Yeasts from glacial ice of Patagonian Andes, Argentina


  • Virginia de Garcia,

    Corresponding author
    • Laboratorio de Microbiología Aplicada y Biotecnología, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, INIBIOMA (CONICET-UNCo), Río Negro, Argentina
    Search for more papers by this author
  • Silvia Brizzio,

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, INIBIOMA (CONICET-UNCo), Río Negro, Argentina
    Search for more papers by this author
  • María Rosa van Broock

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue, INIBIOMA (CONICET-UNCo), Río Negro, Argentina
    Search for more papers by this author

Correspondence: Virginia de Garcia, Quintral 1250, San Carlos de Bariloche, Río Negro C.P. 8400, Argentina. Tel.: +54 2944 428 505; fax: +54 2944 423 111; e-mails:;


Glacial ice and snow are known habitats for cold-adapted microorganisms. Research on cold-adapted yeast biodiversity from Perito Moreno and Mount Tronador glaciers (Patagonia, Argentina), and production of extracellular enzymatic activity at low temperatures (5 and 18 °C), was performed and described in this study. Ninety percent (90%) of the isolates were basidiomycetous; 16 genera and 29 species were identified. Twenty-five percent (25%) of total isolates corresponded to psychrophilic yeasts, whereas 75% were psychrotolerant yeasts. Eighty-five percent (85%) of all isolates had at least one enzymatic activity. Multiple correspondence analysis and cluster classification revealed a relationship between certain genera and some enzymatic activities. Cold-adapted yeast isolates were able to hydrolyze natural compounds (casein, lipids, starch, pectin, and carboxymethylcellulose) at low temperatures, suggesting a significant ecological role of these organisms as organic matter decomposers and nutrient cyclers. These yeasts are especially relevant for metabolic and ecological studies, as well as for yeast-based biotechnological process at low temperatures.


Ice is considered a life-preserving medium that can entrap randomly deposited microorganisms that may remain viable for a long time (Gunde-Cimerman et al., 2003). Recent studies have shown that different types of ice (i.e. snow, sea ice, glacial ice) can be inhabited by psychrotolerant and psychrophile microorganisms (Margesin et al., 2005; Butinar et al., 2007; Turchetti et al., 2008). These microorganisms can use complex biopolymers as energy sources, by synthesizing extracellular enzymes active at low temperatures (Margesin et al., 2007a; Frisvad, 2008). Psychrophilic yeasts play an essential role in nutrient cycling and biomass production processes in cold ecosystems (Margesin et al., 2007a). Studying microbial diversity in extreme environments, such as ice environments, is interesting because these micororganisms exhibit metabolic adaptations to extreme conditions and may provide insight into potential biotechnological applications. Furthermore, psychrophilic and psychrotolerant microorganisms found in these environments may be used as bio-indicators in studies monitoring global warming (Gunde-Cimerman, 2006; Kuhn, 2008).

Currently, most glaciers of the world are retreating rapidly as a result of global warming (Delgado Granados et al., 2007). Glaciers of Mount Tronador Patagonia, Argentina, are a clear example, having experienced constant retreat during the past 100 years (Villarosa et al., 2008). Perito Moreno glacier is one of the few glaciers that are in a stable situation, that is, it is neither advancing nor retreating. The study of psychrotrophic microbial populations in this vanishing or barely stable cold habitat is of increasing scientific interest (Branda et al., 2010).

Glacier/climate interactions in Patagonian region are of relevance to understand the global climate change pattern. In addition, icefields and periglacial areas hold valuable information for Quaternary paleoenvironments (Stuefer et al., 2007). There is also a dense forest cover, associated with recent glacial deposits. Lakes, rivers, and peat-lands are the main landforms that create conditions for the study of past and present environmental variations in the region (Villarosa et al., 2008). Occurrence of cold-adapted yeasts in glacial environments in Patagonia was first studied from meltwater rivers of Mount Tronador (de Garcia et al., 2007). Occurrence and diversity of yeast in glacier ice in Patagonia have not been studied before.

