Minimizing far-extending chromatin perturbation in genome editing preserves stem cell identity [Long Read Amplicon Sequencing]

While CRISPR/Cas9 holds therapeutic promise, broader application demands understanding complications in vast non-coding regions. We found that CRISPR/Cas9 can cause premature differentiation of neural stem cells in vivo and mouse embryonic stem cells in vitro, even when cleavage occurred at distant sites tens of kilobases away from the nearest regulatory elements. To investigate this, we employed an integrated ATAC/RNA approach (AR-seq) and identified editing-induced chromatin accessibility change, with its scale varying by cell types. Cells with stemness are most affected, experiencing perturbations that extend over a hundred kilobases. Furthermore, even local DNA perturbations can disrupt CTCF- and condensate-associated chromatin architecture, causing distal transcriptional rewiring and ultimately loss of stemness identity. To minimize chromatin perturbations and preserve cell identity we refined gene editing strategies, including distance-aware sgRNA design, pharmacological attenuation of DNA resection, and alternative editing systems. This work paves the way for safer and broader application of genome editing technologies. Overall design: To rigorously test whether CRISPR-induced phenotypic disruption near stemness-associated regulatory regions is driven by direct structural damage at the edited locus, we implemented an orthogonal validation strategy centered on long-read amplicon Nanopore sequencing. The design rationale was to overcome the resolution limits of conventional short-read assays and gel-based screening, and to quantify the full spectrum of local repair outcomes at single-molecule scale, with particular sensitivity to larger deletion classes. Conceptually, this assay was positioned as a mechanistic discriminator: if distal transcriptional and cell-state perturbations were primarily a consequence of extensive on-target sequence loss, a substantial burden of long deletions would be expected in edited populations. Accordingly, we profiled locus-spanning long-read amplicons from edited and matched control samples, then quantified deletion-length distributions relative to the nuclease cut site. The analytical framework was designed to distinguish rare large-deletion events from the dominant indel background while controlling for amplification-related representation bias, enabling robust estimation of the structural contribution to phenotype. In parallel with PEM-seq and targeted off-target interrogation, this design provided convergent evidence for interpreting causality: structural lesions were detectable but limited in prevalence, supporting a model in which widespread chromatin/transcriptional rewiring cannot be explained solely by frequent large on-target deletions.