Optimizing 5'UTRs for mRNA-delivered gene editing using deep learning.

mRNA therapeutics are revolutionizing the pharmaceutical industry, but methods to optimize the primary sequence for increased expression are still lacking. Here, we design 5'UTRs for efficient mRNA translation using deep learning. We perform polysome profiling of fully or partially randomized 5'UTR libraries in three cell types and find that UTR performance is highly correlated across cell types. We train models on our datasets and use them to guide the design of high-performing 5'UTRs using gradient descent and generative neural networks. We experimentally test designed 5'UTRs with mRNA encoding megaTALTM gene editing enzymes for two different gene targets and in two different cell lines. We find that the designed 5'UTRs support strong gene editing activity. Editing efficiency is correlated between cell types and gene targets, although the best performing UTR was specific to one cargo and cell type. Our results highlight the potential of model-based sequence design for mRNA therapeutics.

Crystallographic analyses illustrate significant plasticity and efficient recoding of meganuclease target specificity.

The retargeting of protein-DNA specificity, outside of extremely modular DNA binding proteins such as TAL effectors, has generally proved to be quite challenging. Here, we describe structural analyses of five different extensively retargeted variants of a single homing endonuclease, that have been shown to function efficiently in ex vivo and in vivo applications. The redesigned proteins harbor mutations at up to 53 residues (18%) of their amino acid sequence, primarily distributed across the DNA binding surface, making them among the most significantly reengineered ligand-binding proteins to date. Specificity is derived from the combined contributions of DNA-contacting residues and of neighboring residues that influence local structural organization. Changes in specificity are facilitated by the ability of all those residues to readily exchange both form and function. The fidelity of recognition is not precisely correlated with the fraction or total number of residues in the protein-DNA interface that are actually involved in DNA contacts, including directional hydrogen bonds. The plasticity of the DNA-recognition surface of this protein, which allows substantial retargeting of recognition specificity without requiring significant alteration of the surrounding protein architecture, reflects the ability of the corresponding genetic elements to maintain mobility and persistence in the face of genetic drift within potential host target sites.

Indirect DNA Sequence Recognition and Its Impact on Nuclease Cleavage Activity.

LAGLIDADG meganucleases are DNA cleaving enzymes used for genome engineering. While their cleavage specificity can be altered using several protein engineering and selection strategies, their overall targetability is limited by highly specific indirect recognition of the central four base pairs within their recognition sites. In order to examine the physical basis of indirect sequence recognition and to expand the number of such nucleases available for genome engineering, we have determined the target sites, DNA-bound structures, and central four cleavage fidelities of nine related enzymes. Subsequent crystallographic analyses of a meganuclease bound to two noncleavable target sites, each containing a single inactivating base pair substitution at its center, indicates that a localized slip of the mutated base pair causes a small change in the DNA backbone conformation that results in a loss of metal occupancy at one binding site, eliminating cleavage activity.

Recapitulation of the forward nuclear auxin response pathway in yeast.

Auxin influences nearly every aspect of plant biology through a simple signaling pathway; however, it remains unclear how much of the diversity in auxin effects is explained by variation in the core signaling components and which properties of these components may contribute to diversification in response dynamics. Here, we recapitulated the entire Arabidopsis thaliana forward nuclear auxin signal transduction pathway in Saccharomyces cerevisiae to test whether signaling module composition enables tuning of the dynamic response. Sensitivity analysis guided by a small mathematical model revealed the centrality of auxin/indole-3-acetic acid (Aux/IAA) transcriptional corepressors in controlling response dynamics and highlighted the strong influence of natural variation in Aux/IAA degradation rates on circuit performance. When the basic auxin response circuit was expanded to include multiple Aux/IAAs, we found that dominance relationships between coexpressed Aux/IAAs were sufficient to generate distinct response modules similar to those seen during plant development. Our work provides a new method for dissecting auxin signaling and demonstrates the key role of Aux/IAAs in tuning auxin response dynamics.

A synthetic approach reveals extensive tunability of auxin signaling.

Explaining how the small molecule auxin triggers diverse yet specific responses is a long-standing challenge in plant biology. An essential step in auxin response is the degradation of Auxin/Indole-3-Acetic Acid (Aux/IAA, referred to hereafter as IAA) repressor proteins through interaction with auxin receptors. To systematically characterize diversity in degradation behaviors among IAA|receptor pairs, we engineered auxin-induced degradation of plant IAA proteins in yeast (Saccharomyces cerevisiae). We found that IAA degradation dynamics vary widely, depending on which receptor is present, and are not encoded solely by the degron-containing domain II. To facilitate this and future studies, we identified a mathematical model able to quantitatively describe IAA degradation behavior in a single parameter. Together, our results demonstrate the remarkable tunability conferred by specific configurations of the auxin response pathway.

Slk19p of Saccharomyces cerevisiae regulates anaphase spindle dynamics through two independent mechanisms.

Slk19p is a member of the Cdc-14 early anaphase release (FEAR) pathway, a signaling network that is responsible for activation of the cell-cycle regulator Cdc14p in Saccharomyces cerevisiae. Disruption of the FEAR pathway results in defects in anaphase, including alterations in the assembly and behavior of the anaphase spindle. Many phenotypes of slk19Δ mutants are consistent with a loss of FEAR signaling, but other phenotypes suggest that Slk19p may have FEAR-independent roles in modulating the behavior of microtubules in anaphase. Here, a series of SLK19 in-frame deletion mutations were used to test whether Slk19p has distinct roles in anaphase that can be ascribed to specific regions of the protein. Separation-of-function alleles were identified that are defective for either FEAR signaling or aspects of anaphase spindle function. The data suggest that in early anaphase one region of Slk19p is essential for FEAR signaling, while later in anaphase another region is critical for maintaining the coordination between spindle elongation and the growth of interpolar microtubules.

The genome of epsilon15, a serotype-converting, Group E1 Salmonella enterica-specific bacteriophage.

The genome sequence of the Salmonella enterica serovar Anatum-specific, serotype-converting bacteriophage epsilon15 has been completed. The nonredundant genome contains 39,671 bp and 51 putative genes. It most closely resembles the genome of phiV10, an Escherichia coli O157:H7-specific temperate phage, with which it shares 36 related genes. More distant relatives include the Burkholderia cepacia-specific phage, BcepC6B (8 similar genes), the Bordetella bronchiseptica-specific phage, BPP-1 (8 similar genes) and the Photobacterium profundum prophage, P Pphipr1 (6 similar genes). epsilon15 gene identifications based on homologies with known gene families include the terminase small and large subunits, integrase, endolysin, two holins, two DNA methylase enzymes (one adenine-specific and one cytosine-specific) and a RecT-like enzyme. Genes identified experimentally include those coding for the serotype conversion proteins, the tail fiber, the major capsid protein and the major repressor. epsilon15's attP site and the Salmonella attB site with which it interacts during lysogenization have also been determined.