How to make a domesticate

Transcription

1 contingent on a clear, quantitative understanding of the integration of organism environment relations by structural functional trait complexes, and may help safeguard the future of this fascinating and diverse group of plants. FURTHER READING Arakaki, M., Christin, P.-A., Nyffeler, R., Lendel, A., Eggli, U., Ogburn, R.M., Spriggs, E., Moore, M.J., and Edwards, E.J. (2011). Contemporaneous and recent radiations of the world s major succulent plant lineages. Proc. Natl. Acad. Sci. USA 108, Borland, A.M., Griffi ths, H., Hartwell, J., and Smith, J.A.C. (2009). Exploiting the potential of plants with crassulacean acid metabolism for bioenergy production on marginal lands. J. Exp. Bot. 60, Evans, M., Aubriot, X., Hearn, D., Lanciaux, M., Lavergne, S., Cruaud, C., Lowry, P.P., and Haevermans, T. (2014). Insights on the evolution of plant succulence from a remarkable radiation in Madagascar (Euphorbia). System. Biol. 63, Heyduk, K., McKain, M.R., Lalani, F., and Leebens-Mack, J. (2016). Evolution of a CAM anatomy predates the origins of Crassulacean acid metabolism in the Agavoideae (Asparagaceae). Mol. Phylogenet. Evol. 105, Ihlenfeldt, H.-D., (1985). Lebensformen und Übelebensstrategien bei Sukkulenten. P. Biol. 98, Klak, C., Reeves, G., and Hedderson, T. (2004). Unmatched tempo of evolution in Southern African semi-desert ice plants. Nature 427, Lüttge, U. (2004). Ecophysiology of crassulacean acid metabolism (CAM) plants. Ann. Bot. 93, Males, J. (2017). Secrets of succulence. J. Exp. Bot. (in press). Nyffeler, R., and Eggli, U. (2010). An up-to-date familial and suprafamilial classifi cation of succulent plants. Bradleya 28, Ogburn, R.M., and Edwards, E.J. (2010). The ecological water-use strategies of succulent plants. Adv. Bot. Res. 55, Ogburn, R.M., and Edwards, E.J., (2013). Repeated origin of three-dimensional leaf venation releases constraints on the evolution of succulence in plants. Curr. Biol. 23, Ripley, B.S., Abraham, T., Klak, C., and Cramer, M.D. (2013). How succulent leaves of Aizoaceae avoid mesophyll conductance limitations of photosynthesis and survive drought. J. Exp. Biol. 64, Schwinning, S., and Ehleringer, J.R. (2001). Water use trade-offs and optimal adaptation to pulse-driven arid ecosystems. J. Ecol. 89, Smith, S.A., Beaulieu, J.B., Stamatakis, A., and Donoghue, M.J. (2011). Understanding angiosperm diversifi cation using small and large phylogenetic trees. Am. J. Bot. 98, Von Willert, D.J., Eller, B.M., Werger, M.J.A., and Brinckmann, E. (1990). Desert succulents and their life strategies. Vegetatio 90, Physiological Ecology, Department of Plant Sciences, University of Cambridge, Cambridge, UK. * jom23@cam.ac.uk Primer How to make a domesticate Markus G. Stetter 1, *, Daniel J. Gates 1, Wenbin Mei 1, and Jeffrey Ross-Ibarra 1,2, * The Neolithic Revolution brought about the transition from hunting and gathering to sedentary societies, laying the foundation for the development of modern civilizations. The primary innovation that facilitated these changes was the domestication of plants and animals. In the case of plants, this involved the cultivation and selection of individuals with larger edible parts, easier harvesting, and decreased defenses, traits that allowed for the production of a food surplus and occupational specialization. Plant domestication is a process which started approximately 10,000 years ago and has thereafter been repeated independently in many locales around the world. Here, we offer a perspective that seeks to predict what factors influence the success of domestication, how many genes contributed to the process, where these genes originated and the implications for de novo domestication. What is a domesticate? Defining domestication is not straightforward, and it is likely that no one concept fits all species. Here, we define domestication as the process of adaptation to agro-ecological environments and human preferences by anthropogenic selection. The advantage of this definition is that it views the domestication status of a crop as a continuum rather than a binary trait, allowing for a spectrum of domestication from the simple tolerance or cultivation of wild plants (e.g., hops and many herbs) to semi-domesticated crops showing a number of agronomic adaptations (e.g., amaranth, flax and olive) and fully domesticated crops such as maize, barley and soybean. These and other crop species demonstrate that domestication is often gradual, ongoing and without easily defined start and end points. While much of the initial selection by humans was likely unintentional, fully domesticated species have also adapted to intentional selection as well. Part of the reason why domestication may be difficult to define is that it generally does not act upon a single trait