ABSTRACT

This paper reports the results of studies on the effects of fungal disease on the composition of rapeseed. Samples of seed from fungicide-treated and untreated crops of a range of double-low (OO) and industrial cultivars, from several sites in the UK, were analysed for selected indicators of quality. In OO cultivars infection, mainly light leaf spot, usually decreased oil content and yield and resulted in significant qualitative and quantitative changes in oil fatty acids, particularly by increasing the proportion of [18:2] and [18:3] at the expense of [18:1] acids. Oil from OO plants with severe stem rot contained hexadecatrienoic and erucic acids, which were absent from that of healthy plants. Disease generally decreased the triacylglycerol fraction of the oil but increased the diacylglycerol fraction. The composition of the oil of high erucic acid rape (HEAR) was usually modified as the result of infection, but without an overall uniformity of change. Disease tended to enhance the protein content of OOs but lowered that of HEARs. Fungicide- treated plots tended to have higher oil yields even when disease levels were low, implying that spray treatments affected crop physiology independently of disease. Our results suggest that controlling diseases will be critical if growers are to consistently provide processors with seed of the required quality.

INTRODUCTION

Oilseed rape (Brassica napus L.) produces an oil that is currently used mainly for human consumption but also, on a smaller scale, for industrial applications. There is great potential for the industrial use of rape oil to expand, and special cultivars are being bred to produce oils specifically suited to a particular application. Such crops are required to produce seed containing as high a content as possible of oil with a fatty acid composition that is optimal for the intended end-use, while containing the lowest possible levels of contaminants removal of which increase processing costs. However, there are economic constraints on the inputs that can be made to the crop, and among inputs, fungicide applications are often targets for saving. The fungal diseases that develop more freely in the absence of fungicides can have profound effects on the yield and physiology, and possibly also on the quality of the crop. This study investigated how fungal disease affected rapeseed quality factors that contribute to the yield of important end-products (seed oil content, oil fatty acid composition, group separation, molecular species pattern, and seed protein content) and those factors that devalue the seed by increasing processing costs (chlorophyll content; acid value and peroxide value).

MATERIALS AND METHODS

Seed samples of various double-low and industrial winter and spring rape cultivars were taken from experiments at IACR-Rothamsted and from variety trials plots co-ordinated by the National Institute of Agricultural Botany at Cambridge (NIAB) and the Scottish Agricultural College at Aberdeen (SAC) in 1994-5, and at Aberdeen, Brampton, Bridgets, Edinburgh and Morley in 1995-6. Plots were either: treated with fungicides to control disease (F), or inoculated with straw taken from a previous oilseed rape crop or left untreated to encourage disease (I or U).

In 1994-95 three winter-rape experiments were done at IACR-Rothamsted: two conventional, double-low cultivars (Capricorn and Falcon) were grown in F and I plots; the high erucic acid rape (HEAR) cultivar Askari was grown in F and I plots; a plot of the double-low cultivar Apex was used as a source of sclerotinia-infected and uninfected plants, collected by hand immediately before combine harvest. One winter-rape experiment was done at Aberdeen in which 18 double-low cultivars were grown in F and U plots.

Experiments in 1995-96 had a greater number of varieties, including both winter and spring types, and were done at more sites. At Aberdeen, fourteen winter type double-low cultivars (Bristol, Corporal, Ecuador, Envol, Express, Falcon, Gazelle, Inca, Jazz, Lizard, Mandarin, Nickel, Rapier, and Synergy, a varietal association) and two HEAR cultivars (Askari and Martina) were grown in F and U plots. Five cultivars (Synergy, Bristol, Nickel, Askari and Martina) were grown in F and U plots at three other sites (Brampton, Bridgets and Morley). In addition, the cultivar Express was grown at the Bridgets site. Three double-low spring type cultivars (Starlight, Mars and Global) were grown alongside the HEAR cultivar Industry and the high oleic acid cultivar Hyola 401 in corresponding F and U plots at Aberdeen and Edinburgh.

Seed samples were analysed at the University of Nottingham for selected indicators of quality. All seed samples were cleaned to remove chaff and weed seeds before analysis. Samples for thousand seed weight measurement were dried at 105°C for 24h. Those destined for other analyses were dried at 45°C for 24h before storage. The following quality parameters were measured for each seed sample: thousand seed weight (TSW); oils content; fatty acid composition; molecular species composition; protein content; chlorophyll content, acid value and peroxide value. The methods used to make these analyses are given elsewhere (Doughty et al, 1998).

