Exploring the AlphaFold Predicted Conformational Properties of Human Diacylglycerol Kinases.
Sukhvershjit S AulakhJosé Carlos BozelliRichard M EpandPublished in: The journal of physical chemistry. B (2022)
Diacylglycerol kinases (DGKs) are important enzymes in molecular membrane biology, as they can lower the concentration of diacylglycerol through phosphorylation while at the same time producing phosphatidic acid. Dysfunction of DGK is linked with multiple diseases including cancer and autoimmune disorders. Currently, the high-resolution structures have not been determined for any of the 10 human DGK paralogs, which has made it difficult to gain a more complete understanding of the enzyme's mechanism of action and regulation. In the present study, we have taken advantage of the significant developments in protein structural prediction technology by artificial intelligence (i.e., Alphafold 2.0), to conduct a comprehensive investigation on the properties of all 10 human DGK paralogs. Structural alignment of the predictions reveals that the C1, catalytic, and accessory domains are conserved in their spatial arrangement relative to each other, across all paralogs. This suggests a critical role played by this domain architecture in DGK function. Moreover, docking studies corroborate the existence of a conserved ATP-binding site between the catalytic and accessory domains. Interestingly, the ATP bound to the interdomain cleft was also found to be in proximity of the conserved glycine-rich motif, which in protein kinases has been suggested to function in ATP binding. Lastly, the spatial arrangement of DGK, with respect to the membrane, reveals that most paralogs possess a more energetically favorable interaction with curved membranes. In conclusion, AlphaFold predictions of human DGKs provide novel insights into the enzyme's structural and functional properties while also paving the way for future experimentation.
Keyphrases
- endothelial cells
- artificial intelligence
- high resolution
- induced pluripotent stem cells
- transcription factor
- machine learning
- molecular dynamics
- molecular dynamics simulations
- mass spectrometry
- multiple sclerosis
- squamous cell carcinoma
- binding protein
- deep learning
- oxidative stress
- single molecule
- protein protein
- small molecule
- amino acid
- liquid chromatography