View all hidden authors and organizations

Edited by Philip N. Benfey, Duke University, Durham, North Carolina, approved on July 13, 2021 (review received on January 25, 2021)

To date, the potential to utilize root traits in plant breeding has remained largely unexploited. In this study, we cloned and characterized the ENHANCED GRAVITROPISM2 (EGT2) gene of barley, which encodes a protein containing the STERILE ALPHA MOTIF domain. We proved that EGT2 is a key gene for the regulation of root growth angle in response to gravity. It is conserved in barley and wheat and may be a promising target for cereal crop improvement.

The root growth angle defines how the root grows toward the gravity vector and is one of the most important determinants of root structure. It controls water absorption capacity, nutrient use efficiency, and resistance to stress, and therefore controls crop yields. We demonstrated that, compared with wild-type plants, the barley egt2 (enhanced heaviness 2) mutant exhibited steeper root and lateral root growth, and a higher responsiveness to gravity independent of auxin. We cloned the EGT2 gene through a combination of batch separation analysis and whole-genome sequencing. Subsequent verification experiments performed by independent CRISPR/Cas9 mutant alleles showed that egt2 encodes a protein containing the STERILE ALPHA MOTIF domain. In situ hybridization experiments showed that EGT2 was expressed from root cap to elongation zone. By knocking out EGT2 orthologs in the genomes of tetraploid durum wheat A and B, we demonstrated the evolutionary conservation role of EGT2 in the control of the growth angle of barley and wheat roots. By combining laser capture microdissection with RNA sequencing, we observed that seven expansin genes were transcriptionally down-regulated in the elongation region. This is consistent with the role of EGT2 in this root region, where the effect of gravity sensing is performed by differential cell elongation. Our results indicate that EGT2 is an evolutionary conservative regulator of the growth angle of barley and wheat root systems, and may be a valuable target for root-based crop improvement strategies in cereals.

Population growth and climate change are the main challenges facing food security (1, 2). Many studies propose to modify the root structure to improve water and nutrient use efficiency, crop yield, and resilience to stress events (3, 4). The most important determinant of root structure is the root growth angle, that is, the angle at which the roots grow toward the ground.

The response to gravity or supergravity increases, so the root growth angle is steeper, indicating that it is related to the improvement of rice drought resistance, which may be achieved by increasing the access to deep soil water (5). At the same time, deeper root systems help to absorb more abundant nitrogen and other mobile nutrients in deeper soil layers (6). Root geotropism is regulated by sensing geotropism stimulation and subsequent differential cell elongation, so that the root geotropism carrier can grow. Removal of root caps mechanically or genetically will significantly reduce the response to gravity (7⇓-9), which indicates that gravity induction mainly occurs in the root caps. However, there is evidence that there is a sensing site outside the root cap, located in the elongation zone (10, 11). There are different hypotheses about how cells perceive gravity. The prevailing view is that starch-containing bodies in root caps act as static rocks and settle in response to gravity. In doing so, they trigger a signal cascade on the surface of organelles (12⇓-14) through mechanically sensitive channels or through direct protein interactions. This signaling pathway ultimately leads to the rearrangement of the auxin export vector, which leads to the recombination of the maximum auxin in the root tip (15). At the same time, the pH value in the root cap changes, and the pH value of the root meristem and the upper and lower sides of the elongation zone changes asymmetrically (16, 17). This ultimately leads to an increase in cell elongation on the side avoiding the gravity vector in the elongation zone of the root, thereby causing the root to grow downward (18). So far, only a single component of the signal cascade that regulates root gravity has been unraveled. Examples include the actin binding protein RICE MORPHOLOGY DETERMINANT, which is located on the surface of the static stone in the root cap cells of rice and controls the angle of root growth based on external phosphate (19). Another protein related to geotropism is ALTERED RESPONSE TO GRAVITY1, which is localized in the middle membrane of Arabidopsis thaliana, which is expressed in the root cap and participates in the gravity-induced lateral auxin gradient (20). Both proteins seem to act as signals immediately after root cap gravity induction. In contrast, rice DEEPER ROOTING1 (DRO1) acts as an early auxin response gene in the late stage of heavy signaling. The DRO1 gene encodes a plasma membrane protein expressed in root meristems and was identified because of its effect on the angle of root growth (5). The role of DRO1 may not be conserved in primary roots of different plant species, because Arabidopsis homologues do not affect the ground response of primary roots, but affect the growth angle of lateral roots (21).

Barley (Hordeum vulgare) is the fourth largest food crop in the world, after wheat (Triticum aestivum), rice (Oryza sativa) and corn (Zea mays) (2019, http://www.fao.org) /faostat/en /). It is grown in a wide geographical area because it can adapt to a wide range of climatic conditions, so it is an excellent model for studying the response to climate change (22). In this study, we used forward genetics to clone EGT2, a gene involved in the barley root-reaction response, and its role is conserved in wheat. EGT2 encodes a protein containing the STERILE ALPHA MOTIF (SAM) domain and may play a role in a regulatory pathway that counteracts the auxin-mediated forward gravity signaling pathway.

