Ref. Ares(2016)1831941 - 18/04/2016
Ref. Ares(2016)2053620 - 29/04/2016
Genetic Engineering in Plants and the
“
New Breeding Techniques (NBTs)”
xxxx@xxxxxxxx.xxxx
www.econexus.info
Inherent risks and the need to regulate
Briefing
December 2015
Dr. Ricarda A. Steinbrecher
Summary and conclusions
Over the last 5-‐10 years there have been rapid developments in genetic engineering
techniques (genetic modification). Along with these has come the increasing ability to make
deeper and more complex changes in the genetic makeup and metabolic pathways of living
organisms. This has led to the emergence of two new fields of genetic engineering that
overlap with each other: synthetic biology and the so-‐called New Breeding Techniques
(NBTs).
As regards NBTs, it is of concern that many efforts seem designed primarily to avoid having
to go through the regulatory process for GMOs, whilst choosing names that make it difficult
for the public to see that genetic engineering (genetic modification) is being used. This goes
alongside efforts to weaken the precautionary principle, which is there to guard against
adopting technologies that are considered likely to bring negative impacts on human and/or
environmental health in the future.
Currently there is a list of 7 “new” genetic engineering techniques before the European
Commission, which is deciding whether or not the products of these techniques, when
applied to plants, are covered by the EU GMO laws. Claims are being made by the industry
either that they are not GMOs according to the current legal definition of a GMO, are made by
techniques exempted from coverage, or that the final product, even if genetic engineering was
used at some point during its development, does not contain GM material and so is no longer
a GMO. The EC is currently working on the legal interpretation, as are many lawyers from
industry and civil society. It is important to be aware -‐ both in terms of legal interpretation
and of risks -‐ that some of theses techniques may also be used in combination with each other
or the same technique may be used several times over in order to achieve the intended effect.
This briefing looks at these 7 techniques from the scientific rather than the legal perspective
and aims to help readers to better understand the techniques and the inherent risks
associated with them. Whilst examining the likely unintended effects it has become evident
that all of the techniques claiming great precision are also found to have off-‐target effects
with unpredictable consequences. In fact, so called precision is actually a very imprecise
notion and does not equate to predictability.
In conclusion, the seven new genetic engineering techniques referred to as NBTs each
bring their own set of risks and uncertainties. Whilst many of these are the same as
with older GM techniques there are also serious additional concerns, such as the
potential environmental and health impacts of RNA dependent DNA methylation
(RdDM). Equally a new degree of uncertainty and risk of unintended effects arise from
the use of gene editing techniques (ZFN and ODM as well as CRISPR and TALENs). This
briefing concludes that there is a scientific case for classifying all these techniques as
GM and regulating their use with as much rigour as previous and current GM
techniques.
1
The 7 techniques under scrutiny are (following the EC’s own titles)1:
1) Zinc Finger Nuclease Technology (ZFN-‐1/2/3)
2) Oligonucleotide Directed Mutagenesis (ODM)
3) Cisgenesis/Intragenesis
4) RNA-‐dependent DNA methylation (RdDM);
5) Grafting (onto a GMO rootstock);
6) Reverse Breeding (RB);
7) Agro-‐infiltration (both Agro-‐infiltration ‘sensu stricto’ & Agro-‐inoculation)
An 8th technique was considered by the EC working group:
Synthetic genomics. However it
is generally regarded as a field within synthetic biology. Furthermore, no link to current
plant breeding programs has been reported, though its application is being researched for
example in microorganisms.
1) Zinc Finger Nucleases (ZFN) types -1, -2 and -3 (Gene–editing techniques) ZFN techniques are genetic engineering techniques that aim for deliberate changes to the
genetic make up and traits of an organism. They are also known as gene-‐editing techniques.
Other gene editing techniques are now becoming prominent, but ZFN is the only one
mentioned on the EU list.
The aim is to be able to change the sequence of the DNA, in order to delete, substitute or
insert DNA sequences at pre-‐determined locations in the genome. In this way, the
objectives are no different to any other engineering technique. In the case of the ‘editing’
techniques, this can mean small changes to 1-‐10 nucleotides2 (ZFN-‐1 and 2), or large
insertions of whole genes, including transgenes (ZFN-‐3).
