Effects of Transgenic Plants on Soil and Plant Microorganisms


Katherine K. Donegan1* and Ramon J. Seidler2

National Health and Environmental Effects Research Laboratory
Western Ecology Division
200 SW 35th Street
Corvallis, Oregon, USA 97333

*Corresponding author: phone (541) 754-4809, FAX (541) 754-4799, E-mail kellyd@mail.cor.epa.gov

The information in this article has been funded by the U.S. Environmental Protection Agency. It has been subjected to the Agency’s peer and administrative review, and it has been approved for publication as an EPA document. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.



The genetic engineering of plants has facilitated the production of valuable agricultural and forestry crops. Transgenic plants have been created that have increased resistance to pests, herbicides, pathogens, and environmental stress, enhanced qualitative and quantitative traits, and the ability to produce industrial and pharmaceutical compounds.1,2 The commercial use of transgenic plants has greatly increased worldwide. In the United States, approximately 3.6 million acres were planted with transgenic crops during 1996 and by 1998 the acreage planted with transgenic crops had increased to 50 to 60 million.3 Fifty-two transgenic plant species, that produce 35 different gene products, have been, or are in the process of being, commercialized in the United States.3 Because some transgenic plants carry genes and produce compounds foreign to their environment, can grow and establish outside of their natural habitat, and have enhanced survival, persistence, and competitive capabilities, there are several concerns about their environmental use and potential ecological effects. These concerns include invasiveness, gene flow to indigenous organisms, development of resistance in target pests, and direct or indirect effects on nontarget organisms and ecosystems.4,5,6,7 The issue of direct and indirect effects on nontarget organisms and ecosystems is particularly important because many transgenic plants are being developed that have new or enhanced antimicrobial properties for protection from phytopathogens.8 These transgenic plants express such antimicrobial compounds as chitinases, glucanases, lysozymes, thionins, defensins, and systemic acquired resistance (SAR) gene products.9

The majority of field and risk assessment studies on transgenic plants have focused on the efficacy of the engineered traits and/or on gene flow and plant invasiveness. Because of studies we conducted on the persistence of transgenic plant products in soil, we became concerned about the effects of transgenic plants on soil and plant microorganisms. In studies using cotton plants engineered for production of the Bacillus thuringiensis var. kurstaki (B.t.k.) endotoxin, we observed that the B.t.k. endotoxin within the transgenic cotton plants persisted after the plants were incorporated into soil and that the endotoxin retained its immunological and biological activity.10,11 Consequently, we were convinced that the potential direct and indirect effects of transgenic plants and their products on microorganisms is an important component of the risk assessment of transgenic plants. In response to this concern, we conducted several microcosm and field studies and demonstrated that exposure to transgenic plants produced changes in the population levels and composition of some soil and plant microorganisms.12,13,14,15 During the course of these studies, we realized the importance of using different methods to evaluate the persistence of the transgenic plants products, and the effects of the transgenic plants on microorganisms, due to differences among the methods in sensitivity and scope.

In this paper, we summarize several microcosm and field studies we have performed using different transgenic plants to evaluate the persistence of their products and their effects on soil and plant microorganisms. These studies used cotton, potato, tobacco, and alfalfa plants that were engineered for the production of pesticidal products (Bacillus thuringiensis endotoxins and proteinase inhibitors) and industrial compounds (alpha-amylase and lignin peroxidase). We discuss our studies in relation to the findings of other researchers who have considered the effects of transgenic plants on soil and plant microorganisms. The different methods we have applied for studying the persistence of the transgenic plant products (enzyme-linked immunoabsorbent assay, polymerase chain reaction) and for determining the effects on microorganisms (DNA fingerprinting, Biolog pure culture and community analysis, substrate-induced respiration assay, enzymatic assays, and culturing on selective media) are also evaluated. Finally, we make recommendations for the design and execution of risk assessment studies of transgenic plants.



Microcosm experiments with transgenic cotton producing the B.t.k. endotoxin

A method using an enzyme-linked immunoabsorbent assay (ELISA) was developed to allow extraction and quantification of B.t.k. endotoxin from soil.11 The importance of ionic and hydrophobic interactions between the soil and the B.t.k. toxin protein for recovery of the protein was investigated by varying different buffer components. The different buffers were used to extract purified B.t.k. toxin, and also toxin from the leaves of two lines of transgenic B.t.k. producing cotton, that were incorporated into three types of soil. The highest recovery of the B.t.k. toxin from soil occurred with a high salt, high pH buffer. For some of the soil types, the addition of a non-ionic detergent was required for optimal recovery. The percentage recovery of the B.t.k. toxin from soil was dependent upon the soil type and also on the toxin concentration in the soil. The highest percentage recovery was achieved with the low organic matter, high sand content soils. It was found that the higher the clay content of the soil, the lower the percentage recovery of the toxin.

