Wine Yeast’s Secret Global Highways Revealed by Population Genomics

A humble yeast that helps give wine its fruity aromas has been hiding a remarkable secret: despite living on different continents, its populations have been quietly swapping genes across the globe. A new large-scale genomic study of the non-Saccharomyces yeast Hanseniaspora uvarum, a species that dominates the early stages of spontaneous wine fermentation, has uncovered extensive intercontinental gene flow, region-specific signatures of natural selection, and an unexpectedly open pangenome. The findings, published in the journal Stress Biology, provide the most comprehensive picture yet of the evolutionary dynamics of a yeast that is rapidly becoming a darling of both food engineering and evolutionary biology.

The research team, led by scientists at Northwest A&F University in Yangling, China, sequenced the whole genomes of 65 wine-related H. uvarum strains isolated from major wine-producing regions of northwestern China, including Shaanxi, Ningxia, Gansu, and Xinjiang. These newly generated sequences were combined with 86 publicly available genome datasets, yielding a total of 151 strains for population genomic analysis, with samples drawn from China, Australia, Europe, the United States, and other locations. The newly sequenced strains showed exceptionally deep sequencing coverage, ranging from roughly 726-fold to more than 1,500-fold, ensuring high confidence in the variants detected.

From the combined dataset, the researchers identified 575,222 high-quality genetic variants, including more than half a million single nucleotide polymorphisms, after masking repetitive and low-complexity regions of the reference genome. Downstream analyses focused on 483,564 biallelic SNPs. Three independent approaches—phylogenetic tree construction, ancestry coefficient estimation, and principal component analysis—told a consistent story: strains sampled in China are broadly distinguishable from those sampled on other continents, confirming that geographic isolation has played a major role in shaping intercontinental population divergence in this species.

Yet the picture is far more tangled than a simple tale of isolated populations drifting apart. Using the TreeMix software to model migration between clades, the team detected substantial gene flow between geographically paired clades after their divergence. In the optimal model, gene flow was inferred from a Chinese clade to an Australian clade, from another Chinese clade to a European clade, and from the European clade back to a third Chinese clade. The f-branch statistic, a method that detects excess allele sharing between paired lineages, corroborated these signals: the highest intercontinental value, 49.84 percent, linked a Chinese clade with an Australian clade, with additional elevated values connecting Chinese, European, and Australian lineages. Together, these results point to ongoing intercontinental dispersal of H. uvarum, likely facilitated by human activity or insect vectors, a pattern reminiscent of what has been documented in the better-known wine yeast Saccharomyces cerevisiae.

Beyond connectivity, the study revealed that local environments have left distinct evolutionary fingerprints on different populations. Scanning for selective sweeps with the RAiSD tool, the researchers identified 117 candidate genes under positive selection in strains from Ningxia, China, and 141 in Australian strains, with only 23 genes shared between the two sets. These positively selected regions showed significantly elevated genetic differentiation and reduced diversity compared to the genomic background—classic hallmarks of recent adaptation. The Ningxia genes were significantly enriched for Gene Ontology terms related to the positive regulation of filamentous growth in response to external stimuli, such as starvation, pH stress, and biotic stimulus. The authors suggest these adaptations may help Ningxia strains colonize fruit surfaces, echoing earlier proposals that the yeast’s ability to form pseudohyphae in rich media confers an advantage for fruit colonization.

Intriguingly, the stress-related GO term “cytoplasmic stress granule” was significantly enriched among positively selected genes in both the Ningxia and Australian subsets, but the two regions shared only two of the six stress granule-associated genes in each set. Stress granules are cytoplasmic assemblies that sequester inactive messenger RNAs under conditions such as heat shock, oxidative stress, and nutrient deprivation, allowing cells to preferentially translate stress-response proteins. The fact that different genes underpin this same function in the two regions suggests that Ningxia and Australian populations have faced distinct environmental pressures and have evolved different molecular solutions to cope with stress.