The aim of this investigation was to assess the occurrence and biodiversity of cold-adapted yeasts in ice samples from Frias glacier (Mount Tronador) and Perito Moreno glacier (Patagonian icefields), and study their extracellular enzymatic activities.

Materials and methods

Area description and sampling

Ice samples were aseptically collected, in summer, from: (i) Frias glacier, Mount Tronador (71°50′W, 41°11′S) in February 2007; (ii) Perito Moreno glacier (73°51′W, 49°15′S) in March 2008.

Air temperature and GPS positioning were recorded in situ. Ice samples were melted aseptically at room temperature and then filtered through Millipore® membrane filters (pore size 0.45 μm, diameter 47 mm).

Yeast isolation and quantitative analysis

Volumes of 100 and 150 mL of melted ice were filtered through Millipore® nitrocellulose membranes, using a sterilized Nalgene® device. Filters were placed on MYP agar (malt extract 7 g L−1; yeast extract 0.5 g L−1; soytone 2.5 g L−1; agar 15 g L−1; pH 4) containing chloramphenicol 100 mg L−1 and in agar media with different substrates (proteins, lipids, starch, carboxy-methylcellulose, or pectin; see Extracellular Enzymatic Activity description below); Petri dishes were incubated at 10 °C for 1 week and at 5 °C for up to 1 month. Plates were periodically examined, and all emerging yeast colonies were transferred to MYP agar plates without antibiotics and purified. The isolates were stored at −80 °C and included in the CRUB Yeast Collection (CRUB: Yeast Collection of Centro Regional Universitario Bariloche). Yeast colony-forming units (CFU) were registered for quantitative analysis of yeast occurrence. Yeast cell counts from three replicates were used to calculate mean values and SDs.

Yeast characterization and identification

Yeast characterization and identification up to genus level (morphological features and physiological tests) were performed according to standard methods, as described by Kurtzman et al. (2011).

In a previous study on yeasts in glacial meltwater (de Garcia et al., 2007), 38 strains remained unidentified. These 38 strains were included in this study for identification, phylogenetic analyses, extracellular enzymatic activity and were also included in the discussions of yeast diversity.

PCR fingerprinting analysis was performed using mini/microsatellite-primed PCR technique (MSP-PCR). Protocols for DNA extraction, PCR, and electrophoresis conditions were those described by Libkind et al. (2003), and primer M13 was used. For DNA sequence analysis, D1/D2 domains of the large subunit of ribosomal DNA (LSU rDNA) was studied, primers NL1 (5′-GCA TAT CAA TAA GCG GAG GAA AAG-3′) and NL4 (5′-GGT CCG TGT TTC AAGACG G-3′) were employed, and internal transcribed spacer (ITS) region was sequenced using the forward primer ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and the reverse primer ITS4 (5′-TCCTCCGCTTATTGATATGC-3′).

Sequencing was performed by the Sequencing Service Macrogen (Korea). BigDye terminator cycle sequencing kits were used in sequence reactions (Applied Biosystems, Foster City, CA). Sequences were obtained with an ABI Prism 3700 (Applied Biosystems). Sequences downloaded from GenBank are indicated in the gene trees by their GenBank accession number; newly generated sequences are indicated by their strain number (CRUB) and GenBank accession number (Table S1, Supporting information).

Phylogenetic analyses

Sequences were automatically aligned using clustalx, and alignments were adjusted manually using mega5 (Tamura et al., 2011). To estimate phylogenetic relationships on the basis of LSU rDNA (D1/D2 domains), neighbor-joining analysis (K2P) was performed using mega software, version 5 (Tamura et al., 2011).

