Genes to Cells

BRIEF REPORT
MYB transcription factor gene involved in sex
determination in Asparagus officinalis
Kohji Murase1*, Shuji Shigenobu2, Sota Fujii1, Kazuki Ueda1, Takanori Murata1,
Ai Sakamoto1, Yuko Wada1, Katsushi Yamaguchi2, Yuriko Osakabe3, Keishi Osakabe3,
Akira Kanno4, Yukio Ozaki5 and Seiji Takayama1*a
1
Graduate School of Biological Sciences, Nara Institute of Science and Technology, Ikoma, Nara 630-0192, Japan
2
National Institute for Basic Biology (NIBB) Core Research Facilities, NIBB, Okazaki, Aichi 444-8585, Japan
3
Center for Collaboration among Agriculture, Industry and Commerce, The University of Tokushima, 3-18-15 Kuramoto-cho,
Tokushima 770-8503, Japan
4
Graduate School of Life Sciences, Tohoku University, Katahira 2-1-1, Aoba-ku, Sendai 980-8577, Japan
5
Faculty of Agriculture, Kyushu University, Kasuya, Fukuoka 811-2307, Japan

Dioecy is a plant mating system in which individuals of a species are either male or female.
Although many flowering plants evolved independently from hermaphroditism to dioecy, the
molecular mechanism underlying this transition remains largely unknown. Sex determination
in the dioecious plant Asparagus officinalis is controlled by X and Y chromosomes; the male
and female karyotypes are XY and XX, respectively. Transcriptome analysis of A. officinalis
buds showed that a MYB-like gene, Male Specific Expression 1 (MSE1), is specifically expressed
in males. MSE1 exhibits tight linkage with the Y chromosome, specific expression in early
anther development and loss of function on the X chromosome. Knockout of the MSE1
orthologue in Arabidopsis induces male sterility. Thus, MSE1 acts in sex determination in
A. officinalis.

Introduction into females and hermaphrodites, and androdioecy, in

which plants separate into males and hermaphrodites
To preserve genetic variety within species, flowering
(Charlesworth & Charlesworth 1978). The first step of
plants have evolved various systems to prevent self-fer-
the evolution of gynodioecy is a recessive male-sterile
tilization. In one such system, dioecy, individuals of a
mutation, followed by a dominant female-sterile muta-
species are either male or female. In angiosperms,
tion (or gain of suppressor function) near the male
approximately 15 000 species (~6%) of 160 families are
mutation locus, thus creating a sex chromosome. Con-
dioecious, and the evolution of dioecy is thought to
versely, a female mutation is the first step in the evolu-
have occurred independently more than 800 times
tion of androdioecy, but this pathway is not
(Charlesworth 2002; Renner 2014). With a current
predominant because female mutations and androdioe-
theoretical model, the transition from hermaphrodit-
cious plants are very rare in nature. However, molecu-
ism to dioecy can proceed by two evolutionary path-
lar mechanisms of sex determination and its evolution
ways: gynodioecy, in which individual plants separate
in flowering plants are largely unknown. The recent
identification of sex determination genes in persim-
mon is the only example that an autosomal homeobox
Communicated by: Toshio Hakoshima
transcription factor gene, MeGI, dominantly suppresses
*Correspondence: [email protected] or [email protected]
a
Present address: Department of Applied Biological
male organ development, whereas OGI on the Y
Chemistry, The University of Tokyo, 1-1-1 Yayoi, chromosome encodes a small RNA that targets MeGI
Bunkyo-ku, Tokyo 113-8657, Japan for gene silencing (Akagi et al. 2014).

DOI: 10.1111/gtc.12453
© 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd Genes to Cells (2017) 22, 115–123 115
 K Murase et al.

