Sie sind auf Seite 1von 16

Journal of Experimental Botany, Vol. 66, No. 12 pp.

3435–3450, 2015
doi:10.1093/jxb/eru547  Advance Access publication 22 January 2015

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: mckersie@nc.rr.com

Received 3 November 2014; Revised 11 December 2014; Accepted 14 December 2014

Abstract

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


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 innova-
tions 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) increase, but for most crops in most regions, negative effects
has concluded that human-induced emission of greenhouse on yield are expected. Higher temperatures are expected
gases including CO2 has increased global temperatures and to exacerbate the effects of water shortage in regions that
predicts that our climate will be continually changing at a receive reduced rainfall and to negate any advantage in those
relatively rapid rate during this century. Higher temperatures, regions that receive increased precipitation. Higher tempera-
altered precipitation patterns, and the increased incidence of tures are also expected to reduce yields by directly impact-
drought will directly impact crop production and threaten ing the biochemistry and physiology of the crops and at
our food supply (IPCC, 2007, 2013). Their common conclu- the same time encouraging weed and pest proliferation. In
sion and other analyses conducted over the past decade is addition, sea levels are predicted to rise, affecting drainage in
that crop production everywhere runs some risk of being coastal areas, particularly in low-lying deltas, and may result
negatively affected by this climate change (Hatfield et  al., in saline intrusion into coastal agricultural regions. Erosion
2008, 2014; Karl et  al., 2009; Gornall et  al., 2010; Muller is likely to increase due to the increased incidence of severe
et  al., 2011; Turral et  al., 2011). For some crops grown in storms. Soil moisture profiles and runoff will also change in
cooler regions, higher temperatures will provide a yield 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: journals.permissions@oup.com
3436  | McKersie

Not only is agriculture likely to be impacted by climate change, development of new products and increased food secu-
but agriculture also contributes significantly to the greenhouse rity. I  hope this overview helps both young and established
gas problem by producing emissions that drive climate change. researchers as they organize and communicate their research
Estimates of all human-induced greenhouse gas emissions from programmes to address regional food security issues.
global agriculture vary from 9% (Karl et al., 2009; http://www.
epa.gov/climatechange/ghgemissions/sources/agriculture.html, Options to mitigate the effects on crop
last accessed 2 January 2015) to 30% (Smith and Gregory, 2013)
production
in various regions. In the United USA, agriculture produces 8.6%
of the nation’s total greenhouse gas emissions, including 80% of Society has several options to ensure food security and to
the nitrous oxide emissions and 31% of the methane emissions mitigate the risks to crop production posed by climate change
(Karl et al., 2009). These gases in combination with CO2 are the and by societal changes. However, not all will be effective in
main gases responsible for global climate change. Deforestation the short term, and some may not be practical due to alter-
in the Brazilian Amazon to enable soybean [Glycine max (L.) native demands on limited resources, such as water. The fol-
Merr.)] and cattle production contributes 2–5% of the global lowing is a brief summary of some options, which are not
carbon emissions (Nepstad et al., 2009). mutually exclusive.
In addition, the global agricultural system must provide
about 70% more food for a global population estimated to Reduce CO2 emissions
be 9 billion or more by 2050 (Fischer et al., 2005; Smith and
Gregory, 2013). A  further complication is that food prefer- Global political efforts to mitigate the effects of climate
change have focused on reducing greenhouse gas emissions.

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


ences are changing. The consumption of relatively low-cost
carbohydrates from cereals is being replaced by increased Unfortunately, these reductions may be too late to prevent
demand for milk, meat, fruits, and vegetables. Since the water- an impact on crop production. Even if emission of green-
use efficiency of animal production is much lower than that house gases was stopped today, climate change and its result-
of cereal crops, not only is extra primary production from ing impacts on agriculture will continue to occur due to the
pastures, rangelands, and arable land required to meet these effects of gases that have already been released (Karl et al.,
demands but also more water (Turral et al., 2011). For exam- 2009). Although this a critical goal that may prevent cata-
ple, the water used per unit of energy in the food produced is strophic effects in the next century, other short-term mitiga-
10 times greater in dairy products than cereals. Consequently, tions are needed to meet the immediate challenges to crop
as the population’s demand for food quality changes, the production.
demand for water in agricultural production systems is likely
to increase. Improvements to the global agricultural infra- Utilize high CO2 in photosynthesis
structure for storage and distribution of food will help but
will not completely solve the problem. The physiology, biochemistry, and morphology of crop
Therefore, the agricultural community must address three plants is impacted directly and indirectly by the environment,
opposing demands on crop production simultaneously to notably by CO2, temperature, and water. Consequently, the
ensure food security in the future (Fig.  1). The demand for same crop is expected to grow, develop, and function differ-
more high-quality food must be reconciled with the need to ently in a future environment. The effects of environment
have less environmental impact and to use fewer resources on the physiology, growth, and ultimately yield of crops are
in a more stressful and uncertain environment caused by cli- complex. Consequently, mathematical models are often used
mate change. My thesis is that an integrated research plan to predict the physiological effects and to estimate future crop
merging agronomic and genetic approaches has the greatest yields in specific environments (Karl et al., 2009; Zhu et al.,
probability of success. In this article, I present a template for 2013). Some propose that these responses to CO2 may par-
a research plan that includes an overview of the current bio- tially ameliorate the negative impacts of climate change and
technology tools that might be used in this quest. I  use the water deficits on yield (Wullschleger et al., 2002; Shanker and
development of drought tolerance in maize (Z. mays L.) for Venkateswarlu, 2011). Others propose that plant breeding
the US corn belt as an example that has recently released new programmes (Ziska et al., 2012) or genetic engineering pro-
hybrids to illustrate this research plan. Regional adaptation grammes (Ainsworth et al., 2008) should focus on capturing
of the plan for each crop will be needed to ensure commercial opportunities from the high-CO2 environment.
Concerns have been expressed that these computer simula-
tion 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
Fig. 1.  The three constraints that will be imposed on crop production in by rising CO2 in the absence of drought. Even soybean,
the next 40 years. which is a C3 plant, does not respond to increased CO2 as
Planning food security  |  3437

predicted by the kinetics of ribulose-1,5-bisphosphate car- done for soybean and winter wheat (Triticum aestivum L.)
boxylase/oxygenase (RuBisCo) due to energy limitations rotation in some regions might become more widespread;
(Crafts-Brandner and Salvucci, 2000). The latest IPCC •• altered planting and/or harvest dates to avoid stressful peri-
(2013) and US Department of Agriculture (USDA) (Hatfield ods, such as drought or flooding;
et al., 2014) reports recognize that negative effects of climate •• water conservation by altered tillage practices, such as
change on crop production may be larger than previously no-till;
anticipated. Therefore, we cannot rely on the existing physiol- •• cropping on marginal land that has poor water or nutrient
ogy and biochemistry of our crops that have evolved in past availability;
low-CO2 environments to prepare them for the future high- •• yield stability in environments experiencing periodic heat or
CO2 environments. drought stress; and
•• increased use of biological or chemical additives to enhance
Increase crop irrigation stress tolerance and improve the overall health of the plant.

An obvious mitigation to reduce the impact of drought is to These changes will require changes in agricultural infrastruc-
increase the use of irrigation for crop production by expand- ture for storage and distribution, farming equipment, crop
ing the water management infrastructure. Where possible, management, and most likely the crop’s physiology, growth,
more irrigation will meet regional needs leading to higher and development, as well as its response to the environment.
crop productivity. However, increased irrigation is unlikely to There is considerable risk in this approach, and concerns
be a global solution. A report by the Food and Agriculture have been expressed as to whether any of these changes in
crop production will be successful (Sayer and Cassman,

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


Organization of the United Nations (FAO) (Turral et  al.,
2011) estimated that, based on population growth alone 2013). First, there are biophysical and environmental limits
to 2050, without factoring in global warming and climate on the conversion of light energy into biomass (i.e. photosyn-
change, crop production will require 11% more water for irri- thesis) and then into economic yield (i.e. seed biomass) that
gation. However, agriculture is not the only segment of soci- may restrict further increases in crop yield. Secondly, the need
ety that will require more water resources in the future due to to intensify agriculture on existing farmland may destroy the
population growth. Strzepek and Boehlert (2010) concluded environment and contribute to climate change and loss of
that agriculture’s primary competition for water will come biodiversity (Phalan et  al., 2014). Thirdly, institutional and
from legislative environmental flow requirements, and from regulatory obstacles may prevent implementation (e.g. dereg-
municipal and industrial water demands. They estimated that ulation of genetically modified plants).
18% less water is likely to be available for global agriculture
by 2050. Consequently, this competition for our dwindling Develop new cultivars
water resources is likely to prevent increased use of irrigation,
Most reports on climate change propose the development of
except in very specific cases.
new crop cultivars that have increased yield with fewer inputs
or have increased yield stability. These hypothetical crop culti-
Increase cultivated land area vars have a higher tolerance to drought and high temperature,
increased resistance to more fungal diseases and insect pests, and
Another option is to expand production of our current
a higher yield potential on both marginal and good farmland.
crop cultivars and hybrids to marginal and deforested land.
Currently, most plant breeding strategies use high-through-
Increasing global food production to meet the projected
put field evaluation of thousands of lines grown in the tar-
demand in 2050 could potentially require an area equiva-
get environment on small field plots (Richards et  al., 2010;
lent to that of the Indian subcontinent (Phalan et al., 2014).
Cooper et  al., 2014). This poses a potential problem if the
Deforestation in several tropical regions is a large source of
breeding programme is targeting a high-CO2 future environ-
greenhouse gas emissions (Galford et al., 2010), contributes
ment that does not exist currently. Our options in developed
to climate change (Soares-Filho et al., 2010), and negatively
countries are to locate testing sites in highly water-limited, hot
impacts biodiversity (Phalan et  al., 2014). Thus, continued
regions or to use controlled environments with CO2 supple-
deforestation is clearly not a preferred option, but this option
mentation. These options may not be available in developing
will be implemented even more extensively if or when food
countries with limited resources. Nonetheless, we must find
shortages change the economic and political circumstances.
ways to optimize the existing biochemical and physiological
traits in our crops for this future field environment.
Develop new crop management technology A new strategy is needed to create genetic diversity for
The ‘yield gap’ between the theoretical and actual crop yields use in plant breeding if we are to successfully meet society’s
obtained in our current production areas is often cited as a expectations. We must do the following:
target for innovation and technological change. Some of the
changes envisioned include: •• predict what the future crop environments will be;
•• envision new crop production systems to capture any
•• more crops grown per year on the same land: double crop- opportunities and to mitigate the risks in those potentially
ping of summer and winter annual crops such as currently different environments;
3438  | McKersie

