Journal of Experimental Botany Advance Access published January 22, 2015 Journal of Experimental Botany doi:10.1093/jxb/eru547

Review Paper

Planning for food security in a changing climate Bryan McKersie* 11032 Fair Chase Ct, Raleigh, NC 27617, USA *  To whom correspondence should be addressed. E-mail: [email protected] Received 3 November 2014; Revised 11 December 2014; Accepted 14 December 2014

The Intergovernmental Panel on Climate Change and other international agencies have concluded that global crop production is at risk due to climate change, population growth, and changing food preferences. Society expects that the agricultural sciences will innovate solutions to these problems and provide food security for the foreseeable future. My thesis is that an integrated research plan merging agronomic and genetic approaches has the greatest probability of success. I present a template for a research plan based on the lessons we have learned from the Green Revolution and from the development of genetically engineered crops that may guide us to meet this expectation. The plan starts with a vision of how the crop management system could change, and I give a few examples of innovations that are very much in their infancy but have significant potential. The opportunities need to be conceptualized on a regional basis for each crop to provide a target for change. The plan gives an overview of how the tools of plant biotechnology can be used to create the genetic diversity needed to implement the envisioned changes in the crop management system, using the development of drought tolerance in maize (Zea mays L.) as an example that has led recently to the commercial release of new hybrids in the USA. The plan requires an interdisciplinary approach that integrates and coordinates research on plant biotechnology, genetics, physiology, breeding, agronomy, and cropping systems to be successful. Key words:  Climate change, cropping systems, DroughtGard, drought tolerance, genetic engineering, marker-assisted selection, maize, plant breeding.

Introduction: predictions for the future The Intergovernmental Panel on Climate Change (IPCC) has concluded that human-induced emission of greenhouse gases including CO2 has increased global temperatures and predicts that our climate will be continually changing at a relatively rapid rate during this century. Higher temperatures, altered precipitation patterns, and the increased incidence of drought will directly impact crop production and threaten our food supply (IPCC, 2007, 2013). Their common conclusion and other analyses conducted over the past decade is that crop production everywhere runs some risk of being negatively affected by this climate change (Hatfield et  al., 2008, 2014; Karl et  al., 2009; Gornall et  al., 2010; Muller et  al., 2011; Turral et  al., 2011). For some crops grown in cooler regions, higher temperatures will provide a yield

increase, but for most crops in most regions, negative effects on yield are expected. Higher temperatures are expected to exacerbate the effects of water shortage in regions that receive reduced rainfall and to negate any advantage in those regions that receive increased precipitation. Higher temperatures are also expected to reduce yields by directly impacting the biochemistry and physiology of the crops and at the same time encouraging weed and pest proliferation. In addition, sea levels are predicted to rise, affecting drainage in coastal areas, particularly in low-lying deltas, and may result in saline intrusion into coastal agricultural regions. Erosion is likely to increase due to the increased incidence of severe storms. Soil moisture profiles and runoff will also change in region-specific patterns.

Abbreviations: AHAS, acetohydroxyacid synthase; FAO, Food and Agriculture Organization of the United Nations; IPPC, Intergovernmental Panel on Climate Change; QTL, quantitative trait loci; IMI, imidazolinone herbicides; MAS, marker-assisted selection; USDA, US Department of Agriculture. © The Author 2015. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: [email protected]

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Abstract

Page 2 of 16 | McKersie

Fig. 1.  The three constraints that will be imposed on crop production in the next 40 years.

development of new products and increased food security. I  hope this overview helps both young and established researchers as they organize and communicate their research programmes to address regional food security issues.

Options to mitigate the effects on crop production Society has several options to ensure food security and to mitigate the risks to crop production posed by climate change and by societal changes. However, not all will be effective in the short term, and some may not be practical due to alternative demands on limited resources, such as water. The following is a brief summary of some options, which are not mutually exclusive.

Reduce CO2 emissions Global political efforts to mitigate the effects of climate change have focused on reducing greenhouse gas emissions. Unfortunately, these reductions may be too late to prevent an impact on crop production. Even if emission of greenhouse gases was stopped today, climate change and its resulting impacts on agriculture will continue to occur due to the effects of gases that have already been released (Karl et al., 2009). Although this a critical goal that may prevent catastrophic effects in the next century, other short-term mitigations are needed to meet the immediate challenges to crop production.

Utilize high CO2 in photosynthesis The physiology, biochemistry, and morphology of crop plants is impacted directly and indirectly by the environment, notably by CO2, temperature, and water. Consequently, the same crop is expected to grow, develop, and function differently in a future environment. The effects of environment on the physiology, growth, and ultimately yield of crops are complex. Consequently, mathematical models are often used to predict the physiological effects and to estimate future crop yields in specific environments (Karl et al., 2009; Zhu et al., 2013). Some propose that these responses to CO2 may partially ameliorate the negative impacts of climate change and water deficits on yield (Wullschleger et al., 2002; Shanker and Venkateswarlu, 2011). Others propose that plant breeding programmes (Ziska et al., 2012) or genetic engineering programmes (Ainsworth et al., 2008) should focus on capturing opportunities from the high-CO2 environment. Concerns have been expressed that these computer simulation models may have overestimated the positive effects of increased CO2 levels and therefore underestimated the threats to food supply posed by climate change (Long et al., 2006a). For example, Leakey et  al. (2006) suggested that ‘…rising CO2 may not provide the full dividend to North American maize production anticipated in projections of future global food supply…’ because photosynthesis in maize is unaffected by rising CO2 in the absence of drought. Even soybean, which is a C3 plant, does not respond to increased CO2 as

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Not only is agriculture likely to be impacted by climate change, but agriculture also contributes significantly to the greenhouse gas problem by producing emissions that drive climate change. Estimates of all human-induced greenhouse gas emissions from global agriculture vary from 9% (Karl et al., 2009; http://www. epa.gov/climatechange/ghgemissions/sources/agriculture.html, last accessed 2 January 2015) to 30% (Smith and Gregory, 2013) in various regions. In the United USA, agriculture produces 8.6% of the nation’s total greenhouse gas emissions, including 80% of the nitrous oxide emissions and 31% of the methane emissions (Karl et al., 2009). These gases in combination with CO2 are the main gases responsible for global climate change. Deforestation in the Brazilian Amazon to enable soybean [Glycine max (L.) Merr.)] and cattle production contributes 2–5% of the global carbon emissions (Nepstad et al., 2009). In addition, the global agricultural system must provide about 70% more food for a global population estimated to be 9 billion or more by 2050 (Fischer et al., 2005; Smith and Gregory, 2013). A  further complication is that food preferences are changing. The consumption of relatively low-cost carbohydrates from cereals is being replaced by increased demand for milk, meat, fruits, and vegetables. Since the wateruse efficiency of animal production is much lower than that of cereal crops, not only is extra primary production from pastures, rangelands, and arable land required to meet these demands but also more water (Turral et al., 2011). For example, the water used per unit of energy in the food produced is 10 times greater in dairy products than cereals. Consequently, as the population’s demand for food quality changes, the demand for water in agricultural production systems is likely to increase. Improvements to the global agricultural infrastructure for storage and distribution of food will help but will not completely solve the problem. Therefore, the agricultural community must address three opposing demands on crop production simultaneously to ensure food security in the future (Fig.  1). The demand for more high-quality food must be reconciled with the need to have less environmental impact and to use fewer resources in a more stressful and uncertain environment caused by climate change. My thesis is that an integrated research plan merging agronomic and genetic approaches has the greatest probability of success. In this article, I present a template for a research plan that includes an overview of the current biotechnology tools that might be used in this quest. I  use the development of drought tolerance in maize (Z. mays L.) for the US corn belt as an example that has recently released new hybrids to illustrate this research plan. Regional adaptation of the plan for each crop will be needed to ensure commercial

Planning food security  |  Page 3 of 16 predicted by the kinetics of ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCo) due to energy limitations (Crafts-Brandner and Salvucci, 2000). The latest IPCC (2013) and US Department of Agriculture (USDA) (Hatfield et al., 2014) reports recognize that negative effects of climate change on crop production may be larger than previously anticipated. Therefore, we cannot rely on the existing physiology and biochemistry of our crops that have evolved in past low-CO2 environments to prepare them for the future highCO2 environments.

Increase crop irrigation

Increase cultivated land area Another option is to expand production of our current crop cultivars and hybrids to marginal and deforested land. Increasing global food production to meet the projected demand in 2050 could potentially require an area equivalent to that of the Indian subcontinent (Phalan et al., 2014). Deforestation in several tropical regions is a large source of greenhouse gas emissions (Galford et al., 2010), contributes to climate change (Soares-Filho et al., 2010), and negatively impacts biodiversity (Phalan et  al., 2014). Thus, continued deforestation is clearly not a preferred option, but this option will be implemented even more extensively if or when food shortages change the economic and political circumstances.

Develop new crop management technology The ‘yield gap’ between the theoretical and actual crop yields obtained in our current production areas is often cited as a target for innovation and technological change. Some of the changes envisioned include: •• more crops grown per year on the same land: double cropping of summer and winter annual crops such as currently

These changes will require changes in agricultural infrastructure for storage and distribution, farming equipment, crop management, and most likely the crop’s physiology, growth, and development, as well as its response to the environment. There is considerable risk in this approach, and concerns have been expressed as to whether any of these changes in crop production will be successful (Sayer and Cassman, 2013). First, there are biophysical and environmental limits on the conversion of light energy into biomass (i.e. photosynthesis) and then into economic yield (i.e. seed biomass) that may restrict further increases in crop yield. Secondly, the need to intensify agriculture on existing farmland may destroy the environment and contribute to climate change and loss of biodiversity (Phalan et  al., 2014). Thirdly, institutional and regulatory obstacles may prevent implementation (e.g. deregulation of genetically modified plants).