but instead leads to a suite of morphological and physiological modifications that may differ among taxa. These changes typically affect traits related to production and human preferences (e.g., taste, seed and fruit size), and together are referred to as the domestication syndrome (Figure 1). The domestication syndrome frequently overlaps between crops with similar purposes, but may differ dramatically between those with distinct purposes. In cereals, for example, the domestication syndrome includes larger seeds as well as reduced seed shattering and dormancy, but these traits were likely of lesser importance for plants domesticated for leaves or fiber. In addition to traits common to the domestication syndrome, many domesticates may also exhibit unique phenotypic changes as well as adaptations that have allowed them to spread outside of their initial geographic region of origin. Which plants were domesticated? Successful and widespread crops comprise only a tiny fraction of angiosperm species. From the over 250,000 described angiosperms only about 2,500 crops have been partially or fully domesticated, and of these only a dozen provide more than 90% of human staple food. In the following we discuss potential explanations for the selection of a species to be domesticated, including geography, life history, and genetics. The domestication and adoption of crops was likely influenced by a number of regional and cultural factors. At least 15 centers of plant domestication have been robustly identified by archaeological and other work, each giving rise to a different assemblage of domesticates. Often, several complementary crops were domesticated alongside in a single center of domestication. For instance, energy rich cereals such as wheat and barley were domesticated together with the protein rich legumes lentil and chickpea in the Fertile Crescent, a pattern mirrored by rice and soybean in Southeast Asia or maize and common bean in the Americas. This suggests R896 27, R853 R909, September 11, 2017 Published by Elsevier Ltd.

2 A B C D Figure 1. The domestication syndrome. (A) Conversion of teosinte to maize ear involved a change from a few small, loosely connected seeds with thick fruitcases to a large maize cob with many naked seeds (photo by Hugh Iltis). (B) Loss of seed shattering during rice domestication. (C) Fruit size increase in tomato. (D) Loss of branching in sunflower leading to a single, large flower head per plant. Photos B, C and D from Doebley et al. (2006). The molecular genetics of crop domestication. Cell 127, that domestication followed similar patterns independently in distinct regions by various cultures and that a major determinant of the success of domesticates was the utility a plant offered to early societies. And while geography undoubtedly influenced the early spread of domesticates, successful domesticates nonetheless hail from diverse geographical and cultural origins. In addition to its geographic origin, a plant s life history may also influence the process of domestication. Annual plants have been very successful as domesticates, likely both because many annuals were ruderal species already adapted to disturbed environments and because the shortened generation time speeds up response to selection. Several crops show increased rates of self-fertilization compared to their wild ancestors, and self-fertilization also facilitates the maintenance of desired genotype combinations and lessens inbreeding depression. Nonetheless, the complexity of adaptation during domestication and the polygenic nature of many domestication traits suggests that at least some outcrossing likely played an important role even in primarily self-fertilizing species, providing an influx of new variation and the opportunity to combine favorable alleles on different genetic backgrounds. Asexual reproduction plays an important role in many perennial crops such as sweet potato, cassava and banana, and may allow a sort of instant domestication by immediately fixing particular combinations of traits while maintaining heterozygosity and avoiding inbreeding depression. But clonal propagation dramatically increases the effective generation time, and many modern crops that are propagated clonally probably reproduced sexually during much of their domestication history. Factors such as polyploidy have also likely contributed to the success of 27, R853 R909, September 11, 2017 R897

