Introduction

The yeast genus Mrakia comprises sexual and asexual yeast species. They share a remarkable adaptation to low temperatures (Buzzini et al. 2018), the ability to ferment sugars (Fell 2011) and to produce diverse extracellular enzymes such as lipases, amylases, proteases, pectinases, cellulases, chitinase and ligninolytic enzymes (Tasselli et al. 2017). These abilities were applied to wastewater treatment, bioethanol and beer production at low temperature (Tsuji et al. 2013a; Tsuji et al. 2013b; De Francesco et al. 2018).

The majority of reported isolations of yeasts of the genus Mrakia were made from cold habitats such as polar soils (Arctic, Antarctic), glaciers (ice, sediment, cryoconite, meltwater), caves (cave ice), permafrost and snow (Buzzini et al. 2017; Sannino et al. 2017). A more exotic habitat of these yeasts includes frozen foods (fish and meet) and arctic goose feathers (Fell 2011; Kabisch et al. 2016; Singh et al. 2016). Reports of isolation of these psychrophilic yeasts outside cold habitats are scarce and limited to a few studies in European boreal and temperate forests, from either plant material or soils (e.g. Glushakova and Chernov 2004; Maksimova and Chernov 2004; Yurkov et al. 2016). However, it must be mentioned that isolation of Mrakia yeasts in association with plants and lichens in polar climates was repeatedly published (Babjeva and Reshetova 1998; Fell 2011; Ferreira et al. 2019; Santiago et al. 2015; Zhang et al. 2015).

Tree (slime) fluxes are sugar-rich substrates occurring on sites where phloem or xylem sap is leaking through damages of tree bark. High sugar (glucose, fructose and sucrose) concentration makes tree sap a suitable substrate for propagation of microorganisms, including yeasts (reviewed in Péter et al. 2017). Slime fluxes (also referred to as wetwood by phytopathologists) are caused by an infection (often bacterial) which hinders injury (mechanical or by insects) healing. These fluxes are not temporarily restricted and may last for years. Conversely, spring sap flows in some temperate tree species like birch are restricted to a short period of time in spring when roots create substantial pressure that pushes tree sap upstream (Essiamah 1980). Yeast communities found in different types of tree exudates differ accordingly (reviewed in Péter et al. 2017). Numerous basidiomycetous yeasts were isolated in the past from injured trees (often birches, Betula spp.) (reviewed in Péter et al. 2017). The accuracy of techniques employed for yeast identification increased greatly allowing more reliable identification of phenotypically indistinguishable (cryptic) species. Tree fluxes caused by tree injuries (pruning or selected tree removal) during winter time were collected in spring of years 2013, 2016 and 2019 in Braunschweig (Lower Saxony, Germany). Several psychrophilic yeasts were isolated from xylem sap and thick microbial matts on stems and branches. Three species of the genus Mrakia were identified in this study. One of them represented a potential novel species highly similar to two isolates from superficial glacial melting water of Calderone Glacier, Italy (Branda et al. 2010) originally identified as Mrakia aquatica. A detailed examination of the five isolates (three from Germany and two from Italy) proved the strains to be a novel yeast species, which is described in this manuscript as Mrakia fibulata sp. nov.

Materials and methods

Tree fluxes caused by tree injuries during winter time were collected in spring of years 2013 and 2016 in the city Braunschweig, Lower Saxony, Germany (Fig. S1). In March 2013, two locations in Stöckheim were sampled: a stump from a recently cut down silver birch Betula pendula (52.209635 N, 10.527774 E) and from a young beech Fagus sylvatica pruned in the winter (52.213777 N, 10.531985 E). In April 2016, stumps from recently cut down silver birch trees B. pendula in Gartenstadt (52.238836 N, 10.507971 E) were sampled. In April 2019, young birch, hornbeam Carpinus betulus and dogwood (Cornus sp.) plants, which were pruned during winter, were sampled in Stöckheim in the same locality as in 2013. Tree fluxes and thick mucous microbial matts (Fig. S1) were aseptically collected in 50 mL tubes and kept at 4 °C until processing.

Yeasts were isolated by streaking out samples on surface of YM plates supplemented with chloramphenicol or by serial dilution (1:10, 1:100, 1:1000) plating on the same medium. Plates were incubated 2 weeks at 10 °C and were inspected every 2–3 days. Colonies were differentiated into macro-morphological types with dissection microscopy, counted and 1–2 representatives of each colony type per plate were transferred into pure culture.

