Institute of Science in Society <>

Horizontal Gene Transfer – The Hidden Hazards of Genetic Engineering

   Mae-Wan Ho - Institute of Science in Society and Department of Biological Sciences,
             Open University, Walton Hall, Milton Keynes, MK7 6AA, UK

 A version of this paper will appear on the website of SCOPE - a NSF-funded research project involving Science Journal and groups
               at the University of California at Berkeley and the University of Washington in Seattle.


Genetic engineering involves designing artificial constructs to cross species barriers and to invade genomes. In other words, it
enhances horizontal gene transfer – the direct transfer of genetic material to unrelated species. The artificial constructs or transgenic
DNA typically contain genetic material from bacteria, viruses and other genetic parasites that cause diseases as well as antibiotic
resistance genes that make infectious diseases untreatable. Horizontal transfer of transgenic DNA has the potential, among other
things, to create new viruses and bacteria that cause diseases and spread drug and antibiotic resistance genes among pathogens.
There is an urgent need to establish effective regulatory oversight to prevent the escape and release of these dangerous constructs into
the environment, and to consider whether some of the most dangerous experiments should be allowed to continue at all.

Key words: antibiotic resistance genes, dormant viruses, CaMV promoter, cancer, naked DNA, transgenic DNA,

 Transgenic pollen and baby bees

Prof. Hans-Hinrich Kaatz from the University of Jena, is reported to have new evidence, as yet unpublished, that genes engineered
into transgenic plants have transferred via pollen to bacteria and yeasts living in the gut of bee larvae(1).

If Prof. Kaatz’ claim can be substantiated, it indicates that the new genes and gene-constructs introduced into transgenic crops and
other transgenic organisms can spread, not just by ordinary cross-pollination or cross-breeding to closely related species, but by the
genes and gene-constructs invading the genomes (the totality of the organisms’ own genetic material) of completely unrelated species,
including the microorganisms living in the gut of animals eating transgenic material.

This finding is not unexpected. Some scientists have been drawing attention to this possibility recently(2), but the warnings actually
date back to the mid-1970s when genetic engineering began. Hundreds of scientists around the world are now demanding a
moratorium on all environmental releases of transgenic organisms on grounds of safety(3), and horizontal gene transfer is one of the
major considerations.

Some of us have argued that the hazards of ‘horizontal’ gene transfer to unrelated species are inherent to genetic engineering(4). The
genes and gene-constructs created in genetic engineering have never existed in billions of years of evolution. They consist of genetic
material originating from bacteria, viruses and other genetic parasites that cause diseases and spread drug and antibiotic resistance
genes. They are designed to cross all species barriers and to invade genomes. The spread of such genes and gene-constructs have the
potential to make infectious diseases untreatable and to create new viruses and bacteria that cause diseases.

 Horizontal gene transfer may spread transgenes to the entire biosphere

Horizontal gene transfer is the transfer of genetic material between cells or genomes belonging to unrelated species, by processes
other than usual reproduction. In the usual process of reproduction, genes are transferred vertically from parent to offspring; and
such a process can occur only within a species or between closely related species.

Bacteria have been known to exchange genes across species barriers in nature. There are three ways in which this is accomplished. In
conjugation, genetic material is passed between cells in contact; in transduction, genetic material is carried from one cell to another
by infectious viruses; and in transformation, the genetic material is taken up directly by the cell from its environment. For horizontal
gene transfer to be successful, the foreign genetic material must become integrated into the cell’s genome, or become stably
maintained in the recipient cell in some other form. In most cases, foreign genetic material that enters a cell by accident, especially if it
is from another species, will be broken down before it can incorporate into the genome. Under certain ecological conditions which are
still poorly understood, foreign genetic material escapes being broken down and become incorporated in the genome. For example,
heat shock and pollutants such as heavy metals can favor horizontal gene transfer; and the presence of antibiotics can increase the
frequency of horizontal gene transfer 10 to 10 000 fold(5).

While horizontal gene transfer is well-known among bacteria, it is only within the past 10 years that its occurrence has become
recognized among higher plants and animals(6). The scope for horizontal gene transfer is essentially the entire biosphere, with bacteria
and viruses serving both as intermediaries for gene trafficking and as reservoirs for gene multiplication and recombination (the process
of making new combinations of genetic material (7)).