•• create new cultivars that will have consistently higher yield agronomy will develop new cultivars and new crop manage-
in these new crop production systems; and ment practices to stabilize or increase yield. These will be cou-
•• use these new cultivars to implement the new food produc- pled with infrastructure improvements to transportation and
tion system. irrigation systems to increase food supply for a growing pop-
ulation (Karl et al., 2009; Nelson et al., 2009; Muller et al.,
In other words, as many others have proposed previously 2011; Turral et al., 2011). This opinion has been re-enforced
(Beddington, 2010; Piesse and Thirtle, 2010; Zeigler and most recently by the US Climate Change Science Program in
Mohanty, 2010; McAllister et  al., 2012; Pingali, 2012), their draft report (Hatfield et al., 2014): ‘Agriculture has been
I believe that we need a Second Green Revolution. The fol- able to adapt to recent changes in climate; however, increased
lowing are some suggestions on how this might be achieved. innovation will be needed to ensure the rate of adaptation
of agriculture and the associated socioeconomic system can
Planning the Second Green Revolution keep pace with climate change over the next 25 years.’ In other
words, society (on the advice of many agricultural experts)
Agronomists around the world are working with farmers and expects that a Second Green Revolution is simply a matter
achieving considerable success in developing new crop produc- of allocating financial and human resources to the problem.
tion practices for a changing climate, a recent example being This may be one reason why society has fewer concerns about
reported by Kirkegaard et al. (2014) in Australia. At the same the future impacts of climate change on food security than is
time, plant breeders are using genetics and biotechnology to warranted.
develop crops that have greater yield stability in our current The initial strategy employed in the first Green Revolution
production systems (for example, in maize: Castiglioni et al.,

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


to increase yield in rice and wheat was to increase the appli-
2008; Cooper et al., 2014; Habben et al., 2014). These initia- cation of nitrogen fertilizer. The problem was that applying
tives may be sufficient to meet our needs for food security in more nitrogen to the existing cultivars increased the height
a hot, dry, high-CO2 environment. Nonetheless, I recommend of the plants, which increased lodging and actually reduced
that we do not take the risk and become complacent. My the- harvestable yields. The solution was to grow short-statured
sis is that is if we coordinate the use of both agronomic and cultivars that did not lodge as readily at high rates of nitrogen
genetic approaches to achieve the shared goal, we will greatly fertilization. However, short-statured, agronomically accept-
improve the probability of food security in the future. able cultivars did not exist, so it was essential to introgress
The first Green Revolution led by Norman Borlaug applied mutated alleles of genes controlling plant height from non-
high inputs of nitrogen fertilizer to responsive short-statured, agronomic germplasm into new cultivars.
short-season cultivars of rice (Oryza sativa L.) and wheat, Once nitrogen was removed as a limiting factor, the larger
often with irrigation, to realize the potential yield (Nelson plants transpired more. Water now became the new growth-
et al., 2009; Pingali, 2012). The greatest impact was in Asia limiting factor in many regions. Thus, the full yield potential
and it is credited as the fundamental factor leading to the eco- of the new cultivars in these environments was achieved only
nomic development of that region today. However, the Green by supplying more nitrogen under irrigation. Infrastructure
Revolution had very little impact on Africa, either in terms of improvements to irrigation systems, fertilizer manufacturing,
food security or wealth creation, perhaps due to the relatively and distribution, and to seed production and handling were
small potential for irrigation in Africa (Turral et  al., 2011). also required. In other words, the whole production system
Increases in cereal yields as a result of widespread adoption had to change to achieve the goals set out in the first Green
of these improved crop management practices and new culti- Revolution.
vars saved natural ecosystems from being converted to agri- Borlaug was limited at the time to using existing genetic
culture (Stevenson et al., 2013). Although their calculations variability in rice and wheat to create his new cultivars. Since
are complex, they made a conservative estimate that, in the then, genetic engineering technology has emerged as a power-
absence of the first Green Revolution, an additional 18–26 ful tool leading to the commercial release of herbicide and
Mha, including 2 MHa of additional deforestation, would insect tolerance in maize, soybean, and other crops. This tool
have been required to produce the food needed in 2004. of plant biotechnology can now be used to tailor other genetic
Infrastructure, including storage and transportation facilities, changes that enable alternative crop management practices.
were critical to success. However, the FAO reports that, even Merging these two approaches into what may eventually
today, the amount of cereals lost during harvesting, threshing, be called a Second Green Revolution is required to meet the
and storage in India is equivalent to about 10% of their annual demands for food security over the next several decades. The
production, or about 15 Mt (http://www.fao.org/wairdocs/ lessons learned from the success of the first Green Revolution
x5002e/x5002e00.htm; http://online.wsj.com/news/articles/SB1 and from the first generation of genetically modified crops
0001424052702304356604577339402041821044, last accessed provide a template for research strategies on how we might
2 January 2015). This is comparable to the annual wheat pro- approach implementation in six sequential steps (Fig. 2):
duction of Australia, which is about 20 Mt annually (http://
en.wikipedia.org/wiki/International_wheat_production_statis- 1. Envision changes in crop management that will increase
tics, last accessed 2 January 2015). crop productivity in the predicted future environment.
It is society’s expectation that crop productivity research 2. Identify physiological traits that enable those changes in
in plant biotechnology, genetics, physiology, breeding, and crop management.
Planning food security  |  3439

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


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.

3. Identify genes that regulate those physiological traits. that might be used in the future to increase yield is beyond
4. Create genetic diversity for those physiological traits using the scope of this article. The reader is referred to the follow-
the identified genes. ing articles on this topic (Fischer et al., 2014; Passioura and
5. Breed new regionally adapted cultivars to incorporate Angus, 2010) as well as a special issue of Field Crops Research
those physiological traits into agronomically acceptable on yield gap analysis (van Ittersum and Cassman, 2013). The
cultivars. following sections discuss only a few of many potential new
6. Implement the envisioned changes in crop management crop production practices to illustrate the concept of envi-
using the newly developed cultivars. sioning in step 1.  Whether any represent significant new
global opportunities or whether any require new crop culti-
The restrictions are that any changes in crop production
vars has yet to be established.
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 Innovation example 1: double cropping and intercropping
yield stability across diverse environments from year to year Double cropping of soybean after harvesting wheat is cur-
and from location to location. So the targets of the Second rently a common production system in the mid-southern
Green Revolution differ from the first because increased USA (Kyei-Boahen and Zhang, 2006). Cropping systems
nitrogen fertilization and increased irrigation are unlikely to based on a single crop waste large proportions of key inputs
be sustainable practices in the future climate. 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) com-
Step 1: envision changes in crop management system
pared wheat–soybean single and double-cropping systems
To begin, we need to envision a new crop management system and confirmed that double cropping improved resource
with the potential to be a breakthrough that leads to an incre- capture and efficiency, but the impact was much greater for
mental yield increase in the predicted future environment. water than for radiation. They attributed the difference to
This step is essential in the research plan to enable commu- storable (water) versus non-storable resources (radiation)
nication to stakeholders and funding agencies and to provide and proposed that further research in their region empha-
focus, context, and mandate for the interdisciplinary teams. size improvements in radiation capture. Many studies have
One way to categorize and prioritize alternatives is to dis- been conducted on this cropping system in the USA and
sect yield into components (Fig.  3) that might be simpler have shown that tillage practices and the availability of
targets to change by modifying crop management. One or adequate soil moisture when planting soybean are among
more of the following opportunities might be a target but the the more critical factors to consider http://oilseeds.okstate.
selection must be tailored for specific regional environments. edu/production-information/soybean/PSS-2137%20(dou-
A  comprehensive review of the crop production practices ble%20crop%20soybean).pdf, last accessed 2 January 2015;
3440  | McKersie

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 com-
bined 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 dis-
tribution system, relevant policy changes, farmer and stake-
holder training, and infrastructure improvement.
Fig. 3.  Targets for improvement in new crop management systems. These double and intercropping systems use current culti-
Yield potential (Yp) is the theoretical maximal yield of the crops(s) in
vars that were developed for single annual crop production
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 systems. It is envisioned that different genetics for maturity or
efficiency of conversion of that light energy into biomass, and its harvest for other agronomic attributes would enable a better fit into
index, which is the proportion of biomass portioned into economic (e.g. new double or intercropping systems.

