Active network of antioxidants as a key element of successful barley microspore transition from gametophytic to embryogenic development

Active network of antioxidants as a key element of successful barley microspore transition from gametophytic to embryogenic development Anna Nowicka1, Zbyněk Milec2, Monika Krzewska1, Przemysław Kopeć1, Agnieszka Springer1, Ewa Dubas1, Iwona A. Żur1* 1The Franciszek Górski Institute of Plant Physiology, Polish Academy of Sciences, Krakow, Poland, 2Ustav vyzkumu globalni zmeny Akademie ved Ceske republiky, Brno, Czechia Submitted to Journal: Frontiers in Plant Science Specialty Section: Plant Cell Biology Article type: Original Research Article Manuscript ID: 1735720 Received on: 30 Oct 2025 Journal website link: www.frontiersin.org In review Scope Statement Our manuscript investigates microspore embryogenesis (ME) as a model of plant cell totipotency, focusing on the redox-based regulation of embryogenic reprogramming. By combining transcriptomic profiling with functional categorization of antioxidant and thiol-redox systems, we reveal genotype-specific redox dynamics underlying cellular fate transition. The study provides novel insights into oxidative signaling and redox homeostasis during stress-induced cell reprogramming, which fits directly within the scope of *Frontiers in Plant Science*—**Plant Cell Biology** specialty section. Conflict of interest statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest Credit Author Statement Agnieszka Springer: Formal Analysis, Writing – review & editing. Anna Nowicka: Formal Analysis, Visualization, Writing – original draft. Ewa Dubas: Formal Analysis, Writing – review & editing. Iwona Anna Żur: Conceptualization, Funding acquisition, Project administration, Writing – original draft. Monika Krzewska: Formal Analysis, Writing – review & editing. Zbyněk Milec: Data curation, Formal Analysis, Writing – review & editing. Przemysław Kopeć: Investigation, Writing – review & editing. Keywords Antioxidant Defense, Hordeum vulgare, microspore embryogenesis, Redoxhomeostasis, RNA-Seq Abstract Word count: 218 Microspore embryogenesis (ME) relies on the cellular reprogramming of the default gametophytic developmental pathway, which normally directs microspores toward pollen formation, into an embryogenic pathway that leads to the development of embryos and, subsequently, haploid or doubled haploid (DH) plants. To test how redox control underpins this switch, we have carried out an extended analysis of previously published RNA-seq data from two barley cultivars differing in ME competence (Igri, responsive; Golden Promise, recalcitrant) across four early induction stages (0–III). A curated set of 472 antioxidant/redox genes—core detoxification enzymes, the ASC–GSH cycle, TRX/GRX/PRX systems and GSTs—was examined. The analysis revealed that the expression of antioxidative defense genes is dynamically modulated during ME induction, underscoring the importance of redox homeostasis in successful microspore reprogramming. Both cultivars shared a late (stages II– III) program with increased SODs, selected CAT/GPX genes, rising MDHARs, deployment of specific TRX/GRX/PRX members and broad GST upregulation. Divergence emerged during progression: Igri showed a pronounced stage-III rise of GRs and targeted TRX/GRX/PRX transcripts, together with stronger activation of multiple GSTs. These stage-restricted, cultivar-biased signatures support a model in which reinforcing ASC–GSH recycling and thiol-redox hubs sustains H₂O₂ signaling while limiting oxidative damage. Targeting MDHARs, GRs, selected TRX/GRX/PRX genes, and GST subsets could improve ME efficiency and accelerate the integration of DH technology into modern crop breeding programs. Funding information This work was supported by the National Science Centre in Poland (grant 2015/18/M/NZ3/00348 to Iwona Żur). Zbyněk Milec was supported by the Ministry of Education, Youth and Sports of the Czech Republic (MEYS ČR) through the project AdAgriF (CZ.02.01.01/00/22_008/0004635). Funding statement The author(s) declare that financial support was received for the research and/or publication of this article. In review Ethics statements Studies involving animal subjects Generated Statement: No animal studies are presented in this manuscript. Studies involving human subjects Generated Statement: No human studies are presented in the manuscript. Inclusion of identifiable human data Generated Statement: No potentially identifiable images or data are presented in this study. Data availability statement Generated Statement: The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material. Generative AI disclosure No