Jasmonates, gibberellins, and powdery mildew modify cell cycle progression and evoke differential spatiotemporal responses along the barley leaf

Abstract Barley (Hordeum vulgare) is an important cereal crop, and its development, defence, and stress responses are modulated by different hormones including jasmonates (JAs) and the antagonistic gibberellins (GAs). Barley productivity is severely affected by the foliar biotrophic fungal pathogen Blumeria hordei. In this study, primary leaves were used to examine the molecular processes regulating responses to methyl-jasmonate (MeJA) and GA to B. hordei infection along the leaf axis. Flow cytometry, microscopy, and spatiotemporal expression patterns of genes associated with JA, GA, defence, and the cell cycle provided insights on cell cycle progression and on the gradient of susceptibility to B. hordei observed along the leaf. Notably, the combination of B. hordei with MeJA or GA pre-treatment had a different effect on the expression patterns of the analysed genes compared to individual treatments. MeJA reduced susceptibility to B. hordei in the proximal part of the leaf blade. Overall, distinctive spatiotemporal gene expression patterns correlated with different degrees of cell proliferation, growth capacity, responses to hormones, and B. hordei infection along the leaf. Our results highlight the need to further investigate differential spatial and temporal responses to pathogens at the organ, tissue, and cell levels in order to devise effective disease control strategies in crops.


Introduction
Barley is a widely grown cereal that is cultivated for animal feed, malting, and human consumption (Akar et al., 2004).In plants, growth, development, and stress responses are modulated by different levels and gradients of phytohormones (Kupke et al., 2022;Cao et al., 2017).Numerous studies have shown how such hormones affect the signalling networks that modulate cell division and thus modulate growth and development, including abscisic acid (ABA; Świątek et al., 2002;(Skirycz and Inzé, 2010), auxin (Perrot-Rechenmann, 2010), brassinosteroids (reviewed by Clouse, 2011), cytokinins (reviewed by Haberer and Kieber, 2002), gibberellins (GAs; Achard et al., 2009), and jasmonates (JAs, including active precursors and derivatives of jasmonic acid; Noir et al., 2013;Bömer et al., 2018).The levels of stress-related hormones such as JA and salicylic acid (SA) differ along the maize leaf (Nelissen et al., 2018), whilst in barley JAs are relatively more abundant in the root tip, the scutella node, and in the leaf base, but not at the leaf tip (Maucher et al., 2000(Maucher et al., , 2004)).These studies therefore also suggest a developmental role for JA and SA.
Gibberellins (GAs) derived from gibberellic acid are a large family of tetracyclic diterpenoids, first identified as secondary metabolites in the pathogenic fungus Gibberella fujikuroi, and they are ubiquitous in higher plants where they function as endogenous growth regulators (Hedden and Thomas, 2012;Binenbaum et al., 2018;Salazar-Cerezo et al., 2018).The interaction between GA and JA signalling through DELLA proteins has been extensively reviewed (Navarro et al., 2008;Hong et al., 2012;Zhong and Yang, 2012;Pérez-Salamó et al., 2019).Briefly, GAs play crucial roles in promoting growth and development, notably by triggering the degradation of the growth-repressor DELLA proteins in angiosperms (reviewed by Harberd et al., 2009).DELLAs integrate plant hormone signalling pathways and environmental changes through physical interactions with regulatory proteins such as JAZs (Hou et al., 2010;Wild et al., 2012;Yang et al., 2012) and the MYC2 transcription factor (Hong et al., 2012).Similar to the action of JA on JAZs, through its GA-INSENSITIVE DWARF1 (GID1) receptor, GAs recruit DELLA proteins for ubiquitination and degradation via the 26S proteasome, thereby resulting in activation of the GA response (Qi et al., 2014).GA and JA synergistically and antagonistically interact to regulate seedling growth and resistance to pathogens through the interaction between JAZs and DELLAs (Song and Bent, 2014;de Vleesschauwer et al., 2016).
The coordination of cell division and cell expansion are critical to the normal growth and development of plant tissues and organs.The regulation machinery of the cell cycle is tightly temporally and spatially coordinated during plant development by the action of phytohormones (Gonzalez et al., 2010;Dudits et al., 2011;Nelissen et al., 2012Nelissen et al., , 2018)).In monocots, such as maize and wheat, dividing cells are found at the base of the leaf, while expanding and mature cells are located towards the tip (Nelissen et al., 2016;Loudya et al., 2021).In dicots such as Arabidopsis, cell division continues for longer at the leaf base and ceases earlier at the tip (Nelissen et al., 2016).In comparison, the characterization of cell-cycle regulation in barley is less advanced, but several orthologues of cyclins and cyclin-dependent kinases of the G1/S and G2/M transitions are known.As in Arabidopsis, Cyclin Dependent Kinase A1 (CDKA1) contributes to initiating the cell cycle at G1/S as well as being active in the G2/M phase in rice (Umeda et al., 1999;Lee et al., 2003;Qu et al., 2018), barley (Gendreau et al., 2012;Jin et al., 2020), and maize (Dante et al., 2014).For full CDKA1 activity, the combination with Cyclin Dependent Kinase D1 (CDKD1) is essential (Sofroni et al., 2020).The Cyclin Dependent Kinase B1 (CDKB1) subgroup of cell-cycle markers functions from the onset of the S-phase until mitosis and regulates cell division in several plants including Antirrhinum (cdc2c) (Fobert et al., 1996), alfalfa (cdc-2MsD) (Magyar et al., 1997), tobacco (Sorrell et al., 2001), Arabidopsis (Segers et al., 1995;Menges and Murray, 2002) and barley (Gendreau et al., 2008(Gendreau et al., , 2012)).
Cell-cycle checkpoints maintain genomic integrity in proliferating cells to prevent aberrant replication following DNA damage or stress.In Arabidopsis, WEE1 inhibits plant growth by arresting cells in the G2-phase (De Schutter et al., 2007), and it also has a role in endoreplication in maize endosperm and in tomato fruit (Sun et al., 1999;Gonzalez et al., 2004Gonzalez et al., , 2007)).Members of the Poaceae family such as maize, barley, and wheat show a lower endoreduplication index in the leaf in comparison to Brassicaceae and Cucurbitaceae family members (Barow and Meister, 2003).
Barley is a host for numerous pathogens and insect pests that attack the plant at different growth stages.Blumeria hordei causes barley powdery mildew and results in considerable loss in productivity despite fungicide treatment (Kusch and Panstruga, 2017;Stam et al., 2017, Preprint;Cowger et al., 2018;Liu et al., 2019).Powdery mildews belong to the Helotiales (Johnston et al., 2019), an ascomycete family representing an ancient lineage that evolved over 100 million years ago and that has diversified to more than 400 species colonizing nearly 10 000 plant species (Takamatsu, 2004;Johnson et al., 2010;Kusch and Panstruga, 2017).Blumeria hordei is an obligate biotroph, and its molecular and genetic interactions have been extensively studied (Kuhn et al., 2016).
The main aim of this study was to gain insights into the spatiotemporal regulation of the responses to hormones and to B. hordei infection along the barley leaf.To this end, we carried out flow cytometry to monitor cell-cycle stages and ploidy, microscopy to quantify hyphae formation, and qRT-PCR to profile spatiotemporal expression patterns of different categories of genes associated with JA, GA, defence, and the cell cycle.We found that different parts of the leaf possess different growth capacities and intrinsic abilities to respond to hormones and B. hordei infection, and that these are further fine-tuned by MeJA and GA.

