The Plant Genome o r i g i n a l r es e a rc h
Identification of the First Nuclear Male Sterility Gene
(Male-sterile 9) in Sorghum
Junping Chen,* Yinping Jiao, Haydee Laza, Paxton Payton, Doreen Ware, and Zhanguo Xin
J. Chen, H. Laza, P. Payton, Z. Xin, Plant Stress and Germplasm Development Unit, USDA–ARS, Lubbock, TX 79415; Y. Jiao, D. Ware, Cold
Spring Harbor Lab., Cold Spring Harbor, NY; D. Ware, Plant, Soil and Nutrition Lab. Research Unit, USDA–ARS, Cornell Univ., Ithaca, NY.
ABSTRACT Nuclear male sterility (NMS) is important for
understanding microspore development and could facilitate the
development of new strategies to control male sterility. Several
NMS lines and mutants have been reported in sorghum [Sorghum
bicolor (L.) Moench] previously. However, no male-sterile gene
has been identified, hampering the utility of NMS in sorghum
breeding. In this study, we characterized a new NMS mutant,
male sterile 9 (ms9), which is distinct from all other reported NMS
loci. The ms9 mutant is stable under a variety of environmental
conditions. Homozygous ms9 plants produced normal ovaries
but small pale-colored anthers that contained no pollen grains.
Microscopic analyses revealed abnormal microspore development
of ms9 at the midmicrospore stage, causing degeneration of
microspore inside the anther lobes and male sterility of ms9
plants. Using MutMap, we identified the Ms9 gene as a plant
homeotic domain (PHD)-finger transcription factor similar to Ms1
in Arabidopsis thaliana (L.) Heynh. and Ptc1 in rice (Oryza sativa
L.). Ms9 is the first NMS gene identified in sorghum. Thus, the
Ms9 gene and ms9 mutant provide new genetic tools for studying
pollen development and controlling male sterility in sorghum.
Abbreviations: CMS, cytoplasmic male sterility; LBK, USDA–ARS Cropping
Systems Research Laboratory in Lubbock, TX; NMS, nuclear male sterility;
PHD, plant homeotic domain; PR, winter nursery farm in Guayanilla, PR; SEM,
scanning electron microscopy; WT, wild-type.
the pl ant genome  vol . 12, no . 3  november 2019 1
core ideas
• The male-sterile 9 (ms9) is a novel nuclear male-sterile
mutant in sorghum.
• The Ms9 gene encodes a PHD-finger transcription
factor critical for pollen development.
• The identification of the Ms9 gene provides a strategy
to control male sterility in sorghum.
S orghum is a major grain crop used for human con-
sumption and animal feed as well as a promising
bioenergy crop for sugar, biomass, and biofuel produc-
tion (de Siqueira Ferreira et al., 2013; Rooney et al., 2007).
Like maize (Zea mays L.), sorghum is a diploid C4 plant
that, in contrast to maize, is well adapted to drought-
prone and high-temperature environments and can
thrive on marginal soils (Morris et al., 2013). Sorghum
also has a much smaller genome than maize (~800 vs.
2500 Mb), and after the recent completion of a high-
quality diploid genome sequence, sorghum has become
an emerging model for highly productive C4 crops
(McCormick et al., 2018; Paterson et al., 2009).
Heterosis, or hybrid vigor, the ability of hybrids to
outperform the best inbred line parents is probably the
most important strategy to increase grain yield in many
crops including sorghum (Kim and Zhang, 2018). A criti-
cal requirement of using hybrids to increase yield is the
ability to produce a pure male-sterile female parent that
Chen, J., Y. Jiao, H. Laza, P. Payton, D. Ware, and Z. Xin.
2019. Identification of the first nuclear male sterility gene (Male-
sterile 9) in sorghum. Plant Genome 12:190020. doi: 10.3835/
plantgenome2019.03.0020
Received 19 Mar. 2019. Accepted 3 May 2019.
*Corresponding author (junping.chen@ars.usda.gov).
© 2019 The Author(s). This is an open access article distributed under
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-
nd/4.0/).
of 12

can be used for production large quantities of hybrid
seeds by cross-pollination with a fertile male parent.
At present, cytoplasmic male sterility (CMS) is the pre-
dominant strategy to produce male-sterile parents in
most crops for which mass production of hybrid seed is
possible. Most grain sorghum varieties currently used
in agricultural production are hybrids. The A1 CMS
reported in the 1950s (Stephens and Holland, 1954)
remains the predominant CMS line used for producing
hybrids in sorghum, although other types of CMS lines
are available (Jordan et al., 2011). The CMS breeding sys-
tem is a three-line system in which the CMS female par-
ent A line is pollinated by a maintainer B line to maintain
cytoplasmic male sterility of the A line, or by a fertility
restorer R line to produce a fertile hybrid (Rooney, 2004).
The three-line breeding system requires a perfect B line
for each A–B pair to maintain the absolute male steril-
ity of the A line. It also requires a perfect R line that can
fully restore the fertility of the hybrid plants by comple-
menting the cytoplasmic defect of the A line (Jordan et
al., 2011). Simultaneous development of all three lines is
time consuming, expensive, and complicated. Moreover,
a significant amount of sorghum germplasm, especially
bioenergy sorghums, is neither characteristic B line nor
R line, and therefore, does not meet the strict require-
ments for the three-line sorghum hybrid breeding system.
The limitation in the number of accessions that can be
used in hybrid breeding prevents sorghum breeders from
creating all possible hybrids to exploit heterosis. Further-
more, since almost all current commercial grain sorghum
hybrids are produced using the A1 CMS, the homogeneity
of A1 cytoplasm could predispose sorghum hybrids prone
to severe diseases and other biotic and abiotic stresses.
One example was the widespread use of the T-cytoplasm
maize in 1960s and in early 1970s. The devastating epi-
demic of southern corn leaf blight disease caused by race
T of Bipolaris maydis caused over 80% yield loss in maize
production in 1970 (Ullstrup, 1972). To avoid catastrophic
yield losses similar to those that occurred in T-cytoplasm
maize, it is necessary to develop new sorghum hybrids by
using a variety of sources of male sterility.
Recently, two-line hybrid breeding systems using
NMS have been explored in rice (Chang et al., 2016).
The two-line breeding system not only simplifies the
hybrid seed production procedure but also dramatically
expands the possibilities for making hybrids between
accessions that cannot be generated with the current
three-line breeding system (Dong et al., 2000; Zhang
et al., 2010, 2013). Furthermore, hybrids made with the
two-line breeding system have a 5 to 10% higher grain
yield over similar hybrids produced using a CMS three-
line breeding system (Chang et al., 2016; Kim and Zhang,
2018; Zhou et al., 2014). Therefore, the two-line breeding
system for hybrid production of major crops holds great
potential for increasing future agricultural production.
The identification of nuclear genes involved in regulating
male sterility in crops is the first and most essential step
in the development of a two-line breeding system.
2 of 12 the pl ant genome
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
In sorghum, eight NMS lines, ms1 through ms8, and
msal (antherless) have been reported (Andrews and Web-
ster, 1971; Xin et al., 2017) previously. The ms4, ms5, and
ms6 lines are no longer available, but the other lines have
been preserved and introduced into different genetic back-
grounds (Pedersen and Toy, 2001). These NMS lines could
make important contributions to the elucidation of male
gametophyte development and aid in the development of
new breeding systems for making sorghum hybrids. How-
ever, to date, none of the genes that mediate male sterility
in these lines has been identified, which greatly hinders
the potential of using NMS in sorghum breeding.
In this study, we reported the characterization of a
new NMS mutant isolated from an ethyl methane sul-
fonate–mutagenized mutant population in the sorghum
inbred line BTx623 (Jiao et al., 2016; Xin et al., 2009) and
identification of the first NMS gene in sorghum. Since the
result of our genetic study indicated that the ms mutant
represents a new locus for NMS, we named it as male
sterile 9 (ms9) and the casual gene as Ms9. Morphologi-
cal analyses of ms9 flowers and spikelets revealed that
the ovary of ms9 develops normally and sets seed after
manual pollination with wild-type (WT) pollen. Detailed
light and electron microscopic analyses revealed that
the male-sterile phenotype in ms9 is caused by a defect
in pollen development that occurs at microspore stage
in the anther. The microspores inside the developing
anthers begin to degenerate prematurely at the midvacu-
olated stage, leading to small, pale-colored anthers with
no pollen grains at anthesis. The causal gene mutation for
male sterility in ms9, identified using MutMap, encodes
a PHD-finger transcription factor critical for tapetum
degeneration and pollen formation. The Ms9 gene and its
mutants represent new genetic tools for dissecting pollen
development and controlling male sterility in sorghum.
MATERIALS AND METHODS
Growing Condition and Management
All plant materials used in this study were planted annu-
ally on the research farm of the USDA–ARS Cropping
Systems Research Laboratory in Lubbock (LBK), TX (33°
35¢ N, 101° 53¢ W, 958 m asl), during the normal growing
season (May–September) and in a winter nursery farm in
Guayanilla, PR (PR) (18.0373° N, 66.7963° W, 49 m asl)
during the winter months (December–April). Soil type
was Amarillo (fine-loamy, mixed, superactive, thermic
Aridic Paleustalfs) fine sandy loam at LBK and Caguabo
(loamy, mixed, active, isohyperthermic, shallow Typic
Eutrudepts) well-drained loamy soil at PR. The plot size
at both locations was 4.6 m long. For genetic analysis, the
mutant and F1 seeds were planted in a single plot, and the
F2 seeds (segregating population) were planted in eight
plots. Approximately 60 sorghum seeds were planted per
plot using a John Deere MaxEmerge planter at a planting
depth of 3 cm. Plots were furrow irrigated 1 wk before
planting to provide saturated soil moisture for optimal
seed germination. Two weeks after germination, the field
 vol . 12, no . 3  november 2019

