Occurrence of Beauvericin and Enniatins in Wheat Affected by Fusarium avenaceum Head Blight

Abstract

We evaluated Fusarium contamination and the levels of hexadepsipeptide mycotoxins in 13 wheat samples affected by head blight in Finland. Fusarium avenaceum was the dominant species (91%) isolated from all samples, but isolates of F. culmorum (4%), F. tricinctum (3%), and F. poae (2%) also were recovered. Beauvericin (0.64 to 3.5 μg/g) was detected in all 13 samples. Enniatin B (trace to 4.8 μg/g) was detected in 12 samples, enniatin B₁ (trace to 1.9 μg/g) was detected in 8 samples, and enniatin A₁ (trace to 6.9 μg/g) was detected in 10 samples. Ten of 13 strains of F. avenaceum and 2 strains of F. poae and F. tricinctum produced beauvericin in culture on rice (trace to 70, 9.4, and 33 μg/g, respectively). All strains also produced enniatins (trace to 2,700 μg/g). This is the first report on the natural cooccurence of beauvericin and enniatins in wheat infected predominantly by F. avenaceum.

Head blight of small grain is an important cereal disease and may result in the accumulation of toxins in scabby grains. Among Fusarium species causing this disease, F. avenaceum (Fr.) Sacc., the anamorphic state of Gibberella avenacea R. Cook, is one of the most important species worldwide (1, 4). This species can produce significant quantities of toxic secondary metabolites in vitro, including moniliformin (14) and enniatins (5, 19). Fusarium avenaceum isolated from maize can produce beauvericin (12), and we think that beauvericin could be a major cereal contaminant as well. Beauvericin and enniatins are well-known cyclic hexadepsipeptides with a specific cholesterol acyltransferase inhibitor activity (20). Beauvericin is toxic to several human cell lines (13) and can induce apoptosis and DNA fragmentation (13, 17). Furthermore, beauvericin exerted a negative inotropic effect (decrease in cardiac contraction strength) as well as a negative chronotropic effect (decrease in frequency of cardiac spontaneous beating activity) in isolated guinea pig hearts (9), suggesting a potential risk of cardiotoxicity when beavericin is present as a contaminant in grains and foods. Only moniliformin has been reported as a natural contaminant of wheat affected by F. avenaceum head blight (6, 7, 18).

The extent of human, animal, and plant exposure to these cyclodepsipeptides has not been well established. Our objective in this study was to investigate the natural occurrence of beauvericin and enniatins in wheat affected by F. avenaceum head blight. In Finland during 1998, climatic conditions favored the development of Fusarium head blight. We identified the Fusarium species involved in the epidemic, determined the levels of beauvericin and enniatins present, and assessed the ability of the isolated Fusarium strains to produce beauvericin and enniatins.

(An abstract of this work was presented at the Sixth European Fusarium Seminar, Berlin, Germany, 11 to 16 September, 2000.)

MATERIALS AND METHODS

Wheat samples.

Thirteen samples (300 g each) of wheat kernels were randomly collected at harvest time during August and September from head blight-affected wheat fields in four areas (Artjärvi, Vantaa, Orimattila, and Renko) of southern Finland during the 1998 crop season. The samples were transported in sterile plastic bags to the laboratory and stored at 4°C for 3 days.

Mycological analysis.

One hundred infected kernels from each of the 13 samples were placed on the surface of petri plates (five kernels per plate) of Fusarium-selective agar medium containing pentachloronitrobenzene (16). Plates were incubated in the dark at 25°C for 1 week. Fusarium colonies were transferred to plates of potato dextrose agar (800 ml of filtrate from 200 g of peeled, sliced, and autoclaved potatoes; 20 g of dextrose; and 15 g of Bacto agar [Difco, Detroit, Mich.], brought to 1,000 ml with distilled water) and were incubated at 25°C for 10 days under fluorescent- and black-light lamps (2,700 lx; 12-h photoperiod). The identification of 773 colonies of Fusarium species was made according to the criteria and synoptic keys of Nelson et al. (16). Single conidia on carnation leaf agar (16) were used for fermentation studies. To preserve the cultures, mycelia and conidia from wild strains grown on carnation leaf agar were transferred aseptically in 1 ml of sterile 18% glycerol-water and frozen at −75°C. The isolates were deposited in the collection of the Istituto Tossine e Micotossine da Parassiti Vegetali.

