Risk assessments for field release of transgenic plants should evaluate containment of the engineered genetic traits, persistence of the novel genetic information and gene products, as well as potential effects on nontarget organisms.

As a model system we initiated a project to study the fate in soil of plasmid and linear DNA encoding tomato proteinase inhibitor I (TPI-I) and neomycin phosphotransferase II (NPT-II). The fate of DNA in 100 mg soil samples was monitored over days to weeks using bulk DNA extraction and purification, Southern blotting, and PCR. Using these tools we have determined the effects of soil moisture content and temperature on the stability of the added DNA. In this system, the DNA was initially rapidly degraded but Southern blotting and polymerase chain reaction revealed persistence of the DNA for weeks, thus potentially making it biologically available for microbial transformation.

In a second part of the study, a transformation assay for plant derived transgenes will be developed based on marker rescue using the transgenic plant encoded NPT-II gene. Genetic transformation based on DNA uptake and homologous recombination will be monitored by plating on selective medium. The assay system will be characterized for laboratory and soil conditions to simulate conditions of actual field releases.

Key Words: soil DNA extraction, DNA stability, transgenic plants, risk assessment, PCR amplification

Plants are genetically engineered in order to introduce new properties such as herbicide tolerance, pest resistance, and altered fruit quality (Beer, 1993). In advance of extensive use in agriculture, various aspects of risk associated with field releases of transgenic plants have to be assessed, including i) unexpected, disadvantageous behavior of transgenic plants ii) transfer, stabilization, and spread of the engineered genes in indigenous plant and microbial populations, and iii) potential for disturbance of the soil ecosystem by residual plant material following harvesting and tillage (Dietz, 1993; Stotzky, 1992).

In order to assess the potential risk of using transgenic plants in agriculture it is important to know the fate of the engineered genetic information in the environment. So far, no information is available on the fate of DNA derived from decomposing transgenic plants in the field. However, data from studies on the fate of naked bacterial DNA in a defined soil (Lorenz and Wackernagel, 1987; Khanna and Stotzky, 1992) and in more complex environmental samples (Chamier et al., 1993; Romanowsky et al., 1993) indicate that free DNA can bind to soil particles, resist degradation, and maintain transformation activity. This raises the possibility that DNA released during the process of decomposition of plant tissue may be stabilized on soil particles and persist for some time. A component (NPT-II gene) of the DNA inserted into the plant genome is derived from a procaryotic transposon, Tn5, that confers resistance to kanamycin. Soil microorganisms may be exposed to and possibly transformed by this free DNA. Such a mechanism may result in the persistence and spread of engineered genes in the microbial community.

The objectives of this study are to provide data on the fate of plasmid and transgenic plant derived DNA in an agricultural soil and to evaluate environmental factors affecting DNA stability in soil. Furthermore it will evaluate the potential of transgenic plant-derived neomycin phosphotransferase II (NPT-II) marker gene construct, encoding a kanamycin resistance, to genetically transform Bacillus subtilis in soil.

In this first progress report we present data on the fate of the model plasmid pFW1 in an agricultural soil and illustrate how its stability is affected by temperature and moisture. A methodology has been developed to extract, purify, and specifically detect recombinant NPT-II sequences from small scale environmental samples.

Plasmids and bacteria. Plasmid pJN3 was a kind gift of Dr. Ryan, Washington State University, Pullman, WA. The transferable DNA between the right and left border of pJN3 encodes the two genes that are expressed in transformed plant cells (An et al., 1985; Johnson et al., 1989). One is the marker gene for the selection of transformed plant cells and encodes recombinant NPT­II conferring kanamycin resistance. This coding region is fused to the nopaline synthase (NOS) promotor. The other gene encodes tomato proteinase inhibitor I (TPI­I) and is fused to the cauliflower mosaic virus (CaMV) 35S promoter.

From plasmid pJN3 the 2652 bp KspI/HinDIII fragment, containing the engineered NPT­II gene was isolated. This fragment was ligated into pUCBM21 (Boehringer Mannheim, Indianapolis, I.N.) from which the NcoI site has been deleted. The resulting plasmid, pFW1 (5.3 kb), was maintained in E. coli HB 101.

