Truth: Genetic engineering is crude and imprecise, and the results are unpredictable

Myth at a glance

GMO proponents claim that GM is a precise technique that allows genes coding for the desired trait to be inserted into the host plant with no unexpected effects. But the genetic engineering and associated tissue culture processes are imprecise and highly mutagenic. They lead to unpredictable changes in the DNA, proteins, and biochemical composition of the resulting GM crop, which can result in unexpected toxic or allergenic effects and nutritional disturbances, as well as unpredictable effects on the environment.

GMO proponents claim that GM is a precise technique that allows genes coding for the desired trait to be inserted into the host plant with no unexpected effects.

The first steps of making a GM plant – isolating the desired gene and cutting and splicing it to form the GM gene cassette in the laboratory – is indeed precise. But the subsequent steps are not. In particular, the process of inserting a GM gene cassette into the DNA of a plant cell is crude, uncontrolled, and imprecise. It causes mutations – inheritable changes – in the plant’s DNA blueprint.1 These mutations can alter the functioning of the natural genes of the plant in unpredictable and potentially harmful ways.2,3 Other procedures associated with producing GM crops, including tissue culture, also cause mutations.1

In addition to the unintended effects of mutations, there is another way in which the GM process generates unintended effects. Proponents of GM crops paint a simplistic picture of GM technology that is based on a naïve and outdated understanding of how genes are organised within DNA and how they work. They imply that they can insert a single gene with laser-like precision and insertion of that gene will have a single, predictable effect on the organism and its environment.

But manipulating one or two genes does not just produce one or two desired traits. Instead, just a single change at the level of the DNA can give rise to multiple changes within the organism.2,4 Such changes are known as pleiotropic effects. They occur because genes do not act as isolated units but interact with one another and are regulated by a highly complex, multi-layered network of genetic and epigenetic processes (epigenetic effects are inheritable changes in gene expression or cells caused by mechanisms other than changes in the underlying DNA sequence). Components of the GM gene, and the functions and structures that the GM genes confer on the organism, interact with other functional units of the organism.

Because of these diverse interactions, and because even the simplest organism is extremely complex, it is impossible to predict the impacts of even a single GM gene on the organism. The complexity of living systems makes it even more challenging to predict the impact of any given GMO on its environment.

In short, unintended, uncontrolled mutations and complex interactions at multiple levels within the organism occur during the GM process, giving rise to unpredictable changes in function as a result of the insertion of even a single new gene.

A seemingly simple genetic modification can give rise to unexpected and potentially harmful changes in the resulting GMO and the foods produced from it. The unintended changes could include alterations in the nutritional content of the food, toxic and allergenic effects, poor crop performance, and the emergence and spread of characteristics that harm the environment.

It is unlikely that potentially harmful changes would be picked up by the inadequate tests carried out in support of GMO authorizations. Even when changes are detected, they are often dismissed as irrelevant without further investigation.

These unexpected changes are especially dangerous because the release of GMOs into the environment is irreversible. Even the worst chemical pollution diminishes over time as the pollutant is degraded by physical and biological mechanisms. But GMOs are living organisms. Once released into the ecosystem, they do not degrade and cannot be recalled, but propagate and multiply in the environment, passing on their GM genes to future generations. Each new generation creates more opportunities for the GMO to interact with other organisms and the environments, generating even more unintended and unpredictable side-effects.

The GM process is highly mutagenic

The process of creating a GM plant is highly mutagenic. This means it damages the DNA, creating changes in the genome. Mutations can be beneficial or harmful. Very infrequently, a specific mutation can benefit the functioning of the organism. Such changes are the basis of evolution through natural selection. Much more frequently, mutations can harm the organism, for example, by giving rise to birth defects and cancer.