Generative AI was used in the preparation of this manuscript. In review 1 Research article Active network of antioxidants as a key element of successful barley microspore transition from gametophytic to embryogenic development Anna Nowicka1,2, Zbyněk Milec3, Monika Krzewska1, Przemysław Kopeć1, Agnieszka 1 Springer1, Ewa Dubas1, Iwona Żur1* 2 3 1The Franciszek Górski Institute of Plant Physiology Polish Academy of Sciences, 4 Niezapominajek 21, 30-239 Kraków, Poland 5 6 2Centre of Plant Structural and Functional Genomics, Institute of Experimental Botany of the 7 Czech Academy of Sciences, Šlechtitelů 31, 779 00 Olomouc, Czech Republic 8 9 3Global Change Research Institute Czech Academy of Sciences, Bělidla 986/4a, 603 00 Brno, 10 Czech Republic 11 12 * Correspondence: 13 Corresponding author: Iwona Żur 14 [email protected] 15 16 Keywords: Antioxidant defense, Hordeum vulgare; Microspore embryogenesis; Redox 17 homeostasis; RNA-seq 18 19 Abstract 20 Microspore embryogenesis (ME) relies on the cellular reprogramming of the default 21 gametophytic developmental pathway, which normally directs microspores toward pollen 22 formation, into an embryogenic pathway that leads to the development of embryos and, 23 subsequently, haploid or doubled haploid (DH) plants. To test how redox control underpins this 24 switch, we have carried out an extended analysis of previously published RNA-seq data from 25 two barley cultivars differing in ME competence (Igri, responsive; Golden Promise, 26 recalcitrant) across four early induction stages (0–III). A curated set of 472 antioxidant/redox 27 genes—core detoxification enzymes, the ASC–GSH cycle, TRX/GRX/PRX systems and 28 GSTs—was examined. The analysis revealed that the expression of antioxidative defense genes 29 is dynamically modulated during ME induction, underscoring the importance of redox 30 homeostasis in successful microspore reprogramming. Both cultivars shared a late (stages II– 31 III) program with increased SODs, selected CAT/GPX genes, rising MDHARs, deployment of 32 specific TRX/GRX/PRX members and broad GST upregulation. Divergence emerged during 33 progression: Igri showed a pronounced stage-III rise of GRs and targeted TRX/GRX/PRX 34 transcripts, together with stronger activation of multiple GSTs. These stage-restricted, cultivar35 biased signatures support a model in which reinforcing ASC–GSH recycling and thiol-redox 36 hubs sustains H₂O₂ signaling while limiting oxidative damage. Targeting MDHARs, GRs, 37 selected TRX/GRX/PRX genes, and GST subsets could improve ME efficiency and accelerate 38 the integration of DH technology into modern crop breeding programs. 39 40 1. Introduction 41 In review 2 Every living cell operates as a dynamic network of biochemical reactions governed by the 42 cellular redox state. This balance—maintained through the production and removal of reactive 43 oxygen, nitrogen, and sulfur species (ROS, RNS, and RSS)—is essential for homeostasis and 44 underpins virtually all life processes. In plants, redox signals shape growth and development 45 by modulating photosynthesis, respiration, and hormone signaling, as well as the activities of 46 transcription factors and stress-related enzymes. Because plants are constantly exposed to 47 environmental stress, rapid and accurate redox signalling is critical for survival. Redox cues 48 have also been proposed as triggers of cellular reprogramming, including the shift from 49 gametophytic to embryogenic development. 50 Our earlier studies revealed a central role for ROS in the induction of microspore 51 reprogramming; whereby immature pollen grains are redirected towards embryogenic 52 development (Żur et al., 2014a; 2021a). The resulting embryo-like structures (ELSs) have the 53 ability to regenerate into haploid and doubled haploid (DH) plants, highly valuable for breeding 54 purposes and basic research. Microspores also represent an attractive target for genetic 55 engineering, including genome editing, because modifications introduced into haploid cells can 56 be stably fixed by chromosome doubling. However, the efficiency of microspore 57 embryogenesis (ME) is strongly genotype-dependent and often varies even between closely 58 related genotypes (Krzewska et al., 2012; Żur et al., 2014a). Trial-and-error approaches to 59 improve effectiveness of ME are therefore laborious, highlighting the need to better understand 60 the underlying molecular mechanisms. 