The role of ERFVIIs as oxygen-sensing transducers in the evolution of land plant response to hypoxia
Laura Dalle CarbonareiD, Hans van VeeniD, Vinay ShuklaiD, Monica PerriiD, Liem T. BuiiD, Michael John HoldsworthiD, Francesco LicausiiD
Abstract
The transcriptional response to low oxygen (hypoxia) in the flowering plant Arabidopsis thaliana is transduced through group VII Ethylene Response Factor (ERFVII) transcription factors, whose proteolysis is oxygen-dependent via the PLANT CYSTEINE OXIDASE (PCO) N-degron pathway. When and how this response to hypoxia evolved in land plants remains unknown. Here we investigated the conservation and divergence of transcriptional responses to hypoxia in major land plant clades. We identified induction of gene functions associated with glycolysis and fermentation as part of a conserved response across all land plant divisions. Our results indicate that ERFVIIs appeared in the last common ancestor of vascular plants with true roots, concurrently with oxygen-dependent destabilisation, to regulate hypoxia-adaptive genes. Proteins from other ERF groups have been recruited multiple times in different clades as substrates of the PCO N-degron pathway. Our results demonstrate that the response of land plants to hypoxia has been refined in derived clades through the evolution of ERFVIIs as transcriptional transducers, that occurred concomitantly with the appearance of vascular systems and roots as foraging systems through hypoxic soil.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 28
- FWCI
- 11.78
- Field percentile
- top 1%
- Organism
- Assay
- Research Focus
- Approach
In-depth summary
Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (28 Sept 2024) · CC BY 4.0
Hypoxic induction of glycolysis and fermentation genes spans the land plant divisions tested, while oxygen-regulated group VII Ethylene Response Factors (ERFVIIs) carrying Cys2 occur only in vascular plants.
Flowering-plant hypoxia signalling relies on ERFVII transcription factors whose N-terminal cysteine is oxidised by plant cysteine oxidases (PCOs) when oxygen is present, targeting them for PRT6 N-degron proteolysis, but sensing outside angiosperms was barely studied and no phylogeny of ERFVIIs as oxygen-dependent substrates existed. This work compares hypoxia transcriptomes from liverwort to angiosperms and tests ERF stability experimentally to date when ERFVII-based sensing arose.
Key results
ShownHypoxia transcriptome responses are weaker in the liverwort, moss, lycophyte and ferns than in Arabidopsis and rice, yet a shared core of induced orthogroups, mostly metabolic and including glycolysis and fermentation enzymes, recurs across divisions.
P. patens had the fewest differentially expressed genes (DEGs) after 8 h at 1% versus 21% O2 in darkness, while A. thaliana and O. sativa had the most regulated orthogroups. Requiring regulation in at least three clades gave 78 up- and 54 down-regulated orthogroups; those induced in six of seven species were mostly metabolic, e.g. sucrose synthase (SUS), phosphofructokinase (PFK) and pyruvate decarboxylase (PDC). · Transcriptional responses to hypoxia in land…; The molecular function of conserved…
ShownERFVII-like proteins with similarity beyond the AP2 domain occur in every tracheophyte examined but in few bryophytes and in neither alga; lycophyte and fern ERFVIIs carry Cys2, whereas liverwort ERFVII-like proteins lack it.
Reciprocal best-hit searches with A. thaliana RAP2.12 covered 104 land plant species (12 bryophytes, 92 tracheophytes) and two algae; most bryophyte and algal hits matched only the AP2 domain, and the algal genomes had no Cys2-ERFs. The tracheophyte tree resolved lycophyte, fern, gymnosperm and angiosperm clades; after Cys2, lycophyte and fern N-termini vary from residue 5-6 onward. · ERFVII evolution in land plants; Non-ERFVII ERF proteins have been recruited as…
ShownLycophyte (Selaginella) and fern (Equisetum, Pteris) ERFVII-like proteins are degraded through the oxygen- and nitric oxide (NO)-dependent PCO N-degron pathway when expressed in Arabidopsis.
