Ó Springer-Verlag 2000
Mol Gen Genet (2000) 263: 648±654
ORIGINAL PAPER
Y.-M. Bi á N. BrugieÁre á Y. Cui á D. R. Goring
S. J. Rothstein
Transformation of Arabidopsis with a Brassica SLG /SRK region
and ARC1 gene is not suf®cient to transfer the self-incompatibility
phenotype
Received: 11 November 1999 / Accepted: 14 February 2000
Abstract Self-incompatibility (SI) promotes outbreeding in ¯owering plants, and in Brassica SI is genetically
controlled by the S locus. Self-incompatible Brassica
and self-fertile Arabidopsis belong to the same crucifer
family. In addition, a comparative analysis reveals a
high degree of microsynteny between the B. campestris S
locus and its homologous region in Arabidopsis ± with
the notable exception that the Brassica SI genes, SLG
and SRK, are missing. Brassica ARC1 encodes a component of the SRK signal transduction pathway leading
to self-pollen rejection, and no closely related ARC1
homolog has been identi®ed in Arabidopsis. The purpose
of the research reported here was to introduce Brassica
SI components into Arabidopsis in an attempt to compensate for the missing genes and to investigate whether
the SI phenotype can be transferred. Inserts of approximately 40 kb from the fosmid clones F20 and F22,
which span the B. napus W1 SLG-SRK region, were
cloned into the plant transformation vector pBIBAC2.
Transgenic plants were generated that expressed the
Brassica SI genes in the ¯ower buds. In addition, the
endogenous, SLG-like, gene AtS1 was not co-suppressed
by the Brassica SLG transgene. No SI phenotype was
observed among the T1 BIBAC2-F20 and BIBAC2-F22
transgenic plants. When the ARC1 gene was transformed into BIBAC2-F20 or BIBAC2-F22 plants, the
resulting BIBAC2-F20-ARC1 and BIBAC2-F22-ARC1
Communicated by G. Haughn
Y.-M. Bi á N. BrugieÁre á Y. Cui á S. J. Rothstein
Department of Molecular Biology and Genetics,
University of Guelph, Guelph, Ontario, N1G 2W1 Canada
D. R. Goring
Biology Department, York University,
4700 Keele Street, North York,
Ontario, M3J 1P3 Canada
S. J. Rothstein (&)
Department of Agronomic Traits,
Pioneer Hi-Bred International Inc.,
P.O. Box 552, Johnston, IA 50131, USA
e-mail: rothsteinsj@phibred.com; Fax: +1-515-334-4778
plants still set seeds normally, and no rejection response
was observed when self-incompatible B. napus W1 pollen was placed on BIBAC2-F20-ARC1 or BIBAC2-F22ARC1 Arabidopsis stigmas. Taken together, our results
suggest that complementing Arabidopsis genome with
Brassica SLG, SRK and ARC1 genes is unlikely to be
suf®cient to transfer the SI phenotype.
Key words Self-incompatibility á SLG (S locus
glycoprotein) á SRK (S locus receptor kinase) á ARC1
(Arm repeat containing) á Transgenic Arabidopsis
Introduction
Self-incompatibility (SI) is one of the mechanisms present in ¯owering plants that prevents self-fertilization
and promotes out-crossing. In Brassica, SI is controlled
genetically by multiple alleles of the S locus, and is
sporophytic in nature, in that the pollen phenotype is
determined by the parental genotype (Dodds et al.
