Gibberellin-induced DELLA recognition by
the gibberellin receptor GID1
Kohji Murase1,2*, Yoshinori Hirano1*, Tai-ping Sun2& Toshio Hakoshima1
Gibberellins control a range of growth and developmental processes in higher plants and have been widely used in the
agricultural industry. By binding to a nuclear receptor, GIBBERELLIN INSENSITIVE DWARF1 (GID1), gibberellins regulate
gene expression by promoting degradation of the transcriptional regulator DELLA proteins, including GIBBERELLIN
INSENSITIVE (GAI). The precise manner in which GID1 discriminates and becomes activated by bioactive gibberellins for
thaliana GID1A, a bioactive gibberellin and the amino-terminal DELLA domain of GAI. In this complex, GID1A occludes
gibberellin in a deep binding pocket covered by its N-terminal helical switch region, which in turn interacts with the DELLA
domain containing DELLA, VHYNP and LExLE motifs. Our results establish a structural model of a plant hormone receptor
that is distinct from the mechanism of the hormone perception and effector recognition of the known auxin receptors.
More than 70years have passed since gibberellin (GA) was first iden-
tified as a fungus toxin from Gibberella fujikuroi (Sawada), which
causes the ‘bakanae’ (foolish seedling) disease of rice, a condition
lowers the yield of rice crops1. In the 1950s, GAs were recognized as
extremely important endogenous hormones in higher plants2. Since
then, the precise mechanism by which the GA signal is perceived and
central in plant biology.
developmental processes throughout the life cycle of a plant3. Out of
the 136 GAs identified from plants, fungi and bacteria (http://
www.plant-hormones.info/gibberellins.htm)4, only a few such as
GA1, GA3, GA4and GA7function as bioactive hormones. These
bioactive GAs are all hydroxylated at C3 and contain a lactone ring
between C4 and C10 as well as a carboxyl group at C6 (Fig. 1a and
Supplementary Fig. 1a). Addition of a hydroxyl group at the C2
position deactivates GAs. However, the lack of structural knowledge
activity relationships pertaining to these molecules.
Breakthroughs inGA-signalling research haveincludedidentifica-
tion of the soluble GID1 GA receptors5,6, the transcriptional regula-
tors DELLA proteins, which negatively regulate GA signalling7,8, and
an F-box protein9,10. DELLA proteins contain conserved DELLA and
VHYNP sequences in the N-terminal regulatory region, which are
important for GID1 binding and proteolysis of the DELLA pro-
SCARECROW) domain16in the carboxy-terminal half.
On GA binding, the GA receptor GID1 is activated, which leads to
GRAS (forGAI,RGA and
1Structural Biology Laboratory, Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0192, Japan.2Department of Biology, Duke University, Durham, North
Carolina 27708, USA.
*These authors contributed equally to this work.
Figure 1 | Structure of the GA3–GID1A–DELLA complex. a, Chemical
structures and numbering of GA3and GA4that were used in our studies.
DELLA domain (pink), the GID1A N-terminal extension (N-Ex, cyan) and
the GID1A a/b core domain (light blue). The bound GA3molecule is
represented as a space-filling model with carbon in green and oxygen in red.
c, As in b, but with a rotation of ,90u around a vertical axis. Three
N-terminal extension helices aa–ac of GID1A are loosely packed against
each other to form a flat cover on the pocket.
Vol 456|27 November 2008|doi:10.1038/nature07519
©2008 Macmillan Publishers Limited. All rights reserved
Protein expression and purification. A polymerase chain reaction (PCR)-
amplified GID1A fragment (residues 1–344) was cloned into the BamHI/SacI
ment containing GID1A with an N-terminal hexahistidine tag (His-tag) and
pET47GID1A2 was subcloned into the NdeI/AvrII site of pCDFDuet-1
(Novagen), resulting in pCDFa2x. PCR-amplified GAI fragments coding for
the DELLA domain (residues 11–113) were cloned into the BamHI/SacI site of
pET-47b to generate pET47GAI6. pCDFa2x and pET47GAI6 were co-
transformed into E. coli strain BL21Star (DE3, Invitrogen).
containing 0.1mM isopropyl-b-D-thiogalactopyranoside (IPTG) and 0.1mM
GA3(Tokyo Chemical Industry). Cells were collected, suspended in lysis buffer
(20mM Tris-HCl, pH8.0, 150mM NaCl and 1mM DTT) containing 0.1mM
GA3, and lysed by sonication. After ultracentrifugation at 100,000g for 30min,
lysis buffer containing 250mM imidazole and treated with PreScission protease
(GE Healthcare) for His-tag removal. Proteins were further purified by anion
exchange (HiTrap Q HP, GE Healthcare) and gel-filtration chromatography
(Superdex 75pg, GE Healthcare). For preparation of the selenomethionine
(SeMet)-labelled GID1A–GAI DELLA complex, pCDFa2x and pET47GAI6 plas-
mids were co-transformed into E. coli strain B834 (DE3, Novagen). Protein
expression was performed in LeMaster medium containing 0.18mgl21SeMet42.
