The team began with classical plant anatomy, sectioning rhizomes at different ages to see exactly when and where the ring pattern emerges. Their analysis revealed that the rings are not laid down gradually throughout the plant’s life. Instead, they arise during a sharply defined critical period, from 126 to 139 days after germination, which falls in early autumn. During this window, the rhizome undergoes uneven secondary growth, the same kind of lateral thickening that widens tree trunks. Crucially, the growth is not uniform around the organ’s circumference. Localized patches of cork cambium, the meristematic tissue that produces protective periderm, become active in some regions while remaining dormant in others. The result is a segmented architecture in which swollen, lignified bands alternate with constricted zones.

The anatomical work also uncovered a second, more dramatic process: programmed cell death in the cortex, the outer tissue layer of the rhizome. As the secondary xylem, the water-conducting wood at the core, expands outward, it exerts mechanical pressure on the surrounding cortex. The cortical cells, already primed for destruction by the plant’s own genetic instructions, are compressed until they collapse, leaving hollow cavities inside the constriction zones. This combination of localized cork cambium activity, xylem-driven compression and cortical cell death produces the multilayered rings and internal voids that give the rhizome its silkworm appearance. In effect, the plant sculpts itself from the inside out, sacrificing its own cortical tissue to create a segmented, hollow-strengthened organ.

To understand what drives this sculpting at the molecular level, the researchers turned to metabolomics, profiling the full complement of small molecules in the developing rhizomes. The analysis revealed a pronounced shift in carbon allocation toward the phenylpropanoid pathway, the metabolic assembly line that plants use to build lignin, the tough polymer that reinforces cell walls. Lignin precursors such as coniferyl alcohol and its storage form coniferin accumulated substantially during the critical period, providing abundant substrate for secondary wall deposition and the lignification of the forming rings. At the same time, the team measured a decrease in core energy metabolites such as succinate, a key intermediate of the citric acid cycle. This pattern points to a metabolic trade-off: the plant is diverting carbon and energy away from general metabolism and toward structural reinforcement, much as a builder diverts materials from interior finishing to fortify a load-bearing wall.

The transcriptomic side of the study identified who is giving the orders. By sequencing the messenger RNA present in rhizomes across the developmental series, the researchers constructed a regulatory network and found a transcriptional hub composed of three families of transcription factors: NAC, ERF and bHLH. This NAC-ERF-bHLH module, active specifically in the root and rhizome tissue, appears to orchestrate the entire ring-forming process. It activates genes involved in phenylpropanoid and flavonoid biosynthesis as well as genes governing starch metabolism, thereby steering the plant’s carbon flux toward structural construction. In other words, a small set of master regulators coordinates the reallocation of sugars, the ramp-up of lignin synthesis and the mobilization of stored starch, all in service of building the segmented architecture.

The multi-omics integration also linked specific enzyme-coding genes to the visible anatomy. Genes encoding phenylpropanoid enzymes such as phenylalanine ammonia-lyase (PAL) and cinnamate 4-hydroxylase (C4H), the entry points of the lignin pathway, showed expression patterns strongly correlated with the timing of ring formation and the accumulation of lignin precursors. Genes encoding ABC transporters, membrane proteins that shuttle molecules across cell membranes, were likewise correlated with the anatomical and metabolic changes, suggesting a role in moving lignin precursors and other metabolites to the sites of wall deposition. This tight correlation between gene expression, metabolite abundance and tissue structure is what gives the study its persuasive power: three independent layers of evidence all point to the same underlying mechanism.

Why would a plant go to such elaborate lengths? The authors propose that the silkworm-like ring is the product of a carbon flow allocation strategy triggered by early autumn environmental signals. As days shorten and temperatures cool, these signals activate the plant’s phospholipid signaling system, which in turn switches on the NAC-ERF-bHLH transcriptional network. The redirected phenylpropanoid carbon flux leads to lignin precursor accumulation, which regulates both cork cambium activation and secondary xylem proliferation. The lignified rings that result improve the mechanical strength of the rhizome, allowing it to resist compression and maintain its shape in the soil. Meanwhile, the hollow cortex created by cell death may serve as an insulating layer that buffers the organ against heat stress, an adaptation that could matter in the exposed mountain habitats where N. incisum grows.

The compartmentalized structure may also serve the plant’s chemistry. N. incisum accumulates medicinal compounds in its rhizome, and the segmented architecture, with its alternating solid and hollow zones, could allow the plant to store these components in discrete compartments, separating them from the transport streams that move water and nutrients through the xylem. The study’s authors suggest that the combination of lignified rings and a buffered, hollow cortex together maintain organ stability while enabling this compartmentalized storage. If correct, the ring pattern is not merely an aesthetic curiosity but a functional adaptation that couples mechanical engineering with pharmaceutical chemistry in a single organ.

