3 Differences between Traditional and Molecular Breeding and Their Significance for Evaluating Crops
The controversy over GMOs is premised to some degree on the idea that they are constructed differently from traditional crops (such as by methods of hybridization or mutagenesis) and that these differences are relevant to the assessment of risk or quality. Both in the United States and Europe, some level of government oversight for new commercial crops is based on a distinction between traditional breeding and molecular breeding. Very simply, by traditional breeding I mean all methods of crossing among sexually compatible crops, mechanical methods of hybridization (including embryo rescue), and mutagenesis by chemicals or radiation (see chapter 1). Molecular breeding involves the use of recombinant DNA, CRISPR, or other forms of transgene engineering that involve the transfer foreign genes into a plant by laboratory methods. The idea that the method of breeding a crop should factor into how we assess its risks and benefits or how it should be regulated has drawn a lot of criticism. The schism has sometimes been framed as product-based versus process-based regulation.1
From the product-based perspective, the safety of a product should not be affected by the method used to assemble or construct it.2 What should matter are the properties that it possesses after it is constructed. A plant cell can undergo mutations from chemicals, radiation, or gene editing by CRISPR. According to the product-based perspective, the mutations themselves may result in a safe or unsafe plant, but the methods of creating them do not. Based on this idea, “the United States had adopted a regulatory stance toward agricultural biotechnology that declined to single out GM organisms for enhanced scrutiny based solely on their method of production.”3 This product-based approach has been reinforced by several reports from the National Academies of Sciences, Engineering, and Medicine (NASEM).
From the process-based perspective, traditional breeding has evolved over a long period of time. Breeders have had opportunities to understand what can go wrong in new crop development and to become alerted to a toxic crop before it is put on the market. There is no precedent for the speed at which molecular breeding has altered the germ plasm of crops. According to the process-based perspective, the possibilities of mixing genetic material from widely diverse organisms (plants, animals, fish, and bacteria) are very likely to create a higher number of adverse effects. Thus, it is argued, the process of molecular breeding should be looked at as a special case.
Europe and the United States have adopted different perspectives on these two approaches to risk-assessment regulation. The European community has chosen a process-based approach, and the United States has chosen a product-based approach. Unlike the American regulators, the Europeans largely believe that, compared with traditional breeding, molecular breeding is likelier to introduce unique, unanticipated risks and that because of the methods used to create new crops, its products should be evaluated as a special class. Under European Union rules, all genetically engineered (GE) crops have to undergo risk assessment, and all products made from those crops must be labeled.
In its 1977 white paper titled Research with Recombinant DNA: An Academy Forum, March 7–9, 1977, the National Academy of Sciences (NAS) concluded that there is no evidence that unique hazards exist either in the use of rDNA techniques or in the movement of genes between unrelated organisms.4 According to the NAS, the risks associated with introducing rDNA-engineered organisms are the same in kind as those associated with introducing unmodified organisms and organisms modified by other methods. By this view, assessment of the risks of introducing rDNA-engineered organisms into the environment should be based on the nature of the host organism, the genes transferred, and the environment into which the GMO is released and not on the method by which it was produced.5 In its most recent and comprehensive report on GE crops, NASEM writes: “While recognizing the inherent difficulty of detecting subtle or long-term effects in health or the environment, the study committee found no substantiated evidence of a difference in risks to human health between currently commercialized genetically engineered (GE) crops and conventionally bred crops.”6
The first issue addressed here is how scientists characterize the differences between traditional and molecular crop breeding. Then the question of whether the differences are relevant to assessing the health or environmental risks from the bioengineered crops is examined.
Wendy Harwood of the John Innes Centre in Norwich in the United Kingdom provides a useful framework for distinguishing between traditional and molecular breeding by classifying four sources of genetic variation for crop improvement. First, the primary gene pool represents the genes from the same and closely related species. These genes can be moved around by natural crosses. Second, the secondary gene pool consists of the genes from more distant species, where natural crosses are difficult but can be achieved through embryo culture methods. Third, the tertiary gene pool is only marginally sexually compatible with the plant of interest. Natural and cell culture crosses are usually not successful, but they can be done. Fourth, the quaternary gene pool consists of all organisms, including animals and microbes. Gene transfer can be done only by genetic engineering.
