Potential role of DNA methylation and histone H3 acetylation in parasitism of Cuscuta

Cuscuta campestris, or field dodder, is a non-photosynthetic, parasitic weed and a prominent agricultural pest. It is widely distributed in various climatic conditions and parasitizes over 100 different host plants in Bulgaria. The objective of the present study was to test the hypothesis that epigenetic regulation is key to the adaptation of Cuscuta spp. to different environments and host plants, as well as the response of host plants to parasitism. First, we tested 12 populations of C. campestris. Still, we found no correlation between the percentage of methylated DNA or acetylated histones and the most common host plant species, although significant differences were found. However, the effect of Cuscuta parasitism was much more pronounced on the enzymes, responsible for both DNA methylation and histone acetylation in the host plants. While changes were not so notable in the susceptible model host, Arabidopsis thaliana, the resistant host Solanum lycopersicum responded to parasitism with increased DNA methyltransferase and histone deacetylase activities and decreased DNA demethylation. All this suggests adaptation to parasitism, which is partially guided by epigenetic regulation. Further studies, including mapping of the methylated regions, are needed to better understand the epigenetic mechanisms of host adaptation to Cuscuta parasitism.


Introduction
Plant parasitism represents the ability of one plant to feed and grow at the expense of another plant, called the host.The connection between them is mediated by an organ, called haustoria which provides a direct connection with the vascular systems of the two plants [1].Cuscuta campestris Yunck., or field dodder is a widespread parasitic plant species with origin in North America but currently distributed worldwide.It has a wide spectrum of host species, including crop plants, and thus it represents a significant agricultural pest [2].Some plant species are, however, naturally resistant to infection with Cuscuta sp.Such are the members of the family Poaceae, which is supposed to be due to the anatomical characteristics of monocots [3].In dicots, a certain resistance to Cuscuta infection was reported in tomatoes (Solanum lycopersicum L.).When subjected to infection with Cuscuta reflexa Roxb., a hypersensitive response was triggered, in the haustoria penetration sites.On the molecular level, the cell wall of the tomato lignifies, auxin levels increase and 1305 (2024) 012016 IOP Publishing doi:10.1088/1755-1315/1305/1/012016 2 phenylpropanoids and long-chain fatty acids are being accumulate, thus preventing the parasite from successful haustoria formation [4].
Parasitism in Cuscuta is known to be induced by different stimuli, including light quality [5] and chemical clues [6].Haustoria formation is of course driven by differential gene expression, especially related to hormonal signaling, cell-wall loosening processes, etc. [5].On the other hand, Cuscuta parasitism induces a wide spectrum of gene expression changes in the host plant, either as a result of defense reaction [4], or on the contrary, a beneficial shift in gene expression, orchestrated by the parasite itself [7].It would not be exaggerated if we conclude that the process of haustoria formation and further parasitism is driven by extensive changes in gene expression of both the parasite and the host.Recent studies showed, that Cuscuta may manipulate the host gene expression profile by targeting mRNAs by its microRNAs, which are transferred through the haustoria into the host [8].However, the classical epigenetic mechanisms of gene expression control, DNA methylation and histone modification, seem to be understudied in the Cuscuta -host plant interactions.
DNA methylation represents the reaction of binding a methyl group to the 5`C of cytosine (5-methyl cytosine, 5mC) [9].Donor of methyl groups is S-adenosyl-L-methionine.This modification provides binding sites for proteins, which can prevent association of initiation transcription factors, and a decrease in gene expression is observed.DNA methyltransferases (DNMT, EC 2.1.1.37)are the enzymes that control DNA methylation.They are responsible for genome imprinting (copying the mother methylome in the next generation) [9].The number of DNMTs in plants may vary significantly between species, but there are three major classes, each with different number of representatives and differing in function.Domains Rearranged Methylases (DRM1 and DRM2) are mainly involved in de novo DNA methylation during plant development, primarily targeting cytosine residues in CpG and CpNpG contexts [10].Chromomethylases (CMTs) are plant-specific DNA methyltransferases that target CpNpG methylation.Finally, Methyltransferase 1 (MET1) is the enzyme, responsible for maintaining CpG methylation during DNA replication, e.g., it is the responsible for inheritance of the methylation pattern of the mother strand [10].There are accumulating evidence that all of them are involved in both abiotic and biotic stress response in plants [11].In plants there is also another mechanism of DNA methylation where RNAs are involved -RNA-dependent DNA methylation [12].
Demethylation of DNA can be passive and active.Passive demethylation occurs when there are low concentrations of S-adenosyl-methionine.Active demethylation is controlled by DNA demethylases.In plants there are bifunctional 5-methyl-cytosine-glycosylases and apurinic/apyrimidinic lyases which recognize 5mC and generate abasic site.A base excision repair mechanism is then activated.In A. thaliana there are four families of bifunctional glycosylases -REPRESSOR OF SILENCING 1 (ROS1, EC 4.2.99.18),TRANSCRIPTIONAL ACTIVATOR DEMETER (DME, EC 3.2.2.X), DEMETER-LIKE PROTEIN 2 (DML2, EC 3.2.2.X) and DML3 (EC 3.2.2.X).The mechanism consists of removing the base with the glycosylase activity then a breakage of the backbone of DNA is being made and the 5mC is removed.Then DNA polymerase and DNA ligase repair the chain [13].
Posttranslational histone modifications play a crucial role in regulation of gene expression by affecting the degree of binding of the histones to the DNA, thus modifying the access of transcription factors to it [14].Out of dozens of different known modifications, acetylation of histone H3 is one of the best-known and widely affected by environmental factors [15].Histone acetylation has a tendency of activating gene expression, while respective histone hypoacetylation decreases the levels of expression.Histone acetyltransferases (HAT, EC 2.3.1.48)add acetyl groups, while histone deacetylases (HDAC, 3.5.1.98)remove them.Considering the central importance of DNA methylation and Histone H3 acetylation for the regulation of gene expression in general, we could expect a significant shift in the activities of all enzymes, involved in these modifications during parasitism of Cuscuta in both the parasite and the host plant.