61 Advances in RNA sequencing (RNA-seq) have enabled detailed exploration of 62 transcriptional networks regulating developmental reprogramming. Earlier studies provided 63 insights into the transcriptomic changes associated with ME induction in barley (Bélanger et 64 al., 2018; 2020). Building on this, our recent work compared two cultivars that differ strongly 65 in embryogenic potential, providing a detailed map of the gene networks that orchestrate the 66 shift in cellular machinery during ME induction (Nowicka et al., 2024). Based on our previous 67 work in triticale (Żur et al., 2014a), we postulated a role for ROS in ME induction, a hypothesis 68 supported by growing evidence. Tight control of redox homeostasis therefore appears to be a 69 prerequisite for efficient microspore reprogramming. It is mediated by antioxidant and redox 70 systems that regulate transfer of electrons from donor molecules to target proteins. Within this 71 framework, antioxidant defense represents a key regulatory layer that scavenges reactive 72 molecules and reduces oxidized substrates to protect microspores, while ROS themselves 73 function as signaling molecules involved in growth, development, and stress adaptation. 74 Recently published data highlight the complexity of the interactions among ROS, 75 transcriptional and epigenetic regulators, plant hormones, metabolites, and suggest the potential 76 mechanisms underlying ROS-mediated effects (Auverlot et al., 2024; Karpinska and Foyer, 77 2024; Lin et al., 2025). 78 Core antioxidant defenses include SUPEROXIDE DISMUTASES (SODs), 79 CATALASES (CATs), and GLUTATHIONE PEROXIDASES (GPXs), which detoxify 80 superoxide anion (O₂•–) and hydrogen peroxide (H₂O₂) (Fig.1A) (Mittler, 2002). The electrons 81 required for ROS reduction are supplied by metal ions, which serve as essential cofactors for 82 enzymes such as SODs and CATs. In the case of peroxidases, which require electron donors, 83 these electrons can be provided by ASCORBATE (ASC), THIOREDOXINS (TRXs) or 84 REDUCED GLUTATHIONE (GSH), all of which also participate directly in redox reactions. 85 Additional protection is provided by the ascorbate–glutathione cycle (ASC–GSH), where 86 enzymes including MONODEHYDROASCORBATE REDUCTASE (MDHAR), 87 DEHYDROASCORBATE REDUCTASE (DHAR), GLUTATHIONE REDUCTASE (GR), 88 and ASCORBATE PEROXIDASE (APX) recycle oxidized forms of ASC and glutathione 89 (glutathione disulfide, GSSG) (Foyer and Noctor, 2011). Glutathione also protects proteins 90 against irreversible oxidation through S-glutathionylation, catalysed by GLUTATHIONE S91 In review 3 TRANSFERASES (GSTs). Many GSTs display also peroxidase activity (Noctor et al., 2024 92 and references therein). Other important electron transmitters include GLUTAREDOXINS 93 (GRXs), PEROXIREDOXINS (PRXs) and TRXs which catalyze reversible disulfide bond 94 formation and protect central metabolic pathways (Sevilla et al., 2023). Notably, the barley 95 genome harbors multi-member gene families for every antioxidant class (Fig. 1B) (Monat et 96 al., 2019). 97 Here, we profile the transcriptional antioxidant network underpinning microspore ME 98 in barley by an extended analysis of previously published RNA-seq data from microspores and 99 microspore-derived multicellular structures sampled across four early ME stages in two 100 cultivars with contrasting embryogenic competence (Igri, responsive; Golden Promise, 101 recalcitrant (Nowicka et al., 2024). We delineate antioxidant pathways, identify stage-specific 102 markers of reprogramming, and nominate candidate regulators whose expression distinguishes 103 the superior ME efficiency of Igri from Golden Promise. Consistent with this, we observe 104 pronounced cultivar-dependent differences in antioxidant defense gene expression that likely 105 contribute to divergent ME responsiveness and may tip the balance between successful 106 reprogramming and stress-induced cell death. Establishing causality will require targeted 107 experiments to determine whether reduced antioxidant activity is a primary driver of microspore 108 mortality or a secondary consequence of broader metabolic and physiological reconfiguration 109 during ME. 110 111 2. Materials and methods 112 113 Plant material, ME-induction stages, and RNA-seq data 114 This study builds on previously published RNA-seq data (Nowicka et al., 2024) from two barley 115 (Hordeum vulgare L.) cultivars: Igri (winter type; HOR 10596; responsive to ME) and Golden 116 Promise (spring type; HOR 16645; recalcitrant to ME). Full details of plant cultivation, ME 117 induction, sample isolation for RNA-seq, and data processing are provided in Nowicka et al., 118 (2024). 119 ME-induction stages (Fig. 1C): 120 Stage 0: Microspores isolated from freshly harvested tillers. 