Exploitation of the late flowering species <Emphasis Type="Italic">Brassica oleracea</Emphasis> L. for the improvement of earliness in <Emphasis Type="Italic">B. napus</Emphasis> L.: an untraditional approach

The oilseed Brassica rapa flowers and matures earlier than B. oleracea, as well as their amphidiploid B. napus. Therefore, earliness of B. rapa has been investigated as a source of variation for earliness in B. napus breeding programs. Variation for days to flower exists in B. oleracea; however, its earliest flowering variant B. alboglabra flowers 2–3 weeks later than B. napus. We hypothesized that the C genome of B. alboglabra carries alleles for early flowering which are different from the C-genome alleles of B. napus; and these alleles can be used for the improvement of B. napus. To test this, we examined flowering time in pedigree and DH populations from two B. napus × B. alboglabra crosses. A B. napus line with about a week earlier flowering than the B. napus parent was achieved through reconstitution of its C genome following pedigree selection. Introgression of the B. alboglabra allele in the early flowering pedigree lines is also evident from the presence of B. alboglabra-specific SSR alleles in this line. However, application of doubled haploidy failed to generate any line that flowered earlier than the B. napus parent, which is probably due to the difficulty of obtaining large numbers of euploid B. napus DH lines from this interspecific cross. Thus, we demonstrate that a trait of the diploid species, which apparently looks undesirable, might in fact be highly valuable for the improvement of amphidiploids; and knowledge from this research can also be applied for other traits.     

Genetic and histological characterization of a novel recessive genic male sterile line of <Emphasis Type="Italic">Brassica napus</Emphasis> derived from a cross with <Emphasis Type="Italic">Capsella bursa</Emphasis>-<Emphasis Type="Italic">pastoris</Emphasis>

In a previously made cross Brassica napus cv. Oro (2n = 38) × Capsella bursa-pastoris (2n = 4x = 32), one F₁ hybrid with 2n = 38 was totally male sterile. The hybrid contained no complete chromosomes from C. bursa-pastoris, but some specific AFLP (amplified fragment length polymorphism) bands of C. bursa-pastoris were detected. The hybrid was morphologically quite similar to ‘Oro’ except for smaller flowers with rudimentary stamens but normal pistils, and showed good seed-set after pollination by ‘Oro’ and other B. napus cultivars. The fertility segregation ratios (3:1, 1:1) in its progenies indicated that the male sterility was controlled by a single recessive gene. In the pollen mother cells of the male sterile hybrid, chromosome pairing and segregation were normal. Histological sectioning of its anthers showed that the tapetum was multiple layers and was hypertrophic from the stage of sporogenic cells, and that the tetrads were compressed by the vacuolated and disaggregated tapetum and no mature pollen grains were formed in anther sacs, thus resulting in male sterility. The possible mechanisms for the production of the male sterile hybrid and its potential in breeding are discussed.     

Overcoming obstacles to interspecific hybridization between <Emphasis Type="Italic">Gossypium hirsutum</Emphasis> and <Emphasis Type="Italic">G. turneri</Emphasis>

Gossypium turneri, a wild cotton species (2n = 2X = 26, D₁₀D₁₀) originating from Mexico, possesses invaluable characteristics unavailable in the cultivated tetraploid cotton gene pool, such as caducous involucels at anthesis, resistance to insects and tolerance to abiotic stresses. However, transferring desired characteristics from wild species into cultivated cotton is often fraught with diverse obstacles. Here, Gossypium hirsutum (as the maternal parent) and G. turneri were crossed in the Hainan Province of China, and the obtained hybrid seeds (2n = 3X = 39, ADD₁₀) were treated with 0.075% colchicine solution for 48 h to double the chromosome complement in order to overcome triploid F₁ sterility and to generate a fertile hexaploid. Chromosome doubling was successful in four individuals. However, the new synthetic hexaploids derived from these individuals were still highly sterile, and no seeds were generated by selfing or crossing. Therefore, an embryo rescue technique was employed in an attempt to produce progenies from the new synthetic hexaploids. Consequently, a total of six large embryos were obtained on MSB2K medium supplemented with 0.5 mg l^?1 KIN and 250 mg l^?1 CH using ovules from backcrossing that were 3 days post-anthesis. Four grafted surviving seedlings were confirmed to be the progenies (pentaploids) of the new synthetic hexaploids using cytological observations and molecular markers. Eight putative fertile individuals derived from backcrossing the above pentaploids were confirmed using SSR markers and generated an abundance of normal seeds. This research lays a foundation for transferring desirable characteristics from G. turneri into upland cotton.     

