Mol Breeding (2009) 24:375–385

DOI 10.1007/s11032-009-9298-3

Characterization of the first dominant dwarf maize mutant

carrying a single amino acid insertion in the VHYNP
domain of the dwarf8 gene
Elena Cassani Æ Edoardo Bertolini Æ Francesco Cerino Badone Æ
Michela Landoni Æ Dario Gavina Æ Alberto Sirizzotti Æ
Roberto Pilu

Received: 2 March 2009 / Accepted: 25 May 2009 / Published online: 10 June 2009
Ó Springer Science+Business Media B.V. 2009

Abstract The ‘‘green revolution’’ involving mainly
wheat and rice was based on the use by breeders of
semidominant mutations involved in the signal trans-
duction pathway of Gibberellin (GA). In particular,
mutations in the Reduced height (Rht) gene of wheat
have been used to reduce plant height and conse-
quently to avoid storm damage and lodging. These
genes have been cloned and they encode for DELLA
proteins which contain an N-terminal DELLA and a
VHYNP domain essential for GA-dependent degra-
dation of these proteins. In maize several mutations
have been isolated which affect gibberellin biosyn-
thesis and perception and in particular, mutations in
Dwarf8 (D8) gene cause a severe dwarfing phenotype.
D8 gene has been identified as an orthologue of Rht
(Reduced height), Slr1(Slender rice 1) and Gibberellic
Acid Insensitive (GAI) genes, this latter is a negative
regulator of GA response in Arabidopsis. In this work,
for the first time, we isolated and characterized a
single amino acid insertion in the VHYNP domain of
D8 maize gene causing the appearance of a dominant
E. Cassani
E. Bertolini
F. Cerino Badone

D. Gavina
A. Sirizzotti
R. Pilu (&)
Dipartimento di Produzione Vegetale, Università degli
Studi di Milano, Via Celoria 2, 20133 Milan, Italy
e-mail: salvatore.pilu@unimi.it
M. Landoni
Dipartimento di Scienze Biomolecolari e Biotecnologie,
Università degli Studi di Milano, Via Celoria 26, 20133
Milan, Italy
dwarf mutation. This spontaneous mutation, named
D8-1023, showed a phenotype which is less severe in
comparison with the other D8 mutants previously
isolated which have modifications in the DELLA
domain. This mutant appears to be an useful tool
either to study the mechanism of GA-modulated
growth in plants or to lower the height of maize
tropical germplasm for breeding purposes.
Keywords Maize
Mutant
Dwarf8 gene

Gibberellins
VHYNP domain

PCR based molecular marker
Introduction
The huge increase in cereal yields during the years of
the ‘‘green revolution’’ was enabled by the applica-
tion of large amounts of pesticides and fertilizer in
combination with the introduction of semidominant
dwarfing mutations into plants causing height reduc-
tions associated with yield increases in several
different crop species (Hedden 2003). The varieties
used today, in particular wheat, rice and sorghum, are
shorter, more resistant to storm damage and have
increased grain yields compared to earlier ones.
In maize several dwarf mutations have been
isolated since the first work of maize genetics (Emm-
erson 1912), but none of them have been used to
improve yield as yet because of the excessively severe

