A rare SNP mutation in Brachytic2 moderately reduces plant height and increases yield potential in maize
Journal of Experimental Botany
A rare SNP mutation in Brachytic2 moderately reduces plant height and increases yield potential in maize
Anqi Xing 2
Yufeng Gao 2
Lingfeng Ye 1
Weiping Zhang 2
Lichun Cai 2
Ada Ching 0
Victor Llaca 0
Blaine Johnson 4
Lin Liu 1
Xiaohong Yang 2
Dingming Kang 1
Jianbing Yan 3
Jiansheng Li 2
0 DuPont Co., Agricultural Biotechnology , 200 Powder Mill Road, Wilmington, DE 19805 , USA
1 College of Agronomy and Biotechnology, China Agricultural University , Beijing 100193 , China
2 National Maize Improvement Centre of China, China Agricultural University , Beijing 100193 , China
3 National Key Laboratory of Crop Genetic Improvement, Huazhong Agricultural University , Wuhan 430070 , China
4 Pioneer Hi-Bred Intl , 1501 Road P, York, NE 68467-8234 , USA
Plant height has long been an important agronomic trait in maize breeding. Many plant height QTLs have been reported, but few of these have been cloned. In this study, a major plant height QTL, qph1, was mapped to a 1.6 kb interval in Brachytic2 (Br2) coding sequence on maize chromosome 1. A naturally occurring rare SNP in qph1, which resulted in an amino acid substitution, was validated as the causative mutation. QPH1 protein is located in the plasma membrane and polar auxin transport is impaired in the short near-isogenic line RIL88(qph1). Allelism testing showed that the SNP variant in qph1 reduces longitudinal cell number and decreases plant height by 20% in RIL88 (qph1) compared to RIL88 (QPH1), and is milder than known br2 mutant alleles. The effect of qph1 on plant height is significant and has no or a slight influence on yield in four F2 backgrounds and in six pairs of single-cross hybrids. Moreover, qph1 could reduce plant height when heterozygous, allowing it to be easily employed in maize breeding. Thus, a less-severe allele of a known dwarf mutant explains part of the quantitative variation for plant height and has great potential in maize improvement.
Brachytic2; major QTL; maize (Zea mays); mild mutation; plant height; rare allele
Short stature, erect leaf angle, disease resistance, and high
yield are traits that have been pursued by breeders for
decades. Cereal production sharply increased in the 1960s as
the ‘Green Revolution’ popularized the use of dwarf and
semi-dwarf cultivars. Likewise, maize (Zea mays)
production improved dramatically due to the adoption of hybrids
and use of moderately short varieties that are more
resistant to lodging and compatible with higher planting density
(Duvick, 2005). Reduced ear height and an increased plant
height/ear height ratio also have the potential to increase
dry matter accumulation (Yang et al., 2010a). Many of the
Green Revolution genes, such as sd-1 in rice (Sasaki et al.,
2002) and rht1 in wheat (Peng et al., 1999), have been
identified and utilized in crop improvement. These genes encode
proteins that either regulate the synthesis of plant hormones
or modulate their signalling pathways. Several genes that
strongly influence plant height and are qualitatively
inherited have also been cloned in maize (Winkler and Helentjaris,
1995; Thornsberry et al., 2001; Multani et al., 2003); however,
these mutants have not found applications in maize breeding
due to their adverse impact on grain yield. Identification of
alleles moderately reducing plant height is highly desirable.
Plant height QTLs are favourable candidates as mild height
modulators. A number of such QTLs have been reported in
cereal crops but only a few of these have been cloned (Xue
et al., 2008; Yan et al., 2011; Teng et al., 2012). The large
genome size of maize makes QTL cloning a time-consuming
task, although some progress has been made (Wang et al.,
2005; Salvi et al., 2007; Zheng et al., 2008). Moreover,
planting density affects the grain yield of maize more than other
members of the grass family (Vega et al., 2001). In the past
few decades, maize yield has been increased mainly by
adopting modern hybrids that are more tolerant to high planting
density and other biotic or abiotic stresses, and have higher
efficiency in resource usage (Liang et al., 1992; Vafias et al.,
2006). Due to competition for light, plant height increases
with planting density. This tends to decrease stalk diameter
and increase the potential of stalk lodging, leading to yield
losses (Park et al., 1989; Hashemi et al., 2005). Modifying
plant height and other architecture components, while not
influencing grain yield, is among the key factors in
developing cultivars for compact planting.
We previously identified a major QTL affecting multiple
traits in bin7 on chromosome 1 using F2:3 (a series of F3
families derived from F2 individuals), immortalized F2, and
recombinant inbred line (RIL) populations derived from
Zong3 and 87-1, a hybrid known as ‘Yuyu22’ that has been
widely planted in China for the past two decades. This QTL,
designated qph1, was mapped near marker umc1122
(chromosome 1: 201677909- 201678070) and explains ~10% of the
phenotypic variation for plant height and ear height in the
RIL, immortal F2, and F2:3 population of Zong3/87-1. qph1
was significantly associated with plant height and marginally
affects yield and yield components (Yan et al. 2003; Yan et al,
2006 Tang et al, 2007a, b; Ma et al, 2007; Tang et al, 2010).
The objectives of this study were to fine map and clone qph1
and to evaluate its use in maize breeding.
Materials and methods
Mapping population of qph1 and near-isogenic lines
A RIL population of 294 lines was constructed previously from a
Zong3 × 87-1 cross. Two RILs, RIL88(qph1) and RIL279(QPH1), which
differed in plant height but shared a large portion of the same
genetic background, were chosen to generate the near-isogenic
line RIL88(QPH1), with the QPH1 allele in a RIL88 background
(Supplementary Figure S1). RIL88(qph1) was used as the recurrent
parent to backcross RIL279 and plants were selected based on
phenotype (tall plants were kept) in each generation until BC4F1.
