Research ArticleLARGE-SCALE BIOLOGY ARTICLES
Open Access

Desiccation Tolerance Evolved through Gene Duplication and Network Rewiring in Lindernia

Robert VanBuren, Ching Man Wai, Jeremy Pardo, Valentino Giarola, Stefano Ambrosini, Xiaomin Song, Dorothea Bartels

Published December 2018. DOI: https://doi.org/10.1105/tpc.18.00517

Abstract

Although several resurrection plant genomes have been sequenced, the lack of suitable dehydration-sensitive outgroups has limited genomic insights into the origin of desiccation tolerance. Here, we utilized a comparative system of closely related desiccation-tolerant (Lindernia brevidens) and -sensitive (Lindernia subracemosa) species to identify gene- and pathway-level changes associated with the evolution of desiccation tolerance. The two high-quality Lindernia genomes we assembled are largely collinear, and over 90% of genes are conserved. L. brevidens and L. subracemosa have evidence of an ancient, shared whole-genome duplication event, and retained genes have neofunctionalized, with desiccation-specific expression in L. brevidens. Tandem gene duplicates also are enriched in desiccation-associated functions, including a dramatic expansion of early light-induced proteins from 4 to 26 copies in L. brevidens. A comparative differential gene coexpression analysis between L. brevidens and L. subracemosa supports extensive network rewiring across early dehydration, desiccation, and rehydration time courses. Many LATE EMBRYOGENESIS ABUNDANT genes show significantly higher expression in L. brevidens compared with their orthologs in L. subracemosa. Coexpression modules uniquely upregulated during desiccation in L. brevidens are enriched with seed-specific and abscisic acid-associated cis-regulatory elements. These modules contain a wide array of seed-associated genes that have no expression in the desiccation-sensitive L. subracemosa. Together, these findings suggest that desiccation tolerance evolved through a combination of gene duplications and network-level rewiring of existing seed desiccation pathways.

INTRODUCTION

Comparative systems are a powerful tool for dissecting the molecular basis of complex biological traits. The origins of desiccation tolerance in resurrection plants are largely unknown, but the underlying genetic signatures could be traced using pairs of closely related desiccation-sensitive and -tolerant species. Such an approach has been applied to Eragrostis (Vander Willigen et al., 2001), Selaginella (Yobi et al., 2013), and Sporobolus (Oliver et al., 2011) at the morphological and biochemical levels to identify signatures that distinguish drought and desiccation responses. Detailed pairwise comparisons have identified changes in cell wall composition (Plancot et al., 2014), metabolite and osmoprotectant accumulation (Oliver et al., 2011; Yobi et al., 2013), and physical properties unique to desiccation-tolerant species. Although genomes are available for several resurrection plants (VanBuren et al., 2015, 2018; Xiao et al., 2015; Costa et al., 2017), genomic resources in these comparative lineages are limited, and no genomes of closely related desiccation-sensitive species have been sequenced. High-quality reference genomes are available for the desiccation-sensitive Selaginella moellendorffii (Banks et al., 2011) and the desiccation-tolerant Selaginella lepidophylla (VanBuren et al., 2018), but their estimated divergence 248 million years ago prevents detailed genomic comparisons (Baniaga et al., 2016).

Resurrection plants endure extreme and prolonged drought events through vegetative desiccation, entering a preserved and protected quiescent state that functionally mirrors seed dormancy in angiosperms. Desiccation tolerance was a critical adaptation during early land plant evolution, and many early-diverging fern, moss, and lycophyte lineages have retained or convergently evolved these ancestral resilience mechanisms (Proctor, 1990; Oliver et al., 2000; Lüttge et al., 2011). Vegetative desiccation tolerance is comparatively less common in angiosperms, and recent genomic and metabolic studies suggest that it evolved through rewiring seed desiccation pathways (Costa et al., 2017; VanBuren et al., 2017). Resurrection plants have a conserved set of molecular signatures associated with desiccation tolerance (Illing et al., 2005; Zhang and Bartels, 2018), but the underlying genomic basis of this trait is largely unknown. The cooption of seed- and desiccation-associated pathways in resurrection plants can occur through a broad range of mechanisms at the gene, pathway, or network level. Changes in gene regulation at network hubs could drive pathway-level rewiring to upregulate a cascade of desiccation-related mechanisms. Changes in cis-regulation at isolated nodes in a network could shift the stoichiometry or abundance of endpoint metabolites and proteins related to desiccation. Gene and genome duplication also can drive adaptive evolution by providing additional copies for pathway- and gene-level subfunctionalization and neofunctionalization.

Figure1

Desiccation tolerance is prominent in Linderniaceae (order Lamiales) within the clade spanning Craterostigma and Lindernia (Rahmanzadeh et al., 2005). Craterostigma plantagineum is a model resurrection plant (Bartels and Salamini, 2001) native to rocky outcrops of sub-Saharan Africa. All Craterostigma and some desert-adapted Lindernia species are desiccation tolerant, but most Lindernia species are desiccation sensitive. Lindernia brevidens is unusual in that it displays desiccation tolerance despite an endemic habitat in the montane rainforests of eastern Africa that never experience seasonal drying (Phillips et al., 2008). Desiccation tolerance is likely an ancestral trait in this group (Fischer et al., 2013) and was retained in L. brevidens before its radiation to the tropical rainforest. Lindernia is paraphyletic (Fischer et al., 2013), and species outside of the clade containing Craterostigma and L. brevidens, such as Lindernia subracemosa, are desiccation sensitive. This diversity in desiccation tolerance makes Lindernia an excellent comparative system in which to test the contribution of gene duplication, cis-elements, and pathway rewiring in the evolution of desiccation tolerance. Here, we assembled high-quality reference genomes for the desiccation-tolerant L. brevidens and desiccation-sensitive L. subracemosa. Detailed comparative genomics and differential coexpression network analysis allowed us to survey the genetic basis of desiccation tolerance in Lindernia."

RESULTS

Comparative Grade Reference Genomes for Lindernia

C. plantagineum is a well-studied model for the evolution of desiccation tolerance in eudicots, but its highly complex, octoploid genome has hindered genome-scale analyses. L. brevidens and L. subracemosa are diploid with relatively small genomes (270 and 250 Mb, respectively), providing an excellent alternative system (Figure 1).

Figure 1.
Figure 1.

Comparative Desiccation Tolerance System within Linderniaceae.

Inferred phylogeny from Fischer et al. (2013) shows the two model desiccation-tolerant species (C. plantagineum and L. brevidens) and the desiccation-sensitive outgroup (L. subracemosa). Ploidy, karyotype, and genome size are shown on branches.

We generated high-quality reference genomes for both Lindernia species using a PacBio-based, single-molecule real-time sequencing approach. In total, we generated 21.7 and 17.9 Gb of filtered PacBio data, collectively representing 80.3× and 71.6× coverage for L. brevidens and L. subracemosa, respectively (Supplemental Figure 1). Raw PacBio reads were error corrected and assembled using the long-read assembler Canu (Koren et al., 2017), which is optimized to avoid collapsing highly repetitive and tandemly duplicated regions. Contigs were polished using high-coverage Illumina data with Pilon (Walker et al., 2014) to remove residual errors. The L. brevidens assembly spanned 265 Mb across 267 contigs with a contig N50 (length where half or more of the assembly is contained) of 3.6 Mb. The L. subracemosa assembly was slightly smaller, at 246 Mb with 328 contigs and an N50 of 1.9 Mb (Table 1). The total assembly sizes were consistent with the estimated genome sizes of 270 and 250 Mb based on flow cytometry. Lindernia species are primarily self-pollinated with low residual within-genome heterozygosity, which contributed to the high contiguity and relatively simple graph-based assembly structures (Supplemental Figures 2 and 3).

View this table:
Table 1. Lindernia Genome Assembly Metrics

We used high-throughput chromatin conformation capture (Hi-C) to generate a chromosome-scale assembly of L. brevidens. The Hi-C-based Illumina reads were mapped to the draft assembly using bwa (Li, 2013) followed by filtering and proximity-based clustering using the Juicer pipeline (Durand et al., 2016) (Supplemental Table 1). This approach yielded 14 high-confidence clusters corresponding to the haploid chromosome number in L. brevidens (2n = 2x = 28; Figure 2). In total, 121 contigs were ordered and oriented into 14 scaffolds collectively representing 94.7% of the assembly (249 out of 263 Mb; Supplemental Table 2). This included anchoring 98.8% of the predicted gene models. The repetitive element density was inversely correlated with gene density, and most chromosomes contained large tracts of retrotransposons (RTs), which likely correlate with centromere position (Figure 3).

Figure 2.
Figure 2.

Hi-C Clustering for Pseudomolecule Construction in L. brevidens.

A postclustering heat map of Hi-C-based intrachromosomal interactions in L. brevidens is shown. Pseudomolecules corresponding to the 14 haploid chromosomes are delineated by gray boxes.

Figure 3.
Figure 3.

Landscape of the L. brevidens Genome.

LTR-RT and coding sequence (CDS) density are plotted in sliding windows of 50 kb with 25-kb step size for the 14 L. brevidens scaffolds (chromosomes). Red indicates high density and blue indicates low density of coding sequence and LTR-RTs in the heat map below each landscape.