123
376
phenotypes of these mutants. Dwarf maize mutants are
usually classified as GA-sensitive, or GA-insensitive
on the basis of their capacity to be restored to wild type
by exogenous application of the GA hormone (Phin-
ney 1956). Members of the class of GA-sensitive
mutants are: dwarf l (d1), d2, d3, d5 and anther ear1
(an1), that identify defective enzymes of the GA
biosynthetic pathway (Fujioka et al. 1988a; Hedden
and Kamiya 1997; Galbiati et al. 2002). Unlike the
class of GA-sensitive mutations that are all recessive,
the class of GA-insensitive mutants is represented by
only two mutants Dwarf8 (D8) and D9 (a D8 duplicate
gene) that are dominant dwarf mutations (Phinney
1956; Harberd and Freeling 1989; Winkler and
Freeling 1994). In these dominant dwarfing mutants,
anther development in the ear, increased tillering, dark
green shorter leaves and delayed flowering are features
of all D9 and D8 alleles.
In recent years, the genes Reduced height (Rht) of
wheat and Slender rice 1 (Slr1) of rice, used by
breeders to reduce plant height, D8 of maize and
Slender1 (Sln1) of barley have been isolated and they
encode DELLA repressor proteins, characterized by
the N-terminal DELLA domain (Peng et al. 1999;
Chandler et al. 2002; Monna et al. 2002; Spielmeyer
et al. 2002). All these genes are homologous ortho-
logues of the GA INSENSITIVE (GAI) and REPRES-
SOR OF ga1-3 (RGA) genes of Arabidopsis (Peng
and Harberd 1997; Richards et al. 2001; Willige et al.
2007). In Arabidopsis it has been shown that
gibberellins are bound to a nuclear receptor, GIB-
BERELLIN INSENSITIVE DWARF1 (GID1), that
interacting with GAI protein promotes the degrada-
tion of this protein complex (Willige et al. 2007;
Murase et al. 2008).
Sequence analysis of D8 gene has shown that D8
maize mutants are the result of modification of the
DELLA domain caused by dominant gain of function
mutations (Peng et al. 1999). As result of this
molecular lesion all D8 mutants accumulate rela-
tively high levels of GA in all the tissues and the
levels of GA seem to be related to D8 mutation
dosage (Fujioka et al. 1988b). DELLA proteins are
negative regulators of genes promoting plant growth,
the interaction with GA causes the inactivation of
DELLA proteins by ubiquitin proteasome–dependent
protein degradation (Fu et al. 2002; Sasaki et al.
2003). The data supporting the role of DELLA
proteins as repressors of genes involved in plant
123
Mol Breeding (2009) 24:375–385
growth are also strengthened by the fact that the
Arabidopsis GAI loss-of-function alleles, isolated by
mutagenesis, confer a tall phenotype (Peng and
Harberd 1993).
Thus it seems that GA hormone is central for the
control of plant development even though elongation
of plant organs is a complex phenomenon mediated
by many plant hormones, including, beside Gibber-
ellins, auxins and brassinosteroids (Vogler and Kuh-
lemeier 2003). In fact the dwarf 3 (dw3) gene used by
breeders to reduce the height of commercial sorghum
was isolated and it encodes for a transporter involved
in polar movement of auxins (Multani et al. 2003).
Another important aspect of D8 gene concerns the
effect on flowering time, a key trait in maize climatic
adaptation. Several papers suggest that D8 might be a
major gene for this feature (Thornsberry et al. 2001;
Andersen et al. 2005; Camus-Kulandaivelu et al.
2006). In particular, using an association approach,
several polymorphisms in D8 gene sequences coming
from 92 inbred lines have been found associated to
flowering time (Thornsberry et al. 2001) and some of
these polymorphisms have been tested as markers in
an independent set of elite European inbred lines
(Andersen et al. 2005).
These papers also suggested that most likely
human selection occurred for this gene as reported
by Camus-Kulandaivelu et al. who presented an
extensive molecular study on hundreds of inbred lines
and landraces from America and Europe, analyzed
for flowering time in short and long day conditions
(Camus-Kulandaivelu et al. 2006). In this paper the
association between a 6-bp insertion/deletion poly-
morphism in the Dwarf8 (D8idp) and flowering time
and climatic adaptation in maize has been confirmed.
Even more recently, a region upstream of the D8
gene, involved in the genetic regulation of flowering
time variation in maize has been found (Camus-
Kulandaivelu et al. 2008). Thus this gene appears to
be involved in the perception of GA hormone and so
to play an important role in several crucial events in
the life of the plant.
Since much of the current progress with regard to
GA metabolism comes from the isolation and char-
acterization of single-gene mutants, in this work we
have isolated and characterized a new maize domi-
nant dwarf mutant that is allelic to D8 gene. This
mutation is a mild D8 allele mutation not previously
analyzed. The novel mutant allele was cloned and the
Mol Breeding (2009) 24:375–385
alignment with d8(?) wild type alleles present in the
database has shown a molecular lesion: an insertion
of 3 bp within the VHYNP domain, localized in the
50 of the gene, near the DELLA domain, responsible
for the GA response. This finding represents the first
evidence of a dominant dwarfing mutation which
does not involve the DELLA domain but is in the not
yet well-characterized VHYNP domain involved in
protein degradation. We found a new and interesting
phenotype and we suggest a possible future modifi-
cation of the VHYNP domain of D8 gene to modulate
plant growth and to shorten excessively tall germ-
plasm with the aim of improving crop production.
Materials and methods
Plant materials
The dwarf mutant here analyzed (previously named
D*-1023 and after genetic and molecular analysis
renamed D8-1023) of spontaneous origin, was orig-
inally isolated from a commercial hybrid (Aliseo,
Limagrain) in Landriano (PV) Italy. Near isogenic
lines (NIL) were generated by backcrossing the dwarf
mutant four times into B73 and W23 lines. Dwarf8-1
(D8-1) and Knotted1 (Kn1) mutant were provided by
Stock Center Resources of MaizeGDB (http://www.
maizegdb.org/stock.php).
Dwarf mapping
D*-1023 was mapped in the segregating F2 popula-
tion using SSR markers chosen on the chromosome 1
(bin 1.09/1.10) from MaizeGDB (http://www.maizeg
db.org/ssr.php). F2 plants (obtained by selfing B73 ?/
D*-1023 9 W23 ?/?plants) were screened for
dwarf phenotype and a piece of leaf of each plant was
used for DNA extraction (Dellaporta et al. 1983).
Polymerase chain reactions (PCR) and gel running
conditions were performed as described in the SSR
Methods Manual by MaizeGDB (http://www.maizeg
