An electronic version of this report, or
any other ISIS report, with full references, can be sent to you via e-mail for a donation
of £3.50. Please e-mail the title of the report to: firstname.lastname@example.org
A version of this article was submitted on behalf of ISIS to the Secretariat of the United Nations Convention on Biological Diversity 26 November 2006 in support of a moratorium on the environmental release of GM trees. Please circulate widely.
Genetically modified trees without caution
There is growing pressure to commercialise the numerous GM tree species that have been modified with a variety of transgenes. One major reason is that GM trees have been proposed for plantations on the mistaken assumption that they can offset carbon emissions, and more so, qualify for subsidies under the Kyoto Protocol's Clean Development Mechanism . At the same time, rising worldwide demand for biofuels has opened up an opportunity for proponents to rescue genetically modified (GM) crops from chronic market failure by promoting them as ‘energy' crops (see Box 1). Unfortunately, energy crops, including GM tree plantations, are far from sustainable or environmentally benign [1-3] ( Biofuels for Oil Addicts , SiS 30; Biofuels: Biodevastation, Hunger & False Carbon Credits , Biofuels Republic Brazil , SiS 33). But in the rush to exploit GM trees, caution will be scattered to the winds, like the pollen of the GM trees currently being tested.
Industry's spin on GM energy crops for saving carbon emissions
Biotech industry sponsored International Service for the Acquisition of Agrobiotechnology Applications (ISAAA) continues its yearly inflated estimates of area planted with GM crops  ( Global GM Crops Area Exaggerated , SiS 33), and makes unsubstantiated, very likely false claims on how GM crops can contribute to saving greenhouse gas emissions on the coat tails of the Stern report on The Economics of Climate Change  ( SiS 33). To set the record straight, the Stern report does not support GM crops nor does it favour biofuels from energy crops, and for good reasons (see main text).
Nevertheless, the ISAAA says that GM crops save carbon emissions by reducing pesticide use through insecticidal Bt crops and by sequestering carbon in the soil through conservation tillage with herbicide tolerant crops . In 2005, it claims, the combined savings were equivalent to 9 million tonnes of carbon dioxide, or removing 4 million cars from the road. And looking to the future, even greater contributions could be made through cultivation of additional areas of GM energy crops to produce ethanol and biodiesel.
Commercial releases and field tests
Even though the first GM tree, papaya, was approved for commercial release more than ten years ago there have been only two petitions for non-regulated status, the first for another papaya GM event and the other for virus resistant plums [6, 7] ( USDA Proposes to Deregulate Its Own Transgenic Plum , SiS 31). However, the United States has undertaken about 264 field test releases of numerous GM trees spread over most of the states and possessions. Modified species include tropical trees (banana, avocado, grapefruit, lime, papaya and coffee), horticultural fruits (apple, plum, pear and walnut), and numerous forest and shade trees such as eucalyptus, American chestnut, American elm, poplar, cottonwood, aspen, white spruce and pine. Transgenic traits range from disease or insect resistance and herbicide tolerance, to lignin modifications, sterility, and bioremediation .
Canada has undertaken 33 field trial releases of GM trees mainly near Quebec City; and these are limited to insect resistant or herbicide tolerant poplar, black spruce and white spruce .
Of the 205 permit applications listed at the end of 2003, 73.5 percent originated in the USA, 23 percent in other OECD member nations (in particular, Belgium, Canada, France, Finland, New Zealand, Norway, Portugal, Spain and Sweden) and 3.5 percent elsewhere (Brazil, China, Chile, South Africa and Uruguay) . Four traits accounted for 80 percent of the permit applications: herbicide tolerance (32 percent), marker genes (27 percent), insect resistance (12 percent), and lignin modification (9 percent). Of the tree species involved, Populus , Pinus , Liquidambar (Sweet Gum Tree) and Eucalyptus account for 85 percent of applications.
Potential hazards of GM trees
Genetically modified (GM) trees have all the potential hazards of GM crops and genetically modified organisms (GMOs) in general (Box 2) ( Genetic Engineering Dream or Nightmare ) , only worse.
Potential Hazards of GMOs
Synthetic genes and gene products new to evolution could be toxic for humans and other animals or provoke serious immune reactions
The uncontrollable, imprecise process involved in making GMOs can generate unintended toxic and immunogenic products, exacerbated by the instability of the transgenic lines
Endogenous viruses that cause diseases could be activated by the transgenic process
The synthetic genes in GMOs, including copies of genes from bacteria and viruses that cause disease as well as antibiotic resistance genes, may be transferred to other species via pollen, or by direct integration into other genomes in horizontal gene transfer
Disease-causing viruses and bacteria are created by horizontal transfer and recombination of the synthetic genes and genetic modification is nothing if not facilitated and greatly enhanced horizontal gene transfer and recombination
GM DNA are designed to invade genomes and insertion into the genome of animals including human beings results in insertion mutagenesis some of which may trigger cancer
Herbicide tolerant GM crops accumulate herbicide and herbicide residues that could be highly toxic to humans and animals as well as plants
Trees are larger and longer lived, and therefore can spread transgenes further and wider, while their extensive root systems are a hotbed for horizontal gene transfer and recombination.
