8

Future Genetically Engineered Crops

With a foundation in emerging genetic-engineering technologies provided in the preceding chapter, the purpose of this chapter is to offer a preliminary assessment of the potential to develop new traits and crops on the basis of the emerging technologies and evaluate how the traits themselves could affect agriculture and society. The committee also considers the social and economic forces that may foster or slow the development of these new traits and crops. The committee defines new traits as ones that have yet to be commercialized as of June 2015 even if the traits had received regulatory approval by that time or if there were published descriptions in the literature of the traits in crop plants.

As discussed in Chapter 3, few genetically engineered (GE) traits have been commercialized. They have mostly been incorporated into major commodity crops grown on millions of hectares worldwide. Many predicted GE traits are expected to be introduced into crops planted over smaller areas than the major commodities such as maize (Zea mays) and soybean (Glycine max). Some of those traits will be variations on GE traits that existed in 2015, such as resistance to new herbicides or the stacking of currently available traits. New and more diverse traits are also expected (Parisi et al., 2016).

It is not possible to predict with certainty the traits that will and will not make it to market or be diffused through nonmarket mechanisms in the future. The outcome will depend on environmental challenges that need to be addressed (for example, climate change), political–economic drivers, the regulatory landscape, and the rate of scientific advances, which is in part a function of the availability of public and private sci-

ence funding. Genetic-engineering technologies have advanced rapidly, as outlined in Chapter 7, but the genetic basis of complex traits—such as drought tolerance (Yue et al., 2006), water-use efficiency (Easlon et al., 2014), and nitrogen-use efficiency (Rothstein et al., 2014)—is not fully understood. Only continued public funding of basic research will enable further advances in understanding of the physiological, biochemical, and molecular basis of these important traits. Furthermore, the potential for any of the new scientific insights and technologies to yield public benefits will depend on the social, political, and economic contexts in which they are generated and diffused.

IS GENETIC ENGINEERING NECESSARY TO DELIVER THE NEXT GENERATION OF PLANT TRAITS?

As noted in Chapter 7, new discoveries in molecular biology and genomics are offering novel tools for increasing the efficiency of both conventional plant breeding and genetic engineering of crops. The committee heard from invited speakers who concluded that the molecular approaches to conventional breeding were safer than and superior to genetic engineering and who raised the question of whether genetic engineering was necessary even to deliver useful new crop varieties (Cotter, 2014; Gurian-Sherman; 2014; Shand, 2014). It is the case for many crops that the genetic potential that exists in sexually compatible germplasm is quite large, and much crop improvement can be achieved by conventional breeding. However, conventional breeding, whether or not it incorporates such modern tools as marker-assisted selection (MAS; discussed in Chapter 7 section “Modern Plant-Breeding Methods”), is limited by the genetic potential in sexually compatible germplasm. Thus, conventional breeding could never achieve some traits, such as the incorporation of a natural insecticide from another species into a particular crop plant (for example, a Cry protein from Bacillus thuringiensis). As discussed in Chapter 4, progress in crop improvement has been brought about by the combined use of conventional breeding and genetic engineering. Furthermore, the committee finds that the distinction between conventional breeding and genetic engineering is becoming less obvious. For example, genome editing can now be used to make a trait-changing, single-nucleotide substitution in a specific gene, and the same nucleotide change can be made with the TILLING (targeting induced local lesions in genomes) method described in Chapter 7. The TILLING method would be considered conventional breeding within the context of most national regulatory systems because it uses radiation-induced or chemical-induced mutations linked with genomic screening to find and isolate the specific nucleotide change. These identical changes made by genome editing (genetic engineering) or TILLING (con-

ventional breeding) would have the same effect on crop qualities or yield, and either could have unintended effects.¹

Proponents of the argument that MAS can and should replace genetic engineering as a tool to deliver new traits (Vogel, 2009; Cotter, 2014) posit that more traits have been successfully introduced with MAS than with genetic engineering and that this approach is better suited to developing crop varieties with traits conducive to sustainable agricultural production (Cotter, 2014). As noted above, if the genetic potential for a trait of interest is present in sexually compatible germplasm, then conventional breeding with or without the efficiency provided by MAS could be used to introduce that trait. However, if the genetic potential is not present in sexually compatible germplasm, MAS will be of no value. For example, plant breeders have worked for decades to develop an alfalfa (Medicago sativa) that is “bloat-safe” for cattle. Such a variety would contain higher concentrations of condensed tannins to protect protein from being too rapidly metabolized by microorganisms in the cow’s digestive tract (Lees, 1992; Coulman et al., 2000). However, despite years of plant-breeding efforts, alfalfa germplasm with sufficient tannin in vegetative tissues has not been obtained, and forage species that have naturally high levels of tannin in their leaves or stems are not sexually compatible with alfalfa. Therefore, on the basis of current knowledge, genetic engineering is still the only way to introduce this trait into alfalfa (Lees, 1992).

With an ever-increasing understanding of plant biology, commercial plant-breeding programs can now be armed with knowledge of the exact physical nature of the genes that need to be manipulated at the outset. Knowing the phenotypic effects of changing the expression of the target gene is critical for generating a desired trait. Such information can be obtained through the application of -omics technologies (discussed in Chapter 7) and could be of great value in both genetic engineering and conventional breeding of the trait, depending on which is most efficient for the specific trait.

At the level of the expression of a specific target gene, genetic engineering can bring about four types of change: reduced expression of a gene, complete loss of gene function, increased expression of a gene (including novel expression of a gene that is already in the plant in a tissue type in which it is not typically expressed), and introduction and expression of a new gene that is not found naturally in the target species or accession. Genome-editing approaches also make it possible to alter the coding sequence of an endogenous gene so that it encodes a protein with altered function (for example, an enzyme with altered catalytic properties or sub-

__________________

¹ As noted in Chapter 7, the TILLING (conventional breeding) approach has a higher probability of unintended effects that result from changes to nontarget genes.

strate specificity); this altered gene could be a novel allele of a gene in the target species or a favorable allele of a gene present in another genotype of the target species. In all cases except the introduction of a novel gene into the target species, it is theoretically possible—although not necessarily practical—to introduce the trait without genetic-engineering approaches by using TILLING or genetic selection in most crop species. Genetic engineering may accelerate the process but would not be essential for reaching the desired end-point.

FINDING: New molecular tools are further blurring the distinction between genetic modifications made with conventional breeding and those made with genetic engineering.

FINDING: Treating genetic engineering and conventional breeding as competing approaches is a false dichotomy; more progress in crop improvement could be brought about by using both conventional breeding and genetic engineering than by using either alone.

FINDING: In some cases, genetic engineering is the only avenue for creating a particular trait. That should not undervalue the importance of conventional breeding in cases in which sufficient genetic variation is present in existing germplasm collections, especially when a trait is controlled by many genes.

PROJECTION OF HOW EMERGING GENETIC-ENGINEERING TECHNOLOGIES WILL AFFECT TRAIT DEVELOPMENT

The emerging genetic-engineering technologies discussed in Chapter 7 have the potential to change future crop production substantially with respect to quality, quantity, and applications because of the following three paradigm-changing capabilities in precision, complexity, and diversity.

Increased Precision of Genome Alteration and Gene Insertion

As discussed in Chapter 7, most GE crops commercially available in 2015 were engineered by using Agrobacterium tumefaciens-mediated or gene gun-mediated transformation, both of which result in insertion of the DNA into semirandom locations in the genome. Variation in expression of the transgene was observed routinely because of positional effects that depend on where in the genome the DNA was inserted; this required screening of large numbers of GE plants to identify the optimal transgenic individuals. The emerging technologies discussed in Chapter 7 enable insertions into specific locations in the genome that are selected to enable appropriate expression.

Like earlier genetic-engineering approaches, emerging genetic-engineering technologies can insert a gene from a very distantly related organism into the genome of the target species, but they also enable the modification of endogenous genes within a species. Such technologies as genome editing discussed in Chapter 7 permit a change or changes to one or more nucleotides in a genome and thereby focus trait alteration on a single targeted gene in a precise and predictable manner with few if any off-target changes.

Increased Complexity of Genomic Changes

Emerging genetic-engineering technologies have the potential to increase the complexity of engineered changes substantially because multiple genes can be introduced or “stacked” into a single target locus in which genes (transgenes, native genes, and cisgenes) are inserted as a “cassette” into the target locus. New paradigms—such as multiple genes for one trait, multiple genes for multiple traits, or, in some cases, one gene for multiple traits—are expected to be seen in future GE crops.

The ability to insert multiple genes expands the possibilities for the alteration of plants’ natural metabolic pathways (metabolic engineering; Box 8-1). Metabolic engineering ranges from simple (modification of a single gene that changes flux in an existing pathway²) to complex (multiple genes manipulated to introduce a new pathway). If multiple genes have been used and are from different organisms or have been modified or synthesized de novo outside the organism of origin, the approach may be referred to as synthetic biology.

Increased Diversity of Engineered Crops and Traits

Perhaps the most dramatic change due to emerging genetic-engineering technologies is the diversity of the crops and traits that will be engineered. Although commercialized GE crops until 2015 were predominantly high-production commodity crops (maize, soybean, and cotton), the committee expects an increase in crop species that are genetically engineered and an expansion in the application of crop genetic engineering for food, feed, and human health.

FINDING: The emerging genetic-engineering technologies outlined in Chapter 7 will lead to increased precision, complexity, and diversity in GE crop development. Although genome editing is a new technique and its regulatory status was unclear at the time the committee was

__________________

² Flux is the flow of intermediate metabolites toward the final products.

writing this report, the committee expects that its potential use in crop improvement in the coming decades will be substantial.

FUTURE GENETICALLY ENGINEERED TRAITS

Throughout this report, the committee has emphasized the idea that GE crops that were commercially available in 2015 had few engineered traits. That is expected to change in the future. Figure 8-1 summarizes a selection

of potential future GE traits, and specific examples are described below. The list in Figure 8-1 is far from exhaustive. It contains examples that span the spectrum of development: initial trait identification or conceptualization, empirical proof-of-trait expression and function in plants, initial field testing, and preparation for commercialization.

Tolerance of Biotic Stress

Stress caused in plants by other living organisms is termed biotic stress. Resistance to insects through the incorporation of Bacillus thuringiensis (Bt) proteins into crops is an example of a GE form of tolerance of biotic stress. Most GE crops commercially available when the committee was writing its report were engineered to have resistance to herbicides or to biotic stress in the forms of insects and, to a much smaller extent, viruses. Research was being conducted on other ways to combat those stressors and on ways to provide protection against sources of biotic stress that had not yet been addressed with genetic engineering.

Resistance to Fungi and Bacteria

Resistance to many fungi and bacteria is controlled in plants by major resistance genes (R genes) that confer qualitative resistance (that is, resistance to a particular strain of a pathogen). However, pathogens can readily mutate to overcome resistance conferred by a single major gene—an example of the “evolutionary arms race.” There is a long history of pathogens evolving to overcome resistance (R) genes introduced into crop varieties by plant breeders (Sprague et al., 2006; Stall et al., 2009). Plant breeders have therefore kept ahead of pathogens by introducing new varieties with combinations of different R genes that are found in plants in gene banks (Gururani et al., 2012). One problem with that approach is the time that it takes to discover and introgress new sources of resistance into elite varieties with conventional-breeding approaches. A detailed understanding of a plant species’ repertoire of R genes, coupled with the ability to stack different natural resistance traits into a single variety by using emerging genetic-engineering technologies, could make it more feasible to respond rapidly to evolving pathogen virulence in the field (Jones et al., 2014), given appropriate infrastructure.

A group at Wageningen University in the Netherlands developed potato varieties with resistance to various strains of the late blight fungus Phytophthora infestans, the pathogen that caused the Irish potato famine in the 19th century (Box 8-2). The approach involved introduction of a set of different R genes, using only DNA sequences from potato (Solanum tuberosum), so that the resulting plants would be classified as cisgenic (Haverkort et al., 2008, 2009; see section “Transgenics versus Cisgenics versus Intragenics”

in Chapter 7). Even though multiple R genes can now be engineered into a crop from wild potato, it is still critical to use a rigorous management strategy to ensure that resistance is not overcome through pathogen evolution. The approach taken by the Wageningen team uses cassettes of different groups of R genes that can be deployed depending on field and computer simulation studies of pathogen spread and evolution. To be effective, such a deployment strategy would require a correspondingly rapid and flexible regulatory-approval system and the infrastructure to replace quickly varieties that are in use.

Box 8-2 provides an example of public research for public good—the cisgenic potato. This example illustrates concern for sustainability by conducting resistance-management research up front; this has not always been the case in conventional breeding of pathogen resistance.

An alternative approach to the engineering of single or combined major qualitative R genes involves the engineering of novel defense mechanisms. Proof-of-concept studies aimed at the reintroduction of the American chestnut (Castanea dentata) into the United States constitute an interesting example of this strategy from both the scientific and public perspectives. The American chestnut, which at one time accounted for 25 percent of the trees in eastern forests of the United States, was almost destroyed by chestnut blight (Cryphonectria parasitica), a fungus introduced from Asia in the early 1900s (Paillet, 2002). Some 3–5 billion trees have been killed by the fungus, and the fungus survives—but is nonpathogenic—on oaks (Quercus spp.); thus, the pathogen is fixed in the ecosystem. The Chinese chestnut (Castanea mollissima) has quantitative resistance to blight, but this involves over 20 genes, making backcrossing in trees with such long generation times difficult (Bauman et al., 2014). Resistance to chestnut blight has been engineered by introduction of a single gene, which encodes the enzyme oxalate oxidase, from wheat (Triticum spp.) (Zhang et al., 2013). Oxalate is produced by the fungus; fungal strains producing the highest levels of oxalate tend to be the more virulent strains. The products of the oxidase are carbon dioxide and hydrogen peroxide, the latter acting as the antifungal agent. Hydrogen peroxide is regulated by the U.S. Environmental Protection Agency (EPA) as a plant-incorporated protectant. Non-native chestnut and American chestnut lines generated by mutation breeding are also being planted. According to the American Chestnut Research and Restoration Project, should the GE chestnut be submitted to and ultimately approved by EPA (and any other regulatory agencies if needed), the GE trees will be initially introduced in areas currently devoid of trees, in botanical gardens, and on private land (Powell, 2015). On the basis of an estimated rate of spread of 8–10 trees that generated 4,000 trees over a period of 125 years in Wisconsin, the originators of the project predict that 40–400 million new chestnut trees could be established within 100 years if there is strong public support for planting the trees. That would represent about 10 percent of the trees lost to the disease (Powell, 2015).

Crops that are important staples in the diets of many people in developing countries include vegetatively propagated crops—such as sweet potato (Ipomoea batatas), banana and plantain (Musa spp.), and yam (Dioscorea spp.)—that are particularly susceptible to microbial infections. Despite decades of research, sources of natural resistance to devastating diseases in banana, such as Panama disease (a wilt caused by Fusarium oxysporum f.sp. cubense) and black sigatoka (a leaf-spot disease caused by Mycosphaerella

fijiensis) remain elusive. Genetic-engineering approaches have potential to increase the speed with which new germplasm can be developed. For example, resistance to banana Xanthomonas wilt, a bacterial disease of increasing importance in east and central Africa, was enhanced by inserting the Hrap gene from sweet pepper (Capsicum annuum) (Tripathi et al., 2010). Of course, use of those genetic-engineering approaches will depend in part on public acceptance of the newly developed germplasm, in terms of both the technology and the actual varieties into which it is incorporated.