db.org/documentation/maizemap/ssr_protocols.php).
Recombinant values were converted to map
distances using MAPMAKER 3 (Lander et al.
1987). Linkage analysis between D*-1023 and the
Kn1 dominant mutant that maps closely linked to D8-
1 was determined by scoring the progeny obtained
from the cross D*-1023/? ?/Kn1 9 ?/? ?/?. The
377
recombinant values were obtained as the ratio
between recombinant individuals (D*-1023 Kn1
and??) and the entire screened population.
Cloning and sequence analysis
D8-1023 double stranded 30 genomic sequence was
determined by sequencing two amplified overlapping
products (genomic DNA was amplified by high-
fidelity PCR, Pfu polymerase; Stratagene, La Jolla,
CA, USA). The 1560 and 184 bp genomic PCR
fragments obtained from homozygous D8-1023/D8-
1023 B73 were subcloned into a PCR-Blunt II-TOPO
vector (Invitrogen). The primers used, respectively
were: D8F5 (upstream primer 50 -TTAGCTGGCTAG
CTAGGCCTGT-30 , position -34) and D8R4 (down-
stream primer 50 -TCGAGAAATCGAACATGGTG
GA-30 , position ?1526); D8F8 (upstream primer 50 -C
AGTCCACCGACGCCTCCC-30 , position ?1549)
and D8R3 (downstream primer 50 -CCAGGTGCACG
GGCGCGAA-30 , position ?1732).
Four independent clones were sequenced, DNA and
deduced amino acid sequences were analyzed using
freely available computer software like CLUSTALW
(http://www.ebi.ac.uk/clustalw/) and BLAST (http://
www.ncbi.nlm.nih.gov/BLAST/).
Cosegregation analysis
Analysis was performed using specific primers
designed on the insertion region to obtain allele-
specific amplified products from ?(d8) and D8-1023
alleles.
In order to perform cosegregation analysis, F2 and
BC1 populations, obtained, respectively by selfing D8-
1023/?B73 plants and by crossing D8-1023/
?B73 9 ?/? B73 plants, were screened for wild
type or dwarf phenotype and a piece of leaf was used
for DNA extraction (Dellaporta et al. 1983). Polymer-
ase chain reactions (PCR) were performed using two
sets of primers, set 1 specific for ?(d8) allele: D8FWT
(upstream primer 50 -GACCTGTCGTCCTGGGTCG
A-30 , position ?292) and D8R6 (downstream
primer 50 -CCGTTGGCCGCCGCGGATG-30 , posi-
tion ?677); set 2 specific for D8-1023 allele: primers
D8FIN (upstream primer 50 -GACCTGTCGTCCTG
GGTGGT-30 , position ?292) and D8R6. The ampli-
fication fragments of 389 bp for D8-1023 and 386 bp
123
378
for d8(?) alleles were fractionated by electrophoresis
using 1.5% (w/v) agarose gels.
Semiquantitative RT-PCR
Total RNA was extracted from D8-1023 B73 homo-
zygous seedlings and wild type B73 inbred line as
control (5 individuals), using the method described
by van Tunen et al. (1988). Reverse transcriptase
polymerase chain reaction (RT-PCR) was used to
detect the D8 gene transcript. First strand cDNA was
synthesized with an oligo (dT) primer from total
RNA. The primer used was a 35-base oligonucleotide
with 17dT residues and an adapter (50 -GGGAATTCG
TCGACAAGC-30 ) (Frohman 1990). All RNA sam-
ples were treated with DNase (1 unit/lg) before
cDNA synthesis. The different samples of cDNA
were then diluted to obtain a uniform concentration.
First-strand cDNA was used as template for PCR
amplification. Amplification reactions were carried
out on samples containing an aliquot of cDNA
synthesized from 5 lg of total RNA, 19 Promega
polymerase buffer, 2.5 mM MgCl2, 200 lM each
dATP, dCTP, dGTP, and dTTP, 0.1 lM each primer,
and 1 unit of Taq DNA polymerase (Promega,
Madison, WI) and were performed in a final volume
of 50 ll. A set of primers specific for the orange
pericarp-1 (orp-1) gene, which encodes the l-subunit
of tryptophan synthase (Wright et al. 1992), was used
to standardize the concentration of the different
samples. orp-1 specific sequences were amplified
using the following primers: upstream primer, 50 -
AAGGACGTGCACACCGC-30 , and downstream
primer, 50 -CAGATACAGAACAACAACTC-30 . The
length of the amplified product was 207 bp. Several
cycles of successive cDNA dilutions and orp-1
amplifications were done in order to obtain similar
amplification signals in the different samples and to
ensure that amplification reactions were within linear
ranges. For mRNA detection of the D8 gene under
analysis, D8F8 (See ‘‘Cloning and sequence analy-
sis’’) and D8R2 (50 -AGGCATTGGAGCCCAGGTG
CA-30 , position ?1744) specific primers were used.
The amplified product was 196 bp and the identity of
the D8 product was confirmed by sequencing (cDNA
was amplified by high-fidelity PCR, Pfu polymerase;
Stratagene, La Jolla, CA, USA). PCR products were
fractionated on 1.2% (w/v) agarose gels.
123
Mol Breeding (2009) 24:375–385
Exogenous GA3 applications
For early GA treatments, seeds obtained from selfed
?/D8-1023 B73 heterozygous plants were germi-
nated in glass boxes on wet filter paper imbibed either
with water or with GA3 10-5 M (Sigma, St. Louis,
Mo, USA product no. G7645) in the dark at 25°C for
4 days. Seedlings were then grown under a light
source consisting of four cool white (F36T12/CW/
HO) fluorescent lamps from GTE SYLVANIA
(Lighting Products Group, Danvers, MA) for 6 days.
For prolonged GA3 treatments, seeds were germi-
nated directly in soil and plants grown in plant
growth chambers at 25°C for 35 days. Starting from
15 DAG (Day After Germination), plants were
sprayed every other day with a 10-5 M GA3 solution.
Histological analyses
To determine epidermal cell sizes, D8-1023 and wild
type control plants in the same background were
grown in normal conditions and at maturity leaves
and silks were collected and treated with a clearing
solution (160 g chloral hydrate, 20 ml glycerol in
60 ml water). Cleared leaves and silks were mounted
on slides, and interference contrast images were taken
using a Zeiss IMAGE R.D1 microscope equipped
with a AxioCam MRc1 digital camera.
Results
Isolation and characterization of the D*-1023
mutant
The dominant dwarf maize mutant isolated and
described in this work, at maturity, shows delayed
flowering, reduced stature, ranging from 60 to 70% in
a W23 NIL (Fig. 1a) to a 40–45% in a W23 9 B73 F1
Hybrid (Fig. 1b), caused by reduced internode length.
This mutant also shows thick broad leaves, that are
25–30% larger than wild type (Fig. 1c), a strong gene
dosage effect on phenotype and a less severe pheno-
type in comparison with the D8-1 dominant mutant as
shown in Fig. 1d. The dwarf phenotype is easily
detectable also in the first stage of plant growth
(Fig. 1e) and at maturity shows a tendency to produce
tillers (Fig. 1f). The dwarf mutant, in addition, is
altered in its floral development, in fact in the terminal
Mol Breeding (2009) 24:375–385 379

