SUMMARYS45A, a double recessive mutant at both the BnMs1 and BnMs2 loci in Brassica napus, produces no pollen in mature anthers and no seeds by self-fertilization. The BnMs1 and BnMs2 genes, which have redundant functions in the control of male fertility, are positioned on linkage groups N7 and N16, respectively, and are located at the same locus on Arabidopsis chromosome 1 based on collinearity between Arabidopsis and Brassica. Complementation tests indicated that one candidate gene, BnCYP704B1, a member of the cytochrome P450 family, can rescue male sterility. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) of the developing anther showed that pollen-wall formation in the mutant was severely compromised, with a lack of sporopollenin or exine. The phenotype was first evident at the tetrad stage (stage 7) of anther development, coinciding with the maximum BnCYP704B1 mRNA accumulation observed in tapetal cells at stages 7-8 (haploid stage). TEM also suggested that development of the tapetum was seriously defective due to the disturbed lipid metabolism in the S45A mutant. A TUNEL assay indicated that the pattern of programmed cell death in the tapetum of the S45A mutant was defective. Lipid analysis showed that the total fatty acid content was reduced in the S45A mutant, indicating that BnCYP704B1 is involved in lipid metabolism. These data suggest that BnCYP704B1 participates in a vital tapetum-specific metabolic pathway that is not only involved in exine formation but is also required for basic tapetal cell development and function.

A recessive genic male sterility (RGMS) system, S45 AB, has been developed from spontaneous mutation in Brassica napus canola variety Oro, and is being used for hybrid cultivar development in China. The male sterility of S45 was controlled by two duplicated recessive genes, named as Bnms1 and Bnms2. In this study, a NIL (near-isogenic line) population from the sib-mating of S45 AB was developed and used for the fine mapping of the Bnms1 gene, in which the recessive allele was homozygous at the second locus. AFLP technology combined with BSA (bulked segregant analysis) was used. From a survey of 2,560 primer combinations (+3/+3 selective bases), seven AFLP markers linked closely to the target gene were identified, of which four were successfully converted to sequence characterized amplified region (SCAR) markers. For further analysis, a population of 1,974 individuals was used to map the Bnms1 gene. On the fine map, Bnms1 gene was flanked by two SCAR markers, SC1 and SC7, with genetic distance of 0.1 cM and 0.3 cM, respectively. SC1 was subsequently mapped on linkage group N7 using doubled-haploid mapping populations derived from the crosses Tapidor x Ningyou7 and DH 821 x DHBao 604, available at IMSORB, UK, and our laboratory, respectively. Linkage of an SSR marker, Na12A02, with the Bnms1 gene further confirmed its location on linkage group N7. Na12A02, 2.6 cM away from Bnms1, was a co-dominant marker. These molecular markers developed from this research will facilitate the marker-assisted selection of male sterile lines and the fine map lays a solid foundation for map-based cloning of the Bnms1 gene.

S45AB, a recessive genic male sterile (RGMS) line, originated as a spontaneous mutant in Brassica napus cv. Oro. The genotypes of sterile (S45A) and fertile plants (S45B) are Bnms1ms1ms2ms2 and BnMs1ms1ms2ms2, respectively. In our previous studies, Yi et al. (Theor Appl Genet 113:643-650, 2006) mapped the BnMs1 locus to a region of 0.4 cM, candidates of which have been identified and genetic transformation is in progress. We describe the fine mapping of BnMs2 exploiting amplified fragment length polymorphism (AFLP) and amplified consensus genetic marker (ACGM) methodologies, and the identification of a collinear region probably containing BnMs2 orthologue in Arabidopsis thaliana. A near isogenic line (NIL) population S4516AB which segregated for BnMs2 locus was generated by crossing, allelism testing and repeated full-sib mating. From the survey of 1,024 AFLP primer combinations, 12 tightly linked AFLP markers were obtained and five of them were successfully converted into co-dominant or dominant sequence characterized amplified region (SCAR) markers. A population of 2,650 sterile plants was screened using these markers and a high-resolution map surrounding BnMs2 was constructed. The closest AFLP markers flanking BnMs2 were 0.038 and 0.075 cM away, respectively. Subsequently, an ACGM marker was developed to delimit the BnMs2 locus at an interval of 0.075 cM. We extended marker sequences to perform BlastN searches against the Arabidopsis genome and identified a collinear region containing 68 Arabidopsis genes, in which the orthologue of BnMs2 might be included. We further integrated BnMs2 linked AFLP or SCAR markers to two doubled-haploid (DH) populations derived from the crosses Tapidor x Ningyou7 (Qiu et al., Theor Appl Genet 114:67-80, 2006) and Quantum x No.2127-17 (available in our laboratory), and BnMs2 was mapped on N16. Molecular markers developed from these investigations will facilitate the marker-assisted selection (MAS) of RGMS lines, and the fine map and syntenic region identified will greatly hasten the process of positional cloning of BnMs2 gene.

Clubroot is an economically important disease affecting plants in the family Cruciferae worldwide. In this study, a collection of 50 Cruciferae accessions was screened using Plasmodiophora brassicae pathotype 4 in China. Eight of these demonstrated resistance, including three Chinese cabbages, two cabbages, one radish, one kale, and one Brassica juncea. The three clubroot-resistant Chinese cabbages (1003, 1007 and 1008) were then used to transfer the clubroot resistance genes to B. napus by distant hybridization combined with embryo rescue. Three methods including morphological identification, cytology identification, and molecular marker-assisted selection were used to determine hybrid authenticity, and 0, 2, and 4 false hybrids were identified by these three methods, respectively. In total, 297 true hybrids were identified. Clubroot resistance markers and artificial inoculation were utilized to determine the source of clubroot resistance in the true hybrids. As a result, two simple sequence repeat (SSR) and two intron polymorphic (IP) markers linked to clubroot resistance genes were identified, the clubroot resistance genes of 1007 and 1008 were mapped to A03. At last, 159 clubroot-resistant hybrids were obtained by clubroot resistance markers and artificial inoculation. These intermediate varieties will be used as the ‘bridge material’ of clubroot resistance for further B. napus breeding.

Male sterility in a near-isogenic line S45AB after 25 generations of subcrossing is controlled by two pairs of duplicate genes. The genotype of S45A is Bnms1Bnms1Bnms2Bnms2, and that of S45B is BnMs1Bnms1Bnms2Bnms2, respectively. Histological observations revealed that abnormal anther development appeared in the tapetum and pollen exine during the tetrad stage. This male sterility was characterized by hypertrophy of the tapetal cells at the tetrad stage and a complete lack of microspore exine after the release of microspores from the tetrads. To elucidate the mechanism of this recessive genic male sterility, the flower bud expression profiles of the S45A and S45B lines were analyzed using an Arabidopsis thaliana ATH1 oligonucleotide array. When compared with the S45B line, 69 genes were significantly downregulated, and 46 genes were significantly upregulated in the S45A line. Real-time polymerase chain reaction (PCR) was then used to verify the results of the microarray analysis, and the majority of the downregulated genes in the S45A line were abundantly and specifically expressed in the anther. The results of the real-time PCR suggest that Bnms1 might be involved in the metabolism of lipid/fatty acids, and the homologous mutation of Bnms1 may either block the biosynthesis of sporopollenin or block sporopollenin from being deposited on the microspore surface, thus, preventing pollen exine formation. The role of Bnms1 in the regulatory network of exine formation is also discussed as well.