The egt2-1 mutant was found in the sodium azide mutagenized population of the barley variety Morex, which is based on the supergravity growth of its fine root system in paper rolls, and was shown as a single-gene recessive Mendelian locus (23 , twenty four). We studied the phenotype in more detail in a two-dimensional root box, where plants grow vertically on flat filter paper. In the Morex wild type, the fine roots grow at a shallow angle toward the gravity vector and cover a larger area, while the fine roots in the egt2-1 mutant grow sharply downwards (Figure 1A and D and SI Appendix, Figure S1A). This phenotype is consistent in soil-filled root canals and plants grown in pots, the latter shown by MRI (Figure 1 B and C and SI appendix, Figure S1 EG). In addition, the lateral roots from the seminal roots also showed increased growth angles (Figure 1 B and E and SI appendix, Figure S1 A and H). Apart from the increased root growth angle, we did not detect any other abnormal root phenotypes, neither the number of altered fine roots nor the difference in root length (SI appendix, Figure S1 B and C). To further investigate the reasons for the steep root phenotype, we tested the response of the root system to gravity. After rotating by 90°, we monitored the angle of the root tip over time (Figure 1 F and G). The roots of the egt2-1 mutant bend faster and stronger than the wild-type roots, approaching 90° after 3 days, while the wild-type roots bend only 30° (Figure 1G). However, the root growth rate did not change (SI appendix, Figure S1D). Therefore, we conclude that the steep root angle of the egt2-1 mutant may be caused by the higher response to gravity. Since gravity sensing and signal transduction occur in the root cap and meristem (16⇓ –18), we compared the root cap and meristem and measured the size of the root meristem through a microscope, but we did not find a significant difference (SI Appendix, Figure S2 AD). Other mutants with disrupted root gravity showed a different rate of sedimentation of starch granules in root cap cells than the wild type (19). However, we did not detect this difference between wild-type and egt2-1 mutants (SI appendix, Figure S2 E and F).

The root phenotype of egt2-1. (A) Wild type and egt2-1 roots grown on germination paper, 7 days after germination (DAG). (Scale bar: 2 cm.) (B) Wild type and egt2-1 roots grown in rhizotrons 26 DAG. (Scale bar: 10 cm.) (C) MRI pictures of wild-type and egt2-1 plants grown in soil 3 DAG. (Scale bar: 4 cm.) (D) Root angle of 7 DAG of fine root; in an experiment, each genotype n = 40; two-tailed t test, **P <0.01. (E) Lateral root angle 14 DAG; in two independent experiments, n = 8 to 9 for each genotype; two-tailed t test, **P <0.01. (F) Wild type and egt2-1 roots rotated at the specified time point (time point 0). (Scale bar: 1 cm.) (G) Root tip angle after rotation; Rotate the plant 5 DAG by 90° (time 0), and measure the root tip angle over time; In three independent experiments, each genotype n = 38; The two genotypes were compared by a two-tailed t test at their respective time points, **P <0.01. The SD is depicted; in order to illustrate the different starting angles of the root, all measured values ​​are normalized to the starting angle of the root at time 0.

It has previously been shown that the plant hormone auxin is involved in ground response signal transduction (15) and the external supply of auxin transport inhibitors or auxin affects the root response to rotation (25). In order to analyze whether the egt2-1 mutant is sensitive to manipulation of auxin status in roots, we treated wild-type and mutants with auxin and auxin transport inhibitors, and recorded the 90° rotation and root extension in the time course experiment. Long response. Compared with the simulated treatment, in the wild-type and egt2-1 mutants, the application of the naturally occurring auxin indole-3-acetic acid at low concentrations resulted in a similar weight response and root elongation (SI appendix, Fig. S3 AD) and inhibit root growth, which results in a slower response to gravity stimulation at higher concentrations (SI appendix, Figure S3 E and F). Compared with the simulated treatment, treatment with a low concentration of auxin transport inhibitor 1-N-naphthylphthalic acid (NPA) produced similar gravitational response and root extension in wild-type and egt2-1 within 48 hours. Compared with the simulation treatment, high concentration of NPA significantly reduced the weight response and root elongation of wild-type and egt2-1, reaching a similar degree (SI appendix, Figure S3 G-J). , Figure S3 K and L). In summary, we demonstrated that egt2-1 responds to auxin treatment to the same degree as wild-type, and we concluded that the mutation in egt2-1 does not disrupt the main auxin signaling pathway. This view is consistent with the results of tissue-specific RNA sequencing (RNA-seq) analysis of wild-type and egt2-1 fine roots. We did not find any auxin-related genes in the differentially expressed genes (see results).

In order to locate and clone the EGT2 gene, the F2 population derived from the cross between the high-gravity egt2-1 carrier line TM2835 (in the Morex background) and the cultivar was used to perform a single nucleotide polymorphism-based bulk segregation analysis (BSA) (cv.) Barke, the latter shows a typical wild-type shallow root structure. egt2 is mapped to the 312 Mbp interval on the short arm of the 5H chromosome between markers SCRI_RS_222345 and SCRI_RS_13395 (data set S1) (Figure 2A). Subsequently, TM2835 performed whole-genome sequencing and identified seven genes in the egt2 interval. Compared with the wild-type Morex sequence, these genes carry missense, splice site or stop codon gain mutations (SI Appendix, Table S1). One of these genes encodes a SAM domain protein containing 252 amino acids [HORVU5Hr1G027890 (26) or HORVU.MOREX.r2.5HG0370880.1 (27)], and its mutation (G447A) causes an early termination of the codon domain ( W149*) (Figure 2B and SI appendix, Figure S4 A and B) (27). Except for the SAM domain, no other functional domains are predicted (28). The SAM domain sequence between EGT2 and the previously described SAM domains of other plant species is highly conserved (SI appendix, Figures S4B and S6) (29, 30).

EGT2 encodes the SAM protein. (A) The relationship between the SNP marker established by BSA in F2 hybridization TM2835 (egt2-1, supergravity root) × cv. and the fine root angle of the entire barley genome. Buck (weight root). The y-axis reports Δθ, which explains the difference in allele-specific fluorescence signals between the two BSA DNA blocks for each SNP. (B) The gene structure of EGT2 (HORVU.MOREX.r2.5HG0370880.1) with egt2 mutation (egt2-1: G to A conversion and egt2-2: deletion); the translation start site in the wild type is shown as The black arrow, and the start site in the egt2-2 mutant is shown as a gray arrow; exons are represented by gray boxes, introns are represented by lines, and UTR is represented by white boxes. The red box indicates the sequence code of the SAM domain. (C) Sample pictures of wild type (cv. Golden Promise) and mutant egt2-2 root 7 DAG. (Scale bar: 2 cm.) (D) Fine root angle of wild type (cv. Golden Promise) and mutant egt2-2 7 DAG; n = 15 to 17 in two independent experiments. (E) Root angle of lateral root 14 DAG; n = 16 to 18 in two independent experiments; two-tailed t test, *P <0.05, **P <0.01. (F) Sample pictures of wheat wild type (WT/WT) and egt2 (mut/mut) roots, 7 DAG. (Scale bar: 1 cm.) (G) Root angle between the second and third fine roots of wild-type (WT/WT) and egt2 (mut/mut) wheat seedlings at 7 DAG; n = 18 and 39 respectively for Wild type and mutant. The wheat plants come from two separate segregated groups.