We have seen that, through human intervention, both traditional and molecular breeding methods exchange genes and other DNA across different species. These processes extend beyond the natural sexual reproduction of plants. In traditional breeding, clusters of genes are exchanged to an extent that depends on their genetic linkage or proximity on the chromosomes, whereas in molecular breeding, specific genes are transferred. In the former case, breeding is a hit-or-miss situation where the breeder cannot control how many and which genes are transferred. When a desired phenotype is selected, it comes with a cluster of other genes whose functions may not be understood. According to the National Academy of Sciences, when traditional breeders cross two sexually reproducing plants, tens of thousands of genes are mixed: “The major differences between traditional breeding and molecular biological methods of gene transfer lie neither in goals or processes, but rather in speed, precision, reliability and scope.”7 The NAS notes that molecular breeders move one gene at a time, whereas traditional breeders have to undertake many crosses before they observe the desired recombination of genes. There is no dispute about this characterization of the differences between the two breeding processes.8
There are disagreements about whether molecular breeding is a qualitatively different process—that is, whether molecular breeding radically transforms food production or whether it is a gradual extension of prior methods. Stephen P. Moose and Rita H. Mumm see ancient breeding as a form of biotechnology: “Prehistoric selection for visible phenotypes that facilitated harvest and increased productivity led to the domestication of the first crop varieties and can be considered the earliest examples of biotechnology.”9 Michael K. Hansen, senior staff scientist at Consumers Union, argues that genetic engineering is not an extension of conventional plant breeding but represents a “quantum leap” in the transformation of plants.10
Most commentators acknowledge that the essence of breeding under biotechnology involves cutting and splicing genes (recombinant DNA) to create genetic exchanges and that traditional breeding capitalizes primarily on reproduction between sexually compatible species to create genetic variability from which to select desirable phenotypes.
The concept of wide crosses in traditional breeding is discussed in chapter 2. It is extremely rare that breeders can cross species from two distinct families or any two levels in the taxa of biological classification. This does not mean that nature does not at times exchange DNA across widely separated taxa. As Nina Federoff and Nancy Marie Brown note, “Even crossing kingdoms to put a bacterial gene into a plant is not new: Agrobacterium has done it for millennia.”11
Bacteria and viruses that have evolved to infect plants and deposit some of their DNA into the plant’s chromosome or in its cytoplasm through circular plasmids do not create genetic variability for sexual reproduction of plants. However, neither natural nor human-activated traditional breeding can create crosses between higher taxonomic categories above genus (such as between kingdoms and orders) (see chapter 1, figure 1.1). Whether molecular breeding is qualitatively unique from traditional breeding methods is not an empirical question but rather depends on the criteria that are used for assessing uniqueness and the technology available for measuring “qualitative uniqueness.” For scientists who claim that molecular breeding is qualitatively distinct from traditional breeding, the next question is, What difference does that make for the safety and quality of the plant, and why should consumers be interested?
Michael K. Hansen of Consumers Union is among those scientists who believe that because molecular breeding draws DNA from varied sources across wide biological taxa, it will experience greater unpredictability. This unpredictability will include the safety of the product: “Because conventional breeding, including hybridization and wide crosses, permits the movement of only an extremely tiny fraction of all the genetic material that is available in nature, and only allows mixing and recombination of genetic material between species that share a recent evolutionary history of interacting together, one would expect that the products of conventional breeding would be more stable and predictable.”12
It is generally recognized that the fungibility of genes through recombinant DNA escapes any barriers that might exist to natural gene exchange between biological organisms. These barriers include organisms that are generally not in contact with one another, that are reproductively incompatible, or that are not part of a genetic exchange regime in nature.
But the fact that genetic exchanges are novel and unpredictable does not necessarily make them dangerous or ineffective. Molecular breeding must work at some level to make commercially successful products. If the desired phenotype is herbicide resistance, then the seed manufacturer will have to demonstrate its efficacy for that property. Others have asked what happens when the transgene construct (marker gene, promoter, transgene, and terminator sequence) induces other properties. Is there any reason to believe that molecular breeding is likely to induce more variable or more harmful properties than traditional breeding? Will the introduction of a single, highly specified gene construct create more unexpected properties than clusters of genes combined from compatible crosses?
