4 | Discussion
This study reveals functional characteristics of lichen symbioses by contrasting gene expression patterns of the fungal partner growing in associations with different photobionts, i.e., thallus sectors containing cyanobacteria or green algae as predominant photosynthetic partners. We focused specifically on the fungal genes differentially expressed between morphs as well as three different temperature treatments, and on temperature-related differential expression of green algal and cyanobacterial genes. Our analyses reveal fungal gene expression differences mediated by different photobionts and temperatures. In the threesome partnership of P. britannica , stress responses are triggered at markedly different temperatures in cyanobacteria, lichen-forming fungi and green algal symbionts.
4.1 | Ascomycete genes / photomorph
Differential gene expression in lichen-forming fungi is mediated by interactions with different photosynthetic partners. For example, the upregulation of an ascomycetous isopenicillin N synthetase in the cyanomorph could be attributable to mycobiont-photobiont interactions. Metabolic interactions between lichenized fungi and their photosynthetic partners have been shown to be able to affect the production of lichen substances (Shrestha & St. Clair, 2013), some of which have antibiotic (Gazzano et al., 2013; Shrestha & St. Clair, 2013) or growth-inhibitory properties (Ocampo-Friedmann & Friedmann, 1993; Ranković & Mišić, 2008). Penicillin is a known fungal β-lactam antibiotic which mainly controls gram positive bacteria (Holtman, 1947) and isopenicillin N synthetase is essential for the production of penicillin (Müller et al., 1991). By producing this antibiotic, the lichenized fungus in our study might be able to control growth and population size of the gram-positive part of its bacterial microbiome and potentially also of its intrathalline cyanobionts. This is further indicated by the upregulation of fungal genes encoding velvet domain-containing proteins in the cyanomorph, as these genes also play a role in secondary metabolism and antibiotic biosynthesis in model fungi Penicillium andAspergillus (Kopke et al., 2013; Kato et al., 2003). Excessive growth of Nostoc photobionts may be detrimental to the symbiosis, as Nostoc produces toxins in various lichen species, especially in those growing in humid climates (Kaasalainen et al., 2012), such asP. britannica . The mycobiont could be able to avoid cyanotoxin-induced damage by expressing isopenicillin N synthetase and velvet domain proteins in the cyanomorph. Subsequent studies are necessary to test this hypothesis. Currently it remains unclear if otherPeltigera species living in symbiosis with only cyanobacterial photobionts also express genes involved in the production of antibiotics. We hypothesize that the regulation of cyanobacterial growth could be important for maintaining long-term cohesion of largePeltigera thalli under varying environmental conditions. The upregulation of the antibiotic proteins depended on temperature as well, but the pattern was not uniform. This suggests upregulation results from a combination of factors, like the presence of Nostoc and temperature changes.
In either photomorph, different genes responsible for cell wall synthesis or cell wall modification were upregulated, e.g. SUN domain proteins in the cyanomorph and chitin synthase in the tripartite morph (downregulated at 25 °C) (Gastebois et al., 2013; Garcia-Rubio et al., 2020; Bowman & Free, 2006). Fungal hyphal structures and their cell wall play a vital part in the infection of host organisms such as plants (Hopke et al., 2018) and in the process of host interaction (Geoghegan et al., 2017) – as is the case in lichen symbioses (Kono et al., 2020; Honegger, 1986). In addition, various environmental factors influence the remodeling of fungal cell walls (Patel & Free, 2019). The lichenized fungus might interact differently with its two photosynthetic partners and, mediated by its respective partner, different aspects of cell wall (trans)formation may be required to establish contact sites. The green algal partner Coccomyxa contains resilient sporopollenin biopolymers in its cell walls, which the fungus cannot penetrate or degrade (Honegger & Brunner, 1981). Instead of haustoria, the mycobiont forms wall-to-wall appositions (Honegger, 1984). The cyanobacterial partner Nostoc usually isn’t penetrated either (Honegger, 1984, 1985), but the hyphae encircle the Nostoc cells tightly and sometimes even invaginate them (Pawlowski & Bergman, 2007). Also, in Peltigera species with Nostoc , intrawall haustoria have been described, which penetrate the membrane of the cyanobacterial cell wall (Koriem & Ahmadjian, 1986). The establishment of these contact sites has an effect on the symbiotic relationship as they allow nutrient transfer between the symbionts (Kono et al., 2020) which is often considered the functional core of lichen symbioses. However, differential gene expression of cell wall modifying genes induced by photosynthetic partners may impact lichens on a much broader scale. They could be involved in the formation of the strikingly different phenotypes which P. britannica (alongside other lichens) develops with the goal to optimize different aspects of the symbiotic relationship. In foliose lichens (such as Peltigera ) the mycobiont actively positions the photobionts within the thallus to ensure optimal acclimatization to its environment (Honegger, 2012). The mycobiont might also control growth and proliferation of the photobionts (Honegger, 2012; Hyvärinen et al., 2002). However, co-development of the symbionts is needed for the successful establishment of a lichen thallus (Hill, 1985). Conclusively, the differential expression of cell wall (trans)formation genes between morphs could cause the different phenotypes or at least contribute to their formation.