Resynthesizing <Emphasis Type="Italic">Brassica napus</Emphasis> from interspecific hybridization between <Emphasis Type="Italic">Brassica rapa</Emphasis> and <Emphasis Type="Italic">B. oleracea</Emphasis> through ovary culture

Using three varieties of Brassica rapa, cv. Hauarad (accession 708), cv. Maoshan-3 (714) and cv. Youbai (715), as the maternal plants and one variety of B. oleracea cv. Jingfeng-1 (6012) as the paternal plant, crosses were made to produce interspecific hybrids through ovary culture techniques. A better response of seed formation was observed when ovaries were cultured in vitro at 9–12 days after pollination on the basal MS and B5 media supplemented with 6-benzylaminopurine (BA) and naphthylacetic acid (NAA). The best response was observed for cross 714×6012 with the rate of seeds per ovary reaching 43.0%. Seeds for cross 715×6012 showed the best germination response (66.7%) on the regeneration medium (MS+1.0 mg l^–1 BA+0.05 mg l^–1 NAA). In all three cross combinations, good response in terms of root number and length of plants was observed on the root induction medium (MS+1.0 mg l^–1 BA+0.1 mg l^–1 NAA). A better response was observed for the regenerated plants cultured for 14 days than for 7 days. The ovary-derived plants with well-developed root system were hardened for 8 days and their survival rate reached over 80%. Cytological studies showed that the chromosome number of all plants tested was 19 (the sum of both parents), indicating that these regenerated plants were all true hybrids of B. rapa (n = 10) × B. oleracea (n = 9). The regenerated plants were doubled with colchicine treatment, and the best response in the crosses 708×6012, 714×6012 and 715×6012 was observed when treated with 170 mg l^–1 colchicine for up to 30 h and their doubling frequency reached 52, 56 and 62%, respectively.     

In vitro polyploidisation of <Emphasis Type="Italic">Helleborus</Emphasis> species

Helleborus species are members of the family of the Ranunculaceae. These popular perennials are all diploids (2n = 2x = 32). This study investigates polyploidy induction by different antimitotic agents. Colchicine, oryzalin and trifluralin were tested in vitro on shoots of Helleborus niger, H. orientalis and H. × nigercors. Furthermore the effect of the antimitotic agents on the viability and the multiplication rate of cultured plantlets were analyzed. Flow cytometry demonstrated that polyploidisation was genotype dependent: using H. niger, tetraploids were obtained using either oryzalin (3 μM) or trifluralin (3 or 10 μM), whereas for H. × nigercors only trifluralin (3 or 10 μM) induced polyploidisation. For H. orientalis neither treatment was effective to produce tetraploids or mixoploids. For these three species, colchicine (100 μM) was ineffective. The polyploidisation events in H. niger and H. × nigercors were confirmed by chromosome counts of mounted nuclei derived from root tips (2n = 4x = 64).     

