The Plant Cell, Vol. 14, 491–504, February 2002, www.plantcell.org © 2002 American Society of Plant Biologists
The Dominance of Alleles Controlling Self-Incompatibility
in Brassica Pollen Is Regulated at the RNA Level
Hiroshi Shiba,a Megumi Iwano,a Tetsuyuki Entani,a Kyoko Ishimoto,a Hiroko Shimosato,a Fang-Sik Che,a
Yoko Satta,b Akiko Ito,c,1 Yoshinobu Takada,c Masao Watanabe,c Akira Isogai,a and Seiji Takayamaa,2
a Graduate
School of Biological Sciences, Nara Institute of Science and Technology, Ikoma 630-0101, Japan
of Biosystem Science, Graduate University for Advanced Studies, Hayama 240-0193, Japan
c Faculty of Agriculture, Iwate University, Morioka 020-8550, Japan
b Department
Self-incompatibility (SI) in Brassica is controlled sporophytically by the multiallelic S-locus. The SI phenotype of pollen
in an S-heterozygote is determined by the relationship between the two S-haplotypes it carries, and dominant/recessive relationships often are observed between the two S-haplotypes. The S-locus protein 11 (SP11, also known as the
S-locus cysteine-rich protein) gene has been cloned from many pollen-dominant S-haplotypes (class I) and shown to
encode the pollen S-determinant. However, SP11 from pollen-recessive S-haplotypes (class II) has never been identified by homology-based cloning strategies, and how the dominant/recessive interactions between the two classes
occur was not known. We report here the identification and molecular characterization of SP11s from six class II
S-haplotypes of B. rapa and B. oleracea. Phylogenetic analysis revealed that the class II SP11s form a distinct group
separated from class I SP11s. The promoter sequences and expression patterns of SP11s also were different between
the two classes. The mRNA of class II SP11, which was detected predominantly in the anther tapetum in homozygotes,
was not detected in the heterozygotes of class I and class II S-haplotypes, suggesting that the dominant/recessive relationships of pollen are regulated at the mRNA level of SP11s.
INTRODUCTION
Many species of hermaphrodite plants have evolved mechanisms to prevent self-fertilization. Self-incompatibility (SI) is
one physiological means of avoiding self-fertilization through
recognition of self-pollen in or on the female pistil. Classic
genetic analyses have revealed the presence of two major
types of homomorphic SI systems, gametophytic and sporophytic (de Nettancourt, 1977). Although the recognition of
self-pollen is controlled genetically by a single highly polymorphic locus called the S-locus in both of these systems,
the SI phenotype of pollen (gametophyte) is determined by
its own S-haplotype in the gametophytic system, whereas in
the sporophytic system, the SI phenotype is controlled by
the S-haplotypes of the diploid parent (sporophyte).
The majority of the members of the cruciferous plant genus Brassica possess a strong sporophytic SI system. Thus,
the SI phenotype of pollen as well as stigma is determined
1 Current address: Iwate Biotechnology Research Center, Kitakami
024-0083, Japan.
2 To whom correspondence should be addressed. E-mail takayama
@bs.aist-nara.ac.jp; fax 81-743-72-5459.
Article, publication date, and citation information can be found at
www.plantcell.org/cgi/doi/10.1105/tpc.010378.
by relationships between the two S-haplotypes carried by
its parent (Bateman, 1955). In other words, a codominant or
a dominant/recessive relationship between the two S-haplotypes influences the ultimate SI phenotype of both pollen
and stigma (Thompson and Taylor, 1966). The following observations have been made about dominance relationships
among S-haplotypes: (1) codominance is common; (2) dominance/recessiveness is frequent in pollen; (3) dominance
relationships among stigmas are different from those among
pollen; and (4) dominance relationships are nonlinear
(Thompson and Taylor, 1966; Ockendon, 1975; Visser et al.,
1982; Hatakeyama et al., 1998a).
Recent molecular studies have revealed that the S-locus
of Brassica encodes three highly polymorphic molecules
that are involved in SI recognition: S-receptor kinase (SRK),
S-locus protein 11 (SP11, also known as S-locus cysteinerich protein [SCR]), and S-locus glycoprotein (SLG). SRK is
a membrane-spanning serine/threonine receptor kinase predominantly present at the papilla cell membrane that functions as the sole determinant of the SI phenotype of stigma
(Stein et al., 1991; Delorme et al., 1995; Takasaki et al.,
2000). SP11/SCR is a small cysteine-rich basic protein predominantly present at the pollen coat that functions as the
sole determinant of the SI phenotype of pollen (Schopfer et
al., 1999; Suzuki et al., 1999; Takayama et al., 2000a; Shiba
492
The Plant Cell
et al., 2001). SLG is a secreted glycoprotein predominantly
present in the papilla cell wall that, although not involved directly in the determination of the SI phenotype, enhances
the SI response by an unknown mechanism (Nasrallah et al.,
1987; Takayama et al., 1987; Kandasamy et al., 1989; KishiNishizawa et al., 1990; Takasaki et al., 2000). Now that the
determinants of both stigma and pollen have been identified, it is important to analyze the connection between the
expression or activities of these determinants and dominance relationships. Recently, Hatakeyama et al. (2001) investigated whether the stigmatic S-determinant of SRK was
involved in determining the dominance relationships of
stigma using five S-homozygotes carrying an SRK28 transgene. They showed that the dominance relationship between
the SRK28 transgene and each of the endogenous S-haplotypes was identical to that between the S28-haplotype and
the respective endogenous S-haplotype. Moreover, in the
S43S43-homozygote carrying the SRK28 transgene, in which
the S43 phenotype in the stigma was masked by the presence of SRK28, the transcript level of SRK28 was found to be
much lower than that of SRK43. These results suggest that
the dominant/recessive relationships between S-haplotypes
in the stigma are determined by SRK itself, but not as a result of its relative transcription level.
Regarding dominance relationships in pollen, the pollen
determinant SP11/SCR from pollen-recessive S-haplotypes
has not been identified, and how the relationships are determined is not known. However, it has been suggested that
pollen-recessive S-haplotypes have a set of SLG and SRK
called class II, whereas pollen-dominant S-haplotypes have
a different set called class I (Nasrallah and Nasrallah, 1993).
To date, class II SLGs have been found only in pollen-recessive S-haplotypes, which include S29-, S40-, S44-, and S60haplotypes of B. rapa (syn. campestris) (Hatakeyama et al.,
1998a; Takasaki et al., 2000) and S2-, S5-, and S15-haplotypes of B. oleracea (Chen and Nasrallah, 1990; Scutt and
Croy, 1992; Cabrillac et al., 1999). The pairwise sequence
identity among class II SLGs is in excess of 86%, whereas
that between class I and class II SLGs is only 60 to 70%.
The extracellular domains of pollen-dominant and pollenrecessive SRKs also can be placed into these two classes
in a similar manner. These analytical results on stigmatic
S-locus proteins suggest that class II S-haplotypes have an
origin quite different from that of class I S-haplotypes
(Uyenoyama, 1995; Kusaba et al., 1997; Schierup et al.,
2001). In accordance with this assumption, attempts to isolate SP11s from class II S-haplotypes based on their sequence similarity with class I SP11s have been unsuccessful
(Watanabe et al., 2000; Kusaba et al., 2001).
We analyzed the SRK flanking region of the class II S60haplotype of B. rapa and succeeded in identifying the allelic
S60-SP11. Based on their similarity to this sequence, we
also identified the class II SP11 alleles from the S29-, S40-,
and S44-haplotypes of B. rapa and the S2- and S5-haplotypes of B. oleracea. These SP11 alleles from class II S-haplotypes showed relatively greater sequence similarity with
each other than with SP11 alleles from class I S-haplotypes,
as was the case with SLG and SRK. In situ hybridization
analysis revealed that class II SP11 was expressed predominantly in anther tapetum in the homozygote of the class II
S-haplotype. In contrast, the expression of the class II SP11
was found to be greatly suppressed in the heterozygote of
class I and class II S-haplotypes, suggesting that the dominant/recessive relationship in pollen is determined by the
expression level of SP11.
