2.1. DNA methylation in Plants
Plants are very specific organisms with unique abilities that lead them to maximize their potential. This feature is due to the non-mobility of plants. Admittedly, plants live in one specific environment but are exposed to a wide range of fluctuations, including climatic fluctuations, which can permanently or temporarily change their living environment. Consequently, plants continually produce various evolutionary biological adaptations. It is known that environmental conditions can also induce gene activity through epigenetic processes. Plants have developed a number of extremely complex epigenetic mechanisms that they use to control and regulate the genome. Genome potential control is possible through changes in chromatin structure. Epigenetic modifications alter the ability of genes to be expressed and directly coordinate with primary metabolism, which regulates plant growth and development. Epigenetic processes affect both the phenotype and fitness of plants and contribute to their ability to colonize and evolve in changing environments
[1][4][5]. Environmentally induced epigenetic modifications represent an important adaptive strategy. In this context, the study of epigenetic modifications in plants is extremely interesting and demonstrates that forest trees have enormous adaptive potential.
DNA methylation is a post-replicative enzymatic modification of DNA. It is a stable process leading to the silencing of gene expression and resulting in the covalent attachment of so-called methyl groups (one carbon atom joined to three hydrogen atoms, -CH3) to the nitrogenous bases of nucleotides (cytosine and adenine). In higher eukaryotes, methyl groups attach to a carbon atom located at position five of the cytosine ring of the DNA double helix, forming C5-methylcytosine (m5C). They can also attach to the amino group of cytosine (N4, m4C). In some higher plants, N6-methyladenine (m6A) is also formed as a methylation effect
[6] (
Figure 1).

Figure 1. Structural models of substrates and products of DNA methylation. Cytosine (CYT or C) is a nitrogenous base from the pyrimidine group that forms via three hydrogen bonds in double-stranded nucleic acids and a complementary pair with guanine (GUA or G). Adenine (ADE or A) is a nitrogenous base from the purine group that, by means of two hydrogen bonds in double-stranded nucleic acids, forms a complementary pair with thymine (THY or T in DNA) or uracil (URA or U in RNA). In methylation, the methyl donor is most often S-adenosyl-L-methionine (SAM), yielding S-adenosyl homocysteine (SAH), which is an inhibitor of methyltransferases; the entire process is catalyzed by DNA methyltransferases (DNMT) and is cytosine- and adenine-specific. DNMTs transfer methyl groups to the appropriate positions on the rings of nitrogenous bases: into carbon (at position 5 of the cytosine ring) and amino groups (at position 4 of the cytosine ring and position 6 of the adenine ring)
[6].
In plants, DNA methylation occurs in all contexts of the CpG, CpHpG, and CpHpH dinucleotide sequences (H represents any A, T, or C nucleotide, except G) located in a linear 5′ to 3′ DNA sequence, with the cytosine and guanine nucleotide adjacent or one base apart
[7][8][9]. CpG dinucleotides are randomly distributed throughout the genome and are mostly methylated. However, they also form sites with lower levels of methylation where CpG sequences are highly concentrated; these are referred to as CpG islands (CGIs). CGIs are regions more than 200 bp long
[10], and CpG sites in CGI are mostly unmethylated, thus avoiding mutational deamination of C5-methylcytosine (m5C) to thymine. CGIs often overlap with transcription start sites (TSSs), and their hypermethylation may be associated with transcriptional repression
[11]. It was previously thought that methylation could only silence genes at CGI sites. However, this dogma was overturned by the recent scientific reports that have documented that CpG islands are not needed to turn off genes by methylation
[12].
Chromatin can be decondensed (open chromatin or euchromatin), enriched in genes and promoting transcription, or condensed (closed chromatin or heterochromatin), enriched in repetitive sequences and silenced sequences
[1]. Changes in chromatin are most often introduced by small RNAs (sRNAs) and longer non-coding RNAs (lnRNAs), which are conduits for mechanisms of DNA methylation, chromatin modification, or transcript degradation/amplification
[13]. Reversible changes in chromatin structure, such as cytosine methylation or histone modifications, directly affect the transcriptional efficiency of genes
[14][15].
The maintenance of methylation is controlled by several pathways (
Figure 2). CpG methylation is maintained by DNA methyltransferase 1 (MET1). In addition, more recently, MET1 has been assigned a broader function, as studies show that it is also required for the establishment of CpHpH methylation
[16].