Plant cultivation and pathogen maintenance
The barley (Hordeum vulgare) cultivars Golden Promise and Haruna Nijo were grown in 9 cm pots in soil (John Innes No.1) with 15-20 seeds per pot.Golden Promise was also grown in hydroponic conditions in magenta vessels (Sigma GA-7 V8505) containing 100 ml of sterile perlite and 50-60 ml of Half-strength Hoagland solution No. 2 (Sigma H2395), with nine seeds per vessel.All the barley seedlings were grown under a 16/8 h photoperiod (~140-170 μmol m -2 s -1 ) at 22 °C and relative humidity of ~60-70%.Blumeria hordei strain DH14 (provided by Dr L. Bindschedler, RHUL) was maintained on Golden Promise seedlings that were inoculated at 7 days after sowing (DAS) and grown in soil under the same conditions

Measurement of nuclei of infected leaf epidermal cells
Primary leaves of plants were inoculated with B. hordei at 5 DAS (see below), and then at 7 DAS the leaves of infected and equivalent uninfected plants were used for DAPI staining (adapted from Chazotte, 2011).Microscopy observations and disease scoring in the infected leaves were according to an adapted protocol outlined by Lambertucci et al. (2019).The infected leaf samples were cleared with ethanol (80% v/v) and acetic acid (20% v/v) solution overnight at 4 °C in the dark, and then rinsed with phosphate buffered saline (PBS) for 30 mins.DAPI solution (2 ng µl -1 ) was added to leaf tissues for 20 min and then they were de-stained for 30 min with PBS.The DAPI-stained leaves were also stained with 2 μg ml -1 of propidium iodide (PI ; adapted from Scheler et al., 2016).The leaves with their adaxial side up were placed on glass slides with PBS and sealed with nail varnish.Nuclei were imaged using a Nikon Eclipse Ni-E Upright microscope with preselected filters for DAPI (358/461 nm) and TxRed (535/617 nm) with a Nikon Intensilight C-HGF1 UV light source.i-stack images were taken of uninfected epidermal cells and infected nuclei (near the haustoria) and analysed using ImageJ software.

Flow cytometry analysis
Ploidy levels were measured by flow cytometry (Sysmex CyFlow ® Space) using a Cystain UV Precise P high-resolution DNA staining kit (Partec, 05-5002) according to the manufacturer's instructions.The protocol for barley sample preparation was adapted from Doležel et al. (2007) and Noir et al. (2013): fine-cutting of the tissue was carried out in Cystain UV Precise P Nuclei Extraction Buffer (Sysmex), staining with DAPI was conducted using Cystain UV Precise P Staining Buffer (Sysmex), and debris was filtered using a 20-30 µm filter (CellTrics, Sysmex) before running the sample through the flow cytometer.To identify the nuclear DNA content in developing barley, untreated Golden Promise seedlings were collected at 7 DAS and the sheath was cut into 5-10 mm sections while the proximal and distal portions of the leaf blade were cut into 20 mm sections.Golden Promise seedlings grown in soil were collected 7 DAS, infected with B. hordei, and then sampled at 3 days postinoculation (DPI) and 5 DPI.The proximal and distal leaf-blade sections of these seedlings were then subjected to flow cytometry.The analysis was performed to measure the DNA content from at least 15 000 nuclei per sample based on relative fluorescence intensities, and recorded using the Sysmex FloMax software.The flow cytometry experiments were repeated for three biological replicates, with two technical replicates for each.Cell-cycle analysis was performed in parallel with the ploidy measurements, by fitting frequency histogram outcomes into the cell-cycle analysis tool to obtain the frequency of nuclei in the G0/G1, S, and G2/M phases of the cell cycle.

Characterization of B. hordei infection along the primary barley leaf blade
The infection protocol was adapted from Haugaard and Collinge (2001), Achuo et al. (2004), Nottensteiner et al. (2018), andLambertucci et al. (2019).The primary leaf blade of soil-grown Golden Promise at 7 DAS was detached and placed with the adaxial surface upwards on 0.6% agar (Melford P1003) containing 20 ng ml -1 benzimidazole (Sigma 194123) in 12 cm plates.Hydroponically grown Golden Promise seedlings were also collected 7 DAS, and their primary leaves were detached and placed on similar plates.In addition, whole seedlings of hydroponically grown Golden Promise at 7 DAS were also placed on similar plates.Spores of B. hordei from fresh inoculum previously propagated on Golden Promise for 7 d were delivered by shaking to the horizontally placed leaves on the plates as described by Nottensteiner et al. (2018).Characterization and quantification of B. hordei infection in the primary leaves was followed as described by Peterhansel et al. (1997) and Li et al. (2020), with a spore density of 15-40 mm -2 in a sealed chamber and the infection was allowed to develop under the plant growth conditions described above.Conidia, appressoria, and hyphae formation in the distal and proximal leaf-blade sections were observed and counted using a light microscope.The data were converted to percentages of hyphae formed relative to the total number of ungerminated conidia, appressoria, and hyphae-forming conidia.

Phytohormone treatments
Either 0.1 µM or 10 µM GA 3 (Sigma) was added to the medium of the hydroponically grown Golden Promise seedlings from seed germination until 7 DAS (concentrations based on physiological responses determined by Uçarli, 2021).Similarly, 50 μM MeJA was applied to the hydroponically grown barley seedlings from 4 DAS to 7 DAS (concentration based on Noir et al., 2013;Bömer et al., 2018).Sampling of the treated seedlings was performed 7 DAS, when the sheaths were dissected into segments that were 0-0.5 cm, (A) 0.5-1.5 cm (B), 1.5-2.5 cm (C), and 2.5-3.5 cm (D) from the seed.Similarly, the leaf blade was dissected into 2 cm sections in the proximal and distal portions.The samples were used for flow cytometry analysis of nuclear DNA content and cell-cycle stages.
Another batch of Golden Promise seedlings grown in hydroponic conditions was treated with either 10 µM GA 3 as described above, or with 50 μM MeJA for 24 h (from 5 DAS to 6 DAS) after which they were transferred to fresh half-strength Hoagland solution.At 7 DAS, leaves were cut from the seedlings and infected with B. hordei as described above.Proximal and distal leaf-blade samples were taken at 5 DPI for assessment of the spread of infection.
A third batch of Golden Promise seedlings were grown hydroponically and treated as described above with either 10 μM GA 3 for 7 d or 50 μM MeJA for 24 h, and were then infected with B. hordei.Proximal and distal leaf-blade samples were collected 2, 3, and 5 DPI for molecular analysis.

Gene expression analysis by quantitative real-time PCR
For gene expression analysis, the sheath and leaf-blade samples from three seedlings per treatment were each pooled to form a biological replicate.Total RNA was isolated using RNeasy Plant Mini Kit (Qiagen) following the manufacturer's protocol.cDNA preparation was performed using a QuantiTect Reverse Transcription Kit (Qiagen).Real-time amplification was performed using SYBR Green JumpStart (Sigma-Aldrich) according to the manufacturer's instructions.Transcript analysis was performed using RNA samples derived from at least three independent biological replicates.The comparative 2 -ΔΔCT method (Livak and Schmittgen, 2001;Pfaffl, 2001;Vandesompele et al., 2002;Noir et al., 2013) was used to evaluate the transcript abundance of each gene in the samples relative to the reference genes Ubiquitin and GAPDH.Primer sequences are listed in Supplementary Table S1.

Statistical analysis
One-way ANOVA was used to identify significant differences in the molecular and flow-cytometry experiments.One-way ANOVA followed by Tukey's post-hoc test was used in the experiments involving B. hordei infection in the leaf blade with or without phytohormone treatment, and for assessment of nuclear enlargement in epidermal cells following B. hordei infection.Statistical analysis was performed using Graphpad Prism as well as the Xrealstats analysis tool in Microsoft Excel.