There are no simple answers to these questions. Without laboratory experiments or other methods of risk assessment, there can be only speculation or hypotheses. Cases of unexpected outcomes have been observed from both transgenic experiments and traditional breeding.13 Biological systems are always defying predictable outcomes.
Yet generalizations from past experience are commonly heard, including the claim that “the many thousands of plants that had been made using these methods had not revealed unexpected hazards.”14 In one study that looked at the effect the transgene has, Maria Montero, Anna Coll, Anna Nadal, Joaquima Messeguer, and Maria Pia state that “around 35% of the unintended effects could be attributed to the process used to produce GM plants.”15
Does nature have any protective methods for avoiding untoward genetic exchange that can be overridden by human intervention? An answer to this question may be found in the DNA constructs created for the transgenes and in the unnatural methods of delivery. Scientists have offered different reasons that a genetically modified crop is distinct from traditional breeding, which direct outcomes of the molecular breeding process.16 First, the transgene insert may occur in the middle of a gene and may disrupt the genetic code of a recipient organism. This is referred to as the pleiotropic effects of integrated DNA on the host plant genome.17 In any of the methods for delivering a gene construct previously discussed, it is generally understood that the precise location of the transgene construct entering the chromosome is not predictable and that multiple copies of the transgene construct may be inserted on the chromosome. Some scientists assert that the method of transgenesis is “imprecise, uncontrollable and unpredictable” and that “This lack of precision, control, and predictability means that the genetic engineering process can, and almost always does, result in unintended effects.”18 The second wave of gene engineering applying CRISPR gene editing may reduce the imprecision and the unintended effects, but no one believes it will eliminate them.
Second, the inserted gene is very likely to disturb the action of neighboring genes. Transgene sequences are found to influence distantly located genes as far as 8 kilobases from the insertion site.19 The reason offered for this assertion is that a special promoter used in the transgene construct is designed exclusively for aiding foreign genes to be correctly expressed into the target plants. As previously mentioned, the cauliflower mosaic virus (CaMV) promoter can affect other genes in the plant chromosome: “Since the CaMV 35S is so strong, not only can it affect the introduced transgenes; it can also affect genes (either turn them ‘on’ or turn them ‘off’) thousands of base pairs upstream and downstream from the insertion site on a given chromosome and even affect behavior of genes on other chromosomes. Consequently, depending on the insertion site, a gene that codes for a toxin could be turned ‘on,’ leading to production of that toxin.”20
Third, scientists argue that the inserted gene can produce a new protein that may be alien to the recipient organism. This is based on the idea of posttranslational modification, where the new cellular environment for a foreign gene can alter the protein structure that is synthesized by the transgenic cell from its original form found in the source (or donor) organism.
Traditional breeding involves crosses from plants that have shared a large gene pool. Even in wide crosses there is sexual compatibility. Breeders are rearranging and exchanging genes in the preexisting gene pool rather than adding and creating totally new genes. Thus, traditional breeding involves the purposeful reshuffling of genes within sexually compatible species. Why are the chances of disturbing the host genome greater with foreign genes than with reshuffling genes in a closely related species? Some new methods of biotechnology respond to the argument about foreign genes being added (or inserted) to a plant genome. The term cisgenesis has been coined to describe an approach to plant breeding that uses only the existing genetic resources of the plant and related species. Hongwei Hou, Neslihan Atlihanm, and Zhen-Xiang Lu begin their journal article by discussing conventional plant breeding: “Traditional plant breeding uses crossing, mutagenesis and somatic hybridization for genome modification to improve crop traits,” with a goal of introducing “beneficial alleles from crossable species.”21 For transgenesis in molecular breeding, the authors claim, the sites of insertion are random and may have unpredictable side effects. They define a cisgenic plant as “a crop plant that has been genetically modified with one or more genes from a crossable donor plant.”22 One of the benefits of cisgenesis is that it avoids “linkage drag” where the transferred sequence is linked to unwanted sequences that may be transferred in traditional breeding due to their proximity to the target gene. They distinguish cisgenesis from transgenesis, although both use gene splicing and the tools of molecular genetics: “Although both transgenesis and cisgenesis use the same genetic modification techniques to introduce gene(s) into a plant, cisgenesis introduce[s] only genes of interest from the plant itself or from a crossable species, and these genes could also be transferred by traditional breeding techniques. Therefore, cisgenesis is not any different from traditional breeding or that which occurs in nature.”23
In some ways, this technique is closer to traditional breeding because it operates within a gene pool of sexually compatible plants. But it also may include promoters and markers, which are not found in conventional plant breeding. Cisgenesis may also be combined with new gene editing technologies, like CRISPR/Cas9, which are specific to a site and known to reduce the uncertainties of random insertions.
