Modulation of seed vigor

Abstract

The present invention provides a polynucleotide which enables the modulation of the seed vigor, and in particular enhances the seed vigor, and more particularly enables the modification of the speed of germination. A plant seed comprising the said polynucleotide is also provided. A method of producing the plant seed, method for improving the germination and vigor of plant seed, transgenic plant and the use of the polynucleotide of the invention for producing plants growing seeds with improved germination and vigor characteristics are also provided. The invention particularly concerns Brassica, more particularly Brassica oleracea.

Claims

1. A cultivated hybrid Brassica plant or plant part which contains within its genome an introgression comprising the nucleotide sequence of SEQ ID NO: 16 at the Speed of Germination (SOG1) QTL from Brassica oleracea line SL101, representative seed of line SL101 having been deposited at NCIMB under deposit number NCIMB 41951, and which exhibits an increase in speed of seed germination as compared to a Brassica plant or plant part that does not comprise the introgression from Brassica oleracea line SL101.

2. The cultivated Brassica plant or plant part according to claim 1 that is selected from the group consisting of a Brassica oleracea, Brassica napus, and Brassica rapa plant or plant part.

3. The cultivated Brassica plant or plant part according to claim 1 that is a Brassica oleracea plant or plant part.

4. A cultivated hybrid Brassica plant or plant part which contains within its genome an introgression comprising the nucleotide sequence of SEQ ID NO: 16 at the Speed of Germination (SOG1) QTL from Brassica oleracea line SL101, representative seed of line SL101 having been deposited at NCIMB under deposit number NCIMB 41951, and which exhibits an increase in percentage of seed germination as compared to a Brassica plant or plant part that does not comprise the introgression from Brassica oleracea line SL101.

5. The cultivated Brassica plant or plant part according to claim 4 that is selected from the group consisting of Brassica oleracea, Brassica napus, and Brassica rapa.

6. The cultivated Brassica plant or plant part according to claim 4 that is a Brassica oleracea plant or plant part.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 is a graphic representation of the cumulative germination curves of seeds from the substitution line SL101 (?) and parent lines (A12DHd (?) and GDDH33 (?) at 15? C. on water. Vertical lines are standard errors.

(2) FIG. 2 is a graphic representation of the cumulative germination curves of SL101 (?) and the recurrent A12DHd (?) parent and reciprocal F.sub.1 backcross lines (A12DHd?SL101 (?) and SL101?A12DHd (?). Vertical lines are standard errors.

(3) FIG. 3 is a graphic representation of the cumulative seedling emergence in the field from seeds of the substitution line SL101 (?) and parental lines A12DHd (?) and GDD33H (?).

(4) FIG. 4 is a diagram illustrating endogenous concentrations of ABA during germination in seeds of the substitution line SL101 (white column) and parent lines, GDDH33 (black column) and A12DHd (grey column).

(5) FIG. 5 is a diagram illustrating the speed of seed germination in 3, for gene At3g01060 (black columns), and 2, for gene At3g01150 (grey columns), KO lines. The wild type control line (white column) is col0.

(6) FIG. 6 is a diagram illustrating the speed of seed germination in KO lines 102 (light grey line) and 18 (low medium grey line) (for gene At3g01060) and in KO lines 15 (high medium grey line) and 3 (dark grey line) (for gene Atg01150) compared to wild type control line Col0 (black line), following seed production in the glasshouse (more stressful conditions).

