SLG Gene Research Articles

Integration of Korean S haplotypes controlling self-incompatibility into the unified nomenclature in radish (Raphanus sativus L.)

ABSTRACT Self-incompatibility in radish (Raphanus sativus L.) is controlled by the S haplotype consisting of three tightly linked genes: SRK, SCR/SP11, and SLG. Since many radish S haplotypes have been reported with different nomenclature by several research groups, such discrepancy has resulted in confusion. To integrate S haplotypes identified in previous studies of Korean radish breeding lines into a unified nomenclature system, the nucleotide sequences of SRK and SLG genes isolated by different research groups were compared with each other. Two radish accessions whose draft whole-genome sequences were previously reported were shown to harbour class II RsS-9 and class I RsS-19 haplotypes, respectively. Nucleotide sequences of SLG and S domains of SRK genes were obtained from diverse breeding lines using four degenerate and 13 haplotype-specific primer pairs. A total of 27 out of 31 S haplotypes isolated in our previous study showed that SRK and SLG sequences were almost identical to those reported by other research groups. In addition, eight novel putative SRK or SLG sequences showing high homology with other SRK genes were identified. Finally, 27 S haplotypes were successfully integrated into the unified nomenclature, and the four S haplotypes containing unique sequences were renamed following the unified nomenclature.

Comparison of different classical and molecular methods for identifying self-incompatibility in two olive cultivars

Aim of study: To determine compatibility relationships and select suitable pollinizers for two olive (Olea europaea L.) cultivars (‘Amygdalolia’ and ‘Konservalia’).Area of study: The Kazeroun Olive Research Station, Fars, Iran.Material and methods: Emasculated flowers from ‘Amygdalolia’ and ‘Konservalia’ cultivars were treated with self-pollination, open-pollination, and cross-pollination with pollen from cultivars such as ‘Dacal’, ‘Amygdalolia’,‘Konservalia’, ‘Koroniki’, and ‘Manzanilla’. Controlled pollination, pollen tube growth, and molecular analysis were employed.Main results: Controlled pollination, pollen tube growth, and molecular analysis showed that cross-pollination was beneficial for ‘Amygdalolia’ compared to self-pollination. The results showed that this cultivar is self-incompatible, and its best pollinator is the ‘Dacal’ cultivar. Experiment results indicated that ‘Konservalia’ behaves as a self-compatible cultivar. The highest fruit percentage and higher pollen tube growth rates were found in self-pollination treatments. Molecular attempts to isolate candidates for sporophytic self-incompatibility (SSI) led the researchers to analyze the expression of SRK and SLG genes.Research highlights: The results indicated an antagonist transcriptional expression pattern in the flowers of ‘Amygdalolia’, classified as a self-incompatible cultivar, and ‘Konservalia’, classified as a self-compatible cultivar, for the SRK and SLG genes.

Identification and classification of S haplotypes in radish (Raphanus sativus)

AbstractRadish (Raphanus sativus L.) is a typical cross‐pollinated crop that exhibits obvious heterosis. Self‐incompatibility is an important character for F1 hybrid breeding of radish. Knowledge of the S haplotypes of breeding lines is very important for breeders to avoid cross‐incompatibility of the parental lines. In the present study, the S haplotypes of 63 radish inbred lines, which were independently cultivated by our research group, were identified by PCR amplification, sequencing and BLAST analyses of the SRK and SLG genes. Finally, fifty‐four inbred lines were classified into 15 class I S haplotypes, including three new types, RsS‐38, RsS‐39 and RsS‐40. Additionally, three class II S haplotypes were identified in nine radish inbred lines. Partial SRK or SLG sequences were completed, such as RsS‐11 Lim (SRK‐S), RsS‐26 (SRK‐K), RsS‐5 Lim (SRK‐K and SRK‐S) and RsS‐9 (SRK‐K). The identified S haplotypes were verified with a cross‐pollination test, and RsS‐9 has weaker self‐incompatibility than other S haplotypes. These information will not only contribute to the production of hybrid seeds but also to the development of new self‐compatible inbred lines, which were advantageous of the production of maintain line and male line in CMS breeding system.

