N. Magan, Applied Mycology Group, Biotechnology Centre, Cranfield University Silsoe, Bedford MK45 4DT, UK (e-mail: email@example.com).
Aims: Comparisons were made of the effect of water activity (aw 0·99–0·85), temperature (15 and 25°C) and time (40 days) on growth/production of the trichothecene mycotoxin deoxynivalenol (DON) by Fusarium culmorum and Fusarium graminearum on wheat grain.
Methods and Results: Studies examined colonization of layers of wheat grain for 40 days. Fusarium culmorum grew optimally at 0·98 aw and minimally at 0·90 aw at 15 and 25°C. Colonization by F. graminearum was optimum at 0·99 aw at 25 and 0·98 aw at 15°C. Overall, temperature, aw and their interactions significantly affected growth of both species. Production of DON occurred over a much narrower range (0·995–0·96 aw) than that for growth. Optimum DON was produced at 0·97 and 0·99 aw at 15 and 25°C, respectively, by F. culmorum, and at 0·99 aw and 15°C and 0·98 aw at 25°C for F. graminearum. Statistically, one-, two- and three-way interactions were significant for DON production by both species.
Conclusions: This suggests that the ecological requirements for growth and mycotoxin production by such species differ considerably. The two-dimensional profiles on grain for DON production by these two species have not been examined in detail before.
Significance and Impact of the Study: This type of information is essential for developing climate-based risk models for determining the potential for contamination of cereal grain with this trichothecene mycotoxin. It will also be useful information for monitoring critical control points in prevention of such toxins entering the wheat production chain.
Fusarium ear blight (FEB) causes significant reduction in yield and quality of wheat grain throughout the world. FEB is a preharvest disease, but Fusarium species can grow postharvest if wet grain is not dried efficiently and quickly. In addition to the degradation in grain quality, Fusarium species produce a range of mycotoxins which contaminate the grain (Jennings et al. 2000; Magan et al. 2002; Magan and Olsen 2004). Fusarium culmorum, and more recently, F. graminearum are the most common causes of FEB in the UK and both produce trichothecenes and the latter species also produces zearalenone. Many of these mycotoxins are harmful to both animals and humans, causing a wide range of symptoms of varying severity, and are possible immunosuppressants.
Control Fusarium spp. has relied on the application of fungicides preharvest which are at most 70% effective. However, the timing and application of these applications are critical. For instance, some fungicides are ineffective against FEB and in some cases result in a stimulation of deoxynivalenol (DON) and nivalenol (NIV) production, particularly at suboptimal fungal growth conditions and low fungicide doses (D'Mello et al. 1999; Jennings et al. 2000; Magan et al. 2002; Ramirez et al. 2004). It has been shown that moisture conditions during the critical anthesis stage is crucial in determining infection and mycotoxin production by F. culmorum on wheat during grain ripening (Lacey et al. 1999). Very few studies have determined the effect of key environmental factors such as available water, temperature and time may have on growth and mycotoxin production. Some studies have identified the water activity (aw) range for germination and growth of F. culmorum and other Fusarium species (Magan and Lacey 1984a,b; Sanchis and Magan 2004). Recently, Hope and Magan (2003) reported on the environmental profiles of growth and DON/NIV production by F. culmorum on wheat-based media. However, less detailed information is available on wheat grain. Hooker et al. (2002) have shown that for F. graminearum infection of ripening cereals the environmental parameters are crucial in determining infection and contamination with DON. Detailed information on interacting environmental conditions as they affect F. culmorum and F. graminearum have not often been examined or compared. The objective of this study was to compare the impact of water availability, temperature and time interactions on growth and DON production by an isolate of F. culmorum and F. graminearum on wheat grain with retained germinative capacity.
Materials and methods
Fungal isolates and culture
A representative strain of Fusarium culmorum (98WW4.5FC; Rothamsted Research Culture Collection, Harpenden, Herts., UK) and F. graminearum (Chemotype I, IBST1003) isolated from UK wheat grain, with a known history of trichothecene production (Lacey et al. 1999; Magan et al. 2002) were used. The strains produced DON levels similar to other strains from the UK and other parts of Europe. Cultures were maintained on a 2% milled wheat grain agar (Tech. agar no 3; Oxoid, Basingstoke, UK). Similar studies were also conducted with three additional strains of F. culmorum (IBST1068, IBST1171, IBST1418) and two of F. graminearum (Chemotype I, IBST1004, IBST1005). These are all held in the Applied Mycology Group culture collection. While the amounts of DON produced varied from 0·31 to 19·33 μg g−1 for F. culmorum strains and between 0·95 and 12·65 μg g−1 for F. graminearum strains at 0·995–0·95 aw at 25°C/40 days, the range of environmental conditions over which production occurred were similar for all strains examined.