There are many potential routes for horizontal gene transfer to plants and animals. Transduction is expected to be a main route as
there are many viruses which infect plants and animals. Recent research in gene therapy indicates that transformation is potentially very
important for cells of mammals including human beings. A great variety of ‘naked’ genetic material are readily taken up by all kinds of
cells, simply as the result of being applied in solution to the eye, or rubbed into the skin, injected, inhaled or swallowed. In many
cases, the foreign gene constructs become incorporated into the genome(8).

Direct transformation may not be as important for plant cells, which generally have a protective cell wall. But soil bacteria belonging to
the genus Agrobacterium are able to transfer the T (tumour) segment of its Tumour-inducing (Ti) plasmid (see below) into plant cells
in a process resembling conjugation. This T-DNA is widely exploited as a gene transfer vehicle in plant genetic engineering (see
below). Foreign genetic material can also be introduced into plant and animal cells by insects and arthropods with sharp mouthparts.
In addition, bacterial pathogens which enter plant and animal cells may take up foreign genetic material and carry it into the cells, thus
serving vectors for horizontal gene transfer(9). There are almost no barriers preventing the entry of foreign genetic material into the
cells of probably any species on earth. The most important barriers to horizontal gene transfer operate after the foreign genetic
material has entered the cell(10).

Most foreign genetic material, such as those present in ordinary food, will be broken down to generate energy and building-blocks for
growth and repair. There are many enzymes which break down foreign genetic material; and in the event that the foreign genetic
material is incorporated into the genome, chemical modification can still put it out of action and eliminate it.

However, viruses and other genetic parasites such as plasmids and transposons, have special genetic signals and probably overall
structure to escape being broken down. A virus consists of genetic material generally wrapped in a protein coat. It sheds its overcoat
on entering a cell and can either hi-jack the cell to make many more copies of itself, or it can jump directly into the cell’s genome.
Plasmids are pieces of ‘free’, usually circular, genetic material that can be indefinitely maintained in the cell separately from the cell’s
genome. Transposons, or ‘jumping genes’, are blocks of genetic material which have the ability to jump in and out of genomes, with
or without multiplying themselves in the process. They can also land in plasmids and be propagated there. Genes hitch-hiking in
genetic parasites, ie, viruses, plasmids and transposons, therefore, have a greater probability of being successfully transferred into cells
and genomes. Genetic parasites are vectors for horizontal gene transfer.

Natural genetic parasites are limited by species barriers, so for example, pig viruses will infect pigs, but not human beings, and
cauliflower viruses will not attack tomatoes. It is the protein coat of the virus that determines host specificity, which is why naked viral
genomes (the genetic material stripped of the coat) have generally been found to have a wider host range than the intact virus(11).
Similarly, the signals for propagating different plasmids and transposons are usually specific to a limited range of host species, although
there are exceptions.

As more and more genomes have been sequenced, it is becoming apparent that gene trafficking or horizontal gene transfer has played
an important role in the evolution of all species(12). However, it is also clear that horizontal gene trafficking is regulated by internal
constraints in the organisms in response to ecological conditions(13).

 Genetic engineering is unregulated horizontal gene transfer

Genetic engineering is a collection of laboratory techniques used to isolate and combine the genetic material of any species, and then
to multiply the constructs in convenient cultures of bacteria and viruses in the laboratory. Most of all, the techniques allow genetic
material to be transferred between species that would never interbreed in nature. That is how human genes can be transferred into pig,
sheep, fish and bacteria; and spider silk genes end up in goats. Completely new, exotic genes are also being introduced into food and
other crops.

In order to overcome natural species barriers limiting gene transfer and maintenance, genetic engineers have made a huge variety of
artificial vectors (carriers of genes) by combining parts of the most infectious natural vectors – viruses, plasmids and transposons -
from different sources. These artificial vectors generally have their disease-causing functions removed or disabled, but are designed to
cross wide species barriers, so the same vector may now transfer, say, human genes spliced into the vector, to the genomes of all
other mammals, or of plants. Artificial vectors greatly enhance horizontal gene transfer (see Box 1).(14)