The expression conditions and purification steps were the same as those used for
the native protein. The protein complex, comprising the GA3-bound GID1A
PAGE) and matrix-assisted laser desorption/ionization time-of-flight mass spec-
plex exists as a monomer in solution (data not shown). MALDI-TOF MS of the
native protein complex gave peaks at 39,221Da (calculated 39,228Da) and
12,273Da (12,272Da) for the GID1A and the DELLA domain, respectively, and
Crystallization and data collection. Crystallization conditions were searched
for using the sitting-drop vapour diffusion method and Hydra II-Plus-One
crystallization robot (Matrix Technology) with a commercial crystallization
solution kit. The best crystals were obtained from a 2:1 mixture solution com-
prising 50mgml21of the complex solution (in 20mM Tris-HCl, pH8.0,
150mM NaCl, 1mM DTT with 0.1mM GA3) and reservoir solution (0.1M
Tris-HCl, pH8.0, 0.8M LiCl2 and 26% polyethylene glycol 4000). The
GID1A–GAI complex crystals appeared within 3–4weeks at 20uC. The 300–
liquor with 10% polyethylene glycol 200 and then flash-cooled in liquid nitro-
gen. The SeMet-labelled complex was crystallized under the same conditions as
for the native complex. Diffraction tests of the crystals obtained were performed
generator. For structure determination, diffraction data of native and SeMet-
labelled crystals were collected at 100K on a BL41XU beamline at the SPring-8
synchrotron facility. Data were collected at the absorption peak point. The total
oscillation ranges were 360u for peak and 180u for native data sets, respectively.
The diffraction data were processed using the HKL2000 program43. The crystal
belongs to space group P43212 with unit cell parameters a582.0A˚, b582.0A˚
and c5130.1A˚and a Matthews coefficient (VM) of 2.1A˚3perDa, suggesting a
solvent content of 42% assuming that one complex is present in the asymmetric
Structure determination and refinement. Phases were calculated by a single-
the program BnP44and solvent flattening was performed with RESOLVE45. The
built model was refined through alternating cycles using the Coot46and CNS
programs47. The model has been refined to 1.8A˚resolution with Rworkand Rfree
values of 20.5% and 22.9%, respectively. Peptide chains corresponding to resi-
molecule and 217 water molecules. The first five N-terminal residues and one
C-terminal residue of GID1A, in addition to the three regions corresponding to
residues 11–24, 61–67 (a loop between helices aB and aC (loop B–C)) and 92–
113 of GAI, were omitted from the current model.
The GA4–GID1A–DELLA complex was also prepared and crystallized in the
same manner as the GA3–GID1A–DELLA complex. The structure was deter-
ture and refined at 1.8A˚resolution. The crystallographic data and refinement
statistics are summarized in Supplementary Table 1.In the Ramachandran plots
using MolProbity48, no outliers were flagged.
Pull-down assays. GAI DELLA mutants were produced by site-directed muta-
genesis49from pET47GAI6 (63His fusion) to generate pET47GM1 (L31A),
pET47GM2 (Y36A), pET47GM3 (V38A), pET47GM4 (M43A), pET47GM5
(E51R), pET47GM6 (E54R), pET47GM7 (L31A, V38A), pET47GM8 (Y36A,
M43A) and pET47GM9 (E51R, E54R). Protein expression and preparation of
glutathione sepharose 4 beads (GE Healthcare) binding with glutathione
S-transferase (GST)–GID1A and other pull-down procedures were followed as
described previously13except for using anti-63His antibodies (Santa Cruz) for
detection of the GAI protein.
Trypsin digestion assay. GST–GID1A protein was purified by affinity chro-
matography (glutathione sepharose 4) and gel-filtration chromatography
(Superdex 200pg, GE Healthcare). GST–GID1A (80mgml21) and trypsin
(2.5mgml21) were mixed in lysis buffer in the presence or absence of 100mM
was stopped by mixing with the SDS–PAGE sample buffer. The proteins were
separated by 10% SDS–PAGE gel and stained by SimplyBlue SafeStain
(Invitrogen). GST–GID1A was cleaved into GST and GID1A within a few min-
utes. MALDI-TOF MS of the resultant GID1A gave a peak at 37,204Da (calcu-
lated 37,200Da) corresponding to residues 14–345. GA-unbound GID1A is
degraded rapidly within 10min, whereas GA- and DELLA-bound GID1A were
resistant for more than 20–30min.
Circular dichroism measurements. The PCR-amplified GAI DELLA domain
(11–97) was cloned into the BamHI/SacI site of pET-47b to generate
iously. The protein sample solution contains 3mgml21GAI DELLA in 20mM
42. LeMaster, D. M. & Richards, F. M. 1H–15N heteronuclear NMR studies of
Escherichia coli thioredoxin in samples isotopically labeled by residue type.
Biochemistry 24, 7263–7268 (1985).
43. Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in
oscillation mode. Methods Enzymol. 276, 307–326 (1997).
44. Weeks, C. M. et al. Towards automated protein structure determination: BnP, the
SnB-PHASES interface. Z. Kristallogr. 217, 686–693 (2002).
45. Terwilliger, T. SOLVE and RESOLVE: Automated structure solution, density
modification and model building. J. Synchrotron Radiat. 11, 49–52 (2004).
46. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta
Crystallogr. D 60, 2126–2132 (2004).
47. Bru ¨nger, A. T. et al. Crystallography & NMR system: a new software suite for
48. Davis, I. W. et al. MolProbity: all-atom contacts and structure validation for
proteins and nucleic acids. Nucleic Acids Res. 35, W375–W383 (2007).
49. Weiner, M. P. et al. Site-directed mutagenesis of double-strand DNA by the
polymerase chain reaction. Gene 151, 119–123 (1994).
©2008 Macmillan Publishers Limited. All rights reserved