Ninetyfour BC4F1 individuals were screened by 101 SSR markers all over
the maize genome and a tall line with the QPH1 allele, 05YH175-2,
was selected on the basis of having the smallest segments introgressed
from RIL279 on chromosome 1. 05YH175-2 was then selfed
repeatedly to produce BC4F3 and its progeny, 08YB036-7, was selected
on the basis of containing a single segment of RIL279 on
chromosome1 exclusively in RIL88 background. 08YB036-7 was then used
to generate BC5F2 and BC6F2 fine-mapping populations, and to
generate the near-isogenic line of QPH1 (Supplementary Figure S12).
8030 BC6F2 kernels were chipped and genotyped with two flanking
markers, umc2396 and MHC412 (Supplementary Table S5).
F2 populations and single-cross F1 hybrids
Four F2 populations, 4F1 × 81162, Ye107 × Zheng32, Ye107 × B73,
and Zong3 × Chuan48-2, were generated. Two markers closely
linked to qph1, umc2396 and MHC412, were used for genotyping
(umc2396 for the Zong3/Chuan48-2 F2 population and MHC412
for the other three F2 populations). A small piece of each F2 kernel
was chipped and genotyped using the soda boiling DNA extraction
method (Gao et al., 2008) before planting; more than 50, 100, and
50 seeds in the QPH1/QPH1, QPH1/qph1, and qph1/qph1 genotype
classes were planted and phenotyped, respectively. Zong3, 81162,
Ye107, W138, B73, and Chang7-2 were crossed with RIL88(qph1) and
RIL88(QPH1) to generate six pairs of hybrids; >50 seeds were planted
and analysed in each genotype class. Two replicates of the F2
population and hybrids were planted to collect data for plant height, ear
height, and yield components. Plots were designed with 50 rows per
plot and 13 individuals per row.
br2 mutant-derived populations
The br2 mutants 117A, 114F, 114G, and 121B were provided by
MGCSC (Maize Genetics Cooperation Stock Centre); 117A carries
the Hahn6 allele of br2 (Leng and Vineyard, 1951). 114G and 114F
are linkage stocks from two different sources that each carried an
unspecified br2 mutant; these two lines also have hm1, hm2
mutations besides the br2 mutation (http://www.maizegdb.org). 121B was
originally named mi8043 and was found to be an allele of the br2
gene (Marty Sachs, personal communication). 117A, 114F, 114G,
and 121B were crossed with RIL88(qph1) and RIL88(QPH1) to develop
four pairs of single-cross hybrids. 117A × RL88(qph1) was selfed to
generate an F2 population; umc2396 was used for genotyping. Plant
height and ear height data were obtained for 36, 89, and 54
individuals in br2-117A/br2-117A, qph1/br2-117A, and qph1/qph1 genotype
classes, respectively. Plant height and ear height of individuals in
these populations were measured and analysed. Plot design was the
same as used for the four F2 populations and six pairs of single-cross
Dwarf line N546 is derived from Mexican super dwarf (Johnson
et al., 1998). Nine crosses, 93NEX501 × PHVRZ, 93NEX501 ×
PHHHN, 93NEX501 × PHVNV6, Y93NEX504 × PHHHN,
YN546 × PHF0D, YN546 × PHCCW, YN546 × PHVRZ6,
YN546 × PHVNV, and YN546 × PH128S were made between
parents carrying the br2-bj allele (the former) and normal elite
inbred lines (the latter). F2 individuals were genotyped with
markers PZE-101155635 and PZE-100001759, which are adjacent to
br2-bj, and phenotyped for plant height and ear height (Ganal
et al., 2011).
DNA preparation and sequence analysis
DNA was extracted from young seedlings using the CTAB method
(Dellaporta et al., 1983). For BC6F2 mapping population and the
four F2 populations, DNA was extracted from chipped kernels using
the soda boiling DNA extraction method (Gao et al., 2008). The
genomic sequences of qph1 and QPH1 were amplified with H109F/
H106R, H114F/R, H103F/R, H212F/233R, and H115F/116R. The
Br2-bj allele was amplified with HF1F/R and HF2F/R. Primer
sequences are listed in Supplementary Table S5. PCR was performed
with Phusion High Fidelity Master Mix with HF buffer (Thermo
Fisher Scientific, Pittsburgh, PA, USA) according to standard
protocol. PCR products were ligated into the T-easy vector and colonies
containing the desired PCR fragment were picked and sequenced;
multiple colonies for each PCR product of each recombinant were
sequenced and analysed to eliminate PCR errors.
Scanning electron microscopy
The second and sixth internodes at the mid-elongation stage (15
expanded leaves and 19 visible leaves) and the uppermost
internodes at the adult stage (19 expanded leaves) of RIL88(qph1) and
RIL88(QPH1) were subjected to scanning electron microscopic
examination. Stem tissues from corresponding internodes of RIL88(qph1)
and RIL88(QPH1) were cut into 1mm longitudinal and transverse
sections and fixed in FAA (Formalin:acetic acid:70% ethanol, 1:1:18,
v/v/v). Fixed samples then went through dehydration with a series
of graded ethanol (15 min in 70%, 80%, 90%, and 100% ethanol).
Samples were then treated with isoamyl acetate for 15 min twice to
replace the remaining ethanol and subjected to critical point
drying (HITACHI, HCP-2). Dried samples were mounted on a suitable
working stage and coated with Pt using a high vacuum (Eiko IB.3,
ION COATER). Scanning electron microscope HITACHI S-3400N
was used for imaging.