Extracellular enzymatic activity

The isolated strains were tested for their ability to degrade starch, protein (casein), pectin, carboxymethyl-cellulose, and Tween-80 according to the procedures described by Brizzio et al. (2007). Calibrated suspensions of 106 cells mL−1, grown for 24–48 h, were surface inoculated on agar plates using a multipoint inoculation device. Plates containing each substrate were incubated at 5 and 18 °C. Enzymatic activity was recorded after 5 days in samples incubated at 18 °C and after 21 days in those incubated at 5 °C.

Statistical analyses

Yeast diversity in each glacier was studied using the Shannon–Weaver (H) index with Hutcheson's t-test (a = 0.05) as described in Moreno (2001). Similarities between communities were studied with Jaccard index (J) according to Chao et al. (2005), using the occurrence frequency of each isolated species.

Enzyme production at both temperatures was compared using Mann–Whitney–Wilcoxon nonparametric test for two independent samples, and the analysis was carried out using sigmastat V2.03 program. Analysis of multiple correspondences and hierarchical classification were carried out to evaluate the results of semiquantitative extracellular enzymatic activity. statistica 6.0 software package was used.

Results and discussion

Yeast occurrence and quantitative analyses

Average yeast counts for each sampling site are shown in Table 1. Yeast counts between sampling sites were not significantly different (P > 0.117), nor were counts for different selective media within each sampling site (P > 0.117). No yeasts colonies were observed in selective media with casein and Tween 80 substrates (Table 1). Yeast counts were similar to those obtained for meltwater rivers of Mount Tronador (de Garcia et al., 2007) and in other aquatic environments of Patagonia Argentina (1–2 × 103 ± 1–4 × 10CFU L−1; Libkind et al., 2003; Brandao et al., 2011). Studies in similar environments of the Italian Alps also registered similar values of yeast counts, 1 × 101 and 4 × 103 CFU L−1 (Buzzini et al., 2005; Turchetti et al., 2008; Branda et al., 2010). These values are relatively low when compared with coastal or polluted aquatic systems (Nagahama, 2006), because of the oligotrophic nature of the ice samples (Hagler & Ahearn, 1987; Foght et al., 2004).

Table 1. Location of sampling sites and viable yeast counts
Sampling sitesFrías Glacier (Mount Tronador)Perito Moreno Glacier
  1. a

    No yeasts colonies were observed in selective media with casein and Tween 80 substrates.

Location41°08′43″S, 71°49′01″W49°S, 73°W
pH of the sample6.56.4
Yeast viable counts in general and selective media * Media ± SD (CFU L−1)
 MYP3.3 × 103 ± 9.6 × 1023.3 × 103 ± 2.1 × 103
 YNB + Pectin pH 52.5 × 103 ± 8.0 × 1025.3 × 103 ± 3.7 × 103
 YNB + Pectin pH 72.1 × 103 ± 6.8 × 1023.6 × 103 ± 1.0 × 103
 YNB + CMC04.0 × 103 ± 2.8 × 103

A total of 153 isolates were classified. Of these, 115 corresponded to the ice samples of the Frias and Perito Moreno glaciers. These were assigned to 16 genera and 29 species (Table 2). The remaining 38 corresponded to the unidentified isolates from a previous study on yeasts in meltwaters of Mount Tronador (de Garcia et al., 2007). These were classified into five genera and 16 different species (Table 2). Results in this section will refer to all 153 isolates. Ninety percent of the total yeast strains studied (153 isolates) belonged to the phylum Basidiomycetes.

Table 2. Yeast species isolated from glacial meltwater and ice from Patagonia Argentina, number of isolates and origin
SpeciesTotal number of isolatesOrigin of strains
Mount TronadorFrias GlacierPerito Moreno Glacier
  1. Occurrence frequencies percentage are shown in parentheses.

  2. a

    Possible new species.

  3. b

    New species recently described.

  4. c

    These were sampled in a previous study on yeasts in glacial meltwater (de Garcia et al., 2007), but had remained unidentified in that study.

