Can genomics, functional analysis and genome editing help build the bridge between orphan crops and modern agriculture?
The world’s food supply depends on a few crop species, such as rice, wheat, maize, soy and potato, on which research and breeding efforts are concentrated. In addition, small farmers grow a variety of orphan crops, a set of species that are tasty, nutritious and well adapted, but mostly unsuited for intensive agriculture because of their wild characteristics. In this issue of Nature Plants, Lemmon et al.1 edit the genome of orphan crop Physalis pruinosa (groundcherry) to explore domestication of this species. Specifically, they modify genes whose orthologues control domestication traits in the close relative, tomato. The authors’ results demonstrate both the power of this approach and the importance of identifying mechanisms and gene targets. Thanks to genetic and genomic analyses2, the path to domestication from wild ancestor to modern crop is becoming clearer for several cultivated species. Evidence indicates that mutations altering the function of a few, selected loci, called domestication genes, played a determining role. For example, alleles at a few major loci are responsible for much of the difference between wild teosinte and modern maize2. Domestication genes have been identified in other key crop species: they control flowering and fruit development, increase harvest index (more product per plant), facilitate harvesting by inhibiting abscission of fruits, or make the final product easier to store, chew and digest2. Manipulation of these traits stands as one of the great human achievements. Some traits, such as loss-of-shattering, were unknowingly selected by Neolithic gatherers. Other traits, such as branching and determinate growth, required keen observation and intent to save the variant. By increasing food availability, crop domestication has enabled the flourishing of sciences, arts and technology. While the basic chassis of our staple species is Neolithic, some improvements are recent, such as semi-dwarfism in wheat and rice. Notably, tomato was radically altered to enable mechanical harvesting by combining a spontaneous mutation in SELF-PRUNING (Fig. 1)3 with alleles of other genes that make the fruit hard to bruise and rich in solids2.
The availability of genomic information and efficient genome editing tools represents a novel opportunity for crop domestication and improvement2,4. Wild species and unimproved orphan crops can now, in theory, be modified rapidly and in a targeted manner, to provide novel and improved crops. Consider groundcherry (P. pruinosa), a solanaceous species that produces a small, but tasty berry. A garden curiosity5, groundcherry cannot be grown on an agricultural scale because of wild characteristics such as sprawling habit, small, husked fruit and strong fruit abscission. The growth habit and production of small fruits unsuited for agriculture resemble the characteristics of the wild ‘currant’ tomato Solanum pimpinellifolium, which was domesticated to become tomato. Lemmon and co-workers saw an opportunity: would modification of the known gene targets of tomato domestication achieve corresponding gains in this sister species? Through gene editing, they targeted repressors of the florigen pathway to increase flower numbers and delimit flowering time, both on primary and axillary shoots. Knockout of SELF-PRUNING, a classical improvement gene that controls indeterminate versus determinate growth in tomato, was too severe to be useful, resulting in extreme compactness. Knockout of another florigen repressor, SP5G, resulted in increased axillary flowering, although caused no changes to the primary shoot; nonetheless, fruit density increased. The authors next targeted the CLAVATA pathway, which regulates shoot apical meristem size by the interaction of a small peptide, CLV3, with its receptors (CLV1 and others). Knockout of CLV1 resulted in increased flower meristem size, additional flower organs and conversion from two-locule to a larger, three-locule fruit. These manipulations produced variants better suited to, although well short of, full agricultural exploitation and constitute an impressive demonstration of what is possible through a combination of genomics and gene editing of just a small number of loci.
The study, as successful as it was, also demonstrates the challenges that ‘domesticators’ will encounter. Domesticating an orphan crop plant requires multiple tools: a well elucidated genome sequence, including the understanding of paralogue structure and gene expression, and a delivery system for genome editing, the simplest being a transformation system. Just as important is the ability to predict what targeted modification will achieve the ideal phenotype. Target identity can be inferred by understanding the domestication history of crops closely related to the orphan crop. Nonetheless, the structure of gene networks varies according to node number, type and connections6. A change yielding the desired outcome in one species may be too severe, insufficient or completely devoid of effect in another. Breeders and geneticists have long known that genetic modifiers present in populations can dramatically alter a mutation phenotype. This was demonstrated here by the dwarfing effect of the SELF-PRUNING knockout in groundcherry, or by the inability to modify primary shoot flowering by SP5G manipulation. In many cases, a knockout may be inadequate and a subtler allele may be needed instead, such as altering promoter activity or protein structure. Editing promoter segments should facilitate the production of alleles with new and useful expression properties7. Finally, the accelerated domestication envisaged here may involve manipulation of several genes with the connected combinatorial challenge of testing many variables. In fact, it is possible that domesticated species may owe their fate in part to their relative ease of genetic manipulation: if key domestication traits were monogenic and variable in the ancestor, they would have been easily apparent and selectable by breeders. Orphan crops, in contrast, may have resisted domestication because of multigenic regulation of the same traits8. All things considered, what are the prospects for ‘taming wild species’? The work by Lemmon et al.1 demonstrates both the feasibility of the approach and the importance of investing in research that furthers our knowledge of genomes, genes and the cellular mechanisms behind plant traits. Some orphan crops may be better candidates than others and some traits will be easier targets than others. This knowledge, together with the necessary tools for manipulating DNA, will be required ingredients for success. Optimistically, accelerated domestication will be an important element of the survival toolbox, the set of technologies necessary to ensure a lasting human civilization. At a minimum, the basic elucidation of how domestication genes act in different species will help improve established crops2.
Lemmon et al. Nat. Plants https://doi.org/10.1038/s41477-018-0259-x (2018).
Doebley, J. F., Gaut, B. S. & Smith, B. D. Cell 127, 1309–1321 (2006).
Pnueli, L. et al. Development 125, 1979–1989 (1998).
Østerberg, J. T. et al. Trends Plant Sci. 22, 373–384 (2017).
Riddle, S. Grow ground cherries! What a ground cherry is. Physalis. https://www.youtube.com/watch?v=LbpYx-pAwwE (2017).
Alonso-Blanco, C. et al. Plant Cell 21, 1877–1896 (2009).
Rodríguez-Leal, D., Lemmon, Z. H., Man, J., Bartlett, M. E. & Lippman, Z. B. Cell 171, 470–480 (2017).
Diamond, J. Nature 418, 700–707 (2002).
The author declares no competing interests.
About this article
Molecular Plant (2019)