Subcellular localization of QPH1
QPH1 coding sequence was amplified by PCR using primer pair
H235/GFP-R and cloned into a GFP vector with the 35S promoter
to express the QPH1–GFP fusion protein. Plasmid containing the
QPH1–GFP construct was transformed into onion epidermal cells
through gene gun bombardment (Bio-Rad PDS-1000). Transformed
cells were then incubated in 1/2 MS media for 20 h at 28°C and
examined by confocal laser scanning microscopy (Nikon EZ-C1).
Stem tissues of RIL88(qph1) and RIL88(QPH1) in three stages, at the
beginning, middle, and end of elongation, was collected. In each
developmental stage, 10 biological replicates of RIL88(qph1) and
RIL88(QPH1) were sampled. Total RNA was extracted with Trizol
reagent (Invitrogen, Carlsbad, CA, USA) and complementary
DNA was synthesized using the AMV reverse transcription system
(Promega, Madison, WI, USA) with Oligo (dT) primer. RT12F/R
were used to amplify QPH1 (qph1) using SYBR Premix Ex TaqTM
(TAKARA, Shuzo, Kyoto, Japan), with β–actin1 as the endogenous
control. Real-time PCR was performed with Real Time PCR
system 7500 (Applied Biosystems) using the 2-ΔΔCT method (Livak and
Schmittren, 2001) according to the standard procedure.
Sequence analysis in diverse maize lines and teosinte
The association panel with 527 inbred lines used in this study is
described by Yang et al. (2010b). The 1.4 kb qph1 target region
containing all five SNPs was amplified with primers 5N3F and 211R.
The 839 bp SNP5259-containing region was amplified with 212F
and 211R from 192 teosinte accessions. The CIMMYT genebank
represented a major portion of the diversity present in the teosinte
Gene transformation in Arabidopsis T-DNA insertional mutant
The Arabidopsis mutant SAIL_716_H02 (Columbia ecotype),
designated atpgp1-2 (At2g36910), was obtained from ABRC (Ohio
State University, Columbus, OH) and genotyped with primers 716F,
716R, and LB2 for the homozygous mutant according to
standard procedure (http://signal.salk.edu/tdnaprimers.2.html). Coding
regions of maize qph1 and QPH1 alleles were amplified with primer
pair 232F/233R. Site-directed mutagenesis was performed on qph1
coding sequence to change the T to G (wild-type genotype). QPH1,
qph1, and mqph1 (the mutated qph1) were overexpressed using
PBI121 under the CaMV 35S promoter in the atpgp1-2 mutant.
Transgenic plants were selected with kanamycin (50mg l–1). Plant
height was measured at 28 days (16/8 light/dark photoperiod,
21–23°C). Coleoptile length for 7-day-old plants grown on 1/2MS
solid medium was measured with ImageJ software (National
Institutes of Health, http://imagej.nih.gov/ij) with the pictures of
the plants taken by digital camera.
Polar auxin transport measurements
Auxin transport assays were performed using the protocol described
by Li et al. (2007) and Multani et al. (2003) with some
modifications. Ten replicates of RIL88(qph1) and RIL88(QPH1) seeds were
grown in sand for 5 days and harvested for coleoptiles in length of
3 cm. Coleoptile segments were equilibrated in1/2MS (pH 5.8)
liquid media for 2 h and the apical portion was submerged in 1/2 MS
solid medium containing 0.35% phytogel, 500 nM unlabelled IAA
(Sigma-Aldrich), and 500 nM [3H]-labelled IAA (specific
activity 20 Ci mmol–1; American Radiochemical, St Louis, MO, USA)
and incubated in the dark for 5 h at 25°C. Coleoptiles were then cut
into 0.5 cm sections and washed twice with 1/2 MS liquid medium.
Washed coleoptile sections were incubated in 1 ml scintillation fluid
for 16 h and counts were made with a scintillation counter
Effects of qph1 on plant height and other traits
To identify the underlying gene for QTL qph1, RIL88(qph1),
and RIL279(qph1), two lines carrying qph1 and QPH1 alleles,
respectively, and sharing a large proportion of the same genetic
backgrounds, were used to generate a BC4F2 fine-mapping
population (Supplementary Figure S1). The previously
identified plant height QTL qph1 was further mapped between
markers umc1035 (chromosome 1: 195219753-195219900)
and umc2236 (chromosome 1: 198266651-198266734) on
chromosome 1 in BC4F2 (Supplementary Figure S2). In
BC4F2:3, the numbers of qph1/qph1, qph1/PQH1, and QPH1/
QPH1 individuals were 21, 42, and 32; segregation of tall to
short plants was ~3:1 for both plant height and ear height
(χ2 = 0.196, P > 0.05), indicating the presence of a single
recessive Mendelian factor (Fig. 1). In contrast, flowering
time and yield components displayed continuous variation
(Supplementary Figure S3), consistent with the hypothesis
that qph1 has major effects on plant height and ear height
and minor effects on other traits. Plant height and ear height
differences were highly significant with P-values of
9.30E48 and 3.40E-44 (t-test) with the dominant allele of QPH1
contributing 90% to plant height and 87% to ear height.
Differences in days to silking and days to tasseling were also
marginally significant (P = 0.0303 and 0.0002, respectively),
revealing that the QPH1/QPH1 individuals flower slightly
earlier. Although ear length, diameter, and weight were all
significantly different (P = 0.0002, 9.50E-05 and 9.00E-08),
the effects of qph1 on these yield components were relatively
small (Table 1).