Develop new cultivars Most reports on climate change propose the development of new crop cultivars that have increased yield with fewer inputs or have increased yield stability. These hypothetical crop cultivars have a higher tolerance to drought and high temperature, increased resistance to more fungal diseases and insect pests, and a higher yield potential on both marginal and good farmland. Currently, most plant breeding strategies use high-throughput field evaluation of thousands of lines grown in the target environment on small field plots (Richards et  al., 2010; Cooper et  al., 2014). This poses a potential problem if the breeding programme is targeting a high-CO2 future environment that does not exist currently. Our options in developed countries are to locate testing sites in highly water-limited, hot regions or to use controlled environments with CO2 supplementation. These options may not be available in developing countries with limited resources. Nonetheless, we must find ways to optimize the existing biochemical and physiological traits in our crops for this future field environment. A new strategy is needed to create genetic diversity for use in plant breeding if we are to successfully meet society’s expectations. We must do the following: •• predict what the future crop environments will be; •• envision new crop production systems to capture any opportunities and to mitigate the risks in those potentially different environments;

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

An obvious mitigation to reduce the impact of drought is to increase the use of irrigation for crop production by expanding the water management infrastructure. Where possible, more irrigation will meet regional needs leading to higher crop productivity. However, increased irrigation is unlikely to be a global solution. A report by the Food and Agriculture Organization of the United Nations (FAO) (Turral et  al., 2011) estimated that, based on population growth alone to 2050, without factoring in global warming and climate change, crop production will require 11% more water for irrigation. However, agriculture is not the only segment of society that will require more water resources in the future due to population growth. Strzepek and Boehlert (2010) concluded that agriculture’s primary competition for water will come from legislative environmental flow requirements, and from municipal and industrial water demands. They estimated that 18% less water is likely to be available for global agriculture by 2050. Consequently, this competition for our dwindling water resources is likely to prevent increased use of irrigation, except in very specific cases.

done for soybean and winter wheat (Triticum aestivum L.) rotation in some regions might become more widespread; •• altered planting and/or harvest dates to avoid stressful periods, such as drought or flooding; •• water conservation by altered tillage practices, such as no-till; •• cropping on marginal land that has poor water or nutrient availability; •• yield stability in environments experiencing periodic heat or drought stress; and •• increased use of biological or chemical additives to enhance stress tolerance and improve the overall health of the plant.

Page 4 of 16 | McKersie •• create new cultivars that will have consistently higher yield in these new crop production systems; and •• use these new cultivars to implement the new food production system. In other words, as many others have proposed previously (Beddington, 2010; Piesse and Thirtle, 2010; Zeigler and Mohanty, 2010; McAllister et  al., 2012; Pingali, 2012), I believe that we need a Second Green Revolution. The following are some suggestions on how this might be achieved.

Planning the Second Green Revolution

1. Envision changes in crop management that will increase crop productivity in the predicted future environment. 2. Identify physiological traits that enable those changes in crop management.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Agronomists around the world are working with farmers and achieving considerable success in developing new crop production practices for a changing climate, a recent example being reported by Kirkegaard et al. (2014) in Australia. At the same time, plant breeders are using genetics and biotechnology to develop crops that have greater yield stability in our current production systems (for example, in maize: Castiglioni et al., 2008; Cooper et al., 2014; Habben et al., 2014). These initiatives may be sufficient to meet our needs for food security in a hot, dry, high-CO2 environment. Nonetheless, I recommend that we do not take the risk and become complacent. My thesis is that is if we coordinate the use of both agronomic and genetic approaches to achieve the shared goal, we will greatly improve the probability of food security in the future. The first Green Revolution led by Norman Borlaug applied high inputs of nitrogen fertilizer to responsive short-statured, short-season cultivars of rice (Oryza sativa L.) and wheat, often with irrigation, to realize the potential yield (Nelson et al., 2009; Pingali, 2012). The greatest impact was in Asia and it is credited as the fundamental factor leading to the economic development of that region today. However, the Green Revolution had very little impact on Africa, either in terms of food security or wealth creation, perhaps due to the relatively small potential for irrigation in Africa (Turral et  al., 2011). Increases in cereal yields as a result of widespread adoption of these improved crop management practices and new cultivars saved natural ecosystems from being converted to agriculture (Stevenson et al., 2013). Although their calculations are complex, they made a conservative estimate that, in the absence of the first Green Revolution, an additional 18–26 Mha, including 2 MHa of additional deforestation, would have been required to produce the food needed in 2004. Infrastructure, including storage and transportation facilities, were critical to success. However, the FAO reports that, even today, the amount of cereals lost during harvesting, threshing, and storage in India is equivalent to about 10% of their annual production, or about 15 Mt (http://www.fao.org/wairdocs/ x5002e/x5002e00.htm; http://online.wsj.com/news/articles/SB1 0001424052702304356604577339402041821044, last accessed 2 January 2015). This is comparable to the annual wheat production of Australia, which is about 20 Mt annually (http:// en.wikipedia.org/wiki/International_wheat_production_statistics, last accessed 2 January 2015). It is society’s expectation that crop productivity research in plant biotechnology, genetics, physiology, breeding, and

agronomy will develop new cultivars and new crop management practices to stabilize or increase yield. These will be coupled with infrastructure improvements to transportation and irrigation systems to increase food supply for a growing population (Karl et al., 2009; Nelson et al., 2009; Muller et al., 2011; Turral et al., 2011). This opinion has been re-enforced most recently by the US Climate Change Science Program in their draft report (Hatfield et al., 2014): ‘Agriculture has been able to adapt to recent changes in climate; however, increased innovation will be needed to ensure the rate of adaptation of agriculture and the associated socioeconomic system can keep pace with climate change over the next 25 years.’ In other words, society (on the advice of many agricultural experts) expects that a Second Green Revolution is simply a matter of allocating financial and human resources to the problem. This may be one reason why society has fewer concerns about the future impacts of climate change on food security than is warranted. The initial strategy employed in the first Green Revolution to increase yield in rice and wheat was to increase the application of nitrogen fertilizer. The problem was that applying more nitrogen to the existing cultivars increased the height of the plants, which increased lodging and actually reduced harvestable yields. The solution was to grow short-statured cultivars that did not lodge as readily at high rates of nitrogen fertilization. However, short-statured, agronomically acceptable cultivars did not exist, so it was essential to introgress mutated alleles of genes controlling plant height from nonagronomic germplasm into new cultivars. Once nitrogen was removed as a limiting factor, the larger plants transpired more. Water now became the new growthlimiting factor in many regions. Thus, the full yield potential of the new cultivars in these environments was achieved only by supplying more nitrogen under irrigation. Infrastructure improvements to irrigation systems, fertilizer manufacturing, and distribution, and to seed production and handling were also required. In other words, the whole production system had to change to achieve the goals set out in the first Green Revolution. Borlaug was limited at the time to using existing genetic variability in rice and wheat to create his new cultivars. Since then, genetic engineering technology has emerged as a powerful tool leading to the commercial release of herbicide and insect tolerance in maize, soybean, and other crops. This tool of plant biotechnology can now be used to tailor other genetic changes that enable alternative crop management practices. Merging these two approaches into what may eventually be called a Second Green Revolution is required to meet the demands for food security over the next several decades. The lessons learned from the success of the first Green Revolution and from the first generation of genetically modified crops provide a template for research strategies on how we might approach implementation in six sequential steps (Fig. 2):

Planning food security  |  Page 5 of 16

3. Identify genes that regulate those physiological traits. 4. Create genetic diversity for those physiological traits using the identified genes. 5. Breed new regionally adapted cultivars to incorporate those physiological traits into agronomically acceptable cultivars. 6. Implement the envisioned changes in crop management using the newly developed cultivars. The restrictions are that any changes in crop production must not negatively impact the environment or contribute to further climate change by increasing global greenhouse gas production. These changes must provide a greater overall yield stability across diverse environments from year to year and from location to location. So the targets of the Second Green Revolution differ from the first because increased nitrogen fertilization and increased irrigation are unlikely to be sustainable practices in the future climate.

Step 1: envision changes in crop management system To begin, we need to envision a new crop management system with the potential to be a breakthrough that leads to an incremental yield increase in the predicted future environment. This step is essential in the research plan to enable communication to stakeholders and funding agencies and to provide focus, context, and mandate for the interdisciplinary teams. One way to categorize and prioritize alternatives is to dissect yield into components (Fig.  3) that might be simpler targets to change by modifying crop management. One or more of the following opportunities might be a target but the selection must be tailored for specific regional environments. A  comprehensive review of the crop production practices

that might be used in the future to increase yield is beyond the scope of this article. The reader is referred to the following articles on this topic (Fischer et al., 2014; Passioura and Angus, 2010) as well as a special issue of Field Crops Research on yield gap analysis (van Ittersum and Cassman, 2013). The following sections discuss only a few of many potential new crop production practices to illustrate the concept of envisioning in step 1.  Whether any represent significant new global opportunities or whether any require new crop cultivars has yet to be established.

Innovation example 1: double cropping and intercropping Double cropping of soybean after harvesting wheat is currently a common production system in the mid-southern USA (Kyei-Boahen and Zhang, 2006). Cropping systems based on a single crop waste large proportions of key inputs on an annual basis, including radiation and water. Growing more crops per year theoretically improves resource capture and productivity. As an example, Caviglia et al. (2004) compared wheat–soybean single and double-cropping systems and confirmed that double cropping improved resource capture and efficiency, but the impact was much greater for water than for radiation. They attributed the difference to storable (water) versus non-storable resources (radiation) and proposed that further research in their region emphasize improvements in radiation capture. Many studies have been conducted on this cropping system in the USA and have shown that tillage practices and the availability of adequate soil moisture when planting soybean are among the more critical factors to consider http://oilseeds.okstate. edu/production-information/soybean/PSS-2137%20(double%20crop%20soybean).pdf, last accessed 2 January 2015;

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Fig. 2.  Schematic flow chart showing a six-step process to develop new crop cultivars to meet the challenges in crop production that are anticipated in the near future. MAS, marker-assisted selection.

