Discovery of a novel fragrant allele and development of functional markers for fragrance in rice

Abstract

Fragrance in rice (Oryza sativa L.) results from the loss of function of the betaine aldehyde dehydrogenase (Badh2) gene on chromosome 8. An 8-bp deletion in exon 7 of Badh2 was reported to be the main allele functionally associated with fragrance. The discovery of new functional alleles will provide additional genetic resources to improve fragrant rice. In this study, we sequenced the Badh2 gene in 30 rice accessions and filtered the Badh2 polymorphisms from whole-genome re-sequence data of 295 rice accessions. Seven alleles were detected from the sequence data. Six of the seven were known alleles and one was a novel allele (badh2-E12). The novel allele was a 3-bp deletion in exon 12. Five functional markers, targeting six of the seven alleles, were identified. Fourteen accessions were selected to test the utility of these markers. The five molecular markers reliably distinguish this fragrant rice from other fragrant or non-fragrant rice accessions. Analysis of two F2 rice population validated the genetic markers FME12-3 and FME14I as functional markers. These two markers co-segregated with the fragrance phenotype. These markers will be used in a Badh2 diversity study and to breed improved fragrant rice accessions via marker-assisted selection.

Full text

Discovery of a novel fragrant allele and development
of functional markers for fragrance in rice
Qiang He .Yong-Jin Park
Received: 23 April 2015 / Accepted: 5 November 2015 / Published online: 11 November 2015
ÓSpringer Science+Business Media Dordrecht 2015
Abstract Fragrance in rice (Oryza sativa L.) results
from the loss of function of the betaine aldehyde
dehydrogenase (Badh2) gene on chromosome 8. An
8-bp deletion in exon 7 of Badh2 was reported to be the
main allele functionally associated with fragrance.
The discovery of new functional alleles will provide
additional genetic resources to improve fragrant rice.
In this study, we sequenced the Badh2 gene in 30 rice
accessions and filtered the Badh2 polymorphisms
from whole-genome re-sequence data of 295 rice
accessions. Seven alleles were detected from the
sequence data. Six of the seven were known alleles
and one was a novel allele (badh2-E12). The novel
allele was a 3-bp deletion in exon 12. Five functional
markers, targeting six of the seven alleles, were
identified. Fourteen accessions were selected to test
the utility of these markers. The five molecular
markers reliably distinguish this fragrant rice from
other fragrant or non-fragrant rice accessions. Anal-
ysis of two F2 rice population validated the genetic
markers FME12-3 and FME14I as functional markers.
These two markers co-segregated with the fragrance
phenotype. These markers will be used in a Badh2
diversity study and to breed improved fragrant rice
accessions via marker-assisted selection.
Keywords Fragrant rice Novel allele Marker-
assisted selection
Introduction
Rice (Oryza sativa L.) is the most important staple
food crop for approximately half of the world’s
population. The demand for fragrant rice has increased
markedly in recent years, due to its favorable flavor
(Myint et al. 2012; Bradbury et al. 2005b). The
premium price and high consumer acceptance of
fragrant rice have received increasing attention in rice-
producing countries (Shi et al. 2014; Shao et al. 2013).
More than 200 volatiles have been identified in rice
(Mahattanatawee and Rouseff 2014; Champagne
2008). 2-Acetyl-1-pyrroline (2-AP) is widely known
as the main volatile in scented rice, which produces a
‘‘popcorn-like’’ odor that could be detected at very low
concentrations (Buttery et al. 1982; Mathure et al.
2014; Mahattanatawee and Rouseff 2014). Several
sensory methods have been developed for researchers
and breeders to select the scented trait in rice. The two
Electronic supplementary material The online version of
this article (doi:10.1007/s11032-015-0412-4) contains supple-
mentary material, which is available to authorized users.
Q. He Y.-J. Park (&)
Department of Plant Resources, College of Industrial
Science, Kongju National University, Yesan 340-702,
Republic of Korea
e-mail: yjpark@kongju.ac.kr
Y.-J. Park
Legume Bio-Resource Center of Green Manure
(LBRCGM), Kongju National University,
Yesan 340-702, Republic of Korea
123
Mol Breeding (2015) 35:217
DOI 10.1007/s11032-015-0412-4

most common methods are tasting rice grain and
smelling KOH-incubated leaves or grains (Bradbury
et al. 2005b; Sood and Siddiq 1978). However, these
two sensory methods are labor-intensive and often
inaccurate (Bradbury et al. 2005b).
The fragrance trait in rice was reported to be
controlled by one to three dominant or recessive genes
or by quantitative trait loci (QTLs) (Lorieux et al. 1996;
Amarawathi et al. 2008). Studies have identified fgr as a
recessive gene located on chromosome 8 (Ahn et al.