             Box 1

             Artificial vectors enhance horizontal gene transfer

                 They are derived from natural genetic parasites that mediate
                 horizontal gene transfer most effectively.
                 Their highly chimaeric nature means that they have sequence
                 homologies (similarities) to DNA from viral pathogens,
                 plasmids and transposons of multiple species across Kingdoms.
                 This will facilitate widespread horizontal gene transfer and
                 They routinely contain antibiotic resistance marker genes which
                 enhance their successful horizontal transfer in the presence of
                 antibiotics, either intentionally applied, or present as xenobiotic
                 in the environment. Antibiotics are known to enhance horizontal
                gene transfer between 10 to 10 000 fold.
                 They often have ‘origins of replication’ and ‘transfer
                 sequences’, signals that facilitate horizontal gene transfer and
                 maintenance in cells to which they are transferred.
                 Chimaeric vectors are well-known to be structurally unstable,
                 ie, they have a tendency to break and join up incorrectly or with
                 other DNA, and this will increase the propensity for horizontal
                 gene transfer and recombination.
                 They are designed to invade genomes, to overcome mechanisms
                 that breakdown or disable foreign DNA and hence will increase
                 the probability of horizontal transfer.

Although different classes of vectors are distinguishable on the basis of the main-frame genetic material, practically every one of them
is chimaeric, being composed of genetic material originating from the genetic parasites of many different species of bacteria, animals
and plants. Important chimaeric ‘shuttle’ vectors enable genes to be multiplied in the bacterium E. coli and transferred into species in
every other Kingdom of plants and animals. Simply by creating such a vast variety of promiscuous gene transfer vectors, genetic
engineering biotechnology has effectively opened up highways for horizontal gene transfer and recombination, where previously the
process was tightly regulated, with restricted access through narrow, tortuous footpaths. These gene transfer highways connect
species in every Domain and Kingdom with the microbial populations via the universal mixing vessel used in genetic engineering, E.
coli. What makes it worse is that there is currently still no legislation in any country to prevent the escape and release of most artificial
vectors and other artificial constructs into the environment (15).

 What are the hazards of horizontal gene transfer?

Most artificial vectors are either derived from viruses or have viral genes in them, and are designed to cross species barriers and
invade genomes. They have the potential to recombine with the genetic material of other viruses to generate new infectious viruses that
cross species barriers. Such viruses have been appearing at alarming frequencies. The antibiotic resistance genes carried by artificial
vectors can also spread to bacterial pathogens. Has the growth of commercial-scale genetic engineering biotechnology contributed to
the resurgence of drug and antibiotic infectious diseases within the past 25 years (16)? There is already overwhelming evidence that
horizontal gene transfer and recombination have been responsible for creating new viral and bacterial pathogens and for spreading
drug and antibiotic resistance among the pathogens. One way that new viral pathogens may be created is through recombination with
dormant, inactive or inactivated viral genetic material that are in all genomes, plants and animals without exception. Recombination
between external and resident, dormant viruses have been implicated in many animal cancers (17).

As stated earlier, the cells of all species including our own can take up foreign genetic material. Artificial constructs designed to invade
genomes may well invade our own. These insertions may lead to inappropriate inactivation or activation of genes (insertion
mutagenesis), some of which may lead to cancer (insertion carcinogenesis)(18). The hazards of horizontal gene transfer are
summarized in Box 2.

             Box 2

             Potential hazards of horizontal gene transfer from genetic engineering

                 Generation of new cross-species viruses that cause disease
                 Generation of new bacteria that cause diseases
                 Spreading drug and antibiotic resistance genes among the viral
                 and bacterial pathogens, making infections untreatable
                 Random insertion into genomes of cells resulting in harmful
                 effects including cancer
                 Reactivation of dormant viruses, present in all cells and
                 genomes, which may cause diseases
                 Spreading new genes and gene constructs that have never existed
                 Multiplication of ecological impacts due to all of the above.

 Transgenic DNA may be more likely to transfer horizontally than non-transgenic DNA

Both the artificial vectors used in genetic engineering and the genes transferred to make transgenic organisms are predominantly from
viruses and bacteria associated with diseases, and these are being brought together in combinations that have never existed in billions
of years of evolution.

Genes are never transferred alone. They are transferred in unit-constructs, known as an ‘expression cassettes’. Each gene has to be
accompanied by a special piece of genetic material, the promoter, which signals the cell to turn the gene on, ie, to transcribe the DNA
gene sequence into RNA. At the end of the gene there has to be another signal, a terminator, to end the transcription and to mark
the RNA, so it can be further processed and translated into protein. The simplest expression cassette looks like this:


Typically, each bit of the construct: promoter, gene and terminator, is from a different source. The gene itself may also be a composite
of bits from different sources. Several expression cassettes are usually linked in series, or ‘stacked’ in the final construct. At least one
of the expression cassettes will be that of an antibiotic resistance marker gene to enable cells that have taken up the foreign construct
to be selected with antibiotics. The antibiotic resistance gene cassette will often remain in the transgenic organism.

