See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225149795

Cytological and mapping analysis of a novel male sterile type resulting from
spontaneous floral organ homeotic conversion in marigold (Tagetes erecta L.)

Article in Molecular Breeding · June 2010

DOI: 10.1007/s11032-009-9372-x

CITATIONS READS

22 3,527

6 authors, including:

Yanhong He Guogui Ning

Huazhong Agricultural University 107 PUBLICATIONS 1,340 CITATIONS
44 PUBLICATIONS 314 CITATIONS
SEE PROFILE
SEE PROFILE

All content following this page was uploaded by Yanhong He on 17 February 2016.

The user has requested enhancement of the downloaded file.

Mol Breeding (2010) 26:19–29
DOI 10.1007/s11032-009-9372-x

Cytological and mapping analysis of a novel male sterile

type resulting from spontaneous floral organ homeotic
conversion in marigold (Tagetes erecta L.)
Yan-Hong He • Guo-Gui Ning • Ya-Lin Sun •

Yan Hu • Xing-Yu Zhao • Man-Zhu Bao

Received: 25 July 2009 / Accepted: 5 December 2009 / Published online: 20 December 2009
Ó Springer Science+Business Media B.V. 2009

Abstract A male sterile line was isolated in mari- map-based cloning strategies for the male sterility
gold (Tagetes erecta L.) and cytological analysis locus. We discuss the significance of this spontane-
determined this to be a novel genic male sterility trait ously derived genic male sterility trait relating to the
(Tems). Through the use of amplified fragment length homeotic conversion of floral organs in marigold.
polymorphisms (AFLPs) and bulked segregant anal-
ysis (BSA), tightly linked markers of Tems were Keywords Marigold (Tagetes erecta L.) 
identified with a view towards a map-based cloning Genic male sterility (GMS)  Homeotic conversion 
strategy. It was found that spontaneous homeotic Cytological characteristics  Mapping
conversion of floral organs was the underlying cause
of the male sterility in this marigold line. Thus, petals
of male sterile plants resembled sepal-like structures Introduction
and the stamens were partially converted to styles,
although without the full characteristics or function Male sterility is believed to play an important role in
of the true style organs. We have constructed a fine plant adaptation and evolution (Darwin 1877) as the
marker-based map for the Tems gene. This is separation of the sexual types between different
intended to provide a tool for marker assisted individuals (dioecy) achieves an obligate cross-
selection (MAS) strategies in hybrid breeding and fertilizing mating system. In plant breeding, male
sterility has been applied as an effective and
economical means of pollination control. The male
Yan-hong He and Guo-gui Ning contributed equally to this sterility trait has been derived following intergeneric
work.
crosses, interspecific crosses and intraspecific crosses,
Y.-H. He  G.-G. Ning  Y. Hu  X.-Y. Zhao  and has also arisen spontaneously or through the
M.-Z. Bao (&) effects of mutagenesis (Edwardson 1970; Serieys
Key Laboratory of Horticultural Plant Biology, Ministry 1996). Natural male sterility in a hermaphrodite
of Education, College of Horticulture and Forestry species can occur at any stage of stamen development
Sciences, Huazhong Agricultural University,
430070 Wuhan, People’s Republic of China and fully normal female functions are usually main-
e-mail: [email protected] tained. Marigold (Tagetes erecta L.), an annual
multipurpose plant (Vasudevan et al. 1997) belonging
Y.-L. Sun to the Asteraceae family, has a terminal capitulum
Institute of Vegetable Science, Wuhan Academy
of Agricultural Sciences, 430064 Wuhan, composed of hundreds of florets with two different
People’s Republic of China types: ray (sterile) florets in the periphery, and disk

