Impact of Transgenic Bt Maize Residues on the Mycotoxigenic ... - ENVIS

Published online May 31, 2006

Impact of Transgenic Bt Maize Residues on the Mycotoxigenic Plant Pathogen Fusarium graminearum and the Biocontrol Agent Trichoderma atroviride Andreas Naef, Thierry Zesiger, and Genevie`ve De´fago* (Baumgarte and Tebbe, 2005; Sims and Holden, 1996; Zwahlen et al., 2003) but some of it may resist microbial degradation by binding to clay minerals and humic acid fractions, and retain its insecticidal activity (Koskella and Stotzky, 1997; Venkateswerlu and Stotzky, 1992). Recent studies reported effects of tissue from Bt transgenic maize (Flores et al., 2005) and Bt transgenic rice (Wu et al., 2004) on biological activities in the soil and changes in the microbial population were associated with decomposing Bt cotton leaves (Donegan et al., 1995). The observed effects could be caused by altered chemical composition of plant tissue as a result of the Bt gene construct insertion. Saxena and Stotzky (2001) found a higher lignin content in 10 Bt maize hybrids, representing three different transformation events, than in their respective non-Bt isolines. An increased lignin content decreases the rate of crop residue decomposition (Hopkins et al., 2001; Parton et al., 1996) and thus may influence the residue-associated microbial population. One suitable approach in risk assessment of transgenic crops is the assessment of a potential impact on microbial indicators with special relevance to the particular agricultural ecosystem (Kowalchuk et al., 2003). The plant pathogen Fusarium graminearum Schwabe is such a microorganism which has not yet been assessed in association with Bt maize cultivation. The fungus aggressively colonizes crop residues particularly from maize (Yi et al., 2002) which then serve as major inoculum source for diseases like Fusarium ear rot of maize and Fusarium head blight of wheat (Sutton, 1982). Food and feed from infected crops are contaminated with different mycotoxins, which is a threat to human and animal health (Bennett and Klich, 2003; Miller and Trenholm, 1994). An increasing incidence of Fusarium head blight of wheat has been associated with the precrop maize in reduced tillage systems (Dill-Macky and Jones, 2000; Miller et al., 1998), suggesting that maize residues above the soil surface are particularly suitable for the survival of F. graminearum. Since Bt maize can be grown several years in succession without fallow, and each year deposits about 8 Mg dry residue per ha for grain maize and 2 Mg per ha for silage maize (Zscheischler et al., 1990), it is important to investigate whether the Bt transformation affects the saprophytic growth of this deleterious Fusarium species on maize residues. Changes in the composition of maize residues could also affect the balance between deleterious and beneficial microorganisms such as antagonists to plant pathogens. Trichoderma spp. are aggressive fungal colonizers

Reproduced from Journal of Environmental Quality. Published by ASA, CSSA, and SSSA. All copyrights reserved.

ABSTRACT Transformation of maize with genes encoding for insecticidal crystal (Cry) proteins from Bacillus thuringiensis (Bt) could have an impact on the saprophytic survival of plant pathogens and their antagonists on crop residues. We assessed potential effects on the mycotoxin deoxynivalenol (DON)-producing wheat and maize pathogen Fusarium graminearum and on the biocontrol agent Trichoderma atroviride. Purified Cry1Ab protein caused no growth inhibition of these fungi on agar plates. Cry1Ab concentrations above levels common in Bt maize tissue stimulated the growth of F. graminearum. The fungi were also grown on g-radiation-sterilized leaf tissue of four Bt maize hybrids and their non transgenic isolines collected at maize maturity on a field trial in 2002 and 2003. Both fungi degraded the Cry1Ab protein in Bt maize tissue. Fungal biomass quantification with microsatellitebased polymerase chain reaction (PCR) assays revealed differential fungal growth on leaf tissue of different maize varieties but no consistent difference between corresponding Bt and non-Bt hybrids. Generally, year of maize tissue collection had a greater impact on biomass production than cultivar or Bt transformation. The mycotoxin DON levels observed in maize tissue experiments corresponded with patterns in F. graminearum biomass, indicating that Bt transformation has no impact on DON production. In addition to bioassays, maize leaf tissue was analyzed with a mass spectrometer–based electronic nose, generating fingerprints of volatile organic compounds. Chemical fingerprints of corresponding Bt and non-Bt leaf tissues differed only for those hybrid pairs that caused differential fungal biomass production in the bioassays. Our results suggest that Cry1Ab protein in maize residues has no direct effect on F. graminearum and T. atroviride but some corresponding Bt/non-Bt maize hybrids differ more in composition than Cry protein content alone, which can affect the saprophytic growth of fungi on crop residues.

T

HE GLOBAL AREA cultivated with transgenic plants grew rapidly during recent years (James, 2005). One of the dominant transgenic crops is maize (Zea mays L.) with genes encoding for insecticidal crystal proteins from Bacillus thuringiensis Berliner (Bt), being cultivated on 11.3 million hectares worldwide in 2005 (James, 2005). Most Bt maize varieties express the Cry1Ab protein which protects the plant against lepidopteran insect pests, particularly the European corn borer Ostrinia nubilalis Hu¨bner. Cry1Ab protein from transgenic maize is released to the soil ecosystem through root exudates (Saxena et al., 1999, 2004) and decomposition of residues left after harvest (Zwahlen et al., 2003). Much of the Cry protein is quickly decomposed in soil

A. Naef and G. De´fago, Plant Pathology, Institute of Integrative Biology, ETH Zurich, 8092 Zurich, Switzerland. T. Zesiger, SMart Nose SA, 2074 Marin-Epagnier, Switzerland. Received 2 Sept. 2005. *Corresponding author ([email protected]). Published in J. Environ. Qual. 35:1001 – 1009 (2006). Technical Reports: Ecological Risk Assessment doi:10.2134/jeq2005.0334 ª ASA, CSSA, SSSA 677 S. Segoe Rd., Madison, WI 53711 USA

Abbreviations: Bt, Bacillus thuringiensis; DON, deoxynivalenol; ELISA, enzyme-linked immunosorbent assay; PCR, polymerase chain reaction; SNA, synthetic nutrient-poor agar.

