A chromosome-scale genome assembly of Hordeum erectifolium: genomic, transcriptomic and anatomical adaptations to drought in a wild barley relative

58 To ensure food security in the face of a changing climate, it is essential to develop and implement 59 innovative strategies that enhance crop resilience to extreme weather events, diversify food sources, 60 and optimize resource use. Crop wild relatives (CWR), the close wild relatives of domesticated crop 61 species, harbour novel traits and alleles that serve as a valuable resource for breeding stress-resilient 62 crops and for advancing research into the genetic and genomic basis of plant adaptation (Feuillet et 63 al., 2008; Bohra et al., 2022). 64 65 CWRs are categorized into three gene pools. The primary gene pool includes the cultivated crop and 66 its closest wild relatives that can cross easily and produce fertile offspring. The secondary relatives 67 can be crossed with som e difficulty and reduced fertility, while tertiary relatives require advanced 68

like genetic engineering or wide hybridization to overcome strong reproductive barriers 69 (Harlan and de Wet, 1971; Baker et al., 2020; Kashyap et al., 2022). For over a ce ntury, CWR have 70 been used for crop improvements, in particular as a resource for disease and pest resistance genes, 71 but also to improve resistance to abiotic stresses (Hajjar and Hodgkin, 2007; Renzi et al., 2022). The 72 introgression of stress escape and avoidance strategies into elite germplasm has often been achieved 73 by introducing exotic alleles altering phenology, the timing of germination and reproductive 74 development (Duc et al. 2015; Farooq et al., 2025 ). More recently, CWRs have been used to introduce 75 perennial growth, a complex trait that extends survival and allows for reproduction over many 76 seasons, into annual crops (Gruner and Miedaner, 2021; Zhang et al., 2023). Perennial crops promise 77 to reduce labor costs, decrease input requirements, reduce soil erosion and enhance soil health, 78 thereby supporting more sustainable agriculture (Zhang et al., 2023). However, the introgression of 79 novel traits such as perenniality , disease and abiotic stress resistance from CWR is hampered by 80 hybridization barriers, low fertility, and time and labor costs for generating advanced lines (Wendler 81 et al., 2015). The availability of reference genomes and genetic tools for gene identification and 82 transfer now promises to deliver novel, efficient approaches for using CWR in research and breeding 83 (Brozynska et al., 2016; Gao et al., 2024). 84 85 The annual crop barley ( Hordeum vulgare ssp. vulgare) was domesticated from the wild barley 86 ancestor H. vulgare ssp. spontaneum in the Fertile Crescent (Badr et al. , 2000, Pankin et al. 2018 ). 87 Wild and cultivated barley belong to the primary gene pool and are characterised by natural 88 continuous gene flow in the Fertile Crescent, where wild and cultivated barley co -occur (Pankin et 89 al. 2018). Wild barley has been succ essfully used to introgress novel alleles for disease resistance, 90 abiotic stress tolerance and flowering time into elite barley (Matus et al. 2003; von Korff et al. 2004; 91 Hernandez et al. 2020). In addition, H. bulbosum from the secondary CWR gene pool was a source 92 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 4 of resistance genes to powdery mildew, leaf rust, barley mild and yellow mosaic virus for breeding 93 elite barley (Walther et al., 2000, Vincent et al., 2012). However, CWRs from the tertiary gene pool 94 within the Hordeum genus have not yet been used for barley improvement, not only because of 95 hybridization barriers but also because of a lack of genetic and genomic resources. The Hordeum 96 genus, comprising about 33 annual and perennial species, originated in the Mediterranean some 12 97 million years ago (Mya) and has since migrated globally, with 16 species in South America, which 98 have evolved in the last 1.5 Mya (Blattner, 2018). The South American clade of the Hordeum genus 99 contains closely related species adapted to different ecological niches and with different life history 100 strategies, making them an interesting resource for trait diversification and research into the 101 physiological and genetic underpinnings of adaptation (Bothmer and Komatsuda, 2010). Among the 102 South American clade, Hordeum erectifolium is one of the younger species , with the last common 103 ancestor of H. erectifolium and its closest sister species, H. stenostachys, dating back to ca. 1.3 Mya, 104 whereas the split of the branches separating H. erectifolium and H. vulgare occurred approximately 105 9.2 Mya (Brassac & Blattner, 2015). H. erectifolium is endemic to the southernmost part of the 106 semiarid Pampas, Argentina, where it grows among halophytes near a salt lake (Bothmer et al., 1985). 107 This region is impacted by frequent irregular periods of drought that can extend to up to a year 108 (Bothmer et al., 1985; Cai et al., 2020; Sgroi et al., 2021; WMO, 2020). Due to its small distribution 109 area, it is considered a near-threatened species (IUCN 2024) and is listed as a high priority for ex-situ 110 conservation within the Hordeum genus; only a single accession of H. erectifolium is available in 111 gene banks (Vincent et al., 2012). Being a perennial species, H. erectifolium has to persist through 112 long and unpredictable periods of drought and saline soil. It has thus evolved unique stress avoidance 113 traits, such as very glaucous erect basal leaves and pubescence on the leaf abaxial side, a thick wax 114 coating, and large suberized silica cells (Bothmer et al., 1985, 1995). These unique adaptations make 115 H. erectifolium an interesting genetic resource to explore the physiological and genetic underpinnings 116 of stress adaptation compared to the closely related cultivated crop barley. 117 118 We present a chromosome- scale, annotated reference genome of H. erectifolium , together with a 119 comprehensive tissue -specific gene expression atlas. By comparing genomic, transcriptomic, and 120 morphological responses to water limitation with those of cultivated barley, we highlight stress -121 adaptive strategies employed by H. erectifolium . Our findings underscore the potential of this 122 perennial close relative of barley as a model for dissecting the morphological, physiological, and 123 genetic bases of stress adaptation. 124 125 126 127 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 5

128 Here is a brief description of methods and material , a more details are found in the Online methods 129 and materials (Online methods). 130 131 Plant material and growth conditions 132 For all experiments we used single descent propagated seeds of H. erectifolium acc. NGB6816 133 (Nordic Genetic Resource Center, Sweden), and H. v. spontaneum acc. B1K-04-12 from the Barley1K 134 collection (Hübner et al., 2009), H. v. spontaneum acc. HID-4 (Liller et al., 2017), and H. vulgare cv. 135 Morex. H. erectifolium and Morex were used in all experiments, but B1K -01-12 was used for leaf 136 anatomical phenotyping, and HID-4 for specific leaf area (SLA) and elemental carbon and nitrogen 137 measurements. Unless specified otherwise, plants were consistently grow n under the following 138 conditions. Seeds were sown in a mixture of 93 % (v/v) Einheitserde ED73 ( Einheitserdewerke 139 Werkverband e.V ., Sinntal-Altengronau, Germany), 6.6 % (v/v) sand, and 0.4 % (v/v) Osmocote exact 140 standard 3-4M (Scotts Company LLC). Stratifi ed at 4 °C before being placed in a growth chamber 141 with long day conditions (16 h light, 8 h dark, at 20 °C day/16 °C night, 60 % relative humidity) for 142 germination. Ten days after germination, they were then vernalized for 8 weeks at 4 °C under short 143 day conditions (8 h light, 16 h dark, at 4 °C day/4 °C night), before being transferred back to the long 144 day conditions. Plants used for specific leaf area and elemental carbon and nitrogen measurements 145 were cultivated as described above. After vernalization, they were repotted to 7.5 L pots and grown 146 further in a common garden at the Botanical Garden of the Heinrich -Heine-Universität Düsseldorf 147 (HHU), data collected were from three summer seasons of 2021 – 2023. 148 149 Phenotyping 150 To visualize the leaf anatomy, we used toluidine blue to stain the lignin of transversely cut leaves. We 151 collected the flag leaf and the leaf below the flag leaf, and 1 cm sections at the midpoint of the leaf 152 were thinly sliced and fixed for staining with toluidine blue for microscope image capture. The 153 specific leaf area (SLA) was measured by scanning the area (cm2) of a main culm flag leaf with 154 Petiole (Petiole LTD, U.S.A.), an application on a mobile device. The dry weight (mg) was measured 155 after drying at 65 °C, and SLA was calculated as SLA = cm 2 / mg. We measured elemental carbon 156 and nitrogen by pooling the flag leaves from the first three reproductive tillers per plant at the grain 157 filling stage. They were dried at 65 °C before being pulverized to a fine powder, and approximately 158 2 mg were accurately measured . The samples were provided to the Metabolomics and Metabolism 159 Laboratory (CMML, He inrich-Heine University Düsseldorf , Germany) for measurements of 160 elemental carbon and nitrogen. 161 162 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 6 Generation of an annotated chromosome-scale genome assembly 163 A single plant of H. erectifolium acc. NGB6816 grown under control conditions , was used for 164 extracting high molecular weight (HMW) DNA. HMW DNA was extracted from young leaf material 165 and sequenced on 27 Oxford Nanopore Technology (ONT) flow cells ( Table S1), and with 10x 166 Genomics Linked-Reads sequencing for assembly polishing and the first scaffolding. Bionano optical 167 genome mapping was made with material of the same plant that also provided the seeds used to 168 generate seedlings for Hi -C chromosome conformation capture sequencing. We processed the 10x 169 Genomics Linked- Reads data with L ong Ranger (10x Genomics, U.S.A.), generating paired- end 170 Illumina (PE150) short- reads data with Linked- Reads information in headers (short -reads). We 171 performed initial genome kmer -based characterization (size, ploidy level, heterozygosity), by 172 analyzing the short-reads data using Jellyfish and findGSE (Marçais & Kingsford, 2011; Sun et al., 173 2018). 174 175 The ONT sequencing was basecalled with Guppy (ONT, United Kingdom) using a quality threshold 176 of ≥Q7 and subsequently trimmed sequencing adapters with Porechop (Wick et al., 2017), before 177 assembling the ONT long-reads with Flye (Kolmogorov et al., 2019). The assembly was progressively 178 polished; first, we used minimap2 to map the ONT long-reads to the assembly before each polishing 179 step, first with Racon and then with Medaka (ONT, United Kingdom) (Vaser et al., 2017; Li, 2018). 180 Thereafter, we mapped the short-reads with BWA-MEM2 to the long- read polished assembly and 181 performed two rounds of polis hing with Hapo-G and repeated the mapping between rounds 182 (Vasimuddin et al., 2019; Aury & Istace, 2021). The polished assembly was progressively scaffolded 183 from contigs to pseudomolecules. Initially, Tigmint was used with the 10x Genomics Linked-Reads 184 short-read data (Jackman et al., 2018), followed by hybrid-scaffolding using Bionano optical genome 185 mapping (Šimková et al., 2023). Finally, the assembly was organized into pseudomolecules using the 186 TRITEX pipeline with Hi-C chromosome conformation capture sequencing (Monat et al., 2019). 