Usnic acid, as a biotic factor, changes the ploidy level in mosses

Abstract Lichens and mosses often share the same environmental conditions where they compete for substrate and other essential factors. Lichens use secondary metabolites as allelochemicals to repel surrounding plants and potential rivals. In mosses, endoreduplication leads to the occurrence of various ploidy levels in the same individual and has been suggested as an adaptation to abiotic stresses. Here, we show that also biotic factors such as usnic acid, an allelochemical produced by lichens, directly influenced the level of ploidy in mosses. Application of usnic acid changed the nuclei proportion and significantly enhanced the endoreduplication index in two moss species, Physcomitrella patens and Pohlia drummondii. These investigations add a new aspect on secondary metabolites of lichens which count as biotic factors and affect ploidy levels in mosses.


| INTRODUCTION
Bryophytes (i.e., mosses, liverworts, and hornworts) represent the first green plants which colonized land millions of years ago (Nickrent, Parkinson, Palmer, & Duff, 2000;Shaw, Szövényi, & Shaw, 2011). They developed various adaptations to survive in such a harsh environment such as alternation of gametophytic (haploid) and sporophytic (diploid) generations, elaborate gametophytes or specialized gametangia (Renzaglia & Garbary, 2001). A unique property of bryophytes is their rather small genome size (Goffinet & Shaw, 2009;Vanderpoorten & Goffinet, 2009) as well as their adaptation to various habitats during phylogeny (Bainard & Newmaster, 2010). In fact, they colonize all habitat types except salt water. Apart from facing unfavorable abiotic, environmental conditions, bryophytes also faced biotic stress factors and developed relationships with other organisms to obtain necessary resources.
Mosses and lichens often share similar habitats, where they compete for substrate as well as water, nutrients, and light (Lawrey, 1977;Macías, Molinillo, Varela, & Galindo, 2007). Lichens segregate secondary metabolites to the substrate to disadvantage their competitors. These secondary metabolites represent allelochemicals, which negatively influence vascular plants, mosses, or even lichens nearby (Armstrong & Welch, 2007;Molisch, 1938). As mosses and lichens have a long evolutionary history of allelopathic relations, they achieved survival strategies for cohabitation.
Endopolyploidy is defined as the existence of various ploidy levels in the same individual created by endoreduplication. This endoreduplication is a prerequisite of endomitosis and appears when DNA replication is not followed by mitosis. In general, nuclei with one chromosome set (1C) are monoploid, with two (2C) are diploid and those with three or more chromosome sets are defined as polyploid (Darlington, 1956). For many flowering plants, the ploidy levels of nuclei can vary in different cells, organs, or even tissues (Barow & Meister, 2003;D′Amato, 1964). First insights into endopolyploidy in bryophytes were given by Bainard and Newmaster (2010) who documented its presence in a dozen moss species. Interestingly, endopolyploidy is absent in liverworts and peat-mosses (Sphagnum spp.).
In vascular plants, abiotic factors such as temperature (Engelen-Eigles, Jones, & Phillips, 2000;Jovtchev, Barow, Meister, & Schubert, 2007), light (Kinoshita, Sanbe, & Yokomura, 2008), drought (Setter & Flannigan, 2001), or salinity (Ceccarelli, Santantonio, Marmottini, Amzallag, & Cionini, 2006) influence the level of endopolyploidy thereby causing effects on growth development as well as stress response (Inzé & De Veylder, 2006). Biotic factors as symbiotic and parasitic relationships also influence endoreduplication in plants (Callow, 1975;Lingua, D'Agostino, Fusconi, & Berta, 2001) fungi, bacteria, and roundworms can all have an impact on the endopolyploidy patterns. For bryophytes, this information is still rare; in particular, the effects of biotic triggers on the ploidy levels are missing. The moss Physcomitrella patens is a model organism in plant biology (Cove & Knight, 1993;Reski, 1999). Goga, Antreich, Bačkor, Weckwerth, and Lang (2017) showed effects of lichen secondary metabolite on its development and growth. Pohlia drummondii grows usually on heavy metal rich substrates and shares this habitat with lichen species of the genus Cladonia (Bačkor, 2011). Cladonia is documented to be rich in usnic acid (among other secondary metabolites). The aim of the study was to test lichen compound usnic acid as a biotic factor, which affects the ploidy level in the above moss species.