Detection of self-incompatible oilseed rape plants (Brassica napus L.) based on molecular markers for identification of the class I S haplotype.

The selection of desirable genotypes with recessive characteristics, such as self-incompatible plants, is often difficult or even impossible and represents a crucial barrier in accelerating the breeding process. Molecular approaches and selection based on molecular markers can allow breeders to overcome this limitation. The use of self-incompatibility is an alternative in hybrid breeding of oilseed rape. Unfortunately, stable self-incompatibility is recessive and phenotype-based selection is very difficult and time-consuming. The development of reliable molecular markers for detecting desirable plants with functional self-incompatible genes is of great importance for breeders and allows selection at early stages of plant growth. Because most of these reliable molecular markers are based on discrimination of class I S-locus genes that are present in self-compatible plants, there is a need to use an internal control in order to detect possible PCR inhibition that gives false results during genotyping. In this study, 269 double haploid F2 oilseed rape plants obtained by microspore embryogenesis were used to verify the applicability of an improved PCR assay based on the detection of the class I SLG gene along with an internal control. Comparative analysis of the PCR genotyping results vs. S phenotype analysis confirmed the applicability of this molecular approach in hybrid breeding programs. This approach allows accurate detection of self-incompatible plants via a different amplification profile.

RFLP mapping of loci controlling self-incompatibility in Brassica campestris and their comparative mapping with B. napus and B. oleracea

RFLP analysis of a cDNA probe SLG6, governing self incompatibility (SI) in Brassica oleracea, using a recombinant inbred population of Brassica campestris followed by genetic linkage analysis led to the detection of two marker loci, SLG6a and SLG6b controlling SI. SLG6a was mapped in linkage group (LG) 9 and was flanked by the RFLP markers ec4f10 (6.4 cM) and wg5b9 (4.2 cM). SLG6b positioned in LG 2 and was flanked by the RFLP markers wg2d11 (9.9 cM) and ec4e7 (26.9 cM). These results indicated the scope of marker-aided introgression of these genes into self-compatible genotypes for production of SI lines suitable for hybridization in B. campestris. Comparative mapping of LG 9 containing SLG6b with corresponding linkage groups of B. oleracea (BO 2) and B. napus (BN 16) led to the detection of small homologous regions with SLG6 locus linked with another RFLP locus. This evidenced for homology of the SLG genes across Brassica species and possibility of using any single cloned SLG gene for development of SI lines in any Brassica species.

Molecular Basis of S-layer Glycoprotein Glycan Biosynthesis in Geobacillus stearothermophilus

The Gram-positive bacterium Geobacillus stearothermophilus NRS 2004/3a possesses a cell wall containing an oblique surface layer (S-layer) composed of glycoprotein subunits. O-Glycans with the structure [-->2)-alpha-L-Rhap-(1-->3)-beta-L-Rhap-(1-->2)-alpha-L-Rhap-(1-->](n) (= 13-18), a2-O-methyl group capping the terminal repeating unit at the nonreducing end and a -->2)-alpha-L-Rhap-[(1-->3)-alpha-L-Rhap](n) (= 1-2)(1-->3)- adaptor are linked via a beta-D-Galp residue to distinct sites of the S-layer protein SgsE. S-layer glycan biosynthesis is encoded by a polycistronic slg (surface layer glycosylation) gene cluster. Four assigned glycosyltransferases named WsaC-WsaF, were investigated by a combined biochemical and NMR approach, starting from synthetic octyl-linked saccharide precursors. We demonstrate that three of the enzymes are rhamnosyltransferases that are responsible for the transfer of L-rhamnose from a dTDP-beta-L-Rha precursor to the nascent S-layer glycan, catalyzing the formation of the alpha1,3- (WsaC and WsaD) and beta1,2-linkages (WsaF) present in the adaptor saccharide and in the repeating units of the mature S-layer glycan, respectively. These enzymes work in concert with a multifunctional methylrhamnosyltransferase (WsaE). The N-terminal portion of WsaE is responsible for the S-adenosylmethionine-dependent methylation reaction of the terminal alpha1,3-linked L-rhamnose residue, and the central and C-terminal portions are involved in the transfer of L-rhamnose from dTDP-beta-L-rhamnose to the adaptor saccharide to form the alpha1,2- and alpha1,3-linkages during S-layer glycan chain elongation, with the methylation and the glycosylation reactions occurring independently. Characterization of these enzymes thus reveals the complete molecular basis for S-layer glycan biosynthesis.