187 Assembly quality, completeness, and metrics were assessed at each step using complementary tools: 188 Merqury for k -mer–based evaluation of assembly accuracy and k-mer completeness , BUSCO for 189 gene space completeness, and QUAST for standard assembly statistics (Mikheenko et al., 2018; Rhie 190 et al., 2020; Manni et al., 2021). A unique genome assembly identifier, lpHorErec1.1, was registered 191 at ToLID – Tree of Life Identifiers, id.tol.sanger.ac.uk. 192 193 We conducted a comprehensive tissue -time specific transcriptome profiling of H. erectifolium . All 194 mature RNA plant tissue samples and seeds used for seedlings and germinating seed samples were 195 collected from the same plant as was previously harvested for H MW DNA. We selected twelve 196 tissues, ten of which were sampled at two time points during the day, in the morning (MOR, ZT 1-3) 197 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 7 and evening (EVE, ZT 13-16), a total of 22 samples. The vegetative tissues contained three -day-old 198 germinating seeds (GS3), whole roots (RO8, 8 days post germination), and whole shoots (LS8, 8 days 199 post germination). At anthesis, the third nodes (NOD), fourth internodes (INT), and flag leaves (FLF) 200 were collected. The six reproductive samples consisted of developing spikes at the spikelet induction 201 phase (ESP, W3.0-4.5), at floral development (MSP, W5.0- 6.5), and during rapid spike and floret 202 growth (LSP, W7.0-8.0), anthers (ANT) and ovules (OVU) were individually dissected at flowering 203 (W10) and caryopses (CAR, 10 days post anthesis) (Waddington et al., 1983). RNA was extracted, 204 and each tissue was barcoded and sequenced with PacBio IsoSeq (Pacific Biosciences, U.S.A). 205 206 For the gene annotation, we first mapped the 22 PacBio IsoSeq samples to the assembled genome and 207 processed them with the isoseq3 pipeline (Pacific Biosciences, U.S.A). We predicted the open reading 208 frames (ORF) for each isoform using TransDecoder (Haas, 2023), and additional ab initio structural 209 gene annotation was made with Helixer (Holst et al., 2023). The IsoSeq-Transdecoder and Helixer 210

were merged for a final gene annotation, and functionally annotated them with InterProScan 211 and Mercator4 (Jones et al., 2014; Bolger et al., 2021). We further predicted the presence of lncRNA 212 in the IsoSeq data. Transcript abundance in 22 tissue - and time-specific samples was quantified as 213 normalized transcripts per million (TPM) using IsoQuant (Prjibelskwe et al., 2023), and the samples 214 were clustered using Principal Component Analysis (PCA). We visualized the top 20 contributing 215 genes, separating the 22 tissues along the principal components 1 (PC1) and PC2. 216 217 For genomic and genetic comparisons between H. erectifolium acc. NGB6816 and barley, we used 218 the barley reference genome MorexV3, and the genome of a wild barley H. v. spontaneum acc. B1K-219 04-12 (Jayakodi et al., 2020; Mascher et al., 2021). We annotated transposable elements (TE) and 220 repetitive regions of the H. erectifolium and barley genomes with EDTA and further classified LTR 221 superfamilies with TEsorter (Ou et al., 2019; Zhang et al., 2022). The telomeric ends were identified 222 by aligning the telomeric sequence TTTAGGG 8 and centromeres using the CRM coding domains, 223 extracted with TEsorter from the EDTA annotation of LTR retrotransposon (Ou et al., 2019; Zhang et 224 al., 2022). The sequences were aligned with BLASTN to the three genomes (Camacho et al., 2009). 225 We identified the centromere midpoints from characteristic alignment density patterns using hdrcde 226 (Hyndman et al., 2023). We calculated the chromosomal synteny among the genomes by mapping 227 them against one another with minimap2 (Li, 2018), and structural variations were calculated with 228 Synteny and Rearrangement Identifier (SyRI) before visualization with plotsr (Goel et al., 2019; Goel 229 & Schneeberger, 2022). 230 231 232 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 8 Gene family evolution analysis with Orthofinder and CAFE5 233 A study of unique and shared genes and gene famil ies was made between H. erectifolium and eight 234 additional species, including the two barley genotypes. The proteomes of six species were retrieved 235 from the JGI Phytozome database : Arabidopsis thaliana, Sorghum bicolor, Zea mays, Oryza sativa, 236 Brachypodium distachyon, and Triticum aestivum cv. Chinese Spring (Goodstein et al., 2012). We 237 assigned the proteomes to Hierarchical Orthologous G roups (HOG) with Orthofinder, and gene 238 family evolution analyzed with CAFE5 (Emms & Kelly, 2019; Mendes et al., 2021), with a species 239 separation time of 59 million years retrieved from TimeTree5 ( Kumar et al., 2022). 240 241 Cross-species drydown experimental setup and differential gene expression analysis 242 We set up a drydown and recovery experiment to study the cross -species transcriptomic and 243 physiological responses to a reduction in relative leaf water content . One seed was sown per 7×7×8 244 cm pot with 150±1 g of soil, and plants were germinated and further cultivated under 12 h light/ 12 h 245 dark and temperatures 20 °C day/16 °C night, 60 % relative humidity. H. erectifolium plants were 246 grown for eight weeks, and Morex plants for two weeks to ensure comparable biomass at the start of 247 the treatment. Soil moisture was adjusted to 50 % field capacity (FC), control FC, and water was 248 withheld for six days before rewatering to control FC. Four replicate samples of the second leaves 249 from the main tillers were collected for transcriptomic analyses at four days of treatment (DOT) at 250 ZT 8, on DOT 2, 5, 6, and 7, 24 hours after re -watering. Leaf relative water content (RWC) was 251 measured at each DOT . Total RNA was extracted , and the transcriptome was Illumina PE150 252 sequenced. 253 254 The RNAseq reads were mapped to their respective genome with STAR, and quantification was done 255 using featureCounts (Dobin et al., 2013; Liao et al., 2014). The quantified reads were processed and 256 normalized with edgeR, and differentially expressed genes (DEG) analyzed at individual timepoints 257 with edgeR. maSigPro was used for time-course and co-expression analysis of DEGs in response to 258 treatment over time ( Nueda et al., 2014; Chen et al., 2024). Gene Ontology ( GO) functional 259 enrichment was made with Clusterprofiler (Xu et al., 2024). Single copy orthologs (SCO) between 260 H. erectifolium and Morex were retrieved from HOGs and were used for direct DEG comparison 261 between species. 262 263 Data analysis 264 Data wrangling, statistics, and visualization was done in R (v. 4.4.3) (R Core Team, 2024) were made 265 with tidyverse (v 2.0.0), and statistics with agricole (v. 1.5) in (Wickham et al., 2019; Mendiburu, 266 2023). 267 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 9

268 Leaf morphology of H. erectifolium, cultivated and wild barley 269 H. erectifolium leaves were erect and had a glaucous appearance (Fig. 1a,b). We observed that upon 270 reduced water availability, H. erectifolium rolled its leaves , whereas wild and cultivated barley 271 maintained flat leaves (Fig. 1b,c). Leaf rolling is typically supported by specific leaf anatomies, such 272 as reduced or a lack of bulliform cells that provide structural support (Redmann, 1985). We therefore 273 compared the leaf anatomy of the flag leaf and the first leaf below the flag leaf of H. erectifolium, 274 cultivated ( Morex), and wild barle y. Here we present the results for the flag leaf , which were 275 confirmed in the first leaf below (Fig. S1). The flag leaf was characterised by a prominent ribbed leaf 276 structure with short and long trichomes on the adaxial side , with no discernible bulliform cells. We 277 also observed that as the flag leaf of H. erectifolium develops and extends, the abaxial side generally 278 faces upwards. The white appearance of the flag leaf indicated a layer of cuticular wax (Fig. 1a,b). 279 In contrast, cultivated and wild barley both had a flat leaf surface, much shorter trichomes, and 280 pronounced bulliform cells (Fig. 1d). We quantified the traverse vein density, the minor: major vein 281 ratio, leaf thickness and width, as well as specific leaf area and elemental carbon: nitrogen ratio . 282 Major veins are distinguished from minor veins by their additional bundle sheath extensions (BSE) 283 (Perico et al., 2022). The average number of veins in H. erectifolium per mm was 6.2 vs 3.4 and 3.9 284 veins/mm, in H. erectifolium versus cultivated and wild barley genotypes, respectively (Fig. 1e, S1). 285 The minor: major vein ratio had significantly shifted in H. erectifolium compared to cultivated and 286 wild barley, from 1.3 vs. 3.0 and 3.1, respectively. While cultivated and wild barley had three minor 287 veins flanked by major veins, the central minor vein had become a major vein in H. erectifolium (Fig. 288 1d,f, S1). We noted that H. erectifolium had overall more extensive sclerenchyma cells within their 289 BSE on the adaxial side of the veins compared to barley. Sclerenchyma cells are connected with 290 increased photosynthetic activity, water transport, and more rapid stomatal closure in addition to 291 structural support (Buckley et al., 2011). We found that Specific Leaf Area (SLA) was significantly 292 lower and carbon: nitrogen (C:N) ratios were significantly higher in H. erectifolium than in cultivated 293 and wild barley; SLA with 1.5 mm 2/mg vs 2.4 mm/mg and 2.2 mm/mg, and C:N ratio of 13 vs 7.0 294 and 7.8, in H. erectifolium and cultivated and wild barley, respectively (Fig. 1g,h). Low SLA and high 295 C:N ratio indicate a slow investment in leaf growth and the photosynthetic apparatus, and thus a 296 conservative growth strategy with improved resource use efficiency ( Poorter et al., 2009; On oda et 297 al., 2017). Leaf thickness and width were overall not different between species ; however, the leaf 298 width of Morex was significantly greater than that of wild barley and H. erectifolium (Fig. 1i,j, S1). 299 In short, H. erectifolium e xhibited significantly higher leaf venation and a greater number of major 300 veins—likely due to the conversion of minor veins into major ones —as well as a higher abundance 301 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 10 of sclerenchyma cells on both the abaxial and adaxial sides of the vascular bundles, reflecting greater 302 photosynthetic tissue compartmentalization and water distribution efficiency. 303 304 A complete, annotated H. erectifolium reference genome 305 Motivated by the unique leaf morphology and drought response of H . erectifolium, we investigated 306 its genomic and transcriptomic differences relative to the cultivated barley cultivar Morex . H. 307 erectifolium, like barley, has a large diploid (2n=14) genome, reported with a haploid size of 2n = 308 9.49 ±0.05 pg, ~4.6 Gbp by flow cytometry, while H. vulgare is 5.04 Gbp (Jakob et al., 2004; Doležel 309 et al. 2018). 310 311 Using short-read sequencing, 10x Genomics Linked-Reads, we characterized the haploid genome of 312 H. erectifolium via k-mer analysis, estimating a genome size of 4.4 Gbp, a repeat content of 75 %, 313 and no detectable heterozygosity (Fig. 2a). For a high-quality chromosome scale genome assembly, 314 we used a combination of ONT long (321 Gb, 72× coverage, read N50 of 42 kb), and Illumina short 315 (244.6 Gb, 55.6× coverage) read data, Bionano optical genome mapping (546 optical genome maps, 316 map length of 3.8 Gbp, N50 of 20.1 Mbp) for hybrid scaffolding and Hi-C chromosomal conformation 317 capture sequencing for pseudomolecule construction (72.2 Gb, 16.4× coverage) (Table S1 -S4). The 318 final assembly size was 3.94 Gbp with 3.85 Gbp (97.7 %) being anchored in seven pseudomolecules 319 and 89 Mbp of unassigned contigs. (Fig. 2b,c) (Table S5). The genome had a k-mer completeness and 320 QV estimate score of 94 % and QV 34.6 and a BUSCO estimate of gene space completeness of 98.2 % 321 (Single: 93.4 %, Duplicate:4.8 %, Poales database, n = 4,896) (Table S5) (Rhie et al., 2020; Manni et 322 al., 2021). 