Genome dependent Cas9/gRNA search time underlies sequence dependent gRNA activity

Functional genomic screening in Komagataella phaffii enabled by high-activity CRISPR-Cas9 library

CRISPR-based high-throughput genome-wide loss-of-function screens are a valuable approach to functional genetics and strain engineering. The yeast Komagataella phaffii is a host of particular interest in the biopharmaceutical industry and as a metabolic engineering host for proteins and metabolites. Here, we design and validate a highly active 6-fold coverage genome-wide sgRNA library for this biotechnologically important yeast containing 30,848 active sgRNAs targeting over 99% of its coding sequences. Conducting fitness screens in the absence of functional non-homologous end joining (NHEJ), the dominant DNA repair mechanism in K. phaffii, provides a quantitative means to assess the activity of each sgRNA in the library. This approach allows for the experimental validation of each guide's targeting activity, leading to more precise screening outcomes. We used this approach to conduct growth screens with glucose as the sole carbon source and identify essential genes. Comparative analysis of the called gene sets identified a core set of K. phaffii essential genes, many of which relate to metabolic engineering targets, including protein production, secretion, and glycosylation. The high activity, genome-wide CRISPR library developed here enables functional genomic screening in K. phaffii, applied here to gene essentiality classification, and promises to enable other genetic screens.

CRISPR-COPIES: an in silico platform for discovery of neutral integration sites for CRISPR/Cas-facilitated gene integration

The CRISPR/Cas system has emerged as a powerful tool for genome editing in metabolic engineering and human gene therapy. However, locating the optimal site on the chromosome to integrate heterologous genes using the CRISPR/Cas system remains an open question. Selecting a suitable site for gene integration involves considering multiple complex criteria, including factors related to CRISPR/Cas-mediated integration, genetic stability, and gene expression. Consequently, identifying such sites on specific or different chromosomal locations typically requires extensive characterization efforts. To address these challenges, we have developed CRISPR-COPIES, a COmputational Pipeline for the Identification of CRISPR/Cas-facilitated intEgration Sites. This tool leverages ScaNN, a state-of-the-art model on the embedding-based nearest neighbor search for fast and accurate off-target search, and can identify genome-wide intergenic sites for most bacterial and fungal genomes within minutes. As a proof of concept, we utilized CRISPR-COPIES to characterize neutral integration sites in three diverse species: Saccharomyces cerevisiae, Cupriavidus necator, and HEK293T cells. In addition, we developed a user-friendly web interface for CRISPR-COPIES (https://biofoundry.web.illinois.edu/copies/). We anticipate that CRISPR-COPIES will serve as a valuable tool for targeted DNA integration and aid in the characterization of synthetic biology toolkits, enable rapid strain construction to produce valuable biochemicals, and support human gene and cell therapy applications.

Strategies to reduce the risks of mRNA drug and vaccine toxicity

mRNA formulated with lipid nanoparticles is a transformative technology that has enabled the rapid development and administration of billions of coronavirus disease 2019 (COVID-19) vaccine doses worldwide. However, avoiding unacceptable toxicity with mRNA drugs and vaccines presents challenges. Lipid nanoparticle structural components, production methods, route of administration and proteins produced from complexed mRNAs all present toxicity concerns. Here, we discuss these concerns, specifically how cell tropism and tissue distribution of mRNA and lipid nanoparticles can lead to toxicity, and their possible reactogenicity. We focus on adverse events from mRNA applications for protein replacement and gene editing therapies as well as vaccines, tracing common biochemical and cellular pathways. The potential and limitations of existing models and tools used to screen for on-target efficacy and de-risk off-target toxicity, including in vivo and next-generation in vitro models, are also discussed.

