Combined molecular graph neural network and structural docking selects potent programmable cell death protein 1/programmable death-ligand 1 (PD-1/PD-L1) small molecule inhibitors

The Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD-1/PD-L1) interaction is an immune checkpoint utilized by cancer cells to enhance immune suppression. There exists a huge need to develop small molecules drugs that are fast acting, cheap, and readily bioavailable compared to antibodies. Unfortunately, synthesizing and validating large libraries of small-molecule to inhibit PD-1/PD-L1 interaction in a blind manner is a both time-consuming and expensive. To improve this drug discovery pipeline, we have developed a machine learning methodology trained on patent data to identify, synthesize and validate PD-1/PD-L1 small molecule inhibitors. Our model incorporates two features: docking scores to represent the energy of binding (E) as a global feature and sub-graph features through a graph neural network (GNN) to represent local features. This Energy-Graph Neural Network (EGNN) model outperforms traditional machine learning methods as well as a simple GNN with an average F1 score of 0.997 (± 0.004) suggesting that the topology of the small molecule, the structural interaction in the binding pocket, and chemical diversity of the training data are all important considerations for enhancing model performance. A Bootstrapped EGNN model was used to select compounds for synthesis and experimental validation with predicted high and low potency to inhibit PD-1/PD-L1 interaction. The new potent inhibitor, (4-((3-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)-2-methylbenzyl)oxy)-2,6-dimethoxybenzyl)-D-serine, is a hybrid of two known bioactive scaffolds, and has an IC50 values of 339.9 nM that is comparatively better than the known bioactive compound. We conclude that our EGNN model can identify active molecules designed by scaffold hopping, a well-known medicinal chemistry technique and will be useful to identify new potent small molecule inhibitors for specific targets.

Download Full-text

Combined Molecular Graph Neural Network and Structural Docking Selects Potent Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD-1/PD-L1) Small Molecule Inhibitors

10.26434/chemrxiv.12083907 ◽

2020 ◽

Author(s):

Prageeth R. Wijewardhane ◽

Krupal P. Jethava ◽

Jonathan A Fine ◽

Gaurav Chopra

Keyword(s):

Neural Network ◽

Machine Learning ◽

Cell Death ◽

Small Molecule ◽

Small Molecule Inhibitors ◽

Molecular Graph ◽

Model Performance ◽

Chemical Diversity ◽

Training Data ◽

Cell Death Protein

The Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD-1/PD-L1) interaction is an immune checkpoint utilized by cancer cells to enhance immune suppression. There exists a huge need to develop small molecules drugs that are fast acting, cheap, and readily bioavailable compared to antibodies. Unfortunately, synthesizing and validating large libraries of small-molecule to inhibit PD-1/PD-L1 interaction in a blind manner is a both time-consuming and expensive. To improve this drug discovery pipeline, we have developed a machine learning methodology trained on patent data to identify, synthesize and validate PD-1/PD-L1 small molecule inhibitors. Our model incorporates two features: docking scores to represent the energy of binding (E) as a global feature and sub-graph features through a graph neural network (GNN) to represent local features. This Energy-Graph Neural Network (EGNN) model outperforms traditional machine learning methods as well as a simple GNN with an average F1 score of 0.997 (± 0.004) suggesting that the topology of the small molecule, the structural interaction in the binding pocket, and chemical diversity of the training data are all important considerations for enhancing model performance. A Bootstrapped EGNN model was used to select compounds for synthesis and experimental validation with predicted high and low potency to inhibit PD-1/PD-L1 interaction. The new potent inhibitor, (4-((3-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)-2-methylbenzyl)oxy)-2,6-dimethoxybenzyl)-D-serine, is a hybrid of two known bioactive scaffolds, and has an IC50 values of 339.9 nM that is comparatively better than the known bioactive compound. We conclude that our EGNN model can identify active molecules designed by scaffold hopping, a well-known medicinal chemistry technique and will be useful to identify new potent small molecule inhibitors for specific targets.