Fig. 1 Phenotype of the

new dwarf mutant. a and b
wild type (left) and D*-
1023/? mutant (right)
whole plants at maturity,
respectively in a W23 near-
isogenic line and a
W23 9 B73 F1 hybrid
genetic background. c
Leaves, wild type above
and mutant below in a W23
near isogenic line. d From
left to right D*-1023/?, D*-
1023/D*-1023, D8-1/?
whole plant in a W23
genetic background. e Wild
type seedling (left) and
dwarf (right). f Dwarf
tillering growth habit in a
B73 near isogenic line. g
Wild type ear (left) and
dwarf anthered ear (right). h
Mutant silks (above) and
wild type silks (below). i
Wild type anthers (left) and
mutant anthers (right)

flowers of the ears stamens are present (andromonoe-
cious ear) that, however, are sterile (Fig. 1g) and the
dimensions of the silks and anthers in the inflores-
cence are bigger, respectively by about 40 and 17%,
compared to the wild type (Fig. 1h, i).
Histological analyses performed using differential
interference contrast imaging microscopy showed, at
the mature stage of development, a different size of
epidermal leaf cells, stomata guard cells and silk cells
in dwarf plants compared with the wild type (Fig. 2).
Statistical analysis revealed that, at maturity, stomata
guard cell length, leaf epidermal cell width and silk
cell width in the homozygous mutant were larger
compared to the wild type (Table 1).
Genetic analysis and mapping of the D8*-1023
trait
This mutant was isolated in an F1 hybrid field,
suggesting that the mutation occurred spontaneously
in one parent in the previous generation and may be a
dominant factor. In fact segregation analysis of the
mutant phenotype indicated that D8*-1023 is a
monogenic dominant trait with a 100% penetrance
of the mutation (Table 2), although a big inter-line
variability of expression was observed (data not
shown). Dwarf F1 plants show a remarkable increase
of stature compared with the dwarf from a selfed
isogenic line (Fig. 1a, b). We also observed a dosage
effect of the mutant allele, in fact D8-1023/?
heterozygous plants were higher (by about 50%)
compared to D8-1023/D8-1023 individuals in a B73
NIL (Fig. 3).
With the aim of finding out whether this dwarf
mutation was previously isolated we mapped the
mutation. Considering the fact that this dwarf was
dominant and andromonoecious we thought that this
new mutation might represent a new allele of D8
gene with these characteristics. To test this hypoth-
esis, D*-1023 mapping was achieved by the analysis

123
380 Mol Breeding (2009) 24:375–385

Fig. 2 Effects of the dwarf

mutation on leaf epidermal
and silk cell size at
maturity. a Wild type leaf
epidermis cells. b Mutant
leaf epidermis cells. c Wild
type silk cells. d Mutant silk
cells. Bars = 20 lm

Table 1 Measurements at maturity of stomata cells length,

leaf epidermal and silk cells size in homozygous dwarf plants
vs. wild type
Wild type Dwarf

Stomata cell length (lm)a 15.1 ± 0.4 16.6 ± 0.4b

Epidermal cell length (lm) 53.1 ± 5.6 50.5 ± 2.7
Epidermal cell width (lm) 9.4 ± 0.7 15.1 ± 0.5b
Silk cell length (lm) 280.7 ± 33 220.3 ± 45
Silk cell width (lm) 20.3 ± 5.4 35.5 ± 6.8b
a
Abaxial side of 3rd leaves
b
Confidence interval significantly different from wild-type Fig. 3 Effects of the dwarf mutation dosage (D*-1023) on the
(wt) plants at P \ 0.05. Mean calculated from [100 plant height at maturity in a B73 genetic background compared
measurements to D8-1. Confidence intervals at 95% are shown

Table 2 Segregation of the dwarf phenotype observed in

genetic tests A polymorphism for the SSR markers umc1082 and
Cross Segregation v2
P bnlg1512 confirmed this hypothesis and established
the position of the mutation on bin 1.09 at a distance
wt Dwarf
of about 4.3 cM from umc1082 (Fig. 4). Recombi-
?/D*-1023  232 713 0.10 0.9–0.7 nant values were converted to map distances using
?/? 9 ?/D*-1023 423 429 0.04 0.9–0.8 MAPMAKER 3 (Lander et al. 1987). These mapping
data were confirmed using the Knotted1 (Kn1)
The data show that D*-1023 is a monogenic dominant
mutation dominant mutation, that maps closely linked to D8,
as a genetically dominant classical marker (Kn1
plants have an easy detectable phenotype causing
of simple sequence repeat (SSR) marker-distribution proliferation of tissue at vascular bundles on the leaf).
(chosen on the long arm of chromosome 1 where D8 With this aim 283 individuals obtained from the cross
maps) in an F2 segregating population of 105 plants. D*-1023/??/Kn1 9 ?/? ?/? were screened for the