123
20 Mol Breeding (2010) 26:19–29

(fertile) florets in the centre. In the disk florets of the concluded in a shorter time and in a more cost-
male sterile plants obtained in our laboratory (He effective manner (Tanksley et al. 1989). Thus, the
et al. 2009), the stamens consisted of a yellow identification of molecular markers that are tightly
filament lacking pollen inside, and this was also linked to the male sterility locus would permit the
associated with the petals of the ray and disk florets early and efficient identification of individual plant
appearing degenerated or as white filament-like genotypes within the breeding populations (Tanksley
petals. The male sterility was confirmed to be a et al. 1989). Genic male sterile genes, induced by
recessive genic trait (He et al. 2009), but it remains to temperature or photoperiod, have been successfully
be established at what point in floral development this isolated from rice through the application of map-
spontaneous male sterility of marigold occurs. based cloning strategies (Lu et al. 2005; Yang et al.
Molecular biology techniques can be applied to 2007), so demonstrating the potential of molecular
identify the causes and mechanisms of male sterility. markers for the GMS trait.
In fact, there are multiple possible causes but, under We have previously identified a sequence charac-
natural conditions, cytoplasmic male-sterility (CMS) terized amplified region (SCAR) marker linked to the
is the most frequently observed type. Molecular marigold recessive male sterile gene Tems at a
studies have shown that CMS is determined by distance of 2.4 cM (He et al. 2009). However, we
extra-nuclear genes, created within plant mitochon- wish to identify more tightly linked markers, including
drial genomes as a result of their high recombinogenic co-dominant markers to distinguish the genotypes of
activity (Budar and Pelletier 2001). Such mitochon- TeMsMs, TeMsms and Temsms and flanking markers
drial-encoded genes are suppressed or counteracted by located closely on either side of the Tems gene.
nuclear genes known as restorer-of-fertility genes Amplified fragment length polymorphism (AFLP)
(Chase and Gabay-Laughnan 2004; Hanson and technology represents one of the most efficient
Bentolila 2004). To date, at least 14 mitochondrial molecular marker systems for mapping genes of
genes relating to CMS have been characterized (Chase interest (Lu et al. 2004; Ke et al. 2004). However, it
2006), and numerous restorer genes have also been is a relatively expensive and complicated procedure,
cloned, including from Zea mays (Wise et al. 1999), and not easily applied to large-scale selection screens
Petunia hybrida (Bentolila et al. 2002), Oryza sativa (Huang et al. 2007). Therefore, we aim to convert such
(Komori et al. 2004), Raphanus sativus (Brown et al. AFLPs to provide convenient and relatively inexpen-
2003) and Sorghum bicolor (Klein et al. 2005). In sive PCR-based SCAR markers for the Tems gene.
contrast to CMS, genic male sterility (GMS) is In this study, histological observations were con-
generally associated with recessive mutations of ducted to investigate the cytological and morpholog-
nuclear-encoded genes and such loss of gene function ical characteristics of a novel spontaneous male
commonly affects multiple cellular activities (Budar sterility trait in marigold. AFLP technology was
and Pelletier 2001). utilized to identify tightly linked markers with a view
Male sterility is an important aid to ensuring cross- to constructing a genomic map for the Tems gene. In
fertilisation and, hence, the production of hybrid addition, we describe the pattern of spontaneous
lines. However, the GMS system has the limitation genic male sterility as achieved through the homeotic
that it is difficult to obtain a 100% male sterile conversion of floral organs in marigold and discuss
population, making it necessary to manually remove the significance of this trait.
the 50% component of fertile plants in order to
prevent contamination of the F1 hybrid progeny. This
must be achieved prior to flowering and consumes Materials and methods
significant time, labor and money (He et al. 2009).
The alternative approach of in vitro cloning of male Plant materials and population construction
sterile plants has inherent problems and is also not
economically viable (Kumar et al. 2004; He et al. An F2 segregating population of 494 progeny plants
2009). Molecular markers may be used to facilitate was generated following the self-pollination of a
the detection of important genes, thereby permitting single F1 plant from an intervarietal cross between
breeding programmes for elite cultivars to be f53f as pollen parent and M525A as seed parent. The

123
Mol Breeding (2010) 26:19–29 21

f53f parent is an inbred line developed through 5 AFLP analysis

generations of selfing. M525A is a male sterile type
plant from the two-type (male sterile/male fertile) The AFLP procedure was performed following the
line M525AB, that had been derived from an protocol developed by Vos et al. (1995) with minor
individual plant mutation of an exotic line, and was modifications (Lu et al. 2004). Genomic DNA was
isolated in our laboratory in 2004. digested with EcoRI and MseI and the fragments
ligated to EcoRI and MseI adaptors. Tenfold dilutions
of the adapter-ligated DNA were pre-amplified using
Microscopic observations
two pairs of primers. The pre-amplified DNA was
diluted (1:30) and aliquots were used in selective
Mature flowers were observed with a stereoscope
amplification reactions employing 512 pairs of
(SZH; Olympus Optical Co., Tokyo, Japan) equipped
selective primers, each of which contained three
with a color CCD camera (DP70; Olympus) using the
additional nucleotides at the 30 end (i.e. EcoRI ?
software Image-Pro Plus Version 5.1. Semi-thin
ANN and MseI ? GNN; EcoRI ? ANN and MseI ?
sections were prepared according to the methods of
CNN). The selective amplification products were
Uchida et al. (2003) with minor modifications.
separated on a 6% w/v polyacrylamide denaturing
Different size florets, ranging from 0.5 to 7 mm,
sequencing gel and visualized by silver staining
were fixed for 24 h in formalin-acetic acid-alcohol
system (Lu et al. 2001).
and embedded with Technovit 7100 (Heraeus Kulzer,
Wehrheim/Ts., Federal Republic of Germany)
Conversion of AFLP markers to SCAR markers
according to the manufacturer’s instructions. Sections
of approx. 1,500 nm thickness were cut with glass
Polymorphic AFLP fragments were cloned (Yi et al.
knives on an ultramicrotome (Leica Ultracut R; Leica
2006). For each polymorphic fragment, three inde-
Microsystems) and stained with 5 g/l toluidine bule.
pendent clones were sequenced through the Shanghai
Mature filament-like stamens and pistils of male
Invitrogen Biological Technology Co., Ltd. The
sterile marigold flowers were sampled 4, 6, 24, 48, 72
Universal GenomeWalker kit (Clontech, USA) was
and 96 h after pollination and were fixed in formalin-
used to extend the flanking sequences of AFLP
acetic acid-alcohol in order to observe the germina-
fragments (Siebert et al. 1995), and the final PCR
tion of pollen grains, according to the methods of
products were sequenced as described above. Based
Dafni and Motte (1998). Each treatment was repli-
on the sequences of AFLP markers, SCAR primer
cated using 10 samples. Nomarski and fluorescence
sets were designed using the software ‘primer 5’ and
microscopy were performed using an epifluorescence
synthesized commercially (Invitrogen, Shanghai).
microscope (BX61; Olympus) equipped with the
These primers were used to amplify genomic DNA
color CCD camera.
from 20 individual plants. PCR was performed using
the conditions: 94°C for 2 min; 35 cycles of 94°C for
DNA extraction and bulked segregant analysis 45 s, 60°C for 45 s, 72°C for 45 s; 72°C for 10 min
(and 12°C holding temperature) and was performed
Fertile and sterile plants were distinguished when at within 20 ll reaction volumes containing 50 ng
the flowering stage and the young leaves from each template DNA, 19 amplification buffer, 200 lM
plant were collected separately for genotyping. DNA dNTPs, 2 mM MgCl2, 5 pmol of each special primer
extraction from leaf tissues was performed using the and 1 U Taq DNA polymerase (Fermentas, Vilnius).
modified cetyltrimethyl ammonium bromide method
(Doyle and Doyle 1990) and final DNA concentration Linkage analysis
was adjusted to 50 ng/ll. For bulked segregant
analysis (BSA) (Michelmore et al. 1991), equal Linkage analysis of AFLP markers was performed
amounts of DNA were taken from either 10 fertile using the 494 plants of the F2 segregating population
plants or 10 sterile plants and were pooled to create and the JoinMap 3.0 software (Van Ooijen and
‘F bulk’ (BF) and ‘S bulk’ (BS), respectively. The Voorrips 2001). Markers originating from each parent
two bulked samples were subjected to AFLP analysis. were scored according to the standard coding system