Morphological and cytological observations
To study the effect of QPH1 and qph1 alleles on plant height
and its components in the RIL88 background, RIL88(QPH1)
(BC6F3), which only has a 5 kb segment of RIL279 introgressed
into RIL88 background, was generated (Supplementary
126 130 134 138 142 146 150 154 158 162 166 170 174
Plant height (cm)
Fig. 1. Plant height and ear height variation in BC4F2:3. Blue, green and red bars represent the number of qph1/qph1, QPH1/qph1, and QPH1/QPH1
individuals in the BC4F3 population developed from RIL88 and RIL279. Plant height (A) and ear height (B) segregations are consistent with the segregation
of a single Mendelian factor.
an amino acid substitution from arginine (R) to leucine (L)
To evaluate the allelic effect between the natural variation
present and the known br2 mutants, four br2 mutants (114F,
114G, 117A, and 121B) were obtained and crossed with
RIL88(QPH1) and RIL88(qph1), respectively. Significant
differences in plant height and ear height between the hybrids were
observed; the phenotypic defects of the four br2 mutants
(1, 0.3) (1, 0.3) (0, 0.3) (1)
(6, 0.1) (4, 0.2) (0, 0.3) (1)
F2 individual F3 progeny (left, BJ; right, HN)
135.1111247549185 10.6 111116636305966.....33079116.6611..513..9246 111112555364392.....1641611106611...41..99342
113594 115786..69121.5.33 116675..04171.7.31
114248..8918.2. 72 115729..8798..23 113657..7819.0. 15
F2 individual F3 progeny (left, BJ; right, HN)
116450 117774..33 1186.. 42 116619..951155..41
111473758 111368087...412 959... 966 111964172...3641115..995.1
111111567673883349 111111676625013171......461684 182192.76.81.... 553222 111111665645680924......63639611111835790.5.....026473
112746..9426161.9.34 2 117272..50 71.0. 42 113728..451100..89
the control. As revealed by transient expression results, the
untargeted GFP was expressed throughout the whole
transformed cell in cytoplasm, nucleus, and membrane. However,
the QPH1–GFP fusion protein was expressed exclusively in
membrane (Supplementary Figure S8); this result is
consistent with the known localization of br2 homologues and
cellular function of ABC transporters.
were completely compensated by crossing with RIL88(QPH1)
and only partially reversed by crossing with RIL88(qph1)
(Supplementary Figure S7), implying that qph1 is allelic to the
Br2 gene and is defective compared with QPH1. Moreover,
in the F2 population of 117A and RIL88 (qph1), plant height
and ear height segregated in a 3:1 ratio (qph1/qph1 and qph1/
br2-117A plants to br2-117A/br2-117A plants), indicating
that the qph1 allele is dominant to br2-117A (Supplementary
Figure S7). Moreover, although RIL88(qph1) is not as short
as the four br2 mutants, it has comparably low ear height
(Supplementary Figure S7), which is a favourable feature
directly related to lodging resistance.
beginning of elongation, reached a higher level in
mid-elongation at period B, and began to decrease in period C before
the end of elongation. T-test showed no significant
differences in expression between RIL88(qph1) and RIL88(QPH1)
in any of the three stages, indicating that the plant height
and ear height difference between the near-isogenic lines is
not caused by the difference in transcription level of qph1
(Supplementary Figure S9).
Association analysis using the qph1 SNPs
Based on the association result of plant and ear height in a
population consisting of 527 inbred lines (Yang et al, 2010b),
SNP5259 is a rare SNP that only exist in five lines and the four
synonymous SNPs (SNP4519, 4999, 5041, and 5320) were
shown not to be associated with phenotype. Due to the
low number of inbred lines that have nucleotide T at
position 5259, the non-synonymous SNP5259 could not be
validated by association analysis (Supplementary Table S2). The
five inbred lines that harbour SNP5259 (T), 81162, Ye107,
Dan9046, W138, and Zong3 are all temperate lines that
belong to the same heterotic group. All five lines have plant
heights shorter than average: 113, 144, 133, 136, and 153 cm
for 81162, Ye107, Dan9046, W138, and Zong3, respectively.
(Data was collected from >50 plants grown in five different
locations; the average plant height of 527 inbred lines was
A total of 192 teosinte entries (data not shown) were also
sequenced and analysed for the qph1 confidence interval.
The four synonymous SNPs were found existing in either
homozygous or heterozygous states in teosinte; however,
only the homozygous G allele (wild-type genotype) was
identified for SNP5259, suggesting that the causative mutation
in qph1 most likely occurred as part of the temperate maize
breeding program after the domestication of maize. The rare
frequency of SNP5259 (T) also implies that the mutation
occurred very recently and is not widely used in breeding
Validation of qph1 and its functional site with
Arabidopsis mutant atpgp1-2
To validate the function of qph1 and its functional site
SNP5259, site-directed mutagenesis was performed on the
maize qph1 allele to mutate SNP5259 (T) to SNP5259 (G).
QPH1, qph1, and mqph1 (mutagenized qph1) were cloned
into vector PBI121 and overexpressed with the 35S
promoter in Arabidopsis T-DNA insertional mutant atpgp1-2
(AtPGP1, At2g36910, is the homologue of the maize Br2
gene in Arabidopsis), which has reduced plant height and
coleoptile length (Ye et al, 2013). QPH1 and mqph1 could
restore the plant height and coleoptile length of the
atpgp12 mutant to that of the wild type; in contrast,
transformation of atpgp1-2 with the qph1 allele led to an intermediate
level of rescue (Fig. 5). No significant difference was observed
between QPH1 and mqph1 transgenic plants in plant height
and coleoptile length, but a significant difference was detected
between qph1 and mqph1 transgenic plants (P = 0.0007 for
plant height and P = 0.005 for coleoptile length) (Fig. 5AB);
the only difference between mqph1 and qph1 coding sequence
is the G/T polymorphism at SNP5259; this result is
consistent with the hypothesis that SNP5259 (T) is the underlying
causative mutation in qph1. Moreover, statistical analysis
showed that the difference in plant height and coleoptile
length between qph1 transgenic plants and the atpgp1-2
control also reached significant levels (P = 0.00023 and
2.93E11), suggesting that qph1 is not a complete loss of function
allele (Fig. 5A, B).
