Science in Society Archive

Unregulated Release of GM Poplars and Hybrids

USDA rubberstamps the largest ever collection of transgenic poplars with uncharacterized and dangerous constructs Prof. Joe Cummins and Dr. Mae-Wan Ho

This report has been submitted to the USDA on behalf of I-SIS

U.S. Department of Agriculture Animal and Plant Health Inspection Service (USDA/APHIS) Biotechnology Regulatory Services prepared an Environmental Assessment (EA) in response to permit application (06-250-01r) received from Oregon State University for field-tests of transgenic Populus alba and Populus hybrids. Comments on the EA are due by 17 August 2007 at Docket APHIS-2007-0018

The field tests to be carried out are in relatively small plots on a single 320 acre open site, broadly separated into three areas. Location A includes trials for reproductive sterility, gibberellin (GA) metabolism, reporter gene constructs and ‘activation tagging’ mutants, and also includes the clone bank containing trees that will not be allowed to flower. Two pairs of ramets (vegetative clones) per event will be planted in a completely randomized design with trees spaced at a distance of 6 to 10 ft. For certain trials, at least two pairs of ramets per event will be planted in a completely randomized design with trees spaced at 10 x 10 ft or 7 x7 ft. Location B will contain trials planted in two blocks to measure competition and yield of transgenic trees modified in GA metabolism. Each plot within respective blocks has 25 trees arranged in a completely randomized design. Location C contains trees engineered with the lignin modification gene. Two pairs of ramets per event are planted in a completely randomized design with trees spaced at a 10 X 10 ft distance.

The list of the trees is given below, followed by a brief description of the risks left out of consideration, based on which we object to the field releases.

Trees allowed to flower

There are five categories of transgenic poplars that will be allowed to flower, each category containing multiple genes and gene constructs.           

Genes conferring reproductive sterility

The transgenic plants engineered with the barstar and barnase genes derived from the bacterium Bacillus amyloliquifaciens are designed to confer male sterility. Barnase (a ribonuclease) is expressed only in the tapetum cells of the anther’s pollen sac during pollen development, resulting in the degradation of host RNAs and arrest of cell development. This blocks pollen formation and results in a male sterile plant. As barnase can kill Agrobacterium (the vector organism used for the transformations) and can inhibit the regeneration of transgenic plant cells, the barnase gene construct is accompanied by the barstar gene that produces the specific inhibitor of barnase.

Transgenic plants are also engineered with the DTA gene derived from the bacterium Corynebacterium diphtheriae (the causal agent of diphtheria) encodes the A-chain of the diphtheria toxin. Like the barnase/barstar construct, the preferential expression of the DTA genes in floral tissues results in reproductive sterility.

Further reproductive sterility genes for PTLF, PTD, PTAG, PTAP, PMFT, PCENL, PSV, PFCL, and PAGL24, all from Populus trichocarpa and genes for AG and AP1from Arabidopsis thaliana affect flowering and flower formation, and are designed to reduce or eliminate flowering, pollen production, or seed formation.

Genes conferring reduced stature/light response

Genes conferring reduced stature include those for GA metabolism: GAI and RGL1 from Arabidopsis thaliana, PcGA2 OXI1 from Phaseolus coccineus and PtaGA2 OXI1 from Populus tremula x alba.

Phytochrome receptor genes for PHYB1 and PHYB2 from the Populus trichocarpa were inserted to affect light response.

Gene that modify tree chemistry

The sequence for 4CL1 from Populus tremuloides was inserted to alter lignin levels. This is an antisense copy of the 4-Coumarate CoA ligase (4CL) gene, which results in a reduction in the messenger RNA of the target 4CL gene. The 4CL leads to a major branch point in phenylpropanoid metabolism. The product 4-coumaroyl:CoA is a precursor for lignin and flavonoids. Lignin, cellulose, and hemicellulose form the cell walls of xylem, which transports water and supports the tree.

Activation tagging

Populus clone 717-1B4 were engineered with a putative transcriptional regulator gene and a putative AP2 domain-containing transcription factor, both from Populus tremula x alba. These trees have been randomly hyperactivated (activation tagging) with genetic mutations in an attempt to develop “experimental domesticates”.

Other activation tagging mutant constructs contain only non-coding CaMV 35 S promoter regions and a selectable marker gene, which are expected to create random mutations via native gene disruption or upregulation.

