| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
* Physiology Weihenstephan, Technical University Munich, Weihenstephaner Berg 3, D-85354 Freising, Germany; and
and
Department of Animal Sciences, Division of Animal Nutrition, Technical University Munich, Hochfeldweg 6, D-85350 Freising, Germany
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: cattle • enzyme-linked immunosorbent assay • in situ disappearance kinetics • polymerase chain reaction • recombinant DNA • recombinant protein
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Einspanier et al., 2004; Lutz et al., 2005b). However, because those studies did not consider the long mean retention time of feed in the rumen of cows, additional data on the time-dependent degradation of plant DNA and the recombinant cry1Ab gene, as well as the Cry1Ab protein are crucial. In addition, knowledge about the degradation of transgenic DNA over time is necessary to estimate the likelihood of functional genes to transfer from feed to ruminal microorganisms. In the current study, we adopted a nylon bag technique (Ørskov and McDonald, 1979; Flachowsky et al., 1988; Madsen and Hvelplund, 1994) to analyze in situ degradation of the chloroplast-specific rubisco gene, the recombinant cry1Ab gene, and the Cry1Ab protein in the rumen of cows. |
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In Situ Experiments and Animal Performance
The study was performed according to strict federal and international guidelines on animal experimentation. The experiment was set up according to the requirements of the Bavarian State animal welfare committee.
Genetically modified Bt176 corn (Navares) and the non-Bt isoline (Antares; Syngenta International AG, Basel, Switzerland) were grown on adjacent experimental fields at the Bavarian State Research Center for Agriculture (Poing-Grub, Germany) to minimize variation based on different environmental conditions. After harvest, the whole corn plants were chopped (particle size of approximately 1 cm) and either frozen to –20°C or ensiled. Subsequently, the following 4 substrates were used for the in situ experiment: whole-plant isogenic (53% DM; 90 g of CP, 27 g of crude fat, and 178 g of NDF/kg of DM); whole-plant transgenic (37.9% DM; 80 g of CP, 28 g of crude fat, and 186 g of NDF/kg of DM); ensiled isogenic (51.7% DM; 90 g of CP, 24 g of crude fat, and 183 g of NDF/kg of DM); and ensiled transgenic corn (36.3% DM; 84 g of CP, 29 g of crude fat, and 182 g of NDF/kg of DM). Samples were lyophilized and milled to pass a 3-mm screen. Five grams of an individual lyophilized sample on an as-fed basis were weighed into bags of precision woven nylon cloth (10 cm x 20 cm) with an aperture of 53 ± 10 µm (BG1020; Bar Diamond Inc., Parma, ID). All nylon bags were sealed and connected to a 55-cm, coated flexible steel cable with lacing cords. Before the 0700 feeding, this plastic carrier was placed in the rumen and attached to the cap of the fistula on a 40-cm nylon cord. To exclude the possibility of distinct degradation patterns of DNA and protein dependent on the sample’s position on the carrier, triplicates of each variety were attached randomly to the carrier. After 2, 4, 8, 16, 24, and 48 h of incubation in the rumen, the bags were quenched in iced water to stop the microbial activity and subsequently washed in a washing machine using a standard cold rinse cycle. The bags were then lightly squeezed to eliminate excess water and lyophilized. To estimate the initial DNA and protein concentration, bags containing either whole-plant or ensiled isogenic or transgenic corn were washed and lyophilized without ruminal incubation in duplicate.
To ascertain equal ruminal fermentation conditions in all cows, we determined the pH values (Schott CG 842 pH meter, Mainz, Germany) and the ammonia content of ruminal fluid collected from the ventral sac according to Voigt and Steeger (1967) before and 30, 60, 90, 150, 210, and 270 min after feeding. Furthermore, ruminal VFA were measured according to Geissler et al. (1976) in samples taken before and 210 min after feeding.