Figure 2. Maintenance methylation in plants. DNA METYLOTRANSFERASE 1 (MET1) is required to maintain DNA methylation in the context of CpG dinucleotides. Methylation variant (VIM) proteins are required here—VIM proteins 1–3 function in transcriptional regulation through their role in the MET1 DNA methylation pathway. Maintenance of DNA methylation in the context of CpHpG requires CHROMOMETHYLASE 3 (CMT3) activity. In the context of CpHpH, this activity is mainly controlled by CHROMETYLASE 2 (CMT2) but also by DOMAINS REARRANGED METHYLTRANSFERASE 2 (DRM2) through the RNA-directed DNA methylation (RdDM) pathway. Plant-specific chromometallases are both “readers” of histone methylation and “writers” of DNA methylation. They interact with SUVH histone methyltransferases. MBD domain proteins in plants can act both as “readers” (MBD5, 6 for CpG context) and “erasers” of DNA methylation (MBD7).
Maintenance of DNA methylation in the CpHpG context requires chromomethylase 3 (CMT3) activity. Methylation in the CpHpH context requires chromomethylase 2 (CMT2) activity [17][18], and a large amount of CpHpH methylation is maintained by domains rearranged methyltransferase 2 (DRM2) in the RNA-dependent DNA methylation (RdDM) pathway [15][19][20][21], which is also responsible for de novo methylation in all three sequence contexts [8][9][18]. CpHpH methylation is asymmetric, meaning methylation will be lost in one progeny strand. Chromomethylases contain both a chromodomain and a DNA methyltransferase domain and interact with some proteins (suppressor of variegation su(var) homolog, SUVH) to ensure proper deposition of histone H3K9 (H3K9me2) methylation, as well as CpHpG or CpHpH in transposable elements (TEs) [20]. SUVH proteins are essential for accessing the regulatory mechanisms of genes located in close proliferating transposable elements (TEs) [20]. DRM2 and MET1 proteins share significant homology with mammalian methyltransferases (DNMT3 and DNMT1). The CMT3 protein is unique to plants and belongs to a family of chromomethylases that are both “readers” of histone methylation and “writers” of DNA methylation [20][22]. Most methylation in plants occurs in transposable elements (TEs), but also in the bodies of active genes where it is restricted to the CpG context [9][18]. The first step in de novo DNA methylation pattern formation is the RNA-dependent DNA methylation pathway, which relies on specialized, plant-specific RNA polymerases POL IV or POL V (POL VI is also specific for grasses) [20][22] (Figure 3).
Figure 3. RNA-directed DNA methylation (RdDM) pathway in plants. The first step in de novo DNA methylation pattern formation is the RNA-dependent DNA methylation (RdDM) pathway, which relies on the specialized plant-specific RNA polymerases POL IV and POL V (in all three sequence contexts). RNA POLYMERASE IV (POL IV) cuts into short (26–45 nt), single-stranded RNAs (ssRNAs) that serve as a substrate for RNA-dependent RNA POLYMERASE 2 (RDR2). RDR2, together with POLI V, converts them into double-stranded RNA (dsRNA). The dsRNAs are then converted into 24-nucleotide small interfering RNAs (siRNAs) by DICER-LIKE 3 (
DCL3), methylated at their 3′-end by HUA ENHANCER 1 (HEN1), and recruited by ARGONAUTE 4 (AGO4) or other ARGONAUTE proteins (AGO6 and AGO9). AGO4-siRNA complexes, interacting with POL V, then recruit DOMAINS REARRANGED METHYLTRANSFERASE 1 and 2 (DRM1, DRM2) DNA methyltransferases to the target DNA. POL V can be recruited by indirectly interacting with histone methyltransferases SU(VAR)3–9 homolog 2 (SUVH2 and SUVH9), which act as de novo methylation “readers” through interaction with the DDR complex (DRD1, DMS3, RDM1). The DDR complex consists of DEFECTIVE IN MERISTEM SILENCING 3 (DMS3), DEFECTIVE IN RNA-DIRECTED DNA METHYLATION 1 (DRD1), and RNA-DIRECTED DNA METHYLATION 1 (RDM1). The DNA methyl-readers SUVH2 and SUVH9, along with the DDR complex, are required for POL V recruitment to chromatin. The SAWADEE HOMEODOMAIN HOMOLOG 1 (SHH1) histone reader is required for POL IV association to chromatin. AGO4-siRNA complexes are then targeted to transcripts generated by POL V and recruit DOMAINS REARRANGED METHYLTRANSFERASE (DRM1, DRM2) DNA methyltransferases to the target DNA
[22][23].
