Crystal Structure Determination of Nucleotide-sugar Binding Domain of Human UDP-glucuronosyltransferases 2B10.

BACKGROUND: UDP-glucuronosyltransferases (UGTs) play a crucial role in maintaining endobiotic homeostasis and metabolizing xenobiotic compounds, particularly clinical drugs. However, the detailed catalytic mechanism of UGTs has not been fully elucidated due to the limited availability of reliable protein structures. Determining the catalytic domain of human UGTs has proven to be a significant challenge, primarily due to the difficulty in purifying and crystallizing the full-length protein. OBJECTIVES: This study focused on the human UGT2B10 C-terminal cofactor binding domain, aiming to provide structural insights into the fundamental catalytic mechanisms. METHODS: In this study, the C-terminal sugar-donor binding domain of human UGT2B10 was purified and crystallized using the vapor-diffusion method. The resulting UGT2B10 CTD crystals displayed high-quality diffraction patterns, allowing for data collection at an impressive resolution of 1.53 Å using synchrotron radiation. Subsequently, the structure of the UGT2B10 CTD was determined using the molecule replacement method with a homologous structure. RESULTS: The crystals were monoclinic, belonging to the space C2 with unit-cell parameters a = 85.90 Å, b = 58.39 Å, c = 68.87 Å, α = γ = 90°, and ß = 98.138°. The Matthews coefficient VM was determined to be 2.24 Å3 Da-1 (solvent content 46.43%) with two molecules in the asymmetric unit. CONCLUSION: The crystal structure of UGT2B10 CTD was solved at a high resolution of 1.53 Å, revealing a conserved cofactor binding pocket. This is the first study determining the C-terminal cofactor binding domain of human UGT2B10, which plays a key role in additive drug metabolism.

Multipath Cross Graph Convolution for Knowledge Representation Learning.

In the past, most of the entity prediction methods based on embedding lacked the training of local core relationships, resulting in a deficiency in the end-to-end training. Aiming at this problem, we propose an end-to-end knowledge graph embedding representation method. It involves local graph convolution and global cross learning in this paper, which is called the TransC graph convolutional network (TransC-GCN). Firstly, multiple local semantic spaces are divided according to the largest neighbor. Secondly, a translation model is used to map the local entities and relationships into a cross vector, which serves as the input of GCN. Thirdly, through training and learning of local semantic relations, the best entities and strongest relations are found. The optimal entity relation combination ranking is obtained by evaluating the posterior loss function based on the mutual information entropy. Experiments show that this paper can obtain local entity feature information more accurately through the convolution operation of the lightweight convolutional neural network. Also, the maximum pooling operation helps to grasp the strong signal on the local feature, thereby avoiding the globally redundant feature. Compared with the mainstream triad prediction baseline model, the proposed algorithm can effectively reduce the computational complexity while achieving strong robustness. It also increases the inference accuracy of entities and relations by 8.1% and 4.4%, respectively. In short, this new method can not only effectively extract the local nodes and relationship features of the knowledge graph but also satisfy the requirements of multilayer penetration and relationship derivation of a knowledge graph.