The genomes of L. brevidens and L. subracemosa were of similar size and the same karyotype, suggesting that they should have comparable repetitive elements and gene composition. Long terminal repeat (LTR)-RTs were the most abundant repetitive elements in both genomes, and they collectively spanned 34% (92.0 Mb) and 31% (77.4 Mb) of the L. brevidens and L. subracemosa genomes, respectively (Table 1). Despite the similar overall LTR composition, L. subracemosa had significantly more intact LTRs compared with L. brevidens (1972 versus 1025; Wilcoxon rank sum, P < 0.05). The distribution of LTR-RT insertion time was similar in both species, and most intact elements inserted within the last 1 million years (Figure 4). These findings suggest that LTR-RTs are similarly active in both genomes but may fractionate more quickly in L. brevidens.

Figure 4.
Figure 4.

Insertion Time of Intact LTR-RTs.

The average insertion time calculated from the divergence of LTR pairs is plotted for the 1025 intact LTR-RTs in L. brevidens and the 1972 intact LTR-RTs in L. subracemosa. MYA, million years ago.

The overall gene composition was similar in both Lindernia species, though L. subracemosa had more annotated gene models. Ab initio gene prediction using the dehydration time-course RNAseq data and protein similarities to other angiosperms identified 27,204 and 33,344 gene models in L. brevidens and L. subracemosa, respectively (Table 1). We assessed annotation quality using the BUSCO pipeline and found 91% and 90% (1319 and 1298) of the 1440 genes in the Embryophyta data set present in the L. brevidens and L. subracemosa assemblies. This proportion is comparable with results from other recent PacBio-based genomes.

Comparative Genomics of Lindernia

The L. brevidens and L. subracemosa genomes were largely collinear based on whole-genome alignment, and 24,053 L. brevidens genes had syntenic orthologs in L. subracemosa. Roughly 70% of the genomes were conserved in 2:2 syntenic blocks, supporting a shared, ancient whole-genome duplication (WGD) event in both species (Figure 5; Supplemental Figures 4 and 5). Six of the seven ancestral homeologous chromosome pairs from the WGD were intact in L. brevidens, including modern chromosome pairs: 1 and 13, 2 and 14, 3 and 5, 6 and 9, 7 and 10, and 8 and 11 (Figure 5A). Two of the ancestral homeologous chromosomes were fused in modern chromosome 5, and chromosome 12 contained fragments from several ancestral chromosomes. Chromosomal rearrangements were difficult to identify in L. subracemosa given its contig-level assembly, but there were no obvious rearrangements based on macrosynteny with L. brevidens (Supplemental Figure 5). The ancestral subgenomes were heavily fractionated, and only 7742 gene pairs were retained in duplicate in L. brevidens and 8452 in L. subracemosa based on synteny. Gene-level fractionation was biased toward a dominant subgenome that contained significantly more genes (Figure 5B; Supplemental Figure 6).

Figure 5.
Figure 5.

Comparative Genomics of L. brevidens and L. subracemosa.

(A) Syntenic dot plot of L. brevidens showing retained gene pairs from the recent WGD event. Each black dot represents a pair of retained genes.

(B) Microsynteny between syntenic blocks of the L. brevidens (top) and L. subracemosa (bottom) genomes. Syntenic gene pairs between L. brevidens versus L. subracemosa are shown by brown connections, and retained WGD gene pairs within each genome are shown by gray connections. Genes are colored by orientation in L. brevidens (light blue are forward, dark blue are reverse) and L. subracemosa (light red are forward, dark red are reverse).

We identified patterns of gene duplication and loss that may be related to the evolution of desiccation tolerance and other lineage-specific traits. Most gene pairs from the WGD were either retained in duplicate or fractionated to single copies in both species, including 11,874 single-copy genes (1:1) and 7568 duplicated genes (2:2) in both genomes (Table 2). We identified 3200 lineage-specific genes in L. brevidens (1:0 or 2:0) and 7067 lineage-specific genes in L. subracemosa (0:1 or 0:2) based on synteny. The higher number of lineage-specific genes in L. subracemosa was likely related to differences in total annotated gene number (27,204 versus 33,344). The lineage-specific genes in L. brevidens were enriched in Gene Ontology (GO) terms related to chlorophyll biosynthesis and metabolism, regulation of mitosis, and response to heat, which may suggest a role for these pathways in desiccation tolerance (Supplemental Table 3).

View this table:
Table 2. Comparison of Biased Fractionation Following the Shared WGD in Lindernia

New genes can arise through tandem gene duplication, and tandem gene duplications are associated with adaptive evolution (Cannon et al., 2004), including of desiccation tolerance in other resurrection plants (VanBuren et al., 2015, 2018). L. brevidens and L. subracemosa had a similar overall number of tandem genes but major differences in array size. L. brevidens had 2673 tandem arrays containing 5345 genes with array sizes ranging from 2 to 24 members. L. subracemosa had 3404 tandem arrays across 6809 genes with array sizes ranging from 2 to 31 (Figure 6). Through cross-referencing with syntenic gene pairs, we found that most tandem arrays were conserved between L. brevidens and L. subracemosa. Only 153 tandem arrays were specific to L. brevidens and 247 arrays were specific to L. subracemosa. Although tandem gene arrays were generally conserved, array sizes were highly variable and few contained the same number of genes between species (Figure 6B). Together, these data suggest that most tandem gene duplication events are ancestral but that each species has undergone unique array expansion and contraction.

Figure 6.
Figure 6.

Comparison of Tandem Gene Arrays in Lindernia.

(A) Histogram of tandem array sizes.

(B) Heat map of tandem array size in syntenic orthologs between L. brevidens and L. subracemosa. Values are plotted as the proportion of tandem genes in each category against all the genes in that array size.

Global Expression Patterns and Desiccation-Related Network Rewiring

To construct a comparative framework of genes related to desiccation, we conducted parallel sampling of leaf tissue during desiccation and rehydration time courses in L. brevidens and L. subracemosa. Parallel sampling between species allowed us to distinguish between genes involved in typical dehydration responses and those related specifically to desiccation tolerance. Sampling ranged from mild dehydration stress (relative water content [RWC] 53–56%; 3 d) through severe dehydration (RWC 23–27%; 7 d) and desiccation (RWC 6–9%; 10 and 14 d), followed by 24 and 48 h post rehydration (Figure 7). RWC was 53 to 56% at day 3 and fell below 10% after 10 d of drought in both species (Figure 7A). L. subracemosa plants were largely dead upon rehydration, and L. brevidens plants were mostly viable and physiologically active at 48 h post rehydration (RWC 44%) (Figure 7).

Figure 7.
Figure 7.

Overview of Desiccation and Rehydration Processes in Lindernia.

(A) RWC of fresh leaf tissues (F), 3, 7, 10, and 14 d of drought (D), and 24 and 48 h post rehydration (R) in L. brevidens and L. subracemosa. Error bars represent the se with three replicates for each RWC measurement.

(B) Representative L. brevidens (top) and L. subracemosa (bottom) at various RWC.

The greatest changes in gene expression occurred at two time points during the transition from mild to severe dehydration stress and from desiccated to rehydrated (Supplemental Table 4). The number of differentially expressed genes between well-watered and mild dehydration (F versus D3) were relatively similar in both species (5322 versus 4824 in L. subracemosa and L. brevidens). Many syntenic gene pairs had similar expression levels, with 581 upregulated and 133 downregulated in both species. Significantly more genes were differentially expressed between mild and severe dehydration stress (D3 versus D7), with 4329 and 9227 differentially expressed genes, respectively. A similar proportion of syntenic gene pairs was upregulated in both species at D7 (581), but significantly more gene pairs were similarly downregulated (1396) compared with mild dehydration stress. This pattern suggests that there is conservation of downregulated pathways in desiccation-sensitive and -tolerant species.

In L. brevidens, gene expression was relatively stable from severe dehydration to desiccation (D7, D10, and D14), whereas a high proportion of genes were differentially expressed in L. subracemosa during the transition to desiccation. This reflects the stability of desiccated L. brevidens and the imminent death of L. subracemosa. Few genes were similarly differentially expressed in both species under severe dehydration and desiccation (D7 versus D10: 50 and 62; D10 versus D14: 3 and 0, upregulated and downregulated, respectively; Supplemental Table 4). A substantial proportion of syntenic gene pairs (2065) were similarly upregulated in both species during early rehydration (24 h), supporting the conclusion that there is conserved activation of repair pathways. Expression changes in both species were minimal between 24 and 48 h post rehydration. Although L. subracemosa had some transcriptional response post rehydration, this was not sufficient to repair the extensive desiccation-induced damage. Together, the divergent expression patterns suggest that there is extensive upregulation of distinct pathways with desiccation-specific roles.

We conducted GO enrichment analysis of the gene pairs that were uniquely upregulated in L. brevidens, with no change or a decrease in expression in L. subracemosa. We reasoned that such genes are likely to be specific to the induction of desiccation tolerance. Most of the GO terms enriched among genes upregulated in mild dehydration stress (D3) were related to responses to abiotic stress and secondary metabolite biosynthesis (Supplemental Table 5), suggesting early activation of protective mechanisms. There were only a few GO terms enriched among genes upregulated in severe dehydration (D7) and desiccation (D10 and D14), including terms related to transport, vacuole organization, ion homeostasis, and RNA modification (Supplemental Table 5). Most GO terms of genes uniquely downregulated in L. brevidens under mild and severe dehydration stress were related to photosynthesis processes, suggesting that the photosynthetic apparatus is inactivated early under mild dehydration compared with L. subracemosa (Supplemental Table 6).

