PSBP-Fused Prime Editing Platform Dramatically Improves Precision Base Transition in Human Target Genes
Background: Molecular limitations of existing reverse transcription rate-limiting steps and therapeutic efficacy data bottlenecks in rare genetic disease drug R&D.
Prime Editing (PE) technology has emerged as an innovative solution for correcting single base substitutions, insertions, and deletions by targeting a narrow range of genetic variations while minimizing the induction of double-strand breaks (DSBs). However, the correction efficiency in actual human primary cells and clinical efficacy models is extremely heterogeneous. In particular, the spontaneous dissociation tendency and low processivity of reverse transcriptase (RT) have acted as critical limitations, preventing the complex from overcoming the physical access barrier to the target genomic site. Existing, linear, and static gene editing analysis standard guidelines cannot predict the differences in chromatin three-dimensional structure density, the active noise of DNA replication enzyme complexes at each cell cycle stage, and steric interference with repair proteins. As a result, it has consistently failed to control the preventive and therapeutic efficacy concentrations that exert actual target protein activity under post-engraftment negative feedback flux or fluctuations in the molecular density of the microenvironment. In particular, genomic structural collapse noise occurring during cell dissociation and the mean value bias of microarrays and bulk sequencing induce different off-target variations in each batch, creating a significant barrier to achieving toxicological safety data milestones for IND approval. This molecular dynamic uncertainty has resulted in a critical bottleneck in the development of pipelines that require precise substitution of target genetic variations, such as cystic fibrosis and chronic granulomatous disease (CGD), where the therapeutic efficacy threshold is high, and the optimal candidate docking flux cannot be derived.
Discovery: Activation of Polymerase Substrate Binding Protein (PSBP) fusion modality and demonstration of single-cell resolution genome editing tensor synchronization.
In this study, we mimic the natural polymerase's inherent substrate binding module mechanism and activated an innovative PSBP-RT fusion architecture by chemically linking NHP6A, a small HMG box domain-derived protein, to reverse transcriptase (RT). The introduction of PSBP precisely down-regulates the binding free energy between the prime editor complex and the 3' terminal template of pegRNA, autonomously adjusting the initiation rate constant of the reverse transcription reaction to match the theoretical model predicted in silico. The NHP6A-RT module, identified through screening of 12 subunits, demonstrated performance that disruptively surpassed the thermodynamic fusion limitations of existing stand-alone prime editors (PE2) in single base substitution, multiple base deletion, and insertion pathways. Furthermore, this system maximizes the synergy with MLH1dn (PE4), which inhibits mismatch repair mechanisms, epegRNA, which maintains 3' structural stability, and the pegRNA microenvironment stabilization infrastructure (PE7) via La protein, dramatically increasing the correction yield within the complex genomic gradient. The resulting micro-indels are maintained below the lowest baseline, and the topological network variation curve of the entire transcriptome is tracked in real-time, demonstrating at the molecular biological level that the frequency of off-target mutations is dramatically reduced compared to existing systems.
Establishment of a model for regulating reverse transcription processivity and reversible, layered homeostasis.
The fusion behavior of the PSBP-RT platform continuously alleviates the energy barrier of the single nucleic acid docking step, enabling the establishment of an optimal drug delivery model for each patient group based on the genomic omics matrix. In particular, by controlling the rate-limiting step constant derived from the differential equation based on the single nucleotide incorporation rate with the molecular energy down-clamping and up-clamping mechanisms, the stable expression rate of the target gene can be freely set. This autonomous rate control acts as a backbone data platform that enables precise stratification of patients with unique genetic variations. Even when the DNA replication stress molecular feedback pathway is disrupted by external stress, the modified prime editor complex maintains autonomous chemical equilibrium, demonstrating unique molecular biological precision in restoring reversible homeostasis in real-time.
Prospects: Establishing a standard for programmable therapeutic genomics and launching a next-generation IND digital governance.
With this, we have reached a commercialization milestone where we can completely replace the traditional molecular biology R&D governance's chronic bottleneck, the static, post-hoc candidate selection framework, with a programmable computational genomics infrastructure based on a multi-dimensional tensor model. In the competitive landscape of the next-generation prime editing therapeutics market, such as PM-359 (Phase 1/2 clinical trial) currently being developed by Prime Medicine as a treatment for chronic granulomatous disease, this architecture perfectly synchronizes the in silico gradient correction coefficient for zeroing out batch-to-batch variations from the high-throughput screening (HTS) stage. This provides a computational moat that completely blocks the genetic batch effect noise that multinational pharmaceutical companies experience when scaling up (cGMP). Ultimately, it will fully meet the stringent validity specifications of the precision companion diagnostic (CDx) biomarker panel and serve as a unique core digital asset that will disruptively shorten the toxicological test validation timeline in the clinical trial protocol (IND) and marketing approval evaluation stages, leading the next-generation bio-market.
Prime editing (PE) enables precise small DNA changes yet often shows modest efficiency in human cells. Recent advances indicate that the activity of the reverse transcriptase (RT) component is a key determinant of PE performance. Here, we adapt a principle from natural polymerases by fusing a polymerase substrate binding protein (PSBP) with RT to enhance PE. In a twelve-member screen, multiple PSBPs increased editing efficiency, with a compact HMG box module, NHP6A, emerging as a representative lead fusion. NHP6A fused to RT significantly increases intended edits across substitutions, insertions, and deletions while indels remain low. The enhancement is broadly compatible with a nicking sgRNA (PE3), transient mismatch repair inhibition via MLH1dn (PE4), and structured 3' pegRNA extensions (epegRNA). The PSBP module also synergizes with the La protein-assisted pegRNA tail stabilization (the PE7 system) for additional improvements. Using NHP6A together with La, we efficiently installed four clinically relevant alleles with high product purity. Finally, we validate that PSBP fusion can be consolidated into a single component prime editor without loss of activity. These results establish PSBP fusion as a precise route to improve prime editing outcomes and support integrating compact accessory modules into next-generation PE platforms.
This study's PSBP-fused prime editing technology goes beyond theoretical molecular replication mechanism exploration and is directly applied to the actual global finished pharmaceutical market and the next-generation precision personalized bio-business line.
First, in the clinical setting, by instantly scanning the rate of mutations causing chronic granulomatous disease and cystic fibrosis using a Python algorithm and computational structure scanning, the temporal blank noise that occurs during conventional microarray analysis is eliminated at the source, and a concrete protective barrier is maintained to ensure patient-specific therapeutic efficacy.
At the same time, by linking to the open-source gnomAD database, which aggregates large-scale variation datasets and omics matrices, a companion diagnostic (CDx) panel interface is realized that can virtually simulate confounding variables of off-target mutations in clinical trial design and real-time reverse-calculate the effective docking concentration of the target genome.
Furthermore, when multinational companies conduct large-scale approval clinical trials for next-generation base insertion and substitution therapeutics, by linking the reverse transcription processivity binding free energy of the NHP6A-RT fusion as a correction coefficient, the batch-to-batch genome editing efficiency variation is eliminated, and the backbone infrastructure that maximizes the probability of obtaining clinical trial protocols and cGMP commercial operation approvals from global regulatory agencies is provided.