The GM process involves three kinds of mutagenic effects, as follows.1,2

Insertional mutagenesis

Genetic modification or the genetic engineering of an organism always involves the insertion of a foreign GM gene cassette into the genome (DNA) of the recipient organism. The insertion process is uncontrolled, in that the site of insertion of the foreign gene is random. The insertion of the GM gene cassette interrupts the normal sequence of the letters of the genetic code within the DNA of the plant, causing what is called insertional mutagenesis. This can occur in a number of different ways:

  • The GM gene can be inserted into the middle of one of the plant’s natural genes. Typically this blocks the expression of – “knocks out” – the natural gene, destroying its function. Less frequently the insertion event will alter the natural plant gene’s structure and the structure and function of the protein for which it encodes.
  • The GM gene can be inserted into a region of the plant’s DNA that controls the expression of one or more genes of the host plant, unnaturally reducing or increasing the level of expression of those genes.
  • Even if the GM gene is not directly inserted into a gene of the host plant or its control elements, its mere presence within a region of the plant’s DNA where host genes are located and active can alter the normal pattern of gene function – that is, the level at which a given gene is switched on. Thus it can alter the balance of the genes’ resulting protein products. The inserted gene can compete with gene expression control elements within the DNA of the host plant for the binding of regulatory proteins. The result will be marked disturbances in the level and pattern of expression of the host plant’s natural genes.

Since the insertion of the GM gene is an imprecise and uncontrolled process, there is no way of predicting or controlling which of the plant’s genes will be influenced and how.

Genome-wide mutations

In most cases, the insertion process is not clean. In addition to the intended insertion, fragments of DNA from the GM gene cassette can be inserted at multiple random locations in the genome of the host plant. Each of these unintended insertions is a mutational event that can disrupt or destroy the function of other genes in the same ways as the full GM gene, described under “Insertional mutagenesis”, above.

It is estimated that there is a 53–66% probability that any insertional event will disrupt a gene.1 Therefore, if the genetic modification process results in one primary insertion and two or three unintended insertions, it is likely that at least two of the plant’s genes will be disrupted.

Evidence from research indicates that the genetic modification process can also trigger other kinds of mutations – rearrangements and deletions of the plant’s DNA, especially at the site of insertion of the GM gene cassette1 – which are likely to compromise the functioning of genes important to the plant.

Mutations caused by tissue culture

Three steps of the genetic modification process take place while the host plant cells are being grown in a process called cell culture or tissue culture. These steps include:

  1. The initial insertion of the GM gene cassette into the host plant cells
  2. The selection of plant cells into which the GM gene cassette has been successfully inserted
  3. The development of GM plant cells into GM plantlets with roots and leaves with the help of plant hormones.

The process of tissue culture is itself highly mutagenic, causing hundreds or even thousands of mutations throughout the host cell DNA.1,2 Since tissue culture is obligatory to all three steps described above and these steps are central to the genetic engineering process, there is abundant opportunity for tissue culture to induce mutations in the plant cells.

In the case of plants that are vegetatively propagated (that is, not through seeds but through tubers or cuttings), such as potatoes, all the different types of mutations in a given GM plant resulting from the GM transformation process will be present in the final commercialized crop.

In the case of soy, maize, cotton, and oilseed rape (canola), the initial GM plant can be back-crossed (bred) with the non-GM parent variety to achieve closer genetic similarity. This back-crossing enables many, but not all, of the mutations incurred during the GM transformation process to be “bred out”.

However, given the fact that hundreds of genes may initially be mutated during insertion of the GM gene cassette and during tissue culture, there is a significant risk that a gene or genes crucial to some important property, such as disease- or pest-resistance, could be damaged. In another example, a gene that plays a role in controlling biochemical reactions in the plant could be damaged, making the plant allergenic or toxic, or altering its nutritional value.

The genetic engineer will not be able to detect and eliminate many such harmful mutations because their effects will not be obvious under the conditions of the development process. But these mutations would still be present in the commercialized crop and could cause problems. For instance, the non-GM parent crop may contain a gene that confers resistance to an insect pest. In the laboratory and greenhouse where the GM crop is developed, that insect will not be present and so the genetic engineers would have no way of knowing that the insect resistance gene present in the GM plants had been damaged. Only after the crop has been commercialized would it be discovered that the plants were no longer able to resist the insect pest.

How GM selects for host gene mutational effects

The GM gene cassette that is inserted into the host plant’s DNA (step 1 in “3. Mutations caused by tissue culture”, above) normally carries a selectable marker gene. Most commonly the marker gene confers antibiotic resistance on cells that have successfully incorporated the GM gene cassette into their DNA and expressed the genes in that cassette. As discussed in Myth 1.1, the antibiotic resistance marker gene enables the genetic engineer to identify which plant cells have successfully incorporated the GM gene cassette into their genome. Alternatively, a GM gene conferring tolerance to a herbicide can be used for selection of transformed plants.