1996). Several S locus genes have been discovered, and
there is evidence that two of these are required for the SI
phenotype. They encode, respectively, an S locus glycoprotein (SLG) and a plasma membrane-spanning receptor protein kinase (SRK). Both genes are expressed
speci®cally in the stigma papillae cells and they are
thought to be responsible for the female side of the SI
phenotype when the communication between pistil and
pollen begins (Nasrallah and Nasrallah 1993; Nasrallah
et al. 1994a). The expression of SLG and SRK is necessary for the SI function, as plants carrying mutated or
down-regulated SLG or SRK genes are self-fertile
(Nasrallah et al. 1992; Goring et al. 1993; Nasrallah
et al. 1994b; Conner et al. 1997; Stahl et al. 1998). The
current model for SI in Brassica proposes that the pollen-borne ligand interacts with SLG and SRK, activates
the SRK protein kinase and thus triggers a signal cascade that eventually leads to the rejection of self-pollen
(Nasrallah et al. 1994a). According to the model, at least
one gene required for the SI phenotype on the male side
649
should be encoded in the S locus complex. Although S
locus genes that are speci®cally expressed in the anther
have been isolated, such as SLL1 and SLA (Boyes and
Nasrallah 1995; Yu et al. 1996), they do not ful®ll all the
requirements that one would expect for the male determinant (Yu et al. 1996; Pastuglia et al. 1997). Very recently, a new gene named SCR (S locus cysteine-rich)
was isolated in Brassica (Schopfer et al. 1999). It is
encoded at the S locus and is a highly polymorphic,
anther-expressed gene. Functional studies proved that it
is necessary and suf®cient for determining pollen SI
speci®city, suggesting that it is a ligand for the stigmatic
S receptor complex. Meanwhile, in an attempt to dissect
the SRK-mediated signal transduction pathway, an
ARC1 (Arm Repeat Containing) gene was isolated by
employing the yeast two-hybrid system. It encodes a
protein that interacts speci®cally with the SRK kinase
domain, and its expression is restricted to the stigma, the
site of the SI response (Gu et al. 1998). In addition,
down-regulation of ARC1 mRNA levels in transgenic
plants that express ARC1 antisense RNA is correlated
with a partial breakdown of SI, providing evidence that
ARC1 is a positive effector of this response (Stone et al.
1999).
Recently, a fosmid contig of 65 kb from a functional S-haplotype (B. napus W1 line) was isolated and
analyzed (Cui et al. 1999). In our preliminary studies on
transgenic B. napus plants, the 40-kb fragments (from
fosmid F20 or F22) each appeared to be capable of
conferring the SI female phenotype (Cui et al. 2000). In
this work, we attempted to transform these SLG/SRKspanning regions into Arabidopsis, based on the following considerations. First, the out-crossing Brassica
and the self-fertilizing Arabidopsis belong to the same
crucifer family. A high-resolution comparative analysis
of the organization of the S locus region in one genotype
of B. campestris (S8 haplotype) and a region of the
Arabidopsis genome has already been conducted by
Conner et al. (1998). The Arabidopsis region was identi®ed as homologous to the Brassica S locus and their
results revealed a high degree of microsynteny between
the Brassica S locus region and its homolog in Arabidopsis, with the notable exception that sequences related
to Brassica SI genes were missing in the Arabidopsis
genome (Conner et al. 1998). Thus, by addressing
whether these Brassica SI components could transfer the
phenotype into Arabidopsis, we were hoping to shed
some light on the mechanisms underlying transitions in
mating systems in this crucifer family. Second, the life
cycle of Arabidopsis is signi®cantly shorter than that of
Brassica, so that further investigations could be done
considerably more quickly if an interesting change in
phenotype was observed in the transgenic plants.
Therefore, we transformed Arabidopsis with the genomic
fragment containing the SLG and SRK genes from this
910 S-haplotype (B. napus W1 line) and with the Brassica ARC1 gene, and analyzed the resulting transgenic
plants for the expression of the Brassica genes as well as
examining them for any changes in phenotype.
Materials and methods
Vector constructs and Agrobacterium strains used for plant
transformation
F20 and F22 were identi®ed from a fosmid library (Cui et al. 1999).
The F20, F22 and BIBAC2 (Hamilton 1997) DNAs were prepared
using a Qiagen kit. NotI digestion released the F20 or F22 insert.
After pulsed ®eld gel electrophoresis (PFGE), the 40-kb insert band
was excised from the agarose gel and subjected to digestion with
GELase (Epicentre). The BIBAC2 vector was digested with NotI
and dephosphorylated with shrimp alkaline phosphatase (Epicentre). Aliquots (100 ng) of F20 or F22 insert were co-precipitated
with 40 ng of NotI-digested, dephosphorylated BIBAC2 vector.
The precipitates were dissolved in 20 ll of distilled water. Ligation
was carried out under standard conditions and the ligated DNA
was electroporated into E. coli strain DH10B (GIBCO-BRL Life
Technologies). The resulting construct (pBIBAC2-F20 or pBIBAC2-F22) was introduced into the Agrobacterium strain COR338
(Hamilton 1997) by triparental mating (Draper et al. 1988).