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


seed) yield. Photosynthetic leaf area and its duration over the growing
season are factors that modulate Yp. Therefore, management practices Innovation example 2: biological soil additives
that increase the duration of light capture on an annual basis, such as
double cropping, or intercropping, might be expected to increase Yp.
Microbial symbionts associated with plant roots confer biotic
Annual yield (Ya) is the economic yield delivered into the food chain per and abiotic stress tolerance to their plant hosts. Positive
unit of land area per year. The difference between Yp and actual yield is effects have been reported for both plant growth-promoting
due to various losses, which are represented as pies in the chart. The rhizobacteria (Timmusk and Wagner, 1999; Yang et al., 2009;
size of the pie slice for each will vary from region to region and from Glick, 2012; Mengual et al., 2014) and arbuscular mycorrhi-
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
zal fungi (Redman et al., 2002; Waller et al., 2005; Rodriguez
converted into economic yield in a grower’s field. Yg is large if a factor et  al, 2008; Celebi et  al., 2010). For example, inoculating
other than light energy limits yield, which is almost always the case. switchgrass seeds with plant growth-promoting rhizobacteria
Yg includes factors inherent to the crop such as nutrient-use efficiency resulted in more and taller tillers and a 40% higher yield com-
and environmental factors such as nutrient availability that may classify pared with than those not inoculated in a low-nitorgen input
certain regions as marginal. Management practices that conserve water,
such as no-till, or which enhance uptake of growth-limiting nutrients,
production system (Ker et al., 2012).
such as biological soil additives, would be expected to reduce the The review of Trabelsi and Mhamdi (2013) on the impact of
yield gap. Yield volatility (Yv) is a function of the plant’s response to its microbial inoculants on soil microbial communities concluded
environment. Management practices that reduce the impact of biotic and that both soil or seed treatments may change the indigenous
abiotic stress, such as altered planting dates or the application of health- microbial populations, at least temporarily, and thereby improve
promoting chemicals, would be expected to reduce volatility. Losses in
seed yield during harvest occur due to shattering or lodging that leave
plant performance. The current working hypothesis is that
seed in the field, or due to deterioration such as sprouting in wheat. microbes provide the crop with novel nutritional and defence
Direct combining of wheat to avoid swathing and the risk of sprouting pathways, and at the same time modulate plant biochemical
damage, which is common in some regions, would reduce harvest loss. pathways, thereby altering plant phenotypes. This potential
Other losses in storage or distribution can also be quite significant before benefit of microbial additives may provide a novel strategy for
the grain enters the food chain.
reducing the yield gap and mitigating the impacts of global
climate change on crops (Berendsen et  al., 2012). Genomic
http://www.lsuagcenter.com/NR/rdonlyres/5673F10F- technologies have the potential to facilitate the development of
32C8-4465-9A40-4D6BD8D6FCA9/45621/pub3053dou- these microbial approaches (Coleman-Derr and Tringe, 2014).
blecroppingsoybeanswheatHIGHRES.pdf, last accessed 2 Although still very much in its technical infancy, microbial soil
January 2015). inoculants may be developed as an agronomic management tool
In another example, winter cereal production systems that could be used to mitigate the effects of climate change and
in the northern USA are considered to be inefficient with provide greater yield stability in future environments. Whether
respect to the capture of radiation during the year because plant genetics can be used to enhance the microbial–plant inter-
the majority of fields lie fallow after grain harvest until the action commercially has yet to be established.
following cropping season. To maximize radiation capture
and biomass production, a common cropping system utilizes Innovation example 3: plant health-promoting chemical
frost-seeding techniques to intercrop the winter cereal with additives
an annual legume, usually red clover (Trifolium pratense L.). A long-term goal of research on plant growth regulators
In such a system, winter triticale (×Triticosecale Wittmack) has been improved tolerance of drought and other abiotic
and wheat differ in their effect on the forage crop because red stresses, but there have been few commercial successes until
clover’s plant growth rate and radiation-use efficiency interact recently. The strobilurin fungicides are promoted not only
Planning food security  |  3441

for their fungicidal properties but also for their effects on biomass production. This not only reduces evaporative losses
stress tolerance and yield. Strobilurins are effective against but also inhibits the growth of weeds (Lemerle et al., 2001).
several different plant-pathogenic fungi and give high levels Other innovations involving better rooting patterns and a
of protection against a wide range of crop diseases. (For healthier root system can be postulated to further improve
more details on these fungicides, see the following reviews: this cropping system. Small amounts of subsoil water can
Bartlett et  al., 2002; Balba, 2007; Zhao et  al., 2010). In enhance grain yield (Kirkegaard et al., 2007). Consequently,
addition, the strobilurin fungicides have been reported in plants that have deep roots are expected to have greater yield
numerous studies to increase seed yield in the absence of but only if there is water available in the subsoil, if there is
detectable levels of fungal infection, prompting the com- no hardpan from soil compaction, and if there is no subsoil
mercial application of strobilurins to improve tolerance of salinity. Thus, the potential for this target is very regional.
drought and other stress resulting in reduced variable yield
(Grossmann and Retzlaff, 1997; Grossmann et  al., 1999; Step 2: identify traits that enable changes in crop
Jabs et al., 2002; Nason et al., 2007; Nelson and Meinhardt, management
2011; Ishikawa et al., 2012). A growing number of produc-
ers have responded by incorporating fungicide treatments Once the new crop management system is envisioned, it is
into their management programmes in attempts to increase necessary to determine which (if any) physiological trait(s)
yield (http://agproducts.basf.us/products/headline-fungi- of the crop should be changed by plant biotechnology and/
cide.html, last accessed 2 January 2015). Like biological or plant breeding to enable the new crop management sys-
additives, the use of chemical additives to reduce variable tem. Several recent articles have proposed various targets for

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


yield may be a management option to evaluate as part of a genetic manipulation to increase yield in crops (for example,
new crop management system. Reynolds et  al., 2005; Long et  al., 2006b; Ainsworth et  al.,
2008, 2012; Long and Ort, 2010; Yang et al., 2010; Zhu et al.,
Innovation example 4: avoiding drought stress 2010; Kant et al., 2012; Slattery et al., 2013; Wang et al., 2013).
If water is the predominant limitation to yield, improvements Very few of these proposals are linked to proposed crop man-
can be made by matching the crop’s water requirements to agement changes. I am proposing that the target crop man-
the pattern of water supply, thereby avoiding the impact of agement system be used to identify the required physiological,
drought (Passioura, 2006). Cereal crops are very sensitive to developmental, or biochemical modifications to the crop that
water deficits at anthesis, and require the most water during are needed to implement that new crop management system.
grain filling. Thus, conservation of water during the veg- This is not a novel proposal but is in fact exactly the same
etative growth stage may avoid the effects of a later chronic strategy as used in the first Green Revolution, except now
drought. Several agronomic practices affect the infiltration of we have more genetic tools. Let us take drought tolerance in
water into the soil, its water holding capacity, and evapora- maize grown in the US corn belt as an example to illustrate
tive losses. These include planting date, rate and uniformity the subsequent steps in this strategy because this has been the
of establishment, weed control, fertilization, stubble manage- focus of several recent commercial releases.
ment, and the previous crop rotation. More drought-tolerant maize hybrids may stabilize yield
The direct evaporation of moisture from the soil surface (reduce yield volatility) in existing production areas but may
can be a significant loss, but it can be minimized by rapid also enable several alternative crop management systems
canopy closure, which requires good seedling establishment including:
and rapid leaf development. Because leaf growth increases
with air and soil temperature, planting winter annual crops •• reduced irrigation;
early, when the soil and air are still warm, creates more •• increased plant population densities; and
canopy cover during late autumn and winter, which reduces •• expansion into more marginal and rain-fed regions/soils.
evaporative losses from the soil surface and conserves water
for growth during grain filling (Passioura, 2006). ‘Drought’ has different meanings to a meteorologist,
Repeated cultivation of the soil to control weeds and to farmer, agronomist, and molecular biologist (Passioura,
make a fine seed bed not only damages the soil structure 2007). Consequently, ‘drought tolerance’ has quite different
but allows a greater evaporative loss of water from the soil. meanings as well. In the following sections, the target ‘drought
Directly planting seed into the soil without cultivation (no- tolerance’ trait in maize is defined as the kg of seed produced
till), combined with using a broad-spectrum herbicide (e.g. per unit of water transpired. A drought-tolerant maize hybrid
glyphosate) to achieve better weed control, improves the would therefore produce more grain from the same amount
timeliness of sowing, reduces evaporative losses, and thereby of water transpired, or the same amount of grain from less
improves yields in water-limited environments. This agro- water transpiration.
nomic innovation requires a range of herbicide-resistant
(usually genetically engineered) cultivars that are specifically Picking drought-tolerance traits
suited to being planted at different times during the season There are several factors to consider when picking the traits
(Passioura, 2006). that will impact ‘drought tolerance’ in maize for the US mar-
Another innovation is to select for large early leaves in ket. First is the phenotyping method. Lack of good phenotyp-
wheat combined with reduced tiller development to optimize ing tools that provide quantitative data is a major reason why
3442  | McKersie

plant breeding for knowledge-based stress avoidance and tol- law that is often used to this day in extension publications
erance mechanisms has not progressed well, even though there that provide fertilizer recommendations. The barrel has staves
is genetic variation for these traits in germplasm collections of unequal length, with each stave representing a nutrient.
(Richards et al., 2010; Araus et al., 2012; Araus and Cairns, The shortest stave limits the capacity of the barrel, indicat-
2014). Salekdeh et al. (2009) proposed a set of rigorous cri- ing that this nutrient should be supplemented to the crop to
teria for phenotyping in both controlled and field situations, increase the capacity of the barrel. Once that stave is length-
and they proposed several phenotypes associated with differ- ened (i.e. nutrient supplied), another stave becomes the short-
ent yield components under drought. They noted that drought est and limits the capacity of the barrel.
environments are diverse, and several biotic and abiotic stresses This concept is implicit in crop production research. For
affect yield in these environments. Thus, they did not propose example, in the first Green Revolution, once the dwarf wheat
the use of a single environment to impose a drought stress cultivars were supplied with more nitrogen, water became
or a single set of phenotypes to quantify drought tolerance, the next limiting factor. In a more recent example, Passioura
but suggested that yield, water use, water-use efficiency, and (2006) noted that if cereal grain yield per water used is mark-
harvest index be included as reference phenotypes in future edly less than 20 kg ha−1 mm−1, it is likely that other stresses
studies. Some commercial breeding programmes advocate the such as weeds, diseases, poor nutrition, or inhospitable soil
use of managed-drought-environment technologies (Cooper are limiting yield. He noted that any benefit from improved
et al., 2014), whereas others have concluded that most traits water management can be achieved only if these constraints
of importance in dry environments are selected best in favour- are eliminated first.
able moist environments (Richards et al., 2010). Perhaps this The concept of Liebig’s law seems especially useful when

