This is the accepted version of the article: Preece, Catherine; Peñuelas, Josep. «A return to the wild : root exudates and food security». Trends in plant science, Vol. 25, issue 1 (Jan. 2020), p. 14-21. DOI 10.1016/j.tplants.2019.09.010 This version is avaible at https://ddd.uab.cat/record/218219 under the terms of the license 1 2 A return to the wild: root exudates and food security 3 4 5 6 Catherine Preece1,2,* and Josep Peñuelas1,2 7 1 8 CREAF, Cerdanyola del Vallès, 08193, Catalonia, Spain; 2 CSIC, Global Ecology Unit CREAF- 9 CSIC-UAB, Bellaterra 08193, Catalonia, Spain 10 *Corresponding author 11 12 Keywords: wild ancestors, rhizodeposition, agriculture, pesticide, fertiliser, traits. 13 Abstract 14 Challenges to food security under conditions of global change are forcing us to increase global crop 15 production. Focusing on belowground plant traits, especially root exudation, has great promise to 16 meet this challenge. Root exudation is the release of a vast array of compounds into the soil. These 17 exudates are involved in many biotic and abiotic interactions. Wild relatives of crops provide a large 18 potential source of information and genetic material and have desirable traits that could be 19 incorporated into modern breeding programmes. Root exudates, however, are currently under- 20 exploited. We highlight how the traits of root exudates of crop wild relatives could be used to improve 21 agricultural output and reduce environmental impacts, particularly by decreasing our dependence on 22 pesticides and fertilisers. 1 23 Roots and their exudates in the fight for food security 24 Challenges to food security (see Glossary) are forcing us to increase global crop production in order 25 to feed the growing population. The ability to provide enough food to humans around the world, in a 26 sustainable way, is endangered by a range of environmental and social factors including climate 27 change causing widespread droughts and other extreme weather events, soil erosion, changing diets 28 (such as increased meat consumption) and high food waste. An array of strategies can be implemented 29 to try to meet this task including increased cropping efficiency, closing yield gaps through improved 30 management, reducing waste and encouraging plant-based diets [1]. Characteristics of crops 31 themselves can also be modified, and the first Green Revolution increased yields by modifying 32 important aboveground traits. These ‘improvements’, however, are only possible when water and 33 nutrients are not limited, e.g. at high levels of irrigation and fertilisation. These conditions are 34 impossible for subsistence farmers and are becoming increasingly undesirable even for farmers that 35 can afford them, due to issues with sustainability and pollution [2-4]. New ideas and techniques are 36 thus needed to meet these challenges. Ecologists and plant scientists must therefore target new 37 beneficial plant traits to boost food production. Increasing the focus on belowground plant 38 characteristics may thus offer great promise [5-7]. Belowground traits vary greatly between species 39 and individuals and with environmental conditions [8, 9], so they represent a mostly unexplored area 40 in the search for tools to ameliorate the current threats to food security. Roots are the main plant 41 structures responsible for the acquisition of both water and nutrients but have largely been ignored, 42 mainly due to challenges associated with sampling within the soil [10, 11]. A wide range of traits 43 related to the roots and rhizosphere (see Glossary) could potentially be targeted to improve yields and 44 lessen the inputs of fertilisers and pesticides in the future, and attention is finally turning belowground 45 [12]. 46 Root exudation remains particularly under studied. Exudation is the release of a vast array of 47 compounds into the rhizosphere, including sugars, organic acids, amino acids, secondary metabolites, 48 and structural carbohydrates [13]. Root exudates are involved in a wide range of biotic interactions 2 49 with other plants, microbes in the rhizosphere, and abiotic components of the soil, including nutrients 50 [14]. More specifically, root exudates can be an important source of carbon (C) for bacteria and fungi. 51 They also contain compounds that repel pathogens or attract beneficial microbes, such as nitrogen 52 (N) fixers [13]. Exudates may also increase the availability of some nutrients, such as phosphorus 53 (P), due to the release of phosphatases and chelating organic acids that render P available for plant 54 uptake [15]. Exudates can also negatively affect neighbouring plants such as through the production 55 of allelochemicals (see Glossary) [16]. 