1992;Yanoetal.1991; Tragoonrung et al. 1996;
Lorieux et al. 1996; Garland et al. 2000; Cordeiro et al.
2002; Jin et al. 2003). Bradbury et al. (2005a) reported
that fgr encodes betaine aldehyde dehydrogenase
homologue 2 (Badh2) on chromosome 8. The gene fgr
is a fragrance-related gene, with an 8-bp deletion in exon
7. Chen et al. (2006)alsofoundfgr within 69 kb on
chromosome 8 and cloned the gene (Chen et al. 2008).
Results from these independent studies suggested that
variation of the fgr gene, leading to functional lossof the
Badh2 protein, is associated with fragrance in rice
(Bradbury et al. 2005a;Chenetal.2006,2008). This
hypothesis was supported by results from studies that
employed RNA interference (RNAi) and transcription
activator-like effector nuclease (TALENs) technology
to knock out the fgr gene (Chen et al. 2008; Niu et al.
2008;Shanetal.2013). The fgr gene encodes the Badh2
enzyme and is usually referred to as badh2.
After the function of Badh2 gene was reported, the
fragrant rice researchers set off a new study rush on
Badh2. First of all, novel Badh2 alleles have been
identified successively from studies that focused on
the Badh2 gene. A 7-bp deletion in exon 2 and a 7-bp
insertion in exon 8 were detected in 2008 (Shi et al.
2008; Amarawathi et al. 2008). Kovach et al. (2009)
sequenced badh2 in 280 O. rufipogon and 242 O.
sativa accessions and detected an additional eight
alleles in exons 1, 10, 13 and 14. Shao et al. (2011)
found a large deletion (803-bp) between exons 4 and 5.
Shao et al. (2013) discovered one single nucleotide
polymorphism (SNP) in exons 10-, 75- and 806-bp
deletions in exon 2 and between exons 4 and 5,
respectively. Shi et al. (2014) discovered a 3-bp
deletion in the 50UTR and an 8-bp insertion in the
promoter. Lately, one SNP at the exon1–intron1
junction, as a splicing donor, was detected in Japanese
fragrant rice landraces (Ootsuka et al. 2014).
The function of Badh2 has also been studied
extensively. In addition to fragrance in rice (Niu
et al. 2008; Shan et al. 2013; Chen et al. 2008), Badh2
has also been associated with salt tolerance, since the
fragrance is associated with reduced yield in response
to high-salt treatment (Fitzgerald et al. 2010). By
establishing the single origin of the badh2.1 allele (8-
bp deletion in exon 7) within the Japonica group,
Kovach et al. (2009) demonstrated that badh2 is a
domesticated gene. Kovach et al. (2009) proved that
this allele was introgressed from Japonica to Indica
rice. This understanding of the relationship between
badh2 and fragrance will be useful in identifying new
alleles for fragrant rice breeding programs.
Introducing fragrance into elite rice accessions will
result in rice with a high market value. In a typical rice
breeding program, fragrance is measured on a per
plant basis and in early generations (Jin et al. 2010).
The sensory approach to improving fragrance is often
time-consuming, expensive and unreliable. Recently,
gas chromatography–mass spectrometry (GC–MS)
and e-nose (electronic nose) have been used to
distinguish fragrant and non-fragrant rice and
improved the reliability. While these two approaches
were more expensive, breeders require an easy and
effective method of distinguishing fragrant and non-
fragrant rice. With the rapid development of various
types of DNA marker, MAS has played a prominent
role in plant breeding (Wang et al. 2011). MAS has
also been widely used in fragrant rice breeding, due to
the convenience and accuracy of this selection
method. Numerous markers have been developed for
fragrant rice breeding. Among the 17 reported badh2
alleles, 7 have been used in MAS (Table 1; Bradbury
et al. 2005b; Shi et al. 2008,2014; Sakthivel et al.
2009; Shao et al. 2011; Dissanayaka et al. 2014;
Ootsuka et al. 2014). Five of these seven functional
markers were tested in a segregating population
(Table 1).
In this study, we aimed to find novel alleles of
Badh2 by screening 325 rice accessions and develop
functional markers to select favorable alleles in a
fragrant rice breeding program.
Materials and methods
Plant materials
PCR-based sequencing was applied to 25 fragrant and
5 non-fragrant rice accessions (Table S1). Whole-
217 Page 2 of 10 Mol Breeding (2015) 35:217
123

genome re-sequencing was performed on 295 rice
accessions (Table S1). These accessions were planted
during the 2012 rice-growing season at Kongju
National University. During 2013, an F2 population
was generated from the cross between ‘‘Dongji-wx’’
and ‘‘Mongdonjaerae’’ (3-bp deletion in exon 12), and
the seeds were planted in 2014. Another F2 population
was generated from the cross between non-fragrant
rice and the accession A30 (1-bp insertion in exon 14).