The large-scale expression changes unique to desiccation in L. brevidens may be driven by changes in cis-regulation. Genes with unique desiccation-related expression in L. brevidens were enriched with cis-regulatory elements associated with dehydration and abscisic acid (ABA)-mediated responses as well as seed development pathways (Supplemental Table 7). Enriched cis-elements associated with typical ABA-mediated dehydration responses included ABF1 and ABF2 (Yoshida et al., 2015) among others. Enriched seed maturation-associated cis-elements included bZIP53 (Alonso et al., 2009), ABA-responsive element binding protein3 (AREB3; Nakashima et al., 2009), and ABI5 (Lopez-Molina et al., 2001). Cis-elements at the interface of heat- and dehydration-mediated ABA responses, such as Heat Stress Factor A6b (HSFA6B; Huang et al., 2016) and HSF7, also were enriched in desiccation-related genes. Enriched cis-elements in genes downregulated during desiccation in L. brevidens had wide roles in plant growth and development, hormone responses, and photosynthesis (Supplemental Table 8).

To compare network-level gene expression in L. brevidens and L. subracemosa, we utilized a weighted correlation network analysis approach (Langfelder and Horvath, 2008) across the dehydration and rehydration RNAseq time courses. This comparative coexpression network approach allowed us to parse conserved dehydration-related patterns from desiccation-specific pathway rewiring. After filtering genes with low expression (see Methods), we constructed two coexpression networks, with 14,246 genes in 10 modules for L. brevidens and 14,075 genes in 9 modules for L. subracemosa (Figure 8; Supplemental Figures 7 and 8). Based on their temporal dynamics, coexpression modules could be broadly classified into three groups: (1) high expression in well-watered tissue but downregulation in dehydration/desiccation; (2) expression during early dehydration; and (3) sustained high expression throughout dehydration and desiccation (Figure 8). Modules 1, 2, 3, and 5 in the L. brevidens network and modules 2, 4, and 5 in L. subracemosa had high expression in well-watered and rehydrating conditions, with decreasing expression throughout dehydration/desiccation time points (Figure 8). Modules 7 and 8 in L. brevidens and modules 3, 7, and 8 in L. subracemosa were involved in early dehydration responses, with peak expression at 3 or 7 d of dehydration. Modules 4, 6, and 10 in L. brevidens and modules 1 and 3 in L. subracemosa had sustained dehydration and desiccation-induced expression.

Figure 8.
Figure 8.

Comparative Coexpression Networks during Desiccation and Rehydration in Lindernia.

The mean expression of genes from modules in the coexpression network in L. brevidens (left) and L. subracemosa (right) are plotted for the seven time points. Conserved modules with upregulation during desiccation are highlighted in yellow in both networks, and modules with downregulation during desiccation are highlighted in gray.

We compared module overlap between the networks to identify patterns of conservation and species-specific divergence. Modules downregulated during severe dehydration and desiccation were largely conserved between the two species, with 74% of gene pairs falling in the same group of modules (Supplemental Figure 9). Modules upregulated during dehydration and desiccation had comparatively little overlap between species, and only 43% of gene pairs were in overlapping modules (Supplemental Figure 9). This indicates that there was significant pathway rewiring during desiccation. Network-wide cis-regulatory element enrichment patterns mirrored the observations comparing pairwise differentially expressed genes. Desiccation-associated modules from the L. brevidens network were enriched in dehydration-associated ABA-responsive cis-elements and seed maturation-associated cis-elements, including bZIP53, AREB3, and ABI5 among others (Supplemental Table 9).

Unique Desiccation-Related Pathways in L. brevidens

The similarities between seed and vegetative desiccation suggest overlapping pathways, which is supported by expression data from several resurrection plant lineages (Costa et al., 2017; VanBuren et al., 2017). We identified a wide range of seed-specific genes and pathways that were expressed only under dehydration in L. brevidens compared with syntenic orthologs in L. subracemosa (Supplemental Table 10). Seed storage proteins serve as a reserve of nitrogen, carbon, and sulfur for germinating seeds, and they likely play a role in seed longevity (Nguyen et al., 2015). Orthologs to genes encoding 2S and 12S seed storage proteins were generally upregulated in L. brevidens under desiccation, and syntenic orthologs in L. subracemosa were not expressed or were expressed highly in well-watered conditions (Supplemental Table 10). Delay of germination1 (DOG1) is an essential component of seed dormancy regulation, and its expression affects hundreds of seed-related genes (Dekkers et al., 2016). DOG1 was highly expressed in well-watered L. brevidens tissues but was downregulated during desiccation. The L. subracemosa DOG1 transcript had a low basal-level expression in all time points.

Oil bodies are lipid organelles filled with triacylglycerols that function as high-density energy reserves during seed germination. Oil bodies accumulate in desiccated leaf tissue of Oropetium and likely play a role in desiccation tolerance. Oil bodies are enveloped with oleosin structural proteins that prevent membrane coalescence and protect membrane integrity during freeze-thaw cycles (Shimada et al., 2008). Oil body membranes also are studded with calcium binding caleosin proteins that are associated with oil body degradation (Poxleitner et al., 2006) and general stress response pathways (Shen et al., 2014). L. brevidens and L. subracemosa had similar numbers of genes for oleosin (eight versus seven) and caleosin (four versus three) proteins, although L. brevidens had more retained whole-genome and tandem duplicates (Figure 9). Most oleosin and caleosin genes had low or undetectable expression in well-watered tissue, but several were induced during progressive dehydration and desiccation. Six oleosin genes in L. brevidens and three oleosin genes in L. subracemosa were upregulated in desiccating tissue, with most having a peak expression of less than 30 transcripts per million (TPM). L. brevidens had a pair of retained duplicated oleosin genes where one ortholog peaked at more than 500 TPMs in desiccating tissue, and the single syntenic ortholog in L. subracemosa had a relatively low expression. No caleosin genes were upregulated in L. subracemosa upon desiccation, but a pair of syntenic orthologs in L. brevidens were abundantly expressed (Figure 9).

Figure 9.
Figure 9.

Subfunctionalization of Oleosin and Caleosin Genes in L. brevidens.

Heat maps of log2 transformed expression of genes encoding oleosins (top) and caleosins (bottom) in L. brevidens and L. subracemosa. Syntenic orthologs are connected by brown lines including 1:2 and 2:2 orthologs between species.

Early light-induced proteins (ELIPs) are predicted to bind chlorophyll and function in photoprotection under high light and other abiotic stresses. The L. subracemosa genome had four genes encoding ELIP proteins, including a pair of syntelogs retained from the WGD event (Figure 10). Expression of two ELIP genes was hardly detectable during the surveyed time points, and two others were highly expressed during dehydration. The number of ELIPs in L. subracemosa was similar to that in other desiccation-sensitive angiosperms, and their dehydration-induced expression was consistent with the hypothesized protective mechanisms (Hayami et al., 2015). By contrast, the L. brevidens genome had undergone a dramatic expansion of ELIP genes, with 26 in total, including a large tandem array of 19 duplicates. This large tandem array was collinear to a pair of retained syntenic orthologs in L. subracemosa and a single retained gene copy in L. brevidens. Nearly all the ELIP genes in this array, and dispersed copies throughout the genome, were highly expressed during severe dehydration, desiccation, and rehydration, but they were hardly expressed in well-watered and mildly dehydrated tissue (Figure 10B). The tandem array was syntenic with the highly expressed ortholog in L. subracemosa, and the single-copy syntelog in L. brevidens and its syntenic ortholog in L. subracemosa were not expressed in dehydrated tissue. This suggests an ancestral subfunctionalization of this duplicated pair where only one gene copy was involved in dehydration-related responses. After the divergence of L. brevidens and L. subracemosa, the dehydration-specific syntelog likely underwent massive tandem proliferation in L. brevidens.

Figure 10.
Figure 10.

Tandem Proliferation of Desiccation-Associated ELIP Genes in L. brevidens.

(A) Microsynteny of a large tandem gene array in L. brevidens compared with the single-gene syntenic ortholog in L. brevidens and the two syntenic regions in L. subracemosa.

(B) Log2 transformed expression patterns of ELIPs in L. subracemosa (top) and L. brevidens (bottom). The large tandem array and syntenic whole-genome duplicates are labeled.

In Arabidopsis (Arabidopsis thaliana), STAY-GREEN (SGR) proteins are key regulators of chlorophyll degradation, and they are typically upregulated under abiotic stresses (Sakuraba et al., 2014b). Syntenic orthologs of SGR were highly expressed in both Lindernia species during dehydration/desiccation. STAY-GREEN LIKE (SGRL) proteins are negative regulators of chlorophyll degradation, and overexpression of SGRL2 leads to a stay-green phenotype (Sakuraba et al., 2014a). The syntenic ortholog of SGRL was highly induced during desiccation in L. brevidens, but the L. subracemosa ortholog had no detectable expression (Supplemental Table 10).

Carbohydrate metabolism is heavily shifted during desiccation, and sucrose, trehalose, and short-chain oligosaccharides function as osmoprotectants to stabilize cellular macromolecules. Sucrose is the most abundant carbohydrate in most resurrection plants, and the accumulation of sucrose distinguishes desiccation-sensitive and -tolerant Eragrostis species (Illing et al., 2005). Craterostigma and L. brevidens accumulate the unusual C8 sugar 2-octulose in photosynthetic tissues, which serves as a reservoir of sucrose accumulation during desiccation (Bianchi et al., 1991; Phillips et al., 2008). Transketolase7 and -10 catalyze the formation of octulose-8-phosphate in Craterostigma (Zhang et al., 2016), and the orthologous transketolase genes were highly expressed in L. brevidens leaf tissue (Supplemental Table 10). This included two pairs of syntenic 1:1 orthologs and a trio of retained 2:1 duplicates, with upregulation in well-watered and rehydrating tissue in L. brevidens and no or little expression in L. subracemosa.