Calderone Glacier (42.470278 N, 13.565556 E) can be considered a debris-covered glacier. The lower part of the glacier is completely covered by white limestone rock debris, whereas the top is characterized by the presence of seasonal snow. Above the glacier two of the highest peaks are present with pioneer vegetation. Details on sampling and isolation of yeasts from superficial glacial melting water of Calderone Glacier are provided by Branda et al. (2010).

Growth tests were performed in liquid media (according to Kurtzman et al. 2011) and in BIOLOG MicroStation and YT and FF MicroPlates (BIOLOG Inc., Hayward, CA, USA) at 20 °C. To inoculate BIOLOG MicroPlates, yeasts were pre-incubated on potato dextrose agar (PDA, Difco) at 20 °C, harvested from PDA, suspended in the inoculation solution IF-B (BIOLOG Inc., Hayward, CA, USA) and the turbidity was adjusted to the transmittance value provided by the manufacturer. MicroPlates were sealed to prevent desiccation and incubated at 22 °C and measured after 1, 2 and 3 weeks, with the optical density recorded at 490, 590 and 750 nm. The ability to utilise particular substrates by individual strains was recorded as positive, weak and negative.

DNA was isolated with the MasterPure Yeast DNA Purification Kit (Epicenter, Madison, WI, USA), following the manufacturer’s instructions. Amplification of the ITS and partial LSU regions were performed with primers ITS1F and LR5, while internal primers ITS4 and NL4 were used for sequencing (Spirin et al. 2018). The assembly and editing of sequence data were performed with Sequencher 5.3 (Gene Codes, Ann Arbor, MI, USA). Sequences were compared with sequences deposited in the NCBI GenBank (full dataset and RefSeq) and MycoBank.

Phylogenetic relationships of yeasts of the genus Mrakia were analysed with ITS and LSU nucleotide sequences. Sequence alignments were performed with the online version of the MAFFT algorithm using the Q-INS-i option with the default parameters (Katoh et al. 2019). Sequences were trimmed to obtain strict flanking regions. Phylogenetic relationships were inferred by the maximum likelihood (ML) method based on the general time reversible (GTR) model with RAxML (version 8.1.22) using raxmlGUI 1.5b and the GTRGAMMA option with 1000 rounds of bootstrap replicates (Silvestro and Michalak 2012).

Pair-wise distances were analysed in a 1181 bp long ITS-LSU alignment among type strains of Mrakia aquatica, Mrakia hoshinonis, and Mrakia niccombsii, and strains identified as M. aquatica in other studies. Distances were estimated after trimming ITS and LSU sequences in multiple sequence alignments to obtain strict flanking regions. A parsimony network was constructed from aligned ITS and LSU (D1/D2 domains) sequences with the program TCS 1.21 (Clement et al. 2000) using 95% connection limit and gaps treated as fifth state.

Results

Tree fluxes caused by tree injuries sampled during 2013–2019 in Braunschweig (Lower Saxony, Germany) yielded 19 yeasts. In 2013, a stump of a silver birch tree (Betula pendula) yielded one yeast species, Cystofilobasidium infirmominiatum (Table 1). Fluxes dropping from branches of young beech (Fagus sylvatica) trees which were trimmed during winter, yielded ten more species (Table 1). Sampled in 2016 stumps of recently cut down silver birch trees Betula pendula yielded only basidiomycetous yeasts, including three Mrakia species (Table 1). More recent sampling (April 2019) of birch, hornbeam and dogwood trees resulted in isolation of ten yeasts, five of which were not obtained from previous samplings (Table 1). Cystofilobasidium capitatum, Cystofilobasidium macerans, Cystofilobasidium infirmominiatum, and Tausonia pullulans were isolated in all three sampling years. Only one ascomycetous yeast, Candida sake was isolated from the sampled tree fluxes. Out of 19 isolated yeasts, 13 species were only isolated in one sampling year.

