Host–parasite coevolution can maintain high levels of genetic diversity in traits involved in species interactions. In many systems, host traits exploited by parasites are constrained by use in other functions, leading to complex selective pressures across space and time. Here, we study genome-wide variation in the staple crop Sorghum bicolor (L.) Moench and its association with the parasitic weed Striga hermonthica (Delile) Benth., a major constraint to food security in Africa. We hypothesize that geographic selection mosaics across gradients of parasite occurrence maintain genetic diversity in sorghum landrace resistance. Suggesting a role in local adaptation to parasite pressure, multiple independent loss-of-function alleles at sorghum LOW GERMINATION STIMULANT 1 (LGS1) are broadly distributed among African landraces and geographically associated with S. hermonthica occurrence. However, low frequency of these alleles within S. hermonthica-prone regions and their absence elsewhere implicate potential trade-offs restricting their fixation. LGS1 is thought to cause resistance by changing stereochemistry of strigolactones, hormones that control plant architecture and below-ground signaling to mycorrhizae and are required to stimulate parasite germination. Consistent with trade-offs, we find signatures of balancing selection surrounding LGS1 and other candidates from analysis of genome-wide associations with parasite distribution. Experiments with CRISPR–Cas9-edited sorghum further indicate that the benefit of LGS1-mediated resistance strongly depends on parasite genotype and abiotic environment and comes at the cost of reduced photosystem gene expression. Our study demonstrates long-term maintenance of diversity in host resistance genes across smallholder agroecosystems, providing a valuable comparison to both industrial farming systems and natural communities.

24Host-parasite coevolution can maintain high levels of genetic diversity in traits involved 25 in species interactions. In many systems, host traits exploited by parasites are 26 constrained by use in other functions, leading to complex selective pressures across 27 space and time. Here, we study genome-wide variation in the staple crop Sorghum 28 bicolor (L.) Moench and its association with the parasitic weed Striga hermonthica 29 (Delile) Benth., a major constraint to food security in many African countries. We 30 hypothesize that sorghum landraces are subject to geographic selection mosaics within 31 parasite-prone areas and selection against resistance where S. hermonthica is never 32 found. Supporting this hypothesis, multiple independent loss-of-function alleles at 33 sorghum LOW GERMINATION STIMULANT 1 (LGS1), a locus known to impact 34 resistance, are broadly distributed among African landraces and geographically 35 associated with S. hermonthica occurrence, suggesting a role in local adaptation to 36 parasite pressure. However, the low frequency of these alleles within S. hermonthica-37 prone regions and their absence elsewhere indicates potential trade-offs restricting their 38 distribution. LGS1 impacts stereochemistry of strigolactones, hormones controlling 39 plant architecture, belowground signaling with other organisms, and abiotic stress 40 tolerance. Supporting trade-offs, transcriptome profiling of nutrient-stressed roots 41 revealed differential regulation of several strigolactone biosynthesis and signaling genes 42 in LGS1-deficient sorghum compared to a susceptible line. Signatures of balancing 43 selection surrounding LGS1 and candidates from analysis of genome-wide associations 44 with parasite distribution support long-term maintenance of diversity in parasite 45 resistance genes. Our study of host resistance evolution across smallholder 46 agroecosystems provides a valuable contrast to both industrial farming systems and 47 natural communities. 48 49 KEYWORDS: species distribution modeling, environmental niche modeling, genotype-50 environment association analysis, Red Queen 51 52 SIGNIFICANCE STATEMENT: 53 3 Understanding co-evolution in crop-parasite systems is critical to management of 54 myriad pests and pathogens confronting modern agriculture. In contrast to wild plant 55 communities, parasites in agricultural ecosystems are usually expected to gain the 56 upper hand in co-evolutionary 'arms races' due to limited genetic diversity of host crops 57 in cultivation. Here, we develop a framework for studying associations between genome 58 diversity in global landraces (traditional varieties) of the staple crop sorghum with the 59 distribution of the devastating parasitic weed Striga hermonthica. We find long-term 60 maintenance of diversity in genes related to parasite resistance, highlighting an 61 important role of host adaptation for co-evolutionary dynamics in smallholder 62 agroecosystems. 63 64 INTRODUCTION:

Populations often exhibit genetic diversity in traits involved in responses to abiotic stressors, but what maintains this diversity is unclear. Arabidopsis thaliana exhibits high within-population variation in drought response. One hypothesis is that competition, varying at small scales, promotes diversity in resource use strategies. However, little is known about natural variation in competition effects on Arabidopsis physiology. We imposed drought and competition treatments on diverse genotypes. We measured resource economics traits, physiology, and fitness to characterize plasticity and selection in response to treatments. Plastic responses to competition differed depending on moisture availability. We observed genotype-drought-competition interactions for relative fitness: competition had little effect on relative fitness under well-watered conditions, whereas competition caused rank changes in fitness under drought. Early flowering was always selected. Higher d 13 C was selected only in the harshest treatment (drought and competition). Competitive context significantly changed the direction of selection on aboveground biomass and inflorescence height in wellwatered environments. Our results highlight how local biotic conditions modify abiotic selection, in some cases promoting diversity in abiotic stress response. The ability of populations to adapt to environmental change may thus depend on small-scale biotic heterogeneity.

Biodiversity can affect the properties of groups of organisms, such as ecosystem function and the persistence of colonizing populations. Genomic data offer a newly available window to diversity, complementary to other measures like taxonomic or phenotypic diversity. We tested whether native genetic diversity in field experimental stands of Arabidopsis thaliana affected their aboveground biomass and fecundity in their colonized range. We constructed some stands of genotypes that we a priori predicted would differ in performance or show overyielding. We found no relationship between genetic diversity and stand total biomass. However, increasing stand genetic diversity increased fecundity in high-resource conditions. Polyculture (multiple genotype) stands consistently yielded less biomass than expected based on the yields of component genotypes in monoculture. This under-yielding was strongest in stands with late-flowering and high biomass genotypes, potentially due to interference competition by these genotypes. Using a new implementation of association mapping, we identified genetic loci whose diversity was associated with stand-level yield, revealing a major flowering time locus associated with under-yielding of polycultures. Our field experiment supports community ecology studies that find a range of diversity-function relationships. Nevertheless, our results suggest diversity in colonizing propagule pools can enhance population fitness. Furthermore, interference competition among genotypes differing in flowering time might limit the advantages of polyculture.

Arabidopsis thaliana has a wide elevational range and much of its diversity may be associated with local adaptation to elevation. We took a multi-regional view of the genomics and physiology of elevational adaptation in Arabidopsis, with >200 ecotypes, including 17 newly collected from Africa. We measured plant responses to potential high elevation stressors: low pCO2, high light, and night freezing and conducted genome-wide association studies (GWAS). We found evidence of an adaptive cline in the western Mediterranean with low δ13C/early flowering at low elevations to high δ13C/late flowering at high elevations. By contrast, central Asian high elevation ecotypes flowered earlier. Antioxidants and pigmentation under high light and freezing showed regional differentiation but not elevational clines and may be associated with maladaptive plasticity. We found natural variation in non-photochemical quenching (NPQ) kinetics in response to chilling and fluctuating light, though with an unclear role in local adaptation. There were several candidate genetic loci mapped, including the ascorbate transporter PHT4;4 (associated with antioxidants) that influences the xanthophyll cycle, and may be involved in local adaptation to Morocco. Our study shows how the ecological strategies and genetic loci causing local adaptation to elevation change across regions and contribute to diversity in Arabidopsis.