Doubled haploids of triticale are of interest for plant breeders due to hybrid breeding programs based on cytoplasmic male sterility Tt phenomenon. However, (epi)mutations appearing during in vitro culture regeneration may lead to a phenotypic variation that makes the uniformity of plant materials questionable. Using RP-HPLC genomic DNA methylation of donor doubled haploid plants utilized as a source of tissues for the in vitro regeneration (via androgenesis and somatic embryogenesis) of triticale cv. Bogo and their consecutive generative progeny was evaluated. It was demonstrated that in vitro cultures induced a decrease of the DNA methylation of the regenerants independently of the approach used for plant regeneration. The decrease in DNA methylation of genomic DNA proceeded up to the first/second successive generations followed by the beginning of its reestablishment. Moreover, somatic embryogenesis resulted in a higher level of genomic DNA demethylation in regenerants than androgenesis and the process of methylation seems to be affected by donor plant. It is being speculated that long term changes in genomic DNA methylation may be a source of off-type individuals that may spontaneously arise during plant breeding.

We present the development of the theoretical background of the metAFLP approach which allows for partition of complex variation into sequence changes, de novo methylation and demethylation of the regenerants derived via in vitro tissue culture methods in the case of triticale. It was demonstrated that, independent of whether andro- or embryogenesis was used for plant regeneration, the level of sequence changes identified between regenerants is about 10 %. Moreover, DNA demethylation prevails over de novo methylation of the regenerants compared to the donor plant. The metAFLP approach allows for the evaluation of numerous quantitative characteristics. For instance, one may quantify the number of sites unaffected by tissue culture approaches, global site DNA methylation etc. It is suggested that the approach could be useful for breeders in order to control plant material uniformity or for the evaluation of modified in vitro tissue culture approaches allowing for control of the (epi)mutation level. The extended metAFLP approach presented here delivers sufficient background for the evaluation of software that could facilitate analyses of the tissue culture induced variation.Electronic supplementary materialThe online version of this article (doi:10.1007/s11032-014-0079-2) contains supplementary material, which is available to authorized users.

Plant regeneration via in vitro culture can induce genetic and epigenetic variation; however, the extent of such changes in triticale is not yet understood. In the present study, metAFLP, a variation of methylation-sensitive amplified fragment length polymorphism analysis, was used to investigate tissue culture-induced variation in triticale regenerants derived from four distinct genotypes using androgenesis and somatic embryogenesis. The metAFLP technique enabled identification of both sequence and DNA methylation pattern changes in a single experiment. Moreover, it was possible to quantify subtle effects such as sequence variation, demethylation, and de novo methylation, which affected 19, 5.5, 4.5 % of sites, respectively. Comparison of variation in different genotypes and with different in vitro regeneration approaches demonstrated that both the culture technique and genetic background of donor plants affected tissue culture-induced variation. The results showed that the metAFLP approach could be used for quantification of tissue culture-induced variation and provided direct evidence that in vitro plant regeneration could cause genetic and epigenetic variation.Electronic supplementary materialThe online version of this article (doi:10.1007/s11103-015-0368-0) contains supplementary material, which is available to authorized users.

BackgroundIn vitro plant regeneration via androgenesis or somatic embryogenesis is capable of inducing (epi)mutations that may affect sexual progenies. While epimutations are associated with DNA methylation, mutations could be due to the movement of transposons. The common notion is that both processes are linked. It is being assumed that demethylation activates transposable elements (TEs). Analysis of methylation changes and their relation with TEs activation in tissue cultures requires uniquely derived donor plants (Ds), their regenerants (Rs) and respective progeny (Ps) that would allow discrimination of processes not related to changes introduced via in vitro cultures. Moreover, a set of methods (RP-HPLC, SSAP, and MSTD) is needed to study whether different TEs families are being activated during in vitro tissue culture plant regeneration and whether their activity could be linked to DNA methylation changes or alternative explanations should be considered.ResultsThe in vitro tissue culture plant regeneration in barley was responsible for the induction of DNA methylation in regenerants and conservation of the methylation level in the progeny as shown by the RP-HPLC approach. No difference between andro- and embryo-derived Rs and Ps was observed. The SSAP and MSTD approach revealed that Ds and Rs were more polymorphic than Ps. Moreover, Rs individuals exhibited more polymorphisms with the MSTD than SSAP approach. The differences between Ds, Rs and Ps were also evaluated via ANOVA and AMOVA.ConclusionsStressful conditions during plant regeneration via in vitro tissue cultures affect regenerants and their sexual progeny leading to an increase in global DNA methylation of Rs and Ps compared to Ds in barley. The increased methylation level noted among regenerants remains unchanged in the Ps as indicated via RP-HPLC data. Marker-based experiments suggest that TEs are activated via in vitro tissue cultures and that, independently of the increased methylation, their activity in Rs is greater than in Ps. Thus, the increased methylation level may not correspond to the stabilization of TEs movement at least at the level of regenerants. The presence of TEs variation among Ds that were genetically and epigenetically uniform may suggest that at least some mobile elements may be active, and they may mask variation related to tissue cultures. Thus, tissue cultures may activate some TEs whereas the others remain intact, or their level of movement is changed. Finally, we suggest that sexual reproduction may be responsible for the stabilization of TEs.

No abstract