The study began with a deceptively simple experiment. The researchers decapitated young pea plants and applied a lanolin paste containing indole-3-acetic acid, the principal natural auxin, to just one side of the second internode. Within three days, the treated side elongated faster than the untreated side, bending the stem in a motion that mimics the tropic responses plants use to seek light or support. This setup, refined in earlier work by the same team, avoids the artefacts of dunking excised stem segments in auxin solutions and instead lets the researchers watch cell wall chemistry change in a living, intact organ. The convex, faster-growing flank and the concave, slower flank could then be dissected apart and compared directly.

To read the chemical differences between the two sides, the team used microarray polymer profiling, a high-throughput method in which cell wall materials extracted with two different solvents are printed onto nitrocellulose arrays and probed with a panel of thirteen monoclonal antibodies, each recognising a specific carbohydrate structure. Among all the epitopes surveyed, one stood out. Galactan, detected by the LM5 antibody that binds β-(1→4)-linked D-galactose chains, accumulated preferentially on the convex side of the curvature, and the asymmetry grew stronger over the 24, 48 and 72 hour sampling window, peaking at 48 hours. Extensin epitopes, by contrast, stayed relatively stable across the two flanks. Control stems treated with plain lanolin showed no such asymmetry, confirming that the change was driven by auxin.

Two independent techniques backed up the antibody data. Epitope detection chromatography, which couples anion-exchange chromatography with antibody detection, showed that the galactan molecules on the two sides were structurally similar and differed mainly in quantity rather than in the architecture of the chains. Meanwhile, monosaccharide composition analysis of alcohol-insoluble residues by high-performance chromatography confirmed that galactose levels were significantly higher on the elongating side, with a corresponding drop in glucose that the authors attribute to xyloglucan rather than cellulose. Immunolocalisation on resin sections sealed the case: the LM5 signal lit up in the cell walls of cortical cells on the convex side of auxin-stimulated internodes and was nearly absent on the concave side.

Galactan is a side chain of the pectic polysaccharide rhamnogalacturonan-I, and it is assembled in the Golgi apparatus by β-1,4-galactosyltransferases belonging to the GT92 family, enzymes first characterised in Arabidopsis. Searching pea genome databases, the team identified two candidate orthologues, which they named PsGALS3 and PsGALS2, sharing roughly 61 percent amino acid identity with their closest Arabidopsis counterparts and about 71.6 percent with each other. Phylogenetic analysis showed that the pea proteins form a distinct, legume-specific clade, evidence of lineage-specific duplication and diversification, while sequence alignment confirmed that the conserved transmembrane helix and the GT-A fold catalytic domain are intact in both. The N-terminal carbohydrate-binding module, which in poplar interacts with the RG-I backbone, showed more species-specific variation.

Expression analysis added a crucial clue. Both genes were active across the aerial organs examined, with transcripts generally more abundant in young, actively growing tissues, but only PsGALS3 responded strongly to auxin. When IAA-containing paste was applied to internodes, PsGALS3 expression rose significantly, whereas PsGALS2 barely moved. That auxin inducibility made PsGALS3 the obvious target for a functional test, and the researchers turned to virus-induced gene silencing, using the pea early browning virus vector system, a well-established tool in a crop species where stable genetics remain laborious. A 463-base-pair fragment of PsGALS3 was inserted into the viral vector, and two-week-old plants were infiltrated with Agrobacterium carrying the construct.

The results were striking. Three weeks after inoculation, plants in which PsGALS3 had been knocked down to roughly 15 percent of normal expression were severely stunted compared with controls carrying an empty virus. Internode length fell by up to 50 percent, leaves were smaller, tendrils shorter, and the root system reduced. Microarray polymer profiling of the silenced stems confirmed the molecular consequence: LM5 galactan epitopes dropped to about 30 percent of wild-type levels, while extractable cellulose, homogalacturonan and arabinan were unchanged. But there was a surprise. Epitopes recognised by the LM1 and LM3 antibodies, which mark extensin, were significantly upregulated, and in situ immunolocalisation showed the LM3 signal intensifying at the triangular cell junctions of parenchyma tissue, precisely where the galactan signal had faded.

The contrast with Arabidopsis is what makes the finding so compelling. In the standard reference plant, single and multiple mutants in the GT92 galactan synthases show only mild developmental defects, limited to altered salt-stress responses and reduced freezing tolerance during cold acclimation. In pea, disrupting the same biochemical pathway cripples the plant. The authors suggest this may reflect genuine architectural differences between the cell walls of the fabid clade, which includes pea and other legumes, and the malvid clade that includes Arabidopsis, differences previously hinted at by work showing that polysaccharides partition differently during extraction in the two lineages. If so, Arabidopsis is not a universal model for primary wall function, and findings from one clade cannot simply be transplanted to another.

Intriguingly, the silenced pea stems still bent in response to auxin paste, despite their shortened internodes and depleted galactan. The researchers caution that virus-induced gene silencing is transient, leaves residual expression, and can co-silence other genes, so compensatory mechanisms may be at work. It is also possible that galactan accumulation accompanies, rather than determines, auxin-driven elongation. The rise in extensin is equally enigmatic. Extensins are hydroxyproline-rich glycoproteins known to strengthen walls through tyrosine cross-linking and to respond to mechanical stress and pathogen attack, and non-ionic interactions between extensin and pectin have been documented. The team cannot rule out that truncated galactan chains simply expose previously masked extensin epitopes to the antibodies, and no galactan-specific receptor-like kinase has yet been identified that could mediate a compensatory signalling pathway.

What the study delivers, nonetheless, is the strongest causal evidence yet that pectic galactan is required for normal organ expansion in a crop legume, and a clear invitation to look beyond Arabidopsis when asking how walls are built. Early biochemical work in pea had shown galactan being synthesised as part of a pectin-xyloglucan complex in the Golgi before deposition into the wall, and later studies linked galactan remodelling to xyloglucan structure during elongation. The new data fit neatly into that picture, suggesting coordinated remodelling of pectic and hemicellulosic components during auxin-induced growth. With a reference pea genome now available and VIGS proven effective for cell wall genes, the authors argue that pea is poised for a renaissance in wall biology, one in which the interactions among galactan, xyloglucan and extensin in the native wall, in muro rather than in extraction tubes, will finally be disentangled.

Subject of Research: The role of pectic galactan synthesis in cell wall assembly and organ expansion in pea

Article Title: Loss of galactan synthesis in pea (Pisum sativum) causes defects in organ expansion and is associated with increased extensin content

Article References: Guo, X., Kumar, A., Johansen, I. E., Harholt, J., Willats, W. G. T., Ulvskov, P., & Mravec, J. (2026). Loss of galactan synthesis in pea (Pisum sativum) causes defects in organ expansion and is associated with increased extensin content. Plant Molecular Biology, 116(5), Article 94. https://doi.org/10.1007/s11103-026-01759-x

Image Credits: AI Generated

DOI: 10.1007/s11103-026-01759-x

Keywords: pea, Pisum sativum, galactan, cell wall, auxin, extensin, PsGALS3, virus-induced gene silencing, rhamnogalacturonan-I, cell elongation, microarray polymer profiling, plant molecular biology