Important functions of the RdDM pathway include the control of repeats in heterochromatic regions and dispersed transposons, as well as related sequences in euchromatic regions
[8][24]. Transcriptional repression of actively proliferating transposons for genome defense and activation of the parent-of-origin expression of specific genes in reproductive tissues are also included
[25]. Pol IV and Pol V polymerases uniquely contribute to epigenetic regulation by producing silencing-related non-coding transcripts. The silencing pathway through Pol IV and Pol V activity can be counteracted by active demethylation, thereby creating epigenetic flexibility that is important for environmental adaptation
[13].
2.2. Forest Trees—Ecosystems Important to Humans
Forests cover approximately 31% of the world’s land area, or 4 billion hectares (http://www.fao.org/state-of-forests/en, accessed on 19 July 2021). Almost half of these are intact (natural forests), and more than one-third are naturally regenerating forests of native species where there are no traces of human activities and natural ecological processes are undisturbed (primary forests). Forests are dynamic ecosystems with high environmental, social, and economic importance [1]. However, the global changes currently occurring on Earth, such as desertification, insect invasion, abiotic stresses, deforestation, degradation, and climate change, pose a significant threat to the condition of forests.
In 2017, a database on forests and forest-forming species of the world was established
[26]. GlobalTreeSearch is the first global, authoritative list of tree species created through the multidisciplinary work of many organizations and associated with forest scientists
[26]. One of the goals of GlobalTreeSearch is to protect forests.
Other global organizations have also become involved in addressing the threat to forests from global change, including the United Nations, which has created a forest aid plan—The United Nations Strategic Plan for Forests 2017–2030 (UNSPF) (
https://www.un.org/esa/forests/wp-content/uploads/2016/12/UNSPF_AdvUnedited.pdf, accessed on 19 July 2021). The organization recognizes the need for global cooperation and coordination for sustainable forest development and creating more resilient and adaptable forest communities.
The possibility that the plants themselves, including trees, have to thrive and create an ideal habitat for themselves in terms of adapting to any stress conditions relates to maximizing their adaptability and plasticity. Because of their longevity and exposure to large seasonal changes, trees and perennial woody plants, especially from temperate and boreal regions
[4], have evolved phenotypic modification systems to tolerate changes in climatic conditions. Tree adaptation to the surrounding environment is based on the natural evolution of biological mechanisms (including epigenetics) that lead to the development of plant tolerance and resilience and the avoidance of environmental constraints
[4][5]. Hence, there is a wide range of plant tolerances to edaphic and climatic conditions, from northern subarctic to subtropical zones. Among trees, conifers are considered the most adapted plants with the greatest adaptive potential
[4]. They are also among the longest-lived terrestrial organisms. However, accelerating climate change, with increasing temperatures and decreasing precipitation, is challenging for all plants, including those with the greatest adaptive capacity. Although trees possess adaptive genomes and excellent regulatory mechanisms (genetic and epigenetic) whereby changes in gene activity occur in a plastic manner that allows them to survive and reproduce successfully in changing environments, their future is not known or predictable in terms of their ability to adapt to ongoing changes. Therefore, there is a need for further intensive research, especially in the field of epigenomics of forest trees, which is the least known area among plants and, in light of recent scientific reports, seems to be one of the key issues surrounding climate change.
The vast majority of studies devoted to plant epigenetics focus only on model herbaceous plants, such as
Arabidopsis thaliana L.
[18][27][28][29][30][31][32][33][34][35]. Studies using Arabidopsis also include work on global climate change
[36][37][38][39][40][41]. Similarly, issues concerning the role of DNA methylation itself in the ability of plants to acclimate and adapt to changing environmental conditions or stress memory have also recently been quite extensively described
[13][42][31][43][44][45].
In contrast, little research has been conducted on long-lived woody plants, omitting useful trees. Despite the growing climate problem, in recent years, there have been few studies on forest tree epigenetics in the context of stricte climate change
[5][46][47][48][49]. As can be seen from the herbaceous plant research reports cited above, we have research tools that can expand our knowledge of trees and forest ecosystems. Therefore, there is still much work required to discover the mechanisms responsible for adaptation and all the environmental processes involved in forest trees.
Forest trees are globally dispersed, modular organisms with a complex life cycle. They are subjected to multiple environmental pressures during their long lives. Tree populations that survive environmental changes are the result of complex, interacting, and advanced evolutionary mechanisms, such as migration, adaptation, and phenotypic plasticity
[1]. Trees, as long-lived organisms, may specifically use epigenetics to facilitate phenotypic modifications in response to environmental change
[1]. Identifying genomic loci that undergo epigenetic changes in response to environmental conditions is an important research goal. By expanding this knowledge, it will become possible to understand the processes involved in epigenetic adaptation, that is, how epigenetic modifications heritably alter a plant’s gene expression and thereby refine its responses to environmental stimuli and its ability to survive under altered conditions
[13].
