Earlier this year, I visited the Centre’s University of Queensland Node, where I brainstormed with members of the Centre – especially Professor and Chief Investigator, Daniel Ortiz-Barrientos – about how evolutionary approaches could aid in crop adaptation to climate change. These discussions led to a review that recently appeared in Molecular Plant.
The disciplines of evolutionary biology and plant and animal breeding have been intertwined since their inception. Plant and animal domestication served as both inspiration and evidence for Darwin’s famous monograph, On the Origin of Species. Conversely, evolutionary biologists, mainly R.A. Fisher and S. Wright, developed the statistical theory of quantitative inheritance and the mathematical framework for predicting responses to artificial and natural selection.
Evolutionary theory has also contributed to our understanding of characteristics likely to contribute to success in crop versus wild plants. Australian agronomist, C.M. Donald, recognized that while natural selection favours the competitiveness of individuals, crop productivity is enhanced by traits (i.e., the crop ideotype) that minimize individual competition and maximize group performance. For example, branching and height are traits that maximize individual competitiveness at the expense of group productivity. These ideas underpinned the development of the short, unbranched crop plants that helped drive yield gains during the green revolution.
From a crop improvement perspective, evolutionary thinking is especially useful for understanding and minimizing trade-offs, such as those often seen between resistance to biotic and abiotic stressors and productivity under ideal conditions. In his book on Darwinian Agriculture, R.F. Denison argues that plant researchers and breeders are unlikely to find simple trade-off free solutions because natural selection has had ample time to discover them over the course of plant evolution. Fortunately, there are work-arounds. If the crop environment differs from the environment in which the targeted trait initially evolved, then trade-off free improvements should be possible. An example is the recent increase in CO2 concentration due to the burning of fossil fuels. Improvements in physiological processes that take advantage of higher CO2 concentrations should be feasible without incurring yield penalties. Another example involves the evolution of RubisCO, which plays a key role in photosynthesis and global carbon cycle by converting carbon into biomass. RubisCO evolved in an atmosphere without oxygen. As a consequence, it cannot reliably distinguish CO2 from O2, reducing photosynthetic carbon fixation by about 30%. Recent genetic engineering studies have solved this problem by modifying plastid glycolate metabolic pathways, leading to large increases in productivity.
Trade-offs associated with increased resistance to biotic and abiotic stress can be minimized if resistance is inducible rather than constitutive. Inducible resistance should not be expressed under ideal conditions and thus is likely to be of little or no cost to the plant. Even in stressful environments, limiting the timing and duration of the resistance response would reduce its costs. Given this, why have plants evolved constitutive resistance at all? One explanation is that inducible resistance may act too slowly to be effective or the cues needed to induce responses are unreliable. Also, natural selection is more efficient in responding to a single, predictable environment than to multiple environments or to environments that vary in unpredictable ways. This means that there are many circumstances where constitutive resistance may evolve in natural populations – even when inducible resistance would be less costly – which opens a large window for molecular biologists and plant breeders to improve crop varieties via the engineering or breeding of inducible resistance. While speed of response and reliability of cues may also be an issue for inducible resistance in crops, there are possible solutions, including the exogenous application of bio-stimulants such as primary metabolites that induce resistance to abiotic stresses or double-stranded RNAs that induce disease resistance.
Local adaptation is another concept from evolutionary biology that is potentially useful when considering crop improvement strategies. Most natural populations of plants are adapted to their local environments. Evolutionary biologists typically document local adaptation using reciprocal transplant experiments, common gardens, or provenance trials; the latter are similar to the multi-environment evaluations carried out by breeders and agronomists. Crop varieties often exhibit local adaptation as well, especially landraces that have developed in situ and interbreed with local wild populations. In contrast, modern breeding programs aim to develop cultivars that perform well over diverse environments, which is expected to reduce the strength of local adaptation. From an evolutionary perspective, we know that human-guided selection is better able to minimize evolutionary trade-offs than natural selection (see previous paragraph), easing the development of generalist crop varieties. On the other hand, breeding for local adaptation is preferred if the environment is tightly controlled (e.g., greenhouse production).
In addition to evolutionary thinking, evolutionary biologists have developed methods that are proving to be useful for reconstructing the evolutionary history of crops and their wild relatives, detecting and mitigating the load of deleterious mutations that have accumulated in crops during their evolution, and identifying candidate genes underlying domestication and improvement traits or adaptation to local environments. I encourage crop biologists, plant breeders, and agronomists to consider evolutionary principles when interpreting their results and to incorporate evolutionary methods and thinking when designing strategies to expedite crop improvement.
Loren Rieseberg
Partner Investigator, The University of British Columbia