For this purpose the DNA molecule first needs to be ‘cut’ at a specific location. ZFNs are
proteins that are custom-‐designed and utilised for this purpose. The “zinc finger” (ZF)
component can recognise a specific short stretch of DNA (9-‐12 bases) and the nuclease (N)3
component will cut the DNA at that site. It requires two ZFNs – each to dock diagonally
across the double stranded DNA – to cut through both strands. This DNA cut will then
trigger one of the cell’s two DNA repair mechanisms to stick the loose ends together again,
with a number of possible outcomes.
3 categories of ZFNs:
• ZFN-‐1: small site-‐directed random DNA changes, which may be small deletions,
substitutions or insertions of nucleotides. In this case the cell will ‘repair’ the break in a
random fashion, using a repair mechanism called ‘NHEJ’ (non-‐homologous end joining).
• ZFN-‐2: small site-‐directed intended DNA changes, such as ‘point mutations’ (one
nucleotide change). Here the repair will follow the instructions provided by a DNA
‘template’ that has been added (a stretch of DNA that has the same sequence as the
target area but with one or two small alterations or a short insertion). Here the repair
mechanism used is called ‘HR’ (homologous recombination).
• ZFN-‐3: large site-‐directed insertions of genes or regulatory sequences. In the genetic
engineering process a DNA template will be added as in ZFN-‐2, but the template will
also contain an additional long DNA sequence (eg one or more genes) for integration.
1 http://ec.europa.eu/food/plant/gmo/legislation/plant_breeding/index_en.htm
2
Nucleotides are the building blocks of DNA and RNA, which is the genetic material also referred to as
nucleic
acids. The nucleotides are made up in DNA of A, C, G and T (Adenine, Cytosine, Guanine, Thymine); and in
RNA of A, C, G and U (Adenine, Cytosine, Guanine, Uracil). It is the sequence of these letters that will
determine which protein is being produced or which instruction is given.
3 Nuclease: enzyme capable of cutting DNA.
Inherent risks and the need to regulate: ‘New Breeding Techniques (NBTs)
2
The gene for the specifically designed ZFNs will commonly be introduced into the plant
through genetic engineering with standard GM transformation, making it a GMO at this
stage. Once the ZFN proteins have been expressed and done their work, plant lines will be
selected that do not carry the transgene for the ZFN proteins. Alternatively, in a declared
effort to avoid being designated a GMO, plant virus expression systems have been
developed where the ZFN gene is meant to stay within the viral expression system. The
intention is that the ZFN transgene will not integrate into the plant’s own DNA – and thus
would not be passed on to future generations.
Commercial Applications ZFN-‐1, 2: The loss, change or insertion of a single nucleotide
(point mutation) can be sufficient to change traits in a plant, such as: herbicide tolerance,
male or female sterility, flower colour, delayed fruit ripening.
Unintended changes and risks:
• Off-‐target effects: ZFN technology is known for its non-‐specific binding to non-‐target
DNA and thus result in a significant level of off-‐target mutations in the genome. These
mutations can a) if in the coding sequence, result in changes of function of proteins, or
b) if in regulatory sequences, result in changes in the expression of genes, such as
increased presence of plant toxins, or absence of proteins important for nutrition, plant
defence or disease resistance.
• Template DNA (ZFN-‐2 and 3) may integrate randomly into the genome, as do transgenic
insertions, either as a whole or in parts, disrupting genes and regulatory sequences or
potentially resulting in altered proteins. This may lead to a decrease in performance,
heightened disease susceptibility, accumulation of toxins and residues, increase in
allergens.
• Transformation and transfection4 processes, including tissue culture5, are used in the
production of ZFN genetically modified plants. Such processes are known to lead to
additional mutations (with risks as detailed in the bullet points above).6
Conclusion: All three ZFN techniques are genetic engineering techniques, aiming for
deliberate changes to the genetic make up and traits of an organism. They are laboratory
techniques. All three are prone to off-‐target effects due to the ZFN activity, as well as the
effects of the genetic engineering processes, resulting in hundreds of mutations and
unintended effects. Further, the plant’s own repair mechanisms are not fully understood,
giving rise to additional uncertainties. Due to the process, modifications and risks, ZFNs are
GMOs and require full risk assessments.