123
Mol Breeding (2009) 24:375–385
Fig. 4 Mapping of the dwarf mutant isolated. Approximate
distance of D*-1023 from umc1082 SSR molecular markers
and Kn1 classical marker is shown on long arm of chromosome
1, bin 1.09. Genetic distance is expressed in cM
presence of recombinants ?/? ?/? (6 plants) and
D8/? Kn1/? (5 plants). The map distance between
Kn1 and D*-1023 was 3.8 cM, confirming the
position estimated using molecular markers (Fig. 4).
These map values are similar to those reported for
the D8 allele previously isolated (Harberd and
Freeling 1989), strengthening the hypothesis that
D8*-1023 is allelic to D8. Thus it became more likely
that D8*-1023 represents a mutation in the D8 gene.
Following the guidelines provided by the Maize-
GDB, we renamed the new mutation D8-1023.
Cloning, sequencing and gene expression analysis
of D8-1023 allele
With the aim of finding the molecular lesion in the
D8 gene which causes the mutant phenotype, we
designed specific primers on the basis of the D8 gene
sequence deposited in GenBank (Peng et al. 1999).
The genomic sequence of D8-1023 was deter-
mined by sequencing two amplified subcloned over-
lapping products of 1560 and 184 bp obtained from
homozygous D8-1023/D8-1023 (See ‘‘Materials and
methods’’). Alignment of D8-1023 allele vs d8 (?)
sequences present in GenBank, performed using the
CLUSTALW program, showed the presence of an
Fig. 5 Partial alignment between d8 wild type allele and
predicted proteins encoded by dominant mutant alleles. Wild
type d8 allele is compared with D8-1, D8-2023, D8-Mpl and
D8-1023 dominant mutant alleles encoded proteins with
381
insertion of 3 nucleotides (?307 GTG ?309 respect
to the start codon) in the coding region compared
with the d8 allele of A554 inbred line (accession
number AF413203). This small insertion causes the
addition of a valine residue at the position 103 in the
putative VHYNP domain, predicted by translating the
nucleotide sequence using the ORFINDER program,
and this would be the molecular lesion causing the
dwarf phenotype observed. In Fig. 5 we show the
partial proteins alignment (from ?33 E to N ?111)
between d8(?), D8-1, D8-2023, D8-Mpl and D8-
1023 alleles to point out DELLA and VHYNP
mutations causing dominant dwarf maize mutations
so far characterized.
Furthermore, in order to investigate D8 gene
expression we used specific primers D8F8 and
D8R2 (See ‘‘Materials and methods’’) to amplify
retro transcribed RNA extracted from seedlings.
Semi-quantitative RT-PCR analysis showed that the
expression level of D8 gene is the same in the wild-
type and the mutant, indicating that the phenotype
associated to the D8-1023 allele is not caused by a
difference in the gene expression level of D8 gene
(Fig. 6a).
Cosegregation analysis between D8-1023 allele
and the dominant dwarf phenotype
To verify the association between the dwarf pheno-
type and the D8-1023 insertion mutation, PCR-based
cosegregation analysis was performed using allele-
specific primers designed on the three nucleotide
insertion in the D8-1023 allele. Genotyping was
performed using F2 and BC1 populations: F2 plants
were obtained by selfing D8-1023/?B73 plants and
back-cross plants were obtained by crossing D8-
1023/?B73 9 ?/? B73 plants. The offspring were
screened for dwarf phenotype vs wild type and a
piece of leaf was used for genomic DNA extraction.
mutant N-terminal. Differences between wild-type and mutant
sequences (deletions, insertions and substitutions) are high-
lighted in white and the previously identified highly conserved
DELLA and VHYNP domains are shown
123
382 Mol Breeding (2009) 24:375–385

Fig. 6 Expression analysis of D8-1023 allele and genotyping

of plants. a Semi-quantitative RT-PCR analysis of D8 and orp1
genes were carried out using total RNA prepared from
seedlings. Orp-1 was used as control. b D8-1023 and wild
type (?) allele specific amplified products, obtained using,
respectively D8FWT and D8FIN forward specific primers,
used for cosegregation analysis

A 389 bp for D8-1023 and 386 bp for d8(?) specific

amplification product was obtained using, respec-
tively D8FIN/D8R6 and D8FWT/D8R6 primers
(Fig. 6b).
On the whole the results of this analysis showed
that all the 130 dwarf plants analyzed carried one or
two doses of D8-1023 allele and all the 558 wild type
individuals were homozygous for d8(?) allele, con-
firming a relationship between D8-1023 allele and the
presence of the dwarf phenotype.
GA3 treatment
Fig. 7 Effects of GA3 treatment on the elongation of wild type
and ?/D8-1023 and D8-1023/D8-1023 dwarf plants. a
Seedling elongation was evaluated after 10 days of GA
treatment (seeds germinated on filter paper imbibed with or
without 10-5 M GA3 solution), by measuring the distance
between the scutellar node and the first leaf. For each
determination at least 20 seedlings were measured. Confidence
intervals at 95% are shown. b Dwarf plants grown in soil were
sprayed for 20 days either with a 10-5 M GA3 solution (right)
or with water (left), starting at day 15 after germination

To test the effects of GA administration on D8-1023
mutant growth, we compared the elongation of
homozygous and heterozygous D8-1023 seedlings
and wild type siblings following treatments with
10-5 M GA3. Seeds obtained from selfed ?/D8-
1023 B73 heterozygous plants were germinated on
filter paper with or without gibberellins. Ten days after
germination, mutant and wild type seedling elongation
was measured and either D8-1023/D8-1023 or D8-
1023/? seedlings revealed a slight increase in elon-
gation when cultured in the presence of GA3 (Fig. 7a).
To test if the GA responsiveness of the dwarf mutants,
as established at the seedling stage, was maintained
during later stages of development, mutants were
sprayed every day with a 10-5 M GA3 solution. The
results obtained confirmed the positive response to GA
treatment previously observed (Fig. 7b).
Discussion
Height reduction in several different crops (in
particular wheat and rice) associated with increases
in yield has been achieved first of all by the use of

