Innovations continuously enhance crop breeding and demand new strategic planning

Global Food Security 12 (2017) 15–21

Contents lists available at ScienceDirect

Global Food Security journal homepage: www.elsevier.com/locate/gfs

Innovations continuously enhance crop breeding and demand new strategic planning

MARK

Richard B. Flavell Ceres, Inc., Thousand Oaks, CA 91320, USA

A R T I C L E I N F O

A BS T RAC T

Keywords: Crop improvement Genomics Microbes Genome editing Strategic leadership and planning

Food security relies on continuous supplies of improved products from plant breeding and their assimilation into agriculture. Extraordinary innovations in the life sciences have brought plant breeding into a new phase of opportunity. These include the means to discover, manage and select better versus poorer versions of genes and the ability to change gene sequences in situ by gene editing. Genomics is also revealing the thousands of different microbes in all plants and the roles that their genomes play in determining crop traits that can be further improved by addition of the right microbes. Assimilation of such innovations into breeding strategies can have major impacts on rates of breeding gain but to achieve this will require comprehensive strategic leadership, planning and investments by scientists, leading global agencies and all governments based on appreciation of (i) the continuous streams of innovations underpinning crop improvement and (ii) the necessities for more rapid crop improvements everywhere to help avert otherwise inevitable catastrophes.

1. Introduction The annual rate of gain of production in farmers’ fields has slowed for major cereal crops in recent years (Fisher and Edmeades, 2010; Fisher et al., 2009). This has added to the global food security concerns. World food production must increase substantially to avoid catastrophies and the breeding of new crop varieties must play its part (Tilman et al., 2011). This puts a huge responsibility on plant breeders. What limits faster rates of gain via plant breeding? The many constraints include deployment of new genetic variation from the crop or another species; the ability to make commercially improved genotypes much faster and cheaper; the ability to recognize superior genotypes more easily; the organization of plant breeding resources and the movement of germplasm efficiently around the world; better cost input versus output ratios and preferences of consumers. Plant breeding is based on bringing together different sets of chromosomes and the reassortment of genes into different combinations during gamete formation. The best combinations can then be selected amongst the offspring in relevant environments. The processes involve the constituents of all chromosomes comprising a huge number of the informational units created and selected during evolution, domestication and modern breeding. Therefore it is necessary to have complete genome sequences, or their proxy, of large numbers of strains of every crop to both interpret what has been going on in plant breeding and selection and to design/select preferred forms of plants. Fortunately, there have been remarkable innovations over the past 40 years in genomics, coming from huge investments across the life sciences, that enable new understanding of what chromosomal inforhttp://dx.doi.org/10.1016/j.gfs.2016.10.001 Received 2 June 2016; Received in revised form 1 October 2016; Accepted 4 October 2016 2211-9124/ © 2016 Published by Elsevier B.V.

mation determines plant properties. The application of genome sequencing to plants has also revealed the presence of thousands of bacterial and fungal strains that live inside plants as endophytes. These discoveries are radically changing our views on what a plant is and how its properties are determined. Finally, breakthroughs in changing individual gene sequences in chromosomes to preferred designs have emerged that have extraordinary potential in plant breeding when preferred forms of genes are known (National Academy, 2016). Consequently, a new nexus between genomics, genetics and breeding technologies has arrived that promises increased efficiency in crop improvement for future food security. This is therefore a special era in the history of plant breeding and reminds us that the future will not be like the past or the present. However, the evolving opportunities demand new investment and planning in breeding, regulations, consumer awareness and politics to avoid failure, disappointment and unnecessary controversy. This paper draws attention to some of the major current and future innovations in plant improvement and key global challenges for deploying them. Comprehensive innovation-driven, strategic planning and leadership by international bodies and governments will be essential for the innovations to play their part in achieving increased food production. 2. Innovations bringing together genomics and breeding, 1975–2015 Innovations from conception to full deployment can sometimes take several decades. To illustrate one forty-year path of innovations

Global Food Security 12 (2017) 15–21

R.B. Flavell

not be wasted growing plants to be thrown away. More resources can then be spent on finding even better traits. This should significantly reduce a previous large problem when handling many genetically complex traits. Finally, a more-or-less complete DNA sequence from a hexaploid wheat accession has been produced using advanced sequencing and bioinformatics (Brenchly et al., 2012; Mayer et al., 2014). This reference sequence is a landmark that will, in time, make an enormous difference to wheat breeding, especially when variants found in a range of individuals are mapped upon it (Huang et al., 2013) and knowledge of the diversity in defined chromosomal segments is understood. In summary, it has taken some 40 years of innovations from defining the first DNA markers to the routine application of DNA polymorphisms for selection of genetically complex traits (Eathington et al., 2007; Fig. 1). This is not only because enough markers were unavailable until recently but mainly because methods were too expensive and statistically uncertain to apply routinely in breeding programs by applied breeders. While huge cost reductions have occurred in recent years, cost/low investment is still a major reason why marker assisted breeding is not widely used in small breeding programs. To achieve the recent reductions in costs and allow applications at the required scale it has taken many millions of dollars, many entrepreneurs and small startup companies with an eye to profitability from the human genetics/medical world to provide the means of using genomics for applied plant breeding. In fact, nearly all the landmark innovations that have created what we know as plant genomics have been based on innovations outside plant sciences and agriculture. This will continue. Thus it should be recognized that future developments in applied plant breeding will be dependent on discoveries and investments elsewhere, especially in medical sciences. Given that these will surely emerge in the richer countries we should look at projections of funding and innovation strategies in the medical and IT worlds to see the future opportunities for plant breeding based on high throughput genetic and genomic analysis beyond those noted in Fig. 1. The frontiers are in an exciting, very different and powerful position compared with ten years ago. Also, the innovations in genomics are being complemented by innovations in the means of measuring traits and the performance of plants in fields by drones equipped with cameras able to detect different properties of plants (Pieruschka and Poorter, 2012).