Table 1 Yeast species isolated from spring fluxes of sampled trees during this study
Full size table

Identification of most of yeasts isolated from tree fluxes was successful using either nucleotide sequences of LSU or ITS fragments. A few strains showed one to three nucleotide substitutions to type strains of respective species (Fig. 1). Nucleotide sequences of the ITS region of strains DSM 27892 (GenBank MK693026) and DSM 103936 (GenBank MK693025) preliminary identified as M. gelida differed in five nucleotide positions from that of the type strain (GenBank AF144485). Sequences of strains recovered from tree fluxes matched with strains of M. gelida previously isolated from Calderone (Apennines) and Miage (Alps) glaciers, respectively (GenBank GQ911543, KC455909). A detailed investigation of strains identified as M. gelida will be made in a separate study. Four strains of Mrakia sp. from tree fluxes showed almost equal sequence similarity values to strains comprising the M. aquatica sub-clade, namely M. aquatica, M. hoshinonis, and M. niccombsii. Three of the four strains DSM 103931 (GenBank MK372216), DSM 109846 (GenBank MN597447), and DSM 109847 (MN597448) were studied further. Strain BS-7 isolated from birch in 2013 was identified but not preserved and not further studied. All four strains shared identical ITS and LSU sequences. The best match was obtained with strain DBVPG 4990, one gap in ITS and identical LSU sequences (GenBank GQ911547 and GQ911522, respectively). Strain DBVPG 4990 was isolated from superficial meltwater collected at Calderone glacier and preliminary identified as Mrakia aquatica by Branda et al. (2010). Among Mrakia aquatica strains preserved in the DBVPG collection, strain DBVPG 4999, which shared ITS and LSU sequences identical with DBVPG 4990 (data not shown), was included in the analysis. Phylogenetic analyses of concatenated ITS and LSU alignments using a maximum likelihood and parsimony network analyses were performed to determine the placement of this potential new yeast (Figs. 1, 2).

Fig. 1
figure 1

Maximum likelihood analysis of concatenated sequences of the internal transcribed spacer (ITS) regions and D1/D2 domains of the LSU rRNA gene of the genus Mrakia. The numbers given on branches are frequencies (> 60%) with which a given branch appeared in 1000 bootstrap replications. The scale indicates the number of expected substitutions accumulated per site. The tree is rooted with Tausonia pullulans CBS 2532T (AF444417, NG_042352)

Full size image

Phylogenetic analysis showed that strains representing potential new species were placed in the Mrakia aquatica sub-clade comprised by M. aquatica, M. hoshinonis and M. niccombsii (Fig. 1). This clade received good (ML: 96%) statistical support in the analysis but internal nodes were not resolved. The most recent phylogenetic study of the genus Mrakia was performed by Tsuji et al. (2019), who described M. hoshinonis from a glacier in Canadian Arctic Archipelago. These authors considered only type strains and did not include other strains labelled in GenBank as M. aquatica or unidentified Mrakia sp. isolates. Three strains designated as M. aquatica (H1, H2, and DBVPG 4994) clustered close to M. hoshinonis (Fig. 1). To resolve phylogenetic relationships in this group, we analysed pair-wise distances in a trimmed 1181 bp long ITS-LSU alignment among type strains comprising the M. aquatica sub-clade and strains identified as M. aquatica with parsimony network analysis with 95% connection limit. This type of analysis proved to be useful to delimit ascomycetous (Lachance et al. 2010, 2011) and basidiomycetous (e.g. Yurkov and Golubev 2012) yeast species. Although Lachance et al. (2010, 2011) treated gaps as missing data in the aforementioned parsimony network analyses, we included gaps in the analysis in order to compare distances in the network with results of pair-wise sequence comparisons performed by Tsuji et al. (2019) to support the erection of the species M. hoshinonis.

All strains were connected in the network at 95% connection limit and gaps treated as missing data (Fig. S2). Mrakia niccombsii CBS 8917 was excluded from the network at 95% connection limit and gaps treated as fifth character state but not at 90% (Fig. 2). Parsimony network analysis showed that strains Mrakia sp. DSM 103931, 109846, 109847 and DBVPG 4990 differed from the type strain of M. hoshinonis JCM 32575 and M. aquatica CBS 5443 in 12-15 nucleotide positions and 19-20 nucleotide positions, respectively (Fig. 2). In the network, the type strain of M. hoshinonis JCM 32575 differed in eight positions from “M. aquatica” H2 (GenBank AY052488, AY052480); 13 positions from the type strain of M. aquatica CBS 5443 and in 15 positions from “M. aquatica” H1 (GenBank AY052487, AY052479). Type strains of M. aquatica CBS 5443 and M. niccombsii CBS 8917 differed in 16 positions at 90% connection limit (Fig. 2). It is important to note that these pair-wise distances are lower than those reported by Tsuji et al. (2019), possibly due to sequencing artefacts (read errors and gapped positions in homopolymer repeats) in flanking regions, which were trimmed in the present analysis. But even after trimming, nucleotide sequences of M. hoshinonis strains (GenBank LC335798- LC335801) still contained 4 unique positions (2 mismatches and two insertions) which were not present in any other analysed sequence (members of the genus Mrakia and outgroup Tausonia pullulans). Considering unclear nature (natural variation or sequencing artefacts) of these differences, M. aquatica strain from Iceland (H2; GenBank AY052488, AY052480) may represent M. hoshinonis. Another presumptive M. aquatica strains from Calderone glacier (DBVPG 4994: GQ911548, GQ911523) was identified as M. hoshinonis in our study (Fig. 2). According to these results, M. hoshinonis is not restricted to the Canadian high Arctric and has been previously isolated in Iceland (Birgisson et al. 2003) and Italy (Branda et al. 2010).