Production of interspecific hybrids between <Emphasis Type="Italic">Lilium longiflorum</Emphasis> and <Emphasis Type="Italic">L. lophophorum</Emphasis> var. <Emphasis Type="Italic">linearifolium</Emphasis> via ovule culture at early stage

Interspecific hybridization was carried out between Lilium longiflorum and L. lophophorum var. linearifolium by using the cut style method of pollination, as a contrast, intraspecific hybridization between L. longiflorum ‘Gelria’ and L. longiflorum was also made, but no mature seeds and offspring were obtained from the two combinations under in vivo condition. Ovules excised from each carpel 5–35 days after pollination (DAP) were cultured on B5 or half-strength B5 medium containing sucrose at different concentrations in vitro. In L. longiflorum × L. lophophorum var. linearifolium, only 1.17% of ovules excised at 10 DAP developed into seedlings, and in L. longiflorum ‘Gelria’ × L. longiflorum, only 0.99% of ovules excised at 25 DAP developed into seedlings; none of the ovules excised at other different DAP in the two cross combinations produced any seedlings. The results showed that interspecific hybridization had a more serious post-fertilization barrier than the intraspecific hybridization, and that a lower concentration (3%) of sucrose led to better embryo development and higher percentage of seedlings in ovule cultures. All hybrid seedlings obtained were successfully transplanted to soil and grew normally. The progenies investigated were identified as true hybrids based on inter-simple sequence repeat (ISSR) analysis.     

Morphological,cytological and reproductive characterization of tri-species hybrids (GOS) between <Emphasis Type="Italic">Pennisetum glaucum</Emphasis>, <Emphasis Type="Italic">P. orientale</Emphasis> and <Emphasis Type="Italic">P. squamulatum</Emphasis>

New tri-species hybrids (GOS) in the genus Pennisetum involving the cultivated species pearl millet (P. glaucum L.) and two wild species, viz. P. squamulatum Fresen and P. orientale L. C. Rich, are reported. Six hybrid plants were recovered after crossing a backcross hybrid (2n = 3x = 23, GGO) between P. glaucum (2n = 2x = 14, GG) and P. orientale (2n = 2x = 18, OO) with F₁s (2n = 6x = 42, GGSSSS) between P. glaucum (2n = 4x = 28, GGGG) and P. squamulatum (2n = 8x = 56, SSSSSSSS). The hybrids were perennial, morphologically intermediate to their parents, and represented characters from the three contributing species. The hybrids contained 2n = 44 chromosomes (GGGSSO) representing 21, 14 and nine chromosomes from P. glaucum, P. squamulatum and P. orientale, respectively. Meiotic and flow-cytometric analysis suggested origin of these hybrids from unreduced female and reduced male gametes. Average chromosome configuration (8.42_I + 14.32_II + 1.62_III + 0.52_IV) at Meiosis showed limited inter-genomic pairing indicating absence of significant homology between the three genomes. The hybrids were male sterile (except one) and highly aposporous. P. orientale was identified to induce apospory in hybrid background with P. glaucum at diploid and above levels, though it was quantitatively affected by genomic doses from sexual parent. A case of inducible and recurrent apospory is presented whereby a transition from Polygonum-type sexual embryo-sacs to Panicum-type aposporous embryo-sacs was observed in diploid interspecific hybrids. Results supported independent origin and partitioning of the three apomixis-components (apomeiosis, parthenogenesis, and functional endosperm development), reported for the first time in Pennisetum. Potential utilization of GOS hybrids in understanding genome interactions involved in complex traits, such as perenniality and apomixis, is discussed.     

Breeding for salinity resistance in mulberry (<Emphasis Type="Italic">Morus</Emphasis> spp.)