Deoxyribonucleic Acid Extraction
Lyophilized and frozen samples (50 mg) of at least two different positions on the rumen-placed carrier per time point and substrate (5 cows x 2 bags; n = 10) were processed. Using the bead-beating FastPrep technique (BIO101, Carlsbad, CA), 50 mg were repeatedly ground with 0.8 g of Matrix Green ceramic beads (BIO101) at 5.0 m/s for 40 s. To improve DNA yield, samples were refrozen in liquid N2 for 10 min before being reground. The resulting fine powder was dissolved in 600 µL of lysis buffer (C1) and 10 µL of RNase-A (Nucleo Spin Plant Kit; Macherey-Nagel GmbH & Co. KG, Düren, Germany), mixed thoroughly, and incubated for at least 30 min at 60°C. All succeeding DNA purification steps were performed using a silica spin column following the manufacturer’s protocol. The DNA was finally eluted in 50 to 100 µL of CE-buffer (Nucleo Spin Plant Kit) depending on the concentration.
Concentrations of DNA were determined by UV absorption at 260 nm, and the DNA integrity was estimated by 260/280 UV absorption ratio.
PCR Analysis
Oligonucleotides.
Primer sets used for PCR and the respective DNA fragments amplified are described in Table 1. Five primer pairs were designed to detect different fragment lengths (173, 430, 896, 1,197, and 1,753 bp) of the highly abundant chloroplast-specific rubisco gene (Zea mays complete chloroplast genome; GenBank Accession No. X86563). Four primer pairs were used to amplify fragments (211, 420, 727, and 1,423 bp) of the transgenic construct in Bt176 (cry1Ab gene from US Patent 5625136; GenBank Accession No. I41419). Primers were designed using the computer software PRIMER-3 (freely available at http://www.wi.mit.edu).
|
.
|
All PCR assays included both positive and negative controls. As a positive control, DNA from Bt176 corn leaf tissue (Navares) was used. Negative controls for the rubisco and cry1Ab primer sets did not contain template DNA. Additional negative controls for the amplification of cry1Ab fragments consisted of non-recombinant corn DNA (Antares).
Agarose Gel Electrophoresis.
The PCR amplicons (15.0 µL) were electrophoresed at 100 V on a 1.5 to 2% (wt/vol) agarose gel and visualized using an UV transilluminator. The gels were digitized using a video documentation system (Vilber Lourmat, Marne-la-Vallée Cedex 1, France).
Sequencing.
The PCR products were commercially sequenced to confirm nucleotide sequence identity (Medigenomix, Martinsried, Germany).
Relative Quantification of Rubisco and Cry1Ab DNA (Real-Time PCR)
Quantification of the rubisco and cry1Ab gene was performed on at least duplicate samples of two different positions on the rumen-placed carrier using the LightCycler instrument (Roche, Mannheim, Germany). rubisco and the cry1Ab gene concentrations of samples incubated in the rumen were expressed as a percentage of initial values.
For rubisco gene quantification, the reaction mixture (final volume = 10 µL) contained 50 ng of extracted DNA, 1 µL of LightCycler DNA Master SYBR Green I (10x), 3 mM of MgCl2 (Roche), and 0.4 µM of forward and reverse primer (Rub173F and Rub173R). Amplification involved one cycle at 95°C for 10 min for initial denaturation and 45 cycles of 95°C for 15 s, 61°C for 10 s, and 72°C for 20 s. Amplified products underwent melting curve analysis by slow heating with a 0.1°C/s increment from 65 to 95°C with fluorescence collection at 0.1°C intervals after the last cycle to specify the integrity of amplification. Additionally, the product size of all samples was verified by electrophoresis after the PCR run. Dilutions of purified cloned DNA were used to construct gene-specific calibration curves.
For relative quantification of genetically modified Bt176 corn, a commercially available Bt176 quantification kit (Roche) was used according to the manufacturer’s instructions.
In addition to the cry1Ab gene, a reference gene (corn invertase, provided in the kit) was measured and the ratio between the cry1Ab and invertase gene was calculated to ensure unvarying degradation patterns of both genes. To exclude possible unspecific amplification for the cry1Ab gene quantification, isogenic corn samples of all time points and substrates also were analyzed.
Cry1Ab Protein Analysis
Enzyme-Linked Immunosorbent Assay.