323 324 Taken together, we assembled a high-quality chromosomal-scale reference genome of H. erectifolium 325 using ONT long reads and polished with Illumina short reads, attaining a consensus score of QV 34.6. 326 The final genome consisted of 3.85 Gbp assigned to 7 pseudomolecules. 327 328 Tissue-time specific long read transcriptome sequencing and gene annotation 329 Transcriptome profiling of 22 samples from 12 different tissues harvested at two diurnal timepoints, 330 morning (MOR) and evening (EVE) yielded 3,645,621 full-length and non-chimeric reads used for 331 annotation and analyses (Table S6, S7). Principal component analysis (PCA) grouped the MOR and 332 EVE tissue pairs closely together , vegetative and reproductive samples differentiated along PC1 333 (35 %) with roots ( RO8) and nodes ( NOD) at the center . High expression of PROLIFIN and 334 EXP ANSIN genes separated anthers ( ANT) from other reproductive samples (Fig. 2d). When the 335 caryopsis (CAR) was included, it accounted for 49 % of the PC1 variance, primarily due to high 336 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 11 expression of multiple GLIADIN genes, which encode the major storage proteins in cereal grains (Fig. 337 S2) (Anderson et al., 2012). The trajectory of developing shoot apical meristem (ESP, MSP, LSP) and 338 ovules (OVU) were most influenced by histones ( H2A and H3/CENP-A) and a non-specific lipid-339 transfer protein (nsLTP). Photosynthesis-related genes, particularly CAB (Chlorophyll a-b binding 340 protein) and RuBisCo (Ribulose-1,5- bisphosphate carboxylase/oxygenase), were the main drivers 341 distinguishing above-ground vegetative samples, including whole shoots (LS8) and flag leaves (FLF; 342 Fig. 2d, S2). We combined two structural gene annotation methods , where IsoSeq yielded 29,099 343 protein-coding genes (with 5’ and 3’ UTRs) and 8,993 putative lncRNAs , Helixer predicted 55,287 344 protein-coding genes, of which 20,875 fully and 1,029 partially overlapped with IsoSeq. W e 345 annotated a total of 71,475 genes, thereof 8,993 lncRNAs, and 157,682 isoforms. IsoSeq sequencing 346 captured 98,238 unique protein-coding isoforms, with 3.4 isoforms per gene, 5.9 and 4.5 exons per 347 mRNA and CDS, respectively. Mean gene, mRNA and CDS lengths were 5,835 bp, 3,808 bp and 826 348 bp, respectively, while lncRNA genes were considerably shorter at 1,857 bp, and only 1.8 isoforms 349 and two exons per gene (Table S8). 350 351 In summary, with IsoSeq mRNA sequencing in 22 tissue samples, we could generate high- accuracy 352 gene predictions with UTR s and accurate isoform information, as well as lncRNA predictions. 353 Complemented with ab initio gene prediction, we identified and annotated a final set of 71,475 genes 354 (thereof 8,993 lncRNAs) and 157,682 isoforms. Furthermore, we provided a comprehensive tissue 355 and time of the day specific gene expression atlas in H. erectifolium. 356 357 Comparative genome analysis between H. erectifolium and H. vulgare 358 We further characterized the genome of H. erectifolium in relation to the reference genomes of the 359 spring barley cultivar Morex and the wild barley accession B1K -04-12 (Jayakodi et al., 2020 ; 360 Mascher et al., 2021). We compared the chromosome organization and structure, repeat content and 361 identified telomeric sequences, centromere locations, and transposable element (TE) composition. 362 The assembly of telomeres and centromeres is one of the remaining challenges in gapless genome 363 assemblies, due to the presence of long stretches of satellite repeats (Navrátilová et al., 2022). Here 364 we report that the telomeric ends were only found on the short arms of five out of seven chromosomes 365 (missing on 4H and 5H) in H. erectifolium (Fig. 2c). No telomeric ends were found on long and short 366 arms in either of the two barley genomes. We found CRM (homolog of Cereba in barley) enrichments 367 in the pericentromeric regions spanning on average ~37 Mbp per chromosome in H. erectifolium, and 368 ~100 intact CRMs per chromosome, but only ~18 intact in barley (Fig. 2c, S3) (Presting et al., 1998; 369 Hudakova et al., 2001; Neumann et al., 2011). The CRM enrichment midpoints also colocalized with 370 the regions identified in the Hi-C matrices (Fig. 1b) (Navratilova et al., 2022). 371 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 12 372 Large-scale chromosomal synteny among H. erectifolium , Morex, and B1K -04-12 was overall 373 conserved, although numerous small - to medium- scale structural rearrangements were detected . 374 However, two notable large pericentric inversions were found, a near full chromosomal inversion on 375 3H that spanned 471 Mbp (81.4 %) in H. erectifolium and 570 Mbp (91.7 %) in Morex, and on 4H a 376 pericentric inversion of 56 Mbp (10.2 %) in H. erectifolium and 109 Mbp (17.9 %) in Morex (Fig. 377 2c). Neither inversion had previously been reported in barley (Mascher et al., 2024; Jayakodi et al., 378 2024). Between H. erectifolium and Morex, we found 59 inversions >1 Mbp, and 17 inversions >10 379 Mbp and thus more than between Morex and B1K -04-12, with 32 inversions > 1 Mbp and one 380 inversion >10 Mbp. 381 382 We performed de novo annotation of transposable elements (TEs) in the H. erectifolium genome and 383 re-annotated TEs in the barley genomes Morex and B1K-04-12. TE content was relatively consistent 384 across genomes, covering 85.7 % in H. erectifolium, 87.5 % in Morex, and 86.9 % in B1K-04-12. In 385 H. erectifolium, the majority of repeats were LTR retrotransposons (68.2 %), while non-LTR elements 386 included terminal inverted repeats ( TIRs, 11.3 %) and Helitrons (5.3 % ) (Table S9). Among LTRs, 387 Gypsy elements were most abundant (20.1 % ), followed by Copia (11.5 %), with 37.6 % classified 388 as mixed or unclassified LTRs . Notably, CACTA transposons were enriched in H. erectifolium 389 (7.2 %), but were less prevalent in Morex (1.3 % ) and B1K -04-12 (2.2 % ) (Table S9). Although 390 overall LTR content was comparable across genomes, the number of intact LTRs varied by clade (Fig. 391 3b). In H. erectifolium , Retand (39 %), Angela (23 % ), Tekay (15 %), and SIRE (8 %) dominated, 392 accounting for 85 % of all intact LTRs. In contrast, Morex and B1K-04-12 were primarily composed 393 of Angela (39 %, 35 %), Athila (29 %, 35 %), Tekay (10 %, 8 %), and SIRE (7 %, 9 %), contributing 394 85 % and 87 % of their LTR content, respectively. 395 396 Insertion times of intact LTRs were estimated using a nucleotide substitution rate of 1.3×10⁻⁸ to infer 397 their activation periods (Ma and Bennetzen, 2004; Ou and Jiang, 2018). We observed two distinct 398 Gypsy insertion peaks: one at ~0.5 Mya in H. erectifoliu m, and another at ~1.8 Mya in Morex and 399 B1K-04-12. Copia showed a concurrent insertion burst around 0.5 Mya in all three genomes during 400 the Chibanian, coinciding with intensified and asynchronous glacial cycles (Fig. 3a) (Sun et al., 401 2021). At this time, C opia insertions were predominantly driven by the Angela clade across all 402 genomes, while H. erectifolium also showed lineage -specific bursts of the Gypsy clades Tekay and 403 Retand (Fig. 3c). The older burst at ~1.8 Mya, unique to Morex and B1K -04-12, involved Athila 404 insertions, corresponding with the Gelasian- Calabrian transition and early Northern Hemisphere 405 glaciation (Gibbard et al., 2010; Cita et al., 2012). Recent Copia insertions were more frequent near 406 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 13 chromosomal ends in all genomes, though also dispersed across chromosomes . Additionally, in H. 407 erectifolium, Gypsy insertions exhibited a similar distribution pattern , recent (0.6 Mya) (Fig. S4). 409 410 In summary, we identified numerous small - to medium-scale structural rearrangements between H. 411 erectifolium and Morex, along with two large -scale pericentric inversions on chromosomes 3H and 412 4H. Despite similar TE amounts, each species displayed distinct LTR retrotransposon profiles, 413 indicating independent TE -driven genome evolution. Additionally, we detected two major LTR 414 insertion bursts across the Hordeum genus, occurring around 0.5 and 1.8 million years ago, coinciding 415 with significant geological and climatic events. 416 417 Genetic signatures of adaptive abiotic stress-related gene family expansions in H. erectifolium 418 We established hierarchical orthologous groups (HOGs) across the Poaceae family: Sorghum bicolor 419 Zea mays, Oryza sativa, Brachypodium distachyon, Triticum aestivum, and H. vulgare , along with 420 Arabidopsis thaliana as an outgroup, to explore unique or shared gene families, expansions, or 421 contractions. 422 423 We identified 35,722 HOGs, 9,680 of which were shared across the Poaceae and Arabidopsis and 424 3,501 exclusive to Poaceae, while 1,472 HOGs were unique to H. erectifolium (Fig. 4a). There were 425 7,182 genes in the 1,472 HOGs unique to H. erectifolium with Gene Ontology (GO) enrichments 426 related to biological functions such as methylation, ribosome biogenesis, cell redox homeostasis, and 427 signal peptide processing (Fig. S5a). Using a total of 18,869 HOGs with representative genes in at 428 least two Poaceae species excluding A. thaliana, we identified 327 HOGs significant expanded or 429 contracted in H. erectifolium, 288 expansions and 39 contractions (Fig. S6, Table S10, S11). The 430 3,968 genes from H. erectifolium belonging to the 327 HOGs were enriched for terms related to 431 anatomical structure development, cell surface receptor signaling, and root development (Fig. S5b ). 432 Of those, we explored three expanded HOGs, with putative functions in abiotic stress resistance that 433 had undergone expansions in H. erectifolium (Fig. S7, Table S12). 434 435 The EARLY LIGHT-INDUCIBLE PROTEIN (ELIP) gene family, which enhances desiccation toler-436 ance by mitigating photooxidative damage (VanBuren et al., 2019), is expanded in H. erectifolium 437 with 25 paralogs, compared to 20 in barley. The primary expansion occurred through tandem dupli-438 cation on chromosome 5H, with 18 copies in H. erectifolium and 12 in barley. Of the 25 ELIP genes, 439 20 showed tissue- and time-specific expression: eleven were exclusively expressed in vegetative tis-440 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 14 sue, predominantly in leaves in the morning (LS8, FLF) whereas nine were expressed both in vege-441 tative and reproductive tissue (Fig. 4b, Table S13). We also observed an expansion of LATE EMBRY-442 OGENESIS ABUNDANT (LEA) PROTEIN 6 -RELATED genes (group 4 LEA s), which protect pro-443 teins and membranes during desiccation (Candat et al., 2014). H. erectifolium carried eight paralogs 444 versus four in barley, with six showing expression in at least one tissue or time point (Fig. 4b, Table 445 S13). Similarly, the DEHYDRATION-RESPONSIVE ELEMENT-BINDING PROTEIN 1C (DREB1C, 446 N0.HOG0002366) family underwent tandem duplication on chromosome 7H in H . erectifolium, re-447 sulting in four paralogs compared to two in barley; however, only two were expressed in the IsoSeq 448 data, each in a single tissue (Fig. 4b, Table S13). 449 450 In summary, gene sets unique to H. erectifolium were enriched for functions including methylation, 451 ribosome biogenesis, cell redox homeostasis, and signal peptide processing. HOGs expanded in H. 452 erectifolium were enriched for terms related to anatomical structure development, cell surface 453 receptor signaling, and root development. Coupled with expansions of gene families associated with 454 abiotic stress adaptation, i.e., the desiccation -related ELIP, LEA, and DREB1C genes, these 455 enrichments indicate that H. erectifolium is genetically primed to tolerate environmental stress. 