**1. Introduction**

A considerable number of studies have confirmed that epigenetic mechanisms play a crucial role in the ability of plants to adapt to the environment during stress conditions [1–3]. Heritable epigenetic variability can lead to physiological, morphological and ecological changes. Alterations of the epigenome can improve the plant features valuable for agronomical reasons [4]. Precise manipulation of the DNA methylation status enables one to control chromatin condensation, and thus gene expression. The assessment of detailed mechanisms that regulate the epigenetic changes under particular stress factors could provide information about the characteristic epigenetic marks that lead to adaptation in encountered stress conditions [5].

Epigenetic modifications induced under stress factors can be reversed and epigenomic status can return to the initial state, when the stress conditions are no longer present. However, the latest findings have shown that some of these changes can be preserved as stress memory leading to inheritance of the epigenetic status and thus enhancement of the adaptable ability of the progeny to the environment [5]. In comparison to mammalian cells, in plant cells, massive erosion of the epigenetic marks during gametogenesis occurs [6]. Particular short RNA molecules are assumed to crucially contribute to the inheritance of the epigenetic changes. Due to the RNA-dependent DNA methylation, the methylation implemented to the genome of parental cells can be maintained after cell division. The maintenance of the DNA methylation after replication is the effect of the feed-forward maintenance of de novo methylation in plants. The essence of this process is the establishment of two processes. The first mechanism involves methylation in the CpG and CpHpG contexts (H indicates cytosine, adenine or thymine) through methyltransferase 1 (MET1) and chromomethylase 3 (CMT3) in the newly synthesized DNA strand. However, cytosine methylation is also maintained through RNA-dependent DNA methylation by short RNA molecules stimulated by DNA methylation. The occurrence of these two processes leads to independence of the DNA methylation mechanism in the plant cell [7].

The latest research has shown that apart from environmental stress, epigenetic changes may occur as an effect of short oligodeoxynucleotides' (ODNs/OLIGOs) activity. The mechanism of ODNs is based on the development of particular short non-coding RNAs, which may lead to the elevated accumulation (RNA activation—RNAa) or degradation (RNA interference—RNAi) of the gene transcript and the RNA-directed DNA methylation (RdDM) [8,9].

In our previous work [9], we demonstrated the effectiveness of the short oligodeoxynucleotides in the induction of epigenetic changes in flax. The plant infiltration with ODNs complementary to the regulatory regions (ODN1 to 5'UTR, ODN6 to intron, ODN11 to 3'UTR) of the chalcone synthase (CHS) genes *LuCHS6* and *LuCHS7* led to changes in the CHS genes' expression. The alteration of the *CHS* transcript levels was assumed to be triggered by the changes in the DNA methylation. The identification of the 5-methylcytosine (5-mC) in the CCGG motifs located in the CHS genes made it possible to distinguish the stably unmethylated motifs, independently of the analyzed plant (*LuCHS6*/*LuCHS7*: −232/−, +217/−, +1606/+1156) and variable methylated motifs (*LuCHS6*/*LuCHS7*: +996/+552, +1219/+775, +1273/+829). The in silico analysis of nucleosome formation energy also indicates the alteration of nucleosome position upon the changes in the methylation status of *LuCHS6*. The preliminary examination of this theoretical presumption suggests that the increased level of methylation in the +1273 CCGG motif of the transgenic plants led to the visible changes in the accessibility of DNA for the restriction enzymes. Thus, the obtained results suggested the variable susceptibility of DNA upon methylation to the interaction with the particular effector proteins (e.g., transcription machinery) [9].

The subject of our research was the linseed variety of common flax *Linum usitatissimum* L.–Linola. Linseed is widely known for its biomedical applications due to an appropriate ratio of unsaturated fatty acids in the linseed-derived oil (the ω-6:ω-3 ratio is 1:4) [10], as well as high content of phenylpropanoid compounds (flavonoids, lignans and phenolic acids) [11,12] and fiber [13]. Flavonoids present in the flaxseed are popular in the context of health-promoting properties, including antioxidative [14] and antimicrobial activity [15]. Recent findings indicate a preventive or therapeutic role of flavonoids from flax in metabolic disorders [16] and carcinogenesis [17,18].

Chalcone synthase (CHS) catalyzes the first step of flavonoid biosynthesis [19]. In many plant species, there occur more than one gene encoding CHS (multigenomic family). It was demonstrated that there are 5 CHS genes (Phytozome ID)—*LuCHS3* (Lus10041508), *LuCHS4* (Lus10023670), *LuCHS5* (Lus10011746), *LuCHS6* (Lus10033717), *LuCHS7* (Lus10031622)—and 4 CHS-like genes (Phytozome ID): *LuCHS1* (Lus10039904), *LuCHS2* (Lus10042388), *LuCHS8* (Lus10026286), *LuCHS9* (Lus10002187) in the genome of *Linum usitatissimum* L. [20]. The transcripts derived from the 5 mentioned CHS genes were also confirmed by our group [9,21]. Moreover, our findings indicated the presence of 2 additional CHS genes, named in order: *LuCHS10* and *LuCHS11* (recognized fragment of *LuCHS11* sequence is presented in Supplemental Figure S3). Initially, they were only suspected to be transcriptional variants of *LuCHS3* and *LuCHS4,* respectively. However, the in silico analysis showed the diverse localization of these genes on the chromosomes. The systematized names, numbers in databases and genomic localization are presented in Table 1. Apart from the fact that CHS genes encode the key enzyme in flavonoid formation, they display a susceptibility to the response to stress factors and can be specifically regulated under particular environmental conditions. The expression of the chalcone synthase genes may also depend on the plant developmental stage and tissue specificity. In our research, we focused on two CHS genes: *LuCHS6* and *LuCHS7*. The genomic sequences of these genes are similar to the CHS-A encoding gene from *Petunia* x *hybrida*, used for the generation of genetically modified (GM-CHS) flax, which served as the model for changes in CHS gene expression in our initial research.


**Table 1.** Genes encoding CHS in flax.

<sup>a</sup>—indicates the CHS genes (sequences and names) published by [20]; <sup>b</sup>—indicates the CHS genes (sequences and names) published by [21]; <sup>c</sup>—recently identified CHS gene.

The main purpose of our research was the monitoring of changes induced by the treatment with ODNs after a time longer than 48 h since the beginning of incubation. The plants from subsequent generations derived from ODN-treated flax were analyzed in the context of the *LuCHS6* and *LuCHS7* gene expression and methylation of CCGG motifs, in order to verify the perpetuation of the induced changes. With a view to the relative susceptibility to ODNs' degradation by nucleases, modification of the oligodeoxynucleotide molecules was applied—by methylation of cytosines present in the ODN sequence (met) and by the replacement of oxygen by a sulfur at the phosphodiester bond (thiophosphate–pto). Stability of the epigenetic changes induced by the modified ODNs was also analyzed in the next generation of plants. Moreover, the chromatin accessibility assay has shown the differences between the control (Linola) and plants with stable epigenetic modification (W92.40). Performed experiments contributed to the understanding of the processes involved in the introduction of exogenous oligonucleotides into plant cells.
