Machine learning guided batched design of a bacterial Ribosome Binding Site

Optimisation of gene expression levels is an essential part of the organism design process. Fine control of this process can be achieved through engineering transcription and translation control elements, including the ribosome binding site (RBS). Unfortunately, design of specific genetic parts can still be challenging due to lack of reliable design methods. To address this problem, we have created a machine learning guided Design-Build-Test-Learn (DBTL) cycle for the experimental design of bacterial RBSs to show how small genetic parts can be reliably designed using relatively small, high-quality data sets. We used Gaussian Process Regression for the Learn phase of cycle and the Upper Confidence Bound multi-armed bandit algorithm for the Design of genetic variants to be tested in vivo. We have integrated these machine learning algorithms with laboratory automation and high-throughput processes for reliable data generation. Notably, by Testing a total of 450 RBS variants in four DBTL cycles, we experimentally validated RBSs with high translation initiation rates equalling or exceeding our benchmark RBS by up to 34%. Overall, our results show that machine learning is a powerful tool for designing RBSs, and they pave the way towards more complicated genetic devices.

Crystal structure and ligand-induced folding of the SAM/SAH riboswitch

While most SAM riboswitches strongly discriminate between SAM and SAH, the SAM/SAH riboswitch responds to both ligands with similar apparent affinities. We have determined crystal structures of the SAM/SAH riboswitch bound to SAH, SAM and other variant ligands at high resolution. The riboswitch forms an H-type pseudoknot structure with coaxial alignment of the stem–loop helix (P1) and the pseudoknot helix (PK). An additional three base pairs form at the non-open end of P1, and the ligand is bound at the interface between the P1 extension and the PK helix. The adenine nucleobase is stacked into the helix and forms a trans Hoogsteen–Watson–Crick base pair with a uridine, thus becoming an integral part of the helical structure. The majority of the specific interactions are formed with the adenosine. The methionine or homocysteine chain lies in the groove making a single hydrogen bond, and there is no discrimination between the sulfonium of SAM or the thioether of SAH. Single-molecule FRET analysis reveals that the riboswitch exists in two distinct conformations, and that addition of SAM or SAH shifts the population into a stable state that likely corresponds to the form observed in the crystal. A model for translational regulation is presented whereby in the absence of ligand the riboswitch is largely unfolded, lacking the PK helix so that translation can be initiated at the ribosome binding site. But the presence of ligand stabilizes the folded conformation that includes the PK helix, so occluding the ribosome binding site and thus preventing the initiation of translation.

Peptide Mimics of the Ribosomal P Stalk Inhibit the Activity of Ricin A Chain by Preventing Ribosome Binding

Ricin A chain (RTA) depurinates the sarcin/ricin loop (SRL) by interacting with the C-termini of the ribosomal P stalk. The ribosome interaction site and the active site are located on opposite faces of RTA. The interaction with P proteins allows RTA to depurinate the SRL on the ribosome at physiological pH with an extremely high activity by orienting the active site towards the SRL. Therefore, if an inhibitor disrupts RTA–ribosome interaction by binding to the ribosome binding site of RTA, it should inhibit the depurination activity. To test this model, we synthesized peptides mimicking the last 3 to 11 amino acids of P proteins and examined their interaction with wild-type RTA and ribosome binding mutants by Biacore. We measured the inhibitory activity of these peptides on RTA-mediated depurination of yeast and rat liver ribosomes. We found that the peptides interacted with the ribosome binding site of RTA and inhibited depurination activity by disrupting RTA–ribosome interactions. The shortest peptide that could interact with RTA and inhibit its activity was four amino acids in length. RTA activity was inhibited by disrupting its interaction with the P stalk without targeting the active site, establishing the ribosome binding site as a new target for inhibitor discovery.

A Codon Deletion at the Beginning of Green Fluorescent Protein Genes Enhances Protein Expression

Recombinant protein expression is one of the key issues in protein engineering and biotechnology. Among the different models for assessing protein production and structure-function studies, green fluorescent protein (GFP) is one of the preferred models because of its importance as a reporter in cellular and molecular studies. In this research we analyze the effect of codon deletions near the amino terminus of different GFP proteins on fluorescence. Our study includes Gly4 deletions in the enhanced GFP (EGFP), the red-shifted GFP and the red-shifted EGFP. The Gly4 deletion mutants and their corresponding wild-type counterparts were transcribed under the control of the T7 or Trc promoters and their expression patterns were analyzed. Different fluorescent outcomes were observed depending on the type of fluorescent gene versions. In silico analysis of the RNA secondary structures near the ribosome binding site revealed a direct relationship between their minimum free energy and GFP production. Integrative analysis of these results, including SDS-PAGE analysis, led us to conclude that the fluorescence improvement of cells expressing different versions of GFPs with Gly4 deleted is due to an enhancement of the accessibility of the ribosome binding site by reducing the stability of the RNA secondary structures at their mRNA leader regions.