Dioszegia crocea 24 (15.6)213
Psychrophilic strain sp. 1a19 (12.5)19
Sporobolomyces ruberrimus 15 (9.8)15
Dioszegia fristingensis 12111
Udeniomyces pannonicus 88
Cryptococcus victoriae 761
Udeniomyces pyricola 55
Mrakiella aquatica 44
Rhodotorula sp. 1a33
Rhodotorula mucilaginosa 33
Dioszegia butyracea 33
Cr. spencermartinsiae b 33
Udeniomyces megalosporus 33
Mrakia robertii 33
Aureobasidium pullulans 33
Candida sp. 1a33
Cryptococcus sp. 1a22
Cryptocccus sp. 2a22
Holtermanniella festucosa 22
Candida mesenterica 211
Mastigobasidium intermedium 11
Rhodotorula sp. 2 CRUB 1756a11
Cryptococcus sp. 3 CRUB 1267a11
Holtermanniella sp. 1 CRUB 1256a11
Udeniomyces sp. 1 CRUB 1695a11
Mrakia sp. 1 CRUB 1707a11
Mrakia sp. 2 CRUB 1706a11
Guehomyces pullulans 11
Bensingtonia yamatoana 11
Cryptococcus sp. 4 CRUB 1245a11
Cryptococcus terrícola 11
Cryptococcus wieringae 11
Ascomycota sp. 1 CRUB 1755a11
Phaeococcomyces sp. 1 CRUB 1760a11
Wickerhamomyces patagonicus b 11
Debaryomyces hansenii 11
Candida sp. 2 CRUB 1719a11
Candida sp. 3 CRUB 1220a11
Candida sp. 4 CRUB 1295a11
Candida maritima 11

All isolates were adapted to living at cold temperatures, 75% were psychrotolerant (growth at 5–25 °C), while the remaining 25% were psychrophilic (growth at 5–15 °C). The occurrence of psychrophilic yeasts in cold environments reported here was similar to that found in Alpine glaciers (17%) by Turchetti et al. (2008) and Branda et al. (2010), and in Arctic glaciers (30%) by Pathan et al. (2010).

Species diversity values, measured with the Shannon–Weaver index, were higher in the ice from Frias glacier and meltwaters from Mount Tronador (H = 2.23 and H = 2.52, respectively) than in ice from Perito Moreno (H = 1.72). There were no significant differences for Shannon–Weaver diversity indices among all compared communities (P > 0.05). Jaccard index for community similarity analysis showed that Frias and meltwater yeast communities and Perito Moreno and meltwater yeast communities were the most similar (J = 0.15 and J = 0.13, respectively); however, the index values were low for all the compared communities, and no similarity was found when yeasts communities of Perito Moreno and Frias were compared.

A relatively higher richness index of taxa among ice and meltwater samples was observed, compared to the values reported for soil samples in Patagonian forest by Mestre et al. (2011). In addition, Brandao et al. (2011) mentioned similar values in water samples from Nahuel Huapi Lake (Patagonia, Argentina; coast sites H = 2.2 and pelagic sites H = 2.8).

Yeast identification, diversity, and ecology

Basidiomycetous yeasts were the predominant group in ice from Patagonian glaciers, belonging mostly to subphylum Agaromycotina, particularly to Tremellales (55 isolates) and Cystofilobasidialles orders (27 isolates). Similar results from different cold environments of the world (Antarctica, Alpine glaciers, Alaska, and Arctic) have been reported (Thomas-Hall et al., 2002, 2010; Bergauer et al., 2005; Margesin et al., 2005, 2007b; Butinar et al., 2007; Connell et al., 2008, 2010; Turchetti et al., 2008; Shivaji & Prasad, 2009; Branda et al., 2010; Uetake et al., 2011; Vaz et al., 2011) and also from other aquatic environments of Patagonia (Libkind et al., 2003; de Garcia et al., 2007; Brandao et al., 2011).