RESULTS

Disease incidence

In general, fungicide-treated (F) plots were never free of infection, but epidemics were generally later, and less severe in those plots than in untreated (U) plots, and much less severe than in inoculated (I) plots. Inoculation encouraged disease and induced earlier epidemics that culminated in more severe pod infection. But the effect was selective, favouring light leaf and pod spot and dark leaf and pod spot in particular. For example at Rothamsted the incidence of both light leaf spot and dark pod spot was greater in I plots, but dark pod spot in I plots was about four times that in F plots. At Aberdeen, in both seasons, light leaf spot infection was nearly always lower in fungicide treated plots for all cultivars. In 1995-6 the incidence of light leaf spot at Brampton, Bridgets and Morley was generally low in both untreated and sprayed plots. At Brampton and Bridgets stem canker was the predominant disease found, but at Bridgets disease incidence was similar on untreated and treated plots. Disease incidence in the spring rape experiments was extremely low in both sprayed and unsprayed plots.

Seed dry matter, thousand seed weight (TSW), oil content and protein content.

Disease tended to decrease dry matter yield of winter cultivars: at Aberdeen yield was reduced by about 13% and 23% in 1994-5 and 1995-6 respectively when averaged over all cultivars; at Rothamsted the yields of, Capricorn, Falcon and Askari (HEAR) were decreased by an average of 25, 9 and 23% respectively.

The effect of disease on TSW was less clear cut. Although TSW tended to be reduced in disease susceptible winter cultivars at Aberdeen, the response was more variable for other cultivars and other sites.

In winter rape experiments, disease appeared to affect seed oil content, but there were differences between double-low and HEAR cultivars. In 1994-5 and 1995-6, seed from U or I plots of winter and spring double-low cultivars (including the hybrid varietal association Synergy) usually had lower oil contents (up to a difference of about 5%) than corresponding F plots. In some experiments, the combination of lower dry matter yield and lower oil content led to substantial losses in oil yield between I or U plots, up to 30% in some of the susceptible varieties. In contrast, seed from HEAR cultivars in 1995-6 had no consistent pattern of difference in oil content between F and U plots. For example, for the four sites, cultivar Martina had a similar average oil content in F and U plots, whereas Askari had a higher average oil content in U plots.

The differences in oil content between U and F plots among spring cultivars was less consistent than for the winter types. With one exception (Mars), all the F treated cultivars at Aberdeen had slightly higher oil contents than the U treatments. More variable responses were obtained from the Edinburgh trials.

Comparison of disease incidence with oil content suggests that where light leaf spot was predominant and severe, it could be a major factor in the loss in oil content suffered by a double-low cultivars if fungicides were withheld. In contrast, sclerotinia stem rot had relatively little impact on oil content: severely-infected plants of the winter cultivar Apex (Rothamsted, 1994-5) and the spring cultivar Hyola 401 (Aberdeen, 1995-6) produced seed with an oil content similar to that of seed from uninfected plants.

Fungicide treated plots tended to have higher oil yields even when disease levels were low, implying that spray treatments could affect crop growth independent of disease.

Seed from heavily-infected plots of double-low cultivars usually had higher protein contents, but seed from heavily-infected plots of HEAR cultivars had lower protein contents than that from corresponding fungicide-treated plots. In double-low cultivars the decrease in oil content tended to equal the increase in protein content suggesting that the effect of disease on the two components were interrelated but opposite.

Oil fatty acid composition, group separation and molecular species pattern.

Severe disease tended to cause changes in the fatty acid composition of the oil. In 1994-5, the greatest changes occurred among winter double-low cultivars that had suffered the greatest reductions in oil content as a result of severe infection, but in 1995-6 this pattern was not repeated. Thus, like oil content, the extent of the changes depended on the type of variety and the local disease pattern. The greatest changes in fatty acid composition among double-lows in 1994-5 were associated with light leaf spot susceptible varieties, where this disease was severe. Oil from more heavily-diseased plots generally had up to about 6% less oleic acid [18:1] but more polyunsaturated fatty acids (up to about 4% linoleic acid [18:2] and up to about 1.4% alpha-linolenic acid [18:3]) than that from F plots. No changes in the relative proportions of the saturated fatty acids palmitic acid [16:0] and stearic acid [18:0] were detected. The same overall pattern occurred in 1995-6, but the differences between F and U plots were not as great, possibly because disease incidence tended to be lower in that season. Some of the double-low cultivars grown at SAC-Aberdeen and NIAB sites had a greater proportion of [18:1] in the oil from U plots than from sprayed plots in 1995-6. There was little difference in the fatty acid profile in the oils from U and F plots from Brampton, Bridgets and Morley in 1995-6, but incidence of light leaf spot was low at these sites, although some stem canker was recorded.