Download Full-text

Recent progress in small-molecule inhibitors for critical therapeutic targets of necroptosis

Future Medicinal Chemistry ◽

10.4155/fmc-2020-0386 ◽

2021 ◽

Author(s):

Zhennan Fang ◽

Huiqiang Wei ◽

Wenfeng Gou ◽

Leyuan Chen ◽

Changfen Bi ◽

...

Keyword(s):

Cell Death ◽

Small Molecule ◽

Future Development ◽

Recent Progress ◽

Small Molecule Inhibitors ◽

Cell Necrosis ◽

Significant Progress ◽

Regulated Cell Death ◽

Associated Proteins ◽

Widespread Interest

Nonapoptotic types of regulated cell death have attracted widespread interest since the discovery that certain forms of cell necrosis can be regulated. In particular, research into cell necroptosis has made significant progress in connection with kidney, inflammatory, degenerative and neoplastic diseases. Inhibitors targeting the critical necroptosis-associated proteins RIPK1/3 and MLKL have been in development for more than a decade. Herein the authors compile a list of the known small-molecule inhibitors of these enzymes and representative structures of compounds co-crystallized with these proteins and put forward some thoughts regarding their future development.

Download Full-text

Small-Molecule Inhibitors of the Programmed Cell Death-1/Programmed Death-Ligand 1 (PD-1/PD-L1) Interaction via Transiently Induced Protein States and Dimerization of PD-L1

Journal of Medicinal Chemistry ◽

10.1021/acs.jmedchem.7b00293 ◽

2017 ◽

Vol 60 (13) ◽

pp. 5857-5867 ◽

Cited By ~ 86

Author(s):

Katarzyna Guzik ◽

Krzysztof M. Zak ◽

Przemyslaw Grudnik ◽

Katarzyna Magiera ◽

Bogdan Musielak ◽

...

Keyword(s):

Cell Death ◽

Programmed Cell Death ◽

Small Molecule ◽

Small Molecule Inhibitors ◽

Programmed Death Ligand 1 ◽

Programmed Death ◽

Programmed Cell Death 1

Download Full-text

Discovery of quinazoline derivatives as novel small-molecule inhibitors targeting the programmed cell death-1/programmed cell death-ligand 1 (PD-1/PD-L1) interaction

European Journal of Medicinal Chemistry ◽

10.1016/j.ejmech.2021.113998 ◽

2021 ◽

pp. 113998

Author(s):

Yu Wang ◽

Kun huang ◽

Yali Gao ◽

Dandan Yuan ◽

Lin Ling ◽

...

Keyword(s):

Cell Death ◽

Programmed Cell Death ◽

Small Molecule ◽

Small Molecule Inhibitors ◽

Programmed Cell Death 1 ◽

Quinazoline Derivatives

Download Full-text

Cell Death: Biological Mechanisms and Small Molecule Inhibitors

Wiley Encyclopedia of Chemical Biology ◽

10.1002/9780470048672.wecb059 ◽

2008 ◽

Author(s):

Olga Korkina ◽

Alexei Degterev

Keyword(s):

Cell Death ◽

Small Molecule ◽

Small Molecule Inhibitors ◽

Biological Mechanisms

Download Full-text

A phenolic small molecule inhibitor of RNase L prevents cell death from ADAR1 deficiency

Proceedings of the National Academy of Sciences ◽

10.1073/pnas.2006883117 ◽

2020 ◽

Vol 117 (40) ◽

pp. 24802-24812 ◽

Cited By ~ 1

Author(s):

Salima Daou ◽

Manisha Talukdar ◽

Jinle Tang ◽

Beihua Dong ◽

Shuvojit Banerjee ◽

...