RESULTS
Cloning and Sequence Analysis of S60-SP11
We previously amplified 18 alleles of SP11 from class I S-haplotypes of B. rapa by reverse transcription–polymerase
chain reaction (RT-PCR) using a primer designed from the
conserved signal peptide region and an oligo(dT) primer
(Watanabe et al., 2000). However, the same combination of
primers failed to amplify allelic SP11 from any class II S-haplotypes. Various attempts to amplify the class II SP11 genomic fragment using primers designed from the conserved
5 and 3 noncoding regions of class I SP11 alleles also
failed (data not shown), suggesting that the entire genomic
sequences of SP11s are rather different between class I and
class II S-haplotypes. This idea was supported by the observation that DNA gel blot analysis using class I SP11 probes
detected no or only a weak hybridization signal with DNA of
any class II S-haplotype of B. oleracea and B. rapa (Schopfer
et al., 1999; A. Ito and M. Watanabe, unpublished results).
Therefore, we decided to adopt a genome-walking strategy to identify SP11 from a class II S-haplotype. The genomic organization of the S-locus has been analyzed
intensively in class I S-haplotypes of B. rapa (S8-, S9-, and
S12-haplotypes [Schopfer et al., 1999; Suzuki et al., 1999;
Takayama et al., 2000a]) and B. napus (S910- and SA10-haplotypes [Cui et al., 1999; Brugière et al., 2000]). Although the
relative positions of the three S-locus genes are different in
different S-haplotypes, these genes are located within an
30-kb genomic region. Because our preliminary analysis
of the class II S60-haplotype of B. rapa also suggested that
both SLG60 and SRK60 were localized within a 35-kb mluIdigested genomic fragment (see below), we decided to analyze the S-locus region of this S-haplotype.
We first cloned the SRK60 cDNA from stigmas of the S60homozyogote by PCR and used it as a probe to isolate a genomic SRK60 clone (see Methods). The SRK60 genomic
clone, named N4p4, was 12 kb in size and contained 3.4 kb
of the upstream side of SRK60 (Figure 1A). Using an end probe
of N4p4, we next isolated a clone, designated Bp120, that extends 17 kb on the upstream side of SRK60 (Figure 1A).
Nucleotide sequence analysis of Bp120 revealed the
presence of a short open reading frame (ORF) encoding a
putative cysteine-rich SP11 protein 6.5 kb upstream of
Recessive SP11 in Brassica SI
493
Figure 1. Genomic Structure of the SRK Flanking Region of the S60-Haplotype of B. rapa and the DNA Sequence of S60-SP11.
(A) Genomic organization of the SRK flanking region of the S60-haplotype. Thick bars represent the phage clones covering the S60-SP11/SRK60
region. Arrows indicate the direction of transcription of each gene. Exons are indicated by solid boxes, and introns are indicated by dips.
(B) Nucleotide sequence of S60-SP11 and its deduced amino acid sequence. The noncoding regions are shown in lowercase letters, with the intron splice donor/acceptor sequences (gt and ag in boldface letters) demarcating the intron. The coding regions are shown in uppercase letters,
with the underlined TAA sequence indicating the stop codon. Nucleotides are numbered from the transcription start site (a in boldface), which
was determined in a 5 rapid amplification of cDNA ends experiment. A putative TATA box and inverted sequence homologs of the CAAT motif
and the LAT52/56 box are denoted in underlined boldface letters.
SRK60 (Figure 1B). Another ORF encoding a putative signal
peptide was found just upstream of this ORF.
RT-PCR analysis using primers based on the putative signal peptide and an oligo(dT) primer revealed that this genomic region is expressed in anther tissue and produces a
secreted cysteine-rich small basic protein, as shown in Figure 2. Because this gene is located close to SRK, is ex-
pressed in anther tissue, and encodes a SP11/SCR-type
pollen coat protein (PCP) with the distinguishing characteristic position of the fourth cysteine residue (Doughty et al.,
1998, 2000; Schopfer et al., 1999; Suzuki et al., 1999;
Takayama et al., 2000a, 2000b), we determined that this
was an allelic SP11 of the S60-haplotype of B. rapa and designated it S60-SP11.
494
The Plant Cell
Figure 2. Alignment of the Predicted Amino Acid Sequences of Six Allelic Variants of Class II SP11 and Class I S9-SP11.
Gaps (hyphens) were introduced to optimize the alignment. Alternative signal peptide cleavage sites of the two alternative transcriptional forms
for class II SP11s are indicated by arrows. Conserved amino acid residues, eight cysteine residues (C1 to C8), a glycine residue (#), and an aromatic amino acid residue (†) are boxed. B.r., B. rapa; B.o., B. oleracea.
Detailed RT-PCR analysis revealed the presence of two
types of S60-SP11 transcripts in anther with ORFs of 285
and 279 bp, both of which could be produced from S60SP11 by alternative splicing (Figure 3A). The former (the
AL[] form) encodes a 95–amino acid protein and is expected to produce a 66–amino acid mature form of S60SP11 after removal of the putative 29–amino acid hydrophobic signal peptide. The latter (the AL[] form) encodes a 93–
amino acid protein in which the predicted cleavage site of
signal peptide in the former is deleted; it is expected to produce a 61–amino acid mature form of S60-SP11 after removal of a putative 32–amino acid signal peptide at an
alternative cleavage site (Figure 3A). Although the presence
of the alternative form of SP11 has not been detected in
class I S-haplotypes (Watanabe et al., 2000), the genomic
organization of SP11s having an intron close to the 3 end of
the signal peptide coding region is common to both classes
(Takayama et al., 2000a).
Biological Activity of Recombinant S60-SP11
To confirm that S60-SP11 works as the pollen determinant of
the S60-haplotype of B. rapa, we prepared a recombinant
S60-SP11 protein and tested its biological activity using a
pollination bioassay system (see Methods). The results are
summarized in Table 1. When the stigma of S60S60-homozygote was pretreated with recombinant S60-SP11 and then
pollinated with compatible pollen (from S12S12-homozygote),
the penetration of compatible pollen tubes into stigma was
inhibited in a dose-dependent manner. Because the recombinant S60-SP11 did not affect the compatible pollination when applied on the stigma of S8S8-homozygote, this
inhibitory effect of S60-SP11 was not caused by the toxicity
of this protein but rather by an S-haplotype–specific induction of the SI response. These results clearly demonstrated
that S60-SP11 was an allelic SP11 of the S60-haplotype of
B. rapa.
Cloning and Sequence Analysis of SP11s from Other
Class II S-Haplotypes of B. rapa and B. oleracea
Despite the polymorphic nature of SP11s, the nucleotide sequences of class I alleles are highly conserved in the region
coding for the signal peptide (Watanabe et al., 2000). Suspecting that this region also might be conserved within class
II SP11s, we attempted to amplify SP11 cDNAs from other
class II S-haplotypes of B. rapa using an oligonucleotide
primer (SP11-60-F3) based on the signal peptide region of
S60-SP11 and an oligo(dT) primer (NotI-dT). The PCR products were amplified further using a set of nested primers
(SP11-60-F4 and NotI-dT). Each of the tested class II S-homozygotes, S29S29-, S40S40-, and S44S44-homozygotes, gave a
single band at 400 to 450 bp, whereas the class I S-homozygotes, S8S8-, S9S9-, and S12S12-homozygotes, exhibited
no amplification (data not shown). The full-length cDNA of
each class II SP11 was obtained by the 5 rapid amplification of cDNA ends strategy. The deduced amino acid sequences of each class II SP11 are shown in Figure 2. As
with S60-SP11, two types of transcripts (AL[] and AL[]
forms) were obtained in S29- and S40-SP11 that presumably
were generated by alternative splicing, as shown in Figure
3B. Because this alternative splicing site is conserved
among all of the class II SP11s analyzed, the presence of
two types of SP11 transcripts might be a common feature of
class II S-haplotypes. However, because to date we have
not detected the AL() form from the S44-haplotype, a more
careful analysis should be performed before making this
conclusion.
DNA gel blot analysis showed that each of these class II
SP11 alleles of B. rapa was present as a single-copy gene
(Figure 4A). Polymorphisms of the restriction fragment
length suggested a high degree of sequence diversity in the
vicinity of the SP11s. Using HindIII restriction fragment
length polymorphism, the genetic linkage between SP11s
and the corresponding S-haplotypes was confirmed in
S29S40-heterozygote progeny (Figure 4B). Pulsed-field gel
Recessive SP11 in Brassica SI
electropohoresis (PFGE) gel blot analysis suggested that
each SP11, together with its cognate SLG and SRK, was localized within a 30- to 35-kb mluI-digested genomic fragment
in the S29-, S44-, and S60-haplotypes of B. rapa (Figure 4C).