56 57 Crop wild relatives as a source of advantageous traits of root exudation 58 Modern agriculture is centred on a very small number of crop species, but their wild relatives provide 59 a large potential source of information and genetic material [17, 18]. Understanding the processes 60 that have led to the development of modern crops can help us to understand differences in root 61 exudation due to domestication (see Glossary). Most current crops are the result of both thousands of 62 years of natural evolution and intensive selective breeding to maximise yields, leading to clear trait 63 differences from their wild relatives [19]. Crop wild relatives (CWRs) (see Glossary) have desirable 64 traits linked to mycorrhizal (see Glossary) associations [20], pest resistance [21], and tolerance of 65 challenging climatic conditions [22]. The effect of crop domestication on the provision of ecosystem 66 services, however, has not been well studied, leaving many gaps in our knowledge [23, 24], especially 67 relating to root exudation. 68 Exudation varies between species and supplies about 15% of belowground C in cereals and 69 grasses [25] and can therefore be an expensive process to maintain. Exudate production thus 70 represents a balance between the possible advantageous functions that exudates bestow on plants 71 against the loss of energy that could otherwise be allocated elsewhere. The more we learn about root 72 exudates, the clearer it becomes that they can show a high level of variation, in both amount and 73 composition, at all levels of organisation. For example, patterns of exudation vary between species 74 and individuals, but root exudation also varies within individuals over time and in response to abiotic 3 75 factors, such as the availability of water or nutrients [26]. Evidence suggests that the amount of 76 exudation increases when plants experience drought [27-29] and low P availability [30-32], although 77 results also vary depending on the experimental conditions. Exudate composition also varies in 78 response to the availability of nutrients such as N, P, and potassium [26, 33, 34]. Plant breeders are 79 now focusing more on CWRs to provide important new developments in the next generation of crops 80 but identifying differences in root exudation between crops and their wild relatives will be a 81 challenge. These differences may be general patterns that are repeated across plant families (e.g. 82 grasses, legumes, and brassicas) or may be much more idiosyncratic, with different, potentially 83 advantageous traits in some species but not others. The large potential benefit of these new findings, 84 combined with the current lack of knowledge in this area, will likely generate an area of great interest 85 for future research. 86 By reviewing the available literature on root exudation, we hope to identify sources of 87 variation between crops and their wild relatives and to learn how this variation may offer 88 opportunities for plant breeders to develop the next generation of crops. We will therefore discuss the 89 differences in root exudation and the interactions within the rhizosphere between crops and their wild 90 relatives. This discussion will allow us to assess the potential to identify advantageous traits of root 91 exudation in CWRs that could improve agricultural productivity and sustainability, with a focus on 92 reducing fertilisation and pesticide use. We will also discuss potential obstacles to this process, which 93 could be focal areas for improving the likelihood of success. 94 95 Improving tolerance to pests 96 Traits from CWRs may benefit crop breeding, in order to reduce pesticide use. Root exudates 97 participate in a wide range of positive and negative interactions with soil organisms. Primarily these 98 associations are with microorganisms in the rhizosphere [35], but important interactions also occur 99 with invertebrates [36]. Studies of plant defence generally provide support for the hypothesis that 100 crop exudates can be modified to improve pest resistance. Crops tend to have lower resistance to 4 101 attacks from herbivores compared with their wild ancestors [37]. This is perhaps due to direct 102 selection to remove protective but undesirable food traits, such as a bitter taste, toxicity, toughness, 103 or hairiness, and/or because breeding for increased yield has led to trade-offs with defensive 104 mechanisms [38, 39]. Domestication can reduce the level of resistance of aboveground tissues to 105 herbivores [37], for example with lower silicon concentration in the leaves of cereals [40], lower 106 foliar toughness in domesticated maize (Zea mays) compared with wild relatives [41], and lower 107 levels of glucosinolate defences in domesticated compared with wild cabbage (Brassica oleracea) 108 [42]. There are also examples of the loss, during domestication, of herbivore-induced VOCs from the 109 leaves of maize [43] and cranberry (Vaccinium macrocarpon) [44]. These VOCs are beneficial to the 110 plant as they attract other species that predate the herbivore. These results, however, contrast with 111 others demonstrating that two parasitoids were more attracted to domesticated kale (Brassica 112 oleracea) and scarlet runner bean (Phaseolus coccineus) than their wild relatives, implying that the 113 foliar VOCs of the crops had not been removed by selection during domestication [45]. 