LATE EMBRYOGENESIS ABUNDANT (LEA) proteins are predicted to have protective functions that are essential for desiccation tolerance (Hoekstra et al., 2001; Goyal et al., 2005; Hundertmark and Hincha, 2008). We identified 77 and 82 LEA protein-encoding genes in L. brevidens and L. subracemosa, respectively (Supplemental Table 11). Orthologs were assigned for 70 of these genes. About half of the identified LEAs were classified to the LEA_2 group, whereas the second largest LEA group was the LEA_4 group (14 and 12 genes in L. brevidens and L. subracemosa, respectively). Nine LEA genes in L. brevidens (one Dhn, five LEA_2, two LEA_4, and one LEA_5) were derived from gene duplication events, which suggests that these genes may have functions related to desiccation tolerance in L. brevidens. For example, LEA5-2 occurs in one copy in L. subracemosa (LsLEA5-2) and has two orthologs in L. brevidens (i.e., LbLEA5-2 and LbLEA5-3). The LsLEA5-2 gene showed negligible expression in L. subracemosa under control and dehydration conditions, whereas LbLEA5-2 was among the highest expressed LEA genes upon dehydration in L. brevidens (Supplemental Data Set ).

Almost one-third of L. brevidens LEA genes showed expression levels 30 times higher or more than L. subracemosa orthologs during late dehydration (10 and 14 d) (Supplemental Data Set). Most L. subracemosa orthologs had reduced or no expression in all surveyed time points, suggesting that there was a massive rewiring of expression networks (Supplemental Data Set ). For example, LEA1-3, LEA1-4, LEA2-14, LEA2-19, LEA4-1, LEA4-6, LEA4-7, and LEA5-2 showed very high expression in fully hydrated L. brevidens but no expression in fully hydrated L. subracemosa.

DISCUSSION

Genomic resources are abundant for resurrection plants, but the lack of suitable dehydration-sensitive outgroups has limited genomic insights into the origin and pathways controlling desiccation tolerance. Here, we leveraged a unique comparative system of closely related desiccation-tolerant and -sensitive species to identify gene- and pathway-level changes associated with the evolution of desiccation tolerance. This approach allowed us to distinguish dehydration pathways conserved in all plants from desiccation-specific processes observed only in resurrection plants.

L. brevidens and L. subracemosa have similar overall genome size and gene number, and most genes were likewise retained as singletons (1:1) or duplicates (2:2) after their shared WGD event. The genomes have no significant differences in architecture, rRNAs, repetitive element composition, or clustering of desiccation-related genes. These features were proposed previously to contribute to desiccation tolerance in other resurrection plant lineages (Xiao et al., 2015; Costa et al., 2017). Instead, our data indicate that desiccation tolerance in L. brevidens is driven by a complex cascade of cis-regulatory element-mediated pathway rewiring, tandem duplication, and preferential gene retention and neofunctionalization.

Gene expression patterns are dramatically divergent in dehydration and rehydration time-course data between L. brevidens and L. subracemosa. Only a few of the syntenic orthologs are similarly expressed in both species, and coexpression network modules are largely rewired. Early dehydration responses have surprisingly little overlap, suggesting that the gene expression program and signals leading to tolerance are already apparent upon mild dehydration. Gene expression is most dynamic between the day-3 and -7 time points, when the plants shift from moderate to severe dehydration stress. This likely reflects major shifts in leaf water potential, photosynthesis, oxidative stress, and cellular damage. Gene expression is stabilized after moderate drought and desiccation in L. brevidens, which reflects the successful deployment of protective mechanisms. By contrast, the dynamic and chaotic expression patterns in desiccating L. subracemosa may reflect last-ditch efforts to avoid imminent senescence.

Drought and seed development are linked by the common stress of water deficit. Vegetative and seed desiccation processes are strikingly similar, and overlapping pathways have been identified in resurrection plants (Costa et al., 2017; VanBuren et al., 2017). These include accumulation of osmoprotectants, expression of LEA proteins, and free radical scavenging systems as well as downregulation of photosynthesis and dismantling the photosynthetic apparatus. Drought responses and seed development are similarly regulated by ABA-related signaling, and both elicit comparable downstream responses (Nakashima and Yamaguchi-Shinozaki, 2013). Several important transcription factors involved in dehydration and seed-related processes are preferentially retained in L. brevidens compared with L. subracemosa, which may allow high-level pathway rewiring. Desiccation-related genes such as ELIPs or LEAs have increased in copy number in L. brevidens via tandem gene duplication. For genes encoding proteins with structural, enzymatic, or chaperone functions, tandem duplications may serve to increase their absolute abundance to surpass a threshold required for desiccation tolerance. The expansion of ELIPs has been observed in several resurrection plants, including C. plantagineum (Bartels et al., 1992), S. lepidophylla (VanBuren et al., 2018), and Boea hygrometrica (Xiao et al., 2015). ELIPs likely play an important role in protecting the photosynthetic apparatus and bind excess chlorophyll during prolonged desiccation (Alamillo and Bartels, 2001). The repeated duplication of ELIPs may be a hallmark of the convergent evolution of desiccation tolerance across land plants. Expression patterns also can be shifted by changes in cis-regulatory elements, as was observed previously in LEA genes from C. plantagineum and L. brevidens (van den Dries et al., 2011; Giarola et al., 2018). The enrichment of seed-related cis-regulatory elements in modules uniquely upregulated in L. brevidens is likely the result of novel cis-element acquisition in desiccation-related genes and activation of seed-related transcription factors.

Desiccation tolerance likely evolved from a complex, additive series of gene duplications and pathway rewiring rather than a simple master regulatory switch. Naturally drought-tolerant species could undergo favorable duplication of ELIPs or rewiring of LEA proteins to promote a quasi-desiccation-tolerant state. These responses could be refined further through the accumulation of additive features to surpass the threshold required for surviving anhydrobiosis. This step-wise hypothesis is supported by the continuum of desiccation tolerance, where the magnitude and duration of tolerance varies across species. Craterostigma can tolerate more rapid desiccation and recover more completely than L. brevidens, and older leaf tissue in L. brevidens is often desiccation sensitive.

The recovery rate in L. brevidens is related to environmental factors, including developmental stage, rate of drying, and dehydration priming. This comparatively weak desiccation tolerance may reflect relaxed selection in the drought-free rainforest habitat of L. brevidens. Desiccation tolerance is ancestral in the clade spanning L. brevidens and Craterostigma (Fischer et al., 2013), and some protective mechanisms were likely present in the shared ancestor with L. subracemosa. This also may explain the partial induction of seed and vegetative desiccation-associated pathways in L. subracemosa compared with the typical dehydration responses in other species. The trajectory from sensitive to desiccation tolerant is a complex, multistep process, and future work in intermediate or weakly desiccation-tolerant species will help uncover the origins of this trait.

METHODS

Growth Conditions and Sampling

Lindernia brevidens and Lindernia subracemosa were grown as described previously (Phillips et al., 2008). Voucher specimens have been deposited: L. brevidens Kenya, Taita Hills, E. Fischer 8022 (KOBL;=Herbarium Koblenz) and L. subracemosa Rwanda, Uwinka, E. Fischer 1350 (BG-Bonn 19990-2, KOBL). Plants were propagated via cuttings and maintained under day/night temperatures of 22 and 18°C, respectively, under fluorescent lighting with an intensity of 80 μE m−2 s−1 and a 16/8-h photoperiod. L. brevidens and L. subracemosa were grown in the same chamber to minimize environmental variance. For the desiccation and rehydration time courses, plants were allowed to gradually dry for a period of 30 d, with sampling in triplicate with three independent plants at D3, D7, D14, D21, and D30. Plants were rehydrated and sampled at 24 and 48 h post rehydration. Samples were always taken at the same time of the day, 6 h after the onset of light, to minimize effects associated with circadian oscillation. Leaf tissue for RNAseq was flash frozen in liquid nitrogen and stored at −80°C. RWC measurements were calculated using the equation RWC = [(FW − DW)/(SW − DW)], where FW, DW, and SW indicate fresh weight of the leaf tissue, dry weight, and saturated weight. Dry weight was obtained after drying tissue at 80°C for 48 h, and saturated plant weights were obtained after submerging leaf tissue in water for 24 h. Three replicates of RWC measurements were collected for each time point.

Nucleic Acid Extraction, Library Construction, and Sequencing

High molecular weight genomic DNA for PacBio and Illumina sequencing was isolated from young leaf tissue of growth chamber-grown L. brevidens and L. subracemosa. DNA was isolated using a modified nucleus preparation (Zhang et al., 1995) followed by phenol chloroform purification to remove residual contaminants. PacBio libraries were constructed and size selected for 25-kb fragments on the BluePippen system (Sage Science) followed by purification using AMPure XP beads (Beckman Coulter). Libraries were sequenced on a Sequel platform with V4 software and V2 chemistry. In total, 2,054,566 filtered subreads spanning 21.7 Gb were sequenced for L. brevidens and 1,615,065 reads spanning 17.9 Gb were sequenced for L. subracemosa. This represents 80.3× and 71.6× coverage for L. brevidens and L. subracemosa, respectively. Illumina DNAseq libraries for error correction were constructed using the KAPA HyperPrep Kit (Kapa Biosystems) following the manufacturer’s instructions. Libraries were sequenced on an Illumina HiSeq4000 device under paired-end 150-bp mode.