Anti-CRISPR proteins trigger a burst of CRISPR-Cas9 expression that enhances phage defense

CRISPR-Cas immune systems provide bacteria with adaptive immunity against bacteriophages, but they are often transcriptionally repressed to mitigate auto-immunity. In some cases, CRISPR-Cas expression increases in response to a phage infection, but the mechanisms of induction are largely unknown, and it is unclear whether induction occurs strongly and quickly enough to benefit the bacterial host. In S. pyogenes, Cas9 is both an immune effector and auto-repressor of CRISPR-Cas expression. Here, we show that phage-encoded anti-CRISPR proteins relieve Cas9 auto-repression and trigger a rapid increase in CRISPR-Cas levels during a single phage infective cycle. As a result, fewer cells succumb to lysis, leading to a striking survival benefit after multiple rounds of infection. CRISPR-Cas induction also reduces lysogeny, thereby limiting a route for horizontal gene transfer. Altogether, we show that Cas9 is not only a CRISPR-Cas effector and repressor but also a phage sensor that can mount an anti-anti-CRISPR transcriptional response.

HyCas9-12aGEP: an efficient genome editing platform for Corynebacterium glutamicum

Corynebacterium glutamicum plays a crucial role as a significant industrial producer of metabolites. Despite the successful development of CRISPR-Cas9 and CRISPR-Cas12a-assisted genome editing technologies in C. glutamicum, their editing resolution and efficiency are hampered by the diverse on-target activities of guide RNAs (gRNAs). To address this problem, a hybrid CRISPR-Cas9-Cas12a genome editing platform (HyCas9-12aGEP) was developed in C. glutamicum in this study to co-express sgRNA (corresponding to SpCas9 guide RNA), crRNA (corresponding to FnCas12a guide RNA), or hfgRNA (formed by the fusion of sgRNA and crRNA). HyCas9-12aGEP improves the efficiency of mapping active gRNAs and outperforms both CRISPR-Cas9 and CRISPR-Cas12a in genome editing resolution and efficiency. In the experiment involving the deletion of the cg0697-0740 gene segment, an unexpected phenotype was observed, and HyCas9-12aGEP efficiently identified the responsible genotype from more than 40 genes. Here, HyCas9-12aGEP greatly improve our capability in terms of genome reprogramming in C. glutamicum.

Interpretable neural architecture search and transfer learning for understanding CRISPR-Cas9 off-target enzymatic reactions

Finely tuned enzymatic pathways control cellular processes, and their dysregulation can lead to disease. Developing predictive and interpretable models for these pathways is challenging because of the complexity of the pathways and of the cellular and genomic contexts. Here we introduce Elektrum, a deep learning framework that addresses these challenges with data-driven and biophysically interpretable models for determining the kinetics of biochemical systems. First, it uses in vitro kinetic assays to rapidly hypothesize an ensemble of high-quality kinetically interpretable neural networks (KINNs) that predict reaction rates. It then employs a transfer learning step, where the KINNs are inserted as intermediary layers into deeper convolutional neural networks, fine-tuning the predictions for reaction-dependent in vivo outcomes. We apply Elektrum to predict CRISPR-Cas9 off-target editing probabilities and demonstrate that Elektrum achieves improved performance, regularizes neural network architectures and maintains physical interpretability.

Interpretable neural architecture search and transfer learning for understanding CRISPR/Cas9 off-target enzymatic reactions

Finely-tuned enzymatic pathways control cellular processes, and their dysregulation can lead to disease. Creating predictive and interpretable models for these pathways is challenging because of the complexity of the pathways and of the cellular and genomic contexts. Here we introduce Elektrum, a deep learning framework which addresses these challenges with data-driven and biophysically interpretable models for determining the kinetics of biochemical systems. First, it uses in vitro kinetic assays to rapidly hypothesize an ensemble of high-quality Kinetically Interpretable Neural Networks (KINNs) that predict reaction rates. It then employs a novel transfer learning step, where the KINNs are inserted as intermediary layers into deeper convolutional neural networks, fine-tuning the predictions for reaction-dependent in vivo outcomes. Elektrum makes effective use of the limited, but clean in vitro data and the complex, yet plentiful in vivo data that captures cellular context. We apply Elektrum to predict CRISPR-Cas9 off-target editing probabilities and demonstrate that Elektrum achieves state-of-the-art performance, regularizes neural network architectures, and maintains physical interpretability.