MeJA and GA treatments differentially regulate the cell cycle in barley sheath and leaf blade
In monocotyledon leaves, the sheath and the leaf blade undergo distinct developmental processes, with the sheath emerging last from the meristem (Ramírez-González et al., 2018).These two primary leaf portions from the barley cultivar Golden Promise were separated and dissected at 7 DAS), and the cell cycle stages along their length axes were examined using flow cytometry (Fig. 1A).The frequency of nuclei with 4-16 C ploidy was higher in the sheath than in the proximal and distal portions of the primary leaf blade (Fig. 1B), and this is consistent with a relatively higher mitotic activity in the sheath, where a higher frequency of cells in the S/G2-M phase of the cell cycle was observed compared to the leaf blade (Fig. 1C; Table 1; Supplementary Table S2).Similarly, higher ploidy levels were observed in the sheath of the fast-growing barley cultivar Haruna Nijo (Groszyk and Szechynska-Hebda, 2021; Supplementary Fig. S1), suggesting that cell ploidy and cell cycle transition stages in barley are probably independent of the growth rate of the cultivar.
Flow cytometry was used to determine the effects of JA and GA on the progression of the cell cycle in the sheaths and leaf blades.Compared with the control, the MeJA treatment significantly increased the frequency of 2C nuclei (P=2.0 × 10 - 6 ) and reduced the frequency of 4C (P=2.6 × 10 -8 ) and 8C (P=5.0 × 10 -4 ) nuclei in the sheath, while smaller differences were observed in the leaf blade (Fig. 1F; Table 1).In the leaf blade, the MeJA treatment significantly reduced the frequency of 4C (P=4.3 × 10 -3 ) and 8C (P=9.1 × 10 -3 ) nuclei compared with the untreated control.The 0.1 µM GA treatment did not show a significant effect on the nuclear DNA content in the sheath; however, the 10 µM GA treatment resulted in significant increases in 4C (P=0.035) and 8C (P=9.6 × 10 -3 ) nuclei, and a corresponding decrease in 2C nuclei (P=0.026).In the leaf blade, there were significant dose-dependent increases in the frequency of 4C nuclei at both GA concentrations, and a significant increase in 8C nuclei at 10 µM GA.These results suggested that GA and JA have opposite effects on the cell cycle and growth in the barley leaf, with GA causing increases in elongation and nuclear DNA content (ploidy), whilst JA reduces them.

MeJA and GA treatments differentially regulate the expression of cell-cycle marker genes in the barley leaf
To gain insights into the molecular mechanisms regulating the barley response to the hormone treatments, we next used qRT-PCR to examine the expression of signature genes associated with the cell cycle and proliferation along the sheath and leaf blade.The genes have previously been characterized by Gendreau et al. (2008Gendreau et al. ( , 2012)), included the following: Cyclin A3 (CYCA3, HORVU.MOREX.r2.5HG0363660; a checkpoint regulator of S, S/G2 transition; Gendreau et al., 2012;Qi and Zhang, 2020); Cyclin B1 (CYCB1, HORVU.MOREX.r2.3HG0246040; a checkpoint regulator at the G2/M transition; Dewitte et al., 2003); Cyclin-Dependent Kinase A1 (CDKA1, HORVU.MOREX.r2.6HG0457700; known to be constitutively expressed throughout the cell cycle and acts as a positive regulator of cell proliferation; Iwakawa et al., 2006;Gendreau et al., 2012), Cyclin-Dependent Kinase B1 (CDKB1, HORVU.MOREX.r2.7HG0565000; functions during G2/M and the S/G2 phase of the cell cycle; Dewitte et al., 2003;Gendreau et al., 2012); Cyclin-Dependent Kinase D1 (CDKD1, HORVU.MOREX.r2.7HG0565000; its expression increases from G1 to S phase, and it mediates cell-cycle progression through the activation of CDKs such as CDKA1 and CDKB1; Gendreau et al., 2012;Gutierrez et al., 2009;Takatsuka et al., 2015); and the nuclear kinase WEE1 (HORVU.MOREX.r2.6HG0461230; a negative regulator of the G2/M phase that inhibits the activities of CDKA and CDKB; De Schutter et al., 2007;Qi and Zhang, 2020).The MYB3R4 transcription factor (HORVU.MOREX.r2.3HG0206530) has not been formally identified as a cell-cycle marker, but it controls the expression of G2/M phase-specific genes and maintains diploidy (Saito et al., 2015;Haga et al., 2007Haga et al., , 2011)).For a schematic depiction of the known roles of these genes in the cell cycle, see Fig. 2. Using the distal leaf-blade segment as the reference, the overall expression of these cell cycle-related genes under control conditions was higher in the sheath than in the blade, especially at its base, the region closest to the meristem and more actively dividing (Fig. 3).
Compared to the control, treatment with 10 µM GA and 50 µM MeJA both significantly reduced the expression of these genes in the sheath and blade tissues (Fig. 3).These results supported a role, either direct or indirect, of JA and GA in modulating leaf growth by acting on common cell-cycle regulators.
In the controls, the expression of the gibberellin receptor Gibberellin-Insensitive Dwarf1 (GID1, HORVU.MOREX.r2.1HG0049800; regulating growth and development; Ueguchi-Tanaka et al., 2005) and the defencerelated gene Pathogenesis Related 5 [PR5, HORVU.MOREX.r2.5HG0351950; belonging to the thaumatin-like protein family, suggested to interact with (1,3)-β-d-glucans commonly found in fungal walls; Osmond et al., 2001] were broadly comparable to JIP23 in the sheath (Fig. 3).The expression levels of GID1, PR5, and JIP23 in the controls were particularly low in all segments of the sheath.Notably, GID1 expression was reduced by ~1.5-fold in the distal leaf blade by both the GA and JA treatments compared to the control (GA, P=1.84 × 10 -3 , MeJA, P=2.81 × 10 -4 ), while that of PR5 was 1.5-fold higher in the sheath and ~5-fold in the proximal section (P=3.06 × 10 -7 ) of the leaf blade following MeJA treatment.Taken together, these results showed the existence of Fig. 3. Expression of genes related to the cell cycle, JA, GA, and plant defence in the sheaths and leaf blades of barley seedlings in response to JA and GA treatment.Seedlings of the cultivar Golden Promise were grown hydroponically.For the JA treatment, 50 µM MeJA was added to the solution at 4 days after sowing (DAS).For the GA treatment, either 0.1 µM GA 3 or 10 µM GA 3 was present in the solution from the time of germination.Samples of expression signatures associated with the sheath and leaf blade that could be linked to cell's distinct abilities to respond to hormone-induced stress.

The proximal and distal sections of the leaf blade show different susceptibility to B. hordei infection
Barley seedlings were cultivated either in soil or hydroponic conditions and then detached leaves or whole seedlings were infected with B. hordei in order to evaluate the distribution of infection along the leaf blade and to test whether the process was affected by the excision of the leaf from the plant (Fig. 4A).Infection at the macroscopic scale was clear at 5 DPI, with colony clusters more abundant in the distal section (Fig. 4B).Hyphae formation from the germinated conidia was quantified 2 DPI, when it was possible to distinguish individual colonies according to (Li et al., 2020) (Fig. 4C).Irrespective of growth conditions or cutting, the hyphae formation was greater on the distal leaf blade (Fig. 4D), mimicking the natural process of infection.Compared to the proximal leaf blade of cut leaves from hydroponic seedlings, ~58% (P=4.0 × 10 -2 ) more hyphae were observed in the distal leaf blade.The hyphae formation in hydroponically grown barley seedlings was ~30% lower in the proximal leaf section whether the leaf was infected as part of the whole seedling (P=1.9 × 10 -3 ) or detached (P=4.2 × 10 -2 ).
To examine their effects on B. hordei infection, seedlings were pre-treated with either 50 μM MeJA or 10 μM GA.Compared with the control, MeJA treatment reduced hyphae formation (P=8.0 × 10 -4 ) in the proximal leaf blade by an average of ~33% but had no detectable effect in the distal leaf blade (Fig. 4E).In contrast, the GA pre-treatment had a positive effect on hyphae formation in the leaf.
The distribution pattern of B. hordei infection along the leaf blade therefore reflected the gradient of JA-and GA-related gene expression along the leaf (Figs 3, 4D) and it was not affected by wounding; instead it reflected the existence of an intrinsic regulatory system that was possibly modulated by JA and GA to protect the cells near the meristem from pathogen attack.