Sex determination in the dioecious plant Asparagus low levels (Table S2 in Supporting Information).
officinalis is controlled by a single locus, the Mating Mapping of RNA sequence reads against the MSE1
(M) locus, located on chromosome 5 (L€ optien 1979; contig showed that female reads only mapped to the
Telgmann-Rauber et al. 2007). The sex chromosome 50 end of the transcript, whereas male reads covered
karyotypes of males and females are XY and XX, the whole transcript (Fig. S3 in Supporting Informa-
respectively. Relatively large vestiges of organs corre- tion). Reverse transcription (RT)-PCR using primers
sponding to the opposite sex are observed in both that amplified the full-length MSE1 transcript con-
male and female flowers in A. officinalis, suggesting firmed male-specific expression of MSE1 (Fig. 1C).
that morphological sex differentiation occurs at a late To determine whether MSE1 is on the Y chro-
stage of flower development (Fig. 1A,B). In fact, at mosome, we PCR-amplified MSE1 from the gen-
early developmental stages, the male and female flow- omes of male and female individuals of A. officinalis
ers look like those of hermaphrodites; the morpho- cv. Super Welcome. MSE1 specifically amplified
logical differences between male and female flowers from male individuals, but not from females
appear later when the stylar tube is formed on carpels (Fig. 1D). Subsequent PCR analysis of 112 indepen-
in female flowers, and during or just before meiosis dent plants confirmed male-specific amplification of
in male flowers (Caporali et al. 1994). To generate MSE1 (Fig. S4 in Supporting Information). These
these morphological differences, sex determination results suggest that MSE1 is on the Y chromosome
genes must be expressed in the appropriate tissues at gene, tightly linked to the M locus.
the appropriate developmental stages. Genetic analysis MSE1 encodes a 276-amino acid protein contain-
suggested the involvement of two sex determination ing two MYB domains at the N-terminus (Figs 1E,
genes, called ‘male activator’ and ‘female suppressor’, S5 in Supporting Information). MSE1 belongs to the
located in the M locus of the Y chromosome (Marks R2R3-MYB class of proteins, which includes MYB
1973). transcription factors involved in metabolism, cell fate
and identity, development, and biotic and abiotic
stress responses (Stracke et al. 2001; Dubos et al.
Results and discussion 2010; Ambawat et al. 2013). To study the spatial and
To search for sex determination genes, a transcrip- temporal pattern of MSE1 expression, we measured
tome analysis was carried out during early develop- the levels of MSE1 mRNA in each tissue of male
ment of male and female flowers of A. officinalis cv. plants by quantitative RT-PCR. MSE1 mRNA was
Super Welcome. De novo assembly of 10.5 Gb of specifically expressed in small buds, but not in other
male paired-end sequences by Trinity (Grabherr et al. tissues (Fig. 1F). Detailed analysis of MSE1 expres-
2011) yielded 104 937 contigs and 51 525 unigenes sion in young buds showed that MSE1 was predomi-
(Fig. S1, Table S1 in Supporting Information). Map- nantly expressed in anther (Fig. 1G). These results
ping of 52.8 and 56.6 million reads from males and suggest that MSE1 acts in early stages of male organ
females against the assembled contigs, respectively, development.
showed that 149 contigs (114 unigenes) are expressed RNA and genome sequence mapping data of
in a male-biased manner. Because the previous tran- MSE1 transcripts suggested that the vestige of MSE1
scriptome analysis of A. officinalis failed to identify Y still exists on the X chromosome (Table S2 in Sup-
chromosome genes (Harkess et al. 2015), we carried porting Information). To test this hypothesis, whole-
out further screening by mapping each of 316 million genome sequencing and assembly was carried out on
reads from the male and female genome sequencings a male genome. BLAST searches against the assembly
against the 114 candidate genes. Ultimately, seven scaffolds showed four scaffolds with high sequence
contigs were obtained as candidates for male-specific similarity to MSE1 cDNA. One completely matched
genes (Table S2 in Supporting Information). To con- the genomic sequence of MSE1 in which male DNA
firm that the candidate genes were male-specific, we sequence reads were specifically mapped, and was
amplified them by polymerase chain reaction (PCR) judged to represent the MSE1 sequence on the Y
from bulked male and female genomes (Fig. S2 in chromosome (Figs 1H,S6 in Supporting Information).
Supporting Information). Only one gene, which we The other three scaffolds shared partial similarity with
named Male Specific Expression 1 (MSE1), showed the MSE1 genome sequence on the Y chromosome
male-specific amplification. This result was unex- (Fig. 1H). The high conservation of intergenic
pected because both female RNA and genome regions and introns of MSE1 between these scaffolds,
sequence reads mapped to the MSE1 contig, albeit at and the fact that they could be amplified from both

116 Genes to Cells (2017) 22, 115–123 © 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd
 Sex determination gene in asparagus

(A) (B) (F)

0.5

0.4

Relative expression
MSE1/Actin
0.3

0.2

0.1

0
(C)

Bud (mm)
MSE1 (G)
0.5

Relative expression
(D) 0.4
Male Female

MSE1/Actin
0.3
MSE1
0.2

0.1

(E) 0
1 276
R2 R3
MYB Bud (mm) 1.0–1.5 0.8–1.0 0.5–0.8

ATG TAG
(H) Scaffold_47312
MSE1 locus Y
97% 93% 90% 89% 92%
ATG
X 1 kbp
Y28→Stop

Scaffold_44689 Scaffold_51332
Scaffold_29228
PB_3
5 kbp

Figure 1 Characterization of the MSE1 gene. (A and B) Photographs of male (A) and female (B) flowers of Asparagus officinalis
cv. Super Welcome. Arrows show the vestiges of opposite sex organs. (C) RT-PCR analysis of full-length MSE1 using mRNA
extracted from early buds. (D) PCR amplification of MSE1 from genomic DNA extracted from male and female individuals of
A. officinalis cv. Super Welcome. (E) Domain structure of MSE1 protein. R2- and R3-type MYB domains are shown. (F and G)
Quantitative RT-PCR analysis of MSE1 expression using mRNA extracted from each tissue of A. officinalis cv. Super Welcome.
Expression levels were normalized against the corresponding levels of Actin. Means and SEs of three (F) and nine (G) replicates are
shown. (H) Comparison of the MSE1 locus between the X and Y chromosomes. Gray boxes show protein-coding regions. Dot-
ted lines show the genomic regions that share DNA sequence similarities. Scaffolds were extracted from the assembly data of Illu-
mina short-read sequences from a single male DNA of A. officinalis cv. Super Welcome. PB means a long-read sequence generated
by PacBio sequencer using female genome.

