Metal-driven nanoassembly of hexahistidine-tagged melittin enables superior phytopathogen biofilm degradation with attenuated toxicity Authors: Xingcheng Zhou, Jie Hong, Li Yang, Satyabrata Nanda, Youbo Yu, You Wang, Krongthong Kamonsuangkasem, Ming Chang, Janaki Mohotti, Gefei Hao, Peiyi Wang, Libo Zhang Published: 2026 ABSTRACT AbstractBacterial biofilms formed by phytopathogens confer formidable resistance to chemical pesticides, underscoring an urgent need for innovative antimicrobial solutions. Antimicrobial peptides (AMPs), exemplified by the potent bee venom derivative melittin, offer a promising alternative owing to their broad-spectrum activity and intrinsic biofilm-disrupting capacity. However, the agricultural application of melittin is severely hindered by rapid environmental degradation, susceptibility to enzymatic degradation, and non-selective cytotoxicity. Here, we report a metal-coordination-driven nanoassembly strategy to enhance the stability and efficacy of melittin. Engineering an N-terminal hexahistidine tag enabled a one-step assembly of melittin into uniform nanoparticles (NanoMel) via Zn²⁺ coordination. This nanoformulation improved the antibacterial potency, lowering the half-maximal effective concentration (EC₅₀) values against Xanthomonas oryzae pv. oryzae (Xoo), Xanthomonas oryzae pv. oryzicola (Xoc) to 3.795 µg/mL and 3.202 µg/mL, representing a 1.59- and 1.38-fold enhancement over its linear counterpart. Furthermore, NanoMel demonstrated superior biofilm eradication, degrading 86.9% of mature Xoo biofilms at 24 µg/mL, significantly outperforming the free peptide. In planta assays revealed that NanoMel provided 68.2% curative and 65.9% protective efficacy against rice bacterial leaf blight at 200 µg/mL, surpassing the commercial bactericide thiodiazole-copper 20% suspension concentrate (TC-20% SC). Furthermore, the nanoassemblies effectively attenuated the inherent toxicity of melittin, as evidenced by significantly improved safety profiles in the zebrafish model. Collectively, these findings establish the metal-coordination-driven nanoassembly as a platform for constructing effective and eco-friendly AMP-based bionanobactericides, demonstrating a potent and practical strategy for sustainable plant protection.Graphical Abstract Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12951-026-04732-4.Keywords: Antimicrobial peptide, Melittin, Metal, Nano-assembly, Biofilm, Nanobactericide Introduction Plant pathogenic bacteria pose a major threat to global food security and natural ecosystems, with disease management often hindered by the formation of resilient bacterial biofilms [1, 2]. Biofilm-associated infections contribute substantially to annual crop yield losses estimated at 20–40%, affecting staple crops such as rice [3–5]. For instance, Xoo, the causative agent of bacterial blight in rice, can lead to severe crop damage with disease incidence reaching up to 70–80%. [6, 7, 8] The biofilm matrix enhances bacterial resistance to conventional bactericides, making infections difficult to eradicate [9,]. To date, no commercially available bactericides have been specifically developed to disrupt Xanthomonas biofilms, highlighting the need for targeted anti-biofilm strategies [9, 10]. Antimicrobial peptides (AMPs) are small amphipathic molecules ubiquitous in nature and represent promising alternatives to traditional pesticides [11, 12]. Their cationic and amphipathic properties enable microbial cell membrane disruption, conferring broad-spectrum activity against bacteria, fungi, and viruses [13]. This membrane-targeting mechanism is advantageous to minimize the risk of resistance development [14, 15]. Besides, AMPs exhibit strong anti-biofilm activity, inhibiting initial bacterial adhesion and disrupting established biofilms through membrane permeabilization and killing of embedded cells [16–18]. Such dual functionality, demonstrated even against multidrug-resistant pathogens [19, 20], highlights their potential for controlling biofilm-forming plant pathogens. Among these, melittin, a 26-amino-acid peptide from honeybee venom, has attracted considerable attention for its potent and rapid antimicrobial activity [21]. Melittin inhibits diverse plant pathogens, including Xoo, by disrupting biofilm and cell membrane [7, 14]. Furthermore, melittin has proven effective against biofilms formed by various clinical pathogens, such as Pseudomonas aeruginosa and Staphylococcus aureus, by lysing embedded cells and preventing surface attachment [22]. However, the agricultural application of melittin remains hampered by its non-selective cytotoxicity, environmental instability, susceptibility to proteolytic degradation [23]. Nanobiotechnology presents a promising avenue to overcome these limitations. The integration of AMPs into nanoscale systems (NanoAMPs), including polymers, liposomes, or inorganic materials, can significantly enhance their stability and bioavailability [24, 25]. Such nanosystems protect AMPs from proteolytic degradation and harsh environmental conditions, while facilitating controlled release at the target sites. Thus, use of NanoAMPs improves efficacy and minimizes off-target toxicity by reducing the required dosage and application frequency [26, 27]. Here, we report the first application in phytopathogen biofilm control using Zn²⁺-driven His-tag melittin nanoassembly (NanoMel) as a potent nanobactericide. (Fig. 1a). Extensive characterization by X-ray absorption spectroscopy (XAS), Fourier transform infrared spectroscopy (FTIR), zeta potential measurements, and circular dichroism spectroscopy confirmed the role of peptide-Zn interaction in nanoassembly and the conformational changes associated with nanoassembly. NanoMel displayed 37.2% and 27.6% enhanced antimicrobial activity against Xoo and Xoc, respectively, and enhanced biofilm inhibition and eradication relative to free melittin (Fig. 1b). Moreover, NanoMel exhibited an excellent safety profile in rice and markedly attenuated melittin-associated toxicity in a zebrafish embryo model (Fig. 1c). These findings establish an efficient nanotechnological approach to overcoming the key barriers to AMP application, demonstrating a potent and practical strategy for sustainable plant protection. Zinc-driven nanoassembly of hexahistidine-fused melittin: A facile and efficient strategy for combating biofilm-associated infections. a Schematic illustration of NanoMel synthesis. b Mechanistic diagram of NanoMel action against rice bacterial diseases. c Safety profile of NanoMel in rice and its attenuation of melittin-associated cytotoxicity Results and discussion Fusion of a hexahistidine tag at the N-terminus retains the antimicrobial activity of melittin The wild-type melittin sequence (NH2-GIGAVLKVLTTGLPALISWIKRKRQQ-COO⁻)28 features a hydrophobic N-terminus and a cationic, hydrophilic C-terminus, forming an amphipathic α-helix with a + 5 charge at neutral pH [29, 30]. Its structure typically consists of two α-helices connected via a flexible linker segment [31]. To assess the impact of hexahistidine tag fusion at either terminus of the peptide, N-terminal (H-M) and C-terminal (M-H) hexahistidine-fused peptides were synthesized (Table S1). An in vitro bioactivity assay against Xoo revealed that H-M retained activity comparable to that of wild-type melittin, with a slight enhancement against Xoc, whereas M-H showed significantly reduced antimicrobial activity (Table S2). We hypothesized that the cationic C-terminal region of melittin is critical for bacterial cell membrane disruption. Consequently, H-M was selected for subsequent assembly and characterization. Preparation and characterization of NanoMel A metal-coordination-driven hexahistidine-tagged peptide assembly was used to prepare NanoMel. This approach exploits the intrinsic coordination between the divalent cation (e.g., Zn²⁺) and the hexahistidine tag to drive the nanoassembly of tagged proteins or peptides [32, 33]. The optimal peptide-to-Zn²⁺ molar ratio for nanoassembly was determined to be 2:1, achieving a NanoMel loading efficiency exceeding 95% (Fig. S1). The prepared NanoMel exhibited nanoparticle morphology with an average size of ~25 nm (Fig. 2a and b). The polydispersity index (PDI) of NanoMel was 0.21, which indicates a uniform dispersion of the particles. The PDI value rose to 0.32 during long-term stability testing, confirming