Page 6 of 16 | McKersie

http://www.lsuagcenter.com/NR/rdonlyres/5673F10F32C8-4465-9A40-4D6BD8D6FCA9/45621/pub3053doublecroppingsoybeanswheatHIGHRES.pdf, last accessed 2 January 2015). In another example, winter cereal production systems in the northern USA are considered to be inefficient with respect to the capture of radiation during the year because the majority of fields lie fallow after grain harvest until the following cropping season. To maximize radiation capture and biomass production, a common cropping system utilizes frost-seeding techniques to intercrop the winter cereal with an annual legume, usually red clover (Trifolium pratense L.). In such a system, winter triticale (×Triticosecale Wittmack) and wheat differ in their effect on the forage crop because red clover’s plant growth rate and radiation-use efficiency interact

Innovation example 2: biological soil additives Microbial symbionts associated with plant roots confer biotic and abiotic stress tolerance to their plant hosts. Positive effects have been reported for both plant growth-promoting rhizobacteria (Timmusk and Wagner, 1999; Yang et al., 2009; Glick, 2012; Mengual et al., 2014) and arbuscular mycorrhizal fungi (Redman et al., 2002; Waller et al., 2005; Rodriguez et  al, 2008; Celebi et  al., 2010). For example, inoculating switchgrass seeds with plant growth-promoting rhizobacteria resulted in more and taller tillers and a 40% higher yield compared with than those not inoculated in a low-nitorgen input production system (Ker et al., 2012). The review of Trabelsi and Mhamdi (2013) on the impact of microbial inoculants on soil microbial communities concluded that both soil or seed treatments may change the indigenous microbial populations, at least temporarily, and thereby improve plant performance. The current working hypothesis is that microbes provide the crop with novel nutritional and defence pathways, and at the same time modulate plant biochemical pathways, thereby altering plant phenotypes. This potential benefit of microbial additives may provide a novel strategy for reducing the yield gap and mitigating the impacts of global climate change on crops (Berendsen et  al., 2012). Genomic technologies have the potential to facilitate the development of these microbial approaches (Coleman-Derr and Tringe, 2014). Although still very much in its technical infancy, microbial soil inoculants may be developed as an agronomic management tool that could be used to mitigate the effects of climate change and provide greater yield stability in future environments. Whether plant genetics can be used to enhance the microbial–plant interaction commercially has yet to be established. Innovation example 3: plant health-promoting chemical additives A long-term goal of research on plant growth regulators has been improved tolerance of drought and other abiotic stresses, but there have been few commercial successes until recently. The strobilurin fungicides are promoted not only

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Fig. 3.  Targets for improvement in new crop management systems. Yield potential (Yp) is the theoretical maximal yield of the crops(s) in one year. Yp is determined by the total amount of light energy captured by the crop(s), its radiation-use efficiency, which is a measure of the efficiency of conversion of that light energy into biomass, and its harvest index, which is the proportion of biomass portioned into economic (e.g. seed) yield. Photosynthetic leaf area and its duration over the growing season are factors that modulate Yp. Therefore, management practices that increase the duration of light capture on an annual basis, such as double cropping, or intercropping, might be expected to increase Yp. Annual yield (Ya) is the economic yield delivered into the food chain per unit of land area per year. The difference between Yp and actual yield is due to various losses, which are represented as pies in the chart. The size of the pie slice for each will vary from region to region and from crop to crop. The relative sizes shown here are simply illustrative and not quantitative. The yield gap (Yg) is the proportion of Yp that is lost and not converted into economic yield in a grower’s field. Yg is large if a factor other than light energy limits yield, which is almost always the case. Yg includes factors inherent to the crop such as nutrient-use efficiency and environmental factors such as nutrient availability that may classify certain regions as marginal. Management practices that conserve water, such as no-till, or which enhance uptake of growth-limiting nutrients, such as biological soil additives, would be expected to reduce the yield gap. Yield volatility (Yv) is a function of the plant’s response to its environment. Management practices that reduce the impact of biotic and abiotic stress, such as altered planting dates or the application of healthpromoting chemicals, would be expected to reduce volatility. Losses in seed yield during harvest occur due to shattering or lodging that leave seed in the field, or due to deterioration such as sprouting in wheat. Direct combining of wheat to avoid swathing and the risk of sprouting damage, which is common in some regions, would reduce harvest loss. Other losses in storage or distribution can also be quite significant before the grain enters the food chain.

with the cereal species and environmental conditions during specific growth periods (Singer et al., 2007). Another crop management system being introduced into East Africa is to address the yield stability issue caused by combined stresses from insect pests, striga weeds, and degraded soils. The new cropping system intercrops cereal crops with a forage legume, desmodium (Desmodium uncinatum Jacq.) and with Napier grass (Pennisetum purpureum Schumach) as a border crop (Khan et al., 2014). On-farm field trials of the technology have shown significant grain yield increases, but improvements to infrastructure are also required. These include a more efficient desmodium seed production and distribution system, relevant policy changes, farmer and stakeholder training, and infrastructure improvement. These double and intercropping systems use current cultivars that were developed for single annual crop production systems. It is envisioned that different genetics for maturity or for other agronomic attributes would enable a better fit into new double or intercropping systems.

Planning food security  |  Page 7 of 16

Innovation example 4: avoiding drought stress If water is the predominant limitation to yield, improvements can be made by matching the crop’s water requirements to the pattern of water supply, thereby avoiding the impact of drought (Passioura, 2006). Cereal crops are very sensitive to water deficits at anthesis, and require the most water during grain filling. Thus, conservation of water during the vegetative growth stage may avoid the effects of a later chronic drought. Several agronomic practices affect the infiltration of water into the soil, its water holding capacity, and evaporative losses. These include planting date, rate and uniformity of establishment, weed control, fertilization, stubble management, and the previous crop rotation. The direct evaporation of moisture from the soil surface can be a significant loss, but it can be minimized by rapid canopy closure, which requires good seedling establishment and rapid leaf development. Because leaf growth increases with air and soil temperature, planting winter annual crops early, when the soil and air are still warm, creates more canopy cover during late autumn and winter, which reduces evaporative losses from the soil surface and conserves water for growth during grain filling (Passioura, 2006). Repeated cultivation of the soil to control weeds and to make a fine seed bed not only damages the soil structure but allows a greater evaporative loss of water from the soil. Directly planting seed into the soil without cultivation (notill), combined with using a broad-spectrum herbicide (e.g. glyphosate) to achieve better weed control, improves the timeliness of sowing, reduces evaporative losses, and thereby improves yields in water-limited environments. This agronomic innovation requires a range of herbicide-resistant (usually genetically engineered) cultivars that are specifically suited to being planted at different times during the season (Passioura, 2006). Another innovation is to select for large early leaves in wheat combined with reduced tiller development to optimize

biomass production. This not only reduces evaporative losses but also inhibits the growth of weeds (Lemerle et al., 2001). Other innovations involving better rooting patterns and a healthier root system can be postulated to further improve this cropping system. Small amounts of subsoil water can enhance grain yield (Kirkegaard et al., 2007). Consequently, plants that have deep roots are expected to have greater yield but only if there is water available in the subsoil, if there is no hardpan from soil compaction, and if there is no subsoil salinity. Thus, the potential for this target is very regional.

Step 2: identify traits that enable changes in crop management Once the new crop management system is envisioned, it is necessary to determine which (if any) physiological trait(s) of the crop should be changed by plant biotechnology and/ or plant breeding to enable the new crop management system. Several recent articles have proposed various targets for genetic manipulation to increase yield in crops (for example, Reynolds et  al., 2005; Long et  al., 2006b; Ainsworth et  al., 2008, 2012; Long and Ort, 2010; Yang et al., 2010; Zhu et al., 2010; Kant et al., 2012; Slattery et al., 2013; Wang et al., 2013). Very few of these proposals are linked to proposed crop management changes. I am proposing that the target crop management system be used to identify the required physiological, developmental, or biochemical modifications to the crop that are needed to implement that new crop management system. This is not a novel proposal but is in fact exactly the same strategy as used in the first Green Revolution, except now we have more genetic tools. Let us take drought tolerance in maize grown in the US corn belt as an example to illustrate the subsequent steps in this strategy because this has been the focus of several recent commercial releases. More drought-tolerant maize hybrids may stabilize yield (reduce yield volatility) in existing production areas but may also enable several alternative crop management systems including: •• reduced irrigation; •• increased plant population densities; and •• expansion into more marginal and rain-fed regions/soils. ‘Drought’ has different meanings to a meteorologist, farmer, agronomist, and molecular biologist (Passioura, 2007). Consequently, ‘drought tolerance’ has quite different meanings as well. In the following sections, the target ‘drought tolerance’ trait in maize is defined as the kg of seed produced per unit of water transpired. A drought-tolerant maize hybrid would therefore produce more grain from the same amount of water transpired, or the same amount of grain from less water transpiration.

Picking drought-tolerance traits There are several factors to consider when picking the traits that will impact ‘drought tolerance’ in maize for the US market. First is the phenotyping method. Lack of good phenotyping tools that provide quantitative data is a major reason why

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

for their fungicidal properties but also for their effects on stress tolerance and yield. Strobilurins are effective against several different plant-pathogenic fungi and give high levels of protection against a wide range of crop diseases. (For more details on these fungicides, see the following reviews: Bartlett et  al., 2002; Balba, 2007; Zhao et  al., 2010). In addition, the strobilurin fungicides have been reported in numerous studies to increase seed yield in the absence of detectable levels of fungal infection, prompting the commercial application of strobilurins to improve tolerance of drought and other stress resulting in reduced variable yield (Grossmann and Retzlaff, 1997; Grossmann et  al., 1999; Jabs et al., 2002; Nason et al., 2007; Nelson and Meinhardt, 2011; Ishikawa et al., 2012). A growing number of producers have responded by incorporating fungicide treatments into their management programmes in attempts to increase yield (http://agproducts.basf.us/products/headline-fungicide.html, last accessed 2 January 2015). Like biological additives, the use of chemical additives to reduce variable yield may be a management option to evaluate as part of a new crop management system.