male and female genomes, suggested that these three is fragmented in at least 30-kb region of X chromo-
scaffolds represented X chromosome sequences. We some rather than MSE1Y is encoded within 2.5 kb
designated MSE1 on Y chromosome as MSE1Y and (Fig. 1H).
the putative MSE1 sequence on the X chromosome This result could explain the misamplification of
as MSE1X. Two scaffolds of MSE1X were assembled MSE1 genome fragment by PCR from female indi-
with a 19-kb PacBio sequence showing that MSE1X viduals because the amplicon is too large (Fig. 1D).

© 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd Genes to Cells (2017) 22, 115–123 117
 K Murase et al.

In A. officinalis cv. Super Welcome, three insertions, dioecious species from A. officinalis, have no deleteri-
five deletions and 16 point mutations are present in ous mutation in MSE1 coding regions (Figs 2A,S11
the coding region of MSE1X relative to MSE1Y in Supporting Information). This result suggests that
(Fig. S7 in Supporting Information). Some of these the origin of male mutation has occurred in outside
are likely to be deleterious mutations: a one-base of MSE1 coding region or these three species have
deletion at tyrosine 28 of MSE1Y induces a frame evolved in independent pathway.
shift, resulting in a premature stop codon; a deletion If MSE1 mutation is responsible for the transition
at the end of the second exon causes the loss of 26 from hermaphroditism to dioecy, artificial mutation
bases of protein-coding region and a splicing signal; of MSE1 orthologues in other plant species should
and a large deletion at the end of third exon also convert hermaphrodites into female plants. Phyloge-
causes a 200-bp deletion of protein-coding region netic analysis of MSE1-like MYB transcription factors
(Figs 1H,S7,S8 in Supporting Information). Various showed that MSE1 orthologues are widely conserved
mutations were observed among A. officinalis cultivars in monocot and dicot species, including the model
in MSE1X, but no SNPs were detected in MSE1Y, plant Arabidopsis thaliana (Fig. S12 in Supporting
suggesting that MSE1X is no longer under selection Information). The second most similar MYB protein
pressure to maintain its function (Fig. S7 in Support- in A. thaliana, AtMYB103, is outside the MSE1
ing Information). These results suggest that loss of clade, suggesting that ancestral MSE1 and AtMYB103
function of MSE1 has occurred on the X chromo- branched before the monocot–dicot divergence
some.
The Asparagus genus contains up to 300 species
distributed widely around the world (Kubitzki & (A)
Rudall 1998). Phylogenetic analysis of these species M F
A. officinalis
showed that the dioecious species form a single clade, cv. Super Welcome MSE1
suggesting that the evolutionary event leading from cv. Mary Washington MSE1
hermaphroditism to dioecy in Asparagus occurred cv. Pole Tom MSE1
only once (Fig. S9 in Supporting Information) (Kub-
ota et al. 2012). Therefore, if MSE1 acts as ‘male acti- A. pseudoscaber MSE1
vator’ in sex determination during male organ A. kiusianus MSE1
development, the system is likely to be conserved in
A. verticillatus MSE1
dioecious Asparagus species. To test whether the
MSE1 system is conserved in dioecious Asparagus spe-
cies, PCR amplification and sequencing of MSE1Y
and MSE1X from genomic DNA of the male and
female individuals were carried out. MSE1 genes
could be amplified from the genomes of all male M F M F M F M F
individuals, but not those from female individuals in
three cultivars of A. officinalis, Asparagus pseudoscaber, MSE1
Asparagus kiusianus, Asparagus schoberioides and Aspara-
gus verticillatus (Fig. 2A). Three conserved deleterious (B)
A. densiflous MSE1
mutations, which are caused by the frameshift muta-
tions, were observed in MSE1X sequences from the A. plumosus MSE1
female individuals of these related species (Fig. S7 in A. virgatus MSE1
Supporting Information). Furthermore, MSE1 could
A. asparagoides MSE1
also be amplified from all tested hermaphroditic spe-
cies, and the coding protein sequences were highly A. scandens MSE1
conserved among dioecious and hermaphroditic spe-
cies (Figs 2B,S10 in Supporting Information). These Figure 2 Conservation of MSE1 in genus Asparagus. PCR
results suggest that the arrest of male organ develop- amplification of MSE1 from male (M) and female (F) individ-
ment in female flowers in these Asparagus species is uals of Asparagus officinalis three cultivars and dioecious (A) and
caused by loss of MSE1 function. Interestingly, eight individuals of hermaphroditic species (B) in genus
Asparagus acutifolius, Asparagus stipularis and Asparagus Asparagus. Their sexes were determined by male-specific PCR
cochinchinensis, which are phylogenetically most distant markers or flower phenotypes.