Fig. 1. Initial adoptions of genomics-based technologies into plant breeding over the past 40 years. The technologies that were assimilated into leading plant breeding programs are positioned in time approximately as they were initially adopted. This does not mean that they have been widely adopted. Costs as well as other factors have commonly resulted in slow rates of adoption by most plant breeding organizations.

that has finally come to have major impact on applied plant breeding I use the development of genomics-based breeding in wheat from the 1970s to the present, (Fig. 1). Many diverse innovations had to be integrated. Studies into the isolated DNAs that comprise plant genomes started extensively in the 1970s (e.g. Rimpau et al., 1978). This time can be considered the beginnings of crop plant genomics. The innovations that resulted in being able to clone DNA matured in the late 1970s (Gerlach and Bedbrook, 1979; Bedbrook et al., 1980) and launched the search to clone plant genes and so identify their DNA sequences. By the early 1980s, copies of plant genes had been sequenced and published (e.g Forde et al., 1981; Smith et al., 1983). Many scientists were motivated to start defining plant genes in the belief that it would help plant breeding but the way this would happen was not clear. It was obvious that it would be possible to define genetic variation in DNA but not how this could be used directly by breeders at scale.. Large numbers of polymorphisms in DNA sequences between different plants revealed by treatments with restriction enzymes, or differences in the lengths of tandemly organised DNA sequences due to variation in numbers of copies of repetitive units (microsatellite DNAs) were gradually discovered from the late 1970s and mapped extensively using segregating populations during the early-mid 1990s (Röder et al., 1998; Somers et al., 2004). Geneticists used molecular marker maps from the 1990s but breeders only for specialist uses where known polymorphisms were associated with known phenotypes. Disease resistances, encoded by single, mapped alleles, were ideal candidates to be selected in breeding programs based on genetically linked DNA polymorphisms (Li et al., 2015). It took high throughput, acceptably cheap methods to assay large numbers of polymorphisms to bring use of whole genome mapping and the discovery of better alleles into a broad-based breeding context. By 2007 Monsanto had reported that marker assisted breeding had become their conventional norm and increased the mean performance of progeny compared to conventional breeding methodologies (Eathington et al., 2007; Mammadov et al., 2012). Then genome-wide association studies (GWAS) between segregating markers associated with better alleles and phenotypes could be made and documented for elite populations (Huang and Han, 2014; Sukumaran et al., 2015). From this “Genomic Selection”, i.e. the selection of plants based on the markers (better alleles) for all chromosome segments found to be co-inherited with a trait in segregating populations, promises to be a huge step forward in efficiency when applied appropriately in known populations (Heslot et al., 2012). It will enable breeders to select seeds that have desired traits without growing the whole plant. Thus what breeders evaluate in the field will be enriched for key traits and so land, time and money will

3. Innovations in trait improvement by crop genome editing A “holy grail” for breeders has long been the wish to be able to replace alleles ( specific variant forms of a gene) in situ with better alleles for specific traits. The need to do this remains and will be increased by the need to adjust existing crops where increasing temperature, drought, cold, new insects and diseases occur due to climate change (Jaggard, Qi and Ober,2010). If advances in genomics coupled with phenotyping are giving us the means of identifying better alleles, how can the alleles be assembled more effectively? Where favourable alleles are known, this has been routinely achieved by backcrossing but this process is lengthy and still suffers from deleterious alleles linked to the favourable genes that cannot be easily separated by recombination. In the last few years the challenge of modifying defined genes to alternative alleles in situ has leapt ahead with the developments of truly ground-breaking technologies that have changed plant science and breeding opportunities (Kumar and Jain, 2014; Voytas, 2013; Ran, 2014). There are various versions but all are based on deliberate cleavage of chromosomes at specific target sites and then modification of the DNA during repair of the broken sites. The different techniques all enable gene-specific “editing”. The changes are made in cells from which whole plants are regenerated, as in the initial steps of making transgenic plants. The preferred method of cleaving DNA and modifying a gene sequence is now the CRISPR/cas9 system emanating from Clustered 16