Fig. 2
figure 2

Parsimony network analysis of the LSU (D1/D2 domains) rRNA gene and the internal transcribed spacer (ITS) regions of yeasts in Mrakia aquatica subclade (95% connection limit and gaps treated as fifth state). Each connecting line represents one substitution, numbers indicate a position in the alignment and each small circle represents a missing intermediate sequence. Mrakia niccombsii CBS 8917 was disconnected from the network at 95% connection limit but not at 90% connection limit (dashed line)

Full size image

Our results showed that pair-wise distances between type strains of M. hoshinonis, M. aquatica and strain Mrakia sp. DSM 103931 are equal and lay within 14–20 nucleotide positions (Fig. 2). These results suggest that Mrakia sp. DSM 103931 represents a yet undescribed species in the genus Mrakia, for which the name Mrakia fibulata sp. nov. is proposed. We isolated six strains of this species (four from Germany and two from Italy) which showed identical LSU and similar (one gap) ITS sequences. These yeasts were isolated from tree fluxes (this study) and from superficial meltwater collected at Calderone glacier (Branda et al. 2010). Strain DSM 103931 produced true branched hyphae with regularly occurring septa and clamp connections and teliospores after a prolonged incubation at 10 °C on PDA (Fig. 3). Teliospores of a 3- to 6- month-old culture germinated in agar with hyphal segments in addition to the typical basidia-like structures (Fig. 3). Unfortunately, technical circumstances prohibited a clear-cut photo documentation of basidium formation because teliospore germination occured in agar blocks. Other strains reproduced strictly asexually. The sexually reproducing strain was selected as the type for the new species.

Fig. 3
figure 3

Light microscopic images illustrating the different stages of Mrakia fibulata sp. nov. (strain DSM 103931) after 6 months at 10 °C on PDA: true hyphae with clamp connections and teliospores (a–c); germinating teliospores (a, d); teliospores produced with a bud-like projection (b, e); teliospores in pairs (f); asexual cells (g, h). Bars, 20 µm (a–c) and 10 µm (d–h)

Full size image

Description of Mrakia fibulata A.M. Yurkov & B. Turchetti sp. nov., MB 830398

Etymology The species epithet ‘fibulata’ is derived from fibula (L. adj. f., with clamp) and refers to the presence of hyphae with clamp connections.

After 1 week at 15 °C on PDA, streak culture is whitish to cream-coloured, raised, dull, mucoid with a smooth surface. Margins are smooth and entire and the profile is flat. Cells are ovoid to elongate 2–4 × 4–7 μm in size, occur singly or in pairs, and proliferating by polar budding (Fig. 3). After 1 month at 15 °C, the streak culture is cream to light brownish-grey, raised, dull, butyrous with smooth surface. The margin may be fringed with true hyphae. Pseudohyphae and true hyphae with clamp connections are developed after 6 months at 10 °C on PDA (Fig. 3). True, branched hyphae with regularly occurring septa and clamp connections are present after 6 months at 15 °C on PDA. Blastoconidia are ovoid and occur singly, in chains or in clusters. Teliospores are spherical, 5–15 μm in diameter, terminal or intercalary, single and in short chains of 2–4 spores (Fig. 3). Teliospore germination mode is variable. Teliospores germinate with a one- to three-celled metabasidium, 8–12 μm long followed a transfer of an agar block to CMA and PDA. Teliospores also germinate with hyphal segments.