123
22 Mol Breeding (2010) 26:19–29

using A, B, C, D and H. Ambiguous genotypes were closely similar to that of sepal structures. In male
resolved by assigning a blank score (-) to the sterile flowers, the top edge structure of both the
individual locus for map construction. In general, petals and sepals comprised a centrally placed
linkages considered for mapping were those with a vascular bundle surrounded by larger diameter
recombination frequency lower than 0.4 and a LOD (‘bubble-like’) cells (Fig. 2h), and sections taken
score larger than 3.0. Recombination frequencies were approx. half-way down the structures revealed 3–5
converted to map distances in centimorgans (cM) sets of vascular bundles (Fig. 2i). In male sterile
using the Kosambi mapping function (Kosambi 1944). plants it was observed that the development of the
stamen primordia failed to result in the differentiation
of archesporial cells, sporogenous cells, microspore
Results mother cells, microspore tetrads and pollen grains.
Cytological observation of the development of male
Histological analysis of male sterile and fertile sterile stamens showed that the morphology changed
flowers of marigold from ellipsoid (Fig. 2a), through reniform (Fig. 2b, c)
and then to a lunate form (Fig. 2d, e), and was
The comparative morphologies of male sterile and accompanied by the formation of ‘bubble-like’ cells,
male fertile flowers of marigold were observed by papilla-like cells and three sets of vascular bundles.
stereoscope (Fig. 1), and the results were consistent The transverse cytological structure of the male
with the descriptions given previously by He et al. sterile stamens indicated that they resembled style
(2009). In addition, we observed that in the male organs, possessing pistil structures, with the excep-
sterile flower the color, luster and shape of the petal tion that true style organs contained only one
whorl organs more closely resembled sepal struc- (central) vascular bundle (Fig. 2f, g). Thus, male
tures, and also the stamens lacked a cleft at the top sterility in the marigold flowers was associated with a
and appeared somewhat similar to style structures failure to develop normal archesporial cells and the
(Fig. 1b). stamens followed a developmental pathway more
From the observation of transverse semi-thin similar to that of style organs.
sections of the sterile flowers it was seen that the Longitudinal paraffin sections of floret structures,
development of the petal and stamen organs of male i.e. through the true style organs of both male sterile
sterile flowers differed significantly from the devel- and male fertile flowers, exposed the ovary contain-
opment of the corresponding organs in male fertile ing a single ovule and showed the stylar canal of the
plants. Transverse sections of petal organs taken from pistils connecting to the ovary (Fig. 3a, b).
male sterile plants revealed a morphological structure Fluorescence microscopy under ultraviolet radia-
tion was used to observe the germination of pollen
grains held on the pistils and stamens of male sterile
flowers. The results showed that the pollen success-
fully germinated, both on pistil and stamen structures,
and attached to papilla cells or papilla-like cells after
4 h. The pollen tubes entered the stylar canal 6 h after
germination (Fig. 4b) and extended into the ovule
72 h after germination (Fig. 4d). However, the pollen
tubes failed to enter the style-like structures of male
sterile flowers (Fig. 4a, c), and by 72 h post-germi-
nation the tubes had withered without demonstrating
cytoplasm fluorescence (Fig. 4c).
Fig. 1 Floret morphological characteristics of male sterile and Thus, it appears that the cytological structure of
male fertile marigold. a The male fertile floret of marigold. The the stamen-whorl organs in the male sterile plant is
stamen is hidden by the petal. b The male sterile floret of
transformed to style-like structure, but the trans-
marigold. The ovary structure in a and b is hidden from view
by the other floret organs; a, b are shown at the same formed organs lack the functionality of true style
magnification. Se sepal, Pe petal, St stamen, Pi pistil organs.

