Genetic engineering and the production of food stuffs

Biosafety Aspects

Beatrix Tappeser, presented at Discovery 98, international conference 28 -30.9.1998, Kulmbach, Germany

Food stuffs produced with the help of genetic engineering are reaching the market place. Industry and the main regulation bodies in Europe and the United States have decided that these foodstuffs are safe for the consumer and do not pose a threat for the environment. However, there is an ongoing worldwide discussion about the validity of these judgements.

To start let me emphasize that our knowledge about the safety and digestibility of traditional food is mainly based on the history of safe use and experience. This history of safe use cannot be transferred to genetically engineered food. There are at least four main reasons why this cannot be done.

Often described as extremely precise, the results of genetic engineering are only partly predictable. Position effects create secondary effects (dependent on the mostly uncontrolled integration of the new genetic material into the genome of the host organism) that affect the physiology of the organism in question and consequently the food produced from it. In the complicated network of an organism's physiology, pleiotropic effects of a newly introduced gene can result in unintentional changes in the contents or modifications of specific ingredients (toxic, allergenic or influencing the nutritional value) or in the synthesis of products not previously present.

Insufficiently characterized genetic material that is transferred through genetic engineering can also present a safety risk. Genes have been identified whose coding area only comprises 21 nucleotides and consequently could be transferred unknowingly. GONZALEZ-PASTOR et al., (1994) describe such a gene that codes for Microcin and is effective as an antibiotic.

The possibility of horizontal gene transfer is an especially important factor in evaluating the safety of food produced with the help of genetically engineered microorganism (OECD, 1993, p. 33). This applies especially to antibiotic-resistant genes. But not only microorganisms contribute to horizontal transfer, engineered plants can also. Gene constructs with replication units and promoter sequences effective in plants and microorganisms open up the possibility of transfer and expression in both organisms, a situation which is new in evolution and is found only in certain pathogens. Horizontal gene transfer is also of special concern when discussing the environmental safety of transgenic organism. As a case study I will explore this topic in more detail later on.

Another result of genetic engineering is that gene products and organisms that were not previously present in food are now used in food production. Consequently, we are confronted with an increasing number of completely new food components. The increase of variety in the food sector is considered one of the main reasons for the increase of food allergies (Dr. Przyrembel from the BgVV in an interview with Greenpeace Magazine 2/96).

A Swiss study which reviewed the allergenic potential of transgenic plants came to the conclusion that these increase the probability of an unwanted allergenic reaction. In essence these authors argued that while molecular breeding can use the totality of all organisms as gene-resources, traditional breeding was restricted to cross-hybridizing relatives with similar allergenic potential. The directed alteration of the expression levels of the species own genes is also only possible by genetic engineering. The risk of unwanted and unprecedented allergies is therefore higher in transgenic plants than in traditionally bred plants (FRANCK-OBERASPACH & KELLER, 1996).

Genetically modified microorganisms in food production.

It is now widely acknowledged that microorganisms can exchange genetic material on a varied manner though little is known on the factors such as selection pressure, nutrient requirements etc. which facilitate and promote transfer.

All those microorganisms genetically engineered and used in food production may contribute to transfer events. If they are used as such they may be part of transfer events in the gastrointestinal tract of human beings or animals. If they are used in food production they can contribute to horizontal gene transfer via waste water and sewage.

Current security measures for research and production are based on earlier assumptions about the survival and transfer abilities of microorganisms. These assumptions are no longer valid. It has been shown that GMOs have the capacity to survive and persist in very different environments as it has been shown for recombinant DNA. At the European level the contained use directive is under revision. Instead of adapting the directive to the new scientific data a further deregulation is planned. Whole groups of organisms shall be exempted of any regulation especially those „generally regarded as safe" and used in food production. I will present two examples why this is extremely premature and violates the precautionary principle adapted by the European Union as a whole and additionally by each European member state when signing the Rio Declaration of 1992.

