TLCrys: Transfer Learning Based Method for Protein Crystallization Prediction

X-ray diffraction technique is one of the most common methods of ascertaining protein structures, yet only 2–10% of proteins can produce diffraction-quality crystals. Several computational methods have been proposed so far to predict protein crystallization. Nevertheless, the current state-of-the-art computational methods are limited by the scarcity of experimental data. Thus, the prediction accuracy of existing models hasn’t reached the ideal level. To address the problems above, we propose a novel transfer-learning-based framework for protein crystallization prediction, named TLCrys. The framework proceeds in two steps: pre-training and fine-tuning. The pre-training step adopts attention mechanism to extract both global and local information of the protein sequences. The representation learned from the pre-training step is regarded as knowledge to be transferred and fine-tuned to enhance the performance of crystalization prediction. During pre-training, TLCrys adopts a multi-task learning method, which not only improves the learning ability of protein encoding, but also enhances the robustness and generalization of protein representation. The multi-head self-attention layer guarantees that different levels of the protein representation can be extracted by the fine-tuned step. During transfer learning, the fine-tuning strategy used by TLCrys improves the task-specialized learning ability of the network. Our method outperforms all previous predictors significantly in five crystallization stages of prediction. Furthermore, the proposed methodology can be well generalized to other protein sequence classification tasks.

Phase Separation of Shell Protein and Enzyme: An Insight into the Biogenesis of a Prokaryotic Metabolosome

Bacterial microcompartments are substrate specific metabolic modules that are conditionally expressed in certain bacterial species. These all protein structures have size in the range of 100-150 nm and are formed by the self-assembly of thousands of protein subunits, all encoded by genes belonging to a single operon. The operon contains genes that encode for both enzymes and shell proteins. The shell proteins self-assemble to form the outer coat of the compartment and enzymes are encapsulated within. A perplexing question in MCP biology is to understand the mechanism which governs the formation of these small yet complex assemblages of proteins. In this work we use 1,2-propanediol utilization microcompartments (PduMCP) as a paradigm to identify the factors that drive the self-assembly of MCP proteins. We find that a major shell protein PduBB tend to self-assemble under macromolecular crowded environment and suitable ionic strength. Microscopic visualization and biophysical studies reveal phase separation to be the principle mechanism behind the self-association of shell protein in the presence of salts and macromolecular crowding. The shell protein PduBB interacts with the enzyme diol-dehydratase PduCDE and co-assemble into phase separated liquid droplets. The co-assembly of PduCDE and PduBB results in the enhancement of catalytic activity of the enzyme. A combination of spectroscopic and biochemical techniques shows the relevance of divalent cation Mg2+ in providing stability to intact PduMCP in vivo. Together our results suggest a combination of protein-protein interactions and phase separation guiding the self-assembly of Pdu shell protein and enzyme in solution phase.

The Effect of Ethanol on Disassembly of Amyloid-β1-42 Pentamer Revealed by Atomic Force Microscopy and Gel Electrophoresis

The most common type of dementia, Alzheimer’s disease, is associated with senile plaques formed by the filamentous aggregation of hydrophobic amyloid-β (Aβ) in the brains of patients. Small oligomeric assemblies also occur and drugs and chemical compounds that can interact with such assemblies have attracted much attention. However, these compounds need to be solubilized in appropriate solvents, such as ethanol, which may also destabilize their protein structures. As the impact of ethanol on oligomeric Aβ assembly is unknown, we investigated the effect of various concentrations of ethanol (0 to 7.2 M) on Aβ pentameric assemblies (Aβp) by combining blue native-PAGE (BN-PAGE) and ambient air atomic force microscopy (AFM). This approach was proven to be very convenient and reliable for the quantitative analysis of Aβ assembly. The Gaussian analysis of the height histogram obtained from the AFM images was correlated with band intensity on BN-PAGE for the quantitative estimation of Aβp. Our observations indicated up to 1.4 M (8.3%) of added ethanol can be used as a solvent/vehicle without quantitatively affecting Aβ pentamer stability. Higher concentration induced significant destabilization of Aβp and eventually resulted in the complete disassembly of Aβp.

