Abstract

Introduction

The extreme sensitivity of fully mature pods to opening during harvesting, results in seed dispersal before harvest ('pod shatter’) of at least 10% of the yield each year, with up to 50% losses in some seasons (Child and Evans, 1989). Improved seed recovery would increase yield levels per unit area of production and reduce the need to use herbicides to control volunteers. Little variation in shatter susceptibility is found amongst current cultivars of Brassica napus. L. Therefore, conventional breeding procedures using current commercial lines are unlikely to result in significant improvement in developmental synchrony and shatter resistance.

The oilseed rape pod (silique or siliqua) consists of two valves (carpels) whose walls are separated by a replum containing the main vascular tissue of the pod. A layer of two to three parenchymatous cells between the valve edges and the replum forms the separation or dehiscence-zone (DZ) and can be identified because adjacent replum tissue contains vascular elements and, from about four weeks after anthesis, adjacent cells, which form the valve edge, become heavily thickened.

A transitory climacteric of ethylene is detectable at about five weeks after anthesis, prior to senescence, much of which is attributable to the seeds (Meakin and Roberts, 1990). Many studies have supported the concept that indole-3-acetic acid (IAA) desensitises abscission zones to ethylene, so that when IAA levels decrease abscission zone cells become responsive to ethylene. The interaction of IAA with ethylene may determine the onset of cell separation in the DZ.

The decline in climacteric ethylene in pod tissues is accompanied by a rise in cellulase (β-1,4-glucanase) (Meakin and Roberts, 1990). Biochemical analysis of maturing pods has shown a correlation between cellulase activity and DZ cell separation, but it is clear other hydrolases are involved this process. Temporal and spatial correlation of the activity of a polygalacturonase (PG) has now been confirmed in DZ tissue of oilseed rape (Petersen et al., 1996). A significant role for PG in middle lamella breakdown in cells of the DZ in oilseed rape pods is apparent.

Separation of the cells in the DZ takes place about seven weeks after anthesis, just before moisture loss is complete. The pods open as a result of external force supplied by contact with other pods, racemes or harvesting machinery, which severs the vascular connections. Halting or slowing down the degradation of the cell walls may be expected to prevent or reduce the amount of cell separation and increase the force necessary to open the pod. The conclusions of the study of the developmental and biosynthetic processes that take place prior to and during cell separation in the DZ and the identification of target processes for genetic manipulation are outlined in this paper.

Ethylene and auxin as targets for manipulation by genetic engineering.

Experimental approach.

Production of uniform plant material. Plants of the spring-sown cultivar Fido were grown in a glasshouse. Flowers at the base of the terminal and next three racemes were labelled at anthesis and hand-pollinated. Pods were removed at the labelled positions for analysis at known intervals until maturity, which took place 50-60 d after anthesis (DAA). Other pods were dissected to separate tissue strips approximately 2 mm wide which comprised the DZ together with adjacent tissue, from the remaining pod wall (PW) tissue and the seeds. This material was frozen in liquid N₂ stored at –80^oC and used to determine hormone and enzyme levels.

Induction of parthenocarpy. Treatments were developed that allowed parthenocarpic (seedless) pod growth, thereby allowing cell separation to be studied in the absence of seed-produced ethylene. Flowers were effeminated at anthesis and a droplet of 50% ethanol containing 1 mg ml^-1 gibberellic acid (GA) and 6-benzylaminopurine (BAP) was placed in the receptacle. This treatment induced the pericarp development to about two-thirds of the size of seeded fruits.

Growth regulation treatments. Ethylene levels throughout the climacteric were reduced by approximately 50%, following spray-treatment with aminoethoxyvynileglycine (AVG) at 500 mg l^-1. AVG is known to be an inhibitor of the precursor of ethylene, 1-aminocyclopropane-1-carboxylic acid (ACC). The synthetic auxin 2-methyl-4-chlorophenoxyacetic acid (4-CPA) is known to exibit physiological action similar to the endogenous auxin indole-3-acetic acid (IAA). It has also been shown to delay senescence and plant maturation in oilseed rape. In a separate experiment, immediately prior to the onset of senscence (35 DAA), plants were sprayed overall with an aqueous solution containing 500mg l^-1 4-CPA.

Hormone and enzyme analysis. Ethylene was measured throughout development using whole, freshly-harvested pods which were incubated in 15cm³ tubes for 1 h. 1.0 cm³ sample of gas was injected into a Pye Unicam 204 gas chromatograph incorporating a hydrogen flame ionization detector. The following methods of analysis are described by Chauvaux et al., (1997): levels of free and conjugated levels ACC; the extraction and measurement of free and conjugated IAA; glucanase activity, which was measured viscometrically.