We also identified the S2- and S5-SP11 genomic sequences of B. oleracea from commercial F1 hybrid cultivars
containing the class II S2- or S5-haplotypes (Sakamoto et al.,
2000). In the case of S2-SP11, a completely identical genomic sequence was obtained from two independent commercial broccoli cultivars, Three Main (S2S13) and Ryokurei
(S2S18), supporting its identity. The deduced amino acid sequences of S2- and S5-SP11 are aligned with those from B.
rapa in Figure 2.
The amino acid sequences of SP11s corresponding to
their putative signal peptides are highly conserved among
class II S-haplotypes (amino acid identity, 82.8 to 100%),
as is the case with class I, although the actual sequences
are rather different between class I and class II. In contrast,
the mature protein region of class II SP11s is polymorphic
(53.0 to 93.7%), although like the mature regions of class I
SP11s, it features hydrophilic and basic (pI 8.1 to 8.5)
properties. All eight cysteine residues are conserved in an
arrangement that is characteristic of SP11/SCR. Two other
495
residues that are conserved among most of the class I
SP11s, a glycine residue between C1 and C2 and an aromatic amino acid residue between C3 and C4 (Watanabe et
al., 2000), also are conserved among all of the class II
SP11s (Figure 2).
Molecular Phylogenies of the SP11s of Class I and
Class II S-Haplotypes
A phylogenetic tree of class I/class II SP11 sequences (Figure 5) was constructed with the deduced amino acid sequences of the mature protein region using the program
PROTML in MOLPHY (Adachi and Hasegawa, 1994). The
class II SP11s form a distinct cluster from class I SP11s, as
do stigmatic SLGs and SRKs, consistent with the proposal
that class II S-haplotypes have a different origin than do
class I S-haplotypes (Hinata et al., 1995; Kusaba et al.,
1997). The amino acid sequence identities among class II
SP11s are 63.2 to 94.6%, rather high compared with those
of class I SP11s, which range from 19.5 to 76.1%. These
observations suggest that the coalescence time of class II
SP11 alleles is shorter than that of class I and/or that the
Figure 3. Alternative Transcripts of Class II SP11s.
(A) Exon-intron structure and alternative transcripts of S60-SP11. The alternative splicing produces S60-SP11 proteins with or without an alanineleucine (AL) in the sequence. They are designated AL() and AL(), respectively. Coding regions are shown in uppercase letters, and the amino
acid residues they encode are shown below. Introns are shown in lowercase letters, and the intron splice donor/acceptor sequences (gt and ag)
are shown in boldface letters. The expected signal peptide cleavage sites are indicated by arrows.
(B) Alignment of the exon/intron junction of six class II SP11s. The intron splice donor/acceptor sequences (gt and ag) and an alternative possible acceptor sequence (AG) are shown in boldface letters. The alternative transcripts were detected experimentally in the S60-, S29-, and S40haplotypes, but not in the S44-haplotype, of B. rapa. The S2- and S5-haplotypes of B. oleracea were not tested.
496
The Plant Cell
Table 1. Effect of the Pretreatment of Stigma with Recombinant S60-SP11 on Cross-Pollination
Recombinant S60-SP11 (pmol per Stigma)
Pretreatment of Stigma
S60S60-stigma S12S12-pollen
compatible pollination (%)
S8S8-stigma S12S12-pollen
compatible pollination (%)
Tween 20
100 (45)
100 (28)
5
0.5
0a (24)
100 (21)
33a (33)
100 (27)
0.05
38a (29)
100 (17)
0.005
82a (22)
100 (10)
Numbers in parentheses indicate the number of pollination events examined.
a P 0.01 with Fisher’s exact probability test compared with the Tween 20–treated control.
amino acid substitution rate in class II genes is much slower
than that in class I. Class II SP11s of B. rapa and B. oleracea
do not exhibit a monophyletic relationship to each other. For
example, B. rapa S40-SP11 clusters with B. oleracea S5SP11, and B. rapa S44-SP11 clusters with B. oleracea S2SP11 (Figure 5). This observation shows that class II SP11
divergence also occurred before speciation, as suggested for
class I SLGs and SRKs (Dwyer et al., 1991; Hinata et al., 1995).
To gain insight into the evolutionary forces affecting the
divergence of SP11s, we calculated the levels of synonymous and nonsynonymous nucleotide substitutions for the
mature coding regions of six class II SP11s (192 sites of total independent comparisons). The mature coding regions of
class II SP11s from both species exhibited a synonymous
substitutions per site (Ps) value of 0.053 and a nonsynonymous substitutions per site (Pn) value of 0.189. The extremely high Pn/Ps ratio of 3.58 is highly significant (Z 4.90, P 0.001). This high ratio indicates that the nonsynonymous changes in SP11s are not selectively neutral but
suggests some natural selection (Hughes and Nei, 1988;
Lee et al., 1995; Nei and Kumar, 2000). One such selection
is diversifying selection, which probably operates through
disassortative mating in Brassica.
Class II SP11 Transcripts Are Regulated Recessively by
Class I S-Haplotypes
The promoter region of S60-SP11, which can affect gene expression, is shown in lowercase letters in Figure 1B. A putative TATA box is located 30 bp upstream of where the 5
end of the mRNA begins. Inverted sequence homologs of
the CAAT motif and the LAT52/56 box, both of which have
been found in the promoter of the anther/pollen-expressed
respiratory chain complex I genes (Zabaleta et al., 1998),
were identified in the promoter sequence of S60-SP11 (Figure 1B). However, no other known regulatory elements,
common repeats, or palindromic sequences were found.
We previously analyzed the promoter sequences of class I
SP11s and found that the immediate upstream region within
200 bp of the coding region is highly conserved among
different class I S-haplotypes. Using promoter deletion analysis, we further demonstrated that the region between 124
and 192 is essential for the sporophytic/gametophytic expression of class I SP11s (Shiba et al., 2001). However, the
promoter sequence of S60-SP11 showed little homology and
no motif conservation with that of class I SP11s.
To determine the temporal and spatial expression pattern
of the class II SP11, we first performed RNA gel blot analysis. Total RNAs were prepared from the anthers at several
developmental stages and from the stigma and the leaf tissue of a B. rapa S60S60-homozygote. A full-length S60-SP11
cDNA was used as a probe. As shown in Figure 6A, transcripts were detected only in anther, with the highest expression at an early stage of development, when the tapetal
cells were fully intact (stage 5), and little expression at late
stage, when tapetal degradation was virtually completed
(stage 7). This is in contrast to the expression of the class I
SP11s, which continues at a high level throughout these developmental stages of the anther (Takayama et al., 2000a).
In situ hybridization analysis showed that S60-SP11 was
expressed predominantly in tapetal cells of the anther at
early stages (Figure 7A). However, the gametophytic expression of S60-SP11 was not detected in pollen at late developmental stages (Figure 7B), a time when class I SP11s
are known to be expressed (Schopfer and Nasrallah, 2000;
Takayama et al., 2000a). This finding suggests that the gene
is expressed either weakly or not at all in pollen. Control experiments using a sense S60-SP11 probe showed no signal
at any stages (Figures 7C and 7D).
Next, we analyzed the expression of S60-SP11 in a heterozygote of class I and class II S-haplotypes, the S52S60heterozygote of B. rapa, to gain insight into the mechanisms
of dominance/recessiveness. RNA gel blot analysis showed
no expression of S60-SP11 in the anther (mixture of stages
5, 6, and 7) of S52S60-heterozygote, whereas strong expression was observed in the anther of S60S60-homozygote (Figure 6B). A similar result was obtained in S40-SP11, which
showed clear expression in anther tissue of the S40S40homozygote but no expression in the S35S40-heterozygote,
another class I/class II heterozygote (Figure 6C).
In situ hybridization analysis revealed that the expression
Recessive SP11 in Brassica SI
497
DISCUSSION
Structural Features on SP11s of Class II S-Haplotypes
Figure 4. DNA Gel Blot Analyses of the Class II SP11s.
(A) DNA gel blot analyses of SP11s using the corresponding SP11
cDNAs as probes. Genomic DNA (1 g each) of the S29-, S40-, S44-,
or S60-haplotypes was digested with EcoRI (E), BamHI (B), or HindIII
(H) and analyzed. The length of the DNA fragments in kilobases is indicated at left.
