Advanced Topics in DNA Editing: Repair Manipulation, Base Editing, and Large Serine Recombinases

Learning Outcomes and Theme Overview

• Overall Theme: The core theme of the lecture series is making directed, precise changes to genome sequences.

• Lecture 1 Learning Outcomes:     * Understanding the critical importance of cellular DNA repair pathways in the context of gene editing.     * Appreciating the specific limitations associated with Homology-Directed Repair (HDR).     * Understanding the systematic improvement of base editing outcomes through repair pathway manipulation.     * Re-introduction to the mechanisms and logic of Prime Editing.

• Lecture 2 Learning Outcomes:     * Understanding the mechanics of site-specific recombination facilitated by serine recombinases.     * Learning how Large Serine Recombinases (LSRs) are utilized for the targeted insertion of large DNA cargoes.     * Understanding methods for screening and identifying improved LSR variants.     * Understanding the combination of Prime Editing and LSR technologies for programmable cargo insertion.

Double-Strand Break (DSB) Repair and Its Limitations

• DSB-Dependent Strategies:     * Traditional editing utilizes Cas9 and Cas12a to create Double-Strand Breaks (DSBs).     * Cas9: Uses a gRNA, and contains RuvC and HNH nuclease domains.     * Cas12a: Uses a crRNA and contains RuvC domains.     * Homology-Independent Targeted Integration (HITI): A strategy for targeted integration that does not require homology, though it often results in random orientation of the insert.

• Toxicities and Risks of DSBs:     * Creation of DSBs for NHEJ (Non-Homologous End Joining) or HR (Homologous Recombination) can lead to deleterious outcomes, including:         * Deletions recorded at the target site.         * Chromosomal translocations.         * Chromothripsis (shattering and reassembly of chromosomes).         * Insertion and activation of retrotransposons.         * Activation of the $p53$ DNA damage response pathway.

• The Precision Goal: To reduce DSB-related toxicity, strategies seek to make small precision changes (typically < 30 ext{ nt}) without inducing full DSBs.

Development of Base Editing and First Generation Models (BE1)

• Base Editor Fundamentals: These systems typically use a Cas9 nickase ($\Delta RuvC$) fused to a deaminase enzyme to effect point mutations without DSBs.

• Types of Mutations: Base editors can address six out of the $12$ possible types of point mutations:     * $C \bullet G ightarrow T \bullet A$     * $A \bullet T ightarrow G \bullet C$     * $C \bullet G ightarrow G \bullet C$

• First Generation (BE1) Architecture:     * Components: Dead Cas9 (dCas9) fused to a Cytidine deaminase.     * rAPOBEC1: Rat-derived cytosine deaminase (Apolipoprotein B MRNA Editing Enzyme Catalytic Subunit 1).     * Mechanism: Converts Cytosine ($C$) to Uracil ($U$) within a specific editing window in the R-loop.

• BE1 Limitations in Human Cells:     * Editing efficiency is low, ranging from $0.8 ext{--} 7.7\%$.     * Results in unintended INDELs (insertions/deletions).     * Repair Interference: Cellular DNA repair pathways often subvert the desired edit. Uracil is rapidly recognized as an error.

Second Generation Cytosine Base Editors (BE2) and Uracil Repair Inhibition

• The DNA Repair Problem:     * $C ightarrow U$ deamination creates a $U:G$ mismatch.     * Base Excision Repair (BER): Uracil DNA Glycosylase (UNG) identifies and excises the Uracil, creating an abasic (AP) site.     * Desired vs. Undesired Outcomes: If replication occurs before repair, the desired $C:G ightarrow T:A$ transition happens. If BER occurs, the cell restores the original $C:G$ pair or uses error-prone polymerases, leading to INDELs.

• BE2 Architecture:     * Incorporates a Uracil DNA Glycosylase Inhibitor (UGI) at the C-terminus of the BE1 construct.     * UGI Source: Derived from the Bacillus subtilis bacteriophage PBS1. Phage PBS1 contains uracil in its genome and uses UGI to prevent the host's UNG from degrading its DNA.     * UGI Characteristics: A small protein ($9.5 ext{ kDa}$) that acts as a DNA mimetic, reversibly inhibiting host UNG.

• Efficiency Gains: Addition of UGI leads to a threefold increase in editing efficiency in human cells compared to BE1.

Third and Fourth Generation CBEs: Manipulating Mismatch Repair (MMR)

• Third Generation (CBE3) Logic:     * Uses Cas9 nickase (D10A) which has a mutated RuvC but active HNH domain.     * The nick occurs on the non-edited Target Strand (TS).     * Bias Strategy: In Eukaryotes (using MSH2-MSH6, MLH1-PMS2, ExoI, etc.), the repair machinery is directed to the strand containing the "nick" (the daughter strand). By nicking the non-edited strand, the cell is tricked into using the edited (U-containing) strand as a template for repair.

• CBE3 Performance:     * Increases editing efficiency by six-fold over BE2.     * Results in a slight increase in INDEL frequency ($1.1\%$) compared to BE2 ($0.1\%$), though still less than standard DSB methods.

• Fourth Generation (CBE4) Logic:     * Aims to prevent INDELs resulting from abasic sites.     * Architecture: Features $2$ copies of UGI with optimized linkers.     * Outcome: Reduced conversion of $C:G$ into $G:C$ or $A:T$ (unintended transitions) and significantly reduced INDEL frequency.