COR338 is the Agrobacterium strain GV3101 (Koncz and Schell
1986) containing the plasmid pCH32, which carries the extra virulent genes required for the transfer of large fragments (Hamilton
et al. 1996; Hamilton 1997). After several generations, Southern
hybridization was performed to con®rm that the plasmid was
stably maintained in COR338 (data not shown). The resulting
Agrobacterium strains, pBIBAC2-F20/COR338 and pBIBAC2F22/COR338, were used for Arabidopsis transformation. The
transformation vector carrying the ARC1 sense cDNA under the
direction of the SLR1 promoter was introduced into Agrobacterium
strain GV2260 (Deblaere et al. 1985) by triparental mating (Draper
et al. 1988).
Arabidopsis transformation and screening for transgenic plants
Transgenic Arabidopsis plants of ecotype Columbia were generated
by using the A. tumefaciens-mediated whole-plant in®ltration protocol (Bechtold et al. 1993), except that ¯owers were dipped for 6
min in the in®ltration medium supplied with Silwet L-77 (Lehle
Seeds) instead of using vacuum in®ltration. Seeds were harvested
and then germinated on germination medium (1/2 MS, 10 g/l sucrose, pH 5.8; Clarke et al. 1992) containing 50 mg/l kanamycin
for selection. One week later, kanamycin-resistant seedlings were
transferred into soil.
Southern analysis
Genomic DNA was extracted from leaf tissues of transgenic plants
and non-transformed control plants using the DNA extraction kit
Phytopure (Amersham). Aliquots (1 lg) of DNA were digested
with either EcoRI or NotI. EcoRI-digested DNA was fractionated
on a 0.7% agarose gel and NotI-digested DNA was fractionated by
PFGE. The DNA was transferred to a nylon membrane (Boehringer Mannheim) and probed with the DIG-labeled SLG cDNA
(for EcoRI digestion) or with the radioactively labeled SLG and
SRK cDNAs (for NotI digestion). Hybridizations were carried out
under standard conditions and ®lters were washed twice at 65 °C
with 0.1 ´ SSC, 0.1% SDS for 20 min (Sambrook et al. 1989).
Detection of the signal from the DIG probe was done according to
the manufacturer's (Boehringer Mannheim) protocol.
Northern analysis
Total RNA was extracted according to Verwoerd et al. (1989) from
transformed and untransformed Arabidopsis ¯ower buds or from
W1 stigmas. Aliquots of total RNA from Arabidopsis (15 lg) and
from W1 stigmas were loaded and fractionated on formaldehyde
gels as described in BrugieÁre et al. (1999), and transferred to nylon
650
®lters for hybridization. Probes were prepared using a randomprimer DNA labeling kit (Boehringer Mannheim). Hybridization
was carried out under standard conditions and ®lters were washed
under the same conditions as for Southern analysis, described
above.
molecular-weight T-DNAs into plants (Hamilton 1997).
The F20 and F22 inserts were each cloned into BIBAC2,
and the resulting pBIBAC2-F20 and pBIBAC2-F22
(Fig. 1) were individually transformed into A. tumefaciens strain COR338 (see Materials and methods).
Pollination assay
Arabidopsis pollen or Brassica pollen was placed on Arabidopsis
stigmas and allowed to develop overnight before the pistils were
removed. These were ®xed in a mixture of 4% formaldehyde in
1 ´ PBS for 2 h at room temperature (HuÈlskamp et al. 1995).
Following two washes in 1 ´ PBS, the pistils were placed in 4 M
NaOH for 10 min to soften the tissue before being squashed in a
drop of 0.05% aniline blue in 70 mM sodium phosphate buffer pH
8.0. Squashes were examined using a ¯uorescence microscope.
Results and discussion
Construction of pBIBAC2-F20 and pBIBAC2-F22
for Arabidopsis transformation
Fosmid clones, F20 and F22, isolated from a fosmid
library of B. napus W1 genomic DNA (Cui et al. 1999)
were used in this study. As shown in Fig. 1, the F20 or
F22 inserts both contain the SLG and SRK genes, which
are missing from the region of the Arabidopsis genome
that is otherwise homologous to the Brassica S locus.
