Yeasts from an oligotrophic lake in Patagonia (Argentina): diversity, distribution and synthesis of photoprotective compounds and extracellular enzymes

Authors

  • Luciana R. Brandão,

    1. Departamento de Microbiologia, ICB, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
    Search for more papers by this author
  • Diego Libkind,

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Instituto de Investigaciones en Biodiversidad y Medioambiente, UNComahue-CONICET, Bariloche, Rio Negro, Argentina
    Search for more papers by this author
  • Aline B.M. Vaz,

    1. Departamento de Microbiologia, ICB, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
    Search for more papers by this author
  • Lília C. Espírito Santo,

    1. Departamento de Microbiologia, ICB, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
    Search for more papers by this author
  • Martín Moliné,

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Instituto de Investigaciones en Biodiversidad y Medioambiente, UNComahue-CONICET, Bariloche, Rio Negro, Argentina
    Search for more papers by this author
  • Virginia de García,

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Instituto de Investigaciones en Biodiversidad y Medioambiente, UNComahue-CONICET, Bariloche, Rio Negro, Argentina
    Search for more papers by this author
  • Maria van Broock,

    1. Laboratorio de Microbiología Aplicada y Biotecnología, Instituto de Investigaciones en Biodiversidad y Medioambiente, UNComahue-CONICET, Bariloche, Rio Negro, Argentina
    Search for more papers by this author
  • Carlos A. Rosa

    1. Departamento de Microbiologia, ICB, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
    Search for more papers by this author

  • Editor: Gary King

Correspondence: Diego Libkind, Laboratorio de Microbiología Aplicada y Biotecnología, Instituto de Investigaciones en Biodiversidad y Medioambiente, UNComahue-CONICET, Quintral 1250, Bariloche 8400, Rio Negro, Argentina. Tel.: +154 2944 428505; fax: +154 2944 423111; e-mail: libkind@crub.uncoma.edu.ar

Abstract

Nahuel Huapi (NH) Lake is an oligotrophic temperate lake of glacial origin with high transparency, surrounded by well-developed forests and located at San Carlos de Bariloche, Nahuel Huapi National Park, in Patagonia, Argentina. In this lake, we characterized yeast distribution and diversity along a south-to-north transect and established a relationship between the ability to produce photoprotective compounds (PPCs) (carotenoid pigments and mycosporines) and the occurrence of yeast at different collection points. Subsurface water samples were filtered for yeast isolation. Total yeast counts ranged between 22 and 141 CFU L−1, and the highest values corresponded to the most impacted sites. Littoral sites had a low proportion of yeast-producing PPCs and this group prevailed in pelagic sites. This is probably a result of the high transparency of the water and the increased UV exposure. The yeast community from NH Lake showed a high species richness and a uniform distribution of taxa between pelagic and border collection points. Yeasts were identified as belonging to 14 genera and 34 species. Rhodotorula mucilaginosa and Cryptococcus victoriae were the most frequently found species, representing 14.4% and 13.6% of the total yeast isolates, respectively. Most of the yeast isolates demonstrated at least one extracellular enzymatic activity (mainly cellulase and lipase activities), which suggested that these microorganisms are metabolically active in the lake.

Introduction

Yeasts occur in a variety of freshwater systems, including eutrophic to ultra-oligotrophic lakes, lagoons, rivers, groundwater, glaciers and glacial meltwaters (Hagler & Ahearn, 1987; Libkind et al., 2003; Nagahama, 2006; de García et al., 2007; Medeiros et al., 2008; Brandão et al., 2010). The yeast diversity in these ecosystems is highly affected by a variety of abiotic and biotic factors, such as temperature, pressure, UV radiation (UVR), salinity, fauna, flora, soil run-off and anthropogenic effluents. The conditions prevailing in these natural habitats determine the metabolic activity, growth and survival of the yeast populations (Deak, 2006). Despite the fact that yeasts are common in different aquatic systems, our knowledge of the ecology of freshwater yeasts remains incipient. Notable gaps in our knowledge are mainly related to the factors that drive the distributional patterns, diversity and functional significance of yeasts in aquatic systems (Spencer & Spencer, 1997).