(B) Restriction fragment length polymorphism linkage analysis of an
F2 population segregating for S29- and S40-haplotypes. Genomic
DNAs isolated from parental (P) plants homozygous for either the
S29- or S40-haplotype, their F1 heterozygote, and F2 plants were digested with HindIII and hybridized with a mixed-SP11 probe (a mixture of S29-SP11 and S40-SP11 cDNA probes). The incompatibility
phenotype of each plant (H, S29S40-heterozygote; 29, S29S29-homozygote; 40, S40S40-homozygote) was determined by pollination tests
(Hatakeyama et al., 1998a).
(C) PFGE gel blot analysis of three S-haplotypes of B. rapa. High
molecular mass genomic DNA was digested with mluI and separated by PFGE. The DNA was hybridized with an SRK29 kinase domain (SRK29-KD) probe, an SLG29 probe, or SP11-mix probes (a
mixture of the S29-SP11 and S60-SP11 probes or the S29-SP11 and
S44-SP11 probes). DNA size markers are shown at left in kilobases.
of recessive class II S60-SP11 was below the level of detection in both anther tapetum and pollen through all three
stages of development (Figures 7E and 7F), whereas the
dominant class I S52-SP11 transcript was detected clearly in
both anther tapetum at an early developmental stage (Figure
7G) and pollen at a late stage (Figure 7H). Together, these
results suggest that the dominant/recessive relationships
between class I and class II S-haplotypes are regulated by
the mRNA level of SP11s.
In this work, we have identified SP11 from four class II S-haplotypes of B. rapa and two class II S-haplotypes of B. oleracea. Class II SP11s show several similarities with class I
SP11s: (1) they are small, secreted proteins with conserved
putative signal peptides; (2) their mature proteins show S-haplotype–specific polymorphisms in spite of their common hydrophilic and basic properties; (3) like other PCPs, they have
eight conserved cysteine residues whose positioning is characteristic of SP11s but different from other PCPs; and (4)
they have a glycine residue between C1 and C2 and an aromatic residue between C3 and C4 that are conserved.
However, the amino acid sequence alignment revealed extensive differences between SP11s from class I and class II.
In contrast, pairwise identities among the products of class II
alleles were relatively high, resulting in a tight cluster in the
phylogenetic tree. This relatively high similarity of class II
SP11s gives some insight into the amino acid residues that
determine the S-haplotype specificities of SP11s, because all
four of the class II S-haplotypes of B. rapa and two of the
class II S-haplotypes of B. oleracea have been shown to be
cross-compatible by reciprocal crosses within each species
(Hatakeyama et al., 1998a; Cabrillac et al., 1999).
Conservation of the eight cysteine residues suggests a
common three-dimensional protein structure of SP11s that
is stabilized by intramolecular disulfide bonds, similar to the
Figure 5. Phylogenetic Tree of Eight Alleles of SP11 from Brassica
Species.
Bootstrap probabilities for clusters are shown as percentages. The
bar under the tree indicates the number of amino acid substitutions
per site. B.r. B. rapa; B.o., B. oleracea.
498
The Plant Cell
type specificity in SP11s, although the involvement of the
conserved region cannot be excluded completely.
Hydrophilicity analysis of the class II SP11s does not suggest a hydrophilic (surface-exposed) structure in the C3-C4
region (data not shown), which has been suggested for
class I SP11s (Schopfer and Nasrallah, 2000). However, this
region does contain a conserved aromatic amino acid residue and more than one basic amino acid residue, which
have been shown to be involved in the bactericidal potency
of defensins (Yang et al., 2000). Three other regions, C1-C2,
C2-C3, and C5-C6, are highly divergent and contain two or
three amino acid residues that are completely variable
across the four S-haplotypes of B. rapa (and also between
the two S-haplotypes of B. oleracea). We speculate that
some of these four connecting regions are involved in the
specific interaction between SP11 and its cognate SRK.
Evolutionary Implications
Figure 6. RNA Gel Blot Analyses of the Class II SP11s.
(A) RNA gel blot analysis of S60-SP11 in an S60S60-homozygote of B.
rapa. The total RNAs of anther (A), stigma (S), and leaf (L) were used.
Numbers represent developmental stages of anther classified by
bud sizes, where 5 4 to 5 mm, 6 5 to 7 mm, and 7 7 to 10 mm
in length (Takayama et al., 2000a).
(B) RNA gel blot analysis of S60-SP11 and S52-SP11 in an S60S60homozygote (60) and an S52S60-heterozygote (H) of B. rapa. The
same amount of total RNA of anthers (mixture or stages 5 to 7) was
loaded in each lane. Similar results were obtained on three independent S60S60-homozygotes and S52S60-heterozygotes, and the results
of a representative experiment are shown.
(C) RNA gel blot analysis of S40-SP11 in an S40S40-homozygote (40)
and an S35S40-heterozygote (H) of B. rapa. The same amount of total
RNA of anthers (stages 5 to 7, mixture) was loaded in each lane. The
blot shown is representative of two independent experiments.
The bottom gel of each blot shows ethidium bromide (EtBr)–stained
rRNA bands.
defensin family of antimicrobial proteins (Broekaert et al.,
1995). In proteins of this family, such as defensin A and
MGD-1, positively charged and hydrophobic side chains,
which are positioned so that they are exposed to the molecular surface and stabilized by the backbone structure, are supposed to generate bactericidal potency and Gram-positive
specificity (Cornet et al., 1995; Yang et al., 2000). In class II
SP11s, N-terminal amino acid sequences before C1 and the
single amino acid spacer between C4 and C5 are well conserved. The C-terminal sequences after C6 are conserved
completely among all of the class II SP11s. Therefore, it is evident that at least one connecting region of C1-C2, C2-C3,
C3-C4, or C5-C6 is involved in the determination of S-haplo-
The phylogenetic tree of SP11s confirmed that the class II
genes form a monophyletic group separating from a cluster
of class I genes, as observed in SRKs and SLGs (Hinata et
al., 1995; Uyenoyama, 1995; Schierup et al., 2001). The sequence identity among class II SP11s is relatively high. According to the evolutionary model of Brassica SI proposed
by Uyenoyama (2000), class II S-haplotypes are expected to
invade populations at lower rates and decline to extinction
at higher rates than class I S-haplotypes. This predicts a
lower survival rate of a newly arisen mutation and less divergence among alleles in class II than class I. The most diverged pair of alleles (B. rapa S40-SP11 and S60-SP11)
reveals a synonymous divergence rate per site of 0.111 0.049. Using 0.5 108 per site per year (Li, 1997) as a representative of the synonymous nucleotide substitution rate
in plant nuclear genes, the divergence time of class II SP11s
is estimated at 10 5 million years ago. This estimate is
consistent with the previous estimate, 6.4 0.2 million
years ago, based on SRK and SLG sequences (Uyenoyama,
1995).
The diversification of S-haplotypes in Brassica has been
suggested to predate the speciation of B. rapa and B. oleracea (Dwyer et al., 1991; Hinata et al., 1995). The phylogenetic tree of SP11s (Figure 5) showed that class II SP11s
also diverged before speciation of these species. Among six
SP11 alleles from B. rapa and B. oleracea, the most closely
related interspecific pair of alleles is B. rapa S40-SP11 and
B. oleracea S5-SP11, which are identical at the 46 synonymous sites. Although the number of synonymous sites is
small, the presence of an allele pair identical at the synonymous sites implies that B. rapa and B. oleracea are closely
related. No synonymous changes at 46 sites gives an upper
limit of species divergence time of 2.6 million years with a
95% confidence limit, based on the assumption of 0.5 108 per site per year as a synonymous nucleotide substitution rate. This species divergence is much younger than ex-
Recessive SP11 in Brassica SI
499
pected, which has tentatively been estimated as on the
order of 10 million years ago (Uyenoyama, 1995). However,
an alternative hypothesis, such as the introgression of S-haplotypes (Schierup et al., 1998), also must be considered before drawing any conclusion on the evolutionary trail of these
S-haplotypes.
The conservation in the sequence of class II SP11s allowed us to calculate the levels of synonymous and nonsynonymous nucleotide substitutions for the coding regions of
mature proteins, which was impossible to do in class I
SP11s because of their great diversity and consequent ambiguous alignment (Watanabe et al., 2000). The high Pn/Ps
ratio (3.58) for the mature coding region of class II SP11s indicates an involvement of diversifying selection. In fact,
when we align six SP11s in amino acid sequences among
63 comparable sites, 28 sites are identical between distinct
alleles. On the other hand, some amino acid residues are
highly variable among six SP11s (Figure 2), suggesting that
these sites are targets of the selection and probably functionally important for SI determination.