The F20 insert carries more DNA downstream of the
SRK and F22 includes the region upstream from the
SLG. Detailed sequence analysis of these two fragments
was done and no obvious additional S-genes were
identi®ed (Cui et al. 1999). Since the F20 and F22 inserts
both span the SLG and SRK region, which is absent in
the Arabidopsis genome, we wished to determine
whether these fragments might be suf®cient to transfer
the SI phenotype. The binary-BAC (BIBAC) vectors
were constructed for the purpose of transferring high-
Fig. 1 Schematic diagram of the T-DNA region of the plasmids
pBIBAC2-F20 and pBIBAC2-F22. The F20 or F22 insert was cloned
into BIBAC2 at the NotI site to form pBIBAC2-F20 and pBIBAC2F22. Both inserts are about 40 kb long. EcoRI digestion can release a
25-kb fragment from F22 and a band of variable size from F20
(depending on the insertion site) that can be detected by hybridization
with the SLG gene
Identi®cation of transgenic BIBAC2-F20 and
BIBAC2-F22 Arabidopsis
Agrobacterium-mediated in planta transformation (see
Materials and methods) resulted in 19 kanamycinresistant plants for the pBIBAC2-F20 construct and 18
for the pBIBAC2-F22 construct after screening about
80,000 and 85,000 seeds, respectively. The transformation frequency seems to be 5±10 times lower than when
this method is used for transformation of small fragments. The transgenic nature of these plants was con®rmed by GUS histochemical assay, PCR (data not
shown) and by Southern analysis (Fig. 2). Genomic
DNA extracted from two randomly selected transgenic
lines (one for each construct) and a non-transformed
control plant was digested with EcoRI. As expected, a
25-kb band from the pBIBAC2-F22 construct and a
band smaller than 25 kb from the pBIBAC2-F20 construct were detected with the SLG probe (Fig. 2a). Since
transfer of the T-DNA is initiated at the right border
and ends at the left border and the nptII gene lies at the
end of the left border, it is very likely that the entire TDNA region was present in kanamycin resistant transgenic Arabidopsis. It was later con®rmed by PFGE blot
analysis that the inserts of fosmid clones were transferred into Arabidopsis intact. As shown in Fig. 2b, genomic DNA isolated from two randomly selected
transgenic lines and a control plant was digested with
NotI and the whole 40-kb insert was released. In both
Southerns, no signal was obtained from the non-transformed control plant. Perhaps due to the large size of the
insert, all the transgenic plants have a single insertion
since they all showed kanamycin resistance with a segregation ratio around 3:1 (data not shown).
Fig. 2a, b Southern analysis of transgenic BIBAC2-F20 and BIBAC2-F22 plants. a Genomic DNA extracted from a wild-type (wt)
plant or transgenic plant (t-1 from BIBAC2-F22 and t-2 from
BIBAC2-F20) was digested with EcoRI and probed with DIG-labeled
SLG cDNA. b Genomic DNA extracted from a wild type (wt) plant
or transgenic plant (t-3 from BIBAC2-F20 and t-4 from BIBAC2F22) was digested with NotI and probed with 32P-labeled SLG and
SRK cDNAs
651
Phenotype of the transgenic plants
SI in Brassica is controlled sporophytically; hence, if
Brassica SI genes are the only missing components in
Arabidopsis, we should expect to see an SI phenotype in
the T1 transgenic Arabidopsis plants. However, the T1
transgenic plants (37 in total for two constructs) were
phenotypically normal and set seeds well, apart from
three mutant plants. Among these three plants, one was
derived from the pBIBAC2-F20 construct and two from
the pBIBAC2-F22 construct. The plant transformed
with the pBIBAC2-F20 construct grew as vigorously as
wild-type plants, but it had very small siliques and the
seed yield was only 10% of the wild type. It was excluded later that the phenotype of this mutant plant
might be SI-related because it was found that the mutant
phenotype was not co-segregating with the transgenes.
The other two mutant plants from the pBIBAC2-F22
transformation were sterile so that we were unable to
obtain any descendants for further characterization. A
second transformation experiment was performed with
the pBIBAC2-F22 construct. Thirty-®ve kanamycin-resistant plants were generated after screening approximately 200,000 seeds. Apart from one plant which
appeared unhealthy and died eventually, the remaining
34 plants were all phenotypically normal and set seeds
like the wild type.