Late maturity α-amylase (LMA) is a genetic defect that is fairly widely spread in bread wheat (Triticum aestivum L.) germplasm, and recently detected in durum cultivars, which can result in unacceptably high α-amylase activity (low falling number) in ripe grain. LMA has also been observed at unexpectedly high frequency and severity in synthetic hexaploid wheats derived from the interspecific hybridisation of Triticum durum (AABB) and Aegilops tauschii (DD). Since synthetic hexaploids represent an important new source of resistances/tolerances to a range of biotic and abiotic stresses for wheat breeders, there is a pressing need to understand the mechanisms involved in LMA in synthetics and develop strategies for avoiding its adverse effects on grain quality. The objectives of this study were to firstly, compare the LMA phenotype of synthetics that varied for plant height, secondly, to characterise the LMA phenotype in groups of synthetics derived from the same durum parents and finally to determine whether LMA in primary synthetics is associated with the QTL previously reported in conventional bread wheat. More than 250 synthetic hexaploids, a range of durum cultivars and a doubled haploid population derived from Worrakatta (non-LMA) × AUS29663 (high LMA synthetic) were phenotyped and genotyped with markers reported to be linked to LMA in conventional bread wheat and markers diagnostic for the semi-dwarfing gene, Rht1. More than 85% of synthetics were prone to LMA, approximately 60% ranked as very high. Genetic control of LMA in synthetic hexaploids appeared to involve QTL located on 7B, and to a lesser extent 3B, similar to bread wheats. However, the LMA phenotype of many synthetic hexaploids appeared to be more extreme than could be explained by comparisons with bread wheat even taking into account the apparent absence of Rht1 in most genotypes. Other mechanisms, possibly triggered by the interaction between the AABB and DD genomes cannot be excluded. The presence of wild type rht1 in most synthetic hexaploids and their extreme height is difficult to reconcile with the semi-dwarf, Rht1, stature of many of the durums used in the interspecific hybridisation process. Mechanisms that could explain this observation remain unclear.     

Biochemical markers for cabbage seedpod weevil (<Emphasis Type="Italic">Ceutorhynchus obstrictus</Emphasis> (Marsham)) resistance in canola (<Emphasis Type="Italic">Brassica napus</Emphasis> L.)

In the last decade, the cabbage seedpod weevil (Ceutorhynchus obstrictus (Marsham)) has become a major insect pest of canola (Brassica napus L.) in Canada reducing seed yields up to 35%. Therefore, the benefits of developing weevil resistant germplasm to canola breeders and the industry would reduce input costs, pesticide use, environmental degradation and increase yield. Yellow mustard (Sinapis alba L.) is resistant to C. obstrictus (CSPW), although the exact mechanism is not known (McCaffrey et al. 1999). A unique canola population was generated at the University of Guelph from a cross between B. napus and S. alba through embryo rescue and backcrossed to canola several times prior to double haploid (DH) production. Approximately one-half of this DH population had canola quality glucosinolate concentration (<16 μmol/g) and was used for further breeding. The hypothesis was that some DH progeny from this cross inherited resistance to CSPW from S. alba. Weevil infestation levels were assessed for the B. napus × S. alba BC2 and BC3 DH populations in the field over 7 years in Alberta where weevil pressure is strong to establish the resistant or susceptible status of these lines. The basic objectives for this study were to confirm field resistance in the B. napus × S. alba germplasm in Ontario and to identify any biochemical markers associated with resistance/susceptibility. Canola doubled haploid lines derived from BC2 or BC3 families were field screened for resistance (R) followed by chemical analysis of glucosinolates to detect biochemical polymorphisms correlated with CSPW resistance using High Performance Liquid Chromatography (HPLC). Two polymorphic peaks were found, one each, from extracts of upper cauline leaves and Stage 3 pod seed, with retention times of ~23 and 19 min, respectively. These HPLC peaks consistently correlated with larval infestation data and the peak differences between R and S DH lines were significant. Therefore, these two peaks can be considered as biochemical markers in this breeding germplasm and may play a role in rapid and early detection of CSPW resistance.     