1001


1002

J. ENVIRON. QUAL., VOL. 35, JULY – AUGUST 2006

of residues of maize and other crops throughout the decomposition process (Broder and Wagner, 1988) and have been shown to antagonize Fusarium on wheat and black oat straw (Fernandez, 1992). No studies have been reported about non-target effect assessment of Bt transformed crop residues on these important antagonists of plant pathogens. In this study, we used purified Cry1Ab protein and leaf residues of transgenic Bt maize hybrids to assess nontarget effects of the Cry protein itself and of maize Bt transformation on well investigated model fungi. We assessed the growth and the mycotoxin production of the maize and wheat pathogen F. graminearum strain GZ3639 and the growth of the biocontrol agent Trichoderma atroviride Karsten strain P1. Quantitative microsatellite-based PCR assays which are not affected by plant tissue decomposition (Naef et al., 2006) were used to monitor the growth of these fungi in flasks with leaf tissue of four different Bt/non-Bt maize hybrid pairs representing two different transformation events. Variation due to changing environmental conditions was taken into consideration by using maize tissue from a field trial in two different years. Fingerprints of volatile organic compounds from maize leaf residues were generated with an electronic nose to elucidate differences in the chemical composition other than Cry protein content. Furthermore, we tested whether these fungi are able to degrade the Cry1Ab protein in maize leaf residues. MATERIALS AND METHODS Fungal Strains and Culture Conditions Fusarium graminearum GZ3639 (NRRL 29169) is a wheat and maize pathogenic strain (Harris et al., 1999) isolated from wheat in the U.S. Midwest. GZ3639 was stored on 1.5% malt extract agar slants at 38C [15 g malt extract (Oxoid, Hampshire, United Kingdom), 12 g agar (Oxoid)]. For agar plug inoculation of agar plate assays, GZ3639 was grown for 5 d on 1.5% malt extract agar and on synthetic nutrient-poor agar [SNA: 1 g KH2PO4, 1 g KNO3, 0.5 g MgSO47H2O, 0.5 g KCl, 0.2 g glucose, 0.2 g saccharose, 12 g agar (Oxoid)] at 248C in darkness. For spore inoculation of experiments with maize tissue, it was grown for 14 d on SNA under the same conditions. Trichoderma atroviride strain P1 (ATCC 74058) is an isolate from wood chips and was classified for a long time as T. harzianum based on morphological features, but recent genome sequence analysis has shown that it is more closely related with T. atroviride (Kullnig et al., 2001). Strain P1 possesses well investigated biocontrol activity against fungal plant pathogens (Harman and Bjo¨rkman, 1998; Hjeljord et al., 2001) and has been used in competition assays with Fusarium graminearum strain GZ3639 on maize tissue (Lutz et al., 2003; Naef et al., 2006). P1 was stored as spore suspension in 20% (v/v) glycerol at 2208C. For agar plug inoculation, P1 was grown for 5 d on 1.5% malt extract agar and on SNA at 248C in darkness. For spores inoculation, it was grown for 14 d on 1/5 PDA [4.8 g potato dextrose broth (Difco, Detroit, MI), 12 g agar (Oxoid), pH 6.5] under the same conditions.

Agar Plate Assays with Cry1Ab Protein The Cry1Ab protoxin (130 kDa) from B. thuringiensis subsp. kurstaki HD-1 was expressed as a single gene product in Escherichia coli and purified as previously described (Masson

et al., 1989). To obtain the activated Cry1Ab toxin (65 kDa), the protoxin was trypsinized, and high performance liquid chromatography (HPLC) purified (Pusztai-Carey et al., 1994). Sterile Cry1Ab toxin solutions (250 mL of 0, 0.1, 1, and 10 mg mL21) were evenly spread over the surface of 25 mL malt extract agar or synthetic nutrient-poor agar (formulae see above) in Petri dishes. Then, these agar plates were incubated for 48 h at 88C to obtain even Cry1Ab concentrations of 0, 1, 10, and 100 mg mL21 thorough the agar medium. An agar plug from an actively growing culture of T. atroviride P1 or F. graminearum GZ3639 was placed inverted in the middle of each plate and incubated at 248C in the dark. Radial mycelium growth of T. atroviride and F. graminearum was measured on two orthogonal diameters after 3 and 5 d, respectively. Mean growth per day from three experiments with three replicate plates for T. atroviride and five replicate plates for F. graminearum were analyzed for each fungus and medium separately with a two way ANOVA on trial means including the factors experiment and Cry1Ab concentration [GLM procedure in Systat (Systat, 2000)]. In case of a significant effect of Cry1Ab concentration (P , 0.05), means were compared with Fisher’s LSD test.

Maize Cultivars and Crop Residue Preparation Maize cultivars used in this study were the transgenic Bt hybrids, Valmont, X0920 RT, Novelis, TXP 138, and their nearisogenic non-transgenic commercial hybrids Prelude (Syngenta Seeds, Basel, Switzerland), Benicia (Pioneer Hi-Bred International, Des Moines, IA), Nobilis (Euralis, Lescar, France), and DKC 3420 (Dekalb c/o Monsanto, St. Louis, MO), respectively. Valmont carries the transformation event Bt176 (Syngenta Seeds) causing expression of native Cry1Ab protein in pollen and green tissue only (AGBIOS, 2005), whereas X0920 RT, Novelis, and TXP 138 contain the transformation event MON810 (Monsanto) causing expression of a truncated version of Cry1Ab in the whole plant (AGBIOS, 2005). For easier comprehension, we use the term isolines for corresponding Bt/non-Bt hybrids and call the Bt hybrids by the name of their corresponding non-transgenic hybrid followed by the suffix Bt in this study. Maize hybrids were planted on a sandy loam soil in adjacent strip plots of 0.5 ha area in an experimental field in the Rhine valley in Germany. Bt/non-Bt isolines of Prelude and Benicia were grown in 2002 and 2003, isolines of Nobilis were grown in 2002, and isolines of DKC 3420 were grown in 2003. At maturity stage, the fifth leaf from the top was collected from 50 randomly selected maize plants across the variety plot. The leaves were air-dried and ground in two steps using an A11 analytical mill (IKA, Staufen, Germany) and a ZM1 centrifuge mill (Retsch, Haan, Germany) with a ring sieve with slot openings of 0.1 mm. Powder from the 50 leaves was pooled, mixed thoroughly, and sterilized in 140-mL polystyrene screw cap containers in an industrial cobalt chamber with g-radiation at a dose of 25 kGray. The sterile leaf powder was stored hermetically sealed at 2208C until it was used as substrate in flask experiments, mimicking maize crop residues.

Cry1Ab Protein Quantification in Maize Tissue Content of Cry1Ab protein in the maize leaf powder from the different varieties was measured before and after g-irradiation with a commercial enzyme-linked immunosorbent assay (ELISA) kit (EnviroLogix, Portland, ME) with a limit of quantification of 0.25 mg g21 for maize leaves. Cry1Ab protein was extracted from 30 mg air-dried leaf powder with 1.5 mL extraction buffer. The extracts were 1:50 diluted to meet the

NAEF ET AL.: IMPACT OF TRANSGENIC Bt MAIZE RESIDUES ON FUNGI

range of calibrators. Three replicates were measured for each sample. Means were compared with Fisher’s LSD test (Systat).


Mycotoxin Deoxynivalenol (DON) Quantification The DON levels were quantified in 50 mL of the aqueous extracts from flasks inoculated with F. graminearum GZ3639 (see below), using a commercial ELISA kit (R-Biopharm, Darmstadt, Germany). If necessary, the extracts were diluted to fall into the range of calibrators and measured again. The test detects DON down to 0.2 ng mg21.