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


is a difference between maize and wheat breeding, or the we are developing strategies to improve yield-related traits in
American and Australian environments. crops using genetics. The staves can be considered to represent
Second is the genetics. Most current plant breeding pro- different traits or genes (alleles) and the capacity of the barrel
grammes select for yield in field environments (Richards et al., to represent yield (Fig. 4). Although it may be overly simplis-
2010; Cooper et al., 2014). To be useful in a plant breeding tic, it does not incorporate the interactions among traits, and
programme, physiological traits must have genetic variability, it is not quantitative, the law of the minimum model provides
be genetically correlated with yield, have higher heritability a conceptual model to categorize the physiological factors
than yield, and be quantifiable in a high-throughput manner limiting drought tolerance and raises several questions about
(Fig. 2). Knowledge of the genomics of the plants and pre- our current research strategy. For example:
diction methodology has led to the identification of molec-
1. Does the same stave limit ‘drought tolerance’ regardless
ular markers in the maize genome that are associated with
of what the barrel contains? In other words, does what we
drought-related traits, and these surrogates for the physiolog-
measure dictate what we observe and conclude? If we meas-
ical traits have been applied in maize breeding (Cooper et al.,
ure greenness of a leaf as a proxy for drought tolerance, do
2014). Similarly, in rice (O. sativa L.), breeders first charac-
we identify the same biological processes as we would if we
terized the available germplasm under drought at the mor-
measured seed yield in a water-limited field trial?
phological, genetic, and molecular levels and found genetic
2. Does the same stave limit ‘drought tolerance’ in all plant
variation for drought tolerance within the available gene pool.
species?
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 interac-
tion of multiple traits for their combined effects on drought
response and yield in maize (Cooper et al., 2014).

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) 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
available, but by the scarcest resource (http://en.wikipedia. same concept can be applied to yield traits. Reproduced from Wikipedia
org/wiki/Liebig’s_law_of_the_minimum, last accessed 2 at http://en.wikipedia.org/wiki/Liebig%27s_law_of_the_minimum (last
January 2015). Liebig’s barrel (Fig. 4) is an illustration of this accessed 2 January 2015) with permission.
Planning food security  |  3443

3. Does the same stave limit ‘drought tolerance’ under mild, 600 distinct types of transgenic plants. In another example,
transient water deprivation as following a long period of Yang et al. (2010) summarized the regulatory genes that have
water deprivation? been identified in Arabidopsis and reported positive results
4. Does the same stave limit ‘drought tolerance’ regardless of from eight transcription factor gene families, two post-tran-
the plant’s stage of development? scription gene families, and nine osmoprotectant metabolite
5. Does the same stave limit ‘drought tolerance’ in all tissues? classes. This is not a positive outcome from our science.
Concerns have also been expressed about the quality of
The literature indicates that the answer to all of the above this gene discovery research because overexpression of very
questions is a clear ‘No’. Our original research strategy was many single transgenes seems to slightly increase drought tol-
to characterize drought tolerance traits in model plant sys- erance in transgenic plants under extreme conditions. Lawlor
tems grown in artificial environments and extrapolate these (2013) has expressed concern about the quality of these reports
results to crops. We now know that we would make differ- because they neglect the physiology of plant water relations,
ent conclusions based on how we define drought tolerance, use unspecific definitions and criteria, and use inadequate
what we measure, how we measure it, and in which species we methods to assess drought tolerance. He noted that in these
measure it. Since our inductive logic has not been validated reports there is a trend in which transgenic plants exhibit a
by the empirical data, we need to change our research strat- form of ‘drought resistance’ that he calls ‘delayed stress onset’.
egy. It now appears that research in a model plant species, In experiments that deprive plants of water to demonstrate
such as Arabidopsis, is exceedingly useful to define potential drought tolerance, the transgenic plants develop stress symp-
fundamental biological processes that impact drought toler- toms later than in wild-type plants. These ‘drought-tolerant’

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


ance. Nonetheless, we cannot use inductive logic to create plants tend to have slower and less water loss, less total leaf
a hypothesis on drought tolerance in crops based on these area, decreased stomatal conductance, and thicker laminae,
observations. Experiments using model systems will not iden- regardless of the transgene that was overexpressed. When rewa-
tify traits that limit yield in crops. If we accept this argument, tered, these transgenic plants have faster and greater recovery
it is not surprising that a genetic tool (e.g. vector construct) than wild-type plants. Biochemical and metabolic changes are
that was efficacious in a model species rarely impacts the often observed in the transgenic plants. These observations are
crop’s drought tolerance or yield (Lawlor, 2013). Therefore, then proposed to indicate a cause–effect relationship among
the target physiological traits that may enable new crop man- the transgene, the biochemical changes, and drought tolerance.
agement and improve drought tolerance must be selected Lawlor (2013) considers this conclusion to be invalid because
based on knowledge of crop growth and development in the the degree of water deprivation experienced by the cells in the
target crop grown in field environments. This becomes exceed- control and transgenic plants is not equivalent.
ingly difficult in a changing climate, since those hot, dry, high- Thus, there seems to be a high rate of false-positive gene
CO2 field environments do not exist yet. Consequently, we discovery for drought tolerance transgene, and the stringency
have no alternative but to compromise and extrapolate from of our gene discovery research needs to increase. A  set of
controlled environments, noting the risk involved in doing so. clearly defined terms for the relevant physiological traits that
are commonly accepted needs to be implemented (Lawlor,
Step 3: identify genes that modify traits 2013). Preliminary results in one set of experiments need to be
validated in more thorough follow-up studies and the results
A prerequisite for biotechnology approaches to crop improve- published, whether they support or refute the initial findings.
ment is the identification of genes that modify or regulate Editors, reviewers, and granting agencies need to recognize
the physiological traits of interest. One approach often used, the importance of this confirmation research because it will
especially in academic research, is knowledge based. The minimize costly commercial development by providing focus
researcher starts with a hypothesis on the mode of action to the most efficacious transgenes in each crop.
(function) of a gene that can be tested by overexpression
(or knockout) of a specific gene in model transgenic plants.
Step 4: create genetic diversity
This approach has been successfully used commercially by
DuPont Pioneer in maize for drought tolerance targeting The next step in this research plan is to create the genetic
ethylene metabolism (Habben et  al., 2014). An alternative diversity using one or more biotechnology tools for the new
approach is to conduct large-scale screening, starting with physiological or biochemical traits in the target crop (Fig. 2).
a population of genes overexpressed (or knocked out) in These tools include:
a model plant such as Arabidopsis, and then selecting the
most efficacious transgene based on empirical results. This •• MAS;
approach has also been used successfully leading to the •• genetic engineering;
release of DroughtGard by Monsanto/BASF (Castiglioni •• mutagenesis; and
et al., 2008). •• transformation.
However, the list of transgene candidates identified by
these approaches for drought tolerance is huge. For example, In general, if the genetic diversity exists already within the
Lawlor (2013) summarized the positive drought tolerance crop for the desired trait, then it is simpler, faster, and cheaper
phenotypes into nine major categories that included about to use MAS coupled with traditional breeding to introgress
3444  | McKersie

the trait into regionally adapted germplasm than it is to use engineered trait than a trait from exotic germplasm. A single
either genetic engineering approach (Collard and Mackill, marker may be sufficient for the engineered trait, but if the
2008). Thus, the first task is to screen the available germ- trait originated in exotic germplasm or is quantitative (multi-
plasm for the target trait, and based on these results decide genic), several markers may be required to track the required
whether to use MAS or an engineering approach (Fig. 2). To genes, and to remove linked genes, which adds complexity.
use MAS, the trait must meet the requirements listed in step Unlike the previously engineered traits of glyphosate resist-
2: have genetic variability, be genetically correlated with yield, ance or root worm resistance that introduced novel protein
have higher heritability than yield, and be quantifiable in a functions into the crops, changes to yield and stress-related
high-throughput manner. These are not simple requirements. traits will most likely require the modification of existing
Both genetic engineering techniques, mutagenesis and molecular networks within the plant that involve many inter-
transformation, require detailed genetic, physiological, and acting genes. It is anticipated that there is allelic variance for
biochemical knowledge about the target trait to efficiently these networks within the crops, and that these ‘native’ alleles
engineer the trait. In some cases, success has been achieved may interact differently with the engineered gene. This creates
without this knowledge. Instead a high-throughput screen is the potential to optimize the trait using breeding methodology
required to identify an individual with the trait from within a in commercial germplasm by combining favourable alleles. It
large population, which is not usually feasible for most agro- also raises a potential commercial challenge if there is a sig-
nomic traits (herbicide resistance being the obvious excep- nificant (genotype×transgene)×environment interaction, as
tion). The more complete our understanding and knowledge reported by Habben et al. (2014) for the downregulated ACC
about the target trait, the more likely the project will be synthases in maize. This approach is highly analogous to the