Belated environmental impact studies show GM tree pollen cannot be contained
ISIS alerted the public to the serious health and environmental impacts of GM trees in forestry  ( GM Forest Trees - The Ultimate Threat , SiS 26) and earlier, in bioremediation and low lignin applications  ( GM Trees Alert , SiS 16). Numerous field releases were approved in the absence of information on the spread of pollen and seed in forest and orchard ecology. Only recently have models of pollen dispersal from forest trees begun to appear. Significant amounts of oak pollen were deposited up to 30 km downstream from a stand of oak trees, and lower quantities deposited up to 100 km . It has been claimed that conifer pollen dispersed to between 6 and 800 m from a source; but a more comprehensive study revised this figure upwards to between 8 and 33 km [14, 15].
Eucalyptus pollen is spread by small insects, which can carry pollen to distances of 1.6 km, although most of the hybridisation is found within 200 m of the plantation . It is essentially impossible to contain GM trees; the probability of spreading transgenes from GM conifers is 100 percent at a distance of one km from a source . Pine seeds, too, are transported over great distances, the probability that seeds are transported further than one km from a source was nearly 100 percent . Canadian regulators, recognizing that transgene containment is not possible for GM forest trees, are now suggesting that regulations should be altered to accommodate the uncontrolled release of GM trees with transgenes for herbicide tolerance, insect resistance or low lignin content !
The low lignin trait is one much desired by foresters as it provides greatly reduced costs in preparing fibre for paper. However, reduced lignin results in reduced strength to resist wind damage in the GM trees, and tends to make the trees susceptible to disease  ( Low Lignin GM Trees and Forage Crops , SiS 23). A recent field study showed that the trees with reduced lignin decomposed more rapidly in the soil and that decay was associated with major restructuring of the soil microbial communities, the adverse impacts of which have yet to be fully evaluated .
Terminator trees no answer to containment
‘Terminator trees' are trees genetically modified to produce either no flowers or no pollen. For the most part, the methods to control flowering interfere with the genetic programme for floral development, or kill cells involved in floral development  ( Terminator Trees , SiS 26). Controlled cell killing is achieved using an enzyme barnase that breaks down RNA, in combination with a specific inhibitor called barstar . The barnase–barstar system has been approved for some transgenic food crops, but its toxicity and immunogenicity have been ignored or dismissed  ( Chronicle of An Ecological Disaster Foretold , SiS 18).
Much effort is dedicated to producing male-sterile or sterile modification events, which are supposed to prevent the spread of the transgenes. A male-cone specific promoter from Pinus radiata was used to drive a stilbene synthase gene from grape transferred to tobacco (as a first step to modifying pine), leading to greatly decreased pollen viability in the transgenic tobacco. The stilbene synthase inhibits flavonol synthesis resulting in sterile pollen . The system is still in preliminary development and seems quite ‘leaky' in that viable pollen is produced. The killing gene used in this male-sterile system is far less toxic to humans and animals than are many of the others, but the male-sterility trait will more readily spread to contaminate non-GM crops and natural species.
If and when GM trees are released for commercial use, many releases are likely to use terminator genes. Such genes, regardless of their inherent toxicity, will produce trees that do not sustain many mammal, bird and insect species that eat seeds or pollen. The plantations and contaminated natural forests will both become huge green desserts.
There has been a suggestion of using old forests as buffer zones to ‘contain' GM trees , and Dr. Claire Williams at Duke University, North Carolina in the United States sees transgenic contamination as inevitable, and introducing GM forest trees as opening a Pandora's box in ecological term. It could be a recipe for disaster as GM pollen contaminates indigenous species in the old forest and undermine its tightly balanced circular ecology that's vital for regulating climate  ( Why Gaia Needs Rainforests , SiS 20).
Gene therapy for trees could bring nightmares
Gene therapy uses vectors to deliver genes to treat disease or to enhance growth in humans or animals. Viral gene vectors have also been developed to rapidly produce large quantities of pharmaceutical proteins in plants. A locally replicating gene-silencing vector based on Poplar mosaic virus was developed to deliver gene-silencing RNA sequences . Gene silencing provides a means of regulating metabolic pathways and controlling plant diseases, and small synthetic RNA molecules have been developed to control plant viruses [29, 30]. Such synthetic RNA molecules are readily delivered using viral vectors, which could be sprayed onto forest stands from helicopters, for example, similarly to the current delivery of herbicides and fertilizers. Small RNA molecules require careful and extensive safety evaluations, as mice receiving ‘gene therapy' from small interfering RNA died in droves [31, 32] ( Gene Therapy Nightmare for Mice , SiS 31 ) . Forests sprayed with small RNA vectors could have disastrous effects on bystander plants and animals including humans.
New modifications of forest trees
The main focus of genetic modifications in forest trees has been on herbicide tolerance, insect resistance, and flowering discussed earlier [10, 20, 22], but there are some other new developments.
Transgenic poplar with enhanced growth was constructed using a maize uridinediphosphoglycosyltransferase gene accompanied by an Arabidopsis gene for acyl-CoA-binding protein, which enhanced the production of the growth hormone indoleacetic acid. The transgenic poplar grew much faster than the unmodified poplar .