Genes associated with the cell cycle, hormones, and defence show different expression patterns in the leaf blade following B. hordei infection
To understand of the spatiotemporal effect of hormones on B. hordei infection of barley leaf, the expression of genes related to the cell cycle, JA, and defence in uninfected and B. hordeiinfected, proximal and distal leaf blades at 3 DPI and 5 DPI was next examined.The same set of cell-cycle marker genes described in Fig. 3 was analysed to assess whether the pathogen altered their expression and interfered with cell proliferation in planta.Relative to the uninfected control, the expression of the two G2/M phase marker genes CYCB1 and CDKB1 increased by ~1.3-fold in infected distal leaf blades at 3 DPI, although it was only significant for CYCB1 (P=2.06 × 10 -2 ; Fig. 5A).Expression of the diploidy maintenance transcription factor MYB3R4 also showed a significant 3-fold increase in the infected distal leaf blade at 3 DPI (P=2.1 × 10 -3 ).The expression of the three genes subsequently decreased at 5 DPI, but MYB3R4 expression remained significantly higher in the distal infected leaf blade by ~1.2-fold compared to the control (P=2.69 × 10 -2 ).The expression of the S/SG2 marker CYCA3 in infected distal leaf blades increased from 3 DPI (P=0.66) to 5 DPI(P=0.435), and it was higher than in the control at 5 DPI, although this was not statistically significant.The transcript levels of CDKA1, CDKD1, and WEE1 in infected distal leaf blades either decreased or remained the same at 3 DPI and 5 DPI.
In infected proximal leaf blades, expression of CYCA3, CDKA1, and CDKB1 was higher than that of the control at 5 DPI, but not at a significant level (Fig. 5A).The expression of CYCA3, CDKA1, CDKB1, and CDKD1 increased from 3 DPI to 5 DPI.The expression of MYB3R4 was significantly lower than in the control at 5 DPI.The overall increase in the expression of a sub-set of cell-cycle marker genes alongside the decrease in MYB3R4 expression in the proximal leaf blade during the infection process, illustrates the ability of the fungus to induce a stress response while modulating host cell growth.
Exogenous MeJA reduced B. hordei infection in the proximal leaf blade (Fig. 4E) and therefore the expression of JA-biosynthesis and -signalling genes was examined to gain clues about the manipulation of the JA pathway by the pathogen during the infection process.In infected distal leaf blades the expression levels of the JA biosynthesis genes AOC and AOS were ~0.6-fold and ~2-fold higher (P=1.1 × 10 -16 ), respectively, than in the control at 3 DPI (Fig. 5B).Similarly, the expression of the JA-inducible genes JIP23 and JIP60 was ~1.4-fold higher (JIP23, P=1.44 × 10 -2 ).In contrast, the expression of the JA biosynthesis genes AOS (P=1.3 × 10 -5 ) and AOC (P=8.2 × 10 -3 ) and the JA-induced genes JIP60 (P=2.6 × 10 -5 ) was lowered significantly by B. hordei infection different sections of the sheath (S) and leaf blade (B) were taken at 7 DAS, as illustrated in Fig. 1A.Pr, proximal; Di, distal.Expression was determined by qRT-PCR relative to the reference genes Ubiquitin and GAPDH.The values for the untreated leaf and sheath controls (black bars) are shown as the log 2 fold-change relative to the expression in the distal blade; the results for MeJA and GA (red and green lines respectively) are the log 2 fold-change relative to the corresponding control value.Data are means (±SE) of three biological replicates (n=3 experiments, with 3-4 seedlings per experiment, with at least three technical replicates).One-way ANOVA was used to determine significant differences compared with the distal leaf section for the control values, and compared with the corresponding control value for the MeJA and GA treatments: *P<0.05,**P<0.01,***P<0.001.Genes associated with the cell cycle: CYCA3, CYCB1, CDKA1, CDKB1, CDKD1, WEE1, and MYB3R4.Genes associated with GA and JA biosynthesis and signalling: AOC, AOS, JAZ1, JIP60, JIP23, SLN1, and GID1.Genes associated with defence: PR5.
in the proximal leaf blade at 3 DPI compared to the uninfected control.The expression of the JA-signalling gene JAZ1 increased slightly in the infected proximal leaf blade from 3 DPI to 5 DPI.Compared to the control at 5 DPI, AOS expression was 1.4-fold higher (P=1.3 × 10 -5 ) in the infected distal leaf blade, AOC and JIP23 expression remained relatively unchanged, while JIP60 expression was significantly reduced (P=1.7 × 10 -2 ).From 3 DPI to 5 DPI, the expression of AOC, JIP23, JIP60, and JAZ1 decreased in the distal leaf blade.SLN1 has been shown to affect node growth, GA/JA interactions, and infection (de Vleesschauwer et al., 2016).In infected proximal leaf blades, SLN1 expression was lower at 3 DPI and 5 DPI than in the control whereas in infected distal leaf blades it was higher at 3 DPI but unchanged 5 DPI.Taken together, the decrease in the overall expression of JA-related genes from 3 DPI to 5 DPI in the infected distal leaf blade might be indicative  of an activated, but not maintained, host stress response at the earlier time-point.
Mildew Locus O (MLO, HORVU.MOREX.r2.4HG0342080) is a well-known susceptibility factor in powdery mildew infection (Devoto et al., 1999;Kusch and Panstruga, 2017) and PR5 is associated with resistance against it (Lambertucci et al., 2019).In our study, PR5 and MLO expression increased significantly by ~0.6-fold (P=2.03 × 10 -2 ) and 1.1-fold (P=1.2 × 10 -3 ), respectively, in infected distal leaf blades at 3 DPI relative to the control (Fig. 5C).At 5 DPI, PR5 expression remained significantly higher than in the control (~1.3-fold;P=2.0 × 10 -4 ), while MLO expression was lower, albeit not significantly (P=0.156).A proton pump in the plasma membrane, ATPase, regulates plant growth and JA signalling (Visnovitz et al., 2012), and its expression is enhanced by bacterial infection (Zhou et al., 2015).We found that ATPase expression was higher in the infected distal leaf blades than in the control at 3 DPI and 5 DPI (P=2.7 × 10 -2 ).Whether the inducibility of ATPase was the result of B. hordei manipulation of the host pump to facilitate its infection remains to be determined.

The size of nuclei increases in B. hordei-infected epidermal cells
The nuclear DNA content differed between the barley leaf blade and sheath (Fig. 1B), and no significant changes were observed following infection with B. hordei (Supplementary Fig. S2).We used fluorescence microscopy to determine the changes in nuclear size (Chandran et al., 2010(Chandran et al., , 2013;;Scheler et al., 2016) in infected epidermal cells adjacent to the B. hordei haustoria and in uninfected cells (Fig. 6A).The mean volume of the nucleus in uninfected epidermal cells was 527 µm 3 (median 440 µm 3 ) and this increased significantly by 40% in infected cells (P=1.0 × 10 -4 ; Fig. 6B, C).The nuclear enlargement in the cells adjacent to the B. hordei infection site was not accompanied by any apparent increase in their fluorescence (P=0.803),suggesting that the increased size of the nuclei in the infected epidermal cells did not result from higher DNA content.

