An RNA-based brake may help Arabidopsis keep putrescine in balance

A regulatory region at the beginning of an Arabidopsis messenger RNA appears to help limit expression of a gene involved in putrescine production. The findings connect experiments on RNA-controlled gene expression with altered growth and putrescine accumulation in genome-edited plants, although important steps in the proposed mechanism remain unresolved.

The study by Yuta Hiragori and colleagues, posted on bioRxiv on November 2, 2026, is a preprint and has not been peer reviewed.

Testing a feedback response

Putrescine belongs to a group of molecules called polyamines, which contribute to plant growth, development and stress responses. Too much can be harmful. In Arabidopsis thaliana, the ADC2 gene encodes arginine decarboxylase 2, an enzyme involved in putrescine biosynthesis, and is a major determinant of putrescine abundance.

The researchers focused on ADC2’s 5′ untranslated region, an RNA segment preceding its main protein-coding sequence. They placed this region ahead of a gene encoding emerald luciferase, a light-producing reporter, so that changes in reporter activity could reveal its regulatory effects.

In isolated leaf cells, increasing the amount of externally supplied putrescine reduced activity from the ADC2-linked reporter. A control reporter lacking the ADC2 region did not respond significantly to the tested treatment, arguing against a general effect on the reporter system.

The response also worked in the opposite direction. Reporter activity was higher in cells from two adc2 mutant lines. In a separate suspension-cell experiment, an inhibitor of arginine decarboxylase increased reporter activity, and adding putrescine counteracted that increase. Together, these interventions support feedback regulation linked to putrescine availability. However, intracellular putrescine was not directly quantified in these reporter experiments.

Most of the response points to translation

To distinguish changes in RNA abundance from changes in protein production, the team examined two independently generated Arabidopsis lines carrying the reporter. Ten-day-old seedlings received 2 millimolar putrescine for 24 hours.

Reporter activity fell to approximately 40% of untreated levels, while reporter RNA abundance declined more modestly, to roughly 70–80%. Dividing relative reporter activity by relative RNA abundance gave an estimated translation efficiency of about 50–60% of untreated levels. A control line carrying a different RNA leader showed no significant response.

These results support repression mainly at the translation stage, when cells use RNA to produce protein, with a smaller contribution from reduced RNA abundance. The translation-efficiency measure was nevertheless an indirect estimate based on luciferase activity and RNA measurements, not a direct observation of ribosomes translating native ADC2.

Two overlapping regulatory elements

Mutation experiments implicated two features of the ADC2 leader. One is an upstream open reading frame, a short coding segment preceding the main coding sequence, with unusual AUU and CUG initiation codons. Altering either codon eliminated statistically significant putrescine-responsive repression under the assay conditions.

The other feature is the ADC-box, a GC-rich sequence overlapping that upstream coding segment. Computational predictions suggested that it forms a folded RNA structure. Mutations designed to disrupt predicted base pairing removed the significant reporter response. Additional, compensatory mutations designed to restore pairing brought putrescine responsiveness back despite extensive sequence changes.

That restoration provides experimental support for a role of RNA structure rather than simply the original nucleotide sequence. But the structure itself was predicted, not directly measured in cells, and the overlapping elements cannot be completely separated experimentally by these mutations.

The study does not establish how putrescine triggers repression. Possibilities discussed by the authors include stabilization of the RNA fold or effects on ribosomes. Direct putrescine binding to this RNA region was not demonstrated.

A physiological link, with limits

The researchers also generated two genome-edited lines with deletions of 25 or 31 nucleotides spanning the overlapping elements. Both showed delayed germination, reduced shoot fresh weight and shorter primary roots.

In the more severely affected line, putrescine abundance was approximately 3.5 times that of wild-type seedlings. Spermidine and the combined spermine-plus-thermospermine fraction did not differ significantly. Polyamine measurements used three pooled seedling samples per genotype and were reported for only this edited line, not both.

These findings associate disruption of the regulatory region with putrescine accumulation and developmental defects. They do not establish that excess putrescine caused those defects: the deletions altered both overlapping elements, and no experiment rescuing growth by correcting putrescine levels was reported.

Overall, the evidence supports a putrescine-responsive RNA brake and a contribution to normal growth and metabolic balance in Arabidopsis. Whether the same feedback operates in other plants, and precisely how it changes ribosome behavior, remain open questions.