123
Mol Breeding (2010) 26:19–29 23

Fig. 2 Developmental stages of male sterile floret of marigold form of the mature male sterile stamen. The size of the
as observed by semi-thin section. a The ellipse form stage of floret = approx. 7 mm. f The lunate form of the male sterile
male sterile stamen and style primordium. The size of the floret flower’s style with papilla cells at either end, bubble-like cells
B0.5 mm. b The reniform stage of male sterile stamen and around and one set of vascular bundles. The size of the
style with high frequency cell division at either end. The size of floret = approx. 4 mm. g The lunate form of the mature male
the floret = approx. 0.5 mm. c The reniform stage of male sterile style. The size of the floret is about 7 mm. h Transverse
sterile stamen and style with cells at either ends differentiated section at the top of sepal and petal of male sterile floret,
into papilla cells, high frequency division of cells in the middle showing similar structure. i Transverse section in the middle of
and larger cells around. The size of the floret = 0.7 mm. d The sepal and petal of male sterile floret, showing similar structure.
lunate form of the male sterile stamen with papilla-like cells at Bar = 20 lm, Pe petal, St stamen, Pi pistil, Vb vascular
either end, bubble-like cells around and three sets of vascular bundles, Bu-like bubble-like cells, Pa papilla cells, Pa-like
bundles. The size of the floret = approx. 4 mm. e The lunate papilla-like cells

Screening AFLP markers linked to the Tems gene for polymorphisms between the two parent lines and
the ‘F’ and ‘S’ bulk populations. In total, 11
AFLP analysis was conducted using the selective ?3 polymorphic fragments were identified that were
nucleotide PCR primers (described in ‘‘Materials and present for the fertile parent and ‘F’ bulk while absent
methods’’). Each primer pair produced products which from the sterile parent and ‘S’ bulk. To verify the
separated as approx. 50–100 bands with sizes ranging implied linkage of these polymorphisms with the male
from 50 to 900 bp. For bulked segregant analysis, the sterility locus, twenty individuals from the ‘F’ and ‘S’
primer pairs 256 EcoRI ? ANN/MseI ? GNN and bulks were examined in more detail with respect to the
256 EcoRI ? ANN/MseI ? CNN were used to screen polymorphic bands. Subsequently, 64 individuals

123
24 Mol Breeding (2010) 26:19–29

expected from the PCR-origin of the fragments.

Based on the sequence data of the four AFLP
markers, primers were designed to permit the direct
amplification of the corresponding loci from genomic
DNA. Since the amplified fragments for these loci
were extremely small in size, they were not easily
visualised as markers of polymorphism between male
sterile and fertile individuals. Thus, the PCR-walking
technique was employed in order to extend the
flanking sequences of the four AFLP markers. AF1,
AF2 and AF4 were successfully extended in this
manner. We compared the marker flanking sequences
of male fertile and sterile plants and identified a
single nucleotide polymorphism within the MseI
restriction site of AF1 and AF2 markers and also a
192-bp deletion within the AF4 flanking sequence
(data not shown). Based on the latter deletion
sequence, a specific primer pair was designed,
namely: SC4F: 50 -CGTCTAGTCCATGAAGGAT
GACTCT-30 ; SC4R: 50 -CGAGCGGTATATGCTTG
CATGT-30 . PCR reactions with these primers were
performed amongst the two parent lines, the filial (i.e.
first) segregating generation and twenty individuals
comprising the ‘F’ and ‘S’ bulks. The results of this
Fig. 3 Longitudinal structure of male fertile and male sterile analysis demonstrated that the AF4 marker had been
floret as observed by paraffin section. a Male fertile floret. b
Male sterile floret. Bar = 0.1 mm, Se sepal, Pe petal, St
successfully converted to a SCAR marker designated
stamen, Pi pistil, O ovule SC4. SC4 represents a co-dominant marker and it was
detected as a 528-bp band in the m525A parental line
and the recessive homozygotes of male sterile
from the F2 segregating population were screened in a individuals (marked as an arrow in Fig. 6). By
primary linkage analysis. Such screening eventually contrast, the SC4 marker was detected as a 335-bp
identified four polymorphic markers linked to the band in the f53f parental line and the homozygotes of
Tems gene, which we designated AF1-4 (Table 1). fertile individuals (marked as an arrow in Fig. 6).
Figure 5 shows a representative amplification profile Both of these marker bands were detected in TeMsms
obtained using the primer pair E-AAC/M-GCT and heterozygous individuals.
identifying a 262-bp band (marked by an arrow) that
was present in male fertile individuals but absent in the Mapping
male sterile individuals (this marker was designated as
AF4). The strong correlation of these polymorphic To analyze the genetic distances and map the order of
bands with the male fertility/sterility trait established the developed markers, a total of 494 F2 individuals
that these AFLP markers were closely linked to the from the cross hybridisation of f53f and m525A lines
target (Tems) gene. were screened with the four AFLP markers. The
results revealed that four sterile and ten fertile
Conversion of AFLP markers into SCAR markers individuals showed recombination for the AF1
marker, four fertile individuals showed recombina-
In order to obtain a SCAR marker, the AFLP tion for the AF2 marker, two sterile and eight fertile
fragments linked to the Tems gene were cloned and individuals showed recombination for the AF3
sequenced. The sequence data revealed the presence marker, and two sterile and two fertile individuals
of the selective amplification primer sequences, as showed recombination for the AF4 marker. The four

