Deep learning models incorporating endogenous factors beyond DNA sequences improve the prediction accuracy of base editing outcomes.

Adenine base editors (ABEs) and cytosine base editors (CBEs) enable the single nucleotide editing of targeted DNA sites avoiding generation of double strand breaks, however, the genomic features that influence the outcomes of base editing in vivo still remain to be characterized. High-throughput datasets from lentiviral integrated libraries were used to investigate the sequence features affecting base editing outcomes, but the effects of endogenous factors beyond the DNA sequences are still largely unknown. Here the base editing outcomes of ABE and CBE were evaluated in mammalian cells for 5012 endogenous genomic sites and 11,868 genome-integrated target sequences, with 4654 genomic sites sharing the same target sequences. The comparative analyses revealed that the editing outcomes of ABE and CBE at endogenous sites were substantially different from those obtained using genome-integrated sequences. We found that the base editing efficiency at endogenous target sites of both ABE and CBE was influenced by endogenous factors, including epigenetic modifications and transcriptional activity. A deep-learning algorithm referred as BE_Endo, was developed based on the endogenous factors and sequence information from our genomic datasets, and it yielded unprecedented accuracy in predicting the base editing outcomes. These findings along with the developed computational algorithms may facilitate future application of BEs for scientific research and clinical gene therapy.

Undetectable off-target effects induced by FokI catalytic domain in mouse embryos.

The FokI catalytic domain can be fused to various DNA binding architectures to improve the precision of genome editing tools. However, evaluation of off-target effects is essential for developing these tools. We use Genome-wide Off-target analysis by Two-cell embryo Injection (GOTI) to detect low-frequency off-target editing events in mouse embryos injected with FokI-based architectures. Specifically, we test FokI-heterodimers fused with TALENs, FokI homodimers fused with RYdCas9, or FokI catalytic domains alone resulting in no significant off-target effects. These FokI genome editing systems exhibit undetectable off-target effects in mouse embryos, supporting the further development of these systems for clinical applications.

PHD3 inhibits colon cancer cell metastasis through the occludin-p38 pathway.

Prolyl hydroxylase 3 (PHD3) hydroxylates HIFα in the presence of oxygen, leading to HIFα degradation. PHD3 inhibits tumorigenesis. However, the underlying mechanism is not well understood. Herein, we demonstrate that PHD3 inhibits the metastasis of colon cancer cells through the occludin-p38 MAPK pathway independent of its hydroxylase activity. We find that PHD3 inhibits colon cancer cell metastasis in the presence of the PHD inhibitor DMOG, and prolyl hydroxylase-deficient PHD3(H196A) suppresses cell metastasis as well. PHD3 controls the stability of the tight junction protein occludin in a hydroxylase-independent manner. We further find that PHD3-inhibited colon cancer cell metastasis is rescued by knockdown of occludin and that occludin acts as a negative regulator of cell metastasis, implying that PHD3 suppresses metastasis through occludin. Furthermore, knockdown of occludin induces phosphorylation of p38 MAPK, and the p38 inhibitor SB203580 impedes cell migration and invasion induced by occludin knockdown, indicating that occludin functions through p38. Moreover, knockdown of occludin enhances the expression of MKK3/6, the upstream kinase of p38, while overexpression of occludin decreases its expression. Our results suggest that PHD3 inhibits the metastasis of colon cancer cells through the occludin-p38 pathway independent of its hydroxylase activity. These findings reveal a previously undiscovered mechanism underlying the regulation of cancer cell metastasis by PHD3 and highlight a noncanonical hydroxylase-independent function of PHD3 in the suppression of cancer cells.

Cytosine base editors induce off-target mutations and adverse phenotypic effects in transgenic mice.

Base editors have been reported to induce off-target mutations in cultured cells, mouse embryos and rice, but their long-term effects in vivo remain unknown. Here, we develop a Systematic evaluation Approach For gene Editing tools by Transgenic mIce (SAFETI), and evaluate the off-target effects of BE3, high fidelity version of CBE (YE1-BE3-FNLS) and ABE (ABE7.10F148A) in ~400 transgenic mice over 15 months. Whole-genome sequence analysis reveals BE3 expression generated de novo mutations in the offspring of transgenic mice. RNA-seq analysis reveals both BE3 and YE1-BE3-FNLS induce transcriptome-wide SNVs, and the numbers of RNA SNVs are positively correlated with CBE expression levels across various tissues. By contrast, ABE7.10F148A shows no detectable off-target DNA or RNA SNVs. Notably, we observe abnormal phenotypes including obesity and developmental delay in mice with permanent genomic BE3 overexpression during long-time monitoring, elucidating a potentially overlooked aspect of side effects of BE3 in vivo.

