduction

Type 2 diabetes mellitus (T2DM) is a degenerative disease caused by impairment in insulin action and pancreatic β-cell function, characterized by the inability to maintain glucose homeostasis. The general objective of all therapeutic modalities for T2DM is to decrease the circulating blood glucose levels. The mechanisms of action involve (i) insulin receptor ligands (insulin analogs), (ii) reduction of insulin resistance (biguanides and thiazolidinediones), GPR40 Peptide Agonists Design Rationale (iii) stimulation of β-cells by insulin secretagogues (sulfonylureas and meglitinides), (iv) lowering postprandial blood glucose level via alpha-glucosidase inhibitors (acarbose and miglitol), and (v) blocking renal glucose reabsorption via a sodium-glucose cotransporter-2 (SGLT2) inhibitor (dapagliflozin) (Chatterjee et al, 2017). Most treatments can manage glucose levels in T2DM patients, the progressive decline in β-cell function leads to inevitable dependence on exogenous insulin supply. The novel mechanism and potent candidates should draw our attention in the treatment of T2DM (Chen et al, 2016)

Methods

Results

Conclusion