456 457 Cross-species transcriptional drought response and recovery 458 To explore differences and similarities in the transcriptional responses to drought between H. 459 erectifolium and barley cv. Morex , we set up a drydown drought and recovery experiment at the 460 vegetative stage. Using a controlled drydown assay, we ensured that the reduction in leaf relative 461 water content (RWC) was equal between plants of the two species in the control and water -limited 462 conditions. 463 464 During drydown, both leaf RWC and soil field capacity (FC) decreased at the same rate in both 465 species, with no significant differences between either species at each respective day -of-treatment 466 (DOT) timepoint (Fig. S8, S9). Control samples were kept at 50 % soil FC while water was withheld 467 for 6 days before rewatering, and samples were collected on days 2, 5, 6, and 7 of treatment. On the 468 last day of drydown, DOT6, the average leaf RWC and soil FC for H. erectifolium were at 42 % ± 9 469 and 11 % ± 2, and for Morex at 51 % ± 9 and 13 % ± 0.6, respectively. Strong phenotypic differences 470 at DOT5 and DOT6 were observed; leaves of H. erectifolium had rolled inward and remained erect, 471 whereas Morex leaves wilted and required structural support ( Fig. S8b,c). Both species recovered 472 leaf RWC to control levels and regained leaf structure on DOT7, 24 hours after re-watering. No visible 473 symptoms of senescence, i.e., no yellowing of leaves, were observed at DOT7 nor seven days after 474 re-watering, DOT13. 475 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 15 476 In total, 64 leaf samples from the two species, four time points and two treatments with four biological 477 replicates per sampling point, were RNA-sequenced with an average of 21 million reads per sample 478 (Table S14). The RNAseq data were mapped to their respective genomes with STAR and quantified 479 with featureCounts (Table S15) (Dobin et al., 2013; Liao et al., 2014). We employed two methods for 480 transcriptomic analysis: maSigPro for time-course analysis, taking into account day of sampling and 481 treatment variables (Nueda et al., 2014) , and edgeR for pairwise expression comparison between 482 control and drydown for each time point separately, to also identify transiently expressed genes (Chen 483 et al., 2025). 484 485 With both methods, we identified 15,036 and 13,119 differentially expressed genes (DEGs) in H. 486 erectifolium and Morex, respectively (Table S1 6-S18). The time -course analysis identified 7,941 487 DEGs in H. erectifolium and 7,382 DEGs in Morex (Fig. S10). In addition, we detected 7,095 DEGs 488 in H. erectifolium and 5,737 DEGs in Morex, which were expressed only at specific DOTs and not 489 identified in the time-course analysis. 490 491 To compare DEGs between the two species, we focused on their shared single-copy orthologs (SCO). 492 Since transient DEGs showed a low overlap between species, we only included DEGs identified by 493 the timecourse analysis. Between H. erectifolium and Morex, we identified 19,204 SCO, of which 494 8,451 were differentially expressed in at least one species, thereof 5,487 in H. erectifolium and 5,743 495 in Morex, with an overlap of 2,799 (33 % ) DEGs detected in both H. erectifolium and Morex (Fig. 496 5a, Table S19). 497 498 DEGs shared by both species were enriched for photosynthesis, translation (ribosome), amide 499 biosynthesis, carboxylic acid and tetrapyrrole metabolism, suggesting that both species adapt 500 metabolically and coordinate energy production, carbon and nitrogen as similation, and oxidative 501 stress management in response to water limitations (Fig. 5b). However, while both species 502 downregulated genes with functions in photosynthesis during water limitation, they exhibited 503 markedly different responses in genes related to biosynthesis and translation. Notably, differentially 504 expressed genes (DEGs) associated with biosynthesis (amide biosynthetic processes) and translation 505 (ribosome-related processes) were strongly downregulated in H. erectifolium , whereas they were 506 upregulated in Morex at DOT5 and DOT6 (Fig. 5c). In H. erectifolium, the downregulation of these 507 DEGs indicated a repression of biosynthesis and translational activity in response to drought. The 508 downregulation of amide biosynthesis, i.e., decreased transcript levels of glutamine and asparagine 509 synthase genes (Table S17, S18), likely reflect ed a metabolic shift towards energy conservation to 510 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 16 prioritize survival over growth. The concurrent upregulation of ubiquitination, macromolecule 511 catabolism, and vesicle-mediated transport specifically in H. erectifolium indicated mobilization of 512 proteins for recycling or stress adaptation and active degradation of damaged proteins and cellular 513 components. In contrast, Morex upregulated genes involved in amide biosynthesis and translation, 514 potentially as part of a protective response to counteract ROS -induced protein denaturation and to 515 replace damaged proteins to maintain cell functions . The transcriptional responses in Morex were 516 thus reminiscent of proteostasis responses, which commonly lead to massive transcriptional and 517 translational upregulation of protective proteins (Mittler et al., 2012). The Morex-specific enrichment 518 of DEGs with functions in transmembrane transport and chloride channel activities suggested that 519 Morex maintained ion and osmotic balance to support active metabolism under water limitations. 520 Morex thus invested in maintaining cellular functions and supporting future recovery rather than 521 shutting down. In line with this, at recovery (DOT7), transcripts related to biosynthesis and translation 522 were strongly induced in H. erectifolium but not strongly altered compared to DOT5 and 6 in Morex 523 (Fig. 5c). 524 525 Given the different transcriptional responses to water limitation between both genotypes, we further 526 explored the expression of the three expanded desiccation-tolerance gene families, encoding DREB, 527 ELIP, and LEA proteins, which function in cellular protection and stabilization , particularly during 528 dehydration or oxidative stress. Among the ELIP (N0.HOG0000208) genes, 13 of 25, and eight of 20 529 were differentially expressed during desiccation in H. erectifolium and Morex, respectively (Fig. 5d, 530 S11a). We observed a strong upregulation of the four DREB1C genes (N0.HOG0002366) in H. 531 erectifolium compared to only a mild upregulation of one DREB1C copy in Morex (Fig. 5f, S11b). 532 DREB transcription factors act as upstream regulators that activate LEA genes, which were also more 533 strongly expressed in H. erectifolium than in Morex ( Sakuma et al. , 2006). Of the LEA 534 (N0.HOG0001214) genes, five of the eight copies were upregulated in H. erectifolium, whereas only 535 three of four copies were upregulated in Morex in response to drought (Fig. 5e, S11c). One LEA 536 (lpHorErec1.1-7HG30647) had a high constit utive expression under control and drought in H. 537 erectifolium and was also expressed in nearly all vegetative tissues (Fig. 4b). 538 539 Taken together, the differential regulation of amide biosynthesis and translation in H. erectifolium and 540 Morex suggested contrasting metabolic responses to water limitations . Morex upregulated both 541 pathways, thereby maintaining active nitrogen metabolism and detoxification and proteostasis. In 542 contrast, H. erectifolium downregulated both pathways, indicative of a conservative, survival-focused 543 strategy, minimizing energy expenditure and growth processes to preserve cellular integrity during 544 drought. The expanded and highly inducible DREB, LEA, and ELIP families suggested that H. 545 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 17 erectifolium is genetically prime d for drought, enabling strong protective responses with low 546 metabolic cost. 547 548

549 The unique leaf morphology of H. erectifolium 550 H. erectifolium displayed distinct leaf morphological traits indicative of drought adaptation, including 551 dense trichome coverage, high vein density with a near-equal major-to-minor vein ratio, and a thick 552 cuticular wax layer, all features of xeromorphic leaf traits (Shields, 1950). The prominently ribbed 553 adaxial surface bore dense short trichomes and longer trichomes along major veins, features involved 554 in microclimate stabilization, dew capture, and enhanced structural rigidity ( Liakoura et al., 1997; 555 Galdon-Armero et al., 2018). A compact, uniform epidermal cell layer and substantial wax coating 556 likely strengthened the transpiration barrier, UV protection and prevented biotic intrusion (Kasapligil, 557 1961; Xue et al., 2017). Major veins were reinforced with extensive bundle sheath extensions (BSE), 558 sclerenchymatous tissue causing increased mesophyll compartmentalization which has been associ-559 ated with enhanced light penetration, photosynthetic efficiency, and water -use optimization through 560 more responsive BSE -connected stomatal regulat ion (Barbosa et al., 2019). H. erectifolium and 561 Morex both maintained comparable leaf RWC levels as soil moisture declined during drydown. How-562 ever, while Morex leaves wilted and required external support, H. erectifolium leaves rolled tightly 563 and remained erect (Fig. 1, S8). Leaf rolling under water deficit reduces exposed surface area, mini-564 mizing transpiration, photodamage, and heat stress (Kadioglu et al., 2012). Although bulliform cells 565 are typically associated with this response, they were largely absen t in H. erectifolium (Cal et al., 566 2019; Zhu et al., 2022). We propose that leaf rolling in this species is facilitated by the ribbed struc-567 ture, numerous major veins with extensive BSE and sclerenchyma, and a compact epidermal cell 568 architecture (Guo et al., 2024). The distinctive morphology of H. erectifolium was mirrored at the 569 genomic level by an enrichment of expanded gene families involved in anatomical structure devel-570 opment. 571 572 Triticeae genomes are typically large (~5 Gbp per genome or subgenome) and consist of up to 90 % 573 repetitive elements, primarily transposable elements (Middleton et al., 2013; Winterfeld et al., 2025). 574 While the total proportion of repetitive content was relatively stable across species (85.7–87.5 % ), 575 we observed notable differences in LTR clade composition and insertion histories. H. erectifolium 576 showed a five-fold enrichment of CACTA transposons compared to barley, elements that influence 577 genome restructuring and gene evolution (Lisch, 2013; Caton i et al., 2019; Liu et al., 2020). We 578 discovered two LTR retrotransposon insertion bursts at ~0.5 Mya and ~1.8 Mya, which coincide with 579 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 18 known geological events. The Chibanian stage (~0.5 Mya), characterized by intense, globally asyn-580 chronous glacial cycles, and the Calabrian stage (Early Pleistocene, ~1.8 Mya), marked by the onset 581 of Northern Hemisphere glaciation affecting Eurasian species (Cita et al., 2012; Sun et al., 2021). 582 There was a synchronous peak at ~0.5 Mya, where Angela (Copia) was active in both species, while 583 the Gypsy families Retand and Tekay were specific to H. erectifolium. The second unilateral peak at 584 ~1.8 Mya, which was observed only in barley, was exclusively associated with Athila (Gypsy) (Fig. 585 4). LTR insertion bursts 1.8 Mya were also detected in two other Mediterranean species, wheat and 586 Brachypodium distachyon (Wicker et al., 2018; Stritt et al., 2020), suggesting that Eurasian species 587 experienced more severe glaciation and biome turnover than South American species (Vuilleumier 588 1971; Connor 2009; Gibbard et al., 2010). Accordingly, phylogenetic and biogeographic studies sug-589 gested high extinction rates for the Eurasian Hordeum species during the Early Pleistocene, while 590 high speciation rates were maintained in the South American clade (Jacob and Blattner, 2006). 591 592 The higher copy number of desiccation-related genes and the distinct transcriptome response to water 593