favorable structural stability with only mild time-dependent aggregation. (Fig. S2, S3). The simple procedure with low metal concentrations and good stability endows this system with great potential for translational applications. Characterization of NanoMel nanoassemblies. a SEM image of NanoMel (Inset: TEM image of NanoMel). b DLS analysis to characterize the hydrodynamic size distribution of NanoMel. c Zeta potential analysis of melittin, ZnCl2, and NanoMel. d FTIR spectra. e XPS spectra. f CD spectra of melittin and NanoMel. g Zn K-edge XAS analysis of the atomic-level coordination architecture of NanoMel Elemental analysis further confirmed the presence of Zn within the assembled structures (Fig. S4), consistent with metal-coordination-driven assembly. Notably, the surface charge of the nanoparticles was significantly enhanced compared to that of pristine melittin (Fig. 2c). Collectively, these findings confirm the successful fabrication of metal- ion-driven hexahistidine-fused melittin nanoassemblies. Fourier-transform infrared (FTIR) spectra (Fig. 2d) revealed distinct changes following nanoassembly. The broad N-H stretching band (3,200–3,600 cm⁻¹), associated with amide and amino groups, displayed distinct peak broadening and a wavenumber shift in NanoMel relative to free melittin. For the C = O stretching vibration peak of the amide I band (1,600–1,700 cm⁻¹), a wavenumber blue shift and diminished intensity were observed in NanoMel. Meanwhile, the C-N stretching vibration peak of the amide III band (1,200–1,400 cm⁻¹) also exhibited a measurable wavenumber shift. These bond-position shifts stemmed from the coordination interactions between Zn²⁺ ions and the N/O atoms of amide bonds in melittin molecules, which modulate the electron cloud density and bond force constants of the N-H, C = O, and C-N bonds, ultimately inducing characteristic alterations in their infrared vibration modes. Overall, these findings validated the successful coordination assembly of Zn²⁺ with melittin. X-ray photoelectron spectroscopy (XPS) further confirmed Zn incorporation within NanoMel (Fig. 2e). The spectrum of free melittin displayed only peptide-associated signals (C1s, N1s, and O1s), whereas NanoMel exhibited additional Zn2p spin-orbit doublet peaks along with Zn LMM Auger features, indicating the presence of coordinated Zn species. Concurrently, the binding energies of C1s, N1s, and O1s in NanoMel shifted relative to those of free melittin, suggesting electronic redistribution upon nanoassembly. This phenomenon arises because the formation of coordination bonds between Zn²⁺ ions and the N atoms (of amide bonds) as well as O atoms (of carboxyl/hydroxyl groups) in melittin alters the electron cloud density of these atoms, thereby inducing variations in their photoelectron binding energies. This result is mutually corroborated by the changes in functional group vibration modes observed in the FTIR analysis. CD spectroscopy revealed structural changes following nanoassembly (Fig. 2f). The peptide secondary structure underwent significant changes after nanoassembly: in the far-UV region (180–260 nm), NanoMel exhibited a pronounced negative ellipticity at ~208 nm, suggesting a higher α-helix content or a more ordered conformational arrangement. Further, secondary structure composition was quantified using the CONTINLL algorithm in CDPro (Table S3). The α-helix dominated both samples: NanoMel had a higher total α-helix content (84.3%) with more regular α-helix (50.2%) than distorted (34.1%), while melittin’s total α-helix content (82.7%) featured more distorted α-helix (42.3%) than regular α-helix (40.4%). Moreover, NanoMel showed a reduced proportion of unordered structure (4.2%) compared with melittin (8.2%). To further understand the composition and structure of NanoMel, the atomic-level coordination architecture was analyzed by Zn K-edge X-ray absorption spectroscopy (XAS) (Fig. 2g). Normalized XANES (1) confirms Zn2+ oxidation state (absorption edge 9659.7 eV), with white-line asymmetry suggesting coordination heterogeneity relative to crystalline ZnCl₂ and ZnO. The k³-weighted EXAFS (2) oscillations of NanoMel differed markedly from all references—neither matching ZnO’s strong crystalline features (amplitude ±12) nor ZnCl2’s simple salt pattern, suggesting the absence of oxide formation, residual ZnCl2, or metallic aggregation. R-space Fourier transform (3) revealed a dominant first-shell peak at R + α ≈ 2.35 Å (1.25 Å−3·k−3), corresponding to 2.0–2.4 Å interatomic distances consistent with mixed Zn-N(His)/Zn-Cl coordination, while broadened second-shell features (R + α > 3Å) reflect peptide backbone conformational flexibility. Wavelet transform analysis (4) provided critical coordination fingerprinting. NanoMel exhibited a broad WT signature at R + α ≈ 1.5 Å spanning k = 5–10 Å−1, indicating contributions from both Zn–Cl (k = 5–7 Å−1) and Zn–N/O (k = 8–10 Å−1) scatterers in a mixed first coordination shell. This extended k-range coverage confirmed that hexahistidine-mediated assembly partially but incompletely displaces chloride, yielding a heterogeneous coordination sphere estimated to contain 2–3 imidazole nitrogens atoms plus 1–2 chloride/water ligands in distorted octahedral geometry. The weak secondary scattering feature (R + α ≈ 3.5 Å, k ≈ 5 Å−1) arose from disordered second-shell scattering through the flexible peptide framework. Driving force, stability, antibacterial activity, and proteinase resistance analysis of NanoMel The structural analyses suggest that hexahistidine-Zn²⁺ coordination mediates nanoassembly (Fig. 2d-g). To confirm this, a pH-dependent assembly assay was conducted. The size distribution and zeta potential of NanoMel were monitored by DLS. As shown in Fig. 3a, nanoparticle size increased progressively with increasing pH. At pH 6, no nanoparticles were detected. At pH < 7, the coordination groups of the peptides undergo protonation and H⁺ competes for binding sites. Additionally, the positively charged peptides repel each other, thereby hindering the formation and assembly of Zn²⁺-protein nanoparticles [34]. At pH 7, nanoparticles with an average size of ~25 nm were formed, accompanied by an elevated zeta potential indicative of stable nanoassemblies. In contrast, larger aggregates were formed at pH 8 and pH 10 owing to histidine deprotonation, enhanced Zn²⁺ cross-linking effects, and reduced electrostatic repulsion between nanoparticles, thereby promoting the aggregation and enlargement of nanoassemblies [32]. Characterization of driving force, stability, antibacterial activity, and proteinase resistance of NanoMel. a Size distribution and zeta potential of NanoMel particles assembled at different pH values (Inset: Photographs of NanoMel samples synthesized at different pH). b Size distribution and zeta potential of NanoMel stored at room temperature for 7 days. c, d In vitro EC50 comparison of NanoMel and melittin against Xoo and Xoc. e Electrical conductivity of Xoo treated with different concentrations of NanoMel and melittin. f Proteinase resistance: Inhibition rates of Xoo by NanoMel and melittin after incubation with different concentrations of trypsin. For (f), an independent Student’s t-test was used for analysis. Significance levels are indicated as *p < 0.05, **p < 0.01, ***p < 0.001; ns. denotes no significant difference Collectively, these results indicate that the nanoassembly is driven by histidine-metal coordination. These findings are consistent with the formation of peptide-Zn coordination clusters that remain stable at neutral pH, likely due to a balance between coordination cross-linking and electrostatic repulsion. As shown in Fig. 3b, the nanoparticles maintained high dispersibility and stable surface charge at room temperature (25℃) for at least 7 days, with only a slight increase in particle size and a mild decrease in surface charge. These observations indicated excellent structural stability of NanoMel. The antibacterial activity of NanoMel and free melittin was characterized by determining their EC50 on Xoo, Xoc, and Escherichia coli. As shown in Fig. 3c-d and S7, the EC50 values of NanoMel on Xoo, Xoc and E. coli were 3.795 µg/mL, 3.202 µg/mL and 2.722 µg/mL, respectively, representing 37.2%, 27.6% and 31.6% higher potency than that of melittin. The enhanced antibacterial activity of nanoassemblies can be attributed to the increased local