114 Whether consistent reductions in root defence to pests are due to domestication remains to be 115 confirmed, but good evidence of impacts on belowground VOCs produced by plants has been found. 116 Maize commonly suffers herbivory from the larvae of the beetle Diabrotica virgifera virgifera, also 117 known as the Western corn rootworm. The wild ancestor of maize, teosinte (Zea mays subsp. 118 parviglumis), and European maize varieties produce and emit the sesquiterpene (E)--caryophyllene 119 from their roots, which attracts an entomopathogenic nematode that feeds on the beetle larvae. This 120 chemical signal, however, is not released by most North American maize cultivars, which are thus 121 more susceptible to beetle attack; this trait may have been lost during breeding [46, 47]. 122 Reincorporation of the ability to produce this sesquiterpene can greatly benefit the productivity of 123 North American maize and is an example of a wild trait that has been successfully restored to a crop 124 [48]. This indicates the potential for using gene editing or classical breeding techniques for 125 manipulating the traits of root exudates to reduce damage by pests, thereby lowering the requirement 126 for high rates of pesticide use. 5 127 128 Opportunities to reduce fertilisation 129 Contemporary intensive farming systems use high levels of fertilisation to counteract the constraints 130 to growth due to too little soil N, P, or other elements. Reducing fertiliser use, however, is desirable 131 for environmental and economic reasons [49]. Global levels of fertilisation are higher than ever but 132 are not a completely new phenomenon. Humans became less nomadic at the beginning of agriculture 133 with the cultivation of plants, and the soils in these early settlements would have been richer in 134 nutrients compared with natural soils due to the use of human and animal waste [50]. Nutrient 135 acquisition is therefore not the main requirement in most crops that have evolved and been bred in 136 these conditions [20]. Wild species that have not been domesticated may therefore have traits that are 137 better for acquiring soil nutrients. Crop wild relatives might thus produce higher amounts of exudates 138 compared to crops, different extracellular enzymes with higher efficiency in phosphorus 139 solubilisation, and/or a higher proportion of organic anions, whereby crops would not need to invest 140 as much in mobilising mineral P. The very limited study of crops and their wild relatives, however, 141 has not yet supported this premise. The composition of exudate metabolites in one recent study 142 differed between modern durum wheat (Triticum turgidum subsp. durum), domesticated emmer 143 wheat (T. turgidum subsp. dicoccum), and wild emmer (T. turgidum subsp. dicoccoides), but the 144 pattern was not clear for organic acids, and the effects depended on the substrate in which the plants 145 were grown [51], indicating the complexity of the potential effects of domestication on exudation 146 composition. 147 Other root exudates that can directly mobilise nutrients include those involved in the uptake 148 of iron, an element that often has low bioavailability in soils. In non-graminaceous plants, rhizosphere 149 acidification combined with the release of phenolics (such as coumarins), and flavins facilitates iron 150 uptake (strategy I) [52, 53]. In grasses, chelators called phytosiderophores are exuded into the 151 rhizosphere (strategy II) greatly increasing the uptake of iron and, to a lesser degree, other 6 152 micronutrients such as copper, cobalt, nickel and zinc [53, 54]. Although comparisons of 153 phytosiderophore production between crops and CWRs are few, one study comparing Triticum 154 aestivum and T. durum with six Aegilops species found exudation of phytosiderophores was three to 155 four times higher in the wild Aegilops cultivars compared with domesticated wheat cultivars, and 156 moreover, Aegilops plants had higher shoot and root concentrations of iron and zinc [55]. The 157 Aegilops genus is closely related to bread wheat and has already significantly contributed to wheat 158 breeding [56], making this a good target for improvements relating to root exudation. 159 Examples comparing crops and CWRs are still lacking, but much evidence suggests that the 160 nutrient content of soil affects the composition of exudates [57]. Plant species from the family 161 Proteaceae notably produce different types of roots when growing in soil low in nutrients. For 162 example, white lupin (Lupinus albus) can grow thicker “cluster roots” (or “proteoid roots”) in the P- 163 poor soils of Australia and elsewhere [58, 59]. These roots have larger surface areas and exude higher 164 amounts of carboxylates (organic anions, disassociated forms of the organic acid carboxylic acid), 165 which are important for changing soil pH and can chelate soil minerals and mobilise P in the soil [31, 166 58]. An increase in exudation of carboxylates has been experimentally demonstrated under conditions 167 in low P soils. For example, the production of root exudates in low P conditions in an experiment 168 with sugar beet increased 4-5-fold [30]. Exudation in a study with white lupin even increased 25-fold 169 (mostly malate and citrate) when the plants were grown in low P conditions [60]. While the addition 170 of cluster roots is not possible or looked-for in all crops, greater understanding of organic acid 171 exudation in CWRs, and the potential for increasing P-mobilising abilities in crops are clearly 172 desirable. 