indicated that H. erectifolium and Morex diverge not only in their morphological but also 594 in their molecular responses to stress. Gene family expansion through tandem duplication can enable 595 faster transcriptional activation and greater dosage effects, providing an adaptive advantage during 596 rapid declines in relative water content (Dassanayake et al., 2011; Wu et al., 2012). Large tandem 597 arrays of ELIP s, which stabiliz e and protect photosynthetic components during dehydration and 598 enhance thermal energy dissipation, have been repeatedly associated with independent origins of 599 desiccation tolerance across diverse plant lineages (VanBuren et al., 2019; Pardo et al. 2020; Marks 600 et al., 2024). Similarly, the observed expansion of DREB1C and LEA gene families in H. erectifolium 601 is consistent with a genetic architecture that enhances resilience across multiple stresses, including 602 drought, salinity, and cold (Hernández -Sánchez et al., 2022; Wang et al., 2022). Such expansions 603 likely genetically primed H. erectifolium for rapid adaptation to fluctuating water availability, offering 604 a selective advantage in environments where desiccation events are frequent and unpredictable. This 605 genetic priming for rapid stress response is reminiscent of mechanisms seen in resurrection plants, 606 where tandem gene duplications of protective proteins underpin efficient tolerance with minimal 607 metabolic burden (VanBuren et al., 2019; Pardo et al. 2020). 608 609 The drydown experiment corroborates this hypothesis, showing fast and strong activation of all three 610 gene families followed by rapid deactivation upon rehydration, a pattern indicative of efficient 611 regulatory control. Furthermore, the constitutive high expression of one LEA copy may have 612 conferred an immediate protective buffer at the onset of dehydration stress, minimizing cellular 613 damage during early water loss (Cheng et al., 2002; Liu et al., 2009). 614 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 19 615 The transcriptional responses of H. erectifolium and Morex to water limitation suggested contrasting 616 drought‐adaptation strategies: metabolic downregulation and survival prioritization in H. erectifolium 617 versus maintenance of metabolic activity and competitiveness in Morex (Skirycz et al., 2011; Claeys 618 & Inzé, 2013). Both genotypes showed enrichment of DEGs related to photosynthesis, translation, 619 and nitrogen metabolism, but their regulation differed markedly. In H. erectifolium, strong repression 620 of amide biosynthesis and ribosome‐related processes, including reduced expression of glutamine 621 and asparagine synthetases, indicates energy conservation through suppression of growth and nitro-622 gen assimilation (Diaz et al. , 2010, Nagy et al., 2013) . Concurrent induction of ubiquitination and 623 macromolecule catabolism suggests active recycling of proteins and degradation of damaged compo-624 nents to protect cellular integrity (Skirycz & Inzé, 2010; Eckardt et al., 2024). Furthermore, H. erec-625 tifolium showed unique activation of pathways linked to cellular localization, vesicle -mediated 626 transport, and oxidative stress mitigation, including ROS metabolism, which together indicate pre -627 emptive preparation for drought (Mazel et al., 2004; Jarzyniak & Jasiński, 2014; Noctor et al., 2014). 628 By contrast, Morex upregulated biosynthesis and translation, consistent with a strategy of sustaining 629 metabolic activity and proteostasis to replace damaged proteins and maintain cellular function (Mit-630 tler et al., 2012). The enrichment of transmembrane transport and chloride channel activity indicates 631 active osmotic regulation, supporting growth and recovery under moderate stress ( Nieves-Cordones 632 et al., 2019; Franco-Navarro et al., 2021). The absence of strong post -drought induction of biosyn-633 thetic and translational processes in Morex at recovery, compared to the rapid reactivation seen in H. 634 erectifolium, further highlights differences in resource allocation and stress recovery dynamics. 635 636 Overall, H. erectifolium appears genetically primed for survival under severe drought through gene 637 family expansion and rapid protective responses, efficient shutdown of growth- related metabolism, 638 and rapid recovery. Such a strategy is likely advantageous in habitats with frequent and severe 639 desiccation events, favoring survival over productivity. Morex, in contrast, invests in maintaining 640 metabolic activity during drought, which may favor productivity and recovery potential under mild 641 stress but is energetically costly under prolonged or severe water deficits (Skirycz et al., 2011). 642 643 Combining both strategies, stress avoidance through active metabolism with drought tolerance via 644 growth suppression, is a central challenge for breeding crops that can both survive severe drought 645 and maintain yield under moderate stress. The exploration of desiccation-tolerant crop relatives might 646 provide novel genetic insights and tools, such as stress -inducible protective pathways (e.g., DREB, 647 LEA, ELIP) and stress -responsive promoters to activate key protective genes while minimizing 648 growth penalties under transient or moderate stress (Nelson et al., 2007; Peleg et al., 2011). 649 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 20 650

651 We greatly thank Rebekka Schüller and Nina Döring for excellent technical assistance, and Dominik 652 Brilhaus for data management support. This work was supported by the by the DFG Research Infra-653 structure NGS_CC (West German Genome Center (WGGC), project 407493903) as part of the Next 654 Generation Sequencing Competence Network (project 423957469). Computational infrastructure and 655 support were provided by the Centre for Information and Media Technology at Heinrich Heine Uni-656 versity Düsseldorf. This work was funded by the European Research Council (ERC) under the Euro-657 pean Union's Horizon Europe research and innovation program me (PERLIFE, No. 101002085 to 658 MvK and TRANSFER, No. 949873 to MM), the Deutsche Forschungsgemeinschaft (DFG) under 659 Germany's Excellence Strategy —EXC -2048/1—Project ID: 390686111, and the Collaborative Re-660 search Centre/Transregio (TRR 341, Project ID: 456082119), and the GRK 2064: Water use effi-661 ciency and drought stress responses: From Arabidopsis to Barley, Project ID: 252965955. H. Šimková 662 and Z. Tulpová were supported from the project TANGENC, reg. no. 663 CZ.02.01.01/00/22_008/0004581 of the ERDF Programme Johannes Amos Comenius. 664 665 Competing interests 666 None declared. 667 668 Author contributions 669 EBH and MvK conceived and designed the project and experiments. EBH performed all plant growth, 670 DNA extractions, data processing, genome assembly, experiments, and analysis, with the following 671 exceptions: HŠ, HT, and ZT made the optical genome map and did the hybrid scaffolding. MA 672 provided the carbon-nitrogen elemental and Specific Leaf -Area measurements. TR performed RNA 673 extractions for sequencing. MM and JF helped with the chromosome -scale genome sequence 674 assembly. EBH wrote the manuscript with the help of MvK. 675 676 Data availability 677 The data that support the findings of this study will be made openly available in D ataPLANT and 678 new sequence data generated in the study will be deposited at the European Nucleotide Archive 679 (ENA). 680 681

682 Anderson OD, Dong L, Huo N, Gu YQ. 2012. A New Class of Wheat Gliadin Genes and Proteins. 683 PLOS ONE 7: e52139. 684 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 21 Aury J-M, Istace B. 2021. Hapo-G, haplotype-aware polishing of genome assemblies with accurate 685 reads. NAR Genomics and Bioinformatics 3: lqab034. 686 Badr A, M K, Sch R, Rabey HE, Effgen S, Ibrahim HH, Pozzi C, Rohde W, Salamini F . 2000. 687 On the Origin and Domestication History of Barley ( Hordeum vulgare). Molecular Biology and 688 Evolution 17: 499–510. 689 Baker L, Grewal S, Y ang C, Hubbart -Edwards S, Scholefield D, Ashling S, Burridge AJ, 690 Przewieslik-Allen AM, Wilkinson PA, King IP, et al. 2020. Exploiting the genome of Thinopyrum 691 elongatum to expand the gene pool of hexaploid wheat. Theoretical and Applied Genetics 133: 2213–692 2226. 693 Barbosa MAM, Chitwood DH, Azevedo AA, Araújo WL, Ribeiro DM, Peres LEP, Martins SCV , 694 Zsögön A . 2019. Bundle sheath extensions affect leaf structural and physiological plasticity in 695 response to irradiance. Plant, Cell & Environment 42: 1575–1589. 696 Blattner FR. 2018. Taxonomy of the Genus Hordeum and Barley (Hordeum vulgare). In: Stein N, 697 Muehlbauer GJ, eds. The Barley Genome. Cham: Springer International Publishing, 11–23. 698 Blum M, Chang H -Y, Chuguransky S, Grego T, Kandasaamy S, Mitchell A, Nuka G, Paysan -699 Lafosse T, Qureshi M, Raj S, et al. 2021. The InterPro protein families and domains database: 20 700 years on. Nucleic Acids Research 49: D344–D354. 701 Bohra A, Kilian B, Sivasankar S, Caccamo M, Mba C, McCouch SR, Varshney RK. 2022. Reap 702 the crop wild relatives for breeding future crops. Trends in Biotechnology 40: 412–431. 703 Bolger M, Schwacke R, Usadel B. 2021. MapMan Visualization of RNA-Seq Data Using Mercator4 704 Functional Annotations. In: Dobnik D, Gruden K, Ramšak Ž, Coll A, eds. Solanum tuberosum: 705

and Protocols. New York, NY: Springer US, 195–212. 706 Bothmer R von, Jacobsen N, Baden C, Jorgensen RB, Linde -Laursen I . 1995. An 707 ecogeographical study of the genus Hordeum (2nd edition). 708 Bothmer R von, Jacobsen N, Jørgensen RB . 1985. Two New American Species of Hordeum 709 (Poaceae). Willdenowia 15: 85–90. 710 Bothmer R von, Komatsuda T. 2010. Barley Origin and Related Species. In: Ullrich SE, ed. Barley. 711 Wiley, 14–62. 712 Brassac J, Blattner FR . 2015. Species-Level Phylogeny and Polyploid Relationships in Hordeum 713 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 22 (Poaceae) Inferred by Next-Generation Sequencing and In Silico Cloning of Multiple Nuclear Loci. 714 Systematic Biology 64: 792–808. 715 Brozynska M, Furtado A, Henry RJ. 2016. Genomics of crop wild relatives: expanding the gene 716 pool for crop improvement. Plant Biotechnology Journal 14: 1070–1085. 717 Buchfink B, Reuter K, Drost H -G. 2021. Sensitive protein alignments at tree -of-life scale using 718 DIAMOND. Nature Methods 18: 366–368. 719 Buckley TN, Sack L, Gilbert ME . 2011. The Role of Bundle Sheath Extensions and Life Form in 720 Stomatal Responses to Leaf Water Status. Plant Physiology 156: 962–973. 721 Cai W, McPhaden MJ, Grimm AM, Rodrigues RR, Taschetto AS, Garreaud RD, Dewitte B, 722 Poveda G, Ham Y-G, Santoso A, et al. 2020. Climate impacts of the El Niño–Southern Oscillation 723 on South America. Nature Reviews Earth & Environment 1: 215–231. 724 Cal AJ, Sanciangco M, Rebolledo MC, Luquet D, Torres RO, McNally KL, Henry A. 2019. Leaf 725 morphology, rather than plant water status, underlies genetic variation of rice leaf rolling under 726 drought. Plant, Cell & Environment 42: 1532–1544. 727 Camacho C, Coulouris G, Avagyan V , Ma N, Papadopoulos J, Bealer K, Madden TL . 2009. 728 BLAST+: architecture and applications. BMC Bioinformatics 10: 1–9. 729 Candat A, Paszkiewicz G, Neveu M, Gautier R, Logan DC, Avelange-Macherel M-H, Macherel 730 D. 2014. The Ubiquitous Distribution of Late Embryogenesis Abundant Proteins across Cell 731 Compartments in Arabidopsis Offers Tailored Protection against Abiotic Stress. The Plant Cell 26: 732 3148–3166. 