A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets

The CRISPR/Cas9 nuclease from Streptococcus pyogenes (SpCas9) can be used with single guide RNAs (sgRNAs) as a sequence-specific antimicrobial agent and as a genome-engineering tool. However, current bacterial sgRNA activity models struggle with accurate predictions and do not generalize well, possibly because the underlying datasets used to train the models do not accurately measure SpCas9/sgRNA activity and cannot distinguish on-target cleavage from toxicity. Here, we solve this problem by using a two-plasmid positive selection system to generate high-quality data that more accurately reports on SpCas9/sgRNA cleavage and that separates activity from toxicity. We develop a machine learning architecture (crisprHAL) that can be trained on existing datasets, that shows marked improvements in sgRNA activity prediction accuracy when transfer learning is used with small amounts of high-quality data, and that can generalize predictions to different bacteria. The crisprHAL model recapitulates known SpCas9/sgRNA-target DNA interactions and provides a pathway to a generalizable sgRNA bacterial activity prediction tool that will enable accurate antimicrobial and genome engineering applications.

A landing pad system for multicopy gene integration in Issatchenkia orientalis

The robust nature of the non-conventional yeast Issatchenkia orientalis allows it to grow under highly acidic conditions and therefore, has gained increasing interest in producing organic acids using a variety of carbon sources. Recently, the development of a genetic toolbox for I. orientalis, including an episomal plasmid, characterization of multiple promoters and terminators, and CRISPR-Cas9 tools, has eased the metabolic engineering efforts in I. orientalis. However, multiplex engineering is still hampered by the lack of efficient multicopy integration tools. To facilitate the construction of large, complex metabolic pathways by multiplex CRISPR-Cas9-mediated genome editing, we developed a bioinformatics pipeline to identify and prioritize genome-wide intergenic loci and characterized 47 gRNAs located in 21 intergenic regions. These loci are screened for guide RNA cutting efficiency, integration efficiency of a gene cassette, the resulting cellular fitness, and GFP expression level. We further developed a landing pad system using components from these well-characterized loci, which can aid in the integration of multiple genes using single guide RNA and multiple repair templates of the user's choice. We have demonstrated the use of the landing pad for simultaneous integrations of 2, 3, 4, or 5 genes to the target loci with efficiencies greater than 80%. As a proof of concept, we showed how the production of 5-aminolevulinic acid can be improved by integrating five copies of genes at multiple sites in one step. We have further demonstrated the efficiency of this tool by constructing a metabolic pathway for succinic acid production by integrating five gene expression cassettes using a single guide RNA along with five different repair templates, leading to the production of 9 g/L of succinic acid in batch fermentations. This study demonstrates the effectiveness of a single gRNA-mediated CRISPR platform to build complex metabolic pathways in a non-conventional yeast. This landing pad system will be a valuable tool for the metabolic engineering of I. orientalis.

Position-dependent sequence motif preferences of SpCas9 are largely determined by scaffold-complementary spacer motifs

Streptococcus pyogenes Cas9 (SpCas9) nuclease exhibits considerable position-dependent sequence preferences. The reason behind these preferences is not well understood and is difficult to rationalise, since the protein establishes interactions with the target-spacer duplex in a sequence-independent manner. We revealed here that intramolecular interactions within the single guide RNA (sgRNA), between the spacer and the scaffold, cause most of these preferences. By using in cellulo and in vitro SpCas9 activity assays with systematically designed spacer and scaffold sequences and by analysing activity data from a large SpCas9 sequence library, we show that some long (>8 nucleotides) spacer motifs, that are complementary to the RAR unit of the scaffold, interfere with sgRNA loading, and that some motifs of more than 4 nucleotides, that are complementary to the SL1 unit, inhibit DNA binding and cleavage. Furthermore, we show that intramolecular interactions are present in the majority of the inactive sgRNA sequences of the library, suggesting that they are the most important intrinsic determinants of the activity of the SpCas9 ribonucleoprotein complex. We also found that in pegRNAs, sequences at the 3' extension of the sgRNA that are complementary to the SL2 unit are also inhibitory to prime editing, but not to the nuclease activity of SpCas9.

acCRISPR: an activity-correction method for improving the accuracy of CRISPR screens