Plant material and growth conditions
Vegetative material of Cuscuta campestris, determined morphologically, was collected from 12 localities on the territory of Bulgaria and the main host plants were also determined (Table 1).Seeds of the Ribnovo population were used for laboratory experiments.For germination C. campestris seeds were subjected to chemical scarification with concentrated H2SO4 for 30 min.After several washings with water, the seeds are sown on commercial soil and transferred near a host for infection after germinated and reached length of 2-3 cm.Arabidopsis thaliana L. (Ecotype Colombia, Col-0, Nottingham Arabidopsis Stock Centre) seeds were first placed on moistened filter paper in a petri dish in the dark at 4°C for 24 to 48 hours for stratification.Then they were transferred to soil.Solanum lycopersicum L. cv.Oxheart were grown from commercially available seeds and sprouts directly in the soil.The plants were cultivated in greenhouse conditions with a natural photoperiod of 14/10 hours.

DNA isolation
DNA was isolated from the samples with Plant Genomic DNA Kit, Abbexa according to manufacturer's instruction.DNA concentration (absorbance at λ = 260 nm) and DNA purity (A260/A280 ratio) were measured on NanoDrop 2000 (ThermoScientific), and DNA integrity was additionally checked on 0.8% agarose gel electrophoresis.

Protein isolation
Histone isolation was conducted according to the method, described by Rodriguez-Collazo [16].Briefly, 100 mg of plant material was powdered with liquid nitrogen and dissolved in 0.075 M NaCl in 0.005 M Tris HCl pH 8.After centrifugation (10 min at 10,000 g, 4 o C) 0.2 M H2SO4 is added to the pellet to dissolve acid-soluble proteins.Samples were incubated on ice for 2 hours, centrifuged for 10 min at 10,000 g, 4 o C and proteins in the supernatant were precipitated with 0.5 volume of 50% trichloroacetic acid.After centrifugation the pellet was washed with acetone and finally dissolved in 8 M urea with 5% CH3COOH.Proteins for enzymatic activities were isolated using the EpiQuik™ Nuclear Extraction Kit (EpigenTek).Protein concentration was measured by the method of Bradford [17].An equal amount of 20 μg of nuclear extract was used in all experiments.However, it is important to note, that all four kits detect either accumulation of product or decrease in substrate concentrations, irrespective of the enzyme, e.g., they are not specific toward a particular DNMT, for example.

Software and statistical analysis
All figures were constructed using GraphPad Prism version 8.0.0 for Windows, GraphPad Software, San Diego, California USA.The same software was used for statistical analyses -One-way ANOVA with Tukey's multiple comparison test.

Ethical statement
Permission for the field study conducted on public land was not provided because the studied locations do not fall into protected areas and no specific permission is required.The studied species is not protected.On the contrary, it is a non-native, invasive weed.