Keyword(s):

Cell Death ◽

Protein Kinase ◽

Small Molecule ◽

Small Molecule Inhibitors ◽

Therapeutic Potential ◽

Protein Kinase Inhibitors ◽

Response To Treatment ◽

Rnase L ◽

Oligoadenylate Synthetase ◽

In Cells

The oligoadenylate synthetase (OAS)–RNase L system is an IFN-inducible antiviral pathway activated by viral infection. Viral double-stranded (ds) RNA activates OAS isoforms that synthesize the second messenger 2-5A, which binds and activates the pseudokinase-endoribonuclease RNase L. In cells, OAS activation is tamped down by ADAR1, an adenosine deaminase that destabilizes dsRNA. Mutation of ADAR1 is one cause of Aicardi-Goutières syndrome (AGS), an interferonopathy in children. ADAR1 deficiency in human cells can lead to RNase L activation and subsequent cell death. To evaluate RNase L as a possible therapeutic target for AGS, we sought to identify small-molecule inhibitors of RNase L. A 500-compound library of protein kinase inhibitors was screened for modulators of RNase L activity in vitro. We identified ellagic acid (EA) as a hit with 10-fold higher selectivity against RNase L compared with its nearest paralog, IRE1. SAR analysis identified valoneic acid dilactone (VAL) as a superior inhibitor of RNase L, with 100-fold selectivity over IRE1. Mechanism-of-action analysis indicated that EA and VAL do not bind to the pseudokinase domain of RNase L despite acting as ATP competitive inhibitors of the protein kinase CK2. VAL is nontoxic and functional in cells, although with a 1,000-fold decrease in potency, as measured by RNA cleavage activity in response to treatment with dsRNA activator or by rescue of cell lethality resulting from self dsRNA induced by ADAR1 deficiency. These studies lay the foundation for understanding novel modes of regulating RNase L function using small-molecule inhibitors and avenues of therapeutic potential.

Download Full-text

PD-1-Targeted Discovery of Peptide Inhibitors by Virtual Screening, Molecular Dynamics Simulation, and Surface Plasmon Resonance

Molecules ◽

10.3390/molecules24203784 ◽

2019 ◽

Vol 24 (20) ◽

pp. 3784 ◽

Cited By ~ 2

Author(s):

Yuanqiang Wang ◽

Haiqiong Guo ◽

Zhiwei Feng ◽

Siyi Wang ◽

Yuxuan Wang ◽

...

Keyword(s):

Molecular Dynamics ◽

Cell Death ◽

Surface Plasmon Resonance ◽

Virtual Screening ◽

Programmed Cell Death ◽

Surface Plasmon ◽

Small Molecule ◽

Plasmon Resonance ◽

Small Molecule Inhibitors ◽

Dynamics Simulation

The blockade of the programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1) pathway plays a critical role in cancer immunotherapy by reducing the immune escape. Five monoclonal antibodies that antagonized PD-1/PD-L1 interaction have been approved by the Food and Drug Administration (FDA) and marketed as immunotherapy for cancer treatment. However, some weaknesses of antibodies, such as high cost, low stability, poor amenability for oral administration, and immunogenicity, should not be overlooked. To overcome these disadvantages, small-molecule inhibitors targeting PD-L1 were developed. In the present work, we applied in silico and in vitro approaches to develop short peptides targeting PD-1 as chemical probes for the inhibition of PD-1–PD-L1 interaction. We first predicted the potential binding pocket on PD-1/PD-L1 protein–protein interface (PPI). Sequentially, we carried out virtual screening against our in-house peptide library to identify potential ligands. WANG-003, WANG-004, and WANG-005, three of our in-house peptides, were predicted to bind to PD-1 with promising docking scores. Next, we conducted molecular docking and molecular dynamics (MD) simulation for the further analysis of interactions between our peptides and PD-1. Finally, we evaluated the affinity between peptides and PD-1 by surface plasmon resonance (SPR) binding technology. The present study provides a new perspective for the development of PD-1 inhibitors that disrupt PD-1–PD-L1 interactions. These promising peptides have the potential to be utilized as a novel chemical probe for further studies, as well as providing a foundation for further designs of potent small-molecule inhibitors targeting PD-1.

Download Full-text