123
Mol Breeding (2010) 26:19–29 25

Fig. 4 Germination of pollen grains on stamen organs pollination. The pollen tube could not fully penetrate the
(homeotically transformed to style structures) and true pistil style-like structure and withered without showing fluorescence.
organs of male sterile marigold. a The germination of pollen d The germination of pollen grains on pistil of male sterile
grains on stamen of male sterile marigold 6 h after pollination. marigold 72 h after pollination. A number of pollen tubes fully
b The germination of pollen grains on pistil of male sterile penetrated the style. Bars = 0.1 mm, Po pollen, Tu pollen
marigold 6 h after pollination. c The germination of pollen tube, Pa papilla, Pa-like papilla-like, Vb vascular bundles
grains on stamen of male sterile marigold 72 h after

Table 1 Description of AFLP markers tightly linked to the about 0.7 cM from Tems. These results indicated that
Tems gene the molecular markers were close enough to the Tems
AFLP Primer Size (bp) Map locus to be applied in MAS.
marker combination distance (cM)

AF1 E-ATC/M-CCA 152 3.2

Discussion
AF2 E-ACG/M-CAC 292 0.7
AF3 E-AAA/M-GTA 119 2.1
Histological analysis of flower structure indicated that
AF4 E-AAC/M-GCT 262 0.3
spontaneous floral organ homeotic conversion was the
E = EcoRI primer, 50 -GAC TGC GTA CCA ATT CA-30 ; cause of the male sterility trait in marigold. The ‘petal’
M = MseI primer, 50 -GAT GAG TCC TGA GTA AC-30 organs of the male sterile flowers appeared more
similar to sepals in terms of colour, luster and shape,
AFLP markers, the SC4 marker developed from one and detailed cytological and morphological observa-
of the AFLP markers and the SCS84 marker devel- tions confirmed that the petals had converted to sepal-
oped from a SRAP marker (He et al. 2009) were all like structures. More critically, the stamen primordia
mapped in a 5.3 cM region around the Tems gene deviated from the normal polarization and developed
(Fig. 7). Of these flanking markers of the Tems gene, as style-like structures, thereby failing to produce any
AF2 and AF4/SC4 were the most closely linked. of the cell-types leading to pollen formation. This
SCAR marker SC4 and AFLP marker AF4 occupied stamen-to-style transformation was morphologically
the same locus, approximately 0.3 cM from the Tems incomplete and the organs were non-functional.
gene. On the other side of the gene, AF2 was placed Thus, although extraneously introduced pollen could

123
26 Mol Breeding (2010) 26:19–29

Fig. 5 AFLP profile for sterile individuals fertile individuals

marker AF4 in 32 fertile M
plants and 32 sterile plants.
The arrows indicate the
polymorphic band present
in fertile individuals but not
in sterile individuals. M
100 bp DNA ladder 300

Fig. 6 The amplification bp M P1 P2 F1 sterile individuals fertile individuals CK

results of SC4 on
individuals plants. Lanes: M 2000
DNA molecular weight
marker 2000, P1 m525A,
P2 f53f, F1 Filial(1) 1000
generation, CK control 750
500 528 bp

250 335 bp

100

Fig. 7 Linkage map any step of its development, the failure of stamens to
around the Tems locus dehiscence, and the inability of mature pollen to
consisting of four AFLP germinate on compatible stigma (Laser and Lersten
markers and two SCAR
markers. Left, map
1972; Budar and Pelletier 2001). Here, we have shown
distances (Kosambi, cM); that male sterility in marigold is due to the homeotic
right, names of markers conversion of stamens into other floral organ struc-
tures, i.e. corresponding to the abnormal male organ
category. There has been extensive research on
homeotic conversion in CMS lines, where the male
sterility was found associated with genes of the
mitochondrial organelles, mtDNA (Wan et al. 2008;
Yang et al. 2008; Robison and Wolyn 2002). To the
best of our knowledge, however, there have been no
such reports relating spontaneous male sterility to
genomic loci, i.e. GMS. Therefore, we believe that this
germinate and attach to the papilla-like cells of the novel male sterile type in marigold, resulting from
converted stamen/style organs, the germinated pollen spontaneous floral organ homeotic conversion, is of
tube could not fully penetrate the style-like structure to great significance. The homeotic transformation of
reach the site of the ovary. The male sterile marigold stamens into pistil-like structures, a phenomenon
flower maintained fully normal female functions as the called pistillody, has been observed in some alloplas-
true style was unaffected by homeotic conversion. mic lines of Nicotiana tabacum, Brassica napus,
The associated phenotypic manifestations of male Daucus carota and Triticum aestivum (Farbos et al.
sterility are very diverse and include the absence or 2001; Murai et al. 2002; Linke et al. 2003; Teixeira
abnormality of male organs, the failure to develop et al. 2005). Causes of pistillody in Triticum aestivum
normal sporogenous tissues, the abortion of pollen at have been identified to the orf260cra mitochondrial