Other gene-editing techniques:
There are several other gene-‐editing techniques, called TALENs, meganucleases and
CRISPR/Cas.7
Though different in detail, they all consist of nucleases directed to specific DNA sequences
where they will cut the DNA strand and evoke/trigger a natural repair system of the cell, as
detailed above.
4
Transformation of plant cells is the process of getting the external DNA into the cell and incorporated into
the plants DNA. The term
transfection of plants is more common when viruses are being used or when the
external DNA is not meant to integrate.
5
Tissue culture is the growing of plant cells in growth medium away from the plant. Through the use of
nutrients, special compounds, enzymes and various growth hormones, cells can 1) reach the right stage for
the transformation (insertion of the new gene sequence) and 2) then made to re-‐grow into a full plant. Tissue
culture – especially the type used for plant transformation – is known to cause genome-‐wide mutations.
6 See also Wilson et al. (2006), referenced in Background Information at the end.
7 TALENs (transcription activator-‐like effector nucleases), MN (meganucleases) and CRISPR/Cas (clustered
regularly interspaced short palindromic repeat system) – see also Agapito-‐Tenfen (2015) referenced in
Background Information at the end.
EcoNexus December 2015
3
The same considerations that lead to ZFN techniques being classified as producing GMOs
also apply to these gene-‐editing techniques. That is, they aim for deliberate changes to the
genetic make up and traits of an organism; they are laboratory techniques and are prone to
off-‐target effects, as well as unintended effects from the genetic engineering processes.
2) Oligonucleotide Directed Mutagenesis (ODM);
The aim is to create small and predesigned changes within very specific sites in genes, to
either change the function of the gene product or to stop its production. For the purposes
of ODM, an oligonucleotide8 that is a short stretch of a single-‐stranded nucleic acids
composed of a small number of nucleotides is synthetically produced.
It is designed to be almost identical to the DNA sequence of the target gene, except for 1-‐4
nucleotides. This will create a sequence mismatch when the oligonucleotide binds to the
target gene, inducing a site-‐specific DNA change (mutation) once the cell’s own DNA repair
mechanism is triggered, preserving the sequence of the oligonucleotide rather than the
original sequence.
Unintended changes and risks:
• Off-‐target effects: The oligonucleotide can bind to other DNA sites to which it is
sufficiently similar, where it is likely to cause unintended mutations. These in turn can
result in changes or loss of function of proteins, or changes in the expression of genes,
leading to problems such as increased presence of plant toxins.
• The oligonucleotide can also integrate into the plant DNA, in a manner similar to
transgenic insertions, disrupting genes and regulatory sequences or potentially
resulting in altered proteins.
• The utilisation of tissue culture and GM type transformation or transfection9 methods
are known to lead to genome-‐wide unintended mutations.
• Near target site mutations have been observed in ODM derived GM organisms.
• Depending on the oligonucleotides used, there is a risk that the oligonucleotides may
interfere with a cell’s regulation of gene expression, by triggering the RNAi pathway10,
which can lead to gene silencing. This can manifest in heritable changes, that may last
for many generations, and which depend on various factors that are not well
understood. This may be more the case for oligonucleotides that contain RNA
nucleotides.
Conclusion: ODM is a genetic engineering technique that can give rise to the same or to
similar direct and indirect negative impacts as current GMOs, both due to the intended
traits (eg herbicide tolerance, as performed by CIBUS for sulfonylurea herbicides in oilseed
rape)11, the processes and methods used and the potential integration of the
oligonucleotides. It thus requires full risk assessment.
8
Nucleotides are the building blocks of DNA and RNA, the genetic material also referred to as
nucleic acids.
The nucleotides are made up in DNA of A, C, G and T (Adenine, Cytosine, Guanine, Thymine); and in RNA of A,
C, G and U (Adenine, Cytosine, Guanine, Uracil).
Oligonucleotides are a stretch of genetic material (nucleic acid) and are commonly 20-‐200 nucleotides long;
they may consist of DNA, RNA or nucleotide analogues, or any combination of those. They are commonly
single stranded, but not always.
9 see footnote 5 & 6
10 RNAi pathway. The RNA interference (RNAi) pathway is an internal cell process in which RNA molecules of
various forms can, through a number of steps, lead to the silencing of genes.
