On Monday we covered the van der Ent group’s Metallomics paper showing kale crystallizing thallium chloride on its leaf surface. Four days later the same group — van der Ent, Jakovljević, Salinitro, Brueckner and Gonnelli — has published the companion question for the genuine hyperaccumulator: Biscutella laevigata, the Brassicaceae that has held the European Tl record since 1999, where does the metal actually sit inside the plant? The answer, in Annals of Botany (mcag306, published 2026-09-24), is the cell-scale detox map that phytoextraction agronomy has been guessing at for a quarter century: in the vacuoles of the leaf epidermis, concentrated at the base of the trichomes.
What the tomography showed
Plants were grown hydroponically and dosed with 5 µM Tl — a mild, non-toxic exposure by Tl-accumulator standards. Bulk concentrations came from monochromatic X-ray fluorescence (MXRF); the cellular maps from synchrotron X-ray fluorescence microtomography, the same DESY/P06-style approach the group used on kale. Two numbers anchor the result. First, the treatment produced shoot concentrations exceeding 3,000 µg g⁻¹ — comfortably above the classical 100 µg g⁻¹ hyperaccumulation threshold (the abstract states no dry-weight basis; we quote it as printed). Second, the spatial pattern: thallium sits predominantly in the vacuoles of the leaf epidermal cells and at the base of the trichomes; in roots it accumulates in the epidermis; in petioles most of it is in the vasculature. The authors’ conclusion is unambiguous: B. laevigata compartmentalizes Tl in foliar epidermal vacuoles as the main detoxification mechanism for high shoot thallium.
Why epidermal-vacuole storage matters
This is tonoplast sequestration — the same fundamental strategy Ni hyperaccumulators use — but with an agronomically convenient address. The metal is parked in the outer cell layer of the very tissue a phytoextraction harvest removes, not in root or wood that stays in the ground. The trichome-base hotspot is the non-obvious detail: glandular trichomes are a known detox sink in several metal-accumulating species, and a reproducible subcellular address like “base of the trichome” is exactly the kind of trait that could be selected or bred for — here against a species that exists as diploid hyperaccumulators at Les Malines (France) and tetraploid ones at Cave del Predil (Italy), per the paper’s framing of the trait as facultative and population-specific. Note that tension against our own species record: the database has long carried B. laevigata as an obligate hyperaccumulator on the strength of the 1999 reports; the current literature — this paper, the group’s 2024 companion work, and earlier population surveys — treats hyperaccumulation as restricted to metallicolous populations, i.e. facultative. We have corrected the record accordingly.
The risk section: three reasons not to extrapolate
One hydroponic run is not a field. This is one accession set, one 5 µM dose, hydroponics, no soil chemistry, no rhizosphere. That bound matters more than usual here because of the second point:
The same lab’s earlier root map disagrees. The group’s 2024 Annals of Botany companion study on the same species (Salinitro et al., mcae115, open access) — same beamline, same 5 µM design — found root Tl “widespread in the vascular tissues and the cortex, but not the epidermis,” the opposite of the new paper’s root result. Different accessions and growth setups can plausibly explain it, but it means the compartmentation pattern is protocol- and genotype-dependent, and any engineering target drawn from a single map should be treated as provisional.
The field runs an order of magnitude below the legend. The species’ database record — 1.94% Tl (19,400 µg g⁻¹) above-ground dry tissue, LaCoste et al. 1999, with Anderson et al. 1999 independently reporting >1.5% (15,200 µg g⁻¹) — comes from Tl-amended pot trials. Field surveys at the same French district found far less: Leblanc et al. 1999 measured a maximum of ~495 µg g⁻¹ in Biscutella leaves at Les Malines, and Escarré et al. 2011 (as reviewed by the 2024 companion paper) found ≤1,490 µg g⁻¹ DW at Saint-Laurent-le-Minier. The genuine European field record, ~32,700 µg g⁻¹ DW, comes from Raibl by Cave del Predil in Italy (Fellet et al. 2012). That spread — 495 to 32,700 µg g⁻¹ across sites, with the famous numbers at the pot-trial end — is the honest uncertainty band for any Tl agromining spreadsheet, and at roughly US$8,800 kg⁻¹ Tl (2024, per the Metallomics paper discussed Monday) the difference between the low and high end is the difference between a footnote and a business case.
The conservation edge
The facultative framing cuts toward conservation, not just agronomy. The genotypes that hyperaccumulate are concentrated in a handful of tiny metallophyte populations on calamine and mineralized ground — Les Malines, Cave del Predil/Raibl — the same habitats that mining, road-building and reclamation routinely destroy, sometimes under the legal obligation to “clean” metal-rich soils that are exactly what these populations need. If B. laevigata ever becomes a Tl crop, its breeding base is a few hectares of European mine ground. That is a small, defensible reason to list metallophyte sites as germplasm reserves, not just contamination liabilities.
What this study does not show
Bounds first: this piece is grounded in the paper’s complete structured abstract — the OUP full text is paywalled and we could not read it, so tissue details, statistics and any additional elements mapped beyond Tl are beyond our reach today. Within the abstract: no dry-weight basis is stated for the 3,000 µg g⁻¹ figure; the root-epidermis result conflicts with the group’s own 2024 data as noted above; and the Les Malines-diploid framing is the new paper’s claim, with older taxonomy (Tremetsberger et al. 2002) having grouped those populations as tetraploid — newer flow-cytometry data supports the paper, but ploidy is a breeding variable, not settled trivia. None of this dulls the core result: Europe’s classic thallium hyperaccumulator defends itself by locking the metal in leaf epidermal vacuoles, and we now have the cellular map. For thallium coverage this is the strongest compartmentation evidence the genus has ever had — read it alongside Monday’s TlCl-crystal kale piece as the two ends of the Brassicaceae Tl story: one species excretes the metal onto its surface, the other buries it in its skin.
Source: van der Ent, Jakovljević, Salinitro, Brueckner & Gonnelli (2026), Annals of Botany, mcag306, DOI 10.1093/aob/mcag306 (paywalled; full text not read). All load-bearing claims — the 5 µM dose; shoot
3,000 µg g⁻¹; vacuoles of leaf epidermal cells and trichome bases; root epidermis; petiole vasculature; facultative status; the Les Malines (diploid) and Cave del Predil (tetraploid) populations — were verified verbatim by four independent reviewers (three vendor-separated fact-checkers plus an independent research re-retrieval) pulling the abstract separately from Crossref and PubMed; no numeric or wording disagreements were found. Context figures from the open companion literature were verified against primary or near-primary sources: Salinitro et al. 2024 (PMC11560374) for the root-map contrast, the 32,700 µg g⁻¹ Raibl record (Fellet et al. 2012), and the ≤1,490 µg g⁻¹ field value (Escarré et al. 2011); LaCoste 1999’s 1.94% and Anderson 1999’s >1.5% via their own abstracts and independent scholarly quotation; Leblanc 1999’s ~495 µg g⁻¹ field maximum from full text. Reviewer caveats reflected above: the 3,000 µg g⁻¹ figure carries no stated dry-weight basis; the root-epidermis claim conflicts with Salinitro 2024; and the 1999 maxima are pot-trial values, not field measurements. See /methodology/ for how we source and check analysis pieces.