The majority of the studies on yeasts from freshwater ecosystems have focused on the association of yeasts with contaminated waters (de Almeida, 2005; Hagler, 2006; Nagahama, 2006; Medeiros et al., 2008; Brandão et al., 2010). These studies identified the presence of opportunistic yeast pathogens belonging to species such as Candida albicans, Candida parapsilosis, Candida krusei, Candida guilliermondii, Candida glabrata and Candida tropicalis in polluted water ecosystems. These yeasts are part of the faecal microbiota of many animals, including humans (Medeiros et al., 2008). Less research has been performed on yeast occurrence in pristine aquatic environments. The oligotrophic to ultra-oligotrophic lakes of glacial origin in Patagonia, Argentina, are examples of this type of minimally polluted water ecosystem (Díaz et al., 2000). Lakes near the city of San Carlos de Bariloche, which is located within Nahuel Huapi National Park (NHNP), can be considered to be pristine and have minimal anthropogenic influence. These Patagonian natural lakes are typically exposed to elevated UVR due to their transparency and ultra-oligotrophic conditions (Villafañe et al., 2001). Recent reports have demonstrated that UVR is a selective factor that favours the occurrence of more UV-resistant yeast species in these lakes (Libkind et al., 2009). To minimize UV-induced damage, several yeast species synthesize antioxidants and UV sunscreen compounds such as carotenoids and mycosporines (Libkind et al., 2009; Molinéet al., 2009, 2010b).

The biodiversity of basidiomycetous yeasts in certain lakes in Andean Patagonia (Argentina) was investigated by Libkind et al. (2003, 2005b, 2009). These studies exclusively focused on carotenoids and mycosporine-producing basidiomycetous yeasts and detected several novel species, including Rhodotorula meli, Sporidiobolus longiusculus, Sporobolomyces patagonicus and Cystofilobasidium lacus-mascardii (Libkind et al., 2005a, 2009, 2010). Studies of the entire cultivable yeast community of these pristine lakes have not been conducted previously. In this work, we studied the diversity and frequencies of yeast species along a south-to-north transect of Nahuel Huapi (NH) Lake in Andean Patagonia. In addition, we investigated extracellular enzyme production and photoprotective compound (PPC) synthesis by yeast species and correlated our results with their distribution in the lake.

Materials and methods

Collection area and yeast sampling

NH Lake is the largest water body in NHNP (41°S–71°W), Patagonia, Argentina, with an area of 557 km2 and a maximum depth of 465 m (Calcagno et al., 1995). It is considered an oligotrophic temperate lake of glacial origin (Pedrozo et al., 1993; Díaz et al., 2000) and has an elevated transparency (20.9 m) (Zunino & Diaz, 2000). Its waters have a very low dissolved organic carbon (<0.5 g m−3), which results in a low vertical attenuation of UVR in the water column (Morris et al., 1995). Most energetic wavelengths of UVB radiation can occur down to 10-m deep, which means that organisms living or migrating to the upper 10 m would be affected by this deleterious band or the indirect effect of UVR (i.e. reactive oxygen species generation) (Balseiro et al., 2008).

The city of San Carlos de Bariloche (c. 100 000 inhabitants) is located on the southern coast of the lake, and this coast has a significant periphyton growth (Baffico, 2001). Although the southern coast of the lake is strongly influenced by urban discharges, no relevant industrial or extensive agricultural activities exist in the area (Guevara et al., 2002), tourism being the main commercial activity. The northern coast of the lake is surrounded by a dense native forest of Nothofagus dombeyi and Austrocedrus chilensis and the anthropogenic impact is negligible.