EXAMPLE 1

Material and Methods

(7) Seed Production and Comparison of Lines

(8) Seed samples were obtained from Birmingham University, UK, for a range of Brassica oleracea chromosome substitution lines derived from the doubled haploid parent lines A12DHd (var. alboglabra) and GDDH33 (var. italica; Rae et al., 1999). Bulk seeds were then produced and collected from 10 individual replicate plants of the substitution lines and the GDDH33 parent and 20 plants of the recurrent A12DHd parent as substitution lines are compared to the latter in germination experiments. Plants were laid out in a randomized block with 10 replicates in a glasshouse at 16-18? C. during the 16 h day and at 10-15? C. at night as described by Bettey et al. (2000). Supplementary lighting (400 W high pressure sodium lamps; Osram Ltd, St Helens, UK) was supplied when light intensity fell below 300 w m.sup.2 during the 16 h day. Plants were self-pollinated by enclosing the inflorescences in perforated polyethylene bags containing blowflies before the flowers opened. The seedpods were allowed to dry completely on the plant within the enclosing bags before harvest. The seeds were cleaned, equilibrated at 15% rh and 15? C., and then stored at ?20? C. before germination experiments were carried out. Cumulative germination on moist filter paper was recorded at 15? C. on 4 replicates of 15 seeds collected from each of the 10 replicate plants (or 20, A12DHd) described above. Previous work had shown this to be sufficient seeds (Bettey et al., 2000). Frequent counts were made to allow an accurate calculation of the time to 50% germination from these measurements. Percentage germination was high in all seed lots.

(9) In later experiments, F1 seeds from reciprocal backcrosses were produced in the same manner as described above. Bud pollination was performed to make the cross resulting in the F1. Seeds from the parent lines were produced at the same time for comparison to minimize the influence of environmental differences during seed production. In addition, on a number of occasions at different times of the year seeds were produced from replicate plants of both the parent A12DHd and substitution line SL101 in glasshouses as described above. Although glasshouse heating, venting and lighting settings remained the same as those described above ambient temperature differed and was recorded.

(10) Germination Assays

(11) Three biological replicates of 50 seeds from substitution and parent Brassica oleracea lines or 3 to 15 biological replicates of 50 seeds from Arabidopsis wild type and mutant lines were placed to germinate on 2 layers of filter paper (Whatman International Ltd., UK) kept moist throughout treatment with water.

(12) In all germination experiments, seeds on filter paper were held in clear polystyrene boxes laid out in randomized blocks and kept in the dark. No evidence of fungal infection was observed and so seeds were not sterilized to avoid influencing their germination. Germination (radicle emergence) was recorded at intervals to construct cumulative germination curves.

(13) Field Emergence

(14) As part of a larger unpublished comparison of seedling emergence from B. oleracea genotypes, 100 seeds of the parent lines GDDH33, A12DHd and substitution line SL101 were sown on 31 May, in 4 replicate 1 m rows arranged in a randomized block. Seeds were sown by hand in a 15 mm deep furrow, covered with sieved soil (sieve hole size <4 mm) and the surface rolled once with a Stanhay seed drill press wheel. The soil was a sandy loam and irrigation was applied to maintain soil moisture throughout seed germination and seedling emergence. The latter was recorded at regular intervals until no more seedlings emerged.

(15) Hormone Analysis

(16) Samples of seeds from substitution and parent lines were taken as non-imbibed dry seeds and seeds imbibed for 24 and 48 hours at 15? C. on moist filter paper. Each sample contained 1 g of dry seed measured before imbibition. The samples were placed immediately in liquid nitrogen, freeze dried and then placed in a domestic freezer at ?20? C. until extraction. The samples were in 10 ml of cold (4? C.) 80% methanol (containing 20 mg I-I of BHT), then 500 ng ABA and 100 ng each of GA standards were added. The samples were stirred overnight in a cold cabinet and then centrifuged. The supernatant was decanted and the pellet re-suspended in a further 10 ml of 80% methanol. The samples were stirred for 4 h, centrifuged and the supernatants combined. The supernatants were evaporated to aqueous (c. 3-4 ml) and 10 ml of 0.5 M pH 8.2 phosphate buffer was added, samples were partitioned with 2?15 ml of dichloromethane, and the dichloromethane discarded. The aqueous fraction was adjusted to pH 3 with 1 M phosphoric acid partitioned with 3?15 ml of ethylacetate.

