Abstract

Introduction

The development of an ideotype, which describes oilseed rape more resistant to seed losses at harvest and maintains high performance agronomically, depends on the identification of relevant characteristics of the plant architecture and pod anatomy.

Prior to harvest, losses occur because of adverse weather conditions and during harvest, because of the impact of the combine. It is known that the structure and distribution of the components of the canopy affect seed recovery as well as yield potential (Child and Evans, 1989). In addition, canopy height and stem stiffness may affect the efficiency of seed recovery. Seed losses may also be affected by characteristics of the canopy architecture which are determined by raceme morphology such as pod angles, length, thickness and width. The relative importance of these canopy characteristics has been described by Morgan et al., (1998).

Fully mature pods of current commercial cultivars are extremely sensitive to opening. Sutures situated on both sides of the bi-valved pod contain dehiscence zones (DZ’s) which are composed of a layer, 3-4 cells wide, of simple, parenchymatous cells. The DZ is situated between the pod valve edge and a central replum that contains the main vascular bundle to the stalk or pedicel. Small, widely spaced vascular bundles pass from the pod wall (PW) through the DZ and connect with the main vascular bundle. Late in senescence, the cells of the DZ separate and the valves are connected only by the vascular traces, which fracture in contact with other canopy components, or with combine surfaces, releasing seeds. Reduction in the extent of cell separation has been proposed as a target for increasing resistance to opening (Jenkins et al., 1996). Current progress towards achieving this goal by genetic engineering methods is described by Child et al., (1999).

Individual plants in cultivars that have been developed for specific agronomic characters probably vary genetically in some of their apparently minor characters. In the cultivar Jet Neuf, shatter susceptibility varies significantly in pods from different plants (Child and Huttly, 1999). Increased resistance appeared to be associated with increased lignification and vascular tissue in the DZ. It is not known whether this variation is genetically or environmentally determined.

However, greater uniformity of individual characters may be expected in completely true breeding lines that are derived from doubled haploid parents. Therefore, the extent of variation in plant and pod characters, as well as DZ structure, was investigated amongst a range of synthetic lines derived from doubled haploid lines of synthetic B. napus developed at the John Innes Centre. The relative importance of plant and pod characters in determining shatter susceptibility is summarised in this paper.

Materials and methods

Experimental approach.

Derivation of plants. The material for this project was derived from synthetic lines developed from several wild B. oleracea and B. rapa parents. Microspore culture was used to derive doubled haploid lines from crosses of the synthetic lines onto B.napus cv Tapidor and doubled haploid oilseed rape breeding lines.

Production of material for characterisation. Plants that had produced five true leaves were vernalised for 8 weeks at 4^oC in short days. They were then planted in lines in the field at the John Innes Centre or were transferred to pots and grown under cold glass at IACR-Long Ashton. In addition to 210 lines derived from synthetic material, twelve oilseed rape cultivars including Apex and Tapidor were grown as control plants.

Assessments of plant characters. Plant height, width, stem diameter and number of primary branches were measured on a selected “average” plant. At maturity, the resistance to shattering of each line was estimated visually and manually by bending the pods and giving scores from zero (susceptible) to five (resistant). The terminal raceme from three plants of each of 36 lines was cut off, and with careful handling, stored in the laboratory and assessed for pod and raceme characters before pods were removed for shatter assessment.

Shatter assessments. The force needed to separate a pod valve from the septum was measured using a tensile separation test (TST) devised at Silsoe Research Institute (Davies and Bruce, 1996). A single pod was glued to a fixed base. The upper valve was connected to a universal test machine and steadily raised through 1.5 mm to record force-deflection data. From the results, the energy used to propagate the crack was calculated and expressed per unit area of DZ. This enabled comparison of the toughness of DZ’s between the lines. Pods were subjected to a random impact test (RIT) procedure that consisted of agitation with steel balls. 20 intact mature pods were placed in a cylinder 20cm diameter, on a reciprocal shaker for a standard time and amplitude. The numbers of opened pods were counted at intervals. This test aimed to subject pods to conditions that were similar to those experienced in the canopy during combining.

Comparison of dehiscence zone structure. Pods from each of 20 plants, selected to represent the range of shatter susceptibility, were carefully opened by hand. One-cm portions of the two valves from the pedicel end of each pod were mounted on aluminium specimen stubs using plastic/carbon cement. The preparations were coated with gold under reduced pressure for eight minutes using a Poleron series II Sputter coater and the structure of the valve edges (which included the DZ) examined in a scanning electron microscope (SEM) using an electron beam at 5kV. Images were viewed on a television monitor and photographed for later evaluation.

Results and discussion

Relevance of plant and pod characters to shatter susceptibility.

As well as resistance to shattering, the range of variation observed, was greater in the synthteic lines than in the cultivars for all the plant morphology characters measured in this study. The extent of the variation was dependent on the parents used in the crosses and the combination of diploid parents used in the synthesis of the doubled haploid lines of B. napus.