123
Mol Breeding (2009) 24:375–385
semi-dominant mutations having an abnormal
response to the gibberellin (GAs) hormones. Also
much of the current progress in understanding GA
metabolism comes from the isolation and character-
ization of single-gene mutants.
Several independent dwarf mutants have been
mapped in maize, altogether representing a very
heterogeneous category of dwarfing mutations (Coe
et al. 1988; Winkler and Freeling 1994; Galbiati et al.
2002).
In this work we have isolated a new dominant
Dwarf8 mutant allele named originally D*-1023 that
arose in an F1 maize population. It is characterized by
reduced stature, due to shortening of internodes, and
dark green, crinkly leaves (Fig. 1). Mutant plants
exhibit ectopic anthers in the ear (Fig. 1g) and show a
strong variation in the flowering time (data not
shown) as reported for the dominant dwarf mutants
previously isolated (Winkler and Freeling 1994).
Flowering time is an important agronomic trait
involved in the early process of domestication of
the modern maize plant, and among all the adaptive
traits it plays a central role in the quick expansion of
maize through America and later in Europe. Flower-
ing time is a character determined by many genes and
on the basis of QTL analysis at least 60 loci are
involved in the regulation of this trait (Chardon et al.
2004). Among these QTLs only Dwarf8 (Thornsberry
et al. 2001; Andersen et al. 2005; Camus-Kulandaiv-
elu et al. 2006) and Vegetative to generative transi-
tion 1 (Salvi et al. 2007) genes have been identified.
Furthermore, at the cellular level the leaf and silk
cells are bigger than those of the wild type (Fig. 2)
suggesting that the increase in organ size observed in
mutant leaves and silks is caused primarily by the
cells being larger while still maintaining the correct
organization inside the tissues (Table 1). It is an open
question whether there is in the D*1023 mutant a
change in cellular number as previously reported for
the d1 recessive mutation where the rates of stem cell
division and elongation are both affected (Abbe and
Phinney 1940, 1942). In fact the cell files in the d1
mutant leaves, contained only 40% of the number of
cells present in the normal siblings (Phinney 1946).
However, our data suggest that the reduction in
leaf length and the increase in leaf width character-
istic of the D8 phenotype (Fig. 1c) are due to a
reduction in the number and a widening of the shape
of the cells in the longitudinal files within the leaf.
383
The genetic analysis performed to understand the
inheritance of this dwarf mutation, demonstrated a
monogenic dominant inheritance of this trait
(Table 1) and the map position of this dwarf has been
established on the long arm of chromosome 1 (Fig. 4).
The results obtained from this analysis show that
D*1023 maps where D8-ref. was localized (Winkler
and Freeling 1994) and thus it was renamed D8-1023.
To test the relationship between our dwarf mutant
and the previously isolated D8 mutant alleles, D8-
1023 gene was cloned and the alignment with d8(?)
wild type alleles present in the database has shown up
the molecular lesion: an insertion of 3 bp within the
VHYNP domain, localized in the 50 of the gene, near
the DELLA domain. This insertion causes the
addition of one valine residue at the position 103 in
the putative VHYNP domain, predicted by translating
the nucleotide sequence (Fig. 5). The association
between the presence of this insertion in the coding
region labelling the D8-1023 allele was also con-
firmed by using allele-specific primers (Fig. 6b) on a
segregating population of 688 plants that did not
show any recombinant. Taken together these data
strongly suggest that this molecular lesion causes the
dwarf phenotype although a linkage disequilibrium
with one or more closely linked molecular lesion/
polymorphisms outside the sequenced region causing
a change of the D8 gene expression level could be
responsible for the phenotype observed. In fact a
recent work by Camus-Kulandaivelu et al. has shown
that a region upstream of the D8 coding sequence is
involved in the control of flowering time variation in
maize (Camus-Kulandaivelu et al. 2008). Neverthe-
less in our case it would be hard to explain the
dominant habit of this mutation because so far all the
dominant dwarf mutations isolated are due to a
molecular lesion in the N-terminal DELLA domain
causing the loss of GA perception and consequently a
constitutive repression of growth plant genes. Fur-
thermore, expression analysis carried out using RT-
PCR did not show any apparently significant differ-
ence between the level of expression of D8 gene in
our dwarf mutant and wild type (Fig. 6a).
Finally these data strongly suggest that the inser-
tion of valine 103 in D8 protein causes a mild
dwarfing effect on the plant’s phenotype, resembling
the phenotype previously decrypted for D8-Miniplant
(Harberd and Freeling 1989; Winkler and Freeling
1994).
123