The interaction between mycobiont and photobiont can also be seen in the upregulation of ascomycete stress response genes, e.g., that of glutathione-S-transferase (GST) in the tripartite morph and that ofpara -aminobenzoic acid (PABA) synthase in the cyanomorph; the expression of both of these enzymes was temperature-dependent as well, as they were downregulated at 25 °C. GSTs are, inter alia, active during oxidative stress (Morel et al., 2009). As desiccation leads to an accumulation of reactive oxygen species (ROS), oxidative stress occurs in desiccating organisms, including lichens (Kranner et al., 2008; Holzinger & Karsten, 2013). A major player providing tolerance towards oxidative stress is glutathione (GSH), an antioxidant which reduces ROS, whereby GSH itself is being oxidized into glutathione disulfide (GSSG) (Kranner et al., 2008). The enzyme GST is a major driver of this detoxification process as it is responsible for the conjugation of GSH onto ROS (Kammerscheit et al., 2019). Lichens that tolerate desiccation to a higher degree usually contain an increased GSH pool in their hydrated state, enabling a rapid oxidation to GSSG during desiccation and causing an increased GSSG pool (Kranner, 2002). The upregulation of GST in our samples could therefore be a result of initial desiccation and desiccation-induced accumulation of ROS, even though the thalli were watered regularly and were sampled while fully hydrated. The upregulation of GST occurred in the tripartite morph only – although both morphs received identical treatment. This pattern might be due to an increased desiccation tolerance of lichens with green algal photobionts, relative to those with cyanobacteria (Kranner, 2002), or it might reflect inherent differences in physiological properties among photomorphs, e.g. water holding capacities of the gelatinous sheath ofNostoc sp. (Liang et al., 2014). Future studies should conduct specific experiments to characterize these putative physiological differences. Furthermore, the expression of PABA synthase in the cyanomorph indicated stress response, as this enzyme has been shown to improve tolerance to thermal stress in Agaricus bisporus (Lu et al., 2014) and to enhance UV-C resistance in Arabidopsis thaliana(Hu et al., 2019). As cyanobacteria are more susceptible to heat stress (see below) as well as high light when hydrated than green algae (Gauslaa et al., 2012), the expression of PABA synthase could be a response of the lichenized fungus to environmental stress in which the mycobiont supports the photobiont in keeping the holobiont vital.
Photomorph-dependent differential expression of genes involved in fungal carbon metabolism were expected as both partners produce distinctive carbon-based secondary compounds. Green algae like Coccomyxaproduce polyols such as ribitol (Richardson & Smith, 1968), whereas cyanobionts produce glucans and glucose (Hill, 1972), all of which are translocated and taken up by the mycobiont. In the mycobiont, these carbohydrates are converted into the energy-storing compound mannitol (Palmqvist et al., 2008; Grzesiak et al., 2021). The obtained sugars either serve a nutritional purpose (i.e. growth and respiration) (Palmqvist et al., 2008; Smith, 1963) or they are conducive to stress tolerance (e.g. protection during desiccation) (Farrar, 1976; Spribille et al., 2022). As the carbon-based substrates are distinct, the mycobiont requires different enzymes for substrate transport and transformation. Although various genes responsible for carbohydrate metabolism were found upregulated in both morphs, none of these could be assigned directly to glucan or ribitol metabolism. Yet, a number of studies have shown the complex nature of carbohydrate movement within lichens and have emphasized photobiont-induced distinctions (Smith et al., 1969; Hill, 1972; Richardson et al., 1967; Palmqvist et al., 2008; Hill & Ahmadjian, 1972; Kono et al., 2020). In our study, the expression of ascomycete genes involved in carbon metabolism was temperature-dependent in some cases, and some genes (e.g., carbohydrate esterase family 4 and galactonate dehydratase) were upregulated at 4 °C_1 and 4 °C_2, which might have to do with cold tolerance mechanisms, as carbon metabolism plays a role in tolerance and acclimation to cold. In photosynthetic organisms, primary as well as secondary carbon metabolites have been proven to be essential to withstand cold temperatures (Fürtauer et al., 2019; Calzadilla et al., 2019; Tarkowski & Van den Ende, 2015). The lichenized fungus might respond similarly to cold temperatures by metabolizing the carbohydrates it obtains from its photosynthetic partners; indeed, previous studies have shown that polyols, such as ribitol and mannitol, serve as cryoprotectants (Fontaniella et al., 2000; Hájek et al., 2009).