Estimation of the Cry1Ab protein was performed using a commercially available ELISA kit according to the manufacturer’s instructions (Agdia Inc., Elkhart, IN). Frozen whole-plant or ensiled transgenic corn samples (20 mg) obtained from all animals at all time points and originating from at least 2 positions on the rumen-placed plastic carrier (n 
10) were ground using the FastPrep homogenizing instrument with 0.8 g of Matrix Green ceramic beads at 6 m/s for 40 s until the material was pulverized. The resulting powder was dissolved in 1 mL of MultiEvent buffer (provided in the kit). Dilutions of 0.015, 0.03, 0.06, 0.125, 0.25, 0.5, 1.0, and 1.6 ng/mL of a control Cry protein (provided in the kit) were used to create a standard curve. Results are expressed as Cry1Ab protein (ng/g of fresh weight). Samples of non-recombinant corn served as negative controls. All samples were measured at least in duplicates.
Immunoblotting.
An immunoblotting technique (Lutz et al., 2005b) was used to determine the fragment size of the Cry1Ab protein detected by ELISA. Briefly, whole-plant and ensiled transgenic corn samples were prepared as described for the ELISA assay, except that PBS (pH 7.4) with protease inhibitors (Merck KGaA, Darmstadt, Germany) was used as extraction buffer. The SDS-PAGE conditions included a 4 to 12% gradient Bis-Tris gel (NuPage, Invitrogen GmbH, Karlsruhe, Germany), 17.75 µL of extracted protein, 1.0 µL of 1,4-Dithiothreitol (1 mM; Merck KGaA), and 6.25 µL of SDS sample buffer (4x). After separation and transfer onto a nitro-cellulose membrane, Cry1Ab protein was detected using a polyclonal rabbit antiCry1Ab/1Ac antibody (Agdia Inc.; final concentration 5 µg/mL; 60 min) followed by a secondary antibody solution (biotinylated antirabbit IgG in casein solution; final concentration 1.5 µg/mL; 30 min). For signal amplification, membranes were incubated in Vectastain ABC-AmP Reagent (Vector Laboratories, Inc., Burlingame, CA; 10 min). Signals were visualized by chemiluminescence (DuoLuX, Vector Laboratories, Inc.). Isogenic corn served as a negative control; the sample without incubation time in the rumen (0 h) and Cry1Ab/1Ac protein included in the ELISA kit served as positive controls.
Statistical Analyses
Results of PCR for a specific time point were regarded as positive when >50% of all extracted DNA samples yielded amplicons of the correct size (bp) and identity (sequence analysis).
Data of ELISA were analyzed by using the MIXED procedure of SAS (Version 8.2; SAS Inst., Inc., Cary, NC). At 2 h, the influence of ruminal position was tested for significance for the transgenic whole-plant corn samples. The model included animal (n = 5) and position (n = 3) as class variables. Because degradation was not significantly influenced by position, further evaluations were restricted to two ruminal positions. The model of the MIXED procedure included animal as a random effect, incubation time as repeated measure, and corn type and ruminal position as class variables. If the F-test of the MIXED procedure was significant, differences were separated using the Duncan’s multiple range test.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Fragments of 896 and 197 bp were found up to 24 h in whole-plant samples but could not be detected in any of the ensiled samples tested. Rubisco gene fragments of 1,753 bp could be amplified up to 8 h after rumen incubation in whole-plant corn. Analyzing all amplifiable fragments of the rubisco gene did not show a qualitative difference between isogenic and transgenic corn.
|
). The 420-bp fragment could be amplified in all samples of whole plant but only up to 8 h of ensiled corn. Fragment lengths of 727 and 1,423 bp were detected in whole-plant corn up to 16 and 4 h, respectively, but were not found in any sample of ensiled corn. A summary of all tested amplicons in whole-plant and ensiled corn samples as well as a comparison between the rubisco and the cry1Ab gene is shown in Figure 3.
|
|
). In addition, conventional PCR (and ELISA measurements, see subsequent) yielded similar results for all cows. For technical reasons, quantitative PCR was performed in 3 animals only, and results are shown for one representative animal.