Page 8 of 16 | McKersie

Law of the minimum Liebig’s law of the minimum was originally used to promote the efficacious use of fertilizers. His law states that plant growth is controlled not by the total amount of resources (nutrients) available, but by the scarcest resource (http://en.wikipedia. org/wiki/Liebig’s_law_of_the_minimum, last accessed 2 January 2015). Liebig’s barrel (Fig. 4) is an illustration of this

law that is often used to this day in extension publications that provide fertilizer recommendations. The barrel has staves of unequal length, with each stave representing a nutrient. The shortest stave limits the capacity of the barrel, indicating that this nutrient should be supplemented to the crop to increase the capacity of the barrel. Once that stave is lengthened (i.e. nutrient supplied), another stave becomes the shortest and limits the capacity of the barrel. This concept is implicit in crop production research. For example, in the first Green Revolution, once the dwarf wheat cultivars were supplied with more nitrogen, water became the next limiting factor. In a more recent example, Passioura (2006) noted that if cereal grain yield per water used is markedly less than 20 kg ha−1 mm−1, it is likely that other stresses such as weeds, diseases, poor nutrition, or inhospitable soil are limiting yield. He noted that any benefit from improved water management can be achieved only if these constraints are eliminated first. The concept of Liebig’s law seems especially useful when we are developing strategies to improve yield-related traits in crops using genetics. The staves can be considered to represent different traits or genes (alleles) and the capacity of the barrel to represent yield (Fig. 4). Although it may be overly simplistic, it does not incorporate the interactions among traits, and it is not quantitative, the law of the minimum model provides a conceptual model to categorize the physiological factors limiting drought tolerance and raises several questions about our current research strategy. For example: 1. Does the same stave limit ‘drought tolerance’ regardless of what the barrel contains? In other words, does what we measure dictate what we observe and conclude? If we measure greenness of a leaf as a proxy for drought tolerance, do we identify the same biological processes as we would if we measured seed yield in a water-limited field trial? 2. Does the same stave limit ‘drought tolerance’ in all plant species?

Fig. 4.  Liebig’s barrel illustrating the law of the minimum. The shortest stave of the barrel limits the capacity of the barrel to hold water, but the same concept can be applied to yield traits. Reproduced from Wikipedia at http://en.wikipedia.org/wiki/Liebig%27s_law_of_the_minimum (last accessed 2 January 2015) with permission.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

plant breeding for knowledge-based stress avoidance and tolerance mechanisms has not progressed well, even though there is genetic variation for these traits in germplasm collections (Richards et al., 2010; Araus et al., 2012; Araus and Cairns, 2014). Salekdeh et al. (2009) proposed a set of rigorous criteria for phenotyping in both controlled and field situations, and they proposed several phenotypes associated with different yield components under drought. They noted that drought environments are diverse, and several biotic and abiotic stresses affect yield in these environments. Thus, they did not propose the use of a single environment to impose a drought stress or a single set of phenotypes to quantify drought tolerance, but suggested that yield, water use, water-use efficiency, and harvest index be included as reference phenotypes in future studies. Some commercial breeding programmes advocate the use of managed-drought-environment technologies (Cooper et al., 2014), whereas others have concluded that most traits of importance in dry environments are selected best in favourable moist environments (Richards et al., 2010). Perhaps this is a difference between maize and wheat breeding, or the American and Australian environments. Second is the genetics. Most current plant breeding programmes select for yield in field environments (Richards et al., 2010; Cooper et al., 2014). To be useful in a plant breeding programme, physiological traits must have genetic variability, be genetically correlated with yield, have higher heritability than yield, and be quantifiable in a high-throughput manner (Fig. 2). Knowledge of the genomics of the plants and prediction methodology has led to the identification of molecular markers in the maize genome that are associated with drought-related traits, and these surrogates for the physiological traits have been applied in maize breeding (Cooper et al., 2014). Similarly, in rice (O. sativa L.), breeders first characterized the available germplasm under drought at the morphological, genetic, and molecular levels and found genetic variation for drought tolerance within the available gene pool. Rice genome sequence information, genome-wide molecular markers, and low-cost genotyping platforms are now being applied in marker-assisted selection (MAS) approaches to improve grain yield under drought by stacking grain yield quantitative trait loci (QTL) (Swamy and Kumar, 2013). Third is response to the environment. Climate change is being driven in part by higher CO2 levels in the atmosphere. Because of the complexity of the plant’s response to CO2, and to water deprivation, many have utilized mathematical simulation models to predict the effects of climate and of physiology on crop growth and yield (Karl et al., 2009; Zhu et al., 2013). Crop modelling is used to evaluate the interaction of multiple traits for their combined effects on drought response and yield in maize (Cooper et al., 2014).

Planning food security  |  Page 9 of 16 3. Does the same stave limit ‘drought tolerance’ under mild, transient water deprivation as following a long period of water deprivation? 4. Does the same stave limit ‘drought tolerance’ regardless of the plant’s stage of development? 5. Does the same stave limit ‘drought tolerance’ in all tissues?

Step 3: identify genes that modify traits A prerequisite for biotechnology approaches to crop improvement is the identification of genes that modify or regulate the physiological traits of interest. One approach often used, especially in academic research, is knowledge based. The researcher starts with a hypothesis on the mode of action (function) of a gene that can be tested by overexpression (or knockout) of a specific gene in model transgenic plants. This approach has been successfully used commercially by DuPont Pioneer in maize for drought tolerance targeting ethylene metabolism (Habben et  al., 2014). An alternative approach is to conduct large-scale screening, starting with a population of genes overexpressed (or knocked out) in a model plant such as Arabidopsis, and then selecting the most efficacious transgene based on empirical results. This approach has also been used successfully leading to the release of DroughtGard by Monsanto/BASF (Castiglioni et al., 2008). However, the list of transgene candidates identified by these approaches for drought tolerance is huge. For example, Lawlor (2013) summarized the positive drought tolerance phenotypes into nine major categories that included about

Step 4: create genetic diversity The next step in this research plan is to create the genetic diversity using one or more biotechnology tools for the new physiological or biochemical traits in the target crop (Fig. 2). These tools include: •• MAS; •• genetic engineering; •• mutagenesis; and •• transformation. In general, if the genetic diversity exists already within the crop for the desired trait, then it is simpler, faster, and cheaper to use MAS coupled with traditional breeding to introgress

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

The literature indicates that the answer to all of the above questions is a clear ‘No’. Our original research strategy was to characterize drought tolerance traits in model plant systems grown in artificial environments and extrapolate these results to crops. We now know that we would make different conclusions based on how we define drought tolerance, what we measure, how we measure it, and in which species we measure it. Since our inductive logic has not been validated by the empirical data, we need to change our research strategy. It now appears that research in a model plant species, such as Arabidopsis, is exceedingly useful to define potential fundamental biological processes that impact drought tolerance. Nonetheless, we cannot use inductive logic to create a hypothesis on drought tolerance in crops based on these observations. Experiments using model systems will not identify traits that limit yield in crops. If we accept this argument, it is not surprising that a genetic tool (e.g. vector construct) that was efficacious in a model species rarely impacts the crop’s drought tolerance or yield (Lawlor, 2013). Therefore, the target physiological traits that may enable new crop management and improve drought tolerance must be selected based on knowledge of crop growth and development in the target crop grown in field environments. This becomes exceedingly difficult in a changing climate, since those hot, dry, highCO2 field environments do not exist yet. Consequently, we have no alternative but to compromise and extrapolate from controlled environments, noting the risk involved in doing so.

600 distinct types of transgenic plants. In another example, Yang et al. (2010) summarized the regulatory genes that have been identified in Arabidopsis and reported positive results from eight transcription factor gene families, two post-transcription gene families, and nine osmoprotectant metabolite classes. This is not a positive outcome from our science. Concerns have also been expressed about the quality of this gene discovery research because overexpression of very many single transgenes seems to slightly increase drought tolerance in transgenic plants under extreme conditions. Lawlor (2013) has expressed concern about the quality of these reports because they neglect the physiology of plant water relations, use unspecific definitions and criteria, and use inadequate methods to assess drought tolerance. He noted that in these reports there is a trend in which transgenic plants exhibit a form of ‘drought resistance’ that he calls ‘delayed stress onset’. In experiments that deprive plants of water to demonstrate drought tolerance, the transgenic plants develop stress symptoms later than in wild-type plants. These ‘drought-tolerant’ plants tend to have slower and less water loss, less total leaf area, decreased stomatal conductance, and thicker laminae, regardless of the transgene that was overexpressed. When rewatered, these transgenic plants have faster and greater recovery than wild-type plants. Biochemical and metabolic changes are often observed in the transgenic plants. These observations are then proposed to indicate a cause–effect relationship among the transgene, the biochemical changes, and drought tolerance. Lawlor (2013) considers this conclusion to be invalid because the degree of water deprivation experienced by the cells in the control and transgenic plants is not equivalent. Thus, there seems to be a high rate of false-positive gene discovery for drought tolerance transgene, and the stringency of our gene discovery research needs to increase. A  set of clearly defined terms for the relevant physiological traits that are commonly accepted needs to be implemented (Lawlor, 2013). Preliminary results in one set of experiments need to be validated in more thorough follow-up studies and the results published, whether they support or refute the initial findings. Editors, reviewers, and granting agencies need to recognize the importance of this confirmation research because it will minimize costly commercial development by providing focus to the most efficacious transgenes in each crop.

