1. Introduction
Heavy metals are amongst the most hazardous pollutants to living organisms due to their high toxicity and persistence in the environment1–4. Multiple studies showed that heavy metals damage the cell membrane system and negatively affect plant development, growth, and reproduction 5–8, generate reactive oxygen species (ROS) 9,10, and may alter the structure of essential plant biomolecules 11–14. Still, some species have adapted to living in heavy metal-enriched environments, and are able to survive and accumulate concentrations of heavy metals that greatly exceed the tolerance limits of most other plants15–17.
Metal tolerant species often rely on the enhanced expression of genes related to heavy metal homeostasis – i.e. transport, chelation, and sequestration of metals – relative to closely related non-tolerant ones18–23. Thus, constitutive overexpression of heavy metal homeostasis-related genes (e.g. ABC metal transporters and enzymes involved in oxidative stress relief 20,21) often underlies differences between tolerant and non-tolerant plants (but see24,25 for the contribution of gene copy number variation). Yet, the regulatory mechanisms determining the differences in gene expression are still poorly understood. Gene expression in eukaryotic genomes is, in part, tightly regulated by epigenetic mechanisms, the set of chromatin modifications that control chromatin structure, and thus, the accessibility of the transcriptional machinery to genes 26,27. In plants, epigenetic mechanisms, such as DNA methylation, histone modifications, and small non-coding RNAs, are known to regulate gene expression changes in response to developmental and environmental cues (e.g. 28–33). Based on this evidence, epigenetic mechanisms are currently considered potentially important regulators of plants response to stress.
So far, experimental evidence on the contribution of epigenetic mechanisms to plant heavy metal tolerance has shown that DNA methylation could have a direct protective role. For example, metal tolerant plants like Arabidopsis halleri - in response to Cd - and Noccaea caerulescens - in response to Ni showed hypermethylation of DNA and no heavy metal-induced DNA damage 34,35. Non-tolerant species like the moss Physcomitrium patens , however, showed overall DNA hypomethylation and significant signs of DNA strand breaks in response to Mn 36. Studies employing anonymous DNA methylation markers or immunolabelling techniques reported DNA hypermethylation in response to exposure to high Al levels in wheat (Triticum aestivum ) 37 and maize38, to Pb in maize (Zea mays)39, to Cd in Posidonia oceanica40 and Nicotiana benthamiana41, and to Cd, Ni and Cr in Trifolium repensand Cannabis sativa 42. In contrast, DNA hypomethylation was reported in response to Cd in rapeseed (Brassica napus ) 43 and the red algaGracilaria dura 44, to Al in triticale (xTriticosecale ) 45,46, to Cu and Zn inPopulus alba 47, and to low Al levels in wheat37. Epigenetic mechanisms like DNA methylation can thus be modified in response to heavy metal exposure.
More targeted studies have demonstrated that epigenetic changes induced by heavy metals can lead to changes in the expression of genes involved in heavy metal tolerance. For example, work with methylation mutants confirmed that CG and H3K9me2 hypomethylation upstream of the metal responsive metallochaperone, OsHMP , led to its overexpression and enhanced Cd tolerance in rice (Oryza sativa )48. Similar DNA and histone hypomethylation led to increased expression of a Cd tolerance factor (OsCTF ) that enhanced Cd tolerance in rice 49. Other studies used transformation of Arabidopsis thaliana with an S-adenosylmethionine synthetase (SAMS) gene, which encodes an enzyme that catalyzes the biosynthesis of SAM, the main methyl group donor for DNA methylation in plants. This transformation conferred Al as well as Cd, Cu, and Zn tolerance to this species 50. The overexpression of specific heavy metal detoxification transporters in a Pb, Cd, and Zn tolerant variety of wheat has been linked to CG hypomethylation in their promoter region in response to all three metals51. Prolonged Cr exposure during selection of a Cr-tolerant strain of the green alga Scenedesmus acutus led to new DNA methylation and expression patterns in protein-coding genes involved in Cr tolerance. MicroRNAs also seem to contribute to plant response to Cr stress in rice 52, Cd in rice53,54 and rapeseed 55,56, and Al, Cd, and Hg in Medicago truncatula 57. Finally, some studies showed that DNA methylation changes induced during heavy metal exposure can be transmitted to the offspring in rice58,59 and Arabidopsis 60potentially leading to heritable changes in gene expression patterns. Collectively, these data suggest that epigenetic regulation could play an important role in the plant response to heavy metal stress. However, these studies are based on a phylogenetically restricted set of species, largely from the Brassicaceae and Poaceae (i.e. crops and model plants like Arabidopsis ), which may not be representative of plants as a whole.
Bryophytes, the sister group to the vascular plants, have long been known for their capacity to tolerate high concentrations of heavy metals in their tissues 61. Most of what is presently known about their tolerance mechanisms comes from biochemical and physiological studies. Mechanisms such as retention of heavy metals in extracellular wax-like substances 62, sequestration in the cell wall 63–66, preferential accumulation in specific parts/organs of the leafy plant – i.e. gametophore67–69, or the activation of the ROS scavenging machinery and induction of metal-chelating proteins70, have been reported to mitigate heavy metal stress in bryophytes. However, we lack data concerning the molecular underpinnings of these anatomical or physiological phenomena. Filling this gap will enable us to test whether the land plants employ conserved pathways to tolerate heavy metal stress, or whether each lineage has evolved novel mechanisms.
In this study, we used a reduced representation bisulfite sequencing technique (hereafter epiGBS 71), and RNA sequencing (RNAseq) to quantify the effect of Cd and Cu exposure on DNA methylation and gene expression in the copper moss Scopelophila cataractae(Mitt.) Broth, a species with high affinity for heavy metal-, especially Cu-enriched substrates 72. We generated laboratory cultures, in control or test (either Cd or Cu) conditions, of samples from four field microsites (maximum distance between microsites <500 m) within a former copper mine. Using this experimental set up, we previously found that phenotypic differences for Cd and Cu tolerance (defined as the ability to maintain vegetative growth in metal stressed vs. control conditions, sensu 73) in this species were maintained in the laboratory 68. The more tolerant plants were found in the center of the mine growing in microhabitats devoid of any vegetation and exposed to higher metal concentrations, i.e., harsher environmental conditions, while less tolerant plants were from the mine edges, located in milder, less contaminated microhabitats. Here we used epigenetic and transcriptomic analyses to identify the mechanisms driving intraspecific differentiation for heavy metal tolerance in S. cataractae . Specifically, we addressed the following questions: (1) What are mechanisms driving intraspecific differentiation for heavy metal tolerance in S. cataractae ? (2) Does heavy metal exposure affect DNA methylation? (3) What are the main functional changes in S. cataractae in response to heavy metal exposure? (4) Is there any association between methylation and expression changes? Our results provided evidence for non-genetically-based intraspecific phenotypic differentiation for heavy metal tolerance in this species with the more tolerant plants investing more in constitutive protection mechanisms and being more efficient in entering a conservative state when faced with acute Cu stress.