To verify HORVU.MOREX.r2.5HG0370880.1 as EGT2, we used CRISPR/Cas9 to create an additional mutant allele (egt2-2) in the barley cv. Gold promise. We targeted the 5'untranslated region (5' UTR) and two sites in exon 1, which were separated by 196 bp, and restored the 215 bp deletion including the start codon, resulting in truncated protein translation (Figure 2B and SI appendix, Figure S4A). We analyzed the root phenotype of the homozygous T1 line and determined that the root angle of the seminal and lateral roots in the mutant was significantly higher than that of the wild type (Figure 2 CE). Therefore, we confirmed that the change in the root angle phenotype of egt2-2 was caused by the truncation of HORVU.MOREX.r2.5HG0370880.1. Like the egt2-1 mutant in the Morex background, the root length of egt2-2 is similar to that of the wild type (SI appendix, Figure S4C). The response of egt2-2 roots after rotation was faster than that of the wild type, but it was not statistically significant (SI appendix, Figure S4E). It is worth noting that although Golden Promise and Morex both carry wild-type EGT2 alleles (compare Figures 1A and D and 2C and D), they have differences in the growth of fine root horns. In addition, the reorientation of roots after rotation occurs much faster in the wild-type Golden Promise than in Morex (compare Figure 1G with the SI appendix, Figure S4E). Therefore, in addition to EGT2, other genetic factors also affect the growth angle of roots.

In order to further verify the function of the EGT2 gene, we identified mutant lines carrying premature stop codons from the sequencing mutant population of tetraploid wheat (31). We combined the mutations in two durum wheat EGT2 orthologs (the homologs of the A and B genomes) to generate a complete egt2 knockout line. Compared with sib lines carrying wild-type alleles in the two homologues, these double mutants showed a narrower sperm root growth angle in the root box (Figure 2F and G). Similar to barley, the number and length of fine roots of 7-day-old seedlings are not affected (SI appendix, Figure S4 GI).

To investigate the spatial expression pattern of EGT2 in roots, we performed RNA in situ hybridization experiments. EGT2 is expressed throughout the root tip, including root cap, meristem and elongation zone (Figure 3A). The negative (sense) control showed background staining, mainly in the root cap; therefore, we confirmed this expression pattern by investigating our RNA-seq data, and we found that EGT2 is expressed in the root cap, meristem and elongation zone. The qRT-PCR analysis of wild-type and egt2 mutant root tips, including root cap, meristem and elongation zone, rotated 90° for 9 hours, did not indicate any transcriptional regulation of EGT2 on gravity stimulation (Figure 3B). In addition, We measured the expression of EGT2 on the upper and lower sides of the elongation zone 6 hours after the root was rotated by 90° (Figure 3C and SI appendix, Figure S5A), but did not detect any significant changes. At all time points, EGT2 expression was significantly down-regulated in the mutant, which is consistent with the observation that premature stop codons in this gene may lead to nonsense-mediated transcript decay (Figure 3B). The control experiment for unrotated roots is described in Figure S8B of the SI Appendix.

EGT2 expression. (A) RNA in situ hybridization of EGT2; the negative control (sense probe) is shown on the right. (Scale bar: 200 μm.) (B) qRT-PCR of EGT2 expression in root cap, meristem and elongation zone samples rotated 90°; normalized to tubulin; two-tailed t test, *P <0.05, **P <0.01. (C) qRT-PCR of EGT2 expression on the upper and lower flanks of the root elongation zone after rotating 90° for 6 hours (as shown in the SI appendix, Figure S5A); normalized to tubulin; two-tailed t test, **P <0.01 .

SAM domain-containing proteins from animals and plants have many different functions, including their role as transcription factors (29). To analyze the effect of EGT2 mutation on the root transcriptome, we isolated RNA from different root tissues. To this end, we applied laser capture microdissection technology to separate the root cap, meristem and part of the elongation zone from the wild-type and egt2-1 fine roots. This allows us to distinguish between gravity induction (root cap), signal transduction (meristem) and signal execution (elongation zone) (Figure 4D and SI appendix, Figure S5B). For this reason, we selected the most vertically growing fine roots in the two genotypes, and they showed similar root growth angles (Figure 4A). By doing this, we ruled out secondary effects caused by different root growth angles. In addition, we use roots of similar length to rule out age differences because the barley roots do not grow at the same time (32). The RNA-seq experiment produced an average of 41 million 100 bp double-ended reads per sample (SI Appendix, Table S2). We determined the transcriptome relationship between two genotypes and three tissues through principal component analysis (PCA) (Figure 4B). In PCA, the two principal components PC1 and PC2 explain 82% of the total variance (Figure 4B). The biological replicates of each tissue consist of four wild-type and four mutant samples closely clustered together. This indicates that the transcriptomic differences between genotypes are small, but the differences between tissues are large. In order to identify differentially regulated genes, we calculated the pairwise comparisons between the genotypes of each tissue uniquely mapped to the genes on chromosomes 1 to 7 (false discovery rate [FDR] <5% and log2FC >|1|; see materials and Method) (33). This resulted in 67 differentially regulated genes in all tissues, some of which were shared between all or two tissues (Figure 4C and SI appendix, Figure S7 and Table S3). Strikingly, we found that 7 genes encoding expand proteins were down-regulated in the extension region of mutant egt2 (SI appendix, Figure S7 and S8A). Among them, HORVU3Hr1G076620 and HORVU3Hr1G076650 have high homology, with a sequence identity of 99.5%. Time course experiments using the other five expansins showed no difference in expression between wild-type and egt2 when the roots were rotated 90° (SI appendix, Figure S8 CG). Gene Ontology (GO) terms are only assigned to genes that are down-regulated in the elongation zone, all of which are related to the term cell wall (Figure 4E). At the same time, this validated our data set, because expansins are expressed in elongation zones and differentiated root tissues (34). In addition, we found that several genes classified as members of the peroxidase superfamily proteins were up-regulated in meristems or elongation regions (HORVU2Hr1G026420, HORVU7HR1G020300, HORVU3Hr1G036820) (SI Appendix, Figure S7). The differential regulation of peroxidase superfamily protein-coding genes has been found in studies related to gravitational mutants in Arabidopsis (35). In addition, we demonstrated that the gene encoding calmodulin, a major plant calcium receptor, is down-regulated in the meristem and elongation zone (HORVU1Hr1G068440) (SI appendix, Figure S7). Finally, we identified a gene and annotated it as a component of the outer capsule complex up-regulated in the meristem region7. The components of extracellular vesicles are involved in directing exocytotic vesicles to the fusion site on the plasma membrane, and may be involved in the distribution of the auxin transporter PINFORMED4 in Arabidopsis (36, 37).