In vitro mycotoxin production.

Sixteen representative single-conidium isolates of F. avenaceum (13), F. poae (Peck) Wollenw. (1), F. tricinctum (Corda) Sacc. (1), and F. culmorum (W. G. Smith) Sacc. (1) were cultured on 10 g of autoclaved rice, adjusted to about 45% moisture in a 50-ml plastic tube, and inoculated with 0.5 ml of an aqueous suspension containing approximately 10⁷ conidia/ml. Cultures were incubated at 25°C for 3 weeks in the dark. This procedure was repeated two more times to account for possible variability in toxin production by fungi. Culture material was dried in a forced-draft oven at 60°C for 48 h, finely ground, and stored at 4°C until use. Controls were treated the same way except that they were not inoculated.

Mycotoxin analysis.

Water for the high-pressure liquid chromatography (HPLC) mobile phase was purified in a Milli-Q system (Millipore, Bedford, Mass.). Organic solvents (HPLC grade) were purchased from Merck (Darmstadt, Germany). The standards for beauvericin (catalog no. B7510) and an enniatin mixture (catalog no. E3643) were purchased from Sigma Chemical Co. (St. Louis, Mo.). The enniatin mixture contained enniatin B (19%), enniatin B₁ (54%), enniatin A (3%), and enniatin A₁ (20%), as described by Monti et al. (15). To extract beauvericin and enniatin, 10 g of each sample was ground and homogenized in a Ultraturrax model T25 basic from IKA Werke (Staufen, Germany) for 3 min at 13,500 rpm with 50 ml of methanol (HPLC grade). Samples were filtered through Whatman no. 4 filter paper, and all methanol was removed under reduced pressure. Extracts were resuspended in 3 ml of methanol, prepurified once on a C₁₈ column (500 mg, 3 ml, 40 μm) (Varian, Inc., Palo Alto, Calif.), concentrated to 1 ml, and filtered through an Acrodisk filter (pore size, 0.22 μm) (Millipore, Jonezawa, Japan) before HPLC analysis. Beauvericin and enniatin (B, B₁, and A₁) analyses were performed as described by Monti et al. (15) with minor modifications. HPLC analyses were performed using LC-10AD pumps and a diode array detector from Shimadzu (Kyoto, Japan). A Shiseido Capcell Pak C₁₈ column (250 by 4.6 mm, 5 μm) was used. HPLC conditions included a constant flow at 1.5 ml/min and acetonitrile-water (65:35, vol/vol) as the starting eluent system. The starting ratio was kept constant for 5 min and then linearly modified to 70% acetonitrile in 10 min. After 1 min at 70% acetonitrile, the mobile phase was returned to the starting conditions in 4 min. Beauvericin and enniatins were detected at 205 nm. Mycotoxins were identified by comparing retention times and UV spectra of samples with those of authentic standards. Further confirmation was obtained by coinjecting pure standards with each sample. Mycotoxins were quantified by comparing peak areas from samples to a calibration curve of standards. Chemical structures were confirmed by liquid chromatography-mass spectrometry. A Perkin-Elmer series 200 liquid chromatograph connected to a 785A UV-visible detector was coupled with an API-100 single-quadrupole mass spectrometer (Perkin-Elmer, Sciex Instruments, Ontario, Canada). The detection limits were 20, 1.3, 3.6, and 1.2 ng/g for beauvericin, enniatin A₁, enniatin B₁, and enniatin B, respectively. All analyses were run in triplicate, and the mean values are reported. The calculated standard deviation was always less than 5%.

RESULTS

All 13 wheat samples in this study were infested with Fusarium species at from 5 to 100% of kernels. In some cases the kernels were infected with more than one species. The predominant species was F. avenaceum (91%), which was isolated from all samples. The other Fusarium species encountered, which made up 4% of the recovered species, included F. tricinctum, F. poae, and F. culmorum (Table 1). Although we used a medium selective for Fusarium, other fungi, belonging mostly to the Epicoccum, Alternaria, and Microdochium genera, also were recovered. Beauvericin was detected in all 13 samples. Enniatin B, enniatin B₁ and enniatin A₁ were detected in 12, 8, and 10 samples, respectively. Beauvericin concentrations ranged from 0.64 to 3.5 μg/g. Enniatin B concentrations ranged from trace to 4.8 μg/g, enniatin B₁ concentrations ranged from trace to 1.9 μg/g, and enniatin A₁ concentrations ranged from trace to 6.9 μg/g. Eight samples contained all four toxins. The total cyclodepsipeptide concentration ranged from 1.7 to 17 μg/g.