|
for transgenic whole-plant and ensiled corn samples, no consistent difference in the degradation pattern between the positions was apparent. Therefore, DNA originating from the 2 different bags was extracted and pooled for quantification of rubisco and cry1Ab gene fragments.
|
) or ensiled corn (Figure 4D). Less than 20% of the initial rubisco gene fragment was measured after 2 h of ruminal incubation both in isogenic and transgenic whole-plant corn. Values were <5% of initial by 4 h; final values were <0.1% for Bt and non-Bt corn by 48 h of ruminal incubation. Initial values for the rubisco gene measured in ensiled plant samples were approximately 0.9% of initial values found in whole-plant corn samples. A marked decrease to <6% of the initial values for both isogenic and transgenic corn after 4 h of incubation in the rumen was measured. Values dropped to <0.5% by 48 h of ruminal incubation.
Quantifying the cry1Ab gene using real-time PCR (Figure 5) showed a sharp decrease during the first 4 h of incubation to <1% of the initial value for whole-plant corn samples, which ended up in final values of <0.5% after 48 h. Quantification of the cry1Ab gene fragment in ensiled corn samples reached the detection limit of the commercial kit after 8 h of ruminal incubation; therefore, data are not shown. Results after that time point were not reliably distinguishable from values of isogenic corn samples; however, the 0-h value of ensiled corn was only 0.62% of the initial value of whole-plant transgenic corn. Comparing the degradation pattern of rubisco DNA (Figure 4) and cry1Ab DNA (Figure 5) did not reveal obvious differences between both genes.
|
). Quantities of protein diminished gradually to 18 ± 5.2% (28.2 ± 9.31 ng/g) after 4 h and ended with final values of 2.6 ± 1.0% (4.1 ± 1.52 ng/g) after 48 h.
|
Immunoblotting.
Evaluation and validation of the immunoblotting assay was performed on fresh transgenic corn leaves and showed a detection limit of 50 mg (fresh weight) of transgenic corn, equivalent to approximately 2.0 ng of Cry1Ab protein/g of fresh weight according to ELISA concentration measurement (Lutz et al., 2005a). The full-sized protein of 60 kDa could be detected up to 8 h in whole-plant corn (Figure 6B). Bands of immunoactive fragments with the length of 17 kDa were observed after 2 h and increased thereafter. After 16 h of incubation, bands of approximately 34 kDa were evident and intensified at later time points.
Bands for Cry1Ab protein were not detected in any of the ensiled corn samples (data not shown).
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
To study precise degradation profiles, an in situ technique using nylon bags was adopted. This method has been applied previously to investigate the ruminal degradability of feed from Bt and non-Bt corn and revealed no effect of a genetic modification on the determined values (Aulrich et al., 2001; Folmer et al., 2002; Donkin et al., 2003).
In our studies, conventional PCR techniques using the highly abundant rubisco DNA as a marker gene provided comparable results to those using cry1Ab DNA. Only small fragments of the rubisco gene and the cry1Ab gene were amplifiable in samples of whole-plant and ensiled corn. In ensiled corn samples, fragments spanning the size of 896 bp for rubisco and 727 bp for cry1Ab could not even be detected in the 0-h samples. This parallels observations of Hupfer et al. (1999) and our own laboratories (Lutz et al., 2005a), in which only small DNA fragments persisted after 60 or 106 d of ensiling. Compared with ensiled corn, rubisco and cry1Ab gene segments of whole-plant corn samples were detectable for longer periods of ruminal incubation. Fragments of DNA with comparable sizes to a potential functional bioactive gene (e.g., beta-lactamase [bla] introduced in Bt176 corn with a size of 861 bp; Accession No. U03991) were detectable up to 16 h for rubisco (896 bp) and 24 h for cry1Ab (727 bp). However, Badosa et al. (2004) demonstrated no transfer of the bla gene from Bt176 to corn-associated bacteria under field conditions. Even under optimized laboratory conditions, the transformation of highly competent bacteria with transgenic plant DNA extracts was only observed when highly homologous regions were present (Schluter et al., 1995; Nielsen et al., 1998; de Vries et al., 2001). Thus, we showed that feed DNA can survive in ruminal fluid for a significant time, but functional activity of that DNA is unlikely to remain after exposure to the ruminal environment.