Global Food Security 12 (2017) 15–21

R.B. Flavell

nucleases to change specific genes is dependent on knowing the genomic sequences in and around genes. This was not so easily determined before the major achievements of sequencing complete genomes that have emerged over the past twenty years. Thus not only was the targeting of nucleases to specific genes unknown when transgenic biology took off in plants but also if the targeting systems had been known the molecular genetic information to exploit them was unavailable. 4. Innovations for finding significant alleles to change by “Genome Editing” The achievements of determining complete genome sequences for the major crops (http://www.phytozome.net; over 50 species listed) enable annotations of all the genes and their potential regulatory sequences, though this remains challenging, especially for the potential regulatory sequences. The development of genome descriptions of all the genes and how to change them are somewhat ahead of knowing a large catalog of genes which need to be selected, inactivated, modified in novel ways or replaced with new alleles to make major crop yield gains. However, with genomics-based breeding linked to assaying traits this major deficiency can be eroded. Future studies on the genes around the Quantitative trait loci (QTL) and molecular DNA markers will increasingly reveal the genes involved and the genetic variation to be exploited in targeted genetic changes. Surveys of recent journals show associations between defined alleles and traits are growing exponentially for the major crops. Databases of alleles and combinations of alleles that change traits will be the fountain of many breeding editing approaches in the future. However, it should be noted that while commercially ready varieties clearly have many important QTLs incorporated, many of the QTLs described in the literature from academic studies do not provide further gains in elite crops because either the QTLs have been already incorporated or genetically alternative routes to trait improvement have been selected. However, once gene-trait links are clearly established it may be possible to generate much new variation in elite germplasm by gene editing at these gene loci to find novel variants that enhance elite varieties. To develop such knowledge bases will take a long time. While this is a major challenge it is one that lies on the “holy grail” of plant improvement and so is worthy of the long investment. Some examples of where elimination of functional copies of crop genes would be useful include removal of ricin toxins from castor bean, anti-nutritionals such as trypsin inhibitors from soybean, allergenic proteins from nuts and cereals and removal of the pathway that creates bruising discoloration in fruits (Voytas, 2013). Many others exist, but much needs to be discovered for more complex traits where multiple genes may be required to be changed. Large catalogs of knockout mutants have been established in Arabidopsis (The Arabidopsis Information Resource, TAIR) and rice and screened for many traits. Many favourable gene-trait associations have been established. This catalog becomes additionally useful now that genome editing is established. Because model species are not routinely relevant to crops new approaches have emerged to discover the phenotypic effects of knockout mutations in any gene in any species. To achieve this TILLING (Targeting Induced Local Lesions in Genomes) technology was devised (McCallum et al., 2000; Comai et al., 2004). It allows directed identification of mutations in any specific gene and has since been used to find mutant genes in corn, wheat, rice, soybean, tomato, sorghum and lettuce. Alternatively, starting with plants that have a desired phenotype it is possible to find the mutant gene responsible. This is particularly useful in a breeding context. An updated system has been set up in wheat (Henry et al., 2014) in which lines of wheat have been created carrying ethyl methanesulfonate-induced mutations in every gene. Hybridization chips have been made that carry every gene in wheat based on near-complete DNA sequencing. When DNA from the heavily mutated plant is hybridized to the chip it is possible to

Fig. 2. The CRISPR/cas9 system for changing any chromosomal DNA sequence.. The Cas9 (CRISPR-associated, 9) nuclease protein (grey/blue) is localized in the cell to a given chromosomal sequence by the guide RNA, supplied from the laboratory, that contains a 20 base sequence complementary to the plant chromosomal target DNA sequence to be modified. The sites cleaved by the Cas9 nuclease are given by the red crosses. Once cleaved repair can occur creating mutations in the target gene or the target gene can be modified to a different sequence by recombination with another DNA molecule supplied experimentally. More details are given in Kumar and Jain (2014) and Ran (2014).

Regularly Interspaced Short Palindromic Repeats in bacteria and DNA cleavage by the cas9 nuclease (Doudna and Charpentier,2014; Ran, 2014; Kumar and Jain, 2014). The DNA nuclease is guided to the targeted gene by a complementary short RNA (Fig. 2) that is easily made in the laboratory from a genome sequence (Svitashev et al., 2015). Because targeting is achieved simply through RNA/DNA base pairing, CRISPR/cas9 has emerged as the system of choice. It is much easier to use, cheaper and much better suited to multiplex gene targeting with greater efficiencies than previous methods using TALENs or Zinc Fingers (Doudna and Charpentier, 2014). It is noteworthy that the arrival of CRISPR/cas9 has taken thirty years of development since yeast gene replacement was achieved in 1979 (Doudna and Charpentier, 2014).. After DNA cleavage by any of these technical approaches the cell's own DNA repair mechanisms work on the break. The repair systems, which join non homologous chromosome ends together but are error prone, are found in all organisms. The errors in the repair lead to loss of nucleotides or variations in the repaired DNA sequence. This then leads to loss of the protein, generation of an altered protein or altered regulation if the DNA being repaired has regulatory functions. Additionally, and importantly, if extra copies of a modified gene are supplied at the time of cleavage (by inserting the gene into the cells at the same time as the CRISPR/cas9 molecules), then insertion of this new copy can occur by homologous recombination while repair is taking place (Svitashev et al., 2015). The CRISPR/cas9 technology has been used successfully to achieve defined gene changes in Arabidopsis, Nicotiana benthamiana, rice, wheat, corn, canola, orange and sorghum (Kumar and Jain, 2014; Svitashev et al., 2015; Li et al., 2013; Shan et al., 2013; Wang et al., 2014a; 2014b). Products with changes in one, two or more homologous genes can be recovered from genome editing. One of the early examples in plants of gene editing resulted in nonfunctional mutations in wheat at all six recessive genes determining sensitivity to a fungus, so making the plant resistant to the fungus (Wang et al., 2014a; 2014b). The example illustrated the power of the technique to change several loci in one experiment, something that would be essentially impossible by conventional breeding. When this does not occur then homologous mutations recovered separately can be stacked by breeding to create the full set of null alleles. It is most important to note that the approach of using targeted