TABLE 1.

Occurrence of beauvericin and enniatins in wheat head blight caused by Fusarium species in Finland

Samplea	Fusarium spp. (no. of kernels infected)b	Concn (μg/g)c
		Beauvericin	Enniatin B	Enniatin B1	Enniatin A1	Total cyclodepsipeptides
Kem 192	F. avenaceum (10)	2.3	1.6	0.66	2.6	7.1
Kem 195	F. avenaceum (5)	2.9	TRd	1.8	2.8	7.5
Kem 194	F. avenaceum (7)	0.64	NDe	ND	ND	0.64
Kem 201	F. avenaceum (100)	2.6	3.3	ND	TR	5.9
Kem 202	F. avenaceum (100), F. culmorum (3)	3.1	3.9	1.6	4.9	13
Kem 203	F. avenaceum (100)	2.5	TR	ND	ND	2.5
Kem 204	F. avenaceum (94)	3.5	4.5	1.9	6.9	17
Kem 205	F. avenaceum (100)	1.3	3.1	TR	5.5	9.9
Kem 206	F. avenaceum (100), F. tricinctum (22)	0.91	2.9	ND	TR	3.8
Kem 207	F. avenaceum (38)	3.1	TR	ND	ND	3.1
Kem 219/1	F. avenaceum (13), F. poae (12)	1.5	4.8	1.4	TR	7.7
Kem 219/3	F. avenaceum (9)	1.7	TR	TR	TR	1.7
Kem 219/4	F. avenaceum (32), F. culmorum (28)	1.1	4.7	TR	5.3	11

^aKem, Finnish cereal sample from National Veterinary and Food Research Institute. The Kem samples were collected in southern Finland from the following four districts: Artjärvi (Kem 192), Vantaa (Kem 195), Orimattila (Kem 194), and Renko (Kem 201 to Kem 219/4).

^bBased on 100 kernels per sample; kernels can be infected by multiple species.

^cDetection limits for beauvericin, enniatin A₁, enniatin B₁, and enniatin B were 20, 1.3, 3.6, and 1.2 ng/g, respectively.

^dTR, trace.

^eND, not detected.

Of the 16 strains tested for mycotoxin production, 13 were F. avenaceum, 1 was F. tricinctum, 1 was F. poae, and 1 was F. culmorum (Table 2). When the strains were cultured on rice, beauvericin was produced by 10 of 13 strains of F. avenaceum (trace to 70 μg/g.). All 13 strains of F. avenaceum also produced enniatin B (3.4 to 2,700 μg/g), enniatin B₁ (trace to 1,200 μg/g), and enniatin A₁ (trace to 94 μg/g). The F. poae and F. tricinctum strains also produced beauvericin (9.4 and 33 μg/g, respectively), enniatin B (3.4 and 690 μg/g), enniatin B₁ (trace and 1,200 μg/g), and enniatin A₁ (trace and 94 μg/g).

TABLE 2.

Production of beauvericin and enniatins on rice by strains of Fusarium species isolated from wheat head blight in Finland^a

Fusarium sp.	ITEMb strain no.	Source	Concn (μg/g)c
			Beauvericin	Enniatin B	Enniatin B1	Enniatin A1	Total cyclodepsipeptides
F. avenaceum	3397	Kem 201	6.4	9.0	12	TRd	27.4
F. avenaceum	3410	Kem 201	57	1,900	150	6.9	2,114
F. avenaceum	3411	Kem 201	3.6	2,600	1,000	7.8	3,611
F. culmorum	4154	Kem 202	NDe	ND	ND	ND
F. avenaceum	3401	Kem 203	TR	2.6	3.0	TR	5.6
F. avenaceum	3398	Kem 204	1.9	170	74	2.2	248.1
F. avenaceum	3409	Kem 204	70	2,000	220	41	2,331
F. avenaceum	3402	Kem 205	TR	24	13	TR	37
F. avenaceum	3403	Kem 205	3.2	580	26	14	623
F. avenaceum	3404	Kem 205	5.3	450	33	18	506
F. tricinctum	3405	Kem 206	33	690	1,200	94	2,017
F. avenaceum	3408	Kem 206	60	1,200	270	1.8	1,532
F. poae	3399	Kem 219/1	9.4	3.4	TR	TR	12.8
F. avenaceum	3400	Kem 219/1	ND	56	74	TR	130
F. avenaceum	3406	Kem 219/4	ND	860	130	TR	990
F. avenaceum	3407	Kem 219/4	ND	260	180	15	455

^aStrains were grown on rice for 3 weeks at 25°C in the dark.