Connell et al. (2008) mentioned that in Antarctic soils (South Victoria Land), 90% of yeast isolates were basidiomycetous, 43% of these corresponded possible new species. In this study, 40% (11 species) of basidiomycetous isolates were also possible new species. Several authors have suggested that the predominance of basidiomycetous yeasts in these extreme environments is because they are more nutritionally versatile and more tolerant to extreme environmental conditions than ascomycetous yeasts (Sampaio, 2004; Brandao et al., 2011).

The species most frequently recovered from Perito Moreno glacier ice was a yeast identified as psychrophilic yeasts sp. 1 (19 isolates), followed by and Sporobolomyces ruberrimus (15 isolates) identified by MSP-PCR fingerprinting (data not shown) isolated from ice from Frias glacier. While Dioszegia species were most frequently recovered (D. crocea 24 and D. fristingensis 12 isolates) from Mount Tronador Glaciers (ice and meltwaters). Species belonging to the Dioszegia genus are frequently found associated with plants and terrestrial substrates (Inácio et al., 2005), while D. crocea species more commonly related to cold environments.

Some cosmopolitan species, such as Rhodotorula mucilaginosa (three strains from Perito Moreno glacier), Cryptococcus victoriae (six strains from meltwaters of Mount Tronador and one strain from Perito Moreno Glacier), and Aureobasidium pullulans (three strains from Frias glacier) were isolated.

Psychrophilic yeasts sp. 1 was shown to be identical in D1/D2 sequence to yeasts isolated in Alaska (Basidiomycota sp. GU 74), and considering ITS region, the closest related strain was an Antarctic yeast (CBS 8941), with 94% similarity in blast result. This species is closely related to Camptobasidium hydrophilum and together with the unidentified species from Alaska and Antarctica (GU 54 and CBS 8941) formed a new clade within the class (Fig. 1). The presence of teliospores without mating was observed in some isolates of this species.

Figure 1.

Phylogenetic placement of psychrophilic basidiomycetous Patagonian yeasts species obtained by neighbor-joining (distance K2P method) of the LSU rRNA gene D1/D2 domains. Names in bold type are strains described in this work. Bar, substitutions accumulated every 100 nucleotides. Bootstrap values higher than 50% are shown (1000 replicates). TType strain.

Strains related to the genus Rhodotorula were found (Rhodotorula sp. 1 and Rhodotorula sp. 2 CRUB 1756). Species Rhodotorula sp. 1 was related to Rh. glacialis and had four nucleotide differences in the D1/D2 region. Further analyses are needed to determine whether these strains represent new species. Rhodotorula sp. 2 CRUB 1756 had 15 nucleotide differences in D1/D2 region and 11 in ITS region, with the closest species Rhodotorula laringysT, having a basal position in Rh. laringys clade.

Three potential new Cryptococcus species, related to Tremellales order, were identified, Cryptococcus sp. 1 (related to Cr. foliicola), Cryptococcus sp. 2 (related to Cr. laurentii clade), and Cryptococcus sp. 3 (related to Kwoniella clade formal Cr. heaveanensis clade). Formal description of these species is in progress. Species Cr. spencermartinsiae sp. nov. has been recently described (de Garcia et al., 2010a).

Species of genera Udeniomyces and Guehomyces were identified (U. pannonicus, U. pyricola, U. megalosporus, Udeniomyces sp. 1 CRUB 1697 and G. pullulans); Udeniomyces sp. 1 CRUB 1697 had 11 nucleotide differences in the D1/D2 region and 16 in the ITS region with the closest species being U. pseudopyricolaT.

Hortemanniella festucosa (Cryptococcus festucosus) of the recently described order Holtermanniales (Wuczkowski et al., 2010) was also isolated.