17

Global Food Security 12 (2017) 15–21

R.B. Flavell

cultivars. The microbiomes inside plants and their offspring seeds are modified each generation by microbes taken up from the soil. Their propagation through the plant is dependent on the host plant and microbe-microbe interactions. The populations of microbial genomes are dynamic, unlike the genomes of a plant. The results of plant– microbe and microbe-microbe interactions are consequently dynamic. It will be complex to sort out the genetics of consortia of microbes inside crops. The metagenomics studies on large populations of endophytes in plants has begun (Sessitsch et al., 2012; Lundberg et al., 2012) but will grow tremendously over the coming years, given that development of the molecular biology of microbes is tractable. Many of the microbes cannot be cultured on defined media and may not have free-living forms. Others are readily cultured. It is facile to get a plant to take up bacteria and fungi providing that they are compatible with the plant and microbial biology established through evolution and selection. There are many reports that show that the effects of microbes on plants are related to the consortia of microbes present (Turner et al., 2013; Porras-Alfaro and Bayman, 2011). However, there are some single bacteria and fungi that have major effects on plant growth when added to consortia in and around seeds. Thus large discovery projects to find strains that provide new or enhanced traits to our crop plants coupled with the means of adding them to seeds, soils or other ways are underway to provide a new and exciting addition to plant breeding. Many of the principles involved are well understood from agricultural application of rhizobia to enhance nitrogen fixation. Microbiomes may turn out to be large sources of novel genetic information available to breeders and their deployment should not suffer from many of the issues associated with targeted genetic changes to plant chromosomes. The fact that microbes can be added to elite cultivars and do not recombine with plant genomes means that this genetic variation can be exploited more easily. The applications may be very suitable for developing countries with poorer soils to adopt relatively rapidly and new industries based on microbes will become sources of such innovations. Such innovations should be built into projections of crop performance and agronomy, noting that their deployment may mean the addition of different microbe variants to crops and food. Thus research programs involving the isolation of endophytes from soils and crops growing in various conditions that influence microbiome composition are being developed. Testing for efficacy is reasonably straightforward, assuming that a relevant assay is available by which to screen large numbers of microbes. Microbes suitable for routine use in agriculture also need to be stable during storage and delivery and easily applied to seeds. This is another area of important research to realize the value of plant–microbe relations. No doubt regulatory agencies, farmers and the general public will need to be satisfied that the microbes, some of which may be closely related to pathogens, are not pathogenic to crops, wild organisms or humans.

select which genes have been mutated. These innovations could not have been done rapidly for any gene without the full set of DNA sequences or the advanced chip technologies to find the mutated genes. Thus the sequencing of complete genomes and chip technologies built up over the past decade have revolutionized the ability to build traitmutation associations. Catalogs of trait-mutation associations combined with genome editing technologies will provide a powerful basis for plant improvement, especially when the mutant lines are made and screened in elite genotypes. There are large catalogs of transgenes that have resulted in increases or decreases in gene expression and improved traits (James, 2014). These are also useful sources of knowledge of which genes to modify, given that gene edits to achieve the same ends with endogenous genes can be achieved. They include new stacks of herbicide and insect resistance genes, nutritional enhancements, nematode resistance, Asian soybean rust resistance, drought tolerance, nitrogen use efficiency and other stress tolerances. Also numerous transgenes conferring resistance against specific fungal and bacterial diseases are already known and surely modifications of endogenous genes conferring disease susceptibility to forms that confer disease resistance will become early candidates for modification by gene editing, given the huge cost of managing diseases (Jones et al., 2014; Dangl et al., 2013). Such modifications would be extremely valuable for developing countries, reducing costs and providing better yield stabilities. However, it has to be recognized that the huge investment into deployment of transgenes into corn, for example, in the USA has not realized large consistent improvements across hybrids, although the genes clearly have the potential to augment traits in other plant species such as Arabidopsis or rice (Flavell, 2015). This may also be due to different hybrids having become optimized by different genetic routes for complex traits and single genes are therefore unable to provide the preferred universal improvements across different hybrids. More knowledge needs to be gained on such hypotheses. Here again deploying different variants by gene editing approaches, rather than a single transgene with a single regulatory sequence may be more rewarding. In summary there needs to be much more knowledge a gained on the relationships between allele structure and function and traits, especially complex traits, before gene editing becomes a major contributor to plant breeding, but the potential for directly improving already elite plants is enormous. 5. Innovations from the addition of targeted endophytes to modify traits Genes that influence/regulate plant traits are not only in plant chromosomes. Millions of microbes, comprised of many taxa, families and strains live within every plant in apparent mutualistic harmony as endophytes and many plant traits are determined in part by interactions between plant-endophyte and endophyte-endophyte genomes. (Hallmann et al., 1997; Turner, et al., 2013; Porras-Alfaro and Bayman, 2011; Gaiero et al., 2013). Thus these endophytes need to be considered as part of the functional plant genome, a concept new to plant breeding programs today. The microbes associated with plants are known to aid germination, facilitate uptake of key metabolites from the soil, provide plant hormones to modulate plant development, to suppress pathogens, provide additional tolerance to stresses of many kinds and many other benefits (Hallmann et al., 1997; Gaiero et al., 2013; Zuniga et al., 2012; Poupin et al., 2013; Naveed et al., 2014). The Burkholdaria phytofirmans strain PsJN is a prominent example and appears to exert its effects across many plant species and growing conditions (Zuniga et al., 2012; Poupin et al., 2013; Naveed et al., 2014). It promotes growth, changes in flowering time, enhances drought resistance and many other features throughout the life of the plant. From surveys of the literature, it is probably sound to assume that critical traits in all plants can be modified by changing the composition of the endogenous endophyte communities even in elite

6. Innovations create new visions for plant improvement The foregoing on just three major areas of innovations- gene identification and management based on genomics, gene editing and deployment of endophytes- has illustrated that innovations devised over decades, in plant and non-plant research settings, have finally reached points of integration where they can transform a whole process, such as plant breeding, assuming adequate investment. Innovations will continue and as they become integrated will make crop breeding more efficient. Such innovations should be built into global planning scenarios. We can now imagine that in years to come hundreds of genes and chromosome structures will have been optimized by genomics-based approaches. This is an extremely important point. The approaches are very much the holistic, preferred scientific future for breeding every crop. The extent to which they are adopted will determine whether or how new innovations will be assimilated broadly into plant improvement over many decades to come. The 18