The most commonly used promoters are from viruses associated with serious diseases. The reason is that such viral promoters give
continuous over-expression of genes placed under their control. The same basic construct is used in all applications of genetic
engineering, whether in agriculture or in medicine, and the same hazards are involved. There are reasons to believe that transgenic
DNA is much more likely to spread horizontal than the organisms’ own DNA (see Box 3) (19).

             Box 3

             Reasons to suspect that transgenic DNA may be more likely to spread horizontally
             than non-transgenic DNA

                 Artificial constructs and vectors are designed to be invasive to
                 foreign genomes and overcome species barriers.
                 All artificial gene-constructs are structurally unstable (20), and
                 hence prone to recombine and transfer horizontally.
                 The mechanisms enabling foreign genes to insert into the genome
                 also enable them to jump out again, to re-insert at another site,
                 or to another genome.
                 The integration sites of most commonly used artificial vectors
                 for transferring
                 genes are ‘recombination hotspots’, and so have an increased
                 propensity to transfer horizontally.
                 Viral promoters, such as that from the cauliflower mosaic virus,
                 widely used to make transgenes over-express, contain
                 recombination hotspots (21), and will therefore further enhance
                 horizontal gene transfer.
                 The metabolic stress on the host organism due to the continuous
                 over expression of transgenes may also contribute to the
                 instability of the insert (22).
                 The foreign gene-constructs and the vectors into which they are
                 spliced, are typically mosaics of DNA sequences from
                 numerous species and their genetic parasites; that means they
                 will have sequence homologies with the genetic material of
                 many species and their genetic parasites, thus facilitating
                 wide-ranging horizontal gene transfer and recombination.

 Additional hazards from viral promoters

We have recently drawn attention to additional hazards associated with the promoter of the cauliflower mosaic virus (CaMV) most
widely used in agriculture (23). It is in practically all transgenic plants already commercialized or undergoing field trials, as well as a
high proportion of transgenic plants under development, including the much acclaimed ‘golden rice’ (24).

CaMV is closely related to human hepatitis B virus, and less so, to retroviruses such as the AIDS virus (25). Although the intact virus
itself is infectious only for cruciferae plants, its promoter is promiscuous in function, and is active in all higher plants, in algae, yeast,
and E. coli (26), as well as frog and human cell systems (27). Like all promoters of viruses and of cellular genes, it has a modular
structure, with parts common to, and interchangeable with promoters of other plant and animal viruses. It has a recombination
hotspot, flanked by multiple motifs involved in recombination, similar to other recombination hotspots including the borders of the
Agrobacterium T DNA vector most frequently used in making transgenic plants. The suspected mechanism of recombination
requires little or no DNA sequence homologies. Finally, viral genes incorporated into transgenic plants have been found to recombine
with infecting viruses to generate new viruses (28). In some cases, the recombinant viruses are more infectious than the original.

Proviral sequences – generally inactive copies of viral genomes - are present in all plant and animal genomes, and as all viral
promoters are modular, and have at least one module – the TATA box - in common, if not more. It is not inconceivable that the
CaMV 35S promoter in transgenic constructs can reactivate dormant viruses or generate new viruses by recombination. The CaMV
35S promoter has been joined artificially to copies of a wide range of viral genomes, and infectious viruses produced in the laboratory
(29). There is also evidence that proviral sequence in the genome can be reactivated (30).

These considerations are especially relevant in the light of recent findings that certain transgenic potatoes - containing the CaMV 35S
promoter and transformed with Agrobacterium T-DNA - may be unsafe for young rats, and that a significant part of the effects may
be due to "the construct or the genetic transformation (or both) (31)" The authors also report an increase in lymphocytes in the
intestinal wall, which is a non-specific sign of viral infection (32).

 Evidence for horizontal transfer of transgenic DNA

It is often argued that transgenic DNA, once incorporated into the transgenic organism, will be just as stable as the organism’s own
DNA. But there is both direct and indirect evidence against this supposition. Transgenic DNA is more likely to spread, and has been
found to spread by horizontal gene transfer.