Genetic analysis reveals a dominant <Emphasis Type="Italic">S</Emphasis> locus and an <Emphasis Type="Italic">S</Emphasis> suppressor locus in natural self-compatible <Emphasis Type="Italic">Brassica napus</Emphasis>

A self-incompatible (SI) line, S-1300, and its maintainer 97-wen135, a self-compatible (SC) line, were used to study the inheritance of maintenance for self-incompatibility in B. napus. The ratio of SI plants to SC plants from S-1300 × 97-wen135 F₂ and (S-1300 × 97-wen135) × 97-wen135 was 346:260 and 249:232, fitting the expected ratio of 9:7 and 1:1, respectively. Based on these observations, here we propose a genetic model in which two independent loci, S locus and S suppressor locus (sp), are predicted to control the inheritance of maintenance for self-incompatibility in B. napus. The genotypes of S-1300 and 97-wen135 are S ₁₃₀₀ S ₁₃₀₀ sp ₁₃₀₀ sp ₁₃₀₀ and S ₁₃₅ S ₁₃₅ sp ₁₃₅ sp ₁₃₅ , respectively. S ₁₃₅ is dominant to S ₁₃₀₀ , but coexistence of sp ₁₃₀₀ and sp ₁₃₅ fails to suppress S locus. Both S ₁₃₀₀ and S ₁₃₅ can be suppressed by sp ₁₃₅ , while sp ₁₃₀₀ can suppress S ₁₃₅ but not S ₁₃₀₀ . The model contains two characteristics: that a dominant S locus exists in self-compatible B. napus, and that co-suppression will occur when sp loci are heterozygous. The model has been validated by the segregation of S phenotypes in the (S-1300 × 97-wen135) × S-1300, the progenies of SC S-1300 × 97-wen135 F₂ plants and DH population developed from S-1300 × 97-wen135 F₁. This is the first study to report co-suppression of S suppressor loci in B. napus. The genetic model will be very useful for developing molecular markers linked to maintenance for self-incompatibility and for dissecting the mechanism of SI/SC in B. napus.     

Interspecific hybridization of fig (<Emphasis Type="Italic">Ficus carica</Emphasis> L.) and <Emphasis Type="Italic">Ficus erecta</Emphasis> Thunb., a source of Ceratocystis canker resistance

Ceratocystis canker, which is caused by the fungus Ceratocystis fimbriata Ellis et Halsted, is one of the most severe diseases of the common fig (Ficus carica L.). In contrast, the wild fig species F. erecta Thunb. is resistant to this fungus. We performed interspecific hybridization between the common fig (seed parent) and F. erecta (pollen parent) through artificial pollination. Even though hybrid seeds showed high germination rates, the seedling survival rates were low. All of the seedlings contained the expected simple sequence repeat (SSR) alleles from both common fig and F. erecta at each of the three loci tested, thus confirming the parent–offspring relationships of the interspecific hybrids. The leaf morphological characters of hybrid seedlings were intermediate between those of the parents. Cuttings of cultivars of common fig, F. erecta, and hybrid seedlings were inoculated with C. fimbriata by direct wounding of the shoot. All of the common fig cultivars tested withered and died within 10 weeks. Leaves and shoots of the hybrids and F. erecta were healthy 100 days after inoculation. Our results suggest that interspecific hybridization between the common fig and the wild species F. erecta is a breakthrough in the breeding of a new fig rootstock source with resistance to Ceratocystis canker.     

Microsporogenesis of <Emphasis Type="Italic">Rps</Emphasis>8/<Emphasis Type="Italic">rps</Emphasis>8 heterozygous soybean lines

Phytophthora root and stem rot caused by Phytophthora sojae, is one of the most damaging diseases of soybean, for which management is principally done by planting resistant cultivars with race specific resistance which are conferred by Rps (Resistance to Phytophthora sojae) genes. The Rps8 locus, identified in the South Korean landrace PI 399073, is located in a 2.23 Mbp region on soybean chromosome 13. In eight cv. Williams (rps8/rps8) × PI 399073 (Rps8/Rps8) populations, this region exhibited strong segregation distortion. In a cross between the South Korean lines PI 399073 (Rps8/Rps8) and PI 408211B (multiple Rps genes) this region segregated in a Mendelian fashion. In this study, microsporogenesis was evaluated to identify meiotic abnormalities that may be associated with the segregation distortion of the Rps8 region. Pollen was collected from greenhouse-grown plants of the parental genotypes: Williams, PI 399073, and PI 408211B; as well as selected Rps8/rps8 RILs from Williams × PI 399073 BC₄F_2:3 and PI 399073 × PI 408211B F_4:5 populations. There were no differences for pollen viability among the genotypes. However, for PI 399073, a mix of dyads, triads, tetrads and pentads was observed. A high frequency of meiotic abnormalities including fragments, laggards, multinucleated microspores; and microcytes containing DNA was also observed in Rps8/rps8 Williams × PI 399073 BC₄F_2:3 RILs. These meiotic abnormalities may contribute to the high degree of segregation distortion present in the Williams × PI 399073 populations.     