For RNAseq analysis, total RNA was extracted from 200 mg of ground L. brevidens and L. subracemosa leaf tissues using an Omega Bio-tek E.Z.N.A. Plant RNA Kit (Omega Bio-tek) according to the manufacturer’s instructions. RNA quality was validated using gel electrophoresis and Qubit RNA IQ Assay (ThermoFisher). Two micrograms of total RNA was used to construct Illumina TruSeq-stranded mRNA libraries following the manufacturer’s protocol (Illumina). Libraries were pooled and sequenced on the Illumina HiSeq4000 device under paired-end 150-bp mode. Three replicatfinnes were sequenced for each time point in each species.

Genome Assembly

Genome sizes for L. brevidens and L. subracemosa were estimated using flow cytometry as described previously (Arumuganathan and Earle, 1991). The flow cytometry-based estimates of 270 and 250 Mb for L. brevidens and L. subracemosa, respectively, were consistent with k-mer-based analysis using Illumina WGS data. Lindernia are mostly selfing, and unimodal k-mer distribution suggests low within-genome heterozygosity for both species. Raw PacBio reads were error corrected and assembled using Canu (V1.4; Koren et al., 2017). Based on previous experience (Edger et al., 2018; VanBuren et al., 2018), Canu produced the most contiguous and accurate assembly for homozygous, diploid species compared with other leading long-read assemblers. The following Canu parameters were modified and all others were left as default: minReadLength = 2500, GenomeSize = 270 Mb (or 250 Mb), minOverlapLength = 1000. Assembly graphs were visualized in Bandage (Wick et al., 2015). Draft contigs were polished with Pilon (V1.22; Walker et al., 2014) using 79× and 58× coverage of Illumina paired-end 150-bp data for L. brevidens and L. subracemosa, respectively. Illumina reads were quality trimmed using Trimmomatic (V0.33; Bolger et al., 2014) and aligned to the draft contigs using bowtie2 (V2.3.0; Langmead and Salzberg, 2012) with default parameters. Alignment rates in the first round of corrections were 96 and 97%, respectively, suggesting that both the L. brevidens and L. subracemosa assemblies were largely complete. Parameters for Pilon were modified as follows: –flank 7, –K 49, and –mindepth 20. Pilon was run recursively three times with minimal corrections in the third round, supporting accurate insertion/deletion correction.

Hi-C Library Construction and Analysis

The L. brevidens draft genome was anchored into a chromosome-scale assembly using a Hi-C proximity-based assembly approach. The Hi-C library was constructed using 0.2 g of young leaf tissue from well-watered L. brevidens plants with the Proximo Hi-C Plant kit (Phase Genomics) following the manufacturer’s protocol. The final library was size selected for 300 to 600 bp and sequenced on the Illumina NexSeq 500 device under paired-end 75-bp mode. In total, 178 million reads were used as input for the Juicer and 3d-DNA Hi-C analysis and scaffolding pipelines (Supplemental Table 1; Durand et al., 2016; Dudchenko et al., 2017). Quality-filtered reads were aligned to the PacBio contigs using bwa (V0.7.16; Li, 2013) with strict parameters (-n 0) to prevent mismatches and nonspecific alignments, and the resulting sam files were used as input into the Juicer pipeline. Read pairs were merged and duplicates or near duplicates were removed prior to constructing the distance matrix. Contigs were ordered and oriented and assembly errors were identified using the 3d-DNA pipeline with default parameters (Dudchenko et al., 2017). The resulting hic contact matrix was visualized using Juicebox, and misassemblies and misjoins were manually corrected based on neighboring interactions. This approach identified 14 high-confidence clusters representing the haploid chromosome number in L. brevidens. The manually validated assembly was used to build pseudomolecules using the finalize-output.sh script from 3d-DNA, and chromosomes were renamed and ordered by size.

Genome Annotation

Prior to genome annotation, LTR-RTs were predicted using LTR harvest (genome tools V1.5.8; Ellinghaus et al., 2008) and LTR finder (V1.07; Xu and Wang, 2007), and the LTR library was refined using LTR retriever (V1.8.0; Ou and Jiang, 2018). RTs were classified as intact if they were flanked by full-length LTRs. The insertion time for each intact element was calculated using LTR retriever with the formula T = K/2μ, where K is the divergence rate approximated by percentage identity and μ is the neutral mutation rate estimated as μ = 1 × 10−8 mutations per bp per year.

The filtered, nonredundant LTR library from LTR retriever was used as input for whole-genome annotation of LTR-RTs using RepeatMasker (http://www.repeatmasker.org/; Chen, 2004). The L. brevidens and L. subracemosa genomes were annotated using the MAKER-P pipeline (Campbell et al., 2014). Transcript-based evidence for gene predictions was produced using the desiccation/rehydration time-course RNAseq data. RNAseq reads were aligned to the L. brevidens and L. subracemosa genomes using the splice aware aligner STAR (V2.6; Dobin et al., 2013). Transcripts were identified using StringTie (V1.3.4; Pertea et al., 2015) with default parameters, and –merge flag was used to combine the output from individual libraries. The sets of nonredundant transcripts were used as EST evidence, and protein sequences from Arabidopsis (Arabidopsis thaliana; Lamesch et al., 2012) and UniprotKB plant databases (Boutet et al., 2007) were used as protein evidence. The custom LTR-RT library produced by LTR retriever and Repbase libraries were used for repeat masking. Ab initio gene prediction was done using SNAP (Korf, 2004) and Augustus (3.0.2; Stanke and Waack, 2003), with two rounds of iterative training for each species. The raw gene models were filtered to identify any residual repetitive elements using BLAST with a nonredundant transposase library. After filtering, a final set of 27,204 and 33,344 gene models was produced for L. brevidens and L. subracemosa, respectively. Annotation quality was assessed using BUSCO (V.2; Simão et al., 2015) with the plant-specific data set (embryophyta_odb9).

Comparative Genomics

Syntenic gene pairs within and between L. brevidens and L. subracemosa were identified using the MCScan toolkit (V1.1; Wang et al., 2012) implemented in python [https://github.com/tanghaibao/jcvi/wiki/MCscan-(Python-version)]. Gene models were aligned using LAST, and hits were filtered to find syntenic blocks. Tandem gene duplicates were identified using all-versus-all BLAST with a minimum e-value of 1e-5 and maximum gene distance of 10 genes. Macrosynteny and microsynteny plots and syntenic block depths were plotted using the python version of MCScan. Genes were classified as lineage specific if they had no syntenic orthologs between the two species or hits from LASTAL with greater than 70% nucleotide identity. The WGD event within Lindernia was identified using a combination of synteny and synonymous substitution rate estimation between duplicated gene pairs. Comparison of L. brevidens and L. subracemosa in MCScan identified a 2:2 syntenic pattern with 7742 and 8452 duplicated gene pairs retained, respectively. Duplicated regions span 70% of the L. brevidens genome and 72% of the L. subracemosa genome. Most duplicated regions were retained in large blocks, allowing chromosome pairs to be identified. Substitution rate was estimated using KaKs_calculator with the NG model (Zhang et al., 2006), and peaks of 0.65 in L. brevidens and 0.69 for L. subracemosa were identified, which indicates that the event is shared between both species.

RNAseq Analysis

Paired-end Illumina RNAseq reads were trimmed by quality score and by adapter contamination using Trimmomatic (V0.33; Bolger et al., 2014) with default parameters. The expression level of each gene was quantified using the pseudoaligner Kallisto (Bray et al., 2016) against the final gene models for L. brevidens and L. subracemosa. Parameters were left as default with 100 bootstraps per sample. Expression was quantified in TPM, and a mean across the three replicates was used for single-gene analysis and to construct log2 transformed expression-based heat maps. Pairwise differentially expressed genes were identified using sleuth (Pimentel et al., 2017) implemented in R.

Coexpression Network Construction

The time-course RNAseq data were clustered into gene coexpression networks using the R package WGCNA (Langfelder and Horvath, 2008). Genes with less than an average TPM of 5 across all seven time points were filtered prior to network construction. A signed coexpression network was constructed for each species using a soft-thresholding power of 8 and a tree cut height of 0.15. All remaining parameters were left as default. In total, 14,246 genes were clustered into 10 modules for L. brevidens and 14,075 genes were clustered into 9 modules for L. subracemosa.

Cis-Element Identification

Cis-regulatory elements were identified using the Hypergeometric Optimization of Motif EnRichment program (V4.10; Heinz et al., 2010) using cis-elements from 529 plant transcription factors (O’Malley et al., 2016). Cis-elements were identified in the 1-kb region upstream of the transcriptional start site when known or directly upstream of the start codon of each gene model. Promoters of gene models with detectable expression (TPM > 1) were used as background. Enrichment tests were performed using syntenic gene pairs with differential expression specific to L. brevidens or gene models unique to modules upregulated or downregulated during desiccation in L. brevidens. P < 0.00001 was used as the cutoff for identifying enriched motifs in any comparison.