121 Stage I: Microspores isolated from anthers pre-treated with 0.4 M mannitol for 48 h at 21 °C. 122 Stages II–III: Microspores isolated from anthers cultured in KBP medium (Kumlehn et al., 123 2006) for 24 h (stage II) or 48 h (stage III) following the same mannitol pre-treatment. 124 Because Golden Promise is recalcitrant to ME induction, biochemical stimulation was 125 provided by co-culture with immature pistils (+p) of wheat cv. Bobwhite (Lippmann et al., 126 2015). Longitudinally bisected pistils (three halves per ml KBP medium) were added to isolated 127 Golden Promise microspores, and stage II–III RNA-seq samples were collected after co-culture. 128 Igri did not undergo co-culture. ME-induction efficiency and sample purity were assessed 129 microscopically prior to RNA-seq (Daghma et al., 2014). Samples for RNA-seq were collected 130 using a mannitol/maltose density-gradient method. The gradient was applied once for stage 0 131 and twice for stages I–III; stage III fractions were additionally sieve-filtered. 132 Bicellular pollen grains produced by asymmetric division were used as a gametophytic 133 control. Pollen was isolated from freshly harvested tillers: cells were first enriched by a 134 mannitol/maltose density gradient and then individually picked using a glass micropipette 135 mounted on a micromanipulator under an inverted microscope, coupled to a microinjector for 136 precise aspiration. These RNA-seq data are unpublished (Kopeć P., Nowicka A., Żur I. et al.,). 137 The RNA-seq dataset is publicly available in the NCBI Gene Expression Omnibus 138 (GSE233486). Expression quantification was performed against the H. vulgare cv. Morex 139 reference genome v2 (Monat et al., 2019). 140 141 In review 4 Antioxidant gene selection and classification 142 To investigate redox‐associated transcriptional responses during ME induction, we have 143 completed the RNA-seq dataset with a targeted set of antioxidant and redox-related genes. 144 Candidate functions were curated manually from biochemical pathway knowledge and primary 145 literature (Mittler, 2002; Foyer and Noctor, 2011; Sevilla et al., 2023). In total, 472 genes (Fig. 146 1B) were curated and assigned to four functional categories: 147 i) core antioxidants: SUPEROXIDE DISMUTASES (SODs), CATALASES (CATs), 148 GLUTATHIONE PEROXIDASES (GPXs) 149 ii) ASC–GSH cycle genes: ASCORBATE PEROXIDASES (APXs), 150 MONODEHYDROASCORBATE REDUCTASES (MDHARs), DEHYDROASCORBATE 151 REDUCTASES (DHARs) and GLUTATHIONE REDUCTASE (GRs) 152 iii) redox-regulated thiol genes: THIOREDOXINS (TRXs), THIOREDOXIN REDUCTASE 153 (TRXRs), GLUTAREDOXINS (GRXs), PEROXIREDOXINS (PRXs) 154 iv) genes involved in S-glutathionylation: GLUTATHIONE S-TRANSFERASES (GSTs) 155 156 Expression analysis and marker gene criteria 157 Expression values (FPKM) were taken from the processed dataset (Nowicka et al., 2024). 158 Stage-specific markers were genes that, at a given induction stage, met all of the following 159 criteria: (i) FPKM > 10 in both cultivars; (ii) no significant cultivar effect at that stage (FDR160 adjusted P > 0.05); (iii) log₂ fold change (log₂FC) > 2 relative to the subsequent stage; and (iv) 161 absent or very low expression in the gametophytic control (bicellular pollen, P). Cultivar162 specific markers were genes that, at a given stage, met: (i) FPKM > 5; (ii) a significant 163 difference between cultivars (FDR-adjusted P < 0.05); (iii) |log₂FC| > 2 relative to the other 164 cultivar at the same stage; and (iv) absent or very low expression in P. Genes that did not meet 165 these thresholds but showed directionally consistent patterns are reported as trend-level 166 candidates for stage or cultivar specificity. 167 168 Data analysis and visualization 169 Gene filtering, selection, and summarization were performed in R (v4.2.2) using custom scripts. 170 Heatmaps were generated with Heatmapper (http://heatmapper.ca/expression/). Matrix bubble 171 charts were produced in R using ggplot2. 172 173 3. Results 174 175 Cytological composition across ME-induction stages 176 Using the published dataset of Nowicka et al., (2024), we assessed oxidative-stress-related 177 transcriptomic responses across four ME-induction stages in two barley cultivars, Igri 178 (responsive) and Golden Promise (recalcitrant): stage 0 (untreated), stage I (48 h in 0.4 M 179 mannitol at 21 °C), and stages II–III (24 h and 48 h after transfer to KBP, respectively). Each 180 stage comprised mixed cell populations with cultivar-specific composition (Fig. 1C; Fig. S1). 