An ethanol-inducible promoter from the fungus Aspergillus driving a GUS colour marker gene was used to transform aspen. Ethanol or ethanol vapour at concentrations as low as 0.5 percent induced the marker gene , and this presumably has applications in both the laboratory and in the field.
A bacterial gene for producing mannitol from fructose was used to induce salt- tolerance in Chinese white poplar ( Populus tomentosa ). The transgenic poplar grew about half as fast both in the presence and absence of high salt levels, but the untransformed poplar did not survive in the high salt environment .
Transformation of a poplar hybrid with the tryptophan decarboxylase gene from Camptotheca acuminate (tree of life, cancer tree) caused the gene to over-express. The tryptophan decarboxylase converts tryptophan into tryptamine, which provides resistance to caterpillars of Malacosoma disstria . Excess of tryptamine may result in hallucinogenic tryptamines, but that aspect was not explored in the report.
A transcription factor from Capsicum annuum (pepper) transferred to pine trees resulted in enhanced multiple stress tolerance (drought, salt and freezing). The transcription factor increases polyamine biosynthesis [37-39]. But polyamines such as putresine and cadaverine are toxic to humans.
China has planted over one million transgenic poplars since 2002. The plantations are located mainly in the northwest regions of Xinjang province, while a further 400,000 trees are planted in the headlands of the Yellow and Yangtze rivers  ( GM Trees Lost in China's Forests , SiS 26). China has an extensive programme of poplar genetic improvement including transgenic technology and marker assisted selection. Poplars modified with the Bt Cry1Ac gene or with a Cry1Ac gene fusion with the cowpea protease inhibitor gene have been most extensively deployed in China. The level of resistance of the transgenic trees to the main target insects has not dropped since deployment, but some insect pests are tolerant to the transgenic trees . There have been no reports on whether or not the resistant insect pests have proliferated since the transgenic trees were released.
Transgenic fruit trees
Fruit trees are much targeted by genetic engineers. Papaya and plum trees resistant to virus were the first trees approved, or petitioned for commercial release in the United States, with flagrant disregard of safety [6, 42] ( Allergenic GM Papaya Scandal , SiS 18).
A long term study of transgenic marker gene stability in Higan weeping cherry ( Prunus subhirtella ) showed that the markers were relatively stable but 91 percent of the transformation events also contained various lengths of the bacterial plasmid vector backbone, as Agrobacterium transformation is far from precise .
A grape stilbene synthase gene accompanied by a bar gene for herbicide tolerance was used to transform apple to enhance picied (reveratrol glucoside) production in the apple. Picied is both a phytoelexin for pest control and a health-promoting antioxidant .
Bacterial fire blight disease is a significant problem in pear and apple. Pears were transformed with a gene from a bacteria phage that dissolves the extracellular polysaccharide of the bacterial pest. The transgenic pears were only partially resistant to the bacterial pathogen but researchers thought improvements in the process might be possible .
In a pilot experiment, transgenic orange trees with a GUS marker gene driven by a CaMV promoter accompanied by a neomycin antibiotic resistance gene bore fruit that was harvested. The fruit was processed to make juice, to which was added bacterial plasmid DNA, yeast DNA and additional transgenic orange DNA. The orange juice-DNA soup was then pasteurized and stored. The pasteurization and acidic environment of the orange juice degraded all of the added and endogenous DNA molecules to molecular sizes smaller than the size required for bacterial transformation . The experiment would have been more informative if the ability of the transgenic orange juice to actually transform bacteria were investigated. Transformation may well occur before all of the DNA was degraded.
Trifoliate orange ( Poncirus trifoliate ) is a member of the family Rutaceae closely related to Citrus, and sometimes included in that genus, being sufficiently closely related for it to be used as a rootstock for Citrus. The plant is fairly hardy and will tolerate moderate frost and snow, making a large shrub or small tree 4-8 m tall. Because of the relative hardiness of Poncirus , citrus grafted onto it are usually hardier than when grown on their own roots. A gene from Arabidopsis CiFT that promotes transition from vegetative to floral development was transferred to trifoliate orange. The transgenic trifoliate oranges flowered as early as 12 weeks of growth in a green house while the untransformed plants takes several years . Reducing the generation time can greatly facilitate genetic improvement of the rootstock for commercial citrus production, subject to satisfactory safety assessment.
The biotechnology of temperate fruit trees and grapevines was reviewed in 2005 along with marker-assisted selection . It seems likely that marker assisted selection may provide the most long lasting and best fruit-tree improvement.
Moratorium on releases of all GM trees and GM forest trees should be banned
In conclusion, even though most of the work on transgenic forest and fruit trees is well meant and promises rich financial reward, no GM trees should be commercialised or released at this time. A moratorium on release of all GM trees is essential, and GM forest trees, in particular, should be banned. The inevitable spread of transgenes in pollen and seed cannot be prevented. Sterile trees promise no real remedy, as sterile forests will be green desserts at best, at worst, it will turn them from effective carbon sinks into massive carbon sources, thereby greatly exacerbating global warming .