^bITEM, Istituto Tossine e Micotossine Culture Collection.

^cDetection limits for beauvericin, enniatin A₁, enniatin B₁, and enniatin B were 20, 1.3, 3.6, and 1.2 ng/g, respectively. Data are averages from three cultures of each strain.

^cTR, trace.

^dND, not detected.

DISCUSSION

This report is the first of the natural occurrence of beauvericin and enniatins B, B₁, and A₁ in wheat grain infected by Fusarium. Previous reports indicated that beauvericin is primarily a natural contaminant of maize infected by F. subglutinans (Wollenw. et Reinking) Nelson, Toussoun et Marasas (10) and F. proliferatum (Matsushima) Nirenberg (11). Wheat contamination with these toxins is probably underestimated, since F. avenaceum is a widespread wheat head blight agent and can produce high levels of these toxins. To date, the main concern regarding mycotoxins associated with F. avenaceum in wheat has been moniliformin (6, 7), which can have at least an additive effect with beauvericin and enniatins eventually cooccurring. On the other hand, wheat head blight caused by F. graminearum Schwabe [Gibberella zeae (Schw.) Petch] and F. culmorum should result in little, if any, beauvericin and enniatin contamination, because neither species is known to produce these toxins (12). Field survey reports clearly indicate that the mycotoxins most frequently produced in cereal head blight by F. graminearum and F. culmorum in all European countries are deoxinivalenol, 3-acetyldeoxynivalenol, and zearalenone (1). However, since Fusarium head blight often is caused by a complex of toxigenic species which occur at different levels depending on environmental conditions and geographic locations, toxin analyses should be addressed with respect to the species most frequently isolated.

Our results on the occurrence of Fusarium in wheat kernels in Finland confirm those of several previous surveys in northern Europe (1, 6, 7), including Finland (21), which found that F. avenaceum is one of the dominant species on wheat. Although F. avenaceum is considered by some to be a saprophyte or a weak parasite (2), under favorable conditions, it may be an aggressive wheat pathogen, causing scab (3, 6, 8).

In the present study, most of the strains of F. avenaceum collected from Finnish wheat produced beauvericin and enniatins. These results confirm earlier reports of beauvericin- and enniatin-producing strains of F. avenaceum (12, 19). The level of beauvericin production varied by strain, indicating that this phenotype is polymorphic in field populations of F. avenaceum. Recently, Yli-Mattila et al. (22) showed that F. avenaceum from Finland is polymorphic for randomly amplified polymorphic DNA markers and that the examined population could be divided into five main groups by using this character. Further analyses of toxigenic and genetic variation within and between populations of F. avenaceum, also from different hosts (e.g., potatoes, legumes, and conifers), and pathogenicity tests in particular, will be necessary to resolve their genetic relationship and relative importance in producing symptoms of the disease with relative mycotoxin accumulation.

In this study the type B enniatins were much more common than the type A enniatins in F. avenaceum. However, this relationship is reversed for other F. avenaceum strains, where the type A enniatins predominate (19).

We found that F. poae, F. tricinctum, and F. culmorum can be recovered from wheat kernels in addition to F. avenaceum. Of these species, only F. poae and F. tricinctum synthesize hexadepsipeptides that might accumulate in the colonized wheat kernels. The production of beauvericin by F. poae is consistent with a previous study in which we found that three strains from Poland could produce this toxin (12). This is the first report of the production of beauvericin by F. tricinctum.

In conclusion, our results indicate that beauvericin and enniatins can occur at significant levels in wheat affected by F. avenaceum head blight in Finland and that further study of their spread in cereals and their toxicity is needed. It also will be important to determine if there are synergistic effects with other toxins, e.g., moniliformin and trichothecenes, to better evaluate the toxicological risk to humans and animals.