181 At stage 0, Igri was enriched for uninucleate microspores (73%) with fewer bicellular structures 182 after symmetric division (17%). Golden Promise contained fewer uninucleates (14%) and more 183 bicellulars (21%) whereas a high proportion of microspores remained unidentified. Stage I 184 showed comparable reprogramming (≈30% symmetric divisions), but uninucleates remained 185 more frequent in Igri (68%) than in Golden Promise (22%). Across stages II–III, Golden 186 Promise maintained ≈30% bicellular structures, while multicellular structures—arising from 187 continued symmetric divisions—increased from 3% at stage II to 18% at stage III. In Igri, 188 bicellulars remained ≈40% across stages II–III, with multicellular structures at 12% in stage II 189 and 6% in stage III. These differences were more pronounced in cultures lacking the stimulatory 190 effect of co-cultured pistils from the highly embryogenic wheat cultivar Bobwhite (−p; Fig. 1C; 191 In review 5 Fig. S1). This staged cytology provides the framework for the antioxidant/redox transcriptomic 192 analyses that follow. 193 194 Core antioxidant enzymes were reprogrammed in a stageand cultivar-dependent manner 195 We first profiled genes encoding the core antioxidant enzymes that constitute the primary 196 defense against ROS — SODs, CATs and GPXs (Fig.2). Of 21 annotated SOD genes, 15 were 197 transcriptionally active (FPKM > 0) in at least one induction stage, together with five of nine 198 CATs and all five GPXs (Fig. S2A). SODs spanned a wide expression range: in Golden Promise, 199 most transcripts were moderate (10–100 FPKM) with few highly expressed (>100 FPKM), 200 whereas in Igri expression was more evenly split between low (1–10 FPKM) and moderate 201 levels. CATs were generally weakly expressed in both cultivars, while GPXs were consistently 202 stronger, predominantly at moderate–high levels (Fig. S3A). Most expressed SODs belonged 203 to Cu/Zn-SOD orthogroups (Arabidopsis CSD1/2/3), with fewer Fe-SODs and a single 204 mitochondrial MnSOD, two CATs aligned with CAT2 and several GPXs with GPX1/6 205 (Supplementary Dataset 1). 206 Family-level averages revealed distinct trends (Fig. 2A). Mean SOD expression ranged 207 from ~40 FPKM (Igri, stage II) to ~60 FPKM (Golden Promise, stage I), with maxima at stage 208 0 in Golden Promise and stage III in Igri; bicellular pollen (P; gametophytic control) showed 209 SOD levels comparable to induction stages. CATs averaged 12–28 FPKM and peaked at stage 210 III in both cultivars. GPXs exceeded SODs and CATs overall, tended to be higher in Golden 211 Promise, and peaked at stage III (~180 FPKM). In contrast to induction stages, CAT and GPX 212 transcripts were low in mature pollen. 213 Gene-level profiles highlighted discrete regulatory modes (Fig. 2B,C). Among SODs, 214 some transcripts were induced early (stages 0–I; e.g. MnSOD 215 HORVU.MOREX.r2.2HG0173140.1; Cu/Zn-SOD HORVU.MOREX.r2.2HG0154740.1), 216 whereas others rose later (stages I–III; e.g. Cu/Zn-SODs HORVU.MOREX.r2.4HG0329510.1 217 and HORVU.MOREX.r2.6HG0513460.1—the latter strongly upregulated in both cultivars at 218 stages II–III with FPKM > 10 but not detected in mature pollen, supporting its candidacy as an 219 ME-stage marker). Additional cultivar specificity was evident: Cu/Zn-SOD 220 HORVU.MOREX.r2.7HG0573050.1 was moderate at stages 0–II in both cultivars but became 221 Igri-specific at stage III, and HORVU.MOREX.r2.3HG0190910.1 was confined to Igri. 222 Three CATs showed clear stage specificity: HORVU.MOREX.r2.1HG0067700.1 at 223 stage 0 (uninucleate microspores), HORVU.MOREX.r2.2HG01094980.1 at stage I (osmotic 224 treatment), and HORVU.MOREX.r2.4HG0341470.1 at stage III, with higher expression in 225 Golden Promise. Another CAT (HORVU.MOREX.r2.7HG0623480.1) increased progressively 226 across Stages II–III in both cultivars and served as an II–III marker. All CATs were very low in 227 mature pollen (Fig. 2B,C). GPXs also exhibited stageand cultivar-dependent regulation: 228 HORVU.MOREX.r2.4HG0304790.1 peaked at stages I and III, and 229 HORVU.MOREX.r2.2HG0156400.1 marked Golden Promise at stage III. As with CATs, GPX 230 transcripts were low in mature pollen. 231 Collectively, core antioxidant genes show dynamic regulation during ME, with stage232 and cultivar-specific patterns that distinguish the responsive Igri from the recalcitrant Golden 233 Promise. 234 235 ASC–GSH cycle genes showed late MDHAR rise and stronger GR induction in responsive 236 cultivar 237 We next analyzed APX, MDHAR, and GR families (Fig.3). DHAR was not annotated. Most 238 genes were expressed (APX: 4/5; MDHAR: 5/7; GR: 2/2; 14 total; Fig. S2B), mapping to 239 orthogroups containing the corresponding Arabidopsis genes (Supplementary Dataset 1). 240 At the family level, APXs dominated transcript abundance and were relatively stable 241 In review 6 across stages and in pollen in both cultivars (≈300–460 mean FPKM; Fig. 3A, Fig. S3B). 