Expression of Brassica SI genes in transgenic
plants and expression of the endogenous AtS1 gene
Since no SI phenotype was observed in the transgenic
plants, it was important to determine whether the
Brassica SI genes were expressed properly in the transgenic Arabidopsis plants. Northern analysis was performed, and the 1.6-kb transcript corresponding to the
introduced SLG gene and the 2.8-kb transcript corresponding to the introduced SRK gene were detected in
the transgenic ¯ower buds, but not in the buds of a nontransformed control plant (Fig. 3). The expression level
in Arabidopsis was slightly lower than that in the W1
control, which may be due to the different materials used
for RNA extraction (stigmas for W1 and buds for
Arabidopsis). To determine whether the expression pattern of these two genes was still tissue speci®c, RNA was
extracted from leaf tissues and ¯ower buds, and hybridized with the SLG and SRK genes. The expression of
these two genes was limited to the buds, as in B. napus
(data not shown).
Four genes have been isolated from Arabidopsis
based on sequence homology with SLG and SRK probes
(Dwyer et al. 1992, 1994; Tobias et al. 1992). Among
these, the three SRK-like genes ARK1, ARK2 and ARK3
are expressed in vegetative tissue. The fourth gene AtS1
is a SLG-like gene and is speci®cally expressed in reproductive tissues (Dwyer et al. 1992). The expression of
AtS1 was checked to determine if its expression was
altered by the transgenes. RNA extracted from wild-type
Fig. 3 Northern analysis of SLG and SRK genes. Total RNA was
extracted from B. napus W1 stigmas (W1), wild-type Arabidopsis
¯ower buds (wt) and buds of various transgenic BIBAC2-F20 plants
(a) (F20-1, 2, 5, 8, 11 and 20) or BIBAC2-F22 plants (b) (F22-2, 6, 7,
10, 11). Aliquots of RNA (15 lg for Arabidopsis samples and 4 lg for
W1) were loaded and hybridized with SLG and SRK cDNAs. Both
blots were re-hybridized with the 18S rRNA gene isolated from B.
napus to show that equal amounts of Arabidopsis samples had been
loaded
plants and two transgenic plants were hybridized with
the SLG probe and re-probed with an AtS1 probe. An
SLG signal was obtained only from transgenic plants
but AtS1 signal was present in both wild-type and
transgenic samples (Fig. 4). The same results were obtained from other transgenic lines (data not shown),
indicating that there was no change in the expression of
the Arabidopsis AtS1 gene.
Transformation of the Brassica ARC1 gene
into BIBAC2-F20 or BIBAC2-F22 Arabidopsis
Despite the fact that the Brassica SI genes were well
transcribed in the transgenic Arabidopsis buds, no SI
phenotype was detected. This implies that (an)other
component(s) required for the SI response is (are) absent
in the Arabidopsis genome. The ARC1 gene was isolated
from Brassica through the use of the yeast two-hybrid
system (Gu et al. 1998). Later, in vivo functional assays
demonstrated that it is indeed one of the components
needed in the signal transduction pathway leading to the
rejection of self-pollen in Brassica (Stone et al. 1999).
Since no closely related ARC1 homologs have been
identi®ed in the Arabidopsis genome (data not shown),
the ARC1 gene was transformed into Arabidopsis and
introduced into the BIBAC2-F20 or BIBAC2-F22
transgenic plants via genetic crosses. ARC1 was cloned
652
Fig. 4 Northern analysis of AtS1 expression. Total RNA extracted
from wild-type Arabidopsis ¯ower buds (wt) and transgenic buds
(F20-2, F22-10) were hybridized with SLG cDNA and subsequently
with an AtS1 probe. Total RNA stained with ethidium bromide is
shown at the bottom to demonstrate the approximately equal loading
of the total RNA
into a binary vector in the sense orientation, driven by
the SLR1 (S locus related) promoter (Hackett et al.
1996). The SLR1 gene also encodes a secreted glycoprotein and belongs to the same gene family as the SLG
gene, although it is not located at the S locus (Lalonde
et al. 1989; Hinata et al. 1995). Nevertheless, SLR1
shows the same developmental pattern of expression as
the SLG gene during ¯ower maturation (Umbach et al.
