The hazards of genetically-engineered foods

With the release of genetically-engineered food on to the world market, a new source of diseases has emerged. Dr Mae-Wan Ho explains why such foods are unsafe.

I WAS, for some time, a molecular geneticist and taught the subject for the Open University until seven years ago when I changed my field of research, just as commercial gene biotechnology was taking over the subject. I began to review the literature again in 1994 as a member of an international group of scientists helping the non-government organisation, Third World Network, assess UN policy on gene biotechology. We produced a Scientists' Statement on what we perceived to be the ecological, socio-economic and health hazards of gene biotechnology, calling for a moratorium on commercial releases of transgenic organisms and immediate action on establishing legally binding international biosafety regulation.

A number of us also put together an independent experts' report on biosafety last year after we lost confidence in the official UN experts' report. Apart from the hazards, there are also many global ethical issues, including the patenting of life and the intellectual property rights of indigenous peoples, which are dealt with elsewhere.

Today, I shall concentrate on genetically engineered foods. By that, I include both food produced with genetically-engineered additives, and transgenic food plants and animals such as the Flavr Savr tomato and Zeneca's tomato puree which is now on the shelves in Safeway and Sainsbury, and animals such as the transgenic salmon produced in Canada and now reared in Scotland. I am particularly concerned about transgenic foods though genetically- engineered food additives are already problems by themselves, like Bovine somstotropin (Bst) milk featured in the last issue of The Splice of Life.

Built-into hazards

My thesis is that the hazards of transgenic foods are built into the technology, and that new evidence confirms this, suggesting that transgenic foods are neither safe to grow nor safe to eat. Therefore, there is no case for relaxing existing, already inadequate, guidelines for environmental releases of transgenic organisms, and for marketing transgenic foods. On the contrary, a moratorium on both environmental releases of transgenic organisms and marketing of transgenic foods should be imposed as a precautionary measure until the evidence can be fully assessed, and appropriate legally binding biosafety regulations firmly established.

Genetic engineering bypasses conventional breeding by using artificially constructed parasitic genetic elements as vectors to carry and smuggle genes into cells. Once inside cells, these vector slot themselves into the host genome. In this way, transgenic organisms are made carrying the desired transgenes. The most common vectors are a mosaic recombination of natural genetic parasites from different sources, including viruses causing cancers and other diseases in animals and plants, and tagged with one or more antibiotic resistance 'marker' genes.

Unlike natural parasitic genetic elements which have various degrees of host specificity, vectors used in genetic engineering are designed to overcome species barriers, and can therefore infect a wide range of species. Critics have warned that these vectors in the transgenic organisms constitute major sources of genetic pollution with drastic ecological and public health hazards that cannot be contained, once the transgenic organisms are released into the environment.

Genetic engineering is also known as recombinant DNA or rDNA technology, as it uses enzymes to cut and join, and therefore recombine genetic material from different sources. Let me summarise why rDNA technology differs radically from conventional breeding methods.

1. rDNA technology recombines genetic material in the laboratory between species that have very little probability of exchanging genes otherwise.

2. While conventional breeding methods shuffle different forms (alleles) of the same genes, rDNA technology enables completely new (exotic) genes to be introduced with unpredictable effects on the physiology and biochemistry of the transgenic organism. The insertion of foreign genes into the host genome is known to have many harmful and fatal effects including cancer.

3. Gene transfers are mediated by vectors which have three undesirable characteristics:

a. they are derived from disease-causing viruses, plasmids and mobile genetic elements - parasitic DNA that have the ability to invade cells and insert themselves into the cell's genome. In plant genetic engineering, the vector most widely used is derived from a tumour-inducing plasmid carried by the bacterium Agrobacterium tumefaciens. In animals, the most common vectors are constructed from retroviruses which are known to cause cancers and other diseases.

b. they are designed to breakdown species barriers so that they can shuttle genes between a wide range of species. Their wide host range means that they can infect many animals and in the process pick up genes from viruses of all these species to create new pathogens. Thus, a vector currently used in fish has a framework from the Moloney murine leukaemic virus, which causes leukaemia in mice, but can infect all mammalian cells. It has bits from The Rous Sarcoma virus, causing sarcomas in chickens, and from the vesicular stomatitis virus, causing oral lesions in cattle, horses, pigs and humans.

c. they carry genes for antibiotic resistance. This will speed up the evolution of antibiotic resistance which is already a big public health problem.

Vectors, the real danger

The vectors for gene transfer are where most of the dangers lie. Unlike ordinary pieces of DNA, they are resistant to enzymic degradation, and can survive indefinitely and independently in the environment where they infect cells, multiply in them, and jump in and out of their genomes.

Much of the current concern regarding the health hazard of transgenic foods centres on toxicity or allergies from the exotic gene, while the ecological hazards are focussed on the secondary gene transfer by conventional hybridisation of transgenic plants with weedy relatives. The role of vector-mediated horizontal gene transfer by infection has been down-played or ignored in current guidelines, and is not generally monitored in field releases. This is most unfortunate in view of the rapid advances in genetics within the past 20 years, which so radically alters the subject that it is legitimate to contrast the old, pre-rDNA genetics with the new post-rDNA genetics.