11 CIBUS is using ODM under the name of Rapid Trait Development System (RTDS™)
Inherent risks and the need to regulate: ‘New Breeding Techniques (NBTs)
4
3) Cisgenesis and Intragenesis
Cisgenesis and Intragenesis are basically the same as transgenesis, but instead of sourcing
the DNA sequence from totally different species or inventing a new synthetic DNA
sequence, the sequence of the DNA inserted will be sourced from the same or closely
related species, those with which the plant would, in theory at least, be able to interbreed.
In ‘Cisgenesis’ the DNA inserted will have been made according to the exact sequence of a
gene found in a related donor organism.12 In ‘Intragenesis’ the inserted gene sequence is a
composite, made up of sequences and elements from different genes of one or more closely
related species; (see footnote for details)13
Unintended changes and risks:
• Whether or not the DNA sequences come from closely related species is irrelevant, the
process of genetic engineering is the same, involving the same risks and
unpredictabilities, as with transgenesis. There will be:
o random integration of the transferred DNA, capable of disrupting another gene or
interfering with the regulation of neighbouring genes (
positional effects).
o insertion-‐site mutations and genome-‐wide mutations resulting from the
transformation processes, including the effects of tissue culture14. These can include
deletions, rearrangements and multiplications of DNA sequences.
o potential for gene silencing of the introduced gene or the plant’s own genes if
promoter sequences share high similarity (homology).
• re cisgenesis: The fact that the inserted gene comes from a related species is no
guarantee that there are no unintended or unpredictable effects, as neither this
particular gene nor its product would have been present before in this genetic context
or position. Hence it may express in a different way from the way it did in the plant
from which it is taken and/or interact (eg interfere) with wider gene regulation or
metabolic pathways. This can give rise to altered behaviour and performance, higher
susceptibility to disease, increased fitness and/or invasiveness, altered composition of
signalling molecules15, nutrients, toxins and allergens.
• re intragenesis: the DNA sequences assembled in such a gene will never have existed in
this composition and in this regulatory context before. Their behaviour and
interactions cannot be predicted simply by knowing the DNA sequence or by knowing
that these sequences are derived from related organisms. Only a full analysis and strict
assessment of the actual effects and impacts can provide answers.
Conclusions: with regard to risks and potential negative impacts, there is little to
distinguish these techniques from transgenesis, therefore full molecular
characterisation and full risk assessments, including comprehensive feeding trials for
food and feed, are necessary.
4) RNA-dependent DNA methylation (RdDM); One aim is to obtain a new trait for a number of generations of seed, and to do so without
changing any DNA sequences, ie the sequence of nucleotides, within the organism, in the
hope of avoiding it being classified as GM. Instead, a process of RdDM16 can be utilised
12 The DNA inserted will not be taken directly from the donor organisms but rather be synthesised in vitro or
amplified within the microorganism
E. coli.
13 e.g. the promoter, coding and terminal sequences may be derived from different genes and species.
14 Tissue culture: see footnote 5
15
Signalling molecules will transmit information between the cells and tissues of multicellular organisms.
They can be simple molecules or complex proteins, such as growth hormones. Another category are the so-‐
called
semio-chemicals, that carry messages between different individuals either of the same species or
between different species.
16 RdDM is a form of RNA interference (RNAi).
EcoNexus December 2015
5
within the cell to silence a specific gene, so there will be no gene product from that gene.
This in turn can give rise to desired traits such as delayed fruit ripening, different coloured
flowers, enhanced content of specific nutrients, male sterility.
RNA-‐directed DNA methylation (RdDM) is a process where RNA molecules direct the cell to
add methyl groups (-‐CH3 groups)17 to certain nucleotides along a specific stretch of DNA in
order to silence a gene. (for details see footnote)18
The methylation of the promoter region of a gene will stop the expression of that gene.
Whilst such gene silencing is not a permanent alteration, it will be inherited for many
generations. In plants it is thought to eventually fade, but this is not true for all organisms,
e.g. in the nematode
C. elegans. However the triggers for this reversal of the methylation
are not known or understood.
How it works:
Any small double-‐stranded RNA with a sequence that matches the sequence of a stretch of
DNA will initiate the methylation of these DNA sequences, and thus silence the associated
gene. There are a number of ways to get specific sequences of double-‐stranded RNA into a
cell, for example:
(a) genetically engineering the plant with a gene that will produce such an RNA (with an
‘inverted’/reversed sequence) -‐ intended for permanent or transient gene silencing.