118 Genes to Cells (2017) 22, 115–123 © 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd
 Sex determination gene in asparagus

(Fig. S12 in Supporting Information). The conserva- Washington from Takii Seed (Kyoto); cv. Niagara from Nitto
tion of MSE1 is assumed to reflect the functional Nousan (Yokohama); and A. pseudoscaber, A. verticillatus,
importance of this gene in the life cycle of flowering A. densiflorus and A. virgatus from B & T World Seeds
plants. The MSE1 orthologue of A. thaliana is (Aigues-Vives). A. plumosus and A. asparagoides were purchased
from public garden centers. A. kiusianus was collected from
AtMYB35/TDF1 (Tapetal Development and Function
Keyakaigan (Fukuoka, Japan). A. schoberioides, A. acutifolius,
1), which is essential for normal anther development A. stipularis and A. cochinchinensis are described in Kubota et al.
(Fig. S12 in Supporting Information) (Zhu et al. 2012.
2008). Because a T-DNA insertion line was not
available, genome editing knockouts of TDF1 were
produced using the CRISPR/Cas9 system targeting Transcriptome analysis
three sites in the TDF1 gene (Fig. 3A,B). Each trans- Total RNAs were extracted from early developmental buds
formant exhibited normal vegetative growth and (0.5–0.8 mm) of male and female A. officinalis cv. Super Wel-
flowering, but seedless siliques (Fig. 3C–F). No pol- come using the RNeasy Plant Mini Kit (Qiagen, Tokyo,
len grains were observed in the transformants, sug- Japan). Library preparations and RNA sequencing by HiSeq
gesting that the sterility is caused by a defect in male 2000 (Illumina, San Diego, CA, USA) were outsourced to
organ development (Fig. 3G,H). These features are Hokkaido System Science (Hokkaido). Adapter sequences
consistent with the previously reported phenotype of were removed from raw sequence data using the cutadapt pro-
gram (Martin 2011). Next, four bases of 30 -terminal sequences
the tdf1 mutant (Zhu et al. 2008). These results sup-
of the treated sequences were removed using FastX-Toolkit
port the idea that MSE1 functions in male organ
<http://hannonlab.cshl.edu/fastx_toolkit/index.html>. De novo
development. assembly of RNA sequences from males was carried out using
Charles Darwin considered male (or female) organ Trinity with default parameters except that minimum k-mer
abortion to be the first step in the evolution of coverage was set to 3 (Grabherr et al. 2011). Mapping of
dioecy (Darwin 1877). Our results strongly suggest RNA and genome sequences from males and females were
that MSE1 acts as the ‘male activator’ in A. officinalis conducted using bowtie (Langmead et al. 2009). Mapping data
sex determination and that the loss-of-function muta- were processed by SAMtools (Li et al. 2009) and visualized by
tion in MSE1X was an important step in the evolu- the INTEGRATIVE GENOMICS VIEWER (IGV) software (Robinson
tion of dioecy in Asparagus. Although gynodioecious et al. 2011). Gene functions were annotated by BLAST2GO
(or androdioecious) Asparagus species have not been (Conesa et al. 2005).
identified to date, our data may provide the first
molecular evidence that this species evolved via the RT-PCR and real-time quantitative RT-PCR
gynodioecy (or androdioecy) by the mutation of gene Total RNA was extracted from each tissue as described above.
involved in male organ development pathway (Char- Reverse transcription was carried out using SuperScript III
lesworth & Charlesworth 1978). The MSE1 system is (Invitrogen, Carlsbad, CA, USA). MSE1 cDNA was PCR-
clearly distinct from the OGI–MeGI system involved amplified using Ex Taq (Takara, Kusatsu, Japan) with primers
in persimmon sex determination, in which a small 646 (50 -GATCGGATCCATGGGCAGGCCTCCATGCTGC
RNA acts as the sex determination factor (Akagi GA-30 ) and 647 (50 -GATCGAATTCCTACAGCAAATCAT
et al. 2014). Divergent molecular mechanisms have AAAAAAACTCAGG-30 ), which were designed to amplify
been described in the self-incompatibility system, full-length MSE1.Real-time quantitative RT-PCR was carried
which is also involved in preventing self-fertility in out on a LightCycler 96 system (Roche, Tokyo, Japan) using
flowering plants (Takayama & Isogai 2005). It will be QuantiFast SYBR Green RT-PCR Kit (Qiagen). MSE1 and
Actin were amplified by primers AoMSE1realtimeFw (50 -
interesting to compare the molecular mechanisms and
GCCCTAATTTGAAGCATGAGAG-30 )/AoMSE1realtimeRv
evolution of these systems. Our results will contribute (50 -GATTTGAGAGATGGGTTGTG-30 ) and AoActin1F (50 -
to the understanding of the molecular mechanisms of GTTCCTGCTCATAATCTAGAGCAAC-30 )/AoActin1R (50 -
sex determination, as well as the evolution of dioecy CTTCTCACTGAGGCTCCACTCAAC-30 ), respectively.
from hermaphroditism in flowering plants. MSE1 expression was normalized against expression of Actin.