242 MDHARs increased progressively from stage 0 to stage III, particularly in Igri (up to ~77 243 FPKM), with pollen lower than induction stages. GRs were expressed at moderate levels (~60 244 FPKM) across ME stages and in pollen, except for a marked rise in Igri at stage III (~115 245 FPKM). 246 Gene-specific patterns reinforced these trends (Fig. 3B, C). Within APX, 247 HORVU.MOREX.r2.4HG0320930.1 was high in Igri at stage 0 and re-emerged at stage III, 248 showing stage-III specificity in both cultivars. Several MDHARs 249 (HORVU.MOREX.r2.6HG0503910.1, HORVU.MOREX.r2.7HG0581330.1, 250 HORVU.MOREX.r2.7HG0571990.1) showed progressive induction marking stages II–III, 251 whereas HORVU.MOREX.r2.6HG0503900.1 peaked at stage I in both cultivars. Among GRs, 252 HORVU.MOREX.r2.6HG0521730.1 was specifically upregulated in Igri at stage III. In mature 253 pollen, APXs expression was high in one or both cultivars, whereas MDHARs and GRs were 254 low. 255 In summary, APX transcripts are abundant but largely stage-stable, MDHARs rise with 256 ME progression, and GR showed an Igri-biased Stage-III induction. 257 258 Thiol-redox regulators showed late activation with responsive cultivar enrichment 259 We examined the transcriptional activity of thiol–redox regulatory families (Fig. 4). Of 143 260 annotated genes, most were expressed in at least one stage: 57/63 TRXs, 37/56 GRXs, 14/16 261 TRXRs and all 8 PRXs (Fig. S2C). TRX and GRX transcripts generally accumulated at low–to– 262 moderate levels, with few highly expressed members; TRXR and PRX showed similar 263 distributions (Fig. S3C). 264 At the family level, mean expression was relatively stable across stages 0–III and in 265 pollen, with only minor cultivar differences (Fig. 4A). TRXR was notably invariant: all six 266 members showed no clear induction with ME or genotype effects (Supplementary Dataset1). 267 In contrast, several TRXs and GRXs were selectively mobilized at later stages (II–III), 268 particularly in Igri. Representative trajectories illustrated these contrasts (Fig. 4B,C). For 269 example, TRXs HORVU.MOREX.r2.1HG0023970.1, HORVU.MOREX.r2.4HG0315890.1, 270 and HORVU.MOREX.r2.2HG0091330.1 exhibited biphasic dynamics: high at stage 0, 271 decreased at stages I–II, then re-induced at stage III—Igri-specific for 272 HORVU.MOREX.r2.1HG0023970.1, shared by both cultivars for the other two. Strong 273 cultivar biases were evident: HORVU.MOREX.r2.7HG0534000.1 was consistently higher in 274 Golden Promise, whereas HORVU.MOREX.r2.2HG0174890.1 was higher in Igri across 275 stages. Marker-like behavior included Igri-specific induction at stages II–III 276 (HORVU.MOREX.r2.2HG0093810.1) and stage-III-restricted increases 277 (HORVU.MOREX.r2.5HG0432930.1, HORVU.MOREX.r2.7HG0560380.1). Among GRXs, 278 HORVU.MOREX.r2.2HG0125570.1 acted as a stage-III marker in both cultivars, 279 HORVU.MOREX.r2.7HG0552610.1 was enriched in Golden Promise at stage III, and 280 HORVU.MOREX.r2.3HG0230900.1 was Igri-stage III marker. Several transcripts 281 (HORVU.MOREX.r2.6HG0496700.1, HORVU.MOREX.r2.2HG0095970.1, 282 HORVU.MOREX.r2.1HG0072330.1) accumulated progressively across induction. Within 283 PRXs, HORVU.MOREX.r2.6HG0476250.1 was stage-I specific in both cultivars, whereas 284 3HG0231680.1 showed clear Igri specificity at stage III. 285 Notably, a subset of TRX, GRX and PRX genes was not ME-responsive but was strongly 286 expressed in mature pollen, indicating developmental rather than reprogramming regulation 287 (Fig. S4). 288 Hence, although family-level expression appears stable, distinct TRX/GRX/PRX 289 members are selectively deployed in a stageand cultivar-dependent manner, with late (II–III) 290 induction particularly prominent in Igri. 291 In review 13 in hexaploid triticale (×Triticosecale Wittm.). Plant Cell. Tissue Organ Cult. 116, 261– 577 267. doi: 10.1007/s11240-013-0399-7 578 Żur, I., Dubas, E., Krzewska, M., Zieliński, K., Fodor, J., and Janowiak, F. (2019). 579 Glutathione provides antioxidative defence and promotes microspore-derived embryo 580 development in isolated microspore cultures of triticale (× Triticosecale Wittm.). Plant 581 Cell Rep. 38, 195–209. doi: 10.1007/s00299-018-2362-x 582 Żur, I., Kopeć, P., Surówka, E., Dubas, E., Krzewska, M., Nowicka, A., et al. (2021b). Impact 583 of ascorbate—glutathione cycle components on the effectiveness of embryogenesis 584 induction in isolated microspore cultures of barley and triticale. Antioxidants 10, 1254. 585 doi: 10.3390/antiox10081254 586 Declaration of competing interest 587 The authors declare that they have no known competing financial interests or personal 588 relationships that could have appeared to influence the work reported in this paper. 