173 Root exudates also play an important role in promoting positive interactions with 174 microorganisms, e.g. initiating colonisation with mycorrhizae by the release of strigolactones (see 175 Glossary) [61]. Evidence has already been found of differences due to domestication. Colonisation 176 by mycorrhizae was lower in breadfruit (Artocarpus altils) than its wild relatives [62], and 177 mycorrhizal responsiveness (plant improvement in the presence of mycorrhizae) was lower in modern 7 178 than older cultivars of wheat [63]. Other studies, however, have found that colonisation by 179 mycorrhizae is not necessarily lower in domesticated cultivars [64] and that colonisation in annual 180 crops may be lower than in their wild ancestors, but new cultivars did not lack the capability to form 181 these associations and could even be more responsive to mycorrhizae than wild ancestors [65]. The 182 largest experimental study to date on this topic (Martín-Robles et al., 2018) measured how 183 domestication affected mycorrhizal responsiveness in 27 crop species and their wild progenitors, 184 concluding that crops only benefitted from mycorrhizae in P‐limited conditions, whereas their wild 185 ancestors benefited irrespective of P availability [20]. 186 The symbiotic relationship between plants and N-fixing bacteria is another positive 187 interaction between roots and microorganisms, with the legumes-rhizobia (see Glossary) relationship 188 being the most common example. Leguminous species cannot form symbioses with all types of 189 rhizobia, and the amount of specificity varies between plant species. Root exudation is part of this 190 process, because rhizobia are attracted to roots of suitable host plants by the release of flavonoid (see 191 Glossary) compounds in the exudates [66]. Evidence from pea, faba bean, and chickpea suggests that 192 domestication has reduced the ability of legumes to associate with many rhizobial populations, 193 leading to lower symbiont diversity compared with related wild species [67-70]. We continue to lack 194 information, however, about consistent differences in the amount or type of flavonoids released by 195 crops compared with their wild relatives. Increasing the ability of plants to take up nutrients by 196 increasing beneficial interactions with microbes is another possible route to lower the dependence on 197 fertilisers. 198 199 Further clues of the effects of domestication on the rhizobiome 200 We can infer information about the effects of domestication on root exudation from the differences 201 in rhizosphere microorganisms, because the rhizobiome (see Glossary) is intimately linked to and 202 shaped by root exudates [71, 72]. We can thus use knowledge about the rhizobiome as a mirror to 8 203 reveal changes in exudation. Edaphic factors and other C pools, though, are also very important [72, 204 73]. Abundant evidence supports divergent patterns of exudation between crops and their wild 205 relatives, in a range of species. A general pattern of reduced diversity in the crop rhizobiome has been 206 reported [70], and pot experiments with controlled soil conditions allow the observation of differences 207 due to domestication. For example, small but significant differences in the composition of bacterial 208 root microbiota have been found between wild, traditional, and modern cultivars of barley (Hordeum 209 vulgare) [74]. Bacterial alpha diversity is significantly higher in the rhizospheres of teosinte than 210 domesticated sweet corn [75], wild Agave than domesticated Agave tequilana [76], and wild beet 211 (Beta vulgaris subsp. maritima) than modern sugar beet (Beta vulgaris subsp. vulgaris) [77]. In 212 contrast, bacterial diversity is higher in lettuce (Lactuca sativa) than its wild relative, L. serriola [78]. 