(17) The combined ethylacetate fractions were washed with 2?3.5 ml pH 3.0 water and evaporated to dryness, re-dissolved in 5 ml water and the pH adjusted to 8. The solution was then loaded onto QMA cartridges which were subsequently washed with 5 ml of 15% methanol pH 8.0. GAs and ABA were eluted from the QMA cartridges directly onto C18 21 cartridges with 0.2 M formic acid in 5% methanol. C18 cartridges were then washed with 5 ml of 20% methanol and samples recovered in 5 ml of 80% methanol and evaporated to dryness. Samples were then dissolved in methanol and methylated with excess ethereal diazomethane. Following evaporation to dryness samples were re-dissolved in dry ethylacetate and passed through amino cartridges. The resulting samples were analyzed directly for ABA content by GC-MS, then evaporated to dryness and re-dissolved in BSTFA prior to analysis for GA content by GC-MS.

(18) Marker Analysis

(19) Primer pairs were designed to 30 Arabidopsis gene models that were spread at intervals across the SOG1 region using Primer 3 software (on the World Wide Web at primer3plus.com) and gene data from Tair (on the World Wide Web at arabidpsis.org/) to give PCR products from 200 to 700 bp. The PCR mix used was standard but a touch-down program was used. This consisted of cycling parameters as follows: 94? C. for 5 mins; then annealing at 65? C. to 55? C. for 10 cycles dropping a degree each cycle with 30 s extension at 72? C. and 30 s denaturation at 94? C. over the 10 cycles; followed by 30 cycles of 94? C. 30 s, 55? C. 30 s, 72? C. 45 s; and a final extension at 72? C. for 15 min. Primer sequences for the gene models that gave polymorphic results were selected as markers (Table 1).

(20) Data Analysis

(21) All analyses were performed using the statistical package Genstat 5 (Payne et al. 1993), and where appropriate data were subjected to analyses of variance.

(22) Results

(23) QTL for Speed of Germination (SOG1) on Linkage Group C1 Confirmed and Fine-Mapped

(24) Analysis of variance of germination data comparing substitution and parent lines showed that the GDDH33 parent germinated significantly (P<0.001) faster than the A12DHd parent confirming that the positive speed of germination alleles are provided by GDDH33, as shown by Bettey et al. (2000) (FIG. 1). There were 4 substitution lines which spanned the SOG1 QTL (SL101, SL111, SL118, SL119) and all of these had significantly (P<0.005) faster germination than the A12DHd parent. The substitution line SL101 had the smallest introgressed region (1-9 cM; Rae et al., 1999) that enhanced speed of germination compared to A12DHd and accounted for much of the difference in speed of germination between the parent lines (FIG. 1) and was therefore selected for further study of SOG1.

(25) The SOG1 Fast Germinating Phenotype is not Influenced by the Maternal Genotype

(26) Speed of germination is determined by the embryo, but can also be significantly influenced by the tissues that surround it which are maternal in origin. Reciprocal backcrosses between SL101 and A12DHd and between GDDH33 and A12DHd were carried out to determine the maternal and zygotic genetic components at the SOG1 locus. Germination was recorded from the F1 seeds of each cross and from seeds of the selfed parent lines produced at the same time. There was no significant difference in the speed of germination of SL101 and the F1 from the reciprocal backcrosses with A12DHd (SL101 as mother plant and pollen from A12DHd and vice versa), but germination of seeds from all three were significantly (P<0.01) faster than that from seeds of A12DHd (FIG. 2). This shows the faster germinating GDDH33 allele to be dominant with no genetic maternal influence on inheritance of the trait confirming that it is embryo based.

(27) Differences in Speed of Germination Lead to Differences in the Timing of Seedling Emergence in the Field

(28) The data above show that the speed of germination of GDDH33 and SL101 seeds was significantly greater than that of A12DHd under constant temperature conditions. In the field this resulted in significantly earlier seedling emergence from GDDH33 and SL101 than from A12DHd (FIG. 3).

