henhaobio
Sign in

Proceedings of the National Academy of SciencesPlant BiologyPeer-reviewed journal article

INDETERMINATE DOMAIN-DELLA protein interactions orchestrate gibberellin-mediated cell elongation in wheat and barley

Patrycja Sokolowska, Matthias JostiD, Wolfram BussiD, Brett FordiD, Peter Michael ChandleriD, Wolfgang SpielmeyeriD, Andrew L. PhillipsiD, Alison K. Huttly, Danuše TarkowskáiD, Rocío Alarcón‐ReverteiD, Suzanne J. ClarkiD, Stephen P. PearceiD, Peter HeddeniD, Stephen G. ThomasiD

PDFDOIPreprintCodeData
In-depth full-text digest available · Jump to digest ↓

Abstract

DELLA proteins, members of the GRAS-domain family of transcriptional regulators, are critical for plant growth and development. They modulate transcription indirectly via interactions with hundreds of transcription factors. The phytohormone gibberellin (GA) triggers DELLA degradation, providing a mechanism by which plants can integrate developmental and environmental signals to regulate gene expression and optimize growth responses. In agriculture, DELLA mutations have been instrumental in improving crop performance. Most modern wheat (Triticum aestivum L.) varieties carry Rht-B1b or Rht-D1b alleles that encode DELLA proteins resistant to GA-mediated degradation, resulting in constitutive partial suppression of stem growth, a semi-dwarf stature and lodging resistance. However, these alleles also reduce nitrogen use efficiency and early vigour, limiting their utility in some environments. Understanding how DELLA proteins regulate growth and development is, therefore, critical for refining breeding strategies. In this study, we identified the orthologous C2H2 zinc-finger transcription factors INDETERMINATE DOMAIN 5 (IDD5) in wheat and SEMI-DWARF 3 (SDW3) in barley (Hordeum vulgare) as positive regulators of stem and leaf expansion. Both IDD5 and SDW3 physically interact with, and act downstream of, DELLA proteins as key components of GA-mediated growth responses. Altered expression levels of GA biosynthesis genes suggest that IDD5 helps maintain GA homeostasis in addition to growth regulation. Loss-of-function mutations in IDD5 and SDW3 confer a GA-insensitive semi-dwarf phenotype comparable to that of the Rht-D1b Green Revolution allele, highlighting their potential as novel dwarfing alleles for cereal improvement. Significance statementOur study identifies homologous wheat and barley transcription factors (IDD5 and SDW3) that interact with DELLA proteins to regulate plant height. Unlike conventional Green Revolution DELLA mutations, which can reduce height but also have drawbacks such as lower nitrogen-use efficiency, these IDD genes may provide more targeted approaches to manage plant growth. By showing how IDD proteins promote stem and leaf expansion and by revealing their potential as alternative dwarfing alleles, our research opens new avenues both for fundamental research into plant growth pathways and for applications in cereal breeding. Ultimately, it could help produce crops with improved lodging resistance and fewer negative side effects than current dwarfing alleles.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (18 Apr 2025) · CC BY 4.0

Wheat IDD5 and barley SDW3, orthologous zinc-finger transcription factors, bind DELLA proteins and promote gibberellin (GA)-driven elongation; their loss-of-function mutants are GA-insensitive semi-dwarfs resembling Rht-D1b.

Green Revolution Rht-B1b and Rht-D1b DELLA alleles give wheat lodging-resistant semi-dwarf stature but also lower early vigour and nitrogen use efficiency, and the transcription factors through which DELLA restricts cereal stem growth are poorly understood. Identifying the orthologous IDD5/SDW3 factors as DELLA partners required for GA-driven elongation supplies a mechanistic entry point and candidate dwarfing alleles in which DELLA stays GA-degradable.

Key results

  1. ShownBarley SDW3 is HORVU.MOREX.r3.2HG0131300, a C2H2 zinc-finger gene of the IDD (INDETERMINATE DOMAIN) family. SDW3 binds the barley DELLA SLN1 and its wheat orthologue IDD5 binds RHT-D1; truncation tests in IDD5 showed the C-terminal PAM domain is required, while the ID DNA-binding domain is dispensable.

