Global banana production is currently facing an existential threat from Fusarium wilt, a soil-borne fungal disease. The most aggressive strain, Tropical Race 4 (Foc_TR4), has decimated plantations worldwide, particularly affecting the Cavendish variety, which dominates international trade.
Because modern cultivated bananas are clones with almost no genetic diversity, they lack the natural toolkit to fight off these evolving pathogens.
However, a new study led by the Chinese Academy of Sciences suggested that the answer to this crisis could be found in the “wild”. The study’s research team successfully utilised Musa cheesmanii — a wild relative of the banana — to breed new hybrids that were not only resistant to Foc_TR4 but also produced superior fruit.
The power of wild relatives
Crop wild relatives are essential for modern breeding because they hold the genetic diversity commercial crops have lost over centuries of human selection. While the banana genus Musa contains approximately 70 wild species, very few have been used in commercial breeding programmes. Historically, breeders have relied on a narrow pool of ancestors, leading to a genetic bottleneck that makes the crop vulnerable.
Musa cheesmanii, native to the high-altitude forests of the Himalayas and northeastern India, has long been a species of interest. It is a hardy plant, capable of thriving in cooler temperatures and high humidity. Most importantly, field observations showed that it remained entirely disease-free even when grown in soil heavily contaminated with the deadly Foc_TR4 fungus.
Identifying potential male parents from banana wild relatives is promising for developing superior cultivars, given their significant role in enhancing cultivar diversity and their limited use in current breeding programs.
Breeding for resistance and quality
The research team crossed M. cheesmanii with two popular triploid cultivars in China, Yulin and Jinyi. These cultivars belong to the ABB genome group, which are prized for their cooking and eating qualities but suffer from low fertility and disease susceptibility.
The results of these crosses were remarkable. The hybrid offspring, named Haifen No. 1 and Haijiao No. 1, showed significantly enhanced resistance to Fusarium wilt. In greenhouse tests, while susceptible varieties died within 34 days, the M. cheesmanii hybrids showed either no symptoms or a significantly delayed onset of the disease.
Beyond disease resistance, the hybrids outperformed their parents in several commercial metrics. The bunch weights of the hybrids were nearly double those of the maternal parents. For example, hybrids derived from Yulin reached an average bunch weight of 17.8 kg, compared to much lower yields from other wild crosses.
The plants also exhibited a sturdier architecture, with thicker stems that provided better resistance to wind damage — a major concern for plantation owners in tropical regions.
Improving the consumer experience
A critical finding was the improvement in fruit quality. Historically, breeding for disease resistance has often come at the cost of taste or shelf-life, but this study has shown that does not have to be the case.
Sensory evaluations conducted by a panel of testers revealed that the hybrids scored higher than the original cultivars in overall acceptability. Chemical analysis supported these findings, showing significantly higher levels of soluble sugars, sucrose, and beta-carotene. This suggested that the new hybrids were not only sweeter but also more nutritious.
Logistics and retail sectors may also find value in the extended shelf-life of the new hybrids. The study found that the fruit lasted between six and 10 days after ripening, compared to just four to five days for the standard Yulin variety. The pulp remained firm even as the peel began to colour, reducing waste during transport and on-shelf display.
A genomic roadmap for the future
To understand why M. cheesmanii was so resilient, the researchers produced a ‘telomere-to-telomere’ (T2T) gapless genome assembly. This is the highest standard of genomic mapping available, providing a complete blueprint of the plant’s DNA.
By comparing this genome to other banana species, the team identified specific gene expansions related to defence mechanisms and carbohydrate metabolism. They also investigated the species’ unique black pseudo-stem, identifying the delphinidin-based pathway as the primary driver of its dark pigmentation.
Crucially, the genomic analysis addressed a major hurdle in banana breeding: the Banana Streak Virus (BSV). Many wild relatives carry endogenous versions of this virus within their DNA, which can become active and cause disease in hybrid offspring. The researchers found that in M. cheesmanii, these viral sequences were highly fragmented and non-functional. This made the species a much safer “male parent” for breeding than other wild relatives, as the risk of viral outbreaks was significantly lower.
Implications for the industry
The success of these hybrids marks a significant shift in banana improvement strategies. By moving away from the limited genetic pool of traditional ancestors and “going wild,”, breeders can introduce hardiness and quality traits that were previously thought to be inaccessible.
The study’s authors stated that M. cheemanii was a highly promising candidate for use as a male parent in future banana breeding programs, as its use could significantly broaden the genetic base of bananas, leading to the development of high-yield, disease-resistant varieties.
While they also noted that the tall stature and long flowering cycles of M. cheesmanii mean it might not replace commercial giants like Cavendish immediately, it could be an invaluable source for “improved diploids”. These could then be integrated into existing breeding pipelines to create the next generation of global staple bananas.
Source: Nature Communications
“Going wild in banana breeding enables Fusariumresistant hybrids with improved fruit quality”
https://doi.org/10.1038/s41467-026-70186-9
Authors: Liu Xin, et al