Winter wheat grain was irradiated (12 kGy gamma irradiation, Hamer 1994) to remove contaminant micro-organisms, but conserve germinative capacity (>75%). Different amounts of water were added to the grain and an adsorption curve prepared to facilitate accurate modifications of the water availability of the grain (Pixton 1982). Grain samples were adjusted in the range 0·995–0·850 aw. The equivalent moisture contents of the aw treatments of 0·995, 0·98, 0·95, 0·90 and 0·85 were 30, 25–26, 22–24, 19–20 and 17–18% for wheat grain respectively. Grain was placed in sterile flasks and inoculated with known amounts of sterile water to obtain the necessary treatments. The flasks were sealed and left for 24 h to equilibrate. Grain was then decanted carefully into 90 mm Petri dishes to obtain a monolayer of wheat grain. The aw of the grain was confirmed using an Aqualab (Decagon Inc., Washington, DC, USA). In all cases the aw levels at both temperatures (15 and 25°C) remained within 0·004 of the desired treatment level.
Inoculation and growth measurements
Twenty-four plates of each aw treatment were inoculated centrally with a 5 μl drop of a 105-ml−1F. culmorum or F. graminearum conidial suspension obtained from a 7-day-old colony. Conidia were obtained by flooding cultures with 5 ml of sterile distilled water containing 0·5% Tween 80 and agitating the colony surface with a sterilized glass rod. Grain replicates of the same treatment aw were enclosed in polyethylene chambers together with 2 × 500 ml of a glycerol/water solution of the treatment aw level to maintain equilibrium relative humidity and incubated at 15 or 25°C for 40 days. Temporal growth measurements were taken throughout the incubation period, by taking two diametric measurements of the colonies at right angles to each other. Growth rates were determined subsequently by linear regression of the radial extension rates (R2 values were between 0·9 and 0·99). Three replicates per treatment were removed after 10, 20, 30 or 40 days and analysed for DON. The experiment was carried out twice with similar results.
Mycotoxin extraction and analyses
The DON analysis was performed using a modified method of Cooney et al. (2001). Each sample was finely ground and mixed well. The sample was extracted by mixing with acetonitrile/methanol (14 : 1; 40 ml) shaken for 2 h and then filtered through Whatman No. 1 filter paper. For analysis a 2-ml aliquot was passed through a cleanup cartridge consisting of a 2-ml syringe (Fisher Ltd, Loughborough) packed with a disc of filter paper (No. 1; Whatman International Ltd, Maidstone), a 5-ml luger of glass wool and 300 mg of alumina/activated carbon (20 : 1, 500 mg). The column was washed with acetonitrile/methanol/water (80 : 5 : 15; 500 μl), and the combined eluent was evaporated (compressed air, 50°C) to dryness and then resuspended in methanol/water (5 : 95; 500 μl).
Quantification of DON was performed using a Luna C18 reverse phase column (100 × 4·6 mm; 5 mm particle size; Phenomenex, Macclesfield, UK) connected to a guard column SecuritryGuard (4 mm × 3 mm) filled with the same stationary phase. Separation was achieved using an isocratic mobile phase of methanol/water (12 : 88, v/v) at 1·5 ml min−1. Eluates (50 μl) were detected using a UV detector set at 220 nm with an attenuation of 0·01 AUFS. The retention time for DON was 13·3 min. Quantification was relative to external standards of 1 to 4 μg ml−1 in methanol/water (5 : 95). The quantification limit was 5 ng g−1.
Statistical treatment of results
The data was analysed using anova (SigmaStat; SPSS Inc., Chicago, IL, USA), with significance values of P < 0·01 used. Excel 97 (Microsoft Reading, UK) was used for determination of growth rates by linear regression.
Effect of water and temperature effects on growth
Figure 1 compares growth rates (Kr) for both species over the range of aw × temperatures levels studied. Temperature generally affected growth, with radial extension rates faster at 25°C than 15°C. The highest Kr of F. culmorum was at 0·995 and 0·98 aw and 25°C, while this was at 0·995 and the same temperature for F. graminearum. For F. culmorum, at 15°C the aw range for optimum growth was wider 0·995–0·96, while for F. graminearum growth decreased by c. 50% at 0·96 aw. There was no mycelial growth observed at <0·90 aw. Statistical analysis showed that aw, temperature and their interactions significantly affected growth (data not shown).
Comparison of DON production in relation to water, temperature and incubation time
Figure 2 shows the surface response curves for DON production at 15 and 25°C over 40-day periods for F. culmorum on wheat grain. The highest DON levels were obtained at 0·995 aw, 25°C after 40 days of incubation. However, very little DON was produced during the first 10 days and as aw was reduced DON production declined sharply, with none produced at <0·95 aw. Although DON production was c. 10-fold lower at 15°C than 25°C there was a more rapid increase in production with maximum amounts being produced over a wide range of aw levels (0·995–0·95). The effects of aw, temperature, incubation time and their interactions were found to be statistically significant (Table 1a).
