Type II diabetes, a chronic disease that affects more than 320 million people worldwide, is the fourth or fifth leading cause of death in most developed countries.
At the same time, the incidence is on the rise.
Acarbose is an alpha glucosidase inhibitor that lowers the concentration of sugar in the blood. Hence, it is an effective drug for the treatment of type II diabetes. It is increasingly preferred by patients because of its mild and long-lasting therapeutic effect and non-toxicity. Actinomyces mobilis SE50/110 is known for the production of alpha glucosidase inhibitor and antidiabetic drug acarbose. Strains with high acarbose production have been obtained by mutagenesis screening.
Nevertheless, as the number of diabetics increases, there is an urgent need for higher production of acarbose.
At present, although there is no radical cure for diabetes, diabetes can be controlled through a variety of treatments. It mainly includes five aspects: education of diabetic patients, self-monitoring blood glucose, diet therapy, exercise therapy and drug therapy. Drug treatment includes sulfonylureas, biguanides α-Glucosidase inhibitors (for example, we can use acarbose, one kind of α- Glucosidase inhibitor, which can inhibit the activity of maltase, isomaltase, glucoamylase and sucrase, thereby reducing blood glucose), insulin sensitizer, glinide type insulin rough secretors, etc.
However, the prevalent treatment method by insulin may result in various problems. Insulin treatment of diabetes may lead to hypoglycemia, edema, local allergic reaction, blurred vision, subcutaneous fat hyperplasia or fat atrophy at the injection site and other adverse reactions.
On the one hand, Genome-scale Metabolic Models (GEM) are important tools in systems biology, which analyze complex biological systems from a systems perspective by combining computer models and experimental data.
On the other hand, deep learning method has been applied in chemical space modeling and has shown excellent performance. DLKcat (Deep Learning-based Kcat prediction) using substrate structure and protein sequence as input, has the ability to predict various biological enzyme activities (Kcat) on a large scale.
In addition, according to existing studies, the restriction of enzymes and the 3D structure of proteins have been integrated into GEM, thus expanding its application scope and laying the foundation for whole-cell modeling.
So we have been inspired to develop a software. GEM is used to model the metabolic process of actinomycetes, and the restriction of enzymes is integrated into GEM. At the same time, the Kcat value prediction algorithm based on deep learning is used to optimize the model. Some new protein prediction tools can also be added in the software.
This project plans to develop a generalized modeling software to fill the gaps in the Actinobacteria protein database by adding enzyme constraints to the genome-wide metabolic model of Actinobacteria and combining deep learning to predict the Kcat values of model parameters.
Our software provides a feasible solution for the reconstruction of DL-ecGEM Model (Deep Learning - Enzyme Constraints GEM) for Actinobacteria. The reconstructed model is used to conduct simulation experiments to find key enzymes for target reactions, and to combine with protein hotspot identification tools to find targeted modification solutions for key enzymes.
This project has developed multi-functional software integrating modeling, prediction, and simulation to facilitate rapid use by bioscience researchers, reduce wasted experimental costs, and shorten experimental exploration cycles. The software can also rovide optimized high-yield strategies for existing antibiotics, advance the discovery exploration of novel antibiotics in the research of actinomycetes in extreme environments such as the ocean, and accelerate the industrialization and clinical application of rare antibiotics.
[1] Wang Y, Xu N, Ye C, Liu L, Shi Z and Wu J (2015) Reconstruction and in silico analysis of an Actinoplanes sp. SE50/110 genome-scale metabolic model for acarbose production. Front. Microbiol. 6:632. doi: 10.3389/fmicb.2015.00632
[2] Sánchez, Benjamín & Zhang, Cheng & Nilsson, Avlant & Lahtvee, Petri-Jaan & Kerkhoven, Eduard & Nielsen, Jens. (2017). Improving the phenotype predictions of a yeast genome‐scale metabolic model by incorporating enzymatic constraints. Molecular Systems Biology. 13. 10.15252/msb.20167411.
[3] Gu D, Jian X, Zhang C, Hua Q (2016) Reframed genome-scale metabolic model to facilitate genetic design and integration with expression data. IEEE/ACM Trans Comput Biol Bioinform https://doi.org/10.1109/TCBB.2016.2576456
[4] Domenzain, I., Sánchez, B., Anton, M. et al. Reconstruction of a catalogue of genome-scale metabolic models with enzymatic constraints using GECKO 2.0. Nat Commun 13, 3766 (2022). https://doi.org/10.1038/s41467-022-31421-1
[5] Schellenberger, J., Que, R., Fleming, R. et al. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox v2.0. Nat Protoc 6, 1290–1307 (2011). https://doi.org/10.1038/nprot.2011.308
[6] Lu, H., Li, F., Sánchez, B.J. et al. A consensus S. cerevisiae metabolic model Yeast8 and its ecosystem for comprehensively probing cellular metabolism. Nat Commun 10, 3586 (2019). https://doi.org/10.1038/s41467-019-11581-3
[7] Li, F., Yuan, L., Lu, H. et al. Deep learning-based kcat prediction enables improved enzyme-constrained model reconstruction. Nat Catal 5, 662–672 (2022). https://doi.org/10.1038/s41929-022-00798-z
[8] Bendl J, Stourac J, Sebestova E, Vavra O, Musil M, Brezovsky J, Damborsky J. HotSpot Wizard 2.0: automated design of site-specific mutations and smart libraries in protein engineering. Nucleic Acids Res. 2016 Jul 8;44(W1):W479-87. doi: 10.1093/nar/gkw416. Epub 2016 May 12. PMID: 27174934; PMCID: PMC4987947.