Development of triploid daylily (<Emphasis Type="Italic">Hemerocallis</Emphasis>) germplasm by embryo rescue

Interploidy crosses between diploid daylily (Hemerocallis) cultivars (2n = 22) and tetraploid cultivars (2n = 44) were performed via in vivo hybridization aiming to produce triploid hybrid germplasm (3n = 33). Plant growth regulator-free MS based medium containing 5 levels of sucrose—1, 2, 3, 4 or 5% was used to optimize the embryo rescue medium. It was determined that the medium supplemented with 3% sucrose gave highest rate of immature hybrid embryo rescue. Thirty-seven hypothetically triploid genotypes were subjected to verification of ploidy status by root tip chromosome counting and flow cytometry. Thirty-one of them were confirmed triploids. These genotypes were in vitro propagated, acclimatized to ex vitro conditions, and planted for future field performance evaluation and environmental testing. The newly developed triploid genotypes could open new horizons for further polyploidy breeding.     

Cyto-morphological evidence for segmental allopolyploid origin of Teasle gourd (<Emphasis Type="Italic">Momordica subangulata</Emphasis> subsp. <Emphasis Type="Italic">renigera</Emphasis>)

Teasle gourd [Momordica subangulata Blume subsp. renigera (G. Don) de Wilde, 2n = 56] exhibits morphological characters found in both M. dioica (2n = 28) and M. cochinchinensis (2n = 28). Morphological analysis of M. subangulata subsp. renigera suggests an allopolyploid origin. We present evidence elucidating the genomic relationships between M. dioica, M. cochinchinensis and M. subangulata subsp. renigera. A triploid M. dioica × M. subangulata subsp. renigera hybrid had an average of 12.76 bivalents, 13.84 univalents and 0.88 trivalents at metaphase I, while the M. cochinchinensis × M. subangulata subsp. renigera hybrid had an average of 13.08 bivalents, 12.96 univalents and 0.96 trivalents. F₁ hybrids of the two diploid species (M. dioica × M. cochinchinensis) showed an average of 9.12 bivalents and 9.76 univalents, suggesting that the genomes of these species are only partially homologous. A higher number of bivalents in the triploid hybrids suggests that M. subangulata subsp. renigera is a segmental allopolyploid of M. dioica and M. cochinchinensis and that its genomes have diverged from the parental genomes.     

In vitro induction of tetraploids in ornamental <Emphasis Type="Italic">Ranunculus</Emphasis>

This study investigates the capacity of the antimitotic agents colchicine, oryzalin and trifluralin for inducing polyploidisation of Ranunculus asiaticus ‘Alfa’ in vitro shoots. Flow cytometry was used to evaluate the optimal concentration of each antimitotic agent for polyploidisation. Trifluralin at a concentration of 2 μM resulted in the highest percentage of polyploidisation (27.5%), followed by a colchicine treatment of 200 μM, which induced 23.3% of polyploids. For oryzalin the highest percentage was achieved using a concentration of 1 μM. Different exposure periods were tested and turned out to be an important factor. The maximal exposure period tested (10 weeks) resulted in a significant increase in polyploidisation by oryzalin and trifluralin. In contrast, for colchicine (100 μM) exposure times of either 16 or 24 h did not significantly influence polyploidisation. Additionally the effect of the antimitotic agents on the viability was analysed. For colchicine no significant effect on the survival rate was observed, for trifluralin only a concentration of 10 μM affected viability whereas for oryzalin, concentration as well as exposure period were significant parameters. Flow cytometric data were confirmed by counting chromosomes in root tip cells.     