SmERF1, EhERF1 and PvERF1 were unstable in rabbit reticulocyte pulse-chase, with longer half-lives under bortezomib or after Cys2Ala mutation. In Arabidopsis erfVII seedlings they failed to accumulate, but did so in erfVII prt6, with bortezomib or the NO scavenger cPTIO, and under hypoxia (2 h, 0.1% O2) or submergence (1 h). · ERFVII evolution in land plants; Low oxygen, submergence and chemical treatments
ShownBryophyte ERFs from other groups (Group X and others) independently acquired Cys2 and are N-degron substrates, but in Arabidopsis they are not stabilised by hypoxia, and the liverwort Cys2-ERF MpERF23 is not needed for induction of the hypoxia genes tested.
In Arabidopsis seedlings, MpERF23, PpERF133 and PpERF130 accumulated with prt6, bortezomib or cPTIO but not under hypoxia or submergence, while a RAP2.3 reporter was stabilised by both. A RAP2.12 N-terminal 50 aa-Citrine fusion accumulated more in nuclei under hypoxia in M. polymorpha. Mperf23 CRISPR knockouts showed no change in hypoxia-responsive genes (MpPDC2, MpETHE1, MpL2HGDH; n = 4). · Non-ERFVII ERF proteins have been recruited as…; Figure legends
SuggestedBryophyte, lycophyte and fern Cys2-ERFs do not restore hypoxic gene induction in Arabidopsis erfVII seedlings, but with the RAP2.12 activation domain several activate a Hypoxia Responsive Promoter Element (HRPE) reporter, not a GCC-box one.
Under 1 h submergence, Cys2-ERF transgenes did not significantly change PDC1, ADH1 or PCO2 expression relative to erfVII. In erfVII prt6 protoplasts (n = 5), RAP2.12-activation-domain fusions of P. patens Cys2-ERFs and S. moellendorffii and P. vittata ERFVIIs induced HRPE- but not GCC-box-luciferase, below RAP2.12; SmERF1 activated two of three Selaginella promoters, RAP2.12 all three. · Evolution of Cys2-ERFs as oxygen-dependent…; Luciferase transactivation assay in protoplasts; Figure legends
Methods, data and reproducibility
Hypoxia (1% O2, 8 h) mRNA-seq of a liverwort, moss, lycophyte and two ferns was compared with public Arabidopsis and rice data via OrthoFinder orthogroups from 33 species. ERFVII and Cys2-ERF homologues were mined for phylogenies and selected proteins tested for N-degron stability (in vitro, Arabidopsis mutants), protoplast transactivation and, for MpERF23, CRISPR knockout in Marchantia.
- mRNA-seq of hypoxia (1% O2, 8 h, dark) vs air-treated gametophores (M. polymorpha, P. patens) and sporophyte shoots (S. moellendorffii, E. hyemale, P. vittata) · multiple species (liverwort, moss, lycophyte, two ferns) · n = five species; each replicate pooled three thalli, plantlets or shoots · PRJNA1128557, PRJNA1150224 (SRA; split across datasets not stated)
- RNA-seq (Illumina NovaSeq 6000, 150 bp paired-end) of 7-day-old Col-0 and pentuple erfVII mutant seedlings at 1% O2 or air for 2 and 10 h · Arabidopsis thaliana · PRJNA1128557, PRJNA1150224 (SRA; split across datasets not stated)
- Public hypoxia (1% O2) microarray data (Licausi et al. 2011) · Arabidopsis thaliana
- Public transcriptome data under anoxia · Oryza sativa
- Transcriptomes used for OrthoFinder orthogroup inference across nine clades (angiosperms to hornworts) · multiple species (land plants) · n = 33 species
- Predicted proteomes searched with A. thaliana RAP2.12 for ERFVII reciprocal best hits (tblastn against ONEKP, back-aligned to Phytozome); ERF families retrieved from PlantTFDB · multiple species (land plants and algae) · n = 104 land plant species (12 bryophytes, 92 tracheophytes), two algae