Pre-treatment with MeJA or GA modifies gene expression patterns in the B. hordei-infected leaf blade
As MeJA pre-treatment reduced B. hordei infection in the proximal leaf blade (Fig. 4), the expression patterns of the same cell cycle, JA, and defence genes in infected leaf blades that were pre-treated with 50 µM MeJA for 24 h, was next examined, to gain insights into the phytohormone-mediated resistance mechanisms.Thus, the relative changes in gene expression were compared between MeJA pre-treated and B. hordei-infected proximal and distal leaf blades compared to the phytohormone treatment alone 2, 3, and 5 DPI (Fig. 7A-C).
The expression of the JA-biosynthesis gene AOS (P=1.6 × 10 -2 ) and the JA-inducible gene JIP60 at 2 DPI was lower in the proximal portion of the leaf blade after MeJA pre-treatment and B. hordei infection than in the MeJA-only control (Fig. 7B).In the distal leaf blade, the expression of the two JA-biosynthesis genes and the two JA-inducible genes was higher than in the control at all the time-points (AOC, P=5.5 × 10 -3 ; JIP23, P=2.2 × 10 -9 ).Relative to the control, AOS and AOC expression was similar in the MeJA+B.hordeitreated distal leaf blade at 2 DPI and 3 DPI, but AOC expression was markedly higher at 5 DPI (P=4.6 × 10 -10 ).
Relative to the control, the expression of JIP23 (P=9.0 × 10 - 4 ), JIP60, and JAZ1 (P=1.6 × 10 -2 ) was higher in the distal leaf blade at 5 DPI in the MeJA+B.hordei treatment (Fig. 7B).In the case of the proximal leaf blade, the expression of AOC and AOS was unchanged at 5 DPI, compared with MeJA pretreatment alone.The expression of JIP23 and JIP60 increased with time in MeJA+B.hordei-treated proximal leaf blades, and the same pattern was observed for SLN1 in the distal leaf blade.The increase in the expression of JA-associated genes in response to sustained B. hordei infection following MeJA pretreatment is indicative of the contribution of the latter to the attenuation of pathogen infection.
The expression of PR5 and MLO in the distal leaf blade increased with from 2 DPI to 5 DPI in the MeJA+B.hordei treatment, whereas their expression remained unchanged in the proximal leaf blade (Fig. 7C).
Thus, combining MeJA pre-treatment with B. hordei infection resulted in expression patterns of cell cycle-and hormone-associated genes that provide evidence for different host responses compared with MeJA or B. hordei treatment alone.
The existence of crosstalk between JAs and GA (reviewed by Pérez-Salamó et al., 2019) prompted the analysis of gene expression in the leaf blade following GA pre-treatment and subsequent B. hordei infection (Fig. 7D-F).Relative to pre-treatment with GA alone, the expression of CYCB1, CDKA1, CDKD1, MYB3R4, and WEE1 increased from 2 DPI to 3 DPI in the distal leaf blade of the GA+B.hordei treatment (Fig. 7D).In the proximal leaf blade, the expression of all the cell-cycle markers except for CYCB1 and the expression of MYB3R4 and WEE1 was generally decreased or unchanged from 2 DPI to 3 DPI.The expression of CYCA3 and WEE1 in the proximal leaf blade increased from 3 DPI to 5 DPI.Overall, the expression of all the cell cycle-related genes decreased in the distal leaf blade from 3 DPI to 5 DPI in the GA+B.hordei treatment.
Similar to the genes related to the cell cycle and the hormones, the expression of the defence-related PR5 and susceptibility factor MLO decreased with time from 2 DPI to 5 DPI in the distal leaf blade in the GA+B.hordei treatment (Fig. 7F).In the proximal leaf blade the expression of PR5 was generally low at all the time-points, while the expression of MLO remained relatively unchanged.Contrary to our expectations, whilst MLO expression in the distal leaf blade increased with time in the MeJA+B.hordei treatment (Fig. 7C), it decreased with time in the GA+B.hordei treatment (Fig. 7F).Hence, MeJA and GA treatment might antagonistically regulate barley defence responses following B. hordei infection.