The dTDP-4-dehydro-6-deoxyglucose reductase encoding fcd gene is part of the surface layer glycoprotein glycosylation gene cluster of Geobacillus tepidamans GS5-97T

The glycan chain of the S-layer protein of Geobacillus tepidamans GS5-97(T) consists of disaccharide repeating units composed of L-rhamnose and D-fucose, the latter being a rare constituent of prokaryotic glycoconjugates. Although biosynthesis of nucleotide-activated L-rhamnose is well established, D-fucose biosynthesis is less investigated. The conversion of alpha-D-glucose-1-phosphate into thymidine diphosphate (dTDP)-4-dehydro-6-deoxyglucose by the sequential action of RmlA (glucose-1-phosphate thymidylyltransferase) and RmlB (dTDP-glucose-4,6-dehydratase) is shared between the dTDP-D-fucose and the dTDP-L-rhamnose biosynthesis pathway. This key intermediate is processed by the dTDP-4-dehydro-6-deoxyglucose reductase Fcd to form dTDP-alpha-D-fucose. We identified the fcd gene in G. tepidamans GS5-97(T) by chromosome walking and performed functional characterization of the recombinant 308-amino acid enzyme. The in vitro activity of the enzymatic cascade (RmlB and Fcd) was monitored by high-performance liquid chromatography and the reaction product was confirmed by (1)H and (13)C nuclear magnetic resonance spectroscopy. This is the first characterization of the dTDP-alpha-D-fucopyranose biosynthesis pathway in a Gram-positive organism. fcd was identified as 1 of 20 open reading frames contained in a 17471-bp S-layer glycosylation (slg) gene cluster on the chromosome of G. tepidamans GS5-97(T). The sgtA structural gene is located immediately upstream of the slg gene cluster with an intergenic region of 247 nucleotides. By comparison of the SgtA amino acid sequence with the known glycosylation pattern of the S-layer protein SgsE of Geobacillus stearothermophilus NRS 2004/3a, two out of the proposed three glycosylation sites on SgtA could be identified by electrospray ionization quadrupole-time-of-flight mass spectrometry to be at positions Ser-792 and Thr-583.

Evolution of the self-incompatibility system in the Brassicaceae: identification of S-locus receptor kinase (SRK) in self-incompatible Capsella grandiflora

Self-incompatibility (SI) has been well studied in the genera Brassica and Arabidopsis, which have become models for investigation into the SI system. To understand the evolution of the SI system in the Brassicaceae, comparative analyses of the S-locus in genera other than Brassica and Arabidopsis are necessary. We report the identification of six putative S-locus receptor kinase genes (SRK) in natural populations of Capsella grandiflora, an SI species from a genus which is closely related to Arabidopsis. These S-alleles display striking similarities to the Arabidopsis lyrata SRK alleles in sequence and structure. Our phylogenetic analysis supports the scenario of differing SI evolution along the two lineages (The Brassica lineage and Arabidopsis/Capsella lineage). Our results also argue that the ancestral S-locus lacked the SLG gene (S-locus glycoprotein) and that the diversification of S-alleles predates the separation of Arabidopsis and Capsella.

S-layer glycan-specific loci on the chromosome of Geobacillus stearothermophilus NRS 2004/3a and dTDP-l-rhamnose biosynthesis potential of G. stearothermophilus strains