733 Catoni M, Jonesman T, Cerruti E, Paszkowski J. 2019. Mobilization of Pack-CACTA transposons 734 in Arabidopsis suggests the mechanism of gene shuffling. Nucleic Acids Research 47: 1311–1320. 735 Chen Y, Chen L, Lun ATL, Baldoni PL, Smyth GK. 2025. edgeR v4: powerful differential analysis 736 of sequencing data with expanded functionality and improved support for small counts and larger 737 datasets. Nucleic Acids Research 53: gkaf018. 738 Cheng Z, Targolli J, Huang X, Wu R . 2002. Wheat LEA genes, PMA80 and PMA1959, enhance 739 dehydration tolerance of transgenic rice (Oryza sativa L.). Molecular Breeding 10: 71–82. 740 Cita MB, Gibbard PL, Head MJ, Alloway B, Beu AG, Coltorti M, Gibbard PL, Hall VM, Head 741 MJ, Jiaqi L, et al. 2012. Formal Ratification of the GSSP for the Base of the Calabrian Stage (Second 742 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 23 Stage of the Pleistocene Series, Quaternary System). Episodes 35: 388–397. 743 Claeys H, Inzé D . 2013. The Agony of Choice: How Plants Balance Growth and Survival under 744 Water-Limiting Conditions. Plant Physiology 162: 1768–1779. 745 Connor SE . 2009. Human impact – the last nail in the coffin for ancient plants? Journal of 746 Biogeography 36: 485–486. 747 Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V , Pollard MO, Whitwham A, Keane T, 748 McCarthy SA, Davies RM, et al. 2021. Twelve years of SAMtools and BCFtools. GigaScience 10: 749 giab008. 750 Dassanayake M, Oh D, Hong H, Bohnert HJ, Cheeseman JM . 2011. Transcription strength and 751 halophytic lifestyle. Trends in Plant Science 16: 1–3. 752 Díaz P, Betti M, Sánchez DH, Udvardi MK, Monza J, Márquez AJ. 2010. Deficiency in plastidic 753 glutamine synthetase alters proline metabolism and transcriptomic response in Lotus japonicus under 754 drought stress. New Phytologist 188: 1001–1013. 755 Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras 756 TR. 2013. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29: 15–21. 757 Doležel J, Čížková J, Šimková H, Bartoš J. 2018. One Major Challenge of Sequencing Large Plant 758 Genomes Is to Know How Big They Really Are. International Journal of Molecular Sciences 19: 759 3554. 760 Duc G, Agrama H, Bao S, Berger J, Bourion V , De Ron Antonio M., Gowda Cholenahalli L. L., 761 Mikic Ale ksandar, Millot Dominique, Singh Karam B., et al. 2015. Breeding Annual Grain 762 Legumes for Sustainable Agriculture: New Methods to Approach Complex Traits and Target New 763 Cultivar Ideotypes. Critical Reviews in Plant Sciences 34: 381–411. 764 Eckardt NA, Avin-Wittenberg T, Bassham DC, Chen P, Chen Q, Fang J, Genschik P, Ghifari 765 AS, Guercio AM, Gibbs DJ, et al. 2024. The lowdown on breakdown: Open questions in plant 766 proteolysis. The Plant Cell 36: 2931–2975. 767 Emms DM, Kelly S . 2019. OrthoFinder: phylogenetic orthology inference for comparative 768 genomics. Genome Biology 20: 1–14. 769 Farooq M, Frei M, Zeibig F, Pantha S, Özkan H, Kilian B, Siddique KHM. 2025. Back into the 770 Wild: Harnessing the Power of Wheat Wild Relatives for Future Crop and Food Security. Journal of 771 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 24 Experimental Botany: eraf141. 772 Feuillet C, Langridge P, Waugh R. 2008. Cereal breeding takes a walk on the wild side. Trends in 773 Genetics 24: 24–32. 774 Franco-Navarro JD, Díaz-Rueda P, Rivero-Núñez CM, Brumós J, Rubio-Casal AE, de Cires A, 775 Colmenero-Flores JM, Rosales MA . 2021. Chloride nutrition improves drought resistance by 776 enhancing water deficit avoidance and tolerance mechanisms. Journal of Experimental Botany 72: 777 5246–5261. 778 Galdon-Armero J, Fullana-Pericas M, Mulet PA, Conesa MA, Martin C, Galmes J . 2018. The 779 ratio of trichomes to stomata is associated with water use efficiency in Solanum lycopersicum 780 (tomato). The Plant Journal 96: 607–619. 781 Gao M, Hao Z, Ning Y, He Z. 2024. Revisiting growth–defence trade-offs and breeding strategies 782 in crops. Plant Biotechnology Journal 22: 1198–1205. 783 Gibbard PL, Head MJ, Walker MJC, The Subcommission On Quaternary Stratigraphy. 2010. 784 Formal ratification of the Quaternary System/Period and the Pleistocene Series/Epoch with a base at 785 2.58 Ma. Journal of Quaternary Science 25: 96–102. 786 Goel M, Schneeberger K . 2022. plotsr: visualizing structural similarities and rearrangements 787 between multiple genomes. Bioinformatics 38: 2922–2926. 788 Goel M, Sun H, Jiao W-B, Schneeberger K. 2019. SyRI: finding genomic rearrangements and local 789 sequence differences from whole-genome assemblies. Genome Biology 20: 277. 790 Goodstein DM, Shu S, Howson R, Neupane R, Hayes RD, Fazo J, Mitros T, Dirks W, Hellsten 791 U, Putnam N, et al. 2012. Phytozome: a comparative platform for green plant genomics. Nucleic 792 Acids Research 40: D1178–D1186. 793 Gruner P, Miedaner T. 2021. Perennial Rye: Genetics of Perenniality and Limited Fertility. Plants 794 10: 1210. 795 Guo K, Liu M, Vella D, Suresh S, Hsia KJ. 2024. Dehydration-induced corrugated folding in Rhapis 796 excelsa plant leaves. Proceedings of the National Academy of Sciences 121: e2320259121. 797 Haas B. 2023. TransDecoder (Find Coding Regions Within Transcripts). 798 Hajjar R, Hodgkin T . 2007. The use of wild relatives in crop improvement: a survey of 799 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 25 developments over the last 20 years. Euphytica 156: 1–13. 800 Harlan JR, de Wet JMJ. 1971. Toward a Rational Classification of Cultivated Plants. Taxon 20: 801 509–517. 802 Hernandez J, Meints B, Hayes P . 2020. Introgression Breeding in Barley: Perspectives and Case 803 Studies. Frontiers in Plant Science 11. 804 Hernández-Sánchez IE, Maruri-López I, Martinez-Martinez C, Janis B, Jiménez-Bremont JF, 805 Covarrubias AA, Menze MA, Graether SP, Thalhammer A . 2022. LEAfing through literature: 806 late embryogenesis abundant proteins coming of age —achievements and perspectives. Journal of 807 Experimental Botany 73: 6525–6546. 808 Holst F, Bolger A, Günther C, Maß J, Triesch S, Kindel F, Kiel N, Saadat N, Ebenhöh O, Usadel 809 B, et al. 2023. Helixer–de novo Prediction of Primary Eukaryotic Gene Models Combining Deep 810 Learning and a Hidden Markov Model. : 2023.02.06.527280. 811 Hübner S, Höffken M, Oren E, Haseneyer G, Stein N, Graner A, Schmid K, Fridman E . 2009. 812 Strong correlation of wild barley (Hordeum spontaneum) population structure with temperature and 813 precipitation variation. Molecular Ecology 18: 1523–1536. 814 Hundertmark M, Hincha DK . 2008. LEA (Late Embryogenesis Abundant) proteins and their 815 encoding genes in Arabidopsis thaliana. BMC Genomics 9: 1–22. 816 Hyndman RJ, Einbeck J, Wand MP . 2023. hdrcde: Highest Density Regions and Conditional 817 Density Estimation. 818 Jackman SD, Coombe L, Chu J, Warren RL, Vandervalk BP, Yeo S, Xue Z, Mohamadi H, 819 Bohlmann J, Jones SJM, et al. 2018. Tigmint: correcting assembly errors using linked reads from 820 large molecules. BMC Bioinformatics 19: 1–10. 821 Jakob SS, Blattner FR . 2006. A Chloroplast Genealogy of Hordeum (Poaceae): Long -Term 822 Persisting Haplotypes, Incomplete Lineage Sorting, Regional Extinction, and the Consequences for 823 Phylogenetic Inference. Molecular Biology and Evolution 23: 1602–1612. 824 Jakob SS, Meister A, Blattner FR . 2004. The Considerable Genome Size Variation of Hordeum 825 Species (Poaceae) Is Linked to Phylogeny, Life Form, Ecology, and Speciation Rates. Molecular 826 Biology and Evolution 21: 860–869. 827 Jarzyniak KM, Jasiński M. 2014. Membrane transporters and drought resistance – a complex issue. 828 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 26 Frontiers in Plant Science 5. 829 Jayakodi M, Lu Q, Pidon H, Rabanus-Wallace MT, Bayer M, Lux T, Guo Y, Jaegle B, Badea A, 830 Bekele W, et al. 2024. Structural variation in the pangenome of wild and domesticated barley. Nature 831 636: 654–662. 832 Jayakodi M, Padmarasu S, Haberer G, Bonthala VS, Gundlach H, Monat C, Lux T, Kamal N, 833 Lang D, Himmelbach A, et al. 2020. The barley pan-genome reveals the hidden legacy of mutation 834 breeding. Nature 588: 284–289. 835 Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell 836 A, Nuka G, et al. 2014. InterProScan 5: genome-scale protein function classification. Bioinformatics 837 30: 1236–1240. 838 Kadioglu A, Terzi R, Saruhan N, Saglam A . 2012. Current advances in the investigation of leaf 839 rolling caused by biotic and abiotic stress factors. Plant Science 182: 42–48. 840 Kasapligil B. 1961. Foliar Xeromorphy of Certain Geophytic Monocotyledons. Madroño 16: 43–70. 841 Kashyap A, Garg P, Tanwar K, Sharma J, Gupta NC, Ha PTT, Bhattacharya RC, Mason AS, 842 Rao M. 2022. Strategies for utilization of crop wild relatives in plant breeding programs. Theoretical 843 and Applied Genetics 135: 4151–4167. 844 Kolmogorov M, Yuan J, Lin Y, Pevzner PA. 2019. Assembly of long, error-prone reads using repeat 845 graphs. Nature Biotechnology 37: 540–546. 846 von Korff M, Wang H, Léon J, Pillen K. 2004. Development of candidate introgression lines using 847 an exotic barley accession (Hordeum vulgare ssp. spontaneum) as donor. Theoretical and Applied 848 Genetics 109: 1736–1745. 849 Kumar S, Suleski M, Craig JM, Kasprowicz AE, Sanderford M, Li M, Stecher G, Hedges SB . 850 2022. TimeTree 5: An Expanded Resource for Species Divergence Times. Molecular Biology and 851 Evolution 39: msac174. 852 Li H. 2018. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34: 3094–3100. 853 Liakoura V , Stefanou M, Manetas Y, Cholevas C, Karabourniotis G. 1997. Trichome density and 854 its UV-B protective potential are affected by shading and leaf position on the canopy. Environmental 855 and Experimental Botany 38: 223–229. 856 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 27 Liao Y, Smyth GK, Shi W. 2014. featureCounts: an efficient general purpose program for assigning 857 sequence reads to genomic features. Bioinformatics 30: 923–930. 858 Liller CB, Walla A, Boer MP, Hedley P, Macaulay M, Effgen S, von Korff M, van Esse GW, 859 Koornneef M. 2017. Fine mapping of a major QTL for awn length in barley using a multiparent 860 mapping population. Theoretical and Applied Genetics 130: 269–281. 861 Lisch D. 2013. How important are transposons for plant evolution? Nature Reviews Genetics 14: 49–862 61. 863 Liu Y, Du H, Li P, Shen Y, Peng H, Liu S, Zhou G-A, Zhang H, Liu Z, Shi M, et al. 2020. Pan-864 Genome of Wild and Cultivated Soybeans. Cell 182: 162-176.e13. 865 Liu X, Wang Z, Wang L, Wu R, Phillips J, Deng X. 2009. LEA 4 group genes from the resurrection 866 plant Boea hygrometrica confer dehydration tolerance in transgenic tobacco. Plant Science 176: 90–867 98. 868 Ma J, Bennetzen JL . 2004. Rapid recent growth and divergence of rice nuclear genomes. 869 Proceedings of the National Academy of Sciences 101: 12404–12410. 870 Manni M, Berkeley MR, Seppey M, Simão FA, Zdobnov EM. 2021. BUSCO Update: Novel and 871 Streamlined Workflows a long with Broader and Deeper Phylogenetic Coverage for Scoring of 872 Eukaryotic, Prokaryotic, and Viral Genomes. Molecular Biology and Evolution 38: 4647–4654. 873 Marçais G, Kingsford C . 2011. A fast, lock -free approach for efficient parallel counting of 874 occurrences of k-mers. Bioinformatics 27: 764–770. 875 Marks RA, Van Der Pas L, Schuster J, Gilman IS, VanBuren R . 2024. Convergent evolution of 876 desiccation tolerance in grasses. Nature Plants 10: 1112–1125. 877 Mascher M, Marone MP, Schreiber M, Stein N. 2024. Are cereal grasses a single genetic system? 878 Nature Plants 10: 719–731. 879 Mascher M, Wicker T, Jenkins J, Plott C, Lux T, Koh CS, Ens J, Gundlach H, Boston LB, 880 Tulpová Z, et al. 2021. Long-read sequence assembly: a technical evaluation in barley. The Plant 881 Cell 33: 1888–1906. 