Non-coding sequences

Some of the noncoding regulatory sequences introduced as part of the genetic constructs in these poplars were derived from the plants Nicotiana, Solanum, Arabidopsis, and the plant pathogens cauliflower mosaic virus, tobacco mosaic virus, Aspergillus nidulans, and A. tumefaciens.

Trees that will not be allowed to flower (maintained in the clone bank)

A few transgenic clones from earlier work are Populus species or species hybrids Populus trichocarpa x Populus deltoids of the sections Aigeiros and Tacamahaca. These will be retained in the cloned bank and pruned frequently, and will not be allowed to flower; they contain the Cry3A gene encoding a coleopteran-active insecticidal protein from Bacillus thuringiensis, the Green fluorescent protein  (Gfp) gene from Aequorea victoria, the nptII gene from Escherichia coli, the bar gene from Streptomyces hygroscopicus, and the PTLF gene from Populus trichocarpa in various combinations

Genetic constructions are not adequately explained in the EA

The numerous genetic constructs listed above will need a small army of trained technicians to do the required monitoring. The task is well nigh impossible as the EA gives little or no information on the structure of the transgenic constructs or their functions. One can only conclude that the USDA/APHIS is rubberstamping the open release of the largest collection of transgenes and transgenic constructs without even a pretext at health and environmental risk assessment. We shall attempt to point out some of the risks below, although that too, is well nigh impossible on account of the scant information given, but we do our best.

Risks of transgenes not considered

Reproductive sterility

The barnase toxin for cell ablation to control flowering or pollen production has been used extensively with trees and with crops. Barnase gene alone driven by a floral promoter tends to have leaky activity in tissues other than the flowers, thereby reducing the growth of the transgenic tree. However, when the barnase gene is accompanied by a gene for its natural inhibitor barstar driven by a promoter giving moderately low level expression, there is enough barnase activity left to ablate the floral tissue but not the vegetative tissues, at lease in theory, though this has not been borne out in field trials [1]. It is clear that barnase in the leaves, stems, and roots of trees will adversely affect not only the transgenic plant, but also the fauna and flora of the forest ecosystem. The toxicity of barnase to mammals is well known and we have pointed that out previously [2, 3] (Terminator Trees, SiS 26; Chronicle of An Ecological Disaster Foretold, SiS 18).

The diphtheria toxin A chain has been used as a cell suicide agent in transgenic plants but there does not seem to be any published study on the safety in animals from eating transgenic plants modified with the diphtheria a chain gene.

MADS-box genes are a large family of homeotic (regulators for development) genes that control plant organs; for example, the gene for PTD determines petal and stamen identity, the gene for PTAG determine stamen and carpel identity and the gene for PTAP determine flower and perianth identity. These have been altered by changing their native genetic code [4, 5] and are included in the transgenic poplars proposed for the field trials. MADS-box transgenes should not be presumed safe, as they are related to the extensively studied animal homeotic genes that regulate development [2] and may well be active in animals.

Dominant negative mutants (DNM) are genes altered to produce proteins that interfere with the function of the wild type gene from which they were derived. Arabidopsis DNM genes AP1 and AG with specifically altered MADS domain are in the present collection. Also in the collection are Populus genes for DNM and PTLF to alter flower initiation, the gene for PCEN to repress flowering and the genes for PAGL24 and PAGL to promote flowering. 

RNAi consists of small RNA molecules that regulate genes by turning off their transcripts. Populus RNAi genes have been used to construct RNAi sterility clones that have been co-transformed with multiple transgenes [4, 5].  RNAi constructs raise special concerns as many such RNAi sequences proved fatal to mice in gene therapy experiments [6] (Gene Therapy Nightmare for Mice, SiS

Reduced stature/light response

Dwarf poplars are presumed to be desirable because they cannot compete with tall trees should they escape to the wild. Poplars were engineered with gibberellin (GA) metabolism genes.