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


successful. breeding approach used successfully in the 1970s and 1980s
When the trait can be easily identified in a high-through- to intogress disease resistance from wild relatives into culti-
put screen and when it can be created by mutation to cre- vated barley by E. Reinbergs (University of Guelph, Ontario,
ate a new allele, then mutagenesis is the preferred genetic Canada, personal communication) to find a ‘Happy Home’
engineering method because of the lower cost and faster for the resistance gene.
regulatory approval processes (see the following section on The final commercial product from the plant breeding
imidazolinone tolerance as an example). This method is espe- programme must be robust. The new cultivar or hybrid must
cially applicable in smaller market crops that cannot support function across various environments and across the range of
the large development and regulatory expense required to crop maturities grown within the target region. And it must
deregulate transgenic traits. Consequently, mutagenesis per se function every year. If an engineered trait fails, or even is per-
is unlikely to be applicable to development of drought toler- ceived to fail, in one year because of weather conditions, then
ance in maize, unless it is used to create a specific new allele farmers would be highly unlikely to invest in that seed the fol-
that could be identified by a high-throughput screen. lowing year. Thus, commercial plant breeding will probably
Only when sufficient genetic variability cannot be found be highly conservative in the approach used to develop new
within the available breeding populations does a transfor- cultivars. Also, extensive multiyear testing of the new culti-
mation approach become a viable option. For example, vars or hybrids will be required before commercial launch of
glyphosate resistance did not exist in soybean before it was these products.
engineered. Western corn root worm resistance did not exist It must be recognized by all stakeholders in these projects
in maize before it was engineered. Both traits were created by that these are high-risk approaches because the final product
the introduction of novel transgenes using plant transforma- may not meet our expectations in the future climate. So con-
tion technology and are now hugely successful commercial stant monitoring, evaluation, reassessment, and change man-
products. agement will be critical to the success of this effort.
The product of this step in research is an individual plant
that is sexually compatible with the crop, so it can then be Step 6: implement changes in the crop
used in a plant breeding programme, and is deregulated glob- management system
ally, so it can be used in the global food chain.
The first Green Revolution did not end with the develop-
ment of short-statured rice and wheat cultivars that met the
Step 5: breed diversity into new adapted cultivars
quality and agronomic characteristics required by the local
The next step to implement the crop management change is agricultural systems. Neither will this putative Second Green
to transfer the trait from the original source, whether that Revolution end with the deregulation of a transgenic event
be from exotic germplasm or from an engineered plant, to and its incorporation into a new cultivar. The implementa-
regionally adapted commercial germplasm. The new trait will tion of the new crop management system and the develop-
need to be combined with existing biotechnology traits, such ment of new cultivars will be reiterative. The new cultivars
as herbicide resistance, and with other agronomic traits, such need to be tested in the new management system, and the new
as maturity, disease resistance, and quality to meet agronomic management system needs to be optimized using the new cul-
and quality requirements. Breeding is likely to use MAS for tivars. Because any degree of crop failure cannot be risked,
several (or all) of these traits (Collard and Mackill, 2008). the introduction of any new management system will be slow
In this respect, it will be simpler and faster to transfer an and methodical over several years.
Planning food security  |  3445

Success in developing drought tolerance Richards et al., 2010). QTL for drought-tolerance traits and
in maize their molecular markers have been identified in different
crops, including maize (Cooper et al., 2014). With the devel-
The following are some examples of projects that have suc- opment of comprehensive molecular linkage maps, MAS pro-
cessfully released more drought-tolerant maize hybrids in cedures stack these desirable traits to achieve improvements
recent years. Almost all approaches have had some degree in drought tolerance. The accuracy and preciseness in QTL
of success. Traditional plant breeding has made considerable identification, the significant genetic×environment interac-
progress without applying knowledge of the physiological tion, the large number of genes encoding yield, and the use
basis of yield. MAS using markers correlated with drought of wrong mapping populations have hindered the use of QTL
tolerance has been used successfully, again without specific to select for growth and yield under water-limited conditions
knowledge of the genes and their physiological effects. On the (Ashraf, 2010).
other hand, both empirical screening and knowledge-based Nonetheless, the development of drought tolerance in
approaches using genes with known functions and physiolog- maize in North America has recently progressed with the
ical effects have been successfully used to genetically engineer release of maize hybrids from Syngenta’s Agrisure Artesian
transgenic maize with greater drought tolerance. The lack of (http://www3.syngenta.com/country/us/en/agriculture/
high-throughput screens for a trait as complex and diverse as seeds/agrisure-traits/Pages/agrisure-artesian-4011.aspx, last
drought tolerance has limited any application of mutagenesis accessed 2 January 2015) and Pioneer’s AQUAmax (https://
to drought tolerance, but its potential for other commercial www.pioneer.com/home/site/us/products/corn/seed-traits-
targets has been demonstrated. technologies-corn/optimum-aquamax-hybrids/, last accessed

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


2 January 2015) breeding programmes. Both companies
developed new inbreeds using MAS for physiological pheno-
Success example 1: maize breeding
types, combining their proprietary genetic information with
Breeding for drought tolerance in maize has been diffi- genomic data to identify genes associated with drought tol-
cult (Salekdeh et  al., 2009; Araus et  al., 2012). The large erance. They then developed genetic markers and by MAS
genotype×environment (location–year) interaction for yield brought those markers together in inbreeds. Cause–effect
has limited the success of our current approach (Araus et al., relationships and modes of action were not established, but
2012). Surprisingly, physiological studies comparing hybrids the markers were ‘simply’ correlated with drought tolerance.
from different decades have shown that the genetic gains that Since the new inbreeds and hybrids were derived from previ-
have been achieved in maize yield have been associated with ously engineered and deregulated transgenic events for her-
greater ‘stress tolerance’ (see, for example, Tollenaar and bicide resistance, the new drought-tolerant hybrids did not
Wu, 1999). Individual plants in new maize hybrids are able require USDA approval before commercial release (http://
to set ears under higher levels of competition than those in biotech.about.com/od/Genetically-Modified-Organisms/a/
older hybrids, which enables higher population densities to be Breeding-Versus-Engineering-To-Make-Drought-Tolerant-
planted. Other traits in maize that have been altered by plant Corn.htm, last accessed 2 January 2015).
breeding include: sustained leaf photosynthesis during grain One concern is that the infrastructure, in terms of labo-
filling, a greater number of kernels, a shorter anthesis–silking ratories and knowledge, that is required to identify QTL for
interval under drought, extraction of less water from the soil drought tolerance and use MAS in plant breeding represents
before flowering, a reduction in tassel size (in temperate but a significant expense. This technology therefore may not be
not tropical maize), escape from terminal drought by early readily available to smaller plant breeding programmes in
flowering, deeper rooting, and, importantly for this overview, developing countries.
genetically engineered pest resistance that has reduced root
damage from soil insects. These physiological analyses were Success example 3: DroughtGard
done in retrospect, not by a priori design.
Lobell et al. (2014) recently confirmed that maize and soy- The first commercial release of a crop that was genetically
bean yields in the US midwest over the period 1995–2012 engineered for drought tolerance is DroughtGard maize
increased under all levels of stress. However, the sensitiv- (MON87460) from the Monsanto/BASF collaboration, as
ity of maize to high vapour pressure deficits has increased. a result of an empirical screening approach. These plants
Vapour pressure deficit measures atmospheric water demand express a novel protein that encodes an RNA chaperone. In
and depends on air temperature and humidity. Consequently, Bacillus, this protein acts to maintain RNA structure and
Lobell et  al. (2014) suggested that sensitivity to heat stress biological function. In plants, the endogenous cold shock
has increased in maize hybrids developed in recent decades in domain-containing (CSD) proteins regulate stress responses
the US midwest. It is uncertain how concerned we should be through a post-transcriptional mechanism (Castiglioni et al.,
about this observation given future climate predictions. 2008). The initial discovery of the transgene’s efficacy in
plants was made in Arabidopsis where its overexpression was
Success example 2: MAS shown to improve cold tolerance in transgenic seedlings. This
observation was expanded and the efficacious CSD genes nar-
Genetics research has identified genomic regions useful to rowed in subsequent screening, where it was observed that
a crop under stress conditions (Collard and Mackill, 2008; transgenic rice plants expressing CspB had improved growth
3446  | McKersie

following exposure to cold, heat, and water deficits as dem- 2016/2017 in South Africa, but due to the lack of regulatory
onstrated by greater plant height. Constitutive overexpression approval, release in other parts of Africa will be delayed.
in maize gave a positive improvement in yield under man- Note that this technology transfer is only feasible for trans-
aged water environments and dryland conditions that was genic traits. The transfer of markers and drought-tolerant
predominately the result of increased kernel numbers, not germplasm from breeding programmes, such as Agrisure
kernel weight. The positive effects of transgene expression Artesian and AQUAmax, is unlikely to be successful because
were observed in late vegetative/flowering and grain-filling of the genetic complexity and genotype×environment inter-
periods. Contrary to the effects of many other transgenes that actions involved.
claim effects on drought tolerance, expression of CspB did
not result in detrimental effects on plant size, development, Success example 5: ethylene biosynthesis
or productivity under well-watered conditions. Consequently,
the yield improvements observed under water limiting condi- DuPont Pioneer has used a knowledge-based approach to
tions were not associated with a yield penalty in high-yielding develop more drought-tolerant maize hybrids. Habben et al.
environments. An application for deregulation of a transgenic (2014) observed that maize uses the ethylene signalling system
maize event was made to APHIS in 2011 (http://www.aphis. to abort kernels as part of its survival strategy during severe
usda.gov/brs/aphisdocs/09_05501p_fea.pdf, last accessed drought stress. They hypothesized that the survival response
2 January 2015) and development of commercial maize in maize is too conservative to maintain high yield and is
‘DroughtGard’ hybrids was initiated (http://www.monsanto. probably unnecessary in modern agricultural environments.
com/SiteCollectionDocuments/whistlestop-drought-posters. They proposed that modulation of the response by downreg-