Fragrance determination
The aroma of young leaves was determined by a
sensory evaluation panel, according to the method of
Sood and Siddiq (1978). About 2 g of young leaves
were placed into a 15-ml tube and mixed with 10 ml of
1.7 % KOH solution. The tubes were placed in a 30 °C
water bath, incubated for 10 min, and smelled by
multiple individuals to estimate fragrance.
Genotyping for Badh2 gene
Ten primer sets, designed by Shi et al. (2008), were
used to amplify the whole sequence of Badh2 in 30
rice accessions. HiSeq 2000 and HiSeq 2500 were
employed for the whole-genome re-sequencing of the
other 295 germplasms. The total mapping depth was
about 9X in average. The polymorphisms of Badh2 for
295 germplasms were purified from whole-genome re-
sequencing data by mapping to the reference genome
IRGSP-1.0 (http://rapdb.dna.affrc.go.jp/download/
archive/irgsp1/IRGSP-1.0_genome.fasta.gz). For
Badh2, SNPs and InDels were identified with C3X of
read depth coverage.
Table 1 Summary of the exon polymorphisms in Badh2
Alleles Location Sequence
divergences
Marker
development
Segregation
test
References
badh2-p-50
UTR
50UTR 3-bp deletion N – Shi et al. (2014)
badh2-E1.1 Exon 1 2-bp deletion N – Kovach et al. (2009)
badh2-E1.2 Exon 1–intron 1
junction
G/A snp Y Y Ootsuka et al. (2014)
badh2-E2.1 Exon 2 7-bp deletion Y Y Shi et al. (2008)
badh2-E2.2 Exon 2 75-bp deletion N – Shao et al. (2013)
badh2-E4-
5.1
Exon 4 to exon 5 806-bp deletion N – Shao et al. (2013)
badh2-E4-
5.2
Exon 4 to exon 5 803-bp deletion Y N Shao et al. (2011)
badh2-E7 Exon 7 8-bp deletion Y Y Bradbury et al. (2005a,b), Shi et al.
(2008)
badh2-E8 Exon 8 7-bp insertion N – Amarawathi et al. (2008)
badh2-E10.1 Exon 10 1-bp insertion N – Kovach et al. (2009)
badh2-E10.2 Exon 10 1-bp deletion N – Kovach et al. (2009)
badh2-E10.3 Exon 10 G/T snp N – Kovach et al. (2009)
badh2-E10.4 Exon 10 G/A snp N – Shao et al. (2013)
badh2-E12 Exon 12 3-bp deletion Y Y
a
badh2-E13.1 Exon 13 3-bp insertion Y Y Kovach et al. (2009)
badh2-E13.2 Exon 13 C/T snp N – Kovach et al. (2009)
badh2-E14.1 Exon 14 1-bp insertion Y N Kovach et al. (2009)
badh2-E14.2 Exon 14 G/T snp N – Kovach et al. (2009)
a
Discovered in this study
Mol Breeding (2015) 35:217 Page 3 of 10 217
123

Table 2 Functional alleles detected in this study
Alleles Sequence divergences Sample ID Fragrant
3-bp deletion
in 50UTR
7-bp deletion
in exon 2
8-bp deletion
in exon 7
3-bp deletion
in exon 12
3-bp insertion
in exon 13
One C/T SNP
in exon 13
1-bp insertion
in exon 14
Badh2
a
-–––– – – A01–A05 –
badh2-UTR-
E7
?–?– – – – A09, A10 ?
badh2-E2 –?– – – – – A06, A07, A08 ?
badh2-E7 ––?– – – – A11 *A23, RWG-042, RWG-061,
RWG-064, RWG-069, RWG-179,
RWG-201, RWG-236, RWG-272,
RWG-295
?
badh2-E12 –––?– – – RWG-031 ?
badh2-E13 ––––?– – A24, A25 ?
badh2-E13.2 ––––– ?– A26, A27, A28, RWG-088, RWG-
191, RWG-197
?
badh2-E14 ––––– – ?A29, A30 ?
?/-: Yes/No
a
Wide type
217 Page 4 of 10 Mol Breeding (2015) 35:217
123

Primer design and PCR-based genotyping
of Badh2 alleles
DNA was extracted from rice leaves, following a
CTAB procedure (Doyle 1991). The cleaved amplified
polymorphic sequence (CAPS) marker, which tar-
geted a functional insertion in exon 14 of the Badh2
gene, was developed using Primer Premier v. 5.0
software (Lalitha 2000) and NEBcutter v. 2.0 online
(http://nc2.neb.com/NEBcutter2/). The remaining
four primers were developed using Primer Premier v.