Another approach to improving resistance to pathogens is enhancing phytoalexin production; phytoalexins are antimicrobial compounds synthesized by plants after infection or stress (see Chapter 5, “Endogenous Toxins in Plants”). They make up a broad array of chemical classes that contribute to quantitative disease resistance, the best studied being of flavonoid or terpenoid origin. It is likely that the ability to produce some of those defensive chemicals may have been lost during domestication of major crops (Palmgren et al., 2015). Although proof of concept of plant protection through engineering of inducible defense compounds was obtained more than 20 years ago for a single-step pathway (Hain et al., 1993), the complexity of many phytoalexin pathways and the danger of evolved resistance to single antimicrobial compounds appear to have limited interest in the approach (Jeandet et al., 2013). Studies have begun to decipher the transcriptional control of induced small molecule defense pathways in plants (Mao et al., 2011a; Yamamura et al., 2015; Yogendra et al., 2015), allowing for coordinated expression and tissue-specific targeting of defense metabolites, even in the absence of knowledge of their complete biosynthetic pathways. Engineering transcriptional control mechanisms also facilitates the introduction of multiple antimicrobial defense molecules to reduce the chances of the pathogen evolving to overcome the introduced resistance. Regulatory scrutiny should ensure that the newly introduced or ectopically expressed metabolites do not adversely affect plant quality and food safety.

FINDING: A better understanding of both the nature and regulation of inducible defense pathways coupled with emerging genetic-engineering technologies could enable manipulation of complex metabolic pathways for enhancing plant resistance to disease.

Resistance to Viruses

The mechanism of virus resistance in papaya (Carica papaya)—transgenic expression of viral coat protein—was discussed in Chapter 3. Cassava (Manihot esculenta), a major subsistence crop in sub-Saharan Africa, is susceptible to two serious virus diseases, cassava brown streak disease (CBSD) and cassava mosaic disease (CMD). CBSD first became a

problem in Mozambique and is spreading to central and western Africa. Two research groups were working to develop virus-resistant cassava at the time the committee was writing its report. With a mix of federal, foundation, and corporate funding (from the Monsanto Fund, the U.S. Agency for International Development, the Bill and Melinda Gates Foundation, and the Howard G. Buffett Foundation), scientists at the Donald Danforth Plant Science Center, in collaboration with scientists in Uganda, Kenya, and Nigeria, were developing a high-throughput genetic transformation platform for farmer-preferred cassava varieties (Taylor et al., 2012) and were addressing cassava virus resistance through an RNA interference (RNAi) strategy. A similar approach was being taken by scientists at ETH Zurich (Nyaboga et al., 2013), introducing CBSD resistance into a Nigerian variety of cassava that is naturally resistant to CMD (Vanderschuren et al., 2012). Field trials suggest that those approaches are promising (Ogwok et al., 2012).

FINDING: Genetic engineering can be used to develop crop resistance to plant pathogens with potential to reduce losses for farmers in both developed and developing countries.

Resistance to Insects

The use of single-protein and multiple-protein toxins from Bt in cotton (Gossypium hirsutum) and maize is widespread, but other insecticidal proteins have not become widely available in most commercialized crops. There has been considerable discussion about the use of RNAi, and the first RNAi maize variety was deregulated by the U.S. Department of Agriculture (USDA) in 2015 (USDA–APHIS, 2015). The RNAi strategy derives from usurping a natural cellular defense mechanism that cuts RNA from infectious organisms that invade cells (Fjose et al., 2001). The effectiveness of RNAi constructs such as double-stranded RNA (dsRNA)—introduced through ingestion of plant tissues—was first shown for the western corn rootworm (Diabrotica virgifera virgifera) (Baum et al., 2007) and cotton bollworm (Helicoverpa armigera) (Mao et al., 2007). The strategy was designed to silence genes critical for growth and development of specific insect pests without silencing genes in other organisms. The gene encoding DvSnf7, a protein essential for intercellular movement of membrane vesicles in the insect midgut, has been the most studied insect target (Koči et al., 2014), and DvSnf7 RNAi has been coexpressed with the Cry3Bb1 protein; the DvSnf7 RNAi and the Cry3Bb1 protein act independently against Colorado potato beetle (Leptinotarsa decemlineata) (Levine et al., 2015).

Developments in RNAi technology to protect crops against insect pests have also used novel approaches that depend on the chemical ecology of

the host–herbivore interaction. In a project aimed at controlling cotton bollworm (Mao et al., 2007), dsRNA was targeted specifically against the insect cytochrome P450 enzyme CYP6AE14. Cytochrome P450s are ubiquitous enzymes that in animals and insects are involved primarily in the detoxification of foreign compounds. CYP6AE14 is involved in the detoxification of the cotton sesquiterpene gossypol, a potent antifeedant, and is induced by that compound. Lack of gossypol detoxification as a result of RNAi-targeted silencing of CYP6AE14 expression led to drastic reduction of bollworm larval growth on cotton plants that were engineered to express the CYP6AE14 dsRNA (Mao et al., 2011b).

A more complex example, which also illustrates the potential for genetic engineering to improve the understanding of complex ecological systems, concerns the expression in the tobacco species Nicotiana attenuata of dsRNA targeting the messenger RNA produced from the insect CYP6B46 gene; this gene encodes an enzyme that helps to direct a portion of the nicotine ingested by the insect Manduca sexta from the midgut to the hemolymph (analogous to the blood of vertebrates). Plants expressing the CYP6B46 dsRNA and nicotine-free GE plants were grown in their native habitat. Larvae of M. sexta feeding on either nicotine-free plants or plants expressing the dsRNA were attacked more frequently by nocturnal wolf spiders [Camptocosa parallela (Lycosidae)] than larvae feeding on N. attenuata that had not been engineered because the larvae feeding on nicotine-free plants exhaled no nicotine and those feeding on dsRNA plants exhaled less nicotine through their spiracles (Kumar et al., 2014). As might be expected, not all attempts to modify insect behavior via genetic engineering are successful; wheat engineered to express a gene from peppermint (Mentha × piperita) for production of the volatile insect pheromone E-beta-farnesene (which attracts predators of aphids) was protected from aphid damage in the laboratory but not in the field (Sample, 2015).

A technical limitation of expression of dsRNAs that are long enough to be gene-specific is the presence in plants of machinery to cut the long dsRNAs into short interfering RNAs. As described in Chapter 7, that limitation has been overcome by inserting the gene for the dsRNA into the plant’s chloroplast genome rather than the nuclear genome.³ With that alternative approach, the dsRNA accumulates in the chloroplast and is not exposed to the RNA-cutting enzymes in the cytoplasm. The strategy has proved effective for control of Colorado potato beetle by targeting the caterpillar’s essential actin gene (Whyard, 2015; Zhang et al., 2015). Another report described the expression of dsRNA against the gossypol-detoxifying CYP6AE14 and other targets in tobacco chloroplasts (Jin et al., 2015) and concluded that the strategy was effective for insect control.

__________________

³ The chloroplast genome is part of the plastid genome.

The committee heard from invited speakers (for example, Hansen, 2014) and received and read comments submitted from members of the public voicing concerns about the potential for off-target effects on human gene expression through consumption of dsRNA molecules targeted against insects. A number of studies (Dickinson et al., 2013; Snow et al., 2013; Witwer et al., 2013) have contradicted the results of a previous study (Zhang et al., 2012) that dsRNAs originating in plants can accumulate in human tissues following ingestion, and, in 2014, an EPA scientific advisory panel likewise concluded that there were minimal risks from use of dsRNAs because of the degradation of such molecules in the human gut and the ability to design sequences that lack off-target effects.⁴ As an additional precaution, the committee believed that targeting the dsRNA to a subcellular compartment in a nonedible portion of the plant could prevent exposure. For example, in the case of potato, the dsRNA could be expressed in the functional chloroplasts in the leaves but not in the amyloplasts in tubers. This additional targeting precaution would, of course, only be useful if the edible portion of the plant was not subject to attack by the insect that is targeted by the dsRNA.

The effectiveness of RNAi in controlling insects appears to be both species-specific and tissue-specific. For example, lepidopteran insects are generally poor targets for RNAi, and genes expressed in epidermal tissues are difficult to silence with RNAi approaches (Terenius et al., 2011).

Given that lepidoptera are naturally resistant to the effects of exogenous RNAi, it seems feasible that other insects could evolve resistance to RNAi. Whether other species of insects could downregulate or even lose their natural cellular RNAi machinery and thereby become resistant to this approach is not obvious. Acquired RNAi degradation pathways seem to be one potential mechanism for evolved resistance. Increased RNAi degradation could be a dominant or partially dominant trait based on the general finding that gain-of-function traits have this mode of inheritance (Gould, 1995), and the best resistance-management strategy could be to provide large refuges (see Chapter 4 section “Resistance Evolution and Resistance Management in Bt Crops”). The major concern is that one of these mechanisms could provide resistance against dsRNAs irrespective of sequence, so it would not be possible to overcome resistance by switching to new RNAi targets.

The committee heard concerns from members of the public at the workshop about the environmental effects of different pest-management practices and received and read submitted comments about off-target effects of RNAi on beneficial or endangered species. The concerns, which mirror

__________________

⁴ The meeting materials for the EPA scientific advisory panel on RNAi technology as a pesticide, including the meeting minutes, are available at http://www.epa.gov/sap/meetingmaterials-january-28-2014-scientific-advisory-panel. Accessed March 9, 2016.

concerns addressed in environmental impact studies conducted for any GE crop, were:

• Effects on other (non-pest) herbivores consuming the same crop.
• Effects on predators or parasites that consume the herbivore.
• Gene flow to other plant species (less of an issue for chloroplast transformation).

When the committee was conducting its review, there was insufficient evidence to support or refute any of those concerns beyond the conclusion by USDA’s Animal and Plant Health Inspection Service (APHIS) that a Monsanto transgenic event incorporating RNAi into maize to control corn rootworm did not pose an environmental risk (USDA–APHIS, 2015).

FINDING: Several genetic-engineering approaches involving use of RNAi or exploitation of chemical ecological phenomena are becoming available for the control of insect pests. More research is required to address the sustainability of, and off-target effects arising from, RNAi approaches and to learn how to adapt agroecological manipulation for crop protection.

Tolerance of Abiotic Stress

Abiotic stresses to crops include cold, heat, drought, and soils with high concentrations of salts or other chemicals that inhibit plant growth. Although an understanding of the biochemistry of stress is far from complete, commercialization of some GE stress-resistant plant varieties has begun.

Tolerance of Drought

Plants cannot grow under severe drought, and strategies to enhance drought tolerance generally focus on better survival during drought so that if moisture returns plants can resume growth and yield loss will not be too great. Monsanto offers DroughtGard™ maize, a GE maize that expresses a gene encoding cold-shock protein B (cspB) from Bacillus subtilis (Castiglioni et al., 2008). Under some drought conditions, cspB expression seems to result in higher yield than non-GE controls, but more on-farm tests are needed.⁵ Drought tolerance could be increasingly important in the face of climate change (Box 8-3).

__________________

⁵ Genuity DroughtGard Hybrids. Available at http://www.monsanto.com/products/pages/droughtgard-hybrids.aspx. Accessed November 20, 2015.

Drought tolerance in plants involves multiple responses, so engineering of a single response protein may be less effective than mechanisms that can alert a plant to drought and activate a suite of responses. That may be feasible, as exemplified by a strategy that targets a receptor that responds to the drought hormone abscisic acid (ABA). A variant of the ABA receptor PYRABACTIN RESISTANCE 1 was engineered to possess high sensitivity

to the agrochemical mandipropamid; spraying this compound was shown to enhance drought tolerance in GE plants (Park et al., 2015). A structural model of the ABA receptor suggests that it may be possible to engineer the molecule for activation by other chemicals. One attraction of this approach is that most soil-water loss in agricultural crops is from transpiration; attenuating transpiration in anticipation of drought will help to preserve soil moisture. A drawback of the approach in medium-size commercial farms is that drought often occurs in midseason when plants are large, and spray coverage must usually be accomplished by airborne equipment.

Tolerance of Cold

Another example of engineered abiotic stress tolerance is the development of cold-tolerant eucalyptus (Eucalyptus spp.) trees by the company ArborGen for cultivation in the United States. Cold tolerance was enhanced by GE expression of the C-Repeat Binding Factor 2 (CBF2) gene from Arabidopsis thaliana (Nehra and Pearson, 2011). CBF2 is a regulator of other genes that are involved in cold tolerance. Increasing CBF2 expression in some commercial eucalyptus hybrids results in greater cold tolerance, which permitted cultivation in regions of the southeastern United States where frost can occur. The potential for reducing growth by overexpression of CBF genes was mitigated by placing the CBF gene under the control of a cold-inducible promoter to limit its expression to conditions under which it is desirable. The GE eucalyptus lines also contain a gene-expression cassette that prevents pollen development and thereby restricts gene flow, although ArborGen claimed, in its petition to USDA for deregulation, that the existing biological limitations of eucalyptus species grown in the southeastern United States would themselves serve as an effective barrier to gene flow (Nehra and Pearson, 2011). ArborGen submitted the petition in 2011; when the committee conducted its review in 2015, the petition for deregulation was still pending.

As discussed in Chapter 7, most current examples of GE traits are based on one gene. As the fundamental understanding of the biochemical basis of why a specific stress inhibits plant growth or why some organisms are more resistant to a specific stress than others increases, so does the potential to use genetic engineering to enhance abiotic stress tolerance more effectively. It is increasingly likely that future enhancements of stress tolerance will require the introduction of several genes, possibly associated with metabolic engineering (see Box 8-1).

FINDING: Several approaches to engineering abiotic stress tolerance in plants are available, but, owing to the complexity of plant stress responses, more complex, temporally adjustable approaches than are

now used will probably be necessary, particularly in the face of unpredictable climate change.

Increasing Plant Yield and Efficiency of Production

Increasing plant yield in absolute terms, as opposed to overcoming the yield gap (as defined in Chapter 4), has always been a major goal of conventional plant breeding. In commodity crops—such as maize, soybean, and wheat—additional yield gain as a result of breeding is generally incremental, at an average of about 1–2 percent per year, although there have been historical exceptions when the introduction of such innovations as hybrid maize or semi-dwarf wheat and rice increased yields considerably (see Chapter 2 section “The Development of Genetic Engineering in Agriculture” and Chapter 4 section “Potential versus Actual Yield”). Examples of future genetic-engineering approaches to improve plant yield or increase the efficiency of production include improving nutrient-use efficiency, introducing nitrogen fixation, and re-engineering primary metabolism, particularly increasing the efficiency of photosynthesis.