Polar auxin transport in RIL88(qph1), RIL88(QPH1) and the
br2 mutant 114F
Because the maize Br2 gene is known to function in polar
auxin transport (Multani et al, 2003), an assay of
basipetal transport of [3H] IAA was carried out in the coleoptiles
of RIL88(qph1), RIL88(QPH1), and the null br2 mutant 114F.
Results showed that [3H] IAA translocation to the lower
sections of the coleoptiles was significantly reduced in RIL88(qph1)
and the br2 mutant 114F compared to RIL88(QPH1) (Fig. 6).
As the br2 mutant consistently showed increased loading of
auxin into the upper coleoptile near the site of application in
former studies (Multani et al, 2003), the difference was
probably caused by the impaired polar auxin transport in
defective br2 plants. Consistent with the hypothesis that the qph1
allele is partially defective compared with the br2-117A allele,
a significant difference in [3H] IAA translocation between
RIL88(qph1) and 114F was detected, indicating that qph1 was
not a complete loss of function allele.
Functional assessments of qph1 in F2 populations of
different genetic backgrounds
To evaluate the effect of qph1 under different genetic
backgrounds, four F2 populations, 4F1 × 81162, Ye107 × Zheng32,
Ye107 × B73, and Zong3 × Chuang48-2 were constructed by
crossing the three lines containing the rare SNP (T) (81162,
Ye107, and Zong3) with normal inbred lines (4F1, Zheng32,
B73, and Chuan48-2). The four F2 populations were planted
in two different locations, Beijing and Hainan (200
individuals per population in each location). In each replicate,
50 qph1/qph1, 100 qph1/QPH, and 50 QPH1/QPH1
individuals per population were phenotyped. In both locations,
a significant difference in plant height and ear height was
detected between QPH1/QPH1 and qph1/qph1 individuals
and between QPH1/qph1 and qph1/qph1 individuals. Plant
height difference between QPH1/QPH1 and qph1/qph1
individuals in the B73 × Ye107 F2 population is shown as an
example (Supplementary Figure S10). Single-factor ANOVA
analysis in 4F1/81162 F2 detected a strong dominant effect
for plant height (D/A = 1.3 and 0.9, P = 6.06E-16 and
1.21E18 in 09HN and 10BJ, respectively); similar results were also
observed in the other three F2 populations. 15–49% and
4–37% phenotypic variation could be explained by qph1 for
plant height and ear height, respectively, in different genetic
backgrounds and environments (Supplementary Table S3
and Supplementary Figure 7A). These results show that qph1
Distance from site of application (cm)
Fig. 6. Transport of [3H] IAA in coleoptiles of RIL88(QPH1), RIL88(qph1), and the
br2 mutant 114F. CPM, counts per minute. Error bars indicate SD, n=8.
has a consistent effect in reducing plant height and ear height
in multiple genetic backgrounds.
A semi-dwarf mutant N546 (Johnson et al., 1998) has
features very similar to the br2 recessive lines and a QTL peak for
plant/ear height was detected at the Br2 region in the N546 ×
PHB00 F2 population. The N546 Br2 allele is designated
br2bj; it has a 3.5 kb insertion of En/Spm-like transposon 660 bp
upstream of the predicted TATA box and a complete gag/
pol retro-transposon insertion of 4.7 kb in exon 5 which
truncates the last 153 amino acids of the protein. Mean plant/ear
height for br2-bj/br2-bj individuals are 88% and 50% of the
heterozygous and wild-type plants, respectively. Consistent
with the predicted reduction or loss of function of this
complex allele, br2-bj behaves as a recessive allele in nine F2
populations constructed by crossing N546 and two of its
conversion lines N501 and N504 (with the br2-bj allele) with
normal lines (PHVRZ, PHHHN, PHVNV, PHF0D, OHCCW,
and PH128S). In these segregating populations, heterozygous
and wild-type plants showed no significant difference in plant
height; ear height between wild-type and heterozygous plants
showed a significant difference in some crosses depending on
the background (Supplementary Figure S11). Results
indicate that br2-bj has different effects on plant height and ear
height in F2 segregation populations from qph1.
Functional assessments of qph1 in six pairs of
Because hybrid maize is used in agriculture, the effect of
qph1 was also estimated in hybrid backgrounds. Six pairs of
single-cross hybrids were generated by crossing RIL88(QPH1)
and SNP5259 (T)-containing RIL88(qph1) with the inbred
lines with or without the SNP5259 (T); effects of qph1/qph1
with qph1/QPH1 and QPH1/qph1 with QPH1/QPH1 under
the same hybrid genetic backgrounds were evaluated. Plant
and ear height showed significant differences between each
pair of hybrids (Fig. 7B). In the hybrids derived by
crossing Ye107, 81162, Zong3, and W138 with RIL88(qph1), plant
heights between F1 individuals of qph1/qph1 were reduced by
10–24%, respectively, compared to their QPH1/qph1
counterparts derived from RIL88(QPH1). Reductions in ear height
were also significant, ranging from 26 to 38%. These results
indicate the strong effect of the homozygous recessive qph1/
qph1 in reducing plant height and ear height. Moreover, in
B73 and Chang7-2 F1 hybrids, due to the incomplete recessive
feature of qph1, significant differences in plant height and ear
height between QPH1/QPH1 and QPH1/qph1 hybrids were
also detected; phenotypic variations in plant height and ear
height range from 3 to 12% for plant height and 6 to 23% for
ear height, indicating that qph1 is able to reduce plant height
and ear height in a heterozygous state. qph1 is proved to have
a significant effect on plant height and ear height in hybrids,
and the incomplete recessive feature of qph1 allows the plant
and ear height of a hybrid to be modified by introducing it
into only one parent.