concentration of the peptide and elevated surface charge, which collectively facilitate binding to and disruption of bacterial cell membranes, which possesses more positively charged motifs that facilitate binding to and disruption of the cell membranes. Moreover, with increasing pH, the surface charge of NanoMel diminished, leading to large-scale aggregation. To further elucidate the cell membrane-disruption mechanism, the electrical conductivity of Xoo suspensions following treatment with NanoMel or melittin was analyzed. As shown in Fig. 3e, the relative electrical conductivity of the bacterial samples treated with lower concentrations of NanoMel or melittin gradually increased within 8 h of co-incubation. At concentrations above 24 µg/mL, conductivity rose rapidly within the first hour, consistent with acute membrane permeabilization and cytoplasmic leakage. Notably, NanoMel induced significantly higher conductivity changes than free melittin, indicating enhanced disruption of bacterial membrane integrity and increased permeability. To further verify alterations in membrane permeability, DiBAC₄(3) was employed to measure changes in membrane potential in the presence of NanoMel and free melittin. The results revealed that the degree of membrane damage increased with rising NanoMel concentrations, which was significantly higher than that induced by free melittin (Fig. S5). These findings were further corroborated by the scanning electron microscopy (SEM) images of Xoo treated with NanoMel or melittin, which revealed more severe membrane damage induced by NanoMel (Fig. S6). In agricultural applications, proteolytic degradation represents a major limitation for peptide-based agents, since there are hundreds of proteases in plants [35]. Therefore, we adopted high concentration of protease [36, 37] to test the proteolytic resistance of NanoMel and free melittin by determining the antibacterial activity against Xoo after incubation with different concentrations of trypsin. As shown in Fig. 3f and S8, the inhibitory activity of melittin against Xoo decreased significantly with increasing proteinase concentration. In contrast, NanoMel retained ~70% of its antibacterial activity even at 2 mg/mL of trypsin, which was significantly superior to the residual activity of free melittin. This indicated that the nanoassembly structure effectively protected the peptide from proteolytic degradation, enhancing its environmental stability. NanoMel exhibits enhanced activity to inhibit bacterial biofilm formation In rice bacterial diseases, Xoo is a major biofilm-forming pathogen [38]. Melittin achieves antimicrobial activity primarily through membrane disruption [39]. Hence, we employed Xoo as the pathogen model to assess the activity of NanoMel on the inhibition of biofilm formation and biofilm eradication using the crystal violet staining assay [40, 41]. As shown in Fig. 4a-b, the biofilm inhibition increased in a dose-dependent manner. At lower doses (0.5, 1, 2 × EC50), the biofilm inhibition rates of NanoMel reached 29.1%, 55.87%, and 79.7%, respectively, which were significantly higher than those of free melittin (16.1%, 38.5%, and 66.3%). These results demonstrate that nanoassembly can substantially enhance the antibiofilm activity of melittin. Notably, when the concentration of NanoMel was increased to 24 µg/mL (4 × EC50), the inhibition rate reached 95.6% (Fig. S9), indicating potent biofilm formation inhibitory capacity. Comparison of antibiofilm and bactericidal properties in biofilms between NanoMel and melittin. a, b Photographs of crystal violet staining and quantitative analysis of Xoo biofilms after co-incubation with different concentrations of melittin and NanoMel. c, d Representative photographs and colony count of Xoo within biofilms on agar plates following 48 h treatment with melittin and NanoMel. e Confocal laser scanning microscopy (CLSM) 3D imaging to compare the bactericidal activity of NanoMel and melittin against Xoo in biofilms; live cells in biofilms were stained with AO (green). f Quantification of green fluorescence intensity from CLSM images (indicating Xoo viability), analyzed using ImageJ software. For (b, d, f), Statistical significance was determined using one-way ANOVA with Tukey’s post-hoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant) To further evaluate bacterial viability within established biofilms, we inoculated and cultured Xoo colonies treated with NanoMel or melittin following biofilm formation. As depicted in Fig. 4c-d, NanoMel reduced the bacterial colony count within Xoo biofilms, outperforming the inhibitory effect of melittin. At 1.0–4.0 × EC50, the NanoMel treatment group exhibited significantly fewer colony-forming units (CFUs), underscoring its increased inhibitory effect on biofilm-protected bacterial growth relative to melittin. These findings were consistent with the confocal laser scanning microscopy (CLSM) analysis (Fig. 4e, f). The three-dimensional (3D) imaging revealed that bacteria residing in the Xoo biofilm were stained green by acridine orange (AO), confirming that NanoMel more effectively suppresses bacterial viability within biofilms. NanoMel eradicates established Xoo biofilms During advanced infection stages, pathogens form dense and recalcitrant biofilms that confer increased tolerance to antimicrobial treatments [41, 42]. Thus, eradicating pre-formed mature biofilms is more challenging. Here, the eradication efficacy of NanoMel or melittin was analyzed after crystal violet staining [43]. As shown in Fig. 5a-b, the eradication effect of both NanoMel and melittin was enhanced with increasing concentration. The biofilm eradication rates of NanoMel were significantly higher than that of melittin. For example, at 1 × EC₅₀ and 4 × EC₅₀, NanoMel induced biofilm reduction of 29.1% and 87.1%, respectively, compared with 14.3% and 73.1% for melittin (Fig. S10). Comparison of the eradication efficacy of NanoMel and melittin against pre-established Xoo biofilms. a, b Crystal violet staining images illustration and quantification of pre-established Xoo biofilms eradication following treatment with different concentrations of NanoMel and melittin. c Representative images of Xoo colony development within established biofilms on solid medium after 48 h exposure to different concentrations of NanoMel and melittin. d CLSM images of pre-established Xoo biofilms after 48 h treatment with different concentrations of melittin and NanoMel (live cells: green, AO; dead cells: red, PI). e Quantitative analysis of Xoo CFUs within biofilms after 48 h treatment with different concentrations of melittin and NanoMel. f, g Quantification of green and red fluorescence intensity from CLSM images (live Xoo and dead, respectively) analyzed using ImageJ software. For (b, e, f, g), Statistical significance was determined using one-way ANOVA with Tukey’s post-hoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant) Beyond biofilm eradication, the administration of NanoMel and melittin elicited a concentration-dependent decline in the Xoo cell population within biofilms, based on the CFU analysis (Fig. 5c, e). Additionally, the potency of melittin was consistently lower than that of NanoMel. After treatment with NanoMel, the number of colonies decreased by 23.3%, 24.2%, and 72.6% at concentrations of 2.0 ×, 4.0 × and 8.0 × EC50, respectively, as compared to the melittin, demonstrating significantly increased activity. CLSM 3D imaging was conducted to further validate the biofilm eradication ability of NanoMel (Fig. 5d), in which live and dead bacteria residing in the Xoo biofilm were stained employing a blend of acridine orange (AO) and propidium iodide (PI). Application of NanoMel and melittin induced Xoo cell death within the biofilm. Quantitative analysis of average green and red fluorescence intensities (Fig. 5f, g) confirmed that NanoMel induced significantly higher loss of viability compared with melittin, supporting its superior efficacy in eradicating Xoo biofilms. NanoMel disrupts biofilm matrix integrity and attenuates Xoo virulence To investigate the impact of NanoMel on biofilm matrix composition, extracellular polysaccharides (EPS) and extracellular proteins (EPN) were investigated by using phenol-sulfuric acid method, gravimetric method, and Coomassie Brilliant Blue staining [44]. As shown in Fig. 6a, S11(a-b), NanoMel reduced EPS production