**4. Discussion**

Many reports have shown that RobTs in wheat can arise by centromeric misdivision of univalents at AI, followed by segregation of the derived telocentric chromosomes to the same nucleus, and fusion of the broken parts [26,30,31]. The focus of this study was to exploit the formation of RobTs to transfer of the *Lr54* + *Yr37* gene loci from *Ae. kotschyi* into triticale. We have focused on two essential aspects of this process: the frequency of univalent misdivision, and the frequency at which the broken chromosomes fuse to form a RobT. The control experiment used the standard approach of self-pollinating monosomic substitution (40 + M2R + M2Sk) of triticale. Genomic in-situ hybridization identified that chromosome 2S<sup>k</sup> was present in all meiotic cells assayed, and the centromere was divided during AI in 9% of the PMCs surveyed. The same event was observed in PMCs isolated from the 40 + M2RS + M2RL + M2S<sup>k</sup> plants. The frequency of PMCs where the centromere of the 2S<sup>k</sup> univalent divided was 7.89%. Our results are comparable with other reports, showing that the frequency of univalent misdivision range between 3% up to 51% [26,30,31]. When both approaches were compared (40 + M2R + M2S<sup>k</sup> vs 40 + M2RS + M2RL + M2Sk), no significant di fference in the frequency at which the *Aegilops* univalent misdivided was observed. However, the 2R univalent did not misdivide in the control experiment, and the use of the double ditelosomic D2RS + D2RL in the alternative approach was justified.

It can be assumed that the probability of RobTs formation is dependent on the rate of centromere breakage, and subsequent formation of telocentric chromosomes, occurs during meiosis. In wheat, the frequency of recovery of wheat–alien RobTs is between 2% and 20%, depending on the chromosomes involved [23,26,30,31,40–44]. In our experiment, the progeny of monosomic substitution plants lacked any kind of triticale-*Aegilops* translocation. This could correlate with the lack of the 2R chromosome misdivision observed in these plants. Similar issues were reported by Ghazali et al. [45], who aimed to produce whole arm RobTs involving chromosomes 2B (wheat) and 2E<sup>b</sup> (*Thinopyrum bessarabicum*). Surprisingly, the authors obtained only T2EbS.2BL, while no 2BS.2EbL RobTs were recovered [45]. It has been reported that rye chromosomes rearrange at a higher rate when compared to wheat chromosomes [46], with the long arm of chromosome 2R more readily fused with the chromosome arm of an alien species. For example, Fiebe et al. [47] derived 2BL.2RL RobTs carrying Hessian fly resistance gene *H21* in wheat cv. Hamlet. Rahmatov et al. [48] also obtained 2DS.2RL RobT, which transfered the stem rust resistance gene *Sr59* into wheat.

In our study, we have e ffectively used ditelosomic lines of triticale to increase the frequency of 2Sk.2R RobTs generated. Moreover, we have obtained both 2SkS.2RL and 2RS.2SkL RobTs. The ability to produce both types of RobTs was an important goal as the particular chromosome segments of *Aegilops* may harbor other desirable genes. To summarize, we obtained plants with chromosome translocations, which probably carry the *Lr54* + *Lr37* leaf and stem rust resistance genes. The next step of this project will be to phenotype the eleven RobTs lines for leaf rust or stripe rust resistance. If these tests succeed, such germplasm could be a valuable source for triticale resistance breeding.

**Author Contributions:** Conceptualization, M.T.K.; methodology, M.T.K.; validation, W.U.; formal analysis, W.U.; investigation, W.U., J.B., J.S. and R.S.; resources, M.T.K., H.W.; data curation, M.T.K., W.U.; writing—original draft preparation, W.U., M.T.K.; writing—review and editing, W.U., D.P. and M.T.K.; visualization, W.U., M.T.K.; supervision, M.T.K., D.P. and H.W.; project administration, M.T.K.; funding acquisition, M.T.K.

**Funding:** This research and the APC were funded by NATIONAL CENTRE FOR RESEARCH AND DEVELOPMENT, Poland (Narodowe Centrum Bada ´n i Rozwoju, Polska), gran<sup>t</sup> number LIDER/3/0004/L-8/16/NCBR/2017.

**Acknowledgments:** We would like to acknowledge the administrative support provided by the Institute of Plan Genetics of the Polish Academy of Sciences.

**Conflicts of Interest:** The authors declare no conflict of interest.