The study also touched on one of the most curious features of the genus Hanseniaspora: its extensive loss of DNA repair genes. Previous work showed that the Faster-Evolving Lineage, to which H. uvarum belongs, lost dozens of genes involved in cell-cycle control and DNA repair, and also evolved a novel cis-regulatory mode for its core histone genes. Despite this genetic erosion, the new analysis found that DNA repair terms showed marginal enrichment among positively selected genes in the Ningxia subset, and that two classes of histone acetyltransferase complexes involved in DNA repair—the NuA4 complex and the SAGA and SLIK complexes—were significantly or marginally enriched in both regional subsets. Moreover, the pangenome analysis revealed that a subset of strains may carry DNA repair complex genes not present in the rest of the species, hinting that some populations have partially compensated for ancient gene losses through acquired or retained genetic material.

To capture the full gene repertoire of the species, the team constructed a pangenome from 159 strains, including eight additional genome assemblies. They identified 4,991 gene families, of which 3,340 were present in all strains (the core), while the remainder were divided into soft-core, shell, cloud, and private categories. Perhaps the most striking result came from Heaps’ Law analysis: with a decay parameter of 0.51—well below the threshold of 1—the H. uvarum pangenome is open, meaning new gene families keep appearing as more genomes are sampled. This is notable because the sampling focused almost entirely on anthropogenic environments such as vineyards and wineries, a relatively homogeneous niche. An earlier pangenome analysis based on just eight strains had concluded the pangenome was closed; the new, much larger dataset overturns that view and suggests the species harbors substantial adaptive potential for diverse stresses.

The practical implications reach well beyond evolutionary theory. H. uvarum is valued in winemaking for its ability to increase esters and volatile phenols, reduce volatile acidity, and improve the sensory profile of wine when used in co-fermentation with Saccharomyces cerevisiae. It also participates in fermentations of cider, cocoa, coffee, and pineapple wine, and related species serve as biocontrol agents against postharvest fruit diseases. A genomic understanding of how regional populations differ—in stress tolerance, filamentous growth, and metabolic capacity—could guide the selection of indigenous starter strains tailored to specific terroirs and fermentation goals, including recent efforts to screen Chinese Hanseniaspora strains for ethanol reduction in wine.

The authors caution that their sampling, while broad, remains uneven: wild-niche strains were absent because genomic data were not publicly available, and Chinese sampling was concentrated in Ningxia and Shaanxi. Future studies incorporating strains from more diverse ecological niches, broader geographic regions, and longer time spans will be needed to refine the evolutionary picture. Nevertheless, by combining deep whole-genome sequencing, phylogenetics, admixture modeling, selective sweep detection, and pangenomics, the study delivers high-quality genomic resources and a foundational framework for population genomics of non-Saccharomyces yeasts—a field long overshadowed by work on Saccharomyces cerevisiae. For a species once considered merely a fleeting early fermenter, Hanseniaspora uvarum is proving to be a genetically rich, globally connected, and remarkably adaptable microbe whose evolutionary story is only beginning to unfold.

Subject of Research: Population genomics, gene flow, and positive selection in the wine-related yeast Hanseniaspora uvarum

Article Title: Population genomics reveals gene flow and positive selection patterns in the wine-related yeast Hanseniaspora uvarum

Article References: Ma, R., Wang, H., Wei, Y., Sun, Y., Xue, J., Qin, Y., Tao, S., & Liu, Y. (2026). Population genomics reveals gene flow and positive selection patterns in the wine-related yeast Hanseniaspora uvarum. Stress Biology, 6(1), Article 54. https://doi.org/10.1007/s44154-026-00319-z

Image Credits: AI Generated

DOI: 10.1007/s44154-026-00319-z

Keywords: Hanseniaspora uvarum, non-Saccharomyces yeasts, population genomics, gene flow, introgression, positive selection, pangenome, wine fermentation, selective sweep, stress granules, DNA repair, geographic divergence