Polymerase Chain Reaction Quantification of Fungal Biomass Biomass of F. graminearum strain GZ3639 and T. atroviride strain P1 in experiments with maize leaf tissue was determined with competitive PCR on microsatellite loci with length polymorphism as described by Naef et al. (2006). This method is insensitive to variation in DNA extraction and in PCR amplification which bias quantification of fungi on decomposing plant material with conventional PCR assays (Naef et al., 2006). Briefly, 30-mg subsamples of freeze dried and homogenized experimental repeats (see below) were spiked with 2.5 mg of freeze dried mycelium of a closely related fungus. DNA in this reference mycelium works as internal standard during DNA extraction with the DNeasy Plant mini kit (Qiagen, Hilden, Germany) and as competing template during PCR amplification of a microsatellite fragment. Polymerase chain reaction was performed with primers of the Fusarium-specific microsatellite MS-Fg103 (59-GGTATCCGTACAACCCGATG-39 and 59-TTCTTTGATTTGGACCGAGG-39) and with primers of the Trichoderma-specific microsatellite MS-Ta4 (59-ATCTGGCACTGCTTGGTAGG-39 and 59-TCGATCGCCTTCGTATTAGG-39), respectively. Using fluorescently labeled forward primers, the differently sized PCR products of trial and reference strain were separated and quantified with a CEQ 2000XL capillary sequencer (Beckman-Coulter, Fullerton, CA). Since the PCR product ratio between trial and reference strain is correlated to their biomass ratio in the sample (Naef et al., 2006), the biomass of the trial strain can be calculated from the PCR product ratio and the known biomass of reference strain. In each PCR assay, mycelium mixtures of strains were included for calibration.

Flask Experiments with Leaf Tissue of Bt and Non-Bt Maize Hybrids Three experiments were conducted with Bt and non-Bt maize leaf tissue collected on the field trial in 2002 and three experiments were conducted with Bt and non-Bt maize leaf tissue collected on the field trial in 2003. For each experiment, spores of T. atroviride P1 and F. graminearum GZ3639 were harvested with sterile water from agar plates, passed through sterile glass wool to remove mycelium, and quantified using a Thoma-Kammer. Suspensions were made to contain 10 000 conidia mL21 for P1 or 1000 macroconidia mL21 for GZ3639. For each maize variety, 15 autoclaved 25-mL Erlenmeyer flasks were filled with a half gram of sterilized maize leaf powder. Six of them were inoculated with 5 mL of P1 conidia suspension, six with 5 mL GZ3639 macroconidia suspension, and three were rewetted with 5 mL sterile water. After an initial incubation for 24 h on a shaker at 60 rpm at 248C to ensure spore germination, all samples were incubated at 158C at 100% humidity in the dark. Three of six inoculated replicates were taken for analysis after 7 d of incubation. The remaining replicates and the sterile control samples were taken

1003

after 14 d. All samples were frozen at 2808C and freeze dried. Then, the samples were homogenized in 8-mL plastic tubes with glass balls in a M300 mixer mill (Retsch, Haan, Germany) for 5 min at maximum oscillation frequency. Levels of Cry1Ab protein and biomass of GZ3639 and P1 were quantified in subsamples of 30 mg as described above. An additional step was included for 14-d samples with GZ3639. Before freezing, the flasks were filled with 10 mL of sterile distilled water and shaken for 2 h at 90 rpm. After the solid parts were allowed to settle for 30 min, mycotoxin DON levels were quantified in the supernatant as described above. Trial means of Cry1Ab protein content in non-inoculated and inoculated flasks after 14 d were analyzed for each variety and collection year separately with a factorial ANOVA with experiment and treatment as factors (GLM procedure in Systat). In case of a significant treatment effect (P # 0.05), means were compared with Fisher’s LSD test. Trial means of T. atroviride biomass, F. graminearum biomass and accumulated DON production in flasks with tissue of Bt/non-Bt isolines were compared for each time point and collection year of maize tissue separately with paired t tests (Systat). Trial means of Benicia and Prelude isolines from both years were further analyzed with a linear model including the factors maize hybrid, Bt transformation, year, experiment nested within years, and the interaction maize hybrid 3 Bt transformation (Tables 5 and 7, GLM procedure in Systat).

Volatile Organic Compound Fingerprints of Maize Tissue A mass spectrometry–based electronic nose (Smart Nose, Marin-Epagnier, Switzerland) was used to generate fingerprints of volatile organic compounds (Pillonel et al., 2003) from field-collected maize tissue before and after g-irradiation. Aliquots of 100 mg of homogenized maize leaf powder were filled into 10-mL vials and closed by a magnetic closure with a silicone/PTFE septum. The Smart Nose system is composed of an autosampler with oven, a headspace injector, and a quadrupole mass spectrometer. Operating conditions were as follow: 12-min incubation at 1208C with strong agitation; injection of 2.5 mL headspace with syringe temperature of 1108C and injector temperature of 1708C; purge gas nitrogen with a flow of 150 mL min21; 2-min syringe purge time; ionization at 60 eV; mass spectrometer scan over a mass range of 10 to 160 amu. Three replicates were measured for each sample and analyses occurred in randomized order. Ion mass intensity data were analyzed by principle component analysis and discriminant function analysis with the software supplied by Smart Nose.

RESULTS Agar Plate Assays with Cry1Ab Protein On malt extract agar, Cry1Ab protein concentrations of 10 and 100 mg mL21 significantly stimulated radial mycelium growth of F. graminearum GZ3639 by 5 and 12%, respectively, compared to medium without Cry protein (Table 1). No significant growth effect was observed for 1 mg Cry1Ab mL21. On synthetic nutrientpoor agar (SNA), only a Cry1Ab protein concentration of 100 mg g21 caused a significant growth enhancement of GZ3639 by 8% (Table 1). The radial mycelium growth of T. atroviride P1 was not affected by Cry1Ab protein at the tested concentrations on SNA and on malt extract agar (Table 1).

1004


Table 1. Influence of Cry1Ab protein on the growth of F. graminearum strain GZ3639 and T. atroviride strain P1 on agar plates at 24°C. Radial mycelium growth‡ F. graminearum GZ3639 Cry1Ab concentration†

SNA§

T. atroviride P1

1.5% malt agar


21

mg mL 0 1 10 100

mm d 3.03 6 0.03 a 3.07 6 0.03 a 3.15 6 0.03 ab 3.28 6 0.05 b

5.31 6 0.12 a 5.35 6 0.11 a 5.55 6 0.12 b 5.96 6 0.10 c

SNA

1.5% malt agar

6 0.22 a 6 0.19 a 6 0.30 a 6 0.29 a

10.73 6 0.11 a 10.69 6 0.15 a 10.68 6 0.12 a 10.63 6 0.10 a

21

8.42 8.31 8.42 8.52

† Cry1Ab protoxin was expressed in E. coli, trypsinized, and purified with high performance liquid chromatography (HPLC). ‡ Values are means of three experiments with three agar plates for T. atroviride and five agar plates for F. graminearum. Values in the same column followed by different letters are significantly different according to Fisher’s LSD test (P # 0.05). § Synthetic nutrient-poor agar.