Recently, probable orthologs of Brassica SP11 have been
identified from two S-haplotypes of Arabidopsis lyrata
(Kusaba et al., 2001). A. lyrata belongs to the tribe Arabideae (Price et al., 1994) and is thought to have diverged
from Brassica species 15 to 20 million years ago (Yang et
al., 1999), and its putative SRKs have shown more variability
than those of Brassica. In accord with this, two putative
SP11s obtained from A. lyrata showed great divergence, including their predicted signal sequences (Kusaba et al.,
2001), which are fairly conserved in Brassica SP11s. This
evidence shows that the origin of SP11s existed before
the divergence of Brassicaceae from Arabideae. Recently,
Schierup et al. (2001) analyzed 11 putative SRKs (Aly13) of
A. lyrata together with Brassica class I and class II SRKs. Although the phylogeny was unrooted, the Brassica class II
SRK cluster shares an ancestor with one cluster of Aly13. If
we can determine whether any A. lyrata SP11 shares an ancestor with Brassica class II SP11, we may be able to infer
the origin of the dominant/recessive expression system.
Dominant/Recessive Relationships
Figure 7. Analysis of S60- and S52-SP11 Expression in Anther Using
in Situ Hybridization.
Anther sections (stages 5 and 7) of an S60S60-homozygote or an
S52S60-heterozygote were hybridized with S60-SP11 or S52-SP11 antisense riboprobes or their sense riboprobes.
(A) and (B) S60S60-homozygote hybridized with an antisense S60SP11 probe.
(C) and (D) S60S60-homozygote hybridized with a sense S60-SP11
probe.
(E) and (F) S52S60-heterozygote hybridized with an antisense S60SP11 probe.
Class II S60-SP11 exhibited an exclusively sporophytic expression pattern in the anther tapetum and little or no gametophytic expression in the pollen, which is in contrast to the
sporophytic/gametophytic expression pattern of class I SP11s
(Schopfer and Nasrallah, 2000; Takayama et al., 2000a).
This characteristic expression pattern of class II SP11 was
(G) and (H) S52S60-heterozygote hybridized with an antisense S52SP11 probe.
Bar in (H) 50 m for (A) to (H).
500
The Plant Cell
confirmed by in situ analysis of S29-SP11 expression in
the S29S29-homozygote, which also showed purely sporophytic expression and no signal in pollen (data not shown).
As expected given the difference in the expression pattern of class I and class II SP11s, the promoter sequences
of SP11s showed little homology between these two
classes.
RNA gel blot and in situ analyses demonstrated that the
sporophytic expression of class II SP11s was suppressed in
the class I/class II heterozygotes, suggesting that the dominant/recessive relationship between class I and class II S-haplotypes in the determination of pollen phenotype is
generated at the level of SP11 transcription. This is in contrast to the dominant/recessive relationship in the stigma,
which is determined by SRK itself, but not by virtue of its relative expression level (Hatakeyama et al., 2001).
To our knowledge, there is no other demonstration that
the dominant/recessive relationship between multiallelic
genes is determined by their expression level. The mechanisms of this epigenetic phenomenon, in which an allelic
class II SP11 is suppressed in the presence of another allelic
class I gene, remain to be revealed. One hypothetical explanation is the “enhancer imbalance” model, in which dominant class I SP11s have higher binding affinities for limiting
transcription factors and the sequestration of critical transcription factors by class I SP11 enhancers results in the silencing of class II SP11s. Another possible explanation is
that the selective hypermethylation of recessive class II
SP11s blocks the binding of RNA polymerase II transcription factors, as demonstrated to occur during the development of a number of plant and animal species (Eden and
Cedar, 1994).
Another epigenetic occurrence known as “nucleolar dominance” might be similar to the dominant/recessive phenomenon seen between class I and class II SP11s (Frieman et
al., 1999). Nucleolar dominance describes the phenomenon
in which rRNA genes inherited from only one parent are
transcribed in interspecies hybrids or allopolyploids. This
event was first described in plants (Navashin, 1934) and
then in insects, amphibia, and mammals (Reeder, 1985;
Chen and Pikaard, 1997). Although nucleolar dominance is a
transcriptional event and was shown to be controlled at the
level of RNA polymerase I transcription (Chen and Pikaard,
1997), the mechanisms that discriminate between parental
sets of rRNA genes remain obscure. In interspecies hybrids
between Xenopus laevis and X. borealis, experimental data
support the enhancer imbalance model (Reeder and Roan,
1984), although this model does not fit observations with
hybrids between B. rapa and B. oleracea (Frieman et al.,
1999). The involvement of DNA methylation and histone
modification in the mechanism for the enforcement of nucleolar dominance also has been suggested (Chen and Pikaard,
1997). The dominance of class I over class II SP11s was
very clear and does not require an enforcement process
(i.e., newly formed F1 heterozygotes of class I and class II
S-haplotypes exhibited clearly distinguishable dominant/
recessive relationships). This study demonstrates that these
relationships are determined at the steady state mRNA level.
This newly found silencing system should provide a promising model for studies of general epigenetic phenomena.
METHODS
Plant Materials
The eight S-haplotypes of Brassica rapa (S8-, S9-, S12-, S29-, S40-,
S44-, S52-, and S60-haplotypes) used in this study have been described (Takayama et al., 1987; Nou et al., 1993; Hatakeyama et al.,
1998a, 1998b; Takasaki et al., 1999). The commercial F1 hybrid lines
of B. rapa Kukai (S35S40-heterozygote, classified previously as ST-17
ST-18-heterozygote), B. rapa Osome (S52S60-heterozygote), and B. oleracea Three Main (S2S13-heterozygote) were obtained from Takii
Seed Co. (Kyoto, Japan). B. oleracea Ryokurei (S2S18-heterozygote)
and B. oleracea Kinkei 201 (S5S45-heterozygote) were obtained from
Sakata Seed Co. (Kanagawa, Japan). All of these F1 hybrid lines
have been described (Iwano et al., 1998; Takasaki et al., 1999;
Sakamoto et al., 2000). All of the S-genotypes of plant materials used
in this study were checked through direct cloning of SLG genes by
polymerase chain reaction (PCR) using class I SLG-specific primers
(Brace et al., 1993) and class II SLG-specific primers (Nishio et al.,
1996).
Cloning of SRK60 cDNA
Stigmas of the S60S60-homozygote were collected from buds at 2 to
3 days before anthesis. Total RNA was isolated using Isogen (Nippon
Gene, Toyama, Japan). Approximately 20 g of RNA was subjected
to first-strand cDNA synthesis using Superscript II (GIBCO) with an
oligo(dT)18 primer. The partial sequence of SRK60 (2.4 kb) was amplified by PCR using a set of primers (5-ATGAAAGGGGTACAGAACAT-3 [Nishio et al., 1996] and 5-GTTCTTGAAYACACAAKAGRCCAAT-3) designed from the conserved extracellular domain of class
II SLG and the conserved kinase domain of SRK, respectively. The 5
and 3 rapid amplification of cDNA ends (RACE) cloning strategies
were used to reveal the entire cDNA sequence of SRK60. The genetic
linkage of cloned SRK60 with S60-haplotype was confirmed using
selfed F2 progeny of an S52S60-heterozygote.
Screening of the Genomic Library
Genomic DNA from the B. rapa S60S60-homozygote was digested
partially with Sau3AI and fractionated on a cesium chloride density
gradient. The fraction containing 9- to 23-kb fragments was ligated
with DASH II vector (Stratagene). To isolate the genomic clone containing SRK60, we screened 1.6 105 plaque-forming units under
standard conditions using an SRK60 cDNA as the probe and obtained
one phage clone designated N4p4. The insert DNA was amplified by
PCR with LA Taq DNA polymerase (TaKaRa, Shiga, Japan) using specific primers corresponding to the vector (EMBL3F, 5-CTTGCAGACAAACTGCGCAACTCGTGAAAG-3; EMBL3R, 5-GATCCACTCGTTATTCTCGGACGAGTGTTC-3) and then fragmented using HydroShear
(GeneMachines, San Carlos, CA). Fragments ranging from 1.5 to 3
kb were treated with T4 DNA polymerase (TaKaRa) and subcloned
Recessive SP11 in Brassica SI
into pBluescript SK (Stratagene) for sequence analysis. To isolate
flanking regions of SRK60, we screened 3 105 plaque-forming units
with both end probes of the N4p4 clone. The end probes were amplified by specific primers designed from both of the end sequences.