Gibberellins are plant growth substances (phytohormones) involved in promotion of stem elongation, mobilization of food reserves in seeds and other processes. Its absence results in the dwarfism of some plant varieties. The DELLA proteins are thought to act as repressors of GA-regulated processes, while GA is thought to act as a negative regulator of DELLA protein function. Poplars engineered with GA metabolism genes include the DELLA genes for

GAI and RGL1 from Arabidopsis thaliana, PcGA2 OXI1 from Phaseolus coccineus and PtaGA2 OXI1 from Populus tremula x alba. Arabidopsis DELLA proteins RGA and RGL2 jointly repress petal, stamen and anther development in GA-deficient plants, and this function is enhanced by RGL1 activity. PcGA2 OXI1 reduces gibberellins in the plant cells and down regulates GA synthesis. Phytochrome receptor genes for PHYB1 and PHYB2 from the poplar species Populus trichocarpa were inserted to affect light response, and hence the stature of the trees. Phytochrome is a photoreceptor, a pigment that plants use to detect light. It is sensitive to light in the red and far-red region of the visible spectrum. Many flowering plants use it to regulate the time of flowering based on the length of day and night (photoperiodism) and to set circadian rhythms. It also regulates other responses including the germination of seeds, elongation of seedlings, the size, shape and number of leaves, the synthesis of chlorophyll. PHYB1 and PHYB2 are quantitative trait loci  concerned with bud set and bud flush in Populus. [7,8,9]. The transgenes in these releases have not been studied regarding any potential untoward effects,

Tree chemistry

The 4CL1 gene from Populus tremuloides inserted to alter lignin levels. Genetic modification of lignin is a key alteration in producing trees destined for bioenergy production. Low lignin trees are likely to be more susceptible to pests and to be prone to wind damage because they lack mechanical strength [10] (Low Lignin GM Trees and Forage Crops, SiS 23).  Aspen (Populus tremuloides) modified for reduced stem lignin had normal cellulose content accompanied by reduced lignin content.  The transgenic aspen had reduced root carbon and greatly reduced soil carbon accumulation compared to unmodified aspen.  The trees accumulated 30% less plant carbon and 70% less new soil carbon than unmodified trees [11]. This makes the transgenic tree highly undesirable in terms of reducing carbon in the atmosphere, hence defeating the whole purpose of switching from fossil fuels to biofuels.

Activation tagging

Activation tagging is insertional mutagenesis using insertion vectors that contain strong transcription enhancer to up-regulate a gene near the insertion site. The insertions appear randomly in the genome, resulting in gain of function dominant mutations. The activated gene is easily recovered in organism such as Poplar which has been fully sequenced [12]. AP2 is an activating enhancer binding protein, a member of a large transcription regulator family in plants.  The AP2 domain is a DNA binding recognition signal.  The kind of mutation selection described above is efficient, but is hardly safe for field test releases, as it is likely to cause unintended insertional mutagenesis in a range of microorganisms and animals that interact with the transgenic plants. Insertion mutagenesis is a major cause of human cancers [13] (Slipping Through The Regulatory Net: ‘Naked’ and ‘free’ nucleic acids; ISIS and TWN publication)

Non-coding sequences

This is a potentially a very risky class of transgenes as they contain numerous regulatory sequences affecting all aspects of growth and development that have only been discovered within the past five years [14] (Subverting the Genetic Text, SiS 24).

Trees that will not be allowed to flower

Trees that will not be allowed to flower include poplars modified with the Bacilllus thuringiensis (Bt) Cry 3A gene to control beetles [15] and those modified with the PTLF gene controlling flower initiation [16]. Gene flow leading to transgene dispersal has been studied in hybrid poplars [17].  Mechanical pruning to prevent flowering seems risky in a large complex array of experimental trees it seems inevitable that the transgenes will be dispersed.

Horizontal gene transfer especially important for transgenic trees

As for all transgenes, dispersal by horizontal gene transfer is a distinct possibility, the extensive root system of trees in particular is a hotbed for horizontal gene transfer and recombination, which is why we have called for a Moratorium on all GM Trees and Ban on GM Forest Trees [18] (SiS 35).


Most of the previous field tests and petitions for non-regulated status were for plants or animals with single, or a few relatively simple transgenic constructs involving genes for enzymes and other proteins. The current application is a major departure in that it contains a plethora of complex, uncharacterized constructs, and where the main focus of gene modification is in transcription regulation and worse, non-coding sequences and hyperactivated insertion mutagenesis. The proposal made no attempt to present the genetic modifications of the many transgenic lines in a rational and coherent manner, with diagrams detailing the transgenic constructs in each line being tested along with an explanation of the function of each gene. The presentation of the lines is vague, and the many multiple transgene inserts are not clearly identified. There are so many separate lines being tested in one big 320 acre site that transgene escape form the site is bound to happen due to human error, and there will be plenty of opportunity for enhanced horizontal gene transfer and recombination of transgenes in the extensive root systems of the trees to generate the most exotic new pathogens out of some of the deadly toxin genes used. At the same time, too little information is provided to allow the escaped transgenes to be identified by anyone other than members of the research team proposing the releases. It would be a recipe for disaster and a travesty of regulation to permit the proposed releases.