The approximately 16.5 kb surface layer (S-layer) glycan biosynthesis (slg) gene cluster of the Gram-positive thermophile Geobacillus stearothermophilus NRS 2004/3a has been sequenced. The cluster is located immediately downstream of the S-layer structural gene sgsE and consists of 13 ORFs that have been identified by database sequence comparisons. The cluster encodes dTDP-L-rhamnose biosynthesis (rml operon), required for building up the polyrhamnan S-layer glycan, as well as for assembly and export of the elongated glycan chain, and its transfer to the S-layer protein. This is the first report of a gene cluster likely to be involved in the glycosylation of an S-layer protein. There is evidence that this cluster is transcribed as a polycistronic unit, whereas sgsE is transcribed monocistronically. To get insights into the regulatory mechanisms underlying glycosylation of the S-layer protein, the influence of growth temperature on the S-layer was investigated in seven closely related G. stearothermophilus strains, of which only strain NRS 2004/3a possessed a glycosylated S-layer. Chromosomal DNA preparations of these strains were screened for the presence of the rml operon, because L-rhamnose is a frequent constituent of S-layer glycans. From rml-positive strains, flanking regions of the operon were sequenced. Comparison with the slg gene cluster of G. stearothermophilus NRS 2004/3a revealed sequence homologies between adjacent genes. The temperature inducibility of S-layer protein glycosylation was investigated in those strains by raising the growth temperature from 55 degrees C to 67 degrees C; no change of either the protein banding pattern or the glycan staining behaviour was observed on SDS-PAGE gels, although the sgsE transcript was several-fold more abundant at 67 degrees C. Cell-free extracts of the strains were capable of converting dTDP-D-glucose to dtdp-L-rhamnose. Taken together, the results indicate that the rml locus is highly conserved among G. stearothermophilus strains, and that in the investigated rml-containing strains, dTDP-L-rhamnose is actively synthesized in vitro. However, in contrast to previous reports for G. stearothermophilus wild-type strains, an increase in growth temperature did not switch an S-layer protein phenotype to an S-layer glycoprotein phenotype, via the de novo generation of a new S-layer gene sequence.

Genomic organization of the S core region and the S flanking regions of a class-II S haplotype in Brassica rapa

The nucleotide sequence of an 86.4-kb region that includes the SP11, SRK, and SLG genes of Brassica rapa S-60 (a class-II S haplotype) was determined. In the sequenced region, 13 putative genes were found besides SP11-60, SRK-60, and SLG-60. Five of these sequences were isolated as cDNAs, five were homologues of known genes, cDNAs, or ORFs, and three are hypothetical ORFs. Based on their nucleotide sequences, however, some of them are thought to be non-functional. Two regions of colinearity between the class-II S-60 and Brassica class-I S haplotypes were identified, i.e., S flanking region 1 which shows partial colinearity of non-genic sequences and S flanking region 2 which shows a high level of colinearity. The observed colinearity made it possible to compare the order of SP-11, SRK, and SLG genes in the S locus between the five sequenced S haplotypes. It emerged that the order of SRK and SLG in class-II S-60 is the reverse of that in the four class-I S haplotypes reported so far, and the order of SP11, SRK and SLG is the opposite of that in the class-I haplotype S-910. The possible gene designated as SAN1 (S locus Anther-expressed Non-coding RNA like-1), which is located in the region between SP11-60 and SRK-60, has features reminiscent of genes for non-coding RNAs (ncRNAs), but no homologous sequences were found in the databases. This sequence is transcribed in anthers but not in stigmas or leaves. These features of the genomic structure of S-60 are discussed with special reference to the characteristics of class-II S haplotypes.

Genetic organization of chromosomal S-layer glycan biosynthesis loci of Bacillaceae.

S-layer glycoproteins are cell surface glycoconjugates that have been identified in archaea and in bacteria. Usually, S-layer glycoproteins assemble into regular, crystalline arrays covering the entire bacterium. Our research focuses on thermophilic Bacillaceae, which are considered a suitable model system for studying bacterial glycosylation. During the past decade, investigations of S-layer glycoproteins dealt with the elucidation of the highly variable glycan structures by a combination of chemical degradation methods and nuclear magnetic resonance spectroscopy. It was only recently that the molecular characterization of the genes governing the formation of the S-layer glycoprotein glycan chains has been initiated. The S-layer glycosylation (slg) gene clusters of four of the 11 known S-layer glycan structures from members of the Bacillaceae have now been studied. The clusters are approximately 16 to approximately 25 kb in size and transcribed as polycistronic units. They include nucleotide sugar pathway genes that are arranged as operons, sugar transferase genes, glycan processing genes, and transporter genes. So far, the biochemical functions only of the genes required for nucleotide sugar biosynthesis have been demonstrated experimentally. The presence of insertion sequences and the decrease of the G + C content at the slg locus suggest that the investigated organisms have acquired their specific S-layer glycosylation potential by lateral gene transfer. In addition, S-layer protein glycosylation requires the participation of housekeeping genes that map outside the cluster. The gene encoding the respective S-layer target protein is transcribed monocistronically and independently of the slg cluster genes. Its chromosomal location is not necessarily in close vicinity to the slg gene cluster.