Global Food Security 12 (2017) 15–21

R.B. Flavell

using additional genes, especially for genetically complex traits is challenging (Flavell, 2015). The last 20 years of transgenic crops have also taught us that more than technological improvements and sciencebased regulations are necessary (Haslberger, 2000; Herring, 2008; Chassy, 2007; Lemaux, 2008, 2009). Societies and governments must be supportive of the changes and their value. We have also learned that many lies are spread by anti-science groups to kill scientific innovations and therefore access to and propagation of authoritative information that can be distinguished from the vast quantities of anti-scientific material on the world wide web is vital (Lynas, 2013). Given the technical and non-technical factors there are many strategic, organizational, regulatory and political issues to be debated and resolved. These are massive in complexity and global diversity. I outline here some of the more important general conclusions. They are inspirational, organizational, regulatory, political and strategic issues because I believe these are most important, indeed essential to get right, especially in developing countries. A major reason that transgenic products of sound quality emerged so rapidly in the 1990 s is because scientific leaders created the right vision, comprehensive strategies to fulfil the vision and invested time and resources to overcome most of the roadblocks. They took risks because the prize was big. It was a few companies, small and large, who created and delivered the products, not public breeding programs serving the rich or poor, nor the CGIAR. To equip themselves with the right skills, the large companies reoriented themselves internally, bought smaller companies and created strategic alliances thereby repositioning themselves to create and bring such products to market. They competed and protected investments by IP filings and opened up new issues at patent offices. They pushed for new regulations to be made so that they could sell officially deregulated products, helped establish the safety criteria that should be associated with such products and delivered the necessary documentation. In summary, a committed, comprehensive approach was necessary to bring new kinds of products to market. Similar targeted and holistic approaches based on the new innovations will be needed to aid food security goals. Will it be left again to the private sector to create the advances or can the public sector breeders take the lead? “Scientific business as usual” will fail. Therefore “business as usual” attitudes and plans should be rejected because they fail to recognize the potential extraordinary value of innovations. The processes of genome editing and all the developments that will occur based on these approaches are so fundamental to future plant breeding that they need to be embedded in a new, agreed vision for plant improvement, focused on meeting the goals of “food for all” and based on the widely held beliefs of a moral imperative (Flavell, 2016). Inclusive processes for deliberating on and providing adequate societal oversight of risks, trade-offs and opportunity costs are needed. They must involve everyday people-not only scientists and companies. The United Nations Food and Agriculture Organization, the Organization for Economic Cooperation and Development and the like need to take the lead as they did with the International Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD) whose report now needs updating with the arrival of gene editing (Scoones, 2009). But more than this it will need national and international leadership to enable plant breeding to help address the potential food shortages. All the push backs against transgenic crops have seriously wounded plant science and plant breeding worldwide. Plant science has been vilified in many places. Innovations into transgenic food and transgenic plant breeding have been reduced or stopped, including in Europe, because of reduced investments into approaches that embraced the new technologies (Lynas, 2013). In many places societies have become mobilized against genetic innovations in breeding. It is therefore really important that genome editing for plants and the addition of microbes are debated not just as technologies but in the context of plant improvement and food and nutrition security advances for all. Misunderstandings about plant breeding and its products must be

Fig. 3. Future hypothetical yield gains from new innovations in genomics. The hypothetical graph predicts that many major innovations (blue arrows) will be made that help breeders to produce crops with higher yields in agriculture. Innovation 2 is dependent on Innovation 1 and Innovation 3 is dependent on Innovation 2. These innovations could include those discussed in this paper, in addition to manipulating recombination rates and positions, photosynthesis, nitrogen use efficiency, heterosis, harvest index, disease resistance and many others. The exponential nature of the curve mimics what is frequently observed following major innovations, where one enables the faster deployment of the next (series).

scientific innovations being discussed here are but a few of many that can and should impact plant breeding over the coming decades beyond the known technologies depicted in Fig. 1. Consider the scenario in Fig. 3. Innovations often develop along an exponential curve because new innovations are enabled by previous ones. If this happens and societies benefit from a more efficient development of safe crops, then it is likely that global food and nutritional security can be realized with better local stabilities, reduced poverty and insulation from the price increases that inevitably arise when there are food shortages. However, if we do not assimilate efficiencies available in plant breeding and not proceed up the hypothetical curve in Fig. 3, societies will then have a much higher probability of having failing agricultural systems. Thus adoption of a new innovation is not only important for the innovation itself but also for all the value to be gained from the subsequent innovations that are enabled by the first ones. The issues associated with the adoptions of innovations are much more crucial for the poorer countries and that is why they need to take more independent decisions and be masters of their own food destinies. The situations over transgenics in Africa and Asia (Paarlberg, 2010) surely should not be repeated with genome editing. While individuals should be able to choose what kinds of food they eat, the laws should not prevent choice and deny citizens solutions that can increase the probability of better use of land and the other resources of the planet and environmental services. In the immediate future issues relating to regulations over genome editing are critical (Kershon and Parrott, 2014; Lussa and Davies, 2013). Much has been lost due to confusion over what transgenes, plant genetics and plant breeding entail and the resulting fears and stifling regulations (Chassy, 2007; Hartung and Schieman, 2014).. 7. Ways forward Scientific breakthroughs have once again brought us to a new vision for plant improvement just as they did in 1982 when the first transgenic plants were made. What have we learnt from the development of transgenic crops that can help us with implementing the new vision of plant breeding, especially for applications in developing countries? Scientifically, we have learned that improving elite varieties 19

Global Food Security 12 (2017) 15–21

R.B. Flavell

produce products for sale, export and import in all countries. One only hopes that those who are responsible for the regulations around the world and especially in Africa and Asia will, this time around, grasp the vision of what science can do over the coming years, as depicted in Fig. 3, and the consequences of not going up the innovation curve for each crop and country.