was irrigated as needed by an automated subsurface drip
system at LBK and surface drip lines at PR. Routine farm
practices were applied with regard to fertilizer, pest con-
trol, and other field management.
The environmental conditions at the two field loca-
tions differed greatly. The weather during the summer
growth season at LBK is typically hot and dry with spo-
radic occurrences of heat-wave events (Supplemental Fig.
S1) and relative low humidity (~50% on average), whereas
the tropical weathers in the winter nursery in PR are mild
with high humidity (~75% on average) and much less
temperature fluctuation (Supplemental Fig. S1). The pho-
toperiod (from sun rise to sun set) in winter times in PR
is short day with an average daylength of 12 h, while the
photoperiod during the summer growth season at LBK is
long day (13.5–14.3 h). The greenhouse test was conducted
throughout of the year at LBK with temperatures set at
28°C during the day and 23°C at night and a 14 h light–10
h dark cycle with ambient light and supplemental light.
Isolation of ms9 Mutant
The ms9 mutant was identified in 2015 at LBK from an
M4 mutant line, M2P0110 (ARS178). This mutant was
one of the 256 sorghum mutant lines that were sequenced
and annotated for gene mutations (Jiao et al., 2016). After
identifying potential sterility phenotype of the main
shoot at anthesis, flowered panicle of that particular
mutant plant was bagged immediately, and a line below
the flowering zone was marked on the sorghum bag to
indicate the separation of the open-pollinated part of the
panicle from the self-pollinated part. The seed set of the
main panicle was examined 2 wk after bagging. Sterility
on the main panicle was confirmed by having seed set in
the open-pollinated zone and the absence of seed set in
the part that flowered after being bagged. Approximately
10 d after the main panicle flowered, panicles from two
tillers were bagged prior to anthesis. One tiller panicle
was bagged for 45 d until harvest and set no seed. The
other bagged tiller panicle was checked daily for flower-
ing, pollinated with inbred line BTx623 WT pollen on
the third day of anthesis (~50% of the sessile spikelets
flowered), and continually bagged for 45 d after pollina-
tion. Additionally, another independent cross was made
using a tiller panicle from a different sterile plant identi-
fied from the same plot as M2P0110. The seeds produced
(F1) from the two independent ms plants crossed by the
original WT inbred line BTx623 were harvested at matu-
ration and used for genetic analysis.
Genetic Analysis of the Male Sterile Phenotype of the
ms9 Mutant
The generations notations used for the genetic analysis
of ms9 in this study was in accordance with the standard
breeding terms. The F1 seeds refer to the seeds produced
on the original ms mutant plants crossed by the WT
BTx623 pollen. The F1 plants produced F2 seeds through
self-fertilization. The BC1F1 seeds were those produced
on male-sterile plant in the F2 segregating population
chen et al . 3
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
pollinated with BTx623. Self-fertilization of BC1F1 plants
produced BC1F2 seeds. BC2F1 and BC2F2 were pro-
duced by pollination of male-sterile BC1F2 by BTx623
pollen and self-fertilization of BC2F1, respectively.
For genetic analysis, both the F1 and open-pollinated
seeds, along with the same M4 mutant seeds, from which
the ms plant was identified, were planted in the PR winter
nursery in 2015 to examine the nature of genetic control
of male sterility in the F1 generation and to confirm the
sterility phenotype in the same M4 line. The F1 plants
were self-fertilized, the resultant F2 seeds were planted at
LBK in the 2016 growing season. Segregation for fertile
and sterile phenotypes was analyzed in two F2 popula-
tions. We used three approaches to further examine the
nature of the sterility in the F2 generation. First, at anthe-
sis, we marked a subset of F2 sterile plants and let them
open-pollinate in the field for 1, 2, or 3 d before bagging
the panicles. Second, we bagged another subset of F2 pani-
cles prior to anthesis to produce completely self-pollinated
panicles. Third, on the first day of anthesis, we bagged the
ms panicle after removing the top, manually pollinated
it with BTx623 pollen 4 d later, and then bagged it again
until maturation to check the seed set. Seeds produced
from two manually pollinated F2ms plants in the third
approach (BC1F1 seeds) were harvested and planted in
PR in the winter of 2016 to produce self-pollinated BC1F2
seeds. The BC1F2ms was backcrossed again with BTx623
pollen in the 2017 season at LBK, and the resultant BC2F1
were grown at PR in the winter and self-pollinated to pro-
duce the BC2F2 seeds. The field-grown BC2F2 fertile and
sterile plants as well as the BTx623 WT plants were used
for phenotypical and morphological characterization of
the M2P0110ms mutant in 2018.
To determine the genetic relationship of ms9 to
previous reported male-sterile lines in sorghum, we
manually pollinated the male-sterile plants from other
available NMS lines (ms1, ms2, ms3, ms7, msal, and ms8)
with pollen collected from the heterozygous Ms9 fertile
BC2F1 plants. The resulting complement-crossing F1
seeds (cpF1) were planted in the field and then subse-
quently examined for the male sterility phenotype of
cpF1 plants during flowering time. If ms9 were allelic to
any of these NMS lines, we would expect to observe a 1:1
fertile to sterile segregation ratio among the cpF1 prog-
enies. For an NMS locus that is distinct from Ms9, we
would expect all cpF1 plants to be fertile.
Examination of Female Fertility
The development and fertility of the female floral organs
of the BC2F2 of ms9 plants were examined and compared
with those of BTx623 WT plants. Specifically, on the day
of flowering, ovaries from these plants were dissected from
freshly flowered spikelets and compared morphologically.
The top flowering parts of BTx623 and ms9 BC2F2ms pan-
icles were removed, and the remaining nonflowering parts
of the panicle were bagged. The next day, ~4 h into the
photoperiod, the flowering zones of the bagged panicles
were marked for both BTx623 and ms9 panicles and were
of 12