One clone covering the 18.5-kb upstream region of SRK60 was isolated and designated Bp120. The insert DNA was amplified by PCR
and sequenced as described above.
501
blue staining, as described previously (Shiba et al., 2000). Typically,
100 pollen tubes penetrated the stigma in a 0.05% Tween 20–
treated control. When the number of penetrating pollen tubes was reduced to 10 in the test, pollination was judged to become incompatible.
DNA Gel Blot Analysis
Cloning of Class II SP11s from Brassica Species
Total or poly(A) RNAs were extracted from anthers of S60S60-,
S29S29-, S40S40-, and S44S44-homozygotes of B. rapa and reverse
transcribed to synthesize first-strand cDNA as described (Suzuki et
al., 1999; Takayama et al., 2000a). An S60-SP11 cDNA fragment was
amplified by PCR using a primer (5-ATGAAATCTGCTGTTTATGTTTC-3) designed from the genomic sequences of Bp120 and the
oligo(dT)18 primer and subcloned into the vector pGEM-T Easy
(Promega). The 5 end of S60-SP11 was determined by 5 RACE.
Other class II SP11s of B. rapa were amplified using the primers
SP11-60-F3 (5-ACATTTTTAACAAATATACACTACTTATGT-3) and
NotI-dT (Watanabe et al., 2000). The PCR products were amplified
further using a nested primer, SP11-60-F4 (5-ATACACTACTTATGTTTCATATTTTTGATT-3), and the NotI-dT primer. The amplified
PCR products were cloned into the vector pCR2.1 (Invitrogen, Carlsbad, CA) and sequenced. The 5 end of the SP11s was determined
by 5 RACE.
Genomic DNA was extracted from leaves of S60S60-, S29S29-,
S40S40-, and S44S44-homozygotes of B. rapa and S2S13-, S2S18-, and
S5S45-heterozygotes of B. oleracea as described (Edwards et al.,
1991). The genomic structures of SP11s were determined by PCR
using a set of primers (5-GCGAAAATCTTATATACTCATAAG-3 and
5-TTCGTTGATCAATTATGATT-3) designed from the 5 and 3 ends
of the coding region of SRK60, respectively. The fragments obtained
were subcloned into the vector pGEM-T Easy and then sequenced.
Total DNA was extracted from young leaves by the cetyldimethylammonium bromide method (Murray and Thompson, 1980). The extracted DNA (1 g) was digested with EcoRI, BamHI, or HindIII and
separated by electrophoresis on a 0.8% agarose gel. DNA fragments
were transferred onto nylon membranes, hybridized with digoxigenin-labeled probes, washed, and detected as described (Watanabe
et al., 1999). Digoxigenin-labeled SP11 cDNA probes were prepared
by PCR amplification with a specific primer set. The detection of the
hybridized probe was performed as described (Matsuda et al., 1996).
PFGE Gel Blot Analysis
Megabase DNA embedded in agarose plugs was prepared from
young leaf tissue of S29S29-, S44S44-, and S60S60-homozygotes by the
rapid method (Tanaka et al., 1993; Liu and Whittier, 1994). DNA digestion with mluI, pulsed-field gel electrophoresis (PFGE) of digested
DNA, DNA transfer to a nylon membrane, hybridization with digoxigenin-labeled probes, and detection of the hybridized probes were
performed as described (Watanabe et al., 2000) except that the
membranes were washed twice in 0.1 SSC (1 SSC is 0.15 M
NaCl and 0.015 M sodium citrate) and 0.1% SDS at 68C for 20 min.
The digoxigenin-labeled SLG29 probe and SRK29 kinase domain
probe (Hatakeyama et al., 1998b) were used under the same conditions used for hybridization to SP11.
Sequence Analysis of Class II SP11s
Preparation of Recombinant S60-SP11
The mature coding region of the S60-SP11 cDNA (AL[] form) was
amplified using the primers 5-GGATCCCTAGATGTGGGAGCTTGG-3
and 5-CTCGAGTTCGTTGATCAATTATGATT-3 and cloned between
the BamHI and XhoI sites of pGEX-6P-3 vector (Amersham Pharmacia). This construct was transformed into the Escherichia coli BL21
strain. The induction and purification of the glutathione S-transferase
fusion protein were performed according to the manufacturer’s protocol. The recombinant S60-SP11 with the five–amino acid N-terminal
extension (GPLGS) was obtained after cleavage with PreScission
Protease (Amersham Pharmacia). The recombinant S60-SP11 was
reduced once in the presence of DTT and then refolded in the presence of both oxidized and reduced forms of glutathione (1:10). One
major oxidized form of recombinant S60-SP11 was purified to homogeneity by reverse phase HPLC.
Deduced amino acid sequences of six class II SP11s were aligned
together with previously reported class I S9- and S12-SP11s
(Takayama et al., 2000a). Eight cysteine residues were kept invariable in the alignment. Based on this alignment, the following analyses were performed. Any sites that contained a gap were excluded
from the analysis. The phylogenetic tree was constructed using the
program PROTML in MOLPHY version 2.3 (option JTT-F; Adachi and
Hasegawa, 1994). To evaluate the clustering patterns, bootstrap
probability for each cluster was calculated with 1000 times of resampling. For the calculation of synonymous and nonsynonymous differences per site, taking account of transition/transversion bias, we
used the modified Nei and Gojobori method with R 1.0 (Nei and
Kumar, 2000).
RNA Gel Blot Analysis
Pollination Bioassay
The pollination bioassay was performed as described (Takayama et
al., 2001). Stigmas were treated with 0.5 L of a recombinant S60SP11 solution containing 0.05% Tween 20 and air dried for 1 hr. After
cross-pollination, the stigmas were kept at 20C for 6 hr, and penetration of pollen tubes into each stigma was observed after aniline
Total RNA was isolated from several tissues of B. rapa using Isogen.
Each 10 g of total RNA was electrophoresed on a 1.2% (w/v) agarose/formamide gel and transferred to a nylon membrane (Hybond
N; Amersham-Pharmacia). The membrane was hybridized at 65C
for 12 hr with 32P-labeled S40-, S52-, and S60-SP11 probes specific for
their respective coding regions of the mature proteins and the 3
noncoding regions. After hybridization, the membrane was washed
502
The Plant Cell
in 0.1 SSPE (1 SSPE is 0.115 M NaCl, 10 mM sodium phosphate, and 1 mM EDTA, pH 7.4) and 0.1% (w/v) SDS at 65C for 30
min and exposed on x-ray film. Equal loading of total RNA was assessed by ethidium bromide staining of rRNA bands.
In Situ Hybridization
The anthers at developmental stage 5 (bud length, 4 to 5 mm) and
stage 7 (7 to 10 mm) were collected from a B. rapa S60S60-homozygote and an S52S60-heterozygote. Digoxigenin-labeled sense and antisense RNA probes of S60-SP11 and S52-SP11 (coding region of
mature protein) were prepared using the SP6/T7 digoxigenin RNA
labeling kit (Roche Diagnostics, Mannheim, Germany). In situ hybridization to 10-m-thick Paraplast (Sigma) sections of paraformaldehyde-fixed anthers was performed according to the method
described previously (Doughty et al., 1998).
components of the host defense system. Plant Physiol. 108,
1353–1358.
Brugière, N., Cui, Y., and Rothstein, S.J. (2000). Molecular mechanisms of self-recognition in Brassica self-incompatibility. Trends
Plant Sci. 5, 432–438.
Cabrillac, D., Delorme, V., Garin, J., Ruffio-Châble, V., Giranton,
J.-L., Dumas, C., Gaude, T., and Cock, J.M. (1999). The S15 selfincompatibility haplotype in Brassica oleracea includes three S gene
family members expressed in stigmas. Plant Cell 11, 971–986.
Chen, C.-H., and Nasrallah, J.B. (1990). A new class of S
sequences defined by a pollen recessive self-incompatibility allele
of Brassica oleracea. Mol. Gen. Genet. 222, 241–248.
Chen, Z.J., and Pikaard, C.S. (1997). Epigenetic silencing of RNA
polymerase I transcription: A role for DNA methylation and histone
modification in nucleolar dominance. Genes Dev. 11, 2124–2136.
Accession Numbers
Cornet, B., Bonmatin, J.-M., Hetru, C., Hoffmann, J.A., Ptak, M.,
and Vovelle, F. (1995). Refined three-dimensional solution structure of insect defensin A. Structure 3, 435–448.