Coevolution of theS-Locus GenesSRK,SLGandSP11/SCRinBrassica oleraceaandB. rapa

Brassica self-incompatibility (SI) is controlled by SLG and SRK expressed in the stigma and by SP11/SCR expressed in the anther. We determined the sequences of the S domains of 36 SRK alleles, 13 SLG alleles, and 14 SP11 alleles from Brassica oleracea and B. rapa. We found three S haplotypes lacking SLG genes in B. rapa, confirming that SLG is not essential for the SI recognition system. Together with reported sequences, the nucleotide diversities per synonymous and nonsynonymous site (pi(S) and pi(N)) at the SRK, SLG, and SP11 loci within B. oleracea were computed. The ratios of pi(N):pi(S) for SP11 and the hypervariable region of SRK were significantly >1, suggesting operation of diversifying selection to maintain the diversity of these regions. In the phylogenetic trees of 12 SP11 sequences and their linked SRK alleles, the tree topology was not significantly different between SP11 and SRK, suggesting a tight linkage of male and female SI determinants during the evolutionary course of these haplotypes. Genetic exchanges between SLG and SRK seem to be frequent; three such recent exchanges were detected. The evolution of S haplotypes and the effect of gene conversion on self-incompatibility are discussed.

Expression of the S receptor kinase in self-compatible Brassica napus cv. Westar leads to the allele-specific rejection of self-incompatible Brassica napus pollen.

Expression of an S receptor kinase (SRK910) transgene in the self-compatible Brassica napus cv. Westar conferred on the transgenic pistil the ability to reject pollen from the self-incompatible Brassica napus W1 line, which carries the S910 allele. In one of the SRK transgenic lines, 1C, virtually no seeds were produced when the transgenic pistils were pollinated with W1 pollen (Mean number of seeds per pod = 1.22). This response was specific to the W1 pollen since pollen from a different self-incompatible Brassica napus line (T2) and self-pollinations were fully compatible. Westar plants expressing an S locus glycoprotein transgene (SLG910) did not show any self-incompatibility response towards W1 pollen. Transgenic Westar plants resulting from crosses between the 1C SRK transgenic line and three SLG910 transgenic lines were also tested for rejection of W1 pollen. The additional expression of the SLG910 transgene in the SRK910 transgenic plants did not cause any significant further reduction in seed production (Mean seeds/pod = 1.04) or have any detectable effects on the number of pollen grains that adhered to the pistil. Thus, while the allele-specific SLG gene was previously reported to have an enhancing effect on the self-incompatibility response, no evidence for such a role was found in this study.

Isolation and Characterization of a Class I SLG Gene from Chinese Cabbage ( Brassica campestris)

Isolation and Characterization of a Class I SLG Gene from Chinese Cabbage ( Brassica campestris)

Isolation and characterization of a class I SLG gene from Chinese cabbage (Brassica campestris).

In Brassica species, self-incompatibility in the recognition reaction between self and non-self pollens is determined by two genes, SLG and SRK, at the S locus. We have cloned and characterized a genomic DNA fragment containing a complete open reading frame of the SLG gene from Chinese cabbage. The genomic clone, named BcSLG2, was found to possess the region that shares a homology of 77% in amino acid identity with the SLG46 gene of Brassica campestris. Northern blot analysis revealed that the BcSLG2 gene expression is restricted to the pistil of Chinese cabbage flower. In situ hybridization showed that in the pistil, the gene is expressed predominantly in the stigmatic tissue. Much lower expression in the tapetum was also detectable at an immature stage of the flower development. Southern blot hybridization with the BcSLG2 DNA probe showed polymorphism in the SLG gene organization of the Chinese cabbage plants. These results will provide valuable information in understanding the S gene complex of the Chinese cabbage plants.