5.0 software. The total 20 ll volume of the PCR
consisted of 10 mM Tris–HCl, 50 mM KCl, 0.1 %
Triton X-100, 1.8 mM MgCl
2
, 0.1 mM dNTPs,
0.2 mM primers, 1 U Taq DNA polymerase, and
50 ng of genomic DNA. Due to the high GC content of
exon 2, 1 ll of DMSO per 10 ll was added to the
PCR. PCR cycling was performed in a Thermal Cycler
S1000 and consisted of initial denaturation at 94 °C
for 5 min, followed by 30 cycles for 30 s at 94 °C,
30 s at 50–61 °C, and 1 min at 72 °C, then terminated
with a final extension step for 10 min at 72 °C. The
restriction enzyme BslI was used to digest the PCR
products. A total of 1 lg of DNA was mixed with 1 U
enzyme at 55 °C in a 50-ll reaction volume and
incubated for 1 h. The PCR products of FME2-7,
EME7 and FME14I were separated in a 3 % agarose
gel, while the products of FMU1-2 and FME12-3 were
separated in a 6 % denaturation polyacrylamide gel.
Results
Discovery of rice fragrance allele
According to the fragrance results, 25 of the 30
accessions were fragrant and the remaining 5 were
non-fragrant rice accessions. Thirteen of the 295 rice
accessions were fragrant (Table 2).
A total of 10 overlapping primer pairs were used to
amplify the Badh2 gene sequence in 30 accessions
(Shi et al. 2008). Five alleles were detected after
sequence alignment (Fig. S1; Table 2). Among the 25
fragrant rice accessions, 13 contained an 8-bp deletion
in exon 7, 3 accessions contained a 7-bp deletion in
exon 2, 3 contained one C/T SNP in exon 13, 2
contained a 3-bp insertion in exon 13, and 2 accessions
contained a 1-bp insertion in exon 14. These results
were reported previously. Two accessions that com-
bined an 8-bp deletion in exon 7 and 3-bp deletion in 50
UTR were reported for the first time.
We detected badh2 alleles from whole-genome re-
sequencing data of 295 accessions (unpublished). The
8-bp deletion in exon 7 appeared in nine accessions,
and the C/T SNP in exon 13 was present in three
accessions. A novel allele that contained a 3-bp
deletion in exon 12 was discovered in the Korean
landrace Mongdonjaerae. The 3-bp deletion led to a
loss of one amino acid, which conferred fragrance in
rice (Fig. S2).
Table 3 Development of functional markers for badh2-UTR-E7,badh2-E2,badh2-E7,badh2-E12,badh2-E13,badh2-E14
Alleles Functional
markers
Primer sequence (50–30) Annealing
temperature (°C)
Enzyme PCR products size (bp)
(non-fragrant/fragrant)
badh2-UTR-E7 FMU1-2 F: TCCCACCACCACTCCACA
R: ACGAAGAGCTGCCGCTGC
61 – 163/160
badh2-E2 FME2-7 F: ACGAAGAGCTGCCGCTGC
R: GCGATTGCGCGGAGGTACT
61 – 78/71
badh2-E7 FME7 F: TCCTGTAATCATGTATACCC
R: AATTTGGAAACAAACCTT
50 – 151/143
badh2-E12 FME12-3 F: TTGGTCCAGTGCTCTGTGTG
R: GCACCAGCCAGACCATAAC
58 – 192/189
badh2-E13 FME12-3 F: TTGGTCCAGTGCTCTGTGTG
R: GCACCAGCCAGACCATAAC
58 – 192/195
badh2-E14 FME14I F: TCGATGCCGGAATTATCTGGGTGA
R: TCCCCACGGCTCATCGGAGG
61 BslI 60,205/266
Mol Breeding (2015) 35:217 Page 5 of 10 217
123

Development of functional markers
We designed several functional markers for the 3-bp
deletion in 50UTR, the 7-bp deletion in exon 2, the
8-bp deletion in exon 7, the 3-bp insertion in exon 13,
and one CAP marker for the 1-bp insertion in exon 14.
For our novel allele (badh2-E12), we also designed
several markers that were dependent on the flanking
sequence. We then selected optimal primers that
produced unambiguous polymorphic PCR products of
suitable size.