Transgenic lines are notoriously unstable and often do not breed true (33). There is a paucity of molecular data documenting the
structural stability of the transgenic DNA, both in terms of its site of insertion in the genome and its arrangement of genes, in
successive generations. Instead, transgenes may be silenced in subsequent generations or lost altogether (34).

A herbicide-tolerance gene, introduced into Arabidopsis by means of a vector, was found to be up to 30 times more likely to escape
and spread than the same gene obtained by mutagenesis (35). One way this may happen is by secondary horizontal gene transfer via
insects visiting the plants for pollen and nectar (36). The reported finding that pollen can transfer transgenic DNA to bacteria in the gut
of bee larvae is relevant here.

Secondary horizontal transfer of transgenes and antibiotic resistant marker genes from genetically engineered crop-plants into soil
bacteria and fungi have been documented in the laboratory. Transfer to fungi was achieved simply by co-cultivation (37), while
transfer to bacteria has been achieved by both re-isolated transgenic DNA or total transgenic plant DNA (38). Successful transfers of
a kanamycin resistance marker gene to the soil bacterium Acinetobacter were obtained using total 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) (39). 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.

Despite the misleading title in one of the publications,(40) a high gene transfer frequency of 5.8 x 10-2 per recipient bacterium was
demonstrated under optimum conditions. But the authors then proceeded to calculate an extremely low gene transfer frequency of 2.0
x 10-17 under extrapolated "natural conditions", assuming that 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 transgenic DNA is bound to be readily available in the rhizosphere around the plant roots, which is also an ‘environmental
hotspot’ for gene transfer (41). Other workers have 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 (42).

Defenders of the biotech industry still insist that just because horizontal gene transfer occurs in the laboratory does not mean it can
occur in nature. However, there is already evidence suggesting it can occur in nature. First of all, genetic material released from dead
and live cells, is now found to persist in all environments; and not rapidly broken down as previously supposed. It sticks to clay, sand
and humic acid particles and retains the ability to infect (transform) a range of micro-organisms in the soil (43). The transformation of
bacteria in the soil by DNA adsorbed to clay sand and humic acid has been confirmed in microcosm experiments (44).

Reseachers in Germany began a series of experiments in 1993 to monitor field releases of transgenic rizomania-resistant sugar beet
(Beta vulgaris), containing the marker gene for kanamycin resistance, for persistence of transgenic DNA and of horizontal gene
transfer of transgenic DNA into soil bacteria (45). It is the first such experiment to be carried out; after tens of thousands of field
releases and tens of millions of hectares have been planted with transgenic crops. It will be useful to review their findings in detail.

Transgenic DNA was found to persist in the soil for up to two years after the transgenic crop was planted. Though they did not
comment on it, the data showed that the proportion of kanamycin resistant bacteria in the soil increased significantly between 1.5 and
2 years. Could it be due to horizontal transfer of antibiotic resistance marker gene in the transgenic DNA? Although none of 4000
colonies of soil bacteria isolated – a rather small number - was found to have taken up transgenic DNA by the probes available, two
out of seven samples of total bacterial DNA yielded positive results after 18 months. This suggests that horizontal gene transfer may
have taken place, but the specific bacteria which have taken up the transgenic DNA cannot be isolated as colonies. That is not
surprising as less than 1% of all the bacteria in the soil are culturable. The authors were careful not to rule out transgenic DNA being
adsorbed to the surface of bacteria rather than being tranferred into the bacteria.

The researchers also carried out microcosm experiments to which total transgenic sugar-beet DNA was added to non-sterile soil with
its natural complement of microorganisms. The intensity of the signal for transgenic DNA decreased during the first days and
subsequently increased. This may be interpreted as a sign that the transgenic DNA has been taken up by bacteria and become
amplified as a result.

In parallel, soil samples were plated and the total bacterial lawn allowed to grow for 4 days, after which DNA was extracted. Several
positive signals were found, "which might indicate uptake of transgenic DNA by competent bacteria."

