The critiques of B. Hohn, J. Paszkowski, J. Fuetterer, R. Hull, M. Tepfer, T. Hohn, H. Puchta, Monsanto and M. Cantley were directed at the, as yet, unpublished manuscript posted on the web by the Journal. None of the critics, except for B. Hohn, questions the accuracy of our description of the findings. They disagree with the implications we draw and our conclusions. We shall reply to the substantive points of disagreement and ignore ad hominem arguments and other ill-mannered remarks. We welcome this rare opportunity for open debate.
We apologize for mis-citing the paper of B. Hohn and her colleagues as having demonstrated a recombination hotspot for the 35S promoter instead of another, the 19S promoter, also belonging to the cauliflower mosaic virus (CaMV). The correct reference is Vaden and Melcher, 1990. This mistake has been put right in the proofs.
The substantive points of disagreement are as follows:
The intact, encapsidated CaMV, consisting of the CaMV genome wrapped in its protein coat, is not infectious for human beings nor for other non-susceptible animals and plants, as is well-known, for it is the coat that determines host specificity in the first instance. However, the naked or free viral genomes may be more infectious and have a wider host-range than the intact virus. Human T-cell leukemia viral genomes formed complete viruses when injected into the bloodstream of rabbits (1). Similarly, the genomes of the human polyomavirus BK (BKV) gave a full-blown infection when injected into rabbits, despite the fact that the intact BKV is not infectious (2). Recent developments in gene therapy show that naked DNA can be successfully delivered by direct injection into muscle (3), through the skin (4), as well as by mouth, where the DNA is taken up by cells lining the gut (5). Free DNA can even be taken up by sperm cells of marine organisms and mammals, and transgenic animals created (6). Yet, the potential ecological and health impacts of the increasing amounts of different kinds of free nucleic acids discharged into our environment as the result of genetic engineering biotechnology has never even been considered by the overwhelming majority of the scientific community (7-9).
According to R. Hull, 10% of the cabbages and cauliflowers on sale in supermarkets are infected with CaMV, with an estimated 100 000 fully encapsidated viral particles and 10 000 free viral genomes in each cell. Should we worry about eating infected cabbages and cauliflower and getting infected by the free viral genomes? Perhaps; we simply do not know. This is something we should find out. But there is also a big difference between the CaMV viral genome and its promoter in the transgenic plant.
The 35S promoter in the CaMV genome is an integral part of the virus, the genetic material of which has been adapted to stay together as a whole for millions of years. Why should we think its 35S promoter is the same when taken out of its original context and joined up with genes it has never been linked to before, and inserted into genomes it has never been in before?
It is well-known that pieces of DNA taken out of context and recombined in novel configurations are likely to be unstable, so much so that structural instability of artificial vectors for genetic engineering - made by joining pieces of DNA from different viruses, plasmids, transposons and other sources - is a topic discussed in a text-book (10). This instability also renders the artificial vectors prone to recombination and rearrangement.
The CaMV promoter in the transgenic plant is part of the transgenic DNA introduced, which has quite a complicated structure. The CaMV promoter is joined to a gene it has never been linked to before, to form an 'expression-cassette', several expression-cassettes are stacked in series, and spliced in turn into an artificial vector (T DNA), which is also known to be flanked by recombination hotspots (11, 12, 13). This structure of the transgenic DNA is itself recognized to be unstable and to increase the propensity for horizontal gene transfer, as stated in a recent scientific report commissioned by the UK Health and Safety Executive (13),
'The location of released genes on mobile genetic elements or in close association within IS sequences, transposons, gene cassettes or hot-spots for recombination …. can all increase the probability of horizontal gene transfer.' p. 70.
CaMV promoter in transgenic plant versus the plant's own promoters, integrated viruses and other possible recombination hotspots
The plant's promoters have been adapted, structurally and functionally, to its own genes in the context of the whole genome for hundreds of millions of years, and therefore expected to be much more stable than a novel intruder like the transgenic DNA containing CaMV promoter described above. Integrated, inactive viruses and retrotransposons, similarly, have been 'tamed' by the plant, probably for millions of years and hence, again, more likely to be stable than the new intruder.