Four primer sets were selected for the four known
alleles. They included FMU1-2 for the 3-bp deletion in
50UTR, FME2-7 for the 7-bp deletion in exon 2,
FME7-1 for the 8-bp deletion in exon 7, and FME14I
for the 1-bp insertion in exon 14. For the new allele, we
selected FME12-3 as the optimal primer for detecting
the 3-bp insertion in exon 12. The FME12-3 marker
could also detect the 3-bp deletion in exon 13. FME12-
3 could amplify the 142-bp region from the 3-bp
insertion in exon 13 to the 3-bp deletion in exon 12
(Table 3; Fig. 1).
FMU1-2 generated PCR products of 163/160 bp for
non-fragrant/fragrant rice. FME2-7 and FME7 gener-
ated PCR products of 78/71 and 151/143 bp, respec-
tively. For FME14I, the PCR product sizes for non-
fragrant and fragrant rice were 265 and 266 bp,
respectively. After digestion by BslI, the non-fragrant
fragments separated into 205- and 60-bp fragments
and the fragrant rice fragment was 266 bp. FME12-3
generated PCR products of 192/189 bp for the allele in
exon 12 (badh2-E12), and 192/195 bp for the allele in
exon 13 (badh2-E13; Table 3; Fig. 1).
Application of the functional marker to genotyping
germplasm and F
2
populations
Five functional markers were used to genotype 14
accessions, including two accessions containing a
7-bp deletion in exon 2, two accessions with an 8-bp
Fig. 1 The 9 fragrant and 3 non-fragrant rice varieties were
genotyped by 5 functional markers. 1–3WT, non-fragrant rice;
4–5 badh2-UTR-E7;6–7 badh2-E2;8–9 badh2-E7;10 badh2-
E12;11–12 badh2-E13;13–14 badh2-E14.aPolymorphic PCR
bands were observed using the FMU1-2 marker, 4–5have 3-bp
deletion in 50-UTR; bpolymorphic PCR bands were observed
using the FME2-7 marker, 6–7have 7-bp deletion in exon 2;
cpolymorphic PCR bands were observed using the FME7
marker, 4–5and 8–9have 8-bp deletion in exon 7; dpolymor-
phic PCR bands were observed using the FME12-3 marker,
accession 10 has 3-bp deletion in exon 12, 11–12 have 3-bp
insertion in exon 13; epolymorphic PCR bands were observed
from the CAPs marker FME14I amplify and BslI digested, 13–
14 have 1-bp insertion in exon 14
217 Page 6 of 10 Mol Breeding (2015) 35:217
123

deletion in exon 7, two accessions with an 8-bp
deletion in exon 7 and a 3-bp deletion in 50UTR, two
accessions with a 3-bp insertion in exon 13, two
accessions with a 1-bp insertion in exon 14, one
accession with a 3-bp deletion in exon 12, and three
accessions as the control. All accessions were clearly
distinguished by the corresponding functional markers
(Fig. 1).
The functional markers FME14I and FME12-3
were used to genotype each individual in the F2
families derived from the cross between non-fragrant
rice and the Ar-30 accession (which has the badh2-
E14 allele), and between Dongjin-wx (wild type) and
Mongdonjaerae (3-bp deletion in exon 12), respec-
tively. We selected 13 individuals from each F2
family. There was only one individual with a 1-bp
insertion, eight individuals that were heterozygous,
and four individuals that were the wild type. The
samples with a 1-bp insertion were fragrant, while the
heterozygotes and wild types were not fragrant
(Fig. 2a). For the Badh2/badh2-E12 population,
depending on the banding patterns, two plants had a
3-bp deletion, seven plants were heterozygous with
two bands, and four plants were wild type (Fig. 2b).
The individuals with a 3-bp deletion were fragrant and
the others were not. These results supported the
conclusion that badh2 is recessive for rice fragrance.
Discussion
With the introduction of advanced genetic and
genomics tools, rice breeding has entered the geno-
mics era (Bao et al. 2008). Much phenotype-related
genetic variation can be explained by various markers
and single nucleotide changes. As two important
approaches to dissect the naturally occurring variation
of complex agronomic and quality traits, QTLs and
association mapping have attracted extensive atten-
tion. The QTL markers are population-specific. Alter-
natively, the functional markers associated with
specific genes are invariably associated with specific
traits. Functional markers could be used in any cross
with any parents, after the marker–trait association is
confirmed in a segregating population (Jin et al. 2010).