Assessment of susceptibility of pods to shatter. Shatter resistance of fully matiure pods was determined by a mechanical test which measured the force required to open the pod along the DZ. The procedure has been described by Kadkol et al., (1984). A clamped pod was loaded as a cantilever and the displacement and force necessary to initiate and propagate an opening in the DZ was measured. From this, the energy that was required to fully open the pod was calculated.

Effects of growth regulator treatments on hormone levels and susceptibility to shatter.

Manipulation of ethylene production by the pod tissues. AVG applied during the pre-senescence climacteric reduced ACC levels and ethylene production by the seeds, but did not affect subsequent levels in the PW (Child et al., 1998a). The production of В-1,4-glucanase and the separation of the DZ cells was delayed for 3-4 d by AVG, but the force required to open fully mature pods was unaltered. In parthenocarpic pods, ethylene was produced later than in seeded pods. Although cell separation in parthenocarpic pods took place as in seeded pods, it was delayed by 3-4 d. No significant differences in the energy (mJ x 10^-5) required to open fully mature pods were recorded (seeded pods. 23.8; parthenocarpic pods 28.1; sed 8.81 (18df).

Effect of the auxin mimic 4-CPA. In seeded pods, 4-CPA delayed В-1,4-glucanase activity and cell separation in the DZ by 10 d (Chauvaux et al., 1997). Approximately 70% more energy was needed to separate the valves of fully mature, 4-CPA-treated pods (Table 1). These results suggested that the activity of hydrolytic enzymes involved in cell separation might be regulated by auxin activity. Comparison of parthenocarpic pods with seeded pods pointed to the seeds as the source of IAA. Levels of IAA in the DZ of parthenocarpic pods were low over the entire sampling period, while cell separation, which was delayed by about 4 d, only took place after the increase in ethylene. These results indicated that a low level of auxins in the DZ is necessary for dehiscence.

Table 1. Cantilever bending tests with fully mature pods (8% moisture) of oilseed rape cultivar Fido (**, * =significant differences from unsprayed at P < 5% or1%, respectively).

Conclusions: Target processes for possible genetic manipulation. These results suggest that it is likely that reduced ethylene biosynthesis, whether engineered by chemical manipulation of development or possibly by molecular means, may not be sufficient to increase resistance pod shatter. An alternative strategy for reducing cell separation and, consequently, susceptibility to shatter may be offered by the concept of changing sensitivity to ethylene. However, mature pods from 4-CPA-treated plants required more energy to open them than those from untreated plants. This suggests that an engineering strategy that prolongs IAA activity in the pods and specifically in the DZ, will increase resistance to shatter.

Polygalacturonase (PG) activity in the DZ of Brassica napus.

Anatomical and biochemical charactertisation.

Slices of pod tissue of the cultivar Fido containing the DZ were fixed in a mixture of 4% formaldehyde and 5% gluteraldehyde and post-fixed in 1% osmium tetroxide before being embedded in epoxy resin. The process of cell separation in the DZ was studied using ultrathin sections which were examined using a transmission electron microscope (TEM). The primary cell wall and middle lamella remained intact whilst the pod was still photosynthesising. After pod senescence had begun, dissolution of the cell wall was visible and by the time senescence was complete, total separation had taken place. This was clearly attributable to the breakdown of the middle lamella. A molecular down-shift assay was carried out with polygalacturonate as the substrate. Extracts of wall-bound protein were prepared from DZ cells of pods in which senescence was well under way (approximately 36DAA). The results indicated that wall-bound endo-PG was responsible for the depolymerisation of the substrate (Petersen et al., 1996).

Characterisation of DZ-specific PG.

Seven distinct partial cDNA’s, similar in sequence to previously described PGs, were amplified from DNA derived from the pod wall, dehiscence zone and leaves by polymerase chain reaction (PCR). Northern analysis showed that one clone, PG35-8, was expressed at low levels in the DZ until senescence when levels increased greatly. In contrast, no PG35-8 related RNA was detected in the pod wall. The data suggested that there are temporal and spatial correlations between the breakdown of the DZ tissue and the pattern of synthesis of PG35-8 which indicate a role for this PG in cell separation in the DZ. PG35-8 was used to isolate five cDNA clones from a rape DZ cDNA library. Restriction enzyme analysis and partial sequencing revealed that they were derived from four homologous transcripts which are probably allelic forms of a single gene. One full length clone, RDPG1, was completely sequenced. The predicted protein of RDPG1 showed highest identity with PG from apple with an identitity of 52%. The identification and characterisation of the full length clone and a genomic clone with 5` regulatory (promoter) region has been completed (Sander et al., 1996).