213 Our knowledge of crop rhizobiomes generally suggests a reduction in microbial diversity due 214 to domestication. Exudates are a major source of substrates on which rhizosphere microbes feed so 215 crops may exude a less diverse array of compounds, creating a lower diversity of metabolic niches 216 for microbes to exploit [79]. Domestication has decreased the genetic diversity within crops compared 217 with their wild relatives due to the strong selection of a small proportion of the original wild 218 populations that contained favourable traits [80], and evidence suggests that this loss of diversity has 219 affected various ecological aspects of crop roots and rhizospheres. There is therefore an opportunity 220 to recover some of this lost diversity via breeding programmes with CWRs. 221 222 Concluding remarks and future directions 223 To date, CWRs have been used to improve many crops, especially sunflower, wheat, and 224 potato, with breeding programmes using CWRs to improve resistance to biotic and abiotic stresses, 225 increase yields, and improve fertility [18, 81]. Few programmes, if any, however, have incorporated 226 wild traits associated with root exudation. The quantity and composition of root exudates may differ 227 between domesticated crop species and their wild relatives. This research may be relatively 9 228 undeveloped, but it offers a range of potential benefits to future cultivars. Two key areas could be 229 exploited (Figure 1). Firstly, defence against root pests is a promising goal, through augmentation of 230 beneficial VOCs and other non-volatile secondary metabolites. There is already clear evidence 231 suggesting that VOCs from root exudates protecting against herbivory vary between maize cultivars 232 [46, 47], and indeed, the ability to emit these VOCs was successfully introduced into a non-emitting 233 maize cultivar, leading to much lower damage from a beetle pest [48]. Secondly, reducing the 234 dependence on fertilisers may be possible via two routes: (1) incorporation of traits from CWRs that 235 increase the availability of soil P, perhaps with mechanisms similar to those in Proteaceae such as 236 higher exudation of organic acids; (2) increasing positive associations with microbes, such as N- 237 fixing bacteria and mycorrhizae, because the pairing of symbionts or fungal partners of some current 238 crops may be restricted, and more beneficial partnerships could be nurtured. A third area, which we 239 have not focused on in this review, is the potential for root exudates in CWRs to have traits relating 240 to weed suppression. Although future applications for allelopathy in agriculture have been recognised 241 [82] studies on the difference in these types of allelopathic traits between crops and their wild relatives 242 are still greatly lacking and is a clear topic for future work. 243 Two main complicating factors limit the incorporation of CWR traits of root exudation into 244 crops. Firstly, perhaps the only consistent result from the different studies on root exudation is the 245 inconsistency in exudation patterns. That is, exudation varies greatly, not just at the level of the 246 individual or species, but also depending on soil type, availability of water or soil nutrients, plant age, 247 light intensity, and other factors. Moreover, differences in the growth and sampling conditions affect 248 root exudation, and this is vastly understudied. Growth conditions can vary from hydroponics, to soil 249 in pots and microcosms, to growing in the field. Sampling can be done via many different methods, 250 including direct measurements of exudates collected in solution, soil or via another type of ‘trap’, or 251 with isotope labelling, and with samples taken from a root segment or from the whole root system, 252 and with many other variations (see the review in Oburger et al. 2018 [83] for more details). 253 Identifying differences that are due to domestication and predicting how the incorporation of a new 10 254 trait would function outside the controlled conditions of a laboratory or greenhouse are thus very 255 complicated, and without a large effort to standardise the conditions under which exudates are 256 measured we will likely continue to struggle to draw general conclusions. The effects of 257 domestication on exudation also likely differ between crops where the seeds, fruits, or leaves are 258 consumed and crops where the roots and tubers are eaten, due to differences in above- or 259 belowground. The second challenge is that it remains difficult to reliably measure root exudation, 260 particularly in field conditions, where efforts to find reliable and serviceable methods should be 261 focused [83]. Also, regarding work investigating the composition of exudates, much of the current 262 research is based around targeted analysis, as non-targeted metabolomics analysis is currently 263 expensive both in terms of time and cost of equipment. So, it must be noticed that when only looking 264 at a small fraction of the large diversity of compounds that are released other important patterns may 265 be missed. 266 During the last decades, the replacement of locally adapted, diverse, traditional varieties with 267 cultivars bred for high productivity in intensive agricultural systems means that there is too often a 268 very limited focus on sustainability [84]. The investigation of CWRs to find possible beneficial traits 269 associated with root exudation offers many possibilities for improving both the quality of our food 270 and the agricultural practices that produce it. Efforts may first need to be focused on improving the 271 measurement of root exudation in realistic environments, but the opportunity for substantial advances 272 to improve the resilience of our food production encourages the continued search for beneficial root 273 traits in crop wild relatives. 