(b) to have transient gene silencing, ie for a few generations only, the inserted gene can be
removed (de-‐selected) by back-‐crossing in the breeding process.
(c) infection of plants with genetically engineered plant viruses (containing the targeted
promoter sequence), which will result in the silencing of the targeted gene through
methylation. (‘Virus Induced Gene Silencing’ (VIGS) – RdDM)
(d) spraying of plant with dsRNA (double stranded RNA).
Unintended changes and risks:
• off target effects: silencing of other genes, leading to altered traits, with potential
negative impacts such as the production and accumulation of toxins and allergens,
lowered nutrient content, disease susceptibility.
• the silencing of the target gene may not only stop the manufacture of the gene product
(ie protein), but depending on the possible involvement of this protein in other
pathways, may cause other unpredicted effects (often referred to as pleiotropic
effects). Consequences may include anything that is linked to those pathways, eg
growth factors, defence and signalling mechanisms, accumulation of compounds, etc.
• Specific to dsRNA: Depending on the methodology used, the presence of dsRNA
molecules in the food chain and the environment may negatively impact other
organisms exposed through ingestion or contact, for example in the case of sprays.
They cay be passed down the food chain, and may be amplified and lead to the
switching off of vital genes, which could have wide ecological and health consequences.
This is the intended outcome, for example, with insecticidal dsRNAs produced in
genetically engineered crop plants. This is a new and serious dimension of risk as
compared to older GMOs.
Conclusion: The crucial question is not whether or not the final product (the plant)
contains DNA sequences inserted through genetic engineering. It is rather that RdDM is a
very new and little understood technology with potentially serious negative impacts both
for consumption and the environment. It is a genetic engineering technology that, given the
risks, needs full regulation and risk assessments.
17 A methyl group (-‐CH3 group) is made up of one carbon linked to three hydrogen atoms.
18 In the case of higher organisms like plants and animals, only one of the four DNA nucleotides, the cytosine
base, can be methylated. In lower organisms such as bacteria, the adenosine nucleotide can also be methylated.
Inherent risks and the need to regulate: ‘New Breeding Techniques (NBTs)
6
5) Grafting: of non-GMO graft (scion)19 on GMO rootstock; (and vice versa)
Grafting (eg of fruit trees, grapevines, tomatoes)20 is a way to combine the strength or
desired traits of two organisms into one, without having to cross-‐breed them, eg rootstock
for disease resistance and the graft or scion for fruit flavour. Though in combination a
chimera (a single organism composed of genetically distinct cells), the graft and rootstock
in themselves will largely keep their own genetic identities with regard to the basic
sequence of their DNA.
The aim of using a GM rootstock is to create grafts that would benefit from the GM
characteristics without being defined as GM or sharing the GM DNA, though, as a whole, the
plants are GM.
Thus, strictly speaking, the tissue of the graft would not have been genetically engineered,
while the rootstock has. Yet many of the molecules produced by the GM rootstock, whether
proteins, certain types of RNA (eg: dsRNA), hormones, signalling or defence molecules, can
spread throughout the whole of the chimeric plant.21
Unintended changes and risks:
• impact of the GM rootstock on the environment: genetic engineering processes, such
as transformation and tissue culture (see footnote 1), are known to induce genome
wide mutations, as well as insertion site mutations. These can lead to altered and
unexpected traits, potentially with negative impacts on soil and environment.
Positional effects of inserted genes (e.g.: affecting the expression of neighbouring
genes) may equally lead to negative impacts.
• Compounds and metabolites produced by the GM rootstock will be present in the graft
and its products (eg in fruit) and may alter the composition of the fruit/product,
which in turn may alter the nutrient, allergen or toxin composition.
• If RNAi (RNA interference) methodology has been used in the GM rootstock, the gene
silencing active in the DNA of the rootstocks could transfer to the DNA of the graft via
the movement of small RNA molecules from the rootstock into the graft. This may
silence genes in the graft and alter its traits and vice versa.