The ELISA method was applied to evaluate the persistence of purified B.t.k. endotoxin and B.t.k. endotoxin in transgenic cotton plants.16 The five microcosm experiments were of 28-140 days duration and used initial B.t.k. endotoxin concentrations ranging from 1 to 1600 ng/g soil. In one of the experiments, after 28 days, 23% of the initial extractable toxin was present as measured by ELISA and 20% was present as measured by insect bioassays.10 In all of the experiments, there was a rapid decline in concentration of the extractable toxin during the first 7 days and then the decline became more gradual. In one of the experiments, half-lives were calculated from an exponential curve fit to the data points and showed that the B.t.k. endotoxin half-life in a fine sandy loam soil was 40 days and in a silt loam soil with high organic matter content was 22 days. In some of the experiments, after initial rapid declines, the low concentrations of B.t.k. endotoxin remained unchanged after several weeks; this was likely due to protection from degradation resulting from protein binding to clay particles. Sterile, irradiated soil was also used in one of the experiments to evaluate the role of biotic degradation in the decline of B.t.k. endotoxin concentration. The B.t.k. endotoxin was stable in the sterile soil, remaining at approximately 87% of the initial recoverable amount. In contrast, in the non-sterile soil the B.t.k. endotoxin concentration decreased over 35 days to 28% of the original recoverable concentration. These five microcosm experiments demonstrated that although high amounts of B.t.k. endotoxin did not persist in the soil, low levels may persist for several weeks or months. Such persistence of low concentrations of B.t.k. endotoxin would be of ecological concern if repeated use of B.t.k. endotoxin-producing plants led to an accumulation of the endotoxin and potential adverse impacts on soil organisms.

Field experiments with transgenic B.t.t. endotoxin-producing potato and proteinase-inhibitor-producing tobacco

The persistence of genomic DNA in decaying potato and tobacco plants in the field was measured using molecular methods, including polymerase chain reaction.17 Primers specific for a gene sequence present only in the transgenic potato or tobacco plants were used. The specificity was achieved by using forward primer sequences unique to a fragment of the cauliflower mosaic virus 35S promoter (in potato) or the nopaline synthase (NOS) promoter (in tobacco) and a 20 bp sequence of the N-terminal NPT-II coding region. The samples from the decaying potato plants, which were engineered to express the Bacillus thuringiensis var. tenebrionis cryIIIA protein, came from a field planting at the Oregon State University Agricultural Research & Extension Center in Hermiston, Oregon, USA. The tobacco samples were obtained from a buried litterbag field experiment performed at the E.P.A. laboratory in Corvallis, Oregon, USA, and came from tobacco that was engineered to express tomato proteinase inhibitor I. In the potato field, marker gene amounts declined after 84 days to 2.74% (leaf and stem) and 0.50% (tuber) of the initial levels and by the last sample day (137 days), the levels were 1.98% (leaf and stem) and 0.19% (tuber). In the tobacco litterbag experiment, after 14 days the marker DNA amount dropped to 0.36% of the original level and was detectable up until 77 days when it was at 0.06% of the initial amount. The results from this study demonstrated that plant genomic DNA can persist in soil under field conditions for several months.



Microcosm experiment with transgenic cotton producing the B.t.k. endotoxin

In this study, leaves of three different lines of cotton plants that were genetically engineered to produce the B.t.k. endotoxin were incorporated into soil obtained from a cotton growing area.12 Four sets of experiments, lasting 28 or 56 days, were performed in microcosms with combinations of the following treatments: 1) soil only; 2) soil + purified B.t.k. endotoxin; 3) soil + parental cotton; 4) soil + purified B.t.k. endotoxin + parental cotton; 5) soil + B.t.k. endotoxin-producing cotton. The concentration of the purified B.t.k. endotoxin that was used was comparable to the levels of B.t.k. endotoxin expressed in the transgenic plants.