Transgenic plants. The transgenic and control tobacco plants used in this study are a kind gift of Dr. Ryan, Washington State University, Pullman, WA. They were grown under contained conditions at the EPA Environmental Research Laboratory in Corvallis, OR. The transgenic tobacco plants were engineered with the pJN3 encoded transferable sequence to express TPI­I in order to protect them from insect attack (Graham et al., 1985; Johnson et al., 1989).

Soil. Fresh soil from the Willamette Research Station (Corvallis, OR) was sieved through a 2 mm sieve and stored for 1 day at 4°C in a sealed plastic bag. During this time a soil sample was dried for 12 hrs at 110°C to gravimetrically determine its moisture content. After one day soil moisture content was adjusted to 30% water holding capacity (WHC) and the soil was stored at constant moisture and temperature (25°C).

Experimental conditions to study the fate of linear plasmid DNA. A small scale soil system was used to study the fate of EcoR I linearized pFW1 in agricultural soil. Soil moisture was adjusted to 5-10% below the final intended value. For low moisture content experiments soil was dried over night at 36°C prior to the DNA addition. 100 mg (dry weight equivalent) of nonsterile soil was added to 1.5 ml test tubes. 720 ng linearized pFW1 (1.2x10¹¹ copies) in sterile water was added to each tube. Water only was added to control soil samples. The water addition adjusted the moisture content to the various final levels, i.e. 10%, 40%, and 80% WHC. The closed tubes were incubated for various selected periods of time under controlled temperature conditions, i.e. 4°C, 20°C, and 36°C.

DNA extraction from soil and plant tissue. To extract DNA from soil and plant tissue a modified protocol for small scale DNA extraction from environmental samples (Porteous and Armstrong, 1993; Porteous et al., in press) was used. To 100 mg soil samples 700 l extraction buffer (250 mM NaCl, 100 mM EDTA, pH 8, 2% SDS) was added and vortexed for 15 sec. After addition of 50 l guanidine isothiocyanate (5 M) samples were again vortexed and stored at ­20°C. For plant DNA extraction 100 mg leaf tissue was manually ground with a pestle (Kontes, Vineland, NJ) for 30 sec in a test tube containing 100 l extraction buffer. After addition of 600 l extraction buffer and 50 l guanidine isothiocyanate (5 M) extraction proceeded without freezing. To extract DNA all samples were heated to 68°C for 10 min, sonicated for 3 min (Banson 5200 sonicator-bath, Shelton, CT), and further incubated at 68°C for 1 hr. Samples were centrifuged for 5 min at 12000 x g at room temperature, supernatants saved, and DNA precipitated with 750 l isopropanol over night at ­20°C. After 30 min centrifugation at 4°C pellets were washed with 70% ethanol and resuspended in 100 l TE (10 mM TrisHCl, 1 mM EDTA, pH 8). This DNA solution is referred to as crude soil extract.

Purification of extracted soil DNA. To obtain soil DNA that can be used in PCR amplification a modified protocol (Romanowsky et al., 1993) for purifying small scale DNA samples was used. DNA from 50 l crude soil extract was precipitated using 16.7% polyethylene glycol 8000 (PEG, Sigma, St. Louis, MO) and 0.33 M NaCl. The resulting pellet was washed with 70% ethanol and dissolved in 50l TE. This solution was further purified using adsorption of contaminants to acid washed polyvinyl-polypyrrolidone (PVPP, Sigma, St. Louis, MO) followed by ultrafiltration with Microcon-100 (Amicon, Beverly, MA). Finally DNA was precipitated with ethanol and dissolved in 50 l TE.