We found over 300 photomorph-mediated differentially expressed genes in the lichenized fungus. More than half of these genes were also differentially expressed at different temperatures. Therefore, the differential expression of these ascomycete genes appears to be the result of a combination of factors – photobiont type plus specific stimulus. However, the results clearly indicate photobiont-mediated differential fungal gene expression.
4.2 | Ascomycete genes / temperature
A stepwise temperature increase from 4 to 25 °C resulted in the upregulation of various stress-response genes in the lichenized fungus, with upregulation beginning already at 15 °C. Heat shock proteins and chaperonins are proteins directly involved in stress responses, giving a clear indication that the organism is stressed at elevated temperatures when it has been pre-acclimated to cold; as does the upregulation of proteins which are only indirectly involved in stress response mechanisms, such as the Rad60-SLD domain and ARPC5. SLDs are SUMO-like domains and – as SUMO (small ubiquitin-like modifier) proteins – SLDs are responsible for the SUMOylation of a range of other proteins (Ghimire et al., 2020; Prudden et al., 2009). Protein SUMOylation is of great relevance as it renders targeted proteins useful for various vital biological processes. Heat stress has been described as one of the factors leading to increased SUMOylation activity (Zhou et al., 2004). Shortly after a rise in temperature, SUMO conjugates accumulate, indicating that SUMOylation might be an early stress response system (Kurepa et al., 2003). SUMOylation is the starting point of a cascade of cellular processes in reaction to stress as it activates target proteins, such as heat shock factors, which in turn activate specific proteins, such as heat shock proteins (Kurepa et al., 2003). Activation of HSPs as a consequence of SUMOylation has been described for various organisms, including A. thaliana (Kurepa et al., 2003) andCandida albicans (Leach et al., 2011). The upregulation of Rad60-SLD in P. britannica photomorphs at 15 and 25 °C therefore indicates fungal responses to thermal stress. Similarly, the expression of ARPC5 at 25 °C reflects stress response processes. ARPC5 is a member of the multiprotein complex Arp2/3; in the nucleus, the Arp2/3 complex contributes to DNA repair mechanisms as it promotes migration of DNA double-strand breaks which are to be repaired (Schrank et al., 2018). An upregulation of DNA repair mechanisms at high temperatures is expected because elevated temperatures can cause heat-induced DNA damage (Oei et al., 2015; Steinhäuser et al., 2016).
Furthermore, in both photomorphs, an upregulation of ascomycete transposon genes was detected at 25 °C. Transposons, or transposable elements (TE), are DNA sequences which can change their position in the genome (Muñoz-López & García-Pérez, 2010). Transposon translocations can affect gene functioning, especially when they are inserted into a gene’s coding region. The movements of TEs are subject to prior activation; stress conditions can serve as stimuli for TE activation (Dubin et al., 2018). Increased TE transcription has been described for other organisms experiencing heat stress, such as A. thaliana(Huang et al., 2018). In the pathogenic ascomycete fungusMagnaporthe grisea, heat stress, copper sulfate and oxidative stress cause activation of retrotransposons (Ikeda et al., 2001). Therefore, the upregulation of TEs in our P. britannica specimens at 25 °C might result from thermal stress. The biological consequences of stress-related TE translocations (Negi et al., 2016) would be an interesting area of future studies.
In addition to upregulation of ascomycete stress response genes, downregulation of a large number of genes was observed at 25 °C. Functional annotation of these genes proved difficult, as many could only be annotated roughly (e.g., to enzyme classes). Genes that could be annotated more precisely were mostly part of regular metabolic pathways. GTPase and ATPase activity as well as NAD(P)-binding were major functions downregulated at 25 °C. The same is true for various genes responsible for translation and transcription and for some genes encoding mitochondrial proteins. The latter suggests a reduction of mitochondrial function; similar results have been described for stressedSaccharomyces cerevisiae cells (Sakaki et al., 2003). Curbed metabolism in stress situations might be beneficial to allocate the available energy resources to stress response pathways, allowing organisms to survive under suboptimal conditions (Peredo & Cardon, 2020).