For those who share the views of Federoff and Brown in Mendel in the Kitchen, too much has been made of the distinction between traditional breeding and molecular breeding. In the former case, genes are altered through mutagenesis or cell fusion in order to create new alleles with favorable crop properties. Cisgenesis is just another technique that blurs the distinction between traditional and molecular breeding. It uses molecular techniques (such as the cutting and splicing of genes) but stays within a conventionally acceptable gene pool. This raises the question of how important and acceptable the phylogenic proximity of the plant genes is in breeding to the safety and quality of the product.
In the risk assessment of GMOs, the assumption has always been that greater caution is placed in species’ genetic material exchanges that have never occurred or that are unlikely to occur in nature. This was one of the precautionary consensus positions that came out of the 1975 International Congress on Recombinant DNA Molecules at the Asilomar Conference Center in Pacific Grove, California, which assessed the risks of gene splicing experiments. A coliform bacterium that was transformed with a tumor virus gene it had never seen was considered to be a risky experiment.24
When genetic exchanges have occurred naturally, dangerous outcomes to humans are more likely detectable by breeders or farmers because they have been expressed in nature and thus are not unexpected.
The introduction of gene editing technology to plant breeding is another path to blurring boundaries. The new tools can alter or silence gene function or expression through mutations in the plant’s genome without introducing foreign DNA. Could the mutation have occurred naturally? Is that the core principle on whether the product of gene editing should be regulated? The 2001 EU directive on biotechnology does not address these questions.25
CRISPR/Cas9 also can be used for introducing new genes into the “editing process.” But the precision of the location eliminates some of the uncertainties of the viral vectors and gene gun where the insertions cannot be controlled in the plant genome. Governments are in discussion over whether gene editing changes the regulatory landscape of transgenic plants.
There is a consensus within the scientific community that molecular breeding widens the range of possibilities for developing crops. There remain some disagreements over whether the frequency and seriousness of untoward and unexpected outcomes are increased with molecular breeding over traditional breeding, although as is shown in later chapters, that notion is put to rest by some prominent scientific groups. Mechanical methods of hybridization, embryo rescue, and mutagenesis in traditional breeding and cisgenesis in molecular breeding blur the boundaries between the two forms of crop production. Even so, the outliers of sexual versus nonsexual (recombinant DNA) crop production remain qualitatively distinct. Natural scientists are inclined to search for definitive, empirical answers to the question of whether there is a clear or unclear demarcation between GMOs and non-GMOs. Social scientists are more likely to draw cultural explanations for dividing lines between traditional and molecular breeding. Alonzo Plough and I looked at the cultural and technical rationality of risk and found that the former is less reductionist, less beholden to the prestige principle, and more aligned with political culture than the latter is.26 As an example, political scientist Hannes Stephan at the University of Stirling in the United Kingdom has noted that “In Europe, agbiotech has come to represent a ‘sounding board’ for contemporary anxieties about modernity, globalization, and the decline of national identity. In various combinations, these concerns give rise to a potent moral critique of the ‘unnaturalness’ of GMOs, which often crowds out utilitarian risk/benefit evaluations.”27 Chapter 8 continues this discussion by focusing on the risk of GMOs.
Notes
1. Sheila Jasanoff, “Product, Process, or Program: Three Cultures and the Regulation of Biotechnology,” in Resistance to New Technology: Nuclear Power, Information Technology and Biotechnology, ed. Martine Bauer, 311–331 (Cambridge, UK: Cambridge University Press, 1995).
2. Giovanni Tagliabye, “Product, Not Process! Explaining a Basic Concept in Agricultural Biotechnologies and Food Safety,” Life Science, Society and Policy 13, no. 3 (2017): 1–9.