It is also becoming increasingly clear that every organism has its own integrity, such that the recombination, rearrangement, mobility and mutability of its genes are regulated in complex, entangled genetic networks that maintains the survival of the whole (reviewed in ref. 14). Transgenes are unwanted intruders to these entangled networks, which is why they are unstable. Instability of transgene expression as the result of 'gene-silencing' is now well-recognized and actively researched (see ref. 16 for the latest mechanism discovered). Structural instability involving secondary mobility or rearrangement of integrated transgenes is also a common cause of breakdown of transgenic lines (see ref. 17).
Structural instability of transgenes is generally underestimated, as cells which lose transgenes are automatically eliminated during the selection process which is necessary to produce transgenic lines. However, such instability may arise even in later generations of plant propagation if not during cell proliferation. We are aware of no published data documenting the structural stability and integrity of transgenic lines in successive generations, even though phenotypic instability has been documented in transgenic bt-cotton commercially grown in Southern United States in 1996, for example (17). Such data should give the exact genome location and configuration of the insert(s) as well as phenotypic expression of the desired trait(s). H. Puchta claims to have recently found 'no indication of a single recombination event in over a billion of transgenic plant cells', but this is not (yet) published, and does not address the stability of transgenic plants under a range of growth conditions. For example, one might expect that physiological stress due to extremes of temperature, or drought, which can mobilize transposons, may also increase transgene instability. The continuous (constitutive) over-expression of transgenes linked to the CaMV promoter is also a metabolic stress-factor that may increase transgene instability (18).
It is quite likely that stacking CaMV promoters in three successive expression cassettes, as Kohli et al (1999) have done, will increase structural instability further. The recombinations and rearrangements they have obtained in the different transgenic lines, however, may have occurred both before and after the primary transformation events during propagation and selection in cell and plant culture.
As pararetroviral and retrotransposon sequences are increasingly found in plant genomes (19), and given the increased structural instability of transgenic DNA associated with recombination hotspots as argued above, the potential for generating recombinant infectious viruses cannot be dismissed. In particular, it has already been found that viral coat protein transgenes can recombine with infecting viruses to generate new viruses (20-22). A possible scenario, in the case of the CaMV promoter, is for a sleeping pararetrovirus that has lost its promoter, to recombine with the CaMV promoter to give an infectious recombinant virus. As far as we are aware, there have been no studies addressing the possibility of recombination between transgenic CaMV promoter and dormant viral genomes. Furthermore, it should be noted that the integration of pararetroviral sequences into plant genomes has occurred in the absence of viral integrase (23), presumably because similar enzymes are present in the host cell. Integrases belong to a superfamily of enzymes present in all genomes from viruses and bacteria to higher plants and animals. They can catalyze the insertion of viral DNA into the host genome, and also function as disintegrases, catalyzing the reverse reaction. Hence, viral sequences, and indeed, any extraneously inserted DNA sequence, will have the potential of being mobilised by these enzymes to transfer again.
There is evidence of secondary horizontal gene transfer from transgenic plants, but there is no evidence that the CaMV promoter is involved, if only because this has not been investigated. However, as the T-DNA vector itself is also flanked by recombination hotspots, part or all of the transgenic DNA may be subject to horizontal gene transfer. Secondary horizontal tranfer of transgenes and antibiotic resistant marker genes from transgenic plants into soil bacteria and fungi have been documented in the laboratory. In the case of fungi, the transfer was obtained simply by co-cultivation (24). Successful transfers of a kanamycin resistance marker gene to the soil bacterium Acinetobacter were obtained using DNA extracted from homogenized plant leaf from a range of transgenic plants: Solanum tuberosum (potato), Nicotiana tabacum (tobacco), Beta vulgaris (sugar beet), Brassica napus (oil-seed rape) and Lycopersicon esculentum (tomato) (25). It is estimated that about 2500 copies of the kanamycin resistance genes (from the same number of plant cells) is sufficient to successfully transform one bacterium, despite the fact that there is six million-fold excess of plant DNA present. A single plant with say, 2.5 trillion cells, would be sufficient to transform one billion bacteria.