High throughput CRISPR screens are revolutionizing the way scientists unravel the genetic underpinnings of engineered and evolved phenotypes. One of the critical challenges in accurately assessing screening outcomes is accounting for the variability in sgRNA cutting efficiency. Poorly active guides targeting genes essential to screening conditions obscure the growth defects that are expected from disrupting them. Here, we develop acCRISPR, an end-to-end pipeline that identifies essential genes in pooled CRISPR screens using sgRNA read counts obtained from next-generation sequencing. acCRISPR uses experimentally determined cutting efficiencies for each guide in the library to provide an activity correction to the screening outcomes via calculation of an optimization metric, thus determining the fitness effect of disrupted genes. CRISPR-Cas9 and -Cas12a screens were carried out in the non-conventional oleaginous yeast Yarrowia lipolytica and acCRISPR was used to determine a high-confidence set of essential genes for growth under glucose, a common carbon source used for the industrial production of oleochemicals. acCRISPR was also used in screens quantifying relative cellular fitness under high salt conditions to identify genes that were related to salt tolerance. Collectively, this work presents an experimental-computational framework for CRISPR-based functional genomics studies that may be expanded to other non-conventional organisms of interest.

Strategies and Methods for Improving the Efficiency of CRISPR/Cas9 Gene Editing in Plant Molecular Breeding

Following recent developments and refinement, CRISPR-Cas9 gene-editing technology has become increasingly mature and is being widely used for crop improvement. The application of CRISPR/Cas9 enables the generation of transgene-free genome-edited plants in a short period and has the advantages of simplicity, high efficiency, high specificity, and low production costs, which greatly facilitate the study of gene functions. In plant molecular breeding, the gene-editing efficiency of the CRISPR-Cas9 system has proven to be a key step in influencing the effectiveness of molecular breeding, with improvements in gene-editing efficiency recently becoming a focus of reported scientific research. This review details strategies and methods for improving the efficiency of CRISPR/Cas9 gene editing in plant molecular breeding, including Cas9 variant enzyme engineering, the effect of multiple promoter driven Cas9, and gRNA efficient optimization and expression strategies. It also briefly introduces the optimization strategies of the CRISPR/Cas12a system and the application of BE and PE precision editing. These strategies are beneficial for the further development and optimization of gene editing systems in the field of plant molecular breeding.

Precise DNA cleavage using CRISPR-SpRYgests

Methods for in vitro DNA cleavage and molecular cloning remain unable to precisely cleave DNA directly adjacent to bases of interest. Restriction enzymes (REs) must bind specific motifs, whereas wild-type CRISPR-Cas9 or CRISPR-Cas12 nucleases require protospacer adjacent motifs (PAMs). Here we explore the utility of our previously reported near-PAMless SpCas9 variant, named SpRY, to serve as a universal DNA cleavage tool for various cloning applications. By performing SpRY DNA digests (SpRYgests) using more than 130 guide RNAs (gRNAs) sampling a wide diversity of PAMs, we discovered that SpRY is PAMless in vitro and can cleave DNA at practically any sequence, including sites refractory to cleavage with wild-type SpCas9. We illustrate the versatility and effectiveness of SpRYgests to improve the precision of several cloning workflows, including those not possible with REs or canonical CRISPR nucleases. We also optimize a rapid and simple one-pot gRNA synthesis protocol to streamline SpRYgest implementation. Together, SpRYgests can improve various DNA engineering applications that benefit from precise DNA breaks.

Richardson R. New advances in CRISPR/Cas-mediated precise gene-editing techniques

Over the past decade, CRISPR/Cas-based gene editing has become a powerful tool for generating mutations in a variety of model organisms, from Escherichia coli to zebrafish, rodents and large mammals. CRISPR/Cas-based gene editing effectively generates insertions or deletions (indels), which allow for rapid gene disruption. However, a large proportion of human genetic diseases are caused by single-base-pair substitutions, which result in more subtle alterations to protein function, and which require more complex and precise editing to recreate in model systems. Precise genome editing (PGE) methods, however, typically have efficiencies of less than a tenth of those that generate less-specific indels, and so there has been a great deal of effort to improve PGE efficiency. Such optimisations include optimal guide RNA and mutation-bearing donor DNA template design, modulation of DNA repair pathways that underpin how edits result from Cas-induced cuts, and the development of Cas9 fusion proteins that introduce edits via alternative mechanisms. In this Review, we provide an overview of the recent progress in optimising PGE methods and their potential for generating models of human genetic disease.