The example is on self-cloning in a yeast what means genes of the yeast were duplicated and then reintroduced with the help of genetic engineering. Duplication or amplification of genes is an extremely important approach expecially in the production of additives, enzymes, amino-acids, vitamins and other substances used in food processing.

INOSE AND MURATA (1995) experimented with the yeast S. cerevisiae and discovered that a mere threefold increase of an enzyme in the glycolytic pathway (phosphofructokinase) resulted in a 40-fold increase (even 200-fold in certain mutants) of methylglyoxal (MG), a toxic substance that demonstrates mutating effects in the Ames test. They expressly point out the safety implications of these results regarding the use of GMO in the food sector. The experiments incontestably prove the relevance of the position and pleiotropic effects I already mentioned. They also prove that self-cloning provides no decisive proof of safety.

Inose and Murata conclude their publication with the following paragraph:

„Although, except for the case of microbes, we have no information as to the toxic effect of MG in foods on human beings, the results presented here indicate that, in genetically engineered yeast cells, the metabolism is significantly disturbed by the introduced genes or their gene products and the disturbance brings about the accumulation of the unwanted toxic compound MG in cells. Such accumulation of highly reactive MG may cause a damage in DNA, thus suggesting that the scientific concept of „substantially equivalent" for the safety assessment of genetically engineered food is not always applied to genetically engineered microbes, at least in the case of recombinant yeast cells. […] Thus, the results presented may raise some questions regarding the safety and acceptability of genetically engineered food, and give some credence to the many consumers who are not yet prepared to accept food produced using gene engineering techniques" (INOSE & MURATA, 1995).

The second example is on the possible ecological impacts of genetically modified microorganisms. A soil organism named Klebsiella planticola was engineered to produce ethanol out of plant residues. When tested for survival capacity and competitiveness in different soils no differences to the parent strain could detected. The engineered microbe persisted in any soil where it was introduced. Further experiments looked for the effect of these modified organisms on plant growth. The addition of the engineered organism coincided in some soils with the death of the planted wheat plants - a very unexpected result (HOLMES et al., 1998). This sheds light on the practise of tolerated release out of production plants as a potentially dangerous action.

Horizontal gene transfer

For a long time it was assumed that it is impossible for recombinant plant DNA to establish itself in microorganisms. In the meantime, however, the successful integration and expression of plant transgenes in microorganisms has been demonstrated. HOFFMANN et al. (1994) have proven horizontal gene transfer from plants to fungi, while GEBHARD & SMALLA (1998) and WACKERNAGEL (1997, personal communication) have demonstrated the transfer of plant transgenes to bacteria. The latter two study groups have shown that bacteria bearing a genetic defect are capable of integrating an intact (recombinant) gene previously inserted into a plant by homologous recombination from the transgenic plant genome.

It is reasonable to presume horizontal gene transfer of recombinant plant transgenes to be more likely than that of native plant genes because there are often of bacterial origin or at least have sequences that are derived from bacteria and contain promoters that can be used by bacteria, and include a bacterial origin of replication (oriV). These preconditions are encountered in many plants that have already been approved for commercialisation. A further precondition for successful transfer is that transgenes are present in substantial quantity. This is the case in any transgenic plant because their transgenes are contained in every cell. When these plants are ploughed in after the harvest it has been shown that DNA remains available for gene transfer for a relatively long period because binding to soil particles stabilises free DNA and protects it against degradation (for an overview of the meanwhile very extensive literature on the persistence of isolated DNA and gene transfer processes see ECKELKAMP et al., 1997b). In spite of all this, successful horizontal transfer of recombinant genes from plants to microorganisms is probably rare. On the other hand, when a newly integrated gene does offer a selective advantage, it may well establish itself in a population because, as is now known, the effectivity of interbacterial gene transfer rises sharply in the presence of selective pressure (HEINEMANN, 1991; ECKELKAMP et al., 1997a,b). The low frequency of horizontal gene transfer itself may therefore be of lesser significance than the ensuing selection process (HEINEMANN, 1997). In view of this, calculations on the probability of gene transfer occurrences such as have been performed by SCHLÜTER et al. (1995), are of little informative value and, what is more, permit no conclusion as to their true hazard potential. I like to cite Heinemann who stated at a seminar of the Norwegian Biotechnology Advisory Board on this point as follows: "The extent and consequences of horizontal gene transfer are apparent in the evolution of antibiotic resistant microorganisms and evidence suggests that horizontal gene transfer may be equally frequent among multicellular eukaryotic organisms. But the actual and potential frequencies of gene transfer are poor indicators of risk; very common genes are not maintained in nature if unselected and rare genes become common extremely quickly if they are the subject of selection. What remains essential to assessing risk is identifying all potential selective pressures a recombinant gene might be suited to neutralise. New evidence suggests that current knowledge of evolutionary theory is inadequate to predict the fate of recombinant organisms or recombinant genes." (HEINEMANN, 1997, p.17). Past safety studies on horizontal gene transfer have given little attention to the role of selective pressure (NIELSEN, 1997). Transfer probability calculations are in fact variously seen to be based on experiments totally devoid of selective pressure (e.g., in SCHLÜTER et al., 1995).