in a dose-dependent manner. For instance, at 1 × EC50, 2 × EC50, the EPS production levels in the NanoMel-treated groups were 41.8% and 15.4%, respectively, showing a significantly lower than those in the melittin treatment groups (75.6% and 25.1%). This consistent trend was confirmed by the gravimetric analysis (Fig. S11c). To further elucidate the underlying mechanism, we employed RT-qPCR to detect the expression levels of the gum gene cluster that governs EPS biosynthesis at the molecular level. The results revealed that NanoMel treatment at 4×EC₅₀ downregulated the expression of gumB, gumE, gumG, gumH, gumK and gumM by 76.20%, 90.80%, 48.17%, 77.51%, 75.90% and 36.85%, respectively (Fig. S12). Such inhibitory effects were markedly stronger than those induced by free melittin. Similarly, for EPN production, the levels in the NanoMel treatment groups were 72.5%, 34.4%, and 15.5%, revealing a significant reduction than of that in the melittin treatment groups (48.7%, 21.8%, and 10.7%) at 1 ×, 2 ×, and 4 × EC₅₀, respectively (Fig. 6b, c and S11d). Together, these results indicate that NanoMel more effectively suppresses the biosynthesis of key structural components of the Xoo biofilm matrix relative to free melittin. Investigation of the antibiofilm mechanism of NanoMel. a-c Inhibition of EPS and EPN production in Xoo following treatment with NanoMel and melittin at different concentrations (1, 2, 4 × EC50). d Representative images and swimming zone diameters of Xoo following treatment with NanoMel and melittin at 2 × EC50. e Representative images and hydrolysis zone diameters of Xoo extracellular cellulase following treatment with NanoMel and melittin at 2 × EC50. f Representative images and hydrolysis zone diameters of Xoo extracellular amylase following treatment with NanoMel and melittin at 2 × EC50. g Representative susceptible rice leaves inoculated with Xoo pre-treated with NanoMel and melittin via the leaf-clipping method at an effective dose of 2 × EC50 (12 µg/mL). h Quantitative analysis of leaf lesion lengths (n = 10 leaves per treatment. For (a–c) were analyzed by one-way ANOVA followed by Tukey’s multiple comparisons test, and panels (d–f) and (h) by independent Student’s t-test. Significance levels are indicated as *p < 0.05, **p < 0.01, ***p < 0.001; ns. denotes no significant difference Xoo exhibits motility during rice infection and secretes enzymes such as cellulase and amylase to accelerate the infection process. Bacterial motility is pivotal for biofilm formation and pathogenicity [45]. Its inhibition reduces biofilm development and diminishes the pathogen’s virulence [46]. Therefore, the impacts of NanoMel and melittin on Xoo motility were analyzed. As shown in Fig. 6d, the swimming diameter of Xoo cells notably decreased following NanoMel treatment compared with melittin treatment, with a diameter of only 6.8 mm, whereas the diameters for the melittin treatment group and control group were 12.7 and 25.5 mm, respectively. During pathogen infection, the secretion levels of extracellular enzymes directly determine their pathogenicity, migration ability, and colonization efficiency [47]. Therefore, we further characterized the effect of NanoMel on such key virulence factors of Xoo, using an effective concentration of 2 × EC50 (12 µg/mL) for the treatment. As shown in Fig. 6e-f, at 2×EC50 concentration (12 µg/mL), the cellulase hydrolysis zone diameter was reduced to 12.1 mm following NanoMel treatment, compared with 15.0 mm for melittin and 17.8 mm for the control. Similarly, amylase activity was reduced to 11.13 mm in the NanoMel group, relative to 13.1 mm and 17.4 mm in the melittin and control groups, respectively. These findings indicate that the robust anti-biofilm functionality of NanoMel stems from the inhibition of EPS and EPN production, reduction of the activity of the extracellular enzymes (cellulase, amylase), and inhibition of bacterial swimming motility, thereby potentially reducing the infection and spread of pathogens to plants. This inference was further verified by rice pathogenicity tests. In this assay, rice leaves were inoculated with Xoo pre-treated with NanoMel or melittin at 12 µg/mL via the leaf-clipping method [48]. As shown in Fig. 6g-h, the average lesion lengths of the control group, ZnCl2-treatment group, melittin treatment group and NanoMel treatment group were 20.4 cm, 19.1 cm, 13.7 cm and 8.7 cm, respectively. These results demonstrate that the pathogenicity of Xoo was significantly attenuated after co-incubation with NanoMel, as evidenced by a marked reduction in lesion length. NanoMel effectively controls plant bacterial diseases caused by Xanthomonas (Xoo and Xoc) Based on the excellent bactericidal and antibiofilm activity of NanoMel, we evaluated its in vivo efficacy against Xoo and Xoc. As shown in Fig. 7a, rice plants in the control group and the ZnCl2 treatment group exhibited typical and severe symptoms of rice bacterial leaf blight (BLB), including obvious leaf wilting and chlorosis. The NanoMel treatment group exhibited significantly improved disease control efficacy. Its curative activity reached 68.2%, which was significantly superior to that of the melittin treatment group (48.4%). In addition, the protective activity of NanoMel was significantly greater than that of the melittin treatment group (Fig. S13a). In vivo performance of NanoMel and melittin against bacterial leaf blight caused by Xoo and bacterial leaf streak caused by Xoc on rice. a Disease symptoms and in vivo curative efficacy of NanoMel and melittin on rice bacterial blight. The effective application dose was 200 µg/mL, and rice plants were continuously cultured for 14 days after spraying. b Disease symptoms and in vivo curative effect of NanoMel and melittin on rice bacterial leaf streak. The effective application dose was 200 µg/mL, and rice plants were continuously cultured for 14 days after spraying. For this panel, Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test. Significance levels are indicated as *p < 0.05, **p < 0.01, ***p < 0.001; ns. denotes no significant difference Given the excellent therapeutic efficacy of NanoMel against Xoo-induced BLB, we sought to expand its application scope by evaluating its activity against Xoc-induced BLS using the pressure infiltration method [46]. As shown in Fig. 7b, severe disease symptoms were observed on rice leaves in both control and ZnCl2 treatment groups, with lesion lengths of 13.1 mm and 13.0 mm, respectively. In contrast, NanoMel treatment significantly reduced the lesion length to 3.9 mm, achieving a control efficacy of 69.6%, a value that exceeded that of melittin (48.6%). For comparative purposes, the commercial bactericide thiodiazole-copper (TC-20% SC) was included as a positive control, which exhibited only 23.2% and 29.0% curative efficacy on Xoo and Xoc, respectively. Collectively, these results demonstrate the promising potential of NanoMel for the simultaneous control of two major rice bacterial diseases, bacterial leaf blight and bacterial leaf streak. NanoMel exhibits excellent biological safety The biological safety of bactericides is of great significance for sustainable plant protection. A major limitation of melittin application is its high toxicity toward both the host plant and non-target organisms [49]. Herein, we systematically evaluated the toxic response of NanoMel on rice (the target crop) and zebrafish (as a representative non-target organism). First, the potential toxicity of NanoMel on rice seedlings was investigated. As shown in Fig. S14, rice seedlings treated with 500 µg/mL of NanoMel exhibited no obvious phytotoxic symptoms. No abnormalities such as leaf chlorosis or wilting were observed, and the overall growth status of the plants was not significantly affected compared to other treatment groups. Then, the effects of NanoMel on rice seed germination rate and root-shoot growth were evaluated. As shown in Fig. 8a-b, the NanoMel and melittin treatment groups exhibited comparable growth performance to the control groups, with no statistically significant differences in seed germination rate and root-shoot. These results indicate that the toxicity of NanoMel on rice seed germination and early seedling growth is negligible. Acute toxicity assessment of melittin and NanoMel. a, b Representative images and quantitative analysis of rice seed germination, including root