| Allelopathic assay
Preparation of moss material was carried out according to Goga et al. (2017). In brief, gametophores of P. patens and P. drummondii were collected from the Petri dish after 5 weeks on control medium and transferred to plastic tubes with deionized water (3 ml/3 gametophores). Subsequently, the moss material was homogenized by a tissue grinder (OMNI TH Homogenizer with Omni Tips ™ ). This procedure was carried out for P. patens and for P. drummondii separately. The homogenous suspension was further used in our allelopathic assay.
Sterilized glass fiber disks (Whatman CF/C filters, glass fiber disks, 25 mm in diameter) were placed on the surface of solid control medium. Usnic acid (UA, Aldrich Company 329967 5C) was dissolved in acetone, and stock solutions of different concentrations were prepared for the treatments (control, 0.01 mg of UA/disk, 0.1 mg of UA/disk). 50 μl of UA, corresponding to the respective treatments, was applied on the surface of each disk by an automatic pipette. Petri dishes with treated glass fiber disks were opened in a laminar flow cabinet for 1 hr to allow the acetone to evaporate.
Finally, 40 μl of homogenized moss suspension was applied on each glass fiber disk. Nutrients are able to pass through disk pores from the medium to the plantlets (Bačkor, Klemová, Bačkorová, & Ivanova, 2010). Mosses were cultivated for 5 weeks; each treatment was repeated at least ten times.

| Growth area rate
For growth rate analysis, each fiber disk was photographed after 5 weeks and the area occupied by plant material was measured.
Images were taken with a camera (Nikon D700, objective Nikon AF-S 50 mm f/1, 8G). The area on the fiber that was occupied by protonemata and gametophores was quantified using the GSA Image Analysis software (GSA, Rostock).

| Flow cytometry analysis of endopolyploidy level
Samples for endopolyploidy analysis were prepared from the whole available plant material grown on respective disks. To isolate cell nuclei, plant material was placed in a Petri dish and chopped in 1 ml of general purpose buffer using razor blades (Loureiro, Rodriguez, Doležel, and Santos (2007) (2) endoreduplication index (EI) representing the degree of endopolyploidy. EI was calculated according to Barow and Meister (2003) but modified as indicated below because the moss gametophyte represents a life phase with reduced chromosome complement: where n 1C + n 2C + n 4C + n 8C … represent the numbers of nuclei with the corresponding ploidy level (1C, 2C, 4C, 8C…). Plant samples with EI < 0.1 are not considered endopolyploid (Barow & Meister, 2003).

| RESULTS
The  Figure 4a). Therefore, this moss can be considered as an endopolyploid species, in addition to the species tested by Barow and Meister (2003). The proportion of 2C and 4C nuclei increased with UA application, even 8C nuclei were recorded (Figure 5a, Table 1). This is also reflected by the increasing EI which is significantly different between treatments (ANOVA, p = .014). Tukey`s pairwise post hoc test revealed significant differences between control and 0.1 UA treatment (p = .008241).
In gametophyte of P. drummondii, flow cytometry measurements detected three peaks on FCM histograms (Figure 3b,d) which correspond to 1C, 2C, and 4C nuclei. The data suggest that P. drummondii is a highly endopolyploid species. Application of UA led to an increase in 2C nuclei proportion (Figure 5b, Table 1) and a significant increase in EI (Figure 4b, t test, t = 2.46, p = .039). The 0.1 mg/disk of UA treatment was excluded from statistics because this treatment was mostly lethal and too little material was available for FCM measurements.
F I G U R E 1 Growth area of Physcomitrella patens and Pohlia drummondii after 5 weeks of cultivation on fiber disks F I G U R E 2 Growth area of Physcomitrella patens and Pohlia drummondii after 5 weeks. Physcomitrella patens is more adapted to exposition of usnic acid (UA) in studied concentration. Pohlia drummondii is very sensitive to highest concentration 0.1 mg/disk of UA (n = 12) In conclusion, the effect of UA on gametophyte growth is accompanied by an increasing number of endopolyploid cells in gametophytes. Endopolyploidy increased with increase in UA concentration (0 < 0.01 < 0.1 mg/disk) in both studied mosses. However, the EI of P. drummondii was higher than in P. patens.