So it is impossible to define in advance so called safe-organisms or safe genes. That always depends on the context. An organism without problems in one environment may pose great problems in another environment. That may depend on numbers, on possibilities to counteract, on predators, on coevolution features, but it has to be evaluated for any case in any environment.

Horizontal gene transfer in the gastrointestinal tract of insects and vertebrates

In contrast to earlier views of long standing, DNA is not fragmented in the intestine but rather remains stable surprisingly long. DNA ingested with food can be excreted after only partial digestion. Moreover, it can also pass into the blood to be taken up by leukocytes and cells of the liver and spleen (SCHUBBERT et al., 1994, 1997a). The quoted experiments were performed on naked DNA. DNA ingested with food is normally complexed with proteins and better protected in this form than naked DNA. Beside this, the environment of the human or animal gastrointestinal tract changes in the course of digestion depending on what types of food are eaten together. This means that the resistance of proteins or DNA to digestion is not always constant but can vary. Laboratory experiments on the degradability of DNA in synthetic gastrointestinal liquids as they are usually carried out in studies on transgenes are blind to such effects because they use a constant pH and purified DNA (ECKELKAMP et al., 1997a,b). It has not been possible to date to demonstrate "natural" transformation in simulated mammalian gastrointestinal liquids. This may be attributable to the relatively short half-time of DNA in the chosen experimental set-ups, which is roughly 10 min and hence substantially less than the DNA stability found during passage through the gastrointestinal tract in mice. Whether this greater "in vivo" stability of DNA is sufficient to permit horizontal gene transfer via transformation remains to be examined. ORPIN et al. (1986) were at least able to show that Selenomonas ruminantium, a bacterial species inhabiting the bovine gastrointestinal tract, is naturally transformable. Furthermore, TEBBE et al. (1994) in their laboratory studies found that gene transfer can occur from orally administered GMMs to various recipients (Arthrobacteria) via transformation in the intestine of springtails (Folsomia candida). PERRETEN et al. (1997) documented the development of a plasmid bearing multiple antibiotic resistances which they had isolated from raw milk cheese. These resistances originated from four different microorganisms and probably developed through the action of antibiotics in the microflora of the lactating cows. The authors take a clear stance on the issue: "To preserve the life-saving potential of antibiotics, the spread of resistance genes at all levels must be stopped. Distribution routes like those between animals, food and consumers have to be interrupted." (PERRETEN et al., 1997 p.802).

Vertical gene transfer shown for the example of naturalisation and outcrossing of rape

Another central issues that has figured in the discussion on the cultivation of transgenic useful plants since its very beginning is that of outcrossing of such plants and introgession of the recombinant genes to related weed and wild plants. It was more or less agreed at least in the beginning of the debate that pervasive spread of transgenes should be avoided as best possible, as this may have problematic effects on species composition and biocoenoses in general. A further point now attracting increasing attention are the implications of resistance development through outcrossing (e.g. herbicide resistance or insect resistance), as this may not only have consequences for non-cultivated ecosystems but in particular also for agricultural land use systems.