On 18 December 2007, five independent 300–400-mL water samples were collected in sterile bottles from seven sites (NH0, 1, 2, 3, 4, 5 and 6) that were located along a south-to-north transect of the lake (Fig. 1, Table 1). Samples were collected from the subsurface at a depth of c. 30 cm. The first sampling site (NH0) was located in the city coast, at 4 m from the edge of the lake close to the Ñireco river discharge that collects runoff and urban discharges from the city. Point 1 (NH1) was located 30 m from the city border, at the western Ñireco river inflow, and point 6 was located approximately 5 m from the northern edge of the lake, near a forest of N. dombeyi and A. chilensis (Fig. 1). The distances between sampled points (NH1–NH6) were approximately 1.3 km. The samples were stored at 5 °C and processed until a maximum of 10 h after collection. Sampling was performed at midday (12:00–13:00 hours) of a cloudless and windless day. Water temperature was recorded in situ; pH and conductivity were measured in the laboratory with a 3310 JENWAY metre and an Orion 135 apparatus, respectively.

Figure 1.

 Location of NH Lake and sampling points.

Table 1.   Physicochemical characteristics, yeast counts and percentages of pigmented basidiomycetous yeasts, mycosporine (MYC)-positive yeasts and yeasts producing photoprotective compounds (pigmented and/or mycosporine-positive yeasts)
SitesTemperature (°C)pHConductivity
(μS cm−1)
Yeast counts
(CFU L−1)
% Pigmented
yeasts
% MYC
positive
% Pigmented
and/or MYC+yeasts
NH0166.9034.5141±9424.0±24.014.2±14.140.0±28.4
NH1106.4935.097±7846.9±31.852.1±27.672.9±19.8
NH2116.5836.049±4371.0±23.969.8±15.581.3±22.5
NH3116.4035.043±4367.9±19.077.3±17.781.6±19.1
NH4116.5835.573±4773.6±19.673.5±16.775.5±14.6
NH5116.4236.025±1369.7±27.568.7±36.272.3±20.9
NH6146.5235.022±38.3±12.083.3±23.683.0±24.0

UV irradiance data acquisition

The irradiance at different wavelengths in the UV range (305, 320, 340 and 380 nm) was acquired for the period 16–20 November 2007; the total-day doses for each wavelength was calculated and then averaged. The maximum irradiance detected for each day was registered and then averaged. Ground level irradiance data were obtained by means of a radiometer GUV 511 (Biospherical Instruments) and provided by Laboratorio de Fotobiología (INIBIOMA-CONICET, UNComa, Bariloche). The radiometer was placed 12 km from the city of S.C. Bariloche and 5 km from the NH Lake.

Yeast isolation and quantitative analysis

Variable volumes of water (300–400 mL) were filtered through Millipore nitrocellulose membranes (0.45-mm pore size, 47-mm diameter) with a sterilized Nalgene device. The filters were placed on the surface of yeast extract–malt extract agar plates (YMA, yeast extract 0.3%, malt extract 0.3%, peptone 0.5%, dextrose 1%, agar 2%, pH 4.0) containing 200 mg L−1 chloramphenicol and incubated at 15 °C for up to 1 week. The yeasts were chosen for isolation based on colony morphology. When possible, at least three randomly selected colonies of each different morphotype were purified by repeated streaking on YMA plates and preserved at −80 °C or with liquid nitrogen for later identification. Yeast CFU were registered for quantitative analysis of yeast occurrence using a stereoscopic microscope (Olympus SZX9) on the seventh day of incubation. The percentages of pigmented yeast colonies from each sample were also calculated. The averages and SDs of the total viable yeast cells (CFU) per litre of water of each sampling point of the lake were calculated. The same was carried out for the percentage of pigmented colonies.

Yeast identification

All yeasts were preliminarily grouped based on their cultural morphology, urease production and physiological characteristics based on assimilation tests of carbon and nitrogen sources and the production of amyloid compounds (Yarrow, 1998). The yeasts were also characterized by PCR fingerprinting, using the mini/microsatellite-primed PCR technique (MSP-PCR) (Libkind et al., 2003).