In breeding programs, it is important to efficiently
select individuals that contain the targeted trait within
a segregating population. If breeders are able to select
effectively at a very early stage, the results will likely
be savings in time and money, with overall enhanced
Fig. 2 Segregation pattern, as revealed by aFME14I marker,
of the representative 13 F
2
plants from the cross between A30
(badh2-E14) and non-fragrant rice; 1–4,6,9–10 and 13 are
heterozygous have no fragrance; 5,7–8and 11 are homologous
have no fragrance; 12 is homologous with 1-bp insertion in exon
14 and is fragrance; bFEM12-3 marker, of the representative 13
F
2
plants from the cross between RWG-031 (Mongdonjaerae
with 3-bp deletion in exon 12) and non-fragrant rice (Dongji-
wx); 1and 5have fragrance with homologous allele, 2–5,6–7,
10, and 12 have no fragrance with heterozygous allele; 8–9,11,
and 13 have no fragrance with homologous allele
Mol Breeding (2015) 35:217 Page 7 of 10 217
123

breeding progress. Unfortunately, in rice breeding, the
targeted traits (e.g., yield, cooking quality and eating
quality) usually involve the rice grain. Data for grain-
related traits are typically recorded after harvest,
which results in the loss of one season. Backcrossing is
initiated during the next flowering season, which
slows breeding progress.
MAS saves time and simplifies rice breeding and
facilitates introduction of an important allele to an
elite line in a relatively short period of time. Wang
et al. (2010) crossed the high-yielding rice cultivar
‘‘Wu-xiang-jing 14’’ as the maternal parent to the
high-quality Japonica rice cultivar ‘‘Kantou 194’’ with
low amylose content and translucent endosperm to
develop the new variety ‘‘Nan-jing 46’’ within 9 years,
by using one CAPS marker identify the Wx-mq
homozygous and heterozygous genotypes (Wang
et al. 2009).
In fragrant rice studies, 2-AP was identified as the
main component of fragrance, although more than 114
volatile flavor components have also been identified
(Shi et al. 2008). The chewing, KOH extraction, and
hot water extraction methods are dependent on the
sensory of testers. The traditional methods for qual-
itative or quantitative detection of 2-AP are less
credible and less efficient when 2-AP is at a low
concentration or a large number of samples are tested
simultaneously. Although it is difficult to differentiate
the targeted individuals in a large number of segre-
gating populations, breeders continue to use the
traditional methods to develop high-quality fragrant
rice cultivars (Siddiq et al. 2012), due to the high
economic value and preferred flavor of fragrant rice.
As we discussed, MAS allows the breeder to select
the targeted trait within a short time. Fortunately, the
rice fragrance-related gene (Badh2) was cloned by
Bradbury et al. (2005a,b). Intact Badh2, encoding
betaine aldehyde dehydrogenase, inhibits the synthe-
sis of 2-AP synthesis in non-fragrant rice. In contrast,
the recessive badh2 allele results in 2-AP accumula-
tion and, consequently, fragrant rice (Chen et al.
2008). Detecting the recessive badh2 alleles and
developing functional markers for those alleles will
considerably enhance fragrant rice breeding.
The non-functional badh2 gene is responsible
for rice fragrance. Hence, it is conceivable that
any mutation rendering the Badh2 gene non-
functional would lead to a new functional badh2
allele. Shan et al. (2013) successfully created
fragrant rice by knocking out the Badh2 gene of
non-fragrant rice via TALEN technology, and
validated this theory.
Functional alleles will be used as selection tools to
develop improved fragrant rice varieties. In this study,
we used 10 primers sequenced from 5 non-fragrant
and 25 fragrant rice accessions. Six alleles were
detected in the 25 fragrant rice accessions, 5 of which
were reported previously (Bradbury et al. 2005a; Shi
et al. 2008,2014; Shao et al. 2011; Kovach et al.
2009), while one combined allele (badh2-UTR-7) was
discovered in this study (Table 2).
From 2012 to 2014, we re-sequenced the whole
genome of 295 rice accessions (unpublished). We also
filtered the Badh2 sequence from our whole-genome
re-sequencing data. We detected a novel allele (bad-
h2-E12) from the Korea landrace Mongdonjaerea.
This allele has a 3-bp deletion in exon 12 (Table 2).
Overall, seven alleles were detected in our rice
germplasm (Table 2; Fig. S1). Shi et al. (2008)
developed functional markers for badh2-E2 and
badh2-E7 and confirmed marker co-segregation in a
segregating population. The CAPS marker for badh2-
E14 was used only to screen fragrant rice accessions,
without confirming co-segregation (Dissanayaka et al.