3 some domesticates. The advantages of polyploidy in domestication likely include increased maintenance of genetic diversity, increased recombination products, or increased opportunities for novel adaptations via homeologous genes. Overall, polyploidy appears to have played a direct role in the success of some crops such as wheat and even though most domesticates are considered diploid, many, such as maize, are relatively recent polyploids and may still benefit from similar advantages. A number of studies have demonstrated that domesticated species display an overall decrease in genetic diversity compared to their wild ancestors, consistent with pronounced demographic change during domestication. Early human agriculturalists likely sampled only a fraction of plants from natural populations, often leading to genetic bottlenecks associated with the selection of favorable phenotypes. Such population bottlenecks, combined with strong selection for adaptive traits, undoubtedly reduced the effective size of plant populations and resulted in increased genetic drift and decreased diversity within populations. Species with a larger effective population size at the onset of domestication should be more resilient to such changes, maintaining more diversity upon which subsequent selection can act and thus increasing the likelihood of successful domestication. Consistent with this idea, successful crops exhibit greater genetic diversity than most wild plants surveyed, suggesting that larger effective population size may have played a role in these species success as domesticates. Changing population size during the initial phase of domestication may have also led to an accumulation of slightly deleterious alleles, and this cost of domestication may have proven limiting for species with initially small effective population sizes. How many genes contribute? At the genetic level, it is clear that selection and demographic change during the process of domestication have resulted in the reduction of genetic diversity across the genome in most crops compared to their wild relatives. But how many loci were actually targeted by selection? While early crossing experiments in maize suggested that as few as four or five loci of large effect could explain differences in ear morphology between maize and its wild ancestor teosinte, genome-wide scans of domestication often identify hundreds of loci targeted by selection. Mapping studies seeking to characterize the specific quantitative trait loci (QTLs) contributing to adaptive domestication traits have identified numerous large-effect loci for traits such as seed shattering or branching, but most studies of this nature are under-powered to identify loci of smaller effect. Moreover, even large effect loci usually explain only a fraction of the differences between wild and domesticated taxa: QTL mapping of the classic domestication locus tb1, for example, reveals that the additive effect of the locus is likely less than 20% of the difference in lateral branching between maize and teosinte. Thus, rather than acting to fix a single large-effect locus, selection during domestication has likely acted predominantly on polygenic variation, moving the phenotype mean and reducing but not eliminating genetic variation (Figure 2). Consistent with this idea, recent work in maize has identified substantial variation in domesticated maize for traits selected for during domestication, predominantly driven by loci with small, additive effects as expected for a trait under stabilizing selection to maintain the population mean. This idea is also supported by archaeological evidence in many crops, where even traits with known QTL of large effect such as seed size in rice show continual change in the archaeological record. From whence beneficial alleles? A long-standing question in the study of evolution is whether domestication is limited by genetic variation and thus forced to wait for de novo mutations to generate beneficial variation. Relative to other study systems, domestication offers the advantage that the direct wild ancestor of domesticated species is often known and can be assayed for the relevant variation. Observations of convergent evolution, in which multiple crops show similar genetic changes for similar traits, suggests that the genes that can be targeted by selection may be limited for at least some traits. One such example is the Sh1 gene, important in the reduction of seed shattering in rice, maize and sorghum. And while it is difficult to rule out the existence of domestication alleles at low frequencies in natural populations, causal mutations for some traits such as the nonsynonymous mutation in teosinte glume architecture 1 that contributes to the reduction in hard fruit cases observed in maize have never been observed in wild plants and appear to have been selected from de novo mutations. Given the polygenic nature of most domestication traits, however, it seems unlikely that adaptation to domestication could occur if it required new mutations at each of many loci. Although domestication phenotypes such as reduced shattering or lack of seed dormancy are likely deleterious in wild populations, alleles