Page 10 of 16 | McKersie

Step 5: breed diversity into new adapted cultivars The next step to implement the crop management change is to transfer the trait from the original source, whether that be from exotic germplasm or from an engineered plant, to regionally adapted commercial germplasm. The new trait will need to be combined with existing biotechnology traits, such as herbicide resistance, and with other agronomic traits, such as maturity, disease resistance, and quality to meet agronomic and quality requirements. Breeding is likely to use MAS for several (or all) of these traits (Collard and Mackill, 2008). In this respect, it will be simpler and faster to transfer an

engineered trait than a trait from exotic germplasm. A single marker may be sufficient for the engineered trait, but if the trait originated in exotic germplasm or is quantitative (multigenic), several markers may be required to track the required genes, and to remove linked genes, which adds complexity. Unlike the previously engineered traits of glyphosate resistance or root worm resistance that introduced novel protein functions into the crops, changes to yield and stress-related traits will most likely require the modification of existing molecular networks within the plant that involve many interacting genes. It is anticipated that there is allelic variance for these networks within the crops, and that these ‘native’ alleles may interact differently with the engineered gene. This creates the potential to optimize the trait using breeding methodology in commercial germplasm by combining favourable alleles. It also raises a potential commercial challenge if there is a significant (genotype×transgene)×environment interaction, as reported by Habben et al. (2014) for the downregulated ACC synthases in maize. This approach is highly analogous to the breeding approach used successfully in the 1970s and 1980s to intogress disease resistance from wild relatives into cultivated barley by E. Reinbergs (University of Guelph, Ontario, Canada, personal communication) to find a ‘Happy Home’ for the resistance gene. The final commercial product from the plant breeding programme must be robust. The new cultivar or hybrid must function across various environments and across the range of crop maturities grown within the target region. And it must function every year. If an engineered trait fails, or even is perceived to fail, in one year because of weather conditions, then farmers would be highly unlikely to invest in that seed the following year. Thus, commercial plant breeding will probably be highly conservative in the approach used to develop new cultivars. Also, extensive multiyear testing of the new cultivars or hybrids will be required before commercial launch of these products. It must be recognized by all stakeholders in these projects that these are high-risk approaches because the final product may not meet our expectations in the future climate. So constant monitoring, evaluation, reassessment, and change management will be critical to the success of this effort.

Step 6: implement changes in the crop management system The first Green Revolution did not end with the development of short-statured rice and wheat cultivars that met the quality and agronomic characteristics required by the local agricultural systems. Neither will this putative Second Green Revolution end with the deregulation of a transgenic event and its incorporation into a new cultivar. The implementation of the new crop management system and the development of new cultivars will be reiterative. The new cultivars need to be tested in the new management system, and the new management system needs to be optimized using the new cultivars. Because any degree of crop failure cannot be risked, the introduction of any new management system will be slow and methodical over several years.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

the trait into regionally adapted germplasm than it is to use either genetic engineering approach (Collard and Mackill, 2008). Thus, the first task is to screen the available germplasm for the target trait, and based on these results decide whether to use MAS or an engineering approach (Fig. 2). To use MAS, the trait must meet the requirements listed in step 2: have genetic variability, be genetically correlated with yield, have higher heritability than yield, and be quantifiable in a high-throughput manner. These are not simple requirements. Both genetic engineering techniques, mutagenesis and transformation, require detailed genetic, physiological, and biochemical knowledge about the target trait to efficiently engineer the trait. In some cases, success has been achieved without this knowledge. Instead a high-throughput screen is required to identify an individual with the trait from within a large population, which is not usually feasible for most agronomic traits (herbicide resistance being the obvious exception). The more complete our understanding and knowledge about the target trait, the more likely the project will be successful. When the trait can be easily identified in a high-throughput screen and when it can be created by mutation to create a new allele, then mutagenesis is the preferred genetic engineering method because of the lower cost and faster regulatory approval processes (see the following section on imidazolinone tolerance as an example). This method is especially applicable in smaller market crops that cannot support the large development and regulatory expense required to deregulate transgenic traits. Consequently, mutagenesis per se is unlikely to be applicable to development of drought tolerance in maize, unless it is used to create a specific new allele that could be identified by a high-throughput screen. Only when sufficient genetic variability cannot be found within the available breeding populations does a transformation approach become a viable option. For example, glyphosate resistance did not exist in soybean before it was engineered. Western corn root worm resistance did not exist in maize before it was engineered. Both traits were created by the introduction of novel transgenes using plant transformation technology and are now hugely successful commercial products. The product of this step in research is an individual plant that is sexually compatible with the crop, so it can then be used in a plant breeding programme, and is deregulated globally, so it can be used in the global food chain.

Planning food security  |  Page 11 of 16

Success in developing drought tolerance in maize

Success example 1: maize breeding Breeding for drought tolerance in maize has been difficult (Salekdeh et  al., 2009; Araus et  al., 2012). The large genotype×environment (location–year) interaction for yield has limited the success of our current approach (Araus et al., 2012). Surprisingly, physiological studies comparing hybrids from different decades have shown that the genetic gains that have been achieved in maize yield have been associated with greater ‘stress tolerance’ (see, for example, Tollenaar and Wu, 1999). Individual plants in new maize hybrids are able to set ears under higher levels of competition than those in older hybrids, which enables higher population densities to be planted. Other traits in maize that have been altered by plant breeding include: sustained leaf photosynthesis during grain filling, a greater number of kernels, a shorter anthesis–silking interval under drought, extraction of less water from the soil before flowering, a reduction in tassel size (in temperate but not tropical maize), escape from terminal drought by early flowering, deeper rooting, and, importantly for this overview, genetically engineered pest resistance that has reduced root damage from soil insects. These physiological analyses were done in retrospect, not by a priori design. Lobell et al. (2014) recently confirmed that maize and soybean yields in the US midwest over the period 1995–2012 increased under all levels of stress. However, the sensitivity of maize to high vapour pressure deficits has increased. Vapour pressure deficit measures atmospheric water demand and depends on air temperature and humidity. Consequently, Lobell et  al. (2014) suggested that sensitivity to heat stress has increased in maize hybrids developed in recent decades in the US midwest. It is uncertain how concerned we should be about this observation given future climate predictions.

Success example 2: MAS Genetics research has identified genomic regions useful to a crop under stress conditions (Collard and Mackill, 2008;

Success example 3: DroughtGard The first commercial release of a crop that was genetically engineered for drought tolerance is DroughtGard maize (MON87460) from the Monsanto/BASF collaboration, as a result of an empirical screening approach. These plants express a novel protein that encodes an RNA chaperone. In Bacillus, this protein acts to maintain RNA structure and biological function. In plants, the endogenous cold shock domain-containing (CSD) proteins regulate stress responses through a post-transcriptional mechanism (Castiglioni et al., 2008). The initial discovery of the transgene’s efficacy in plants was made in Arabidopsis where its overexpression was shown to improve cold tolerance in transgenic seedlings. This observation was expanded and the efficacious CSD genes narrowed in subsequent screening, where it was observed that transgenic rice plants expressing CspB had improved growth

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

The following are some examples of projects that have successfully released more drought-tolerant maize hybrids in recent years. Almost all approaches have had some degree of success. Traditional plant breeding has made considerable progress without applying knowledge of the physiological basis of yield. MAS using markers correlated with drought tolerance has been used successfully, again without specific knowledge of the genes and their physiological effects. On the other hand, both empirical screening and knowledge-based approaches using genes with known functions and physiological effects have been successfully used to genetically engineer transgenic maize with greater drought tolerance. The lack of high-throughput screens for a trait as complex and diverse as drought tolerance has limited any application of mutagenesis to drought tolerance, but its potential for other commercial targets has been demonstrated.

Richards et al., 2010). QTL for drought-tolerance traits and their molecular markers have been identified in different crops, including maize (Cooper et al., 2014). With the development of comprehensive molecular linkage maps, MAS procedures stack these desirable traits to achieve improvements in drought tolerance. The accuracy and preciseness in QTL identification, the significant genetic×environment interaction, the large number of genes encoding yield, and the use of wrong mapping populations have hindered the use of QTL to select for growth and yield under water-limited conditions (Ashraf, 2010). Nonetheless, the development of drought tolerance in maize in North America has recently progressed with the release of maize hybrids from Syngenta’s Agrisure Artesian (http://www3.syngenta.com/country/us/en/agriculture/ seeds/agrisure-traits/Pages/agrisure-artesian-4011.aspx, last accessed 2 January 2015) and Pioneer’s AQUAmax (https:// www.pioneer.com/home/site/us/products/corn/seed-traitstechnologies-corn/optimum-aquamax-hybrids/, last accessed 2 January 2015) breeding programmes. Both companies developed new inbreeds using MAS for physiological phenotypes, combining their proprietary genetic information with genomic data to identify genes associated with drought tolerance. They then developed genetic markers and by MAS brought those markers together in inbreeds. Cause–effect relationships and modes of action were not established, but the markers were ‘simply’ correlated with drought tolerance. Since the new inbreeds and hybrids were derived from previously engineered and deregulated transgenic events for herbicide resistance, the new drought-tolerant hybrids did not require USDA approval before commercial release (http:// biotech.about.com/od/Genetically-Modified-Organisms/a/ Breeding-Versus-Engineering-To-Make-Drought-TolerantCorn.htm, last accessed 2 January 2015). One concern is that the infrastructure, in terms of laboratories and knowledge, that is required to identify QTL for drought tolerance and use MAS in plant breeding represents a significant expense. This technology therefore may not be readily available to smaller plant breeding programmes in developing countries.