manually pollinated with BTx623 pollen. This process was
repeated one more time on the following day. The develop-
ment of ovaries from spikelets of the pollinated zones at 0
and 4 d after pollination was examined and photographed
using a Leica MZ6 stereomicroscope equipped with a
Leica Fire Cam program (Microsystems Inc.).
Examination of Anther and Pollen Development
by Staining
Pollen viability was examined by alexander staining
method as described previously (Huang et al., 2016; Zhao
et al., 2002). Briefly, spikelets containing anthers were
excised from the sorghum panicle prior to dehiscence
and immediately submerged into Carnoy’s fixative solu-
tion in glass tubes (60 mL ethanol; 30 mL chloroform;
10 mL acetic acid). The tissues were vacuum-infiltrated
for 15 min and fixed overnight at room temperature.
Anthers from the fixed spikelet tissues were excised,
rinsed in a series of increasing percentage ethanol solu-
tions, placed in Alexander staining buffer (Xin et al.,
2017), and incubated at 37°C for 2 h. The stained anthers
were rinsed several times with staining solution without
dye to remove the free dye and then mounted on a glass
slide. The morphological characteristics of the stained
anthers and pollen grains were examined and photo-
graphed using an Olympus BX60 microscope equipped
with an Olympus DP 80 digital camera.
Light and Scanning Electron Microscopic Analyses
of Anther Development
Cross-sections of flowers and individual anthers were
obtained by paraffin embedding and CRYO-scanning
electron microscopy (SEM) procedures. Whole-flower
samples collected at early developmental stages (prior
to booting) were fixed in FAA (contains 95% ethanol,
water, 37% formaldehyde, and acetic acid at a ratio of
50:35:10:5) at room temperature for 24 h. The fixed
samples were stored for 1 wk at 4°C, followed by dehy-
dration at room temperature in an ethanol concentration
series (from 30 to 100%, increasing concentration by 10%
per hour), and then by 1 h incubation in 100% xylene.
The total dehydration time was 24 to 30 h depending on
the tissue stage. Dehydrated flower tissues were infil-
trated (12–24 h) and embedded in 100% Paraplast Plus
(Sigma-Aldrich) at 56 to 60°C. Wax blocks were cut into
5- to 10-μm sections using a Leica RM2125 RTS (Leica
Biosystems) manual rotatory microtome. Slides contain-
ing sections were submerged three times in 100% xylene
for 5 min followed by rehydration in a series of ethanol
solutions (100% two times, and 95, 80, and 70% for 3
min each). Cross-sections were then stained with 0.1%
toluidine blue or 0.01% Safranin O solution. Flower and
pollen images were photographed and analyzed at 10 to
40´ magnification on an Olympus BX60 microscope
equipped with an Olympus DP 80 digital camera.
For SEM analysis, the fresh anthers dissected from
sorghum spikelets were processed and analyzed accord-
ing to the CRYO-SEM procedures described in Tsou et
4 of 12 the pl ant genome
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
al. (2015). Surface and cross-section images of anther and
pollen were analyzed and photographed at 100´, 400´,
and 800´ magnifications using a Hitachi S-4300 SEM
(Hitachi America).
Whole Genome Sequencing and Identification
of Ms9 Gene
MutMap, which is based on whole-genome sequenc-
ing data of bulked F2 segregants, was used to identify
the causal mutation in the ms9 mutant (Abe et al., 2012;
Jiao et al., 2018). Leaf tissues from 40 confirmed back-
crossed F2ms mutants were sampled and used to isolate
genomic DNA (gDNA) as described previously (Xin and
Chen, 2012). Equal amounts of gDNA from each of the
40 samples were pooled, quality-checked, and subjected
to 150-bp paired-end sequencing on an Illumina X-10
instrument (https://en.novogene.com). Whole-genome
sequencing was performed using three bulked F2 pools:
two BC1F2ms pools (A and B) for the two backcrosses
made on two sterile plants in 2015 and one BC2F2ms
pool containing equal amounts of gDNA of 20 BC2F2ms
plants generated from each of the BC1F2ms A and B pop-
ulations. For each of the F2 pools, we obtained ~10 Gb of
high-quality sequence data, corresponding to ~15´ cov-
erage of the whole genome. Sorghum reference genome
version 2 was used for all sequencing data analyses.
Standards, programs, database and processes used for
bioinformatic analyses of the whole-genome sequencing
data were essentially as described previously (Jiao et al.,
2018). After identification of the candidate causal muta-
tions for each of the BC1F2ms A and B pools, the candi-
date mutations from two F2 populations were compared
to determine the overlap of the mutations. The candi-
date mutations identified for the combined BC2F2ms
pool were also compared with those identified from the
BC1F2ms pools. The bulked segregant analysis sequenc-
ing data has been deposited to NCBI sequence read
archive database under accession number of SRP183026
(https://www.ncbi.nlm.nih.gov/sra/?term=SRP183026).
The orthologs of the Ms9 gene were obtained from
the Gramene database (http://ensembl.gramene.org/Sor-
ghum_bicolor/Gene/Compara_Ortholog?g=SORBI_30
02G234700;r=2:62572601–62576927;t=KXG35825). The
phylogenetic relationship of several selected homologous
genes of Ms9 was constructed using Mega 7 (Kumar et
al., 2016) with ClustalW alignment and the maximum
likelihood method (Tello-Ruiz et al., 2016). Multiple
alignments from ClustalW were visualized with MSAv-
iewer (Yachdav et al., 2016).
RESULTS
Morphological Characterization of the ms9 Mutant
In WT BTx623 plants, the first visible sign of anthesis is
the protrusion of yellow anthers from the sessile spike-
lets on the top of panicles (Fig. 1A). The female stigmas
of WT plants were only transiently and partially visible
because they are overshadowed by the large fluffy yellow
 vol . 12, no . 3  november 2019