Two of the three transgenic cotton lines produced a transient but significant increase in levels of culturable, aerobic bacteria and fungi. In contrast, the third transgenic line of cotton and the purified B.t.k. toxin did not have any significant effects. Transient changes in the bacterial species composition, as measured by Biolog metabolic pure culture analysis and community analysis, and DNA fingerprinting, were also associated with the two transgenic lines of cotton that stimulated bacterial and fungal populations.

Our conclusions from this study were that the B.t.k. endotoxins, both the purified and those produced in the transgenic plants, did not have a direct effect on the soil microorganisms. The plant line specificity of the response, and the lack of effects from the purified B.t.k. endotoxins, suggested that the effects were due to unexpected changes in plant characteristics, aside from the intended B.t.k. endotoxin production, that resulted from genetic manipulation or tissue culturing.

Field experiment with transgenic potato producing the B.t.t. endotoxin

A field experiment using potato plants genetically engineered to produce the Bacillus thuringiensis var. tenebrionis (B.t.t.) endotoxin was performed for 98 days.13 Total culturable, aerobic bacterial and fungal populations, fungal species diversity and abundance, and plant pathogen levels were measured in field plots of 1) commercial potato plants treated with standard chemical pesticides; 2) commercial potato plants treated with a commercial preparation of B.t.t., (M-Trak); and 3) transgenic B.t.t.-producing potato plants. Selective plating was used to determine bacterial and fungal populations on three phenological stages of potato leaves (green, yellow, and brown) and subcultures of the fungi were used for fungal species identification based on standard microbiological diagnostics. Plant pathogens were monitored by surveys and standard assays.

Our results indicated that the populations of culturable, aerobic bacteria and fungi, and the species of fungi, on the transgenic B.t.t.-producing potato plants differed minimally from those on the chemically and microbially treated commercial potato plants. There was a higher incidence of the plant pathogen Verticillium dahliae on the transgenic B.t.t.-producing plants but it was attributed to the longer viability of the transgenic plants.

Field experiment with transgenic tobacco producing proteinase inhibitor I

The decomposition of tobacco genetically engineered to produce the proteinase inhibitor I, a protein with insecticidal activity, was studied in a litterbag field experiment.14 Litterbags containing 1) parental tobacco leaves and 2) transgenic proteinase inhibitor I tobacco leaves, were buried in field plots and sampled, in addition to the soil surrounding the litterbags and soil from control plots without litterbags, at 2- to 4-week intervals over a 5-month period. The litterbag contents were analyzed for proteinase inhibitor concentration, litter decomposition rates, and carbon and nitrogen content, and the litterbag contents and surrounding soil were analyzed for microbial respiration rates, and population levels of nematodes, protozoa, and microarthropods.

The proteinase inhibitor I remained immunologically active in the plant litter for at least 57 days. Although the carbon content of the transgenic tobacco was comparable to that of the parental tobacco at the start of the experiment, it became significantly lower over the course of the experiment. The nematode populations in the soil surrounding the transgenic tobacco litterbags were greater and had a different trophic group composition, including a significantly higher ratio of fungal feeding to bacterial feeding nematodes, than in soil surrounding the parental tobacco. Populations of Collembola, however, were significantly lower in the soil surrounding the transgenic tobacco litterbags. Microbial respiration, measured by the substrate-induced respiration (SIR) assay, did not indicate significant differences between the parental and transgenic plant litter or among soil collected around the litterbags or in control plots without litterbags. The observed effects on the plant carbon content, and on the levels of nematodes and Collembola suggest, however, that longer term sampling may have revealed indirect effects on microbial populations.

Field experiment with transgenic alfalfa producing alpha-amylase or lignin peroxidase

In this study, we performed the first field release in the United States of transgenic plants with associated recombinant microorganisms by inoculating transgenic alfalfa plants with recombinant microorganisms and planting them in an agricultural field plot.15 The transgenic alfalfa plants produced alpha-amylase or lignin peroxidase, industrial compounds used in food and wood processing. Prior to planting, some of the transgenic and parental alfalfa plants were inoculated with wildtype Sinorhizobium meliloti or with S. meliloti that were genetically engineered for antibiotic resistances or for antibiotic resistances and enhanced nitrogen fixation capability. Analyses of the alfalfa plants and field plot soil were conducted over two growing seasons and included plant biomass and chemistries; soil chemistries and enzyme activities; Biolog metabolic analyses and DNA fingerprints of soil bacterial communities; soil microbial respiration; population counts of indigenous soil bacteria, fungi, nematodes, protozoa and microarthropods; and identifications of nematodes and microarthropods.