In 1994-5, the oil from winter HEAR cultivars also showed a shift away from [18:1] in I plots, but towards eicosenoic [20:1] and erucic acids [22:1], rather than [18:2] and [18:3]. Winter HEAR cultivars grown in 1995-6 had changes in fatty acid composition between sprayed and unsprayed plots, but without a consistent pattern of change. At Aberdeen, both Askari and Martina had a greater erucic acid [22:1] content in U plots than F plots, but at NIAB-Brampton the proportion of erucic acid [22:1] was greater (about 10%) in the oil of U plots only for Askari and it was less in the oil from U plots of Martina. There was similar variation at other sites. Most of the change in the proportion of erucic acid in the oil from U plots of HEARs was accounted for by a commensurate change in the proportion of [18:1]. It was not possible to relate differences in the fatty acid composition in oils from HEAR cultivars in F and U plots with differences in disease patterns across sites.

Oil from spring cultivars grown in 1995-6 also showed changes in fatty acid composition as a result of withholding fungicides. At Aberdeen, the oil from U plots had about 3% less [18:1] than that from F plots, whereas at Edinburgh, there was no consistent pattern in the proportion of this fatty acid. The only spring HEAR tested (Industry) had no consistent pattern of change in [22:1] at the two sites.

Infection by Sclerotinia sclerotiorum appeared to cause different changes in oil content compared to infection by other pathogens. At Rothamsted oil from severely infected plants of the winter double-low cultivar Apex differed little from that from uninfected plants in 18-carbon fatty acids, but it contained about 2.3% erucic acid [22:1] and about 0.7% hexadecatrienoic acid [16:3], which were absent from oil from uninfected plants. Levels of eicosenoic acid [20:1] also increased in the oil of infected plants. The spring cultivar Hyola 401 grown at SAC-Aberdeen in 1995-6 responded to sclerotinia infection with a similar shift in fatty acids in its oil away from [18:1] towards [18:2] and [18:3], but it did not produce erucic acid in response to infection.

Fungicide treatment generally increased the triacylglycerol fraction (TAG) of the oil but decreased the diacylglycerol (DAG) fraction of the oil. In the two winter cultivars, Envol and Mandarin, which were examined in detail, there appeared to be a trend in that U plots produced oil containing greater amounts of TAG molecular species rich in polyunsaturated fatty acids, and lesser amounts of those richer in monounsaturates. The changes in the fatty acid composition of the TAG fraction was reflected those in the oil.

Contaminants of the oil (chlorophyll, free fatty acids and peroxides)

Severe disease tended to affect chlorophyll content, but the direction of the change varied between sites and seasons. Seed from heavily-infected plots showed no consistent changes in acid value or peroxide value.

DISCUSSION AND CONCLUSIONS

Our results indicate that diseases can be significant determinants of the oil quality of seed via their effect on the yield of a particular, desirable fatty acid. The loss in oil content in seed from unprotected crops may endanger the saleability of the crop. The increase in protein content in seed from unprotected plots does not offset the loss in oil content, because the meal is much less valuable than the oil. The implications of the effects of disease on seed quality detected in this study are complex. In particular the importance of the changes in fatty acid composition in unprotected plots depends on the use intended for the oil. The general pattern of a shift towards longer-chain unsaturated fatty acids in heavily-infected plots suggests that some industrial cultivars will be more vulnerable to losses in quality when unprotected than others, depending on the required fatty acid spectrum. The increase in the proportion of erucic acid [22:1] in the oil of diseased HEAR plots in some experiments represented an improvement in quality, although the loss in seed dry matter yield offset this ostensible advantage of withholding fungicides. In contrast, the appearance of erucic acid [22:1] in the oil from stem rot-infected crops would be particularly undesirable for food-use. The effects of diseases cannot be generalised, because changes in both oil content and fatty acid composition differ according to the predominant pathogen.

Considered together, our results suggest that controlling diseases will be important if growers are to provide processors consistently with seed of the required quality from crops. They also suggest that unless it is possible to incorporate durable varietal resistance into future, high-value industrial rape cultivars, fungicides are likely to play an important role in their agronomy.

ACKNOWLEDGEMENTS

This work was supported by a grant from the UK Home Grown Cereals Association. IACR-Rothamsted is supported by a grant in aid from the UK Biotechnology and Biological Sciences Research Council. We are grateful to S. Ibrahim and J. Hopkinson, C. Lewis, S. Jobes, A. Todd and A. Sheridan from the three organisation for help in carrying out the experiments and analysing the samples.

REFERENCES

1. Doughty, K.J., Norton, G., Landon, G., West, G., McCartney, H.A., Booth, E.J., Walker, K.C., Kightley, S.P.J. & Thompson, J.E. (1998) Effects of disease on seed quality parameter of oilseed rape grown for industrial uses. H.G.C.A. Project Report OS33, 44p.