Prime Editing Mechanisms

• Prime Editor Architecture:     * Uses a Cas9 nickase ($H840A$) where the HNH domain is dead, but it nicks the non-targeted strand.     * Fused to an engineered Reverse Transcriptase (RT) derived from the Moloney Murine Leukemia Virus (M-MuLV or MMLV).     * pegRNA (Prime Editing gRNA): Includes a spacer, a Primer Binding Site (PBS), and a Reverse Transcription Template (RTT).

• The Process:     1. Cas9 nicks the DNA.     2. The PBS on the pegRNA hybridizes to the nicked strand.     3. The RT synthesizes new DNA using the RTT as a template, incorporating the desired edit.     4. The system enables precise nucleotide changes, insertions, deletions, or combinations thereof.     5. Bias strategies similar to CBEs are used to encourage the cell to retain the edited strand during mismatch repair.

Principles of Serine Recombinases

• Classes of Recombinases:     1. Conservative Site-Specific:         * Serine Recombinases.         * Tyrosine Recombinases (e.g., Lambda Integrase - which has more complex requirements).     2. Transpositional: "Jumping genes."

• General Functions: Recombination involves bringing DNA sites together (synapsis), cleavage, and re-joining.

• Serine Recombinase Categories:     * Integrases: Facilitate insertion.     * Resolvases: Facilitate deletion.     * Invertases: Facilitate inversion.

• The Catalytic Mechanism:     * Utilizes a cut-and-rejoin mechanism.     * Covalent Intermediate: DSBs are not released to the cell; DNA ends are protected by a protein-DNA covalent intermediate via a serine residue.     * Steps: Synaptic tetramer formation (from two pre-synaptic dimers) $\rightarrow$ Cleavage of all four strands $\rightarrow$ $180^{\circ}$ rotation (swapping DNA partners) $\rightarrow$ Rejoining.     * No requirement for endogenous cellular repair factors (NHEJ/HR).

Large Serine Recombinases (LSRs) and Attachment Sites

• Biological Origin: LSRs evolved in temperate bacteriophages to manage the decision between lytic growth and lysogeny.

• Bxb1 System (Mycobacterium smegmatis):     * Attachment Sites (att): The phage genome contains an $attP$ site; the host genome contains an $attB$ site.     * Recombination: Integrase mediates $attP + attB \rightarrow attL + attR$. The integrated prophage is flanked by $attL$ and $attR$.     * Directionality: Recombination is unidirectional. Reversal ($attL + attR \rightarrow attP + attB$) requires the Integrase plus a Recombination Directionality Factor (RDF).

• Editing Applications:     * $\phi C31$: Can target "pseudo-sites" in the human genome (resembling $attB$), but with low efficiency (< 1\%).     * Pre-installed Landing Pads: Using HR or Prime Editing to install an $attB$ site, followed by Bxb1 treatment to insert payloads (e.g., $27 ext{ kb}$) at efficiencies of approximately $10\%$.

Expanding the LSR Toolkit

• Modern Screening Efforts:     * Researchers collected $194,585$ genomes to identify candidates.     * Identified $12,638$ candidate LSRs and their predicted attachment sites.     * Efficiency Benchmarks: Some newly identified LSRs show up to seven-fold higher plasmid recombination than Bxb1.     * Genome Insertion: Specific LSRs (e.g., Ec03, Ec04, Kp03, Pa01) achieved efficiencies of $40\% ext{--} 75\%$ with cargo sizes over $7 ext{ kb}$.

• Off-target Integration:     * $\Phi C31$ exhibits higher off-target integration without a landing pad.     * Kp03 was found to have approximately $730$ off-target sites, whereas Pa01 had around $16$ sites.     * Identified $856$ LSRs with significant matches for at least one site in the human genome.

Programmable Addition via Site-specific Targeting Elements (PASTE)

• The Logic of PASTE: Combines the programmability of CRISPR/Prime Editing with the cargo capacity of LSRs.

• Process Flow:     1. Prime Editing is used to install an $attB$ landing pad at a specific genomic locus.     2. An LSR (Integrase) is used to insert a large DNA cargo into that installed $attB$ site.

• PASTE Generations:     * PASTEv1: Utilized Bxb1 integrase and Prime Editing.     * PASTEv2 & v3: Optimized linkers, RT variants, and attachment sequence lengths.     * PASTEv4: Utilizes Bacillus cereus BceINTc integrase.

• Capabilities and Performance:     * Demonstrated insertion of a $36 ext{-kb}$ cargo at the ACTB locus.     * Comparison to HDR/HITI: PASTE shows significantly higher on-target integration and dramatically lower INDEL rates.     * Cell Cycle Independence: Unlike HDR, which requires the S/G2 phase, PASTE remains effective even in the presence of inhibitors like aphidicolin (which blocks DNA synthesis), making it viable for non-dividing cells.

Future Directions and Phage Defense Systems

• Sources for New Biotech Tools: Phage-host evolutionary conflicts provide numerous enzymes:     * Restriction-Modification Systems: Led to modern molecular cloning.     * CRISPR-Cas: Led to gene editing and diagnostics.     * Recombinant DNA Enzymes: Source of ligases and polymerases.     * LSRs: Current frontier for large-scale genomic insertion.

• Ongoing Research: Exploration of small protein inhibitors and other RNA-based phage defense mechanisms as potential editing regulators.