I shall focus on the last feature, horizontal gene transfer, which was known to be widespread among bacteria and viruses for at least 20 years. Microbes are completely promiscuous in their mating (conjugation). Moreover, a host of parasitic DNA can ferry genes across during the mating process, or independently by transduction, and bits of DNA can also be directly taken up from the environment by transformation. The parasitic genetic elements can jump between cells, slot in and out of the genome, multiply in cells, and exist in a dormant state almost indefinitely in the environment. As they slot in and out of genomes, they disrupt gene function and also take with them genes of the cell or leave other previously acquired genes behind. There are three kinds of parasitic elements - viruses, plasmids and mobile genetic elements - mosaics of all of them currently employed to transfer genes in transgenic technology. Viruses are probably the most infectious as they do not require cell-to-cell contact for infection and can persist in the environment indefinitely. Plasmids and mobile genetic elements are generally exchanged by cell-to- cell contact during conjugation or when one cell ingests (or phagocytoses) another.

For a long time, geneticists supposed that horizontal gene transfers did not involve higher organisms, and certainly not organisms like ourselves, because there are genetic barriers between species and viruses and other genetic parasites are species-specific. After all, genetic engineering involves constructing mosaic vectors to overcome those barriers so that genes can be ferried across kingdoms of organisms.

Within the past two years, however, the full scope of horizontal gene transfer is slowly coming to light. I have done a computer search under 'horizontal gene transfer' and came up with 68 references published in prestigious journals between 1993 and 1996, all but one giving direct or indirect evidence of horizontal gene transfers. Transfers occur between very different bacteria, between fungi, between bacteria and protozoa, between bacteria and higher plants and animals, between fungi and plants, between insects... in short, as one paper states,'The threat of horizontal gene transfer from recombinant organisms to indigenous ones is..very real and mechanisms exist whereby, at least theoretically, any genetically engineered trait can be transferred to any prokaryotic organism and many eukaryotic ones.' If you follow those arrows, you will realise how a gene transferred to any species in a vector can reach every other species on earth, the microbial/viral pool providing the main genetic thoroughfare.

It must be stressed that although horizontal gene transfers have occurred in our evolutionary past, they were relatively rare events among multicellular plants and animals. However, horizontal gene transfer is now made much more likely because the vectors constructed for genetic engineering are designed to infect a wide range of host cells.

Antibiotic resistance

Among those 68 references are documentations for the rapid spread of antibiotic resistance genes carried on plasmids among bacterial populations. As you know, multi-drug antibiotic resistance is already endemic in many UK hospitals. The transgenic tomatoes currently marketed here and the US both carry genes for kanamycin resistance. Kanamycin is widely used to treat tuberculosis which is coming back all over the world including Europe. The single reference which dismisses horizontal gene transfer is a review produced by the staff of Calgene, assuring us that the kanamycin resistance gene used in the Calgene transgenic tomato is completely safe.

As pathogens become antibiotic resistant they also exchange and recombine virulence genes by horizontal gene transfer thereby generating new virulent strains of bacteria and mycoplasm. This has been shown for Vibrio cholerae involved in the new pandemic cholera outbreak in India, Streptococcus involved in the world-wide increase in frequency of severe infections including the epidemic in Tayside Scotland in 1993, and Mycoplasma-genitalium , implicated in urethritis, pneumonia, arthritis, and AIDS progression.

Horizontal gene transfers have been directly demonstrated between bacteria in the marine environment , in the freshwater environment and in the soil. The aquatic environments are known to contain some 108 or more virus particles per millilitre, all capable of transferring genes, of helping endogenous 'crippled' vectors move and recombining with them to generate new viruses. Transfer of transgenes have been experimentally demonstrated from potato to a bacterial pathogen, and between transgenic plants and soil fungi.

Transgenic organisms now include all major crop-plants, engineered to be resistant to herbicides, or to insect pests with transgenes producing a bacterial poison, the Bt toxin, which unfortunately, also attacks many non-pest species. Field trials have shown that herbicide resistance transgenes can spread to weedy relatives within a single growing season, while Bt resistance evolved rapidly among major insect pests due to the continuous presence of Bt toxin in the transgenic plants. Ecologists such as Jane Rissler and Margaret Mellon who have opposed the release of transgenic organisms since the 1980s, have predicted those ecological effects .

A potentially greater danger

A potentially even greater danger lies in the vector-mediated gene transfer, as recent evidence suggests. An obvious route for the vectors to spread - which is not adequately taken into account in existing guidelines - is by infecting the teeming microbial populations in the soil, where transgenic plants are grown, and in aquatic environments, where transgenic fish and shellfish are currently being developed for marketing. These microbial populations form large reservoirs supporting the multiplication of the vectors, enabling them to spread to all other species. There will also be ample opportunity for the genetic elements to recombine with other viruses and bacteria to generate new genetic elements and pathogenic strains of bacteria and viruses, which will, at the same time, be antibiotic resistant.