RNA-seq reveals differences in cell wall-related processes in the elongation zone. (A) Wild type and egt2-1 plant 3 DAG for RNA isolation. (Scale bar: 1 cm.) The arrow points to the sample root (the most vertical root) used for RNA isolation. (B) PCA of 24 RNA-seq samples of two genotypes and three tissues; the first and second principal components together explain 82% of the variance. (C) Venn diagram showing up-regulated (up arrow) and down-regulated (down arrow) differentially expressed genes (DEG) in the corresponding tissues. (D) Experimental device: Isolate RNA from root cap, meristem and 900 μm elongation zone. (E) Rich GO terms that down-regulate DEG in the elongation zone.

The optimization of root structure has been recognized as one of the most important goals in current breeding programs, aiming to improve the resilience and sustainability of crops and agricultural systems (4, 38). Changes in the angle of root growth will affect root anchoring and exploring different soil layers, as well as the way to capture nutrients and water, thereby affecting drought tolerance, such as DRO1 in rice (4). Specifically, modeling and experimental evidence show that steeper root angle growth can increase root depth, thereby helping plants to forage water and move nutrients, such as nitrogen (3⇓ ⇓ –6). A possible trade-off is to obtain less mobile surface nutrients (such as phosphate) or increase sensitivity to waterlogging and salinity (3, 39). Steeper root systems are also expected to reduce the total amount of soil explored, change root competition within and between plants, and contribute to root lodging, although these effects are closely related to crop management factors such as seeding rate (3, 40, 41). However, there is currently limited knowledge about root development genes, gene interactions, and regulatory networks in all major crops (including grains).

Here, we cloned ENHANCED GRAVITROPISM2 (EGT2), a key regulator of root gravity in barley and wheat. The mutation in EGT2 leads to an enhanced heavitrophic response, which results in a steeper root growth angle in the semen and lateral roots. We did not find any other root or branch morphological characteristics affected by this mutation, indicating that EGT2 does not play a role in the ubiquitous signaling pathway, but is specific to root-gravity. In Arabidopsis, most of the ground-oriented mutants have been found because they display the ground-rooted phenotype (20, 42). However, in grasses, some mutants with supergravity roots have been found, such as vln2 and rmd. Villin VLN2 promotes the formation of microfilaments, while the actin binding protein RMD connects actin filaments with gravity-sensing organelles (19, 25).

The only predicted domain in EGT2 is the SAM domain. In animals, proteins containing SAM domains can act as transcription factors, receptors, kinases, or ER proteins (29). In plants, the most famous protein containing the SAM domain is the transcription factor LEAFY (LFY), which is involved in the formation of flower and meristem identity. The modeling of Arabidopsis SAM protein based on structure prediction and LFY characterization showed that most of these proteins can form head-to-tail homo or hetero oligomers/polymers (29). The close phylogenetic relationship between EGT2 and AtSAM5 (At3g07760) indicates that EGT2 has similar oligomerization potential.

EGT2 is also closely related to WEEP. WEEP is a protein containing the SAM domain and was discovered due to the prominent branch phenotype in the peach mutant (30). Peach trees lacking WEEP exhibit a weeping branch growth phenotype; therefore, the branches grow at a larger angle, and after being stimulated by gravity by rotating 90°, the branches no longer move their growth direction upward (30). Therefore, EGT2 and WEEP may be involved in similar pathways that regulate geotropism, but in different plant organs. Peach blossom bud grafting experiments show that WEEP encodes the autonomous determinants of the direction of each branch and does not require movement signals from other parts of the plant (such as plant hormones) (30). In addition, no differences in the concentration of auxin or abscisic acid were detected in the shoots grown between the wild-type peach and the WEEP mutant, and genes related to auxin biosynthesis or perception were not differentially expressed (30).

Similarly, we did not find any changes in the expression of genes related to auxin biosynthesis or perception in the transcriptome comparison between wild-type and egt2-1. Treatment with auxin or auxin transport inhibitors confirmed that the egt2-1 mutant was as sensitive to the interference of auxin balance as the wild type (SI Appendix, Figure S3), indicating that EGT2 functions independently of auxin. However, the auxin transport pathway may still be affected, as demonstrated by the rice mutant villin2 (vln2), which shows that the circulation of the auxin efflux vector PINFORMED2 (PIN2) is disturbed, resulting in a root hypergravity response (25) . However, auxin treatment of the vln2 mutant induced phenotypic restoration, and the mutant was not sensitive to auxin transport inhibitors other than egt2-1. On the other hand, exocyst complex component 7 is transcriptionally upregulated in the egt2-1 mutant. In Arabidopsis thaliana, interference with EXOCYST70A3 expression by knockout or overexpression may regulate the localization of PIN4 in the column cells, thereby regulating the distribution of auxin in the root tip, leading to higher levels of auxin efflux inhibitors Geotropic response (37). It is conceivable that the interference of the expression level in the egt2-1 mutant may cause changes in PIN localization, thereby altering signal transduction; however, this hypothesis remains to be tested. The ubiquitous expression of EGT2 in root cap, meristem and elongation zone indicates that it is involved in gravitational signal transduction, rather than in sensing or differential cell elongation. Since the expression of EGT2 does not change under gravity stimulation, whether it is expression level or distribution, signal transduction is most likely to occur at the protein level.