(29) Endogenous ABA Concentration Differs Between Genotypes at Maturity

(30) The endogenous concentration of ABA in dry and imbibing seeds of the three genotypes was measured using GCMS. There was no significant difference in the endogenous concentration of ABA in the dry seed of SL101 and GD33 or during imbibition to 48 hours just prior to radicle emergence. ABA concentration in these two seed lots remained the same over the first 24 hours and then declined to half that by 48 hours. In contrast, the endogenous concentration of ABA in seeds of A12DHd was initially three fold higher than that in SL101 and GDDH33 and then declined progressively over the 48 hour period of imbibition, but remained significantly above that of the other two seed lots (FIG. 4). Interestingly, if ABA continues to decline at the same speed in A12DHd it would reach the same level after 72 hours, the point of germination, as seen in the other two lines immediately before their germination.

(31) The results for GDDH33, SL101 and the A12DHD lines presented here show that a clear genetically determined relationship between higher endogenous ABA content, and lower speed of germination and vice versa.

(32) A Quantitative Genetic Analysis of the Speed of Germination Trait in Brassica oleracea has been Carried Out.

(33) By fine mapping the QTL with the previously described substitution line SL101, we finally identified two genes underlying the QTL (Speed Of Germination (SOG1)). This line SL101 has a short introgression at the telomeric end of C1 from the fast germinating parent (GD33, Broccoli) in the background of the slow germinating parent (A12, Kale) in Brassica oleracea. By the way of using markers across this introgressed region of SL101, 30 recombinations along this area were identified within 1,300 lines, thanks to a BAC tilling path strategy. This strategy allows gathering the lines into 5 distinct groups and the 2 parent lines. Seed germination was then evaluated throughout these 7 groups and statistical analysis revealed that faster germination was associated with 2 markers at the telomeric end of C1. Co-localization of these markers with a single BAC on the telomeric end of C1 was further assessed by Fluorescence In-Situ Hybridization. The BAC was sequenced and 12 full-length genes were found to be present within.

(34) Putative orthologs of these Brassica genes have been identified in Arabidopsis, at the top arm of chromosome 3, where a SOG1 QTL has been located (Clerkx et al. 2004). Based on the good genetic colinearity between the telomeric end of C1 in Brassica oleracea and the top arm of chromosome 3 in Arabidopsis (see above and below), it is reasonable to think that the genes that were identified in Brassica and in Arabidopsis do share common function in seed germination. In order to strengthen our hypothesis, we have identified Arabidopsis knockout (KO) mutant lines in putative ortholog genes of those that have been discovered in Brassica. Two of these KO lines were found to have a significant germination phenotype (faster germination compared to control Col0 line) suggesting that these genes act as negative regulators of germination (FIG. 5). It is interesting to note that at least two different KO lines were used to assess germination phenotype and that these distinct KO lines showed similar results regarding to the speed of germination. When assessed following seed production in more stressful conditions (FIG. 6), the results are very comparable.

(35) This functional study has thus confirmed the role of Atg01060 and Atg01150 genes in the regulation of seed vigour, in particular in the regulation of the speed of germination. The Brassica orthologs of these genes, BolC.VG1.a and BolC.VG1.b, which have been identified within the SOG1 QTL in Brassica oleracea, can therefore indeed be considered as tools allowing the modulation of the seed vigour, more particularly the modulation of the speed of seed germination in brassicaceae.

(36) The Linkage Between B. oleracea Linkage Group C1 and the Top Arm of Chromosome Three of Arabidopsis is Confirmed

(37) Studies on SL101 above have shown that a single introgressed region at the telomeric end of linkage group C1 (LGC1) contains the QTL for SOG1. In the current work we aimed to establish colinearity between this region in B. oleracea and the Arabidopsis genome to enable comparison with the extensive QTL analysis carried out on this model species. Previously a number of linkages have been shown between the Brassica genome and Arabidopsis (Cogan et al., 2004; Parkin et al., 2005).

