Rhizosphere microorganisms play pivotal roles in mitigating the challenges associated with continuous cropping in melon cultivation. While grafting and plant growth-promoting rhizobacteria (PGPR) independently influence rhizosphere microbial communities, their combined effects remain largely unexplored. This study investigates the synergistic regulation of Paenibacillus polymyxa NBmelon-1 inoculation and grafting on rhizosphere microbiome assembly, plant performance, and disease resistance in melon self-rootstocks. Field experiments demonstrated that NBmelon-1 inoculation significantly enhanced rootstock stem diameter (95.3% increase in spring) and root development, achieving a graft survival rate exceeding 95%. The combined treatment (NB+GJ) increased scion fruit yield by 29.8% in autumn and 36.5% in spring, as well as the single-fruit weight by 22.5% in autumn and 37.3% in spring, while maintaining fruit morphology. Integrated 16S rRNA and ITS sequencing revealed that the NB+GJ treatment selectively enriched antagonistic bacterial phyla (e.g., Firmicutes and Actinobacteriota) and suppressed pathogenic fungi (e.g., Fusarium and Melanconiella). Seasonal shifts in microbial diversity and functional gene profiles underscored the dynamic interplay between treatments and environmental factors. These findings establish a novel strategy for optimizing melon self-rootstock grafting systems and sustainably managing soil-borne diseases.

In May of 2020, November of 2021 and May of 2022, a preharvest fruit rot with white mycelia was observed inside and outside of the fruits of thick skin muskmelon (Cucumis melo L.) growing in about ten greenhouses (each greenhouse had about 320 muskmelons) with disease incidence of 70% in Ningbo, Zhejiang Province of China. In order to identify the causal agent, plant tissues from the margin of the symptomatic tissue were sterilized for 1 min with 1% sodium hypochlorite (NaClO), 2 min with 75% ethyl alcohol, rinsed in sterile distilled water three times (Zhou et al 2019), and then placed on potato dextrose agar (PDA) plates containing streptomycin sulfate (100 μg/mL) at 25℃ for 4 days. Only Fusarium colonies were isolated from all the plant tissues. The growing hyphae were transferred to new PDA plates using the hyphal tip method, putative Fusarium colonies were purified by single-sporing. Six fungal isolates (Fi-1~6) were obtained. The average radial mycelial growth rate of Fusarium isolate Fi-3 was 4.6 mm/day at 25℃ in the dark on PDA, and like other five isolates. The colonies are abnormal, producing lots of aerial hyphae, each isolate was white to light orange. Isolate Fi-3 produced macroconidia with 4 to 6 septa, tapered with pronounced dorsiventral curvature and measured 21 to 30 μm long 4 to 5 μm wide on Spezieller Nährstoffarmer Agar (SNA) medium at 25℃ for 10 days (Leslie and Summerell 2006), but polyphialides and chlamydospores were still not available for 30 days. The pathogen species was further identified by translation elongation factor-1 alpha (EF-1α) sequencing. The EF-1α of six isolates were sequenced, and their EF-1α sequences were 100% identical to each other, and the sequence of strain Fi-3 was deposited in GenBank with accession no. OL782040 and was also compared with sequences in the FUSARIUM-ID database (Geiser et al. 2004), which indicated that it was 100% identical to those of F. pernambucanum strain NRRL 32864 (GenBank accession GQ505613), F. pernambucanum strain LC7040 (GenBank accession MK289626), and F. pernambucanum strain LC12149 (GenBank accession MK289588) within the Fusarium incarnatum - F. equiseti species complex 17 (FIESC17). Two phylogenetic trees were established based on the TEF1-α sequences of Fi-1~6 and other Fusarium spp., Fi-1~6 was clustered with the sequences of F. pernambucanum within the FIESC17. Thus, both morphological and molecular criteria supported identification of the strain as F. pernambucanum. A pathogenicity test was conducted to verify Koch's postulates, mycelium agar plugs (6 mm in diameter) were removed from the colony margin of a 3-day-old culture of strain Fi-3, healthy melon fruits were surface-sterilized with 70% ethanol and rinsed twice with sterile-distilled water. Then, the melons were wounded using a sterile inoculating needle to stab and inoculated by a mycelium agar plug of strain Fi-3 on the wound sites. 5 fruits were inoculated in each treatment, and a mycelium-free PDA plug was used as a negative control, repeated 3 times, at 25℃ with high relative humidity for 10 days. The results show disease symptoms similar to those naturally infected fruits on all inoculated melon fruits. The fungus re-isolated from the diseased fruits, showed the same colony morphology as the original isolate. Koch's postulates were repeated three times with the same results. Strain Fi-3 inoculated fruits without wounding remained healthy. To our knowledge, this is the first report of fruit rot of melon caused by F. pernambucanum in China.