Flask Experiments with Leaf Tissue of Bt and Non-Bt Maize Hybrids The fungi were also grown in spore-inoculated flask experiments with g-radiation-sterilized maize leaf tissue of four different Bt maize hybrids and their nontransgenic isolines. The maize leaves were collected in a field trial at maize maturity in two different years. Cry1Ab protein content varied significantly between different Bt maize hybrids and between collection years for Benicia Bt but not for Prelude Bt (Table 2). No Cry1Ab protein was detected in leaf tissue of nontransgenic isolines. The sterilization of maize tissue with g-rays had no effect on the Cry1Ab protein content measured with ELISA (data not shown). Incubation of Bt maize leaf tissue with T. atroviride P1 for 14 d significantly reduced the Cry1Ab protein content between 23.5 and 76.7% compared to sterile incubation (Table 3) which had no effect on the Cry1Ab protein content. The decline of Cry1Ab protein content in Benicia Bt and Prelude Bt was greater in leaf tissue collected in 2003 than in tissue from 2002. This was also true for the incubation with F. graminearum strain GZ3639 which was a less efficient Cry1Ab protein degrader than T. atroviride P1 under the given conditions. Nevertheless, the presence of GZ3639 caused a significant loss of Cry1Ab protein between 29.6 and 41.1% for all maize leaf tissues from 2003 and for tissue of Nobilis Bt from 2002 (Table 3). The small losses in leaf tissue of Benicia Bt and Prelude Bt from 2002 missed significance at the 0.05 probability level.

Fungal biomass and mycotoxin DON production of F. graminearum and biomass of T. atroviride were measured in flask experiments with irradiated maize tissue after 7 and/or 14 d of incubation. No fungal biomass and no DON were detected in sterile reference treatments. For F. graminearum GZ3639, comparisons between leaf tissues of isolines showed significantly lower biomass production on the Bt transgenic lines of Nobilis from 2002 and of Benicia from 2003, but no significant isoline difference on the other collected maize leaf tissues (Table 4). GZ3639 biomass after 7 d was higher on leaf tissues collected in 2002 than on those collected in 2003 (Table 4). For Benicia and Prelude hybrids, this collection year difference was stronger (P , 0.001) than the difference between hybrid genotypes (P 5 0.026) and between Bt and non-Bt hybrids (P 5 0.100) (Table 5). No year effect was observed after 14 d (P 5 0.550) because GZ3639 biomass slightly decreased on leaf tissue from 2002 and further increased on tissue from 2003 (Table 4). Comparisons of mycotoxin DON production by F. graminearum GZ3639 within 14 d showed significantly Table 3. Cry1Ab protein degradation by F. graminearum or T. atroviride in g-radiation sterilized leaf tissue of transgenic Bt maize varieties collected from a field trial in 2002 and 2003. Cry1Ab protein content‡ Varieties

Cry1Ab protein content‡ Maize variety

2002

Benicia Bt

Benicia Bt Prelude Bt Nobilis Bt DKC 3420 Bt

8.69 6 0.13 b 4.36 6 0.13 d 9.93 6 0.67 a

Nobilis Bt DKC 3420 Bt

2003 21

mg g

6.35 6 0.36 c 4.87 6 0.19 d 6.62 6 0.50 c

† Leaves of non-Bt maize varieties contained no Cry1Ab protein. ‡ Cry1Ab protein concentrations were determined by quantitative enzymelinked immunosorbent assay (ELISA). Values are means of three subsamples from a homogenized powder of 50 air-dried leaves each from a different plant. Values followed by the same letter are not significantly different according to Fisher’s LSD test (P # 0.05).

2002

2003 21

Prelude Bt

Table 2. Concentration of Cry1Ab protein in air-dried leaves of transgenic Bt maize† collected at maturity on a field trial in year 2002 and in year 2003.

Inoculated fungus† no fungus F. graminearum T. atroviride no fungus F. graminearum T. atroviride no fungus F. graminearum T. atroviride no fungus F. graminearum T. atroviride

9.12 8.31 6.30 3.91 3.69 2.99 9.25 6.11 3.90

6 6 6 6 6 6 6 6 6

mg g 0.50 a 6.15 6 0.09 a 0.72 ab 4.33 6 0.40 b 0.33 b 2.75 6 0.45 c 0.11 a 4.59 6 0.16 a 0.42 ab 3.22 6 0.42 b 0.35 b 1.11 6 0.25 c 0.41 a 0.29 b 0.25 c 6.62 6 0.47 a 3.73 6 0.57 b 2.86 6 0.25 b

† Flasks with leaf tissue were inoculated with conidia of either F. graminearum strain GZ3639 or T. atroviride strain P1 and incubated at 15°C for 14 d. ‡ Concentration of Cry1Ab protein per dry weight sample was determined with quantitative enzyme-linked immunosorbent assay (ELISA). Values for Benicia Bt and Prelude Bt are means (6standard error) of three experiments with three flasks per treatment and values for Nobilis Bt and DKC 3420 Bt are means of two experiments with three flasks per treatment. Values of the same variety in the same column followed by different letters are significantly different according to Fisher’s LSD test (P # 0.05).

1005


Table 4. F. graminearum strain GZ3639 biomass and deoxynivalenol (DON) production in flasks† with g-radiation sterilized leaf tissue of transgenic Bt maize varieties and their non-transgenic isolines collected from a field trial in 2002 and 2003. 2002

2003

Biomass‡ Varieties

DON§

7d

14 d


21

Benicia Benicia Bt Prelude Prelude Bt Nobilis Nobilis Bt DKC 3420 DKC 3420 Bt

Biomass‡

14 d 21

mg (mg sample) 0.116 6 0.014 0.112 6 0.005 0.109 6 0.019 0.110 6 0.008 0.147 6 0.024 0.120 6 0.012 0.130 6 0.019 0.115 6 0.006 0.162 6 0.013 0.146 6 0.007 0.120 6 0.022¶ 0.104 6 0.002*

DON§

7d

ng (mg sample) 2.118 6 0.261 1.579 6 0.354* 1.190 6 0.196 1.234 6 0.259 2.077 6 0.466 0.931 6 0.309*

14 d

14 d

21

21

mg (mg sample) 0.088 6 0.012 0.121 6 0.003 0.061 6 0.013¶ 0.102 6 0.002* 0.094 6 0.016 0.128 6 0.013 0.086 6 0.021 0.120 6 0.008 0.069 6 0.009 0.073 6 0.014

ng (mg sample) 0.549 6 0.075 0.520 6 0.075 0.446 6 0.146 0.479 6 0.136

0.089 6 0.012 0.082 6 0.007

0.559 6 0.079 0.573 6 0.069

* Values differ from the values above them at P # 0.05 according to paired t tests on experimental means. † Flasks with leaf tissue were inoculated with macroconidia and incubated at 15°C. Values for Benicia and Prelude isolines are means (6standard error) of three experiments with three flasks per treatment and values for Nobilis and DKC 3420 isolines are means of two experiments with three flasks per treatment. ‡ Biomass was measured with competitive microsatellite polymerase chain reaction (PCR) (Naef et al., 2006). § Accumulated DON production was measured with enzyme-linked immunosorbent assay (ELISA). ¶ Values differ from the values above them at P 5 0.09 for Nobilis after 7 d and at P 5 0.15 for Benicia 2003 after 7 d according to paired t tests on experimental means.