For a long time also the probability of rape outcrossing to neighbouring rape crops was greatly underestimated. This may be attributable to the fact that the hybridisation behaviour of rape varies depending on its hereditary character and on environmental parameters. Moreover, diverging views on the rate of hybridisation can also result from differences in experimental design. The wide range of data to be found in the literature on the rate of cross-fertilisation in rape thus becomes plausible (SCHEFFLER et al., 1993; FELDMANN, 1997; TIMMONS et al., 1995a).

The results of hybridisation experiments also demonstrate the possibility of a gene flow from rape to wild herb populations. Potential hybridisation partners of Brassica napus are not only to be found in the genus of Brassica but also in other groups of the mustard family (SCHEFFLER & DALE, 1994). Potential hybridisation partners of rape include in particular wild herbs, which are probably all subject to a high degree of cross-fertilisation. According to DARMENCY (1994) this cross-fertilisation facilitates the transmission of transgenes from rape to associate herbs. Under field conditions rape has proven capable of hybridisation with wild turnip (Brassica rapa), brown mustard (Brassica juncea), black mustard (Brassica nigra) hoary mustard (Hirschfeldia incana, synonymous with Brassica adpressa), wild radish (Raphanus raphanistrum) and wild mustard (Sinapis arvensis) (reported in detail in ECKELKAMP et al., 1997c).

All experience and data gained in the course of the past years point to a high probability of rape populations prevailing outside cultivated areas and the possibility of gene flow to non-transgenic populations and related wild herbs. Many of Europe's major useful plant species have been equipped with the same herbicide resistance genes. Their large-scale use will therefore produce an enormous selective pressure towards similarly resistant weeds. While rape will be the plant to initiate rapid resistance development, other plants equipped with the same resistance but lacking crossable wild relatives in the region will sustainedly promote the one-sided selection of weeds rendered resistant by the former. This development will also be accompanied by a further impoverishment in farmland-associated floral species and insects.

Furthermore, the cloning of multiple resistances into one and the same useful plant species also gives related wild herbs the opportunity to acquire multiple resistance.

The large-scale resistance management schemes now being discussed in anticipation of herbicide resistance problems may have to accommodate whole regions and extend over several rotation periods in order to be effective (KORELL et al., 1997). That will need a high planning and control effort.

No less in conflict with the requirements of sustainability, and with the principles of sustainable utilisation and conservation laid down in the Convention on Biological Diversity, is the endangerment of species diversity entailed in the present herbicide resistance strategies. In the light of the knowledge on horizontal and vertical gene transfer that has accumulated during the past years, the rapid commercialisation of a multitude of herbicide-resistant transgenic useful plants also constitutes a violation of the precautionary principle. Speakers at international debates are often heard to invoke another principle, namely that decisions should only be made on the basis of scientific knowledge. There is no objection to this, just so long as such knowledge-based decisions really take account of all the relevant scientific evidence available.

In concluding I would like to quote Heinemann again:

"The risk of genetically engineered organisms for commercial preparation is the potential for the engineered product to demonstrate unexpected "monster" qualities or for the genes to escape into the wild fauna and flora and thereby create genetic "monsters". These are risks which cannot be excluded with present data. The potential for escape of a resistance gene introduced into the genome of a commercially desirable plant cannot be gauged by small-scale gene escape experiments. By known examples of gene transfer frequencies in nature, the potential for exchange is too great to be excluded by argument. (...) What we know the least about is the nature of the evolutionary forces that determine the success or failure of monsters. We have limited or no predictive power for the fate of recombinant genes. In the case of an antibiotic resistance, we may be able to say that its known functions pose no additional threats. But we cannot be sure that its known functions are all of its potential functions either or its own or in conjunction with the many novel genomic contexts in which it might be found should it be transferred horizontally. Therefore, as a qualitatively different technology, genetic engineering should be commercialised with extreme caution until the appropriate scientific experiments can be conducted" (HEINEMANN, 1997, p.23).