Growth in barley leaves is spatially separated and differentially regulated by MeJA and GA
To identify and determine the extent of proliferation and endoreduplication along the barley leaf, flow cytometry was used to analyse the nuclear DNA content along the sheath and blade.A higher ratio of nuclei in the S/G2-M phase was observed in the sheath (Fig. 1B, C) and this was associated with elevated expression of cell-cycle genes (Fig. 3).Higher expression of MYB3R4 and WEE1 in the sheath reflected the presence of more proliferating and endoreduplicating cells, respectively, in the sheath compared to the leaf blade (Figs 3).A similar expression pattern for the inhibitory WEE1 was observed by Beemster et al. (2005).WEE1 might have a role in fine-tuning the cell cycle, inhibiting the mitotic cell cycle, and inducing endoreduplication/differentiation in response to environmental stress.The higher expression of the other markers associated with the cell cycle and proliferation could be reconciled within the global context of leaf development.In actively dividing meristematic cells, resources are predominantly directed into transcriptional regulation, cell proliferation, and protein synthesis to support cytoplasmatic growth (Noir et al., 2013;Nelissen et al., 2018;Loudya et al., 2021).Evidence of post-transcriptional regulatory mechanisms that are responsible for the differential activity of cell-cycle components is extremely limited in monocots.The trends that we observed for genes with a role in G2/M (such as CYCB1 and MYB3R4), with steeper down-regulation in comparison with those with a role in G1/S (or both G1/S and G2/M) such as CYCA3 and CDKA, were comparable to those previously observed in Arabidopsis (Beemster et al., 2005).Our results indicate spatial separation of growth-promoting processes in barley and are in line with previous observations in maize and wheat, where the cells transition from expanding to becoming mature (Nelissen et al., 2018;Loudya et al., 2021).Our study provides candidate  In the absence of phytohormone treatment and infection, the expression of cell-cycle markers is generally higher in the sheath compared to the distal (Di) leaf blade (Fig. 3), and this is associated with higher nuclear DNA contents in the proliferating sheath tissue (Fig. 1), reflecting the different developmental stages in the two modules of the leaf.
targets for further investigations that are paramount for research in cereals (Fig. 8).
Treatment with MeJA reduced leaf elongation and the progression of the cell cycle (Fig 1E, F), matching observations in other species (Noir et al., 2013;Świątek et al., 2002, 2004;Pauwels et al., 2008;Bömer et al., 2018).The frequency of nuclei in the G2/M phase decreased after treatment, indicating that the cell cycle was mainly arrested in G1 prior to the S transition.MeJA has previously been found to repress the onset of endoreduplication in Arabidopsis (Noir et al., 2013); however, in our study, effects of MeJA treatment on the frequencies of 8C and 16C nuclear DNA contents in barley were not evident, in line with the lower occurrence of endoreduplication in monocots relative to dicots (Barow and Meister, 2003).In contrast to MeJA, GA promoted leaf elongation and increased, albeit slightly, nuclear DNA contents of 4C and above (Fig 1E,F).This is consistent with previously observed increased mitotic activity and positive effects on endoreduplication in other plant species (Perrazza et al., 1998;Fabian et al., 2000;Yuxi et al., 2021).The opposing effects on the nuclear DNA content of MeJA and GA might reflect a broader control mechanism that balances growth by cell division and expansion.
Both JAs and GA regulate developmental switches that induce the transition from cell division to cell expansion.In this study, both phytohormones down-regulated the expression of genes associated with the cell cycle in the leaf (Figs 3, 7).Their effect was particularly evident in the basal region of the sheath (sections A, B in Fig. 1A) where cell division is more active.This phenomenon can be explained by the known negative effect of MeJA on cell-cycle activity (Świątek et al., 2004;Pauwels et al., 2008;Noir et al., 2013;Bömer et al., 2018) and with the role of GA in promoting cell differentiation and development at the expense of cell proliferation (Achard et al., 2009).In our study, the shift from cell division to differentiation was identified within the proximal sheath as supported by the strongest down-regulation of genes associated with the cell cycle by MeJA and GA.GA and JA interact synergistically and antagonistically with each other in seedling growth and resistance to pathogens through interactions between JAZ and DELLA proteins (Song et al., 2014).Such interactions suggest that JA-triggered growth inhibition could be fine-tuned by binding competition among JAZ, DELLA, and MYC2 proteins (Noir et al., 2013;Pérez-Salamó et al., 2019).
Higher expression of the JA-biosynthesis genes AOS and AOC in the sheath (Fig. 3) is in agreement with previous results for barley, where similar constitutive spatial and temporal expression has been observed (Maucher et al., 2000(Maucher et al., , 2004)).Higher expression levels of these JA biosynthesis genes also matches previously observed elevated content of JAs in lower portions of leaf blade sections (Maucher et al., 2000(Maucher et al., , 2004)).Similar hormone distribution along the growing maize leaf has been observed (Nelissen, 2018).The role of JIP60 in seedling development is not known, but its coordinated expression alongside AOS and AOC might suggest it has a role in leaf growth regulation.
The lack of further AOC and AOS induction in the sheath by MeJA and GA might reflect the existence of a feedback mechanism that keeps tight control over cell proliferation and preservation of meristem identity in the sheath.In the leaf blade, AOC and JIP60 expression are delayed compared to AOS (Maucher et al., 2000(Maucher et al., , 2004)).Whether this reflects the lack of activation of feedback mechanisms remains to be determined.The induction of PR5 in response to MeJA, correlating with that of AOC, JAZ1 and JIP60 suggests a differential ability to activate defence responses in the leaf blade relative to the sheath.
Interestingly, higher GA levels have also been reported in proximal leaf sections alongside JAs (Nelissen, 2018).The patterns of expression of SLN1 and GID1 (Fig. 3), however, were Expression of genes associated with JAs and GA is generally higher in the sheath and decreases towards the distal leaf blade, whereas JIP23, GID1, and PR5 expression is higher in the distal section.The expression of AOC, JAZ1, and JIP60 in the leaf blade increases following both MeJA and GA treatment.The expression of the cell cycle markers of AOS and SLN1 is reduced by the two hormones in both the sheath and leaf blade.GA does not have any effect on the expression of PR5.(B) The effects of B. hordei on gene expression.The expression of MYB3R4 (highlighted in bold) is already relatively high in the distal leaf blade prior to infection (Fig. 3), and together with CYCB1 its expression increases significantly by 3 days post-inoculation (DPI; Fig. 5).The expression of the JAs-related genes AOC, AOS, JIP23, and JIP60 as well as SLN1 is also higher in the distal leaf blade at 3 DPI.The expression of ATPase and PR5 is higher in the distal leaf blade at both 3 DPI and 5 DPI.MLO expression is higher only in the distal leaf blade at 3 DPI.At 5 DPI, MYB3R4 and AOS expression increases in the distal leaf blade.In the proximal leaf blade, a noticeable, albeit not significant, increase in CYCA3, CDKA1, CDKB1, and JAZ1 expression occurs at 5 DPI.Infection by B. hordei causes a reduction in the expression of CYCB1 at 3 DPI in the proximal section and of JIP60 in the distal section at 5 DPI.(C) Cumulative effects of pre-treatment with MeJA or GA combined with sustained B. hordei infection on cell cycle progression in the leaf blade (Fig. 7).Pre-treatment with MeJA negatively affects growth and arrests the cell cycle, affecting the transition checkpoints at G1/S and G2/M (Pauwels et al., 2008;Noir et al 2013;Bömer et al 2018).The increase in expression of the JAs-associated genes AOS, AOC, JAZ1, JIP23, and JIP60 might be associated with stabilization of the GA-signalling inhibitor SLN1 following MeJA treatment and B. hordei infection.The increase in expression of CYCB1 is in line with a previous study showing G2 arrest of Arabidopsis root cells and increased CYCB1;1 expression after γ-irradiation (Ricaud et al., 2007).The increased expression of WEE1, involved in the DNA damage response and G2 arrest, further reinforces the role of MeJA+B.hordei on cell cycle arrest.Pre-treatment with GA has the opposite effect to that of MeJA on cell cycle transitions and B. hordei infection in the leaf blade.The expression of AOS, AOC, JAZ1, and JIP23 is reduced following GA pre-treatment and B. hordei infection, whilst the expression of CYCA3, CYCB1, MYB3R4, and WEE1 is reduced.The reduction of PR5 and SLN1 expression following GA pre-treatment and B. hordei infection might contribute to increasing the susceptibility to B. hordei.
not in apparent agreement with our expectations.Active GA signalling via the GID1 receptor would result in degradation of DELLA proteins/SLN1, leading to the activation of the GA response (Qi et al., 2014).The decrease in expression of SLN1 in response to MeJA and GA might correlate with the induced expression of JAZ1 and/or the JA induction of GID1.However, as reports on the transcriptional regulation of SLN1 and GID1 are sparse, it remains to be determined whether the expression patterns observed in our work, correlate with posttranslational regulation of GA signalling components.Overall, the MeJA and GA expression signatures in the leaf reflected hormone crosstalk to maintain the cell metabolic homeostasis.