Fig. 1. Comparison of sorghum panicles, spikelets, and ovary devel-
opment between BTx623 wild-type and ms9 plants. (A) BTx623 wild-
type panicles at the anthesis stage showing the mature yellow anthers
from sessile spikelets. (B) ms9 panicle at the anthesis stage showing
the small pale anthers extruded from ms9 sessile spikelets. (C,D)
Close-up of sorghum spikelets at the anthesis stage showing (C) pollen-
releasing yellow anthers and pollen-receiving hairy stigmas of BTx623
and (D) unopened small pale anthers and white hairy stigmas of ms9.
(E) A sessile spikelet of wild-type BTx623 exhibits the three normal yel-
low mature anthers releasing pollen and two hairy stigmas with numer-
ous yellow pollen grains. (F) A sessile spikelet of an ms9 plant exhibits
a normal developed spikelet with three pale skinny anthers and two
clear hairy stigmas with no pollen grains. (G) Male and female repro-
ductive organs of a dissected ms9 spikelet shown three skinny pale
no-pollen containing anthers and two hairy stigmas. Female reproduc-
tive organs of (H) ms9 and (I) BTx623 on the day of flowering. Ovary
of BTx623 (left) and ms9 (right) (J) prior to pollination and (K) 4 d after
pollination showing normal ovary development in both BTx623 and
the ms9 mutant after manual pollination of the ms9 panicle.
chen et al . 5
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
anthers and became withered following pollination (Fig.
1A,C). For ms9 plants, however, anthers extruded from
the sessile spikelets were pale in color, shed no pollen, and
were much flatter and thinner than those of WT plants
(Fig. 1B,D). Additionally, the white hairy stigmas of ms9
that emerged from the spikelet were longer, fluffier, and
much more visible (Fig. 1D,F) than those of WT plants
(Fig. 1C,E). The small pale anthers and fluffy white hairy
stigmas of ms9 made it very easy to identify the sterile
plants at a very early stage of anthesis (Fig. 1B). Overall,
other than the sterile phenotype, the development and
morphological structure of ms9 sessile spikelets were sim-
ilar to those of BTx623 (Fig. 1E,F). No noticeable morpho-
logical or developmental differences in any other parts of
the sorghum floral tissues were observed between the ms9
mutant and WT plants (Fig. 1).
Female Fertility is Normal in the ms9 Mutant
To test for defects in female organs, we examined ovary
development and seed set of ms9 plants by manual pol-
lination of ms9 panicles with BTx623 pollen. Dissection
of sorghum sessile spikelets just prior to anther extrusion
revealed that female organs in ms9 sessile spikelets had
normal structures (Fig. 1G). On the day of flowering,
we observed no differences between ms9 and BTx623 in
stigma, style, or ovary size and appearance (Fig. 1H,I),
indicating that development of female organs was nor-
mal in mutant plants. The only difference observed was
the absence of pollen grains on ms9 stigmas (Fig. 1H). In
addition, we examined the development of ovaries before
and 4 d after pollination and found that the pollinated
ovaries of ms9 developed similarly to those of BTx623
plants (Fig. 1J,K). Moreover, the ovaries of ms9 sessile
spikelets were able to complete fertilization processes and
produce normal seeds after pollination with WT pol-
len, providing further evidence of normal development.
Examination of seed set, seed size, and seed morphology
of manually pollinated ms9 panicles revealed normal
seed set on pollinated parts (Fig. 2A–D), no seed set on
nonpollinated parts (Fig. 2A–C), and no seed set on the
bagged, self-pollinated ms9 panicles (Fig. 2F and 2G)
compared with the normal seed set of the self-pollinated
WT BTx623 (Fig. 2E). These results indicated that the
development of female reproductive organs and female
fertility were not affected by the ms9 mutation. Therefore,
we concluded that the ms9 mutation only affected male
sterility, the phenotypes shown in Fig. 1B, 1D, and 1F.
The Sterile Phenotype of ms9 Mutant is a Result
of Male Sterility
The normal seed set on manually pollinated and cross-
pollinated (not-bagged) BC1F2 ms9 panicles (Fig. 2A–D),
in contrast to the lack of seed set on bagged ms9 panicle
prior to anthesis (Fig. 2F,G), suggested that ms9 was a
completely male-sterile mutant. In the absence of a fertile
pollen source, the ms9 panicles were unable to produce
any seed, whether plants were grown in long-day, hot
and dry summer at LBK field, in short-day, moderate
of 12
 19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Fig. 2. Seed set phenotypes of BTx623 and ms9 plants grown under field conditions. (A to D) Seed sets of the ms9 panicles manually pol-
linated with BTx623 pollens in the morning of the (A) first, (B) second, and (C) third day, and (D) at completion of anthesis showing normal seed
set of ms9 panicles after manual pollination and (A–C) the absence of seed set after the panicles were bagged for self-pollination compared
with (E) a self-pollinated BTx623 panicle and (F) the absence of seed set after the panicle was bagged for self-pollination. (F) An ms9 panicle
bagged prior to anthesis. (G) An ms9 panicle bagged after removing the top open flowered spikelets during anthesis. The self-pollinated
BTx623 wild-type panicle set normal seeds (E), whereas the self-pollinated ms9 panicles (F and G) produced no seeds.

temperature and humid PR field site, or under long-day

and optimal temperature greenhouse conditions. These
results indicated that the male sterility of ms9 is stable
under variation in day length and temperature condi-
tions, and that the mutation is fully penetrant.

The ms9 Anthers Lack Mature Pollen as a Result

of Abnormal Microspore Development
Morphologically, the three extruding anthers of the ses-
sile spikelets in ms9 mutant panicle were pale in color and
much thinner than the plump yellow anthers on BTx623
panicles (Fig. 1). At anthesis, the ms9 anthers dispersed
no pollen (Fig. 1D), and therefore the two hairy stigmas
of ms9 were free of pollen grains (Fig. 1F,H) in contrast
to the two pollen-receiving hairy stigmas observed on
BTx623 sessile spikelets (Fig. 1E,I). To investigate the cause
of male sterility in the ms9 mutant, we examined pollen
production and pollen viability in mature anthers of ses-
sile spikelets prior to their extrusion. The BTx623 anther
lobes were full of round, reddish, mature pollen grains
(Fig. 3A,C), whereas the anther lobes of ms9 mutants were
empty (Fig. 3B). A microscopic close-up of the ms9 anther
Fig. 3. Morphological analyses of anther and pollen development
showed no sign of mature pollen being released (Fig. 3D). in BTx623 wild-type and ms9 mutant plants. Alexander staining of
We conclude that the lack of mature pollen formation is anthers and pollen of (A,C) BTx623 and (B,D) ms9. The BTx623 anther
the underlying cause for male sterility in ms9. contains plenty round pollen grains (A), and the mature pollens are
To determine the cause of failed pollen produc- viable (C, viable tissues are stained in reddish-pink in color), whereas
tion in the ms9 mutant, we examined and compared the ms9 anther is visibly empty (B) and contains no pollen grains (D).

6 of 12 the pl ant genome  vol . 12, no . 3  november 2019

 19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Fig. 4. Morphological analyses of anther development by (A–H) light and (I–P) scanning electron microscopic analyses show delayed tapetal cell
degradation and abnormal microspore development at midmicrospore stage, leading to the degeneration of microspores inside the anther lobes
of ms9. Histological thin section of anther lobes of (A–D) BTx623 and (E–H) ms9 and SEM images of anther lobes of (I–L) BTx623 and (M–P) ms9
showing (i) no difference between BTx623 and ms9 at the early anther development stages (A,E; B,F); (ii) delayed degradation of tapetal layer
and premature degeneration of microspores in ms9 at midmicrospore stages (G,N,O); (iii) and an empty anther lobe with no pollen grains in ms9
at the pollen maturation stage (H,P). BMsp, binuclear microspores where the microspore underdo the second mitosis; DMsp, degenerated micro-
spores; E, epidermis; En, endothecium; M, middle layer; PG, mature pollen grains; Msp, microspores; PMC, pollen mother cell; T, tapetal layer.