reduced DON levels for Nobilis Bt and Benicia Bt from 2002 compared to their non-transgenic lines (Table 4) but no significant difference between leaf tissues of other Bt/non-Bt hybrid pairs. For Benicia and Prelude hybrids, DON production by GZ3639 was significantly higher on leaf tissue from 2002 than on leaf tissue from 2003 (P , 0.001) (Table 5). Maize hybrid genotype also had a significant impact (P 5 0.003), since DON production was higher on leaf tissue of Benicia hybrids than on that of Prelude hybrids in both years (Table 4). The detection of 50% lower DON levels in one experiment compared to other experiments caused a significant effect of experiment repetition on DON levels (P 5 0.002). Since the trends between varieties in the respective experiment were the same as in the other experiments, the effect most likely resulted from a deficient ELISA kit and does not indicate a true impact of experiment repetition. Biomass reduction of T. atroviride through Bt transformation was observed on leaf tissue of Benicia Bt from both years (Table 6). A relatively strong trend in the opposite direction was observed on tissue of DKC 3420 isolines after 7 d (P 5 0.20) which was no longer present after 14 d. No differences were seen between isoline tissues of Nobilis and Prelude, in contrast to the above mentioned differences between Nobilis isolines for F. graminearum. Biomass of T. atroviride P1 after 7 d was

higher on leaf tissue from 2002 than from 2003 (Table 6), causing again a significant collection year effect on tissue of Benicia and Prelude hybrids (P , 0.001), which was stronger than the effect of maize hybrid (P 5 0.254) or Bt transformation (P 5 0.039) (Table 7). The effect of Bt transformation after 7 d (P 5 0.039, Table 7) should not be taken as indication for a general impact of Bt transformation because reduced T. atroviride biomass on the Bt transgenic line was only observed for Benicia but not for Prelude resulting in a strong maize hybrid 3 Bt transformation interaction (P 5 0.050) (Table 7). During the subsequent incubation of 7 d, biomass of T. atroviride decreased on all maize tissues but more on those from 2002 than on those from 2003 (Table 6), resulting in a less pronounced year effect after 14 d (P 5 0.063) (Table 7).

Volatile Organic Compound Fingerprints of Maize Tissue The chemical composition of maize tissue which served as substrate in the flask experiments was assessed with the mass spectrometer–based electronic nose Smart Nose, generating fingerprints of volatile organic compounds. Principle component analysis (PCA) of chemical fingerprints showed strong similarity between maize tissues before and after sterilization with g-rays (data

Table 5. Analysis of variance on the biomass of F. graminearum strain GZ3639 after 7 and 14 d and on the deoxynivalenol (DON) concentration after 14 d in flasks with g-radiation sterilized leaf tissue of two transgenic Bt maize varieties and their non-transgenic isolines (Benicia Bt/non-Bt and Prelude Bt/non-Bt) from two years (2002 and 2003). Biomass

DON concentration

7d

14 d

14 d

Source of variation

df

F ratio

P

F ratio

P

F ratio

P

Maize hybrid Bt transformation Maize hybrid 3 Bt transformation Year Experiment (years)† Error

1 1 1 1 4 15

6.104 3.080 0.072 26.277 2.793

0.026 0.100 0.792 ,0.001 0.065

2.446 2.023 0.089 0.374 0.995

0.139 0.175 0.769 0.550 0.441

11.678 1.406 2.417 99.038 6.993

0.004 0.254 0.141 ,0.001 0.002

† Three experiments with three flasks per treatment were performed for each year. Means of experiments were treated as replicates.

1006


Table 6. T. atroviride strain P1 biomass production in flasks† with g-radiation sterilized leaf tissue of transgenic Bt maize varieties and their non-transgenic isolines collected from a field trial in 2002 and 2003. Biomass‡ 2002 Varieties

2003

7d

14 d


mg (mg sample) 0.194 6 0.018 0.175 6 0.024¶ 0.186 6 0.022 0.192 6 0.021 0.194 6 0.012 0.181 6 0.012

Benicia Benicia Bt Prelude Prelude Bt Nobilis Nobilis Bt DKC 3420 DKC 3420 Bt

0.106 6 0.009 0.085 6 0.007¶ 0.077 6 0.008 0.080 6 0.008 0.089 6 0.005 0.092 6 0.008

7d

14 d

0.142 6 0.020 0.103 6 0.005* 0.138 6 0.020 0.131 6 0.015

0.106 6 0.005 0.084 6 0.005* 0.101 6 0.004 0.101 6 0.002

0.111 6 0.027 0.135 6 0.009¶

0.089 6 0.003 0.096 6 0.007

21

* Values differ from the values above them at P # 0.05 according to paired t tests on experimental means. † Flasks with leaf tissue were inoculated with conidia and incubated at 15°C. Values for Benicia and Prelude isolines are means (6standard error) of three experiments with three flasks per treatment and values for Nobilis and DKC 3420 isolines are means of two experiments with three flasks per treatment. ‡ Biomass was measured with a competitive microsatellite polymerase chain reaction (PCR) (Naef et al., 2006). ¶ Values differ from the values above them at P 5 0.19 for Benicia 2002 after 7 d, at P 5 0.18 for Benicia 2002 after 14 d, and at P 5 0.20 for DKC 3420 after 7 d according to paired t tests on experimental means.

not shown). Hence, the data from analyses before and after sterilization were grouped together for further statistical calculations to increase the number of measurements and thus the reliability of results. Principle component analysis on these grouped fingerprints was used to select fourteen non-redundant variables (ion masses) for a discriminant function analysis (DFA). This DFA could discriminate between tissues from the two collection years and between tissues of the Bt and nonBt isolines of Benicia and Nobilis (Fig. 1A). No discrimination could be achieved between tissues of isolines of Prelude and DKC 3420 (Fig. 1B). Discriminant function analyses with different variable selection confirmed the described discrimination pattern.