2014). Shi et al. (2014) discovered the 3-bp deletion in
50UTR. They did not design the functional markers for
this locus, other than another linked locus upstream of
badh2 with an 8-bp insertion in the promoter. In this
study, we developed five functional markers for six of
seven alleles (without badh2-E13.2; Table 3). All five
markers were effectively used to distinguish the
fragrant rice from the other fragrant or non-fragrant
rice accessions (Fig. 1). Moreover, we confirmed the
function of the markers FME12-3 and FME14I in a
segregating population. Those two markers perfectly
co-segregated with the phenotype. All of these func-
tional markers will be useful for a Badh2 diversity
study and for breeding improved fragrant rice acces-
sions using marker-assisted selection.
Acknowledgments We thank Dr. Yoo-Hyun Cho for his help
in collecting rice varieties and developing F
2
population. This
research was supported by Next-Generation BioGreen21
Program (PJ01116101), and Bio-industry Technology
Development Program Ministry for Food, Agriculture,
Forestry and Fisheries (No. 110136-5), Rural Development
Administration, Republic of Korea.
217 Page 8 of 10 Mol Breeding (2015) 35:217
123

References
Ahn S, Bollich C, Tanksley S (1992) RFLP tagging of a gene for
aroma in rice. Theor Appl Genet 84(7–8):825–828
Amarawathi Y, Singh R, Singh AK, Singh VP, Mohapatra T,
Sharma TR, Singh NK (2008) Mapping of quantitative trait
loci for basmati quality traits in rice (Oryza sativa L.). Mol
Breed 21(1):49–65
Bao J, Jin L, Xiao P, Shen S, Sun M, Corke H (2008) Starch
physicochemical properties and their associations with
microsatellite alleles of starch-synthesizing genes in a rice
RIL population. J Agric Food Chem 56(5):1589–1594
Bradbury LM, Fitzgerald TL, Henry RJ, Jin Q, Waters DL
(2005a) The gene for fragrance in rice. Plant Biotechnol J
3(3):363–370
Bradbury LM, Henry RJ, Jin Q, Reinke RF, Waters DL (2005b)
A perfect marker for fragrance genotyping in rice. Mol
Breed 16(4):279–283
Buttery R, Ling L, Juliano B (1982) 2-Acetyl-1-pyrroline: an
important aroma component of cooked rice. Chem Ind
958–959
Champagne ET (2008) Rice aroma and flavor: a literature
review. Cereal Chem 85(4):445–454
Chen S, Wu J, Yang Y, Shi W, Xu M (2006) The fgr gene
responsible for rice fragrance was restricted within 69kb.
Plant Sci 171(4):505–514
Chen S, Yang Y, Shi W, Ji Q, He F, Zhang Z, Cheng Z, Liu X,
Xu M (2008) Badh2, encoding betaine aldehyde dehydro-
genase, inhibits the biosynthesis of 2-acetyl-1-pyrroline, a
major component in rice fragrance. Plant Cell Online
20(7):1850–1861
Cordeiro GM, Christopher MJ, Henry RJ, Reinke RF (2002) Iden-
tification of microsatellite markers for fragrance in rice by
analysis of the rice genome sequence. Mol Breed 9(4):245–250
Dissanayaka S, Kottearachchi NS, Weerasena J, Peiris M (2014)
Development of a CAPS marker for the badh2. 7 allele in
Sri Lankan fragrant rice (Oryza sativa). Plant Breed
133(5):560–565
Doyle J (1991) DNA protocols for plants—CTAB total DNA
isolation. In: Hewitt GM, Johnston A (eds) Molecular
techniques in taxonomy. Springer, Berlin, pp 283–293
Fitzgerald TL, Waters DLE, Brooks LO, Henry RJ (2010)
Fragrance in rice (Oryza sativa) is associated with reduced
yield under salt treatment. Environ Exp Bot 68(3):292–300
Garland S, Lewin L, Blakeney A, Reinke R, Henry R (2000)
PCR-based molecular markers for the fragrance gene in
rice (Oryza sativa L.). Theor Appl Genet 101(3):364–371
Jin Q, Waters D, Cordeiro GM, Henry RJ, Reinke RF (2003) A
single nucleotide polymorphism (SNP) marker linked to
the fragrance gene in rice (Oryza sativa L.). Plant Sci
165(2):359–364
Jin L, Lu Y, Shao Y, Zhang G, Xiao P, Shen S, Corke H, Bao J
(2010) Molecular marker assisted selection for improve-
ment of the eating, cooking and sensory quality of rice
(Oryza sativa L.). J Cereal Sci 51(1):159–164
Kovach MJ, Calingacion MN, Fitzgerald MA, McCouch SR
(2009) The origin and evolution of fragrance in rice (Oryza
sativa L.). Proc Natl Acad Sci 106(34):14444–14449
Lalitha S (2000) Primer premier 5. Biotech Softw Internet Rep
Comput Softw J Sci 1(6):270–272
Lorieux M, Petrov M, Huang N, Guiderdoni E, Ghesquie
`re A
(1996) Aroma in rice: genetic analysis of a quantitative
trait. Theor Appl Genet 93(7):1145–1151
Mahattanatawee K, Rouseff RL (2014) Comparison of aroma
active and sulfur volatiles in three fragrant rice cultivars using
GC–Olfactometry and GC–PFPD. Food Chem 154:1–6
Mathure SV, Jawali N, Thengane RJ, Nadaf AB (2014) Com-
parative quantitative analysis of headspace volatiles and
their association with BADH2 marker in non-basmati
scented, basmati and non-scented rice (Oryza sativa L.)