morphological features of the anthers at different devel-
opmental stages by histological and SEM analyses. We
identified no morphological differences in the lobule
anthers at or prior to the microspore midvacuolated
stages (Fig. 4A,B,E,F). The four distinct layers of anther
wall, epidermis, endothecium, middle layer, and tape-
tum formed normally in ms9 anthers (Fig. 4E,F). While
the tapetum of WT anther started to degenerate dur-
ing microspore stages (Fig. 4C,I,J), the tapetum of ms9
anther remained clearly noticeable (Fig. 4G,M,N). In
addition, analyses revealed that the microspores of the
ms9 plants started to degenerate prematurely at mid-
microspore stage and the microspores became further
shriveled (Fig. 4N) at the late-vacuolated stage, whereas
the microspores of the WT BTx623 increased in size
and became spherical (Fig. 4J). At this stage, the epider-
mal layers of both the WT and the mutant anthers were
concentrically organized and the middle layer became
imperceptible (Fig. 4C,G,J,N). In ms9, however, the tape-
tum layer was still largely noticeable (Fig. 4G,N) whereas
the microspores shriveled (Fig. 4N). During the pollen
maturation stages, WT pollen became round, and the
tapetum and endothecium layers degenerated completely
(Fig. 4D,L). By contrast, the microspores in ms9 anther
lobes degenerated completely, resulting in little or no cell
debris within the empty anthers at pollen maturation
stage (Fig. 4H,P). These results indicate that mutation in
ms9 hinders the normal degeneration process of tapetal
cells of the anther wall layer and causes abnormal devel-
opment of microspores at the midmicrospore develop-
mental stage, resulting in degeneration of microspores
inside the anther lobes.
ms9 is a Novel Recessive Nuclear Male Sterility Mutant
Several lines of genetic evidence revealed in this study
indicated that the male sterility of ms9 was caused by a
recessive mutation on a single nuclear gene. The F1 plants
produced by manual pollination of the ms9 mutant
with BTx623 pollen were all male-fertile and set seeds,
indicating that the ms9 mutation is recessive (Table 1).
Phenotyping of F2 plants derived from self-pollinated F1
progeny identified 161 fertile plants and 43 male-sterile
plants, reflecting a segregation ratio of approximately 3:1.
Phenotyping of the subsequent advanced backcrossing F2

chen et al . 7 of 12
 19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 1. Genetic analysis of the sorghum ms9 nuclear male sterility mutant. Segregation and complementation (cp) analyses indicated that
the male sterility of ms9 is mediated by a novel recessive mutation in a nuclear gene.
c2
Crosses Generation Total no. of plants Sterile plants Fertile plants Sterile/fertile ratios (p-value)
ms9*BTx623 F1 33 0 33 0: 1 –
ms9*BTx623 F2 204 43 161 1: 3 0.94
ms9*BTx623 BC1F2 461 117 344 1: 3 0.87
ms9*BTx623 BC2F2 493 122 371 1: 3 0.90
ms1*ms9BC2F1(Ms9ms9) cpF1 32 0 32 0: 1 –
ms2*ms9BC2F1 cpF1 27 0 27 0: 1 –
ms3*ms9BC2F1 cpF1 36 0 36 0: 1 –
Ms7*ms9BC2F1 cpF1 33 0 33 0: 1 –
msal*ms9BC2F1 cpF1 18 0 18 0: 1 –
ms8*ms9BC2F1 cpF1 39 0 39 0: 1 –
ARS84ms*ms9BC2F1 cpF1 41 20 21 1: 1 0.86

populations (BC1F2 and BC2F2) yielded similar ratios of
fertile to sterile plants (Table 1). This segregation analy-
sis indicated that the male sterility in ms9 is caused by a
recessive mutation in a single nuclear gene.
We then performed complementation analyses
to determine whether ms9 is allelic to any previously
reported sorghum NMS mutants (Pedersen and Toy,
2001; Xin et al., 2018). Phenotyping results revealed
that all cpF1 plants from the complementation crosses
between male-sterile plants of known NMS lines and
heterozygous BC2F1 ms9 plants were fertile (Table 1).
These findings demonstrated that the ms9 is not allelic
to any of the currently available sorghum NMS mutants
and that Ms9 is most likely a new male-sterile locus.
Ms9 Encodes a Plant Homeotic Domain–Finger
Transcription Factor
We used a modified MutMap approach (Jiao et al., 2018)
to identify the causal gene mutation that leads to male
sterility in ms9. Specifically, we performed whole-genome
sequencing of two bulked F2 populations independently
generated from two different original crosses, each con-
taining 40 homozygous F2 ms9 mutants. Bioinformatic
analysis of the whole-genome sequencing datasets using
our in-house pipeline (Jiao et al., 2018) identified homo-
zygous nonsynonymous mutations in nine genes in the
first population and five genes in the second population,
only two of which overlapped (Fig. 5A). Analysis of the
two combined BC2F2ms whole-genome sequencing data-
sets identified homozygous nonsynonymous mutations
in six genes, including mutations in the same two over-
lapping genes (Fig. 5A). Therefore, the number of causal
gene mutations for male sterility in ms9 was reduced to
the two located on chromosome 2. One is a deleterious
mutation located in Sobic.002G221000 gene encoding a
PHD-finger transcription factor. A C-to-T transition at
nucleotide 1207 in Sobic.002G221000 gene in ms9 causes
an amino acid change from arginine to tryptophan
at conserved position 218 (R218W, Fig. 5B). The other
mutation located in Sobic.002G234700 gene encoding
an ubiquitin carboxyl-terminal hydrolase family pro-
tein. A C-to-T transition in Sobic.002G234700 causes an
amino acid changes from proline to leucine at position
98 (P98L). The M4 mutant line (ARS178) from which the
ms9 mutant was isolated has already been sequenced and
annotated for gene mutations (Jiao et al., 2016). There-
fore, we searched the mutant database and confirmed
that the ARS178 M4 mutant line contained the same
nonsynonymous mutations (but in the heterozygous
state) in the two candidate genes identified by MutMap.
We employed four approaches to determine which of
the two gene mutations is responsible for the male-sterile
phenotype in ms9. First, we mined the mutation database
of our 256 sequenced mutant lines to identify mutant lines
that contained independent mutation alleles in one of the
two candidate genes to determine whether they segregated
a male-sterile phenotype. One of the two M4 lines harbor-
ing heterozygous mutations in Sobic.002G221000 segre-
gated a male sterility phenotype [ARS84 (25M2-1505)],
whereas none of the five M4 mutant lines harboring hetero-
zygous mutations in Sobic.002G234700 did. The ARS84 line
harbored a C-to-T mutation at base 208, changing alanine
to valine at conserved position 37 (A37V). The other line,
ARS44, harbors a predicted amino acid change (G419S) in
a nonconserved region in Sobic.002G221000 with a high
SIFT score (0.74), implying that the predicted amino acid
substitution (G419S) in ARS44 would have little effect
on the function of the Sobic.002G221000 protein, and is
therefore unlikely to alter the phenotype. Consistent with
this prediction, no male-sterile plant was observed in the
M4 plants. Together, these results suggest that the R218W
mutation identified in the sorghum Sobic.002G221000 gene
is the likely cause of male sterility in ms9.
Second, we performed complementation analysis
between ARS178 and ARS84 by crossing a male-sterile
plant in ARS84 with pollen collected from a heterozygous
BC2F1 ms9 plant and then phenotyping segregation of
male-sterile and fertile plants in the resulting cpF1 prog-
enies. We obtained 21 fertile cpF1 plants and 20 male-
sterile cpF1 plants, a segregation ratio of approximately
1:1 (Table 1). This result indicated that the male sterility
phenotypes in ARS84 and ms9 were caused by mutations
in the same gene, Sobic.002G221000. Therefore, we desig-
nated Sobic.002G221000 as Ms9. The ms mutation isolated