DNA methylation and histone H3 acetylation in wild Cuscuta campestris populations
Cuscuta campestris specimens from a total of 12 localities were studied (Table 1).Although the parasitic plant can simultaneously parasitize multiple host plant species and is rarely found attached to a single species, the dominant host species could be easily identified.In this respective study, the most preferred host species were Convolvulus arvensis, Portulaca oleraceaea, Lactuca seriola, Polygonum aviculare and Xanthium italicum.Some of them are relatively common hosts for C. campestris in Eastern Europe.For example Polygonum aviculare was established as the most common host of this particular species in Hungary, accounting for nearly 70% of the hosts [18].Our main hypothesis was that levels of DNA methylation and histone H3 acetylation would correlate with the dominant host species.However, the results for both % methylated DNA and acetylated histone H3 (Fig. 1) showed substantial variation among localities, without a clear pattern, related to the dominant host species.Most probably, the level of epigenetic modifications depends on several factors, including the host identity, but also environmental factors [19].

Short-term epigenetic changes in the host
To assess the impact of Cuscuta campestris parasitism on the host epigenetic control, we employed two contrasting host-parasite systems.First, the model plant Arabidopsis thaliana showed little to no defense response (Fig. 2A) and nearly 100% successful infections.In contrast, Solanum lycopersicum responded to Cuscuta infection with characteristic browning (Fig. 2B) and gradual dying out of the parasite.Less than 1% of the infections were successful and lead to development of the parasite.Therefore, we measured the activities of epigenetic regulation involved enzymes during the initial stage of haustoria formation and at the stage of developed parasitic plant (defined by at least 3 cm long secondary stem).The data for both DNA methyltransferase and DNA demethylase actitivies (Fig. 3 A,B) revealed comparatively low activities in Arabidopsis, as compared to tomato.Most importantly, no significant changes occurred either in the initial stages of infection (haustoria formation) or during active development of the parasitic plant.In contrast, during Cuscuta infection the activity of DNA methyltransferase in tomato increased reciprocally to DNA demethylase activity, suggesting selective inhibition of gene expression [9].This finding is in agreement with a relatively old study, showing DNA hypermethylation in host Medicago sativa, subjected to Cuscuta reflexa infection [20].Similarly, the activities of histone acetyltransferase and

Conclusions
Apparently, the presented results are not sufficient for a conclusive conception of the epigenetic control of Cuscuta infection.However, they clearly show that both DNA methylation and histone H3 acetylation play a crucial role in both the parasite and the host plant.Especially in the host, these processes are much more intensive in plant species, in which an active defense reaction occurs, in comparison to fully susceptible host, as Arabidopsis is.Further methylome sequencing and additional studies of histone modifications, in combination with transcriptomics analyses are needed to fully elucidate the epigenetic control of plant-toplant parasitism.

Figure 1 .
Figure 1.Percentage of methylated DNA (A) and acetylated histone H3 (B) in Cuscuta campestris specimens from different localities.The dominant host species is shown at genus level.Mean values +/-SEM from three replicates are shown.Different letters indicate significant differences at P < 0.05, Oneway ANOVA, Tukey's multiple comparison test.

Figure 2 .
Figure 2. Contrasting response of Arabidopsis thaliana (A) and Solanum lycopersicum (B) to Cuscuta campestris infection.The data for both DNA methyltransferase and DNA demethylase actitivies (Fig.3 A,B) revealed comparatively low activities in Arabidopsis, as compared to tomato.Most importantly, no significant changes occurred either in the initial stages of infection (haustoria formation) or during active development of the parasitic plant.In contrast, during Cuscuta infection the activity of DNA methyltransferase in tomato increased reciprocally to DNA demethylase activity, suggesting selective inhibition of gene expression[9].This finding is in agreement with a relatively old study, showing DNA hypermethylation in host Medicago sativa, subjected to Cuscuta reflexa infection[20].Similarly, the activities of histone acetyltransferase and

Table 1 .
Cuscuta campestris populations, used in this study EpigenTek) manual was strictly followed.All samples were measured in triplicates.For global methylated DNA equal amounts of 70 ng isolated DNA were used for each sample and for acetylated H3 100 ng of isolated histones.Both positive (in several dilutions) and negative controls, provided with the kits were used.The absorbance was measured at 450 nm on a Microplate reader DR-200B (Diatek Instruments, Wuxi, China).Percentage of 5mC and acetylated H3 (in ng per mg protein) were calculated using the recommended equations (EpigenTek).Similarly, histone acetyltransferase activity was measured with EpiQuik™ HAT Activity/Inhibition Assay Kit, histone deacetylase with EpiQuik™ HDAC Activity/Inhibition Assay Kit (Colorimetric), DNA methyltransferase with EpiQuik™ DNMT Activity/Inhibition Assay Ultra Kit (Colorimetric) and DNA demethylase with EpiQuik™ DNA Demethylase Activity/Inhibition Assay Ultra Kit, following manufacturer's (EpiGenTek) procedures and calculations.