Udeniomyces pannonicus, G. pullulans and H. festucosa, species are psychrophilic or psychrotolerant and have been isolated from different cold terrestrial regions of the world (Fell & Guého-Kellermann, 2011; Takashima & Nakase, 2011). Wuczkowski et al. (2010) concluded that Holtermanniella species possess significant amounts of polyunsaturated fatty acids, a fatty acid composition typical of yeasts adapted to cold environments.

Regarding the Filobasidialles order, Cryptococcus terricola, Cr. wieringae and a possible new species Cryptococcus sp. 4 CRUB 1245 (13 nucleotide difference in D1/D2 region and 9 in ITS region with the closest species Cryptococcus arrabidensis CBS 8678T) were identified.

Representatives of the psychrophilic genus Mrakia and the anamorphic-related genus Mrakiella were isolated in this survey: three strains of Mrakia robertii, two possible new species (Mrakia sp. 1 CRUB 1706 and Mrakia sp. 2 CRUB 1707), and four strains of Mrakiella aquatica.

Yeasts belonging to phylum Ascomycota were less abundant, being approximately 10% of total isolates. Species Candida maritima, Debaryomyces hansenii, Candida mesenterica, and A. pullulans were identified, and the new species Wickerhamomyces patagonicus was recently described (de Garcia et al., 2010b).

From 11 identified species, seven (63%) failed to match the already-described species, and most probably represent new ones. Four species were related to Candida in Saccharomycetales order, one strain was related to the species Hyalodendriella betulae (Helotiales), and one to the melanin-producing species Phaecoccomyces nigricans (Chetothyriales; Fig. 2).

Figure 2.

Phylogenetic placement of Ascomycetous Patagonian yeasts species obtained by neighbor-joining (distance K2P method) of the LSU rRNA gene D1/D2 domains. Names in bold type are strains described in this work. Bar, substitutions accumulated every 100 nucleotides. Bootstrap values higher than 50% are shown (1000 replicates). TType strain.

As was mentioned, isolates of species Rhodotorula sp. 1 related to Rh. glacialis were identified. Gadanho & Sampaio (2009) proposed the ‘ecoclade’ concept, which refers to species that are phylogenetically related and show metabolic adaptations associated with physicochemical conditions present in the environment from which they were recovered. These authors suggest the existence of two ‘ecoclades’, acidic and psychrophilic. The psychrophilic ecoclade is defined by species from glacial environments from the Alps and mountainous areas from Himalaya. Rhodotorula sp. 1, isolated from glacier meltwater from Mount Tronador belongs to this clade, supporting this ecoclade proposal.

All Mrakiella strains isolated here produced mycelium and teliospores in culture medium without nitrogen (Yeasts Carbon Base). Teliospores were placed in distilled water and incubated at 5 °C for up to 1 year; after this period, agar blocks containing teliospores were placed in agar-water and, after 2 months, germinating teliospores were observed in two strains (Mrakiella sp. CRUB 1272 and M. aquatica CRUB 1209). Hyphae and teliospores developed directly from a single cell, without mating, and clamp connections were not observed. Germinating teliospores of strain CRUB 1272 produced three to five single-celled structures (Fig. 3), a septate structure was observed (Fig. 3c).

Figure 3.

Phase-contrast micrograph of teliospores of Mrakiella strains isolated from Patagonian glaciers. Mrakiella sp. CRUB 1272 (a–c), after 7 weeks in YCB agar media at 10 °C followed by 1 year in distilled water, (a, b) Teliospores germinated, (c) Teliospores germinated, arrow is showing septated structure. (d) Mrakiella aquatica CRUB 1709, germinated teliospores. (e, f) Teliospores of M. aquatica CRUB 1708 and CRUB 1721, respectively. On 2% agar after 2 months. Bar = 10 μm.

Is not clear whether teliospores found here in Mrakiella species were from asexual or sexual origin. Generally, germination of sexual teliospores in Mrakia species is not common and in some species has rarely been observed (Mrakia frigida and M. gelida; Fell, 2011). Further studies will be necessary to determine the sexual state of Mrakiella strains.

