Across 272 progeny grown over two years, skin texture proved 93 per cent heritable, and the chemistry divided cleanly: suberin in smooth skins, lignin in russeted ones.

Russeting is the rough, netted brown skin of a baking potato, and it is not a defect. It is a trait breeders select for or against depending on the market — russets for baking and processing, smooth skins for the fresh-market varieties shoppers prefer to see in a bag. Which makes it useful to know what causes it, and until now the genetics have been unclear.

A study in Horticulture Research from the Chinese Academy of Agricultural Sciences maps the trait and finds it unusually tractable: almost all of the variation is genetic, and most of it sits in one place.

The work was done at the Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences, within the State Key Laboratory of Vegetable Biobreeding and the Ministry of Agriculture and Rural Affairs’ key laboratory for tuber and root crops.

A cross designed to separate the two

The population was 272 first-generation progeny from a cross between Zhongshuzao43, a smooth-skinned cultivar, and Innovator, a russeted one. They were grown in Hebei and Inner Mongolia over two consecutive years.

Two sites and two years is what makes the headline number trustworthy. Broad-sense heritability of skin texture came out at 93.1 per cent, meaning the environment accounted for only a small fraction of the difference between plants — the same genotype produced much the same skin in different soil, in a different climate, in a different season. Traits that behave that consistently are the ones marker-assisted breeding works on.

Bulked segregant analysis then located a major quantitative trait locus on chromosome 4, within an interval of 1.94 megabases, and two independent mapping methods agreed on it. Four insertion/deletion markers were developed from the region; their average agreement with the observed skin phenotype exceeded 80 per cent, and three of the four exceeded 90 per cent.

A marker that calls the trait correctly nine times in ten is immediately useful. A breeder can test a seedling’s DNA rather than growing it to harvest and looking at the tubers, which removes a season from every selection cycle.

What russeting actually is

The anatomy turned out to be a story about cracks that never heal.

In Zhongshuzao43 the skin layers stayed compact and orderly as the tuber grew. In Innovator, cracks appeared and widened progressively, and the tuber retained more skin layers as it expanded. Russeting, in the authors’ description, develops as cracked outer layers remain attached and accumulate during tuber growth.

That is a mechanical account with a chemical basis, and the chemistry divided cleanly. Smooth skin accumulated more suberin; russeted skin accumulated more lignin. The same pattern appeared in selected progeny, which is the check that matters — a difference between two cultivars could be incidental, while a difference that tracks the trait through a segregating population is linked to it.

The two polymers behave differently in a way that explains the phenotype. Suberin is a waxy, flexible material that seals wounds and expands with the tissue; lignin is rigid. A skin built to stretch accommodates a swelling tuber, and a skin built stiff cracks instead — then, because potato periderm keeps producing new layers beneath, the cracked material stays on and builds up.

The candidate gene, and the authors’ caution about it

RNA sequencing across three developmental stages found differentially expressed genes enriched in cell-wall formation and in lignin, fatty-acid and suberin-related pathways — consistent with the chemistry, and with the trait being decided while the skin is being built.

The strongest candidate within the mapped interval was StPXG4, which encodes a peroxygenase, and its expression was consistently higher in smooth-skinned varieties. Co-expression analysis linked it to two MYB-family transcription factors, StMYB103 and StMYB58, which the authors note still require direct functional testing.

They were explicit on this point, stressing that StPXG4 is “a high-priority candidate gene” rather than a functionally confirmed causal gene.

That distinction deserves to be preserved rather than smoothed over, because it is the difference between two kinds of claim. A 1.94-megabase interval in potato contains many genes. Picking the one whose function fits the phenotype and whose expression pattern matches is a reasonable way to generate a hypothesis, and it is not evidence that this gene is the cause. The experiment that would settle it — knocking StPXG4 out, or expressing it in a russeted cultivar, and watching the skin change — has not been done.

The MYB link is a further step removed. Co-expression means two genes rise and fall together across samples, which is consistent with one regulating the other and equally consistent with both responding to something else. Both MYB103 and MYB58 are known regulators of secondary cell-wall and lignin biosynthesis in other plants, which makes the association plausible and does not make it demonstrated.

What is usable now

The practical output does not depend on the gene being right. The four indel markers were developed from the mapped interval and validated against phenotype, and they work whether or not StPXG4 turns out to be the causal gene — a marker needs only to be linked to the locus, not to be in it.

What the markers cannot yet do is travel. They were validated in one biparental population derived from two specific cultivars, and potato is a highly heterozygous, tetraploid crop in which marker–trait associations frequently fail to transfer between genetic backgrounds. Whether these four markers predict skin texture in unrelated breeding material is the next question, and the release does not report that it has been tested.

The release also names no individual authors and contains no direct quotations from any researcher, so the interpretations reported here are the release’s own paraphrase of the paper’s argument.

The paper was published on 13 March 2026. The research was funded by the National Natural Science Foundation of China (grant 32172090) and the China Agriculture Research System (CARS-9). The authors declare no competing interests.

---

Notes

1. “Genetic architecture and major genes for tuber skin texture in potato.” Horticulture Research, 2026. DOI: 10.1093/hr/uhag102. The release does not name the paper’s authors.
2. Nanjing Agricultural University The Academy of Science. “Potato genes reveal why tubers develop smooth or russet skins.” EurekAlert!, 4 August 2026. https://www.eurekalert.org/news-releases/1138761.

Photo by Polina Tankilevitch on Pexels.com