1990). The level of ARC1 expression increased as ¯ower
buds developed and reached its highest level in the large
buds (6±7mm; Gu et al. 1998), which is identical to the
timing of SLG expression (Goring and Rothstein 1992).
Flower buds were harvested from transgenic plants,
which were identi®ed as being kanamycin resistant and
carried the T-DNA integrated at a single locus, to determine the expression level of the ARC1 gene. As shown
in Fig. 5, ARC1 was expressed in most transgenic lines.
Pollen from transgenic lines expressing relatively high
levels of ARC1 (ARC1-2 And ARC1-12) was used to
pollinate stigmas from BIBAC-F20 or BIBAC2-F22
Fig. 6a, b Germination of
Arabidopsis and Brassica pollen
on transgenic Arabidopsis stigmas. a Stigmas of transgenic
Arabidopsis pollinated with
Arabidopsis pollen. b Stigmas of
transgenic Arabidopsis pollinated with B. napus W1 pollen.
Notice that the Brassica pollen
is bigger and the pollen tubes
are fatter than those from Arabidopsis. Both pictures are
shown at the same magni®cation. Bar 0.05 mm
Fig. 5 Northern analysis of ARC1 gene. Total RNA extracted from
B. napus W1 stigmas (W1), wild.type Arabidopsis ¯ower buds (wt) and
buds from various transgenic ARC1 plants (ARC1±2, 5, 6, 10, 12, 18,
21) were hybridized with the ARC1 coding region. Total RNA stained
with ethidium bromide is shown at the bottom to demonstrate the
approximately equal loading of the Arabidopsis samples
transgenic plants. PCR assays were used to detect the
presence of both the ARC1 gene and the SLG/SRK
genes.
Pollination analysis of BIBAC2-F20-ARC1
and BIBAC2-F22-ARC1 transgenic plants
The BIBAC2-F20-ARC1 or BIBAC2-F22-ARC1 transgenic plants still set seeds well. When the transgenic
Arabidopsis plants were fertilized with self-pollen, the
pollen tubes grew very well (Fig. 6a). Thus, even with
the expression of ARC1, no SI reaction seemed to have
occurred. Results obtained with transgenic Brassica
plants carrying the 40-kb fragment (from F20 or F22)
showed that this region is suf®cient to confer the female
part of the SI phenotype, but not the male part (Cui
et al. 2000). Thus, while the transgenic Arabidopsis
plants are not self-incompatible, they still may contain
the components necessary for the female part of the SI
reaction. Since self-incompatible B. napus W1 pollen
should contain the male components, we placed W1
pollen on the stigmas of BIBAC2-F20-ARC1 or BIBAC2-F22-ARC1 transgenic Arabidopsis plants to see if
pollen hydration and penetration would be prevented.
653
When there is a strong SI reaction in Brassica, the pollen
generally does not germinate. For those few that do, the
pollen tubes usually do not penetrate the stigma surface.
According to the well documented unilateral incompatibility phenomenon, pollen of the self-incompatible
species could develop in the styles of the self-compatible
species while inhibition would occur in the reciprocal
cross (Lewis and Crowe 1958). Previous studies also
indicated that Brassica pollen was able to elicit the water
release response, hydrate and germinate when deposited
on Arabidopsis stigmas (HuÈlskamp et al. 1995). However, when W1 pollen was placed on the stigmas of
BIBAC2-F20-ARC1 or BIBAC2-F22-ARC1 transgenic
plants, the SI rejection response did not occur and W1
pollen hydration and tube growth were observed
(Fig. 6b), at similar levels to those seen when the pollen
was placed on wild-type Arabidopsis stigmas (data not
shown). Although the cellular interactions required for
the initial recognition of the pollen grain by the stigmatic
cell are fairly non-speci®c, no seeds would be produced,
though, as the interspeci®c fertilization is blocked at a
later stage (HuÈlskamp et al. 1995).
In summary, we have transformed Brassica SI genes
(SLG, SRK and ARC1 genes) into Arabidopsis and
demonstrated that the Brassica SI genes were well
transcribed in the transgenic Arabidopsis ¯ower buds.