corrected (Royal Society, 2016). There is too much at stake for myths to drive policies. The ethical issues associated with innovations in plant breeding also need to be assessed against the background of what happens if societies fail to produce food that relieves poverty and ill health and crops do not cope with diseases and climate changes. How countries and the international communities respond now to the new scenarios of genome editing and use of microbiomes in plant breeding is crucial. The responsibility on plant breeders is enormous given the potential catastrophies if food insufficiencies accumulate around the world and current investments are far too low for most crops. Therefore the processes of crop improvement and its products need to be championed at the highest levels around the world. Scientific and institutional leaders should clearly see what the new innovations mean for the short, medium and long terms and programs should be restructured and alliances optimized to deliver outputs that meet the right business, societal and environmental criteria. The vision needs to be developed internationally in a harmonized way. International leaders need to embrace it based on the increased wealth, food and nutrition security, health and environmental gains that implementation will generate. Science and business plans need to be based on better product-relevant strategies, not on the way things have been done before the scientific innovations were made. The public sector institutions should adopt the focus, planning, urgency and efficiencies of the best private sector companies (Delmer, 2005). Almost no public sector organization has deregulated a transgenic product from its own discovery program in spite of a huge research base. That is shocking and not to be repeated, surely. Major national agricultural programs, especially in developing countries, and the CGIAR must shoulder responsibilities for establishing the vision and viable strategies in ways that they failed to do with transgenics. Strategic collaborations will be essential. The private sector should be encouraged to feed the world as companies and in public-private sector partnerships. Planning to achieve a faster rate of gain in plant breeding must be done locally and globally for all the critical crops. Fully integrated planning should decide what products will bring the most benefits to customers and national and international societies. Without national or global plans we will continue to be in chaos like over Golden Rice (Goldenrice.org) where science and societies ended up on different sides of whether GMOs should be introduced into societies, Which results in everyone losing and people continuing to suffer malnutrition and ill health unnecessarily.

9. Summary What increased rates of gain in crop yields are attained from new innovations and what improved traits emerge as deployable will depend on the value of the trait, costs, regulations and consumer acceptance, size of market, profitability, public and market acceptance and export potential. All these are critical and need to be assessed better by public institutions as well as industrial companies. However, a failure to embrace today's technologies and anticipate future opportunities to solve food, feed, fiber and fuel problems safely and sustainably by plant biotechnology will surely only increase the probability of human suffering and misery. Today's crop breeding is slow, laborious and complex. It needs to be speeded up. There is now perhaps a window of opportunity to work more closely with publics all over the world to enable scientific innovation to continue to serve the world better in the crucial matters of food security and the best use of land. However, strategic planning and leadership at global and local levels are urgent to find and achieve the best ways forward. The links between innovation and food security need to be fully recognized and built into global and local plannings. Funding The writing of this article did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Bedbrook, J.R., Jones, J.D.G., O’dell, M., Thompson, R.D., Flavell, R.B., 1980. A molecular description of telomeric heterochromatin in Secale species. Cell 19, 545–560. Brenchly, et al., 2012. Analysis of the bread wheat genome using whole-genome shotgun sequencing. Nature 491, 705–710. Chassy, B.M., 2007. The history and future of GMOs in food and agriculture. Cereal Foods World 52, 169–172. Comai, L., Young, K., Till, B.J., Reynolds, S.H., Greene, E.A., Codomo, C.A., Enns, L.C., Johnson, J.E., Burtner, C., Odden, A.R., Henikoff, S., 2004. Efficient discovery of DNA polymorphisms in natural populations by Ecotilling. Plant J. 37 (5), 778–786. Dangl, J., Horvath, D.M., Staskawicz, B., 2013. Pivoting the plant immune system from dissection to deployment. Science. http://dx.doi.org/10.1126/science.1236011. Delmer, D.P., 2005. Agriculture in the developing world: connecting innovations in plant research to downstream applications. Proc. Nat. Acad. Sci. USA 102, 15739–15746. Doudna, J.A., Charpentier, E., 2014. The new frontiers of genome engineering with CRISPR-Cas9. Science 346, 1077–1086. Eaglesham, A., Hardy, R.W.F., 2014. Overview and summary of NABC 26. New DNA editing approaches: methods, applications and policy for agriculture. NABC 26, 3–22. Eathington, S.R., et al., 2007. Molecular markers in a commercial breeding program. Crop Sci. 47, S154–S164. Fischer, R.A., Byerlee, D., Edmeades, G.O., 2009. Can Technology Deliver on the Yield Challenge to 2050? FAO.org. Fisher, R.A., Edmeades, G.O., 2010. Breeding and cereal yield progress. Crop Sci.. http:// dx.doi.org/10.2135/cropsci 2009.10.0564. Flavell, R.B., 2015. Crop Improvement Using Transgenes, Genome Editing and Microbes: A Forward-looking Essay to Celebrate 20 Years of Transgenic Crops, ISAAA, 51, pp. 28–49. Flavell, R.B., 2016. Greener revolutions for all. Nat. Biotechnol., (in press). Forde, B.G., Kreis, M., Bahramian, M.B., Matthews, J.A., Miflin, B.J., Thompson, R.D., Bartels, D., Flavell, R.B., 1981. Molecular cloning and analysis of cDNA sequences derived from polyA+ RNA from barley endosperm: identification of B hordein related clones. Nucleic Acid Res. 9, 6689–6708. Gaiero, J.R., McCall, C.A., Thompson, K.A., Day, N.J., Best, A.S., Dunfield, K.E., 2013. Inside the root microbiome: bacterial root endophytes and plant growth promotion. Am. J. Bot. 100, 1738–1750. Gerlach, W.L., Bedbrook, J.R., 1979. Cloning and characterization of ribosomal RNA genes from wheat and barley. Nucleic Acids Res. 7, 1869–1885. Hallmann, J., Quadt-Hallman, A., Mahaffee, W.F., Kloepper, J.W., 1997. Bacterial endophytes in agricultural crops. Can. J. Microbiol. 43, 895–914.