589 590 Generative AI statement 591 The authors declare that no Generative AI was used in creating this manuscript. 592 593 Authorship contribution statement 594 Anna Nowicka: Formal analysis, Visualization, Writing – original draft. Zbyněk Milec: Data 595 curation, Formal analysis. Monika Krzewska: Formal analysis. Przemyslaw Kopeć: 596 Investigation. Agnieszka Springer: Formal analysis. Ewa Dubas: Writing – review & editing, 597 Formal analysis. Iwona Żur: Concept and experimental design, Funding acquisition, Project 598 administration, Writing – original draft. 599 600 Funding 601 This work was supported by the National Science Centre in Poland (grant 602 2015/18/M/NZ3/00348 to Iwona Żur). Zbyněk Milec was supported by the Ministry of 603 Education, Youth and Sports of the Czech Republic (MEYS ČR) through the project AdAgriF 604 (CZ.02.01.01/00/22_008/0004635). 605 606 Acknowledgments 607 Computational resources were provided by the e-Infrastruktura CZ project (e-INFRA CZ 608 LM2018140), supported by MEYS ČR. 609 610 Supplementary Information 611 Dataset S1. Expression data for all antioxidant-annotated genes in barley. 612 Figure S1. Phenotypic progression of barley microspores during induction of microspore 613 embryogenesis. 614 Figure S2. Proportions of annotated genes with detectable expression (FPKM > 0) versus no 615 expression (FPKM = 0) in at least one stage of microspore embryogenesis. 616 Figure. S3. Distribution of redox-related gene families by expression class across microspore 617 embryogenesis stages. 618 Figure S4. Thiol–redox regulatory genes are involved in gametophytic pollen development in 619 barley. 620 In review 14 Data availability 621 RNA sequencing data were deposited at the Gene Expression Omnibus (GEO; 622 https://www.ncbi.nlm.nih.gov/geo/): GSE233486. All other relevant data can be found within 623 the manuscript and its supporting materials. 624 625 626 Figure description 627 Figure 1. Antioxidant network and sampling stages during microspore embryogenesis 628 (ME). 629 (A) Schematic of the ROS‐scavenging network surveyed in this study. Core antioxidants: 630 SUPEROXIDE DISMUTASE (SOD), CATALASE (CAT) and GLUTATHIONE 631 PEROXIDASE (GPX). Ascorbate–glutathione (ASC–GSH) cycle: ASCORBATE 632 PEROXIDASE (APX), MONODEHYDROASCORBATE REDUCTASE (MDHAR) and 633 GLUTATHIONE REDUCTASE (GR) Thiol–redox regulators: THIOREDOXIN (TRX), 634 GLUTAREDOXIN (GRX), and PEROXIREDOXIN (PRX). S-glutathionylation is mediated 635 by GLUTATHIONE S-TRANSFERASE (GSTs). 636 (B) Gene families and numbers of annotated barley genes (HORVU.MOREX.r2). 637 (C) Cytological composition across ME induction stages in two cultivars, Golden Promise (GP; 638 recalcitrant) and Igri (responsive). Stage 0: microspores/bicellulars isolated from freshly 639 harvested spikes. Stage I: isolates from anthers pre-treated 48 h in 0.4 M mannitol (21 °C). 640 Stage II: isolates after mannitol-pretreatment followed by 24 h culture in KBP medium. Stage 641 III: isolates after mannitol-pretreatment followed by 48 h in KBP. Icons with percentage bars 642 indicate the proportions of uninucleate microspores, bicellular structures (after symmetric 643 division) and multicellular structures (continued symmetric divisions). Detailed percentages 644 and additional phenotypes are shown in Fig. S1. Quantitative data redrawn from (Nowicka et 645 al., 2024). 646 647 Figure 2. Transcriptional dynamics of genes encoding core antioxidant enzymes during 648 early microspore embryogenesis (ME) in barley. 649 (A) Mean transcript abundance (FPKM; mean ± SE) for SUPEROXIDE DISMUTASE (SOD; n 650 = 21), CATALASE (CAT; n = 9), and GLUTATHIONE PEROXIDASE (GPX; n = 5) gene 651 families in Golden Promise (GP) and Igri. Only genes with detectable expression (FPKM > 0) 652 in at least one stage in either cultivar were included. Stages 0–III denote successive steps of ME 653 induction; P denotes bicellular pollen (gametophytic pathway). 654 (B) Heatmaps of ten representative SODs, four CATs and four GPXs illustrating stageand 655 cultivar-specific expression patterns (row Z-scores). The full gene ID name starts with 656 HORVU.MOREX.r2 and is followed by the unique segment presented here. Black triangles 657 mark genes highlighted in (C). 658 (C) Expression trajectories for selected SOD, CAT and GPX genes across ME stages. Asterisks 659 indicate between-cultivar differences at a given stage (DESeq2, FDR-adjusted P < 0.05); ‘ns’ 660 indicates non-significance. Shading indicates the stage where each gene is stage-specific (grey), 661 Igri-specific (green), or Golden Promise-specific (yellow). Dashed horizontal lines show the 662 pollen expression level for the corresponding cultivar. 