DNA extraction

For DNA extraction, yeast colonies were grown on modified Sabouraud agar (glucose 2%, peptone 1%, yeast extract 0.5% and agar 2%) at 15 °C overnight, transferred to 2-mL sterile tubes (Eppendorf) containing 100 μL extraction buffer solution (50 mmol Tris L−1, 250 mmol NaCl L−1, 50 mmol EDTA L−1, 0.3% w/v SDS, pH 8) and incubated at 65 °C for 30 min. After incubation, 100 μL phenol/chloroform/isoamilic alcohol (25 : 24 : 1) was added. The mixture was vortexed vigorously, incubated for 3 min and centrifuged for 3 min at 7558 g. DNA was dried overnight at room temperature, suspended in 100 mL TE buffer (10 mM Tris, 10 mM Na-EDTA, pH 8.0) and stored in a refrigerator.

PCR fingerprinting

The synthetic oligonucleotide (GTG)5 and the core sequence of the phage M13 (GAGGGTGGCGGTTCT) were used in MSP-PCR experiments, respectively. The PCR reactions were performed according to Libkind et al. (2003). Yeast strains with identical DNA banding patterns were grouped and putatively considered to belong to the same species (Gadanho & Sampaio, 2002). At least one representative strain of each MSP-PCR group was subjected to sequence analysis of the D1/D2 domains of the large subunit of the rRNA gene as described below. Physiologically distinct strains with unique MSP-PCR banding patterns were also selected for direct identification by sequencing of the D1/D2 region of the rRNA gene. When necessary, the internal transcribed spacer (ITS) domains of the rRNA gene were also sequenced.

Sequencing analysis

Total DNA was extracted using the methods described above. The D1/D2 variable domains of the large subunit of the rRNA gene were amplified as described previously by Lachance et al. (1999) using the primers NL-1 (5′-GCATATCAATAAGCGGAGGAAAAG-3′) and NL-4 (5′-GGTCCGTGTTTCAAGACGG-3′). The ITS regions of rRNA genes were amplified with the universal primers ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) as described by White et al. (1989). Sequencing of the D1/D2 region and ITS domains was performed directly from purified PCR products using a MegaBace™ 1000 automated sequencing system (Amersham Biosciences). The sequences obtained were compared with those included in the GenBank database using the basic local alignment search tool (blast at http://www.ncbi.nlm.nih.gov) (Altschul et al., 1997).

Mycosporine production and extracellular enzymatic activities

The ability to synthesize mycosporine was tested according to the method described previously by Libkind et al. (2005b). The yeast isolates were tested for their ability to degrade starch, protein (casein), pectin, carboxymethyl-cellulose and lipids (Tween-80) according to the procedures described by Brizzio et al. (2007). Calibrated suspensions of 106 cells mL−1, which were grown for 24–48 h, were inoculated on the surface of agar plates using a multipoint inoculation device (de García et al., 2007). Plates containing each substrate were incubated at 4 or 20 °C. Enzymatic activity was analysed after 5 days in the samples incubated at 20 °C and after 21 days in those incubated at 4 °C. The enzymatic activities for specific substrates were evaluated as described by Brizzio et al. (2007).

Statistical analyses

A one-way anova was used to test the differences in the percentages of pigmented basidiomycetous yeasts and/or mycosporine-producing yeasts and the nonproducing PPCs among the sampling sites. The variable total yeast count (CFU) was analysed with a Mann–Whitney rank sum test. When possible, a post hoc multiple comparison was performed by applying the Tukey test (α=0.05). Otherwise, sampling sites were grouped into coastal (NH0 and 6) and pelagic (NH1, 2, 3, 4 and 5) categories and compared using the Student t-test. The same was done to compare the yeast community composition at impacted sites (NH0 and 1) with that at pristine sites (NH2, 3, 4, 5 and 6).

Species diversity at coastal and pelagic sites was measured in terms of the richness, evenness and dominance given by three indexes: (1) Shannon H=−Σni/n ln (ni/n), (2) Simpson's index=Σ(ni/n)2 and (3) Dominance D=sum((ni/n)2), where ni is the number of individuals of the taxon i and n is the total number of individuals. All results were obtained with 95% confidence, and bootstrap values were calculated from 1000 iterations. Species richness refers to the number of species in a community, and species dominance refers to the contribution of individuals. The index calculations were performed using the computer program past, version 1.90 (Ryan et al., 1995).