Experimental procedures Linkage analysis and amplification of MSE1 from

genus Asparagus
Plant materials
For linkage analysis of MSE1, seeds of A. officinalis cv. Super
Asparagus officinalis cv. Super Welcome and Pole Tom were Welcome were treated with n-propyl N-(3,4-dichlorophenyl)
purchased from SAKATA SEED (Yokohama), cv. Mary carbamate (NPC) to induce early flowering as described

© 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd Genes to Cells (2017) 22, 115–123 119
 K Murase et al.

(A) (B)
TDF1
1 1146

GE1 GE2 GE3

(C) (D) (E) (F)

(G) (H)

Figure 3 Genome editing of MSE1 orthologue, TDF1, in the model plant Arabidopsis thaliana. (A) Genomic region of the TDF1
locus in A. thaliana. Gray boxes show protein-coding regions. Arrows show target sites (GE1, position 32–10 in MSE1 cDNA;
GE2, 463–485; GE3, 610–588) of the guide RNA–CRISPR/Cas9 complex. (B) Transgenes were amplified from genomic DNA
of each transformant with primers for amplifying the guide RNA region. (C–F) Phenotypes of transgenic plants obtained by
TDF1 genome editing. The transformants were designated as tdf1-GE1 to 3. Photographs show 2- to 3-week-old shoots of wild-
type (Col-0) (C), tdf1-GE1 (D), tdf1-GE2 (E) and tdf1-GE3 (F). Arrows show siliques. Bars, 1 cm. (G and H) Male-sterile pheno-
type of tdf1-GE3. Pollen grains were observed on anther and stigma in wild type (G), but not in tdf1-GE3 (H). Bars, 1 mm.

previously (Aneja et al. 1999). Genomes of each individual TGGTCGGTAATCGCACATCACCTCC-30 ) and 647. To
were extracted by Plant DNAzol Reagent (Invitrogen). MSE1 sequence the coding region of MSE1 from other A. officinalis
fragments were amplified using ExTaq with primers 706 (50 - cultivars and Asparagus species, primers U10 (50 -

120 Genes to Cells (2017) 22, 115–123 © 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd
 Sex determination gene in asparagus

AATTGGTTTCATCATCATTGTACCTCAG-30 ) and U21 Beverly, MA, USA) and then used to create two libraries
(50 -CTAAGATCCCAACGCACAAAC-30 ) were used for using the TruSeq DNA PCR-Free Sample Preparation Kit
PCR amplification. Sequencing was carried out using BigDye (Illumina) with insert sizes of ~180 and ~800 bp. These
Terminator v3.1 (Applied Biosystems, Foster City, CA, USA). libraries were sequenced on the Illumina HiSeq platform
To amplify MSE1 fragments from genomes of each individual using a 2x 101-nt paired-end sequencing protocol. The reads
of A. officinalis and related species, we used primer sets 646– were cleaned up with cutadapt. Low-quality ends (<QV20)
647 or 646–648 (50 -GATCGAATTCCTAGGCTAGAGTG and adapter sequences were trimmed, and reads shorter than
GTGATGGTTTCCTTG-30 ). For amplification of MSE1Y 50 bp were discarded. Total sequence of 105.3 Gb (~849
from A. kiusianus, 706 and 647 primers were used. For male- coverage of the genome, assuming a genome size of
specific marker, the Asp1-T7 primer set was used as described 1.26 Gb) was generated from the libraries and then assembled
previously (Jamsari et al. 2004). To check the sex genotypes of using the ALLPATHS-LG assembler (Gnerre et al. 2011).
A. verticillatus individuals, newly developed male-specific mar- The assembly yielded 146 894 scaffolds with an N50 length
ker, designated MSM1 (male-specific marker 1), was created of 5.2 kb. For transcriptome analysis, 316 million reads of
from male-specific scaffold in the genome assembly. For ampli- male and female genome sequences were used for mapping.
fication of MSM1, 814 (50 -CAACTCCAGGTGACAACATT Long-read sequences were generated by PacBio RS II
CATAG-30 ) and 805 (50 -TCGTCAACGTCGACTGCAGG sequencer (Pacific Biosciences, Menlo Park, CA, USA) with
TAGGC-30 ) primers were used. MSE1X fragments were ampli- a 20-kb DNA library prepared from the female asparagus.
fied by primers 752 (50 -ATTGGTTTCATCATCATTGTA Total sequence of 3.72 Gb in 372 292 reads was obtained
CCTC-30 ) and 754 (50 -TTGCCTGTCCATCTCACTTCTG from eight SMRT cells. The N50 length was 13 054 bp.
GAT-30 ) for the first and second exons, and 755 (50 -CTAACC Sequences containing MSE1 locus were searched by BLAST
ATGATCTACACACGATCAC-30 ) and 757 (50 -CCCTTCG program (Camacho et al. 2009).
ACGTGGATTAATCGCTACC-30 ) for the third exon.