274 275 Acknowledgements 276 This research was supported by the European FP7 S-Clima project PIEF-GA-2013-626234, the 277 Spanish Government grant CGL2016-79835, the European Research Council Synergy grant ERC- 278 2013-SyG-610028 IMBALANCE-P, and the Catalan Government grant SGR 2017-1005. 11 279 Fig. 1. A summary of some of the beneficial traits that could be incorporated from crop wild relatives to improve agricultural practices and food 280 security. An increase in the release of volatile organic compounds (VOCs) and other non-volatile secondary metabolites (e.g., phenolics, alkaloids), 281 could attract predators of root pests or directly inhibit herbivores and weeds. This would lead to increased pest resistance, therefore reducing the use of 282 pesticides. An increase in exudation of organic acids and more beneficial associations with microbes (e.g., mycorrhizae and rhizobia) could increase 283 soil nutrient availability (especially N and P), reducing dependence on fertilisers. 284 285 12 286 13 287 References 288 1. Foley, J.A. et al. (2011) Solutions for a cultivated planet. Nature 478 (7369), 337. 289 2. Horlings, L.G. and Marsden, T.K. (2011) Towards the real green revolution? Exploring the 290 conceptual dimensions of a new ecological modernisation of agriculture that could ‘feed the world’. 291 Global Environmental Change 21 (2), 441-452. 292 3. Cole, M.B. et al. (2018) The science of food security. npj Science of Food 2 (1), 14. 293 4. Zhang, X. et al. (2015) Managing nitrogen for sustainable development. Nature 528, 51. 294 5. Bishopp, A. and Lynch, J.P. (2015) The hidden half of crop yields. Nature Plants 1, 15117. 295 6. Faucon, M.-P. et al. (2017) Plant functional traits: soil and ecosystem services. Trends in Plant 296 Science 22 (5), 385-394. 297 7. Laliberté, E. (2017) Below-ground frontiers in trait-based plant ecology. New Phytologist 213 (4), 298 1597-1603. 299 8. Freschet, G.T. et al. (2017) Climate, soil and plant functional types as drivers of global fine-root trait 300 variation. Journal of Ecology 105 (5), 1182-1196. 301 9. Iversen, C.M. et al. (2017) A global Fine-Root Ecology Database to address below-ground challenges 302 in plant ecology. New Phytologist 215 (1), 15-26. 303 10. McCormack, M.L. et al. (2017) Building a better foundation: improving root-trait measurements to 304 understand and model plant and ecosystem processes. New Phytologist 215 (1), 27-37. 305 11. Oburger, E. and Schmidt, H. (2016) New methods to unravel rhizosphere processes. Trends in Plant 306 Science 21 (3), 243-255. 307 12. Meister, R. et al. (2014) Challenges of modifying root traits in crops for agriculture. Trends in Plant 308 Science 19 (12), 779-788. 309 13. Walker, T.S. et al. (2003) Root exudation and rhizosphere biology. Plant Physiology 132 (1), 44- 310 51. 311 14. Bais, H.P. et al. (2006) The role of root exudates in rhizosphere interactions with plants and other 312 organisms. Annu. Rev. Plant Biol. 57, 233-266. 313 15. Dakora, F.D. and Phillips, D.A. (2002) Root exudates as mediators of mineral acquisition in low- 314 nutrient environments. Plant and Soil 245 (1), 35-47. 315 16. Callaway, R.M. and Aschehoug, E.T. (2000) Invasive plants versus their new and old neighbors: a 316 mechanism for exotic invasion. Science 290 (5491), 521-523. 317 17. Brozynska, M. et al. (2016) Genomics of crop wild relatives: expanding the gene pool for crop 318 improvement. Plant Biotechnology Journal 14 (4), 1070-1085. 319 18. Dempewolf, H. et al. (2017) Past and future use of wild relatives in crop breeding. Crop Science 57 320 (3), 1070-1082. 321 19. Meyer, R.S. et al. (2012) Patterns and processes in crop domestication: an historical review and 322 quantitative analysis of 203 global food crops. New Phytologist 196 (1), 29-48. 323 20. Martín-Robles, N. et al. (2018) Impacts of domestication on the arbuscular mycorrhizal symbiosis 324 of 27 crop species. New Phytologist 218 (1), 322-334. 325 21. Chen, Y.H. et al. (2015) Crop domestication and its impact on naturally selected trophic interactions. 326 Annual Review of Entomology 60, 35-58. 327 22. Roy, S.J. et al. (2011) Genetic analysis of abiotic stress tolerance in crops. Current opinion in plant 328 biology 14 (3), 232-239. 329 23. Milla, R. et al. (2017) Looking at past domestication to secure ecosystem services of future 330 croplands. Journal of Ecology 105 (4), 885-889. 331 24. Warschefsky, E. et al. (2014) Back to the wilds: tapping evolutionary adaptations for resilient crops 332 through systematic hybridization with crop wild relatives. American journal of botany 101 (10), 1791- 333 1800. 334 25. Pausch, J. and Kuzyakov, Y. (2018) Carbon input by roots into the soil: Quantification of 335 rhizodeposition from root to ecosystem scale. Global Change Biology 24 (1), 1-12. 336 26. Carvalhais, L.C. et al. (2011) Root exudation of sugars, amino acids, and organic acids by maize as 337 affected by nitrogen, phosphorus, potassium, and iron deficiency. Journal of Plant Nutrition and Soil 338 Science 174 (1), 3-11. 