This document only includes an excerpt of the corresponding thesis or dissertation. To request a digital scan of the full text, please contact the Ruth Lilly Medical Library's Interlibrary Loan Department (rlmlill@iu.edu).

G protein-coupled receptors (GPCRs) activate heterotrimeric G proteins (Gαβγ), which together form one of the most important signaling axes found in the cell. Because GPCRs are very common targets for therapeutic drugs, the mechanisms underlying their regulation are of high biomedical importance. Downstream of GPCR activation, there are many cytoplasmic proteins that regulate the activity of G proteins, providing an opportunity for therapeutic intervention. The neuronal protein GINIP binds directly to Gαi and is believed to play a role in modulating GPCR-mediated neurotransmission, an exquisitely-balanced process whereby dysregulation leads to neurological disorders including chronic pain and epilepsy. However, the molecular and structural determinants of GINIP underlying and required for proper regulation of G protein signaling downstream of GPCRs are unclear. In the studies presented here, we dissect the molecular and structural basis by which GINIP regulates G proteins after receptor activation and contributes to the fine-tuning of neurotransmitter responses in the nervous system. First, we revealed a new paradigm of G protein regulation by GINIP whereby it biases GPCR responses favoring Gβγ-mediated signaling to the detriment of Gα-mediated signaling. Second, we demonstrated that GINIP uses specific residues in the first loop of the PHD domain to physically engage Gαi by adopting a binding mode similar to that of G protein effectors like adenylyl cyclase, which is in turn required for the subsequent modulation of G protein signaling. Together, these insights advance our understanding of how GPCR signaling is fine-tuned by GINIP to set the tone of neurotransmission. Characterizing this layer of G protein regulation after receptor activation is crucial for developing novel therapeutic approaches to target diseases that arise from dysregulated GPCR signaling.

No description available.

Note:

Note:

No description available.

No description available.

No description available.

No description available.

Proteins and their function play one of the most essential roles in various biological processes. The study of PPI is of considerable importance. PPI network data are of great scientific value, however, they are incomplete and experimental identification is time and money consuming. Available computational methods perform well on model organisms’ PPI prediction but perform poorly for a novel organism. Due to the incompleteness of interaction data, it is challenging to train a model for a novel organism. Also, millions to billions of interactions need to be verified which is extremely compute-intensive. We aim to improve the performance of predicting whether a pair of proteins will interact, with only two sequences as input. And also efficiently predict a PPI network with a proteome of sequences as input. We hypothesize that information about cellular locations where proteins are active and proteins' 3D structures can help us to significantly improve predict performance. To overcome the lack of experimental data, we use predicted structures by AlphaFold2 and cellular locations by DeepGoPlus. We believe that proteins belonging to disjoint biological components have very little chance to interact. We manually choose several disjoint pairs and further confirmed it by experimental PPI. We generate new no-interaction pairs with disjoint classes to update the D-SCRIPT dataset. As result, the AUPR has improved by 10% compared to the D-SCRIPT dataset. Besides, we pre-filter the negatives instead of enumerating all the potential PPI for de-novo PPI network prediction. For E.coli, we can pass around a million negative interactions. To combine the structure and sequence information, we generate a graph for each protein. A graph convolution network using Self-Attention Graph Pooling in Siamese architecture is used to learn these graphs for PPI prediction. In this way, we can improve around 20% in AUPR compared to our baseline model D-SCRIPT.