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


pdf, last accessed 2 January 2015). In 2  years of small-scale ulating ethylene production during drought stress would lead
production in the drier regions of the US corn belt, growers to an increase in grain yield. Transgenic maize events were
reported about 280 lb ac–1 (5 bu ac–1 or 310 kg ha–1) more grain created with silenced ACC synthases using an ACS6 RNA
than competitive hybrids under conditions that give typical interference construct. These enzymes catalyse the rate-lim-
yields in the 50–125 bu ac–1 range (Waltz, 2014). iting 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
Success example 4: water-efficient maize for Africa
transgenic events had up to 9.3 bu ac–1 more grain than the
project
null controls, but on the negative side, there was a (genotype×
The successful implementation of genetic engineering tech- transgene)×environment interaction. The authors attributed
nology to improve drought tolerance in maize requires the the yield increase to a shorter anthesis–silking interval and an
creation of regionally adapted hybrids. One example where increase in kernel number per ear during drought. Under a
this is already occurring is the Water Efficient Maize for Africa low-nitrogen stress, the best event had 7.1 bu ac–1 more yield.
(WEMA) project (http://wema.aatf-africa.org/, last accessed 2 Further field testing, deregulation studies, and commercial
January 2015). This is a public sector/private partnership that hybrid production are presumably in progress at DuPont
was launched in March 2008 and is being led by the Kenyan- Pioneer.
based African Agricultural Technology Foundation and
funded by the Bill and Melinda Gates Foundation, Howard Success example 6: imidazolinone tolerance
G.  Buffett Foundation, and US Agency for International
Development. The goal of this multinational project is to As previously mentioned, application of mutagenesis to the
enhance food security in sub-Saharan Africa by deploying development of drought tolerance in maize is not feasible
drought-tolerant maize with insect protection to farmers. The at present because of the lack of high-throughput screens.
transgenic technology and the associated germplasm were Nonetheless, mutagenesis will have application in other crops
provided by Monsanto royalty-free (http://www.monsanto. for other yield-related traits. One trait that has agronomic
com/improvingagriculture/pages/water-efficient-maize-for- value in crops and that was developed using different meth-
africa.aspx, last accessed 2 January 2015; http://www.cimmyt. ods of genetic engineering is herbicide tolerance. The devel-
org/en/projects/water-efficient-maize-for-africa-wema-phase- opment of transgenic plants with resistance to glyphosate
ii, last accessed 2 January 2015). The infrastructure has now is well known (Duke and Powles, 2008), but there are other
been established including scientific staff and the establish- herbicides and other approaches. The imidazolinone (IMI)
ment of drought testing sites that meet the national require- family of herbicides control a broad spectrum of grass and
ments for testing genetically engineered hybrids. So far, the broadleaf weeds by inhibiting the enzyme acetohydroxy-
project reports some very encouraging results using conven- acid synthase (AHAS), also called acetolactate synthase
tional breeding, MAS, and biotechnology tools to create (EC 4.1.3.18). This is a critical enzyme in the synthesis of
new germplasm. In 2010, the project identified several initial branched-chain amino acids in plants (Tan et al., 2005, 2006).
hybrids that showed a yield advantage under drought stress Sulfonylurea herbicides target the same enzyme. The develop-
conditions. Nine new maize inbred lines were released for ment of IMI-tolerant crops is a good example of the poten-
licensing. Compared with non-transgenic hybrids, the trans- tial of different genetic engineering strategies to create novel
genic hybrids produced 8–14% more grain in multiple years genetic variability, because mutagenesis, transformation, and
of testing (Waltz, 2014). Commercial release is expected in MAS have all been used to engineer tolerance.
Planning food security  |  3447

Tolerance to IMI herbicides was introduced through diversity already exists within the crop’s germplasm, conven-
mutation breeding and commercially released by BASF tional plant breeding, complimented by MAS, is the preferred
and its partners as Clearfield crops that were developed route to a commercial product. If diversity does not exist,
using conventional breeding methods beginning in 1992. In these traits become the targets for change using genetic tools.
the Clearfield approach, variant AHAS alleles were created Next, if an engineering approach is selected, the genes
through mutagenesis and selection to confer IMI tolerance in that regulate these physiological traits must be characterized
maize, wheat, rice, oilseed rape (Brassica napus L.) and sun- and understood. This may be achieved initially by empiri-
flower (Helianthus annuus L.) (Tan et  al., 2005, 2006). The cal screening or by knowledge-based approaches, apparently
allele encoding IMI tolerance was then introgressed into com- with equal success to date. Although the basic biological
mercial germplasm. Similar approaches have subsequently principles may be developed in model systems, the hypothesis
been used in other breeding programmes (Oldach et al., 2008; that the traits will enable the envisioned crop management
Lee et al., 2011; Rustgi et al., 2014). changes must be validated in crops under field conditions in
Transformation was used as an alternative strategy to cre- the target region.
ate IMI tolerance in crops. Many patents and publications Genetic diversity for these physiological traits is then cre-
describe the production of herbicide-resistant plants by trans- ated using the identified genetic engineering technology and
forming plants with mutant alleles that encode altered AHAS knowledge of the trait. Either mutagenesis or transformation
enzymes resistant to inhibition by the IMI and sulfonylurea may be used depending on the crop, market, and suitability of
herbicides (Anderson and Hinnerd, 1988; Lee et  al., 1988; screening methods. Once the hypothesis is validated in field
Wiersma et al., 1989; Bedbrook et al., 1991; Sala et al., 2008). trials, the biological changes to the crop must be understood

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


Although technically feasible, none of the transgenic IMI- and documented to deregulate any commercial products.
tolerant plants have been commercially released for various Thus, improving our knowledge of the physiology, biochem-
economic and agronomic reasons. In this case, mutagenesis istry, and molecular biology of the trait is fundamental at
was the commercially successful approach mainly because a this step.
high-throughput screen enabled identification of commercial Next, we need to breed new regionally adapted cultivars
levels of tolerance and because the development and regula- to incorporate these physiological traits into agronomically
tory costs were sufficiently low to justify the expense in rela- acceptable cultivars. Finally, the new cultivars (or hybrids)
tively ‘small market’ but important crops. need to be utilized to implement the envisioned crop manage-
ment system in a potentially reiterative research programme.
Implementation will require not only changes in crop man-
agement but also regional infrastructure improvements and
Summary and conclusions
education if we are to meet society’s needs for food security in
Society expects innovation in crop production to meet its the face of inevitable climate change. An integrated sequen-
demand for food security, despite an uncertain environment tial research plan that combines all of our technologies in
due to climate change, increased population, and changing plant biotechnology, genetics, physiology, agronomy, and
food preferences. The innovations to meet this expectation breeding has the greatest potential to provide food security
will require knowledge of the future climate, what might limit for the future.
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. Acknowledgements
I am proposing that research follows a six-step sequen- The author thanks Stephen Crafts-Brandner, Jeremy Singer, Larry Gusta,
tial plan to create these new commercial products based on Suzanne Cunningham, and Bijay Singh for their comments, advice, and criti-
merging the lessons that we have learned from the Green cal review of this manuscript. The author is retired from BASF Plant Science
and is a former faculty member at the University of Guelph, Ontario,
Revolution and from the development of the first genetically Canada.
modified crops. First, we need a vision for the future to guide
us, based on the current and projected challenges in crop pro-
duction. New crop management technology and new agro- References
nomic practices need to be envisioned in a cropping system Ainsworth EA, Rogers, A, Leakey AD. 2008. Targets for crop
that mitigates risk, meets regional needs, and captures poten- biotechnology in a future high-CO2 and high-O3 world. Plant Physiology
147, 13–19.
tial regional opportunities in a predicted hot, dry, high-CO2
Ainsworth EA, Yendrek CR, Skoneczka JA, Long SP. 2012.
environment. A  global solution across regions and crops is Accelerating yield potential in soybean: potential targets for
unlikely due to the complexity of the biological processes, the biotechnological improvement. Plant, Cell and Environment 35, 38–52.
differences among existing agricultural systems, and the local Anderson PC, Hinnerd KA. 1988. Herbicide resistance in plants. US
impact of climate change. Regional innovation, research, and patent no. 4761373.
leadership are needed. Araus JL, Serret MD, Edmeades GO. 2012. Phenotyping maize for
adaptation to drought. Frontiers in Physiology 3, 305.
The physiological traits that will enable these changes in
Araus JL, Cairns JE. 2014. field high-throughput phenotyping: the new
crop management to be implemented might be missing from crop breeding frontier. Trends in Plant Science 19, 52–61.
our current cultivars. If they are missing, genetics can be Ashraf M. 2010. Inducing drought tolerance in plants: recent advances.
used to create the needed crop cultivars. If sufficient genetic Biotechnology Advances 28, 169–183.
3448  | McKersie
Balba H. 2007. Review of strobilurin fungicide chemicals. Journal biodiversity in the United States , p. 362. A Report by the US Climate
of Environmental Science and Health, Part B: Pesticides, Food Change Science Program and the Subcommittee on Global Change
Contaminants, and Agricultural Wastes 42, 441–451. Research. Washington, DC. http://www.usda.gov/oce/climate_change/
Bartlett DW, Clough JM, Godwin JR, Hall AA, Hamer M, Parr- SAP4_3/CCSPFinalReport.pdf. Last accessed 2 January 2015.
Dobrzanski B. 2002. The strobilurin fungicides. Pest Management Hatfield J, Takle G, Grotjahn R, Holden P, Izaurralde RC, Mader T,
Science 58, 649–662. Marshall E, Liverman D. 2014. Agriculture. In: Melillo JM, Richmond TC,
Bedbrook JR, Chaleff RS, Falco SC, Mazur BJ, Summerville CR, Yohe GW, eds. Climate change impacts in the United States: the Third
Yadav, NS. 1991. Nucleic acid fragment encoding herbicide resistant National Climate Assessment. US Global Change Research Program,
plant acetolactate synthase. US patent no. 5013659. 6–1-nn. http://nca2014.globalchange.gov/report/sectors/agriculture. Last
accessed 2 January 2015.
Beddington J. 2010. Food security: contributions from science to a new
and greener revolution. Philosophical Transactions of the Royal Society B: IPCC. 2007. Climate Change 2007: Synthesis Report. Contribution
Biological Sciences 365, 61–71. of Working Groups I, II and III to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. Core Writing Team, Pachauri
Berendsen RL, Pieterse CM, Bakker PA. 2012. The rhizosphere RK, Reisinger A, eds. Geneva: Intergovernmental Panel on Climate
microbiome and plant health. Trends in Plant Science 17, 478–486. Change. http://www.ipcc.ch/publications_and_data/publications_and_
Castiglioni P, Warner D, Bensen RJ, et al. 2008. Bacterial RNA data_reports.shtml. Last accessed 2 January 2015.
chaperones confer abiotic stress tolerance in plants and improved grain IPCC. 2013. Climate Change 2013: The Physical Science Basis.
yield in maize under water-limited conditions. Plant Physiology 147, Contribution of Working Group I to the Fifth Assessment Report of the
446–455. Intergovernmental Panel on Climate Change. Stocker TF, Qin D, Plattner
Caviglia OP, Sadras VO, Andrade FH. 2004. Intensification of agriculture GK, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM,
in the south-eastern pampas. Field Crops Research 87, 117–129. eds. Cambridge, UK/New York: Cambridge University Press. http://www.
Celebi SZ, Demir S, Celebi R, Durak ED, Yilmaz IH. 2010. The effect ipcc.ch/publications_and_data/publications_and_data_reports.shtml Last
of arbuscular mycorrhizal fungi (AMF) applications on the silage maize accessed 2 January 2015.
(Zea mays L.) yield in different irrigation regimes. European Journal of Soil Ishikawa S, Hare MC, Kettlewell PS. 2012. Effects of strobilurin