Identification of LEA Genes

LEA genes were retrieved from L. brevidens and L. subracemosa transcriptomes by BLAST and HMMER (http://hmmer.org/). Arabidopsis and Craterostigma plantagineum LEA protein sequences were used for BLAST searches. HMM profiles for the eight LEA families (DHN, PF00257; LEA_1, PF03760; LEA_2, PF03168; LEA_3, PF03242; LEA_4, PF02987; LEA_5, PF00477; LEA_6, PF10714; and SMP, PF04927) obtained from the Pfam database (http://pfam.xfam.org; Finn et al., 2016) were used with the program hmmscan to search for LEA domain-containing proteins. Proteins identified with hmmscan were queried against the nr databank to confirm their classification as LEAs. Ortholog pairs were additionally confirmed by pairwise sequence alignments of the predicted protein sequences using EMBOSS Needle (https://www.ebi.ac.uk/Tools/psa/emboss_needle/).

Accession Numbers

The genome assemblies, raw PacBio data, Illumina DNAseq, and RNAseq data are available from the National Center for Biotechnology Information Short Read Archive. The RNAseq reads were deposited to the National Center for Biotechnology Information Short Read Archive under BioProject PRJNA488068. The genome assemblies for L. brevidens and L. subracemosa were deposited under BioProjects PRJNA489464 and PRJNA489465, respectively.

Supplemental Data

Acknowledgments

We thank Eberhard Fischer (University Koblenz Germany) for making Lindernia plants available originally. This work is supported by funding from the National Science Foundation (MCB‐1817347 to R.V.).

AUTHOR CONTRIBUTIONS

R.V. and D.B. designed and conceived research. R.V., C.M.W., and J.P. annotated genome features. X.S. collected desiccation and rehydration data. C.M.W. constructed RNAseq, DNAseq, and Hi-C libraries. R.V., C.M.W., J.P., X.S., S.A., V.G., and D.B. analyzed data. R.V. wrote the article. All authors read and approved the final article.

Footnotes

  • www.plantcell.org/cgi/doi/10.1105/tpc.18.00517

  • The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantcell.org) is: Robert VanBuren (bobvanburen{at}gmail.com).

  • ↵[OPEN] Articles can be viewed without a subscription.

  • Received July 6, 2018.
  • Revised October 2, 2018.
  • Accepted October 23, 2018.
  • Published October 25, 2018.