The GenBank accession numbers for S60-SP11, S29-SP11, S40SP11, S44-SP11, S2-SP11, and S5-SP11 are AB067446, AB067449,
AB067450, AB067451, AB067447, and AB067448, respectively.
Cui, Y., Brugière, N., Jackman, L., Bi, Y.-M., and Rothstein, S.J.
(1999). Structural and transcriptional comparative analysis of the
S locus regions in two self-incompatible Brassica napus lines.
Plant Cell 11, 2217–2231.
ACKNOWLEDGMENTS
Delorme, V., Giranton, J.-L., Hatzfeld, Y., Friry, A., Heizmann, P.,
Ariza, M.J., Dumas, C., Gaude, T., and Cock, M. (1995). Characterization of the S locus genes, SLG and SRK, of the Brassica S3
haplotype: Identification of a membrane-localized protein encoded
by the S locus receptor kinase gene. Plant J. 7, 429–440.
We thank Dr. Katsunori Hatakeyama (Research Institute of Seed Production) and Dr. Koji Sakamoto (Takii Seed Company) for generous
gifts of a B. rapa S60S60-homozygote and a commercial cultivar (Kukai), respectively. We thank Kanako Iwasaki, Hanae Sugita, Taduru
Ueda, and Hiroko Sato for technical assistance. This work was supported in part by Grants-in-Aid for Special Research on Priority Areas
B (No. 11238025) to A.I., B (Nos. 11238201 and 11460001) to M.W.,
and C (No. 13206052) to H.S. from the Ministry of Education, Culture,
Sports, Science, and Technology of Japan, by a grant from the Research for the Future Program (No. JSPS-RFTF 00L01605) to S.T.
from the Japan Society for the Promotion of Science, and by a grant
from the Mitsubishi Foundation to A.I.
de Nettancourt, D. (1977). Incompatibility in Angiosperms. Monographs on Theoretical and Applied Genetics 3. (Berlin: SpringerVerlag).
Doughty, J., Dixon, S., Hiscock, S.J., Willis, A.C., Parkin, I.A.P.,
and Dickinson, H.G. (1998). PCP-A1, a defensin-like Brassica
pollen coat protein that binds the S locus glycoprotein, is the
product of gametophytic gene expression. Plant Cell 10, 1333–
1347.
Doughty, J., Wong, H.Y., and Dickinson, H.G. (2000). Cysteinerich pollen coat proteins (PCPs) and their interaction with stigmatic S (incompatibility) and S-related proteins in Brassica: Putative roles in SI and pollination. Ann. Bot. 85 (suppl. A), 161–169.
Received August 28, 2001; accepted October 30, 2001.
Dwyer, K.G., Balent, M.A., Nasrallah, J.B., and Nasrallah, M.E.
(1991). DNA sequences of self-incompatibility genes from Brassica campestris and B. oleracea: Polymorphism predating speciation. Plant Mol. Biol. 16, 481–486.
REFERENCES
Eden, S., and Cedar, H. (1994). Role of DNA methylation in the regulation of transcription. Curr. Opin. Genet. Dev. 4, 255–259.
Adachi, J., and Hasegawa, M. (1994). MOLPHY: A Computer Program Package for Molecular Phylogenetics, Version 2.3. (Tokyo,
Japan: Institute of Statistical Mathematics).
Edwards, K., Johnstone, C., and Thompson, C. (1991). A simple
and rapid method for the preparation of plant genomic DNA for
PCR analysis. Nucleic Acids Res. 19, 1349.
Bateman, A.J. (1955). Self-incompatibility systems in angiosperms.
III. Cruciferae. Heredity 9, 52–68.
Frieman, M., Chen, Z.J., Saez-Vasquez, J., Shen, L.A., and
Pikaard, C.S. (1999). RNA polymerase I transcription in a Brassica
interspecific hybrid and its progenitors: Tests of transcription factor involvement in nucleolar dominance. Genetics 152, 451–460.
Brace, J., Ockendon, D.J., and King, G.J. (1993). Development of
a method for the identification of S alleles in Brassica oleracea
based on digestion of PCR-amplified DNA with restriction endonucleases. Sex. Plant Reprod. 6, 133–138.
Broekaert, W.F., Terras, F.R.G., Cammue, B.P.A., and Osborn,
R.W. (1995). Plant defensins: Novel antimicrobial peptides as
Hatakeyama, K., Watanabe, M., Takasaki, T., Ojima, K., and
Hinata, K. (1998a). Dominance relationships between S-alleles in
self-incompatible Brassica campestris L. Heredity 80, 241–247.
Hatakeyama, K., Takasaki, T., Watanabe, M., and Hinata, K.
Recessive SP11 in Brassica SI
503
(1998b). Molecular characterization of S locus genes, SLG and
SRK, in a pollen-recessive self-incompatibility haplotype of Brassica rapa L. Genetics 149, 1587–1597.
Navashin, M. (1934). Chromosomal alterations caused by hybridization and their bearing upon certain general genetic problems.
Cytologia 5, 169–203.
Hatakeyama, K., Takasaki, T., Suzuki, G., Nishio, T., Watanabe,
M., Isogai, A., and Hinata, K. (2001). The S receptor kinase gene
determines dominance relationships in stigma expression of selfincompatibility in Brassica. Plant J. 26, 69–76.
Nei, M., and Kumar, S. (2000). Molecular Evolution and Phylogenetics. (New York: Oxford University Press).
Hinata, K., Watanabe, M., Yamakawa, S., Satta, Y., and Isogai, A.
(1995). Evolutionary aspects of the S-related genes of the Brassica self-incompatibility system: Synonymous and nonsynonymous base substitutions. Genetics 140, 1099–1104.
Hughes, A.L., and Nei, M. (1988). Pattern of nucleotide substitution
at major histocompatibility complex class I loci reveals overdominant selection. Nature 335, 167–170.
Iwano, M., Sakamoto, K., Suzuki, G., Watanabe, M., Takayama,
S., Fukui, K., Hinata, K., and Isogai, A. (1998). Visualization of a
self-incompatibility gene in Brassica campestris L. by multicolor
FISH. Theor. Appl. Genet. 96, 751–757.
Kandasamy, M.K., Paolillo, D.J., Faraday, C.D., Nasrallah, J.B.,
and Nasrallah, M.E. (1989). The S-locus specific glycoproteins of
Brassica accumulate in the cell wall of developing stigma papillae.
Dev. Biol. 134, 462–472.
Kishi-Nishizawa, N., Isogai, A., Watanabe, M., Hinata, K.,
Yamakawa, S., Shojima, S., and Suzuki, A. (1990). Ultrastructure of papillar cells in Brassica campestris revealed by liquid
helium rapid-freezing and substitution-fixation method. Plant Cell
Physiol. 31, 1207–1219.
Kusaba, M., Nishio, T., Satta, Y., Hinata, K., and Ockendon, D.
(1997). Striking sequence similarity in inter- and intra-specific
comparisons of class I SLG alleles from Brassica oleracea and
Brassica campestris: Implications for the evolution and recognition mechanism. Proc. Natl. Acad. Sci. USA 94, 7673–7678.
Kusaba, M., Dwyer, K., Hendershot, J., Vrebalov, J., Nasrallah,
J.B., and Nasrallah, M.E. (2001). Self-incompatibility in the genus
Arabidopsis: Characterization of the S locus in the outcrossing A.
lyrata and its autogamous relative A. thaliana. Plant Cell 13, 627–643.
Lee, Y.-H., Ota, T., and Vacquier, V.D. (1995). Positive selection is
a general phenomenon in the evolution of abalone sperm lysin.
Mol. Biol. Evol. 12, 231–238.
Nishio, T., Kusaba, M., Watanabe, M., and Hinata, K. (1996). Registration of S alleles in Brassica campestris L by the restriction
fragment size of SLGs. Theor. Appl. Genet. 92, 388–394.
Nou, I.S., Watanabe, M., Isogai, A., and Hinata, K. (1993). Comparison of S-alleles and S-glycoproteins between two wild populations of Brassica campestris in Turkey and Japan. Sex. Plant
Reprod. 6, 79–86.
Ockendon, D.J. (1975). Dominance relationships between S-alleles
in the stigma of Brussels sprouts (Brassica oleracea var. gemmifera). Euphytica 24, 165–172.