Activity- and Enrichment-Based Metaproteomics Insights into Active Urease from the Rumen Microbiota of Cattle

Regulation of microbial urease activity plays a crucial role in improving the utilization efficiency of urea and reducing nitrogen emissions to the environment for ruminant animals. Dealing with the diversity of microbial urease and identifying highly active urease as the target is the key for future regulation. However, the identification of active urease in the rumen is currently limited due to large numbers of uncultured microorganisms. In the present study, we describe an activity- and enrichment-based metaproteomic analysis as an approach for the discovery of highly active urease from the rumen microbiota of cattle. We conducted an optimization method of protein extraction and purification to obtain higher urease activity protein. Cryomilling was the best choice among the six applied protein extraction methods (ultrasonication, bead beating, cryomilling, high-pressure press, freeze-thawing, and protein extraction kit) for obtaining protein with high urease activity. The extracted protein by cryomilling was further enriched through gel filtration chromatography to obtain the fraction with the highest urease activity. Then, by using SDS-PAGE, the gel band including urease was excised and analyzed using LC-MS/MS, searching against a metagenome-derived protein database. Finally, we identified six microbial active ureases from 2225 rumen proteins, and the identified ureases were homologous to those of Fibrobacter and Treponema. Moreover, by comparing the 3D protein structures of the identified ureases and known ureases, we found that the residues in the β-turn of flap regions were nonconserved, which might be crucial in influencing the flexibility of flap regions and urease activity. In conclusion, the active urease from rumen microbes was identified by the approach of activity- and enrichment-based metaproteomics, which provides the target for designing a novel efficient urease inhibitor to regulate rumen microbial urease activity.

Sequence Coverage Visualizer: A web application for protein sequence coverage 3D visualization

Protein structure is connected with its function and interaction and plays an extremely important role in protein characterization. As one of the most important analytical methods for protein characterization, Proteomics is widely used to determine protein composition, quantitation, interaction, and even structures. However, due to the gap between identified proteins by proteomics and available 3D structures, it was very challenging, if not impossible, to visualize proteomics results in 3D and further explore the structural aspects of proteomics experiments. Recently, two groups of researchers from DeepMind and Baker lab have independently published protein structure prediction tools that can help us obtain predicted protein structures for the whole human proteome. Although there is still debate on the validity of some of the predicted structures, it is no doubt that these represent the most accurate predictions to date. More importantly, this enabled us to visualize the majority of human proteins for the first time. To help other researchers best utilize these protein structure predictions, we present the Sequence Coverage Visualizer (SCV), http://scv.lab.gy, a web application for protein sequence coverage 3D visualization. Here we showed a few possible usages of the SCV, including the labeling of post-translational modifications and isotope labeling experiments. These results highlight the usefulness of such 3D visualization for proteomics experiments and how SCV can turn a regular result list into structural insights. Furthermore, when used together with limited proteolysis, we demonstrated that SCV can help validate and compare different protein structures, including predicted ones and existing PDB entries. By performing limited proteolysis on native proteins at various time points, SCV can visualize the progress of the digestion. This time-series data further allowed us to compare the predicted structure and existing PDB entries. Although not deterministic, these comparisons could be used to refine current predictions further and represent an important step towards a complete and correct protein structure database. Overall, SCV is a convenient and powerful tool for visualizing proteomics results.

A database of calculated solution parameters for the recently released AlphaFold predicted protein structures

Recent spectacular advances by AI programs in 3D structure predictions from protein sequences have revolutionized the field in terms of accuracy and speed. The resulting "folding frenzy" has already produced predicted protein structure databases for the entire human and other organisms' proteomes. However, rapidly ascertaining a predicted structure's reliability based on measured properties in solution should be considered. Shape-sensitive hydrodynamic parameters such as the diffusion and sedimentation coefficients (D0t(20,w),s0(20,w)) and the intrinsic viscosity ([η]) can provide a rapid assessment of the overall structure likeliness, and SAXS would yield the structure-related pair-wise distance distribution function p(r) vs. r. Using the extensively validated UltraScan SOlution MOdeler (US-SOMO) suite we have calculated from the AlphaFold structures the corresponding D0t(20,w), s0(20,w), [η], p(r) vs. r, and other parameters. Circular dichroism spectra were also computed. The resulting US-SOMO-AF database should aid in rapidly evaluating the consistency in solution of AlphaFold predicted protein structures.