123
Mol Breeding (2010) 26:19–29 27

gene, altered patterns of expression of class B MADS- only a single line with a modified flower shape
box genes, and WPPK1 (wheat pistillody-related namely, the transformation of the pistil of the ray
protein kinase 1) (Murai et al. 2002; Meguro et al. floret to resemble corolla-like tissues. Studies with
2003; Hama et al. 2004; Saraike et al. 2007; Zhu et al. Gerbera capitulum have indicated that flower-type
2008). Unfortunately in these instances, the male specific MADS protein complexes play a central role
sterility trait was also accompanied by female sterility. in the differential development of ray and disc
A different approach, based on the ABCDE model of flowers (Laitinen et al. 2006). It remains to be
floral organ determination (Theiben 2001; Ferrario investigated whether the phenotype of the genic male
et al. 2004), created male sterile plants by the genetic sterile marigold is similarly due to mutations within
manipulation of floral organ development genes, and flower organ identity genes. Cloning of the Tems
this technique was proposed as particularly useful for gene will be the critical step towards obtaining a
the effective production of commercial hybrid plants direct explanation of the male sterility trait. Towards
(Mitsuda et al. 2006). Transposon insertion mutations this aim, identification of closely linked DNA mark-
of a MADS-box transcription factor in apple have also ers is a prerequisite for map-based cloning of the
be used to produce male sterile flowers (Yao et al. Tems gene. Here, we have identified flanking markers
2001). However, such artificially induced sterile plants located on either side of the Tems gene. These are
still encounter some barriers to their direct application positioned at 0.3 and 0.7 cM from the target locus
in crossbreeding programs. The genic male sterility and so are suitable for map-based cloning.
observed in marigold is attributed to the spontaneous In addition, the set of markers linked to the Tems
occurrence of floral organ homeotic conversion and is locus are suitable for use in marker assisted selection
suitable for hybrid breeding programmes. It therefore (MAS) strategies. Through a screen of 512 AFLP
seems likely that procedures of alloplasmic isolation, primer pairs, we have identified four AFLP markers
insertion mutations or transformation of floral organ linked to the Tems gene. These AFLP markers
development genes represent feasible ways to realize corresponded to amplicon products in the size range
the creation of man-made male sterile lines for the of just 119–292 bp. Therefore, we used PCR-walking
improvement of important plants. to extend the flanking sequences and, thereby, suc-
The model of floral organ development proposes ceeded in converting the AF4 marker to produce a co-
that mutations of the B class MADS-box genes, dominant SCAR marker. This SCAR marker (SC4)
APETALA3 (AP3)/DEFICIENS (DEF) and PISTIL- was placed at 0.3 cM from the Tems gene and the
LATA (PI)/GLOBOSA (GLO), cause homeotic trans- product sizes were sufficient to permit us to effectively
formation of the petals into sepals and stamens into distinguish between the genotypes TeMsMs, TeMsms
carpels (Bowman et al. 1989; Sommer et al. 1990; and Temsms. The SCAR marker obtained in this study
Tröbner et al. 1992). It seems likely that the homeotic can be employed to identify individual male fertile
conversion seen in marigold might, at least in part, be plants, while still at the seedling stage, and should
due to mutation of the B class MADS-box genes. thereby facilitate the removal of male fertile plants
However, this suggestion needs further investigation. from the mother line, when carrying out crossing
Marigold flower has the typical characteristics of breeding. This marker will also greatly facilitate the
the Asteraceae family with heteromorphic flowers transfer of the recessive GMS allele to chosen genetic
and specialized floral organs. Studies of homeotic backgrounds via MAS. PCR-based selection of plants
transformation in members of the Asteraceae family carrying the Tems allele at the seedling stage also has
have shown that the floral phenotype differs signif- the potential to greatly enhance selection efficiency in
icantly from other plants. Dezar et al. (2003) showed backcross breeding programmes.
that HaAG gene transcripts were accumulated mainly In conclusion, we have reported our findings for a
in the stamens of disc (fertile) flowers, while HaPI novel genic male sterile trait in marigold (Tagetes
and HaAP3 genes were preferentially expressed in erecta L.). This recessive GMS trait was shown by
ray (sterile) flowers and, only weakly in the petals cytological analysis to result from the spontaneous
and stamens of fertile flowers. Aida et al. (2008) floral organ homeotic conversion of the petals and
suppressed the chrysanthemum AGAMOUS (CAG) stamens, but not of the style. Fine mapping of the
gene by genetic transformation, but this resulted in Tems gene, encoding the male sterility trait, was