882 Matus I, Corey A, Filichkin T, Hayes PM, Vales MI, Kling J, Riera-Lizarazu O, Sato K, Powell 883 W, Waugh R. 2003. Development and characterization of recombinant chromosome substitution 884 lines (RCSLs) using Hordeum vulgare subsp. spontaneum as a source of donor alleles in a Hordeum 885 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 28 vulgare subsp. vulgare background. Genome 46: 1010–1023. 886 Mazel A, Leshem Y, Tiwari BS, Levine A. 2004. Induction of Salt and Osmotic Stress Tolerance by 887 Overexpression of an Intracellular Vesicle Trafficking Protein AtRab7 (AtRabG3e). Plant Physiology 888 134: 118–128. 889 Mendes FK, Vanderpool D, Fulton B, Hahn MW. 2021. CAFE 5 models variation in evolutionary 890 rates among gene families. Bioinformatics 36: 5516–5518. 891 Mendiburu F de. 2023. agricolae: Statistical Procedures for Agricultural Research. 892 Middleton CP, Stein N, Keller B, Kilian B, Wicker T . 2013. Comparative analysis of genome 893 composition in Triticeae reveals strong variation in transposable element dynamics and nucleotide 894 diversity. The Plant Journal 73: 347–356. 895 Mikheenko A, Prjibelski A, Saveliev V , Antipov D, Gurevich A. 2018. Versatile genome assembly 896 evaluation with QUAST-LG. Bioinformatics 34: i142–i150. 897 Mittler R, Finka A, Goloubinoff P . 2012. How do plants feel the heat? Trends in Biochemical 898 Sciences 37: 118–125. 899 Mittler R, Vanderauwera S, Suzuki N, Miller G, Tognetti VB, Vandepoele K, Gollery M, 900 Shulaev V , Van Breusegem F. 2011. ROS signaling: the new wave? Trends in Plant Science 16: 901 300–309. 902 Monat C, Padmarasu S, Lux T, Wicker T, Gundlach H, Himmelbach A, Ens J, Li C, 903 Muehlbauer GJ, Schulman AH, et al. 2019. TRITEX: chromosome -scale sequence assembly of 904 Triticeae genomes with open-source tools. Genome Biology 20: 1–18. 905 Nagy Z, Németh E, Guóth A, Bona L, Wodala B, Pécsváradi A . 2013. Metabolic indicators of 906 drought stress tolerance in wheat: Glutamine synthetase isoenzymes and Rubisco. Plant Physiology 907 and Biochemistry 67: 48–54. 908 Navrátilová P, Toegelová H, Tulpová Z, Kuo Y -T, Stein N, Doležel J, Houben A, Šimková H, 909 Mascher M. 2022. Prospects of telomere-to-telomere assembly in barley: Analysis of sequence gaps 910 in the MorexV3 reference genome. Plant Biotechnology Journal 20: 1373–1386. 911 Nelson DE, Repetti PP, Adams TR, Creelman RA, Wu J, Warner DC, Anstrom DC, Bensen RJ, 912 Castiglioni PP, Donnarummo MG, et al. 2007. Plant nuclear factor Y (NF -Y) B subunits confer 913 drought tolerance and lead to improved corn yields on water -limited acres. Proceedings of the 914 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 29 National Academy of Sciences 104: 16450–16455. 915 Neumann P, Navrátilová A, Koblížková A, Kejnovský E, Hřibová E, Hobza R, Widmer A, 916 Doležel J, Macas J . 2011. Plant centromeric retrotransposons: a structural and cytogenetic 917 perspective. Mobile DNA 2: 4. 918 Nieves-Cordones M, García -Sánchez F, Pérez- Pérez JG, Colmenero-Flores JM, Rubio F, 919 Rosales MA. 2019. Coping With Water Shortage: An Update on the Role of K+, Cl -, and Water 920 Membrane Transport Mechanisms on Drought Resistance. Frontiers in Plant Science 10. 921 Nigel Maxted, Laura Rhodes, Isabelle Bradley. 2014. IUCN Red List of Threatened Species: 922 Hordeum erectifolium. IUCN Red List of Threatened Species. 923 Noctor G, Mhamdi A, Foyer CH. 2014. The Roles of Reactive Oxygen Metabolism in Drought: Not 924 So Cut and Dried. Plant Physiology 164: 1636–1648. 925 Nueda MJ, Tarazona S, Conesa A. 2014. Next maSigPro: updating maSigPro bioconductor package 926 for RNA-seq time series. Bioinformatics 30: 2598–2602. 927 Onoda Y, Wright IJ, Evans JR, Hikosaka K, Kitajima K, Niinemets Ü, Poorter H, Tosens T, 928 Westoby M. 2017. Physiological and structural tradeoffs underlying the leaf economics spectrum. 929 New Phytologist 214: 1447–1463. 930 Ou S, Jiang N. 2018. LTR_retriever: A Highly Accurate and Sensitive Program for Identification of 931 Long Terminal Repeat Retrotransposons. Plant Physiology 176: 1410–1422. 932 Ou S, Su W, Liao Y, Chougule K, Agda JRA, Hellinga AJ, Lugo CSB, Elliott TA, Ware D, 933 Peterson T, et al. 2019. Benchmarking transposable element annotation methods for creation of a 934 streamlined, comprehensive pipeline. Genome Biology 20: 1–18. 935 Pankin A, Altmüller J, Becker C, von Korff M . 2018. Targeted resequencing reveals genomic 936 signatures of barley domestication. New Phytologist 218: 1247–1259. 937 Pardo J, Man Wai C, Chay H, Madden CF, Hilhorst HWM, Farrant JM, VanBuren R . 2020. 938 Intertwined signatures of desiccation and drought tolerance in grasses. Proceedings of the National 939 Academy of Sciences 117: 10079–10088. 940 Peleg Z, Reguera M, Tumimbang E, Walia H, Blumwald E. 2011. Cytokinin-mediated source/sink 941 modifications improve drought tolerance and increase grain yield in rice under water -stress. Plant 942 Biotechnology Journal 9: 747–758. 943 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 30 Perico C, Tan S, Langdale JA. 2022. Developmental regulation of leaf venation patterns: monocot 944 versus eudicots and the role of auxin. New Phytologist 234: 783–803. 945 Poorter H, Niinemets Ü, Poorter L, Wright IJ, Villar R . 2009. Causes and consequences of 946 variation in leaf mass per area (LMA): a meta-analysis. New Phytologist 182: 565–588. 947 Presting GG, Malysheva L, Fuchs J, Schubert I . 1998. A TY3/GYPSY retrotransposon- like 948 sequence localizes to the centromeric regions of cereal chromosomes. The Plant Journal 16: 721–949 728. 950 Prjibelski AD, Mikheenko A, Joglekar A, Smetanin A, Jarroux J, Lapidus AL, Tilgner HU . 951 2023. Accurate isoform discovery with IsoQuant using long reads. Nature Biotechnology 41: 915–952 918. 953 R Core Team. 2024. R: A Language and Environment for Statistical Computing. Vienna, Austria: R 954 Foundation for Statistical Computing. 955 Redmann RE. 1985. Adaptation of Grasses to Water Stress-Leaf Rolling and Stomate Distribution. 956 Annals of the Missouri Botanical Garden 72: 833–842. 957 Renzi JP, Coyne CJ, Berger J, von Wettberg E, Nelson M, Ureta S, Hernández F, Smýkal P, 958 Brus J. 2022. How Could the Use of Crop Wild Relatives in Breeding Increase the Adaptation of 959 Crops to Marginal Environments? Frontiers in Plant Science 13. 960 Rhie A, Walenz BP, Koren S, Phillippy AM. 2020. Merqury: reference-free quality, completeness, 961 and phasing assessment for genome assemblies. Genome Biology 21: 1–27. 962 Sakuma Y, Maruyama K, Osakabe Y, Qin F, Seki M, Shinozaki K, Y amaguchi -Shinozaki K. 963 2006. Functional Analysis of an Arabidopsis Transcription Factor, DREB2A, Involved in Drought -964 Responsive Gene Expression. The Plant Cell 18: 1292–1309. 965 Sgroi LC, Lovino MA, Berbery EH, Müller GV . 2021. Characteristics of droughts in Argentina’s 966 core crop region. Hydrology and Earth System Sciences 25: 2475–2490. 967 Shields LM . 1950. Leaf Xeromorphy as Related to Physiological and Structural Influences. 968 Botanical Review 16: 399–447. 969 Šimková H, Tulpová Z, Cápal P . 2023. Flow Sorting–Assisted Optical Mapping. In: Heitkam T, 970 Garcia S, eds. Plant Cytogenetics and Cy togenomics: Methods and Protocols. New York, NY: 971 Springer US, 465–483. 972 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 31 Skirycz A, Inzé D . 2010. More from less: plant growth under limited water. Current Opinion in 973 Biotechnology 21: 197–203. 974 Skirycz A, Vandenbroucke K, Clauw P, Maleux K, De Meyer B, Dhondt S, Pucci A, Gonzalez 975 N, Hoeberichts F, Tognetti VB, et al. 2011. Survival and growth of Arabidopsis plants given limited 976 water are not equal. Nature Biotechnology 29: 212–214. 977 Stritt C, Wyler M, Gimmi EL, Pippel M, Roulin AC. 2020. Diversity, dynamics and effects of long 978 terminal repeat retrotransposons in the model grass Brachypodium distachyon. New Phytologist 227: 979 1736–1748. 980 Sun J, Cui X, Teng S, Kunnong Z, Wang Y, Chen Z, Sun X, Wu J, Ai P, Quick WP, et al. 2020. 981 HD-ZIP IV gene Ro c8 regulates the size of bulliform cells and lignin content in rice. Plant 982 Biotechnology Journal 18: 2559–2572. 983 Sun H, Ding J, Piednoël M, Schneeberger K . 2018. findGSE: estimating genome size variation 984 within human and Arabidopsis using k-mer frequencies. Bioinformatics 34: 550–557. 985 Sun Y, McManus JF, Clemens SC, Zhang X, Vogel H, Hodell DA, Guo F, Wang T, Liu X, An Z. 986 2021. Persistent orbital influence on millennial climate variability through the Pleistocene. Nature 987 Geoscience 14: 812–818. 988 The UniProt Consortium, Bateman A, Martin M-J, Orchard S, Magrane M, Ahmad S, Alpi E, 989 Bowler-Barnett EH, Britto R, Bye -A-Jee H, et al. 2023. UniProt: the Universal Protein 990 Knowledgebase in 2023. Nucleic Acids Research 51: D523–D531. 991 VanBuren R, Pardo J, Man Wai C, Evans S, Bartels D . 2019. Massive Tandem Proliferation of 992 ELIPs Supports Convergent Evolution of Desiccation Tolerance across Land Plants. Plant Physiology 993 179: 1040–1049. 994 Vaser R, Sović I, Nagarajan N, Šikić M . 2017. Fast and accurate de novo genome assembly from 995 long uncorrected reads. Genome Research 27: 737–746. 996 Vasimuddin Md, Misra S, Li H, Aluru S. 2019. Efficient Architecture-Aware Acceleration of BWA-997 MEM for Multicore Systems. In: 2019 IEEE International Parallel and Distributed Processing 998 Symposium (IPDPS). 314–324. 999 Vincent H, Bothmer R von, Knüpffer H, Amri A, Konopka J, Maxted N . 2012. Genetic gap 1000 analysis of wild Hordeum taxa. Plant Genetic Resources 10: 242–253. 1001 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 32 Vuilleumier BS. 1971. Pleistocene Changes in the Fauna and Flora of South America. Science 173: 1002 771–780. 1003 Waddington SR, Cartwright PM, WALL PC. 1983. A Quantitative Scale of Spike Initial and Pistil 1004 Development in Barley and Wheat. Annals of Botany 51: 119–130. 1005 Walther U, Rapke H, Proeseler G, Szigat G . 2000. Hordeum bulbosum -a new source of disease 1006 resistance-transfer of resistance to leaf rust and mosaic viruses from H. bulbosum into winter barley. 1007 Plant Breeding 119: 215–218. 1008 Wang H, Lu S, Guan X, Jiang Y, Wang B, Hua J, Zou B. 2022. Dehydration-Responsive Element 1009 Binding Protein 1C, 1E, and 1G Promote Stress Tolerance to Chilling, Heat, Drought, and Salt in 1010 Rice. Frontiers in Plant Science 13. 1011 Wendler N, Mascher M, Himmelbach A, Johnston P, Picke ring R, Stein N . 2015. Bulbosum to 1012 Go: A Toolbox to Utilize Hordeum vulgare/bulbosum Introgressions for Breeding and Beyond. 1013 Molecular Plant 8: 1507–1519. 1014 Wick RR, Judd LM, Gorrie CL, Holt KE . 2017. Completing bacterial genome assemblies with 1015 multiplex MinION sequencing. Microbial Genomics 3. 1016 Wicker T, Gundlach H, Spannagl M, Uauy C, Borrill P, Ramírez-González RH, De Oliveira R, 1017 Mayer KFX, Paux E, Choulet F . 2018. Impact of transposable elements on genome structure and 1018 evolution in bread wheat. Genome Biology 19: 1–18. 1019 Wickham H, Averick M, Bryan J, Chang W, McGowan LD, François R, Grolemund G, Hayes 1020 A, Henry L, Hester J, et al. 2019. Welcome to the Tidyverse. Journal of Open Source Software 4: 1021 1686. 1022 Winterfeld G, Tkach N, Röser M. 2025. Reductional dysploidy and genome size diversity in 1023 Pooideae, the largest subfamily of grasses (Poaceae). Plant Systematics and Evolution 311: 18. 1024 WMO. 2020. WMO Climatological Normals | World Meteorological Organization. 1025 Wu H-J, Zhang Z, Wang J-Y, Oh D-H, Dassanayake M, Liu B, Huang Q, Sun H-X, Xia R, Wu 1026 Y, et al. 2012. Insights into salt tolerance from the genome of Thellungiella salsuginea. Proceedings 1027 of the National Academy of Sciences 109: 12219–12224. 1028 Xu S, Hu E, Cai Y, Xie Z, Luo X, Zhan L, Tang W, Wang Q, Liu B, Wang R, et al. 2024. Using 1029 clusterProfiler to characterize multiomics data. Nature Protocols 19: 3292–3320. 