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


Biology 46, 302–305. fungicide programmes and fertilizer nitrogen rates on winter wheat: leaf
Coleman-Derr D, Tringe, SG. 2014. Building the crops of tomorrow: area, dry matter yield and nitrogen yield. Journal of Agricultural Science
advantages of symbiont-based approaches to improving abiotic stress 150, 427–441.
tolerance. Frontiers in Microbiology 5, 283. Jabs T, Pfirmann J, Schafer S, Wu YX, Tiedemann A. 2002.
Collard BC, Mackill DJ. 2008. Marker-assisted selection: an approach Antioxidative and antisenescence effects of the strobilurin pyraclostrobin
for precision plant breeding in the twenty-first century. Philosophical in plants: a new strategy to cope with environmental stress in cereals. In:
Transactions of the Royal Society B: Biological Sciences 363, 557–572. BCPC Conference—Pests and Diseases 2002 , pp. 941–946. Brighton,
UK, 18–21 November. https://www.seedquest.com/id/b/basf/vitality/
Cooper, M, Gho, C, Leafgren, R, Tang, T, Messina, C. 2014. Breeding articles/Antioxidative_and_antisenescence_effects.pdf. Last accessed 2
drought-tolerant maize hybrids for the US corn-belt: discovery to product. January 2015.
Journal of Experimental Botany 65, 6191–6204.
Kant S, Seneweera S, Rodin, J Materne M, Burch D, Rothstein SJ,
Crafts-Brandner SJ, Salvucci ME. 2000. Rubisco activase Spangenberg G. 2012. Improving yield potential in crops under elevated
constrains the photosynthetic potential of leaves at high temperature CO2: integrating the photosynthetic and nitrogen utilization efficiencies.
and CO2. Proceedings of the National Academy of Sciences, USA 97, Frontiers in Plant Science 3, 162.
13430–13435.
Karl TR, Melillo JM, Peterson TC. 2009. Global climate change impacts
Duke SO, Powles SB. 2008. Glyphosate: a once-in-a-century herbicide. in the United States. Cambridge: Cambridge University Press.
Pest Management Science 64, 319–325.
Ker K, Seguin P, Driscoll BT, Fyles JW, Smith DL. 2012. Switchgrass
Fischer G, Shah M, Tubiello FN, van Velhuizen H. 2005. Socio- establishment and seeding year production can be improved by inoculation
economic and climate change impacts on agriculture: an integrated with rhizosphere endophytes. Biomass and Bioenergy 47, 295–301.
assessment, 1990–2080. Philosophical Transactions of the Royal Society
B: Biological Sciences 360, 2067–2083. Khan ZR, Midega CA, Pittchar JO, Murage AW, Birkett MA, Bruce
TJ, Pickett JA. 2014. Achieving food security for one million sub-
Fischer RA, Byerlee D, Edmeades GO. 2014. Crop yields and global
Saharan African poor through push-pull innovation by 2020. Philosophical
food security: will yield increase continue to feed the world? ACIAR
Transactions of the Royal Society B: Biological Sciences 369, 20120284.
Monograph No. 158. Canberra: Australian Centre for International
Agricultural Research. http://aciar.gov.au/publication/mn158. Last Kirkegaard JA, Hunt JR, McBeath TM, et al. 2014. Improving water
accessed 2 January 2015. productivity in the Australian grains industry—a nationally coordinated
approach. Crop and Pasture Science 65, 583–601.
Galford GL, Melillo JM, Kicklighter DW, Cronin TW, Cerri CEP, Mustard
JF, Cerri CC. 2010. Greenhouse gas emissions from alternative futures Kirkegaard JA, Lilley JM, Howe GN, Graham JM. 2007. Impact
of deforestation and agricultural management in the southern Amazon. of subsoil water use on wheat yield. Australian Journal of Agricultural
Proceedings of the National Academy of Sciences, USA 107, 19649–19654. Research 58, 303.
Glick BR. 2012. Plant growth-promoting bacteria: mechanisms and Kyei-Boahen S, Zhang L. 2006. Early-maturing soybean in a wheat–
applications. Scientifica (Cairo) 2012, 963401. soybean double-crop system. Agronomy Journal 98, 295–301.
Gornall J, Betts R, Burke E, Clark R, Camp J, Willett K, Wiltshire Lawlor D W. 2013. Genetic engineering to improve plant performance
A. 2010. Implications of climate change for agricultural productivity in the under drought: physiological evaluation of achievements, limitations, and
early twenty first century. Philosophical Transactions of the Royal Society possibilities. Journal of Experimental Botany 64, 83–108.
B: Biological Sciences 365, 2973–2989. Leakey AD, Uribelarrea M, Ainsworth EA, Naidu SL, Rogers A, Ort
Grossmann K, Retzlaff G. 1997. bioregulatory effects of the fungicidal DR, Long SP. 2006. Photosynthesis, productivity, and yield of maize are
strobilurin kresoxim-methyl in wheat (Triticum aestivum). Pesticide Science not affected by open-air elevation of CO2 concentration in the absence of
50, 11–20. drought. Plant Physiology 140, 779–790.
Grossmann K, Kwiatkowski J, Caspar G. 1999. Regulation of Lee H, Rustgi, S Kumar, N, Burke I, Yenish JP, Gill KS, Von Wettstein
phytohormone levels, leaf senescence and transpiration by the strobilurin D, Ullrich SE. 2011. Single nucleotide mutation in the barley acetohydroxy
kresoxim-methyl in wheat (Triticum aestivum). Journal of Plant Physiology acid synthase (AHAS) gene confers resistance to imidazolinone herbicides.
154, 805–808. Proceedings of the National Academy of Sciences, USA 108, 8909–8913.
Habben JE, Bao X, Bate NJ, et al. 2014. Transgenic alteration of Lee KY, Townsend J, Black M, Chuil CF, Mazur B, Dunsmuir
ethylene biosynthesis increases grain yield in maize under field drought- P, Bedbrook J. 1988. The molecular basis of sulfonylurea herbicide
stress conditions. Plant Biotechnology Journal 12, 685–693. resistance in tobacco. EMBO Journal 7, 1241–1248.
Hatfield J, Boote K, Fay P, et al. 2008. Agriculture. In The effects of Lemerle D, Gill GS, Murphy CE, Walker SR, Cousens RD, Mokhtari
climate change on agriculture, land resources, water resources, and S, Peltzer SJ, Coleman R, Luckett DJ. 2001. Genetic improvement and
Planning food security  |  3449
agronomy for enhanced wheat competitiveness with weeds. Australian Assessment of genetic diversity among barley cultivars and breeding lines
Journal of Agricultural Research 52, 527–548. adapted to the US Pacific Northwest, and its implications in breeding
Lobell DB, Roberts MJ, Schlenker W, Braun N, Bertis B, Little barley for imidazolinone-resistance. PLoS One 9, e100998.
BB, Rejesus RM, Hammer GL. 2014. Greater sensitivity to drought Sala CA, Bulos M, Echarte M, Whitt SR, Ascenzi R. 2008. Molecular
accompanies maize yield increase in the U.S. midwest. Science 344, and biochemical characterization of an induced mutation conferring
516–519. imidazolinone resistance in sunflower. Theoretical and Applied Genetics
Long SP, Ort DR. 2010. More than taking the heat: crops and global 118, 105–112.
change. Current Opinion in Plant Biology 13, 241–248. Salekdeh G H, Reynolds M, Bennett J, Boyer J. 2009. Conceptual
Long SP, Ainsworth EA, Leakey AD, Nosberger J, Ort DR. 2006a. framework for drought phenotyping during molecular breeding. Trends in
Food for thought: lower-than-expected crop yield stimulation with rising Plant Science 14, 488–496.
CO2 concentrations. Science 312, 1918–1921. Sayer J, Cassman KG. 2013. Agricultural innovation to protect the
Long SP, Zhu XG, Naidu SL, Ort DR. 2006b. Can improvement in environment. Proceedings of the National Academy of Sciences, USA 110,
photosynthesis increase crop yields? Plant, Cell and Environment 29, 8345–8348.
315–330. Shanker AK, Venkateswarlu B (eds). 2011. Abiotic stress in plants—
McAllister CH, Beatty PH, Good AG. 2012. Engineering nitrogen use mechanisms and adaptations. Rijeka, Croatia: InTech.
efficient crop plants: the current status. Plant Biotechnology Journal 10, Singer JW, Sauer TJ, Blaser BC, Meek DW. 2007. Radiation use
1011–1025. efficiency in dual winter cereal–forage production systems. Agronomy
Mengual C, Schoebitz M, Azco, R, Roldan A. 2014. microbial Journal 99, 1175–1179.
inoculants and organic amendment improves plant establishment and Slattery RA, Ainsworth EA, Ort DR. 2013. A meta-analysis of responses of
soil rehabilitation under semiarid conditions. Journal of Environmental canopy photosynthetic conversion efficiency to environmental factors reveals
Management 134, 1–7. major causes of yield gap. Journal of Experimental Botany 64, 3723–3733.
Muller C, Cramera W, Harea W L, Lotze-Campen H. 2011. Climate Smith P,, Gregory P J. 2013. Climate change and sustainable food
change risks for African agriculture. Proceedings of the National Academy production. Proceedings of the Nutrition Society 72, 21–28.