References

  1. ↵
    1. Alamillo, J.M.,
    2. Bartels, D.
    (2001). Effects of desiccation on photosynthesis pigments and the ELIP-like dsp 22 protein complexes in the resurrection plant Craterostigma plantagineum. Plant Sci. 160: 1161–1170. pmid:11337073
    OpenUrlCrossRefPubMed
  2. ↵
    1. Alonso, R.,
    2. Oñate-Sánchez, L.,
    3. Weltmeier, F.,
    4. Ehlert, A.,
    5. Diaz, I.,
    6. Dietrich, K.,
    7. Vicente-Carbajosa, J.,
    8. Dröge-Laser, W.
    (2009). A pivotal role of the basic leucine zipper transcription factor bZIP53 in the regulation of Arabidopsis seed maturation gene expression based on heterodimerization and protein complex formation. Plant Cell 21: 1747–1761.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Arumuganathan, K.,
    2. Earle, E.
    (1991). Estimation of nuclear DNA content of plants by flow cytometry. Plant Mol. Biol. Rep. 9: 229–241.
    OpenUrlCrossRef
  4. ↵
    1. Baniaga, A.E.,
    2. Arrigo, N.,
    3. Barker, M.S.
    (2016). The small nuclear genomes of Selaginella are associated with a low rate of genome size evolution. Genome Biol. Evol. 8: 1516–1525.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Banks, J.A.,
    2. Nishiyama, T.,
    3. Hasebe, M.,
    4. Bowman, J.L.,
    5. Gribskov, M.,
    6. Albert, V.A.,
    7. Aono, N.,
    8. Aoyama, T.,
    9. Ambrose, B.A.,
    10. Ashton, N.W.
    (2011). The Selaginella genome identifies genetic changes associated with the evolution of vascular plants. Science332: 960–963.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Bartels, D.,
    2. Salamini, F.
    (2001). Desiccation tolerance in the resurrection plant Craterostigma plantagineum: A contribution to the study of drought tolerance at the molecular level. Plant Physiol. 127: 1346–1353.
    OpenUrlFREE Full Text
  7. ↵
    1. Bartels, D.,
    2. Hanke, C.,
    3. Schneider, K.,
    4. Michel, D.,
    5. Salamini, F.
    (1992). A desiccation-related Elip-like gene from the resurrection plant Craterostigma plantagineum is regulated by light and ABA. EMBO J. 11: 2771–2778.
    OpenUrlPubMed
  8. ↵
    1. Bianchi, G.,
    2. Gamba, A.,
    3. Murelli, C.,
    4. Salamini, F.,
    5. Bartels, D.
    (1991). Novel carbohydrate metabolism in the resurrection plant Craterostigma plantagineum. Plant J. 1: 355–359.
    OpenUrlCrossRef
  9. ↵
    1. Bolger, A.M.,
    2. Lohse, M.,
    3. Usadel, B.
    (2014). Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30: 2114–2120.
    OpenUrlCrossRefPubMed
  10. ↵
    1. D Edwards
    1. Boutet, E.,
    2. Lieberherr, D.,
    3. Tognolli, M.,
    4. Schneider, M.,
    5. Bairoch, A.
    (2007). Uniprotkb/swiss-prot. In Plant Bioinformatics. Methods in Molecular Biology, D Edwards, ed (New York City, NY: Humana Press), pp. 89–112.
  11. ↵
    1. Bray, N.L.,
    2. Pimentel, H.,
    3. Melsted, P.,
    4. Pachter, L.
    (2016). Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34: 525–527.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Campbell, M.S.,
    2. Law, M.,
    3. Holt, C.,
    4. Stein, J.C.,
    5. Moghe, G.D.,
    6. Hufnagel, D.E.,
    7. Lei, J.,
    8. Achawanantakun, R.,
    9. Jiao, D.,
    10. Lawrence, C.J.,
    11. Ware, D.,
    12. Shiu, S.H., et al
    . (2014). MAKER-P:A tool kit for the rapid creation, management, and quality control of plant genome annotations. Plant Physiol. 164: 513–524.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Cannon, S.B.,
    2. Mitra, A.,
    3. Baumgarten, A.,
    4. Young, N.D.,
    5. May, G.
    (2004). The roles of segmental and tandem gene duplication in the evolution of large gene families in Arabidopsis thaliana. BMC Plant Biol. 4: 10.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Chen, N.
    (2004). Using Repeat Masker to identify repetitive elements in genomic sequences. Current Protocols in Bioinformatics 5: 4.10.1–4.10.14.
    OpenUrl
  15. ↵
    1. Costa, M.D.,
    2. Artur, M.A.,
    3. Maia, J.,
    4. Jonkheer, E.,
    5. Derks, M.F.,
    6. Nijveen, H.,
    7. Williams, B.,
    8. Mundree, S.G.,
    9. Jiménez-Gómez, J.M.,
    10. Hesselink, T.,
    11. Schijlen, E.G.,
    12. Ligterink, W., et al
    . (2017). A footprint of desiccation tolerance in the genome of Xerophyta viscosa. Nat. Plants 3: 17038.
    OpenUrl
  16. ↵
    1. Dekkers, B.J.,
    2. He, H.,
    3. Hanson, J.,
    4. Willems, L.A.,
    5. Jamar, D.C.,
    6. Cueff, G.,
    7. Rajjou, L.,
    8. Hilhorst, H.W.,
    9. Bentsink, L.
    (2016). The Arabidopsis DELAY OF GERMINATION 1 gene affects ABSCISIC ACID INSENSITIVE 5 (ABI5) expression and genetically interacts with ABI3 during Arabidopsis seed development. Plant J. 85: 451–465.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Dobin, A.,
    2. Davis, C.A.,
    3. Schlesinger, F.,
    4. Drenkow, J.,
    5. Zaleski, C.,
    6. Jha, S.,
    7. Batut, P.,
    8. Chaisson, M.,
    9. Gingeras, T.R.
    (2013). STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29: 15–21.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Dudchenko, O.,
    2. Batra, S.S.,
    3. Omer, A.D.,
    4. Nyquist, S.K.,
    5. Hoeger, M.,
    6. Durand, N.C.,
    7. Shamim, M.S.,
    8. Machol, I.,
    9. Lander, E.S.,
    10. Aiden, A.P.,
    11. Aiden, E.L.
    (2017). De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds. Science 356: 92–95.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Durand, N.C.,
    2. Shamim, M.S.,
    3. Machol, I.,
    4. Rao, S.S.,
    5. Huntley, M.H.,
    6. Lander, E.S.,
    7. Aiden, E.L.
    (2016). Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst. 3: 95–98.
    OpenUrl
  20. ↵
    1. Edger, P.P.,
    2. VanBuren, R.,
    3. Colle, M.,
    4. Poorten, T.J.,
    5. Wai, C.M.,
    6. Niederhuth, C.E.,
    7. Alger, E.I.,
    8. Ou, S.,
    9. Acharya, C.B.,
    10. Wang, J.
    (2018). Single-molecule sequencing and optical mapping yields an improved genome of woodland strawberry (Fragaria vesca) with chromosome-scale contiguity. Gigascience 7: 1–7.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Ellinghaus, D.,
    2. Kurtz, S.,
    3. Willhoeft, U.
    (2008). LTRharvest, an efficient and flexible software for de novo detection of LTR retrotransposons. BMC Bioinformatics 9: 18.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Finn, R.D.,
    2. Coggill, P.,
    3. Eberhardt, R.Y.,
    4. Eddy, S.R.,
    5. Mistry, J.,
    6. Mitchell, A.L.,
    7. Potter, S.C.,
    8. Punta, M.,
    9. Qureshi, M.,
    10. Sangrador-Vegas, A.,
    11. Salazar, G.A.,
    12. Tate, J., et al
    . (2016). The Pfam protein families database: Towards a more sustainable future. Nucleic Acids Res. 44: D279–D285.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Fischer, E.,
    2. Schäferhoff, B.,
    3. Müller, K.
    (2013). The phylogeny of Linderniaceae: The new genus Linderniella, and new combinations within Bonnaya, Craterostigma, Lindernia, Micranthemum, Torenia and Vandellia. Willdenowia 43: 209–238.
    OpenUrl
  24. ↵
    1. Giarola, V.,
    2. Jung, N.U.,
    3. Singh, A.,
    4. Satpathy, P.,
    5. Bartels, D.
    (2018). Analysis of pcC13-62 promoters predicts a link between cis-element variations and desiccation tolerance in Linderniaceae. J. Exp. Bot. 69: 3773–3784.
    OpenUrl
  25. ↵
    1. Goyal, K.,
    2. Walton, L.J.,
    3. Tunnacliffe, A.
    (2005). LEA proteins prevent protein aggregation due to water stress. Biochem. J. 388: 151–157.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Hayami, N.,
    2. Sakai, Y.,
    3. Kimura, M.,
    4. Saito, T.,
    5. Tokizawa, M.,
    6. Iuchi, S.,
    7. Kurihara, Y.,
    8. Matsui, M.,
    9. Nomoto, M.,
    10. Tada, Y.,
    11. Yamamoto, Y.Y.
    (2015). The responses of Arabidopsis early light-induced protein2 to ultraviolet B, high light, and cold stress are regulated by a transcriptional regulatory unit composed of two elements. Plant Physiol. 169: 840–855.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Heinz, S.,
    2. Benner, C.,
    3. Spann, N.,
    4. Bertolino, E.,
    5. Lin, Y.C.,
    6. Laslo, P.,
    7. Cheng, J.X.,
    8. Murre, C.,
    9. Singh, H.,
    10. Glass, C.K.
    (2010). Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38: 576–589.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Hoekstra, F.A.,
    2. Golovina, E.A.,
    3. Buitink, J.
    (2001). Mechanisms of plant desiccation tolerance. Trends Plant Sci. 6: 431–438.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Huang, Y.-C.,
    2. Niu, C.-Y.,
    3. Yang, C.-R.,
    4. Jinn, T.-L.
    (2016). The heat stress factor HSFA6b connects ABA signaling and ABA-mediated heat responses. Plant Physiol. 172: 1182–1199.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Hundertmark, M.,
    2. Hincha, D.K.
    (2008). LEA (late embryogenesis abundant) proteins and their encoding genes in Arabidopsis thaliana. BMC Genomics 9: 118.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Illing, N.,
    2. Denby, K.J.,
    3. Collett, H.,
    4. Shen, A.,
    5. Farrant, J.M.
    (2005). The signature of seeds in resurrection plants: A molecular and physiological comparison of desiccation tolerance in seeds and vegetative tissues. Integr. Comp. Biol. 45: 771–787.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Koren, S.,
    2. Walenz, B.P.,
    3. Berlin, K.,
    4. Miller, J.R.,
    5. Bergman, N.H.,
    6. Phillippy, A.M.
    (2017). Canu: Scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation. Genome Res. 27: 722–736.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Korf, I.
    (2004). Gene finding in novel genomes. BMC Bioinformatics 5: 59.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Lamesch, P.,
    2. Berardini, T.Z.,
    3. Li, D.,
    4. Swarbreck, D.,
    5. Wilks, C.,
    6. Sasidharan, R.,
    7. Muller, R.,
    8. Dreher, K.,
    9. Alexander, D.L.,
    10. Garcia-Hernandez, M.,
    11. Karthikeyan, A.S.,
    12. Lee, C.H., et al
    . (2012). The Arabidopsis Information Resource (TAIR): Improved gene annotation and new tools. Nucleic Acids Res. 40: D1202–D1210.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Langfelder, P.,
    2. Horvath, S.
    (2008). WGCNA: An R package for weighted correlation network analysis. BMC Bioinformatics 9: 559.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Langmead, B.,
    2. Salzberg, S.L.
    (2012). Fast gapped-read alignment with Bowtie 2. Nat. Methods 9: 357–359.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Li, H.
    (2013). Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv 1303.3997.
  38. ↵
    1. Lopez-Molina, L.,
    2. Mongrand, S.,
    3. Chua, N.-H.
    (2001). A postgermination developmental arrest checkpoint is mediated by abscisic acid and requires the ABI5 transcription factor in Arabidopsis. Proc. Natl. Acad. Sci. USA 98: 4782–4787.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Lüttge, U.,
    2. Beck, E.,
    3. Bartels, D.
    (2011). Plant Desiccation Tolerance. (Berlin, Germany: Springer Science & Business Media).
  40. ↵
    1. Nakashima, K.,
    2. Yamaguchi-Shinozaki, K.