In the split-ubiquitin yeast two-hybrid screening, AtSAM5, a close homologue of EGT2, interacts with the calcium-dependent protein kinase AtCPK13 (At3G51850) (43), presumably linking AtSAM5 to the calcium signaling pathway. Inhibition of the main plant calcium receptor calmodulin has been shown to inhibit the response of Arabidopsis to gravity (42). In the egt2-1 mutant, calmodulin 5 is down-regulated in the meristem and elongation region, presumably linking EGT2 to calcium-dependent signal transduction. However, the role of calcium in signaling to gravity is still generally unknown.

If we hypothesize the role of EGT2 in signal transduction, we would expect downstream targets in the elongation zone, where the effects of gravity induction are performed by differential cell elongation (18). Here, we found an alarming number of down-regulation of the extensin gene transcription (SI appendix, Figure S7). Expansins are called acid-induced cell wall relaxases. However, most studies are based on the activity of bacterial enzymes, and the function of expansins in plant cell walls is still unknown (34). Similarly, cell wall-related genes are differentially regulated in the weep mutant peach tree, while the differentially regulated auxin-responsive genes in the weep mutant play a role in mediating cell expansion or regulating H transport (30). In addition to expansins, we also found that three genes encoding peroxidase superfamily proteins are up-regulated in meristems and elongation regions (SI appendix, Figure S7). The down-regulation of genes encoding proteins in the peroxidase superfamily has been confirmed in an Arabidopsis study that compared wild-type inflorescence stems with scarecrow and short-root mutant transcriptomes. These transcriptomes showed There is no gravitational response to the rotation on the ground (35). Peroxidase catalyzes the consumption or release of H2O2 and reactive oxygen species (ROS). A type of peroxidase works outside the cell for loose cell walls or cell wall cross-linking (44), and the transcriptional regulation in egt2 may be related to the regulation of expansins. In addition, studies have shown that ROS acts downstream of auxin signaling in root gravity and may act as a second messenger (45).

Based on the extensive expression pattern of EGT2 in the entire root cap, meristem and elongation zone, as well as the interaction between Arabidopsis AtSAM5 homologue and CPK13, we can assume that EGT2 is involved in the signal transduction to gravity. The lack of interference with auxin-related processes at the transcriptome level and the sensitivity to auxin treatment means that EGT2 does not participate in any signal transduction related to changes in auxin levels and/or transport. EGT2 may play a role in a pathway that counteracts the auxin-mediated positive gravity signal pathway, because in order to grow at an angle toward the gravity vector, a pathway that counteracts the positive response to gravity is needed. By knocking it out, the downward growth of roots will dominate and produce a supergravity phenotype.

In summary, our results indicate that EGT2 is an evolutionary conservative checkpoint for the growth angle of semen and lateral roots in barley and wheat. EGT2 may be a promising target for root-based crop improvement in cereals.

The egt2-1 mutation carrier line TM2835 is derived from cv. Sodium azide mutagenesis. Morex is as mentioned earlier (23, 24). To grow on root boxes and agar plates, the seeds were washed in 1.2% sodium hypochlorite for 5 minutes, and then rinsed with distilled water. They were then incubated in the dark at 30°C overnight to induce germination, and only germinated seeds were used for further experiments. Root box growth for plant phenotype and crop rotation experiments was carried out as described above (46). For the phytohormone treatment, the plants were grown in a half-strength Hoagland solution (47) with a pH of 5.8 on a square petri dish placed at an angle of 45° and supplemented with 0.8% phytogel. The plants were grown at 18°C ​​at night (8 hours) and 22°C (16 hours) in a growth cabinet (Conviron). To grow in root canals filled with peat substrates, wild-type and egt2 mutants were grown in the GrowScreen-Rhizo automated platform for 24 days as previously described (48). For MRI measurements, the seeds are placed in a petri dish on wet filter paper. The petri dishes were sealed with parafilm and stored in a growth room protected from light for 24 hours (16°C/20°C night/day temperature, 14 hours light per day) to induce germination, and only germinated seeds were used for further experiments. The seeds are then sown in field soil (Sp2.1, Landwirtschaftliche Untersuchungs- und Forschungsanstalt). The soil moisture is maintained at 8.9%m/m, which is equivalent to 40% of the maximum water holding capacity (49). For each genotype, 18 seeds were planted in a flower pot (Ø = 12.5 cm, 12 cm high), arranged in a hexagonal grid with a seed spacing of 2.5 cm. The seedlings were imaged after 3 days in the growth chamber. For longer experiments, plant individual seeds in larger pots (Ø = 9 cm, 30 cm high) and grow them for a week before imaging.

Durum wheat (Triticum turgidum) egt2 mutant was identified from the TILLING population developed in tetraploid cv. Kronos (31). Two selected lines (Kronos2138 and Kronos3589) carrying premature stop codons in two EGT2 homologous coding sequences (TraesCS5A01G102000 and TraesCS5B02G164200LC) will be crossed, and F1 plants will be self-pollinated. The offspring of selected wild-type and double-mutant F2 individuals from two independent initial crosses were grown in the root box for fine root angle analysis. The seeds were washed in 70% ethanol for 1 minute, then washed in 1% sodium hypochlorite 0.02% TritonX-100 for 5 minutes, and rinsed with distilled water. The sterilized seeds were pre-germinated in wet filter paper at 28°C for 24 hours. Only the germinated seeds were transferred to the root box for 7 days at 25°C.

