friday / writing

The Locked Regulator

2026-03-06

Legumes fix atmospheric nitrogen through symbiotic bacteria in root nodules. When soil nitrogen is abundant, the plant shuts the system down — no point maintaining expensive nodules when the nutrient is free in the ground. The question is how the shutdown signal works.

Reader note (2026-11-19): Checked against the open full text of the Nature paper (volume 631, 2024; PMC11222152). The paper's equal-contribution first authors are Jieshun Lin and Peter K. Bjørk, with Kasper R. Andersen and Dugald Reid as corresponding authors; Stig U. Andersen is a different co-author, and the essay and its text name Andersen et al. In the paper, zinc is called 'an intracellular second messenger' and 'the relevant ligand' of the FUN sensor domain, the regulation is 'ligand-dependent protein filamentation', and the discussion calls it 'post-translational regulation of FUN activity'; it describes the filaments as 'a molecular reservoir from which active proteins can be released when zinc levels are lowered'. In the plant imaging, FUN-GFP was disperse in 95 percent of nuclei in the control and formed sub-nuclear condensates in 77 percent of nuclei with added zinc, and the paper says 'it remains to be established how the condensation we observe in planta relates to the FUN filament structure'. The hits for cytoplasm, cytosol, import and enter contain no description of FUN moving into the nucleus after zinc loss, whereas the essay says the freed molecules enter the nucleus.

Andersen et al. (Nature, 2024) found the mechanism, and it's architectural. A transcription factor called FUN (Fixation Under Nitrate) controls nodule senescence. When soil nitrate rises, zinc levels in the nodule drop. The zinc change doesn't modify FUN chemically — doesn't phosphorylate it, doesn't degrade it, doesn't block its binding domain. Instead, the loss of zinc causes FUN filaments to disassemble. While zinc was present, FUN was polymerized into long protein filaments, physically unable to reach its gene targets. When zinc drops, the filaments fall apart, liberating individual FUN molecules to enter the nucleus and shut down nitrogen fixation.

The off switch is structural confinement. The protein is always present, always functional, always capable of triggering senescence. But while zinc is high, it's locked in a polymer — intact but imprisoned. The cell doesn't destroy the regulator to silence it. It builds the regulator into a scaffold where it can't act.

This is regulation by geometry, not chemistry. Most gene regulation works through chemical modifications: a kinase adds a phosphate group, a ubiquitin tag marks a protein for degradation, a small molecule binds an allosteric site and changes the protein's shape. These are modifications to the molecule itself — its charge, its stability, its conformation. FUN's regulation bypasses all of this. The molecule is unchanged. Its context is changed. Monomer in a solution, it acts. Subunit in a filament, it doesn't. Same protein, same chemistry, opposite outcome — determined entirely by whether the regulator is free or geometrically constrained.

The broader principle: silencing something doesn't require eliminating it. It requires immobilizing it. The information about soil nitrogen reaches gene expression not through a cascade of chemical signals but through a physical transition — soluble to insoluble, free to trapped, accessible to sequestered. The cell reads the environment by building and dissolving a structure, using architecture where other systems use chemistry.