123
28 Mol Breeding (2010) 26:19–29

achieved through AFLPs in an F2 segregating pop- Edwardson JR (1970) Cytoplasmic male sterility. Bot Rev
ulation. We suggest that these findings represent an 36:341–420
Farbos I, Mouras A, Bereterbide A, Glimelius K (2001)
important contribution towards our understanding of Defective cell proliferation in the floral meristem of al-
novel male sterile types at the molecular level and in loplasmic plants of Nicotiana tabacum leads to abnormal
relation to plant adaptation and evolutionary strate- floral organ development and male sterility. Plant J
gies. The identification of markers offers a powerful 26:131–142
Ferrario S, Immink RGH, Angenent GC (2004) Conservation
tool for the efficient selection of parent lines in hybrid and diversity in flower land. Curr Opin Plant Biol 7:84–91
breeding programmes of marigold or related species Hama E, Takumi S, Ogihara Y, Murai K (2004) Pistillody is
of the Asteraceae family. caused by alterations to the class-B MADS-box gene
expression pattern in alloplasmic wheats. Planta 218:712–
Acknowledgments This research was supported by a grant 720
from the Ministry of Agriculture of China (no. 2003-Z36). We Hanson MR, Bentolila S (2004) Interactions of mitochondrial
thank all past and present colleagues in our lab for constructive and nuclear genes that affect male gametophyte devel-
discussion and technical support, and Dr. Alex McCormac opment. Plant Cell 16:154–169
(Mambo-Tox Ltd., UK) for critical editing of the manuscript. He YH, Ning GG, Sun YL, Qi YC, Bao MZ (2009) Identifi-
Yan-hong He and Guo-gui Ning contributed equally to this cation of a SCAR marker linked to a recessive male sterile
work. gene (Tems) and its application in breeding of marigold
(Tagetes erecta). Plant Breed 128:92–96
Huang Z, Chen YF, Yi B, Xiao L, Ma CZ, Tu JX, Fu TD
(2007) Fine mapping of the recessive genic male sterility
References gene (Bnms3) in Brassica napus L. Theor Appl Genet
115:113–118
Aida R, Komano M, Saito M, Nakase K, Murai K (2008) Ke LP, Sun YQ, Liu PW, Yang GS (2004) Identification of
Chrysanthemum flower shape modification by suppression AFLP fragments linked to one recessive genic male ste-
of chrysanthemum-AGAMOUS gene. Plant Biotech J rility (RGMS) in rapeseed (Brassica napus L.) and con-
25:55–59 version to SCAR markers for marker-aided selection.
Bentolila S, Alfonso AA, Hanson MR (2002) A pentatrico- Euphytica 138:163–168
peptide repeat-containing gene restores fertility to cyto- Klein RR, Klein PE, Mullet JE, Minx P, Rooney WL, Schertz
plasmic male-sterile plants. Proc Natl Acad Sci 99:10887– KF (2005) Fertility restorer locus Rf1 of sorghum (Sor-
10892 ghum bicolor L.) encodes a pentatricopeptide repeat pro-
Bowman JL, Smyth DR, Meyerowitz EM (1989) Genes tein not present in the colinear region of rice chromosome
directing flower development in Arabidopsis. Plant Cell 12. Theor Appl Genet 111:994–1012
1:37–52 Komori T, Ohta S, Murai N, Takakura Y, Kuraya Y, Suzuki S,
Brown GG, Formanova N, Jin H, Wargachuk R, Dendy C, Patil Hiei Y, Imaseki H, Nitta N (2004) Map-based cloning of a
P, Laforest M, Zhang J, Cheung WY, Landry BS (2003) fertility restorer gene, Rf-1 in rice (Oryza sativa L.). Plant
The radish Rfo restorer gene of Ogura cytoplasmic male J 37:315–325
sterility encodes a protein with multiple pentatricopeptide Kosambi DD (1944) The estimation of map distances from
repeats. Plant J 35:262–272 recombination values. Ann Eugen 12:172–175
Budar F, Pelletier G (2001) Male sterility in plants: occurrence, Kumar A, Singh SK, Sharma SK, Raghava SPS, Misra RL
determinism, significance and use. Life Sci 324:543–550 (2004) Comparison of seed-derived with micropropagated
Chase CD (2006) Cytoplasmic male sterility: a window to the male-sterile plants of Tagetes erecta L. for F sub (1)
world of plant mitochondrial–nuclear interactions. Trends hybrid seed production. J Hortic Sci Biotechnol 79:260–
Genet 23:81–90 266
Chase C, Gabay-Laughnan S (2004) Cytoplasmic male sterility Laitinen RAE, Broholm S, Albert VA, Teeri TH, Elomaa P
and fertility restoration by nuclear genes. In: Daniell H, (2006) Patterns of MADS-box gene expression mark
Chase CD (eds) Molecular biology, biotechnology of flower-type development in Gerbera hybrida (Astera-
plant organelles. Springer, Dordrecht, pp 623–634 ceae). BMC Plant Biol 6:11
Dafni A, Motte MM (1998) A rapid and simple procedure to Laser KD, Lersten NR (1972) Anatomy and cytology of
determine stigma receptivity. Sex Plant Reprod 11:177– microsporogenesis in cytoplasmic male sterile angio-
180 sperms. Bot Rev 38:425–454
Darwin C (1877) The different forms of flowers on plants of Linke B, Nothnagel T, Börner T (2003) Flower development in
the same species. Murray, London carrot CMS plants: mitochondria affect the expression of
Dezar CA, Tioni MF, Gonzalez DH, Chan R (2003) Identifi- MADS box genes homologous to GLOBOSA and DEF-
cation of three MADS-box genes expressed in sunflower ICIENS. Plant J 34:27–37
capitulum. J Exp Bot 54:1637–1639 Lu GY, Yang GS, Fu TD (2001) Silver stained AFLP—a novel
Doyle JJ, Doyle JL (1990) Isolation of plant DNA from fresh assay for DNA fingerprinting in Brassica napus. J Huaz-
tissue. Focus 12:13–15 hong Agric Univ 20:413–415