Nutrient-Use Efficiency

Nutrient-use efficiency (NUE) refers to plant yield relative to the nutrients (for example, fertilizer components) used to achieve that yield. Many factors influence NUE, including the extent of the root system, how effectively root cells can take up nutrients, and how readily nutrients are transported from root to shoot. Modifying any of these factors in a manner that increases productivity is likely to require several genes that are specific to the factors. For example, using genetic engineering to alter the extent and pattern of a root system is likely to require changing the expression patterns of many genes that control development, whereas increasing the efficiency with which roots take up nutrients could involve altering membrane transporter-protein concentrations or types.

Concerns surrounding the depletion of the world’s mineable phosphate supply make phosphorus-use efficiency an especially important area for plant science research. Plant roots possess both high-affinity and low-affinity phosphate transporters (Poirier and Jung, 2015). Phosphorus sensing and uptake are under complex regulation (Scheible and Rojas-Triana, 2015) and are linked to root development, so simply overexpressing high-affinity phosphate transporters is not in itself a solution. An emerging understanding of the regulators of phosphate sensing, uptake, and response is leading to strategies that use genetic engineering to improve phosphorus-use efficiency in plants (Gamuyao et al., 2012; Wang et al., 2013). It has also been suggested that phosphorus-use efficiency could be improved by

altering the distribution of phosphorus in the plant (Veneklaas et al., 2012); this would involve the manipulation of multiple genes. Finally, engineering plants to secrete phosphatase enzymes that release phosphate from organic constituents of soil has also been shown to improve plant phosphate-use efficiency (Wang et al., 2009).

Nitrogen Fixation

Nitrogen is a key nutrient that often limits plant growth. “Fixing” nitrogen refers to converting nitrogen gas in the atmosphere into a form that can be incorporated into biological molecules. Plants are not able to fix nitrogen directly, but some bacteria are. However, legumes, such as soybean and common bean (Phaseolus vulgaris), have evolved the ability to associate in an intimate symbiotic relationship with some species of nitrogen-fixing bacteria and thus are able to produce a crop without added nitrogen fertilizer or organic amendments. In spite of the greater nitrogen-use efficiency of legumes (such as soybean) as compared to cereals (such as maize), in agricultural production it is often economically advantageous to grow legumes with added nitrogen fertilizer to ensure reliable yields because the additional cost of the fertilizer is not prohibitive.

Nitrogen-fertilizer production uses a substantial amount of natural gas—a fossil fuel—whereas biological nitrogen fixation does not, so engineering biological nitrogen fixation has an environmental benefit. There are two possible routes to enabling plants to fix sufficient nitrogen to support high yields. One is to introduce genes encoding all the proteins involved in nitrogen fixation. That is a complex undertaking because the nitrogen-fixation system is a bacterial metabolic pathway; many aspects of plant metabolism would need to be modified to support biological nitrogen fixation in plants, and this in turn would require introduction or alteration of many genes. Furthermore, a cellular compartment with low oxygen concentration would have to be engineered or adapted to support the nitrogen-fixing activity. Another route is to enhance the plant–bacterial nitrogen-fixation interaction in the legume species in which it naturally occurs so that more nitrogen fixation is supported or to engineer genetic networks needed for nitrogen-fixing symbiosis into plants that do not normally have this interaction.

Those approaches were discussed at a meeting of the Gates Foundation in 2011 (Beatty and Good, 2011), which led to the foundation’s support of basic research toward the goal of engineering nitrogen fixation in cereals. The Gates Foundation funded two approaches that address different aspects of engineering cereals to fix nitrogen. One focuses on engineering the genes that could allow cereals to form nodule structures similar to those found in legumes that house nitrogen-fixing bacteria (Rogers and Oldroyd,

2014). The other focuses on synthetic engineering of the nitrogen-fixing enzyme nitrogenase with the aim of inserting the necessary genes into plastids or mitochondria (Curatti and Rubio, 2014). The necessary approaches to engineering nitrogen fixation in cereals will involve synthetic biology to re-engineer pathways affecting both cell biology and plant metabolism (Rogers and Oldroyd, 2014). Even with the technologies on the horizon, engineering nitrogen fixation or a mechanism for a novel symbiosis is a daunting challenge, so it is not possible at present to judge whether this will be successful.

The commitment to face that challenge is complemented by the Gates Foundation’s projects aimed at tailoring and adapting existing legume technologies for smallholder farmers in Africa through its N2Africa program.⁶

Increasing Efficiency of Photosynthesis

Another example of altering metabolism to increase productivity is improving the efficiency of photosynthesis, which in turn could increase the rate of plant growth and probably yield (Bräutigam et al., 2014; Weber, 2014). One limitation to photosynthetic efficiency is that the enzyme (commonly referred to as RuBisCo) that initiates the process of converting carbon dioxide to sugar (commonly referred to as carbon fixation) can also react with oxygen in a side reaction that wastes energy. The magnitude of the side reaction with oxygen is a function of the relative amounts of carbon dioxide and oxygen that RuBisCo encounters. Some plants have evolved a way to mitigate the side reaction: RuBisCo and carbon fixation are restricted to cells in which carbon dioxide is accumulated through shuttling by four-carbon compounds such as malate (this type of metabolism is referred to as C4). Maize has C4 metabolism. Introducing C4 metabolism into crops, such as rice, that do not have it could increase yields. Re-engineering rice carbon-fixation metabolism would, however, require the manipulation of many genes, including ones that control leaf development and leaf differentiation (to create specialized cells for carbon fixation) and ones that encode the enzymes of C4 metabolism. The International Rice Research Institute has embarked on a long-term project to develop C4 rice.⁷

Another approach to mitigating the side reaction of RuBisCo with oxygen is to change the enzyme in such a way that it can no longer interact with oxygen but continues to carry out the carbon-fixation reaction. Although in principle that is a simple “single-gene” solution of the problem of the side reaction, in practice it has proved quite difficult to accomplish. Perhaps the difficulty is not surprising inasmuch as throughout the course of evolution

__________________

⁶ N2Africa: Putting nitrogen fixation to work for smallholder farmers in Africa. Available at http://www.n2africa.org/. Accessed December 12, 2015.

⁷ The C4 Rice Project. Available at http://c4rice.irri.org. Accessed December 12, 2015.

there has been an extensive opportunity for the side reaction to be eliminated by the random mutations that continually arise in populations. One would expect if such a mutation did arise that it would provide a selective advantage, but this has not happened. Rather, many organisms have independently evolved a variety of carbon dioxide-concentrating mechanisms that increase photosynthetic efficiency, one of which was discussed above. Nevertheless, if scientists discover a way to make a “better” RuBisCo, it would be a substantial advance toward increasing plant productivity. RuBisCo from the cyanobacterium Synechococcus elongatus fixes carbon dioxide more rapidly than plant RuBisCo and has been expressed in tobacco chloroplasts in place of the native tobacco enzyme as a first step in engineering improved carbon-fixation efficiency (Lin et al., 2014); the plants were photosynthetically competent, but further engineering to raise the local carbon-dioxide concentration in the vicinity of the enzyme and suppress the oxygenation reaction is still needed.

FINDING: Applications of genetic engineering that target basic plant processes, such as photosynthesis and nitrogen fixation, have the potential to result in greater yield gains or increased efficiency but will probably require complex genetic changes and therefore involve long-term projects.

Increased Forage Quality

Lignocellulosic biomass (in the form of plant stems and leaves) is the major feedstock for forage-based milk and meat production. Many of the cattle in the United States and elsewhere are fed alfalfa, a high-protein forage. Genetic engineering can introduce traits into alfalfa and other sources of feed that will improve digestibility and animal nutrition and reduce health risks for ruminant livestock associated with methane production.

Digestibility

Lignification of cell walls in forage adversely affect digestibility by ruminant animals because it restricts access of the digestive system’s microorganisms and enzymes to cellulose and hemicellulose polymers that together make up the bulk components of both primary and secondary cell walls in plants (Ding et al., 2012). Those polymers are made of hexose (six-carbon) and pentose (five-carbon) sugar units, the major carbon sources for animal nutrition after degradation of the polymers in the rumen. The inherent properties of the lignocellulosic materials, collectively called recalcitrance, have proven difficult to alter with conventional-breeding methods (Dixon et al., 2014). Although the phenomenon of cell-wall recalcitrance is

complex and probably involves multiple chemical entities and mechanisms, the presence of a lignin in secondary cell walls is a major contributor (Chen and Dixon, 2007; Ding et al., 2012).

Lignin is a complex aromatic polymer derived primarily from hydroxycinnamyl alcohols (monolignols) that are linked into the lignin polymer by a seemingly random free-radical polymerization process in the plant cell wall (Boudet et al., 1995). Lignin is deposited in secondary cell walls of plants after they have stopped expanding and imparts structural integrity to the cell wall, strength to stems, and hydrophobicity to vascular elements for water transport.

The quality of alfalfa hay is assessed on the basis of protein content and fiber digestibility; however, as an alfalfa-hay crop increases in biomass, the quality of the forage decreases, owing in large part to lignin deposition. Reduced-lignin (RL) alfalfa, developed by Forage Genetics International in partnership with Monsanto, was deregulated in the United States in late 2014; it was engineered to contain reduced amounts of lignin in secondary cell walls through partial silencing of the gene encoding an enzyme involved in the synthesis of the monolignol building blocks of lignin. RL forages should allow the farmer to balance optimal biomass with quality, and this should translate into an increased window during which the crop can be harvested.⁸ Increasing the digestibility of alfalfa should decrease the amount of alfalfa that needs to be grown per kilogram of meat or milk produced, thereby decreasing the amount of land needed per kilogram of meat or milk. It should also reduce the amount of manure generated per kilogram of forage.

The reduction in lignin content (by around 10 percent) places RL alfalfa at the limits of, or even outside, the normal variation in lignin content for this species; this is because alfalfa has been highly improved, and there is not a wide range of natural variation in biomass digestibility among varieties. Alfalfa is indigenous to southwestern Asia, having probably originated in Iran, so gene flow from RL alfalfa in the United States and most of the world would result in transfer of the low-lignin trait only to other commercial alfalfa crops or perhaps to feral populations around commercial plantings. The major risk associated with gene flow from RL alfalfa is probably an economic one: the contamination of non-GE alfalfa, including organic alfalfa. That risk is therefore similar to the one arising from herbicide-resistant (HR) alfalfa and has to be managed to take into account the pollination method of the crop.

The second potential risk associated with RL alfalfa—one that will be relevant for many products of plant metabolic engineering—comes from

__________________

⁸ At the time this report was written, Forage Genetics International was preparing to market RL alfalfa under the brand name of HarvXtra™. RL varieties of alfalfa with and without glyphosate resistance were in development.

the partial disruption of a natural metabolic pathway in the plant, with a potential for flow of intermediates into other pathways and different branches of the targeted pathway (in this case lignin). The lignin in RL alfalfa has an increased proportion of sinapyl alcohol monolignol–derived residues. When that was first observed, it was unexpected and viewed as an unintended effect until it was discovered that alfalfa has a route by which the downregulated enzyme could be bypassed in the synthesis of sinapyl lignin (Zhou et al., 2010). That highlights the importance of basic science to provide understanding to help in risk assessment.

Alfalfa contains a number of natural products with known bioactivity, some of which are biosynthetically related to lignin. As discussed in Chapter 5 (see section “Endogenous Toxins in Plants”), isoflavones are antimicrobial compounds with perceived benefits for human health but might also have adverse effects because of their estrogenic properties. Targeted analysis of isoflavones in RL alfalfa did not detect significant changes compared with those in non-GE commercial lines. The triterpene saponins are molecules with antiherbivore properties; alfalfa contains a wide array of these compounds, which have different profiles in foliage and roots. Triterpene saponins at high concentrations can exhibit hemolytic activity in monogastric animals, such as horses. Triterpenes were unaffected in RL alfalfa relative to their concentrations and compositions in non-GE commercial lines. Other members of the genus Medicago, to which alfalfa belongs, exhibit wide natural variation in triterpene saponin content. It therefore seems logical that saponin concentrations should also be evaluated during the breeding of non-GE alfalfa and other Medicago species, such as the annual Medicago truncatula, which is used as a forage legume in Australia. Finally, alfalfa sprouts contain canavanine, a neurotoxic non-protein amino acid. Canavanine concentrations were somewhat lower in RL alfalfa than in non-GE commercial lines.

Because RL alfalfa results in more biomass than alfalfa with normal lignin levels before quality reduction forces harvest, a new factor is introduced into the management of the crop. The quality of single-trait HR alfalfa decreases with increasing maturity in the same way as in nonreduced lignin alfalfa, and HR alfalfa is therefore harvested earlier than RL alfalfa may be harvested. That raises the question of whether RL alfalfa will be more likely to be left to flower in the field and thereby to increase the risk of pollen flow to non-GE alfalfa. To address that possibility, Monsanto has applied restrictions on the growing of RL alfalfa to “include managing hay to prevent seed production, harvesting at or before 10-percent bloom in areas where seed is produced, and prohibitions on use in wildlife feed plots. Growers who raise alfalfa for seed would be required to follow Forage Genetics International Best Practices” (USDA–APHIS, 2011).

Nitrogen Protection and Mitigation of Methane Production

One of the major disadvantages of high-protein forages is the potential for causing pasture bloat arising from excess methane production in the rumen. Methane is an important greenhouse gas. The presence of natural products called condensed tannins (CTs) can protect ruminant animals from potentially lethal bloat. CTs are polymers of flavonoid units (products of plant secondary metabolism) that bind to proteins and therefore reduce the rate of fermentation in the rumen and reduce methane production. Their protein-binding activity also allows more intact protein to leave the rumen, and this results in improved nitrogen nutrition. Alfalfa foliage lacks CTs, and their introduction has been an important but not yet realized goal of alfalfa breeding. In addition to alfalfa, introducing or increasing CTs into other forages—such as clover (Trifolium spp.), ryegrass (Lolium spp.), and grazed wheat—and seed crops used in animal feed, such as cotton and soybean, is a goal. Higher CT concentrations can also prevent long-term spoilage of silage.

Although not all the biochemical reactions leading to CTs in plants are fully understood, several studies have attempted to increase CTs in plant foliage through genetic engineering. It is a complex metabolic-engineering problem because of the need to express multiple genes in a tissue in which they are not normally expressed. One approach to engineering of changes in expression of multiple genes is to use genes that encode transcription factors (TFs). TFs are proteins that bind to the regulatory regions of genes and modify the magnitude of expression of the genes; they may be positive or negative regulators. Often, a single TF will regulate multiple genes in a biochemical pathway and thus obviate separate genetic manipulation of the individual steps in the pathway. TFs that control the genes for the CT pathway have been discovered, and it has been shown that one such TF, from a species of clover that produces high concentrations of CTs, can turn on accumulation of CTs when expressed in alfalfa leaves (Hancock et al., 2012). It therefore seems likely that the CT traits will be engineered into forage species through modification or synthesis of new TFs that activate expression of genes responsible for synthesis of CTs. That raises several scientific and regulatory issues.