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016


of Sciences, USA 108, 4313–4315.
Soares-Filho B, Moutinho P, Nepstad D, et al. 2010. Role of Brazilian
Nason MA, Farrar J, Bartlett, D. 2007. strobilurin fungicides induce Amazon protected areas in climate change mitigation. Proceedings of the
changes in photosynthetic gas exchange that do not improve water National Academy of Sciences, USA 107, 10821–10826.
use efficiency of plants grown under conditions of water stress. Pest
Stevenson JR, Villoria N, Byerlee D, Kelley T, Maredia M. 2013.
Management Science 63, 1191–1200.
Green Revolution research saved an estimated 18 to 27 million hectares
Nelson GC, Rosegrant M W, Koo J, et al. 2009. Climate change: from being brought into agricultural production. Proceedings of the
impact on agriculture and costs of adaptation. Washington, DC: National Academy of Sciences, USA 110, 8363–8368.
International Food Policy Research Institute.
Strzepek K,, Boehlert B. 2010. Competition for water for the food
Nelson KA, Meinhardt CG. 2011. Foliar boron and pyraclostrobin effects system. Philosophical Transactions of the Royal Society B: Biological
on corn yield. Agronomy Journal 103, 1352–1358. Sciences 365, 2927–2940.
Nepstad D, Soares-Filho BS, Merry F, et al. 2009. The end of Swamy BP, Kumar, A. 2013. Genomics-based precision breeding
deforestation in the Brazilian Amazon. Science 326, 1350–1351. approaches to improve drought tolerance in rice. Biotechnology Advances
Oldach KH, Peck DM, Cheong J, Williams KJ, Nair RM. 2008. 31, 1308–1318.
Identification of a chemically induced point mutation mediating herbicide Tan S, Evans R, Singh B. 2006. Herbicidal inhibitors of amino acid
tolerance in annual medics (Medicago spp.). Annals of Botany 101, biosynthesis and herbicide-tolerant crops. Amino Acids 30, 195–204.
997–1005.
Tan S, Evans RR, Dahmer ML, Singh BK, Shaner DL. 2005.
Passioura J. 2006. Increasing crop productivity when water is scarce— Imidazolinone-tolerant crops: history, current status and future. Pest
from breeding to field management. Agricultural Water Management 80, Management Science 61, 246–257.
176–196.
Timmusk S, Wagner EG. 1999. The plant-growth-promoting
Passioura J. 2007. The drought environment: physical, biological and rhizobacterium Paenibacillus polymyxa induces changes in Arabidopsis
agricultural perspectives. Journal of Experimental Botany 58, 113–117. thaliana gene expression: a possible connection between biotic and abiotic
Passioura JB, Angus JF. 2010. Improving productivity of crops in water- stress responses. Molecular Plant–Microbe Interactions 12, 951–959.
limited environments. Advances in Agronomy 106, 37–75. Tollenaar M, Wu J. 1999. Yield improvement in temperate maize is
Phalan B, Green R, Balmford A. 2014. Closing yield gaps: perils and attributable to greater stress tolerance. Crop Science 39, 1597–1604.
possibilities for biodiversity conservation. Philosophical Transactions of the Trabelsi D, Mhamdi R. 2013. Microbial inoculants and their impact on
Royal Society B: Biological Sciences 369, 20120285. soil microbial communities: a review. BioMed Research International 2013,
Piesse J, Thirtle C. 2010. Agricultural R&D, technology and productivity. 863240.
Philosophical Transactions of the Royal Society B: Biological Sciences Turral H, Burke J, Faurès J-M. 2011. Climate change, water and food
365, 3035–3047. security. FAO Water Report 36. Rome: Food and Agriculture Organization
Pingali PL. 2012. Green revolution: impacts, limits, and the path of the United Nations.
ahead. Proceedings of the National Academy of Sciences, USA 109, van Ittersum MK, Cassman KG. 2013. Yield gap analysis—rationale,
12302–12308. methods and applications—introduction to the Special Issue. Field Crops
Redman RS, Sheehan KB, Stout RG, Rodriguez RJ, Henson JM. 2002. Research 143, 1–3.
Thermotolerance generated by plant/fungal symbiosis. Science 298, 1581. Waller F, Achatz B, Baltruschat H, et al. 2005. The endophytic fungus
Reynolds MP, Mujeeb-Kazi A, Sawkins M. 2005. Prospects for utilising Piriformospora indica reprograms barley to salt-stress tolerance, disease
plant-adaptive mechanisms to improve wheat and other crops in drought- resistance, and higher yield. Proceedings of the National Academy of
and salinity-prone environments. Annals of Applied Biology 146, 239–259. Sciences, USA 102, 13386–13391.
Richards RA, Rebetzke GJ, Watt M, Condon AGT, Spielmeyer W, Waltz E. 2014. Beating the heat. Nature Biotechnology 32, 611–617.
Dolferus R. 2010. Breeding for improved water productivity in temperate Wang L, Feng Z, Schjoerring JK. 2013. Effects of elevated atmospheric
cereals: phenotyping, quantitative trait loci, markers and the selection CO2 on physiology and yield of wheat (Triticum aestivum L.): a meta-
environment. Functional Plant Biology 37, 85–97. analytic test of current hypotheses. Agriculture, Ecosystems & Environment
Rodriguez RJ, Henson J, Van Volkenburgh E, Hoy M, Wright L, 178, 57–63.
Beckwith F, Kim YO, Redman RS. 2008. Stress tolerance in plants via
Wiersma PA, Schmiemann MG, Condie JA, Crosby WL, Moloney
habitat-adapted symbiosis. ISME Journal 2, 404–416.
MM. 1989. Isolation, expression and phylogenetic inheritance of an
Rustgi S, Matanguihan J, Mejias JH, Gemini R, Brew-Appiah RA, acetolactate synthase gene from Brassica napus. Molecular and General
Wen N, Osorio C, Ankrah N, Murphy KM, Von Wettstein D. 2014. Genetics 219, 413–420.
3450  | McKersie
Wullschleger SD, Tschaplinski TJ, Norby RJ. 2002. plant water optimizing interaction with conformationally flexible residues. Journal of the
relations at elevated CO2—implications for water-limited environments. American Chemical Society 132, 185–194.
Plant, Cell and Environment 25, 319–331. Zhu XG, Long SP, Ort DR. 2010. Improving photosynthetic efficiency for
Yang J, Kloepper JW, Ryu CM. 2009. Rhizosphere bacteria help plants greater yield. Annual Review Plant Biology 61, 235–261.
tolerate abiotic stress. Trends in Plant Science 14, 1–4.
Zhu XG, Wang YU, Ort DR, Long SP. 2013. E-photosynthesis: a
Yang S, Vanderbeld B, Wan, J, Huang Y. 2010. Narrowing down the comprehensive dynamic mechanistic model of C3 photosynthesis:
targets: towards successful genetic engineering of drought-tolerant crops. from light capture to sucrose synthesis. Plant, Cell and Environment 36,
Molecular Plant 3, 469–490. 1711–1727.
Zeigler RS, Mohanty S. 2010. Support for international agricultural research: Ziska LH, Bunce JA, Shimono H, et al. 2012. Food security and
current status and future challenges. New Biotechnology 27, 565–572. climate change: on the potential to adapt global crop production by active
Zhao PL, Wang L, Zhu XL, Huang X, Zhan CG, Wu JW, Yang GF. selection to rising atmospheric carbon dioxide. Proceedings of the Royal
2010. Subnanomolar inhibitor of cytochrome bc1 complex designed by Society B: Biological Sciences 279, 4097–4105.

Downloaded from http://jxb.oxfordjournals.org/ by guest on November 29, 2016