Distinct gene expression patterns are associated with both B. hordei infection strategy and host responses
Blumeria hordei is an obligate biotrophic pathogen that proliferates rapidly on the leaf surface (Both et al., 2005), thereby imposing metabolic demands on the host (Chandran et al., 2010(Chandran et al., , 2013)).The molecular and genetic interactions of this pathogen with both host (Kuhn et al., 2016) and non-host plants (Lipka et al., 2005;Kusch and Panstruga, 2017;Spanu and Panstruga, 2017;Panstruga and Moscou, 2020) have been extensively studied.In Arabidopsis, infection by another powdery mildew, Golovinomyces orontii, is correlated with the occurrence of endoreduplication, which potentially promotes the metabolic capacity of host cells at the infection site (Chandran et al., 2010).We therefore examined whether B. hordei induced endoreduplication in the barley leaf blade, and our results showed that epidermal cells of infected leaves had increased epidermal nuclear size but did not have any alteration in their nuclear DNA ploidy distribution (Fig. 6; Supplementary Fig. S2), suggesting that infection does not necessarily induce endoreduplication in the barley leaf blade.It is possible that B. hordei might alter the levels of nuclear envelop proteins and hence remodel the nuclear morphology, thereby affecting nuclear transport and signalling, and ultimately immune responses (Goto et al., 2021).
Remarkably, the cell-cycle genes examined, showed differential expression patterns in the proximal and distal leaf blade, and this was associated with different levels of hyphae formation in these sections ( Figs 4B, D, 5A, 8B).The repression of cell-cycle genes in the proximal leaf blade early during infection suggests that the pathogen might suppress cell proliferation, thereby triggering cell differentiation.Whether this is associated with metabolic reprogramming that satisfies the requirement of the biotrophic B. hordei remains to be determined.Alternatively, it could indicate the occurrence of a host defence mechanism to halt growth in response to B. hordei.During infection, the up-regulation of genes for cell-cycle progression could reflect a release from B. hordei-induced stress.This would be consistent with the absence of induction of the genotoxic stress-related CYCB1 (Ricaud et al., 2007).On the other hand, the greater susceptibility of the distal leaf blade to sustained B. hordei infection (i.e. 5 DPI) was associated with lower expression of genes for cell-cycle progression and higher expression of MYB3R4 (Fig. 5A), implying a further attempt of the pathogen to drive cell differentiation (Haga et al., 2007(Haga et al., , 2011) ) to its advantage for successful infection.In addition, CYCB1 down-regulation over time could be the result of reduced host perception of B. hordei-induced stress, perhaps as a consequence of pathogen activity.
JA signalling plays a vital role in plant defences (Pérez-Salamó et al., 2019).The increased expression of all the JA-associated genes examined, as well as that of SLN1, in the distal leaf blade following B. hordei infection 3 DPI (Figs 5B, 7B), indicated the activation of host stress responses concomitant with modulation of growth.These observations are in agreement with results previously obtained in both monocots and dicots, where pathogen infection has been shown to increase the expression of AOS, AOC, and downstream JA-signalling components such as JAZ1, COI1, PR14, JIP60, and MYB (Maucher et al., 2000(Maucher et al., , 2004;;Polesani et al., 2010;Cao et al., 2015;Guerreiro et al., 2016).Conversely, the reduced expression of some JA biosynthesis and signalling genes at 5 DPI might have been the result of pathogen suppression of the host defence response following its establishment.Thus, B. hordei might have evolved strategies to repress JA signalling, possibly through induction of the antagonistic SA (Achuo et al, 2004), and perhaps to promote GA signalling in order to facilitate exploitation of the barley leaf cellular machinery.
H + -ATPases are plasma-membrane (PM) pumps that establish cellular membrane potential in plants, which is crucial for controlling many transport processes (Palmgren, 2001).These pumps are actively regulated during plant immune responses and are important targets of pathogens (reviewed in Elmore and Coaker, 2011).Infection of tomato with Cladosporium fulvum and of barley with B. hordei increases H + -ATPase protein levels, resulting in the generation of proton gradients between the apoplast and the extrahaustorial space (Vera-Estrella et al., 1994;Szabo and Bushnell, 2001).Thus, manipulation of ATPases in the plasma membrane/apoplast by B. hordei might benefit its growth through increased nutrient transport.This is substantiated by the findings of Lambertucci et al. (2019), where increased abundance of V-type PM-ATPase protein levels following B. hordei infection was observed.
Increased expression levels of the well-characterized susceptibility gene MLO (Kusch and Panstruga, 2017;Devoto et al., 1999) in the distal leaf blade at 3 DPI coincided with the establishment of the infection; however, the establishment of the host response might be reflected by its lack of induction at 5 DPI (Figs 4B, D, 5C).Higher expression of PR5 correlated with B. hordei infection in the distal leaf blade, which agrees with previous observations by Lambertucci et al. (2019).The extent of B. hordei infection along the leaf blade might therefore be affected by the gradient of expression of JA-related and defence genes, contributing to shielding the proliferating cells in the sheath from the invading pathogen.Taken together, our results suggest that blocking or interfering with these B. hordei infection strategies could be used as a strategy to limit powdery mildew disease in barley and other plants.

Pre-exposure to MeJA or GA differentially regulates defence responses along the leaf blade
Stress-signalling hormones can orchestrate gene priming or transcriptional memory (Liu and Avramova, 2016).In Arabidopsis, MeJA treatment induces priming of the plant stress response (Noir et al., 2013), and it triggers developmental and stress responses generally in plants (Wu et al., 2008;Pérez-Salamó et al., 2019); however, the exact mechanism by which JAs modulates the responses to pathogen infection is not fully understood.Here, we attempted to examine the effects of exogenous MeJA as well as GA on B. hordei infection of barley leaf blade, in terms of the expression of genes related to the cell cycle and defence.The classical model for infection resistance states that the JA pathway is activated during necrotrophic infection (Díaz, et al., 2002;Ferrari et al., 2003;Zarate et al., 2007); however we found that exogenous MeJA also seemed to play a role in response to the biotrophic infection by B. hordei.This is consistent with a report that MeJA treatment protects Arabidopsis against the biotrophic pathogens Erysiphe cichoracearum and Peronospora parasitica (Zimmerli et al. 2004).
The localized up-regulation of cell cycle progression genes in the distal leaf blade following MeJA pre-treatment and subsequent B. hordei infection (Fig. 7A) could have resulted from host attempts to trigger a defence response, in agreement with other studies showing stress-inducibility of genes associated with the cell cycle (Ricaud et al., 2007, Vanderauwera et al., 2011;Waterworth et al., 2016;Shimotohno et al., 2021).The enhanced susceptibility of the distal leaf blade to B. hordei could underlie the effect of the pathogen to counteract the host response, exemplified by the induction of the growth repressor WEE1.This was substantiated by the higher up-regulation of the stress marker CYCB1 in the distal portion of the leaf blade compared to the proximal one.CYCB1 plays a crucial role in genotoxic stress responses, production of reactive oxygen species, and double-strand breaks in host-plant DNA following bacterial, fungal, or oomycete infection (Veylder et al., 2007;Song and Bent, 2014;Weimer et al., 2016;Chen et al., 2017).The effect of MeJA pre-treatment on the expression of this class of genes during infection was not evident in the lessinfected proximal leaf blade, highlighting the occurrence of different defence mechanisms along the leaf blade.Whether this can be accounted for by differences in cell development status and competence to respond to MeJA treatment along the leaf blade remains to be determined.
We also examined the expression of JA-associated genes in leaf blades pre-treated with MeJA and then infected with B. hordei, and we found that MeJA pre-treatment significantly increased their expression during infection (Fig. 7B).The activation of the complex ribosome-inactivating protein JIP60 in the MeJA pre-treated leaf blade agrees with its role in JA-dependent plant immunity (Przydacz et al., 2020).Overall, the expression of JAs related genes increased with time, albeit with different levels of magnitude in the proximal and distal leaf sections.The reduced B. hordei infection in the proximal section (Fig. 4D) was associated with a stronger induction response of these genes that might have resulted in more effective activation of stress signalling than in the distal section where the expression was higher to start with.The increased expression of the DELLA gene SLN1 following infection suggested a possible concurrent reduction in GA signalling and an increase in JA signalling.Whether such a mechanism is implemented to prioritize defence over growth by suppressing the GA signalling cascade in barley remains to be determined.
The expression of PR5 was significantly increased by MeJA pre-treatment plus B. hordei in the distal leaf blade at 5 DPI (Fig. 7C), similar to its induction following B. hordei infection alone (Fig. 5).However, the MeJA pre-treatment might have delayed the onset of the infection, as evidenced by the later induction of PR5.The absence of induction of PR5 in the proximal section might reflect the existence of lower susceptibility to infection in this section of the leaf (Fig. 4).Overall, our results suggested that MeJA-induced priming might contribute to the activation of defence responses in barley leaf cells, although it was notable that the mechanism was differently executed along the length of the leaf blade.
JAs play the main role in plant defence responses, while GAs predominantly regulate plant growth and development.Hence, JA-GA crosstalk finely balances the trade-off between growth and stress responses (Hou et al., 2010;Pérez-Salamó et al., 2019).The antagonistic effects of JAs and GAs on each other's signalling output to balance growth and defence has been reported previously (Yang et al., 2012(Yang et al., , 2019;;Um et al., 2018).The down-regulation of the JA-related genes, the DELLA SLN, the stress-marker CYCB1, and defence-related PR5 by pre-treatment with GA followed by B. hordei infection in the distal leaf blade (Fig. 7D-F) was indicative of a GA-mediated reduced ability to perceive the stress whilst continuing growth, leading to enhanced susceptibility relative to the proximal section (Fig. 4E).The opposite expression patterns that were observed for the different categories of genes in the infected plants following pre-treatment with JA or GA (Fig. 7) implies a tight control on growth regulation by the two phytohormones during the pathogen infection.