Our results regarding conventional PCR analysis differ from those of Phipps et al. (2003) and Chowdhury et al. (2004). Phipps et al. (2003) detected high-copy rubisco DNA only up to 1,197 bp and single-copy transgenes of approximately 200 bp in ruminal samples after feeding genetically modified soybean meal and ground, genetically modified corn grains. However, in that study, samples from various time points were bulked into single samples, leading to a decrease in the detection of larger fragments. Chowdhury et al. (2004) showed no time-dependent fragmentation but obtained similar amplicons in ruminal samples after 5 and 18 h of feeding.
Real-time PCR results for rubisco and cry1Ab DNA confirmed those of conventional PCR, showing decreasing amounts with increasing ruminal incubation time. After 4 h, <10% of the initial values for rubisco and cry1Ab genes were found in all substrates, reflecting the rapid microbial and enzymatic digestion of DNA in ruminal fluid (Duggan et al., 2000). This agrees with findings of Einspanier et al. (2004), who reported a significant decrease in the amount of chloroplast DNA after gastric digestion.
It should be noted, however, that generally with real-time PCR, only small DNA fragments are analyzed (in our case <174 bp of plant and transgenic DNA). As shown by conventional PCR, those lengths were still detectable after 48 h, even in ensiled corn samples, whereas longer fragment lengths could not be amplified by 48 h of ruminal incubation. Therefore, the results obtained by quantitative PCR do not provide evidence for the persistence of potential full-sized functional genes during the ruminal fermentation process.
Analyzing the Cry1Ab protein in ensiled and whole-plant transgenic corn by ELISA showed a continual decrease of the immunoactive signal with advancing ruminal incubation time. The initial value of ensiled corn samples constituted only 10% of those of whole-plant corn, demonstrating a degradation of recombinant protein during the ensiling process. Our own investigations revealed a clear decrease of Cry1Ab protein during the ensiling process, probably because of low pH conditions and microbial activity (Lutz et al., 2005a).
To the best of our knowledge, there are no data available on the exact time-dependent degradation kinetics of ruminally incubated recombinant protein. However, Chowdhury et al. (2003) and Lutz et al. (2005b) noted a marked decrease in the content of Cry1Ab protein after ruminal digestion.
Immunoblotting showed the full-size protein (60 kDA) up to 8 h, which is in contrast to our recently acquired in vivo data, where ruminal samples were taken after slaughtering and no full-size protein was found (Lutz et al., 2005b). Moreover, specific fragments of 17 and 34 kDa were detected, which were previously described by Lutz et al. (2005b) in gastrointestinal samples. Chowdhury et al. (2003) did not report detection of these smaller fragments, which could be due to the use of different antibodies or a different fragmentation pattern of Bt176 corn compared with the Bt11 corn evaluated by those authors. In summary, the results obtained by immunoblotting confirm that ELISA measurements require careful interpretation because currently used, commercially available antibodies obviously also bind to immunoactive fragments whose potential activity has yet to be investigated.
Using PCR, ELISA, and immunoblotting techniques, we demonstrated that digestion of corn in the rumen of cows results in extensive time-dependent degradation and fragmentation of recombinant DNA and protein from Bt corn. It is unlikely that immunoactive protein fragments exhibit activity, but the potential bioactivity of Cry1Ab protein fragments warrants further investigation.
|
Footnotes |
|---|
2 Corresponding author: christiane.albrecht{at}wzw.tum.de
Received for publication March 9, 2005. Accepted for publication August 12, 2005.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Wiedemann, P. Gurtler, and C. Albrecht Effect of Feeding Cows Genetically Modified Maize on the Bacterial Community in the Bovine Rumen Appl. Envir. Microbiol., December 15, 2007; 73(24): 8012 - 8017. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