TFs themselves are usually expressed at very low levels in plant cells, and a substantial body of evidence suggests that they may have different effects, depending on the magnitude of their expression, with a potential for off-target effects when expressed at unnaturally high levels (Broun, 2004). Humans consume CTs and their precursors in large quantities in foods and beverages, such as chocolate and red wine, in which they are perceived to provide health benefits (see discussion on high-anthocyanin tomatoes in “Flavonoid Antioxidants” section below). In contrast, high concentrations of CTs can be astringent and therefore act as antifeedants. Assuming that

genetic engineering can lead to the moderate concentrations of CTs required to improve animal health and performance, the major regulatory issue is unlikely to be the CT product itself and more likely ensuring that there are no problematic off-target effects resulting from TF expression.

FINDING: Increased understanding of basic plant sciences will enable better prediction of off-target changes that could be caused by genetic engineering.

FINDING: GE approaches to forage-quality improvement have potential to yield environmental benefits associated with reductions in greenhouse gases and manure.

Improved Biofuel Feedstocks

The last decade has seen intensive research on the development of “second-generation” biofuels derived from lignocellulosic materials, mainly trees, such as poplar (Populus spp.), and grasses, such as switchgrass (Panicum virgatum) and Miscanthus (Himmel et al., 2007; Poovaiah et al., 2014). Use of both GE and non-GE approaches will be necessary to deliver biofuel crops that are sustainable agronomically and economically. In particular, reducing the recalcitrance of plant biomass to enzymatic deconstruction to develop crops that can be directly deconstructed and fermented to liquid biofuels (consolidated bioprocessing), and incorporating coproducts to the residual biomass to add value, have been identified as major goals (Mielenz, 2006; Ragauskas et al., 2006, 2014).

The efficient release of fermentable sugars from plant cell walls is the first critical step in the conversion of lignocellulosic biomass to liquid biofuels, such as ethanol or iso-butanol, or to other industrial chemicals produced by fermentation. The recalcitrance of lignocellulose to deconstruction to its component sugars is an important factor that limits the economics of the cellulosic-ethanol industry (Himmel et al., 2007). Since 2010, many proof-of-concept studies have indicated the feasibility of reducing the recalcitrance of biomass, and thereby increasing ethanol yields, through the engineering of cell-wall polymers in both bioenergy crops and model species such as Arabidopsis thaliana. Most of the studies have targeted lignin (Li et al., 2014; Liu et al., 2014) on the basis of the same mechanistic principles that underlie the improvement of forage digestibility discussed above. However, it is becoming clear that recalcitrance is a complex trait that also involves cellulose, hemicellulose, and pectic cell-wall polymers and cell-wall cross-linking by small molecules. It is therefore likely that future improved bioenergy crops will have genetic changes in multiple pathways that are introduced through the techniques of genome

or RNA editing, transformation with multigene cassettes, or combining of transgenes through conventional-breeding approaches (Kalluri et al., 2014). Furthermore, to avoid potentially deleterious effects of strong lignin reduction in critical tissues, such as water-conducting vessels, the genetic changes will be targeted to specific tissue types by using either tissue-specific gene promoters to drive downregulation of target genes or more complex strategies, such as pathway rewiring by modifying regulatory sequences (Yang et al., 2013). Reduced recalcitrance may also be combined with additional engineered traits, such as increased biomass density to reduce the volume of the crop that needs transportation (Wang et al., 2010). Some of those approaches will require engineering of TFs, which will raise issues similar to those discussed above in connection with metabolic engineering of CTs.

Potential future lignocellulosic bioenergy crops include perennial grasses (such as switchgrass and Miscanthus), rapid-rotation coppiced trees (such as willow, Salix spp.), other trees (such as poplar and eucalyptus), and drought-tolerant succulents (such as Agave spp.). Some of these species are only poorly domesticated, so there might be considerable natural variation in multiple traits, including cell-wall composition and recalcitrance, and this is now being experimentally confirmed. An analysis of over 1,000 natural variants of poplar (Populus trichocarpa) growing in the Pacific Northwest of the United States revealed an extreme range of lignin content (Studer et al., 2011), with low lignin equating to reduced recalcitrance. Natural variation in cell-wall composition is also seen in switchgrass (Vogel et al., 2011). Those observations suggest that reduced recalcitrance can be introduced into some bioenergy crops through conventional breeding. In trees with long generation times, genetic engineering could reduce the time to a commercial product compared with conventional breeding.

Risk assessment of reduced-recalcitrance trees and grasses would probably be similar to that of RL alfalfa discussed above, with the extra issues of preventing and monitoring gene flow to native populations for indigenous species, such as switchgrass and poplar, and the long time scale for assessment of true environmental effects for many trees. Gene flow from outcrossing perennial grasses could be prevented by engineering pollen ablation or sterility, traits that may themselves present additional regulatory issues. An inquiry to USDA–APHIS regarding GE switchgrass from Ceres Inc. sought exemption from regulation, claiming that the cassettes used for transformation were introduced by particle bombardment and therefore contained no plant-pest–derived DNA; the exemption was granted (Camacho et al., 2014). The justification was that the plant event did not meet the requirements of being or containing components of a plant pest (see Chapter 9 section “Regulatory Implications of Emerging Genetic-Engineering Technologies”), rather than evaluation of potential

risks associated with its reduced-lignin trait’s being conferred on native populations of switchgrass.

Another approach to development of bioenergy crops is to produce the fuel directly in the crop. The best-known example is biodiesel, and a number of oilseed crops have been used as sources of biodiesel. Examples are soybean, canola (Brassica napus), Camelina sativa (a relative of canola), and Jatropha curcas. The latter has been grown over wide areas in Asia and sub-Saharan Africa, although the economic viability of the crop, on the basis of seed yields, is a matter of debate (Jingura and Kamusoko, 2014). Genetic improvement of biodiesel properties will involve both oil yield and oil quality. There is a considerable body of work on genetic engineering of plant oil quality, much of it associated with improving nutritional quality by reducing concentrations of saturated and trans-fatty acids (see below). Much of the work on GE improvement of oil quality and yield for biodiesel has been in green algae, yeast, and bacteria, rather than in plants (Hegde et al., 2015). However, Camelina has emerged as a potential new oilseed crop, on which genetic knowledge from the related model plant Arabidopsis thaliana can be translated for production of oils for food, industrial, or biofuel applications (Vollman and Eynck, 2015); in the latter case, high oleic oils are preferred.

Some plant species naturally produce hydrocarbons at varied concentrations, either constitutively or after insect or pathogen attack. A good example is the production of terpenes in trees, such as pine (Pinus spp.) and eucalyptus. Knowledge of the genes involved in terpene synthesis in plants has facilitated synthetic-biology approaches to the engineering of microorganisms to produce sesquiterpenes, such as farnesene, that can be used directly as fuels (Wang et al., 2011), and genetic engineering of increased concentrations of such compounds directly in plants is possible (Beale et al., 2006). A potential problem with that latter approach is the volatile nature of the terpenes, which either are released in small amounts as signal molecules in plant–insect interactions (Beale et al., 2006) or are stored naturally in larger amounts in specialized glands in plants that naturally accumulate high concentrations of the compounds (King et al., 2004). In the absence of such glands in such species as Camelina sativa, it may be necessary to convert the terpenes to less volatile forms, through conjugation to sugars or other molecules, to facilitate accumulation.

FINDING: Recent advances in understanding and overcoming biomass recalcitrance make it more likely that “second-generation” lignocellulosic biofuels will be developed commercially through either conventional breeding or genetic engineering.

Introduced or Enhanced Nutritional Traits

Humans require at least 50 nutrients, including vitamins, essential amino acids, essential fatty acids, minerals, and trace elements. Particularly in developing countries, those requirements are not being fully met by the prevailing agriculture; staple crops are poor sources of some critical nutrients (Welch and Graham, 2005; also see Chapter 5 section “Improved Micronutrient Content”).

Plant breeding has generally focused selection on crop yield and, in the case of fruits and vegetables, on taste and processing properties. During the roughly 10,000 years that humans have been farming, the few common crops that are mostly used for human consumption have lost many of their phytonutrients (Robinson, 2013). A detailed examination of nutrient compositional changes of 43 garden crops from 1950 to 1999 suggested a loss of six nutrients (protein, calcium, phosphorus, iron, riboflavin, and ascorbic acid) (Davis et al., 2004)—with the caveat that one needs to factor in differences in sampling methods, varieties chosen for analysis, analytical methods, and growing environment. An example of a comparison of cultivated versus wild red raspberries collected from one of the germplasm centers in Turkey showed that some wild accessions had higher antioxidant activity and phytonutrient contents (Çekiç and Özgen, 2010). When the genes for high phytonutrients are no longer found in germplasm collections, genetic engineering could be the only practical approach for restoring phytonutrient concentrations.

Engineering of Vitamins and Polyunsaturated Fatty Acids

Among the best-publicized examples of GE improvements in nutritional content are Golden Rice engineered to contain high concentrations of provitamin A beta-carotene and oilseed engineered to contain high concentrations of polyunsaturated fatty acids (PUFAs). Conventional breeding and advances in MAS can lead to improvements in nutritional composition of plants that already contain the nutrients in question and whose germplasm pool varies widely in the trait (Graham et al., 1999; Welch et al., 2000; Welch and Graham, 2005; White and Broadley, 2005; Goldman, 2014). In the cases of provitamin A and PUFAs, genetic engineering has been used when the natural germplasm does not contain the genes necessary to deliver the trait. The pros and cons of Golden Rice have been discussed extensively, and the arguments will not be recapitulated here (see discussion in Chapter 5 section “Improved Micronutrient Content”).

Metabolic engineering of health-promoting PUFAs, with profiles similar to those found in fish oils, has been more complex than the engineering of provitamin A, and in some cases requires the introduction of multiple genes

from different species (Wu et al., 2005; Truska et al., 2009; Ruiz-López et al., 2012). Eicosapentaenoic acid and docosahexaenoic acid have been engineered to accumulate up to 20 percent of the seed oils of GE Camelina sativa with no effects on growth phenotype in the greenhouse (Petrie et al., 2012; Ruiz-Lopez et al., 2014) or the field (Usher et al., 2015). Oil from the GE plants was shown to be effective in replacing fish oils in salmon-feeding trials (Betancor et al., 2015).

Engineering of Mineral Micronutrient Concentrations

Major essential micronutrients for humans that can be deficient in staple diets include iron, zinc, copper, selenium, and iodine. Simple agricultural measures, including soil fertilization and amendments or crop rotations, may have little effectiveness in improving micronutrient concentrations in grains (Rengel et al., 1999); delivering micronutrients to vulnerable human populations through traditional methods, such as supplementation or food fortification, can likewise be ineffective because of complex political and socioeconomic factors (White and Broadley, 2005). Although new technologies have the potential to deliver needed micronutrients, they may be limited by the same political and socioeconomic factors that have made it difficult to deliver improved nutrition through non-GE food.

A large multinational research endeavor to genetically engineer increased iron and zinc concentrations in cassava was going on when the committee was writing its report (Sayre et al., 2011; Sayre, 2014). Those two micronutrients were identified previously as commonly deficient in most regions of the developing world, and such deficiencies could be addressed through genetic engineering (Zimmerman and Hurrell, 2002). Iron was increased in the cassava root by using the iron-specific assimilatory protein from Chlamydomonas reinhardtii, which enhanced iron uptake from the soil. Iron increased from about 10 to 40 ppm. The strategy used for zinc was to introduce additional plasma-membrane zinc transporters for higher accumulation of zinc in the tuber. Introduction of the Arabidopsis ZIP⁹ zinc transporter gene caused increase in root zinc concentrations of 2 to 10 fold, but decreased zinc in the plant leaves adversely affected yield; thus, this approach needs refinement.

Three independent genetic-engineering approaches that target iron transport to and storage in the seed endosperm were combined to produce rice with 440 percent more iron without reducing yield (Masuda et al., 2012).

__________________

⁹ ZRT-IRT-like proteins (ZIPs) are a family of plant membrane transporters, originally assigned to zinc and iron transporters.

Although both GE micronutrient-enhanced cassava and rice are precommercial, there were efforts toward commercialization when this report was being written.

Reduced-Phytate Maize

Phytate is a phosphorus storage sink for essential plant metabolic activities, but its composition of inositol plus up to six phosphates creates a dense negatively charged molecule that binds cationic minerals, particularly iron and zinc, rendering them inaccessible to the body during digestion and absorption. Conventional-breeding and genetic-engineering approaches have been used to reduce phytate content to increase bioaccessibility of those essential minerals. Genome editing with zinc finger nucleases was successfully used to reduce phytate concentrations in maize (Shukla et al., 2009). Edited plants were fertile, and the low phytate content was observed to transfer to next-generation plants. However, the technology had yet to be commercialized as of 2015.

Increasing Availability of Essential Amino Acids

For humans and most mammals, lysine is the limiting essential amino acid in most cereal-based diets and is particularly deficient in maize grain. The latter is because the maize storage protein, zein, is very low in lysine. Efforts to improve protein quality began in the 1950s, when reports of a large protein deficit in populations in the developing world spurred government-led research to find improved protein-quality cereal genotypes. High-lysine maize was found as a naturally occurring mutant and was termed opaque2 maize; lysine and tryptophan (the second limiting amino acid in maize) contents were about double those of normal maize (Mertz et al., 1964). Soft kernels and otherwise poor grain quality made for poor adoption, but conversion to a hard endosperm type by the International Maize and Wheat Improvement Center (CIMMYT) in Mexico increased its potential, and this variety is now used in some parts of Africa and other areas (Vasal, 2000). Genetic-engineering techniques have also been used to increase lysine content of soybean, rapeseed, and maize with a number of approaches (Galili and Amir, 2013). Expression of a bacterial feedback-insensitive dihydrodipicolinate synthase of lysine synthesis, which increases free lysine, was used to make a high-lysine maize (Lucas et al., 2007), but it was never released.

Flavonoid Antioxidants

Anthocyanins are the well-known red and purple pigments found throughout much of the plant kingdom as major natural products in fruits and flowers. They have antioxidant activity and have been linked to a number of potential health benefits (Yousuf et al., 2015). Anthocyanins consist of a flavonoid molecule (an anthocyanidin) conjugated with one or more sugar molecules and sometimes other chemical groups. The anthocyanidins are intermediates in the formation of the condensed tannins discussed above. Because plants naturally contain anthocyanins, their concentrations can be increased in fruits and vegetative tissues through conventional breeding if sufficient natural variation is present. However, attempts are being made to increase the concentrations of anthocyanins beyond the ranges found naturally, to introduce anthocyanins into fruits that generally lack them, and to engineer related flavonoids, such a flavonols, in fruit tissues through genetic-engineering technologies or by isolation of mutants (Dixon et al., 2012). Basic proof-of-concept studies in this field have been conducted on various vegetable and fruit species, including apple (Malus spp.), grape (Vitis spp.), tomato (Solanum lycopersicum), and cauliflower (Brassica oleracea). The related condensed tannins have also been the subjects of similar studies in fruits and vegetables, with the purpose either of reducing their concentrations to decrease astringency in high-tannin fruits, such as persimmon (Diospyros spp.) (Akagi et al., 2009), or of increasing their concentrations to provide health benefits.