Prospect evaluation of the impact of qph1 on
In our previous studies, a pleiotropic effect of qph1 causing a
marginal influence on yield and yield components was detected
in RIL, IF2, F2:3, and BC4F2 populations of Zong3/87-1; this
was probably contributed by the genetic backgrounds (Tang
et al., 2007a, b). To evaluate the potential of qph1 in maize
improvement in terms of increasing yield potential, a number
of agronomic traits of the six pairs of F1 and four F2
populations derived from qph1-containing lines were measured and
analysed. Most yield components exhibited no significant
difference between qph1/qph1, QPH1/qph1, and QPH1/QPH1
individuals in the four F2 populations, and significant
differences could be detected in only one biological repeat for a
few traits in some F2 populations, indicating that qph1 has
no or a slight influence on maize yield under different genetic
backgrounds while maintaining a shorter plant height and
ear height (Supplementary Figures S10 and S12). Likewise,
except for ear kernel weight and 100-kernel weight in 81162
F1, and ear weight and days to shedding in Zong3 F1 and ear
kernel weight in B73 F1, no additional significant difference
was identified between each pair of the single-cross hybrids,
suggesting that qph1 only has a minor impact on yield under
hybrid backgrounds (Supplementary Table S4). These results
suggested that qph1 could significantly reduce plant height
and ear height with no or very little negative impact on yield
under multiple genetic backgrounds. qph1 could affect plant
height and ear height when heterozygous, making it very
useful for hybrid maize breeding.
Molecular mechanism underlying the major plant
height QTL qph1
In this study, we report a rare SNP mutation in the maize
Brachyric2 gene underlying the major plant height QTL qph1.
The maize Br2 (ZmPGP1) gene is an ABC (ATP-binding
cassette) transporter which belongs to the MDR (multi-drug
resistant) class of P-glycoprotein (Noh et al., 2001; Multani
et al., 2003) and functions in polar auxin transport as an
efflux carrier. The protein molecule consists of two
transmembrane domains (TMDs) that provide the translocation
pathway of auxin and two cytoplasmic nucleotide-binding
domains (NBDs) that hydrolyse ATP and drive the transport
reaction (Chang and Roth, 2006; Aller et al., 2009). The two
TMDs span the membrane through the 12 α-helices (six per
domain) and enable membrane insertion and regulation. The
predicted causative mutation of qph1, SNP5259 (T), which
resulted in the arginine to leucine substitution, is located on
the ninth α-helix in the TMD of Br2 and thus affects the
efficiency of the transmembrane channel (Fig. 8). As an efflux
carrier, amino acid residues along the transporter channel
are strictly arranged (in this case, all residues are positively
charged); Arg has a hydrophilic and positively charged side
chain, while Leu is hydrophilic and neutral; the
substantial change from Arg to Leu in qph1 is very likely to have
affected its interaction with negatively charged IAA– inside
the cell. Moreover, auxin is synthesized predominantly in the
shoot apex, young leaves, and developing seeds (Ljung et al.,
2001) then dispensed to other organs by multiple efflux and
influx transporters (Friml et al, 2002; Zhao et al, 2010). The
results are consistent with the hypothesis that the defective
qph1 allele in RIL88(qph1) impaired basipetal auxin transport,
which led to the auxin insufficiency in lower internodes and
resulted in shortened internodes. Reduced cell division and
changes in vascular bundle development were observed in
RIL88(qph1) lower internodes, which is consistent with auxin
deficiency (Galweiler et al., 1998). qph1 affects leaf number,
leaf angle, and flowering time minimally compared with plant
height and ear height, suggesting that it has potential for
Potential application of qph1 in maize improvement
Cereal production went through a dramatic increase due to
the adoption of short cultivars during the last century, known
as the Green Revolution. Underlying genes were later isolated
in rice, wheat, sorghum and several other crops (Galweiler
et al., 1998). Obvious defects in major genes were found to be
responsible for phenotypic variation in most cases, and thus
made them easily manipulated for practical use. In maize,
Fig. 8. Protein structure simulation of qph1 generated using Pymol. The
six active sites of the protein are shown in yellow; the Arginine to Leucine
amino acid substitution on the ninth α-helix in the transmembrane domain
is indicated in blue. TMs, transmembrane domains; NBDs, nucleotide
however, loss of function mutations of the major plant height
genes often led to serious defects and very large yield loss
(Winkler et al., 1995; Thornsberry et al., 2001), so
moderateeffect QTLs were considered to be excellent alternatives. QTL
mapping has long been conducted to localize maize plant
height regulators with desirable effects, but has rarely resulted
in candidate gene cloning. Several maize genes underlying
quantitative traits have been cloned and validated based on
linkage analysis; examples include Vgt1, Tga1, and
DGAT12 (Wang et al., 2005; Salvi et al., 2007; Zheng et al., 2008),
but plant height QTLs were rarely cloned (Teng et al, 2012).
Among the plant height factors identified in maize so far, the
recessive br2 gene is considered to have great potential and
efforts have been made to use it practically (Anderson and
Chow, 1960; Djisbar and Brewabaker, 1987). Introgression
of br2 into normal varieties could reduce plant height and
ear height by shortening each internode (Souza and Zinsly,
1985), but unfortunately all the recessive br2 alleles
identified so far cause severe phenotypes and it has not been
possible to use them in breeding. Here, we provide the detailed
phenotypic and molecular characterization of the naturally
occurring mild allele of br2, qph1; it is a very rare SNP
mutation that might have occurred recently and hasn’t been widely
used in breeding programs. The cloning of qph1 sheds more
light on the molecular nature of natural variation at maize
QTLs; it demonstrates that the naturally occurring allele at a
QTL locus and a strong dwarf mutant are genetic variants of
the same gene. Results of this study revealed qph1 as a major
plant height QTL that has a moderate effect on plant height
and no or minimal negative effects on grain yield under
various genetic backgrounds tested, suggesting its potential in
maize improvement by marker-assisted selection for reduced
plant height and lodging resistance.