The authors were cautious not to claim conclusive results simply because the specific bacteria carrying the transgenic DNA sequences
were not isolated. The results do show, however, that horizontal gene transfer may have taken place both in the field and in the soil

DNA is not broken down sufficiently rapidly in the gut either, which is why transfer of transgenic DNA to microorganisms in the gut of
bee larvae would not be surprising. A genetically engineered plasmid was found to have a 6 to 25% survival after 60 min. of exposure
to human saliva. The partially degraded plasmid DNA was capable of transforming Streptococcus gordonii, one of the bacteria that
normally live in the human mouth and pharynx. The frequency of transformation dropped exponentially with time of exposure to saliva,
but it was still detectable after 10 minutes. Human saliva actually contains factors that promote competence of resident bacteria to
become transformed by DNA (46).

Viral DNA fed to mice is found to reach white blood cells, spleen and liver cells via the intestinal wall, to become incorporated into
the mouse cell genome (47). When fed to pregnant mice, the viral DNA ends up in cells of the fetuses and the new born animals,
suggesting that it has gone through the placenta as well (48). The authors remark that "The consequences of foreign DNA uptake for
mutagenesis and oncogenesis have not yet been investigated (49)." As already mentioned, recent experiments in gene therapy leave
little doubt that naked nucleic acid constructs can readily enter mammalian cells and in many cases become incorporated into the cell’s


Horizontal gene transfer is an established phenomenon. It has taken place in our evolutionary past and is continuing today. All the
signs are that natural horizontal gene transfer is a regulated process, limited by species barriers and by mechanisms that break down
and inactivate foreign genetic material. Unfortunately, genetic engineering has created a huge variety of artificial constructs designed to
cross all species barriers and to invade essentially all genomes. Although the basic constructs are the same for all applications, some of
the most dangerous may be coming from the waste disposal of contained users of transgenic organisms(50). These will include
constructs containing cancer genes from viruses and cells from laboratories researching and developing cancer and cancer drugs,
virulence genes from bacteria and viruses in pathology labs. In short, the biosphere is being exposed to all kinds of novel constructs
and gene combinations that did not previously exist in nature, and may never have come into being but for genetic

There is an urgent need to establish effective regulatory oversight, in the first instance, to prevent the escape and release of these
dangerous constructs into the environment, and then to consider whether some of the most dangerous experiments should be allowed
to continue at all.