Interspecific hybrids between <Emphasis Type="Italic">Chrysanthemum grandiflorum</Emphasis> (Ramat.) Kitamura and <Emphasis Type="Italic">C. indicum</Emphasis> (L.) Des Moul. and their drought tolerance evaluation

Chrysanthemum grandiflorum ‘Yuhuaxingchen’ is an important commercial chrysanthemum cultivar with excellent ornamental quality but low drought tolerance, whereas C. indicum has exceptional drought tolerance. In our earlier study, many hybrid seeds between them were obtained through interspecific hybridization. In the present study, we selected six putative hybrid lines with most drought tolerance from all the hybrid lines by withholding water, indentified their facticity by chromosome counting, and then evaluated their drought tolerance through determining foliar electrolyte leakage (EL), contents of malondialdehyde (MDA) and proline, and plant survival rate after 20% polyethylene glycol 6000 treatment. It was found that 155 out of 282 seeds germinated and only 132 seedlings survived. In addition, chromosome and morphological analysis showed that the six putative hybrids were real hybrids and their morphological features were intermediate between their parents. Furthermore, the density of leaf epidermal hair, proline content, and plant survival rate were the highest in C. indicum and the lowest in C. grandiflorum among the six hybrids and their parents. In contrast, EL value and MDA content were the highest in C. grandiflorum and the lowest in C. indicum. These results suggest that some true hybrids with improved drought tolerance can be obtained through interspecific hybridization in chrysanthemum breeding. Therefore, interspecific hybridization between chrysanthemum cultivars and their wild species may become a promising way to improve their biotic and abiotic resistance in the future breeding.     

Interspecific hybridisation and genome size analysis in <Emphasis Type="Italic">Buddleja</Emphasis>

Interspecific hybrids Buddleja davidii × Buddleja weyeriana, Buddleja weyeriana × Buddleja davidii and Buddleja davidii × Buddleja lindleyana were generated using in vitro embryo rescue 10–11 weeks after manual pollination. The morphological variation within the F1 populations was limited. The F1 progeny of B. davidii × B. lindleyana was almost sterile and no F2 generation was obtained. From the other hybrids, F2 generations were made by self pollinations and back crosses. Hybrid nature of all F1 and F2 seedlings was confirmed by AFLP. Chromosome counting and genome size measurement for B. weyeriana (F2 selection of (diploid) B. globosa × (tetraploid) B. davidii) revealed a higher chromosome number (76 chromosomes) and genome size than expected, indicating 2n-gametes formation occurred during meiosis of B. globosa. The F1 hybrids B. weyeriana × B. davidii (76 chromosomes) had an intermediate genome size compared with the genome size of the parent plants, proving their hybrid nature. However, the F1 and F2 hybrids of B. davidii × B. weyeriana all had 76 chromosomes but had a lower genome size than expected, suggesting the occurrence of chromosome rearrangements in the genome of the hybrids. B. lindleyana had 38 chromosomes, while the F1 hybrids of B. davidii × B. lindleyana had 76 chromosomes. Also genome size measurements revealed that the F1 seedlings B. davidii × B. lindleyana had higher genome sizes than expected. Both the results of chromosome counting and genome size measurement indicate that 2n-gametes formation took place during meiosis of B. lindleyana.     