663 664 Figure 3. Transcriptional dynamics of ascorbate–glutathione (ASC–GSH) cycle genes 665 during early microspore embryogenesis (ME) in barley. 666 667 (A) Family-level mean transcript abundance (FPKM; mean ± SE) for ASCORBATE 668 PEROXIDASE (APX; n=5), MONODEHYDROASCORBATE REDUCTASE (MDHAR; n=7) 669 In review 15 and GLUTATHIONE REDUCTASE (GR; n=2) in Golden Promise (GP) and Igri. Only genes 670 with detectable expression (FPKM > 0) in at least one stage in either cultivar were included. 671 Stages 0–III denote successive steps of ME induction; P denotes bicellular pollen (gametophytic 672 pathway). 673 (B) Heatmaps of two representative APXs, five MDHARs and two GRs illustrating stageand 674 cultivar-specific expression patterns (row Z-scores). The full gene ID name starts with 675 HORVU.MOREX.r2 and is followed by the unique segment presented here. Black triangles 676 mark genes highlighted in (C). 677 (C) Expression trajectories for selected APX, MDHAR and GR genes across ME stages. 678 Asterisks indicate between-cultivar differences at a given stage (DESeq2, FDR-adjusted 679 P<0.05); ‘ns’ indicates non-significance. Shading indicates the stage where each gene is stage680 specific (grey) or Igri-specific (green). Dashed horizontal lines show the pollen expression level 681 for the corresponding cultivar. 682 683 Figure 4. Transcriptional dynamics of thiol–redox regulatory genes during early 684 microspore embryogenesis (ME) in barley. 685 (A) Mean transcript abundance (FPKM; mean ± SE) for THIOREDOXIN (TRX; n=73), 686 THIOREDOXIN REDUCTASE (TRXR; n=6), GLUTAREDOXIN (GRX; n=56), and 687 PEROXIREDOXIN (PRX; n=8) families in Golden Promise (GP) and Igri. Only genes with 688 detectable expression (FPKM > 0 in ≥ 1 stage) in either cultivar were included. Stages 0–III 689 denote successive steps of ME induction; P denotes bicellular pollen (gametophytic pathway). 690 (B) Heatmaps of 18 representative TRXs, 13 GRXs, and four PRXs showing stageand cultivar691 specific expression patterns (row Z-scores). The full gene ID name starts with 692 HORVU.MOREX.r2 and is followed by the unique segment presented here. 693 Black triangles mark genes highlighted in (C). 694 (C) Expression trajectories for selected TRX, GRX, and PRX genes across ME stages. Asterisks 695 indicate between-cultivar differences at a given stage (DESeq2, FDR-adjusted P<0.05); ‘ns’ 696 indicates non-significance. Shading indicates the stage where each gene is stage-specific (grey), 697 Igri-specific (green), or Golden Promise-specific (yellow). Dashed horizontal lines indicate the 698 pollen expression level for each cultivar. 699 700 Figure 5. Transcriptional dynamics of GLUTATHIONE S-TRANSFERASE (GST) genes 701 during early microspore embryogenesis (ME) in barley. 702 (A) Mean transcript abundance (FPKM; mean ± SE) for GST (n = 140) gene family in Golden 703 Promise (GP) and Igri. Only genes with detectable expression (FPKM > 0 in ≥ 1 stage) in either 704 cultivar were included. Stages 0–III denote successive steps of ME induction; P denotes 705 bicellular pollen (gametophytic pathway). 706 (B) Heatmap of 30 representative GSTs showing trends in stageand cultivar-specific 707 expression patterns (row Z-scores). The full gene ID name starts with HORVU.MOREX.r2 and 708 is followed by the unique segment presented here. Black triangles mark genes highlighted in 709 (C). 710 (C) Expression trajectories for selected GST genes across ME stages. Asterisks indicate 711 between-cultivar differences at a given stage (DESeq2, FDR-adjusted P < 0.05); ‘ns’ indicates 712 non-significance. Shading indicates the stage where each gene is stage-specific (grey), Igri713 specific (green), or Golden Promise-specific (yellow). Dashed horizontal lines indicate the 714 pollen expression level for each cultivar. 715 716 Figure 6. Graphical summary of the main findings. 717 Top: Reprogramming shared by both cultivars (Igri = Golden Promise, GP) is characterized by 718 late (stages II–III) increases in SUPEROXIDE DISMUTASE (SODs), CATALASEs (CATs), 719 In review 16 GLUTATHIONE PEROXIDASEs (GPXs) and MONODEHYDROASCORBATE REDUCTASEs 720 (MDHARs), THIOREDOXINs (TRXs), GLUTAREDOXINs (GRXs) PEROXIREDOXINs 721 (PRXs) and GLUTATHIONE S-TRANSFERASEs (GSTs). 722 Bottom: In the responsive cultivar (Igri > GP), late-stage upregulation of GLUTATHIONE 723 REDUCTASE (GR) and thiol–redox components—TRX/GRX/PRX—is accompanied by a more 724 pronounced GST-mediated detoxification response. 725 In review Figure 1.JPEG In review Figure 2.JPEG In review Figure 3.JPEG In review Figure 4.JPEG In review Figure 5.JPEG In review Figure 6.JPEG In review