Cyclic lipopeptides from Bacillus subtilis are potent antimicrobial agents with promising applications in sustainable agriculture and postharvest preservation. While surfactin and plipastatin share glutamic acid as the substrate incorporated by the initiation module, natural plipastatin production lags 10-90-fold behind surfactin in wild-type strains. We hypothesized that substituting plipastatin's glutamic acid domain with surfactin's counterpart could boost yield. Using B. subtilis ΔppsΔsrf as a chassis, we integrated B. amyloliquefaciens HYM12's plipastatin cluster to create monoproducer M-25 (359.55 mg/L titer). Three domain-swapping strategies, XU (C-A-T-C units), XUC (CAsub-A-T-CDsub units), and XUT (exchanges T-domain-defined units), were tested. While XU/XUC disrupted recognition of the d-ornithine residue immediately downstream (abolishing production), XUT's targeted replacement of a T-domain flexible loop (FFERGGHSL) maintained substrate specificity, yielding 583.96 mg/L (62% increase). Fermentation analyses revealed that M-25XUT produced six novel C14-C16-predominant homologs while reducing C17+ variants. Promoter engineering via CRISPR/Cas9 (replacing the native promoter with Psrf09) further increased titer to 612.45 mg/L (70.3% total improvement). Antimicrobial assays confirmed enhanced bioactivity against pathogens. This work demonstrates a novel NRPS engineering paradigm for lipopeptide optimization, advancing both mechanistic understanding and translational applications.