Given all these scientific uncertainties and contradicting views, what does that mean for regulation? Bacteria may have very different effects in different environmental settings, especially in ecosystems belonging to other geographical regions and climates. Each plant has somewhere in the world wild and weedy relative. Outcrossing and introgression of transgenes will happen. We need thorough internationally binding impact assessment schemes and a regime of prior informed consent for the introduction of any transgenic organism, as it is recommended by developing countries and others in the context of the negotiations on a biosafety protocol. And it should be open for each country to decide not to have transgenic crops in its landscapes without injuring international trait contracts.

We need clear and strict labeling so that any organism can be traced throughout the production and trade line and each consumer can decide individually. It will not be possible to give a hundred percent safety because of remaining scientific uncertainity and open questions which can be answered only on an empirical basis and the growing evidence on different possibilities of longterm negative health impacts.

The practice of mixing transgenic with non-transgenic varieties as it is done by American suppliers is one of the reasons why European consumers and the public at large are very upset. This practice should be stopped immediately.

Moreover, regulation systems, especially if they aspire to world-wide validity, have to reflect the different values and perceptions of different societies also in the evaluation of scientific findings. Crucial to this is the meaning of sound science, and the „right" evaluation, and who is defining what is accepted as sound science, and how much prevention and precaution is wanted by the public and will be accepted by science and industry.


DARMENCY, H. (1994): The impact of hybrids between genetically modified crop plants and their related species: introgression and weediness. Molecular Ecology 3, 37-40

ECKELKAMP, C.; JÄGER, M.; TAPPESER, B. (1997b): Verbreitung und Etablierung rekombinanter DNA in der Umwelt. Gutachten im Auftrag des Umweltbundesamt Berlin.

ECKELKAMP, C.; JÄGER, M.; WEBER, B. (1997a): Antibiotikaresistenzgene in transgenen Pflanzen, insbesondere Ampicillin-Resistenz in Bt-Mais. Öko-Institut e.V., Freiburg.

ECKELKAMP, C.; MAYER, M.; WEBER B. (1997c): Basta-resistenter Raps. Vertikaler und horizontaler Gentransfer unter besonderer Berücksichtigung des Standortes Wölfersheim-Melbach. Werkstattreihe Nr. 100, Öko-Institut e.V., Freiburg.

FELDMANN, S. (1997): Begleituntersuchungen des Landes Niedersachsen zur Freisetzung gentechnisch veränderter, herbizidresistenter Rapspflanzen in Gehrden. Abstract und Vortrag anläßlich der Fachgespräche "Stand der Sicherheitsforschung zur Freisetzung gentechnisch veränderter Pflanzen", 16. Dezember 1997 im Hanns-Lilje-Haus in Hannover.

FRANCK-OBERASPACH, S.L.; KELLER, B. (1996): Produktsicherheit von krankheits- und schädlingsresistenten Nutzpflanzen: Toxikologie, allergenes Potential, Sekundäreffekte und Markergene. In: Schulte E, Käppeli O (eds.) Gentechnisch veränderte krankheits- und schädlingsresistente Nutzpflanzen. Eine Option für die Landwirtschaft? Band I, Materialien, Publikation des Schwerpunktprogramms Biotechnologie des Schweizerischen Nationalfonds, Bern

GEBHARD, F.; SMALLA, K. (1998): Transformation of Acinetobacter sp. strain BD413 by transgenic sugar beet DNA. Applied and Environmental Microbiology (in press)

GONZALEZ-PASTOR, J.E.; SAN MILLAN, J.L.; MORENO, F. (1994): The smallest known gene; Nature 369: 281

HEINEMANN, J.A. (1997): Assessing the risk of interkingdom DNA transfer. In: Nordic seminar on antibiotic resistance marker genes and transgenic plants. Norwegian Biotechnology Advisory Board (ed.), Oslo, p. 17-28