| DISCUSSION
We have tested UA, the most common secondary metabolite in lichens, as a biotic trigger for endopolyploidy on two moss species P. patens and P. drummondii. We demonstrated that this lichen Usnic acid is present in many lichen species especially in the genera Alectoria, Cladonia, Evernia, Lecanora, Ramalina, as well as Usnea.
The concentration of UA in Usnea species varies between 0.22% and 6.49% of dry weight (Cansaran, Kahya, Yurdakulol, & Atakol, 2006) plants (Bialczyk, Latkowska, & Lechowski, 2011) or algae (Lokajová, Bačkorová, & Bačkor, 2014). Preexperiments in P. patens showed that very low concentrations of UA (0.001 mg/disk) caused no significant difference to control (Goga, unpublished data). Higher concentrations were lethal for mosses (Goga et al., 2017), and therefore, we focused on the concentrations used here for studied species. Endopolyploidy per se is a phenomenon just at the beginning of exploration and its biological significance is still far from being understood (Bainard & Newmaster, 2010). A comprehensive view of the mechanism(s) and its roles in the plant kingdom in general and bryophytes in particular are rare. Thus, the importance of endoreduplication in the plant kingdom is subject to different hypotheses. One of them suggests that endoreduplication provokes changes in the activity of cyclin-dependent kinases which are responsible for the normal changeover of the cell cycle (Inzé & De Veylder, 2006). Endopolyploidy might therefore be seen as a response in growth and development.
Indeed, Goga et al. (2017) showed that lichen metabolites provoked exactly those effects in P. patens. Although the molecular mechanism remained unknown, polyploidization in mosses definitely plays a role, as is evident here. Another hypothesis relates endopolyploidy to the chromosome size. According to Schrader and Hughes-Schrader (1955) and Leitch, Chase, and Bennett (1998), chromosome size of natural polyploids is smaller than in related diploids. Accordingly, small chromosome size is a preadaptation to the origin of polyploidy (Darlington, 1956). A third hypothesis supports the "nucleotypic theory" and is based on the fact that cell volume and other phenotypic traits are influenced by DNA amounts (Barow & Meister, 2003). These authors suggested for vascular plants that small genome size and endopolyploidy degree are correlated positively. Therefore, one possible function of endopolyploidy is the compensation of nuclear DNA deficiency especially in species with small genomes. This is confirmed by our research in the two moss species as well. Namely, P. drummondii with 11 chromosomes has a higher EI than P. patens with 27 chromosomes. Furthermore, the cell type and age played a role in the degree of endopolyploidy in P. patens as well, and different ploidy levels were expressed in various tissues (Schween, Gorr, Hohe, & Reski, 2003). In addition, endopolyploidy may be a response in the adaptation to various habitats during phylogeny. In mosses, endopolyploidization per se is documented and various abiotic factors can be triggers for its appearance, level, and expression degree. Those abiotic conditions including temperature, light, and drought cause different levels of endopolyploidy in individual bryophytes species (Bainard & Newmaster, 2010). Here, we show that also biotic factors such as lichen secondary compounds affected the endopolyploidization status in mosses and that it differed in the two investigated species.

| CONCLUSION
In this pioneer study, we demonstrated that lichen compounds influenced the level of endopolyploidy in mosses. We confirm the occurrence of endopolyploidization in two additional moss T A B L E 1 Proportion of nuclei classified according to their DNA ploidy level in two mosses Physcomitrella patens and Pohlia drummondii. (means that no data were recorded) species P. patens and P. drummondii, supplementary to 46 bryophyte species previously tested by Bainard and Newmaster (2010).
Furthermore, UA as a new biotic factor is documented to affect the endoreduplication in P. patens and P. drummondii. UA is shown to impose allelopathic effects on both mosses suggesting a role in the natural environment as well. In the competition for the common resources, endoreduplication might enhance the chances of the moss by giving vigorous growth and better development. This is, however, only true for physiological concentrations as higher amounts of UA strongly decrease the developmental potential, at least in P. drummondii.

ACKNOWLEDGMENTS
This work was supported by Slovak Grant Agency VEGA 1/0792/16.
We thank Jimmy Rae for critical reading and reviewing the English correction of the manuscript.

CONFLICT OF INTEREST
None declared.

AUTHOR CONTRIBUTIONS
MG, DR, and IL-conceived and designed the experiments; VK and DR-analyzed data by flow cytometry; MG, MS, and IL-performed cultivation of mosses; MG and MB-performed growth data analysis; MG, DR, IL, and MS-wrote the manuscript.