Plant biology

Gibberellins close the lid

Article metrics

Gibberellins regulate many aspects of plant growth and development. Crystal structures of their receptors provide a view in unprecedented detail of how these hormones operate at the molecular level.

We owe a lot to gibberellins. The green revolution depended on the introduction of semi-dwarfing genes that impair the biosynthesis or signalling systems of these naturally occurring plant hormones, and they or their biosynthetic inhibitors continue to be widely used to regulate crop growth. A clear understanding of how they function has emerged only recently, a notable advance being the identification1 of a gibberellin receptor in 2005. On pages 459 and 520 of this issue, Murase et al.2 and Shimada et al.3 take matters further. They describe the crystal structure of receptors from two plant species, providing deeper insight into how gibberellins are perceived by plant cells.

Gibberellins (GAs) promote plant growth and developmental processes, such as seed germination and flower induction. Their action allows plants to respond to changes in their environment. At the molecular level, they stimulate the destruction of growth-repressing proteins, known as DELLA proteins4, that bind to transcription factors and so prevent them from functioning5,6. The degradation of DELLA proteins requires that they are first tagged by the addition of ubiquitin molecules in a process catalysed by a protein complex known as an SCF E3 ubiquitin ligase4. The ubiquitinated DELLA protein is then recognized and destroyed by another protein complex, the 26S proteasome (Fig. 1).

Figure 1: Gibberellin action.

a, When gibberellin (GA) concentrations are low, so that the GID1 receptor is in the unbound state, DELLA proteins inhibit gene expression by interacting with transcription factors. b, At higher GA concentrations, binding of GA to GID1 results in a change in protein conformation such that GID1 interacts with DELLA, promoting its association with an SCF E3 ubiquitin ligase. The resulting ubiquitination (Ub) and destruction of DELLA by the 26S proteasome frees the transcription factors to promote gene expression.

The discovery of GA receptors has helped clarify how these hormones initiate the process of ubiquitination and degradation: in the presence of GA, the receptor, known as GID1, binds to DELLA proteins1,7 and promotes their association with a component (the F-box) of the SCF E3 ubiquitin ligase8. Murase et al.2 now present the structure of a complex comprising a GID1 receptor from the model plant Arabidopsis thaliana with GA and part of the DELLA protein that interacts with the receptor. Shimada et al.3 describe the structure of the rice GID1–GA complex. In neither case was it possible to determine the structure of the receptor in isolation.

GID1 proteins resemble esterase enzymes such as the hormone-sensitive lipases that break down fat in animals. Although GID1 proteins do not function as esterases owing to a change in a critical amino acid, they have close structural similarity to these enzymes, being globular proteins containing a pocket for the substrate. The GA molecule contains four carbon rings that give it a rigid structure (Fig. 2). It is anchored by its carboxylic acid group to the bottom of the receptor pocket, such that its non-polar surface opposite the carboxylic acid group is held at the opening of the pocket. Uniquely, GID1 contains a loose strand at its amino-terminal end that interacts with the surface of the bound GA, so covering the pocket like a lid (Fig. 1).

Figure 2: Gibberellin structure.

Gibberellins are diterpenoid carboxylic acids, and come in many forms. This is GA4, with the carbon atoms numbered according to convention and as referred to in the text. The two new papers2,3 show that in both GA3 and GA4 the carboxylic group on C6 and the hydroxyl group on C3, which are essential for biological activity, promote binding to the receptors. A hydroxyl group on C13, which is present in GA3 but absent in GA4, neither promotes nor hinders binding, whereas a hydroxyl group on C2 seriously reduces binding affinity. Hydroxylation on C2 is an important mechanism in higher plants for deactivating GAs.

Murase et al.2 show that the DELLA protein interacts with the upper surface of the lid, and they speculate that this interaction may cause a change in the shape of the DELLA protein that allows it to associate with the ubiquitin ligase. Thus, GA functions as an allosteric activator of GID1, causing structural changes that allow the receptor to associate with DELLA proteins, but it does not interact directly with DELLAs itself. The action of GA differs from that of auxin, another plant hormone, which also functions by inducing ubiquitination and degradation of transcriptional regulators known as AUX/IAAs. Auxin, however, associates directly with the F-box of the ubiquitin ligase, acting to promote its interaction with AUX/IAA without changing the structure of either protein or requiring the involvement of a third party9.

Both papers show similar interactions of the receptors with two different GA molecules, GA4 and GA3. These molecules share features, including the carboxylic acid group on carbon atom 6 (C6) and a hydroxyl group on C3 (Fig. 2), that are essential for biological activity and, through interaction with polar amino-acid residues, enhance binding of the GA to the receptor. However, GA3 also contains a hydroxyl group on C13, which contributes little to the binding affinity. Although most plant species predominantly use 3,13-dihydroxylated GAs, the function of the 13-hydroxyl group remains unclear. Its purpose may be to increase the solubility of the molecule and so improve mobility between cells.

The structures determined for GID1–GA indicate that a hydroxyl group on C2, which abolishes growth-promoting activity, would introduce unfavourable steric interactions with the receptor and seriously reduce binding affinity. Hydroxylation on C2 is an important mechanism in higher plants for deactivating GAs, but does not apparently occur in the more primitive club moss Selaginella moellendorffii10. In an extension of their study, Shimada et al.3 replaced selected amino acids in the rice GID1 protein with the corresponding amino acids in Selaginella GID1 and found that, in some cases, the mutated protein had lower affinity for the biologically active GA4, but was more accommodating of its 2-hydroxy derivative. They propose that the receptor evolved from a hormone-sensitive lipase through loss of its catalytic activity and gradual refinement of the substrate pocket to increase affinity and specificity for GA. In higher plants, precise regulation of GA concentration is essential, and the receptor must discriminate between the active hormone and its many structurally similar biosynthetic precursors and deactivation products.

The work by Murase et al.2 and Shimada et al.3 has practical as well as intellectual implications, in that knowledge of the detailed structure of the receptor could help in designing more effective and potentially cheaper GA-like growth regulators for agriculture. On the intellectual front, the next challenge will be to determine how GID1–GA seals the fate of DELLAs by promoting association with the ubiquitin ligase.


  1. 1

    Ueguchi-Tanaka, M. et al. Nature 437, 693–698 (2005).

  2. 2

    Murase, K., Hirano, Y., Sun, T.-p. & Hakoshima, T. Nature 456, 459–463 (2008).

  3. 3

    Shimada, A. et al. Nature 456, 520–523 (2008).

  4. 4

    Ueguchi-Tanaka, M., Nakajima, M., Motoyuki, A. & Matsuoka, M. Annu. Rev. Plant Biol. 58, 183–198 (2007).

  5. 5

    Feng, S. et al. Nature 451, 475–479 (2008).

  6. 6

    de Lucas, M. et al. Nature 451, 480–484 (2008).

  7. 7

    Nakajima, M. et al. Plant J. 46, 880–889 (2006).

  8. 8

    Griffiths, J. et al. Plant Cell 18, 3399–3414 (2006).

  9. 9

    Tan, X. et al. Nature 446, 640–645 (2007).

  10. 10

    Hirano, K. et al. Plant Cell 19, 3058–3079 (2007).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hedden, P. Gibberellins close the lid. Nature 456, 455–456 (2008) doi:10.1038/456455a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.