length and shoot length, after treatment with 100 µg/mL melittin and NanoMel. c Morphological observations of zebrafish embryos exposed to 45 µg/mL melittin and NanoMel at different time points. d, e Survival rate and teratogenicity rate of zebrafish embryos treated with melittin and NanoMel at various concentrations. For (b, d), The data in panel (b) were analyzed by one-way ANOVA with Tukey’s multiple comparisons test, and the data in panel (d) were compared using an independent Student’s t-test. Significance levels were marked as *p < 0.05, **p < 0.01, ***p < 0.001; ns. indicates no statistically significant difference Finally, zebrafish embryos were used to evaluate the acute toxicity of NanoMel at different concentrations. As shown in Fig. 8, NanoMel, melittin, and ZnCl2 were co-incubated with zebrafish embryos for 72 h. At a concentration of 45 µg/mL, we observed that the survival rates of zebrafish embryos in the NanoMel and melittin treatment groups were 73.3% and 25.0%, respectively, while the total teratogenicty rates were 28.3% and 96.7%, respectively (Fig. 8c, e). Our findings indicate that the acute toxicity of NanoMel to zebrafish embryos was significantly lower than that of free melittin. NanoMel also substantially mitigated the teratogenic effects of melittin in zebrafish. To elucidate the molecular mechanism underlying the attenuation of melittin toxicity by NanoMel in zebrafish embryos [50], transcriptomic analysis was conducted to compare gene expression profiles between embryos treated with NanoMel and free melittin. Our results revealed that exposure to free melittin induced extensive alterations in the global gene expression profile of zebrafish embryos (Fig. S15). Specifically, 1,125 differentially expressed genes (DEGs) were identified in the melittin treatment group relative to the time-matched control. Furthermore, NanoMel treatment resulted in a marked reduction in the number of DEGs, with only 439 genes exhibiting differential expression (Fig. 9a). Notably, at 72 h post-exposure, only 81 DEGs were commonly dysregulated in both the NanoMel and free melittin groups (Fig. S16). Collectively, these data suggest that NanoMel exerts a far milder effect on differential gene expression in zebrafish embryos relative to free melittin, a finding further corroborated by the volcano plots presented in Fig. 9b. Transcriptomic analysis of zebrafish embryos exposed to NanoMel and melittin for 72 h. a Stacked bar plot of differentially expressed gene (DEG) counts in zebrafish embryos treated with 45 µg/mL NanoMel and free melittin relative to the control group. b Volcano plots illustrating the distribution of DEGs in the two treatment groups. c GO enrichment analysis, and d KEGG enrichment dot plot for DEGs in zebrafish embryos exposed to NanoMel and melittin To investigate the potential functions of the DEGs identified in zebrafish after NanoMel and melittin exposure, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses [51–53]. As shown in Fig. 9(c-d), the melittin treatment triggered profound, multidimensional toxic perturbations in zebrafish embryos at the functional level. GO enrichment analysis revealed significant upregulation of pathways associated with chromatin organization and nucleosome assembly, which are tightly linked to genomic stability, suggesting onset of genomic toxicity characterized by DNA damage and dysregulated gene expression. Concurrently, enrichment of immune-inflammatory pathways such as mast cell activation reflected a strong acute inflammatory response in melittin-treated embryos, which was accompanied by impaired tissue integrity resulting from abnormal cell adhesion. KEGG pathway analysis further validated these toxic effects, showing marked activation of innate immune and inflammatory signaling pathways including Toll-like receptor and TNF signaling. Enrichment of the MAPK pathway indicated severe cellular stress and damage, while perturbations in xenobiotic metabolism and multiple primary metabolic pathways suggested an elevated detoxification burden and systemic metabolic imbalance in melittin-exposed embryos. The coordinated activation of these distinct pathways revealed the complex toxicological profile of melittin, which encompasses inflammatory injury, genomic toxicity, and metabolic dysregulation. In contrast, NanoMel treatment exhibited a markedly attenuated toxic effect. GO enrichment analysis of NanoMel-treated embryos revealed enrichment of only a limited set of terms related to basic metabolic processes, with no significant enrichment of pathways associated with inflammation, DNA damage, or tissue damage. KEGG pathway analysis yielded consistent results, showing only minimal and low-significance perturbations in molecular pathways and no activation of toxicology-related signaling cascades involving inflammation, cellular stress, or metabolic dysregulation. These findings reveal that NanoMel‑treated zebrafish embryos displayed significantly less deviation from normal physiological status in gene expression and pathway regulation compared with melittin‑exposed controls. To further characterize the differences between free melittin and NanoMel, we analyzed embryonic reactive oxygen species (ROS) levels [54] and cell apoptosis [55]. The results showed that the ROS level in the NanoMel group was markedly lower than that of free melittin, with no significant difference from the control group, which was further validated by cell apoptosis assays (Fig. S17). confirming the efficacy of our nanoformulation strategy in mitigating the intrinsic toxicity of melittin. Collectively, these results validate that the ordered NanoMel formulation possesses excellent biological safety, environmental compatibility, and biocompatibility. Discussion This study successfully engineered NanoMel, a multifunctional bionanobactericide, via the self-assembly of rationally designed hexahistidine-tagged melittin (H-M) and zinc ions. These nanoassemblies spontaneously formed ordered nanostructures with an average particle size of 25 nm. Comprehensive characterization assays validated that the assembly process is driven by specific metal-histidine coordination interactions, with clear elucidation of the underlying formation mechanism. NanoMel exhibits potent dual antimicrobial activity: it effectively eliminates planktonic bacterial cells and penetrates the recalcitrant biofilm matrix to inhibit and kill the biofilm-encased bacteria. Its bactericidal mode of action involves efficient penetration of Xoo cells, leading to rapid intracellular electrolyte leakage and subsequent bacterial lysis. Mechanistic investigations into its anti-biofilm activity revealed that NanoMel disrupts biofilm formation and maturation by suppressing the production of key structural components (extracellular polysaccharides and extracellular proteins). In addition, NanoMel restricts bacterial swimming motility and the secretion of virulence-associated extracellular enzymes (cellulase and amylase), thereby attenuating bacterial virulence, migration ability, and pathogenicity in rice. More importantly, at a concentration of 200 µg/mL, NanoMel exhibited remarkable 68.2% curative efficacy and 65.9% protective efficacy against BLB—significantly outperforming free melittin (48.4% curative, 60.9% protective). Critically, NanoMel exhibits excellent biological safety profile for both target plants (rice) and non-target organisms (zebrafish). Transcriptomic analysis of zebrafish embryos confirmed that NanoMel drastically mitigates the intrinsic toxicity of free melittin. Melittin induced severe multifactorial toxicity via dysregulation of multiple molecular pathways (genomic, immune-inflammatory, and metabolic), whereas NanoMel treatment only caused mild, limited perturbations in a small set of pathways. Gene expression and pathway regulation in NanoMel-exposed embryos are similar to those under normal physiological state, suggesting the effectiveness of nanomodification in attenuating melittin’s cytotoxicity. Recently, various systems have been explored to incorporate melittin and other peptides in order to improve their functionality and performance. To distinguish NanoMel from previously reported peptide delivery systems, we systematically compared their raw materials, fabrication strategies and functional characteristics (Table S4). Most conventional systems suffer from drawbacks including