Fermentation of glucose is variable. Growth on d-glucose, D-galactose, L-sorbose, D-glucosamine, D-xylose (variable, weak), L-arabinose, L-rhamnose (weak), sucrose, maltose (variable), trehalose, methyl-alpha-D-glucoside (variable), cellobiose (variable), salicin (variable), arbutin, melibiose, lactose, raffinose, inulin, glycerol (variable), erythritol (variable), ribitol, xylitol, arabinitol, D-glucitol, D-mannitol, galactitol, myo-inositol, 2-keto-D-gluconate, 5-keto-D-gluconate, D-gluconate, D-glucuronate (variable), succinate, citrate (variable, weak), ethanol, quince acid (weak), D-glucarate, gentiobiose, palatinose, L-malate (variable) and Tween 80. No growth on D-ribose, D-arabinose, melezitose, starch, D-galacturonate, lactate, methanol and hexadecane. Utilization of nitrogen sources: growth on potassium nitrate and L-lysin but not on ethylamine and sodium nitrite. Urea hydrolysis and Diazonium Blue B reaction are positive. Starch-like compounds are produced. Maximum growth temperature: 25 °C.

Molecular characteristics (holotype): nucleotide sequences of ITS and LSU (D1/D2 domains) are deposited in NCBI GenBank under the accession number MK372216.

Deposits: holotype, DSM 103931 isolated from xylem sap of a silver birch Betula pendula collected by Anna Yurkova in April 2016 in Braunschweig, Lower Saxony, Germany, permanently preserved as a metabolically inactive culture in the German Collection of Microorganisms and Cell Cultures (DSMZ), Braunschweig, Germany. Isotype DBVPG 8059 is deposited in a metabolically inactive state in the Industrial Yeasts Collection DBVPG, Department of Agricultural, Food and Environmental Sciences, University of Perugia, Perugia, Italy.

Strains studied: DSM103931; paratypes: DSM 109846 from xylem sap of hornbeam Carpinus betulus, Braunschweig, Germany; DSM 109847 from xylem sap of birch Betula pendula, Braunschweig, Germany; DBVPG 4990 and DBVPG 4999 from superficial meltwater collected at Calderone glacier, Gran Sasso d’Italia (an Apennine mountain massif), Abruzzo, Italy (Branda et al. 2010).

Notes New species can be distinguished from closely related Mrakia aquatica, Mrakia hoshinonis and M. niccombsii by growth on inulin, L-rhamnose, myo-inositol, N-acetyl-D-glucosamine, ribitol, and succinate, and inability to grow on D-ribose. In contrast to other sexually reproducing Mrakia species, M. fibulata produces true hyphae with clamp connections.

Discussion

Tree (slime) fluxes are sugar-rich substrates. Here, we studied yeast communities on sites where xylem sap is leaking from cuts on limbs and stems. Fluxes resulting from tree injuries naturally occur in early spring. They result from winter damages occurring through freeze–thaw cycles and injuries caused by birds and animals. In cities and managed forested areas, much of fluxes result from artificial damages during silvicultural interventions like pruning and clearcutting. Numerous yeasts were isolated in the past from spring sap flows (reviewed in Péter et al. 2017). Yeast communities of the studied tree fluxes were strongly dominated by basidiomycetous species, including slowly fermenting species of the genus Mrakia. Despite repeated attempts, we were unable to isolate Phaffia rhodozyma, though various pigmented yeasts of genera Cystofilobasidium, Filobasidium, Rhodosporidiobolus, Rhodotorula, and Vishniacozyma were obtained from spring fluxes. Because spring sap fluxes usually occur early in spring when temperatures are still low, yeast communities were dominated by psychrophilic and psychrotolerant yeasts (reviewed in Buzzini et al. 2018) such as Cystofilobasidium spp., Filobasidium spp., Goffeauzyma gastrica, Mrakia spp., Leucosporidium creatinivorum, and Holtermanniella wattica.

Global warming has already significant effects on our lives causing heat waves, changing precipitation patterns, rising sea level, increasing the incidence of fires and shrinking areas covered by ice. Most of the world’s ice is accumulated in Antarctica followed by ice and snow accumulated during winter in the Northern Hemisphere. Arctic temperatures have increased at almost twice the rate of the rest of the world (Overland et al. 2018). As a result, Arctic sea ice is now declining at a rate of 12.8% per decade since 1979 (NASA global climate change). Climate change affects glacier mass balance through variations in temperature and snowfall. Rising temperatures are responsible for a negative mass balance of most of glaciers and has led to near-global glacier retreat during the last century (Gardner et al. 2013; Parkes and Marzeion 2018; Roe et al. 2017; Sahade et al. 2015). In some regions such as the Alps and Patagonia glacier retreat rates have accelerated during last decades. In Southern Europe glacier retreat is dramatic. Glaciers in the Pyrenees and Apennines lost 84% and 90% of their ice volume, respectively (Grunewald and Scheithauer 2010). Warming determined the cryosphere vanishing and the devastation of habitats of cold-adapted organisms, changing ranges of animals and plants. The loss of ice cover and glaciers retreat also affects microorganisms (Hotaling et al. 2017).

Climate change and cryosphere reduction in space and time can negatively affect psychrophilic and psychrotolerant species, including prominent cold-dwelling yeast genera Glaciozyma, Leucosporidium, Mrakia, Naganishia. Most isolations of Mrakia yeasts were made in polar and non-polar cold environments such as threatened due to climate change Antarctic, Arctic and different glaciers. Strong glacier recession in the Alps, Apennines and Patagonia suggests that the habitat of psychrophilic and psychrotolerant yeasts discovered during the decade (Branda et al. 2010; De García et al. 2007) may disappear soon (Zekollari et al. 2019). If the present deglaciation trend continues, Europe’s southernmost glacier Calderone is unlikely to survive past 2020 (Grunewald and Scheithauer 2010; Zekollari et al. 2019).

Biodiversity and distribution of psychrophilic and psychrotolerant yeasts outside cold habitats are not well known. Frequent sources of isolation such as ice, glacier sediment, snow and meltwater do not tell much about ecological preferences of these yeasts and their dispersal abilities. Most of our knowledge about these yeasts comes from laboratory experiments, not from field studies. We have some understanding of traits helping these yeasts to survive (e.g. production of specific cold active extracellular enzymes) (Buzzini et al. 2012). However, the range of potential habitats and niches suitable for these yeasts remains largely unknown.

The present study reports isolation of psychrophilic yeasts Mrakia spp. (Mrakia blollopis and M. gelida) also from tree fluxes in the temperate zone. Below we review the distribution of Mrakia outside cold habitats. Mrakia frigida was occasionally isolated in Moscow region from vascular plants (Glushakova and Chernov 2010) and Sphagnum moss (Kachalkin et al. 2008). Because of the inability to grow above 20 °C and because no nucleotide sequences were available for any of these isolates, Fell (2011) suggested this was not M. frigida but a different species. However, other strains of M. frigida (for example representative sequence KM384372) were also isolated from an oak bark in different sites across the northern United States using an enrichment protocol and incubation at 10 °C (Sylvester et al. 2015). This species was also found in soils in Germany (Yurkov et al. 2016) and Japan (Nakagawa et al. 2004). M. aquatica, M. blollopis, Mrakia robertii were isolated in summer period from urban soils in Moscow, but not from adjacent low-managed soils (Tepeeva et al. 2018a). Same authors reported M. aquatica above and below ground in locations situated above hot water heating pipelines in Moscow urban areas (Tepeeva et al. 2018b). Similarly to culture-based experiments, culture-independent detected Mrakia yeasts in various cold habitats (Duo Saito et al. 2018; Rämä et al. 2016). An interesting observation was made in a culture-independent study, which investigated the effects of deglaciation on microbial communities (Rime et al. 2015). Sequence reads corresponding to Mrakia species were more characteristic (association strength) for barren (10 years) and sparsely vegetated (60 years) soils in the Damma glacier (Central Alps of Switzerland) forefield. These nutrient-poor soils were inhabited by metabolically versatile bacteria which are able to grow at oxygen-depleted and oligotrophic conditions.

These examples demonstrate the growing evidence that psychrophilic yeasts of the genus Mrakia live also outside cold habitats. So far only M. fibulata was able to propagate above 20 °C. This is the first record of Mrakia species that exhibits psychrotolerant, not strictly psychrophilic, aptitude. Sylvester et al. (2015) already supposed that although Mrakia seems to have a competitive advantage at 10 °C, it can grow at warmer temperature. Isolation of M. aquatica, M. blollopis, M. gelida, M. robertii and newly described M. fibulata from the temperate climate suggests these yeasts might be common inhabitants of boreal and temperate climates. Whether or not Mrakia spp. and other cold-adapted yeasts survive warm temperatures in a dormant state is yet unknown. Adaptations of such species to cold (e.g. enzymes and cryoprotectors) are better studied than the mechanisms to cope with warm temperatures.

Spring tree fluxes are an ephemeral but rather widespread substrate which can be easily recognized and sampled by amateurs in a Citizen Science project. A good example of such Citizen Science initiative has been reported in a recent study of yeasts from Dutch soils (Groenewald et al. 2018). We believe that new studies and isolates of psychrophilic and psychrotolerant yeasts from temperate climates would improve our understanding of yeast adaptation to the changing environment.