8 of 12 the pl ant genome  vol . 12, no . 3  november 2019

 19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
from ARS178 (M2P110) as ms9-1 and the ms mutation
identified in ARS84 as ms9-2 (Fig. 5B). The nucleotide
sequences of Ms9, Ms9-1, and Ms9-2 have been deposited
into the GenBank with accession numbers of MK908415,
MK908416, and MK908417, respectively.
Third, to obtain additional evidence that the muta-
tions of Ms9 are the cause of male sterility in sorghum, we
compared the functions of homologs of the two candidate
genes. Ms9 homologous genes include male sterile 1 (Ms1,
AT5G22260) in Arabidopsis (Wilson et al., 2001), persistent
tapetal cell 1 (Ptc1, Os09g0449000) in rice (Li et al., 2011),
and the male sterile 7 (zm0001d020680) in maize (Zhang
et al., 2018), all of which cause male sterility in their respec-
tive plant species. The homologs of the other candidate
gene identified have no reported roles in male sterility.
The sorghum Ms9 gene has three exons and two
introns; the mature cDNA is 2444 bp in length (Fig. 5B).
This gene contains a PHD-finger protein domain from the
612 to 658 amino acids according to the Pfam annotation
(Fig. 6). This protein domain has been known to regulate
the plant developmental transitions by recognizing the
histone covalent modification and recruiting chromatin
remodeling complexes and the transcriptional machinery
(Mouriz et al., 2015). The structure of Ms9 in sorghum is
similar to those of its homologs in Arabidopsis and cereal
Fig. 5. Sorghum Ms9 encodes a PHD transcription factor. (A) The
crops, and the gene encodes a protein of 668 amino acids
number of overlapping candidate causal gene mutations identified
(Fig. 6). Phylogenetic analysis of nine homologs selected among the three bulk-sequenced F2 populations. (B) Gene structure
from grass species, one from Arabidopsis, and one from of Sobic.002G221000 (MK908415) showing the causal mutation
tomato (Solanum lycopersicum L.) indicated that the clos- (C1207T) in exon 3 (R218W) in ms9-1 (MK908416) and the causal
est homolog is the maize gene zm0001d020680 (Fig. 5C), mutation (C208T) in exon 1 (A37V) in ms9-2 (MK908417). (C) Gene
which has recently been identified as male sterile 7 in tree of Sobic.002G221000 and its homologs in cereal crops, Ara-
maize (Zhang et al., 2018). Amino-acid sequence align- bidopsis, and tomato, showing that SbMS9 is an ortholog of genes
ment revealed that the Ms9 in sorghum is highly similar known to regulate male sterility: AtMS1 in Arabidopsis, OsPTC1 in
to all nine genes used to construct the alignment (Supple- rice, and MS7 (Zm0001d020680) in maize.
mental Fig. S2). Furthermore, the amino acid sequences in
conserved regions of Ms9 in sorghum are 100% identical
to those of Ms7 in maize (Fig. 6) and almost identical to breeding. The discovery of male sterility traits in plants
those of homologs in other cereal crops (Supplemental Fig. has enabled breeders to produce hybrid seeds much more
S2). Both R218W and A37V amino acid changes resulting efficiently in a wide range of crops (Kim and Zhang,
from the nonsynonymous mutations in ms9-1 and ms9-2, 2018). The direct benefit of hybrid production has been
respectively, occur within conserved regions of SbMS9 demonstrated by the significant yield advantage of maize
(Fig. 6). Together, these results provide strong evidence hybrids over land races as well as improved yields and
that Sobic.002G221000 is Ms9 in sorghum. fitness in many other major crops and bioenergy species
In the fourth approach, the expression pattern of (Zhang et al., 2018). Control of pollen production is criti-
the Ms9 gene was compared with its orthologs in Arabi- cal for hybrid breeding especially for hybrid seed pro-
dopsis and rice, in which the expression pattern and its duction in self-pollinated species. Therefore, identifying
role in programed cell death are well characterized. We male sterility in important crop species and improving
took advantage of the publicly available sorghum gene their use in hybrid breeding systems could make impor-
expression atlas (http://sorghum.riken.jp/morokoshi/ tant contributions to increasing future agricultural pro-
Home.html) and showed that the Ms9 gene has similar duction and food security.
expression pattern as Ms1 in Arabidopsis and Ptc1 in rice No gene that mediates NMS in sorghum has been
(Supplemental Fig. S3). cloned previously despite the discovery of several sor-
ghum NMS lines and mutants (Andrews and Webster,
DISCUSSION 1971; Pedersen and Toy, 2001; Xin et al., 2017). Here,
Pollen production is essential for pollination, which is we report the isolation and characterization of a new
the first step in setting seed. Although failure to pollinate male-sterile mutant and the identification of the first
(male sterility) can prevent seed set in self-pollinated NMS gene in sorghum. This mutant, designated ms9,
plant species, male sterility is highly beneficial in hybrid is distinct from all other sorghum NMS lines reported

chen et al . 9 of 12
 19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Fig. 6. Multiple amino acid alignment of the Sobic.002G221000 gene and its ortholog genes in maize, rice, and in model plant Arabidopsis.
The causal mutations in ms9-1 and ms9-2 are indicated by an arrow. The dark shaded are conserved amino acids.

previously (Andrews and Webster, 1971; Pedersen and
Toy, 2001; Xin et al., 2017). The male-sterile phenotype in
ms9 mutants can be easily recognized at onset of anthesis
because of its thin pale anthers and exaggerated stig-
mas. Other than male sterility, the ms9 mutants develop
similarly to WT BTx623 plants. The characteristics of
ms9 make it ideal for development of a two-line breeding
system in sorghum based on NMS (Chang et al., 2016;
Zhang et al., 2018).
The identification of Ms9 as the causal mutation for the
male-sterile phenotype of the ms9 mutant is supported by
bioinformatic analysis of two independent whole-genome
sequencing data sets of pooled F2 mutants as well as by
identification of another independent allele (Fig. 5) from
the sequenced mutant library (Jiao et al., 2016). The cloned
Ms9 gene encodes a PHD-finger transcription factor with
a gene structure very similar to that of Ms1 in Arabidopsis,
Ptc1 in rice, and Ms7 in maize (Li et al., 2011; Wilson et
al., 2001; Zhang et al., 2018) as well as homologs in other
cereals and plant species (Fig. 5C). The mutations identi-
fied in the two ms9 mutant alleles (R218W and A37V)
cause amino-acid changes in the conserved domains of
this protein (Fig. 6). The amino-acid sequence of SbMS9 is
also very similar to its orthologous counterpart other crops
(Supplemental Fig. S2). MS7, recently identified in maize,
is the closest homolog to SbMS9, with 100% identity in
the conserved region (Fig. 5C, 6). The phenotype of ms9 is
similar to that of the male-sterility phenotypes of Arabi-
dopsis ms1, the rice ptc1 and maize ms7 lines (Li et al., 2011;
Wilson et al., 2001: Zhang et al., 2018), all of which have
significant effects on anther morphology and pollen devel-
opment but no effect on other aspects of floral development
and morphology (Fig. 1, 3, 4).
The expression pattern of the Ms9 gene is also simi-
lar to its orthologs in other species. The expression data
of Ms9 gene was extracted from the Morokoshi Sorghum
Gene Expression Atlas (Makita et al., 2015). SbMs9 is
highly expressed in young inflorescence tissues and anthers
(Supplemental Fig. S3A). This tissue-specific gene expression
pattern is similar to that of Ms1 in Arabidopsis and Ptc1 in
rice (Li et al., 2011; Wilson et al., 2001), further supporting
the role of Ms9 in the development of tapetum and pollen
grains. As with Ms1 and Ptc1 in Arabidopsis and rice, Ms9
may serve as a critical regulator in tapetal cell degeneration
and pollen development during anther development.
At present, the three-line hybrid breeding system
relies exclusively on CMS for the male-sterile female
parent in sorghum (Praveen et al., 2015; Rooney, 2004).
Although several types of cytoplasmic male-sterile lines
are available, commercial hybrid production uses mainly