The data clearly showed that by utilising the variation found within wild species of B. oleracea and B. rapa through the production of synthetic B. napus it is possible to increase the range of variation beyond that available amongst oilseed rape cultivars.

There were significant positive correlations (p<0.001) between values for field scores for shattering, the force needed to open the pods (measured by the TST) and the number of intact pods after 20 s agitation in the RIT. PW thickness showed intermediate positive correlation (p<0.01) with RIT and lesser correlation (p<0.05) with TST and field score for shattering. The pod characters could be split into three separate basic groups representing plant morphology, raceme characters and pod shattering assessment.

Therefore in general, selecting for resistance to pod shattering will not necessarily be associated with a particular raceme type or other characters associated with plant morphology. In other words, these groups of characters are, broadly speaking, independent of each other and can thus be selected for or against independently.

The character with the greatest association with resistance as measured in the TST, was beak length. This is of interest because a radiation-induced mutant of Jet Neuf is known to be very resistant to shatter, also has short, bifurcated styles (although associated with reduced fertility).

Structural modification of the DZ in synthetic lines.

Lines with high values for shatter resistance contained anatomical changes in the DZ which involved either increased numbers of thickened cells adjacent to prominent vascular tissue along the inner edge or failure of cells to separate. These two types of modification varied between lines in the extent to which they were present and occurred alone or in combination.

The presence in some lines of large vascular strands lying along the inner edge of the DZ and adjacent, thickened cells, clearly indicated structural modifications that could be associated with the increased resistance to opening shown by the shatter assessments. This type of modification in structure was similar to those found in the most resistant plants of Jet Neuf.

However, in the recombined lines with greater than average shatter scores, the modification were much more prominent than in the ‘resistant’ Jet Neuf, although there was variation in the extent and position in the DZ. For example, in line 155, increased amounts of vascular tissue were present only near the pedicel end of the pod and did not appear elsewhere along the valve edge. By contrast, in line 139, thickened cells connected the prominent vascular tissue around the whole of the valve edge, forming a complete band of tissue. Many torn cells were found in all the areas containing these structural modifications and also in adjacent unmodified DZ cells.

In many lines the simple, uniform cells dissociated completely throughout the DZ and the vascular traces were small, discrete and widely spaced. The random impact data obtained at SRI showed that these lines shattered quickly and appeared to offer the least resistance to opening amongst the resynthesised lines and in many cases, less resistance than amongst the cultivars.

In other lines, unthickened cells in the DZ did not dissociate. These cells separated along the line of the middle lamella only when pulled apart by hand exposing a smooth surface to the DZ. Occasionally, whole cells pulled out of the DZ and elsewhere, cells fractured leaving exposed jagged edges of the primary cell walls. Failure to separate could only clearly be identified in patches of cells throughout the DZ in line 121, whilst in line 139 separation was incomplete around the main vascular bundle at the pedicel end. In lines129 and 150, both of which were very resistant to opening in the shatter tests, these characters were clearly identified as the sole modification throughout the DZ. However, whilst there was no thickening of the DZ cells and vascular traces were small, widely spaced and discrete in 129, the appearance of vascular tissue varied in 150 from plant to plant (as in the cultivars).

However, SEM examination of the valve surface of fully mature pods of line 142 showed parenchymatous cells throughout the DZ, which failed to separate. In addition, large amounts of prominent vascular traces and associated, thickened cells were situated on the inner valve edge. These anatomical characters were present in all the pods tested. This resulted in torn cells in all parts of the DZ and was associated with consistent, high values for shatter resistance.

Acknowledgements

We thank MAFF for providing financial assistance to support this project (grant nos. NFO306 and NFO506); Zoe Ladbrooke and Helen Baggett for their valuable help in collecting records; Gillian Arnold and Roger White for statistical advice. Margaret Ford and Dr. D. Keith produced the seed of the synthetic lines used in this work.

References

Child, R.D. and Evans, D.E. (1989). Improvement of recoverable yields in oilseed rape (Brassica napus) with growth retardants. In: Production and protection of oilseed rape and other Brassica crops. Aspects of Applied Biology 23, 135-144.

Child, R.D. and Huttly, A.K. (1999). Anatomical variation in the dehiscence zone of oilseed rape pods and its relevance to pod shatter. In: Proceedings of 10^th International Rapeseed Congress.

Child, R.D., Ulvskov, P., Van Onckelen, H.A. and Botterman, J. (1999). Strategies for engineering shatter resistance into oilseed rape pods. In: Proceedings of 10^th International Rapeseed Congress.

Davies, G.C. and Bruce, D.M. (1997). Fracture mechanics of oilseed rape pods. Journal of Materials Science 32, 5895-5899.

Morgan, C.L., Bruce, D.M., Child, R.D., Ladbrooke, Z.L. and Arthur, A.E. Genetic variation for pod shatter resistance among lines of oilseed rape developed from synthetic B. napus (1998). Field Crops Research 58, 153 – 165.

Picart, J.A. and Morgan, D.G. (1984). Pod development in relation to pod shattering. Aspects of Applied Biology 6,101 – 110.