Conclusion: To obtain the GM chimeric plant, by definition, requires genetic engineering,
and the risks arising are due to the genetic engineering (the inserted sequence, its location
and the transformation processes). The fact that the graft does not have any of the
genetically modified DNA does not necessarily reduce the risks to the environment,
ecosystems and/or human and animal health. As molecules/compounds can travel
between rootstock and graft, affecting the behaviour and molecular composition of the
graft, both the plant as a whole and the graft and its products need to be defined as GM and
fully assessed and regulated. This is particularly the case as the processes and interaction
between rootstock and grafts are still poorly understood.
19
Scion: A young part of a plant (shoot, twig or bud) that has been cut for grafting.
20
Grafting: has been common for woody plants for over 2000 years, and is often used for fruit trees, roses
and grapevine. Grafting of vegetables is more recent, mostly used for tomatoes and watermelon, but also
cucumbers and eggplant.
21 the transport is especially via the
phloem, which is a type of vessel tissue that transports water, food and
nutrients up and down (ie in both directions) to growing parts of the plant.
EcoNexus December 2015
7
6) Reverse breeding (RB)
RB is a GM technology intended to reconstitute genetically uniform and pure (homozygous)
parental lines from an existing hybrid whose parental lines are no longer available or no
longer exist. A major hurdle in this is that, every time gametes (reproductive cells) are
produced, the chromosomes previously acquired from the parental lines swap information
during the genetic recombination stage22, thus mixing the DNA. To avoid this, the selected
hybrid seed is genetically engineered to suppress genetic recombination (using RNAi).
With the help of tissue culture, individual resulting gametes are used to reconstitute plants
with two sets of the same chromosomes (called ‘double haploid’). At a later stage the GM
gene is deselected and parental lines chosen that – in combination – will give rise to the
envisaged hybrid.
Unintended changes and risks:
• As the same genetic engineering processes are used, both to insert genes and to
reconstitute plants through tissue culture, the same risks and unpredictable outcomes
are possible as with other GM. There will usually be:
o insertion site and genome wide mutations (eg deletions, rearrangements,
multiplications) resulting from the transformation processes, including tissue
culture, with unpredictable consequences that could lead to altered performance
and disease susceptibility, accumulation of toxins, increased production of allergens,
changes in nutritional composition.
o The vast majority of these mutations would remain present in the reconstituted
parental lines even if the GM gene itself is deselected and with it the mutations most
closely associated with the insertion site itself.
• The GM gene silencing method of RNAi may lead to non-‐target gene silencing of other
genes, effects that will be maintained for many generations of seed. Thus tests for
performance and compositional analysis will need to be carried out, followed by full
risk assessments. These need to take place before initial planting, but also several
generations later, once the intended and unintended gene silencing has faded and it
should include feeding trials.
• Functional components or full sequences of the GE gene may have integrated
themselves elsewhere in addition to the primary insertion. They may thus not be
removed in the de-‐selection process, leaving them potentially still able to initiate gene
silencing in the target region or in off-‐target areas.
Conclusion: Parental lines as well as the combined new hybrids need to be tested for the
presence of GE sequences as well as for unintended effects due to off-‐target gene silencing
and transformation induced mutations, which have the potential to, for example, result in
altered performance and disease susceptibility, accumulation of toxins, increased
production of allergens, changes in nutritional composition. Full risk assessments are
required.
22
Meiosis is a process of cell division resulting in gametes, i.e. the male or female reproductive cells, each
containing half the number of chromosomes (
haploid) as compared to the ordinary plant cells (
diploid), that
has two sets of chromosomes, one from each parent.
Inherent risks and the need to regulate: ‘New Breeding Techniques (NBTs)
8
7) Agro-infiltration: Agro-infiltration ‘sensu stricto’ & Agro-infection This method involves two distinct technologies. It is not intended to result in specific GM
genes being stably inserted and integrated into a plant genome, but rather for such genes to
be present within the plant cell transiently, for a maximum of just one generation.
To this end, genes either coding for specific proteins or for RNAs to interfere with the
plant’s own genes (eg via RNAi) are engineered into the plasmid23 of
Agrobacterium
tumefaciens.24 A solution of such Agrobacteria or their plasmids is then used to treat
specific tissues of living plants (eg leaves) so as to have the plasmids with the GM genes
delivered to the cells in that tissue, where these genes will be expressed in the specific RNA.
