**Background:** Fusarium wilt of banana, caused by Fusarium oxysporum f. sp. cubense (Foc), is the most destructive disease in banana production history. The emergence of tropical race 4 (Foc TR4) threatens the global Cavendish-based banana industry, with projections of a 2.0% drop in global output and 240,000 job losses by 2028. Management is extremely challenging because Foc survives in soil for up to 30 years as chlamydospores, resistant cultivars may not meet market demand, and the pathogen evades non-endophytic control measures once inside the vascular system. Endophytic bacteria, which occupy similar ecological niches to vascular pathogens, are promising biocontrol agents. Banana propagation relies on tissue culture under aseptic conditions, producing plants that are vulnerable upon transfer to the field. This study aimed to isolate an endophytic bacterium from a healthy banana plant in a diseased field, characterize its antagonistic activity against Foc TR4, and evaluate its ability to protect tissue-cultured banana plantlets through bio-priming at the rooting stage.
**Methods:** Endophytic bacteria were isolated from surface-sterilized healthy banana plants collected from a wilt-diseased field in Dongguan, Guangdong, China. Strain EB1 was selected based on dual-culture antagonism assays against Foc TR4 strain II5 (VCG01213). Inhibition was quantified as percentage area reduction of mycelial growth. Morphological and ultrastructural changes in Foc TR4 hyphae were examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Whole-genome sequencing was performed using Oxford Nanopore and Illumina platforms. EB1 was identified via 16S rRNA phylogenetic analysis. Colonization dynamics were assessed by colony-forming unit (c.f.u.) counts in root and shoot tissues over 7 days. For biocontrol assays, banana tissue culture plantlets (cv. 'Brazilian') were inoculated with EB1 (OD600 = 0.2, ~10^6 c.f.u.) at the rooting stage, then after 7 days were acclimatized in pots with sterilized soil. Treatments included: control, EB1 only, Foc TR4 only (1,000 conidia/g soil), and EB1 + TR4. Plant height, shoot and root fresh weight, survival rates, and disease index (0–4 scale) were recorded after 60 days. Expression of defense genes NPR1, PR1, LOX2, and MYC2 was quantified by RT-qPCR using the 2^−ΔΔCt method with qTUB as reference.
**Key Results:** EB1 showed strong in vitro inhibition of Foc TR4 mycelial growth (75.43% area inhibition). SEM and TEM revealed severe hyphal deformities including irregular distortion, thickened cell walls, plasmolysis, vacuolation, and degenerated organelles. Whole-genome sequencing identified EB1 as Bacillus velezensis with a 3,929,912 bp circular chromosome (46.5% GC) containing 3,622 ORFs. Twelve biosynthetic gene clusters for secondary metabolites (surfactin, bacilysin, bacillibactin, difficidin, fengycin, bacillaene, macrolactin, butirosin) were identified. EB1 colonized roots rapidly, reaching 1.20 × 10^8 c.f.u./g by day 5 and 1.53 × 10^8 c.f.u./g by day 7. Shoot colonization reached 1.96 × 10^7 c.f.u./g by day 7. SEM showed EB1 in grooves between root epidermal cells and intercellular spaces. In the greenhouse, EB1 alone significantly increased shoot biomass by 1.22-fold (p = 0.002) and root biomass by 1.49-fold (p = 0.007) versus control. In EB1 + TR4 plants, shoot biomass increased 1.35-fold (p = 0.002) and root biomass 1.94-fold (p = 0.004) compared to TR4-only plants. EB1 + TR4 plants had significantly higher survival rates and lower disease severity than TR4-only plants. Gene expression analysis showed that EB1 alone did not induce NPR1 or PR1 but upregulated MYC2 by 1.67-fold (p = 0.016). In EB1 + TR4 plants, both SA and JA pathways were strongly activated: NPR1 upregulated 1.40-fold (p = 0.015), PR1 35.80-fold (p < 0.001), LOX2 1.50-fold (p = 0.0028), and MYC2 2.44-fold (p < 0.001).
**Clinical Implications:** This study provides evidence that bio-priming banana tissue culture plantlets with endophytic B. velezensis EB1 at the rooting stage is an effective strategy to enhance Fusarium wilt resistance and promote plant growth. The approach leverages the aseptic tissue culture system to establish beneficial plant–microbe interactions before field exposure, addressing a key vulnerability of micropropagated plants. EB1 acts through multiple mechanisms: direct antagonism via antifungal metabolites, competitive exclusion of the pathogen, and priming of both SA and JA defense pathways for enhanced immune response upon pathogen attack. The identification of 12 biosynthetic gene clusters and plant growth-promoting genes (IAA, spermidine) in the EB1 genome supports its multifunctional potential. Future work should focus on understanding molecular mechanisms, developing synthetic microbial communities, and formulating stable bioinoculants for large-scale agricultural application.