Another major class of transgenic plants is now engineered for resistance to viral diseases by incorporating the gene for the virus' coat protein. Viruses are notoriously rapid in their mutation rate. They play a large role in horizontal gene transfer between bacteria and also exchange genes among themselves thus increasing their host range. Molecular geneticists have expressed concerns that transgenic crops engineered to be resistant to viral diseases might generate new diseases by recombination. In a study to test this possibility, Nicotiana benthamiana plants expressing a segment of a cowpea chlorotic mottle virus (CCMV) gene were inoculated with a mutant CCMV missing that gene. The infectious virus was indeed regenerated by recombination. As plant cells are frequently infected with several viruses, recombination events will occur and new and more virulent strains can be generated.

Are transgenic foods safe to eat?

In the light of our new knowledge, one must also ask whether transgenic foods are safe to eat. Although natural viruses and other parasitic genetic elements are to varying degrees specific in the range of host cells they will infect or multiply in, current transgenic vectors are designed to overcome species barriers so that they are much more likely to infect a wide range of hosts. In a study to test for the ability of bacterial viruses and plasmids to infect mammalian cells, it was found that plasmids of E. coli carrying the complete poliovirus can be transferred to mammalian cells and the polioviruses recovered from the cells, even though no eukaryotic signals for reading the genes are contained in the plasmid. In the same paper, the authors review experimental observations made since the 1970s that the lambda phage of bacteria, and the baculovirus, supposedly specific for insect cells, are also efficiently taken up by mammalian cells; and in the case of the baculovirus, transported to the cell nucleus. Similarly, E.coli plasmids carrying the complete Simian virus (SV40) genome were also taken up simply by exposing the cell culture to a bacterial suspension. Mammalian cells accept these foreign DNA parasites so well because they phagocytose bacteria and viral particles directly.

It has long been assumed that our gut is full of enzymes which can digest DNA. However, genes carried by vectors are especially resistant to enzyme action, and are much more infectious than ordinary bits of DNA. In a study designed to test the survival of viral DNA in the gut, mice were fed DNA from a bacterial virus, and large fragments were found to survive passage through the gut and to enter the bloodstream. Within the gut, vectors carrying antibiotic resistance will be taken up by the gut bacteria, which would then serve as a reservoir of antibiotic resistance for invading pathogenic bacteria.

The rapid spread of antibiotic resistance markers has been documented in a long-term study carried out in Eastern Germany. In 1982, streptothricin was administered to pigs. By 1983, plasmids encoding streptothricin resistance were found in the pig gut bacteria. This has spread to the gut bacteria of farmworkers and their family members by 1984, and to the general public and pathological strains of bacteria the following year. The antibiotic was withdrawn in 1990. Yet the prevalence of the resistance plasmid has remained high when monitored in 1993, confirming the ability of microbial populations to serve as stable reservoirs for horizontral gene transfer and recombination.

Bacteria and viruses are also known to survive indefinitely in dormant form as biofilms in the body and in the environment, when they accumulate new mutations to come back with a vengeance.

Let me end by summarising the hazards from transgenic foods.

Hazards of transgenic foods

1. Toxic or allergenic effects due to transgene products or products from interactions with host genes. I just got news via the network that a study published in the latest New England J. of Medicine found that genes transplanted from Brazil nuts to soybeans include the allergenic protein, and the biotech company involved had to drop the project.

2. Spread of transgenes to related weed species, creating superweeds (e.g. herbicide resistance.)

3. Vector-mediated horizontal gene transfer to unrelated species, creating many weed species.

4. Vector-mediated horizontal gene transfer and recombination to create new pathogenic bacteria.

5. Vector recombination to generate new virulent strains of viruses, especially in transgenic plants engineered for viral resistance with viral genes.

6. Vector-mediated spread of antibiotic resistance to bacteria in the environment, greatly exacerbating an already existing public health problem.

7. Vector-mediated spread of antibiotic resistance to gut bacteria and to pathogens.

8. Vector-mediated infection of cells after ingestion of transgenic foods. The vector can regenerate disease viruses or insert itself into the cell's genome, disrupting gene function and causing cancer.

9. The vectors carrying the transgene, unlike chemical pollution, are self-perpetuating, and self-amplifying. Once let loose, they are impossible to control or recall.

I hope this helps to convince you that there is no case for relaxing existing, already inadequate, guidelines for environmental releases of transgenic organisms, and for marketing transgenic foods. On the contrary, a moratorium on both environmental releases of transgenic organisms and marketing of transgenic foods should be imposed on the precautionary principle, until the possibility of vector-mediated horizontal gene transfer and its consequence on biodiversity, agriculture and human health can be fully assessed, and appropriate legally binding biosafety regulations firmly established.

The above is an edited text of a talk presented to The National Council of Women of Great Britain Symposium on Food: Facts, Fallacies and Fears, 22 March 1996, Darlington.

Dr Mae-Wan Ho is a Professor and Director , Bioelectrodynamics Laboratory, Department of Biology, Open University, United Kingdom.

For reasons of space the endnotes supplied with the above article have been omitted. Readers interested in the complete article can contact the editor of Third World Resurgence for a copy.