The aims may be to: test potential transgenes; study the function of the plant’s own genes
(eg through gene silencing via RNAi); express and produce high value proteins in plants (eg
pharmaceuticals); produce plants, seeds, hybrids with altered traits through RdDM (RNA
dependent DNA methylation – see section 4); or use as a delivery system for other GM-‐
based NBT tools, such as site directed nucleases.
Two distinct technologies:
Agro-‐infiltration ‘sensu stricto’ (ie in the narrowest meaning): The intention is to keep the
gene expression and effect localised, thus the genetic construct prepared and used is not
expected to replicate in the receiving cell.
Agro-‐infection: The intention is to spread the specific GM gene throughout the whole plant
into almost all the tissues, but without integrating the gene into the plant’s DNA. For this
purpose, in addition to the chosen gene, the gene construct contains a viral vector
sequence in order to replicate the construct in all infected cells. The gene for the RNA is
meant to be expressed from its location on the vector, ie not from a location on the
plant’s DNA.
Unintended changes and risks:
• Though applied locally, the gene construct can spread throughout the plant, due to the
agrobacteria and/or the viral vector sequences used. Although meant to be transient,
the genetic material may become integrated into the plant’s DNA, including
reproductive tissue, thus unintentionally giving rise to GMOs and to GM progeny.
• Integration may happen at random places within the genome and may also involve any
of the DNA sequences introduced, including vector DNA. Disruption of genes due to
positional effects or due to sequences present in the gene construct could give rise to
negative impacts on plant performance, environment and biodiversity, or on its safety
as food.
• Accidental release of genetically engineered Agrobacteria into the environment could
occur (either due to the spread of and contamination from infiltrated plant material
that has been discarded or removed, or simply through spillage, eg from lab,
greenhouse or test plots). This in turn could give rise to adverse effects if the gene
constructs get transferred to other plants or to microorganisms.
• Replication can occur at levels too low to detect for long periods of time, increasing the
chance for either integration or mutation that makes the DNA stably heritable.
Conclusion: plants subjected to Agro-‐infiltration (including agro-‐infection), along with any
of their parts and products as well as their progeny need to be tested for the presence of
DNA sequences from the vector and/or the gene construct, as well as for the presence
and effects of gene silencing, if that was the initial aim of the agro-‐infiltration.
23 A
plasmid is a circular ring of DNA in a bacterial cell that can replicate independently of the chromosomes
and can be passed on to other bacteria. Here it contains the transgenes.
24 Using
Agrobacterium as a gene shuttle is one of the major means of performing GM).
EcoNexus December 2015
9
For background information and further reading, four of the most important
references:
Eckerstorfer M, Miklau M and Gaugitsch H. (2014). New plant breeding techniques and
risks associated with their application. Technical Report. REP-‐0477. Environmental Agency
Austria. ISBN: 978-‐3-‐99004-‐282-‐3
http://www.umweltbundesamt.at/fileadmin/site/publikationen/REP0477.pdf
Heinemann JA, Agapito-‐Tenfen SZ, and Carman JA. (2013). A comparative evaluation
of the regulation of GM crops or products containing dsRNA and suggested improvements
to risk assessment.
Environment International 55: 43–55
http://gmojudycarman.org/wp-‐content/uploads/2013/06/comparative-‐evaluation-‐of-‐
the-‐regulation-‐of-‐GM-‐crops-‐or-‐products-‐containing-‐dsRNA-‐and-‐suggested-‐improvements-‐
to-‐risk-‐assessments.pdf
Wilson AK, Latham JR and Steinbrecher RA. (2006). Transformation-‐induced mutations in
transgenic plants: analysis and biosafety implications.
Biotechnology and Genetic
Engineering Reviews 23:209–237
http://econexus.info/publication/transformation-‐induced-‐mutations-‐transgenic-‐plants
Agapito-‐Tenfen, SZ and Wikmark, O-‐G (2015). Current status of emerging technologies for
plant breeding: Biosafety and knowledge gaps of site directed nucleases and
oligonucleotide-‐directed mutagenesis
. GenØk Biosafety Report 02/15.
http://genok.com/arkiv/4288/
Inherent risks and the need to regulate: ‘New Breeding Techniques (NBTs)
10