DISCUSSION In this study, we evaluated potential non-target effects of Bt transformation of maize on two agronomically important fungal colonizers of maize residues (Broder and Wagner, 1988). We used well investigated model strains of a deleterious and a beneficial fungal species, namely the maize and wheat pathogen Fusarium graminearum strain GZ3639 and the biocontrol agent Trichoderma atroviride strain P1. Agar plate tests with different Cry1Ab protein concentrations showed no effect on the mycelium growth of T. atroviride and growth stimulation Table 7. Analysis of variance on the biomass of T. atroviride strain P1 after 7 and 14 d in flasks with g-radiation sterilized leaf tissue of two transgenic Bt maize varieties and their nontransgenic isolines (Benicia Bt/non-Bt and Prelude Bt/non-Bt) from two years (2002 and 2003). 7d Source of variation Maize hybrid Bt transformation Maize hybrid 3 Bt transformation Year Experiment (years)† Error

14 d

df

F ratio

F ratio

P

1 1 1

1.462 5.090 4.495

0.245 0.039 0.050

P

0.699 2.795 3.473

0.416 0.115 0.082

1 4 15

76.918 2.432

,0.001 0.093

4.035 0.019

0.063 0.999

† Three experiments with three flasks per treatment were performed for each year. Means of experiments were treated as replicates.

of F. graminearum at high Cry1Ab concentrations (Table 1). This stimulating effect was stronger on the nutrient rich malt extract medium than on nutrient-poor medium, suggesting that it rather results from changes of physical medium properties, for example, a reduction of surface tension (observed by A. Naef), than from the use of Cry1Ab protein as additional substrate. The absence of a fungicidal effect of purified Cry1Ab protein is in agreement with Koskella and Stotzky’s (2002) observations in tests with other selected fungi. In flask experiments, F. graminearum strain GZ3639 and T. atroviride strain P1 were grown on sterilized leaf tissue of four different field-grown Bt maize hybrids and their non-transgenic counterparts for 7 and 14 d at 158C (first day at 248C). This experimental setting has been chosen because F. graminearum survives particularly well in no-tillage systems on maize residues above the soil surface (Yi et al., 2002). Preliminary experiments have further shown very slow fungal growth and no conidia germination at lower temperatures and no increase of fungal biomass after 14 d. We used sterile powder of maize leaf tissue to obtain controlled conditions. Test trials with nonsterile maize residue have shown that the natural microflora varies among residue pieces, which affected the growth of our target fungi. A principle component analysis of volatile organic compound (VOC) fingerprints has shown that the sterilization with g-rays has little impact on the chemical composition of maize tissue in contrast to autoclavation which is known to break down compounds in plant tissue (Schenk et al., 1991). Moreover, g-irradiation caused no loss of Cry1Ab protein as measured with an enzyme immunoassay. However, the g-irradiation could have caused a damage of a part of the protein without being detectable with this assay. The range and variability of Cry1Ab protein contents in leaf tissue from different collection years (Table 2) was in agreement with another field study with the varieties Novelis (5Nobilis Bt) and Valmont (5Prelude Bt) (Nguyen Thu, 2004). Both fungi were able to degrade Cry1Ab protein in Bt maize tissue, but year of tissue collection influenced the Cry protein decline. Interestingly, the loss of Cry protein



Fig. 1. Chemical fingerprint discrimination of leaf tissues of transgenic and non-transgenic maize isolines of Benicia (Bt, .; non-Bt, s) and Nobilis (Bt, x; non-Bt, e) (A). No discrimination of isolines of Prelude (Bt, E; non-Bt, D) and DKC 3420 (Bt, j; non-Bt, h) (B). Discriminant function analysis was performed on volatile organic compound data from the head space of leaf tissues collected in the field at maize maturity stage in 2002 and 2003.

was lower in leaf tissue from the year 2002 which caused faster growth of the fungi. One plausible explanation is a lower substrate quality of tissue from 2002 causing an increased use of the additionally available Cry1Ab protein. Generally, T. atroviride degraded the Cry1Ab protein more efficiently than F. graminearum. Incubation with T. atroviride for 14 d caused a Cry1Ab protein loss in maize tissue of at least 23%. Thus, application of T. atroviride on Bt maize residues in the control of deleterious Fusarium could support the Cry1Ab protein degradation. However, Cry1Ab protein degraders are naturally present in microbial communities of agricultural soils causing Cry1Ab protein recoveries of less than one percent of the initial concentration in fieldoverwintered maize leaves (Baumgarte and Tebbe, 2005; Zwahlen et al., 2003). As we used pulverized leaf tissue in the flask experiments, our Cry1Ab protein degradation rates should not be extrapolated to field conditions. A part of the Cry1Ab protein may escape

1007

degradation in lignin rich parts such as maize stems and roots (Baumgarte and Tebbe, 2005). The growth of T. atroviride and F. graminearum was reduced on leaf tissue of Benicia Bt from year 2003 compared to tissue of non-transgenic Benicia, but this effect was weaker on leaf tissue from year 2002 (Tables 4 and 6). Significant growth reduction compared to the non-Bt isoline was also found on Nobilis Bt but only for F. graminearum and not for T. atroviride. No fungal biomass difference was found in isoline comparisons of DKC 3420 and Prelude, except for a somewhat higher biomass of T. atroviride on DKC 3420 Bt after 7 d (P 5 0.2). Regarding DON production in the flask experiments with maize leaf tissue, the result of isoline comparisons widely corresponded with patterns of F. graminearum biomass. Consequently, lower DON contents were observed on Nobilis Bt and Benicia Bt compared to their corresponding non-transgenic lines (Table 4). However, we found no consistent effect of Bt transformation on F. graminearum GZ3639 or T. atroviride P1. The observed differences within Bt/non-Bt pairs were in the same order of magnitude as the differences between different non-Bt hybrids in the respective collection year. Nevertheless, significantly reduced fungal growth on leaf tissue of those Bt lines with relatively high Cry1Ab protein content (Benicia Bt and Nobilis Bt) could be taken as indicator for growth inhibition through Cry protein. This interpretation is in conflict with the results of our agar plate assays with purified Cry1Ab protein (Table 1). Under consideration of a 90% weight loss during air-drying, the initial Cry1Ab protein concentrations in the maize leaf tissue corresponded to concentrations which showed no effect on F. graminearum growth in the agar assays (#1 mg Cry1Ab per g wet substrate). The hypothesis of fungal growth inhibition by Cry1Ab protein is further abolished by smaller biomass reduction but higher Cry1Ab content for Benicia Bt tissue from 2003 compared to that from 2002 (Table 2). In contrast to Bt transformation, collection year of the maize tissue had a consistent impact on fungal growth and mycotoxin DON production (Tables 4–7). The development of both fungi was delayed on all leaf tissues collected in 2003 compared to tissues collected in 2002. The DON contents also were lower on leaf tissues from 2003 corresponding to slower initial growth of F. graminearum and consequently less biomass over the total period of investigation. The equalization of the biomass differences between years at the second sampling time resulted from a growth stop of F. graminearum and a biomass decline of T. atroviride. This biomass decline is caused by autolysis which occurs in aged hyphal cultures of T. atroviride as a result of hydrolase activity (Brunner et al., 2003; White et al., 2002). We collected leaf tissue at ear maturity in both years and stored it in hermetically sealed containers at 2208C after sterilization. We assumed that under these conditions a different storage time for maize tissue from the two years has no effect on tissue composition and that the observed differences resulted from different environmental conditions. The summer of 2003 was extraordinarily hot and dry causing