However, no SI phenotype was observed among the
transgenic plants obtained. Also, no detectable difference was seen in pollen tube growth when Brassica SI
pollen was placed on either transgenic or wild-type
Arabidopsis stigmas. There are several conceivable reasons for the failure to transfer the SI phenotype. Arabidopsis might lack additional genes involved in the SI
response or the signal transduction process leading to
pollen rejection. Alternatively, some other differences in
the fertilization process might be the cause. Also, it is
possible that the transcripts of the Brassica genes were
not present at adequate levels in the transgenic Arabidopsis or their protein products were not properly
processed; both of these factors are important for
self-fertility in Brassica.
Acknowledgements We thank Dr. C. Hamilton from Cornell
University, who kindly provided the BIBAC2 vector and the
COR338 Agrobacterium strain. We also thank Richard Stahl for
useful discussions. This work was funded by grants from the
Natural Sciences and Engineering Research Council (NSERC) of
Canada to S.J.R.
References
Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium
mediated gene transfer by in®ltration of adult Arabidopsis
thaliana plants. C R Acad Sci 316: 1194±1199
Boyes DC, Nasrallah JB (1995) An anther-speci®c gene encoded by
an S locus haplotype of Brassica produces complementary and
differentially regulated transcripts. Plant Cell 7: 1283±1294
BrugieÁre N, Dubois F, Limami AM, Lelandais M, Roux Y,
Sangwan RS, Hirel B (1999) Glutamine synthetase in the
phloem plays a major role in controlling proline production.
Plant Cell 11: 1995±2011
Clarke MC, Wei W, Lindsey K (1992) High-frequency transformation of Arabidopsis thaliana by Agrobacterium tumefaciens.
Plant Mol Biol Rep 10: 178±189
Conner JA, Tantikanjana T, Stein JC, Kandasamy MK, Nasrallah
JB, Nasrallah ME (1997) Transgene-induced silencing of S locus genes and related genes in Brassica. Plant J 11: 809±823
Conner JA, Conner P, Nasrallah ME, Nasrallah JB (1998)
Comparative mapping of the Brassica S locus region and its
homeolog in Arabidopsis: implications for the evolution of
mating systems in the Brassicaceae. Plant Cell 10: 801±812
Cui Y, BrugieÁre N, Jackman LM, Bi YM, Rothstein SJ (1999) A
Structural and transcriptional comparative analysis of the S
locus regions in two self-incompatible Brassica napus lines.
Plant Cell 11: 2217±2231
Cui Y, Bi YM, BrugieÁre N, Arnoldo M, Rothstein SJ (2000) The
S-locus glycoprotein and the S receptor kinase are sucient for
self pollen rejection in Brassica. Proc Natl Acad Sci USA,
in press
Deblaere R, Bytebier B, De Greve H, Deboeck F, Schell J, Van
Montagu M, Leemans J (1985) Ecient octopine Ti plasmidderived vectors for Agrobacterium-mediated gene transfer to
plants. Nucleic Acids Res 13: 4777±4788
Dodds PN, Clarke AE, Newbiggin E (1996) A molecular perspective on pollination in ¯owering plants. Cell 85: 141±144
Draper J, Scott R, Armitage P, Walden R (1988) Plant genetic
transformation and gene expression: a laboratory manual.