It is important to note that either the antibiotic or herbicide-based selection process relies on the expression of these marker genes. This expression is required in order to make the plant resistant to the antibiotic or tolerant to treatment with the herbicide. If this gene does not express its protein, it will not confer resistance to the antibiotic or herbicide, and the cell will die upon exposure to it.

Not all regions of the plant cell DNA are permissive for the gene expression process to take place. In fact, the vast majority of any cell’s DNA is non-permissive. Any gene present in such a region of the plant’s genome will be silent – that is, it will not be expressed. Because the process of inserting the GM gene cassette (containing the GM gene(s) of interest and any associated antibiotic resistance marker genes) is essentially random, most insertions will occur in non-permissive regions of the plant cell DNA and will not result in expression of either the marker gene or the GM gene. Such cells will not survive exposure to the antibiotic or herbicide. Only when the GM gene cassette, including the antibiotic resistance marker gene, happens to have been inserted into a functionally permissive region of the plant cell DNA will the cell express the marker gene and survive exposure to the antibiotic or herbicide.

Permissive regions are areas of DNA where genes or control elements important to the functioning of the recipient plant cells are present and active. Thus, selection for antibiotic or herbicide resistance selects for cells into which the gene cassette has been inserted into a permissive region of DNA. Since these are also the regions that carry genes and control elements important to the function of the recipient plant cell, insertions in these regions carry a greatly increased likelihood of damaging the expression of genes important to the cellular function and even survival of the recipient plant cell.

In short, the selection for GM gene insertions in the GM transformation procedure maximizes the likelihood that incorporation of the GM gene will damage one or more genes that are active and important to the functioning of the host plant.

We conclude from this analysis of the mechanisms by which genetic modification can cause mutations that genetic modification is not the elegant and precisely controlled scientific process that proponents claim, but depends on a large measure of luck to achieve the desired outcome without significant damage. We also conclude that it is unwise to commercialize GM crop varieties without thorough assessment of potential harmful effects to health and the environment.

Is GM technology becoming more precise?

Technologies have been developed that are intended to target GM gene insertion to a predetermined site within the plant’s DNA in an effort to obtain a more predictable outcome and avoid the complications that can arise from random insertional mutagenesis.5,6,7,8,9,10

Some of these technologies use nucleases or “genome scissors” which allow the cutting of DNA and the insertion of new DNA in any position in the chromosomes. The most popular of these new genome scissors are TALENs (transcription activator-like effector nucleases), ZFNs (zinc finger nucleases), and most recently CRISPR-Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats).

These genome scissors are a combination of a unit to recognize specific regions of the DNA and an enzyme to cut both strands of the DNA at a sequence determined by the genetic engineer. When the cell senses that this double-strand DNA break has occurred, it stimulates the cell’s machinery to repair it.

There are two possible outcomes. First, simply allowing the repair to proceed where the cut ends of the DNA are joined back together again (a process known as “non-homologous end-joining”) introduces a mutation at the site of cutting by the genome scissors. This is because non-homologous end-joining repair is not perfect, and in the majority of cases, base units of DNA are lost from the ends of the DNA during the joining process.

Second, at the same time that the genome scissor gene is introduced into the plant cell, the genetic engineer can also introduce a separate DNA molecule that has the same regions in it as the region that he is trying to modify in the host genome, but which also contains a gene coding for the desired additional trait. The artificial gene that has been introduced can align with the corresponding region of the host cell’s DNA. In some instances the cell uses this second introduced DNA molecule as a guide to repair the double-stand DNA break in a process known as “homologous recombination”. The final result is the repair of the double-strand DNA break, but with the incorporation of the artificial gene at this pre-determined site.

By using these methods, genes can be knocked-out (silenced) or mutated, or new DNA including whole gene units can be inserted.