  1.Thanks to Dr. Beatrix Tappeser, Institute for Applied Ecology, Postfach 6226, D-79038, Freiburg, for
    this information. See also Barnett, A. (2000). GM genes 'jump species barrier' The Observer, May 28,
  2.See Stephenson, J.R., and Warnes, A. (1996). Release of genetically-modified miroorganisms into the
    environment. J. Chem. Tech. Biotech. 65, 5-16; Harding, K. (1996). The potential for horizontal gene
    transfer within the environment. Agro-Food-Industry Hi-Tech July/August, 31-35; Ho, M.W. (1996). Are
    current transgenic technologies safe? In Virgin, I. and Frederick R.J., eds. Biosafety Capacity Building,
    pp. 75-80, Stockholm Environment Institute, Stockholm; Traavik, T. (1999). Too Early May be Too
    Late, Report for the Directorate for Nature Research, Trondheim, Norway.
  4.See Ho, M.W. (1998, 1999). Genetic Engineering Dream or Nightmare? The Brave New World of Bad
    Science and Big Business. Gateway, Gill & Macmillan, Dublin; Ho, M.W., Traavik, T., Olsvik, R.,
    Tappeser, B., Howard, V., von Weizsacker, C. and McGavin, G. (1998). Gene Technology and Gene
    Ecology of Infectious Diseases. Microbial Ecology in Health and Disease 10, 33-59.
  5.See Ho et al, 1998 (note 4) and references therein.
  6.See Lorenz, M.G. and Wackernagel, W. (1994). Bacterial gene transfer by natural genetic transformation
    in the environment. Microbiol. Rev. 58, 563-602.
  7.See Ho,1998, 1999 (note 4; Ho, et al, 1998 (note 4).
  8.See Ho, M.W., Ryan, A., Cummins, J. and Traavik, T. (2000a). Unregulated Hazards: ‘Naked’ and
    ‘Free’ Nucleic Acids, ISIS & TWN Report, London and Penang.
  9.Grillot-Courvalin, C., Goussand, S., Huetz, F., Ojcius, D.M. and Courvalin, P. (1998). Functional gene
    transfer from intracellular bacteria to mammalian cells. Nature Biotechnology 16, 862-866.
  10.See Nielsen, K.M., Bones, A.M., Smalla, K. and van Elsas, J.D. (1998). Horizontal gene transfer from
    transgenic plants to terrestrial bacteria – a rare event? FEMS Microbiology Reviews 22, 79-103.
  11.See Ho et al, 2000a (note 9)
  12.See Doolittle, W.F. (1999). Lateral genomics. Trends Cell Biol 9, 5-8.
  13.See Jain, R., Rivera, M.C. and Lake, J.A. (1999). Horizontal gene transfer among genomes: The
    complexity hypothesis. Proc. Natl. Acad. Sci. USA 96, 3801-3806; Shapiro, J. (1997). Genome
    organization, natural genetic engineering and adaptive mutation. TIG 13, 98-104; Ho, 1998,1999 (note 4).
  14.See Ho et al, 1998 (note 4) for references.
  15.See Ho et al, 2000 (note 8)
  16.Reviewed in Ho et al, 1998 ( note 4).
  17.Reviewed in Ho, 1998, 1999 (note 4) Chapter on "The mutable gene and the human condition".
  18.See Ho et al, 2000 (note 9) and references therein.
  19.See Ho, M.W. (1999). Special Safety Concerns of Transgenic Agriculture and Related Issues Briefing
    Paper for Minister of State for the Environment, The Rt Hon Michael Meacher
  20.See Old, R.W. and Primrose, S.B. (1994). Principles of Gene Manipulation, 5th ed. Blackwell Science,
    Oxford; Kumpatla, S.P., Chandrasekharan, M.B., Iuer, L.M., Li, G. and Hall, T.c. (1998). Genome
    intruder scanning and modulation systems and transgene silencing. Trends in Plant Sciences 3, 96-104.
  21.See Kohli, A., Griffiths, S., Palacios, N., Twyman, R.M., Vain, P., Laurie, D.A. and Christou, P. (1999).
    Molecular characterization of transforming plasmid rearrangements in transgenic rice reveals a
    recombination hotspot in the CaMV 35S promoter and confirms the predominance of microhomology
    mediated recombination. The Plant Journal 17, 591-601.
  22.Finnegan, J. and McElroy, D. (1994). Transgene inactivation, plants fight back! Bio/Technology 12,
  23.Ho, M.W., Ryan, A. and Cummins, J. (1999). The cauliflower mosaic viral promoter – a recipe for
    disaster? Microbial Ecology in Health and Disease 11, 194-197; Ho, M.W., Ryan, A. and Cummins, J.
    (2000). Hazards of transgenic plants containing the cauliflower mosaic viral promoter. Microbial Ecology
    in Health and Disease (in press).
  24.Ye, X., Al-Babili, S., Kloti, A., Zhang, J., Lucca, P., Beyer, P. and Potrykus, I. (2000). Engineering the
    provitamin A (-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 287,
    303-305; see also Ho, M.W. (2000). The Golden Rice – An Exercise in How Not to Do Science. ISIS
    Sustainable Science Audit #1
  25.Xiong, Y. and Eikbush, T. (1990). Origin and evolution of retroelements based upon the reverse
    transriptase sequences. The Embo Journal 9, 3363-72.
  26.Assad, F.F. and Signer, E.R. (1990). Cauliflower mosaic virus P35S promoter activity in E. coli. Mol.
    Gen. Genet. 223, 517-20.
  27.Ballas,N., Broido, S., Soreq, H., and Loyter, A. (1989). Efficient functioning of plant promoters and
    poly(A) sites in Xenopus oocytes Nucl Acids Res 17, 7891-903; Burke, C, Yu X.B., Marchitelli, L..,
    Davis, E.A., Ackerman, S. (1990). Transcription factor IIA of wheat and human function similarly with
    plant and animal viral promoters. Nucleic Acids Res 18, 3611-20.
  28.Reviewed in Ho, et al, 2000 (note 24).
  29.Maiss, E., Timpe,U., and Brisske-Rode, A. (1992). Infectious in vivo transcripts of a plumpox potyvirus
    full lenth c-DNA clone containig the cauliflower mosaic virus 35-S RNA promoter J. Gen. Virol. 73,
    709-13; Meyer, M and Dessens, J. (1997). 35S promoter driven cDNA of barley mild mosaic virus
    RNA-1 and RNA-2 are infectious in barley plants. J. Gen. Viol. 78, 147-51.
  30.Ndowora, T., Dahal, G., LaFleur, D., Harper, G., Hull, R., Olszerski, N.E. and Lockhart, B. (1999).