Price, R.A., Palmer, J.D., and Al-Shehbaz, I.A. (1994). Systematic
relationships of Arabidopsis: A molecular and morphological perspective. In Arabidopsis, C.R. Somerville and E.M. Meyerowitz,
eds (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory
Press), pp. 7–19.
Reeder, R.H. (1985). Mechanisms of nucleolar dominance in animals and plants. J. Cell Biol. 101, 2013–2016.
Reeder, R.H., and Roan, J.G. (1984). The mechanism of nucleolar
dominance in Xenopus hybrids. Cell 38, 39–44.
Sakamoto, K., Kusaba, M., and Nishio, T. (2000). Single-seed
PCR-RFLP analysis for the identification of S haplotypes in commercial F1 hybrid cultivars of broccoli and cabbage. Plant Cell
Rep. 19, 400–406.
Schierup, M.H., Vekemans, X., and Christiansen, F.B. (1998).
Allelic genealogies in sporophytic self-incompatibility systems in
plants. Genetics 150, 1187–1198.
Schierup, M.H., Mable, B.K., Awadalla, P., and Charlesworth, D.
(2001). Identification and characterization of a polymorphic receptor kinase gene linked to the self-incompatibility locus of Arabidopsis lyrata. Genetics 158, 387–399.
Schopfer, C.R., and Nasrallah, J.B. (2000). Self-incompatibility:
Prospects for a novel putative peptide-signaling molecule. Plant
Physiol. 124, 935–939.
Li, W.-H. (1997). Molecular Evolution. (Sunderland, MA: Sinauer
Associates).
Schopfer, C.R., Nasrallah, M.E., and Nasrallah, J.B. (1999). The
male determinant of self-incompatibility in Brassica. Science 286,
1697–1700.
Liu, Y.-G., and Whittier, R.F. (1994). Rapid preparation of megabase plant DNA from nuclei in agarose plugs and microbeads.
Nucleic Acids Res. 22, 2168–2169.
Scutt, C.P., and Croy, R.R.D. (1992). An S5 self-incompatibility
allele-specific cDNA sequence from Brassica oleracea shows high
homology to the SLR2 gene. Mol. Gen. Genet. 232, 240–246.
Matsuda, N., Tsuchiya, T., Kishitani, S., Tanaka, Y., and
Toriyama, K. (1996). Partial male sterility in transgenic tobacco
carrying antisense and sense PAL cDNA under the control of a
tapetum-specific promoter. Plant Cell Physiol. 37, 215–222.
Shiba, H., Kimura, N., Takayama, S., Hinata, K., Suzuki, A., and
Isogai, A. (2000). Alteration of the self-incompatibility phenotype
in Brassica by transformation of the antisense SLG gene. Biosci.
Biotechnol. Biochem. 64, 1016–1024.
Murray, M.G., and Thompson, W.F. (1980). Rapid isolation of high
molecular weight plant DNA. Nucleic Acids Res. 8, 4321–4325.
Nasrallah, J.B., and Nasrallah, M.E. (1993). Pollen–stigma signaling in the sporophytic self-incompatibility response. Plant Cell 5,
1325–1335.
Shiba, H., Takayama, S., Iwano, M., Shimosato, H., Funato, M.,
Nakagawa, T., Che, F.-S., Suzuki, G., Watanabe, M., Hinata,
K., and Isogai, A. (2001). A pollen coat protein, SP11/SCR, determines the pollen S-specificity in the self-incompatibility of Brassica species. Plant Physiol. 125, 2095–2103.
Nasrallah, J.B., Kao, T.-H., Chen, C.-H., Goldberg, M.L., and
Nasrallah, M.E. (1987). Amino acid sequence of glycoproteins
encoded by three alleles of the S-locus of Brassica oleracea.
Nature 326, 617–619.
Stein, J.C., Howlett, B., Boyes, D.C., Nasrallah, M.E., and Nasrallah,
J.B. (1991). Molecular cloning of a putative receptor protein
kinase gene encoded at the self-incompatibility locus of Brassica
oleracea. Proc. Natl. Acad. Sci. USA 88, 8816–8820.
504
The Plant Cell
Suzuki, G., Kai, N., Hirose, T., Fukui, K., Nishio, T., Takayama, S.,
Isogai, A., Watanabe, M., and Hinata, K. (1999). Genomic organization of the S locus: Identification and characterization of
genes in SLG/SRK region of S9 haplotype of Brassica campestris
(syn. rapa). Genetics 153, 391–400.
Takasaki, T., Hatakeyama, K., Watanabe, M., Toriyama, K.,
Isogai, A., and Hinata, K. (1999). Introduction of SLG (S locus
glycoprotein) alters the phenotype of endogenous S haplotype,
but confers no new S haplotype specificity in Brassica rapa L.
Plant Mol. Biol. 40, 659–668.
Takasaki, T., Hatakeyama, K., Suzuki, G., Watanabe, M., Isogai,
A., and Hinata, K. (2000). The S receptor kinase determines selfincompatibility in Brassica stigma. Nature 403, 913–916.
the rice genomic P1 library. Sci. Rep. Kyoto Pref. Univ. Agric. 45,
26–33.
Thompson, K.F., and Taylor, J.P. (1966). Non-linear dominance
relationships between S alleles. Heredity 21, 345–362.
Uyenoyama, M.K. (1995). A generalized least-squares estimate for the
origin of sporophytic self-incompatibility. Genetics 139, 975–992.
Uyenoyama, M.K. (2000). Evolutionary dynamics of self-incompatibility alleles in Brassica. Genetics 156, 351–359.
Visser, D.L., van Hal, J.G., and Verhoeven, W. (1982). Classification of S-alleles by their activity in S-heterozygotes of Brussels
sprouts (Brassica oleracea var. gemmifera (DC.) Schultz). Euphytica
31, 603–611.
Takayama, S., Isogai, A., Tsukamoto, C., Ueda, Y., Hinata, K.,
Okazaki, K., and Suzuki, A. (1987). Sequences of S-glycoproteins, products of the Brassica campestris self-incompatibility
locus. Nature 326, 102–104.
Watanabe, M., Suzuki, G., Toriyama, K., Takayama, S., Isogai, A.,
and Hinata, K. (1999). Two anther-expressed genes downstream
of SLG9: Identification of a novel S-linked gene specifically
expressed in anthers at the uninucleate stage of Brassica campestris (syn. rapa) L. Sex. Plant Reprod. 12, 127–134.
Takayama, S., Shiba, H., Iwano, M., Shimosato, H., Che, F.-S.,
Kai, N., Watanabe, M., Suzuki, G., Hinata, H., and Isogai, A.
(2000a). The pollen determinant of self-incompatibility in Brassica
campestris. Proc. Natl. Acad. Sci. USA 97, 1920–1925.
Watanabe, M., et al. (2000). Highly divergent sequences of the pollen self-incompatibility (S) gene in class I S haplotypes of Brassica
campestris (syn. rapa) L. FEBS Lett. 473, 139–144.
Takayama, S., Shiba, H., Iwano, M., Asano, K., Hara, M., Che,
F.-S., Watanabe, M., Hinata, K., and Isogai, A. (2000b). Isolation
and characterization of pollen coat proteins of Brassica campestris that interact with S locus–related glycoprotein 1 involved in
pollen–stigma adhesion. Proc. Natl. Acad. Sci. USA 97, 3765–
3770.
Takayama, S., Shimosato, H., Shiba, H., Funato, M., Che, F.-S.,
Watanabe, M., Iwano, M., and Isogai, A. (2001). Direct ligand–
receptor complex interaction controls Brassica self-incompatibility. Nature 413, 535–538.
Tanaka, K., Makino, H., and Masumura, T. (1993). Construction of
Yang, Y.-S., Mitta, G., Chavanieu, A., Calas, B., Sanchez, J.F.,
Roch, P., and Aumelas, A. (2000). Solution structure and activity
of the synthetic four-disulfide bond Mediterranean mussel defensin
(MGD-1). Biochemistry 39, 14436–14447.
Yang, Y.W., Lai, K.N., Tai, P.Y., Ma, D.P., and Li, W.H. (1999).
Rates of nucleotide substitution in angiosperm mitochondrial DNA
sequences and dates of divergence between Brassica and other
angiosperm lineages. J. Mol. Evol. 48, 597–604.
Zabaleta, E., Heiser, V., Grohmann, L., and Brennicke, A. (1998).
Promoters of nuclear-encoded respiratory chain complex I genes
from Arabidopsis thaliana contain a region essential for anther/
pollen-specific expression. Plant J. 15, 49–59.