Blackwell Scienti®c, Oxford, UK
Dwyer KG, Lalonde BA, Nasrallah JB, Nasrallah ME (1992)
Structure and expression of AtS1, an Arabidopsis thaliana gene
homologous to the S locus related genes of Brassica. Mol Gen
Genet 231: 442±448
Dwyer KG, Kandasamy MK, Mahosky DI, Acciai J, Kudish BI,
Miller JE, Nasrallah ME, Nasrallah JB (1994) A superfamily of
S locus-related sequences in Arabidopsis: diverse structures and
expression patterns. Plant Cell 6: 1829±1843
Goring DR, Rothstein SJ (1992) The S-locus receptor kinase gene
in a self-incompatible Brassica napus line encodes a functional
serine/threonine kinase. Plant Cell 4: 1273±1281
Goring DR, Glavin TL, Schafer U, Rothstein SJ (1993) An S receptor kinase gene in self-compatible Brassica napus has a 1-bp
deletion. Plant Cell 5: 531±539
Gu T, Mazzurco M, Sulaman W, Matias DD, Goring D (1998)
Binding of an arm repeat protein to the kinase domain of the S
locus receptor kinase. Proc Natl Acad Sci USA 95: 382±387
Hackett RM, Cadwallader G, Franklin FC (1996) Functional
analysis of a Brassica oleracea SLR1 gene promoter. Plant
Physiol 112: 1601±1607
Hamilton CM (1997) A binary-BAC system for plant transformation with high-molecular-weight DNA. Gene 200: 107±116
Hamilton CM, Frary A, Lewis C, Tanksley SD (1996) Stable
transfer of intact high molecular weight DNA into plant
chromosomes. Proc Natl Acad Sci USA 93: 9975±9979
Hinata K, Watanabe M, Yamakawa S, Satta Y, Isogai A (1995)
Evolutionary aspects of the S-related genes of the Brassica selfincompatibility system: synonymous and non-synonymous base
substitutions. Genetics 140: 1099±1104
HuÈlskamp M, Kopczak SD, Horejsi T, Kihl BK, Pruitt RE (1995)
Identi®cation of genes required for pollen-stigma recognition in
Arabidopsis thaliana. Plant J 8: 703±714
Koncz C, Schell J (1986) The promoter of TL-DNA gene 5 controls
the tissue-speci®c expression of chimeric genes carried by a
novel type of Agrobacterium binary vector. Mol Gen Genet 204:
383±396
Lalonde BA, Nasrallah ME, Dwyer KG, Chen CH, Barlow B,
Nasrallah JB (1989) A highly conserved Brassica gene with
homology to the S locus-speci®c glycoprotein structural gene.
Plant Cell 1: 249±258
Lewis D, Crowe LK (1958) Unilateral interspeci®c incompatiblity
in ¯owering plants. Heredity 12: 233±256
654
Nasrallah JB, Nasrallah ME (1993) Pollen-stigma signaling in the
sporophytic self-incompatibility response. Plant Cell 5: 1325±
1335
Nasrallah JB, Kandasamy MK, Nasrallah ME (1992) A genetically
de®ned trans-acting locus regulates S locus function in Brassica.
Plant J. 2: 497±506
Nasrallah JB, Stein JC, Kandasamy MK, Nasrallah ME (1994a)
Signaling the arrest of pollen tube development in self-incompatible plants. Science 266: 1505±1508
Nasrallah JB, Rundle SJ, Nasrallah ME (1994b) Genetic evidence
for the requirement of the Brassica S locus receptor kinase gene
in the self-incompatibility response. Plant J 5: 373±384
Pastuglia M, Ruo-Chable V, Delorme V, Gaude T, Dumas C,
Cock JM (1997) A functional S locus anther gene is not required for the self-incompatibility response in Brassica oleracea.
Plant Cell 9: 2065±2076
Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a
laboratory manual (2nd edn). Cold Spring Harbor Laboratory
press, Cold Spring Harbor, N.Y.
Schopfer CR, Nasrallah ME, Nasrallah JB (1999) The male determinant of self-incompatibility in Brassica. Science 286: 1697±
1700
Stahl RJ, Arnoldo M, Glavin TL, Goring DR, Rothstein SJ (1998)
The self-incompatibility phenotype in Brassica is altered by the
transformation of a mutant S locus receptor kinase. Plant Cell
10: 209±218
Stone SL, Arnoldo M, Goring DR (1999) A breakdown of Brassica
self-incompatibility in ARC1 antisense transgenic plants. Science 286: 1729±1731
Tobias CM, Howlett B, Narallah JB (1992) An Arabidopsis thaliana gene with sequence similarity to the S locus receptor kinase
gene of Brassica oleracea. Plant Physiol 99: 284±290
Umbach AL, Lalonde BA, Kandasamy MK, Nasrallah JB, Nasrallah ME (1990) Immunodetection of protein glycoforms encoded by two independent genes of the self-incompatibility
multigene family of Brassica. Plant Physiol 93: 739±747
Verwoerd TC, Dekker BNM, Hoekema A (1989) A small-scale
procedure for the rapid isolation of plant RNAs. Nucleic Acids
Res 17: 2362
Yu K, Schafer U, Glavin TL, Goring DR, Rothstein SJ (1996)
Molecular characterization of the S locus in two self-incompatible Brassica napus lines. Plant Cell 8: 2369±2380