EpiCas-DL: Predicting sgRNA activity for CRISPR-mediated epigenome editing by deep learning

CRISPR-mediated epigenome editing enables gene expression regulation without changing the underlying DNA sequence, and thus has vast potential for basic research and gene therapy. Effective selection of a single guide RNA (sgRNA) with high on-target efficiency and specificity would facilitate the application of epigenome editing tools. Here we performed an extensive analysis of CRISPR-mediated epigenome editing tools on thousands of experimentally examined on-target sites and established EpiCas-DL, a deep learning framework to optimize sgRNA design for gene silencing or activation. EpiCas-DL achieves high accuracy in sgRNA activity prediction for targeted gene silencing or activation and outperforms other available in silico methods. In addition, EpiCas-DL also identifies both epigenetic and sequence features that affect sgRNA efficacy in gene silencing and activation, facilitating the application of epigenome editing for research and therapy. EpiCas-DL is available at http://www.sunlab.fun:3838/EpiCas-DL.

Development and evaluation of a rapid RPA/CRISPR-based detection of Francisella tularensis

Francisella tularensis is a dangerous pathogen that causes an extremely contagious zoonosis in humans named tularemia. Given its low-dose morbidity, the potential to be fatal, and aerosol spread, it is regarded as a severe threat to public health. The US Centers for Disease Control and Prevention (CDC) has classified it as a category A potential agent for bioterrorism and a Tier 1 Select Agent. Herein, we combined recombinase polymerase amplification (RPA) with CRISPR/Cas12a system to select the F. tularensis target gene (TUL4), creating a two-pronged rapid and ultrasensitive diagnostic method for detecting F. tularensis. The real-time RPA (RT-RPA) assay detected F. tularensis within 10 min at a sensitivity of 5 copies/reaction, F. tularensis genomic DNA of 5 fg, and F. tularensis of 2 × 102 CFU/ml; the RPA-CRISPR/Cas12a assay detects F. tularensis within 40 min at a sensitivity of 0.5 copies/reaction, F. tularensis genomic DNA of 1 fg, and F. tularensis of 2 CFU/ml. Furthermore, the evaluation of specificity showed that both assays were highly specific to F. tularensis. More importantly, in a test of prepared simulated blood and sewage samples, the RT-RPA assay results were consistent with RT-PCR assay results, and the RPA-CRISPR/Cas12a assay could detect a minute amount of F. tularensis genomic DNA (2.5 fg). There was no nonspecific detection with blood samples and sewage samples, giving the tests a high practical application value. For example, in on-site and epidemic areas, the RT-RPA was used for rapid screening and the RPA-CRISPR/Cas12a assay was used for more accurate diagnosis.

CRISPR/Cas9 gRNA activity depends on free energy changes and on the target PAM context

A major challenge of CRISPR/Cas9-mediated genome engineering is that not all guide RNAs (gRNAs) cleave the DNA efficiently. Although the heterogeneity of gRNA activity is well recognized, the current understanding of how CRISPR/Cas9 activity is regulated remains incomplete. Here, we identify a sweet spot range of binding free energy change for optimal efficiency which largely explains why gRNAs display changes in efficiency at on- and off-target sites, including why gRNAs can cleave an off-target with higher efficiency than the on-target. Using an energy-based model, we show that local gRNA-DNA interactions resulting from Cas9 "sliding" on overlapping protospacer adjacent motifs (PAMs) profoundly impact gRNA activities. Combining the effects of local sliding for a given PAM context with global off-targets allows us to better identify highly specific, and thus efficient, gRNAs. We validate the effects of local sliding on gRNA efficiency using both public data and in-house data generated by measuring SpCas9 cleavage efficiency at 1024 sites designed to cover all possible combinations of 4-nt PAM and context sequences of 4 gRNAs. Our results provide insights into the mechanisms of Cas9-PAM compatibility and cleavage activation, underlining the importance of accounting for local sliding in gRNA design.