Although tomato plants can produce anthocyanin pigments in their vegetative tissues, the main pigments giving the red coloration to tomato peel and flesh are carotenoids. The peel contains low concentrations of flavonoids in the form of chalcone glycosides (conjugated precursors of flavonols and anthocyanins) and flavonols, but the fruit is deficient in potentially beneficial anthocyanin antioxidants. Although some heritage tomato varieties have a purplish skin, it was probably not due to the presence of anthocyanins. Conventional-breeding approaches have generated tomatoes with anthocyanins in the skin but not in the flesh.¹⁰ Anthocyanin-rich “purple tomatoes,” with anthocyanin throughout the flesh, are being developed through genetic engineering. The strategy has involved the expression of two TFs that naturally control anthocyanin formation in the flowers of the snapdragon (Antirrhinum majus) (Butelli et al., 2008; Gonzali et al., 2009). In one small study, the high-anthocyanin tomatoes were reported to delay tumor development when fed to cancer-susceptible mice (Butelli et al., 2008), but effects on human health require more detailed studies (Tsuda,

__________________

¹⁰ The Purple Tomato FAQ. Available at http://horticulture.oregonstate.edu/purple_tomato_faq. Accessed November 20, 2015.

2012). Furthermore, the shelf-life of the tomatoes was doubled as a result of a slowing of the over-ripening process, and this led to a reduction in infection by the fungus Botrytis cinerea. When the committee was writing its report, high-anthocyanin tomatoes were in greenhouse trials in Canada. Similar approaches with expression of maize TFs (Bovy et al., 2002) or overexpression of the enzyme chalcone isomerase (Muir et al., 2001) have been used to engineer tomato fruits with high concentrations of flavonols, similar to ones in onions (Allium cepa), but not accompanied by accumulation of anthocyanins. High-flavonol GE tomatoes were as acceptable in consumer taste studies as non-GE varieties (Lim et al., 2014).

Because the enhancement of flavonoid concentrations in the GE tomatoes does not occur at the expense of carotenoids, which are also beneficial to health (Johnson, 2002), high-flavonoid/anthocyanin tomatoes may indeed provide substantial health benefits to consumers. Studies in animal models suggest that high concentrations of anthocyanins can be consumed without adverse health effects (Pojer et al., 2013). Potential risks include the possibility of off-target effects resulting from TF overexpression; transcriptome analysis has shown that multiple pathways, including defense responses, are activated because of the expression of the two TFs in high-anthocyanin tomatoes (Povero et al., 2011). Natural variation in anthocyanin concentrations in several fruits and vegetables has been shown to be controlled by differences in expression or naturally occurring mutations in similar TFs, so conventional breeding could also be used to increase anthocyanin concentrations, although probably not in tomato flesh.

Finally, high-anthocyanin grapes have been granted nonregulated status by USDA–APHIS, on the basis of their being cisgenic and generated with gene gun-mediated transformation (Camacho et al., 2014).

FINDING: Genetic engineering can enhance the ability to increase the nutritional quality and decrease antinutrients of crop plants.

Food-Safety Traits

Two examples of GE traits that led to improved food safety are low acrylamide in potato and an association of Bt in maize with lower fumonisin content (see Chapter 5). The low-acrylamide potato was developed by Simplot Plant Sciences through silencing of the asparagine synthetase-1 gene (Waltz, 2015). Acrylamide is one of the byproducts of the Maillard reaction between asparagine and reducing sugars that can occur in high-heat processing and is associated with cancer in rodents. The company has shown a 50–70 percent reduction in acrylamide in the GE potatoes compared with non-GE potatoes (see Chapter 5 section “Genetically Engineered Crops with Lower Levels of Toxins” for health implications).

Postharvest Traits

Postharvest traits that could be improved with genetic engineering span disease resistance of harvested fruits and grains, resistance to bruising, extended shelf-life, and standardization of taste and other qualities. For forage crops, silage stability could be improved. New GE traits for disease resistance are discussed above and are applicable to fruits and grains. Stability of proteins during ensiling of forages may be improved through engineering of tannins, as discussed above. Damage during bruising can be caused by both oxidative browning and cell lysis as a result of loss of integrity of internal cell membranes. As described in Chapter 3, oxidative browning can be reduced in fruits and other plant parts by downregulation of the polyphenol oxidase gene, as exemplified by the GE nonbrowning apple and potato; this could improve storage quality.

FUTURE GENETICALLY ENGINEERED CROPS, SUSTAINABILITY, AND FEEDING THE WORLD

Agriculture is the largest land use on the planet, accounting for 40 percent of Earth’s land area (Foley et al., 2005). Worldwide demand for food is expected to double by 2050 (Tilman et al., 2011), and there is a risk that more natural habitat will be converted to agriculture to meet this demand with consequently huge losses in biodiversity.

Sustainable Intensification

It is well documented that most new agricultural land in recent decades has been created by deforestation in highly diverse tropics, and this makes the link between agriculture and conservation clear (Gibbs et al., 2010). From 1998 to 2008, cropland expanded by 48,000 km²/year in tropical countries (Phalan et al., 2013). For that reason, many conservationists have suggested that sustainable intensification, by which they mean increasing yields per hectare without adverse environmental effects, is important for biodiversity conservation (Garnett et al., 2013).

The key question for the purposes of this report is the extent to which current and future genetic-engineering technologies can promote sustainable intensification (Godfray et al., 2010). Among the traits that have the potential to enhance sustainable intensification are improved forage quality for livestock (less land required and reduced methane production), disease and insect resistance, nutritional fortification, drought tolerance, salinity tolerance, and increased nitrogen-use efficiency (Godfray et al., 2010). Some of those traits have already been improved through conventional breeding and genetic engineering, but as discussed in the sections above,

emerging genetic-engineering technologies coupled to increased understanding of plant biochemistry are expected to enable breeders to address traits that cannot be further improved by existing technologies.

Because 30–40 percent of the world’s food is never consumed or is lost to waste, modifications that enhance storage properties could also reduce agriculture’s demand for converting natural lands (Godfray et al., 2010). Whereas GE crops may play a role in decreasing yield gaps in ways that support sustainable intensification, it is important to point out that closing yield gaps and boosting agricultural production in many parts of the world will require improvement in agroecologically sustainable farming, irrigation, and fertilizer use (Mueller et al., 2012). In that respect, the International Assessment of Agricultural Knowledge, Science, and Technology for Development emphasized the need to expand agroecological knowledge and practices to meet growing food demands (IAASTD, 2009; Vanloqueren and Baret, 2009). It is also important that investment in GE crops is not made at the expense of developing agroecological approaches to intensification (Vanloqueren and Baret, 2009).

FINDING: It is important to develop policy approaches that enable research on GE crops to assist in achieving sustainable intensification without diminishing resources available for the exploitation of proven existing technologies.

Sustainability and the Food System

A 2015 National Research Council report, A Framework for Assessing Effects of the Food System, asserts that determining how any given change will affect the food system is complex but that most studies seeking to measure how changes in agriculture affect the food system tend to have a narrow focus. The report contained a number of principles that should guide research on the food system. One is that it is important to consider four domains to predict effects on the food system: health, environment, social, and economic (NRC, 2015). That principle is meant to address the tendency to frame concerns about new technologies primarily in terms of health and environmental effects. It has been the goal of the present committee to incorporate these four dimensions into the analysis of experiences and prospects of GE crops.

Questions about the role that future GE crops can play in promoting a more sustainable agriculture and food system need to be addressed within those four domains (Hubbell and Welsh, 1998; Ervin et al., 2011; Macnaghten and Carro-Ripalda, 2015). There is a general recognition that sustainability should be understood to have health, economic, and social dimensions (especially in relation to equity of distribution of risks and

benefits) in addition to environmental dimensions (Ervin et al., 2011). For example, Pfeffer (1992) traced the rise of the sustainability movement itself to farmer resistance throughout the 20th century to “appropriationism,” the replacing of an on-farm activity with an industrial input. Furthermore, after reflecting on perspectives among regulatory agencies, scientists, and members of the public on how GE crops might contribute to sustainable agriculture and food in Mexico, Brazil, and India, Macnaghten and Carro-Ripalda (2015) advocated for moving beyond technical arguments about health and environmental risks and what they called stale debates grounded in polarizing positions. They argued instead that the key question is what institutional frameworks are necessary to generate social benefits from technology and that the social and ethical dimensions of this question need to be central. Marden et al. (2016) also directed attention to institutional concerns in what they referred to as the intellectual-property–regulatory complex. Their edited volume highlighted how intellectual property and regulatory frameworks can hinder or promote particular kinds of agricultural innovation and pointed to the need for creative approaches to managing the intellectual-property–regulatory complex.

There are promising new developments in GE crop research in the context of sustainability as defined above, but in most cases the analysis of the fit of new crops and traits with sustainability is assessed only at the environmental level. Even in considering that one domain, the focus of studies can be narrow. A good example is a high-profile GE crop study (Su et al., 2015) that was aimed at reducing the 7–17 percent of total methane in the atmosphere that is produced by farming rice (Bridgham et al., 2013; Kirschke et al., 2013). In the research, Su et al. (2015) engineered a GE rice line that captures more of its starch production in stems and seeds so that there is less starch in the roots. The goal behind the research was to develop a rice plant that provided a less suitable root environment for methane-producing bacteria and thereby mitigated greenhouse gas emissions. The work of Su et al. (2015) was a small experimental study to determine proof of concept, and clearly additional studies will be needed before deployment. In a commentary on the article, Bodelier (2015) noted areas that such future studies should address. For example, more nitrogen inputs may be needed to compensate for the loss of specific bacteria around the roots that deliver plant nutrients, and this could lead to nitrate leaching into the soil and production of more nitrous oxide, a potent greenhouse gas (and other geochemical cycles could be altered as well). Also, a change in root-associated bacterial populations might favor rice pathogens. Of course, there might also be unexpected benefits that become apparent as additional studies are conducted. Further research should be done to determine how the development and diffusion of a crop with the traits in question would be evaluated with respect to contributing to economic concentration, affecting farm

structure and access by small and medium-size farmers, increasing access to food for all, and other social and economic effects that have been addressed in connection with current GE crops in Chapter 6. The committee cautions that it is important that conclusions based on such analyses not be drawn prematurely without knowledge of the properties of the finally developed crop that is to be deployed.

The committee outlines above (see section “Nutrient-Use Efficiency” one approach to increasing phosphorus-use efficiency by engineering plants to secrete phosphatase enzymes that cause the release of phosphate in the farm soil (Wang et al., 2009). While it may prove effective in the short run, a broader view must consider that the reserves of phosphate in the soil will become depleted with time.

From a similar perspective, it is worth examining GE drought tolerance carefully. This trait is considered difficult to engineer, even with emerging genetic-engineering technologies, but is generally considered desirable. That may be true at a general level, but the committee could find no detailed analyses of the long-term effects of different mechanisms of drought tolerance in different environments. If maize or another crop is engineered to be more drought-tolerant, it could become economically profitable in the short term to plant it in drier uncultivated land and thus decrease biodiversity. In some areas, a crop is grown only every other year so that the moisture in the soil can be replenished. With a drought-tolerant variety that had deeper roots, at least initially, the crop could be grown every year until water availability decreased further. There is clearly a need to conduct detailed systems analyses to anticipate long-term effects on environmental sustainability.

Lobell et al. (2014) found that as drought tolerance of conventionally bred maize varieties increased, farmers in the Midwest United States tended to use higher seeding rates to achieve higher yield. However, the higher plant density extracted more water from the soil, which left maize plantings more sensitive to drought. There is a tradeoff between using conventionally bred or GE maize with drought tolerance to maximize yield and using it to guard against losses from drought.

With respect to sustaining health, it is reasonably clear that increasing micronutrients or decreasing fumonisin and aflatoxin concentrations by using genetic-engineering technologies, though challenging to achieve, can improve health in some of the poorest populations of humans (see Chapter 5 section “Foods with Additional Nutrients or Other Healthful Qualities”). The effects of increasing anthocyanins and phenolics on well-nourished populations are less clear because it is unknown whether there are optimal concentrations and whether they could be exceeded. If emerging genetic-engineering technologies could protect plants against most insect pests and pathogens and thereby dramatically reduce the use of synthetic pesticides, there is little doubt that there would be favorable health effects as long

as the GE foods themselves did not have any adverse health effects. Each trait in each crop, environment, and farming system must be considered carefully.

As discussed in connection with current GE crops in Chapter 6, social and economic effects are hard to evaluate, especially on the basis of data on initial adoption. Future GE crops that include multiple traits could be developed in ways that exclude access by resource-poor farmers or that provide traits that are not of much use to such farmers. Social and economic sustainability are more likely to be determined on the basis of the goals of the trait and crop developers than of the technology itself.

FINDING: Although emerging genetic-engineering technologies have the potential to assist in achieving a sustainable food system, broad and rigorous analyses will be necessary to determine the long-term health, environmental, social, and economic outcomes of adding specific crops and traits to an agroecosystem.

The Role of New Genetically Engineered Crops and Traits in Feeding the World

One of the critical questions about new traits and crops that may be enabled by emerging genetic-engineering technologies is the extent to which these products will contribute to feeding the world in the future. As noted above, Tilman et al. (2011) concluded that by 2050 the global crop demand will be about twice what it was in 2005. It has been predicted that genetic engineering could have a major role in achieving the doubling of crop production (Leibman et al., 2014).

On the basis of the committee’s review of the literature, various degrees of uncertainty are associated with the potential of new GE traits and crops to contribute to increased food production. As judged by accomplishments through commercialized GE traits and crops in decreasing damage from insect pests (and to a lesser extent pathogens) reviewed in Chapter 4 and the progress with emerging technologies described in this chapter, there is a strong expectation that new GE traits will be able to reduce losses from insects and pathogens while decreasing the use of synthetic pesticides as long as care is taken to decrease problems related to pest evolution of resistance to new crop traits. Many new GE traits will not increase yield potential but will decrease yield variation, especially by avoiding crop failures due to outbreaks of insect pests and pathogens. The traits could be especially valuable to resource-poor farmers if they are developed in a way that makes their adoption possible by these farmers and assures the public of their safety and if they are not lost due to the evolution of resistance in insect pests and pathogens.

It has been estimated that damage from insect pests and pathogens decreases crop yield by 45 percent, but this is disputed (Yudelman et al., 1998). It will be important to conduct in-depth studies to gain better estimates of how much protection from insects and pathogens could increase global food production. In an in-depth analysis, Savary et al. (2000) examined losses to rice throughout Asia because of insects, pathogens, and weeds (Savary et al., 2000). They found that total losses were on average 37.2 percent of the yield without biotic stresses. However, over 20 percent of the yield loss was from weeds; the main target for Bt rice, the stem borer (Scirpophaga incertulas), only caused an average loss in yield of 2.3 percent. The three major pathogens caused 1–10 percent of the yield loss. Damage estimates differed among rice production systems. These results emphasize the need to carefully assess how any specific intervention with GE traits will contribute to the goal of doubling food production.

