Key message Effective microspore embryogenesis in triticale is determined by a specific hormonal homeostasis: low value of IAA/cytokinins, IAA/ABA and cytokinins/ABA ratios as well as proper endogenous/exogenous auxin balance, which favours androgenic structure formation and green plant regeneration ability. Abstract The concentration of plant growth regulators (PGRs): auxins (Auxs), cytokinins (CKs) and abscisic acid (ABA) was measured in anthers of eight DH lines of triticale (× Triticosecale Wittm.), and associated with microspore embryogenesis (ME) responsiveness. The analysis was conducted on anthers excised from control tillers at the phase optimal for ME induction and then after ME-initiating tillers treatment (21 days at 4 °C). In control, IAA predominated among Auxs (11–39 nmol g⁻¹ DW), with IBA constituting only 1 % of total Auxs content. The prevailing isoforms of CKs were cis isomers of zeatin (121–424 pmol g⁻¹ DW) and zeatin ryboside (cZR, 146–432 pmol g⁻¹ DW). Surprisingly, a relatively high level (10–64 pmol g⁻¹ DW) of kinetin (KIN) was detected. Cold treatment significantly changed the levels of all analysed PGRs. The anthers of ‘responsive’ DH lines contained higher concentrations of IBA, cis and trans zeatin, cZR and ABA, and lower amount of IAA and KIN in comparison with ‘recalcitrant’ genotypes. However, the effects of exogenous ABA, p-chlorophenoxyisobutyric acid (PCIB) and 2,3,5-triiodobenzoic acid treatments suggest that none of the studied PGRs acts alone in the acquisition of embryogenic competency, which seems to be an effect of concerted PGRs crosstalk. The initiation of ME required a certain threshold level of ABA. A crucial prerequisite for high ME effectiveness was a specific PGRs homeostasis: lower Auxs level in comparison with CKs and ABA, and lower CKs/ABA ratio. A proper balance between endogenous Auxs in anthers and exogenous Auxs supplied by culture media was also essential.

To verify the hypothesis that cell redox status regulates the process of microspore embryogenesis (ME), reactive oxygen species (ROS) generation and the activity of enzymatic and non-enzymatic antioxidants were analyzed in eight doubled haploid lines of triticale with significant differences in embryogenic potential. The analyses were performed in anthers excised from freshly cut tillers (control) and from low temperature (LT) pre-treated tillers (3 weeks at 4 °C) in which ME has been initiated. Significant associations between ME effectiveness and the variables studied were found. In control cultures, high superoxide dismutase (SOD) activity appeared crucial for microspore viability. On the other hand, positive though non-linear correlation between ME effectiveness and H₂O₂ generation, and negative correlation with catalase (CAT) activity suggest that some threshold level of H₂O₂ is important for successful ME initiation. LT tillers pre-treatment significantly increased H₂O₂ accumulation, which had a negative effect on ME effectiveness. However, even high level of H₂O₂ did not endanger cell viability as long as the cells exhibited high activity of ROS-decomposing enzymes (SOD, CAT and POX). The ability to sustain antioxidative enzyme activity under cold stress in the dark was another important requirement for high effectiveness of ME, allowing for the generation of the signal initiating microspore reprogramming and simultaneously protecting the cells from the toxic effects of ROS production. The role of antioxidative enzymes cannot be replaced even by high activity of non-enzymatic antioxidants. In conclusion, genetically controlled but environmentally modified cell tolerance to oxidative stress seems to play an important role in triticale ME.

Quantitative trait loci (QTLs) associated with androgenic responsiveness in triticale were analyzed using a population of 90 DH lines derived from the F1 cross between inbred line ‘Saka 3006’ and cv. ‘Modus’, which was used in a number of earlier studies on molecular mapping in this crop. Using Windows QTL Cartographer and MapQTL 5.0, composite interval mapping (CIM) and association studies (Kruskal–Wallis test; K–W) for five androgenesis parameters (androgenic embryo induction, total regeneration and green plant regeneration ability, and two characteristics describing final androgenesis efficiency) were conducted. For the studied components of androgenic response, CIM detected in total 28 QTLs which were localized on 5 chromosomes from A and R genomes. Effects of all QTLs that were identified at 2.0 or above of the LOD score explained 5.1–21.7 % of the phenotypic variation. Androgenesis induction was associated with seven QTLs (LOD between 2.0 and 5.8) detected on chromosomes 5A, 4R, 5R and 7R, all of them confirmed by K–W test as regions containing the markers significantly linked to the studied trait. What is more, K–W test revealed additional markers on chromosomes: 5A, 2BL, 7B and 5R. Both total and green regeneration ability were controlled by genes localized on chromosome 4A. Some of the QTLs that affected final androgenesis efficiency were identical with those associated with androgenic embryo induction efficiency, suggesting that the observed correlation may be either due to tight linkage or to pleiotropy. Key message Five regions of the triticale genome were indicated as revealing significant marker/trait association. Markers located in these regions are potentially useful for triticale breeding through marker-assisted selection.