384
The different alleles of D8 so far isolated cause
varying degrees of dwarfing to the phenotype,
depending on the N-terminal mutations in DELLA
domain, and only one (the D8-2023 allele) affecting
the VHYNP domain (Peng et al. 1999). Our mutation
is the first VHYNP single amino acid insertion
isolated, indicating the importance of this motif, so
far not well characterized in maize. Indeed the D8-
2023 mutation previously isolated causes the loss of
12 amino acids in the region of the VHYNP motif
(Fig. 5) but this mutation due the closeness of the
DELLA domain and the width of the insertion could
represent an indirect lesion of the DELLA motif due
to a tridimensional change of the D8 protein structure
modifying the DELLA region. In rice, Ueguchi-
Tanaka et al. (2007) demonstrated that DELLA and
VHYNP domains are essential for interaction of
SLR1 with GID1 (GIBBERELLIN INSENSITIVE
DWARF1) protein (a soluble GAs receptor) driving
the GA signalling, while Willige et al. (2007)
demonstrated in Arabidopsis that the VHYNP
domain is not necessary for the GAI/AtGID1 pro-
teins. Our result indirectly confirms in vivo the
Ueguchi-Tanaka results, indicating that a specific
lesion of this domain is able to modify the D8 activity
and the GA signalling, causing a dominant mutation.
Furthermore recent results obtained by Murase
et al., regarding the structural analysis of the
Arabidopsis GA3–GID1–DELLA complex, have
shown that the VHYNP motif is composed by the
aminoacid residues TVhynPxxLxxWxxxM (Murase
et al. 2008). The valine insertion found in D8-1023
mutation changes the distance between the W and M
residues (from WxxxM to WxxxxM) and this would
likely causing the loss of the correct motif confor-
mation determining the altered GA perception in the
dwarf mutant isolated.
The peculiarity of this mutation is also highlighted
by the less severe phenotype shown by D8-1023
mutant in comparison with the D8 mutants in the
DELLA domain previously isolated (Fig. 1b, d). This
could suggest a potential for direct utilization of this
mutation to lower the plant height of maize tropical
germplasm for biomass production for example or to
isolate new mutations in the VHYNP domain by
TILLING. Another aspect of our mutant is the partial
response to treatments with GA3 (Fig. 7), unlike the
other D8 mutant alleles (out of the D8-1591 muta-
tion) that are GA-insensitive, indicating a partial
123
Mol Breeding (2009) 24:375–385
degradation of D8-1023 protein in response to GA
stimuli.
Future molecular and biochemical studies will be
necessary to characterize this new mutation, in
particular at protein level, to better understand the
regulatory mechanism of GA-modulated growth.
Acknowledgments This work was supported by Fondo
Interno Ricerca Scientifica e Tecnologica (F.I.R.S.T. 2005,
2006 and 2007 to R. Pilu). We wish to thank Dr. Davide
Reginelli and Dr. Dario Panzeri for their hard work in the field.
References
Abbe EC, Phinney BO (1940) The effect of the gene dwarf-1
on the developmental pattern and cellular constitution of
the stem in maize. Am J Bot 27(10):1s
Abbe EC, Phinney BO (1942) The action of the gene dwarf1 in
the ontogeny of the stem in maize. Genetics 27:129
Andersen JR, Schrag T, Melchinger AE et al (2005) Validation
of Dwarf8 polymorphisms associated with flowering time
in elite European inbred lines of maize (Zea mays L.).
Theor Appl Genet 111:206–217. doi:10.1007/s00122-
005-1996-6
Camus-Kulandaivelu L, Veyrieras JB, Madur D et al (2006)
Maize adaptation to temperate climate: relationship
between population structure and polymorphism in the
Dwarf8 gene. Genetics 172:2449–2463. doi:10.1534/
genetics.105.048603
Camus-Kulandaivelu L, Chevin LM, Tollon-Cordet C et al
(2008) Patterns of molecular evolution associated with
two selective sweeps in the Tb1–Dwarf8 region in maize.
Genetics 180:1107–1121. doi:10.1534/genetics.108.088
849
Chandler PM, Marion-Poll A, Ellis M et al (2002) Mutants at
the Slender1 locus of barley cv Himalaya: molecular and
physiological characterization. Plant Physiol 129:181–
190. doi:10.1104/pp.010917
Chardon F, Virlon B, Moreau L et al (2004) Genetic archi-
tecture of flowering time in maize as inferred from
quantitative trait loci meta-analysis and synteny conser-
vation with the rice genome. Genetics 168:2169–2185.
doi:10.1534/genetics.104.032375
Coe EH Jr, Neuffer MG, Hoisington DA (1988) The genetics
of corn. In: Sprague GF, Dudley JW (eds) Corn and corn
improvement. Am Soc Agron, Madison, pp 81–258
Dellaporta SL, Wood J, Hicks JB (1983) A plant DNA mini-
preparation: version II. Plant Mol Biol Rep 1:19–21. doi:
10.1007/BF02712670
Emmerson RA (1912) The inheritance of certain abnormalities
in maize. Am Breeders Assoc Rep 8:385–399
Frohman MA (1990) RACE: Rapid Amplification of cDNA
Ends In: Innis MA, Gelfand DH, Sninsky JJ, White TJ
(eds) PCR protocols. A guide to methods and applications.
Academic Press, San Diego, pp 28–38