Few effects associated with the recombinant S. meliloti were observed. Several effects were associated with the transgenic alfalfa plants, however, particularly with the lignin peroxidase producing alfalfa. The transgenic lignin peroxidase alfalfa had significantly lower biomass, and higher nitrogen and phosphorous content, than the parental or transgenic alpha-amylase alfalfa. Substrate utilization of the bacterial communities, measured with the Biolog assay, indicated differences among the three plant types. The bacterial communities associated with the lignin peroxidase plants were the most unusual. Significantly higher levels of culturable, aerobic spore-forming and cellulose-utilizing bacteria, higher soil pH levels, and lower activity of the soil enzymes dehydrogenase and alkaline phosphatase were also associated with the lignin peroxidase alfalfa. We concluded that the major ecological effects were unintended alterations in plant characteristics and changes in soil chemistry and microbiology.



Our studies have shown that pesticidal proteins produced in transgenic plants can persist in soil and that binding of the proteins to soil particles can protect them from biotic degradation. We also found that plant genomic DNA in transgenic plants can persist in a field environment for several months.

Persistence of transgenic plant products in the soil, and retention of their biological activity, has also been evaluated by other researchers. Studies with insecticidal toxins from subspecies of Bacillus thuringiensis demonstrated that the toxins were adsorbed and bound on clays and humic acids in soil yet retained their immunological and insecticidal activity.18,19,20,21,22 In these studies it was shown that the binding of the toxins made them more resistant to microbial degradation. These studies also demonstrated that the truncated toxins, the forms which are expressed in transgenic plants, bound more readily to clay minerals and thus were more resistant to biodegradation than the protoxins, the forms that are contained in microbial insecticide preparations.22

Researchers have also considered the fate, persistence, and effects in plant and soil ecosystems of genomic DNA from transgenic plants. Genes conferring antibiotic resistance, which have been commonly used as selection markers in transgenic plants, have been of particular interest. Of most concern has been the potential for horizontal gene transfer of antibiotic resistance from transgenic plants to microorganisms.23,24 Antibiotic resistance due to microbial adaptation and misuse of antibiotics is already perceived as a major problem in the medical industry. Therefore, the additional potential source of antibiotic resistance generated by the use of transgenic plants is an important consideration.

Our studies on the effects of transgenic plants on soil microorganisms have demonstrated that both the population levels of bacteria and fungi, and the species composition of bacteria, can differ between parental and transgenic plants. It is important to note that these effects were all observed with the use of transgenic plants that produced insecticidal or industrial compounds, rather than antimicrobial compounds. It would be expected that the same studies performed with transgenic plants producing antimicrobial compounds would have resulted in more extensive or severe impacts.

Two recent studies by other researchers demonstrated that transgenic plants that produce opines can influence the composition of rhizosphere bacteria. One study showed that root exudation by a transgenic tobacco of a novel substrate conferred a selective advantage to soil bacteria that were able to utilize the novel substrate.25 When near-isogenic strains of Pseudomonas fluorescens, which differed only in their ability to utilize opines, were coinoculated onto opine-producing Nicotiana tabacum, the catabolizer strain of P. fluorescens obtained a significantly higher population level than that of the non-catobilizing strain. A second experiment demonstrated that transgenic plants that produced opines altered the populations of rhizosphere bacteria.26 The levels of bacteria able to utilize mannopine were 80 times higher in the rhizospheres of the transgenic legume, Lotus corniculatus, than in the rhizospheres of the non-engineered L. corniculatus.

Other research showed that beneficial fungi can be influenced by transgenic plants. A delay in the onset of colonization, and a decrease in the level of colonization, by the arbuscular mycorrhizae Glomus mosseae, was observed on transgenic tobacco plants with chitinase expression.27 Light microscopy revealed distinct differences in the fungal structures on the transgenic plants as compared to the control plants.



The results of our studies and those of other researchers highlight the importance of using reliable, sensitive, and multiple experimental methods to monitor for environmental effects from transgenic plants. We propose that as many methods as practical be used to monitor microorganisms because of the abundance, complexity, and great importance of microorganisms in plant and soil processes. In our studies, we have combined traditional microbiological plating on general and selective media, enzymatic activity assays, a substrate-induced microbial respiration assay, Biolog metabolic fingerprinting, and DNA fingerprinting to access changes in levels and communities of soil microorganisms. These methods provide different information and allow a more accurate and extensive assessment than possible with the use of any one method.