    (2013). ABA signaling in stress-response and seed development. Plant Cell Rep. 32: 959–970.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Nakashima, K.,
    2. Fujita, Y.,
    3. Kanamori, N.,
    4. Katagiri, T.,
    5. Umezawa, T.,
    6. Kidokoro, S.,
    7. Maruyama, K.,
    8. Yoshida, T.,
    9. Ishiyama, K.,
    10. Kobayashi, M.,
    11. Shinozaki, K.,
    12. Yamaguchi-Shinozaki, K.
    (2009). Three Arabidopsis SnRK2 protein kinases, SRK2D/SnRK2.2, SRK2E/SnRK2.6/OST1 and SRK2I/SnRK2.3, involved in ABA signaling are essential for the control of seed development and dormancy. Plant Cell Physiol. 50: 1345–1363.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Nguyen, T.-P.,
    2. Cueff, G.,
    3. Hegedus, D.D.,
    4. Rajjou, L.,
    5. Bentsink, L.
    (2015). A role for seed storage proteins in Arabidopsis seed longevity. J. Exp. Bot. 66: 6399–6413.
    OpenUrlCrossRefPubMed
  43. ↵
    1. O’Malley, R.C.,
    2. Huang, S.-C.,
    3. Song, L.,
    4. Lewsey, M.G.,
    5. Bartlett, A.,
    6. Nery, J.R.,
    7. Galli, M.,
    8. Gallavotti, A.,
    9. Ecker, J.R.
    (2016). Cistrome and epicistrome features shape the regulatory DNA landscape. Cell 165: 1280–1292.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Oliver, M.J.,
    2. Tuba, Z.,
    3. Mishler, B.D.
    (2000). The evolution of vegetative desiccation tolerance in land plants. Plant Ecol. 151: 85–100.
    OpenUrlCrossRef
  45. ↵
    1. Oliver, M.J.,
    2. Guo, L.,
    3. Alexander, D.C.,
    4. Ryals, J.A.,
    5. Wone, B.W.,
    6. Cushman, J.C.
    (2011). A sister group contrast using untargeted global metabolomic analysis delineates the biochemical regulation underlying desiccation tolerance in Sporobolus stapfianus. Plant Cell 23: 1231–1248.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Ou, S.,
    2. Jiang, N.
    (2018). LTR_retriever: A highly accurate and sensitive program for identification of long terminal repeat retrotransposons. Plant Physiol. 176: 1410–1422.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Pertea, M.,
    2. Pertea, G.M.,
    3. Antonescu, C.M.,
    4. Chang, T.-C.,
    5. Mendell, J.T.,
    6. Salzberg, S.L.
    (2015). StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat. Biotechnol. 33: 290–295.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Phillips, J.R.,
    2. Fischer, E.,
    3. Baron, M.,
    4. van den Dries, N.,
    5. Facchinelli, F.,
    6. Kutzer, M.,
    7. Rahmanzadeh, R.,
    8. Remus, D.,
    9. Bartels, D.
    (2008). Lindernia brevidens: A novel desiccation-tolerant vascular plant, endemic to ancient tropical rainforests. Plant J. 54: 938–948.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Pimentel, H.,
    2. Bray, N.L.,
    3. Puente, S.,
    4. Melsted, P.,
    5. Pachter, L.
    (2017). Differential analysis of RNA-seq incorporating quantification uncertainty. Nat. Methods 14: 687–690.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Plancot, B.,
    2. Vanier, G.,
    3. Maire, F.,
    4. Bardor, M.,
    5. Lerouge, P.,
    6. Farrant, J.M.,
    7. Moore, J.,
    8. Driouich, A.,
    9. Vicré-Gibouin, M.,
    10. Afonso, C.,
    11. Loutelier-Bourhis, C.
    (2014). Structural characterization of arabinoxylans from two African plant species Eragrostis nindensis and Eragrostis tef using various mass spectrometric methods. Rapid Commun. Mass Spectrom. 28: 908–916.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Poxleitner, M.,
    2. Rogers, S.W.,
    3. Lacey Samuels, A.,
    4. Browse, J.,
    5. Rogers, J.C.
    (2006). A role for caleosin in degradation of oil-body storage lipid during seed germination. Plant J. 47: 917–933.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Proctor, M.
    (1990). The physiological basis of bryophyte production. Bot. J. Linn. Soc. 104: 61–77.
    OpenUrlCrossRef
  53. ↵
    1. Rahmanzadeh, R.,
    2. Müller, K.,
    3. Fischer, E.,
    4. Bartels, D.,
    5. Borsch, T.
    (2005). The Linderniaceae and Gratiolaceae are further lineages distinct from the Scrophulariaceae (Lamiales). Plant Biol (Stuttg) 7: 67–78.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Sakuraba, Y.,
    2. Kim, D.,
    3. Kim, Y.-S.,
    4. Hörtensteiner, S.,
    5. Paek, N.-C.
    (2014a). Arabidopsis STAYGREEN-LIKE (SGRL) promotes abiotic stress-induced leaf yellowing during vegetative growth. FEBS Lett. 588: 3830–3837.
    OpenUrl
  55. ↵
    1. Sakuraba, Y.,
    2. Lee, S.-H.,
    3. Kim, Y.-S.,
    4. Park, O.K.,
    5. Hörtensteiner, S.,
    6. Paek, N.-C.
    (2014b). Delayed degradation of chlorophylls and photosynthetic proteins in Arabidopsis autophagy mutants during stress-induced leaf yellowing. J. Exp. Bot. 65: 3915–3925.
    OpenUrlCrossRefPubMed
  56. ↵
    1. Shen, Y.,
    2. Xie, J.,
    3. Liu, R.D.,
    4. Ni, X.F.,
    5. Wang, X.H.,
    6. Li, Z.X.,
    7. Zhang, M.
    (2014). Genomic analysis and expression investigation of caleosin gene family in Arabidopsis. Biochem. Biophys. Res. Commun. 448: 365–371.
    OpenUrl
  57. ↵
    1. Shimada, T.L.,
    2. Shimada, T.,
    3. Takahashi, H.,
    4. Fukao, Y.,
    5. Hara-Nishimura, I.
    (2008). A novel role for oleosins in freezing tolerance of oilseeds in Arabidopsis thaliana. Plant J. 55: 798–809.
    OpenUrlCrossRefPubMed
  58. ↵
    1. Simão, F.A.,
    2. Waterhouse, R.M.,
    3. Ioannidis, P.,
    4. Kriventseva, E.V.,
    5. Zdobnov, E.M.
    (2015). BUSCO: Assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics 31: 3210–3212.
    OpenUrlCrossRefPubMed
  59. ↵
    1. Stanke, M.,
    2. Waack, S.
    (2003). Gene prediction with a hidden Markov model and a new intron submodel. Bioinformatics 19 (suppl. 2): ii215–ii225.
    OpenUrlCrossRefPubMed
  60. ↵
    1. VanBuren, R.,
    2. Bryant, D.,
    3. Edger, P.P.,
    4. Tang, H.,
    5. Burgess, D.,
    6. Challabathula, D.,
    7. Spittle, K.,
    8. Hall, R.,
    9. Gu, J.,
    10. Lyons, E.,
    11. Freeling, M.,
    12. Bartels, D., et al
    . (2015). Single-molecule sequencing of the desiccation-tolerant grass Oropetium thomaeum. Nature 527: 508–511.
    OpenUrlCrossRefPubMed
  61. ↵
    1. VanBuren, R.,
    2. Wai, C.M.,
    3. Zhang, Q.,
    4. Song, X.,
    5. Edger, P.P.,
    6. Bryant, D.,
    7. Michael, T.P.,
    8. Mockler, T.C.,
    9. Bartels, D.
    (2017). Seed desiccation mechanisms co-opted for vegetative desiccation in the resurrection grass Oropetium thomaeum. Plant Cell Environ. 40: 2292–2306.
    OpenUrl
  62. ↵
    1. VanBuren, R.,
    2. Wai, C.M.,
    3. Ou, S.,
    4. Pardo, J.,
    5. Bryant, D.,
    6. Jiang, N.,
    7. Mockler, T.C.,
    8. Edger, P.,
    9. Michael, T.P.
    (2018). Extreme haplotype variation in the desiccation-tolerant clubmoss Selaginella lepidophylla. Nat. Commun. 9: 13.
    OpenUrl
  63. ↵
    1. van den Dries, N.,
    2. Facchinelli, F.,
    3. Giarola, V.,
    4. Phillips, J.R.,
    5. Bartels, D.
    (2011). Comparative analysis of LEA-like 11-24 gene expression and regulation in related plant species within the Linderniaceae that differ in desiccation tolerance. New Phytol. 190: 75–88.
    OpenUrlCrossRef
  64. ↵
    1. Vander Willigen, C.,
    2. Pammenter, N.,
    3. Mundree, S.,
    4. Farrant, J.
    (2001). Some physiological comparisons between the resurrection grass, Eragrostis nindensis, and the related desiccation-sensitive species, E. curvula. Plant Growth Regul. 35: 121–129.
    OpenUrl
  65. ↵
    1. Walker, B.J.,
    2. Abeel, T.,
    3. Shea, T.,
    4. Priest, M.,
    5. Abouelliel, A.,
    6. Sakthikumar, S.,
    7. Cuomo, C.A.,
    8. Zeng, Q.,
    9. Wortman, J.,
    10. Young, S.K.,
    11. Earl, A.M.
    (2014). Pilon: An integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS ONE 9: e112963.
    OpenUrlCrossRefPubMed
  66. ↵
    1. Wang, Y.,
    2. Tang, H.,
    3. Debarry, J.D.,
    4. Tan, X.,
    5. Li, J.,
    6. Wang, X.,
    7. Lee, T.H.,
    8. Jin, H.,
    9. Marler, B.,
    10. Guo, H.,
    11. Kissinger, J.C.,
    12. Paterson, A.H.
    (2012). MCScanX: A toolkit for detection and evolutionary analysis of gene synteny and collinearity. Nucleic Acids Res. 40: e49.
    OpenUrlCrossRefPubMed
  67. ↵
    1. Wick, R.R.,
    2. Schultz, M.B.,
    3. Zobel, J.,
    4. Holt, K.E.
    (2015). Bandage: Interactive visualization of de novo genome assemblies. Bioinformatics 31: 3350–3352.
    OpenUrlCrossRefPubMed
  68. ↵
    1. Xiao, L.,
    2. Yang, G.,
    3. Zhang, L.,
    4. Yang, X.,
    5. Zhao, S.,
    6. Ji, Z.,
    7. Zhou, Q.,
    8. Hu, M.,
    9. Wang, Y.,
    10. Chen, M.,
    11. Xu, Y.,
    12. Jin, H., et al
    . (2015). The resurrection genome of Boea hygrometrica: A blueprint for survival of dehydration. Proc. Natl. Acad. Sci. USA 112: 5833–5837.
    OpenUrlAbstract/FREE Full Text
  69. ↵
    1. Xu, Z.,
    2. Wang, H.
    (2007). LTR_FINDER: An efficient tool for the prediction of full-length LTR retrotransposons. Nucleic Acids Res. 35: W265–W268.
    OpenUrlCrossRefPubMed
  70. ↵
    1. Yobi, A.,
    2. Wone, B.W.,
    3. Xu, W.,
    4. Alexander, D.C.,
    5. Guo, L.,
    6. Ryals, J.A.,
    7. Oliver, M.J.,
    8. Cushman, J.C.
    (2013). Metabolomic profiling in Selaginella lepidophylla at various hydration states provides new insights into the mechanistic basis of desiccation tolerance. Mol. Plant 6: 369–385.
    OpenUrlCrossRefPubMed
  71. ↵
    1. Yoshida, T.,
    2. Fujita, Y.,
    3. Maruyama, K.,
    4. Mogami, J.,
    5. Todaka, D.,
    6. Shinozaki, K.,
    7. Yamaguchi-Shinozaki, K.
    (2015). Four Arabidopsis AREB/ABF transcription factors function predominantly in gene expression downstream of SnRK2 kinases in abscisic acid signalling in response to osmotic stress. Plant Cell Environ. 38: 35–49.
    OpenUrlCrossRef
  72. ↵
    1. Zhang, Q.,
    2. Bartels, D.
    (2018). Molecular responses to dehydration and desiccation in desiccation-tolerant angiosperm plants. J. Exp. Bot. 69: 3211–3222.
    OpenUrl
  73. ↵
    1. Zhang, H.B.,
    2. Zhao, X.,
    3. Ding, X.,
    4. Paterson, A.H.,
    5. Wing, R.A.
    (1995). Preparation of megabase‐size DNA from plant nuclei. Plant J. 7: 175–184.
    OpenUrlCrossRef
  74. ↵
    1. Zhang, Q.,
    2. Linnemann, T.V.,
    3. Schreiber, L.,
    4. Bartels, D.
    (2016). The role of transketolase and octulose in the resurrection plant Craterostigma plantagineum. J. Exp. Bot. 67: 3551–3559.
    OpenUrlCrossRefPubMed
  75. ↵
    1. Zhang, Z.,
    2. Li, J.,
    3. Zhao, X.-Q.,
    4. Wang, J.,
    5. Wong, G.K.-S.,
    6. Yu, J.
    (2006). KaKs_Calculator: Calculating Ka and Ks through model selection and model averaging. Genomics Proteomics Bioinformatics 4: 259–263.
    OpenUrlCrossRefPubMed
View Abstract
Back to top

Table of Contents

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Request Permissions
Desiccation Tolerance Evolved through Gene Duplication and Network Rewiring in Lindernia
Robert VanBuren, Ching Man Wai, Jeremy Pardo, Valentino Giarola, Stefano Ambrosini, Xiaomin Song, Dorothea Bartels
The Plant Cell Dec 2018, 30 (12) 2943-2958; DOI: 10.1105/tpc.18.00517
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Extras

Jump to section

In this issue

View this article with LENS