Conclusions
Our study has demonstrated that the barley sheath and the leaf blade possess different growth capacities, intrinsic abilities to respond to hormones, and susceptibilities to B. hordei infection.The differential growth capacity is further fine-tuned by MeJA and GA by regulating genes related to the cell cycle, hormones, and defence.Different gene expression patterns along the barley leaf blade are also associated with both the B. hordei invasion strategy and host responses during sustained infection, and are further modified by priming with the two hormones in an antagonistic manner (Fig. 8).Ultimately, our results suggest that tight control is operated on cell proliferation and the preservation of meristem identity to guarantee plant survival.
Overall, the individual and the combination of treatments applied, as well as the analysis of leaf sheath and blade sections, has demonstrated the need for a more systematic approach to understand the underlying mechanisms that regulate the growth-stress trade-off.Identification of the regulators of the differential spatial responses to pathogens in the leaf and the correlation with growth processes will be instrumental for engineering plant resistance to B. hordei and other pathogens that show organ, tissue, and cell specificity.

Fig. 1 .
Fig. 1.Frequencies of different nuclear DNA content and cell-cycle stages in the sheath and leaf blade of barley seedlings, and effects of MeJA and GA treatments.(A-C) Seedlings of the cultivar Golden Promise were grown hydroponically and sampled at 7 days after sowing (DAS).Di, distal; Pr, proximal.(A) Samples that were taken from different sections of the sheath and leaf blade.(B) Frequencies of nuclear DNA content (ploidy levels) in the different sections.(C) Frequencies of cells in different stages of the cell cycle.(D-F) Seedlings of Golden Promise (GP) were grown hydroponically.For the JA treatment, 50 µM MeJA was added to the solution at 4 DAS.For the GA treatment, either 0.1 µM GA 3 or 10 µM GA 3 was present in the solution from the time of sowing.Samples of whole sheaths and whole leaves were taken at 7 DAS.(D) Representative images of control and treated seedlings at 7 DAS.(E) Relative length of the sheath and leaf blade as a percentage of the control.Data are means (±SE), of a minimum of three biological replicates.(F) Frequencies of nuclear DNA contents in the different treatments.Data in (B, C, F) are means (±SE) from 15 000-20 000 nuclei counted from three biological replicates, each of which consisted of 3-4 pooled seedlings.Significant differences compared with the untreated control in (E, F) were determined using one-way ANOVA: *P<0.05,**P<0.01,***P<0.001.

Fig. 4 .
Fig. 4. Effects of barley culture conditions on formation of Blumeria hordei hyphae on the leaf blade, and effects of pre-treatment with MeJA on B. hordei infection.(A) Barley seedlings were grown either in soil or hydroponically for 7 days after sowing (DAS), after which detached leaves from both were infected with B. hordei (15-40 spores mm -2 ), and attached leaves on hydroponically grown seedlings were also infected.(B) Representative images of detached blades of hydroponically grown plants at 0 days post-inoculation (DPI) and 5 DPI.Di, distal blade section; Pr, proximal blade section.(C) Microscopic images of B. hordei structures at 2DPI at 2 DPI on the leaf blade.Asterisks indicate appressoria (top two panels) and the arrow indicates hyphae formation (bottom panel).(D) Hyphae formation on the proximal and distal sections of the leaf blades shown in (A), expressed as the percentage of hyphae formed relative to the total number of ungerminated conidia, appressoria, and hyphae-forming conidia at 2 DPI.Data are means of at least three biological replicates.(E) Seedlings of the cultivar Golden Promise were grown hydroponically.For the MeJA treatment, 50 µM MeJA was added to the solution for 24 h at 5-6 DAS, and then the medium was substituted with half-strength Hoagland's solution No 2. Samples were taken at 7 DAS.For the GA treatment, 10 M GA 3 was present in the solution from the time of germination.At 7 DAS, leaves were detached and infected with B. hordei, and hyphae formation in the proximal and distal sections was assessed at 2 DPI.(D, E) n=6 experiments with at least three leaves per experiment.Different letters indicate significant differences among means as determined using one-way ANOVA and Tukey's post-hoc test (P<0.05).

Fig. 5 .
Fig. 5. Effects of Blumeria hordei infection on the expression of genes associated with the cell cycle, JAs, GA, and plant defence in barley leaf blades.Detached leaves from seedlings at 7 days after sowing (DAS) were infected with B. hordei and samples from the proximal (Pr) and distal (Di) sections were collected for analysis at 3 d post-inoculation (DPI) and 5 DPI.Expression was determined using qRT-PCR with GAPDH as the reference gene.For each gene analysed, the results are presented as the fold-change in log 2 expression in the infected sample compared to the corresponding control sample.(A) Genes associated with the cell cycle, (B) genes associated with JAs and GA, and (C) genes associated with plant defence.Data are means (±SE) of four biological replicates, each consisting of samples pooled from three leaves.Significant differences in the fold-change in expression were determined using one-way ANOVA: *P<0.05,**P<0.01,***P<0.001.

Fig. 6 .
Fig. 6.Effect of Blumeria hordei infection on barley epidermal cell nuclei size.Primary leaf blades of the cultivar Golden Promise were detached and infected with B. hordei.After 2 d, tissue samples were cleared and stained with propidium iodide (PI) and DAPI.(A, B) Representative images and corresponding diagrams of epidermal cell nuclei of (A) an uninfected leaf and (B) of a leaf infected with B. hordei, showing an appressorium (arrowhead).(C) Nuclear volumes of epidermal cells with (+) or without (-) infection by B. hordei (Bgh).The box-plots show results from a minimum of three biological replicates with 3-4 leaves per experiment.A total of 114 control nuclei and 83 infected nuclei were imaged.The significant difference between the infected and uninfected samples was based on a linear regression model (P=0.0023).

Fig. 7 .
Fig. 7. Antagonistic effects of MeJA and GA pre-treatment on the expression of genes associated with the cell cycle, JAs, GA, and plant defence in barley leaf blade following Blumeria hordei infection.Seedlings of barley Golden Promise were grown hydroponically.For the MeJA treatment, 50 µM MeJA was added to the solution at for 24 h at 5-6 days after sowing (DAS), and then the media substituted with 1/2 strength Hoagland's solution No 2 and sampled at 7 DAS.For the GA treatment, 10 µM GA 3 was present in the solution from the time of germination.At 7 DAS, leaves were detached and infected with B. hordei, and samples from the proximal (Pr) and distal (Di) sections taken 2, 3, and 5 days post-inoculation (DPI).Expression was determined using qRT-PCR with GAPDH as the reference gene.For each gene analysed, the results are presented as the fold-change in log 2 expression in the hormone + B. hordei (Bgh) treatment compared with the corresponding sample from the hormone-only treatment.(A) Genes associated with the cell cycle, (B) genes associated with JA and GA, and (C) genes associated with plant defence.(A-C) Seedlings pre-treated with MeJA.(A) Genes associated with cell cycle, (B) genes associated with JAs and GA, and (C) genes associated with plant defence.(D-F) Seedlings pre-treated with GA 3 .(D) Genes associated with the cell cycle, (E) genes associated with JAs and GA, and (F) genes associated with plant defence.Data are means (±SE) of three biological replicates, each of which consisted of samples pooled from three leaves.Significant differences in the fold-change in expression were determined using one-way ANOVA: *P<0.05,**P<0.01,***P<0.001.

Table 1 .
Frequencies of nuclei exhibiting 2C-16C DNA content and cell-cycle transition states in barley sheath and leaf blade, and effects of MeJA and GA treatments