- Transgenic erfVII and erfVII prt6 seedlings expressing HA-tagged Cys2-ERFs or ERFVII-like proteins; mesophyll protoplasts for luciferase transactivation · Arabidopsis thaliana · n = five replicates per protoplast sample
- Mperf23 CRISPR knockout and RAP2.12 N-terminus reporter lines (13-day-old thalli, gemmae) · Marchantia polymorpha · n = n = 4 (qRT-PCR)
- Methods & tools
- OrthoFinder, Trinity, Kallisto, edgeR, HISAT2, StringTie, featureCounts, DESeq2, MUSCLE, IQ-TREE, FIMO, BLAST (tblastn) against ONEKP
- Data & accessions
- https://db.cngb.org/onekp/https://www.phytozome.net/https://planttfdb.gao-lab.org/index.php
- Reagents & environment
- Trinity (k-mer 31, min k-mer abundance 2); OrthoFinder inflation 1.3 with discontiguous Megablast; Hisat2 v2.0.5; StringTie v1.3.3b; featureCounts v1.5.0-p3; DESeq2 1.20.0; GraphPad Prism version 10
- Reproducibility gaps
- No code or scripts released for the orthogroup, differential expression or motif analyses; deposition of the de novo E. hyemale and P. vittata assemblies and replicate numbers for the five-species RNA-seq are not stated.
Limitations (6)
- Gene family diversification across lineages prevented exhaustive cross-species comparison, and orthogroups often hold several genes per species, so regulation is summarised at orthogroup level. · stated by the authors
- The protoplast transactivation assays could not test whether the ERFs bind the promoters directly. · stated by the authors
- Whether bryophyte Cys2-ERFs contribute substantially to hypoxic transcription is unclear, and what drives the bryophyte hypoxia response remains unresolved. · stated by the authors
- The proposed link between ERFVII origin and root survival in soils of changing water saturation is presented as speculation. · stated by the authors
- Angiosperm comparators are public datasets (Arabidopsis microarray at 1% O2; rice under anoxia with a different P threshold), unlike the new 8 h, 1% O2 mRNA-seq, which may affect the apparent difference in response strength. · noted in this summary
- Protein stability and complementation were tested in Arabidopsis or reticulocyte lysate, not in the native lycophyte, fern or moss species, so native N-degron substrate status is inferred. · noted in this summary
Who should read:Read this if you study flooding or hypoxia tolerance in plants, N-degron oxygen sensing, or how transcription factor families gained signalling roles during land plant evolution.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- A synthetic ERFVII-dependent circuit in yeast sheds light on the regulation of early hypoxic responses of plants
Mikel Lavilla-Puerta, Yuming He, … Beatrice Giuntoli·Proceedings of the National Academy of SciencesPeer-reviewed·Mar 10, 2026
- Severe Fe deficiency promotes hypoxia inducible responses in Arabidopsis thaliana
Yuri Telara, Moez Maghrebi, … Beatrice Giuntoli·bioRxiv·Jan 7, 2025
- MEDIATOR SUBUNIT 25 modulates ERFVII-controlled hypoxia responses in Arabidopsis
Jos H. M. Schippers, Kira von Bongartz, … Romy Schmidt·The Plant JournalPeer-reviewed·Sep 11, 2024
- H2O2 repurposes the plant oxygen-sensing machinery to control the transcriptional response to oxidative stress
Mosammat Salma Akter, Monica Perri, … Francesco Licausi·bioRxiv·Oct 21, 2024
- Ethylene-mediated nitric oxide depletion pre-adapts plants to hypoxia stress
S. Hartman, Z. Liu, … L. A. C. J. Voesenek·bioRxiv·Jul 17, 2019
- Anatomy and habitat shape the oxygen sensing machinery of angiosperms
Ximena Chirinos, Vinay Shukla, … Francesco Licausi·bioRxiv·Sep 17, 2025