This is number 35 of ISIS’ detailed submissions to US regulatory authorities for environmental releases GMOs.

Article first published 17/08/07


  1. Strauss S, Wei H, Meilan R, Brunner A, Skinner J, Ma C, Gandhi H and Strauss S. Field trial detects incomplete barstar attenuation of vegetative cytotoxicity in Populus trees containing a poplar LEAFY promoter:barnase sterility transgene Molecular Breeding 2007, 19, 69-85.
  2. Cummins J and Ho MW. Terminator trees. Science in Society 26, 16-17, 2005.
  3. Ho MW and Cummins J. Chronicle of an ecological disaster foretold. Science in Society 18, 26-27, 2003.
  4. Strauss S. Development &. Validation  of  Sterility  Systems in  Trees  Agenda 2020 Conference, Atlanta, GA. 2006
  5. Meilan,R.Brunner,A,Skinner,J. and Strauss,S. Modification of flowering transgenic trees 2001 eds. Morohoshi,M. and Komamine,A. Eds Molecualr Breeding of Woody Plants  Elsevier Science B.V. 247-57.
  6. Ho MW. Gene therapy nightmare for mice, could humans be next? Science in Society 31, 26, 2006.
  7. Brunner AM, Busov VB and Strauss,S Poplar genome sequence: functional genomics in an ecologically dominant plant species. Trends Plant Sci. 2004, 9(1), 49-56.
  8. Böhlenius H, Huang T, Charbonnel-Campaa L, Brunner AM, Jansson S, Strauss SH and Nilsson O.  CO/FT regulatory module controls timing of flowering and seasonal growth cessation in trees. Science 2006, 312(5776), 1040-3.
  9. Frewen BE, Chen TH, Howe GT, Davis J, Rohde A, Boerjan W and Bradshaw HD Jr. Quantitative trait loci and candidate gene mapping of bud set and bud flush in populus. Genetics 2000, 154(2), 837-45.
  10. Cummins J. Low lignin GM trees and forage crops  Science in Society 23, 38-9, 2004.
  11. Hancock JE, Loya WM, Giardina CP, Li L, Chiang VLand Pregitzer KS. Plant growth, biomass partitioning and soil carbon formation in response to altered lignin biosynthesis in Populus tremuloides. New Phytol. 2007, 173(4), 732-42.
  12. Busov,V,Mathias,F,Groover,A and Strass,S. Insertional mutagenesis in Populus:relevance and feasibility Tree genetics and Genomics 2005, 1, 135-42.
  13. Ho MW. Ryan A, Cummins J and Traavik T. Slipping Through the Regulatory Net: ‘Naked’ and ‘free’ nucleic acids, TWN Biosafety Series, 2001,
  14. Ho MW. Subverting the genetic text. Science in Society 24, 6-8, 2004.
  15. James RR., Croft BA. and  Strauss SH. Susceptibility of the Cottonwood leaf beetle (Coleoptera: Chrysomelidae) to different strains and transgenic toxins of Bacillus thuringiensis Environmental Entomology 1999,  28, 108-15.
  16. Rottmann WH, Meilan R, Sheppard LA, Brunner AM, Skinner JS, Ma C, Cheng S, Jouanin L, Pilate G and  Strauss SH.  Diverse effects of overexpression of LEAFY and PTLF, a poplar (Populus) homolog of LEAFY/FLORICAULA, in transgenic poplar and Arabidopsis. Plant J. 2000, 22(3), 235-45.
  17. Slavov, G.T., and DiFazio, S.P. (2004) Gene flow in forest trees: gene migration patterns and landscape modelling of transgene dispersal in hybrid poplar. In H.C.M. den Nijs, D. Bartsch, and J. Sweet (Eds.), Introgression from Genetically Modified Plants into Wild Relatives. CABI Publishing, Cambridge, MA, USA Pp 89-106.
  18. Cummins J and Ho MW. Moratorium on all GM trees and ban on GM forest trees. Science in Society 35 (to appear).

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?