Page 12 of 16 | McKersie

Success example 4: water-efficient maize for Africa project The successful implementation of genetic engineering technology to improve drought tolerance in maize requires the creation of regionally adapted hybrids. One example where this is already occurring is the Water Efficient Maize for Africa (WEMA) project (http://wema.aatf-africa.org/, last accessed 2 January 2015). This is a public sector/private partnership that was launched in March 2008 and is being led by the Kenyanbased African Agricultural Technology Foundation and funded by the Bill and Melinda Gates Foundation, Howard G.  Buffett Foundation, and US Agency for International Development. The goal of this multinational project is to enhance food security in sub-Saharan Africa by deploying drought-tolerant maize with insect protection to farmers. The transgenic technology and the associated germplasm were provided by Monsanto royalty-free (http://www.monsanto. com/improvingagriculture/pages/water-efficient-maize-forafrica.aspx, last accessed 2 January 2015; http://www.cimmyt. org/en/projects/water-efficient-maize-for-africa-wema-phaseii, last accessed 2 January 2015). The infrastructure has now been established including scientific staff and the establishment of drought testing sites that meet the national requirements for testing genetically engineered hybrids. So far, the project reports some very encouraging results using conventional breeding, MAS, and biotechnology tools to create new germplasm. In 2010, the project identified several initial hybrids that showed a yield advantage under drought stress conditions. Nine new maize inbred lines were released for licensing. Compared with non-transgenic hybrids, the transgenic hybrids produced 8–14% more grain in multiple years of testing (Waltz, 2014). Commercial release is expected in

2016/2017 in South Africa, but due to the lack of regulatory approval, release in other parts of Africa will be delayed. Note that this technology transfer is only feasible for transgenic traits. The transfer of markers and drought-tolerant germplasm from breeding programmes, such as Agrisure Artesian and AQUAmax, is unlikely to be successful because of the genetic complexity and genotype×environment interactions involved.

Success example 5: ethylene biosynthesis DuPont Pioneer has used a knowledge-based approach to develop more drought-tolerant maize hybrids. Habben et al. (2014) observed that maize uses the ethylene signalling system to abort kernels as part of its survival strategy during severe drought stress. They hypothesized that the survival response in maize is too conservative to maintain high yield and is probably unnecessary in modern agricultural environments. They proposed that modulation of the response by downregulating ethylene production during drought stress would lead to an increase in grain yield. Transgenic maize events were created with silenced ACC synthases using an ACS6 RNA interference construct. These enzymes catalyse the rate-limiting step in ethylene biosynthesis, and the transgenic events had ethylene emission levels reduced by approximately 50%. In field trials that imposed drought stress around anthesis, the transgenic events had up to 9.3 bu ac–1 more grain than the null controls, but on the negative side, there was a (genotype× transgene)×environment interaction. The authors attributed the yield increase to a shorter anthesis–silking interval and an increase in kernel number per ear during drought. Under a low-nitrogen stress, the best event had 7.1 bu ac–1 more yield. Further field testing, deregulation studies, and commercial hybrid production are presumably in progress at DuPont Pioneer.

Success example 6: imidazolinone tolerance As previously mentioned, application of mutagenesis to the development of drought tolerance in maize is not feasible at present because of the lack of high-throughput screens. Nonetheless, mutagenesis will have application in other crops for other yield-related traits. One trait that has agronomic value in crops and that was developed using different methods of genetic engineering is herbicide tolerance. The development of transgenic plants with resistance to glyphosate is well known (Duke and Powles, 2008), but there are other herbicides and other approaches. The imidazolinone (IMI) family of herbicides control a broad spectrum of grass and broadleaf weeds by inhibiting the enzyme acetohydroxyacid synthase (AHAS), also called acetolactate synthase (EC 4.1.3.18). This is a critical enzyme in the synthesis of branched-chain amino acids in plants (Tan et al., 2005, 2006). Sulfonylurea herbicides target the same enzyme. The development of IMI-tolerant crops is a good example of the potential of different genetic engineering strategies to create novel genetic variability, because mutagenesis, transformation, and MAS have all been used to engineer tolerance.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

following exposure to cold, heat, and water deficits as demonstrated by greater plant height. Constitutive overexpression in maize gave a positive improvement in yield under managed water environments and dryland conditions that was predominately the result of increased kernel numbers, not kernel weight. The positive effects of transgene expression were observed in late vegetative/flowering and grain-filling periods. Contrary to the effects of many other transgenes that claim effects on drought tolerance, expression of CspB did not result in detrimental effects on plant size, development, or productivity under well-watered conditions. Consequently, the yield improvements observed under water limiting conditions were not associated with a yield penalty in high-yielding environments. An application for deregulation of a transgenic maize event was made to APHIS in 2011 (http://www.aphis. usda.gov/brs/aphisdocs/09_05501p_fea.pdf, last accessed 2 January 2015) and development of commercial maize ‘DroughtGard’ hybrids was initiated (http://www.monsanto. com/SiteCollectionDocuments/whistlestop-drought-posters. pdf, last accessed 2 January 2015). In 2  years of small-scale production in the drier regions of the US corn belt, growers reported about 280 lb ac–1 (5 bu ac–1 or 310 kg ha–1) more grain than competitive hybrids under conditions that give typical yields in the 50–125 bu ac–1 range (Waltz, 2014).

Planning food security  |  Page 13 of 16

Summary and conclusions Society expects innovation in crop production to meet its demand for food security, despite an uncertain environment due to climate change, increased population, and changing food preferences. The innovations to meet this expectation will require knowledge of the future climate, what might limit yield in that environment, a new crop management system to address that limitation, and the development of new crop cultivars that enable the new crop management system. I am proposing that research follows a six-step sequential plan to create these new commercial products based on merging the lessons that we have learned from the Green Revolution and from the development of the first genetically modified crops. First, we need a vision for the future to guide us, based on the current and projected challenges in crop production. New crop management technology and new agronomic practices need to be envisioned in a cropping system that mitigates risk, meets regional needs, and captures potential regional opportunities in a predicted hot, dry, high-CO2 environment. A  global solution across regions and crops is unlikely due to the complexity of the biological processes, the differences among existing agricultural systems, and the local impact of climate change. Regional innovation, research, and leadership are needed. The physiological traits that will enable these changes in crop management to be implemented might be missing from our current cultivars. If they are missing, genetics can be used to create the needed crop cultivars. If sufficient genetic

diversity already exists within the crop’s germplasm, conventional plant breeding, complimented by MAS, is the preferred route to a commercial product. If diversity does not exist, these traits become the targets for change using genetic tools. Next, if an engineering approach is selected, the genes that regulate these physiological traits must be characterized and understood. This may be achieved initially by empirical screening or by knowledge-based approaches, apparently with equal success to date. Although the basic biological principles may be developed in model systems, the hypothesis that the traits will enable the envisioned crop management changes must be validated in crops under field conditions in the target region. Genetic diversity for these physiological traits is then created using the identified genetic engineering technology and knowledge of the trait. Either mutagenesis or transformation may be used depending on the crop, market, and suitability of screening methods. Once the hypothesis is validated in field trials, the biological changes to the crop must be understood and documented to deregulate any commercial products. Thus, improving our knowledge of the physiology, biochemistry, and molecular biology of the trait is fundamental at this step. Next, we need to breed new regionally adapted cultivars to incorporate these physiological traits into agronomically acceptable cultivars. Finally, the new cultivars (or hybrids) need to be utilized to implement the envisioned crop management system in a potentially reiterative research programme. Implementation will require not only changes in crop management but also regional infrastructure improvements and education if we are to meet society’s needs for food security in the face of inevitable climate change. An integrated sequential research plan that combines all of our technologies in plant biotechnology, genetics, physiology, agronomy, and breeding has the greatest potential to provide food security for the future.

Acknowledgements The author thanks Stephen Crafts-Brandner, Jeremy Singer, Larry Gusta, Suzanne Cunningham, and Bijay Singh for their comments, advice, and critical review of this manuscript. The author is retired from BASF Plant Science and is a former faculty member at the University of Guelph, Ontario, Canada.

References Ainsworth EA, Rogers, A, Leakey AD. 2008. Targets for crop biotechnology in a future high-CO2 and high-O3 world. Plant Physiology 147, 13–19. Ainsworth EA, Yendrek CR, Skoneczka JA, Long SP. 2012. Accelerating yield potential in soybean: potential targets for biotechnological improvement. Plant, Cell and Environment 35, 38–52. Anderson PC, Hinnerd KA. 1988. Herbicide resistance in plants. US patent no. 4761373. Araus JL, Serret MD, Edmeades GO. 2012. Phenotyping maize for adaptation to drought. Frontiers in Physiology 3, 305. Araus JL, Cairns JE. 2014. field high-throughput phenotyping: the new crop breeding frontier. Trends in Plant Science 19, 52–61. Ashraf M. 2010. Inducing drought tolerance in plants: recent advances. Biotechnology Advances 28, 169–183.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Tolerance to IMI herbicides was introduced through mutation breeding and commercially released by BASF and its partners as Clearfield crops that were developed using conventional breeding methods beginning in 1992. In the Clearfield approach, variant AHAS alleles were created through mutagenesis and selection to confer IMI tolerance in maize, wheat, rice, oilseed rape (Brassica napus L.) and sunflower (Helianthus annuus L.) (Tan et  al., 2005, 2006). The allele encoding IMI tolerance was then introgressed into commercial germplasm. Similar approaches have subsequently been used in other breeding programmes (Oldach et al., 2008; Lee et al., 2011; Rustgi et al., 2014). Transformation was used as an alternative strategy to create IMI tolerance in crops. Many patents and publications describe the production of herbicide-resistant plants by transforming plants with mutant alleles that encode altered AHAS enzymes resistant to inhibition by the IMI and sulfonylurea herbicides (Anderson and Hinnerd, 1988; Lee et  al., 1988; Wiersma et al., 1989; Bedbrook et al., 1991; Sala et al., 2008). Although technically feasible, none of the transgenic IMItolerant plants have been commercially released for various economic and agronomic reasons. In this case, mutagenesis was the commercially successful approach mainly because a high-throughput screen enabled identification of commercial levels of tolerance and because the development and regulatory costs were sufficiently low to justify the expense in relatively ‘small market’ but important crops.