1030 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 33 Xue D, Zhang X, Lu X, Chen G, Chen Z -H. 2017. Molecular and Evolutionary Mechanisms of 1031 Cuticular Wax for Plant Drought Tolerance. Frontiers in Plant Science 8. 1032 Ye Z, Sawada M, Iwasa M, Moriyama R, Dey D, Furutani M, Kitao M, Hara T, Tanaka A, 1033 Kishimoto J, et al. 2024. Revisiting the early light -induced protein hypothesis in the sustained 1034 thermal dissipation mechanism in yew leaves. Journal of Experimental Botany: erae412. 1035 Zhang H, Dawe RK . 2012. Total centromere size and genome size are strongly correlated in ten 1036 grass species. Chromosome Research 20: 403–412. 1037 Zhang S, Huang G, Zhang Y, Lv X, Wan K, Liang J, Feng Y, Dao J, Wu S, Zhang L, et al. 2023. 1038 Sustained productivity and agronomic potential of perennial rice. Nature Sustainability 6: 28–38. 1039 Zhang R-G, Li G-Y, Wang X-L, Dainat J, Wang Z-X, Ou S, Ma Y. 2022. TEsorter: An accurate 1040 and fast method to classify LTR -retrotransposons in plant genomes. Horticulture Research 9: 1041 uhac017. 1042 Zhu Z, Wang J, Li C, Li L, Mao X, Hu G, Wang J, Chang J, Jing R. 2022. A transcription factor 1043 TaMYB5 modulates leaf rolling in wheat. Frontiers in Plant Science 13. 1044 1045 Supporting Information – Supplemental Tables 1046 Table S1: Oxford Nanopore Technologies long read sequencing data. 1047 Table S2: 10x Genomics linked short read Illumina sequencing data. 1048 Table S3: Hi-C chromosomal conformation capture Illumina sequencing data. 1049 Table S4: Optical genome maps - Bionano. 1050 Table S5: Genome assembly metrics, completeness and quality score. From draft assembly, to finally 1051 pseudomolecules. 1052 Table S6: Overview of PacBio IsoSeq full-length capture transcript RNA sequencing of tissue-time 1053 specific samples. 1054 Table S7: Gene expression in the 22 tissue and time specific samples, read counts TPM normalized 1055 per sample with IsoQuant. 1056 Table S8: Summary of gene annotation, combined from evidence based gene prediction and ab initio, 1057 and lncRNA. 1058 Table S9: Transposable element annotation with EDTA in H. erectifolium acc. NGB6816, H.v. cv. 1059 Morex, and H.v. spontaneum acc. B1K-04-12. 1060 Table S10: Total number of expanded or contracted hierarchical phylogenetic orthologs (HOG) 1061 across eight genomes from seven species. 1062 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint 34 Table S11: All significantly expanded or contracted hierarchical phylogenetic orthologs (HOG) in H. 1063 erectifolium. 1064 Table S12: Desiccation related expanded gene families ELIP, DREB1C, and LEA. 1065 Table S13: Tissue and time specific expression of the expanded gene families: ELIP, DREB1C, and 1066 LEA genes in the 22 IsoSeq samples. 1067 Table S14: Transcriptome sequencing with Illumina PE150 of 64 leaf samples from H. erectifolium 1068 and Morex during drydown experiment. 1069 Table S15: RNAseq mapping rates of each sample to their respective genome. 1070 Table S16: Total number of significantly expressed genes found in H. erectifolium and Morex during 1071 drydown experiment. 1072 Table S17: All expressed genes found in H. erectifolium during drydown experiment. 1073 Table S18: All expressed genes found in Morex during drydown experiment. 1074 Table S19: Cross species comparison of significant differentially expressed single copy orthologs 1075 genes using between H. erectifolium and Morex. 1076 1077 Supporting Information - Supplemental figures 1078 Fig. S1: Quantitative leaf characteristics of the leaf below flag leaf (LBF) in H. erectifolium , 1079 cultivated (Morex) and wild barley. 1080 Fig. S2: Principal component analysis (PCA) of tissue-specific expression profiles. 1081 Fig. S3: Putative centromere locations and pericentric sizes in H. erectifolium, Morex, and B1K-04-1082 12. 1083 Fig. S4: Copia and Gypsy LTR retrotransposon insertions over time in chromosomes 2H and 7H. 1084 Fig. S5: Enriched biological pathways found in hierarchical phylogenetic orthologs. 1085 Fig. S6: Summary of expanded and contracted gene families found by CAFE5. 1086 Fig. S7: Significantly expanded gene families found in H. erectifolium related to drydown adaptation. 1087 Fig. S8: Leaf relative water content and plant morphology during drydown and recovery. 1088 Fig. S9: Soil field capacity and fresh weight biomass during drydown and recovery. 1089 Fig. S10: Time-course analysis of transcriptome changes over time in response to drydown. 1090 Fig. S11: Gene expression of expanded hierarchical phylogenetic orthologs gene families in response 1091 to drydown and recovery. 1092 1093 1094 1095 1096 1097 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint Fig. 1: Leaf morphology of H. erectifolium, cultivated (Morex) and wild barley. (a) A representative of Hordeum erectifolium (left), accession NGB 6816, from which all genetic and transcriptomic data was collected, and an example of H. vulgare cv. Morex (right). (b) Abscised leaf of H. erectifolium, folding inward towards the adaxial side of the midriff. (c) Transverse cut of a H. erectifolium leaf, rolled towards the adaxial side of the midriff. (d) Transverse cuts of H. erectifolium, Morex, and wild barley, lignin of fixated tissue was stained with toluidine blue. Red arrows point at bulliform cells in Morex and wild barley, and the expected location in H. erectifolium. Trichomes are indicated by green arrows in H. erectifolium. Ye l l o warrows point at the adaxial bundle sheath extensions (BSE) on a major vein; the BSE is as well on the abaxial side of the vein. Scale bars for 1000 µm are indicated in red and 100 µm in black. Quantitative leaf characteristics of the flag leaf (FL) in H. erectifolium, Morex, and wild barley: (e) number of veins per mm, n = 5, (f) minor to major vein ration, n = 5, (g) specific leaf area, cm2/mg, n = 8, (h) ratio of elemental carbon and nitrogen, n = 8. (i, j) FL thickness and width, mm, n = 5. Different letters indicate significantly differing groups, ANOVA with post-hoc Tukey HSD, p < 0. 05. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint Fig. 2: Genome characteristics, interchromosomal contact matrix, chromosomal synteny between genomes, and principal component analysis (PCA) of tissue- specific expression profiles. (a) K-mer spectra of 21-mers, calculated form short-read data, red line indicates fitted and observed 21-mers for genome size calculation. (b) Interchomosomal contact matrix of the seven assembled chromosomes. Pixel intensity represents Hi-C link counts normalized over a 1 Mb window on a logarithmic scale. (c) Whole chromosomal synteny between each respective chromosome in H. erectifolium, Morex, and B1K-04-12. Inversions smaller than 1 Mb were excluded for visualization. Purple diamonds indicate putative centromere location and red triangles indicate identified telomeric sequences. (d) PCA clustering of transcript abundance in 21 tissue-time specific samples and the trajectory of the top 20 highest loading genes for PC1 and PC2. Eleven individual tissues of which ten were sampled both in the morning (MOR, ZT 1-3) and evening (EVE, ZT 13- 15). Opposite trajectories of vegetative versus reproductive tissues along PC1 and a third trajectory on PC2 contributed by anthers (ANT) and to a lesser extent, roots (RO8). Abbreviations: flag leaf (FLF), three-day-old germinating seeds (GS3), third internode (IND), fourth node (NOD), whole shoots, (LS 8, 8 days post germination), whole roots (RO8, 8 days post germination), ovules (OVU), anthers (ANT), caryopses (CAR), developing shoot apical meristems: ESP (W3.0-4.5), MSP (W5.0-6.5), LSP (W7.0-8.0) (Waddington et al., 1987). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint Fig. 3: Transposable elements composition, long terminal repeat (LTR) retrotransposon profiles, and frequency of predicted insertion times. (a) Frequency plot of estimated LT R retrotransposon superfamily insertion times, Gypsy (red) and Copia (blue). Insertion times were calculated based on mutation rate in rice (1.3x10-8) of intact LT R retrotransposons in H. erectifolium, Morex, and B1K-04-12. Dotted lines mark LT R retrotransposon insertion burst peaks at 0.5 million years ago (Mya) and ~1.8 Mya. (b) Composition landscape of intact LTR retrotransposons in H. erectifolium, Morex, and B1K-04-12. The percentage of LT R retrotransposons was normalized to the total of intact LTR retrotransposons within a genome. (c) Intact LT Rretrotransposon insertion frequency over time by clade for the five most abundant clades; Copia: Angela and SIRE (blue), Gypsy: Athila, Retand, and Tekay (red). The black dotted lines mark 0.5 Mya and 1.8 Mya, respectively. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint Fig. 4: Shared and unique orthologs of H. erectifolium with Poaceae and A. thaliana. Tissue and time-specific expression of expanded gene families. (a) Shared and unique hierarchical phylogenetic orthologs (HOGs) between nine genomes of eight species; eight Poaceae species plus A. thaliana, the wheat (T . aestivum) subgenomes were separated; A, B, and D. Shared HOGs in Poaceae (green) and unique to H. erectifolium (blue) (b) Heatmap with expressed genes of expanded gene families with found in the 22 tissue-and-time specific samples; EARLY LIGHT-INDUCED PROTEINS (ELIP) (N0.HOG0000208), LATE EMBRYOGENESIS ABUNDANT PROTEIN 6-RELATED (LEA PROTEIN 6-RELATED) (N0.HOG0001214), and DEHYDRATION-RESPONSIVE ELEMENT -BINDING PROTEIN 1C (DREB1C) (N0.HOG0002366). Transcript abundance indicated as log(TPM), and timepoints : morning (red, MOR, ZT 1-3), and evening (blue, EVE, ZT 13 -15). Tissues are divided in to reproductive: developing shoot apical meristems: ESP (W3.0-4.5), MSP (W5.0-6.5), LSP (W7.0-8.0), ovules (OVU), anthers (ANT), caryopses (CAR), and vegetative: three-day-old germinating seeds (GS3), whole roots (RO8, 8 days post germination), whole shoots, (LS 8, 8 days post germination), fourth node (NOD), third internode (IND), flag leaf (FLF) (Waddington et al ., 1987). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint Fig. 5: Differentially expressed genes (DEG) during drydown and recovery in H. erectifolium and Morex. (a) Tot a lnumber of shared and exclusive single-copy ortholog (SCO) genes differentially expressed between H. erectifolium and Morex . (b) GO enrichment of shared and exclusive SCO DEGs : exclusive to H. erectifolium (A), exclusive to Morex (B), and interspecies overlap (AB). GO categories: biological processes (BP), cellular compartments (CC), and molecular functions (MF). (c) Five groups of the enriched biological processes terms in SCO : amide biosynthetic process, vesicle-mediated transport, photosynthesis, ribosome biogenesis, and response to oxygen – containing compound, total number of genes shown in brackets. The heatmap shows scaled log(fold- change), and gene clustering was based on H. erectifolium transcripts with the corresponding SCO in Morex fixed to the same location in the heatmap . Second day -of-treatment (DOT) (D2); fifth DOT (D5); D6, sixth DOT (D6); seventh DOT (D7, recovery). Gene expression of expanded hierarchical phylogenetic orthologs (HOG) gene families in H. erectifolium, and Morex genes found in the same HOG, in response to drydown and recovery. (d) EARLY LIGHT-INDUCED PROTEINS (ELIP) (N0.HOG0000208), (e) LATE EMBRYOGENESIS ABUNDANT PROTEIN 6-RELATED (LEA PROTEIN 6-RELATED) (N0.HOG0001214), (f) DEHYDRATION-RESPONSIVE ELEMENT -BINDING PROTEIN 1C (DREB1C) (N0.HOG0002366). Graphs show normalized counts per million (CPM) values for control and drydown. All genes, except those marked with asterisk, were differentially regulated in response to drydown or recovery . .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 31, 2025. ; https://doi.org/10.1101/2025.08.27.672388doi: bioRxiv preprint