Sea Anemones Responding to Sex Hormones, Oxybenzone, and Benzyl Butyl Phthalate: Transcriptional Profiling and in Silico Modelling Provide Clues to Decipher Endocrine Disruption in Cnidarians

Endocrine disruption is suspected in cnidarians, but questions remain how occurs. Steroid sex hormones are detected in corals and sea anemones even though these animals do not have estrogen receptors and their repertoire of steroidogenic enzymes appears to be incomplete. Pathways associated with sex hormone biosynthesis and sterol signaling are an understudied area in cnidarian biology. The objective of this study was to identify a suite of genes that can be linked to exposure of endocrine disruptors. Exaiptasia diaphana were exposed to nominal 20ppb concentrations of estradiol (E2), testosterone (T), cholesterol, oxybenzone (BP-3), or benzyl butyl phthalate (BBP) for 4 h. Eleven genes of interest (GOIs) were chosen from a previously generated EST library. The GOIs are 17β-hydroxysteroid dehydrogenases type 14 (17β HSD14) and type 12 (17β HSD12), Niemann-Pick C type 2 (NPC2), Equistatin (EI), Complement component C3 (C3), Cathepsin L (CTSL), Patched domain-containing protein 3 (PTCH3), Smoothened (SMO), Desert Hedgehog (DHH), Zinc finger protein GLI2 (GLI2), and Vitellogenin (VTG). These GOIs were selected because of functional associations with steroid hormone biosynthesis; cholesterol binding/transport; immunity; phagocytosis; or Hedgehog signaling. Quantitative Real-Time PCR quantified expression of GOIs. In silico modelling utilized protein structures from Protein Data Bank as well as creating protein structures with SWISS-MODEL. Results show transcription of steroidogenic enzymes, and cholesterol binding/transport proteins have similar transcription profiles for E2, T, and cholesterol treatments, but different profiles when BP-3 or BBP is present. C3 expression can differentiate between exposures to BP-3 versus BBP as well as exposure to cholesterol versus sex hormones. In silico modelling revealed all ligands (E2, T, cholesterol, BBP, and BP-3) have favorable binding affinities with 17β HSD14, 17β HSD12, NPC2, SMO, and PTCH proteins. VTG expression was down-regulated in the sterol treatments but up-regulated in BP-3 and BBP treatments. In summary, these eleven GOIs collectively generate unique transcriptional profiles capable of discriminating between the five chemical exposures used in this investigation. This suite of GOIs are candidate biomarkers for detecting transcriptional changes in steroidogenesis, gametogenesis, sterol transport, and Hedgehog signaling. Detection of disruptions in these pathways offers new insight into endocrine disruption in cnidarians.

Gold (III) Derivatives in Colon Cancer Treatment

Cancer is one of the leading causes of morbidity and mortality worldwide. Colorectal cancer (CRC) is the third most frequently diagnosed cancer in men and the second in women. Standard patterns of antitumor therapy, including cisplatin, are ineffective due to their lack of specificity for tumor cells, development of drug resistance, and severe side effects. For this reason, new methods and strategies for CRC treatment are urgently needed. Current research includes novel platinum (Pt)- and other metal-based drugs such as gold (Au), silver (Ag), iridium (Ir), or ruthenium (Ru). Au(III) compounds are promising drug candidates for CRC treatment due to their structural similarity to Pt(II). Their advantage is their relatively good solubility in water, but their disadvantage is an unsatisfactory stability under physiological conditions. Due to these limitations, work is still underway to improve the formula of Au(III) complexes by combining with various types of ligands capable of stabilizing the Au(III) cation and preventing its reduction under physiological conditions. This review summarizes the achievements in the field of stable Au(III) complexes with potential cytotoxic activity restricted to cancer cells. Moreover, it has been shown that not nucleic acids but various protein structures such as thioredoxin reductase (TrxR) mediate the antitumor effects of Au derivatives. The state of the art of the in vivo studies so far conducted is also described.