(38) In particular, the linkage between the LGC1 and Arabidopsis (Cogan et al., 2004) was utilized to assist in the development of further informative markers. Using this approach, primer pairs were designed to 30 Arabidopsis gene models, which were spread across this region. These primers were tested to determine if they amplified a B. oleracea product and then if there were any polymorphisms between SL101 and the parental lines A12DHd and GD33DHd (table 1). A banding pattern that is the same in SL101 and GD33DHd, but different from that in A12DHd indicates its presence at this locus, and therefore its usefulness as a marker for SOG1. Primers for three gene models were identified as informative markers (At3g01190, At3g07130, At3g02420, table 1) that anchored the linkage between the top arm of Arabidopsis chromosome three and the SOG1 region of B. oleracea LGC1. This confirmation of colinearity justifies comparison of SOG1 with QTL for seed performance located to this region of the Arabidopsis genome.

(39) TABLE-US-00001 TABLE1 PrimersforselectedmarkersusedtoanchorSL101 introgressiontoArabidopsis Genemodel Primers At3g01190 F:TTCTTCCACGACTGCTTCG(SEQIDNO:1) R:CTAACAAAACTGATCCGTCAC(SEQIDNO:2) At3g02420 F:GTTGCGTTGCCATCTGCAG(SEQIDNO:3) R:CAGGCTGAGATAGCCATTGG(SEQIDNO:4) At3g07130 F:CTACTAACCATGGAGTTACC(SEQIDNO:5) R:AACGCTGGTGGGATTCAC(SEQIDNO:6)

(40) Altogether, these results open up the possibility of using either BolC.VG1.a A12/GD33 alleles or BolC.VG2.a A12/GD33 alleles or BolC.VG2.b A12/GD33 alleles according to any of embodiments 1 to 23, in particular to engineer plants in which the seed vigour has been modulated, more particularly plants in which speed of germination has been modulated.

EXAMPLE 2

Experiments Highlighting Additional Phenotypes Linked to Seed Vigour

(41) These experiments underlines the differences in germination characteristics between seeds with (SL101) or without (A12) the GD33 allele.

(42) Seeds of the lines A12 and SL101 originated in the UK and were replicated in 2009 in South Africa by a commercial seed producer. Determination of the germination characteristics of the seeds took place in early 2010 in Enkhuizen, Netherlands.

(43) Seed Performance Under Standard Commercial Conditions

(44) Two replicates of 100 seeds were sown in standard trays filled with soil as in a normal commercial practice. Trays were placed in a germination chamber at 18? C. in the dark for three days. Trays were then transferred to a greenhouse with an average temperature of 20? C. At 10 days after sowing the number of normally developed seedlings and the number of non-emerged seeds were counted.

(45) Performance of SL101 under these conditions was considerably better than for A12 (table 2).

(46) TABLE-US-00002 TABLE 2 Percentage normal seedlings and percentage non germinated seeds as measured under practical conditions of seedling production normal seedlings non germinated seeds A12 19 80 SL101 91 7
Temperature Sensitivity of Germination on Paper

(47) Two replicates of 100 seeds of each A12 and SL101 were sown for each combination of conditions on wet filter paper in transparent plastic germination boxes at temperatures of 10, 20 or 30? C. Boxes were either kept in the dark at the mentioned temperatures or were placed under fluorescent light.

(48) Germination, measured as radical protrusion, was counted daily. Germination was counted until 10 days after the start of the measurement when no additional germination was observed.

(49) The data show that under each condition the percentage germination was higher for SL101 compared to A12 (table 3).

(50) TABLE-US-00003 TABLE 3 Final germination percentage of A12 and SL101 at different temperatures in the light or in the dark temperature 10 20 30 light A12 47 94 59 SL101 94 100 99 dark A12 65 24 2 SL101 95 78 50
Speed of Germination on Paper

(51) Speed of germination was determined under conditions of 20? C. in the light, a condition where both lines had their maximum germination. Two replicates of 50 seeds per line were placed on paper in plastic germination boxes under the mentioned conditions. The t50 (time until 50% of the germinating seeds have germinated) was determined (table 4). As shown in the table the time until 50% germination was considerably shorter for SL101 compared to A12.