Proponents claim that these technologies offer “targeted genome editing”.11 However, these GM transformation methods are not failsafe. Two studies found that ZFNs caused unintended genomic modifications in off-target sites in human cell lines.12,13 The simple word for “modifications in off-target sites” is “mutations”. That is, these techniques can cause unintended mutations in other locations in the genome, causing a range of potentially harmful side-effects. In another investigation using human cells, CRISPR was found to cause unintended mutations in many regions of the genome.14

Biotechnologists still know only a fraction of what there is to be known about the genome of any species and about the genetic, biochemical, and cellular functioning of our crop species. That means that even if they select an insertion site that they think will be safe, insertion of a gene at that site could cause a range of unintended effects, such as disturbances in gene expression or in the function of the protein(s) encoded by that gene.

Even if there is no disturbance at the level of the gene, there may be disturbance at the level of the protein for which the gene encodes. For example, a plant may have an enzyme that is normally inhibited by a herbicide, meaning that the plant will die if that herbicide is applied. If the plant is genetically modified to alter the enzyme so that it is not inhibited by the herbicide (genetic engineered for herbicide tolerance), there may be knock-on effects. Enzymes are not totally specific. If the activity of the enzyme is changed, the plant’s biochemistry could be altered in the process, causing unknown chemical reactions with unknown consequences.

Moreover, because tissue culture must still be carried out for these new targeted insertion methods, the mutagenic effects of the tissue culture process remain a major source of unintended damaging side-effects.

Effects could include:

  • Unexpected toxins or allergens, or an alteration in nutritional value
  • Reduced ability of the GM crop to resist disease, pests, drought, or other stresses
  • Reduced productivity or vigour
  • Unexpected environmental effects, such as increased weediness.

According to a German newspaper, plants produced using these technologies are already being grown in greenhouses. The independent research institute Testbiotech says it is not known whether any of the plants have been released into the environment, adding, “There is, however, a clear lack of regulation to ensure that these plants, which are genetically modified organisms, undergo risk assessment.”15

Rapid Trait Development System: GM or not?

The biotechnology companies BASF and Cibus have developed oilseed rape and canola with a technique called RTDS (Rapid Trait Development System).16 According to Cibus, RTDS is a method of altering a targeted gene by utilizing the cell’s own gene repair system to specifically modify the gene sequence in situ, and does not involve inserting foreign genes or gene expression control sequences. The Gene Repair Oligonucleotide (GRON) that effects this change is a chemically synthesized oligonucleotide,17 a short, single-stranded DNA and/or RNA molecule.

Cibus markets its RTDS crops as non-transgenic and as produced “without the insertion of foreign DNA into plants”. The company adds that crops developed using this method are “quicker to market with less regulatory expense”.16 Cibus says that the RTDS method is “all natural”, has “none of the health and environmental risks associated with transgenic breeding”, and “yields predictable outcomes in plants”.18

However, GM is a process, and the definition of genetic modification does not depend on the origin of the inserted genetic material. Crops created with RTDS can and should be described as GMOs, since RTDS alters the genome in a manner that would not occur naturally through breeding or genetic recombination. The fact that no foreign DNA is inserted into the recipient plant’s genome is immaterial.

In addition, RTDS still involves tissue culture, which introduces genome-wide mutations. Some or all of these mutations (the latter in vegetatively propagated plants, e.g. potatoes) will be present in the final marketed product. Also, there will inevitably be off-target effects from the RTDS process. The intent of the RTDS process is specific targeting, but this technique is new and the research has not been done to assess the frequency and extent of off-target effects. The old saying, “Absence of evidence of harm is not evidence of the absence of harm,” is pertinent here.

To assess the fidelity and efficacy of the RTDS process and the extent to which unintended alterations take place at other locations in the genome during RTDS, many different studies will be needed. For instance, one important class of studies that must be carried out is whole genome sequencing of RTDS GMOs. Structural and functional analysis of the proteins present in RTDS GMOs (proteomics), as well as analysis of metabolites present (metabolomics) would also be required. In parallel, the functional performance of these RTDS GMOs should be assessed. The agronomic performance, the impact on the environment, and the quality and safety of the food derived from these RTDS-derived GMOs all need to be investigated, including via long-term toxicological feeding studies.

Even changing a single gene, whether it encodes an enzyme, a structural protein, a peptide hormone, or a regulatory protein, can cause unintended functional or structural disturbances at the level of the cell and the organism as a whole.