The presence of teliospores has been observed in almost all psychrophilic species described (Mrakia and Glaciozyma; Thomas-Hall et al., 2010; Fell et al., 2011; Turchetti et al., 2011). Production of this structure can enhance survival in a diverse array of harsh environmental conditions, including cold habitats.

The presence of these extremophilic microorganisms in geographically distant regions could be the result of ecological fitting and genetic adaptation that allowed them to increase and improve their survival in these specific environments (Margesin et al., 2007b; Rossi et al., 2009).

In summary, and given the hypothesis that microorganisms in extreme environments could have differential evolutionary ratios compared with those in temperate environments (Skidmore et al., 2000; Rosenberg & Hastings, 2003, 2004), studying and understanding the evolution of extremophiles will increase the basic knowledge of evolutionary processes, allowing a better evaluation of potential ecological consequences of environmental changes and possible effects on human health (Gostincar et al., 2011).

Enzymatic analyses

Several selective media containing different substrates were included in the isolation step, attempting to improve the recovery of yeast strains with the ability to metabolize these different substrates. Results were not significantly different to those obtained with MYP agar, indicating that yeast diversity in the samples was adequately reflected by culturing in this generic medium.

Extracellular activity of selected yeast strains was previously reported (de Garcia et al., 2007). However, in that study, the evaluation was not complete, as it included only some strains for the enzymatic analyses. Those strains were therefore included in this work, for a complete analysis of enzymatic production (from ice and meltwater samples) at 5 and 18 °C. Thus, a total of 212 strains were assessed (115 strains from ice and 97 from meltwaters), and the results are shown in Fig. 4.

Figure 4.

Extracellular enzymatic activity of yeast strains from ice and meltwater rivers from Perito Moreno glacier and Mount Tronador glaciers. Bars indicate the SD of halo/colony mean values. N indicates the number of positive strains (of 212 strains) for the corresponding enzymatic activity. a, Significantly different activity at 5 and 18 °C (P = 0.001).

Eighty-five percent (85%) of the 212 yeasts strains were able to produce at least one enzymatic activity, and 18% produced five different enzymatic activities. Differences in qualitative (number of positive strains) and semiquantitative (intensity of degrading activity) expression of extracellular enzymatic activities at 5 and 18 °C were investigated. Higher activity at 5 °C was observed for all extracellular activities analyzed (both in number of positive strains and in intensity), and statistically significant differences were found for proteolysis (P ≤ 0.001), hydrolysis of carboxymethyl-cellulose (cellulose activity; P = 0.025), and hydrolysis of Tween-80 (P = <0.001; Fig. 4).

Turchetti et al. (2008) reported similar results for yeasts isolated from Alpine glaciers, indicating that 73 selected isolates had enzymatic activity at low temperatures (4 °C). However, Pathan et al. (2010) observed higher degradation at 22 than at 8 °C in psychrotolerant yeasts from Arctic glacier meltwaters, but in this case, the authors used an incubation period of only 10 days at 8 °C, when, in general, 20 days are needed at this temperature for psychrophilic and psychrotolerant strains to achieve stationary growth phase, in which extracellular enzymes are produced.

Hydrolysis of Tween-80 (esterase activity) was the most common activity, being present in 146 isolates (69%). These results are in agreement with those reported by Brandao et al. (2011) for enzymatic activity of yeasts from an oligotrophic lake of Patagonia (71.8% of the total isolates) and by Margesin et al. (2005), Buzzini et al. (2005) and Turchetti et al. (2008) for yeasts isolates from Alpine glacial environments (89%, 46%, and 86% respectively). Hydrolysis of carboxymethyl-cellulose (96 isolates) was the second most frequent activity, followed by proteolysis (60 isolates) and pectinolysis (62 isolates), amylase activity was the least frequent of the activities (27 isolates).