Supplementary data can be found at JXB online.
Supplementary Table S1. Cytology analysis of the
RIL88(qph1) and RIL88(QPH1) stem tissues.
Supplementary Table S2. Association analysis of the five
SNPs within the target region of qph1 in a population of 500
Supplementary Table S3. Plant height and ear height
analysis of the four F2 populations.
Supplementary Table S4. Yield-related trait analysis of
single-cross hybrids derived from RIL88(qph1) and RIL88(QPH1).
Supplementary Table S5. Primers used in this study.
Supplementary Figure S1. Construction of qph1
fine-mapping population BC6F2 and near-isogenic lines RIL88(QPH1)
Supplementary Figure S2. Fine mapping of qph1 in BC4F2
Supplementary Figure S3. Phenotypic variation and
distribution of the yield-related traits in BC4F2:3.
Supplementary Figure S4. Stalk internode length variation
between RIL88(qph1) and RIL88(QPH1).
Supplementary Figure S5. Scanning electron microscopy
examination of the sixth internodes from RIL88(qph1) and
Supplementary Figure S6. Scanning electron microscopy
examination of the uppermost internodes from RIL88(qph1)
Supplementary Figure S7. Allelism test of qph1.
Supplementary Figure S8. Subcellular localization of
Supplementary Figure S9. qph1 expression in RIL88(qph1)
and RIL88(QPH1) internodes from three developmental stages
Supplementary Figure S10. Plant height and yield
performance comparison between individuals of different
genotypes in the B73 × Ye107 F2 population.
Supplementary Figure S11. Plant height and ear height
segregation in nine F2 populations of N546 conversion lines
and normal inbred lines.
Supplementary Figure S12. Yield component analysis of
the four F2 populations.
We thank Dr Kevin Fengler for helping with BAC AC210610 and providing
the physical locations of genetic markers; and Dr J. Antoni Rafalski and Dr
Robert L. Last for editing the paper. We also thank Marty Sachs for
providing the four br2 mutants.
Aller SG, Yu J, Ward A, et al. 2009. Structure of P-glycoprotein reveals a
molecular basis for poly-specific drug binding. Science 323, 1718–1722.
Anderson JC, Chow PN. 1960. Phenotypes and grain yield associated
with br2 gene in single cross hybrids of dent corn. Crop Science 1,
Chang G, Roth CB. 2006. Structure of MsbA from E. coli: A homolog of
the multidrug resistance ATP binding cassette (ABC) transporters. Science
Dellaporta SL, Wood J, Hicks JB. 1983. A plant DNA minipreparation:
version II. Plant Molecular Biology Reporter 1, 19–21.
Djisbar A, Brewabaker JL. 1987. Effects of the brachytic-2 gene on
maize yield and its components. Maydica 32, 107–123.
Duvick DN. 2005. Genetic progress in yield of United States maize (Zea
mays L.). Maydica 50, 193–202.
Friml J, Palme K. 2002. Polar auxin transport – old questions and new
concepts? Plant Molecular Biology 49, 273–284.
Galweiler L, Guan C, Muller A, Wisman E, Mendgen K, Yephremov
A, Palme K. 1998. Regulation of polar auxin transport by AtPIN1 in
Arabidopsis vascular tissue. Science 282, 2226–2230.
Ganal MW, Durstewitz G, Polley A, Bérard A, Buckler ES,
Charcosset A, Joseph D, Clarke JD, Graner E, Hansen M. 2011.
A large maize (Zea mays L.) SNP genotyping array: development and
germplasm genotyping, and genetic mapping to compare with the B73
reference genome. PLoS ONE . 6, e28334.
Souza Jr CL, Zinsly JR. 1985. Relative genetic potential of brachytic
maize (Zea mays L.) varieties as breeding populations. Brazilian Journal of
Genetics 3, 523–533.
Funding This work was supported by a grant from the National High Technology Research and Development Program of China (863 Program , No.
2012AA10A307) Gao S , Martinez C , Skinner DJ , Krivanek AF , Crouch JH , Xu Y .
2008. Development of a seed DNA-based genotyping system for markerassisted selection in maize . Molecular Breeding 22 , 477 - 494 .
Hashemi A , Herbert SJ , Putnam DH . 2005 . Yield response of corn to crowding stress . Agronomy Journal 97 , 839 - 846 .
Johnson BE , Rodríguez-Herrera SA , Donald H. 1998 . A parental line of maize . Crop Science 38 , 574 - 574 .
Leng ER , Vineyard ML . 1951 . Dwarf and short plants . Maize Genetics Cooperation Newsletter 25 , 31 - 32 .
2007. LAZY1 controls rice shoot gravitropism through regulating polar auxin transport . Cell Research . 17 , 402 - 410 .
Liang BC , Remillard M , Mackenzie HF . 1992 . Effects of hybrids, population densities, fertilization and irrigation on grain corn (Zea mays L.) in Qubec. Canadian Journal of Plant Science 72 , 1163 - 1170 .
Livak KJ , Schmittren TD . 2001 . Analysis of relative gene expression data using real-time quantitative PCR and the 2- ΔΔCT Method. Methods 25 , 402 - 408 .
Ljung K , Ostin A , Lioussanne L , Sandberg G. 2001 . Developmental regulation of indole-3-acetic acid turnover in Scots pine seedlings . Plant Physiology . 125 , 464 - 475 .
Ma X , Tang J , Teng W , Yan J , Meng Y , Li J. 2007 . Epistatic interaction is an important genetic basis of grain yield and its components in maize.
Molecular Breeding 20 , 41 - 51 .