cultivars of India. Food Chem 142:383–391
Myint KM, Arikit S, Wanchana S, Yoshihashi T, Choo-
wongkomon K, Vanavichit A (2012) A PCR-based marker
for a locus conferring the aroma in Myanmar rice (Oryza
sativa L.). Theor Appl Genet 125(5):887–896
Niu X, Tang W, Huang W, Ren G, Wang Q, Luo D, Xiao Y,
Yang S, Wang F, Lu B-R (2008) RNAi-directed down-
regulation of OsBADH2 results in aroma (2-acetyl-1-
pyrroline) production in rice (Oryza sativa L.). BMC Plant
Biol 8(1):100
Ootsuka K, Takahashi I, Tanaka K, Itani T, Tabuchi H, Yoshi-
hashi T, Tonouchi A, Ishikawa R (2014) Genetic poly-
morphisms in Japanese fragrant landraces and novel
fragrant allele domesticated in northern Japan. Breed Sci
64(2):115–124
Sakthivel K, Rani NS, Pandey MK, Sivaranjani A, Neeraja C,
Balachandran S, Madhav MS, Viraktamath B, Prasad G,
Sundaram R (2009) Development of a simple functional
marker for fragrance in rice and its validation in Indian
Basmati and non-Basmati fragrant rice varieties. Mol
Breed 24(2):185–190
Shan Q, Wang Y, Chen K, Liang Z, Li J, Zhang Y, Zhang K, Liu
J, Voytas DF, Zheng X (2013) Rapid and efficient gene
modification in rice and Brachypodium using TALENs.
Mol Plant 6(4):1365–1368
Shao G, Tang A, Tang S, Luo J, Jiao G, Wu J, Hu P (2011) A
new deletion mutation of fragrant gene and the develop-
ment of three molecular markers for fragrance in rice. Plant
Breed 130(2):172–176
Shao G, Tang S, Chen M, Wei X, He J, Luo J, Jiao G, Hu Y, Xie
L, Hu P (2013) Haplotype variation at Badh2, the gene
determining fragrance in rice. Genomics 101(2):157–162
Shi W, Yang Y, Chen S, Xu M (2008) Discovery of a new
fragrance allele and the development of functional markers
for the breeding of fragrant rice varieties. Mol Breed
22(2):185–192
Shi Y, Zhao G, Xu X, Li J (2014) Discovery of a new fragrance
allele and development of functional markers for identifying
diverse fragrant genotypes in rice. Mol Breed 33(3):701–708
Siddiq E, Vemireddy L, Nagaraju J (2012) Basmati rices:
genetics, breeding and trade. Agric Res 1(1):25–36
Sood B, Siddiq E (1978) A rapid technique forscent determination in
rice. Indian J Genet Plant Breed (The) 38(2):268–275
Tragoonrung S, Sheng J, Vanavichit A (1996) Tagging an aro-
matic gene in lowland rice using bulk segregant analysis.
IRRI, Rice Genet III
Wang C, Zhang Y, Zhu Z, Chen T, Zhao L, Lin J, Zhou L (2009)
Development of a new japonica rice variety Nanjing 46
with good eating quality by marker assisted selection. Mol
Plant Breed 7(6):1070–1076
Mol Breeding (2015) 35:217 Page 9 of 10 217
123

Wang C, Zhang Y, Zhu Z, et al (2010) Development of a new
japonica rice variety Nan-jing 46 with good eating quality
by marker assisted selection. Rice Genomics Genet 1(3).
doi:10.5376/rgg.2010.01.0003
Wang C, Chen S, Yu S (2011) Functional markers developed
from multiple loci in GS3 for fine marker-assisted
selection of grain length in rice. Theor Appl Genet 122(5):
905–913
Yano M, Shimosaka E, Saito A, Nakagahra M (1991) Linkage
analysis of a gene for scent in indica rice variety, Sur-
jamkhi, using restriction fragment length polymorphism
markers. Jpn J Breed 41(1):338–339
217 Page 10 of 10 Mol Breeding (2015) 35:217
123