To analyze the root angle, the plants were grown in a root box for 7 days (fine root angle) or 14 days (lateral root angle). The fine root angle is measured as the angle from the bud to the root tip, relative to the horizontal. For the angle of the lateral root, measure the angle from the growth point of the main root to the tip of the lateral root compared to the horizontal direction. Each plant measures 20 randomly selected lateral roots. For the rotation test, the plants were grown in the root box for 5 days and then rotated by 90°. For the phytohormone treatment, the plants were grown on agar without phytohormone for 5 days, and then transferred to the agar plate supplemented with phytohormone, as shown in the results. After recovery for 1 hour, rotate the agar plate 90°. The picture was taken at the time point shown in the figure. For analysis, measure the root angle of each root tip relative to the horizontal line, and set the rotated angle to 0. For all measurements, calculate the average of all roots of each plant, display it in a graph, and compare it in a statistical test. In order to analyze the growth in the root canal, root images were collected every 2 days in order to distinguish between fine roots and crown roots. The images at 24 days were used for semen, nodule, and lateral root angle analysis. Use the software ImageJ (50) to collect root angle values.

The probe for EGT2 (HORVU5Hr1G027890) mRNA is prepared from the entire coding sequence (start to stop codon). Cloning and RNA probe synthesis were performed as previously described (32). RNA in situ hybridization on the roots of 7-day-old plants was performed as described above (32).

BSA (51) was performed using plants of F2 population obtained from the cross TM2835 × cv. Isolation of Barke and EGT2 loci. 106 F2 seedlings were grown in a flat root box composed of two 38.5 × 42.5 cm black plastic plates. Place five pre-emerged seeds (1 day, 20 °C, on wet filter paper, in the dark) between the wet filter paper pieces in each root box. Each root box is placed vertically in a larger plastic tank containing deionized water, 3 cm from the bottom, in the growth chamber) at night (8 hours) and day (16 hours) at 18 °C and day (16 hours) It lasted for 13 days at 22 °C. At the end of the growth period, the root growth angle of the seedlings was visually evaluated, and the recorded separation rate was 88:18 (wild type and supergravity type), confirming that egt2 was segregated as a single-gene recessive Mendelian Locus (χ2 3:1 = ns). Immediately after this check, 15 plants showing wild-type root growth angles and 15 plants showing high-gravity angles are selected for DNA preparation, based on single plants, use 2 cm2 leaf part, as previously described (23). The DNA sample of BSA was obtained by mixing equal amounts of each of the 15 bulk components of DNA to a final concentration of 50 ng/ul. The 9k Illumina Infinium iSelect barley SNP array (52) was used to genotype two DNA blocks (duplicates) and single DNA samples from 10 supergravity plants. GenomeStudio (Illumina, San Diego, Inc.) was used to analyze the SNP signal. For DNA blocks, the SNP signal is interpreted using theta value method, as described in Reference 1. 53. Modified to integrate the signals obtained from two large blocks (wild type or/and supergravity or -/-) for each SNP in the "delta theta" value score, as follows "delta theta" = [(theta bulk / )-(thetabulk-/-)]2.

A commercial kit (Macheray-Nagel Nucleospin Plant II) was used to extract the genomic DNA of TM2835 for whole-genome shotgun sequencing from leaf samples. Sequencing of DNA using Illumina HiSeq PE150 yielded 699,353,963 paired-end reads, corresponding to approximately 40-fold coverage. The reading is aligned with the first version of the barley cv. The Morex reference genome (26) with BWA v.7.12 (54) and the variants in the genome space are called using SAMtools v. 1.3 (55, 56), the filter minimum read depth is 5 times, and the PHRED quality is >40. In order to discard background mutations caused by the difference between the official Morex reference and the Morex parental seeds that have been used for mutagenesis, the SNP call took into account the other eight TILLMore mutant whole-genome sequencing data available at the time, and filtered the custom AWK script. The minimum ratio DV/DP of the egt2 mutant is 0.8, and the maximum ratio of each other mutant is 0.2, where DP is the coverage depth of the SNP position, and DV is the number of non-reference bases at the same position. SNPeff v.3.0.7 (57) was used to predict SNP effects.

For coverage analysis, a minimum read depth of 5 times was considered, resulting in a 3.5 GB target area containing a total of 15,805 mutations. Therefore, the mutation load on the entire genome is estimated to be 22,579 mutations, or approximately 1 mutation per 220 kb with the same An order of magnitude mutation density (1 per 374 kb), previously estimated based on TILLING results from the same TILLMore population (23). For provean analysis, a value of <-2.5 is considered harmful, and a value of >-2.5 is tolerable.

The modified pseudo-Schiff propidium iodide staining was performed as described in the reference. 32. For Lugol staining, the roots were fixed overnight in 4% paraformaldehyde in phosphate buffered saline (PBS), embedded in 13% agarose, and sliced ​​on a vibrating microtome with a thickness of 40 microns. Then, they were stained with Lugol solution for 3 minutes and rinsed with PBS buffer.

After 1 day of pre-germination, wild-type and egt2-1 plants were grown in the root box for 7 days and then rotated 180° (buds down). The 0.5 cm root segments of the root apex were collected before and after 1 minute, 2.5 minutes, 5 minutes, and 10 minutes. At each time point, eight plants were used for each genotype. Immediately fix the sample in 4% PFA (diluted with 36% formaldehyde in PBS, VWR Chemicals) and place it in a vacuum for 10 minutes. Subsequently, the solution was changed and the sample was rotated overnight at 4 °C. Then, the fixed root samples were embedded in 13% low melting point agarose (peqlab) and sectioned longitudinally in a vibrating microtome (Leica Biosystems) with a thickness of 40 μm. Root slices were stained with Lugol solution (Roth) for 3 minutes, washed with PBS buffer, and photographed with an optical microscope (Zeiss).