Beyond its evolutionary interest, the work has immediate practical value. Because the rings form during a narrow window in early autumn, growers now have a specific developmental target to aim for when managing cultivation. Understanding the environmental cues and signaling pathways that trigger ring formation could allow farmers to optimize planting schedules, soil conditions and harvest timing to produce rhizomes with the prized silkworm-like morphology. The identified NAC-ERF-bHLH hub and the key biosynthetic genes such as PAL and C4H also provide molecular markers that breeders could use to select lines with strong ring-forming capacity, accelerating the improvement of this rare and increasingly pressured medicinal species. All plant material in the study was cultivated at the team’s controlled experimental base in Tanchang County, Gansu Province, rather than collected from the wild, an approach consistent with conservation concerns for a species whose wild populations face harvesting pressure.

Scientifically, the study is a textbook example of what multi-omics integration can achieve in plant developmental biology. Neither the anatomy alone, nor the metabolite profiles, nor the gene expression data could have revealed the mechanism on their own; it is the convergence of all three that turns a morphological description into a causal model. The finding that a single transcriptional hub can coordinate carbon reallocation, secondary wall deposition and programmed cell death adds to a growing body of evidence that plant architecture is often governed by compact regulatory modules that respond to seasonal signals. For N. incisum, the message of the study is clear: the silkworm-like rings that have made this plant famous in apothecaries for centuries are the visible outcome of an autumn-triggered genetic program that trades energy for armor, hollowing the cortex and hardening the rings in a single coordinated sweep. The research, published open access in BMC Plant Biology, gives both scientists and cultivators a molecular handle on a trait that has been valued but never understood.

Subject of Research: Morphogenesis and molecular regulation of silkworm-like ring formation in Notopterygium incisum rhizomes

Article Title: Research on the morphogenesis of Notopterygium incisum rhizomes and its mechanism: multiomics integration analysis reveals the formation mechanism of silkworm-like rings

Article References: Zhao, W., Gao, J., Chen, H., Zhang, J., Lu, W., & Jin, L. (2026). Research on the morphogenesis of Notopterygium incisum rhizomes and its mechanism: multiomics integration analysis reveals the formation mechanism of silkworm-like rings. BMC Plant Biology. https://doi.org/10.1186/s12870-026-09983-9

Image Credits: AI Generated

DOI: 10.1186/s12870-026-09983-9

Keywords: Notopterygium incisum, rhizome, silkworm-like rings, plant morphogenesis, secondary growth, cork cambium, programmed cell death, lignin, phenylpropanoid pathway, multi-omics, transcription factors, medicinal plant

Trichomes are epidermal appendages found across most flowering plants, as well as some gymnosperms and bryophytes, and they take an astonishing variety of forms, from cucumber fruit spines to cotton fibers. Botanists divide them into two broad classes. Glandular secretory trichomes manufacture and store volatile compounds, resins, and other metabolites, while non-glandular trichomes lack secretory activity but provide crucial physical defense, reducing water loss through transpiration, capturing moisture from fog, buffering against temperature extremes, and shielding leaf tissue from ultraviolet radiation and herbivores. Most of what science knows about non-glandular trichome development comes from unicellular systems such as Arabidopsis rosette leaves and cotton fibers, where each hair arises from a single epidermal cell. Multicellular non-glandular trichomes, by contrast, have lacked a workable model system, leaving a significant gap in plant developmental biology.

Artemisia argyi, a perennial herb in the Asteraceae family, turns out to be an unusually elegant model for closing that gap. Scanning electron microscopy performed by the team revealed a striking asymmetry: the upper leaf surface is dominated by slipper-shaped glandular trichomes measuring roughly 33.3 by 52.1 micrometers, while the lower surface is carpeted with elongated T-shaped trichomes whose branched arms exceed one millimeter in length and intertwine into a dense mat. The stalks of these T-shaped hairs are short, about 28.8 micrometers, but their asymmetrical arms give the structure its name and its function. This clean separation of trichome types between the two leaf surfaces gives researchers a natural experiment in epidermal patterning, and it explains why the species has become a defining subject for studying multicellular hair development.

The commercial stakes are considerable. Mugwort’s dried leaves are the primary medicinal material documented in the Chinese Pharmacopoeia for warming the meridians, stopping bleeding, dispelling cold, and relieving pain, and the sector exceeded 30 billion yuan in industrial output in 2023. Moxa floss, the fibrous product central to moxibustion, consists primarily of these leaf trichomes, and its efficacy is attributed not only to combustion heat but to bioactive volatile compounds released from trichome tissue. Because trichome abundance and morphology directly determine floss yield and quality, understanding the genes that control hair formation has immediate implications for breeding improved cultivars of a plant whose germplasm is distributed across China, Mongolia, and North Korea.

To find those genes, the researchers manually separated T-shaped trichomes from the remaining leaf tissue and sequenced the transcriptomes of both, generating 39.03 gigabases of clean reads mapped to a reference genome at rates between 74.30 and 97.15 percent. The comparison was dramatic: 8,901 genes were upregulated and 4,698 downregulated in the trichome fraction. Gene Ontology enrichment pointed toward membrane components, the plasma membrane, and defense responses, while KEGG analysis flagged pathways including MAPK signaling and isoflavonoid and flavonoid biosynthesis. Several transcription factor families long associated with trichome development, including R2R3-MYB, C2H2 zinc-finger, HD-ZIP IV, bHLH, ethylene-responsive, and WD40 repeat proteins, were markedly enriched in the hairs, confirming that the sequencing had captured a genuine developmental signature rather than random tissue noise.