(52) TABLE-US-00004 TABLE 4 Time until 50% germination at 20? C. in the light for A12 and SL101 line t50 (h) A12 59.0 SL101 47.5
Sensitivity to Low Temperatures During Germination on Paper

(53) Two replicates of 50 seeds per treatment were incubated at 5? C. or 10? C. in the light as described before for a germination test on paper. Incubation was in water or in a 1 mM solution of GA3. After 10 days seeds that had germinated were removed. The boxes with the remaining non-germinated seeds were then placed at 20? C. in the light. After a further 5 days the number of germinated seeds was counted. As a comparison seeds of both A12 and SL101 were germinated at 20? C. in the light without a pretreatment. Germination percentage of these seeds was counted after 10 days.

(54) Especially at 5? C. there was a large difference between the two lines (table 5). The pretreatment in water largely prevented for A12 germination at 20? C. in the light later on. This was not the case for pretreatment in 1 mM GA3. SL101 showed only a minor reduction in germination after the pretreatment at 5? C. in water.

(55) TABLE-US-00005 TABLE 5 Percentage of germination of seeds of A12 and SL101 with or without a pretreatment at 5 or 10? C. in the light with an incubation medium of either water or a 1 mM aqueous solution of GA3 A12 SL101 no pretreatment 99 100 10 d 5 C. in water 15 84 10 d 5 C. in 1 mM GA3 100 100 10 d 10 C. in water 82 99 10 d 10 C. in 1 mM GA3 100 100

EXAMPLE 3

Effect of Genotype on Hybrid Seed Performance

(56) Hybrid seeds were produced under commercial seed production conditions in South Africa by pollinating flowers of two male sterile lines with pollen from 7 other lines.

(57) Female line 1 contained the GD33 allele, female line 2 contained the A12 allele.

(58) The sensitivity to low temperature during germination was tested for these seeds. Seeds were incubated on moistened filter paper at 5? C. in white fluorescent light for 10 days. Percent germination was recorded after 10 days and non germinated seeds were transferred to 20? C. in light. After five days the percentage germinated seeds was recorded.

(59) Seeds produced on the female line containing the GD33 allele showed considerably higher germination at 5? C. and germinated more than 90% after transfer to 20? C. Seeds produced on the female without the GD33 allele showed only low germination percentages and did not recover after transfer to 20? C.

(60) TABLE-US-00006 TABLE 6 Percentage of germination of hybrid seeds produced using a GD33 or A12 female line with a pretreatment for 10 days at 5? C. in the light followed by incubation of non germinated seeds for 5 days at 20? C. female 1 (GD33) female 2 (A12) 10 d 5 C. +5 d 20 C. 10 d 5 C. +5 d 20 C. male1 35 96 0 6 male2 28 96 0 10 male3 72 94 19 44 male4 9 96 0 4 male5 43 100 0 8 male6 28 99 0 4 male7 69 97 35 58 average 41 97 8 19

(61) Seeds from the same batches of hybrid seed were tested in a germination test on paper at 25? C. in white fluorescent light From daily counts of germination the time until 50% germination was calculated (t50). The final germination percentage was determined after 5 days. All hybrids had at least 85% germination, the majority more than 95%. Seeds produced on the female line with the GD33 allele showed on average around 25% faster germination, shown by the lower t50 of 1.33 days versus 1.80 days for seeds produced on the female with the A12 allele.

(62) TABLE-US-00007 TABLE 7 Percentage germination and time taken until 50% germination (t50) of hybrid seed produced using a GD33 or A12 female line in the light at 25? C. female 1 (GD33) female 2 (A12) germ % t50 (d) germ % t50 (d) male 1 100 1.08 99 1.73 male 2 97 1.95 98 1.85 male 3 98 1.85 100 1.90 male 4 100 1.15 95 1.84 male 5 99 1.07 98 1.68 male 6 99 1.14 85 1.84 male 7 95 1.05 98 1.79 average 98 1.33 96 1.80