1008


a 40% lower dry matter yield than in year 2002 (V. Heitz, unpublished data). Obviously, the plants were under drought-stress which affects their chemical composition (Cross et al., 1994; Maiti and Wesche-Ebeling, 1998) and thus the fungal growth on their leaf tissue. We used a mass spectrometer–based electronic nose (Smart Nose) to generate fingerprints of volatile organic compounds of the maize tissues which served as substrate in flask experiments. These fingerprints were analyzed with multivariate statistics, a technique which had been successfully used for chemical differentiation of dairy products (Pillonel et al., 2003). The fingerprint analysis revealed no systematic difference between chemical compositions of Bt and non-Bt maize leaf tissues but fingerprint discrimination was possible between tissues from different years and between Bt and non-Bt tissues of Nobilis and Benicia (Fig. 1). For Benicia, the discrimination between isolines was more pronounced in 2003 than in 2002. Hence, chemical fingerprint differences were in agreement with significant fungal growth differences on sterile maize leaf tissue, suggesting that chemical differences other than Cry1Ab protein content caused the observed fungal growth differences. Since DKC 3420 Bt with the same transformation event (MON810) as Benicia Bt and Nobilis Bt showed no fingerprint discrimination from its non-Bt counterpart, the composition differences result rather from less than complete isogenic hybrid pairs than from non-target effects of the Bt transgene insertion. Our observation of chemical differences between Bt/non-Bt maize isolines is in line with other reports of differential isoline chemical composition. The differences reported in C to N ratio (Escher et al., 2000; Flores et al., 2005; Hopkins and Gregorich, 2003) and in lignin content (Escher et al., 2000; Jung and Sheaffer, 2004; Masoero et al., 1999; Saxena and Stotzky, 2001) were not consistent among studies and among isoline pairs, supporting the conclusion that chemical differences between Bt and non-Bt isolines are hybrid specific and depend on environmental conditions.

CONCLUSIONS This study shows that Cry1Ab protein can be degraded by F. graminearum GZ3639 and T. atroviride P1 and has no direct effect on the fungi in concentrations as present in maize residues. Some maize isolines, however, differed more in composition than Cry1Ab protein content alone, which can affect the saprophytic growth of these agronomically important fungi. The observed differences were within the variability between conventional cultivars and smaller than differences due to different environmental conditions. While we conclude that the impact of transgenic Bt maize residues on F. graminearum and T. atroviride is limited, we suggest that agronomically important fungi should be assessed in a field scale monitoring of Bt transgenic maize since small effects of a differential residue composition could amplify in the field over time. Our results emphasize the importance of multiple Bt/non-Bt hybrid comparisons and consideration of environmental vari-

ation in the evaluation of potential non-target effects of transgenic crops. ACKNOWLEDGMENTS We gratefully acknowledge V. Heitz (Amt fu¨r Landwirtschaft of the Landratsamt Ortenaukreis, Offenburg, Germany) and K.H. Dannemann (Regierungspra¨sidium Freiburg, Germany) for permission to sample on the field trial, K. O’Donnell (ARS culture collection, USDA, Peoria, IL) for providing the F. graminearum strain GZ3639, M. Lorito (University Federico II, Naples, Italy) for providing the T. atroviride strain P1, and M. Senatore for technical assistance. This research was supported by Swiss National Center of Competence in Research (NCCR Plant Survival, Neuchaˆtel).

REFERENCES AGBIOS. 2005. Biotech crop database. Available at http://www.agbios. com/dbase.php (verified 14 Mar. 2006). AGBIOS, Merrickville, ON, Canada. Baumgarte, S., and C.C. Tebbe. 2005. Field studies on the environmental fate of the Cry1Ab Bt-toxin produced by transgenic maize (MON810) and its effect on bacterial communities in the maize rhizosphere. Mol. Ecol. 14:2539–2551. Bennett, J.W., and M. Klich. 2003. Mycotoxins. Clin. Microbiol. Rev. 16:497–511. Broder, M.W., and G.H. Wagner. 1988. Microbial colonization and decomposition of corn, wheat, and soybean residue. Soil Sci. Soc. Am. J. 52:112–117. Brunner, K., C.K. Peterbauer, R.L. Mach, M. Lorito, S. Zeilinger, and C.P. Kubicek. 2003. The Nag1 N-acetylglucosaminidase of Trichoderma atroviride is essential for chitinase induction by chitin and of major relevance to biocontrol. Curr. Genet. 43:289–295. Cross, D.L., W.J. Cladwell, and J.R. Strickland. 1994. Efficacy of drought-stressed whole corn plants as a feed for ruminants. I. Effect of drought stress on nutrient accumulation in whole corn plant. Int. J. Anim. Sci. 9:309–315. Dill-Macky, R., and R.K. Jones. 2000. The effect of previous crop residues and tillage on Fusarium head blight of wheat. Plant Dis. 84:71–76. Donegan, K.K., C.J. Palm, V.J. Fieland, L.A. Porteous, L.M. Ganio, D.L. Schaller, L.Q. Bucao, and R.J. Seidler. 1995. Changes in levels, species and DNA fingerprints of soil-microorganisms associated with cotton expressing the Bacillus thuringiensis var. kurstaki endotoxin. Appl. Soil Ecol. 2:111–124. Escher, N., B. Ka¨ch, and W. Nentwig. 2000. Decomposition of transgenic Bacillus thuringiensis maize by microorganisms and woodlice Porcellio scaber (Crustacea: Isopoda). Basic Appl. Ecol. 1:161–169. Fernandez, M.R. 1992. The effect of Trichoderma harzianum on fungal pathogens infesting wheat and black oat straw. Soil Biol. Biochem. 24:1031–1034. Flores, S., D. Saxena, and G. Stotzky. 2005. Transgenic Bt plants decompose less in soil than non-Bt plants. Soil Biol. Biochem. 37: 1073–1082. Harman, G.E., and T. Bjo¨rkman. 1998. Potential and existing uses of Trichoderma and Gliocladium for plant disease control and plant growth enhancement. p. 229–265. In G.E. Harman and C.P. Kubicek (ed.) Trichoderma and Gliocladium, Vol. 2: Enzymes, biological control and commercial applications. Taylor and Francis Ltd., London. Harris, L.J., A.E. Desjardins, R.D. Plattner, P. Nicholson, G. Butler, J.C. Young, G. Weston, R.H. Proctor, and T.M. Hohn. 1999. Possible role of trichothecene mycotoxins in virulence of Fusarium graminearum on maize. Plant Dis. 83:954–960. Hjeljord, L.G., A. Stensvand, and A. Tronsmo. 2001. Antagonism of nutrient-activated conidia of Trichoderma harzianum (atroviride) P1 against Botrytis cinerea. Phytopathology 91:1172–1180. Hopkins, D.W., and E.G. Gregorich. 2003. Detection and decay of the Bt endotoxin in soil from a field trial with genetically modified maize. Eur. J. Soil Sci. 54:793–800. Hopkins, D.W., E.A. Webster, J.A. Chudek, and C. Halpin. 2001. Decomposition in soil of tobacco plants with genetic modifications to lignin biosynthesis. Soil Biol. Biochem. 33:1455–1462.



