Search
Search articles by title, abstract,author or doi
Advanced search

Plant enzymes build RNA from scratch

Futang Wan , Ming Lei

Vita ››

PDF (2418KB)
Vita > Cutting Edge > DOI: 10.15302/vita.2026.10.0083
Vita Open AccessPublished:

Plant enzymes build RNA from scratch

Author information +
History +
Vita () Cite this article
PDF (2418KB)

Canonical RNA synthesis strictly requires a pre-existing nucleic-acid template or primer. In a recent study published in Vita, Li and colleagues identify a conserved family of plant nucleotidyltransferases that assemble canonical 3′–5′-linked RNA polymers directly from free NTPs, expanding the known biochemical landscape of RNA generation1.

登录浏览全文

4963

注册一个新账户 忘记密码

Every known cellular pathway for making RNA begins with something already there — a DNA template to copy, an RNA primer to extend, or an NDP substrate to reverse-phosphorylate. DNA-dependent RNA polymerases transcribe from templates; terminal nucleotidyltransferases extend pre-existing RNA 3′ ends; polynucleotide phosphorylase builds chains from NDPs. Whether a protein alone could stitch free NTPs into long, canonical RNA polymers — with 3′–5′ linkages and no nucleic-acid cofactor — has remained unknown. Li et al. now answer this question, revealing that de novo RNA synthesis may be hiding in plain sight within a familiar enzyme family1 (Fig. 1).
The authors found their candidates by structural shape rather than sequence. Using human cGAS — whose nucleotidyltransferase core defines a conserved architecture for nucleotide binding and phosphodiester-bond formation2 — as a DALI query, they searched AlphaFold-predicted plant proteomes for proteins sharing that catalytic fold. Filtering 266 initial hits for a conserved acidic catalytic triad yielded nineteen candidates, among which rice OsNTP4 emerged as the lead. Purified OsNTP4, incubated with NTPs alone, generated RNase-sensitive polymers. Two relatives — rice OsNTP10 and Arabidopsis AtURT1 — shared the activity, defining a conserved family of template- and primer-independent RNA polymerases1.
The claim of de novo synthesis demands rigorous exclusion of contaminants, and the authors delivered1. Mass spectrometry revealed no covalently bound nucleotide cofactor; nucleic-acid extraction from proteinase-K-digested samples detected no free nucleotides. Most decisively, extensively washed OsNTP4 crystals retained full RNA-synthesis activity after dissolution, while matched noncrystalline material did not — proving polymerization is intrinsic to the protein. NMR confirmed that products carry authentic 3′–5′ phosphodiester linkages, distinct from the 2′–5′ linkages of immune sensors such as oligoadenylate synthetase2. The reaction requires divalent metal ions and shows ribose specificity; under identical conditions these enzymes cannot generate DNA from dNTPs1.
These enzymes are not newcomers to nucleotidyltransferase biology. AtURT1 was previously characterized as an RNA 3′-end uridylyl transferase in microRNA turnover3, and OsNTP4 likewise retains RNA-tailing activity. De novo synthesis and RNA tailing thus represent alternative catalytic outputs of the same scaffold1. In vitro sequencing revealed heterogeneous product lengths, a pronounced uridine bias (~56% of bases), and recurrent motifs such as GGGGUU and UCCUCU. The authors caution that library-preparation biases and G-quadruplex secondary structures complicate interpretation; these features describe in vitro outputs, not programmable sequence-specific synthesis1.
The central mechanistic puzzle is initiation. How does an enzyme form the first phosphodiester bond without a primer? Does a transient dinucleotide intermediate precede elongation? How is product length controlled, and how does release couple to further synthesis? AlphaFold models of OsNTP4 position the RNA 3′-OH near the UTP α-phosphate, suggesting a geometrically plausible configuration for phosphodiester-bond formation, yet these docking predictions remain to be validated experimentally. The conserved acidic residues Asp100, Asp102, and Asp157 are essential, consistent with a metal-assisted mechanism. High-resolution structures of substrate- or intermediate-bound states will be required to resolve these questions1.
This discovery joins a growing catalogue of unconventional protein-directed nucleic-acid synthesis but occupies a distinct niche. Mammalian TENT4B was recently reported to mediate primer- and template-independent RNA polymerization, though without NMR confirmation of linkage4. DRT-family reverse transcriptases catalyze protein-primed5 or protein-templated6 DNA synthesis. Against these precedents, plant NTP proteins use NTPs to produce RNA — not DNA — without protein priming or immune-sensing activation1. The structure-guided search also recovered sixteen additional candidates with conserved catalytic architectures, suggesting de novo RNA synthesis may be only the first glimpse of a broader enzymatic landscape1.
The largest open question is physiological relevance. Endogenous triggers, substrates, and biological roles remain unknown; do these enzymes generate functional non-coding RNAs in plants? Is their activity mobilized during stress or development? Genetic analysis of loss-of-function alleles and in vivo RNA identification will be essential. Beyond plants, this activity reopens an evolutionary conversation — if modern proteins can synthesize RNA from free nucleotides, could analogous enzymes have bridged prebiotic chemistry and the RNA world7,8? By uncovering de novo RNA polymerization in conserved plant enzymes, Li et al. provide a platform for dissecting mechanism and exploring the evolutionary significance of protein-driven, template-independent RNA production1.

[1]

Li, R.N. et al. Vita https://doi.org/10.15302/vita.2026.10.0078 (2026).

[2]

Kranzusch, P.J. Curr. Opin. Struct. Biol. 59, 178–187 (2019).

[3]

Tu, B. et al. PLoS Genet. 11, e1005119 (2015).

[4]

Nagpal, N., Tai, A.K., Fong, Y.W. & Agarwal, S. Nucleic Acids Res. 54, gkag864 (2026).

[5]

Tang, S. et al. Nature 643, 1352–1362 (2025).

[6]

Deng, P.J., Lee, H., Armijo, C., Wang, H.Q. & Gao, A. Science 392, 1274–1281 (2026).

[7]

Gilbert, W. Nature 319, 618 (1986).

[8]

Joyce, G.F. Nature 418, 214–221 (2002).

RIGHTS & PERMISSIONS

The Author(s) 2026. Published by Higher Education Press. This is an Open Access article distributed under the terms of the CC BY license (https://creativecommons.org/licenses/by/4.0/).

Cite this article

Download citation ▾
Wan, F., Lei, M.  Plant enzymes build RNA from scratch  Vita https://doi.org/10.15302/vita.2026.10.0083 ()
AI Summary AI Mindmap
PDF (2418KB)
Sections
Figures
References

0

Accesses

0

Citation

Detail

Recommended

AI思维导图

/

〈 〉