Optimization of C-to-G base editors with sequence context preference predictable by machine learning methods.

Efficient and precise base editors (BEs) for C-to-G transversion are highly desirable. However, the sequence context affecting editing outcome largely remains unclear. Here we report engineered C-to-G BEs of high efficiency and fidelity, with the sequence context predictable via machine-learning methods. By changing the species origin and relative position of uracil-DNA glycosylase and deaminase, together with codon optimization, we obtain optimized C-to-G BEs (OPTI-CGBEs) for efficient C-to-G transversion. The motif preference of OPTI-CGBEs for editing 100 endogenous sites is determined in HEK293T cells. Using a sgRNA library comprising 41,388 sequences, we develop a deep-learning model that accurately predicts the OPTI-CGBE editing outcome for targeted sites with specific sequence context. These OPTI-CGBEs are further shown to be capable of efficient base editing in mouse embryos for generating Tyr-edited offspring. Thus, these engineered CGBEs are useful for efficient and precise base editing, with outcome predictable based on sequence context of targeted sites.

Programmable RNA editing with compact CRISPR-Cas13 systems from uncultivated microbes.

Competitive coevolution between microbes and viruses has led to the diversification of CRISPR-Cas defense systems against infectious agents. By analyzing metagenomic terabase datasets, we identified two compact families (775 to 803 amino acids (aa)) of CRISPR-Cas ribonucleases from hypersaline samples, named Cas13X and Cas13Y. We engineered Cas13X.1 (775 aa) for RNA interference experiments in mammalian cell lines. We found Cas13X.1 could tolerate single-nucleotide mismatches in RNA recognition, facilitating prophylactic RNA virus inhibition. Moreover, a minimal RNA base editor, composed of engineered deaminase (385 aa) and truncated Cas13X.1 (445 aa), exhibited robust editing efficiency and high specificity to induce RNA base conversions. Our results suggest that there exist untapped bacterial defense systems in natural microbes that can function efficiently in mammalian cells, and thus potentially are useful for RNA-editing-based research.

GOTI, a method to identify genome-wide off-target effects of genome editing in mouse embryos.

Genome editing holds great potential for correcting pathogenic mutations. We developed a method called GOTI (genome-wide off-target analysis by two-cell embryo injection) to detect off-target mutations by editing one blastomere of two-cell mouse embryos using either CRISPR-Cas9 or base editors. GOTI directly compares edited and non-edited cells without the interference of genetic background and thus could detect potential off-target variants with high sensitivity. Notably, the GOTI method was designed to detect potential off-target variants of any genome editing tools by the combination of experimental and computational approaches, which is critical for accurate evaluation of the safety of genome editing tools. Here we provide a detailed protocol for GOTI, including mice mating, two-cell embryo injection, embryonic day 14.5 embryo digestion, fluorescence-activated cell sorting, whole-genome sequencing and data analysis. To enhance the utility of GOTI, we also include a computational workflow called GOTI-seq (https://github.com/sydaileen/GOTI-seq) for the sequencing data analysis, which can generate the final genome-wide off-target variants from raw sequencing data directly. The protocol typically takes 20 d from the mice mating to sequencing and 7 d for sequencing data analysis.

Single C-to-T substitution using engineered APOBEC3G-nCas9 base editors with minimum genome- and transcriptome-wide off-target effects.

Cytosine base editors (CBEs) enable efficient cytidine-to-thymidine (C-to-T) substitutions at targeted loci without double-stranded breaks. However, current CBEs edit all Cs within their activity windows, generating undesired bystander mutations. In the most challenging circumstance, when a bystander C is adjacent to the targeted C, existing base editors fail to discriminate them and edit both Cs. To improve the precision of CBE, we identified and engineered the human APOBEC3G (A3G) deaminase; when fused to the Cas9 nickase, the resulting A3G-BEs exhibit selective editing of the second C in the 5'-CC-3' motif in human cells. Our A3G-BEs could install a single disease-associated C-to-T substitution with high precision. The percentage of perfectly modified alleles is more than 6000-fold for disease correction and more than 600-fold for disease modeling compared with BE4max. On the basis of the two-cell embryo injection method and RNA sequencing analysis, our A3G-BEs showed minimum genome- and transcriptome-wide off-target effects, achieving high targeting fidelity.

A rationally engineered cytosine base editor retains high on-target activity while reducing both DNA and RNA off-target effects.