RTDS is a genetic modification process, albeit more targeted than other recombinant DNA techniques. Any crops or other organisms produced in this way must be treated in exactly the same way as crops altered using old-fashioned recombinant DNA techniques, namely thorough evaluation of functionality, utility, and safety.

“New” does not necessarily mean “better” or “safer”. RTDS and the other methods described above are new and they were designed to be more specific. This is a laudable intention, but empirical evidence needs to be gathered on the safety and efficacy of these new techniques.

It is interesting to note that the biotech company Cibus, in its publicity materials for the RTDS method, acknowledges the imprecision of standard genetic modification using recombinant DNA techniques.18


Genetic engineering and the associated tissue culture processes are imprecise and highly mutagenic. They lead to unpredictable changes in the DNA, proteins, and biochemical composition of the resulting GMOs, which can result in unexpected toxic or allergenic effects and nutritional disturbances, as well as unpredictable effects on the environment.


  1. Latham JR, Wilson AK, Steinbrecher RA. The mutational consequences of plant transformation. J Biomed Biotechnol. 2006;2006:1–7. doi:10.1155/JBB/2006/25376.
  2. Wilson AK, Latham JR, Steinbrecher RA. Transformation-induced mutations in transgenic plants: Analysis and biosafety implications. Biotechnol Genet Eng Rev. 2006;23:209–238.
  3. Schubert D. A different perspective on GM food. Nat Biotechnol. 2002;20:969. doi:10.1038/nbt1002-969.
  4. Pusztai A, Bardocz S, Ewen SWB. Genetically modified foods: Potential human health effects. In: D’Mello JPF, ed. Food Safety: Contaminants and Toxins. Wallingford, Oxon: CABI Publishing; 2003:347–372. Available at:
  5. Kumar S, Fladung M. Controlling transgene integration in plants. Trends Plant Sci. 2001;6:155–9.
  6. Ow DW. Recombinase-directed plant transformation for the post-genomic era. Plant Mol Biol. 2002;48:183-200.
  7. Li Z, Moon BP, Xing A, et al. Stacking multiple transgenes at a selected genomic site via repeated recombinase-mediated DNA cassette exchanges. Plant Physiol. 2010;154:622-31. doi:10.1104/pp.110.160093.
  8. Shukla VK, Doyon Y, Miller JC, et al. Precise genome modification in the crop species Zea mays using zinc-finger nucleases. Nature. 2009;459(7245):437-41. doi:10.1038/nature07992.
  9. Townsend JA, Wright DA, Winfrey RJ, et al. High-frequency modification of plant genes using engineered zinc-finger nucleases. Nature. 2009;459(7245):442-5. doi:10.1038/nature07845.
  10. Shen H. CRISPR technology leaps from lab to industry. Nature. 2013. doi:10.1038/nature.2013.14299.
  11. Wood AJ, Lo T-W, Zeitler B, et al. Targeted genome editing across species using ZFNs and TALENs. Science. 2011;333(6040):307. doi:10.1126/science.1207773.
  12. Pattanayak V, Ramirez CL, Joung JK, Liu DR. Revealing off-target cleavage specificities of zinc-finger nucleases by in vitro selection. Nat Methods. 2011;8(9):765-770. doi:10.1038/nmeth.1670.
  13. Gabriel R, Lombardo A, Arens A, et al. An unbiased genome-wide analysis of zinc-finger nuclease specificity. Nat Biotechnol. 2011;29(9):816-823. doi:10.1038/nbt.1948.
  14. Fu Y, Foden JA, Khayter C, et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol. 2013;31(9):822-826. doi:10.1038/nbt.2623.
  15. Then C. Free trade for “high-risk biotech”? Future of genetically engineered organisms, new synthetic genome technologies and the planned free trade agreement TTIP – a critical assessment. Munich, Germany: Testbiotech; 2013. Available at:
  16. Cibus. BASF and Cibus achieve development milestone in CLEARFIELD® production system [press release]. Undated. Available at:
  17. Cibus. What is RTDSTM? The Rapid Trait Development System in brief. 2013. Available at:
  18. Cibus. The evolution of plant breeding: RTDSTM versus other technologies. Undated. Available at: