Introduction
Lichens form when specific species of fungi contact a compatible photobiont, either a cyanobacterium or eukaryotic green alga (Green & Smith Reference Green and Smith1974; Almendras et al. Reference Almendras, Garcia, Caru and Orlando2018). The fungus encapsulates the algal cells and an unidentified inter-kingdom communication alters the algal cell morphology (Joneson & Lutzoni Reference Joneson and Lutzoni2009). Lichen-forming fungi are known for their ability to produce an arsenal of compounds to protect the symbiosis from environmental stress, competition, pathogens and predation (Giez et al. Reference Giez, Lange and Proksch1994; Solhaug et al. Reference Solhaug, Gauslaa, Nybakken and Bilger2003; Halama & Van Haluwin Reference Halama and Van Haluwin2004). In return the algal cells fix carbon and, for cyanobacterial photobionts, nitrogen, which are then available to the fungus. Carbohydrates from the algal cells, such as hexoses and acyclic polyalcohols, are transferred to the fungus and assimilated (Richardson et al. Reference Richardson, Hill and Smith1968; Yoshino et al. Reference Yoshino, Yamamoto, Masumoto, Degawa, Yoshiwaka, Harada and Sakamoto2020). A successful interaction between two compatible symbionts leads to changes in cell morphology and the formation of a stratified thallus structure (Athukorala et al. Reference Athukorala, Huebner and Piercey-Normore2014). The growth of these thalli produces biomass in harsh or barren environments, allowing other species to enter that niche (Kumpula Reference Kumpula2001; Asplund & Wardle Reference Asplund and Wardle2017). Despite the abundance of lichen species, their importance in the environment, and their potential for some industries, there is still much to learn about their fundamental biology. This is largely due to the difficulties in obtaining axenic cultures of both symbionts, and the notoriously slow growth rate of lichen mycobionts.
Fungi are isolated using a variety of common media such as potato-dextrose (PD), yeast-sucrose, malt-glucose and Czapek-Dox agars (Crous et al. Reference Crous, Verkley, Groenewald and Samson2010). These media are often supplemented with nutrients that mimic the microbe's natural environment to encourage growth more reminiscent of that found in nature. For example, Campbell's™ V8 tomato juice is used to initiate sporulation in the wheat pathogen Parastagonospora nodorum (Abeysekara et al. Reference Abeysekara, Friesen, Keller and Faris2009). However, relying on these media can lead to difficulties in cultivating uncharacterized fungi, or observing their fundamental biology. For example, mycotoxin production by Fusarium species is strictly controlled by specific nitrogen sources in the growth medium (Miller & Greenhalgh Reference Miller and Greenhalgh1985; Maeda et al. Reference Maeda, Nakajima, Tanahashi, Kitou, Miwa, Kanamaru, Kobayashi, Nishiuchi and Kimura2017) and the wheat pathogen Zymoseptoria tritici grows in a filamentous form in planta but is yeast-like in vitro (Francisco et al. Reference Francisco, Ma, Zwyssig, McDonald and Palma-Guerrero2019).
Lichen mycobionts are often cultured using complex media such as potato-dextrose, Sabouraud-glucose, malt-yeast, or defined media such as Lilly and Barnett's medium or Czapek solution (Bando et al. Reference Bando, Sugimoto and Sugino1997; Stocker-Wörgötter Reference Stocker-Wörgötter2001; Stocker-Wörgötter et al. Reference Stocker-Wörgötter, Elix and Grube2004; Brunauer et al. Reference Brunauer, Hager, Grube, Türk and Stocker-Wörgötter2007). These media are often supplemented with extracts from the lichen's natural substratum, such as bark (Stocker-Wörgötter et al. Reference Stocker-Wörgötter, Elix and Grube2004) or soil (Ahmadjian & Heikkilä Reference Ahmadjian and Heikkilä1970), or with polyalcohols (Brunauer et al. Reference Brunauer, Hager, Grube, Türk and Stocker-Wörgötter2007). For example, Brunauer et al. (Reference Brunauer, Hager, Grube, Türk and Stocker-Wörgötter2007) grew Xanthoria elegans on Lilly and Barnett's medium supplemented with either ribitol or mannitol to examine differences in secondary metabolite production. Ribitol is produced by the algal cells associated with Xanthoria, while mannitol is the storage carbohydrate of the fungus. Interestingly, it was mannitol that induced the greatest production of secondary metabolites, rather than the algal carbohydrate. Co-culturing the mycobiont with its associated alga has proved useful. Wang et al. (Reference Wang, Liu, Zhang, Zhou, Zhang, Li, Yu, Zhang, Hao and Wang2014) co-cultured the soil-crust lichen used in this study, Endocarpon pusillum, with its photobiont in a carbon-free medium to determine carbohydrate flux between symbionts, as well as changes in gene expression during the initial stages of symbiosis. Alternatively, cultivating thallus fragments has also proved useful for growth experiments. For example, Bando et al. (Reference Bando, Sugimoto and Sugino1997) grew thallus fragments of Parmotrema tinctorum in a range of media supplemented with various amino acids, vitamins and plant hormones to discover factors governing thallus growth. Interestingly, with the exception of asparagine and aspartate, all supplements tested reduced the growth of the thallus and, counterintuitively, diluting the media improved growth. Unfortunately, few amino acids and no polyalcohols were included in this study.
To discover media supplements that might aid isolating and culturing lichen mycobionts, we performed a six-month growth assay on a strain of the model soil-crust mycobiont, Endocarpon pusillum. This species has been the focus of resynthesis experiments in modern science for several decades (Ahmadjian & Heikkilä Reference Ahmadjian and Heikkilä1970). However, physiological observations and attempts at in vitro culture were published as early as 1878 (Zukal Reference Zukal1878). Its scientific history, cosmopolitan distribution and emerging genetic resources promote E. pusillum as an ideal model for studying lichen growth and resynthesis of lichens (Park et al. Reference Park, Choi, Lee, Park, Jung, Oh, Lee and Hur2014; Wang et al. Reference Wang, Liu, Zhang, Zhou, Zhang, Li, Yu, Zhang, Hao and Wang2014, Reference Wang, Zhang, Zhou, Zhang and Wei2015; Mead & Gueidan Reference Mead and Gueidan2020). Despite the advantages of this species, it is still slow growing compared with other model fungi such as Aspergillus. Accordingly, we undertook a growth assay to determine whether metabolites involved in plant microbe interactions might improve the growth of the mycobionts of an Australian strain of E. pusillum. The genome sequence of this Australian strain (EPUS1.4) has been published in a separate study (Mead & Gueidan Reference Mead and Gueidan2020). The slow growth rate of the fungus made measuring biomass difficult. Therefore, we approximated relative growth by photographing flat black biomass against a white background and counting the proportion of black pixels. Ametrano et al. (Reference Ametrano, Selbmann and Muggia2017) similarly measured the growth of rock-inhabiting fungi in co-culture with lichen photobionts, confirming that the method offers a more efficient alternative to weighing the biomass of slow-growing fungi. This assay limited the nitrogen or carbon sources available to the fungus to single compounds to determine metabolically preferred nutrients. These compounds were chosen on the basis of being commonly found in soils and plant apoplasts, and secreted by lichen photobionts (Richardson et al. Reference Richardson, Hill and Smith1968; Solomon & Oliver Reference Solomon and Oliver2002; Mooshammer et al. Reference Mooshammer, Wanek, Hämmerle, Fuchslueger, Hofhansl, Knoltsch, Schnecker, Takriti, Watzka and Wild2014). Previous studies predominantly report growing E. pusillum on potato-dextrose, either on agar or in shaking liquid cultures (Wang et al. Reference Wang, Liu, Zhang, Zhou, Zhang, Li, Yu, Zhang, Hao and Wang2014; Li & Wei Reference Li and Wei2016; Yang et al. Reference Yang, Bae, Nam, Jeong, Zhou, Park, Taş, Hwang, Park and Chung2018). We assessed a range of carbon sources, ranging from two carbon molecules and essential primary metabolites such as acetate and citrate to more complex molecules such as polyalcohols, cellobiose and amino acids. We investigated the efficiency of inorganic and organic nitrogen compounds but focused on components of the γ-amino butyric acid (GABA) shunt and ornithine polyamine pathway. These compounds have been shown to play pivotal roles in growth, development and stress response in both plants and fungi (Chen et al. Reference Chen, Shao, Yin, Younis and Zheng2019; Rocha & Wilson Reference Rocha and Wilson2019; Singh et al. Reference Singh, Mukherjee and Rajam2020). Furthermore, these compounds have been reported to play varying roles in plant-fungal interactions. For example, the non-proteinaceous amino acid, GABA, was shown to be a highly abundant nutrient source in the apoplast of tomato leaves infected with the biotrophic fungus Cladosporium fulvum (Solomon & Oliver Reference Solomon and Oliver2002).
The formulation of growth media is often crucial to successfully isolating and culturing fungi from the environment. To improve the success of isolating and growing lichen-forming fungi supplements are often added to traditional media. In this experiment we observed the growth response of the lichen mycobiont E. pusillum to various carbon and nitrogen sources with the aim of providing guidance for customizing growth media. This may facilitate the culture of lichen mycobionts and a further understanding of lichen biology.
Materials and Methods
Fungal strain and growth conditions
The specimen of Endocarpon pusillum (C. Gueidan 2364) used for isolation was collected from the CSIRO Black Mountain site in Canberra (35°16ʹS, 149°07ʹE), Australia in 2016. It was deposited in the CANB herbarium as accession CANB 913709. Ascospores were shot on Potato Dextrose Agar (PDA) plates and single ascospores were isolated and grown on liquid PD medium in an incubator with 20/18 °C 12-h day/night cycles. Stock cultures of the strain EPUS1.4 were grown at 22 °C in complete darkness in stationary liquid YSSG medium (yeast extract 5 g l−1, sucrose 10 g l−1 sorbitol 10 g l−1, GABA 1 g l−1).
Growth assay cultures of E. pusillum were grown in a similar way to Li & Wei (Reference Li and Wei2016) but in darkness in sterile multi-well plates. The plates were sealed with parafilm, maintaining constant humidity and inhibiting contamination. Each plate held 12 wells and each having a total volume of 3 ml (Nuncleon Delta coated, cat. 150628, ThermoFisher Scientific). Wells were filled with 2 ml of liquid minimal base medium (Table 1) then supplemented with a stock solution of the carbon or nitrogen source being tested. The nitrogen and carbon source treatments were randomly distributed among wells and the position of plates within the incubator was rotated weekly. The media were poured, inoculated and the plates sealed with micropore surgical tape (3M) under sterile conditions. Uninoculated YSSG medium was included to check for environmental contamination. Base minimal medium contained 10 mM sodium phosphate buffer and trace elements, and was adjusted to pH 6.5 with HCl. The base minimal medium was autoclaved then inoculated from a stock suspension of E. pusillum mycelia. The fungal biomass was washed three times with sterile distilled water before inoculation. The biomass was then finely blended in the base medium with a sterile Omni-Inc tissue homogenizer to a final concentration of 0.02 mg of fresh weight biomass per ml. The base medium for testing nitrogen sources was supplemented to a final concentration of 25 mM sucrose and the desired nitrogen source adjusted to 10 mM nitrogen. The base medium for testing carbon sources was supplemented with 10 mM mono-ammonium phosphate and the desired carbon source adjusted to 150 mM carbon (equivalent to 25 mM glucose). Carbon and nitrogen were supplemented from 1.5 M and 100 mM stock solutions, respectively, each adjusted to pH 6.5 with 1 M HCl or 1 M NaOH prior to sterilization. The components of each medium, at their final concentration, are described in Table 1. Note that GABA, ornithine and glutamine are both carbon and nitrogen sources. Including controls lacking any metabolizable carbon or nitrogen, this totalled 23 conditions. Four cultures in each condition were grown, totalling 92 samples, for 150 days before being quantified.
List of minimal media components. Media were derived from the base medium by the addition of carbon and nitrogen sources. Media used to test carbon sources contained 10 mM mono-ammonium phosphate as a nitrogen source. Conversely, media used to test nitrogen sources contained 25 mM sucrose for carbon. The base medium and stock solutions were adjusted to pH 6.5 with 1 M HCl or 1 M NaOH.
Colony surface area approximation
The colony surface area was used as an approximation of growth to overcome the need to weigh restrictively small biomasses of each sample. Each well was photographed on a white background with six technical replicates using a Leica M205C dissecting microscope at ×16 magnification (Fig. 1A). The images were then manually cropped using the software package GNU Image Manipulation Program (GIMP) to contain only the contents of the well (Fig. 1B). GIMP then converted all pixels to black or white and counted all black pixels as a proportion of total pixels (GIMP 2019). To validate this method, images of exactly 25%, 50% and 75% black pixels were generated in Microsoft PowerPoint. These images underwent the same measurement procedure as the experimental well images and determined the measurement error rate of GIMP.
Unmodified (A) and cropped black and white (B) images of Endocarpon pusillum biomass after six months of growth in minimal medium containing mannitol as the sole carbon source. Scale bar = 5 mm. In colour online.
Growth measurements
The mean area covered in both the carbon and nitrogen test wells was normalised using the mean % area in the control wells containing no carbon and no nitrogen.
Biomass measurements
Media were gently removed from the wells so that the fungal biomass was not removed. The 12-well plates were dried at 65 °C for 4 days to remove all moisture. The plates were weighed before the biomass was removed. Biomass was detached from the plastic floor of the wells with a solution of 5 M HCl and 5% (v/v) hydrogen peroxide overnight at room temperature. The degraded fungal biomass was then gently washed away with hot water. The plates were subsequently wiped clean with paper towels, dried at 65 °C for 4 days and re-weighed. The difference in mass represents the total fungal biomass.
Statistical analyses
The black pixel count of each image was expressed as a proportion of total pixels for that image. The mean of the six technical replicates was taken for each well. The mean of the biological replicates for each of the control media (no carbon or no nitrogen) was determined, and each biological replicate was represented as a proportion of its respective control. A Shapiro-Wilks test using R (R Core Team 2017) determined all data followed a normal distribution, allowing a one-way ANOVA to determine variance between samples. Using R, a multiple comparisons test was performed using Tukey's Honest Significant Difference (HSD) test which determined which means were significantly different (P < 0.05). The R packages ggpubr (Kassambara Reference Kassambara2019), multcompView (Graves et al. Reference Graves, Piepho and Selzer2019), dplyr (Kassambara Reference Kassambara2020), rstatix (Wickham et al. Reference Wickham, François, Henry and Müller2020) and ggplot2 (Wickham Reference Wickham2016) were used to analyze and visualize the data.
Results
Technical error associated with measuring surface area covered by hyphal growth
There was < 1.5% deviation between the standardized images and the measured pixel count using GIMP (2019) (Fig. 2). This analysis did not describe the error associated with measuring biomass but indicated the technical error associated with pixel counts.
Assessment of the error associated with estimating fungal growth. Standard images containing exactly 25, 50 and 75% black pixels were generated and the number of black pixels were measured using GIMP (2019). The results are expressed as the relative proportion of measured pixels compared to the predicted number. Black dots represent the mean, black lines within boxes represent the median, boxes represent the inner quartiles, and whiskers represent the minimum and maximum values.
Growth of Endocarpon pusillum on various carbon sources
Endocarpon pusillum was grown on minimal media containing tri-ammonium phosphate as the sole nitrogen source and a selected sole carbon source (Fig. 3). The total biomass of the experiment, including all treatments and controls, reached 248.5 mg over six months and there was insufficient material to weigh the dry mass of individual samples. The fungus grew flat in minimal multicell layers against the bottom of the wells of 12-well plates in liquid media with similar morphologies between replicates and between treatments, and low variability between replicates. The no carbon and no nitrogen controls produced some growth, equivalent to 27.0% (95% CI = 0.63) and 56.0% (95% CI = 3.42) respectively. These values were used to normalize the other test samples. The hexoses and hexose polymer, fructose, glucose, and cellobiose, produced the most growth of the sugar carbon sources and were comparable to the growth observed using the polyalcohol sugar, sorbitol. Interestingly, growth on sucrose was lower than on its monomer constituents, glucose and fructose. However, the growth on glucose was not significantly different from growth on sucrose. Similarly, growth on mannitol was lower than on its isomer, sorbitol, but not significantly (t-test: P = 0.08) (Fig. 3). Acetate, glycerol and citrate, as well as the amino acids, elicited poor growth whereby citrate and glutamine significantly reduced growth below that of the no carbon control.
Relative growth of Endocarpon pusillum on various carbon sources. Growth is relative to the control lacking any metabolizable carbon. Carbon source was adjusted to 25 mM glucose equivalent. Bars with the same letter are not significantly different based on Tukey's HSD test (P < 0.05). Black dots represent the mean, black lines within boxes represent the median, boxes represent the inner quartiles, and whiskers represent the minimum and maximum values. Grey dots represent outliers.
Growth of Endocarpon pusillum on various nitrogen sources
Endocarpon pusillum was grown on minimal media containing sucrose as the sole carbon source and a selected sole nitrogen source. The non-proteinaceous amino acid, GABA, significantly elicited the most growth of all tested nitrogen sources (Fig. 4). Interestingly, glutamine, a precursor to GABA synthesis, elicited significantly less growth than GABA or the no nitrogen control. Ornithine produced no significant difference to the no nitrogen control, whereas nitrate and ammonia produced significantly less growth.
Relative growth of Endocarpon pusillum on various nitrogen sources. Growth is relative to the control lacking any metabolizable nitrogen. Nitrogen source was adjusted to 10 mM nitrate equivalent. Bars with the same letter are not significantly different based on Tukey's HSD test (P < 0.05). Black dots represent the mean, black lines within boxes represent the median, boxes represent the inner quartiles, and whiskers represent the minimum and maximum values. Grey dots represent outliers.
Discussion and Conclusion
One of the major hindrances to studying lichens is the difficulty associated with isolating and culturing the mycobiont. In this study we tested several carbon and nitrogen sources that might improve the growth of the model lichenized fungal species Endocarpon pusillum. We found that the carbon sources glucose, fructose, cellobiose and sorbitol consistently led to significantly more growth than the no carbon control. This is not surprising since glucose is a key component of primary metabolism, whereas fructose, cellobiose and polyalcohols are plant-derived carbohydrates available to the fungus during its symbiotic association with a photobiont (Richardson et al. Reference Richardson, Hill and Smith1968). Cellobiose is a disaccharide of glucose monomers, which may explain the similar growth observed on these two carbon sources. The fungus may be assimilating cellobiose, when associated with its photobiont, by degrading the algal cell wall. This damage to the photobiont may represent a more parasitic interaction. The continuum between mutualism and parasitism in the lichen symbiosis is not a new concept (Richardson Reference Richardson1999; Hyvärinen et al. Reference Hyvärinen, Härdling and Tuomi2002; Ahmadjian & Jacobs Reference Ahmadjian, Jacobs and Goff2011) and considering the multiple phylogenetic origins of lichenization (Gargas et al. Reference Gargas, DePriest, Grube and Tehler1995; Gueidan et al. Reference Gueidan, Villaseñor, de Hoog, Gorbushina, Untereiner and Lutzoni2008; Schoch et al. Reference Schoch, Sung, López-Giráldez, Townsend, Miadlikowska, Hofstetter, Robbertse, Matheny, Kauff and Wang2009) it is not unexpected that different lichen species may fall within this continuum. Alternatively, the propensity of E. pusillum towards cellobiose might be explained by the fungus having evolved to opportunistically recycle plant biomass in the soil or dead algal cells within the thallus.
Mannitol, sorbitol and fructose elicited similarly high levels of growth, with fructose providing the highest but most variable growth. Growth on these substrates correlates with known fungal metabolism whereby sorbitol can isomerize to mannitol which can then be oxidized to fructose for subsequent glycolysis (Solomon et al. Reference Solomon, Waters and Oliver2007). Sucrose elicited less growth than its monomer constituents which might indicate that this species does not harbour an active invertase and cannot efficiently convert sucrose into monosaccharides. Future genomics and metabolic network modelling may provide some insight into this species’ ability to assimilate disaccharides. Interestingly, the fungus can grow slowly on the common complex growth medium, potato-dextrose, but not when it is supplemented with Campbell's™ V8 tomato juice (results not shown), as is often required by plant pathogens (Abeysekara et al. Reference Abeysekara, Friesen, Keller and Faris2009). The other polyalcohols, glycerol, erythritol and ribitol, did not elicit the same level of growth as sorbitol, indicating that the mycobiont metabolism is selective. This correlates with other reports. Richardson et al. (Reference Richardson, Hill and Smith1968) investigated the carbohydrates produced by photobionts of different lichen species and showed that there is a primary polyalcohol secreted in each case. Wang et al. (Reference Wang, Liu, Zhang, Zhou, Zhang, Li, Yu, Zhang, Hao and Wang2014) found that growing E. pusillum in co-culture with its photobiont, Diplosphaera chodatii, led to the production of sorbitol (12.6 mg/100 ml), glucose (0.8 mg/100 ml) and sucrose (0.8 mg/100 ml) in the medium, but not fructose or mannitol. There were no carbohydrates found in media containing the symbionts in isolation, indicating that E. pusillum induced the release of predominantly sorbitol from the algal cells. Subsequent gene expression analysis of the fungus, when grown on several different carbohydrates, showed that sorbitol, mannitol, glucose and sucrose induced consistently high sugar transporter transcription, while other sugars did not (Wang et al. Reference Wang, Liu, Zhang, Zhou, Zhang, Li, Yu, Zhang, Hao and Wang2014). These reports show that identifying the dominant carbohydrate secreted by species of lichen photobionts may provide direction for which carbon source will improve the growth of desired mycobionts.
Amino acids were supplied as sole carbon or sole nitrogen sources. To test the ability of E. pusillum to grow on an amino acid as a nitrogen source, no other nitrogen sources were supplied in the minimal medium, and glucose was supplied as a carbon source. Likewise, to test the ability of the mycobiont to utilize specific amino acids as carbon sources, all other carbon was omitted from the medium and nitrogen was supplied as ammonium. All amino acids tested were poor carbon sources, despite pH buffering and equimolar carbon concentrations. This might be due to the carbon skeletons themselves proving difficult to metabolize but is more likely due to an imbalance of ammonium and carbon leading to ammonium toxicity. This hypothesis is reinforced by the detrimental effect of ammonia on the growth of E. pusillum. Exogenous nitrogen, including ammonia, has been reported to play a key role in lichen growth and development. The focus of these reports varied from the ability of nitrogen sources to induce metabolite production (Kinoshita et al. Reference Kinoshita, Yamamoto, Kurokawa and Yoshimura2001), uptake (Dahlman et al. Reference Dahlman, Persson, Palmqvist and Näsholm2004; Hauck Reference Hauck2010), physiological effect (Gaio-Oliveira et al. Reference Gaio-Oliveira, Dahlman, Palmqvist, Martins-Loução and Máguas2005; Palmqvist & Dahlman Reference Palmqvist and Dahlman2006; Pirintsos et al. Reference Pirintsos, Munzi, Loppi and Kotzabasis2009) and in relation to nitrogenous air pollution (Pescott et al. Reference Pescott, Simkin, August, Randle, Dore and Botham2015). The response to nitrogen source in these studies was predominantly species dependent; however, ammonia was generally more readily assimilated than nitrate. Most studies used lichens in the symbiotic thallus state, presumably due to the difficulties in cultivating the lichen mycobiont and to gain a more ecological understanding of the symbiosis. Unfortunately, including the photobiont precludes conclusions about the mycobiont's ability to use various nitrogen sources. However, the effect of these compounds on lichen growth provides direction aimed at improving the growth of mycobionts in the future.
Only the amino acid GABA elicited more growth than the no nitrogen control when amino acids were supplied as the sole nitrogen sources. Glutamine and ornithine failed to elicit significantly more growth than the inorganic nitrogen sources, nitrate and ammonium. However, GABA produced 144% of the growth of the no nitrogen control, 194% and 251% of the growth elicited by nitrate and ammonium, respectively. GABA is a primary metabolite found in all eukaryotes. It is an important neurotransmitter in the vertebrate brain (Ge et al. Reference Ge, Goh, Sailor, Kitabatake, Ming and Song2006) and a key stress regulator in plants (Ramesh et al. Reference Ramesh, Tyerman, Xu, Bose, Kaur, Conn, Domingos, Ullah, Wege and Shabala2015), yet the role of this small molecule is relatively unknown in fungi (Mead et al. Reference Mead, Thynne, Winterberg and Solomon2013; Bönnighausen et al. Reference Bönnighausen, Gebhard, Kröger, Hadeler, Tumforde, Lieberei, Bergemann, Schäfer and Bormann2015). GABA is produced from the tricarboxylic acid (TCA) cycle as a bypass from α-ketoglutarate to succinate, negating a key ATP producing step. The production of this molecule links the TCA cycle to several other key primary metabolic pathways, such as glutamine-ammonia dissimilation, alanine metabolism and the ornithine polyamine pathway. GABA and this polyamine pathway have been shown to play roles in fungal growth, development and pathogenicity (Bailey et al. Reference Bailey, Mueller and Bowyer2000; Mead et al. Reference Mead, Thynne, Winterberg and Solomon2013). Plant apoplastic GABA concentrations have been found to increase during pathogen infection leading to the hypothesis that fungal plant pathogens might induce GABA leakage to assimilate this nutrient (Solomon & Oliver Reference Solomon and Oliver2002). Conversely, 100 μM GABA was able to trigger sporulation in Parastagonospora nodorum under normally prohibitive conditions (Mead et al. Reference Mead, Thynne, Winterberg and Solomon2013). The GABA concentration was too low to be considered a nutrient source, suggesting the compound was being used as a signalling mechanism. In the lichen symbiosis, GABA may be acting as both a nutrient source and an indicator molecule, supplying a readily available nitrogen source to the fungus and signalling that a compatible alga has been found.
Mycelia grew in both the negative controls lacking carbon or nitrogen. This apparent growth does not represent an increase in biomass but the recycling of nutrients and metabolite reserves. Hyphae in a colony can fuse to share cytoplasm, known as a syncytium. Syncytia allow extending or exploratory hyphae to progressively move their cellular contents from one hypha to the next in search of nutrients (Plamann Reference Plamann2009; Roper et al. Reference Roper, Lee, Hickey and Gladfelterd2015). This creates a colony but is not indicative of an increase in biomass; instead, mass will be lost to respiration. Therefore, we normalized absolute measurements to their respective negative controls to account for this effect.
Citrate, nitrate, ammonia and glutamine showed decreased growth compared to their no carbon and no nitrogen negative controls. This is an artefact of fungal growth in low nutrient conditions and highlights the limitations of using a visual approximation for growth. The compounds that produced less growth than the negative controls could have had a toxic effect, such as ammonium toxicity, that inhibited growth. Alternatively, they could be poor nutrient sources that elicited minimal growth but did not trigger exploratory hyphae. Likewise, samples with equal growth to the negative controls, such as ornithine, are ambiguous. We cannot determine whether that particular nutrient, coincidentally, produced the same apparent growth as the negative control, or whether that nutrient is unavailable to the fungus and the apparent growth is due to exploratory hyphae. Despite these limitations, and the inherent phylogenetic variations in the mode of trophic assimilation, these experiments identified several metabolites most suited to growing E. pusillum and might be applicable to the culture of other slow-growing lichen-forming fungi.
Acknowledgements
The authors would like to thank the editorial staff and peer reviewers of The Lichenologist for their time and consideration. This study was funded by the CSIRO Synthetic Biology Future Science Platform.
Author ORCIDs
Oliver L. Mead, 0000-0001-7919-9515; Cécile Gueidan, 0000-0001-9295-9996.