Genome editing of MSE1 orthologue in A. thaliana

Phylogenetic analysis
Genome editing of the MSE1 orthologue TDF1 was carried
Protein sequences of the MYB transcription factors were mul- out using the binary vector pEgP226-2A-gfbsd (Osakabe et al.
tiply aligned using Clustal Omega (Sievers et al. 2011), with 2016), which was designed for CRISPR/Cas9 and guide
the Myb_DNA-binding domain HMM matrix (Accession No. RNA-mediated genome editing. Three primer sets [735 (50 -G
PF00249 under Pfam database (Finn et al. 2014) used as the ATTTTGGACTTGTCACAACAAGG-30 )—736 (50 -AAACC
external profile HMM). Conserved selection blocks from the CTTGTTGTGACAAGTCCAA-30 ) for GE1, 737 (50 -GA
alignment were selected using Gblocks (Talavera & Castresana TTTCCATTGCACGAAAGCTTCC-30 )—738 (50 -AAACGG
2007) with default parameters. Phylogenetic tree was con- AAGCTTTCGTGCAATGGA-30 ) for GE2 and 739 (50 -GA
structed based on Bayesian inference using the MRBAYES 3.2.2 TTTAATGTTTCTGAATTCTGCA-30 )—740 (50 -AAACTG
program (Ronquist et al. 2012), using HsMYB as the out- CAGAATTCAGAAACATTA-30 ) for GE3] were annealed to
group sequence. Four chains of the Metropolis-coupled Mar- serve as guide RNA-targeting sequences. The annealed DNA
kov Chain Monte Carlo processes were run for 1 000 000 fragments were subcloned into the BsaI site of pEgP226-2A-
generations, with trees sampled every 1000 generations. The gfbsd. Transformation of A. thaliana (Col-0) was carried out
first 25% of trees were discarded, and the remaining trees were by floral dip method (Clough & Bent 1998) using Agrobac-
used to support the majority-rule consensus tree topology with terium (pMP90) harboring these binary vectors. Transformants
posterior probabilities. For the Asparagus genus taxon phy- were screened in Murashige–Skoog (MS) medium containing
logeny, five chloroplast intergenic sequences (Kubota et al. 0.6% agar and 60 lg/mL kanamycin (Murashige & Skoog
2012) were aligned using MAFFT (Katoh & Standley 2013) 1962) and then transferred into soil. Transformants were con-
and concatenated. The alignment was cleaned by Gblocks and firmed by genomic PCR with a primer set, GEF (50 -ACTTA
subjected to MrBayes 3.2.2 analysis as described above, using AGACCTGACC-30 ) and GER (50 -GATGATTTGGATG
the C. stricta sequence as out-group, to yield the consensus GC-30 ).
tree.

Acknowledgements
Genome sequencing and assembly
We thank A. Takahashi, F. Kodama, H. Asao and A. Akita
For whole-genome assembly, DNA from a single male for technical assistance, and T. Nishimoto and H. Asao for a
asparagus was used for genome sequencing. For screening of kind gift of 3-year-old A. officinalis cv. Super Welcome. Com-
male-specific genes in transcriptome analysis, the female gen- putational resources were provided by the Data Integration
ome was also used. Genome DNA was extracted using the and Analysis Facility, NIBB. This work was supported in part
DNeasy Plant Mini Kit (Qiagen). The purified DNA was by Grants-in-Aid for Scientific Research on Innovative Areas
fragmented on a Covaris S2 sonicator (Covaris, Woburn, (23113002, 16H06467 to S.T.), and Grants-in-Aid for Scienti-
MA, USA), size-selected with Pippin Prep (Sage Science, fic Research (25252021, 16H06380 to S.T.) from the Ministry

© 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd Genes to Cells (2017) 22, 115–123 121
 K Murase et al.