In May of 2024, a stem soft rot disease in melon (Cucumis melo L.) was observed in Ningbo (29.52°N, 121.31°E), Zhejiang Province of China. The disease incidence was approximately 30% in two greenhouses (ca. 8 m × 50 m each). The soft rot yellowish-brown lesions developed on infected stems, soft rotted with a bad foul odor. To identify the pathogen, the diseased stems from four plants were surface sterilized by soaking in 75 % ethanol for 30 s, followed by rinsing three times in sterile distilled water (SDW). The stems were cut in 5 mm fragments and let stand in SDW. After 5 minutes 100 µL of the water was spread on nutrient agar (NA). After incubation at 28°C for 24 h, seven bacterial colonies were selected and individually transferred to new NA plates. Three strains (PB052001, PB052002 and PB052003) were randomly selected as representatives for further study. They were streak-inoculated on NA for formation of single colonies (28°C, 3 d). The colonies were white and round in shape with neat, slightly raised margin. Total genomic DNA was extracted from the three bacterial strains using EasyPure Bacterial Genomic DNA Kit (TransGen, Beijing). It was used as template for PCR amplification of 16S rDNA as well as the five house-keeping genes (pgi, proA, mtlD, mdh and icdA) using specific primers (Lane 1991; Ma et al. 2007; Waleron et al. 2008). The resulting DNA fragments were cloned and sequenced. The three bacterial strains had 100% identical DNA sequences at each of the six loci, therefore, only the DNA sequences from strain PB052001 were deposited in GenBank. Sequence analysis showed that the 16S rDNA in PB052001 (GenBank Acc No. PQ146559) was 99.67% (1501/1506 nt) and 99.60% (1500/1506 nt) identical to those in Pectobacterium brasiliense SR10 (CP084655) and 21PCA AGRO2 (CP113504), respectively. A multilocus phylogenetic analysis was done based on pgi, proA, mtlD, mdh and icdA from PB052001 (GenBank Acc Nos. PQ145967, PQ145968, PQ145969, PQ145970 and PQ145971, respectively). The result showed that the three strains clustered together with P. brasiliense. To confirm pathogenicity, a bacterial suspension of PB052001 (108 cfu/mL) was injected into stems (10 per treatment) of melon seedlings (cultivar Fengsu No.10) at 1 cm below the cotyledons, 10 µL for each plant. Ten seedlings was injected SDW as negative control. The treated seedlings were grown at 28°C under 80% relative humidity. After 24 h, while the control plants remained healthy, the plants inoculated with the bacteria showed soft rot symptoms. Additionally, the bacterial suspension was sprayed on melon leaves (10 per treatment), and SDW was sprayed on melon leaves as a control. The plants were incubated under the same conditions described above. After 3 d, the control leaves remained healthy, the leaves treated with the bacteria exhibited soft-rot symptoms. Bacteria were re-isolated from the diseased tissues, and morphological and molecular identification showed that it was identical to PB052001. P. brasiliense has a broad host range, especially in Cucurbitaceae such as cucumber and watermelon (Ma et al. 2007; Waleron et al. 2008). To our knowledge, this is the first report of P. brasiliense causing soft rot disease on melon in China. The finding expands our knowledge about the host range of P. brasiliense. This disease poses a threat to the local melon industry.