    Only this gene carried SNPs in all eight RNA-sequenced sdw3-allelic mutants. In an F2, wild-type SDW3 seedlings responded to GA (P < 0.001) but sdw3a did not (P = 0.812). IDD5 shares >88% amino acid identity with HvSDW3; yeast two-hybrid showed PAM-deleted IDD5 lost RHT-D1 binding, and bimolecular fluorescence complementation placed the interaction in nuclei. · HvSDW3 encodes an INDETERMINATE DOMAIN (IDD)…; IDD5 interacts with DELLA proteins in wheat

  2. ShownLoss of IDD5 or SDW3 is epistatic to DELLA loss: idd5 suppresses the spindly, low-tillering wheat rht-1 null phenotype and sdw3a curbs sln1c overgrowth in barley. GA still triggers RHT-1 degradation in idd5, placing IDD5 downstream of DELLA turnover.

    Sextuple idd5/rht-1 height did not differ from idd5 (P > 0.05), but rht-1 and idd5/rht-1 stayed infertile. sdw3a/sln1c height matched sdw3a, with full grain set restored; sdw3b/sln1c was significantly taller than sdw3b. Immature-spike western blots showed GA-induced DELLA loss in 'Cadenza' and idd5, but not in Rht-B1c. · TaIDD5 and HvSDW3 are epistatic to DELLA

  3. ShownLoss-of-function sdw3 (barley) and idd5 (wheat) alleles produce semi-dwarf plants with height reductions close to Rht-D1b in the glasshouse and, for wheat, in the field. Severity varies among barley alleles; sdw3b (S40L), which the authors think retains some SDW3 activity, is milder.

    Barley glasshouse: sdw3a, sdw3e and sdw3b reduced height by 24.3%, 22.6% and 10.3% vs 'Himalaya'. Wheat glasshouse: idd5 stem 21.5% shorter than IDD5-NS (P<0.001) vs 25.9% for Rht-D1b vs 'Cadenza'; neither formed a fifth internode. Field: idd5 18.7% shorter vs 21.6% for Rht-D1b; spike length unchanged but fewer spikelets in idd5. · TaIDD5 and HvSDW3 are epistatic to DELLA; idd5 and sdw3 mutants exhibit a semi-dwarf…

  4. ShownWheat idd5 triple mutants have shorter first-leaf sheaths and epidermal cells and, like Rht-D1b, do not elongate in response to applied GA3. Barley sdw3 seedlings are similarly GA-unresponsive under normal conditions.

    First-leaf sheath 32.4% shorter in idd5 than IDD5-NS (P < 0.05) vs 36.0% for Rht-D1b vs 'Cadenza'; wild types saturated at 1 μM to 10 μM GA3, while idd5 and Rht-D1b responded at no dose. Sheath cell length differed between 'Cadenza' and idd5 (P < 0.0001). Barley sdw3a ignored 10 μM and 1 mM GA3 but partly recovered after paclobutrazol plus GA3. · HvSDW3 encodes an INDETERMINATE DOMAIN (IDD)…; Wheat idd5 mutants exhibit reduced…; GA dose-response assays

  5. SuggestedIn elongating leaf sheaths, idd5 phenocopies Rht-D1b at the hormone and transcript level: both accumulate bioactive GA1, keep GA biosynthesis genes high and lose GA feedback. This implies that IDD5 acts in the DELLA pathway and contributes to GA homeostasis.

    GA1 rose 2.7-fold (Rht-D1b) and 2.0-fold (idd5) vs wild-type segregants (P < 0.001), with lower GA44 and GA19; only idd5 had lower GA20. GA3 altered 364 differentially expressed genes (DEGs) in 'Cadenza' vs nine in idd5 and one in Rht-D1b; 33.6% of 'Cadenza'-idd5 DEGs were shared with 'Cadenza'-Rht-D1b. GA3OX2 and GA20OX1 stayed high and GA-unresponsive in both mutants. · IDD5 regulates GA homeostasis in the elongating…; GA hormone extraction and analysis; Transcriptome analysis

Methods, data and reproducibility

Barley SDW3 was cloned by RNA-seq SNP detection in allelic azide-induced mutants, and wheat IDD5 was found in a yeast two-hybrid aleurone screen with truncated RHT-D1. Ethyl methanesulfonate triple-homoeologue idd5 and rht-1 knockouts were assessed by GA dose-response, cell imaging, GA quantification, RNA-seq, western blots and glasshouse/field height trials.