8. Regulations and the future of plant breeding A recent review of international laws pertaining to plant biotechnology (Kershon and Parrot, 2013) included the following “Forecasted impact of the present regulatory systems on the future of agricultural biotechnology ranges from cloudy to devastating”. Some countries are currently (re)considering their biosafety laws and regulations because they are recognized as unfit for purpose (Eaglesham and Hardy, 2014; Shearer, 2014; Schieman and Hartung, 2014, Hoffman, 2013; Kershon and Parrott, 2014). The USA and Argentina are moving to not regulate some products of genome editing, presumably because the genetic changes within them cannot be distinguished from those that could have arisen via conventional breeding and are not made using a plant pest (Whelan and Lema, 2015). Europe, however, is likely to keep them regulated because a laboratory scientist has intervened in the production of their genotype by inserting DNA into plant cells to carry out the genome editing. The Catagena Protocol on Biosafety to the Convention on biological Diversity that governs the movement of GMOs around the world is also a crucial legal instrument. It is built on the precautionary principle and allows any state to ban the import of genetically modified organisms if it believes there is inadequate evidence concerning safety. 167 member states of the United Nations are signatories to the Protocol. These sorts of laws, as they evolve, or not, around the world will have the most profound effect on plant breeding and how it can 20

Global Food Security 12 (2017) 15–21

R.B. Flavell

Porras-Alfaro, A., Bayman, P., 2011. Hidden fungi, emergent properties: endophytes and microbiomes. Ann. Rev. Phytopathol. 49, 291–315. Poupin, M.J., Timmerman, T., Vega, A., Zuriga, A., 2013. Effects of the plant growth promoting bacterium Burkholdaria phytofirmans PsJN throughout the life cycle of Arabidopsis thaliana. PLoS One 8, e69435. Ran, F.A., 2014. Tools and applications for eukaryotic genome editing. NABC 26, 69–81. Rimpau, J., Smith, D., Flavell, R.B., 1978. Sequence organisation analysis of the wheat and rye genomes by interspecies DNA/DNA hybridisation. J. Mol. Biol. 123, 327–359. Röder, M.S., Korzun, V., Wendehake, K., Plaschke, J., Tixier, M.-H., Leroy, P., Ganal, M.W., 1998. A microsatellite map of wheat. Genetics 149, 2007–2023. Royal Society, 2016. GM Plants: Questions and Answers. Royal Society, UK. Schieman, J., Hartnung, F., 2014. EU perspectives on new plant breeding techniques. NABC 26, 201–210. Scoones, I., 2009. The politics of global assessments: the case of the International Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD). J. Peasant Stud.. http://dx.doi.org/10.1080/03066150903155008. Sessitsch, A., et al., 2012. Functional characterization of an endophyte community colonizing rice roots as revealed by metagenomics analysis. Mol. Plant Microbe Interact. 25, 28–36. Shan, Q., Wang, Y., Li, J., Zhang, Y., Chen, K., Liang, Z., Zhang, K., Liu, J., Xi, J.J., Qiu, J.-L., Gao, C., 2013. Targeted genome modification of crop plants using a CRISPRCas system. Nat. Biotechnol. 31, 686–688. Shearer, H., 2014. Regulation of plants with novel traits: canadian perspectives on the “novelty” trigger. NABC 26, 193–200. Smith, S.M., Bedbrook, J.R., Speirs, J., 1983. Characterisation of three cDNA clones encoding different mRNAs for the precursor to the small subunit of wheat ribulosebisphosphate carboxylase. Nucleic Acids Res., 118719–118734. Somers, D.J., Isaac, P., Edwards, K., 2004. A high-density microsatellite consensus map for bread wheat (Triticum aestivum L.). Theor. Appl. Genet. 109, 1105–1114. Sukumaran, S., Dreisigacker, S., Chavez, M.L.P., Reynolds, M.P., 2015. Genome-wide association study for grain yield and related traits in an elite spring wheat population grown in temperate irrigated environments. 2015. Theor. Appl. Genet. 128, 353–363. Svitashev, S., Young, J.K., Schwarz, C., Gao, H., Falco, S.C., Cigan, A.M., 2015. Targeted mutagenesis, precise gene editing and site-specific gene insertion in maize using cas9 and guide RNA. Plant Physiol. 169, 931–945. Tilman, D., Balzer, C., Hill, J., Befort, B.L., 2011. Global food demand and the sustainable intensification of agriculture. Proc. Nat. Acad. Sci. 108, 20260–20264. Turner, T.R., James, E.K., Poole, P.S., 2013. The plant microbiome. Genome Biol. 14, 209–219. Voytas, D., 2013. Opportunities and regulatory challenges for genome engineering in agriculture. NABC 26, 29–37. Wang, et al., 2014b. Characterization of polyploid wheat genomic diversity using a highdensity 90 000 single nucleotide polymorphism array. Plant Biotechnol. 12, 1787–1796. Wang, Y., Cheng, X., Shan, Q., Zhag, Y., Liu, J., Gao, C., Qiu, J.-L., 2014a. Simultaneous editing of three homoeoalleles in hexaploid wheat confers heritable resistance to powdery mildew. Nat. Biotechnol. 32, 947–995. Whelan, A.I., Lema, M.A., 2015. Regulatory framework for gene editing and other new breeding techniques (NBTs) in Argentina. GM Crops Food.: Biotechnol. Agric. Food. Chain. http://dx.doi.org/10.1080/21645698.2015.1114698. Zuniga, A., et al., 2012. Quorum sensing and indole-3 acetic acid degradation play a role in colonization and plant growth promotion of Arabidopsis thaliana by Burkholderia phytofirmans, PsJN. Mol. Plant Micro Interact. 26, 546–553.