2.3. Effects of DNA Methylation on Adaptations of Forest Trees
Epigenetic studies of forest trees have been significantly accelerated with the sequencing of the first tree genomes of
Populus trichocarpa [50],
Picea abies (L.) H. Karst.
[51], and
Picea glauca (Moench) Voss
[52], or
Eucalyptus grandis W. Hill
[53] and
Pinus taeda L.
[54]. Today, there is a much-expanded forest tree genome database, with species such as
Pinus lambertiana Douglas
[55],
Ginkgo biloba L.
[56],
Fraxinus excelsior L.
[57],
Pseudotsuga menziesii (Mirbel) Franco
[58],
Betula pendula Roth.
[59],
Larix sibirica Ledeb.
[60],
Fagus sylvatica L.
[61],
Abies alba Mill.
[62], and
Eucalyptus pauciflora Sieber ex Spreng.
[63]. The more knowledge we acquire about forest tree genes, the better we can learn about regulatory epigenetic mechanisms. DNA methylation in the regulation of gene expression in tree responses to environmental stimuli has been widely studied (abiotic stress), including in droughts
[64][65][66][67][68], heavy metals
[69], extreme temperatures
[70][71][72], and salt stress
[73][74][75]. These changes can occur at the genome level
[65][76]. In most cases, global demethylation of genomic DNA occurs in response to abiotic stress, but whether the same is happening in response to environmental climate change is not known. Methylation is known to play an important role in fine-tuning gene expression during plant development, as well as in response to the environment, enabling relatively rapid adaptation to new conditions without altering the DNA sequence
[73]. Evidence for the involvement of methylation in environmental adaptation is also provided by a study of 1001 Arabidopsis DNA methylomes described in The 1001 Epigenomes Project
[32]. These studies show that methylation levels within transposable elements positively correlate with latitude and precipitation, and negatively correlate with warmer temperatures. The question that arises in this context is whether plants, including forest trees, respond to climate change in the same way.
DNA methylation is essential for plant embryogenesis and seed development. Abnormal methylation in the embryo causes defects in embryogenesis, such as impaired cell division, aberration of the embryo apical domain, and reduced viability
[70][77]. The environment of the parents during reproduction also affects offspring performance. An example is the Norway spruce
Picea abies (L.) H. Karst seedling, which “remembers” the temperatures and photoperiod that prevailed during their embryonic life and seed maturation. This memory affects climatic adaptation in this species and is an epigenetic phenomenon
[70]. The existence of epigenetic memory in spruce may also explain the adaptive skills and rapid acclimatization of the spruce of Central European provenance in Norway
[78]. The researchers suspect that changes in the mRNA of genetically identical, somatic spruce embryos during morphogenesis, under different temperature variants, may be related to chromatin modifications. The significant role of DNA methylation and histone and small RNA (sRNA) methylation in the formation of epigenetic memory in this species may also be indicated by the different expression of epigenetic regulators, variable under different conditions for epitopes
[79][80]. The observation of global climate change makes the phenomenon of epigenetic memory during embryogenesis scientifically significant, as well as practically important in the context of forest research. Knowledge about the epigenetically regulated phenology of the vegetative buds of forest trees provides an idea regarding their ability to improve productivity, adaptability, and distribution potential during ongoing climate change
[70], giving them an advantage over other plants.
Studies of methylation clearly show that it is involved in a number of key plant biological processes and is central to many plant developmental processes
[76]. High-resolution genomic DNA methylation mapping studies based on the KEGG database have shown that methylated genes are involved in 118 metabolic pathways
[81]. Many methylated genes encode proteins involved in chromatin structure and DNA synthesis, cell cycle regulation, nitrogen metabolism, fatty acid synthesis and elongation, starch and sugar metabolism, amino acid metabolism, protein metabolism, brassinosteroid biosynthesis, the tricarboxylic acid cycle pathway, hormone metabolism, and signal transduction pathways. These studies show that DNA methylation is involved in a wide range of biological processes
[81], thus accounting for its great importance in development and environmental adaptation
[49].
In plants, methylation-induced modifications may or may not be reversible but can be retained during cell division (mitosis and intragenerational transmission) in a memory process. An example of epigenetic memory is vernalization or transmission to the next generation during meiosis (identification of natural epivariants or artificially induced epivariants and epigenetic recombinant inbred lines)
[1]. Furthermore, studies using epigenetically recombinant inbred lines (EpiRILs), where recombinant offspring are produced by crossing two parents with similar DNA sequences but strongly contrasting DNA methylation profiles, have shown that some of the DNA methylation variations are inherited in a Mendelian manner
[29].