The traditional microbiological plating method and the more novel Biolog metabolic fingerprinting method both have the disadvantage of detecting only culturable microorganisms. As demonstrated in several studies, a large proportion of microorganisms in an environmental sample may be viable but will be non-culturable.28,29 The plating method does offer, however, a simple and inexpensive assessment of the microbial population and the use of selective media allows one to focus on microbial groups of particular interest. The metabolic fingerprinting method can provide detailed examinations of microbial populations by allowing the identification of individual isolates to species level. A more valuable use of the method, however, may be in generating patterns of substrate utilization for entire microbial communities. In this way, a large number of environmental samples can be relatively quickly evaluated for effects from transgenic plants.

The rDNA fingerprint analyses provide a unique molecular approach for evaluating complex microbial populations in environmental samples.30,31 An advantage of the analysis is that it is not dependent on culturing microorganisms. Primers may be selected to examine the taxonomic groups of greatest interest. A disadvantage of rDNA fingerprinting of complex populations of microorganisms is that subtle changes in banding patterns may go undetected. Furthermore, changes in closely related species may not be seen if their rDNA fragments are highly conserved.

The substrate-utilization assay provides an indirect measurement of the total microbial carbon biomass in an environmental sample. This may be used as an estimate of the microbial population level. The assay is fairly simple to perform and large numbers of samples can be analyzed in a short period of time. The addition of antibiotics has been used in the assay to distinguish the fungal and bacterial contributions to the measurements. Unfortunately, the results can be erratic, making it difficult to obtain statistically significant results.

Another approach for evaluating the effects of transgenic plants on microorganisms is to monitor microbial processes rather than population levels or taxonomic groups. One approach for monitoring microbial processes is to measure enzymatic activity. This can be done through direct measurements of enzymatic activity, such as our measurements of dehydrogenase and phosphatase activity in the transgenic alfalfa experiment. Such measurements can assist in the prediction of the effects of transgenic plants on nutrient cycling. For example, phosphatase is involved in soil organic phosphorous mineralization and plant nutrition, and dehydrogenase is used in the biological oxidation of organic compounds. Indirect measurements of enzyme activity can also be performed. The acetylene-reduction assay is an indirect measurement of nitrogenase activity and is used to estimate nitrogen fixation rates. We have successfully used the acetylene-reduction assay to compare the nitrogen fixation activity associated with different forest management practices and have been able to positively correlate the results with molecular analyses of the same samples for detection of the nifH gene for nitrogen-fixing capability.32 Advantages of this method are its sensitivity, low cost, and ease of performance. Drawbacks include possible overestimation of nitrogenase activity because acetylene is not the physiological substrate of nitrogenase and also other side effects of acetylene such as inhibition of other enzymatic processes.

Based on the methods we have discussed, and their various advantages and disadvantages, it becomes obvious that a single method should not be relied upon to give an accurate assessment of the impacts on microorganisms associated with the environmental use of transgenic plants. By combining several methods, more aspects of the microbial community can be examined and significant effects detected with one method can potentially be confirmed using additional methods.



Microorganisms play many vital roles in plant and soil ecosystems, including mineralization and immobilization of nutrients, biochemical degradation of organic matter, and also serve as food sources for other important organisms. Therefore, it is crucial that risk assessment studies on the environmental use of transgenic plants consider the impacts on microbial communities. Research in this area has been quite limited, however, as demonstrated by the few available references.

Many of the plants being engineered for agricultural or forestry use have antimicrobial properties. Due to natural wounding, senescence, and sloughing-off of root cells, along with tillage of plants into the soil, some of the antimicrobial proteins may be released into the soil ecosystem. Because part of the proteins can remain active due to protective adsorption to clay minerals or humic components, there is the potential for prolonged exposure of the microbial community to the antimicrobial compounds. The repeated use of transgenic plants in an area may also result in the accumulation of antimicrobial compounds in the ecosystem.