The image of the root part is analyzed by ImageJ. The distances from the center of the static stone to the front and distal cell wall and the new lower cell wall were measured respectively. Measure 10 cells in the center of each root and plot the data using cells with a length of 25 μm to 40 μm.

A Zeiss PALM MicroBeam microscope was used to examine RNA in situ hybridization and Lugol stained samples.

The MRI image was obtained on a 4.7 T vertical magnet equipped with a Varian console (58). A multi-layer spin echo sequence is used. The sequence parameters are suitable for different pot sizes. For the 9 cm pot, we used a birdcage RF coil with a diameter of 10 cm and the following sequence parameters: 0.5 mm resolution, 1 mm slice thickness, 9.6 cm field of view, TE = 9 ms, TR = 2.85 s, bandwidth = 156 kHz , Two averages. For the 12.5 cm pot, the following parameters were changed: a birdcage RF coil with a diameter of 140 mm, a field of view of 14 cm, and a resolution of 0.55 mm.

For the CRISPR target sequence, we selected a 20 base pair sequence in the first exon of EGT2 (HORVU5Hr1G027890). Among them, the original spacer adjacent to the motif PAM sequence NGG is available at http://crispr.dbcls. The off-target in the barley genome was examined on jp/ (Barley [Hordeum vulgare] Genome, 082214v1 [March 2012)]. The locus we used has only one 20mer prediction target, while the 12mer target sequence upstream of PAM only has 3 prediction targets at most. The CRISPR guide sequence is marked in Figure S4A in the SI Appendix. The sgRNA shuttle vectors pMGE625 and 627 were used to generate the binary vector pMGE599, as described in the reference. 59. Transform with spring barley varieties. Golden Promise is grown in a climate chamber of 18 °C/14 °C (light/dark), with a relative humidity of 65%, a photoperiod of 16 hours, and a photon flux density of 240 μmol ⋅ m-2 ⋅ s-1. The binary vector pMGE599 was introduced into Agrobacterium tumefaciens AGL-1 strain (60) by electroporation (E.coli Pulser; Bio-Rad). As mentioned earlier, the scutellum tissue of barley caryopsis is used for Agrobacterium-mediated transformation (61). The insertion and integration in the barley genome was confirmed by detecting the hygromycin gene sequence by PCR in the generated T0 line, and analyzing the mutations in EGT2 by PCR and Sanger sequencing, and using the seeds of the T1 generation for the experiment.

For qRT-PCR, use the RNeasy Plant Mini Kit (Qiagen) to extract RNA from plants grown in the root box for 7 days after germination, and use the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher) to synthesize the first-strand complementary DNA (cDNA). Each organism repeatedly collects four plants, from the root tip containing the meristem and elongation zone, about 2 mm, until the root hairs grow out. For each genotype, three biological replicates and three technical replicates are used. For the reaction, mix 2 μL PerfeCTa SYBR Green SuperMix (Quantabio), 1 μL primer mix at a concentration of 1 μM, and 1 μL cDNA. Calculate the primer efficiency for each oligonucleotide using the following dilution series: 1, 1/2, 1/4, 1/8, 1/16, 1/32, 1/64, and 1/128. Refer to the housekeeping gene TUBULIN (HORVU1Hr1G081280) and calculate the relative expression level of the transcript according to the method described in the reference. 62. Oligonucleotide primer sequences are listed in Table S4 in the SI Appendix. Significant differences in gene expression levels are determined by the student's t-test on both sides.

Use the root tip of the most vertically growing fine root of a 3 d-old plant and designate it as a biological replicate. For each genotype, four biological replicates were analyzed. The plants were grown in a root box and fixed on ice with Farmer's fixative (ethanol:acetic acid 3:1) under 500 mbar vacuum for 15 minutes, and then rotated at 4°C for 1 hour. Replace the fixative solution, repeat the procedure twice, then replace the solution with 34% sucrose and 0.01% safranine-O in PBS. The sample was vacuum infiltrated again for 45 minutes and incubated on ice at 4°C for 21 hours. The samples were then carefully dried with tissue paper and embedded in tissue paper freezing medium as previously described (63). The media block containing the tissue is stored at -80°C and adapted to -20°C in a cryostat (Leica CM1850). Mount 20 μm thick longitudinal sections on poly-L-lysine-coated glass slides (Zeiss) and remove them by incubating in 50% EtOH and 70% for 1 minute after 5:30 minutes Tissue freezing medium ethanol, 95% ethanol, 100% ethanol and 100% xylene (without RNase). Use the following settings of the PALM micro-beam laser capture instrument (Zeiss, Germany) to cut tissue (root cap, meristem and 900 μm elongation zone adjacent to meristem, SI appendix, Figure S5B): Energy: 79; Speed: 100; Cutting program: "Center RoboLPC", manually picked up with a sharp needle and transferred to the cover of the RNase-free adhesive cover (Zeiss). According to the manufacturer's organization protocol, including DNase treatment, RNA was isolated using Arcturus PicoPure RNA Isolation Kit (Thermo Fisher). Using Agilent RNA 6000 Pico kit and Agilent 2100 bioanalyzer to measure RNA quality, the RIN value obtained was between 7.1 and 8.9, and the concentration was between 610 and 95,000 pg/μL. Pre-amplification and library preparation were performed as described in the references. 64. Detection and sequencing were performed on the Illumina NovaSeq sequencer using the PE100 protocol. Trimmomatic 0.39 (65) version is used to remove low-quality reads and remaining adapter sequences from each read data set. Specifically, a sliding window method is used, where if the average quality in the 4 bp window is lower than the phred quality score 15, then the read is clipped. Only keep reads ≥30 bp in length for further analysis. The data is stored in the Sequence Reading Archive (SRA) PRJNA589222. The BBDuk of the BBTools suite (https://jgi.doe.gov/data-and-tools/bbtools/) is used to remove rRNA readings from the data set, using a kmer length of 27 as the filtering threshold for purification. After removing the rRNA readings, an average of 8 million paired readings remain. Splicing-aware STAR aligner v.2.7.2b (66) is used to compare the remaining reads with the genome index of the barley reference sequence and the annotation of the genotype Morex (IBSC v2.0) (26). By considering only the reads mapped to a single location (–outFilterMultimapNmax 1), multiple mapped reads mapped to multiple locations are excluded from the subsequent steps. On average, 5 million reads per sample are aligned with a unique location in the IBSC v2.0 barley reference genome genome, which contains 46,272 predicted coding and non-coding gene models [EnsemblPlants release 45 (26)]. The aligned double-ended reads are sorted according to their positions and converted to .bam files by the software SAMtools [Version 1.3.1 (55)]. Use HTSeq [version 0.10.0 (67)] with the parameter "-r pos -i gene_id -s no-secondary-alignments ignore-supplementary-supplementary-supplementary." Use the normalization program rlog() implemented in the R package DESeq2 And the plotPCA() function [Version 1.22.2 (33)] performs PCA on the expression data. By using the fpm() function of DESeq2 to calculate reads per million fragments (FPM), after removing low-expressed genes with reads less than 10 in all samples, the expression value is normalized to the library size. Based on the generalized linear model, the negative binomial distribution in the R package DESeq2 (33) is used to calculate the expression level of genes, and the log2 fold change (log2FC) between wild-type and mutant is calculated by the variance-mean dependence in the count table value. Respective tissue design ~ genotype tissue genotype: tissue. The significance value of the log2FC value is calculated as the Wald test P value and adjusted by the Benjamini-Hochberg program to obtain the FDR (68). Genes with FDR <5% and log2FC >|1| are considered to be differentially expressed. From this gene set, we excluded the gene pairs assigned to chr0 and chr1, which have the same annotation, and their respective gene partners are the gene pairs that have the closest related transcript after BLAST search. Use agriGo (69) to perform GO term enrichment on the resulting gene set. The sequencing data has been stored in the sequencing read archive of the National Biotechnology Information Center (PRJNA589222).