Cytosine base editors (CBEs) offer a powerful tool for correcting point mutations, yet their DNA and RNA off-target activities have caused concerns in biomedical applications. We describe screens of 23 rationally engineered CBE variants, which reveal mutation residues in the predicted DNA-binding site can dramatically decrease the Cas9-independent off-target effects. Furthermore, we obtained a CBE variant-YE1-BE3-FNLS-that retains high on-target editing efficiency while causing extremely low off-target edits and bystander edits.

Cytosine base editor generates substantial off-target single-nucleotide variants in mouse embryos.

Genome editing holds promise for correcting pathogenic mutations. However, it is difficult to determine off-target effects of editing due to single-nucleotide polymorphism in individuals. Here we developed a method named GOTI (genome-wide off-target analysis by two-cell embryo injection) to detect off-target mutations by editing one blastomere of two-cell mouse embryos using either CRISPR-Cas9 or base editors. Comparison of the whole-genome sequences of progeny cells of edited and nonedited blastomeres at embryonic day 14.5 showed that off-target single-nucleotide variants (SNVs) were rare in embryos edited by CRISPR-Cas9 or adenine base editor, with a frequency close to the spontaneous mutation rate. By contrast, cytosine base editing induced SNVs at more than 20-fold higher frequencies, requiring a solution to address its fidelity.

Prolyl hydroxylase 2 is dispensable for homeostasis of intestinal epithelium in mice.

Prolyl hydroxylases (PHD1-3) hydroxylate hypoxia inducible factor α (HIFα), leading to HIFα ubiquitination and degradation. Recent studies indicated that administration of generic inhibitors of PHDs improved mice colitis, suggesting that suppression of PHD activity by these inhibitors may be a potential strategy for the treatment of inflammatory bowel diseases. However, the exact role of each member of PHD family in homeostasis of intestinal epithelium remains elusive. The aim of this work is to study the possible role of PHD2 by using mice with genetic ablation of Phd2 in intestinal epithelial cells (IECs). We found that deletion of PHD2 in IECs did not lead to spontaneous enteritis or colitis in mice. Deletion of PHD2 in IECs did not confer upon mice higher susceptibility to dextran sodium sulfate-induced colitis. Furthermore, in a colitis-associated colon cancer model, the PHD2-conditional knockout mice had similar susceptibility to azoxymethane (AOM)-induced colonic tumorigenesis as control mice did. Our results suggest that PHD2 is dispensable for maintenance of intestinal epithelium homeostasis in mice.

PHD3 Stabilizes the Tight Junction Protein Occludin and Protects Intestinal Epithelial Barrier Function.

Prolyl hydroxylase domain proteins (PHDs) control cellular adaptation to hypoxia. PHDs are found involved in inflammatory bowel disease (IBD); however, the exact role of PHD3, a member of the PHD family, in IBD remains unknown. We show here that PHD3 plays a critical role in maintaining intestinal epithelial barrier function. We found that genetic ablation of Phd3 in intestinal epithelial cells led to spontaneous colitis in mice. Deletion of PHD3 decreases the level of tight junction protein occludin, leading to a failure of intestinal epithelial barrier function. Further studies indicate that PHD3 stabilizes occludin by preventing the interaction between the E3 ligase Itch and occludin, in a hydroxylase-independent manner. Examination of biopsy of human ulcerative colitis patients indicates that PHD3 is decreased with disease severity, indicating that PHD3 down-regulation is associated with progression of this disease. We show that PHD3 protects intestinal epithelial barrier function and reveal a hydroxylase-independent function of PHD3 in stabilizing occludin. These findings may help open avenues for developing a therapeutic strategy for IBD.

The Endoplasmic Reticulum Stress Sensor IRE1α in Intestinal Epithelial Cells Is Essential for Protecting against Colitis.

Intestinal epithelial cells (IECs) have critical roles in maintaining homeostasis of intestinal epithelium. Endoplasmic reticulum (ER) stress is implicated in intestinal epithelium homeostasis and inflammatory bowel disease; however, it remains elusive whether IRE1α, a major sensor of ER stress, is directly involved in these processes. We demonstrate here that genetic ablation of Ire1α in IECs leads to spontaneous colitis in mice. Deletion of IRE1α in IECs results in loss of goblet cells and failure of intestinal epithelial barrier function. IRE1α deficiency induces cell apoptosis through induction of CHOP, the pro-apoptotic protein, and sensitizes cells to lipopolysaccharide, an endotoxin from bacteria. IRE1α deficiency confers upon mice higher susceptibility to chemical-induced colitis. These results suggest that IRE1α functions to maintain the intestinal epithelial homeostasis and plays an important role in defending against inflammation bowel diseases.