An association between yeast genera and the ability to produce extracellular enzymes was found through multiple correspondence and hierarchical classification analysis. Six different classes were proposed:

  • Class 1: Sporobolomyces isolates produce amylase at 5 and 18 °C.
  • Class 2: Leucosporidiella and Udeniomyces isolates produce protease, CMC-cellulose, pectinase, and esterase at 5 and 18 °C.
  • Class 3: Cryptococcus isolates produce CMC-cellulase at 5 and 18 °C.
  • Class 4: Dioszegia isolates produce esterase at 5 and 18 °C.
  • Class 5: Mrakia and Mrakiella isolates produce protease, esterase, and pectinase at 5 °C.
  • Class 6: Ascomycota and Rhodotorula isolates were not associated with enzymatic activity tested in the essay conditions. It must be noted that even though, according to multiple correspondence and hierarchical classification analysis, ascomycetous yeasts were not associated with any enzymatic activity; A. pullulans isolate had more than two extracellular activities, which is in agreement with other reports for this species (Turk et al., 2007; Zalar et al., 2008). Buzzini et al. (2005) found similar results for ascomycetous yeast isolated from Alpine glacier.

In conclusion, results of the laboratory cultures carried out in this study support that the yeasts in these extreme (cold) habitats possess metabolic adaptation to low temperatures. Cold-adapted Cryptococcus isolates with the ability to produce more than one hydrolytic cold-active enzyme were obtained from Patagonian glaciers. Also Mrakia and Mrakiella isolates able to produce up to five different extracellular cold-active enzymes, and resistance structures (teliospores) were recovered and characterized. These microorganisms are heterotrophic, and their ability to degrade organic macromolecules through the secretion of extracellular hydrolytic cold-adapted enzymes suggest that, as proposed by Turchetti et al. (2008), they may have a significant ecological role in organic matter decomposition and nutrients in glacial environments. This role is also supported by the presence of organic carbon and organic and inorganic nitrogen in glacial meltwater and ice (Skidmore et al., 2000; Foght et al., 2004; Margesin et al., 2007a).

The biotechnological (and industrial) relevance of cold enzymes from psychrophilic yeasts has been emphasized (Thomas-Hall et al., 2010). Association observed here between certain taxa of basidiomycetous yeasts and extracellular enzymatic activities facilitates a directed search of genera of interest, both in culture collections and in the environment, to find strains with possible biotechnological applications.

Basidiomycetous yeasts are a diverse group of fungi with considerable industrial and medical importance and have undeniable potential for economic exploitation (Abadias et al., 2003; Qin et al., 2004; Schisler et al., 2011). This study has contributed to the understanding of their biodiversity and ecological roles.

Final remarks

The Patagonian Andes possess unique physical and environmental characteristics. Glaciers present in this area have a high potential for glaciological and paleoclimatic studies, and also for microbiological surveys, as shown in this work. The relevance of studies on the effects of climatic and environmental changes on continental glaciers has been shown elsewhere (Villarosa et al., 2008; Branda et al., 2010). Microorganisms inhabiting these withdrawing glaciers may be released into soil, rivers, and oceans, possibly complementing or changing the existing microbial communities (Butinar et al., 2007).

Glaciers of Patagonia Argentina offer unexplored environments and are true cold-adapted yeast reservoirs. Furthermore, these yeasts possess adaptations, such as cold-active enzymes, which may undeniably contribute to biotechnological research and application.


This work was accomplished with financial aid from Universidad Nacional del Comahue (Project B143), ANPyT (PICT06-1176), and Consejo Nacional de Investigaciones Científicas y Tecnicas (CONICET; PhD fellowship given to and project PIP424). Ministerio de Ciencia y Tecnología (International Bilateral Cooperation MINCYT–MHEST SLO08/11). We would like to thank the authorities of Parques Nacionales (Argentina) for providing permission for water sample collection within National Parks, and Dr. Sonia Fontenla for providing Perito Moreno samples.