339 27. Preece, C. and Peñuelas, J. (2016) Rhizodeposition under drought and consequences for soil 340 communities and ecosystem resilience. Plant and Soil 409 (1), 1-17. 14 341 28. Preece, C. et al. (2018) Thirsty tree roots exude more carbon. Tree physiology 38 (5), 690–695. 342 29. Canarini, A. et al. (2016) Drought effects on Helianthus annuus and Glycine max metabolites: from 343 phloem to root exudates. Rhizosphere 2, 85-97. 344 30. Khorassani, R. et al. (2011) Citramalic acid and salicylic acid in sugar beet root exudates solubilize 345 soil phosphorus. BMC Plant Biology 11 (1), 121. 346 31. Lambers, H. et al. (2015) Plant adaptations to severely phosphorus-impoverished soils. Current 347 Opinion in Plant Biology 25, 23-31. 348 32. Øgaard, A.F. et al. (2015) Contrasting responses of root morphology and root-exuded organic acids 349 to low phosphorus availability in three important food crops with divergent root traits. AoB PLANTS 350 7. 351 33. Kraffczyk, I. et al. (1984) Soluble root exudates of maize: Influence of potassium supply and 352 rhizosphere microorganisms. Soil Biology and Biochemistry 16 (4), 315-322. 353 34. Hoffland, E. et al. (2006) Organic anion exudation by lowland rice (Oryza sativa L.) at zinc and 354 phosphorus deficiency. Plant and Soil 283 (1), 155-162. 355 35. Huang, X.-F. et al. (2014) Rhizosphere interactions: root exudates, microbes, and microbial 356 communities. Botany 92 (4), 267-275. 357 36. Peñuelas, J. et al. (2014) Biogenic volatile emissions from the soil. Plant, Cell & Environment 37 358 (8), 1866-1891. 359 37. Whitehead, S.R. et al. (2017) Domestication impacts on plant–herbivore interactions: a meta- 360 analysis. Phil. Trans. R. Soc. B 372 (1712), 20160034. 361 38. Heil, M. and Baldwin, I.T. (2002) Fitness costs of induced resistance: emerging experimental 362 support for a slippery concept. Trends in plant science 7 (2), 61-67. 363 39. Stenberg, J.A. et al. (2015) Optimizing crops for biocontrol of pests and disease. Trends in Plant 364 Science 20 (11), 698-712. 365 40. Simpson, K.J. et al. (2017) Still armed after domestication? Impacts of domestication and agronomic 366 selection on silicon defences in cereals. Functional Ecology 31 (11), 2108-2117. 367 41. Bellota, E. et al. (2013) Physical leaf defenses – altered by Zea life-history evolution, domestication, 368 and breeding – mediate oviposition preference of a specialist leafhopper. Entomologia Experimentalis 369 et Applicata 149 (2), 185-195. 370 42. Moreira, X. et al. (2018) Plant domestication decreases both constitutive and induced chemical 371 defences by direct selection against defensive traits. Scientific Reports 8 (1), 12678. 372 43. de Lange, E.S. et al. (2016) Comparing the attraction of two parasitoids to herbivore-induced 373 volatiles of maize and its wild ancestors, the teosintes. Chemoecology 26 (1), 33-44. 374 44. Rodriguez-Saona, C. et al. (2011) Tracing the history of plant traits under domestication in 375 cranberries: potential consequences on anti-herbivore defences. Journal of Experimental Botany 62 (8), 376 2633-2644. 377 45. Benrey, B. et al. (1998) The effects of domestication of Brassica and Phaseolus on the interaction 378 between phytophagous insects and parasitoids. Biological Control 11 (2), 130-140. 379 46. Rasmann, S. et al. (2005) Recruitment of entomopathogenic nematodes by insect-damaged maize 380 roots. Nature 434 (7034), 732. 381 47. Köllner, T.G. et al. (2008) A maize (E)-β-caryophyllene synthase implicated in indirect defense 382 responses against herbivores is not expressed in most American maize varieties. The Plant Cell 20 (2), 383 482-494. 384 48. Degenhardt, J. et al. (2009) Restoring a maize root signal that attracts insect-killing nematodes to 385 control a major pest. Proceedings of the National Academy of Sciences 106 (32), 13213-13218. 386 49. Erisman Jan, W. et al. (2013) Consequences of human modification of the global nitrogen cycle. 387 Philosophical Transactions of the Royal Society B: Biological Sciences 368 (1621), 20130116. 388 50. Abbo, S. and Gopher, A. (2017) Near Eastern plant domestication: a history of thought. Trends in 389 Plant Science 22 (6), 491-511. 390 51. Iannucci, A. et al. (2017) Evolution of the crop rhizosphere: impact of domestication on root 391 exudates in tetraploid wheat (Triticum turgidum L.). Frontiers in Plant Science 8 (2124). 392 52. Tsai, H.H. and Schmidt, W. (2017) Mobilization of iron by plant-borne coumarins. Trends in plant 393 science 22 (6), 538-548. 394 53. Chen, Y.-T. et al. (2017) Role of root exudates in metal acquisition and tolerance. Current opinion 395 in plant biology 39, 66-72. 15 396 54. Oburger, E. et al. (2014) Root exudation of phytosiderophores from soil‐grown wheat. New 397 Phytologist 203 (4), 1161-1174. 398 55. Neelam, K. et al. (2010) Identification of Aegilops species with higher production of 399 phytosiderophore and iron and zinc uptake under micronutrient-sufficient and-deficient conditions. 400 Plant Genetic Resources 8 (2), 132-141. 