3. Douglas A. Kysar, “Preference for Processes: The Process/Product Distinction and the Regulation of Consumer Choice,” Harvard Law Review 118, no. 2 (2004): 558.
4. National Academy of Sciences, Research with Recombinant DNA: An Academy Forum, March 7–9, 1977 (Washington, DC: National Academy of Sciences, 1977).
5. National Research Council, Introduction of Recombinant DNA-Engineered Organisms into the Environment: Key Issues (Washington, DC: National Academy Press, 1987), 23.
6. National Academies of Sciences, Engineering, and Medicine (NASEM), “Report in Brief: Genetically Engineered Crops: Experiences and Prospects,” May 2016, 1, https://
7. National Research Council, Introduction of Recombinant DNA-Engineered Organisms, 17.
8. Hakan Ulukan, “The Evolution of Cultivated Plant Species: Classical Plant Breeding versus Genetic Engineering,” Plant System Evolution 280 (2009): 133–142.
9. Stephen P. Moose and Rita H. Mumm, “Molecular Plant Breeding as the Foundation for Twenty-first Century Crop Management,” Plant Physiology 147 (2008): 969.
10. Michael K. Hansen, “Genetic Engineering Is Not an Extension of Conventional Plant Breeding,” Consumer Policy Institute, Consumers Union, January 2000, https://
11. Nina Federoff and Nancy Marie Brown, Mendel in the Kitchen: A Scientist’s View of Genetically Modified Foods (Washington, DC: Joseph Henry Press, 2004), 127.
12. Hansen, “Genetic Engineering,” 3.
13. Joy Bergelson, Colin B. Purrington, and Gale Wichmann, “Promiscuity in Transgenic Plants,” Nature 395 (1998): 25; Tomoko Inose and Kousaku Murata, “Enhanced Accumulation of Toxic Compound in Yeast Cells Having High Glycolytic Activity: A Case Study on the Safety of Genetically Engineered Yeast,” International Journal of Food Science and Technology 30 (1995): 141–146.
14. Federoff and Brown, Mendel in the Kitchen, citing the National Academy of Sciences, 150.
15. Maria Montero, Anna Coll, Anna Nadal, Joaquima Messeguer, and Maria Pia, “Only Half the Transcriptomic Differences between Resistant Genetically Modified and Conventional Rice Are Associated with the Transgene,” Plant Biotechnology Journal 9 (2011): 693–702.
16. Hakan Ulukan, “The Evolution of Cultivated Plant Species: Classical Plant Breeding Versus Genetic Engineering,” Plant Syst Evol 280 (2009):133–142.
17. Marcin Filipecki and Stefan Malepszy, “Unintended Consequences of Plant Transformation: Molecular Insight,” Journal of Applied Genetics 47, no. 4 (2006): 277–286.
18. John Fagan, Michael Antoniou, and Clare Robinson, “What Is Genetic Engineering? An Introduction to the Science,” in The GMO Deception, ed. S. Krimsky and J. Gruber (New York: Skyhorse Press, 2014), xxxiv.
19. Filipecki and Malepszy, “Unintended Consequences.”
20. Hansen, “Genetic Engineering,” 7.
21. Hongwei Hou, Neslihan Atlihan, and Zhen-Xiang Lu, “New Biotechnology Enhances the Application of Cisgenesis in Plant Breeding,” Frontiers in Plant Science 5, no. 389 (2014): 1–6.
22. Hou, Atlihan, and Lu, “New Biotechnology,” 1.
23. Hou, Atlihan, and Lu, 3.
24. Sheldon Krimsky, “A Worst Case Experiment,” chap. 18 in Genetic Alchemy: The Social History of the Recombinant DNA Controversy (Cambridge, MA: MIT Press, 1982).
25. Editorial, “Crop Conundrum: The EU Should Decide Definitively Whether Gene-Edited Plants Are Covered by GM Laws,” Nature 528 (December 17, 2015): 307–308.
26. Sheldon Krimsky and Alonzo Plough, Environmental Hazards: Communicating Risks as a Social Process (Dover, MA: Auburn House, 1988), 306.
27. Hannes R. Stephan, Cultural Politics and the Transatlantic Divide over GMOs (New York: Palgrave Macmillan, 2015).