We suggest that risk assessment studies should evaluate the effects of transgenic plants on both microbial communities and processes. The studies should be of sufficient duration so that the persistence or accumulation of compounds detrimental to microorganisms can be detected. The studies should be designed to monitor for the most probable effects. For example, if the transgenic plants produce antifungal compounds then the focus of the study should be on the fungal community. In addition, the specific ecosystem should be considered and the study should be performed to account for impacts on the key players in that ecosystem. This would mean that if Pseudomonas species, for example, are particularly important in a specific ecosystem then their populations and processes would be evaluated over the course of the study. Finally, as discussed previously, the application of multiple methods will provide the most sensitive and comprehensive assessment of the potential ecological impacts of transgenic plants on microbial communities and processes.



1. Fraley, R. 1992. Sustaining the supply. Bio/Technol. 10, 40-43.

2. Levin, M. A. & Israeli, E. (Editors). 1996. Engineered Organisms in Environmental Settings - Biotechnological and Agricultural Applications. CRC Press, Boca Raton, FL.

3. Schechtman. 1998. Strategic regulation of agricultural biotechnology products in the United States. Proceedings of the 5th International Symposium on the Biosafety Results of Field Trials of Genetically Modified Plants and Microorganisms. 6-10 September 1998. Braunschweig, Germany.

4. Bergelson, J., Purrington, C.B. & Wichmann, G. 1998. Promiscuity in transgenic plants. Nature 395, 25.

5. Seidler, R. & Levin, M. 1994. Potential ecological and nontarget effects of transgenic plant gene products on agriculture, silviculture, and natural ecosystems: general introduction. Molec. Ecol. 3, 1-3.

6. Seidler, R.J., Watrud, L.S. & George, S.E. 1997. Assessing risks from GMO’s to ecosystems and human health. Handbook of Environmental Risk Assessment and Management, pp 110-146. Blackwell Science, England.

7. Watrud, L.S. & Seidler, R.J. 1998. Non-target ecological effects of plant, microbial, and chemical introductions to terrestrial systems. In Soil Chemistry and Ecosystem Health, P.M. Huang (Editor). Special Publication No. 52, pp 313-340. Soil Science Society of American, Madison, Wisconsin.

8. Glandorf, D.C.M., Bakker, P.A.H.M. & Van Loon, L.C. 1997. Influence of the production of antibacterial and antifungal proteins by transgenic plants on the saprophytic soil microflora. Acta. Bot. Neerl. 46, 85-104.

9. Salmeron, J.M. & Vernooij, B. 1998. Transgenic approaches to microbial disease resistance in crop plants. Curr. Opinion in Plant Biol. 1, 347-352.

10. Pratt. G.E., Royce, L.A. & Croft, B.A. 1993. Measurement of toxicity of soils following incorporation of plant residues engineered with Bacillus thuringiensis v. kurstaki endotoxin, using a Heliothis virescens growth bioassay. In Proceedings of the Fifth Investigators Meeting for EPA’s Environmental Release of Biotechnology Research Program, Duluth, MN.

11. Palm, C.P., Donegan, K., Harris, D. & Seidler, R.J. 1994. Quantification in soil of Bacillus thuringiensis v. kurstaki - endotoxin from transgenic plants. Molec. Ecol. 3, 145-151.

12. Donegan, K.K., Palm, C.J., Fieland, V.J., Porteous, L.A., Ganio, L.M., Schaller, D.L., Bucao, L.Q. & Seidler, R.J. 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.

13. Donegan, K.K., Schaller, D.L., Stone, J.K., Ganio, L.M., Reed, G., Hamm, P.B. & Seidler, R.J. 1996. Microbial populations, fungal species diversity and plant pathogen levels in field plots of potato plants expressing the Bacillus thuringiensis var. tenebrionis endotoxin. Transgen. Res. 5, 25-35.

14. Donegan, K.K., Seidler, R.J., Fieland, V.J., Schaller, D.L., Palm, C.J., Ganio, L.M., Cardwell, D.M. & Steinberger, Y. 1997. Decomposition of genetically engineered tobacco under field conditions: Persistence of the proteinase inhibitor I product and effects on soil microbial respiration and protozoa, nematode and microarthropod populations. J. of Appl. Ecol. 34, 767-777.