Table 1. Significance test of experimental factors effect on production of deoxynivalenol by (a) Fusarium culmorum and (b) Fusarium graminearum
(a) F. culmorum
2·3E + 08
1·95E + 08
1·95E + 08
3·46E + 08
1·15E + 08
aw × Temperature
2·04E + 08
aw × Time
3·3E + 08
Temperature × Time
3·18E + 08
1·06E + 08
aw × Temperature × Time
3·25E + 08
2·03E + 09
(b) F. graminearum
aw × Temperature
aw × Time
Temperature × Time
aw × Temperature × Time
1·6E + 08
Figure 3 shows the DON production profiles by F. graminearum. Higher DON concentrations were produced at 25°C than 15°C. However, DON concentrations at 25°C were c. 30% of that produced by F. culmorum under the same conditions. The temporal patterns of production were also different from F. culmorum. At 15°C optimum production was at 0·995–0·98 aw. At 25°C production again began only after 10 days incubation with a peak at 0·98 aw. However, in contrast to F. culmorum, this species produced up to 1 ppm at 0·95 aw after 30–40 days. The aw, temperature and incubation time, and two- and three-way interactions had a statistically significant effect on DON production by F. graminearum (Table 1b). anova of the comparison between the two species shows that there was a statistically significant difference in relation to temperature and aw × temperature (Table 2).
Table 2. Analysis of variance (anova) of the quadratic model for the production of deoxynivalenol by Fusarium culmorum and Fusarium graminearum in relation to water activity (aw) and temperature and their interactions
aw × Temperature
This study has shown the dramatic effect that changes in aw, temperature and time have on growth and DON by F. culmorum and F. graminearum for the first time. Overall, growth of both species only occurred at >0·90 aw. Although DON production by both species was lower at 15 than 25°C the aw range was much wider. In this study we have used the aw and temperature range to simulate those occurring in ripening grain (Magan and Lacey 1985) and harvested grain in a wet year (19–30%; 0·90–0·995 aw). This study has provided parallel information on colonization rates and compared DON production by two important head blight species on wheat grain. Growth and DON production were both optimum at 25°C for both species, although F. graminearum grew faster and produced DON over a slightly wider aw range. Thus, while F. culmorum produces more DON, especially at 25°C, it is worthwhile noting that F. graminearum can grow faster and has recently been shown to be more competitive than F. culmorum (Magan et al. 2003). The profiles obtained suggest that below 0·90 aw no DON is produced by either of these species. Magan and Lacey (1984a, 1984b) showed that the minimum aw for germination was c. 0·88 and for growth was c. 0·89–0·90 aw. The minimum aw for DON production by both these species appears to be more limited at >0·93 aw under optimum temperature conditions.
Previous studies have predominantly examined effects of temperature and time. For example, Versonder et al. (1982) demonstrated that a F. graminearum and the so-called F. roseum (F. culmorum) strains produced DON optimally at 29–30 and 25–26°C, respectively, on cracked moist maize (30% water content = 0·99 aw). Minimum temperatures for production of DON were c. 11°C, dependent on time of incubation for both strains. Their contour maps were very useful but excluded interaction with water availability. Cycling of temperature can have a significant impact on both DON and NIV production. Studies by Ryu and Bullerman (1999) using rice cultures showed that temperature cycling of 15 and 30°C over a 6-week period resulted in the highest biomass. However, steady-state incubation at 25°C for 2 weeks resulted in the highest DON and ZEA production. There was also a correlation between DON and ZEA production, but none between fungal biomass and production of either. This suggests that environmental stress has an important influence on toxin production, often unrelated to total fungal biomass.
In this study colonization was from a point source, thus samples of grain contained mycelium of different ages. This may explain the early detection of DON in some treatments. Recent studies on wheat-based media showed that F. culmorum had differential aw and temperature optima for DON and NIV suggesting that production by this fungus may respond differently to aw × temperature stress (Hope and Magan 2003). The fungus may produce NIV under sub-optimal growth conditions for improving competitiveness. However, while it produces less NIV than DON, the former metabolite is more toxic than the latter.
Misting experiments have demonstrated that Fusarium culmorum infection and trichothecene production was highest during wet periods in the summer (Lacey et al. 1999). Studies by Birzele et al. (2000) suggested that DON was produced by F. culmorum at 17% moisture content (0·80–0·85 aw) in natural wheat grain, depending on the moisture determination method used. These conditions are, at most, marginal for germination of conidia of F. culmorum and most other Fusaria, and under which growth would not normally occur (Magan and Lacey 1984b; Sanchis and Magan 2004). The present study has shown that the range of aw conditions for DON production by both F. culmorum and F. graminearum are much narrower than that for growth. This information should be very useful in developing risk maps in relation to potential conditions under which trichothecenes may be produced by F. culmorum and F. graminearum.
We are grateful to the European Commission, Quality of Life and Management of Living Resources Programme (QOL), Key Action 1 on Food, Nutritional Health, Contract No: QLK1-1999-00996 for supporting this work.