401 56. Kishii, M. (2019) An update of recent use of Aegilops species in wheat breeding. Frontiers in plant 402 science 10, 585. 403 57. Badri, D.V. and Vivanco, J.M. (2009) Regulation and function of root exudates. Plant, Cell & 404 Environment 32 (6), 666-681. 405 58. Watt, M. and Evans, J.R. (1999) Proteoid Roots. Physiology and Development. Plant Physiology 406 121 (2), 317-323. 407 59. Laliberté, E. et al. (2012) Phosphorus-mobilization ecosystem engineering: the roles of cluster roots 408 and carboxylate exudation in young P-limited ecosystems. Annals of Botany 110 (2), 329-348. 409 60. Johnson, J.F. et al. (1996) Root carbon dioxide fixation by phosphorus-deficient Lupinus albus 410 (contribution to organic acid exudation by proteoid roots). Plant physiology 112 (1), 19-30. 411 61. Biate, D.L. et al. (2015) Legume root exudates: their role in symbiotic interactions. In Plant 412 Microbes Symbiosis: Applied Facets (Arora, N.K. ed), pp. 259-271, Springer India. 413 62. Xing, X. et al. (2011) Mutualism breakdown in breadfruit domestication. Proceedings of the Royal 414 Society B: Biological Sciences 279 (1731), 1122-1130. 415 63. Zhu, Y.-G. et al. (2001) Phosphorus (P) efficiencies and mycorrhizal responsiveness of old and 416 modern wheat cultivars. Plant and Soil 237 (2), 249-255. 417 64. An, G.-H. et al. (2010) How does arbuscular mycorrhizal colonization vary with host plant 418 genotype? An example based on maize (Zea mays) germplasms. Plant and Soil 327 (1), 441-453. 419 65. Lehmann, A. et al. (2012) Mycorrhizal responsiveness trends in annual crop plants and their wild 420 relatives—a meta-analysis on studies from 1981 to 2010. Plant and Soil 355 (1), 231-250. 421 66. Hassan, S. and Mathesius, U. (2012) The role of flavonoids in root–rhizosphere signalling: 422 opportunities and challenges for improving plant–microbe interactions. Journal of Experimental Botany 423 63 (9), 3429-3444. 424 67. Mutch, L.A. and Young, J.P.W. (2004) Diversity and specificity of Rhizobium leguminosarum 425 biovar viciae on wild and cultivated legumes. Molecular Ecology 13 (8), 2435-2444. 426 68. Bourion, V. et al. (2018) Co-inoculation of a pea core-collection with diverse rhizobial strains shows 427 competitiveness for nodulation and efficiency of nitrogen fixation are distinct traits in the interaction. 428 Frontiers in Plant Science 8 (2249). 429 69. Kim, D.H. et al. (2014) Phylogenetic diversity of Mesorhizobium in chickpea. Journal of 430 Biosciences 39 (3), 513-517. 431 70. Pérez-Jaramillo, J.E. et al. (2016) Impact of plant domestication on rhizosphere microbiome 432 assembly and functions. Plant Molecular Biology 90 (6), 635-644. 433 71. Sasse, J. et al. (2018) Feed your friends: do plant exudates shape the root microbiome? Trends in 434 Plant Science 23 (1), 25-41. 435 72. Lareen, A. et al. (2016) Plant root-microbe communication in shaping root microbiomes. Plant 436 Molecular Biology 90 (6), 575-587. 437 73. Dennis, P.G. et al. (2010) Are root exudates more important than other sources of rhizodeposits in 438 structuring rhizosphere bacterial communities? FEMS microbiology ecology 72 (3), 313-327. 439 74. Bulgarelli, D. et al. (2015) Structure and function of the bacterial root microbiota in wild and 440 domesticated Barley. Cell Host & Microbe 17 (3), 392-403. 441 75. Szoboszlay, M. et al. (2015) Comparison of root system architecture and rhizosphere microbial 442 communities of Balsas teosinte and domesticated corn cultivars. Soil Biology and Biochemistry 80, 34- 443 44. 444 76. Coleman-Derr, D. et al. (2016) Plant compartment and biogeography affect microbiome 445 composition in cultivated and native Agave species. New Phytologist 209 (2), 798-811. 446 77. Zachow, C. et al. (2014) Differences between the rhizosphere microbiome of Beta vulgaris ssp. 447 maritima—ancestor of all beet crops—and modern sugar beets. Frontiers in Microbiology 5 (415). 448 78. Cardinale, M. et al. (2015) Bacterial networks and co-occurrence relationships in the lettuce root 449 microbiota. Environmental Microbiology 17 (1), 239-252. 16 450 79. Jacoby, R.P. and Kopriva, S. (2018) Metabolic niches in the rhizosphere microbiome: new tools 451 and approaches to analyse metabolic mechanisms of plant–microbe nutrient exchange. 452 80. Shi, J. and Lai, J. (2015) Patterns of genomic changes with crop domestication and breeding. Current 453 Opinion in Plant Biology 24, 47-53. 454 81. Hajjar, R. and Hodgkin, T. (2007) The use of wild relatives in crop improvement: a survey of 455 developments over the last 20 years. Euphytica 156 (1), 1-13. 456 82. Cheng, F. and Cheng, Z. (2015) Research progress on the use of plant allelopathy in agriculture and 457 the physiological and ecological mechanisms of allelopathy. Frontiers in plant science 6, 1020. 458 83. Oburger, E. and Jones, D.L. (2018) Sampling root exudates–mission impossible? Rhizosphere. 459 84. Ficiciyan, A. et al. (2018) More than yield: Ecosystem services of traditional versus modern crop 460 varieties revisited. Sustainability 10 (8), 2834. 461 17