Characterization of extracted DNA. The quality of extracted soil DNA was monitored using agarose gel electrophoresis. The fate of pFW1 was examined using Southern blotting of crude extracts onto nylon membrane under alkaline conditions (Sambrook et al., 1989) and hybridization with a nonradioactive pFW1 probe, labeled and incubated according to the manufacturer recommendation using the Genius random priming and chemiluminescent detection system (Boehringer Mannheim, Indianapolis, IN). PCR amplification was performed for 30 or 40 cycles (15 sec denaturation at 92°C, 1 min annealing at 50°C, and 1 min extension at 72°C) using a TempCycler II (Coy Laboratory Products, Grasslake, MI). 100 l PCR reaction cocktails contained 1x reaction buffer (Boehringer Mannheim, Indianapolis, IN), 0.2 mM of each deoxynucleoside (Boehringer Mannheim, Indianapolis, IN), 5 mg/ml bovine serum albumin, 20 pmol of each oligonucleotide primer (Center for Gene Research, Oregon State University, Corvallis, OR), and purified soil and/or plasmid DNA. After 5 min initial denaturation at 96°C samples were kept at 75°C, 1U Taq polymerase (Boehringer Mannheim, Indianapolis, IN) added to each reaction, and temperature cycling started.

Design of the soil system. The concept of the described soil system was to maintain a complex soil at defined conditions. These demands were reached by using thoroughly mixed soil that had been stored for more than one week under controlled conditions. To ensure sample comparability we have chosen 100 mg individual soil incubations for each time point including separate DNA addition to each sample. This approach ensured homogeneity in the initial amounts of DNA and homogenized soil in each sample. The choice of 100 mg small scale soil incubations facilitated the extractions and purification steps required.

Soil DNA Extraction. To extract soil DNA we have chosen a modification of the direct lysis protocol developed at the Environmental Research Laboratory in Corvallis, OR. This method has previously been shown to be appropriate for isolating DNA from a variety of environmental samples including soil and plant tissue (Porteous et al., in press). The quality of extracted DNA was monitored using agarose gel electrophoresis and ethidium bromide staining. This analysis revealed that the applied extraction procedure yielded reproducible quantity and quality of extracted bulk soil DNA over a forty day incubation period.

Fate of linear plasmid DNA in soil. The effects of environmental factors (i.e. temperature and soil moisture) on the stability of linear plasmid DNA in soil was analyzed in two experimental systems. In a first set of experiments temperature effects on DNA stability were studied. The incubations were performed at 4°C, 20°C, and 36°C while keeping soil moisture at 40% WHC. In the second set of experiments soil moisture effects were studied. In this experiment, soil moisture was adjusted to 10%, 40%, and 80% WHC. Incubations were performed at 20°C.

The agarose gel analysis of 10 l crude extract samples provided an estimate of the fate of the added linear pFW1 DNA in soil and of the effects of temperature and moisture on its stability. By direct inspection of the stained gels, in the temperature variation experiment, the 5.3 kb band of linear pFW1 DNA could be detected up to 1 day at 4°C, 6 hrs at 20°C, and 1.5 hrs at 36°C incubations. In the moisture variation experiments the band of linear pFW1 DNA was detected up to 10 days at 10% WHC, 1.5 hrs at 40% WHC, and 80% WHC incubations. These results indicated that both low temperatures (i.e. 4°C) and low moisture content (i.e. 10% WHC) of the soil increased the stability of the added DNA.

The DNA analyzed by gel electrophoresis was transferred onto nylon membranes and analyzed further using hybridization with a nonradioactive pFW1 probe in combination with chemiluminescent detection. These analyses supported the results from the gel analysis with higher sensitivity. In addition to the intact linear pFW1 band, a smear of degradation products could be detected giving a qualitative representation of the breakdown of pFW1 in the system. With this method, pFW1 sequences could be detected for a longer period of time in both types of experiments. For the temperature variation experiment, pFW1 DNA sequences were detected up to 40 days at 4°C, 2 days at 20°C, and 1 day at 36°C. In the moisture variation experiment, the pFW1 DNA sequences were still detected after 40 days at 10% WHC, 2 days at 40% WHC, and 2 days at 80% WHC.