  • Shoot transcriptomes (2 x 150 bp Illumina) of wild-type 'Himalaya', M287 (BC3 sdw3 line) and seven allelic induced semi-dwarf mutants, used for SNP-based gene cloning · barley (Hordeum vulgare) · n = eight sequenced mutants plus wild type
  • 'Himalaya' sodium azide mutant panel (sdw3a, sdw3b, sdw3e and other alleles) plus sln1c, gse1a (M488) and grd2b (M463) for GA assays and epistasis · barley (Hordeum vulgare)
  • 'Cadenza' EMS lines: triple idd5, triple rht-1, sextuple idd5/rht-1 and IDD5 null segregant, plus Rht-D1b near-isogenic and Rht-B1c lines · wheat (Triticum aestivum)
  • Leaf sheath RNA-seq of seven-day-old 'Cadenza', idd5 and Rht-D1b seedlings, 8 h after water or 100 μM GA3 · wheat (Triticum aestivum) · n = four biological replicates each
  • UPLC-MS/MS GA profiles of leaf sheaths from 'Cadenza', IDD5-NS, idd5 and Rht-D1b · wheat (Triticum aestivum) · n = four biological replicates per genotype
  • Rothamsted field trial (spring 2021), row-column design with Latinization; plant height, internode lengths and grain size · wheat (Triticum aestivum) · n = 108 plots of 1 m^2; 39 lines; three replicate blocks
  • SDW3 protein sequences from barley pangenome proteomes, checked for full-length proteins with intact domains · barley (Hordeum vulgare) · n = 44 accessions
  • Mature aleurone cDNA prey library screened by yeast two-hybrid with a ΔRHT-D1 bait (amino acids 218-623) · wheat (Triticum aestivum)
Methods & tools
Yeast two-hybrid (MaV203 cells, aleurone cDNA library screen), Bimolecular fluorescence complementation in Nicotiana benthamiana (Zeiss LSM 780), Western blot with anti-SLR1 DELLA antibody, UPLC-MS/MS GA quantification (Waters Xevo TQ-XS, Masslynx 4.2), Scanning electron microscopy with ImageJ 1.48v cell measurement, Trimmomatic 0.39, Kallisto (IWGSC RefSeq v1.2 gene models), DESeq v2, TopHat2/Bowtie2 (Morex v2) and MuTrigo SNP calling, Bowtie 2 amplicon read-pileup splicing analysis, GenStat 20th edition ANOVA; REML linear mixed models for the field trial, BLASTp via Ensembl Plants, MUSCLE, PhMyl (BLOSUM62, 100 bootstraps)
Software & environment
Trimmomatic 0.39; Kallisto against IWGSC RefSeq v1.2; bowtie/2.3.4 and tophat/2.1.1 against Morex v2 gene models; ImageJ 1.48v; GenStat 20th edition (2019); Masslynx 4.2.
Reproducibility gaps
The text gives no accession for the wheat or barley RNA-seq reads and no repository for the authors' own analysis scripts; DEG lists are in Additional file 2 and phenotype data are in supplementary tables.
Limitations (6)
  • The authors state that yield-plot phenotyping across multiple genotypes and environments is still needed before the agronomic value of idd5/sdw3 alleles, including any escape from Rht-D1b pleiotropy, can be judged. · stated by the authors
  • Useful wheat dwarfing likely requires loss-of-function mutations in all three IDD5 homoeologues (single or double mutants are only mildly shorter), which the authors note is harder for breeders than deploying one dominant allele such as Rht-D1b. · stated by the authors
  • It remains unresolved whether DELLA binding inhibits IDD5/SDW3 or recruits coactivators, and how TPL corepressors fit in; the authors say direct IDD5/SDW3 target loci must be identified to distinguish these models. · stated by the authors
  • idd5 restores rht-1 height but not fertility in wheat, while sdw3a restores sln1c male fertility in barley; the authors leave open whether this reflects species divergence, other DELLA partners or redundancy with IDD paralogues, which they have not yet tested with mutants. · stated by the authors
  • Some reporting is internally inconsistent: a genotype-by-GA interaction on sheath cell length is described as significant at P = 0.18, and the number of independent sdw3 alleles is given as both seven and eight within the same section. · noted in this summary
  • The wheat field evidence comes from one site and one season (Rothamsted, spring 2021), with height and grain measures but no reported yield, and idd5 set fewer spikelets, so a yield penalty cannot be ruled out. · noted in this summary

Read this if you study gibberellin-DELLA signalling, IDD zinc-finger transcription factors or cereal height genetics and want dwarfing alleles that act downstream of DELLA rather than stabilizing it.

Citations
11
FWCI
36.81
Field percentile
top 1%

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

Similarity Map
SelectedSame fieldCross-field
THISZhang '23Hernandez-G '18Blanco-Tour '20McCarthy-Su '21Islam '25McCarthy-Su '21
72% semantic matchPlant Biology
J. Zhang, C. Li, W. Zhang+5 more·bioRxiv·Jan 5, 2023