PLANT SCIENCE
TaCol-B5modifies spike architecture and enhances
grain yield in wheat
Xiaoyu Zhang^1 †, Haiyan Jia1,2†, Tian Li1,3†, Jizhong Wu1,4, Ragupathi Nagarajan^1 , Lei Lei^1 ,
Carol Powers^1 , Chia-Cheng Kan^1 , Wei Hua^5 , Zhiyong Liu^6 , Charles Chen^7 ,
Brett F. Carver^1 , Liuling Yan^1
Spike architecture influences grain yield in wheat. We report the map-based cloning of a gene
determining the number of spikelet nodes per spike in common wheat. The cloned gene is namedTaCOL-B5
and encodes a CONSTANS-like protein that is orthologous toCOL5in plant species. Constitutive overexpression
of the dominantTaCol-B5allele but without the region encoding B-boxes in a common wheat cultivar
increases the number of spikelet nodes per spike and produces more tillers and spikes, thereby enhancing
grain yield in transgenic plants under field conditions. Allelic variation inTaCOL-B5results in amino acid
substitutions leading to differential protein phosphorylation by the protein kinaseTaK4. TheTaCol-B5
allele is present in emmer wheat but is rare in a global collection of modern wheat cultivars.
C
ommon wheat (Triticum aestivum,2n=
6 x= 42, AABBDD genome) grain yields
are influenced by three major compo-
nents: spikes per unit land area, grains per
spike, and grain weight ( 1 ). An increase in
any one of these components can improve grain
yield. The number of spikes can be increased
through promotion of tillering, as fertile tillers
eventually form spikes ( 2 , 3 ). The number of
grains per spike can be physically and genet-
ically dissected into two subcomponents: spike-
lets per spike and grains per spikelet ( 2 , 3 ). A
normal spike can generate between 16 and 25
spikelet nodes, and within a spikelet, grains at
the first and second positions are larger than
those at the third, fourth, or higher positions
( 4 – 6 ). Therefore, understanding spikelet devel-
opmental patterns and generating more spikelet
nodes per spike (hereafter referred to as SNS)
increases grain number without decreasing
average grain weight ( 7 ). The SNS trait is ge-
netically controlled in any given wheat cultivar
( 4 , 5 ); however, the genetic basis of spikelet
development is largely unknown. In this study,
we mapped a quantitative trait locus (QTL) for
SNS and then cloned the gene responsible
for the QTL. We found that the cloned gene
increased both SNS and spike number and
further increased field-based grain yield in
transgenic wheat.
We initially performed a single cross between
two common wheat cultivars, CItr 17600 and
Yangmai18, which have different spike morphol-
ogies (fig. S1, A and B). A population of 186
F 2 plants was genotyped using the genotyping-
by-sequencing (GBS) approach (table S1), and
the population of F 2 -derived F 3 (F2:3) lines was
phenotyped under field conditions. On the basis
of single-year phenotypic data, a potential major
QTL associated with SNS was mapped to
chromosome 7B (hereafter calledQSns.osu-7B),
having a log of the odds (LOD) value of 15.3 and
accounting for 43% of the total phenotypic
variation in the field-tested population (Fig. 1A).
To cloneQSns.osu-7B, we screened 1857 indi-
vidual F 5 plants derived from a single F 3 plant,
WF112 (fig. S1, C and D), and identified 21 F 5
recombinant plants using two flanking mar-
kers (fig. S2). We also developed simple se-
quence repeat (SSR) and single-nucleotide
polymorphism (SNP) markers (fig. S3) for
fine mapping of the recombinant plants. We
determined the genotypes and phenotypes of
four F 6 populations derived from the recom-
binant F 5 plants (Fig. 1, B to F, and fig. S2). The
gene responsible forQSns.osu-7Bwas flanked
by two markers, SNS-M1 and SNS-G2M3,
which spanned a genomic region of 318,786
base pairs (bp) encompassing two genes,
TraesCS7B02G400600andTraesCS7B02G400700,
according to International Wheat Genome
Sequencing Consortium (IWGSC) RefSeq v2.1
sequences (Fig. 1B).
Next, we focused on allelic variation in the
targeted region sequences (figs. S4 and S5)
and concluded thatTraesCS7B02G400600is
most likely the gene responsible forQSns.osu-7B.