controlling these traits can be maintained at low population frequencies, especially for loss-of-function mutations in outcrossing plants where such alleles can be masked in a heterozygous state. Because most traits are polygenic and may be under stabilizing selection in both wild and domesticated populations, it is also likely that the fitness consequences of an individual allele are not constant through time and may depend considerably on genetic background. Because selection is unlikely to reduce diversity around alleles already present on multiple haplotype backgrounds and alleles segregating in the population may not be present in the parents of individual mapping populations, the available evidence likely underestimates the importance of standing genetic variation, and these challenges are only magnified as the number of genes contributing to a trait increases. Standing genetic variation is not limited to variants that affect phenotypes in the wild ancestor, however. Crossing studies have revealed substantial genetic variation for phenotypes not present in the wild ancestor. Such cryptic variation is seen for cob phenotypes in the maize ancestor teosinte that itself does not have a cob. Selection on these variants may be substantially less, exposed only in certain environments or until sufficient phenotypic change is effected by alleles at other loci, but they may R898 27, R853 R909, September 11, 2017

4 nonetheless be an important source of large effect alleles that would otherwise be rare in the wild. Related wild taxa, which may have novel traits or have adapted to novel environments, provide yet another source for potentially adaptive variation. Adaptive introgression from wild relatives appears to have been important for a number of crops, facilitating local adaptation and even agronomic improvement in a number of species including apple, maize, tomato, and sunflower. How long did it take? The timing of crop domestication is tightly linked to human history, though how and why foragers became farmers is still a matter of some controversy. Domestication plausibly began when hunters and gatherers living in semi-permanent settlements planted desirable plants, eventually creating ecologically novel garden and field niches for those plants that fostered the planting harvesting replanting cycle required for domestication. And while Darwin described domestication as an example of accelerated evolution [3], determining the duration of a continuous process such as domestication is difficult and attempts to do so remain controversial, with studies from multiple angles coming to different conclusions. Population genetic analyses, for example, find that individual large-effect alleles could fix very rapidly, and early experimental studies in the field suggest that single domestication traits could change dramatically in as little as 30 years. In stark contrast to these results, however, archaeological remains indicate that important traits such as seed and infructescence size or seed shattering remained variable over millennia, changing only incrementally over time. Although these results appear contradictory, we argue that they are in fact consistent with a model of selection on a polygenic trait. Loci with the largest effects should experience rapid changes in allele frequency, perhaps moving the population mean considerably over shorter periods of time. But because large effect loci explain a minority of the phenotypic difference between wild and domesticated taxa, phenotypic change would continue to be observed for long periods of time. Archaeological remains provide valuable insights into phenotypic change, even though most early crop remains are small and allow inference of only a few phenotypes of interest. The addition of DNA extracted from archaeological samples, however, offers the opportunity to better understand the timing of selection during domestication. Comparison of known domestication genes in maize and barley, for example, have shown the intermediate domestication state of 5,000 to 6,000-year-old samples. The current outlook is that the above studies, while powerful, may only be scratching the surface of the overall potential of the role of ancient DNA in crop domestication. Recent methodological developments allow study of polygenic traits in ancient samples by looking for coordinated shifts in allele frequency across loci associated with phenotypic variation in extant samples. We predict that effective interrogation of ancient samples, using these and other approaches, will rapidly allow a much more detailed analysis of the duration and process of selection for many important crops. Where to go next? We have proposed that domestication is best thought of as an adaptive process instead of a binary trait, often resulting