of Education, Culture, Sports, Science and Technology of annotation, visualization and analysis in functional genomics
Japan (MEXT), Takeda Science Foundation (to K.M.), and research. Bioinformatics 21, 3674–3676.
the Council for Science, Technology and Innovation (CSTI), Darwin, C. (1877) The Different Forms of Flowers on Plants of
Cross-ministerial Strategic Innovation Promotion Program the Same Species. London, UK: John Murray.
(SIP), ‘Technologies for creating next-generation agriculture, Dubos, C., Stracke, R., Grotewold, E., Weisshaar, B., Martin,
forestry and fisheries’ (funding agency: Bio-oriented Technol- C. & Lepiniec, L. (2010) MYB transcription factors in Ara-
ogy Research Advancement Institution, NARO) to K.O. A bidopsis. Trends Plant Sci. 15, 573–581.
Grant-in-Aid for Science and technology research promotion Finn, R.D., Bateman, A., Clements, J., Coggill, P., Eberhardt,
program for agriculture, forestry, fisheries and food industry R.Y., Eddy, S.R., Heger, A., Hetherington, K., Holm, L.,
(No. 27002B) from the Ministry of Agriculture, Forestry and Mistry, J., Sonnhammer, E.L., Tate, J. & Punta, M. (2014)
Fisheries (MAFF) (Japan) (to A.K. and Y. Ozaki). S.T. super- Pfam: the protein families database. Nucleic Acids Res. 42,
vised this project. K.M. and S.T. designed the experiments. D222–D230.
K.M., K.U., T.M. and A.S. carried out molecular experi- Gnerre, S., Maccallum, I., Przybylski, D. et al. (2011) High-
ments, with assistance from Y.W. S.F. carried out phyloge- quality draft assemblies of mammalian genomes from mas-
netic analysis. Y. Osakabe and K.O. were responsible for sively parallel sequence data. Proc. Natl Acad. Sci. USA 108,
genome editing constructs. Y. Ozaki and A. K. collected and 1513–1518.
maintained the plant materials, observed the sex phenotypes Grabherr, M.G., Haas, B.J., Yassour, M. et al. (2011) Full-
and extracted the DNA. S.S. and K.Y. carried out genome length transcriptome assembly from RNA-Seq data without
sequencing and assembly. K.M. and S.T. wrote the manu- a reference genome. Nat. Biotechnol. 29, 644–652.
script, helped by S.S. and S.F., and all other authors con- Harkess, A., Mercati, F., Shan, H.Y., Sunseri, F., Falavigna,
tributed to editing. The MSE1 sequence and high-throughput A. & Leebens-Mack, J. (2015) Sex-biased gene expression
sequencing data used in this study have been deposited in in dioecious garden asparagus (Asparagus officinalis). New Phy-
DNA Data Bank of Japan (DDBJ) under accession numbers tol. 207, 883–892.
LC190965, SAMD00047009, SAMD00047010, SAMD Jamsari, A., Nitz, I., Reamon-B€ uttner, S.M. & Jung, C.
00047011, SAMD00047624 and SAMD00064393. (2004) BAC-derived diagnostic markers for sex determina-
tion in asparagus. Theor. Appl. Genet. 108, 1140–1146.
Katoh, K. & Standley, D.M. (2013) MAFFT multiple
References
sequence alignment software version 7: improvements in
Akagi, T., Henry, I.M., Tao, R. & Comai, L. (2014) A Y- performance and usability. Mol. Biol. Evol. 30, 772–780.
chromosome-encoded small RNA acts as a sex determinant Kubitzki, K. & Rudall, P.J. (1998) “Asparagaceae” in The Fami-
in persimmons. Science 346, 646–650. lies and Genera of Vascular Plants, Vol. 3, pp. 125–129. New
Ambawat, S., Sharma, P., Yadav, N.R. & Yadav, R.C. (2013) York, NY: Springer.
MYB transcription factor genes as regulators for plant Kubota, S., Konno, I. & Kanno, A. (2012) Molecular phylogeny
responses: an overview. Physiol. Mol. Biol. Plants 19, 307– of the genus Asparagus (Asparagaceae) explains interspecific
321. crossability between the garden asparagus (A. officinalis) and
Aneja, M., Gianfagna, T.J., Garrison, S.A. & Durner, E.F. other Asparagus species. Theor. Appl. Genet. 124, 345–354.
(1999) Rapid sex-typing of Asparagus for male hybrid seed Langmead, B., Trapnell, C., Pop, M. & Salzberg, S.L. (2009)
production using n-propyl N-(3,4-dichlorophenyl)carbamate Ultrafast and memory-efficient alignment of short DNA
(NPC). HortScience 34, 1090–1094. sequences to the human genome. Genome Biol. 10, R25.
Camacho, C., Coulouris, G., Avagyan, V., Ma, N., Papado- Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J.,
poulos, J., Bealer, K. & Madden, T.L. (2009) BLAST+: Homer, N., Marth, G., Abecasis, G. & Durbin, R.; 1000
architecture and applications. BMC Bioinformatics 10, 421. Genome Project Data Processing Subgroup. (2009) The
Caporali, E., Carboni, A., Galli, M.G., Rossi, G., Spada, A. & Sequence Alignment/Map format and SAMtools. Bioinfor-
Marziani Longo, G.P. (1994) Development of male and matics 25, 2078–2079.
female flower in Asparagus officinalis. Search for point of L€optien, H. (1979) Identification of the sex chromosome pair
transition from hermaphroditic to unisexual developmental in asparagus (Asparagus officinalis L.). Z. Pflanzenz 82, 162–
pathway. Sex. Plant Reprod. 7, 239–249. 173.
Charlesworth, B. & Charlesworth, D. (1978) A model for the Marks, M. (1973) A reconsideration of the genetic mechanism
evolution of dioecy and gynodioecy. Am. Nat. 112, 975–997. for sex determination in Asparagus officinalis. In: Proc.
Charlesworth, D. (2002) Plant sex determination and sex EUCARPIA Meeting on Asparagus (Asparagus Officinalis L.),
chromosomes. Heredity 88, 94–101. pp. 123–128. Wageningen, Netherlands: EUCARPIA.
Clough, S.J. & Bent, A.F. (1998) Floral dip: a simplified Martin, M. (2011) Cutadapt removes adapter sequences from
method for Agrobacterium-mediated transformation of Ara- high-throughput sequencing reads. EMBnet J. 17, 10–12.
bidopsis thaliana. Plant J. 16, 735–743. Murashige, T. & Skoog, F. (1962) A revised medium for rapid
Conesa, A., G€ otz, S., Garcıa-Gomez, J.M., Terol, J., Tal
on, growth and bio assays with tobacco tissue cultures. Physiol.
M. & Robles, M. (2005) Blast2GO: a universal tool for Plant. 15, 473–497.