As discussed in this chapter, increasing overall yield by engineering more efficient photosynthesis and nutrient use or by increasing drought tolerance is feasible, but there is uncertainty about the degree of success that will come in developing such traits in the future. Given the early stages of the research, even detailed reviews and assessments will not be able to predict how much increase in overall crop production will come from these efforts.

As explained in Chapter 6, feeding the world involves much more than simply increasing crop production. The choices of who invests in new crops and traits, what specific crops and traits are invested in, and other factors such as availability of institutional support and access to inputs will affect how much emerging genetic-engineering technologies will contribute to feeding the world. The policy-makers who will in part determine those choices have the heavy responsibility of gathering information on ways to optimize investment in emerging genetic-engineering technologies that will address the four domains of sustainability discussed above.

FINDING: Given the uncertainty about how much emerging genetic-engineering technologies will increase crop production, viewing such technologies as major contributors to feeding the world must be accompanied by careful caveats.

RECOMMENDATION: Balanced public investment in emerging genetic-engineering technologies and in a variety of other approaches should be made because it will be critical for decreasing the risk of global and local food shortages.

CONCLUSIONS

Genetic engineering and conventional breeding are complementary approaches, and more progress in crop improvement will be made by using both conventional breeding and genetic engineering than by using either alone. A greater understanding of the genetic basis of complex traits, such as drought tolerance and nitrogen-use efficiency, will require basic research investment in both approaches, although some complex traits are unlikely to be introduced without genetic engineering. The research on complex traits is at too early a stage to predict which traits will and will not be successfully developed. The specific traits that will be available to farmers and the specific crops and varieties in which the traits will be available will depend on the extent of investment in crop improvement by the private and public sectors. Beyond technical hurdles, future ecological, social, and economic issues and the regulatory landscape will affect decisions in research investments by the public and private sectors and the diffusion of new discoveries. It is critical that investment in solutions involving genetic engineering not diminish investment in other technologies already shown to efficiently increase sustainable crop production and nutrition.

REFERENCES

Akagi, T., A. Ikegami, and K. Yonemori. 2009. DkMyb2 wound-induced transcription factor of persimmon (Diospyros kaki Thunb.), contributes to proanthocyanidin regulation. Planta 23:1045–1059.

Bange, M.P., G. Constable, D. McRae, and G. Roth. 2010. Cotton. Pp. 41–58 in Adapting Agriculture to Climate Change: Preparing Australian Agriculture, Forestry and Fisheries for the Future, S.M. Howden and C. Stokes, eds. Canberra: CSIRO.

Baum, J.A., T. Bogaert, W. Clinton, G.R. Heck, P. Feldmann, O. Ilagan, S. Johnson, G. Plaetinck, T. Munyikwa, M.Pleau, T. Vaughn, and J. Roberts. 2007. Control of coleopteran insect pests through RNA interference. Nature Biotechnology 25:1322–1326.

Bauman, J.M., C.H. Keiffer, and B.C. McCarthy. 2014. Backcrossed chestnut seedling performance and blight incidence (Cryphonectria parasitica) in restoration. New Forests 45:813–828.

Beale, M.H., M.A. Birkett, T.J.A. Bruce, K. Chamberlain, L.M. Field, A.K. Huttly, J.L. Martin, R. Parker, A.L. Phillips, J.A. Pickett, I.M. Prosser, P.R. Shewry, L.E. Smart, L.J. Wadhams, C.M. Woodcock, and Y. Zhang. 2006. Aphid alarm pheromone produced by transgenic plants affects aphid and parasitoid behavior. Proceedings of the National Academy of Sciences of the United States of America 103:10509–10513.

Beatty, P.H., and A.G. Good. 2011. Future prospects for cereals that fix nitrogen. Science 333:416–417.

Bennett, E., S.R. Carpenter, L.J. Gordon, N. Ramankutty, P. Balvanera, B. Campbell, W. Cramer, J. Foley, C. Folke, L. Karlberg, J. Liu, H. Lotze-Campen, N. Mueller, G. Peterson, S. Polasky, J. Rockström, R. Scholes, and M. Spierenburg. 2014. Toward a more resilient agriculture. Solutions 5:65–75.

Betancor, M.B., M. Sprague, S. Usher, O. Sayanova, P.J. Campbell, J.A. Napier, and D.R. Tocher. 2015. A nutritionally-enhanced oil from transgenic Camelina sativa effectively replaces fish oil as a source of eicosapentaenoic acid for fish. Scientific Reports 5.

Bodelier, P.L.E. 2015. Sustainability: Bypassing the methane cycle. Nature 523:534–535.

Boudet, A.M., C. Lapierre, and J. Grima-Pettenati. 1995. Tansley Review No. 80. Biochemistry and molecular biology of lignification. New Phytologist 129:203–236.

Bovy, A.G., R. de Vos, M. Kemper, E. Schijlen, M.A. Pertejo, S. Muir, G. Collins, S. Robinson, M. Verhoeyen, S. Hughes, C. Santos-Buelga, and A. van Tunen. 2002. High-flavonol tomatoes resulting from the heterologous expression of the maize transcription factor genes LC and C1. The Plant Cell 14:2509–2526.

Bräutigam, A., S. Schliesky, C. Külahoglu, C.P. Osborne, and A.P.M. Weber. 2014. Towards an integrative model of C4 photosynthetic subtypes: Insights from comparative transcriptome analysis of NAD-ME, NADP-ME, and PEP-CK C4 species. Journal of Experimental Botany 65:3579–3593.

Bridgham, S.D., C.-Q. Hinsby, K.K. Jason, and Q. Zhuang. 2013. Methane emissions from wetlands: Biogeochemical, microbial, and modeling perspectives from local to global scales. Global Change Biology 19:1325–1346.

Broun, P. 2004. Transcription factors as tools for metabolic engineering in plants. Current Opinion in Plant Biology 7:202–209.

Butelli, E., L. Titta, M. Giorgio, H.-P. Mock, A. Matros, S. Peterek, E.G.W.M. Schijlen, R.D. Hall, A.G. Bovy, J. Luo, and C. Martin. 2008. Enrichment of tomato fruit with health-promoting anthocyanins by expression of select transcription factors. Nature Biotechnology 26:1301–1308.

Camacho, A., A. Van Deynze, C. Chi-Ham, and A.B. Bennett. 2014. Genetically engineered crops that fly under the US regulatory radar. Nature Biotechnology 32:1087–1091.

Castiglioni, P., D. Warner, R.J. Bensen, D.C. Anstrom, J. Harrison, M. Stoecker, M. Abad, G. Kumar, S. Salvador, R. D’Ordine, S. Navarro, S. Back, M. Fernandes, J. Targolli, S. Dasgupta, C. Bonin, M.H. Luethy, and J.E. Heard. 2008. Bacterial RNA chaperones confer abiotic stress tolerance in plants and improved grain yield in maize under water-limited conditions. Plant Physiology 147:446–455.

Çekiç, Ç., and M. Özgen. 2010. Comparison of antioxidant capacity and phytochemical properties of wild and cultivated red raspberries (Rubus idaeus L.). Journal of Food Composition and Analysis 23:540–544.

Chen, F., and R.A. Dixon. 2007. Lignin modification improves fermentable sugar yields for biofuel production. Nature Biotechnology 25:759–761.

Cotter, J. 2014. GE Crops—Necessary? Presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, September 16, Washington, DC.

Coulman, B., B. Goplen, W. Majak, T. McAllister, K.J. Cheng, B. Berg, J. Hall, D. McCartney, and S. Acharya. 2000. A review of the development of a bloat-reduced alfalfa cultivar. Canadian Journal of Plant Science 80:487–491.

Coviella, C.E., R.D. Stipanovic, and J.T. Trumble. 2002. Plant allocation to defensive compounds: Interactions between elevated CO₂ and nitrogen in transgenic cotton plants. Journal of Experimental Botany 53:323–331.

Curatti, L., and L.M. Rubio. 2014. Challenges to develop nitrogen-fixing cereals by direct nif-gene transfer. Plant Science 225:130–137.

Davis, D.R., M.D. Epp, and H.D. Riordan. 2004. Changes in USDA food composition data for 43 garden crops, 1950 to 1999. Journal of the American College of Nutrition 23:669–682.

Dickinson, B., Y. Zhang, J.S. Petrick, G. Heck, S. Ivashuta, and W.S. Marshall. 2013. Lack of detectable oral bioavailability of plant microRNAs after feeding in mice. Nature Biotechnology 31:965–967.

Ding, S.-Y., Y.-S. Liu, Y. Zeng, M.E. Himmel, J.O. Baker, and E.A. Bayer. 2012. How does plant cell wall nanoscale architecture correlate with enzymatic digestibility? Science 338:1055–1060.

Dixon, R.A., C. Liu, and J.H. Jun. 2012. Metabolic engineering of anthocyanins and condensed tannins in plants. Current Opinion in Biotechnology 24:329–335.

Dixon, R.A., M.S. Srinivasa Reddy, and L. Gallego-Giraldo. 2014. Monolignol biosynthesis and its genetic manipulation: The good, the bad and the ugly. Pp. 1–38 in Recent Advances in Polyphenol Research, A. Romani, V. Lattazio, and S. Quideau, eds. Chichester, UK: John Wiley & Sons, Ltd.

Easlon, H., K. Nemali, J. Richards, D. Hanson, T. Juenger, and J. McKay. 2014. The physiological basis for genetic variation in water use efficiency and carbon isotope composition in Arabidopsis thaliana. Photosynthesis Research 119:119–129.

Ervin, D.E., L.L. Glenna, and R.A. Jussaume, Jr. 2011. The theory and practice of genetically engineered crops and agricultural sustainability. Sustainability 3:847–874.

Fischer, T., D. Byerlee, and G. Edmeades. 2014. Crop Yields and Global Food Security: Will Yield Increase Continue to Feed the World? Canberra: Australian Centre for International Agricultural Research.

Fjose, A., S. Ellingsen, A. Wargelius, and H.-C. Seo. 2001. RNA interference: Mechanisms and applications. Biotechnology Annual Review 7:31–57.

Folberth, C., H. Yang, T. Gaiser, J. Liu, X. Wang, J. Williams, and R. Schulin. 2014. Effects of ecological and conventional agricultural intensification practices on maize yields in sub-Saharan Africa under potential climate change. Environmental Research Letters 9:044004.

Foley, J.A., R. DeFries, G.P. Asner, C. Barford, G. Barford, S.R. Carpenter, F.S. Chapin, M.T. Coe, G.C. Daily, H.K. Gibbs, J.H. Helkowski, T. Holloway, E.A. Howard, C.J. Kucharik, C. Monfreda, J.A. Patz, I.C. Prentice, N. Ramankutty, and P.K. Snyder. 2005. Global consequence of land use. Science 309:570–574.

Galili, G., and R. Amir. 2013. Fortifying plants with essential amino acids lysine and methionine to improve nutritional quality. Plant Biotechnology Journal 11:211–222.

Gamuyao, R., J.H. Chin, J. Pariasca-Tanaka, P. Pesaresi, S. Catausan, C. Dalid, I. Slamet-Loedin, E.M. Tecson-Mendoza, M. Wissuwa, and S. Heuer. 2012. The protein kinase Pstol1 from traditional rice confers tolerance of phosphorus deficiency. Nature 488:535–539.

Garnett, T., M.C. Appleby, A. Balmford, I.J. Bateman, T.G. Benton, P. Bloomer, D. Burlingame, M. Dawkins, L. Dolan, D. Fraser, M. Herrero, I. Hoffmann, P. Smith, P.K. Thornton, C. Toulmin, S.J. Vermeulen, and H.J.C. Godfray. 2013. Sustainable intensification in agriculture: Premises and policies. Science 341:33–34.

Gibbs, H.K., A.S. Ruesch, M.K. Achard, M.K. Clayton, P. Holmgren, N. Ramankutty, and J.A. Foley. 2010. Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. Proceedings of the National Academy of Sciences of the United States of America 107:16732–16737.

Godfray, H.C.J., J.R. Beddington, I.R. Crute, L. Haddad, D. Lawrence, J.F. Muir, J. Pretty, S. Robinson, S.M. Thomas, and C. Toulmin. 2010. Food security: The challenge of feeding 9 billion people. Science 327:812–818.

Goldman, I.L. 2014. The future of breeding vegetables with human health functionality: Realities, challenges and opportunities. HortScience 49:133–137.

Gonzali, S., A. Mazzucato, and P. Perata. 2009. Purple as a tomato: Towards high anthocyanin tomatoes. Trends in Plant Science 14:237–241.

Graham, R., D. Senadhira, S. Beebe, C. Iglesias, and I. Monasterio. 1999. Breeding for micronutrient density in edible portions of staple food crops: Conventional approaches. Field Crops Research 60:57–80.

Gould, F. 1995. Comparisons between resistance management strategies for insects and weeds. Weed Technology 9:830–839.

Gurian-Sherman, D. 2014. Remarks to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, September 16, Washington, DC.

Gururani, M.A., J. Venkatesh, C.P. Upadhyaya, A. Nookaraju, S.K. Pandey, and S.W. Park. 2012. Plant disease resistance genes: Current status and future directions. Physiological and Molecular Plant Pathology 78:51–65.

Hain, R., H.-J.Reif, E. Krause, R. Langebartels, H. Kindl, B. Vornam, W. Wiese, E. Schmelzer, P. Schreier, R.H. Stöcker, and K. Stenzel. 1993. Disease resistance results from foreign phytoalexin expression in a novel plant. Nature 361:153–156.

Hancock, K.R., V. Collette, K. Fraser, M. Greig, H. Xue, K. Richardson, C. Jones, and S. Rasmussen. 2012. Expression of the R2R3-MYB transcription factor TaMYB14 from Trifolium arvense activates proanthocyanidin biosynthesis in the legumes Trifolium re-pens and Medicago sativa. Plant Physiology 159:1204–1220.

Hansen, M. 2014. The Need for Mandatory Safety Assessment for GE Crops. Presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, September 16, Washington, DC.

Haverkort, A.J., P.M. Boonekamp, P. Hutten, E. Jacobsen, L.A.P. Lotz, G.J.T. Kessel, R.G.F. Visser, and E.A.G. van der Vossen. 2008. Societal costs of late blight in potato and prospects of durable resistance through cisgenic modification. Potato Research 51:47–57.

Haverkort, A.J., P.C. Struik, R.G.F. Visser, and E. Jacobsen. 2009. Applied biotechnology to combat late blight in potato caused by Phytophthora infestans. Potato Research 52:249–264.

Hegde, K., N. Chandra, S. Sarma, S. Brar, and V. Veeranki. 2015. Genetic engineering strategies for enhanced biodiesel production. Molecular Biotechnology 57:606–624.

Himmel, M.E., S.-Y. Ding, D.K. Johnson, W.S. Adney, M.R. Nimlos, J.W. Brady, and T.D. Foust. 2007. Biomass recalcitrance: Engineering plants and enzymes for biofuels production. Science 315:804–807.