complicated synthesis procedures, organic solvent dependence, insufficient stability and moderate antibiofilm efficacy. In comparison, our NanoMel is fabricated via zinc-ion-driven self-assembly under fully aqueous conditions, and possesses prominent merits such as superior protease resistance, favorable stability, potent antibiofilm activity and eco-friendly preparation routes. Overall, our work establishes a simple and robust metal-coordination platform for engineering AMP-based nanomaterials, providing novel conceptual insights and promising application prospects for the development of high-efficacy, low-toxicity AMP-based biopesticide formulations for sustainable plant protection. Furthermore, with the continuous advancement of biomanufacturing technologies, the production costs of melittin and His-tagged melittin are expected to decline markedly, further endowing the proposed strategy with favorable economic practicality. Materials and methods Materials Antimicrobial peptides H-M and M-H (purity: 95%) were synthesized by Anhui Guoping Pharmaceutical Co., Ltd. The following reagents were used in this study: ZnCl2 (98%), PBS buffer (pH 7.2), 10% crystal violet stain, membrane‑sensitive fluorescent probe DiBAC₄(3) (indicator grade),, Coomassie Brilliant Blue G250 (reagent grade), sodium carboxymethylcellulose (USP grade), soluble starch (reagent grade), iodine (99%), potassium iodide (99.5%), absolute ethanol (analytical reagent, AR), sodium chloride (analytical grade), Congo red (indicator grade), 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) (98%), acridine orange (AO), and propidium iodide (PI). All chemicals were obtained from Macklin unless stated otherwise. Rice seeds, as well as strains of Xoo and Xoc, were provided by the Green Pesticide National Key Laboratory, Center for Research and Development of Fine Chemicals, Guizhou University. Methods In vitro antibacterial assay The in vitro antibacterial activity of the test compounds was assessed via a turbidimetric assay. A bacterial suspension was diluted in NA medium to an optical density at 595 nm (OD₅₉₅) of 0.1. NA medium was composed of 1.0 g yeast extract, 3.0 g beef extract, 5.0 g peptone, and 10.0 g glucose per liter of double-distilled water (ddH₂O) (pH 7.2), The medium was then supplemented with test compounds at various concentrations. Aliquots (20 mL) of the inoculated mixture were transferred to 50 mL centrifuge tubes and incubated in a shaking incubator at 28 °C and 180 rpm for 12–24 h, until the OD₅₉₅ of the negative control reached 0.6–0.8. Following incubation, 200 µL of each sample was transferred to a 96-well microplate for turbidity determination. The inhibition rate (I) was calculated using the formula below: where C represents the corrected turbidity value of the control and T represents the corrected turbidity value of the test sample. The EC₅₀ value was then determined using Microsoft Excel 2019 based on the inhibition rates observed at different compound concentrations. All tests were performed in triplicate. Preparation of NanoMel His6-tagged melittin powder was dissolved in ddH₂O to a final concentration of 1 mg/mL. The peptide solution was subsequently mixed with a ZnCl2 solution at a molar ratio of 2:1. The mixture was stirred overnight to ensure complete formation of nanoassemblies. The resulting mixture was centrifuged at 5,000 rpm for 1 h using an ultrafiltration centrifuge tube. The collected precipitate was then lyophilized for subsequent use. Characterization The morphology of the nanoparticles was characterized by scanning electron microscopy (SEM; TESCAN CLARA, Czech Republic) and transmission electron microscopy (TEM; JEOL JEM-F200, Japan). The mean particle size and zeta potential of the nanoparticles were determined via dynamic light scattering (DLS; Malvern Zetasizer Nano ZS). Functional groups were characterized using Fourier-transform infrared spectroscopy (FTIR; Thermo Scientific K-Alpha, USA) over the range 4000–400 cm⁻¹ with a resolution of 4 cm⁻¹. Surface chemical properties were analyzed by X-ray photoelectron spectroscopy (XPS; Thermo Scientific K-Alpha, USA). Circular dichroism (CD; GASCO J-1,700, Japan) spectroscopy was used to assess the secondary structures and their changes in both antimicrobial peptides and nanoparticles. X-ray absorption spectroscopy (XAS) was employed to probe the local atomic structure and chemical state of targeted elements within the materials. XAS data at the Zn K-edge (~9,660 eV) were collected at the Synchrotron Light Research Institute (SLRI) in Thailand, using Beamline 1.1 W in Transmission and Fluorescence mode. The synchrotron operated at 1.2 GeV with a beam current of ~100–150 mA. Samples were measured at room temperature under ambient conditions. Reference spectra for standards (Zn foil, ZnO, and ZnCl2) were acquired simultaneously for energy calibration. Multiple scans (3 times per sample) were averaged to improve signal-to-noise ratio. Trypsin stability assay To evaluate the proteolytic stability of NanoMel and melittin, the samples were incubated with trypsin at concentrations ranging from 0.5 to 2 mg/mL at 37 °C for 4 h. The enzymatic reaction was terminated by heating at 60 °C for 20 min. The treated samples were added to 96-well plates inoculated with a Xoo suspension and incubated at 28 °C until the OD₅₉₅ of the negative control reached 0.6. The absorbance at 595 nm was then measured to determine the remaining antibacterial activity. Determination of bacterial biofilm Biofilm inhibition assay Xoo bacterial solution (OD595=0.6–0.8) was adjusted to OD595 = 0.1 with sterile NB medium, and 200 µL was transferred to a sterile 96-well plate. Melittin and NanoMel compounds at concentrations of 0.5 × EC50 to 8 × EC50 were added, and after incubation at 28 °C for 48 h, the medium was discarded and non-adhered bacteria were removed by washing with sterile water. Each well was stained with 200 µL of 0.1% crystal violet for 30 min, washed with sterile water, air-dried, and the stained biofilm was dissolved in 200 µL of 95% ethanol. The OD570 was measured to quantify biofilm production. Biofilm eradication assay Xoo suspensions were incubated in 96-well microplates at 28 °C for 48 h to allow mature biofilm formation. After incubation, biofilms were washed twice with PBS, then treated with varying concentrations of melittin and NanoMel for 16 h. The supernatant was discarded, and planktonic cells were eliminated by three washes with 200 µL PBS. Using the same staining protocol as described in the Biofilm Inhibition Assay, each well was stained with 200 µL of 0.1% (w/v) crystal violet for 30 min. After removing the dye, the wells were rinsed with sterile water, air-dried, and the bound crystal violet was solubilized in 95% (v/v) ethanol, and the absorbance at 570 nm (OD₅₇₀) was measured to evaluate biofilm eradication. Biofilm-associated colony formation assay Sterile cell culture coverslips (10 × 10 mm) were placed in 6-well plates, and 8 mL of Xoo suspension (OD₅₉₅ = 0.1) was added to each well, followed by treatment with different concentrations of melittin and NanoMel. Following 48 h of incubation, the medium was discarded, 1 mL PBS was added, and the mixture was gently sonicated to dislodge and homogenize biofilm-associated bacteria. The resulting suspension was serially diluted 10⁵-fold and spread onto NA agar plates. After incubation at 28 °C for 3 days, colonies in each treatment group were photographed and counted. Confocal laser scanning microscopy (CLSM) analysis Sterile cell culture coverslips (10 × 10 mm) were placed in 6-well plates, and 8 mL of Xoo suspension (OD₅₉₅ = 0.1) was added. The plates were incubated at 28 °C for 48 h to allow biofilm formation, after which different concentrations of melittin and NanoMel were subsequently added. The samples were further incubated overnight at 28 °C. Planktonic bacteria were removed by aspirating the supernatant, and the biofilms on coverslips were stained with 0.1% (w/v) acridine orange (AO) for 10 min, followed by staining with 0.01% (w/v) propidium iodide (PI) for 5 min. All samples were visualized using a confocal laser scanning microscope (CLSM) equipped with 488 nm and 543 nm excitation lasers. Studies on cell membrane disruption-related mechanisms and other biological functions Conductivity assay Initially, Xoo cells grown in NB medium to an optical density at 595 nm (OD₅₉₅) of 0.6–0.8 were harvested and washed repeatedly with 5% (w/v) glucose solution until the conductivity of the cell suspension was consistent with that of the 5% (w/v) glucose solution, yielding the bacterial isotonic solution for subsequent use. The prepared bacterial isotonic solution was heated to boiling for 5 min, and the measured conductivity was designated L₀ (conductivity reference value for complete cell membrane disruption). Meanwhile, different concentrations of melittin and NanoMel were added to 5% (w/v) glucose solution, and the measured conductivity was recorded as L₁ to eliminate conductivity interference from the compounds themselves. Additionally, the same concentrations of melittin and NanoMel as described above were added to the bacterial isotonic solution, followed by incubation in a shaking incubator (180 rpm, 28 °C) for 8 h. The conductivity of the mixture was accurately measured every hour and denoted as L₂. Finally, the relative conductivity—reflecting changes in cell membrane permeability—was calculated using the formula: (L₂ − L₁)/L₀. Membrane potential Measurement.Membrane potential of plant pathogenic bacteria was determined using the membrane‑sensitive fluorescent probe DiBAC₄(3). Briefly, bacterial suspensions were cultured to the logarithmic phase (OD₆₀₀ = 0.6) and treated with free melittin or NanoMel at designated concentrations for 30 and 60 min at 28 °C. After incubation, bacterial pellets were collected by centrifugation, washed twice with phosphate‑buffered saline (PBS, pH 7.2), and resuspended in PBS. DiBAC₄(3) was added to a final concentration of 5 µM, and the mixture was incubated in the dark for 30 min at room temperature. Fluorescence intensity was measured using a microplate reader with excitation at 488 nm and emission at 516 nm. Untreated bacterial cells were set as the negative control, and the relative membrane potential variation was calculated based on the fluorescence intensity changes. All experiments were performed in triplicate. Determination of Extracellular Polysaccharides (EPS) Xoo suspension was adjusted to an optical density at 595 nm (OD₅₉₅) of 0.1 using NB medium supplemented with different concentrations of melittin and NanoMel. The mixture was incubated continuously in a shaking incubator (180 rpm, 28 °C) for 5 days. After incubation, all samples were centrifuged at 12,000 rpm for 10 min to pellet bacterial cells, and an aliquot of the supernatant was thoroughly mixed with an equal volume of 5 g/L aqueous phenol solution. Following this, 5 volumes of 95% (v/v) concentrated sulfuric acid were added rapidly to the mixture, which was then incubated in an ice bath for 10 min with gentle shaking every 2 min to ensure complete reaction. After the reaction, 200 µL of the reaction mixture was transferred to a 96-well microplate, and the absorbance at 490 nm (OD₄₉₀) was measured using a microplate reader. All experiments were performed as triplicate independent replicates. Bacterial suspensions of Xoo with an OD600 of 0.6 were treated with free melittin and NanoMel at a 4.0 × EC50, respectively, and incubated in a constant temperature shaking incubator at 28 ℃ and 180 rpm for 16 h. After incubation, Xoo cells were collected by centrifugation, and total RNA was extracted using the TransZol Up Kit. Total RNA was reverse transcribed into cDNA using TaKaRa (Takara Bio) reagents, with the gyrB gene encoding the B subunit of DNA gyrase as the internal reference gene for data normalization. Reverse transcription quantitative real-time PCR (RT-qPCR) detection was performed using SYBR Premix Ex TaqII reagents on the Bio-Rad iCycler iQ Real-Time Fluorescent Quantitative PCR System, and the relative expression levels of target genes were calculated using the 2^-∆∆Ct method. Six technical replicates were included for each experimental group. Determination of extracellular protein content Xoo suspension was adjusted to an optical density at 595 nm (OD₅₉₅) of 0.1 using NB medium supplemented with different concentrations of melittin and NanoMel. The mixture was incubated in a shaking incubator (180 rpm, 28 °C) for 5 days. After incubation, samples were centrifuged at 10,000 rpm for 20 min at 4 °C. The supernatant was discarded, and the bacterial pellet was collected. The pellet was resuspended in 10 mL of ddH₂O, followed by ultrasonic disruption with the following parameters: 200 W power, 3 s on/3 s off cycle, and total duration of 10 min. Samples were centrifuged again under the same conditions (10,000 rpm, 20 min, 4 °C), the pellet was discarded, and an aliquot of 0.5 mL of the supernatant was aspirated. Subsequently, 2 mL of Coomassie Brilliant Blue G250 staining solution was added, and the mixture was incubated in the dark for 5 min. Using pure Coomassie Brilliant Blue G250 staining solution as the blank control, the absorbance at 595 nm (OD₅₉₅) was measured with a microplate reader, and the relative extracellular protein (EPN) content was calculated from the absorbance values. Swimming motility assay Xoo was cultured to an optical density at 595 nm (OD₅₉₅) of 1.0. For each assay, 10 mL of semi-solid NA medium supplemented with the target compound (12.0 µg/mL, i.e., 2.0×EC₅₀) was dispensed into a 90 mm sterile Petri dish. After solidification, 2 µL of the prepared Xoo suspension was spotted onto the center of each plate using a sterile pipette. Petri dishes were sealed and incubated in an incubator at 28 ± 1 °C for 3–4 days. After incubation, the colony diameter was measured with a vernier caliper. Determination of amylase and cellulase inhibition activity A 1 mL aliquot of Xoo suspension (OD₅₉₅ = 0.1) was combined with either melittin (at 2.0×EC₅₀) or an equivalent dose of NanoMel in a sterile 2 mL centrifuge tube. The mixture was vigorously vortexed and incubated at room temperature for 60 min to facilitate sufficient compound–bacterial interaction. Separately, a base layer was prepared by pouring 10 mL of 1.5% (w/v) agar solution into a 90 mm sterile Petri dish and allowing it to solidify at room temperature. Four pre-sterilized Oxford cups were then placed evenly on the solidified base agar and gently pressed to ensure secure adhesion. Subsequently, 10 mL of NB solid medium containing 0.1% soluble starch and 1.5% (w/v) agar powder was melted, cooled to ~50 °C, and overlaid to form the top agar. Once the top layer was fully solidified, the Oxford cups were carefully removed to generate four uniform wells. A 2 µL aliquot of each compound–bacterial mixture was dispensed into an individual well. Plates were sealed and incubated statically at 28 ± 1 °C for 3 days. After incubation, the bacterial lawns were rinsed with ddH₂O, stained with 10 mL of 0.5% (w/v) iodine/potassium iodide (I₂/KI) solution for 10 min, and destained with 70% (v/v) ethanol until the background cleared. The diameter of the resulting clear hydrolysis zone around each well was measured. Determination of cellulase inhibition activity Sterile nutrient agar (NA) plates were prepared with 0.5% (w/v) sodium carboxymethyl cellulose (CMC-Na). After solidification, 2 µL of the treated Xoo suspension (prepared as described above) was spotted onto each plate using the same protocol as the amylase inhibition assay. Plates were incubated at 28 °C for 72 h. After incubation, the plates were stained with 0.1% (w/v) Congo red solution for 30 min, followed by two rinses with sterile ddH₂O to remove unbound dye. Decolorization was achieved by incubating the plates with 1 mol/L sodium chloride solution. Finally, the plates were imaged using a documentation system, and the diameter of the cellulose hydrolysis zone surrounding each bacterial colony was measured. Pathogenicity assay Xoo cells cultured to the logarithmic growth phase were adjusted to an optical density at 595 nm (OD₅₉₅) of 0.6 in NB medium. The bacterial suspensions were then incubated with melittin or NanoMel at 2.0 × EC₅₀, with shaking (180 rpm) at 28 °C for 12 h. Inoculation was performed via the leaf-clipping inoculation method. Fourteen days post-inoculation, lesion development on rice leaves was monitored, and the length of each lesion was measured accurately with a vernier caliper. Each treatment was conducted in three biological replicates, with ten rice plants per replicate. In vivo antibacterial assay Control assay for plant bacterial diseases caused by Xoo The protective and curative activity of melittin and