TraesCS7B02G400600encodes a CONSTANS-
like (COL) protein and is orthologous toCOL5
in plant species. We therefore named this
wheat geneTaCOL-B5.We observed dom-
inant effects ofTaCol-B5, representing theCItr 17600 allele, overTacol-B5, represent-
ing the Yangmai18 allele, on SNS and spike
length(fig.S2).Wealsoobserved10SNPs
along the sequenced 2014-bp region between
the two alleles (fig. S4). We validated the func-
tions ofTaCol-B5using a transgenic approach
in wheat.
We transformed Yangmai18 with the cloned
cDNA ofTaCol-B5from CItr 17600 and ob-
tained four independent transgenic events
(T 0 plants) that showed changed phenotypes
in the T 1 generation (Fig. 2A and fig. S6, A to D).
We confirmed the overexpression of transgenic
TaCol-B5in the four independent transgenic
events using quantitative real-time polymerase
chain reaction (qRT-PCR) (fig. S6E). Addi-
tionally, we observed the expression of both
transgenicTaCol-B5and nativeTacol-B5in
the same spike sample of the transgenic plants
(fig. S6F). In the greenhouse, the transgenic
plants, averaged across the four transgenic T 1
families expressingTaCol-B5, produced 3.5
more spikelet nodes per spike (Fig. 2B) and
3.4 more grains per spike (Fig. 2C) than did
nontransgenic plants. Further, overexpression
ofTaCol-B5promoted tillering, resulting in
an additional 1.3 spikes per plant and higher
single-plant productivity (fig. S7). This observed
increase in single-plant productivity (fig. S7)
led us to test the transgenic wheat plants under
field conditions.
First, we tested the effects ofTaCol-B5
in T 2 transgenic plants at a reduced seeding
rate (40 plants/m^2 ), owing to limited grain
availability (Fig. 2D). In comparison with
nontransgenic plants, the transgenic plants
produced larger and longer spikes, showed
similar effects in all spikes (fig. S8, A to H), and
generated longer but narrower grains (fig. S8,
I to K). Averaged across the four T 2 families,
the transgenic plants set an additional 2.4
spikelet nodes per spike (Fig. 2E) and in-
creased spike length by 7.3 cm over the non-
transgenic plants (Fig. 2F). Spikelet density
was therefore lower in transgenic plants, 1.3
spikelets/cm versus 1.9 spikelets/cm in non-
transgenic plants, which could contribute to
greater per-plant productivity ( 8 ). The trans-
genic plants produced an additional 8.1 grains
per spike (Fig. 2G) and 2.3 spikes per plant
(Fig. 2H), with no compensatory loss in
thousand grain weight (fig. S9A). The signif-
icant increases in single-plant productivity
(fig. S9B) and single-row-plot grain yield
(fig. S9C) in the presence ofTaCol-B5(table
S2) led us to further investigate its effect on
grain yield following standard wheat yield
trial procedures.
We analyzed the genetic effects ofTaCol-B5
in four T 3 transgenic lines at a higher seeding
rate (130 plants/m^2 )inthefield,eachina6-m^2
plot with three replicates. The phenotypes of
theTaCol-B5transgenic plants were stable at
the population level (Fig. 2, I and J). Compared180 8 APRIL 2022•VOL 376 ISSUE 6589 science.orgSCIENCE
(^1) Department of Plant and Soil Sciences, Oklahoma State
University, Stillwater, OK 74078, USA.^2 The Applied Plant
Genomics Laboratory, National Key Laboratory of Crop
Genetics and Germplasm Enhancement, Nanjing Agricultural
University, Nanjing 210095, Jiangsu, China.^3 Key Laboratory
of Crop Gene Resources and Germplasm Enhancement,
Institute of Crop Sciences, Chinese Academy of Agricultural
Sciences, Beijing 100081, China.^4 Institute of Germplasm
Resources and Biotechnology, Jiangsu Academy of
Agricultural Sciences, Nanjing 210014, Jiangsu, China.
(^5) Zhejiang Academy of Agricultural Sciences, Hangzhou
310021, China.^6 Institute of Genetics and Developmental
Biology, Chinese Academy of Sciences, Beijing 100101,
China.^7 Department of Biochemistry and Molecular Biology,
Oklahoma State University, Stillwater, OK 74078, USA.
*Corresponding author. Email: [email protected] (L.Y.);
[email protected] (B.F.C.)
These authors contributed equally to this work.
RESEARCH | REPORTS