The team then widened the lens, integrating public RNA-seq datasets from old leaves, young leaves, stems, and glandular trichomes with their own data in a Weighted Correlation Network Analysis covering 53,354 genes. Five co-expression modules emerged, and one, the cyan module, stood out for its enrichment in cutin biosynthetic genes and homologs of known Arabidopsis trichome regulators. Within that module’s network, members of a family the authors named AarMIXTAs behaved as hub genes, strongly connected to MYB, bHLH, WD40, HD-ZIP, and WRKY factors. Homology-based screening against well-characterized regulators from Arabidopsis and the related species Artemisia annua identified 119 mugwort genes tied to trichome development across 27 allelic groups, with evidence of tandem duplication for some genes such as ETC1, TT8, and EGL3, and allelic amplification for others including TTG1, RGA, and the MIXTAs themselves.

The evolutionary story proved to be the most striking finding. Drawing on 144 MIXTA and MIXTA-like protein sequences from NCBI and Phytozome databases, spanning 136 angiosperms, two gymnosperms, and six ANA-grade outgroups, the researchers built a maximum likelihood phylogeny that resolved into four major clades. The gymnosperm sequences occupied the most ancestral position, monocots clustered separately, and two dicot-only clades containing 33 and 77 proteins appeared to have descended from the earlier lineages. Within this framework, mugwort stood out: its genome harbors eight AarMIXTA genes, a substantial expansion compared with A. annua. These fell into two clades, one most closely related to lettuce and the other to A. annua, suggesting the duplication events occurred after the Asteraceae lineage diverged from other eudicots. Elevated synonymous substitution rates between the clades hinted at accelerated evolution, possibly driven by local recombination, relaxed selection, or positive selection following segmental or tandem duplication.

Crucially, the duplicated genes did not simply do the same job twice. Although the AarMIXTAs share conserved exon-intron structures and high sequence similarity with their A. annua homologs, their expression patterns diverged sharply. The clade comprising AarMIXTA1.1 through 1.4 was expressed at significantly higher levels in T-shaped trichomes, while AarMIXTA1.5 through 1.8 were more abundant in glandular trichomes. Since AanMIXTA1 in A. annua is a validated positive regulator of glandular trichome density and artemisinin biosynthesis, its mugwort relatives in the second clade likely inherited that role. The first clade, occupying a distinct phylogenetic position, appears to have been recruited for a different task: building the non-glandular, T-shaped hairs that define the species. This is a textbook case of neofunctionalization after gene duplication, likely facilitated by the polyploid history of A. argyi, which is an autotetraploid.

Functional validation sealed the argument. The team selected AarMIXTA1.2, whose expression rises progressively during leaf development and peaks specifically in T-shaped trichomes, and overexpressed it in Arabidopsis under the 35S promoter. The transgenic plants showed two clear phenotypic changes: increased total leaf number and a roughly twofold increase in trichome density, along with moderately longer trichome branches and stalks. Quantitative PCR revealed that core Arabidopsis trichome regulators, including AtGL2, AtTAR2, AtMYB23, AtTTG1, and AtHD1, were all significantly upregulated in the transgenic lines, and the native AtMIXTA gene was elevated as well, hinting at feedback or synergistic interactions. The results demonstrate that a single mugwort transcription factor can activate a conserved downstream cascade and reshape epidermal patterning even in a distantly related species.

The implications extend well beyond mugwort. For developmental biologists, the study establishes A. argyi as a model for multicellular non-glandular trichome development, a process long overshadowed by unicellular systems, and shows that the ancient MIXTA toolkit, first cloned from snapdragon in 1994 as a controller of petal cell shape, has been repeatedly redeployed across plant evolution to sculpt epidermal surfaces. For the mugwort industry, the AarMIXTA family offers molecular markers for breeding varieties with denser, higher-quality trichomes, potentially through targeted genetic screening and trait selection. The authors point to next steps including yeast two-hybrid screens for AarMIXTA1.2 interactors, epigenetic and chromatin accessibility studies, and functional testing in mugwort itself. They also raise an intriguing ecological question: whether the plant’s remarkable environmental adaptability across diverse Chinese habitats owes something to its trichome armor, connecting a millennia-old therapeutic tradition to the frontiers of modern plant genetics.

Subject of Research: Genetic regulation of T-shaped non-glandular trichome development in Artemisia argyi by MIXTA-like transcription factors

Article Title: Functional characterization of AarMIXTAs as essential regulators in T-shaped non-glandular trichome development of Artemisia argyi

Article References: Functional characterization of AarMIXTAs as essential regulators in T-shaped non-glandular trichome development of Artemisia argyi. (n.d.). https://doi.org/10.1007/s44307-025-00077-5

Image Credits: AI Generated

DOI: 10.1007/s44307-025-00077-5

Keywords: Artemisia argyi, trichome development, AarMIXTA, transcription factors, moxibustion, moxa floss, gene duplication, polyploidy, Arabidopsis, comparative transcriptomics, WGCNA, plant morphogenesis