<span class="paragraphSection"><div class="boxTitle">Abstract</div>Oligodendrocytes (OLs), the myelinating cells of the central nervous system, are particularly prone to pathogenic G-to-A mutations, such as <span style="font-style: italic;">PLP1A243V</span>, which causes Pelizaeus–Merzbacher disease (PMD), a lethal hypomyelinating disorder lacking effective therapy. Although adenine base editors (ABEs) can in principle correct such mutations, their application in OLs is limited by inefficient on-target editing. Here, we develop a spatially concentrated ABE (cABE) strategy that enhances editing by promoting nuclear translocation of tRNA adenosine deaminase (TadA*) from the cytoplasm. Using a SunTag-based multivalent recruitment system, TadA* is locally enriched at genomic targets (cABE-1.0), achieving robust editing <span style="font-style: italic;">in vitro</span>. To enable <span style="font-style: italic;">in vivo</span> delivery while preserving high efficiency and fidelity, SpCas9 is replaced with compact eNme2-C Cas9, generating an AAV-compatible variant (cABE-2.0). Notably, cABE-2.0 forms dynamic nuclear puncta with properties of liquid–liquid phase separation, enhancing on-target editing while substantially reducing transcriptome-wide RNA off-target effects. Functionally, cABE-2.0 efficiently corrects the <span style="font-style: italic;">PLP1A243V</span> mutation in OLs, restores Plp subcellular localization, and rescues myelination-related phenotypes. These findings demonstrate that spatial reorganization, rather than increasing intrinsic catalytic activity of TadA*, provides a distinct principle for improving base editing in difficult-to-edit cell types, such as OLs, offering a mechanistic and technical framework for gene therapy of PMD and related myelin diso