NanoMel against Xoo-induced bacterial blight in rice was evaluated. Rice plants were treated with melittin or NanoMel at 200 µg/mL. After 24 h, a 2 cm diagonal incision was made at the leaf tip with sterile scissors, and the wound was inoculated by immersing it in an Xoo suspension (OD₅₉₅ = 0.8). All treated plants were subsequently maintained under controlled conditions (28 °C, 90% relative humidity, RH) for 14 days. The disease index and control efficacy were calculated based on the diseased area using the designated formula. According to the severity level of symptoms and the prescribed formula, the corresponding disease index (C and T stand for lesion length of negative and treatment groups, respectively) was obtained for each group: The following equation was used to assess the anti-Xoo activity: Disease control assay against Xoc The protective and curative activity of melittin and NanoMel against Xoc-induced bacterial leaf streak was evaluated in a pot experiment via the pressure infiltration method. Rice plants were first sprayed with the test compounds (melittin and NanoMel) at 200 µg/mL. After 24 h, the plants were inoculated with Xoc using a needleless syringe. All treated plants were subsequently maintained in a greenhouse for 14 days. An equivalent dose of aqueous ZnCl₂ solution was used as the negative control, with TC serving as the positive control. Lesion lengths on rice leaves were measured, and control efficacy (I) was calculated using the respective formula, where C denotes the negative control group and T the treatment group. Biosafety evaluation Phytotoxicity assay on target plant (Rice) Rice seeds were first treated with melittin or NanoMel at 100 µg/mL for 7 days, and seed germination rates were monitored and recorded. Subsequently, melittin and NanoMel solutions at 500 µg/mL were uniformly sprayed onto the leaves of mature rice plants, ensuring full coverage of all leaves. An aqueous solution was used as the negative control to exclude potential solvent-induced effects on rice plants. Treated plants were transferred to a greenhouse and maintained under controlled temperature and humidity conditions. 7 days post-treatment, lesion symptoms (if any) were evaluated and recorded. Zebrafish embryo toxicity assay For the toxicity assessment, zebrafish embryos at 24 h post-fertilization (hpf) were placed in a 6-well plate and exposed to various concentrations of NanoMel and melittin. Morphological changes of the embryos were observed and recorded at three time points: 24, 48, and 72 h. Survival rate was calculated as the percentage of survival embryos relative to the total number of embryos, using the following formula. The malformation rate was defined as the percentage of embryos exhibiting any type of morphological abnormality (including both malformed and dead embryos) relative to the total number of embryos, calculated as: All calculations were performed across three independent biological replicates (20 embryos per replicate) to ensure statistical robustness. Transcriptomic analysis of zebrafish embryos exposed to 45 µg/mL melittin and NanoMel Zebrafish embryos at 24 h post-fertilization (24 hpf) were selected for transcriptomic analysis. Using the latest version of the zebrafish genome (GRCz11) as the reference, mainstream bioinformatics software was adopted for the analysis, whose core contents included gene alignment, screening of differentially expressed genes (DEGs), and functional annotation and enrichment analysis. Twenty embryos were assigned to each group, with three biological replicates established for every group. Reactive oxygen species (ROS) detection in zebrafish embryos treated with melittin AND NanoMel Zebrafish embryos at 24 hpf were treated with free melittin and NanoMel at a final concentration of 45 µg/mL, respectively, maintained in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM anhydrous MgCl2, pH 7.2–7.4) and co-incubated at 28 ℃ for a certain period; after the treatment, intracellular reactive oxygen species (ROS) levels were detected using the DCFH-DA fluorescent probe, briefly, the DCFH-DA stock solution (10 mM) was prepared by dissolving DCFH-DA powder in dimethyl sulfoxide (DMSO) and stored at −20 ℃ in the dark, the treated embryos were washed three times with fresh E3 medium to remove residual compounds, subsequently incubated with 20 µM DCFH-DA working solution (diluted from the 10 mM stock solution with E3 medium) at 28 ℃ for 30 min in the dark, and after incubation, the embryos were rinsed 3–5 times with E3 medium to eliminate unbound DCFH-DA probe; for quantitative detection, the embryos were first anesthetized with 0.02% tricaine methanesulfonate (MS-222) and mounted on glass slides, fluorescence images were captured using a confocal fluorescence microscope, the mean fluorescence intensity of individual embryos was quantified using ImageJ software with 10 embryos analyzed per group, and all experiments were performed in triplicate to ensure reproducibility. Cell apoptosis detection in zebrafish embryos Detection of Cell Apoptosis in Zebrafish Embryos Cell apoptosis in zebrafish embryos was detected by acridine orange (AO) fluorescent staining. Briefly, zebrafish embryos at 24 hpf were treated with free melittin and NanoMel at a final concentration of 45 µg/mL, respectively, for 72 h, and the embryos were detected at different time points. After the treatment, the embryos were washed three times with fresh E3 medium to remove residual compounds, and then co-incubated with 2 µg/mL AO solution (diluted with E3 medium before use) for 30 min in the dark; after incubation, the embryos were rinsed three times with E3 medium to eliminate unbound AO probe so as to reduce background fluorescence, then anesthetized with 0.02% tricaine methanesulfonate (MS-222) and mounted on glass slides. Fluorescence images were collected using a confocal fluorescence microscope at the FITC channel, apoptotic cells showed bright yellow-green punctate fluorescence, with 10 embryos repeated per group and all experiments set up with three replicates to ensure reproducibility. Statistical analysis All statistical analyses and plotting of statistical graphs were performed using Origin and GraphPad Prism software. Unless otherwise specified in the figure legends, all data are presented as the mean ± standard error (SE) of three independent biological replicates with a sample size of 3 per replicate. Statistical differences were analyzed using SPSS software, with one-way ANOVA combined with Tukey’s post-hoc test or independent Student’s t-test adopted for comparison. A P-value ≤ 0.05 was regarded as statistically significant. Among them, *P < 0.05, **P < 0.01, and ***P < 0.001 indicate significant, extremely significant, and highly significant differences, respectively; ns represents no statistically significant difference. Supplementary Information Supplementary Material 1 Acknowledgements We thank Qiong Yi from the Southwest Medicinal Bioengineering Center of the Ministry of Education for assistance with confocal laser scanning microscopy (CLSM) measurements. Author contributions X. Z., L. Z. conceived the project and designed the experiments. X. Z., J. H., L. Y., K., K performed the experiments and X. Z., J. H., L. Y., P. W., Y. Y. and L. Z. collected and analyzed the data. X. Z., J. H., L. Y. P. W., S. N., M. C. and L. Z. wrote the manuscript. Y. W., J. M., G. H., P. W. provided technical support and edited the manuscript. All authors discussed the results and reviewed the manuscript. Funding This work was supported by the National Key Research and Development Program of China (No. 2024YFE0214300), Post-Subsidy Fund Project of Guizhou International Science and Technology Cooperation Base (Qiankehe Platform GHJD (2025) 008), the National Natural Science Foundation of China (NSFC) Excellent Young Scientists Fund (Overseas, L.Z.) and the starting grant of Guizhou University (L.Z.). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Ethics approval was not required for this study. Consent for publication All authors involved in this study have provided their consent for the publication of the research findings. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Peiyi Wang, Email: pywang@gzu.edu.cn. Libo Zhang, Email: lbzhang@gzu.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement No datasets were generated or analysed during the current study.