Activity of RDPG1 in the DZ during senescence.

The RDPG1 was amplified by PCR and the product used to immunise rabbits from which, serum containing antibody, was collected. Pod material collected as above but fixed in 25 formaldehyde and 0.5% gluteraldehyde was embedded in epoxy resin. Ultrathin sections were incubated first in the RDPG1 antibody and then in a secondary antibody containing 10-nm gold particles (Child et al., 1998b). The endo-PG was detected in the cytoplasm of DZ cells whilst the pod wall was still green. The middle lamella of the DZ cells broke down when senescence was almost complete and PG was detected between the separated cells. In pods treated with 4-CPA as previously described, PG was detected in DZ cell cytoplasm for 10 d longer than in untreated pods. Middle lamella breakdown was incomplete in the DZ cells of 4-CPA-treated plants.

Approaches for genetic engineering for reduced susceptibility to pod shatter.

The endo-PG35-8 promoter has been used to direct expression of antisense PG constructs as well as for driving the expression of heterologous genes in the late stages of DZ and pod development to control auxin activity and DZ maturation. Transformed plants are being assessed for expression of the constructs and results will be published in due course.

Conclusions

1. Reduction in ethylene levels in the pod slightly delayed maturation but did not affect the susceptibility of the DZ to cell breakdown or to shatter’

2. DZ cell wall breakdown and separation was incomplete in pods in which sustained auxin activity was mimicked by 4-CPA treatment.

3. Correlation has been obtained between the activity of an endo-acting isoform of polygalacturonase (PG35-8) and the breakdown of the middle lamella of DZ cells.

4. Sustained auxin activity in senescing pods is a target for modification by genetic engineering.

Acknowledgements.

The authors wish to acknowledge and thank: at IACR-Long Ashton: Karen John and Helen Baggett; at DIAS, Copenhagen: Bernhard Borkhardt, Lilli Sander; at UIA, Antwerp: Els Prinsen, Nancy Chauvaux;and at Plant Genetic Systems, NV: Pascale Redig, Guy Vancanneyt. This study is supported by grants from the European Commission (AIR CT93 0879; and FAIR 96 PL3072).

References

Chauvaux, N., Child, R., John, K., Ulvskov, P., Borkhardt, B., Prinsen, E., and Van Onckelen, H.A. (1997). The role of auxin in cell separation in the dehiscence zone of oilseed rape pods. Journal of Experimental Botany, 48, No.312, pp.1423 - 1429.

Child, R.D., Chauvaux, N., John, K., Ulvskov, P. and Van Onckelen, H.A. Ethylene biosynthesis in oilseed rape pods in relation to pod shatter (1998). Journal of Experimental Botany, 49, No.322, pp.829 - 838.

Child, R.D. and Evans, D, E. (1989). Improvement of recoverable yields in oilseed rape (Brassica napus) with growth retardants. In: Aspects of Biology, 23: 135-143.

Child, R.D., Dal Degan, F., Baggett, H. and Ulvskov, P. (1998). Polygalacturonase activity and cell separation in the dehiscence zone of oilseed rape pods. In: Abstracts 8th International Cell Wall Meeting, Norwich, UK, 1-5 September. 8.38.

Kadkol, G.P., MacMillan, R.H., Burrow, R.P. and Halloran, G.M. (1984). Evaluation of Brassica genotypes for resistance to shatter. 1. Development of a laboratory test. Euphytica 33: 63-73.

Meakin, P.J. and Roberts, J.A. (1990). Dehiscence of fruit of oilseed rape (Brassica napus L.). II. The role of cell wall degrading enzymes and ethylene. Journal of Experimental Botany 41, No.220, 1003-1011.

Petersen, M., Sander, L., Child, R., Van Onckelen, H., Ulvskov, P. and Borkhardt, B. (1996). Isolation and characterisation of a pod dehiscence zone-specific polygalacturonase from Brassica napus. Plant Molecular Biology 31: 517-527.

Sander, L., Botterman, J., Ulvskov, P. and Borkhardt, B, (1996). Nucleotide sequence of a gene encoding a pod dehiscence zone specific polygalacturonase (Accession No. X98373) from Brassica napus. Plant Gene Register 96-056. Plant Physiology. 111:1354.