Howden, S.M., J. Soussana, F.N. Tubiello, N. Chhetri, M. Dunlop, and H. Meinke. 2007. Adapting agriculture to climate change. Proceedings of the National Academy of Sciences of the United States of America 104:19691–19696.

Hubbell, B.J., and R. Welsh. 1998. Transgenic crops: Engineering a more sustainable agriculture? Agriculture and Human Values 15:43–56.

IAASTD (International Assessment of Agricultural Knowledge, Science and Technology for Development). 2009. Agriculture at a Crossroads: Global Report. Washington, DC: Island Press.

IPCC (Intergovernmental Panel on Climate Change). 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva: IPCC.

Jarvis, A., J. Ramirez-Villegas, B.V. Herrera Campo, and C.E. Navarro-Racines. 2012. Is cassava the answer to African climate change adaptation? Tropical Plant Biology 5:9–29.

Jeandet, P., C. Clement, E. Courot, and S. Cordelier. 2013. Modulation of phytoalexin biosynthesis in engineered plants for disease resistance. International Journal of Molecular Sciences 14:14136–14170.

Jin, S., N.D. Singh, L. Li, X. Zhang, and H. Daniell. 2015. Engineered chloroplast dsRNA silences cytochrome p450 monooxygenase, V-ATPase and chitin synthase genes in the insect gut and disrupts Helicoverpa armigera larval development and pupation. Plant Biotechnology Journal 13:435–446.

Jingura, R.M., and R. Kamusoko. 2014. Experiences with Jatropha cultivation in sub-Saharan Africa: Lessons for the next phase of development. African Journal of Science, Technology, Innovation and Development 6:333–337.

Johnson, E.J. 2002. The role of carotenoids in human health. Nutrition in Clinical Care 5:56–65.

Jones, J.D.G., K. Witek, W. Verweij, F. Jupe, D. Cooke, S. Dorling, L. Tomlinson, M. Smoker, S. Perkins, and S. Foster. 2014. Elevating crop disease resistance with cloned genes. Philosophical Transactions of the Royal Society B: Biological Sciences 369:20130087.

Kalluri, U.C., H. Yin, X. Yang, and B.H. Davison. 2014. Systems and synthetic biology approaches to alter plant cell walls and reduce biomass recalcitrance. Plant Biotechnology Journal 12:1207–1216.

King, D.J., R.M. Gleadow, and I.E. Woodrow. 2004. Terpene deployment in Eucalyptus polybractea; relationships with leaf structure, environmental stresses, and growth. Functional Plant Biology 31:451–460.

Kirschke, S., P. Bousquet, P. Ciais, M. Saunois, J.G. Canadell, E.J. Dlugokencky, P. Bergamaschi, D. Bergmann, D.R. Blake, L. Bruhwiler, P. Cameron-Smith, S. Castaldi, F. Chevallier, L. Feng, A. Fraser, M. Heimann, E.L. Hodson, S. Houweling, B. Josse, P. Fraser, P.B. Krummel, J.-F. Lamarque, R.L. Langenfelds, C. Le Quéré, V. Naik, S. O’Doherty, P.I. Palmer, I. Pison, D. Plummer, B. Poulter, R.G. Prinn, M. Rigby, B. Ringeval, M. Santini, M. Schmidt, D.T. Shindell, I.J. Simpson, R. Spahni, L.P. Steele, S.A. Strode, K. Sudo, S. Szopa, G.R. van der Werf, A. Voulgarakis, M. van Weele, R.F. Weiss, J.E. Williams, and G. Zeng. 2013. Three decades of global methane sources and sinks. Nature Geoscience 6:813–823.

Koči, J., P. Ramaseshadri, R. Bolognesi, G. Segers, R. Flannagan, and Y. Park. 2014. Ultra-structural changes caused by Snf7 RNAi in larval enterocytes of western corn rootworm (Diabrotica virgifera virgifera Le Conte). PLoS ONE 9:e83985.

Kumar, P., S.S. Pandit, A. Steppuhn, and I.T. Baldwin. 2014. Natural history-driven, plant-mediated RNAi-based study reveals CYP6B46’s role in a nicotine-mediated antipredator herbivore defense. Proceedings of the National Academy of Sciences of the United States of America 111:1245–1252.

Lees, G.L. 1992. Condensed tannins in some forage legumes: Their role in the prevention of ruminant pasture bloat. Pp. 915–934 in Plant Polyphenols, R.W. Hemingway and P.E. Laks, eds. New York: Plenum Press.

Leibman, M., J.J. Shryock, M.J. Clements, M.A. Hall, P.J. Loida, A.L. McClerren, Z.P. McKiness, J.R. Phillips, E.A. Rice, and S.B. Stark. 2014. Comparative analysis of maize (Zea mays) crop performance: Natural variation, incremental improvements and economic impacts. Plant Biotechnology Journal 12:941–950.

Levine, S.L., J. Tan, G.M. Mueller, P.M. Bachman, P.D. Jensen, and J.P. Uffman. 2015. Independent action between DvSnf7 RNA and Cry3Bb1 protein in southern corn rootworm, Diabrotica undecimpunctata howardi and Colorado potato beetle, Leptinotarsa decemlineata. PLoS ONE 10:e0118622.

Li, Q., J. Song, S. Peng, J.P. Wang, G.-Z. Qu, R.R. Sederoff, and V.L. Chiang. 2014. Plant biotechnology for lignocellulosic biofuel production. Plant Biotechnology Journal 12:1174–1192.

Lim, W., R. Miller, J. Park, and S. Park. 2014. Consumer sensory analysis of high flavonoid transgenic tomatoes. Journal of Food Science 79:S1212–S1217.

Lin, M.T., A. Occhialini, P.J. Andralojc, M.A.J. Parry, and M.R. Hanson. 2014. A faster Rubisco with potential to increase photosynthesis in crops. Nature 513:547–550.

Liu, C.-J., Y. Cai, X. Zhang, M. Gou, and H. Yang. 2014. Tailoring lignin biosynthesis for efficient and sustainable biofuel production. Plant Biotechnology Journal 12:1154–1162.

Lobell, D.B., M.J. Roberts, W. Schlenker, N. Braun, B.B. Little, R.M. Rejesus, and G.L. Hammer. 2014. Greater sensitivity to drought accompanies maize yield increase in the U.S. Midwest. Science 344:516–519.

Lucas, D.M., M.L. Taylor, G.F. Hartnell, M.A. Nemeth, K.C. Glenn, and S.W. Davis. 2007. Broiler performance and carcass characteristics when fed diets containing lysine maize (LY038 or LY038 x MON 810), control, or conventional reference maize. Poultry Science 86:2152–2161.

Macnaghten, P., and S. Carro-Ripalda, eds. 2015. Governing Agricultural Sustainability: Global Lessons from GM Crops. New York: Routledge.

Mao, Y.-B., W.-J. Cai, J.-W. Wang, G.-J. Hong, X.-Y. Tao, L.-J. Wang, Y.-P. Huang, and X.-Y. Chen. 2007. Silencing a cotton bollworm P450 monooxygenase gene by plant-mediated RNAi impairs larval tolerance of gossypol. Nature Biotechnology 25:1307–1313.

Mao, G., X. Meng, Y. Liu, Z. Zheng, Z. Chen, and S. Zhang. 2011a. Phosphorylation of a WRKY transcription factor by two pathogen-responsive MAPKs drives phytoalexin biosynthesis in Arabidopsis. The Plant Cell 23:1639–1653.

Mao, Y.-B., X.-Y. Tao, X.-Y. Xue, L.-J. Wang, and X.-Y. Chen. 2011b. Cotton plants expressing CYP6AE14 double-stranded RNA show enhanced resistance to bollworms. Transgenic Research 20:665–673.

Marden, E., R.N. Godfrey, and R. Manion, eds. 2016. The Intellectual Property–Regulatory Complex: Overcoming Barriers to Innovation in Agricultural Genomics. Vancouver: UBC Press.

Masuda, H., Y. Ishimaru, M.S. Aung, T. Kobayashi, Y. Kakei, M. Takahashi, K. Higuchi, H. Nakanishi, and N.K. Nishizawa. 2012. Iron biofortification in rice by the introduction of multiple genes involved in iron nutrition. Scientific Reports 2:543.

Mertz, E.T., L.S. Bates, and O.E. Nelson. 1964. Mutant gene that changes protein composition and increases lysine content of maize endosperm. Science 145:279–280.

Mielenz, J.R. 2006. Bioenergy for ethanol and beyond. Current Opinion in Biotechnology 17:303–304.

Mueller, N.D., J.S. Gerber, M. Johnston, D.K. Ray, N. Ramankutty, and J.A. Foley. 2012. Closing yield gaps through nutrient and water management. Nature 490:254–257.

Muir, S.R., G.J. Collins, S. Robinson, S. Hughes, A. Bovy, C.H.R. De Vos, A.J. van Tunen, and M.E. Verhoeyen. 2001. Overexpression of petunia chalcone isomerase in tomato results in fruit containing increased levels of flavonols. Nature Biotechnology 19:470–474.

Nehra, N.S., and L. Pearson. 2011. Petition for Determination of Non-regulated Status for Freeze Tolerant Hybrid Eucalyptus Lines. Available at https://www.aphis.usda.gov/brs/aphisdocs/11_01901p.pdf. Accessed November 20, 2015.

NRC (National Research Council). 2015. A Framework for Assessing Effects of the Food System. Washington, DC: National Academies Press.

Nyaboga, E., J. Njiru, E. Nguu, W. Gruissem, H. Vanderschuren, and L. Tripathi. 2013. Unlocking the potential of tropical root crop biotechnology in east Africa by establishing a genetic transformation platform for local farmer-preferred cassava cultivars. Frontiers in Plant Science 4:526.

Ogwok, E., J. Odipio, M. Halsey, E. Gaitán-Solís, A. Bua, N.J. Taylor, C.M. Fauquet, and T. Alicai. 2012. Transgenic RNA interference (RNAi)-derived field resistance to cassava brown streak disease. Molecular Plant Pathology 13:1019–1031.

Paillet, F.L. 2002. Chestnut: History and ecology of a transformed species. Journal of Bio-geography 29:1517–1530.

Palmgren, M.G., A.K. Edenbrandt, S.E. Vedel, M.M. Anderson, X. Landes, J.T. Osterberg, J. Falhof, L.I. Olsen, S.O. Christensen, P. Sandoe, C. Gamborg, K. Kappel, B.J. Thorsen, and P. Pagh. 2015. Are we ready for back-to-nature crop breeding? Trends in Plant Science 20:155–164.

Parisi, C., P. Tillie, and E. Rodríguez-Cerezo. 2016. The global pipeline of GM crops out to 2020. Nature Biotechnology 34:31–36.

Park, S.Y., F.C. Peterson, A. Mosquna, J. Yao, B.F. Volkman, and S.R. Cutler. 2015. Agrochemical control of plant water use using engineered abscisic acid receptors. Nature 520:545–548.

Petrie, J.R., P. Shrestha, X.R. Zhou, M.P. Mansour, Q. Li, S. Belide, P.D. Nichols, and S.P. Singh. 2012. Metabolic engineering of seeds with fish oil-like levels of DHA. PLoS ONE 7:e49165.

Pfeffer, M.J. 1992 Sustainable agriculture in historical perspective. Agriculture and Human Values 9:4–12.

Phalan, B., M. Bertzky, S.H.M. Butchart, P.F. Donald, J.P.W. Scharlemann, A.J. Stattersfield, and A. Balmford. 2013. Crop expansion and conservation priorities in tropical countries. PLoS ONE 8:1–13.

Pham, A.T., J.G. Shannon, and K.D. Bilyeu. 2012. Combinations of mutant FAD2 and FAD3 genes to produce high oleic acid and low linolenic acid soybean oil. Theoretical and Applied Genetics 125:503-515.

Poirier, Y., and J.-Y. Jung. 2015. Phosphate transporters. Pp. 125–158 in Phosphorus Metabolism in Plants, Annual Plant Reviews Volume 48, W. Plaxton and H. Lambers, eds. Chicester, UK: John Wiley & Sons, Inc.

Pojer, E., F. Mattivi, D. Johnson, and C.S. Stockley. 2013. The case for anthocyanin consumption to promote human health: A review. Comprehensive Reviews in Food Science and Food Safety 12:483–508.

Poovaiah, C.R., M. Nageswara-Rao, J.R. Soneji, H.L. Baxter, and C.N. Stewart. 2014. Altered lignin biosynthesis using biotechnology to improve lignocellulosic biofuel feedstocks. Plant Biotechnology Journal 12:1163–1173.

Povero, G., S. Gonzali, L. Bassolino, A. Mazzucato, and P. Perata. 2011. Transcriptional analysis in high-anthocyanin tomatoes reveals synergistic effect of Aft and atv genes. Journal of Plant Physiology 168:270–279.

Powell, W. 2015. Additional Tools for Solving an Old Problem: The Return of the American Chestnut. Webinar presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, March 27.

Ragauskas, A.J., C.K. Williams, B.H. Davison, G. Britovesk, J. Cairney, C.A. Eckert, W.J.J. Frederick, J.P. Hallett, D.J. Leak, C.L. Liotta, J.R. Mielenz, R. Murphy, R. Templer, and T. Tschaplinski. 2006. The path forward for biofuels and biomaterials. Science 311:484–489.

Ragauskas, A.J., G.T. Beckham, M.J. Biddy, R. Chandra, F. Chen, M.F. Davis, B.H. Davison, R.A. Dixon, P. Gilna, M. Keller, P. Langan, A.K. Naskar, J.N. Saddler, T.J. Tschaplinski, G.A. Tuskan, and C.E. Wyman. 2014. Lignin valorization: Improving lignin processing in the biorefinery. Science 344:709.

Rengel, Z., G.D. Batten, and D.E. Crowley. 1999. Agronomic approaches for improving the micronutrient density in edible portions of field crops. Field Crops Research 60:27–40.

Robinson, J. May 25, 2013. Breeding the nutrition out of our food. Online. New York Times. Available at http://www.nytimes.com/2013/05/26/opinion/sunday/breeding-the-nutritionout-of-our-food.html?pagewanted=all&_r=0. Accessed December 11, 2015.

Rogers, C., and G.E.D. Oldroyd. 2014. Synthetic biology approaches to engineering the nitrogen symbiosis in cereals. Journal of Experimental Botany 65:1939–1946.

Rothstein, S.J., Y.-M. Bi, V. Coneva, M. Han, and A. Good. 2014. The challenges of commercializing second-generation transgenic crop traits necessitate the development of international public sector research infrastructure. Journal of Experimental Botany 65:5673–5682.

Ruiz-López, N., O. Sayanova, J.A. Napier, and R.P. Haslam. 2012. Metabolic engineering of the omega-3 long chain polyunsaturated fatty acid biosynthetic pathway into transgenic plants. Journal of Experimental Botany 63:2397–2410.