Page 14 of 16 | McKersie Balba H. 2007. Review of strobilurin fungicide chemicals. Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants, and Agricultural Wastes 42, 441–451. Bartlett DW, Clough JM, Godwin JR, Hall AA, Hamer M, ParrDobrzanski B. 2002. The strobilurin fungicides. Pest Management Science 58, 649–662. Bedbrook JR, Chaleff RS, Falco SC, Mazur BJ, Summerville CR, Yadav, NS. 1991. Nucleic acid fragment encoding herbicide resistant plant acetolactate synthase. US patent no. 5013659. Beddington J. 2010. Food security: contributions from science to a new and greener revolution. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 61–71. Berendsen RL, Pieterse CM, Bakker PA. 2012. The rhizosphere microbiome and plant health. Trends in Plant Science 17, 478–486. Castiglioni P, Warner D, Bensen RJ, et al. 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. Caviglia OP, Sadras VO, Andrade FH. 2004. Intensification of agriculture in the south-eastern pampas. Field Crops Research 87, 117–129.

Coleman-Derr D, Tringe, SG. 2014. Building the crops of tomorrow: advantages of symbiont-based approaches to improving abiotic stress tolerance. Frontiers in Microbiology 5, 283. Collard BC, Mackill DJ. 2008. Marker-assisted selection: an approach for precision plant breeding in the twenty-first century. Philosophical Transactions of the Royal Society B: Biological Sciences 363, 557–572. Cooper, M, Gho, C, Leafgren, R, Tang, T, Messina, C. 2014. Breeding drought-tolerant maize hybrids for the US corn-belt: discovery to product. Journal of Experimental Botany 65, 6191–6204. Crafts-Brandner SJ, Salvucci ME. 2000. Rubisco activase constrains the photosynthetic potential of leaves at high temperature and CO2. Proceedings of the National Academy of Sciences, USA 97, 13430–13435. Duke SO, Powles SB. 2008. Glyphosate: a once-in-a-century herbicide. Pest Management Science 64, 319–325. Fischer G, Shah M, Tubiello FN, van Velhuizen H. 2005. Socioeconomic and climate change impacts on agriculture: an integrated assessment, 1990–2080. Philosophical Transactions of the Royal Society B: Biological Sciences 360, 2067–2083. Fischer RA, Byerlee D, Edmeades GO. 2014. Crop yields and global food security: will yield increase continue to feed the world? ACIAR Monograph No. 158. Canberra: Australian Centre for International Agricultural Research. http://aciar.gov.au/publication/mn158. Last accessed 2 January 2015. Galford GL, Melillo JM, Kicklighter DW, Cronin TW, Cerri CEP, Mustard JF, Cerri CC. 2010. Greenhouse gas emissions from alternative futures of deforestation and agricultural management in the southern Amazon. Proceedings of the National Academy of Sciences, USA 107, 19649–19654.

Hatfield J, Takle G, Grotjahn R, Holden P, Izaurralde RC, Mader T, Marshall E, Liverman D. 2014. Agriculture. In: Melillo JM, Richmond TC, Yohe GW, eds. Climate change impacts in the United States: the Third National Climate Assessment. US Global Change Research Program, 6–1-nn. http://nca2014.globalchange.gov/report/sectors/agriculture. Last accessed 2 January 2015. IPCC. 2007. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Core Writing Team, Pachauri RK, Reisinger A, eds. Geneva: Intergovernmental Panel on Climate Change. http://www.ipcc.ch/publications_and_data/publications_and_ data_reports.shtml. Last accessed 2 January 2015. IPCC. 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Stocker TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM, eds. Cambridge, UK/New York: Cambridge University Press. http://www. ipcc.ch/publications_and_data/publications_and_data_reports.shtml Last accessed 2 January 2015. Ishikawa S, Hare MC, Kettlewell PS. 2012. Effects of strobilurin fungicide programmes and fertilizer nitrogen rates on winter wheat: leaf area, dry matter yield and nitrogen yield. Journal of Agricultural Science 150, 427–441. Jabs T, Pfirmann J, Schafer S, Wu YX, Tiedemann A. 2002. Antioxidative and antisenescence effects of the strobilurin pyraclostrobin in plants: a new strategy to cope with environmental stress in cereals. In: BCPC Conference—Pests and Diseases 2002 , pp. 941–946. Brighton, UK, 18–21 November. https://www.seedquest.com/id/b/basf/vitality/ articles/Antioxidative_and_antisenescence_effects.pdf. Last accessed 2 January 2015. Kant S, Seneweera S, Rodin, J Materne M, Burch D, Rothstein SJ, Spangenberg G. 2012. Improving yield potential in crops under elevated CO2: integrating the photosynthetic and nitrogen utilization efficiencies. Frontiers in Plant Science 3, 162. Karl TR, Melillo JM, Peterson TC. 2009. Global climate change impacts in the United States. Cambridge: Cambridge University Press. Ker K, Seguin P, Driscoll BT, Fyles JW, Smith DL. 2012. Switchgrass establishment and seeding year production can be improved by inoculation with rhizosphere endophytes. Biomass and Bioenergy 47, 295–301. Khan ZR, Midega CA, Pittchar JO, Murage AW, Birkett MA, Bruce TJ, Pickett JA. 2014. Achieving food security for one million subSaharan African poor through push-pull innovation by 2020. Philosophical Transactions of the Royal Society B: Biological Sciences 369, 20120284. Kirkegaard JA, Hunt JR, McBeath TM, et al. 2014. Improving water productivity in the Australian grains industry—a nationally coordinated approach. Crop and Pasture Science 65, 583–601. Kirkegaard JA, Lilley JM, Howe GN, Graham JM. 2007. Impact of subsoil water use on wheat yield. Australian Journal of Agricultural Research 58, 303.

Glick BR. 2012. Plant growth-promoting bacteria: mechanisms and applications. Scientifica (Cairo) 2012, 963401.

Kyei-Boahen S, Zhang L. 2006. Early-maturing soybean in a wheat– soybean double-crop system. Agronomy Journal 98, 295–301.

Gornall J, Betts R, Burke E, Clark R, Camp J, Willett K, Wiltshire A. 2010. Implications of climate change for agricultural productivity in the early twenty first century. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 2973–2989.

Lawlor D W. 2013. Genetic engineering to improve plant performance under drought: physiological evaluation of achievements, limitations, and possibilities. Journal of Experimental Botany 64, 83–108.

Grossmann K, Retzlaff G. 1997. bioregulatory effects of the fungicidal strobilurin kresoxim-methyl in wheat (Triticum aestivum). Pesticide Science 50, 11–20.

Leakey AD, Uribelarrea M, Ainsworth EA, Naidu SL, Rogers A, Ort DR, Long SP. 2006. Photosynthesis, productivity, and yield of maize are not affected by open-air elevation of CO2 concentration in the absence of drought. Plant Physiology 140, 779–790.

Grossmann K, Kwiatkowski J, Caspar G. 1999. Regulation of phytohormone levels, leaf senescence and transpiration by the strobilurin kresoxim-methyl in wheat (Triticum aestivum). Journal of Plant Physiology 154, 805–808.

Lee H, Rustgi, S Kumar, N, Burke I, Yenish JP, Gill KS, Von Wettstein D, Ullrich SE. 2011. Single nucleotide mutation in the barley acetohydroxy acid synthase (AHAS) gene confers resistance to imidazolinone herbicides. Proceedings of the National Academy of Sciences, USA 108, 8909–8913.

Habben JE, Bao X, Bate NJ, et al. 2014. Transgenic alteration of ethylene biosynthesis increases grain yield in maize under field droughtstress conditions. Plant Biotechnology Journal 12, 685–693.

Lee KY, Townsend J, Black M, Chuil CF, Mazur B, Dunsmuir P, Bedbrook J. 1988. The molecular basis of sulfonylurea herbicide resistance in tobacco. EMBO Journal 7, 1241–1248.

Hatfield J, Boote K, Fay P, et al. 2008. Agriculture. In The effects of climate change on agriculture, land resources, water resources, and

Lemerle D, Gill GS, Murphy CE, Walker SR, Cousens RD, Mokhtari S, Peltzer SJ, Coleman R, Luckett DJ. 2001. Genetic improvement and

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Celebi SZ, Demir S, Celebi R, Durak ED, Yilmaz IH. 2010. The effect of arbuscular mycorrhizal fungi (AMF) applications on the silage maize (Zea mays L.) yield in different irrigation regimes. European Journal of Soil Biology 46, 302–305.

biodiversity in the United States , p. 362. A Report by the US Climate Change Science Program and the Subcommittee on Global Change Research. Washington, DC. http://www.usda.gov/oce/climate_change/ SAP4_3/CCSPFinalReport.pdf. Last accessed 2 January 2015.

Planning food security  |  Page 15 of 16 agronomy for enhanced wheat competitiveness with weeds. Australian Journal of Agricultural Research 52, 527–548.

