Extensive antibody search with whole spectrum black-box optimization.

In designing functional biological sequences with machine learning, the activity predictor tends to be inaccurate due to shortage of data. Top ranked sequences are thus unlikely to contain effective ones. This paper proposes to take prediction stability into account to provide domain experts with a reasonable list of sequences to choose from. In our approach, multiple prediction models are trained by subsampling the training set and the multi-objective optimization problem, where one objective is the average activity and the other is the standard deviation, is solved. The Pareto front represents a list of sequences with the whole spectrum of activity and stability. Using this method, we designed VHH (Variable domain of Heavy chain of Heavy chain) antibodies based on the dataset obtained from deep mutational screening. To solve multi-objective optimization, we employed our sequence design software MOQA that uses quantum annealing. By applying several selection criteria to 19,778 designed sequences, five sequences were selected for wet-lab validation. One sequence, 16 mutations away from the closest training sequence, was successfully expressed and found to possess desired binding specificity. Our whole spectrum approach provides a balanced way of dealing with the prediction uncertainty, and can possibly be applied to extensive search of functional sequences.

Selection of target-binding proteins from the information of weakly enriched phage display libraries by deep sequencing and machine learning.

Despite the advances in surface-display systems for directed evolution, variants with high affinity are not always enriched due to undesirable biases that increase target-unrelated variants during biopanning. Here, our goal was to design a library containing improved variants from the information of the "weakly enriched" library where functional variants were weakly enriched. Deep sequencing for the previous biopanning result, where no functional antibody mimetics were experimentally identified, revealed that weak enrichment was partly due to undesirable biases during phage infection and amplification steps. The clustering analysis of the deep sequencing data from appropriate steps revealed no distinct sequence patterns, but a Bayesian machine learning model trained with the selected deep sequencing data supplied nine clusters with distinct sequence patterns. Phage libraries were designed on the basis of the sequence patterns identified, and four improved variants with target-specific affinity (EC50 = 80-277 nM) were identified by biopanning. The selection and use of deep sequencing data without undesirable bias enabled us to extract the information on prospective variants. In summary, the use of appropriate deep sequencing data and machine learning with the sequence data has the possibility of finding sequence space where functional variants are enriched.

Stabilization of CDK6 by ribosomal protein uS7, a target protein of the natural product fucoxanthinol.

Cyclins and cyclin-dependent kinases (CDKs) regulate the cell cycle, which is important for cell proliferation and development. Cyclins bind to and activate CDKs, which then drive the cell cycle. The expression of cyclins periodically changes throughout the cell cycle, while that of CDKs remains constant. To elucidate the mechanisms underlying the constant expression of CDKs, we search for compounds that alter their expression and discover that the natural product fucoxanthinol downregulates CDK2, 4, and 6 expression. We then develop a method to immobilize a compound with a hydroxyl group onto FG beads® and identify human ribosomal protein uS7 (also known as ribosomal protein S5) as the major fucoxanthinol-binding protein using the beads and mass spectrometry. The knockdown of uS7 induces G1 cell cycle arrest with the downregulation of CDK6 in colon cancer cells. CDK6, but not CDK2 or CDK4, is degraded by the depletion of uS7, and we furthermore find that uS7 directly binds to CDK6. Fucoxanthinol decreases uS7 at the protein level in colon cancer cells. By identifying the binding proteins of a natural product, the present study reveals that ribosomal protein uS7 may contribute to the constant expression of CDK6 via a direct interaction.

A chemoproteoinformatics approach demonstrates that aspirin increases sensitivity to MEK inhibition by directly binding to RPS5.

MEK inhibitors are among the most successful molecularly targeted agents used as cancer therapeutics. However, to treat cancer more efficiently, resistance to MEK inhibitor-induced cell death must be overcome. Although previous genetic approaches based on comprehensive gene expression analysis or RNAi libraries led to the discovery of factors involved in intrinsic resistance to MEK inhibitors, a feasible combined treatment with the MEK inhibitor has not yet been developed. Here, we show that a chemoproteoinformatics approach identifies ligands overcoming the resistance to cell death induced by MEK inhibition as well as the target molecule conferring this resistance. First, we used natural products, perillyl alcohol and sesaminol, which induced cell death in combination with the MEK inhibitor trametinib, as chemical probes, and identified ribosomal protein S5 (RPS5) as their common target protein. Consistently, trametinib induced cell death in RPS5-depleted cancer cells via upregulation of the apoptotic proteins BIM and PUMA. Using molecular docking and molecular dynamics (MD) simulations, we then screened FDA- and EMA-approved drugs for RPS5-binding ligands and found that acetylsalicylic acid (ASA, also known as aspirin) directly bound to RPS5, resulting in upregulation of BIM and PUMA and induction of cell death in combination with trametinib. Our chemoproteoinformatics approach demonstrates that RPS5 confers resistance to MEK inhibitor-induced cell death, and that aspirin could be repurposed to sensitize cells to MEK inhibition by binding to RPS5.

