Exploring Human Diseases and Biological Mechanisms by Protein Structure Prediction and Modeling

PLMC: Language Model of Protein Sequences Enhances Protein Crystallization Prediction

X-ray diffraction crystallography has been most widely used for protein three-dimensional (3D) structure determination for which whether proteins are crystallizable is a central prerequisite. Yet, there are a number of procedures during protein crystallization, including protein material production, purification, and crystal production, which take turns affecting the crystallization outcome. Due to the expensive and laborious nature of this multi-stage process, various computational tools have been developed to predict protein crystallization propensity, which is then used to guide the experimental determination. In this study, we presented a novel deep learning framework, PLMC, to improve multi-stage protein crystallization propensity prediction by leveraging a pre-trained protein language model. To effectively train PLMC, two groups of features of each protein were integrated into a more comprehensive representation, including protein language embeddings from the large-scale protein sequence database and a handcrafted feature set consisting of physicochemical, sequence-based and disordered-related information. These features were further separately embedded for refinement, and then concatenated for the final prediction. Notably, our extensive benchmarking tests demonstrate that PLMC greatly outperforms other state-of-the-art methods by achieving AUC scores of 0.773, 0.893, and 0.913, respectively, at the aforementioned individual stages, and 0.982 at the final crystallization stage. Furthermore, PLMC is shown to be superior for predicting the crystallization of both globular and membrane proteins, as demonstrated by an AUC score of 0.991 for the latter. These results suggest the significant potential of PLMC in assisting researchers with the experimental design of crystallizable protein variants.

4-Phenylbutyrate restored γ-aminobutyric acid uptake and reduced seizures in SLC6A1 patient variant-bearing cell and mouse models

We have studied the molecular mechanisms of variants in solute carrier Family 6 Member 1 associated with neurodevelopmental disorders, including various epilepsy syndromes, autism and intellectual disability. Based on functional assays of solute carrier Family 6 Member 1 variants, we conclude that partial or complete loss of γ-amino butyric acid uptake due to reduced membrane γ-amino butyric acid transporter 1 trafficking is the primary aetiology. Importantly, we identified common patterns of the mutant γ-amino butyric acid transporter 1 protein trafficking from biogenesis, oligomerization, glycosylation and translocation to the cell membrane across variants in different cell types such as astrocytes and neurons. We hypothesize that therapeutic approaches to facilitate membrane trafficking would increase γ-amino butyric acid transporter 1 protein membrane expression and function. 4-Phenylbutyrate is a Food and Drug Administration-approved drug for paediatric use and is orally bioavailable. 4-Phenylbutyrate shows promise in the treatment of cystic fibrosis. The common cellular mechanisms shared by the mutant γ-amino butyric acid transporter 1 and cystic fibrosis transmembrane conductance regulator led us to hypothesize that 4-phenylbutyrate could be a potential treatment option for solute carrier Family 6 Member 1 mutations. We examined the impact of 4-phenylbutyrate across a library of variants in cell and knockin mouse models. Because γ-amino butyric acid transporter 1 is expressed in both neurons and astrocytes, and γ-amino butyric acid transporter 1 deficiency in astrocytes has been hypothesized to underlie seizure generation, we tested the effect of 4-phenylbutyrate in both neurons and astrocytes with a focus on astrocytes. We demonstrated existence of the mutant γ-amino butyric acid transporter 1 retaining wildtype γ-amino butyric acid transporter 1, suggesting the mutant protein causes aberrant protein oligomerization and trafficking. 4-Phenylbutyrate increased γ-amino butyric acid uptake in both mouse and human astrocytes and neurons bearing the variants. Importantly, 4-phenylbutyrate alone increased γ-amino butyric acid transporter 1 expression and suppressed spike wave discharges in heterozygous knockin mice. Although the mechanisms of action for 4-phenylbutyrate are still unclear, with multiple possibly being involved, it is likely that 4-phenylbutyrate can facilitate the forward trafficking of the wildtype γ-amino butyric acid transporter 1 regardless of rescuing the mutant γ-amino butyric acid transporter 1, thus increasing γ-amino butyric acid uptake. All patients with solute carrier Family 6 Member 1 variants are heterozygous and carry one wildtype allele, suggesting a great opportunity for treatment development leveraging wildtype protein trafficking. The study opens a novel avenue of treatment development for genetic epilepsy via drug repurposing.