15. Donegan, K.K., Seidler, R.J., Doyle, J.D., Porteous, L.A., DiGiovanni, G., Widmer, F. & Watrud, L.S. 1999. A field study with genetically engineered alfalfa inoculated with recombinant Sinorhizobium meliloti: effects on the soil ecosystem. J. of Appl. Ecol. (in press)

16. Palm, C.P., Schaller, D.L., Donegan, K.K., & Seidler, R.J. 1996. Persistence in soil of transgenic plant produced Bacillus thuringiensis v. kurstaki endotoxin. Can. J. of Microbiol. 42, 1258-1262.

17. Widmer, F., Seidler, R.J., Donegan, K.K. & Reed, G.L. 1997. Quantification of transgenic plant marker gene persistence in the field. Molec. Ecol. 6, 1-7.

18. Crecchio, C. & Stotzky, G. 1998. Insecticidal activity and biodegradation of the toxin from Bacillus thuringiensis subsp. kurstaki bound to humic acids from soil. Soil Biol. Biochem. 30, 463-470.

19. Tapp, H. & Stotzky, G. 1998. Persistence of the insecticidal toxin from Bacillus thuringiensis subsp. kurstaki in soil. Soil Biol. Biochem. 30, 471-476.

20. Tapp, H. & Stotzky, G. 1995. Dot blot enzyme-linked immunoabsorbent assay for monitoring the fate of the insecticidal toxins from Bacillus thuringiensis in soil. Appl. Environ. Microbiol. 61, 602-609.

21. Tapp, H. & Stotzky, G. 1995. Insecticidal activity of the toxins from Bacillus thuringiensis subspecies kurstaki and tenebrionis adsorbed and bound on pure and soil clays. Appl. Environ. Microbiol. 61, 1786-1790.

22. Venkateswerlu, G. & Stotzky, G. 1992. Binding of the protoxin and toxin proteins of Bacillus thuringiensis subspecies kurstaki and tenebrionis on clay minerals. Curr. Microbiol. 25, 225-233.

23. Droge, M., Puehler, A., Selbitschka, W. 1998. Horizontal gene transfer as a biosafety issue: a natural phenomenon of public concern. J. Biotech. 64, 75-90.

24. Nielsen, K.M., Bones, A.M., Smalla, K. Van Elsas, J.D. 1998. Horizontal gene transfer from transgenic plants to terrestrial bacteria - a rare event? FEMS Microbiol. Rev. 22, 79-103.

25. Savka, M.A. & Farrand, S.K. (1997) Modification of rhizobacterial populations by engineering bacterium utilization of a novel plant-produced resource. Nature Biotechnol. 15, 363-368.

26. Oger, P., Petit, A. & Dessaux, Y. 1997. Genetically engineered plants producing opines alter their biological environment. Nature Biotechnol. 15, 369-372.

27. Vierheilig, H., Alt, M., Lange, J., Gut-Rella, M., Wiemken, A. & Boller, T. 1995. Colonization of transgenic tobacco constitutively expressing pathogenisis-related proteins by the vesicular-arbuscular mycorrhizal fungus Glomus mosseae. Appl. Environ. Microbiol. 8, 3031-3034.

28. Colwell, R.R., Brayton, P.R., Grimes, D.J., Roszak, D.R., Huq, S.A. & Palmer, L.M. 1985. Viable, but non-culturable Vibrio cholerae and related pathogens in the environment: implications for release of genetically engineered microorganisms. Biotechnol. 3, 817-820.

29. Wilson, M. & Lindow, S.E. 1992. Relationship of total viable and culturable cells in epiphytic populations of Pseudomonas syringae. Appl. Environ. Microbiol. 58, 3908-3913.

30. Massol-Deya, A., Weller, R., Rios-Hernandez, L., Zhou, J.Z., Hickey, R.F. & Tiedje, J.M. 1997. Sucession and convergence of biofilm communities in fixed-film reactors treating aromatic hydrocarbons in groundwater. Appl. Environ. Microbiol. 63, 270-276.

31. Porteous L.A., Seidler, R.J. & Watrud, L.S. 1997. An improved method for purifying DNA from soil for polymerase chain reaction amplification and molecular ecology applications. Molec. Ecol. 6, 787-791.

32. Shaffer, B.T., Widmer, F., Hornsby, M.J., Porteous, L.A., & Seidler, R.J. 1998. Characterization of the nifH gene pool and acetylene-reduction activities of N2 fixing bacteria in Oregon forests and clearcuts. Abstracts of the 98th General Meeting of the American Society for Microbiology, pg 366.



1 Dynamac Corporation

2 Environmental Protection Agency