Characterization of Brassica S-haplotypes lacking S-locus glycoprotein

Self-incompatibility (SI) in Brassica is regulated by a single multi-allelic locus, S, which contains highly polymorphic stigma-expressed genes, SLG and SRK. While SRK is shown to be the determinant of female SI specificity, SLG is thought to assist the function of SRK. Here we report that the SLG genes of self-incompatible S 18 and S 60 homozygotes of Brassica oleracea have an in-frame stop codon and a 23 bp deletion resulting in a frame-shift, respectively. The finding that these SLG genes do not encode functional SLG proteins suggests that SLG is not essential for SI. The possible role of SLG in SI was discussed.

Transformation of Arabidopsis with a Brassica SLG/SRK region and ARC1 gene is not sufficient to transfer the self-incompatibility phenotype.

Self-incompatibility (SI) promotes outbreeding in flowering 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 identified 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 flower 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 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-F22-ARC1 Arabidopsis stigmas. Taken together, our results suggest that complementing Arabidopsis genome with Brassica SLG, SRK and ARC1 genes is unlikely to be sufficient to transfer the SI phenotype.

Sequence Diversity of SLG and SRK in Brassica oleracea L.

Abstract DNA sequences of SLG and SRK , believed to function in self-pollen recognition of self-incompatibility, were analysed and compared between many S haplotypes of Brassica oleracea . Some different S haplotypes have highly similar SLGs (more than 97.5% amino acid sequence identity), while their SRKs are less similar (less than 88% identity). Different lines having the same recognition specificity often show a high degree of sequence variation in SLG (88.5% identity in the S 2 haplotype), but their SRKs are very similar to each other (97.3% identity). S haplotypes, in which SLG protein is undetectable by Western blot analysis, have stop codons in the middle of the coding region of SLG , or lack an SLG gene in the genome, suggesting that these haplotypes do not synthesize the functional SLG protein. The SRK alleles in these haplotypes are normal. The function of SLG and SRK in self-incompatibility and the evolution of these alleles are discussed.

Alteration of the self-incompatibility phenotype in Brassica by transformation of the antisense SLG gene.

Self-incompatible (SI) Brassica rapa (syn. B. campestris) was transformed with an antisense SLG gene by using SLG8 cDNA isolated from the B. campestris S8 homozygote. Two transformed lines were obtained and analyzed. Northern blot and Western blot analyses revealed that endogenous SLG and SRK were greatly reduced of the transcriptional and translational levels in the transformant. Pollination experiments confirmed that their SI phenotype had broken down. In addition, the progeny with the antisense SLG gene, resulting from self- or cross-pollination of the transgenic plant, also showed the self-compatible phenotype. The breakdown of SI in the tranformants was due to the change in property of the stigma and not of the pollen. These results provide strong evidence that SLG and/or SRK is implicated in the pollen-stigma recognition of SI and that they act only as stigmatic factors.

Recombination and selection at Brassica self-incompatibility loci.

In Brassica species, self-incompatibility is controlled genetically by haplotypes involving two known genes, SLG and SRK, and possibly an as yet unknown gene controlling pollen incompatibility types. Alleles at the incompatibility loci are maintained by frequency-dependent selection, and diversity at SLG and SRK appears to be very ancient, with high diversity at silent and replacement sites, particularly in certain "hypervariable" portions of the genes. It is important to test whether recombination occurs in these genes before inferences about function of different parts of the genes can be made from patterns of diversity within their sequences. In addition, it has been suggested that, to maintain the relationship between alleles within a given S-haplotype, recombination is suppressed in the S-locus region. The high diversity makes many population genetic measures of recombination inapplicable. We have analyzed linkage disequilibrium within the SLG gene of two Brassica species, using published coding sequences. The results suggest that intragenic recombination has occurred in the evolutionary history of these alleles. This is supported by patterns of synonymous nucleotide diversity within both the SLG and SRK genes, and between domains of the SRK gene. Finally, clusters of linkage disequilibrium within the SLG gene suggest that hypervariable regions are under balancing selection, and are not merely regions of relaxed selective constraint.