The results illustrate that an organism exposed to heat stress does not solely react by means of expression of stress genes but also by downregulation of other genes such as those involved in metabolic pathways under normal conditions. In the lichenized fungus, heat stress responses such as upregulation of HSPs were already induced at 15 °C; at 25 °C, the fungus appeared to be highly stressed. Therefore, long-term exposure to high temperatures could result in severe damage, especially when the lichen’s respiration rate increases, causing a negative carbon balance over extended time periods (Sundberg et al., 1999; Lange & Green, 2005).
Interestingly, of the 200 ascomycete genes most significantly differentially expressed between temperatures, 103 were downregulated at 25 °C, and 27 of these downregulated genes were also differentially expressed between photomorphs. Of the 97 fungal genes upregulated at 25 °C, only seven were photomorph-mediated. Stress-related proteins are highly conserved (Elliott, 1998) and stress responses are vital for survival, which could explain why their upregulation at 25 °C occurs largely independent of associations with a specific partner.
4.3 | Cyanobacterial genes / temperature
The relatively low number of just below 2,500 cyanobacterial transcripts detected in our samples results from the method of library construction involving selection of poly-A mRNA. Cyanobacterial transcripts were found nonetheless due to carry-over, but as a result the number of cyanobacterial genes with significantly different expression was relatively small. Provided that this carry-over is a random process, cyanobacterial transcripts would be sampled depending on their frequency in the RNA pool; nevertheless, the results have to be interpreted with due caution.
Increased temperature led to significant differential expression of (heat) stress and photosynthesis-related cyanobacterial genes. The former category comprises HSPs, chaperonins and Dps (DNA-binding protein from starved cells), as well as genes indirectly involved in stress response mechanisms, e.g. lysine-tRNA ligase. Dps is a highly conserved protein which is part of various stress response pathways (Karas et al., 2015). Its two main functions in stress responses are DNA binding – i.e. shielding the DNA from reactive chemicals – and ferroxidase activity. Ferroxidase oxidizes Fe2+ to Fe3+, and thereby prevents the formation of hydroxyl radicals via the Fenton reaction (Fe2+ + H2O23+ + OH + OH) (Calhoun & Kwon, 2011; Nair & Finkel, 2004). The lysine-tRNA ligase is responsible for the formation of lysyl-tRNA which is of relevance in protein synthesis, transferring lysine into ribosomes (Wu et al., 2007). Besides, the enzyme has another function – as has been assessed forEscherichia coli – as it synthesizes various adenyl dinucleotides, particularly Ap4A. This function of lysine-tRNA ligase is active only under stress and leads to an accumulation of Ap4A. Ap4A serves as a modulator of heat shock response, binding to and modifying stress proteins (Onesti et al., 1995).
The second category of genes upregulated at higher temperatures in our samples are photosynthetic genes. These genes were annotated to various photosynthetic functions of photosystems I and II as well as the cytochrome complex and ATP synthase. Expectedly, higher temperatures led to an increase in photosynthetic activity (Lommen et al., 1971), but the expression of photosynthetic genes might as well be caused by heat-induced structural changes of the photosynthetic machinery, such as protein complexes (Allakhverdiev et al., 2012; Ivanov et al., 2017).
A cyanobacterial gene encoding a bleomycin resistance protein was highly significantly upregulated at 15 °C and 25 °C, but the reason for its upregulation is difficult to determine. This protein confers resistance to the antibiotic bleomycin (Dumas et al., 1994). In E. coli it has been shown that the presence of antibiotic resistance genes likely has been induced by adaptation to stress (like thermal stress) as tolerance towards antibiotics underlies similar mechanisms as tolerance towards heat (Cruz-Loya et al., 2019) This might potentially explain why a gene encoding a bleomycin resistance protein was upregulated at elevated temperatures in the Nostoc cyanobiont of P. britannica . Besides, Keszenman et al. (2000 & 2005) have demonstrated that the upregulation of bleomycin resistance genes is a side effect of heat stress in S. cerevisiae , as the yeast cells proved to be resistant to bleomycin treatment after having been exposed to heat stress. This correlation between heat stress and bleomycin resistance is probably the result of cross-linking of DNA repair mechanisms (Keszenman et al., 2000, 2005). A potential role of bleomycin-resistance genes in DNA repair has already been proposed in previous studies on E. coli (Blot et al., 1991).