Sala CA, Bulos M, Echarte M, Whitt SR, Ascenzi R. 2008. Molecular and biochemical characterization of an induced mutation conferring imidazolinone resistance in sunflower. Theoretical and Applied Genetics 118, 105–112. Salekdeh G H, Reynolds M, Bennett J, Boyer J. 2009. Conceptual framework for drought phenotyping during molecular breeding. Trends in Plant Science 14, 488–496. Sayer J, Cassman KG. 2013. Agricultural innovation to protect the environment. Proceedings of the National Academy of Sciences, USA 110, 8345–8348. Shanker AK, Venkateswarlu B (eds). 2011. Abiotic stress in plants— mechanisms and adaptations. Rijeka, Croatia: InTech. Singer JW, Sauer TJ, Blaser BC, Meek DW. 2007. Radiation use efficiency in dual winter cereal–forage production systems. Agronomy Journal 99, 1175–1179. Slattery RA, Ainsworth EA, Ort DR. 2013. A meta-analysis of responses of canopy photosynthetic conversion efficiency to environmental factors reveals major causes of yield gap. Journal of Experimental Botany 64, 3723–3733. Smith P,, Gregory P J. 2013. Climate change and sustainable food production. Proceedings of the Nutrition Society 72, 21–28. Soares-Filho B, Moutinho P, Nepstad D, et al. 2010. Role of Brazilian Amazon protected areas in climate change mitigation. Proceedings of the National Academy of Sciences, USA 107, 10821–10826. Stevenson JR, Villoria N, Byerlee D, Kelley T, Maredia M. 2013. Green Revolution research saved an estimated 18 to 27 million hectares from being brought into agricultural production. Proceedings of the National Academy of Sciences, USA 110, 8363–8368. Strzepek K,, Boehlert B. 2010. Competition for water for the food system. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 2927–2940. Swamy BP, Kumar, A. 2013. Genomics-based precision breeding approaches to improve drought tolerance in rice. Biotechnology Advances 31, 1308–1318. Tan S, Evans R, Singh B. 2006. Herbicidal inhibitors of amino acid biosynthesis and herbicide-tolerant crops. Amino Acids 30, 195–204. Tan S, Evans RR, Dahmer ML, Singh BK, Shaner DL. 2005. Imidazolinone-tolerant crops: history, current status and future. Pest Management Science 61, 246–257. Timmusk S, Wagner EG. 1999. The plant-growth-promoting rhizobacterium Paenibacillus polymyxa induces changes in Arabidopsis thaliana gene expression: a possible connection between biotic and abiotic stress responses. Molecular Plant–Microbe Interactions 12, 951–959. Tollenaar M, Wu J. 1999. Yield improvement in temperate maize is attributable to greater stress tolerance. Crop Science 39, 1597–1604. Trabelsi D, Mhamdi R. 2013. Microbial inoculants and their impact on soil microbial communities: a review. BioMed Research International 2013, 863240. Turral H, Burke J, Faurès J-M. 2011. Climate change, water and food security. FAO Water Report 36. Rome: Food and Agriculture Organization of the United Nations. van Ittersum MK, Cassman KG. 2013. Yield gap analysis—rationale, methods and applications—introduction to the Special Issue. Field Crops Research 143, 1–3. Waller F, Achatz B, Baltruschat H, et al. 2005. The endophytic fungus Piriformospora indica reprograms barley to salt-stress tolerance, disease resistance, and higher yield. Proceedings of the National Academy of Sciences, USA 102, 13386–13391. Waltz E. 2014. Beating the heat. Nature Biotechnology 32, 611–617. Wang L, Feng Z, Schjoerring JK. 2013. Effects of elevated atmospheric CO2 on physiology and yield of wheat (Triticum aestivum L.): a metaanalytic test of current hypotheses. Agriculture, Ecosystems & Environment 178, 57–63. Wiersma PA, Schmiemann MG, Condie JA, Crosby WL, Moloney MM. 1989. Isolation, expression and phylogenetic inheritance of an acetolactate synthase gene from Brassica napus. Molecular and General Genetics 219, 413–420.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Lobell DB, Roberts MJ, Schlenker W, Braun N, Bertis B, Little BB, Rejesus RM, Hammer GL. 2014. Greater sensitivity to drought accompanies maize yield increase in the U.S. midwest. Science 344, 516–519. Long SP, Ort DR. 2010. More than taking the heat: crops and global change. Current Opinion in Plant Biology 13, 241–248. Long SP, Ainsworth EA, Leakey AD, Nosberger J, Ort DR. 2006a. Food for thought: lower-than-expected crop yield stimulation with rising CO2 concentrations. Science 312, 1918–1921. Long SP, Zhu XG, Naidu SL, Ort DR. 2006b. Can improvement in photosynthesis increase crop yields? Plant, Cell and Environment 29, 315–330. McAllister CH, Beatty PH, Good AG. 2012. Engineering nitrogen use efficient crop plants: the current status. Plant Biotechnology Journal 10, 1011–1025. Mengual C, Schoebitz M, Azco, R, Roldan A. 2014. microbial inoculants and organic amendment improves plant establishment and soil rehabilitation under semiarid conditions. Journal of Environmental Management 134, 1–7. Muller C, Cramera W, Harea W L, Lotze-Campen H. 2011. Climate change risks for African agriculture. Proceedings of the National Academy of Sciences, USA 108, 4313–4315. Nason MA, Farrar J, Bartlett, D. 2007. strobilurin fungicides induce changes in photosynthetic gas exchange that do not improve water use efficiency of plants grown under conditions of water stress. Pest Management Science 63, 1191–1200. Nelson GC, Rosegrant M W, Koo J, et al. 2009. Climate change: impact on agriculture and costs of adaptation. Washington, DC: International Food Policy Research Institute. Nelson KA, Meinhardt CG. 2011. Foliar boron and pyraclostrobin effects on corn yield. Agronomy Journal 103, 1352–1358. Nepstad D, Soares-Filho BS, Merry F, et al. 2009. The end of deforestation in the Brazilian Amazon. Science 326, 1350–1351. Oldach KH, Peck DM, Cheong J, Williams KJ, Nair RM. 2008. Identification of a chemically induced point mutation mediating herbicide tolerance in annual medics (Medicago spp.). Annals of Botany 101, 997–1005. Passioura J. 2006. Increasing crop productivity when water is scarce— from breeding to field management. Agricultural Water Management 80, 176–196. Passioura J. 2007. The drought environment: physical, biological and agricultural perspectives. Journal of Experimental Botany 58, 113–117. Passioura JB, Angus JF. 2010. Improving productivity of crops in waterlimited environments. Advances in Agronomy 106, 37–75. Phalan B, Green R, Balmford A. 2014. Closing yield gaps: perils and possibilities for biodiversity conservation. Philosophical Transactions of the Royal Society B: Biological Sciences 369, 20120285. Piesse J, Thirtle C. 2010. Agricultural R&D, technology and productivity. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 3035–3047. Pingali PL. 2012. Green revolution: impacts, limits, and the path ahead. Proceedings of the National Academy of Sciences, USA 109, 12302–12308. Redman RS, Sheehan KB, Stout RG, Rodriguez RJ, Henson JM. 2002. Thermotolerance generated by plant/fungal symbiosis. Science 298, 1581. Reynolds MP, Mujeeb-Kazi A, Sawkins M. 2005. Prospects for utilising plant-adaptive mechanisms to improve wheat and other crops in droughtand salinity-prone environments. Annals of Applied Biology 146, 239–259. Richards RA, Rebetzke GJ, Watt M, Condon AGT, Spielmeyer W, Dolferus R. 2010. Breeding for improved water productivity in temperate cereals: phenotyping, quantitative trait loci, markers and the selection environment. Functional Plant Biology 37, 85–97. Rodriguez RJ, Henson J, Van Volkenburgh E, Hoy M, Wright L, Beckwith F, Kim YO, Redman RS. 2008. Stress tolerance in plants via habitat-adapted symbiosis. ISME Journal 2, 404–416. Rustgi S, Matanguihan J, Mejias JH, Gemini R, Brew-Appiah RA, Wen N, Osorio C, Ankrah N, Murphy KM, Von Wettstein D. 2014.

Assessment of genetic diversity among barley cultivars and breeding lines adapted to the US Pacific Northwest, and its implications in breeding barley for imidazolinone-resistance. PLoS One 9, e100998.

Page 16 of 16 | McKersie Wullschleger SD, Tschaplinski TJ, Norby RJ. 2002. plant water relations at elevated CO2—implications for water-limited environments. Plant, Cell and Environment 25, 319–331. Yang J, Kloepper JW, Ryu CM. 2009. Rhizosphere bacteria help plants tolerate abiotic stress. Trends in Plant Science 14, 1–4. Yang S, Vanderbeld B, Wan, J, Huang Y. 2010. Narrowing down the targets: towards successful genetic engineering of drought-tolerant crops. Molecular Plant 3, 469–490. Zeigler RS, Mohanty S. 2010. Support for international agricultural research: current status and future challenges. New Biotechnology 27, 565–572. Zhao PL, Wang L, Zhu XL, Huang X, Zhan CG, Wu JW, Yang GF. 2010. Subnanomolar inhibitor of cytochrome bc1 complex designed by

optimizing interaction with conformationally flexible residues. Journal of the American Chemical Society 132, 185–194. Zhu XG, Long SP, Ort DR. 2010. Improving photosynthetic efficiency for greater yield. Annual Review Plant Biology 61, 235–261. Zhu XG, Wang YU, Ort DR, Long SP. 2013. E-photosynthesis: a comprehensive dynamic mechanistic model of C3 photosynthesis: from light capture to sucrose synthesis. Plant, Cell and Environment 36, 1711–1727. Ziska LH, Bunce JA, Shimono H, et al. 2012. Food security and climate change: on the potential to adapt global crop production by active selection to rising atmospheric carbon dioxide. Proceedings of the Royal Society B: Biological Sciences 279, 4097–4105.

Downloaded from http://jxb.oxfordjournals.org/ at Selcuk University on January 31, 2015

Planning for food security in a changing climate.

The Intergovernmental Panel on Climate Change and other international agencies have concluded that global crop production is at risk due to climate ch...
757KB Sizes 2 Downloads 10 Views