REFERENCES

(63) Bettey M., Finch-Savage W. E., King G. J., Lynn J. R. 2000. Quantitative genetic analysis of seed vigour and pre-emergence seedling growth traits in Brassica oleracea L. New Phytol 148: 277-286. Clercks E. J. M., EI-Lithy M. E., Vierling E., Ruys G. J., Blankestijn-De Vries H., Groot S. P. C., Vreugdenhil D., Koornneef M. 2004. Analysis of natural allelic variation of Arabidopsis seed germination and seed longevity traits between the accessions Landsberg errecta and Shakdara, using a new recombinant inbred line population. Plant Physiol 135: 432-443. Finch-Savage W. E. 1995. Influence of seed quality on crop establishment, growth and yield. In AS Basra, ed. Seed quality: Basic mechanisms and agricultural implications. Haworth Press, Inc, New York, pp 361-384. Finch-Savage W. E., C?me D., Ly J. R., Corbineau F. 2005. Sensitivity of Brassica oleracea seed germination to hypoxia: a QTL analysis. Plant Sci 169: 753-759. Finch-Savage W. E, and Leubner-Metzger G. 2006. Seed dormancy and the control of germination. New Phytol. 171: 501-523. Finkelstein R. R., Gampala S. S., and Rock C. D. 2002. Abscisic acid signaling in seeds and seedlings. Plant Cell 14, S15-45. Foolad M. R., Lin G. Y., Chen F. Q. 1999. Comparison of QTLs for seed germination under non-stress, cold stress and salt stress in tomato. Plant Breed 118: 167-173. Groot S. P. C., van der Geest A. H. M., Tesnier K. J. Y., Alonso-Blanco C., Bentsink L., Donkers H., Koornneef M., Vreugdenhil D., Bino R. J. 2000. Molecular genetic analysis of Arabidopsis seed quality. In M Black, J Vasquez-Ramos, eds, Seed Biology: Advances and applications, CAB International., London, pp 123-132. Hilhorst H. W. M. and Toorop P. E. 1997. Review on dormancy, germinability and germination in crop and weed seeds. Advances Agron 61: 115-165. Holdsworth M. J., Bentsink L., Soppe W. J. J. 2008a. Molecular networks regulating Arabidopsis seed maturation, after-ripening, dormancy and germination. New Phytol 179: 33-54. Holdsworth M. J., Finch-Savage W. E., Grappin P., Job J. 2008b. Post-genomics dissection of seed dormancy and germination. Trends Plant Sci 13: 7-13. Ruan S. L., Duan X. M. and Hu W. M. 2000. Occurrence of seed vivipary in hybrid rape (Brassica napus L.) and its effect on seed quality. Journal of Zhejiang University (Agriculture and Life Sciences) 2000 Vol. 26 No. 5 pp. 573-578.

SEQUENCE ALIGNMENTS

(64) DNA sequence alignment for the Brassica oleracea VG2 genes. Alignments were performed using the ClustalW web based program (on the World Wide Web at genome.jp) and the annotation is drawn using the web based program BOXSHADE (on the World Wide Web at ch.embnet.org). A12_BOLC.VG2.A is an A12 full length copy of the gene located in the SOG1 region of linkage group C1. A12_BOLC.VG2.B is a truncated copy of a similar gene located within 50 Kb of the full length gene. A similar annotation has been used to describe the same region in the GD33 genomic background.

(65) TABLE-US-00008 0 0 0 0 0 0 0 0 0

(66) Protein sequence alignment of VG2 Brassica oleracea orthologue proteins from GD33 and A12. Protein sequences were predicted using the web based bioinformatics program FGENESH (on the World Wide web at softberry.com).

(67) TABLE-US-00009

(68) DNA Sequence Alignment for the VG1 Brassica oleracea Genes

(69) TABLE-US-00010 00 01 02 03 04 05 06 07 08 09 0 0 0 0 0

(70) Protein sequence alignment of VG1 Brassica oleracea orthologues proteins from A12 and GD33. Protein sequences were predicted using the web based bioinformatic program FGENESH.

(71) TABLE-US-00011