in gradual change without clear-cut phases. This process is complex, and we have argued that successful domestication depends on a number of intrinsic and extrinsic factors, including life history, utility, polyploidy, and large effective population sizes. Nonetheless, more careful consideration of the relative importance of these factors and how they act in concert could provide a useful basis for considering which plants might make good candidates for domestication and better understanding why domestication of some plants appears to have failed. While it may be ultimately difficult to identify the origin of every functional allele and most work to date has focused on alleles of large effect, we argue that most traits are polygenic and that much of the variation important for domestication existed as standing variation already segregating in wild populations. If domestication indeed proceeded via polygenic Domestication Trait value Figure 2. Evolution of domestication traits. Schematic process of the evolution of domestication traits, such as fruit abscission in apples or seed size in rice. Most traits are quantitative, showing variation in both wild and domesticated taxa. Adaptation from this standing genetic variation often results in gradual change over time, reflected in steadily decreasing variation for the trait. adaptation from standing genetic variation, this implies that breeding, rather than modification of a handful of genes, may prove a more efficient means for future domestication efforts. The experimental reproduction of domestication of wild species is likely a challenging endeavor and has yet to be accomplished. The fact the many crops have been independently domesticated multiple times suggests the idea is plausible, though in most cases multiple domestications were facilitated by gene flow among cultivated populations. De novo domestication of a new wild species may in fact be considerably more difficult, due in part to the complex genetic basis of domestication traits, limitations of life history, and the lengthy time required. Nevertheless, some recently adopted crops like sugar beet have shown dramatic adaptation in only the last few hundred years, and we argue that the early stages of such efforts may be reached rather quickly with careful selection of candidate species and modern breeding methods such as genomic selection and high throughput phenotyping. Finally, we believe that better integrating the considerations discussed above into studies of crop domestication will facilitate our 27, R853 R909, September 11, 2017 R899

5 understanding of plant adaptation to anthropogenic environments and help clarify the utility of studying domestication as an example of experimental evolution an idea championed by Darwin nearly 150 years ago. FURTHER READING Beadle, G.W. (1972). Mystery of Maize. Field Museum Natural History Bulletin. Cornille, A., Gladieux, P., Smulders, M.J., Rold an- Ruiz, I., Laurens, F., Le Cam, B., Nersesyan, A., Clavel, J., Olonova, M., Feugey, L. et al. (2012). New insight into the history of domesticated apple: secondary contribution of the European wild apple to the genome of cultivated varieties. PLoS Genet. 8, e Darwin, C. (1868). The Variation of Animals and Plants under Domestication, 2, (New York: Judd & Co). Diamond, J.M. (1999). Guns, Germs, and Steel: The Fates of Human Societies (New York: W.W. Norton & Co). Doebley, J. (2004). The genetics of maize evolution. Annu. Rev. Genet. 38, Doebley, J., Stec, A., and Gustus, C. (1995). teosinte branched1 and the origin of maize: evidence for epistasis and the evolution of dominance. Genetics 141, Fuller, D.Q., Allaby, R.G., and Stevens, C. (2010). Domestication as innovation: the entanglement of techniques, technology and chance in the domestication of cereal crops. World Archaeology 42, Gaut, B.S. (2015). Evolution is an experiment: assessing parallelism in crop domestication and experimental evolution (Nei Lecture, SMBE 2014, Puerto Rico). Mol. Biol. Evol. 32, Gutaker, R.M., and Burbano, H.A. (2017). Reinforcing plant evolutionary genomics using ancient DNA. Curr. Opin. Plant Biol. 36, Hammer, K. (1984). Das domestikationssyndrom. Die Kulturpflanze 32, Hamrick, J., and Godt, M. (1997). Allozyme diversity in cultivated crops. Crop Sci. 37, Hillman, G.C., and Davies, M.S. (1990). Domestication rates in wild-type wheats and barley under primitive cultivation. Biol. J. Linn. Soc. Lond. 39, Larson, G., Piperno, D.R., Allaby, R.G., Purugganan, M.D., Andersson, L., Arroyo-Kalin, M., Barton, L., Vigueira, C.C., Denham, T., Dobney, K. et al. (2014). Current perspectives and the future of domestication studies. Proc. Natl. Acad. Sci. USA 111, Lin, Z., Li, X., Shannon, L.M., Yeh, C.