Multani DS , Briggs SP , Chamberlin MA , Blakeslee JJ , Murphy AS , Johal GS . 2003 . Loss of an MDR transporter in compact stalks of maize br2 and sorghum dw3 mutants . Science 302 , 81 - 84 .
Noh B , Murphy AS , Spalding EP . 2001 . Multidrug resistance-like genes of Arabidopsis required for auxin transport and auxin-mediated development . Plant Cell . 13 , 2441 - 2454 .
Park KY , Park SU , Moon HG . 1989 . Effects of planting density and tiller removal on growth and yield of sweet corn hybrids . Korean Journal of Crop Science 34 , 192 - 197 .
Peng JR , Richards DE , Hartley NM , Murphy GP , Devos KM , Flintham JE , Beales J , Fish LJ , Worland AJ , Pelica F. 1999 . “ Green revolution” genes encode mutant gibberellin response modulators . Nature 400 , 256 - 261 .
Salvi S , Sponza G , Morgante M , Tomes D , Niu X , Fengler KA , Meeley R , Ananiev EV , Svitashev S , Bruggemann E. 2007 . Conserved noncoding genomic sequences controlling flowering time differences in maize.
Proceedings of the National Academy of Sciences , USA 104 , 11376 - 11381 .
Sasaki A , Ashikari M , Ueguchi-Tanaka M , et al. 2002 . Green revolution: A mutant gibberellin-synthesis gene in rice . Nature 416 , 701 - 702 .
Scott GE , Campbell CM . 1969 . Internode length and brachytic-2 maize inbreds in single cross . Crop Science 9 , 293 - 295 .
Tang J , Teng W , Yan J , Ma X , Meng Y , Li J. 2007a. Genetic analysis of plant height using a set of recombinant inbred line population in maize.
Euphytica 155 , 117 - 124 .
Tang J , Ma X , Teng W , Yan J , Wu W , Dai J , Li J. 2007b. Detection of quantitative trait loci and heterotic loci for plant height using an immortalized F2 population in maize . Chinese Science Bulletin 52 , 477 - 438 .
Tang J , Yan J , Ma X , Teng W , Wu W , Dai J , Dhillon BS , Melchinger AE , Li J. 2010 . Dissection of the genetic basis of heterosis in an elite maize hybrid by QTL mapping in an immortalized F2 population .
Theoretical and Applied Genetics 120 , 333 - 340 .
Teng F , Zhai L , Liu R , Bai W , Wang L , Huo D , Tao Y , Zheng Y , Zhang Z. 2012 . ZmGA3ox2, a candidate gene for a major QTL, qPH3.1, for plant height in maize . The Plant Journal 73 , 405 -416 Thornsberry JM , Goodman MM , Doebley J , Kresovich S , Nielsen D , Buckler ES . 2001 . Dwarf8 polymorphisms associate with variation in flowering time . Nature Genetics 28 , 286 - 289 .
Vafias B , Ispalandis CG , Goulas C , Deligeorgidis PN . 2006 . An approach on yielding performance in maize under varying plant densities .
Asian Journal of Plant Science 5 , 690 - 694 .
Vega CRC , Andrade FH , Sadras VO . 2001 . Reproductive partitioning and seed set efficiency in soybean, sunflower and maize . Field Crops Research 72 , 165 - 173 .
Wang H , Nussbaum-Wagler T , Li B , Zhao Q , Vigouroux Y , Faller M , Bomblies K , Lukens L , Doebley JF . 2005 . The origin of the naked grains of maize . Nature 436 , 714 - 719 .
Winkler RG , Helentjaris T. 1995 . The maize Dwarf3 gene encodes a cytochrome P450-mediated early step in Gibberellin biosynthesis . The Plant Cell 7 , 1307 - 1317 .
Xue W , Xing Y , Zhao Y , Tang W , Wang L , Zhou H , Yu S , Xu C , Li X , Zhang Q. 2008 . Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice . Nature Genetics 40 , 761 - 767 .
Yan J , Tang H , Huang Y , Shi Y , Li J , Zheng Y. 2003 . QTL mapping for developmental behavior for plant height in maize . Chinese Science Bulletin 48 , 2601 - 2607 .
Yan J , Tang H , Huang Y , Zheng Y , Li J. 2006 . Quantitative trait loci mapping and epistatic analysis for grain yield and yield components using molecular markers with an elite maize hybrid . Euphytica 149 , 121 - 131 .
Yan WH , Wang P , Chen HX et al. 2011 . A major QTL, Ghd8, plays pleiotropic roles in regulating grain productivity, plant height, and heading date in rice . Molecular Plant 4 , 319 - 330 .
2010a. Dry matter production and photosynthesis characteristics of three hybrids of maize (Zea mays L.) with super-high-yielding potential . Acta Agronomica Sinica 37 , 355 - 361 .
Yang X , Yan J , Shah T , Warbuton ML , Li Q , Li L , Gao Y , Chai Y , Fu Z , Zhou Y. 2010 . Genetic analysis and characterization of a new maize association mapping panel for quantitative trait loci dissection . Theoretical and Applied Genetics 121 , 417 - 431 .
Ye L , Liu L , Xing A , Kang D. 2013 . Characterization of a dwarf mutant allele of Arabidopsis MDR-like ABC transporter AtPGP1 gene . Molecular Cell Biology Research Communications 441 , 782 - 786 .
Zhao Y. 2010 . Auxin biosynthesis and its role in plant development .
Annual Review of Plant Biology 61 , 49 - 64 .
Zheng P , Allen WB , Roesler K , Williams ME , Zhang S , Li J , Glassman K , Ranch J , Nubel D , Solawetz W. 2008 . A phenylalanine in DGAT is a key determinant of oil content and composition in maize . Nature Genetics 40 , 367 - 372 .