123
Mol Breeding (2010) 26:19–29 29

Lu GY, Yang GS, Fu TD (2004) Molecular mapping of a Theiben G (2001) Development of floral organ identity: stories
dominant genic male sterility gene (Ms) in rapeseed from the MADS house. Curr Opin Plant Biol 4:75–85
(Brassica napus L.). Plant Breed 123:262–265 Tröbner W, Ramirez L, Motte P, Hue I, Huijser P, Lönnig WE,
Lu Q, Li XH, Guo D, Xu CG, Zhang QF (2005) Localization of Saedler H, Sommer H, Schwarz-Sommer Z (1992)
pms3, a gene for photoperiod-sensitive genic male steril- GLOBOSA: a homeotic gene which interacts with DEF-
ity, to a 28.4-kb DNA fragment. Mol Genet Genomics ICIENS in the control of Antirrhinum floral organogene-
273:507–511 sis. EMBO J 11:4693–4704
Meguro A, Takumi S, Ogihara Y, Murai K (2003) WAG, a Uchida W, Matsunaga S, Sugiyama R, Kazama Y, Kawano S
wheat AGAMOUS homolog, is associated with develop- (2003) Morphological development of anthers induced by
ment of pistil-like stamens in alloplasmic wheats. Sex the dimorphic fungus Microbotryum violaceum in female
Plant Reprod 15:221–230 flowers of the dioecious plant Silene latifolia. Planta
Michelmore RW, Paran I, Kesseli RV (1991) Identification of 218:240–248
markers linked to disease-resistance genes by bulked Van Ooijen JW, Voorrips RE (2001) Joinmap version 3.0:
segregant analysis: a rapid method to detect markers in software for the calculation of genetic linkage maps. Plant
specific genomic regions by using segregating popula- Research International, Wageningen
tions. Proc Natl Acad Sci 88:9828–9832 Vasudevan P, Kashyap S, Sharma S (1997) Tagetes: a multi-
Mitsuda N, Hiratsu K, Todaka D, Nakashima K, Yamaguchi- purpose plant. Bioresour Technol 62:29–33
Shinozaki K, Ohme-Takagi M (2006) Efficient production Vos P, Hogers R, Bleeker M, Reijians M, VandeLee T, Hornes
of male and female sterile plants by expression of a chi- M, Frijters A, Pot J, Peleman J, Kuiper M, Zabeau M
meric repressor in Arabidopsis and rice. Plant Biotech J (1995) AFLP: a new technique for DNA fingerprinting.
4:325–332 Nucleic Acids Res 23:4407–4414
Murai K, Takumi S, Koga H, Ogihara Y (2002) Pistillody, Wan ZJ, Jing B, Tu JX, Ma CZ, Shen JX, Yi B, Wen J, Huang
homeotic transformation of stamens into pistil-like struc- T, Wang XJ, Fu TD (2008) Genetic characterization of a
tures, caused by nuclear-cytoplasm interaction in wheat. new cytoplasmic male sterility system (hau) in Brassica
Plant J 29:169–181 juncea and its transfer to B. napus. Theor Appl Genet
Robison MM, Wolyn DJ (2002) Complex organization of the 116:355–362
mitochondrial genome of petaloid CMS carrot. Mol Genet Wise RP, Bronson CR, Schnable PS, Horner HT (1999) The
Gemomics 268:232–239 genetics, pathology, and molecular biology of T-cyto-
Saraike T, Shitsukawa N, Yamamoto Y, Hagita H, Iwasaki Y, plasm male sterility in maize. Adv Agron 65:79–130
Takumi S, Murai K (2007) Identification of a protein Yang QK, Liang CY, Zhuang W, Li J, Deng HB, Deng QY,
kinase gene associated with pistillody, homeotic trans- Wang B (2007) Characterization and identification of the
formation of stamens into pistil-like structures, in allo- candidate gene of rice thermo-sensitive genic male sterile
plasmic wheat. Planta 227:211–221 gene tms5 by mapping. Planta 225:321–330
Serieys H (1996) Identification, study and utilisation in Yang JH, Qi XH, Zhang MF, Yu JQ (2008) MADS-box genes
breeding programs of new CMS sources. FAO progress are associated with cytoplasmic homeosis in cytoplasmic
report (1991–1994). Helia 19 (special issue):144–160 male-sterile stem mustard as partially mimicked by spe-
Siebert PD, Chenchik A, Kellogg DE, Lukyanov KA, Lukya- cifically inhibiting mtETC. Plant Growth Regul 56:191–
nov SA (1995) An improved PCR method for walking in 201
uncloned genomic DNA. Nucleic Acids Res 23:1087– Yao JL, Dong YH, Morris BAM (2001) Parthenocarpic apple
1088 fruit production conferred by transposon insertion muta-
Sommer H, Beltrán JP, Huijser P, Pape H, Lönnig WE, Saedler tions in a MADS-box transcription factor. PNAS
H, Schwarz-Sommer Z (1990) Deficiens, a homeotic gene 98:1306–1311
involved in the control of flower morphogenesis in Yi B, Chen YN, Lei SL, Tu JX, Fu TD (2006) Fine mapping of
Antirrhinum majus: the protein shows homology to tran- the recessive genic male-sterile gene (Bnms1) in Brassica
scription factors. EMBO J 9:605–613 napus L. Theor Appl Genet 113:643–650
Tanksley SD, Young ND, Paterson AH, Bonierbale MW Zhu Y, Saraike T, Yamamoto Y, Hagita H, Takumi S, Murai K
(1989) RFLP mapping in plant breeding: new tools for an (2008) orf260cra, a novel mitochondrial gene, is associ-
old science. Biotechnology 7:257–264 ated with the homeotic transformation of stamen into
Teixeira RT, Farbos I, Glimelius K (2005) Expression levels of pistil-like structures (pistillody) in alloplasmic wheat.
meristem identity and homeotic genes are modified by Plant Cell Physiol 49:1723–1733
nuclear-mitochondrial interactions in alloplasmic male-
sterile lines of Brassica napus. Plant J 42:731–742

123

View publication stats