J. Fuetterer accuses one of us, elsewhere (7) of mis-citing his publication (Schluter et al, 1995, ref. 26) demonstrating that horizontal gene transfer does not occur, as evidence that it does. In fact, the publication reports investigations on horizontal gene transfer under a variety of conditions, some of which gave positive results. For example, a high gene transfer frequency of 5.8 x 10-2 per recipient bacterium was demonstrated for ampicillin resistance transgene - re-isolated from the DNA of transgenic potato - to Erwinia chrysanthem, a bacterial pathogen. This was achieved by 105 copies of the ampicillin resistance gene per potato genome, introduced into 6.4 x 108 bacteria by electroporation. When reduced to one copy of ampicillin resistance gene per potato genome, the gene transfer frequency was still significant at 4 x 10-6. The total genomic DNA from the transgenic potato, estimated to carry two copies of ampicillin resistance gene per potato genome, likewise gave a transfer frequency of 9 x 10-6 With only transgenic potato tissue, it was less than 8.7 x 10-9, effectively nil, according to the limit of sensitivity of the protocol. The same result was obtained by co-cultivation of the transgenic tuber with bacteria for 6 weeks. The negative results were not surprising, given the limited access of the bacteria to plant DNA under those conditions. The authors then 'calculated' an extremely low frequency of gene transfer at 2.0 x 10-17 under extrapolated 'natural conditions', assuming the different factors acted independently. The natural conditions, however, are largely unknown and unpredictable, and even by the authors' own admission, synergistic effects cannot be ruled out.
Free plant DNA is bound to be readily available in the rhizosphere around the plant roots, which is also an 'environmental hotspot' for gene transfer (see ref. 13). Gebbard and Smalla have also found evidence of horizontal transfer of kanamycin resistance from transgenic DNA to Acinetobactor, and positive results were obtained using just 100ml of plant-leaf homogenate (27). Many other factors, such as the density of bacteria, temperature, availability of nutrients, heavy metals and pH, can also greatly influence the frequency of horizontal gene transfer in nature (8, 13). Furthermore, less than one percent of all bacteria in the environment can be isolated (28) and hence monitored for horizontal gene transfer; so any negative result in the field should be interpreted with due caution. There is no ground, therefore, for assuming that horizontal transfer of transgenic DNA from transgenic plants will not or does not take place under natural conditions, once it has been demonstrated in the laboratory.
There is no evidence that CaMV promoter per se is active in animal cells. However, on account of the modularity of its structure, small motifs and elements within the CaMV promoter may be interchangeable with those of the animal viruses. HIV is a retrovirus that depends on integration into the host genome for replication, and its complementary DNA will therefore be present in the genome of humans infected with the virus. Hepatitis B virus, being a pararetrovirus, does not require integration for replication, but its genome has, nevertheless, been found in human chromosomes in liver tissue, and is associated with liver carcinoma (see ref. 19).
The initiation site for the +DNA strand synthesis in CaMV is a polypurine tract (PPT). It has recently been reported that the PPT from HIV-1 gives up to 50% of the efficiency for CAMV +DNA strand synthesis as CaMV's own element (29). All eukaryotes and prokaryotes share the core TATA box promoter element. It is not inconceivable that the TATA box, as well as other elements and motifs within the CaMV promoter, when recombined with dormant animal viral promoters, may reactivate the virus, generate new viruses or give functional viral promoters that make cellular oncogenes over-express, resulting in cancer.
In conclusion, the implications we draw are entirely reasonable given the available scientific findings, although it is clearly not the view of some of the scientists whose work we cite. We regard the findings and the implications drawn as sufficient grounds for invoking the precautionary principle (see ref. 9) to recommend that all transgenic crops and products containing CaMV and similar promoters should be withdrawn from field release and from use for human consumption or animal feed. This recommendation is further justified by the conspicuous lack of evidence that we need such crops and products or that they offer any benefits to society at large.
Article first published 26/07/00
Got something to say about this page? Comment