Fu X, Richards DE, Ait-Ali T et al (2002) Gibberellin-medi-
ated proteasome-dependent degradation of the barley
Mol Breeding (2009) 24:375–385
DELLA protein SLN1 repressor. Plant Cell 14:3191–
3200. doi:10.1105/tpc.006197
Fujioka S, Yamane H, Spray CR et al (1988a) Qualitative and
quantitative analyses of gibberellins in vegetative shoots
of normal, dwarf-1, dwarf-2, dwarf-3 and dwarf-5 seed-
lings of Zea mays. Plant Physiol 88:1367–1372. doi:
10.1104/pp.88.4.1367
Fujioka S, Yamane H, Spray CR et al (1988b) The dominant
non-gibberellin-responding dwarf mutant (D8) of maize
accumulates native gibberellins. Proc Natl Acad Sci USA
85:127–131. doi:10.1073/pnas.85.23.9031
Galbiati M, Landoni M, Consonni G et al (2002) Identification
and analysis of maize mutants defining six new genes
affecting plant stature. Maydica 47:169–180
Harberd NP, Freeling M (1989) Genetics of dominant gibber-
ellin-insensitive dwarfism in maize. Genetics 121:827–
838
Hedden P (2003) The genes of the green revolution. Trends
Genet 19:5–9. doi:10.1016/S0168-9525(02)00009-4
Hedden P, Kamiya Y (1997) Gibberellin biosynthesis:
enzymes, genes and their regulation. Annu Rev Plant
Physiol Plant Mol Biol 48:431–460. doi:10.1146/
annurev.arplant.48.1.431
Lander ES, Green P, Abrahamson P et al (1987) MAP-
MAKER: an interactive computer package for construct-
ing primary genetic linkage maps of experiment and
natural populations. Genomics 1:174–181. doi:10.1016/
0888-7543(87)90010-3
Monna L, Kitazawa N, Yoshino R et al (2002) Positional
cloning of rice semidwarfing gene. sd-1 rice: ‘‘green
revolution gene’’ encodes a mutant enzyme involved in
giberellin synthesis. DNA Res 9:11–17. doi:10.1093/
dnares/9.1.11
Multani DS, Briggs SP, Chamberlin MA et al (2003) Loss of an
MDR transporter in compact stalks of maize br2 and
sorghum dw3 mutants. Science 302:81–84. doi:
10.1126/science.1086072
Murase K, Hirano Y, Sun T-p et al (2008) Gibberellin-induced
DELLA recognition by the gibberellin receptor GID1.
Nature 456:459–463. doi:10.1038/nature07519
Peng J, Harberd NP (1993) Derivative alleles of the Arabid-
opsis gibberellin-insensitive (gai) mutation confer a wild-
type phenotype. Plant Cell 5:351–360
Peng J, Harberd NP (1997) Gibberellin deficiency and response
mutations suppress the stem elongation phenotype of
phytochrome-deficient mutants of Arabidopsis. Plant
Physiol 113:1051–1058. doi:10.1104/pp.113.4.1051
Peng J, Richards DE, Hartley NM et al (1999) ‘Green revo-
lution’ genes encode mutant gibberellin response modu-
lators. Nature 400:256–261. doi:10.1038/22307
385
Phinney BO (1946) Gene action in the development of the leaf
in Zea mays L. Ph.D. thesis, University of Minnesota,
Minneapolis
Phinney BO (1956) Growth response of single-gene dwarf
mutants in maize to gibberellic acid. Proc Natl Acad Sci
USA 42:186–189. doi:10.1073/pnas.42.4.185
Richards DE, King KE, Ait-Ali T et al (2001) How gibberellin
regulates plant growth and development: a molecular
genetic analysis of gibberellin signalling. Annu Rev Plant
Physiol Plant Mol Biol 52:67–88. doi:10.1146/
annurev.arplant.52.1.67
Salvi S, Sponza G, Morgante M et al (2007) Conserved non-
coding genomic sequences associated with a flowering-
time quantitative trait locus in maize. Proc Natl Acad Sci
USA 104:11376–11381. doi:10.1073/pnas.0704145104
Sasaki A, Itoh H, Gomi K et al (2003) Accumulation of
phosphorylated repressor for gibberellin signalling in an
F-box mutant. Science 299:1896–1898. doi:10.1126/
science.1081077
Spielmeyer W, Ellis MH, Chandler PM (2002) Semidwarf (sd-
1), green revolution rice, contains a defective giberellin
20-oxidase gene. Proc Natl Acad Sci USA 99:9043–9048.
doi:10.1073/pnas.132266399
Thornsberry JM, Goodman MM, Doebley J et al (2001)
Dwarf8 polymorphisms associate with variation in flow-
ering time. Nat Genet 28:286–289. doi:10.1038/90135
Ueguchi-Tanaka M, Nakajima M, Katoh E et al (2007)
Molecular interactions of a soluble gibberellin receptor,
GID1, with a rice DELLA protein, SLR1, and gibberellin.
Plant Cell 19:2140–2155. doi:10.1105/tpc.106.043729
van Tunen AJ, Koes RE, Spelt CE et al (1988) Cloning of two
chalcone flavanone isomerase genes from Petunia hybri-
da: coordinate, light regulated and differential expression
of flavonoid genes. EMBO J 7:1257–1263
Vogler H, Kuhlemeier C (2003) Simple hormones but complex
signalling. Curr Opin Plant Biol 6:51–56. doi:10.1016/
S1369-5266(02)00013-4
Willige BC, Ghosh S, Nill C et al (2007) The DELLA domain
of GA INSENSITIVE mediates the interaction with the
GA INSENSITIVE DWARF1A gibberellin receptor of
Arabidopsis. Plant Cell 19:1209–1220. doi:10.1105/tpc.
107.051441
Winkler RG, Freeling M (1994) Physiological genetics of the
dominant gibberellin-non responsive maize dwarf,
Dwarf8 and Dwarf9. Planta 193:341–348. doi:10.1007/
BF00201811
Wright AD, Moehlenkamp CA, Perrot GH et al (1992) The
maize auxotrophic mutant orange pericarp is defective in
duplicate genes for tryptophan synthase. Plant Cell 4:711–
719
123