A laser capture dissecting microscope was used to separate cells to analyze gene expression levels on the upper and lower sides of the rotating root (SI appendix, Figure S5A). The 7-day-old plants grown in the root box were rotated by 90° for 6 hours, and the 5 mm long root tip segment of the first vertical root was collected, and then fixed, embedded and sliced ​​as described. Cortex and epidermal cells in the elongation zone 1.5 mm adjacent to the meristem were cut by PALM Microbeam Platform (Zeiss) with the following settings: energy: 54; speed: 3; and cutting program: "cut", manually with a sharp needle Pick up and transfer to the lid of the RNase-free adhesive cap (Zeiss).

Isolate and analyze RNA as described. The RIN value of RNA is between 6.7 and 7.8, and the concentration is between 9 and 20 ng/μL. RNA from three roots of independent plants was used as one biological replicate; three biological replicates were analyzed for each genotype and flanking.

EGT2 (HORVU.MOREX.r2.5HG0370880.1) protein sequence exploded on Phytozome v12.1 to Brachypodium dystachyon proteome v3.1, rice proteome v7_JGI, corn proteome Ensembl-18, Arabidopsis proteome 18, Prunus The default settings for persica proteome v2.1 and Sorghum bicolor proteome v3.1.1. Consider the hits with E value <3.9 × 10−77. Then use the EnsemblPlants Compara Ortholog tool to confirm the identified orthologs. The retrieved protein sequences are aligned by ClustalW in the software MEGA X, the default value (70): use the maximum likelihood method (71) and the JTT matrix-based model (72) to infer the ancestral state. The tree shows a set of possible amino acids (states) at each ancestral node based on their inferred likelihood at position 1. By applying the Neighbor-Join and BioNJ algorithms to the pairwise distance matrix estimated using the JTT model, the initial tree can be automatically inferred and then the topology with superior log-likelihood value can be selected. The rate between sites is considered to be uniform between sites (flat rate option). The analysis involves 16 amino acid sequences.

RNA-seq data has been deposited in SRA (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA589222).

This work was funded by Deutsche Forschungsgemeinschaft Grant HO2249/21-1 (FH). KAN, DP and RK recognize the Helmholtz Association's support for Forschungszentrum Jülich. According to the grant agreement No. 731013 (EPPN2020), root riser research was funded by the European Union Horizon 2020 Research and Innovation Program. The work described here is partly supported by the “Rooty: A Root Model Toolbox for Increasing Wheat Yield” project, which is funded by the IWYP Alliance (Project IWYP122) through the Biotechnology and Biological Science Research Committee to CU, JS, RT and SS UK (BB/S012826/1). We thank Felix Frey (University of Bonn) for suggestions on RNA-seq data analysis and discussion, Johannes Stuttmann (University of Halle) for sharing the CRISPR/Cas cloning vector, and Shalima H. ​​Orse for supporting the entire project​​. We thank Anna Galinski, Jonas Lentz, Carmen Müller, Bernd Kastenholz, Ann-Katrin Kleinert, Roberta Rossi, and Kwabena Agyei (Forschungszentrum Jülich GmbH) for their assistance during the Rhizotron study. RK and DP are very grateful to Dagmar van Dusschoten and Johannes Kochs for their support and maintenance of the MRI system.

↵1G.KK, SR and LG have made equal contributions to this work.

Author's contribution: GKK, SR, SS and FH design research; GKK, SR, LG, IV, JI, JA, SGM, R. Balzano, DP, CF, R. Bovina and JS conducted research; GKK, KAN, RK , TGS, MM, CU, HS, RT, SS and FH analysis data; GKK, SR, SS and FH wrote this paper.

The author declares no competing interests.

This article is directly contributed by PNAS.

This article contains online support information at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2101526118/-/DCSupplemental.

This open access article is distributed under the Creative Commons Attribution-Non-Commercial-No Derivative License 4.0 (CC BY-NC-ND).

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.