122 Genes to Cells (2017) 22, 115–123 © 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd
 Sex determination gene in asparagus

Osakabe, Y., Watanabe, T., Sugano, S.S., Ueta, R., Ishihara, Supporting Information
R., Shinozaki, K. & Osakabe, K. (2016) Optimization of
CRISPR/Cas9 genome editing to modify abiotic stress Additional Supporting Information may be found online in
responses in plants. Sci. Rep. 6, 26685. the supporting information tab for this article:
Renner, S.S. (2014) The relative and absolute frequencies of
Figure S1 Summary of screening for male-specific genes in
angiosperm sexual systems: dioecy, monoecy, gynodioecy,
A. officinalis.
and an updated online database. Am. J. Bot. 101, 1588–
1596. Figure S2 PCR amplification of male-specific gene candidates
Robinson, J.T., Thorvaldsd ottir, H., Winckler, W., in male and female genomes.
Guttman, M., Lander, E.S., Getz, G. & Mesirov, J.P.
(2011) Integrative genomics viewer. Nat. Biotechnol. 29, Figure S3 Female RNA sequence fragments map to the 50
24–26. terminus of MSE1 cDNA.
Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Figure S4 MSE1 is tightly linked to the Y chromosome in
Darling, A., H€ ohna, S., Larget, B., Liu, L., Suchard, M.A. A. officinalis.
& Huelsenbeck, J.P. (2012) MrBayes 3.2: efficient Bayesian
phylogenetic inference and model choice across a large Figure S5 Alignment of predicted amino-acid sequences of
model space. Syst. Biol. 61, 539–542. MSE1 and other homologues.
Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus,
Figure S6 Male DNA sequence reads are specifically mapped
K., Li, W., Lopez, R., McWilliam, H., Remmert, M.,
to the Scaffold_47312.
S€
oding, J., Thompson, J.D. & Higgins, D.G. (2011) Fast,
scalable generation of high-quality protein multiple Figure S7 Alignment of DNA sequences of MSE1Y and
sequence alignments using Clustal Omega. Mol. Syst. Biol. MSE1X in A. officinalis.
7, 539.
Stracke, R., Werber, M. & Weisshaar, B. (2001) The R2R3- Figure S8 One-base deletion creates a stop codon in MSE1X
MYB gene family in Arabidopsis thaliana. Curr. Opin. Plant in A. officinalis.
Biol. 4, 447–456.
Figure S9 Phylogeny of genus Asparagus.
Takayama, S. & Isogai, A. (2005) Self-incompatibility in
plants. Annu. Rev. Plant Biol. 56, 467–489. Figure S10 Alignment of predicted amino-acid sequences of
Talavera, G. & Castresana, J. (2007) Improvement of phyloge- MSE1 orthologues in asparagus and hermaphroditic species.
nies after removing divergent and ambiguously aligned
blocks from protein sequence alignments. Syst. Biol. 56, Figure S11 Alignment of predicted amino-acid sequences of
564–577. MSE1 orthologues from male and female individuals in three
Telgmann-Rauber, A., Jamsari, A., Kinney, M.S., Pires, J.C. dioecious species that have no deleterious mutation in MSE1
& Jung, C. (2007) Genetic and physical maps around the coding region.
sex-determining M-locus of the dioecious plant asparagus. Figure S12 Phylogeny of MSE1-related MYB domain tran-
Mol. Genet. Genomics 278, 221–234. scription factors.
Zhu, J., Chen, H., Li, H., Gao, J.F., Jiang, H., Wang, C.,
Guan, Y.F. & Yang, Z.N. (2008) Defective in Tapetal develop- Table S1 Summary of transcriptome analysis of developing A.
ment and function 1 is essential for anther development and officinalis flowers
tapetal function for microspore maturation in Arabidopsis.
Table S2 List of contigs enriched in male genome
Plant J. 55, 266–277.
sequencings
Received: 11 October 2016
Accepted: 26 October 2016

© 2016 Molecular Biology Society of Japan and John Wiley & Sons Australia, Ltd Genes to Cells (2017) 22, 115–123 123