HEINEMANN, J.A. (1991): Genetics of gene transfer between species. Trends in Genetics 7, 181-185

HOFFMANN, T.; GOLZ, C.; SCHIEDER, O. (1994): Foreign DNA sequences are received by a wild-type strain of Aspergillus niger after co-culture with transgenic higher plants. Current Genetics 27, 70-76

HOLMES, M.T.; INGHAM, E.R.; DOYLE, J.D.; HENDRICKS, C.W. (1998): Effects of Klebsiella planticola SDF20 on soil biota and wheat growth in sandy soil. Applied Soil Ecology 326, 1-12

INOSE, T.; MURATA, K. (1995): Enhanced accumulation of toxic compound in yeast cells having high glycolytic activity: A case study on the safety of genetically engineered yeast; International journal of food science and technology 30: 141-146

KORELL, M.; SCHITTENHELM, S.; WEIGEL, H.J. (1997): Aufstellen von Kriterien für die nachhaltig umweltgerechte Nutzung gentechnisch veränderter Kulturpflanzensorten. Umweltbundesamt, Texte 88/97

NIELSEN, K.M. (1997): Horizontal gene transfer from genetically modified plants (GMP) to soil associated bacteria. In: Nordic seminar on antibiotic resistance marker genes and transgenic plants. Norwegian Biotechnology Advisory Board (ed.), Oslo, 29-37

OECD (1993): Safety evaluation of foods derived by modern biotechnology. Concepts and principles.

ORPIN, C.G.; JORDAN, D.J.; HAZLEWOOD, G.P.; MANN, S.P. (1986): Genetic transformation of the ruminal bacterium Selenomonas Ruminantium. J. Appl. Bacteriol. 61:xvi

PERRETEN, V.; SCHWARZ, F.; CRESTA, L.; BOEGLIN, M.; DASEN, G.; TEUBER, M. (1997): Antibiotic resistance spread in the food. Nature 389, 801-802

SCHEFFLER, J.A.; DALE, P.J. (1994): Opportunities for gene transfer from transgenic oilseed rape (Brassica napus) to related species. Transgenic Research 3, 263-278

SCHEFFLER, J.A.; PARKINSON, R.; DALE, P.J. (1993): Frequency and distance of pollen dispersal from transgenic oilseed rape (Brassica napus). Transgenic Research 2, 356-364

SCHLÜTER, K.; FÜTTERER, J.; POTRYKUS, I.(1995): „Horizontal" gene transfer from a transgenic potato line to a bacterial pathogen (Erwinia chrysanthemi) occurs - if at all - at an extremely low frequence. Biotechnology 13, 1094-1098

SCHUBBERT, R.; LETTMANN, C.M.; DOERFLER, W. (1994): Ingested foreign DNA persists in the gastrointestinal tract and enters the bloodstream of mice. Molecular and General Genetics 242, 495-504

SCHUBBERT, R.; RENZ, D.; SCHMITZ, B.; DOERFLER, W. (1997a): Foreign (M13) DNA ingested by mice reaches peripheral leucocytes, spleen, and liver via the intestinal wall mucosa and can be covalently linked to mouse DNA. PNAS 94, 961-966

TEBBE, C.C.; VAHJEN, W.; MUNCH, J.C.; FELDMANN, S.D.; NEY, U.; SAHM, H.G.; AMORE, R.; HOLLENBERG, C.P. (1994): Überleben der Untersuchungsstämme und Persistenz ihrer rekombinanten DNA. BioEngineering 6/94, 14-21

TIMMONS, A.M.; O`BRIEN, E.T.; CHARTERS, Y.M.; DUBBELS, S.J.; WILKINSON, M.J. (1995a): Assessing the risk of wind polination from fields of genetically modified Brassica napus ssp. oleifera. Euphytica 85, 417-421

WACKERNAGEL, W. (1997): pers. Mittlg. UBA Workshop, November, 1997.