A rational blueprint for the design of chemically-controlled protein switches.

Small-molecule responsive protein switches are crucial components to control synthetic cellular activities. However, the repertoire of small-molecule protein switches is insufficient for many applications, including those in the translational spaces, where properties such as safety, immunogenicity, drug half-life, and drug side-effects are critical. Here, we present a computational protein design strategy to repurpose drug-inhibited protein-protein interactions as OFF- and ON-switches. The designed binders and drug-receptors form chemically-disruptable heterodimers (CDH) which dissociate in the presence of small molecules. To design ON-switches, we converted the CDHs into a multi-domain architecture which we refer to as activation by inhibitor release switches (AIR) that incorporate a rationally designed drug-insensitive receptor protein. CDHs and AIRs showed excellent performance as drug responsive switches to control combinations of synthetic circuits in mammalian cells. This approach effectively expands the chemical space and logic responses in living cells and provides a blueprint to develop new ON- and OFF-switches.

Rabdosianone I, a Bitter Diterpene from an Oriental Herb, Suppresses Thymidylate Synthase Expression by Directly Binding to ANT2 and PHB2.

Natural products have numerous bioactivities and are expected to be a resource for potent drugs. However, their direct targets in cells often remain unclear. We found that rabdosianone I, which is a bitter diterpene from an oriental herb for longevity, Isodon japonicus Hara, markedly inhibited the growth of human colorectal cancer cells by downregulating the expression of thymidylate synthase (TS). Next, using rabdosianone I-immobilized nano-magnetic beads, we identified two mitochondrial inner membrane proteins, adenine nucleotide translocase 2 (ANT2) and prohibitin 2 (PHB2), as direct targets of rabdosianone I. Consistent with the action of rabdosianone I, the depletion of ANT2 or PHB2 reduced TS expression in a different manner. The knockdown of ANT2 or PHB2 promoted proteasomal degradation of TS protein, whereas that of not ANT2 but PHB2 reduced TS mRNA levels. Thus, our study reveals the ANT2- and PHB2-mediated pleiotropic regulation of TS expression and demonstrates the possibility of rabdosianone I as a lead compound of TS suppressor.

Predicting antibody affinity changes upon mutations by combining multiple predictors.

Antibodies are proteins working in our immune system with high affinity and specificity for target antigens, making them excellent tools for both biotherapeutic and bioengineering applications. The prediction of antibody affinity changes upon mutations ([Formula: see text]) is important for antibody engineering. Numerous computational methods have been proposed based on different approaches including molecular mechanics and machine learning. However, the accuracy by each individual predictor is not enough for efficient antibody development. In this study, we develop a new prediction method by combining multiple predictors based on machine learning. Our method was tested on the SiPMAB database, evaluating the Pearson's correlation coefficient between predicted and experimental [Formula: see text]. Our method achieved higher accuracy (R = 0.69) than previous molecular mechanics or machine-learning based methods (R = 0.59) and the previous method using the average of multiple predictors (R = 0.64). Feature importance analysis indicated that the improved accuracy was obtained by combining predictors with different importance, which have different protocols for calculating energies and for generating mutant and unbound state structures. This study demonstrates that machine learning is a powerful framework for combining different approaches to predict antibody affinity changes.

Development of a Novel Q-body Using an In Vivo Site-Specific Unnatural Amino Acid Incorporation System.

A Q-body capable of detecting target molecules in solutions could serve as a simple molecular detection tool. The position of the fluorescent dye in a Q-body affects sensitivity and therefore must be optimized. This report describes the development of Nef Q-bodies that recognize Nef protein, one of the human immunodeficiency virus (HIV)'s gene products, in which fluorescent dye molecules were placed at various positions using an in vivo unnatural amino acid incorporation system. A maximum change in fluorescence intensity of 2-fold was observed after optimization of the dye position. During the process, some tryptophan residues of the antibody were found to quench the fluorescence. Moreover, analysis of the epitope indicated that some amino acid residues of the antigen located near the epitope affected the fluorescence intensity.