Results and discussion

Physical and chemical characteristics

The physical and chemical characterization of the water samples are summarized in Table 1. The pH was nearly constant along the transect, returning values between 6.4 and 6.9. The conductivity also did not change significantly among the sampled sites. The water temperature was higher at sites near the coast than at pelagic sites due the greater depth of the lake at points NH2 through NH5.

Table 2 shows an estimate of UV doses and maximum irradiances to which the NH Lake surface was exposed during sampling. Due to the low attenuation coefficients (Kd) of the NH water (Morris et al., 1995) and the low sampling depth (30 cm), it can be assumed that the microorganisms under study were exposed to almost 100% of the UV irradiance reaching the water surface.

Table 2.   Daily UV dose and maximum irradiance detected at the time of sampling in the NH lake at different wavelengths
Wavelength (nm)Daily UV dose (kJ m−2)Maximum irradiance
(μW cm−2 nm−1 s−1 )]
NH UV attenuation
coefficient: Kd (m−1)*
  • Media and SD are provided.

  • *

    Obtained from Morris et al. (1995); average values of data from three sampling sites.

3051.857±0.0348.864±0.2620.543
3208.416±0.27931.775±0.9440.407
34020.045±0.72071.476±0.3030.303
38025.370±1.01489.640±0.1770.177

Yeast counts and quantitative analyses

Average yeast counts for each sampling site are shown in Table 1. The counts ranged from 22 to 141 CFU L−1, with the highest values corresponding to the most anthropogenically influenced sites (NH0 and NH1). In general, the yeast counts observed in the NH Lake were typical of clean lakes, which usually contain <100 CFU L−1 and rarely exceed 200 CFU L−1 (Hagler & Ahearn, 1987). The highest yeast counts, which were found at sites NH0 and NH1, could be related to the proximity of these sites to the city border (4 and 30 m, respectively), where the effect of the Ñireco river inflow and human activities is the largest. Significant differences were found (P=0.001) when comparing sites affected by anthropogenic activities (NH0 and NH1) with sites that experienced lesser human influence (NH2, 3, 4, 5 and 6). Yeast occurrence decreased with increasing distance to the south-edge of the lake (Table 1). To find higher yeast counts in the south-coast waters than in pelagic zones is reasonable due to the organic matter (and probably yeast) input of the Ñireco river into the NH waters, which are then subjected to a dilution effect. Baffico (2001) found that the area of influence of the Ñireco river had relatively high levels of readily assimilable forms of phosphorus and nitrogen than other nonanthropogenically impacted coasts of NH. The north coast, even though it has several small inflows, showed the lowest yeast counts. This is probably because such streams are not as anthropogenically impacted as the Ñireco river and do not carry much organic matter (and yeasts).

Additional factors limiting yeast propagation in the water body are the lack of nutrients and the lower water temperatures to which yeasts are subjected once they enter NH waters. These, together with the effect of UVR, which in the case of NH is highly significant in the upper layer due to the high transparency of its waters (Morris et al., 1995), are important factors conditioning yeast survival in the water column and thus determining its distribution in the lake. It was then hypothesized that a significant fraction of the yeasts entering the littoral areas of the lake is rapidly eliminated, and thus in pelagic areas mostly yeasts able to cope with such extreme conditions are found.

