Soil Fungi Borrowed a Bacterial Gene Millions of Years Ago to Freeze Water, Opening a Path to Safer Cloud Seeding
A Science Advances study identifies cell-free ice-nucleating proteins in Mortierellaceae soil fungi, acquired from bacteria via horizontal gene transfer, offering a non-toxic alternative to silver iodide for cloud seeding.
A Hidden Talent in Common Soil Fungi
Scientists have known for decades that certain soil-dwelling fungi can trigger ice formation at temperatures well above the point at which pure water freezes on its own. What they did not know was how the fungi pulled it off. A study published in Science Advances on March 11, 2026, led by Rosemary J. Eufemio and an international team spanning Virginia Tech, Boise State University, the University of Utah, the Air Force Research Laboratory, and Germany’s Max Planck Institutes for Polymer Research and Chemistry, now identifies the proteins responsible and traces their unexpected origin.
The proteins belong to fungi in the Mortierellaceae family, a group widespread in soils around the world. Unlike bacterial ice nucleators, which require intact cell membranes to function, the fungal versions are cell-free and water-soluble, as Phys.org reported. That distinction has immediate practical significance: the proteins can operate independently of the organisms that produce them.
What the Researchers Found
The team sequenced the genomes of ice-nucleation-active Mortierellaceae strains, including Mortierella alpina strain 13A and a culture isolated from the lichen Peltigera britannica. They identified orthologs of the bacterial InaZ gene, a well-characterized ice-nucleation gene, embedded in the fungal genomes. The fungal proteins are predicted to fold into beta-solenoid structures and to multimerize, forming extended ice-binding surfaces that catalyze the phase transition from liquid water to ice at high subzero temperatures, according to the Science Advances paper.
To confirm the genes’ function, the researchers expressed them heterologously in both Escherichia coli and Saccharomyces cerevisiae. In both cases, the engineered organisms gained ice-nucleation activity they did not previously possess, providing direct evidence that the identified genes encode the responsible proteins.
The genomic analysis further revealed that the fungal InaZ orthologs were acquired from a bacterial ancestor through horizontal gene transfer, a process in which genetic material jumps between unrelated organisms rather than passing from parent to offspring. Boris A. Vinatzer, a co-author at Virginia Tech’s School of Plant and Environmental Sciences, noted that while gene transfer from bacteria to fungi is documented, “it’s not something that is common,” as Phys.org reported. The transfer occurred at least hundreds of thousands, and possibly millions, of years ago. In the intervening time, the fungi modified the gene to enhance the protein’s performance.
This finding builds on earlier work establishing the molecular basis of fungal ice nucleation. A 2023 study published in the Proceedings of the National Academy of Sciences by researchers at the University of Utah, Boise State University, and the Max Planck Institute showed that ice-nucleating proteins from the related fungus Fusarium acuminatum consist of small 5.3-kilodalton subunits that assemble into complexes of more than 100 units. Those complexes can trigger freezing at temperatures between negative 2 and negative 10 degrees Celsius, far warmer than the negative 46 degrees Celsius at which pure water spontaneously crystallizes.
Replacing Silver Iodide in Cloud Seeding
The most immediate application under discussion is weather modification. Cloud seeding currently relies on silver iodide particles dispersed into clouds to trigger ice crystal formation, which can induce precipitation. Silver iodide is effective but environmentally toxic. “If we learn how to cheaply produce enough of this fungal protein, then we could put that into clouds and make cloud seeding much safer,” Vinatzer said, according to Phys.org.
Because the fungal proteins function outside of cells, they could in principle be manufactured through recombinant expression systems and deployed as a purified biological reagent, sidestepping the need to release living organisms into the atmosphere.
Beyond Weather: Food and Medicine
The cell-free nature of the proteins also opens doors in food science and biomedicine. In frozen food production, controlling the size and distribution of ice crystals determines texture and quality. Bacterial ice nucleators have been explored for this purpose, but using them means introducing whole bacterial cells into food products. The fungal proteins, secreted freely into solution, avoid that problem.
In cryopreservation, the proteins could improve the freezing of biological materials such as tissues, sperm, eggs, and embryos. Adding a small, soluble ice-nucleation molecule causes water surrounding cells to freeze at higher temperatures, reducing the formation of damaging intracellular ice crystals that occurs when samples are cooled too rapidly past the nucleation point.
Implications for Climate Models
The discovery also carries consequences for atmospheric science. Mortierellaceae fungi are abundant in soils globally, and their spores travel through the atmosphere. If those spores carry cell-free ice-nucleating proteins, as the laboratory results suggest, then biological ice nucleation in the atmosphere may be more prevalent than current climate models assume.
An estimated 99 percent of soil microorganisms remain unstudied, and fungi are among the most underrepresented groups in atmospheric models. Quantifying the contribution of fungal ice nucleators to cloud formation and precipitation patterns could refine projections of regional rainfall and climate feedbacks.
What Remains to Be Determined
The research establishes the molecular identity of the proteins and their evolutionary origin but does not yet address how efficiently they can be produced at industrial scale, what their atmospheric concentrations are under field conditions, or how regulatory frameworks would treat a biologically derived cloud-seeding agent. The study was supported by the National Science Foundation, the Department of Defense, and the Air Force Office of Scientific Research, suggesting that at least some of those questions are already attracting federal research investment.
Whether a gene that jumped from bacteria to fungi millions of years ago will ultimately replace the silver iodide flares fired into storm clouds today depends on engineering and economics that remain years from resolution. What the Science Advances paper establishes is that the biological raw material for that transition exists, is well characterized, and works.