10 of 12 the pl ant genome  vol . 12, no . 3  november 2019


the A1 cytoplasm (Jordan et al., 2011). A main advantage
of the three-line breeding system in sorghum is that good
A–B pairs can produce nearly 100% male-sterile line to
serve as a female parent during production of hybrid
seeds. Furthermore, this system has been used in breeding
grain sorghum since 1940s, and many breeding materi-
als in grain sorghum have been converted for use in this
breeding system (Stephens and Holland, 1954). However,
several aspects of the CMS breeding system need to be
improved. For example, many sorghum accessions are
neither perfect B nor R lines and cannot be used in breed-
ing sorghum hybrids without lengthy period of conver-
sion to B or R lines. In addition, many lines suitable for
breeding biomass or sweet sorghum hybrids have not been
converted and cannot be used directly to breed hybrids
with the CMS-based breeding system. Furthermore, The
A1 cytoplasmic homogeneity may predispose sorghum
hybrids to devastating diseases, as in the T-cytoplasmic
maize hybrids produced in the 1970s (Ullstrup, 1972).
A two-line breeding system that uses NMS can poten-
tially overcome the disadvantage of the CMS-based three-
line breeding system. For example, the male-sterility of a
two-line breeding system based on a nuclear mutation can
be restored by any line that does not carry the same muta-
tion (Chang et al., 2016). This advantage is particularly
useful for breeding biomass and sweet sorghum hybrids
because accessions with the useful bioenergy traits can
be directly used as parents for the hybrids. The two-line
breeding system based on NMS also does not require
male-sterile cytoplasm and, therefore, can avoid homo-
geneity of cytoplasm in hybrid. The main disadvantage
of directly using NMS in hybrid breeding is that a fertile
plant, heterozygous for an NMS mutation, only produces
25% of homozygous male-sterile plants. To remove the fer-
tile plants from a breeder’s field is nearly impossible. For-
tunately, through strategic manipulation of the NMS gene
and its mutation, the two-line breeding system has been
shown to produce pure male-sterile lines in rice (Chang et
al., 2016; Huang et al., 2014; Zhou et al., 2014). In Maize,
Ms7, the closest homolog of the sorghum Ms9 gene, was
identified by map-based cloning and used, along with its
WT gene, to engineer a controllable male-sterile line for
a two-line breeding system (Zhang et al., 2018). A gene
construct with the WT Ms7 gene and multiple control
elements has been used to rescue the male sterility of the
ms7 mutant and produce pure male-sterile seeds (Zhang et
al., 2018). The identification of the Ms9 gene and its causal
mutations provides critical tools to manipulate the pro-
duction of male-sterile parent that is 100% male sterile.
In summary, we characterized a new sorghum
NMS mutant, ms9, and identified the first NMS gene in
sorghum. The identification of the SbMs9 provides an
opportunity to engineer controllable male sterility for
development of a two-line breeding system in sorghum.
Supplemental Information Available
Supplemental information is available with the online
version of this manuscript.
chen et al . 11
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Conflict of Interest
The authors declare that there is no conflict of interest.
Author Contributions
JC and ZX conceived the idea and designed experi-
ment, YJ analyzed the data, HL performed histological
and SEM analyses of anther features, all performed
the experiment. JC and ZX drafted the manuscript
with input from all authors. All authors agree with the
final manuscript.
Disclaimer
Mention of trade names or commercial products in this
publication is solely for the purpose of providing specific
information and does not imply recommendation or
endorsement by the USDA. The USDA is an equal oppor-
tunity provider and employer.
ACKNOWLEDGMENTS
We thank Jing Wang for technical support in the phenotypic analysis and
Scott Sattler for providing the sorghum nuclear male-sterile lines used for
allelic test with ms9 mutant.
REFERENCES
Abe, A., S. Kosugi, K. Yoshida, S. Natsume, H. Takagi, H. Kanzaki, et al. 2012.
Genome sequencing reveals agronomically important loci in rice using
MutMap. Nat. Biotechnol. 30:174–178. doi:10.1038/nbt.2095
Andrews, D.J., and O.J. Webster. 1971. A new factor for genetic male sterility
in Sorghum bicolor (L.) Moench. Crop Sci. 11:308–309. doi:10.2135/crop
sci1971.0011183X001100020046x
Chang, Z., Z. Chen, N. Wang, G. Xie, J. Lu, W. Yan, et al. 2016. Construction
of a male sterility system for hybrid rice breeding and seed produc-
tion using a nuclear male sterility gene. Proc. Natl. Acad. Sci. USA.
doi:10.1073/pnas.1613792113
de Siqueira Ferreira, S., M.Y. Nishiyama, Jr., A.H. Paterson, and G.M. Souza.
2013. Biofuel and energy crops: High-yield Saccharinae take center
stage in the post-genomics era. Genome Biol. 14:210. doi:10.1186/
gb-2013-14-6-210
Dong, N.V., P.K. Subudhi, P.N. Luong, V.D. Quang, T.D. Quy, H.G. Zheng, et
al. 2000. Molecular mapping of a rice gene conditioning thermosensi-
tive genic male sterility using AFLP, RFLP and SSR techniques. Theor.
Appl. Genet. 100:727–734. doi:10.1007/s001220051345
Huang, J., A.R. Smith, T. Zhang, and D.D. Zhao. 2016. Creating completely
both male and female sterile plants by specifically ablating micro-
spore and megaspore mother cells. Front. Plant Sci. 7. doi:10.3389/
fpls.2016.00030
Huang, J.Z., Z.G. E, H.L. Zhang, and Q.Y. Shu. 2014. Workable male sterility
systems for hybrid rice: Genetics, biochemistry, molecular biology, and
utilization. Rice 7:13. doi:10.1186/s12284-014-0013-6
Jiao, Y., J.J. Burke, R. Chopra, G. Burow, J. Chen, B. Wang, et al. 2016. A sor-
ghum mutant resource as an efficient platform for gene discovery in
grasses. Plant Cell 28:1551–1562. doi:10.1105/tpc.16.00373
Jiao, Y., G. Burow, N. Gladman, V. Acosta-Martinez, J. Chen, J. Burke, et al.
2018. Efficient identification of causal mutations through sequencing of
bulked f2 from two allelic bloomless mutants. Front. Plant Sci. 8:2267.
doi:10.3389/fpls.2017.02267
Jordan, D.R., R.R. Klein, K.G. Sakrewski, R.G. Henzell, P.E. Klein, and E.S.
Mace. 2011. Mapping and characterization of Rf5: A new gene condi-
tioning pollen fertility restoration in A1 and A2 cytoplasm in sorghum
(Sorghum bicolor (L.) Moench). Theor. Appl. Genet. 123:383–396.
doi:10.1007/s00122-011-1591-y
Kim, Y.J., and D. Zhang. 2018. Molecular control of male fertility for
crop hybrid breeding. Trends Plant Sci. 23:53–65. doi:10.1016/j.
tplants.2017.10.001
of 12

