Biology subjects

Aoki, W.

Publications and source records attributed to Aoki, W..

5 recordsLinked to original sources

bioRxiv2025

Prediction of quantitative function of artificially-designed protein from structural information

Artificially designed proteins are widely used in applications such as optogenetics and biosensing. While experimental optimization of these proteins is effective, it is also costly and labor-intensive. To address this challenge, computational approaches have been developed, primarily relying on sequence-based features. However, protein function is inherently tied to its three-dimensional (3D) structure, and incorporating structural information could enable more accurate predictions and provide deeper biological interpretability. Here, we proposed a structure-based analysis framework called Foldinsight for predicting protein functionalities. In our framework, we first predict protein structures from sequences using AlphaFold2 and then utilize these structures to predict protein properties. Since proteins vary in the number of atoms and lack direct atomic correspondence, we applied molecular field mapping, which captures the energy states surrounding a protein and converts them into fixed-length numerical vectors. This transformation enables the application of machine learning, allowing protein properties to be predicted from structure-derived features. Applying this framework to channelrhodopsin mutants, we achieved predictive performance comparable to sequence-based models. Additionally, our structure-based analysis successfully identified key structural regions contributing to functional differences, highlighting the advantage of incorporating structural data into predictive modeling.

bioinformatics↗

bioRxiv2022

Reconstitution of ribosome self-replication outside a living cell

Ribosome biogenesis, a recursive process of pre-existing ribosomes self-replicating nascent ones, is pivotal in the self-replication of life. In Escherichia coli, three ribosomal RNAs (rRNAs) are transcribed, and 54 ribosomal proteins (r-proteins) are synthesized by pre-existing ribosomes as structural components1, 2. They are cotranscriptionally assembled in a cooperative hierarchy under the support of [~]100 accessory factors1-3. The reconstitution of ribosome biogenesis outside a living cell is an essential goal to understand the self-replication of life. However, this goal could not have been achieved so far due to its complexity. Here, we report the successful in vitro reconstitution of the entire ribosome biogenesis process. We hypothesized that mimicking in vivo ribosome biogenesis1-6 could result in in vitro ribosome biogenesis. Specifically, we found that coactivating the transcription of an rRNA operon, as well as the transcription and translation of 54 r-protein genes encoding r-proteins, and the coordinated ribosomal assembly in a cytoplasm-mimicking reaction solution, resulted in highly efficient in vitro reconstitution of ribosome biogenesis. Our achievement represents a critical step toward revealing fundamental principles underlying the self-replication of life and creating self-replicating artificial cells7. We also succeeded in engineering rRNA and r-proteins by only adding mutant ribosomal genes in the reaction, enabling high-throughput and unconstrained creation of artificial ribosomes with altered or enhanced functionality8-12.

synthetic biology↗

bioRxiv2020

Identification of a Putative Sensor Protein Involved in Regulation of Vesicle Production by a Hypervesiculating Bacterium, Shewanella vesiculosa HM13

Bacteria secrete and utilize nanoparticles, called extracellular membrane vesicles (EMVs), for survival in their growing environments. Therefore, the amount and components of EMVs should be tuned in response to the environment. However, how bacteria regulate vesiculation in response to the extracellular environment remains largely unknown. In this study, we identified a putative sensor protein, HM1275, involved in the induction of vesicle production in a hypervesiculating Gram-negative bacterium, Shewanella vesiculosa HM13. This protein was predicted to possess typical sensing and signaling domains of sensor proteins, such as methyl-accepting chemotaxis proteins. Comparison of vesicle production between the hm1275-disrupted mutant and the parent strain revealed that HM1275 is involved in lysine-induced hypervesiculation. Moreover, HM1275 has sequence similarity to a biofilm dispersion protein, BdlA, of Pseudomonas aeruginosa PAO1, and hm1275 disruption increased the amount of biofilm. Thus, this study showed that the induction of vesicle production and suppression of biofilm formation in response to lysine concentration are under the control of the same putative sensor protein.

microbiology↗

bioRxiv2020

Selected reaction monitoring for the quantification of Escherichia coli ribosomal proteins

Ribosomes are the sophisticated machinery that is responsible for protein synthesis in a cell. Recently, quantitative mass spectrometry (qMS) based on data-dependent acquisition (DDA) have been widely used to understand the biogenesis and function of ribosomes. However, DDA-based qMS sometimes does not provide the reproducible and quantitatively reliable analysis that is needed for high-throughput hypothesis testing. To overcome this problem, we developed a highly sensitive, specific, and accurate method to quantify all ribosomal proteins (r-proteins) by combining selected reaction monitoring (SRM) and isotope labeling. We optimized the SRM methods using purified ribosomes and Escherichia coli lysates, and verified this approach as a high-throughput analytical tool by detecting 41 of the 54 r-proteins separately synthesized in E. coli S30 extracts. The SRM methods will enable us to utilize qMS as a high-throughput hypothesis testing tool in the research of E. coli ribosomes, and they have potential to accelerate the understanding of ribosome biogenesis, function, and the development of engineered ribosomes with additional functions.

synthetic biology↗

bioRxiv2017

Cellomics approach for high-throughput functional annotation of Caenorhabditis elegans neural network

Even in Caenorhabditis elegans, which has only 302 neurons, relationships between behaviors and neural networks are not easily elucidated. In this study, we proposed a novel cellomics approach enabling high-throughput and comprehensive exploration of the functions of a single neuron or a subset of neurons in a complex neural network on a particular behavior. To realize this, we combined optogenetics and Brainbow technologies. Using these technologies, we established a C. elegans library where opsin is labeled in a randomized pattern. Behavioral analysis on this library under light illumination enabled high-throughput annotation of neurons affecting target behaviors. We applied this approach to the egg-laying behavior of C. elegans and succeeded in high-throughput confirmation that hermaphrodite-specific neurons play an important role in the egg-laying behavior. This cellomics approach will lead to the accumulation of neurophysiological and behavioral data of the C. elegans neural network, which is necessary for constructing neuroanatomically grounded models of behavior.

neuroscience↗