Bulk soil DNA purification and PCR amplification of the recombinant NOS / NPT-II target sequence. To amplify pFW1 derived recombinant NPT-II sequences in the soil extracts, the samples had to be further purified. We have chosen a three step purification, as described in the material and methods section, that can be used for small scale DNA sample preparation. The only step selecting for a fraction of the DNA present in the sample is the final ultrafiltration step using Microcon-100 microconcentrators. These filters have a double stranded DNA fragment size cut off of 125 bp, allowing larger fragments to be saved. Since the primers specifically designed for recombinant NPT-II detection amplify a 901 bp fragment, there is no interference of this DNA fragment length selectivity with PCR detection.

For PCR amplification, 0.1 l of purified soil DNA was added to the PCR reaction. In control experiments, up to 2 l of purified soil DNA could be added to a 100 l PCR reaction without interfering with the amplification of the NPT-II target sequence on pFW1. The PCR analysis of the purified soil samples revealed the presence of intact recombinant NPT-II encoding sequences up to 40 days except for the incubations at 36°C, where this sequence was detectable up to 10 days. Most recombinant NPT-II encoding DNA could be detected in the low temperature (4°C) and low moisture content (10% WHC) samples. These results indicate that naked DNA can be stable for long periods of time in nonsterile soil and that its stability can be affected by environmental conditions i.e. temperature and moisture content of the soil.

The future studies will investigate the stability of DNA derived from decomposing plant tissue in soil, using the PCR amplification described here. The potential bioavailability for microbial transformation of the recombinant NPT-II gene will be investigated using marker rescue in a B. subtilis model system.

An G, Watson BD, Stachel S, Gordon MP, Nester EW (1985) New cloning vehicles for transformation of higher plants. EMBO J 4: 277-284.

Beer A (1993) Transgenic crops approach the US market. AGROW 175: 25-27.

Chamier B, Lorenz MG, Wackernagel W (1993). Natural transformation of Acinetobacter calcoaceticus by plasmid DNA adsorbed on sand and groundwater aquifer material. Appl Environ Microbiol 59: 1662-1667.

Dietz A. (1993) Risk assessment of genetically modified plants introduced into the environment. In: Wöhrmann K and Tomiuk J (eds) Transgenic Organisms, Risk Assessment of deliberate Release, pp 209-227.

Graham JS, Pearce G, Merryweather J, Titani K, Ericsson L, Ryan CA (1985) Wound inducible Proteinase Inhibitors from tomato leaves. J Biol Chem 260: 6555-6560.

Johnson R, Narvaez J, An G, Ryan C (1989) Expression of proteinase inhibitors I and II in transgenic tobacco plants: Effects on natural defense against Manduca sexta larfae. Proc Natl Acad Sci USA 86: 9871-9875.

Khanna M, Stotzky G (1992) Transformation of Bacillus subtilis by DNA bound on montemorillonite and effect of DNase on the transforming ability of bound DNA. Appl Environ Microbiol 58: 1930-1939.

Lorenz M, Wackernagel W (1987) Adsorption of DNA to sand and variable degradation rates of adsorbed DNA. Appl Environ Microbiol 53: 2948-2952.

Porteous LA, Armstrong JL (1993) A simple mini-method to extract DNA directly from soil for use with polymerase chain reaction amplification. Curr Microbiol 27: 115-118.

Porteous LA, Armstrong JL, Seidler RJ, Watrud LS (1994) An effective method to extract DNA from environmental samples for polymerase chain reaction amplification and DNA fingerprinting. Curr. Microbiol. in press.

Romanowsky G, Lorenz MG, Wackernagel W (1993) Use of polymerase chain reaction and electorporation of Escherichia coli to monitor the persistence of extracellular plasmid DNA introduced into natural soils. Appl Environ Microbiol 59: 3438-3446.

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press.

Stotzky G (1992) Gene transfer among and ecological effects of genetically modified bacteria in soil. In: Casper R, Landsmann J (eds) 2nd International Symposium on The Biosafety Results of Field Tests of Genetically Modified Plants and Microorganisms, pp 122-134.