Hartung, F., Shieman, J., 2014. Precise plant breeding using new genome editing techniques; opportunities, safety and regulation in the EU, Plant J. Doi: 10.1111/tpj. 12413. Haslberger, A.G., 2000. Monitoring and labeling for genetically modified products. Science 287, 431–432. Henry, I.M., Nagalakshmi, U., Lieberman, M.C., Ngo, K.J., Krasileva, K.V., VasquezGross, H., Akunova, A., Akhunov, E., Dubcovsky, J., Tai, T.H., Comai, L., 2014. Efficient genome-wide detection and cataloging of EMS-induced mutations using next-generation sequencing and exome capture. Plant Cell 26, 1382–1397. Herring, R.J., 2008. Opposition to transgenic technologies: ideology. Interests and collective action frames. Nat. Rev. Genet. 9, 458–463. Heslot, N., Yang, H.P., Sorrells, M.E., Jannink, J.L., 2012. Genomic selection in plant breeding: a comparison of models. Crop Sci.. http://dx.doi.org/10.2135/ cropsci2011.06.0297. Hoffman, N.E., 2013. USDA Regulation of organisms developed through modern breeding techniques. NABC 26, 185–191. Huang, X., Han, B., 2014. Natural variation and genome wide association studies in crop plants. Ann. Rev. Plant Biol. 65, 531–551. Huang, X., Lu, T., Han, B., 2013. Resequencing rice genomes: an emerging new era of rice genomics. Trends Genet. 29, 225–232. Jaggard, K.W., Qi, A., Ober, E.S., 2010. Possible changes to arable crop yields by 2050. Philos. Trans. R. Soc. B. http://dx.doi.org/10.1098/rstb.2010.0153. James, C., 2014. Global status of commercialized biotech/GM Crops: 2014. ISAAA Brief, 49. Jones, J.D.G., Witek, K., Verweij, W., Jupe, F., Cooke, D., Dorling, S., Tomlinson, L., Smoker, M., Perkins, S., Foster, S., 2014. Elevating crop disease resistance with cloned genes. Philos. Trans. R. Soc. B. http://dx.doi.org/10.1098/rstb.2013.0087. Kershon, D.L., Parrott, W., 2014. Regulatory paradigms for modern breeding. NABC 26, 159–168. Kumar, K., Jain, M., 2014. The CRISPR-Cas system for plant genome editing: advances and opportunities. J. Ex. Bot.. http://dx.doi.org/10.1093/jxb/eru429. Lemaux, P., 2008. Genetically engineered plants and food. A scientist's analysis of the issues. Ann. Rev. Plant Biol. 59, 771–812. Lemaux, P., 2009. Genetically engineered plants and food. A scientist's analysis of the issues. Ann. Rev. Plant Biol. 60, 511–559. Li, H., et al., 2015. A high density GBS map of bread wheat and its applications for dissecting complex disease resistance traits. BMC Genom.. http://dx.doi.org/ 10.1186/s12864-015-1424-5. Li, J.F., Aach, J., Norville, J.E., McCormack, M., Zhang, D., Bush, J., Church, G., Sheen, J., 2013. Multiplex and homologous recombination-mediated genome editing in Arabidopsis and Nicotiana benthamiana using guide RNA and Cas9. Nat. Biotechnol. 31, 688–691. Lundberg, et al., 2012. Defining the core Arabidopsis thaliana root microbiome. Nature 488, 86–90. Lussa, M., Davies, H.V., 2013. Comparative regulatory approaches for groups of new plant breeding techniques. New Biotechnol. 30, 437–446. Lynas, M., 2013. Time to Call Out the Anti-gmo Conspiracy Theory, 〈http://mlynas.org〉. Mammadov, J., Aggarwal, R., Buyyarapu, R., Kumpatla, S., 2012. SNP markers and their impact on plant breeding. Int. J. Plant Genom., (doi:10:1155/2012/728398). Mayer, K.X., 2014. A chromosome-based draft sequence of the hexaploid bread wheat (Triticum aestivum) genome. Science, 1788. http://dx.doi.org/10.1126/science.125. McCallum, C.M., Comai, L., Green, E.A., Henikoff, S., 2000. Targeted Induced Local lesion in Genomes (TILLING) for plant functional genomics. Plant Physiol. 123, 439–442. National Academy, 2016. Genetically-Engineered Crop: Experiences and Prospects. National Academy Press, US. Naveed, M., Mitter, B., Reichenauer, T.G., Wieczorek, K., Sessitsch, A., 2014. Increased drought stress resilience of maize through endophytic colonization by Burkholdaria phytofirmans PsJN and Enterobacter sp. FD 17. Environ. Exp. Bot. 97, 30–39. Paarlberg, R., 2010. GMO foods and crops; Africa's choice. New Biotechnol. 27, 609–613. Pieruschka, R., Poorter, H., 2012. Phenotyping plants: genes, phenes and machines. Funct. Plant Biol. 39, 813–820.

Further reading Bernardo, R., 2008. Molecular markers and selection for complex traits in plants: learning from the last 20 years. Crop Sci. 48, 1549–1664.

21