2.4. Epigenetic Modifications of Trees and Environmental Conditions—A Review of Existing Research and the Current State of Knowledge
A study on
Pinus radiata D. Don
[82] examining seedling tolerance to heat stress and priming, based on evaluations of the nuclear proteome and DNA methylation dynamics, identified proteins involved in epigenomically driven gene regulation. The authors believed that priming-induced epigenetic memory might drive the development of new methods to improve pine survival under extreme heat stress in the context of climate change. Facilitating tree acclimation through environmentally induced epigenetic memory has also been previously suggested for winter dormant shoot apical meristems (SAMs) of poplar field crops
[83]. Understanding the mechanisms underlying phenotypic plasticity and stress memory in trees is extremely important in the context of rapid climate change. DNA methylation provides strong plasticity and modulates plant development, morphology, and physiology by controlling gene expression and transposable element (TE) mobility
[83]. A type of phenotypic plasticity is epigenetic memory in the Norway spruce
Picea abies (L.) H. Karst, and important factors for establishing this memory are DNA and histone methylation and sRNA
[4][70]. There are three categories of stress-memory genes
[84]. The first consists of “transcriptional memory” genes, in which stable transcriptional changes persist after a recovery period. The next contains genes called “epigenetic memory candidates”, in which stress-induced chromatin changes persist longer than the stimulus in the absence of transcriptional changes. The category following this comprises “delayed memory” genes, which are not immediately affected by the stress but receive and store the stress signal for a delayed response.
Memory-affecting climatic adaptation in Norway spruce is fixed at seed maturation during embryonic development and persists throughout the life of the offspring
[70]. This mechanism allows for long-term adaptive phenotypic changes. The authors arrived at such conclusions following a transcriptional analysis of spruce seedlings from seeds of several full-sib families derived from different temperatures of embryogenesis (cold vs. warm) under long- and short-day conditions
[70]. Epigenetic memory has evolutionarily important implications for trees growing in variable environments. In Norway spruce, which occurs over large areas, good adaptation to environmental conditions is provided by the epigenetic memory of temperature conditions during embryogenesis
[80]. The large size of conifer genomes may also indicate a greater need for epigenetic regulation of chromatin structure and maintenance of chromatin in a “dormant” or non-transcriptional state until activated in response to a changing environment
[4][80].
An intense decrease in global DNA methylation has been found in studies on winter-dormant shoot apical meristems of SAMs from natural populations of the black poplar
Populus nigra L. in France subjected to summer drought
[85]. To assess the extent of epigenetic changes, the authors examined common genetic parameters, such as narrow-sense heritability (h
2), the phenotypic differentiation index (P
ST), and the overall genetic differentiation index (F
ST). As the results showed, a significant decrease in DNA methylation in these populations was associated with drought stress. Studies have also quite clearly shown that global DNA methylation, genetically and environmentally determined, can serve as a marker of natural population differentiation under drought stress
[85], as well as performance or selection
[48].
Analysis of the white poplar Populus alba L. DNA methylation profiles from vegetatively propagated populations [86] showed that environmental conditions strongly influence internal cytosine hemimethylation. Eighty-three samples of white poplar at different locations in Sardinia were investigated by MSAP. The analysis was performed on genomic DNA extracted from leaves at the same juvenile stage. The study showed that the genetic biodiversity of poplar is quite limited but is balanced by epigenetic interpopulation molecular variation. The results clearly showed that ramets of the same clone were differentially methylated according to geographical location. In poplar, epigenetic changes are frequent and occur more rapidly in response to environmental stimuli, confirming the molecular model of stress epigenetic memory for plant responses to stress leading to increased overall methylation levels induced by external stimuli [87].
The relationship between environmental adaptation and DNA methylation has also been shown in studies on natural populations of the holm oak Quercus ilex L. of Mediterranean forests [88]. Methylation patterns and levels were assessed in individuals from control forest plots (in southern Catalonia, Spain) and in individuals experiencing drought stress (exposed to several years of drought at levels projected for decades to come). Drought-exposed plants had a percentage of hypermethylated loci lower than the control, while the percentage of fully methylated loci was significantly higher. These results also demonstrate that changes in DNA methylation contribute greatly to the ability of trees to rapidly acclimate to changing environmental conditions.
Nevertheless, this is only the beginning of a full understanding of the function and operation of the epigenome. Many of the processes that occur during epigenetic modifications and the modifications themselves that have been most thoroughly understood in model plants (also mentioned above) have not been confirmed in studies of forest trees. Therefore, there is still a long way to go to fully understand the function of epigenetic modifications in trees in the context of both abiotic stresses and strictly global climate change.