Broadening the genetic base of <Emphasis Type="Italic">Brassica napus</Emphasis> canola by interspecific crosses with different variants of <Emphasis Type="Italic">B. oleracea</Emphasis>

Broadening the genetic base of the C genome of Brassica napus canola by use of B. oleracea is important. In this study, the prospect of developing B. napus canola lines from B. napus?×?B. oleracea var. alboglabra, botrytis, italica and capitata crosses and the effect of backcrossing the F₁’s to B. napus were investigated. The efficiency of the production of the F₁’s varied depending on the B. oleracea variant used in the cross. Fertility of the F₁ plants was low—produced, on average, about 0.7 F₂ seeds per self-pollination and similar number of BC₁ seeds on backcrossing to B. napus. The F₃ population showed greater fertility than the BC₁F₂; however, this difference diminished with the advancement of generation. The advanced generation populations, whether derived from F₂ or BC₁, showed similar fertility and produced similar size silique with similar number of seeds per silique. Progeny of all F₁’s and BC₁’s stabilized into B. napus, although B. oleracea plant was expected, especially in the progeny of F₁ (ACC) owing to elimination of the A chromosomes during meiosis. Segregation distortion for erucic acid alleles occurred in both F₂ and BC₁ resulting significantly fewer zero-erucic plants than expected; however, plants with?≤?15% erucic acid frequently yielded zero-erucic progeny. No consistent correlation between parent and progeny generation was found for seed glucosinolate content; however, selection for this trait was effective and B. napus canola lines were obtained from all crosses. Silique length showed positive correlation with seed set; the advanced generation populations, whether derived from F₂ or BC₁, were similar for these traits. SSR marker analysis showed that genetically diverse canola lines can be developed by using different variants of B. oleracea in B. napus?×?B. oleracea interspecific crosses.     

Creation of new maize germplasm using alien introgression from <Emphasis Type="Italic">Zea mays</Emphasis> ssp. <Emphasis Type="Italic">mexicana</Emphasis>

Zea mays ssp. mexicana, an annual wild relative of maize, has many desirable characteristics for maize improvement. To transfer alien genetic germplasm into maize background, F₁ hybrids were generated by using Z. mays ssp. mexicana as the female parent and cultivated maize inbred line Ye515 as the male parent. Alien introgression lines, with a large range of genetic diversity, were produced by backcross and successive self-pollinations. A number of alien introgression lines with the predominant traits of cultivated maize were selected. Genomic in situ hybridization (GISH) proved that small chromosome segments of Z. mays ssp. mexicana had been integrated into the maize genome. Some outstanding alien introgression lines were evaluated in performance trials which showed 54.6% hybrids had grain yield greater than that of hybrid check Yedan12 which possessed 50% Ye515 parentage, and 17.1, 9.9% hybrids had grain yield competitive or greater than those of Nongda108 and Zheng958, which were elite commercial hybrids in China, respectively. The results indicated that some of the introgression lines had excellent agronomic traits and combining ability for maize cultivar, and demonstrated that Z. mays ssp. mexicana was a valuable source for maize breeding, and could be used to broaden and enrich the maize germplasm.     

Fine mapping of a <Emphasis Type="Italic">Xantha</Emphasis> mutation in rice (<Emphasis Type="Italic">Oryza sativa</Emphasis> L.)

The recessive mutation of the XANTHA gene (XNT) transforms seedlings and plants into a yellow color, visually distinguishable from normal (green) rice. Thus, it has been introduced into male sterile lines as a distinct marker for rapidly testing and efficiently increasing varietal purity in seed and paddy production of hybrid rice. To identify closely linked markers and eventually isolate the XNT gene, two mapping populations were developed by crossing the xantha mutant line Huangyu B (indica) with two wild type japonica varieties; a total of 1,720 mutant type F₂ individuals were analyzed for fine mapping using polymorphic InDel markers and high dense microsatellite markers. The XNT gene was mapped on chromosome 11, within in a fragment of ~100 kb, where 13 genes are annotated. The NP_001067671.1 gene within the delimited region is likely to be a candidate XNT gene, since it encodes ATP-dependent chloroplast protease ATP-binding subunit clp A. However, no sequence differences were observed between the mutant and its parent. Bioinformatics analysis demonstrated that four chlorophyll deficient mutations that were previously mapped on the same chromosome are located outside the XNT region, indicating XNT is a new gene. The results provide useful DNA markers not only for marker assisted selection of the xantha trait but also its eventual cloning.