-T., Wang, M.L., Bai, G., Peng, Z., Li, J., Trick, H.N., Clemente, T.E. et al. (2012). Parallel domestication of the Shattering1 genes in cereals. Nat. Genet. 44, Salman-Minkov, A., Sabath, N., and Mayrose, I. (2016). Whole-genome duplication as a key factor in crop domestication. Nat. Plants 2, Wang, L., Beissinger, T.M., Lorant, A., Ross-Ibarra, C., Ross-Ibarra, J., and Hufford, M. (2017). The interplay of demography and selection during maize domestication and expansion. biorxiv Xue, S., Bradbury, P. J., Casstevens, T., and Holland, J.B. (2016). Genetic architecture of domestication-related traits in maize. Genetics 204, Department of Plant Sciences, University of California, Davis, CA 95616, USA. 2 Genome Center and Center for Population Biology, University of California, Davis, CA 95616, USA. * mstetter@ucdavis.edu (M.G.S.); rossibarra@ucdavis.edu (J.R.-I.) Primer Size-dependent variation in plant form Karl J. Niklas* and Edward D. Cobb The study of organic form has a long and distinguished history going at least as far back as Aristotle s Historia Animālium, wherein he identified five basic biological processes that define the forms of animals (metabolism, temperature regulation, information processing, embryo development, and inheritance). Unfortunately, all of Aristotle s writings about plant forms are lost. We know of them only indirectly from his student Theophrastus s companion books, collectively called Historia Plantarum, wherein plant forms are categorized into annual herbs, herbaceous perennials, shrubs, and trees. The study of plant forms did not truly begin until the romantic poet and naturalist Goethe proposed the concept of a hypothetical Plant Archetype, declared Alles ist Blatt, and first coined the word morphologie, which inspired the French anatomist Cuvier (who established the field of comparative morphology), the English naturalist Darwin (who saw his theory of evolution reinforced by it), and the Scottish mathematician D Arcy Thompson (who attempted to quantify it). Although the variety of animal and plant forms is seemingly endless, at a fundamental level Theophrastus, Cuvier, Darwin, and Thompson saw that this variety conformed to certain rules, which in part conformed to physical laws and principles. This is seen perhaps nowhere more clearly than in plants (here, broadly defined as photosynthetic eukaryotes to include the algae as well as the land plants). With few exceptions, cylinders, flat oblate spheroids, and disks are the basic geometries that can be used to approximate the majority of plant body plans, and for good reason they are excellent at dealing with mechanical forces and for exchanging mass and energy between an object and its surrounding fluid (water or air). Natural selection has tested the worth of these geometries over billions of years, and they have not been found wanting, because there are rules about how size, shape, and geometry must co-vary to remain functional. These rules have quite literally shaped the history of life in subtle and not so subtle ways. The goal of this Primer is to review some of these rules, and to show how they are useful in understanding the biology of plants. Shape and g eometry The first rule to consider is that shape and geometry are not the same thing, and that each can be changed as the size of cell, organ, or organism ontogenetically changes over time, or as a consequence of evolutionary modification by means of natural selection. There are many ways of defining what is meant by shape. However, herein shape is defined simply as any dimensionless quotient constructed using dimensions that define the geometry of the object being considered. The distinction between shape and geometry and the use of dimensional quotients to define shape can be illustrated by considering a simple geometric class of objects, such as a cylinder. The cylinder is constructed by translating a circle orthogonally along a straight line, whereas its shape can change by either increasing the radius of the circle, or by changing the length of the translation axis. In this example, the shape of a cylinder can be defined by the dimensionless quotient of length and radius, otherwise known as an aspect ratio. Translating a circle over a long distance results in a very slender cylinder, whereas translating the same circle over a very short distance results in a circular disk. With the exception of the sphere, the shape of every geometric class can be changed independently of size (e.g., length or volume). Indeed, even a sphere can be deformed to produce an oblate or prolate spheroid. The implications of the separation of shape and geometry are profoundly biologically important at the level of the organelle, cell, organ, and organism because the independence of shape and geometry from each other (and the independence of either from R900 27, R853 R909, September 11, Elsevier Ltd.