Genome-wide functional screens enable the prediction of high activity CRISPR-Cas9 and -Cas12a guides in Yarrowia lipolytica

Genome-wide functional genetic screens have been successful in discovering genotype-phenotype relationships and in engineering new phenotypes. While broadly applied in mammalian cell lines and in E. coli, use in non-conventional microorganisms has been limited, in part, due to the inability to accurately design high activity CRISPR guides in such species. Here, we develop an experimental-computational approach to sgRNA design that is specific to an organism of choice, in this case the oleaginous yeast Yarrowia lipolytica. A negative selection screen in the absence of non-homologous end-joining, the dominant DNA repair mechanism, was used to generate single guide RNA (sgRNA) activity profiles for both SpCas9 and LbCas12a. This genome-wide data served as input to a deep learning algorithm, DeepGuide, that is able to accurately predict guide activity. DeepGuide uses unsupervised learning to obtain a compressed representation of the genome, followed by supervised learning to map sgRNA sequence, genomic context, and epigenetic features with guide activity. Experimental validation, both genome-wide and with a subset of selected genes, confirms DeepGuide's ability to accurately predict high activity sgRNAs. DeepGuide provides an organism specific predictor of CRISPR guide activity that with retraining could be applied to other fungal species, prokaryotes, and other non-conventional organisms.

A Meta-Analysis of gRNA Library Screens Enables an Improved Understanding of the Impact of gRNA Folding and Structural Stability on CRISPR-Cas9 Activity

CRISPR systems are known to be inhibited by unwanted secondary structures that form within the guide RNA (gRNA). The minimum free energy of predicted secondary structures has been used in prediction algorithms. However, the types of structures as well as the degree to which a predicted structure can inhibit Cas9/gRNA activity is not well characterized. Here, we perform a meta-analysis of 39 published CRISPR-Cas9 data sets to understand better the role of secondary structures in inhibiting gRNA activity. We (1) identify two distinct inhibitory structures that can form, (2) measure the prevalence of these structures in existing gRNA library data sets, and (3) provide free energy cutoffs at which these structures become inhibitory. First, we show that hairpins that form within the targeting portion (spacer) of the gRNA, having a minimum free energy of <-5 kcal/mol, negatively impact gRNA activity. Second, we demonstrate that a longer hairpin can form between the spacer and the nexus portion of the gRNA scaffold. A duplex stability of this longer hairpin of <-15 kcal/mol negatively impacts gRNA activity. These cutoffs help to explain conflicting impacts of free energy values in different data sets, as well as provide a guideline for future gRNA designs.

Characterization of 67 Confirmed Clustered Regularly Interspaced Short Palindromic Repeats Loci in 52 Strains of Staphylococci

Staphylococcus aureus (S. aureus), which is one of the most important species of Staphylococci, poses a great threat to public health. Clustered regularly interspaced short palindromic repeats (CRISPR) and their CRISPR-associated proteins (Cas) are an adaptive immune platform to combat foreign mobile genetic elements (MGEs) such as plasmids and phages. The aim of this study is to describe the distribution and structure of CRISPR-Cas system in S. aureus, and to explore the relationship between CRISPR and horizontal gene transfer (HGT). Here, we analyzed 67 confirmed CRISPR loci and 15 companion Cas proteins in 52 strains of Staphylococci with bioinformatics methods. Comparing with the orphan CRISPR loci in Staphylococci, the strains harboring complete CRISPR-Cas systems contained multiple CRISPR loci, direct repeat sequences (DR) forming stable RNA secondary structures with lower minimum free energy (MFE), and variable spacers with detectable protospacers. In S. aureus, unlike the orphan CRISPRs away from Staphylococcal cassette chromosome mec (SCCmec), the complete CRISPR-Cas systems were in J1 region of SCCmec. In addition, we found a conserved motif 5'-TTCTCGT-3' that may protect their downstream sequences from DNA interference. In general, orphan CRISPR locus in S. aureus differed greatly from the structural characteristics of the CRISPR-Cas system. Collectively, our results provided new insight into the diversity and characterization of the CRISPR-Cas system in S. aureus.