A range of cyanobacterial genes of the P. britannicaphotosymbiodemes show temperature-mediated differential gene expression. Most DEGs are responsible for either heat stress responses or photosynthesis. In general, our results demonstrate that in cyanobacteria, heat stress is induced at a temperature of 15 °C already, similar to what we found in the lichen-forming fungus. This coincides with personal observations that P. britannica cyanomorphs and compound thalli only grow in shady and moist habitats where environmental fluctuations are minimal. The preference of these sites might simply reflect the cyanobionts’ limited tolerance towards high temperatures and desiccation. The upregulation of genes of the photosynthetic apparatus at higher temperatures suggests an increase in the cyanobacteria’s photosynthetic activity. Hence, elevated temperatures are not merely a stressor but are beneficial for cyanobacteria to a certain extent. It seems, however, unlikely that the favorable conditions for photosynthesis counterbalance the detrimental effects caused by heat stress, otherwise the cyanomorphs and compound thalli would not be habitat-restricted to the coolest, moistest sites available.
4.4 | Green algal genes / temperature
Similar to the cyanobacterial genes, most of the green algal genes upregulated at higher temperatures were either part of photosynthetic or stress response pathways. HSPs, chaperonins as well as proteins for DNA repair and signal transduction were expressed primarily at 25 °C. This suggests that the green algal partner tolerates heat to a greater extent than the cyanobacterial and fungal partners. This result is consistent with findings of Green et al. (2002) as well as personal observations that P. britannica tripartite morphs grow in relatively open habitats exposed to more fluctuations in temperature, light and water availability. Furthermore, photosynthetic activity of the algae was enhanced at 15 °C and 25 °C, so tripartite morphs may partly benefit from a rise in temperature. However, the actual optimal photosynthetic temperature is difficult to determine, as it is species-dependent (Wagner et al., 2014) as well as dependent on various other environmental factors (Alam et al., 2015; Green et al., 2002). Given the upregulation of photosynthetic genes at 15 and 25 °C, one can conclude that both temperatures are within the range of optimal temperature for net photosynthesis in P. britannica photosymbiodemes. However, as 25 °C leads to an expression of genes relevant for heat shock responses, a long-term exposure to higher temperatures could – at least partly – inactivate the photosynthetic apparatus (Ivanov et al., 2017). A prolonged increase in temperature could also negatively impact carbon balance if the lichen’s respiration rate outweighs its photobiont’s photosynthesis rate. Elevated respiration after temperature increases has been described for lichens and their photobionts (Palmqvist et al., 2008; Sundberg et al., 1999), however, the respiration rate usually normalized after one to three hours (Sundberg et al., 1999). There is no evidence in our dataset that a rise in temperature led to an elevated algal respiration rate; based on this the carbon balance in the lichen is most likely still positive. Gas-exchange measurements would be useful to settle this issue, but were beyond the scope of the current study.
We also observed upregulation of proteins associated with lipid metabolism at 25 °C. This metabolic activity could result from lipid remodeling induced by heat stress, especially in regard to membrane lipids. Heat can compromise the structural integrity of membranes and consequently an organism’s vitality. In order to counteract membrane disintegration, a variety of lipids are synthesized and accumulated in the cell (Zhang et al., 2020), such as saturated fatty acids (Barati et al., 2019), whereas other lipids, mostly polyunsaturated fatty acids, undergo selective degradation or are converted to storage lipids (Légeret et al., 2016). These metabolic conversions of lipids seem to allow the algae to cope with an increase in temperature (Zhang et al., 2020; Song et al., 2018). Therefore, the expression of lipid metabolism proteins at 25 °C might be an indirect response to stress induced by elevated temperatures.
These results help us in understanding the ecological conditions under which lichen symbioses grow in nature. Compound thalli of photosymbiodemes represent an attractive study system as both photomorphs grow under the same environmental conditions, so both photobionts do not only have to tolerate these conditions but must also benefit from them to establish a successful symbiosis. Therefore, photosymbiodemes are restricted in their distribution to certain ecological niches (Green et al., 1993; Lange et al., 1988; Purvis, 2000), a circumstance we were also able to observe in our study, as theP. britannica specimens only grew in damp, relatively hidden spots such as small crevices of lava rocks or under branches of birch shrubs growing on a small slope. Solitary tripartite morphs, on the other hand, grew in open areas that are more exposed to environmental fluctuations, especially regarding light, temperature and humidity. Solitary cyanobacterial thalli were missing in the collection area at Heiðmörk. These distribution patterns of P. britannica seem to be indicative of the respective photobiont’s stress tolerance. More studies of gene expression in lichen photomorphs are needed to understand to which extent the patterns reported here hold true for other taxa as well.