Ruiz-Lopez, N., R.P. Haslam, J.A. Napier, and O. Sayanova. 2014. Successful high-level accumulation of fish oil omega-3 long-chain polyunsaturated fatty acids in a transgenic oilseed crop. The Plant Journal 77:198–208.

Sample, I. June 25, 2015. GM wheat no more pest-resistant than ordinary crops, trial shows. Online. The Guardian. Available at http://www.theguardian.com/environment/2015/jun/25/gm-wheat-no-more-pest-resistant-than-ordinary-crops-trial-shows. Accessed December 12, 2015.

Savary, S., L. Willocquet, R.A. Elazegui, N.P. Castilla, and P.S. Teng. 2000. Rice pest constraints in tropical Asia: Quantification of yield losses due to rice pests in a range of production situations. Plant Disease 84:357–369.

Sayre, R. 2014. Engineering Plants and Algae for Improved Biomass Production. Webinar presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, November 6.

Sayre, R., J.R. Beeching, E.B. Cahoon, C. Egesi, C. Fauquet, J. Fellman, M. Fregene, W. Gruissem, S. Mallowa, M. Manary, B. Maziya-Dixon, A. Mbanaso, D.P. Schachtman, D. Siritunga, N.G. Taylor, H. Vanderschuren, and P. Zhang. 2011. The BioCassava Plus Program: Biofortification of cassava for sub-Saharan Africa. Annual Review of Plant Biology 62:251–272.

Scheible, W-R., and M. Rojas-Triana. 2015. Sensing, signaling and control of phosphate starvation in plants—molecular players and application. Pp. 25–64 in Phosphorus Metabolism in Plants, Annual Plant Reviews Volume 48, W. Plaxton and H. Lambers, eds. Chichester, UK: John Wiley & Sons, Inc.

Shand, H. 2014. Corporate Concentration in GE Crops: What Impact on Farmers, Biodiversity and Food Security? Presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, September 15, Washington, DC.

Shukla, V.K., Y. Doyon, J.C. Miller, R.C. DeKelver, E.A. Moehle, S.E. Worden, J.C. Mitchell, N.L. Arnold, S. Gopalan, X. Meng, V.M. Choi, J.M. Rock, Y.-Y. Wu, G.E. Katibah, G. Zhifang, D. McCaskill, M.A. Simpson, B. Blakeslee, S.A. Greenwalt, H.J. Butler, S.J. Hinkley, L. Zhang, E.J. Rebar, P.D. Gregory, and F.D. Urnov. 2009. Precise genome modification in the crop species Zea mays using zinc-finger nucleases. Nature 459:437–443.

Snow J.W., A.E. Hale, S.K. Isaacs, A.L. Baggish, and S.Y. Chan. 2013. Ineffective delivery of diet-derived microRNAs to recipient animal organisms. RNA Biology 10:1107–1116.

Sprague, S.J., S.J. Marcroft, H.L. Hayden, and B.J. Howlett. 2006. Major gene resistance to blackleg in Brassica napus overcome within three years of commercial production in Southeastern Australia. Plant Disease 90:190–198.

Stall, R.E., J.B. Jones, and G.V. Minsavage. 2009. Durability of resistance in tomato and pepper to Xanthomonads causing bacterial spot. Annual Review of Phytopathology 47:265–284.

Studer, M.H., J.D. DeMartini, M.F. Davis, R.W. Sykes, B. Davison, M. Keller, G.A. Tuskan, and C.E. Wymann. 2011. Lignin content in natural Populus variants affects sugar release. Proceedings of the National Academy of Sciences of the United States of America 108:6300–6305.

Su, J., C. Hu, X. Yan, Y. Jin, Z. Chen, Q. Guan, Y. Wang, D. Zhong, C. Jansson, F. Wang, A. Schnürer, and C. Sun. 2015. Expression of barley SUSIBA2 transcription factor yields high-starch low-methane rice. Nature 523:602–606.

Taylor, N., E. Gaitán-Solís, T. Moll, B. Trauterman, T. Jones, A. Pranjal, C. Trembley, V. Abernathy, D. Corbin, and C. Fauquet. 2012. A high-throughput platform for the production and analysis of transgenic cassava (Manihot esculenta) plants. Tropical Plant Biology 5:127–139.

Terenius, O., A. Papanicolaou, J.S. Garbutt, I. Eleftherianos, H. Huvenne, S. Kanginakudru, M. Albrechtsen, C. An, J.-L. Aymeric, A. Barthel, P. Bebas, K. Bitra, A. Bravo, F. Chevalier, D.P. Collinge, C.M. Crava, R.A. de Maagd, B. Duvic, M. Erlandson, I. Faye, G. Felföldi, H. Fujiwara, R. Futahashi, A.S. Gandhe, H.S. Gatehouse, L.N. Gatehouse, J.M. Giebultowicz, I. Gómez, C.J.P. Grimmelikhuijzen, A.T. Groot, F. Hauser, D.G. Heckel, D.D. Hegedus, S. Hrycaj, L. Huang, J.J. Hull, K. Iatrou, M. Iga, M.R. Kanost, J. Kotwica, C. Li, J. Li, J. Liu, M. Lundmark, S. Matsumoto, M. Meyering-Vos, P.J. Millichap, A. Monteiro, N. Mrinal, T. Niimi, D. Nowara, A. Ohnishi, V. Oostra, K. Ozaki, M. Papakonstantinou, A. Popadic, M.V. Rajam, S. Saenko, R.M. Simpson, M. Soberón, M.R. Strand, S. Tomita, U. Toprak, P. Wang, C.W. Wee, S. Whyard, W. Zhang, J. Nagaraju, R.H. Ffrench-Constant, S. Herrero, K. Gordon, L. Swevers, and G. Smagghe. 2011. RNA interference in Lepidoptera: An overview of successful and unsuccessful studies and implications for experimental design. Journal of Insect Physiology 57:231–245.

Tilman, D., C. Balzer, J. Hill, and B.L. Befort. 2011. Global food demand and the sustainable intensification of agriculture. Proceedings of the National Academy of Sciences of the United States of America 108:20260–20264.

Tripathi, L., H. Mwaka, J.N. Tripathi, and W.K. Tushemereirwe. 2010. Expression of sweet pepper Hrap gene in banana enhances resistance to Xanthomonas campestris pv. musacearum. Molecular Plant Pathology 11:721–731.

Truska, M., P. Vrintin, and X. Qiu. 2009. Metabolic engineering of plants for polyunsaturated fatty acid production. Molecular Breeding 23:1–11.

Tsuda, T. 2012. Dietary anthocyanin-rich plants: Biochemical basis and recent progress in health benefits studies. Molecular Nutrition & Food Research 56:159–170.

USDA–APHIS (U.S. Department of Agriculture–Animal and Plant Health Inspection Service). 2011. Questions and Answers: Roundup Ready Alfalfa Deregulation. Available at https://www.aphis.usda.gov/publications/biotechnology/2011/rr_alfalfa.pdf. Accessed December 12, 2015.

USDA–APHIS (U.S. Department of Agriculture–Animal and Plant Health Inspection Service). 2015. Determination of nonregulated status for MON 87411 corn. Available at https://www.aphis.usda.gov/brs/aphisdocs/13_29001p_det.pdf. Accessed December 12, 2015.

Usher, S., R.P. Haslam, N. Ruiz-Lopez, O. Sayanova, and J.A. Napier. 2015. Field trial evaluation of the accumulation of omega-3 long chain polyunsaturated fatty acids in transgenic Camelina sativa: Making fish oil substitutes in plants. Metabolic Engineering Communications 2:93–98.

Vanderschuren, H., I. Moreno, R.B. Anjanappa, I.M. Zainuddin, and W. Gruissem. 2012. Exploiting the combination of natural and genetically engineered resistance to Cassava Mosaic and Cassava Brown Streak Viruses impacting cassava production in Africa. PLoS ONE 7:e45277.

Vanloqueren, G., and P.V. Baret. 2009. How agricultural research systems shape a technological regime that develops genetic engineering but locks out agroecological innovations. Research Policy 38:971–983.

Vasal, S.K. 2000. The quality protein maize story. Food and Nutrition Bulletin 21:445–450.

Veneklaas, E.J., H. Lambers, J. Bragg, P.M. Finnegan, C.E. Lovelock, W.C. Plaxton, C.A. Price, W.-R. Scheible, M.W. Shane, P.J. White, and J.A. Raven. 2012. Opportunities for improving phosphorus-use efficiency in crop plants. New Phytologist 195:306–320.

Vogel, B. 2009. Marker-Assisted Selection: A Non-Invasive Biotechnology Alternative to Genetic Engineering of Plant Varieties. Amsterdam: Greenpeace International.

Vogel, K., B. Dien, H. Jung, M. Casler, S. Masterson, and R. Mitchell. 2011. Quantifying actual and theoretical ethanol yields for switchgrass strains using NIRS analyses. BioEnergy Research 4:96–110.

Vollmann, J., and C. Eynck, C. 2015. Camelina as a sustainable oilseed crop: Contributions of plant breeding and genetic engineering. Biotechnology Journal 10:525–535.

Waltz, E. 2015. USDA approves next-generation GM potato. Nature Biotechnology 33:12–13.

Wang, C., S.-H. Yoon, H.-J. Jang, Y.-R. Chung, J.-Y. Kim, E.-S. Choi, and S.-W. Kim. 2011. Metabolic engineering of Escherichia coli for α-farnesene production. Metabolic Engineering 13:648–655.

Wang, H., U. Avci, J. Nakashima, M.G. Hahn, F. Chen, and R.A. Dixon. 2010. Mutation of WRKY transcription factors initiates pith secondary wall formation and increases stem biomass in dicotyledonous plants. Proceedings of the National Academy of Sciences of the United States of America 107:22338–22343.

Wang, J., J. Sun, J. Miao, J. Guo, Z. Shi, M. He, Y. Chen, X. Zhao, B. Li, F. Han, Y. Tong, and Z. Li. 2013. A phosphate starvation response regulator Ta-PHR1 is involved in phosphate signalling and increases grain yield in wheat. Annals of Botany 111:1139–1153.

Wang, X., Y. Wang, J. Tian, B.L. Lim, X. Yan, and H. Liao. 2009. Overexpressing AtPAP15 enhances phosphorus efficiency in soybean. Plant Physiology 151:233–240.

Weber, A. 2014. C₄ Photosynthesis—A Target for Genome Engineering. Presentation to the National Academy of Sciences’ Committee on Genetically Engineered Crops: Past Experience and Future Prospects, December 10, Washington, DC.

Welch, R.M., and R.D. Graham. 2005. Agriculture: The real nexus for enhancing bioavailable micronutrients in food crops. Journal of Trace Elements in Medicine and Biology 18:299–307.

Welch, R.M., W.A. House, S. Beebe, and Z. Cheng. 2000. Genetic selection for enhanced bioavailable levels of iron in bean (Phaseolus vulgaris L.) seeds. Journal of Agricultural and Food Chemistry 48:3576–3580.

White, P.J., and M.R. Broadley. 2005. Biofortifying crops with essential mineral elements. Trends in Plant Science 10:586–593.

Whyard, S. 2015. Insecticidal RNA, the long and short of it. Science 347:950–951.

Witwer, K.W., M.A. McAlexander, S.E. Queen, and R.J. Adams. 2013. Real-time quantitative PCR and droplet digital PCR for plant miRNAs in mammalian blood provide little evidence for general uptake of dietary miRNAs: Limited evidence for general uptake of dietary plant xenomiRs. RNA Biology 10:1080–1086.

Wu, G., M. Truksa, N. Datla, P. Vrinten, J. Bauer, T. Zank, P. Cirpus, E. Heinz, and X. Qiu. 2005. Stepwise engineering to produce high yields of very long-chain polyunsaturated fatty acids in plants. Nature Biotechnology 23:1013–1017.

Wu, G., F.-J. Chen, F. Ge, and Y.-C. Sun. 2007. Effects of elevated carbon dioxide on the growth and foliar chemistry of transgenic Bt cotton. Journal of Integrative Plant Biology 49:1361–1369.

Yamamura, C., E. Mizutani, K. Okada, H. Nakagawa, S. Fukushima, A. Tanaka, S. Maeda, T. Kamakura, H. Yamane, H. Takatsuji, and M. Mori. 2015. Diterpenoid phytoalexin factor, a bHLH transcription factor, plays a central role in the biosynthesis of diterpenoid phytoalexins in rice. The Plant Journal 84:1100–1113.

Yang, F., P. Mitra, L. Zhang, L. Prak, Y. Vehertbruggen, J.-S. Kim, L. Sun, K. Zheng, K. Tang, M. Auer, H. Scheller, and D. Loque. 2013. Engineering secondary cell wall formation in plants. Plant Biotechnology Journal 11:325–335.

Yogendra, K.L., A. Kumar, K. Sarkar, Y. Li, D. Pushpa, K.A. Mosa, R. Duggavathi, and A.C. Kushalappa. 2015. Transcription factor StWRKY1 regulates phenylpropanoid metabolites conferring late blight resistance in potato. Journal of Experimental Botany 66:7377–7389.

Yousuf, B., K. Gul, A.A. Wani, and P. Singh. 2015. Health benefits of anthocyanins and their encapsulation for potential use in food systems: A review. Critical Reviews in Food Science and Nutrition 56:2223–2230.

Yudelman, M., A. Ratta, and D. Nygaard. 1998. Pest Management and Food Production: Looking at the Future. Washington, DC: International Food Policy Research Institute.

Yue, B., W. Xue, L. Xiong, X. Yu, L. Luo, K. Cui, D. Jin, Y. Xing, and Q. Zhang. 2006. Genetic basis of drought resistance at reproductive stage in rice: Separation of drought tolerance from drought avoidance. Genetics 172:1213–1228.

Zhang, L., D. Hou, X. Chen, D. Li, L. Zhu, Y. Zhang, J. Li, Z. Bian, X. Liang, X. Cai, and Y. Yin. 2012. Exogenous plant MIR168a specifically targets mammalian LDLRAP1: Evidence of cross-kingdom regulation by microRNA. Cell Research 22:107–126.

Zhang, B., A. Oakes, A. Newhouse, K. Baier, C. Maynard, and W. Powell. 2013. A threshold level of oxalate oxidase transgene expression reduces Cryphonectria parasitica-induced necrosis in a transgenic American chestnut (Castanea dentata) leaf bioassay. Transgenic Research 22:973–982.

Zhang, J., S.A. Khan, C. Hasse, S. Ruf, D.G. Heckel, and R. Bock. 2015. Full crop protection from an insect pest by expression of long double-stranded RNAs in plastids. Science 347:991–994.

Zhou, R., J. Nakashima, L. Jackson, G. Shadle, S. Temple, F. Chen, and R.A. Dixon. 2010. Distinct cinnamoyl CoA reductases involved in parallel routes to lignin in Medicago truncatula. Proceedings of the National Academy of Sciences of the United States of America 107:17803–17808.

Zimmerman, M.B., and R.F. Hurrell. 2002. Improving iron, zinc and vitamin A nutrition through plant biotechnology. Current Opinion in Biotechnology 13:142–145.

This page intentionally left blank.