James, C. 2005. Global status of commercialized biotech/GM crops: 2005. ISAAA Briefs no. 34. ISAAA, Ithaca, NY. Jung, H.G., and C.C. Sheaffer. 2004. Influence of Bt transgenes on cell wall lignification and digestibility of maize stover for silage. Crop Sci. 44:1781–1789. Koskella, J., and G. Stotzky. 1997. Microbial utilization of free and clay-bound insecticidal toxins from Bacillus thuringiensis and their retention of insecticidal activity after incubation with microbes. Appl. Environ. Microbiol. 63:3561–3568. Koskella, J., and G. Stotzky. 2002. Larvicidal toxins from Bacillus thuringiensis subspp. kurstaki, morrisoni (strain tenebrionis), and israelensis have no microbicidal or microbiostatic activity against selected bacteria, fungi, and algae in vitro. Can. J. Microbiol. 48:262–267. Kowalchuk, G.A., M. Bruinsma, and J.A. van Veen. 2003. Assessing responses of soil microorganisms to GM plants. Trends Ecol. Evol. 18:403–410. Kullnig, C.M., T. Krupica, S.L. Woo, R.L. Mach, M. Rey, T. Benitez, M. Lorito, and C.P. Kubicek. 2001. Confusion abounds over identities of Trichoderma biocontrol isolates. Mycol. Res. 105:770–772. Lutz, M.P., G. Feichtinger, G. De´fago, and B. Duffy. 2003. Mycotoxigenic Fusarium and deoxynivalenol production repress chitinase gene expression in the biocontrol agent Trichoderma atroviride P1. Appl. Environ. Microbiol. 69:3077–3084. Maiti, R.K., and P. Wesche-Ebeling. 1998. Influences of the environment. p. 233–294. In R.K. Maiti and P. Wesche-Ebeling (ed.) Maize science. Science Publ., Enfield, NH. Masoero, F., M. Moschini, F. Rossi, A. Prandini, and A. Pietri. 1999. Nutritive value, mycotoxin contamination and in vitro rumen fermentation of normal and genetically modified corn (cry1A(b)) grown in northern Italy. Maydica 44:205–209. Masson, L., G. Prefontaine, L. Peloquin, P.C.K. Lau, and R. Brousseau. 1989. Comparative-analysis of the individual protoxin components in P1 crystals of Bacillus thuringiensis subsp kurstaki isolates NRD-12 and HD-1. Biochem. J. 269:507–512. Miller, J.D., J. Culley, K. Fraser, S. Hubbard, F. Meloche, T. Ouellet, W.L. Seaman, K.A. Seifert, K. Turkington, and H. Voldeng. 1998. Effect of tillage practice on Fusarium head blight of wheat. Can. J. Plant Pathol. 20:95–103. Miller, J.D., and H.L. Trenholm (ed.) 1994. Mycotoxins in grain; compounds other than aflatoxins. Eagan Press, St. Paul, MN. Naef, A., M. Senatore, and G. De´fago. 2006. A microsatellite-based method for quantification of fungi in decomposing plant material elucidates the role of Fusarium graminearum DON production in the saprophytic interaction with Trichoderma atroviride in maize tissue microcosms. FEMS Microbiol. Ecol. 55:211–220. Nguyen Thu, H. 2004. Sicherheitsforschung und Monitoringmethoden zum Anbau von Bt-Mais: Expression, Nachweis und Wirkung von rekombinantem Cry1Ab in heterologen Expressionssystemen. Ph.D. diss. Georg-August-Univ., Go¨ttingen, Germany.

1009

Parton, W.J., D.S. Ojima, and D.S. Schimel. 1996. Models to evaluate soil organic matter storage and dynamics. p. 421–448. In M.R. Carter and B.A. Stewart (ed.) Structure and organic matter storage in agricultural soils. CRC Press, Boca Raton, FL. Pillonel, L., S. Ampuero, R. Tabacchi, and J.O. Bosset. 2003. Analytical methods for the determination of the geographic origin of Emmental cheese: Volatile compounds by GC/MS-FID and electronic nose. Eur. Food Res. Technol. 216:179–183. Pusztai-Carey, M.P., P. Carey, T. Lessard, and M. Yaguchi. 1994. Isolation, quantification and purification of insecticidal proteins from Bacillus thuringiensis. U.S. Patent 5356788. Saxena, D., S. Flores, and G. Stotzky. 1999. Transgenic plants— Insecticidal toxin in root exudates from Bt corn. Nature 402:480. Saxena, D., C.N. Stewart, I. Altosaar, Q.Y. Shu, and G. Stotzky. 2004. Larvicidal Cry proteins from Bacillus thuringiensis are released in root exudates of transgenic B. thuringiensis corn, potato, and rice but not of B. thuringiensis canola, cotton, and tobacco. Plant Physiol. Biochem. 42:383–387. Saxena, D., and G. Stotzky. 2001. Bt corn has a higher lignin content than non-Bt corn. Am. J. Bot. 88:1704–1706. Schenk, N., K.C. Hsiao, and C.H. Bornman. 1991. Avoidance of precipitation and carbohydrate breakdown in autoclaved plant-tissue culture media. Plant Cell Rep. 10:115–119. Sims, S.R., and L.R. Holden. 1996. Insect bioassay for determining soil degradation of Bacillus thuringiensis subsp kurstaki CryIA(b) protein in corn tissue. Environ. Entomol. 25:659–664. Sutton, J.C. 1982. Epidemiology of wheat head blight and maize ear rot caused by Fusarium graminearum. Can. J. Plant Pathol. 4: 195–209. Systat. 2000. Systat Version 10. Systat Software, Richmond, CA. Venkateswerlu, G., and G. Stotzky. 1992. Binding of the protoxin and toxin proteins of Bacillus thuringiensis subsp. kurstaki on clayminerals. Curr. Microbiol. 25:225–233. White, S., M. McIntyre, D.R. Berry, and B. McNeil. 2002. The autolysis of industrial filamentous fungi. Crit. Rev. Biotechnol. 22:1–14. Wu, W.X., Q.F. Ye, H. Min, X.J. Duan, and W.M. Jin. 2004. Bttransgenic rice straw affects the culturable microbiota and dehydrogenase and phosphatase activities in a flooded paddy soil. Soil Biol. Biochem. 36:289–295. Yi, C.L., H.P. Kaul, E. Kubler, and W. Aufhammer. 2002. Populations of Fusarium graminearum on crop residues as affected by incorporation depth, nitrogen and fungicide application. Z. Pflanzenkr. Pflanzenschutz 109:252–263. Zscheischler, J., M.C. Estler, W. Staudacher, F. Gross, G. Burgstaller, H. Streyl, and T. Rechman. 1990. Handbuch Mais: Umweltgerechter Anbau, wirtschaftliche Verwertung. 4th ed. Verlagsunion Agrar, Frankfurt am Main, Germany. Zwahlen, C., A. Hilbeck, P. Gugerli, and W. Nentwig. 2003. Degradation of the Cry1Ab protein within transgenic Bacillus thuringiensis corn tissue in the field. Mol. Ecol. 12:765–775.