Kumar, S., G. Stecher, and K. Tamura. 2016. MEGA7: Molecular evolutionary
genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33:1870–
1874. doi:10.1093/molbev/msw054
Li, H., Z. Yuan, G. Vizcay-Barrena, C. Yang, W. Liang, J. Zong, et al. 2011.
PERSISTENT TAPETAL CELL1 encodes a PHD-finger protein that
is required for tapetal cell death and pollen development in rice. Plant
Physiol. 156:615–630. doi:10.1104/pp.111.175760
Makita, Y., S. Shimada, M. Kawashima, T. Kondou-Kuriyama, T. Toyoda, and
M. Matsui. 2015. MOROKOSHI: Transcriptome database in Sorghum
bicolor. Plant Cell Physiol. 56:E6. doi:10.1093/pcp/pcu187
McCormick, R.F., S.K. Truong, A. Sreedasyam, J. Jenkins, S. Shu, D. Sims, et
al. 2018. The Sorghum bicolor reference genome: Improved assembly,
gene annotations, a transcriptome atlas, and signatures of genome orga-
nization. Plant J. 93:338–354. doi:10.1111/tpj.13781
Morris, G.P., P. Ramu, S.P. Deshpande, C.T. Hash, T. Shah, H.D. Upadhyaya,
et al. 2013. Population genomic and genome-wide association studies of
agroclimatic traits in sorghum. Proc. Natl. Acad. Sci. USA 110:453–458.
doi:10.1073/pnas.1215985110
Mouriz, A., L. Lopez-Gonzalez, J.A. Jarillo, and M. Pineiro. 2015. PHDs gov-
ern plant development. Plant Signal. Behav. 10:e993253. doi:10.4161/155
92324.2014.993253
Paterson, A.H., J.E. Bowers, R. Bruggmann, I. Dubchak, J. Grimwood, H.
Gundlach, et al. 2009. The Sorghum bicolor genome and the diversifica-
tion of grasses. Nature 457:551–556. doi:10.1038/nature07723
Pedersen, J.F., and J.J. Toy. 2001. Registration of N316-N320 sorghum
nuclear male-sterility genetic stocks. Crop Sci. 41:607. doi:10.2135/
cropsci2001.412607x
Praveen, M., G. Anurag Uttam, N. Suneetha, A. Umakanth, J.V. Patil, and R.
Madhusudhana. 2015. Inheritance and molecular mapping of Rf6 locus
with pollen fertility restoration ability on A1 and A2 cytoplasms in sor-
ghum. Plant Sci. 238:73–80. doi:10.1016/j.plantsci.2015.05.020
Rooney, W.L. 2004. Sorghum improvement—Integrating traditional and new
technology to produce improved genotypes. Adv. Agron. 83:37–109.
doi:10.1016/S0065-2113(04)83002-5
Rooney, W.L., J. Blumenthal, B. Bean, and J.E. Mullet. 2007. Designing sor-
ghum as a dedicated bioenergy feedstock. Biofuels Bioprod. Biorefin.
1:147–157. doi:10.1002/bbb.15
Stephens, J.C., and R.F. Holland. 1954. Cytoplasmic male-sterility for hybrid
sorghum seed production. Agron. J. 46:20–23. doi:10.2134/agronj1954.0
0021962004600010006x
Tello-Ruiz, M.K., J. Stein, S. Wei, J. Preece, A. Olson, S. Naithani, et al. 2016.
Gramene 2016: Comparative plant genomics and pathway resources.
Nucleic Acids Res. 44:D1133–D1140. doi:10.1093/nar/gkv1179
Tsou, C.H., P.C. Cheng, C.M. Tseng, H.J. Yen, Y.L. Fu, T.R. You, et al. 2015.
Anther development of maize (Zea mays) and longstamen rice (Oryza
longistaminata) revealed by cryo-SEM, with foci on locular dehydra-
tion and pollen arrangement. Plant reproduction 28:47–60. doi:10.1007/
s00497-015-0257-3
12 of 12 the pl ant genome
19403372, 2019, 3, Downloaded from https://acsess.onlinelibrary.wiley.com/doi/10.3835/plantgenome2019.03.0020 by INASP - TANZANIA, Wiley Online Library on [14/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Ullstrup, A.J. 1972. The impacts of the Southern corn leaf blight epidemics
of 1970-1971. Annu. Rev. Phytopathol. 10:37–50. doi:10.1146/annurev.
py.10.090172.000345
Wilson, Z.A., S.M. Morroll, J. Dawson, R. Swarup, and P.J. Tighe. 2001.
The Arabidopsis MALE STERILITY1 (MS1) gene is a transcriptional
regulator of male gametogenesis, with homology to the PHD-finger
family of transcription factors. Plant J. 28:27–39. doi:10.1046/j.1365-
313X.2001.01125.x
Xin, Z., and J. Chen. 2012. A high throughput DNA extraction method with
high yield and quality. Plant Methods 8:26. doi:10.1186/1746-4811-8-26
Xin, Z., J. Chen, Y. Jiao, N. Gladman, C. Hayes, G. Burow, et al. 2018. Reg-
istration of BTx623ms8, a new and easily identifiable nuclear male
sterile mutant in sorghum. J. Plant Reg. 12:278–281. doi:10.3198/
jpr2017.09.0063crgs
Xin, Z., J. Huang, A.R. Smith, J. Chen, J. Burke, S.E. Sattler, et al. 2017.
Morphological characterization of a new and easily recognizable
nuclear male sterile mutant of sorghum (Sorghum bicolor). PLoS One
12:e0165195. doi:10.1371/journal.pone.0165195
Xin, Z., M. Wang, G. Burow, and J. Burke. 2009. An induced sorghum mutant
population suitable for bioenergy research. BioEnergy Res. 2:10–16.
doi:10.1007/s12155-008-9029-3
Yachdav, G., S. Wilzbach, B. Rauscher, R. Sheridan, I. Sillitoe, J. Procter, et
al. 2016. MSAViewer: Interactive JavaScript visualization of multiple
sequence alignments. Bioinformatics 32:3501–3503. doi:10.1093/bioin-
formatics/btw474
Zhang, D., S. Wu, X. An, K. Xie, Z. Dong, Y. Zhou, et al. 2018. Construc-
tion of a multicontrol sterility system for a maize male-sterile line and
hybrid seed production based on the ZmMs7 gene encoding a PHD-
finger transcription factor. Plant Biotechnol. J. 16:459–471. doi:10.1111/
pbi.12786
Zhang, H., W. Liang, X. Yang, X. Luo, N. Jiang, H. Ma, et al. 2010. Carbon
starved anther encodes a MYB domain protein that regulates sugar par-
titioning required for rice pollen development. Plant Cell 22:672–689.
doi:10.1105/tpc.109.073668
Zhang, H., C. Xu, Y. He, J. Zong, X. Yang, H. Si, et al. 2013. Mutation in CSA
creates a new photoperiod-sensitive genic male sterile line applicable
for hybrid rice seed production. Proc. Natl. Acad. Sci. USA 110:76–81.
doi:10.1073/pnas.1213041110
Zhao, D.Z., G.F. Wang, B. Speal, and H. Ma. 2002. The EXCESS MICROSPO-
ROCYTES1 gene encodes a putative leucine-rich repeat receptor protein
kinase that controls somatic and reproductive cell fates in the Arabidop-
sis anther. Genes Dev. 16:2021–2031. doi:10.1101/gad.997902
Zhou, H., M. Zhou, Y. Yang, J. Li, L. Zhu, D. Jiang, et al. 2014. RNase ZS1 pro-
cesses UbL40 mRNAs and controls thermosensitive genic male sterility
in rice. Nat. Commun. 5:4884. doi:10.1038/ncomms5884
 vol . 12, no . 3  november 2019