Pigmented basidiomycetous yeasts are common in most aquatic yeast communities and often comprise >50% of the yeast population, especially in oligotrophic marine or fresh waters (Hagler & Ahearn, 1987; Libkind et al., 2003). In the present study, these yeasts were present in all water samples, and relatively high numbers were found in the pelagic sampling points. Carotenoid pigments (antioxidants), which are synthesized by several pigmented basidiomycetous yeast species from oligotrophic aquatic environments, have been reported to have a photoprotective function in yeasts (Molinéet al., 2009, 2010b). For example, the yeast Rhodotorula mucilaginosa, an abundant species in patagonian lakes (Libkind et al., 2003), produces large quantities of torularhodin, a carotenoid that affords UVB photoprotection (Molinéet al., 2010b). Significant differences were found when comparing the percentage of pigmented yeasts found at pelagic and border lake sites (P<0.005). These differences could be related to the general lower susceptibility of carotenogenic yeasts to UV than nonpigmented ones (Moline, 2004). The latter possibly survive shorter periods in the water column than the former. In a previous study, focused mainly on pigmented yeasts of high-altitude lakes (mountain lakes) from Patagonia, Libkind et al. (2009) found that pigmented yeasts prevailed only in highly transparent lakes. Due to the high transparency of NH water, allowing an extraordinary penetration of solar radiation (Morris et al., 1995; Balseiro et al., 2008), UV appears to be a strong selective factor in favour of more UV-resistant yeast species. This has also been demonstrated for planktonic organisms (Villafañe et al., 2001; Marinone et al., 2006).

Another type of PPC that can be synthesized by yeasts is mycosporines, which are UVB screening compounds that also have antioxidant properties (Molinéet al., 2010a). The percentage of mycosporine-positive yeasts found at the south coast (NH0) was significantly (P<0.005) lower (14%) than the rest of the sampling sites (>52%). Contrary to what was observed for pigmented yeasts, the highest percentage of mycosporine-positive yeasts was observed in the north coast (83%; NH6), while pelagic sites had values ranging 52–77%. When the distribution of yeasts producing at least one of the two PPCs was analysed, a trend similar to that found for mycosporine-positive yeasts was observed (Table 1). A situation similar to that postulated for pigmented yeasts is also observed for mycosporine-positive yeasts, in which the extreme conditions (particularly UV) may select for mycosporine-synthesizing yeasts, explaining their prevalence in pelagic sites. Even though, the high values observed for the north coast (NH6 site is 5 m from the forest border) are not in agreement with such hypotheses, this could be due to the influx of yeasts from phylloplane run-off. The phyllosphere is a notable and common niche for yeasts (Fonseca & Inácio, 2006), which is highly exposed to solar radiation. We have recently found that the surface of Nothofagus spp. leaves harbour a peculiarly large proportion of mycosporine-positive yeasts (Muñoz, 2010) similar to that observed in NH6. Interestingly, the proportion of pigmented yeasts in such leaves rarely exceeded 10%. Libkind et al. (2009) found that mycosporine-synthesizing species were poorly represented in high-altitude lakes, an environment exposed to high UVR in which such PPC could be a useful adaptation for survival. A plausible explanation arises from the fact that due to the high altitude, the vegetation surrounding those mountain lakes is limited to a few Nothofagus shrubs, and therefore a much lower run-off from the phylloplane (and thus of mycosporine-positive yeasts) is expected.

It can be hypothesized that the north coast receives already UV-adapted (mycosporine-positive) yeasts from the nearby Nothofagus phyllosphere, while the south coast receives mostly yeasts without PPC (less adapted) or ubiquitous pigmented yeasts normally related to human activity (e.g. R. mucilaginosa and Aureobasidium pullulans) from urban discharge through the Ñireco river. Additionally, our data support the idea that UVR is an important factor that determines yeast community structures in Andean oligotrophic lakes and that yeasts producing carotenoids and/or mycosporine possess an adaptative advantage in highly UV-exposed habitats than those incapable of producing them.

Comparative analyses among pelagic and coastal zones using the Shannon H, Simpson's, and Dominance indexes are shown in Table 3. The pelagic zone of the lake presented the highest Shannon H and Simpson's indexes. However, the Dominance index was higher at the coastal points of the lake than at the pelagic points. The values of these indexes showed that the yeast community from NH Lake has a relatively higher richness index (H=2.5±0.2) and a uniform distribution of taxa among pelagic and border lake sites (Simpson's index=0.85±0.6). Consequently, a very low species dominance (D=0.09±0.06) was observed.

Table 3.   Diversity indexes of yeasts from coastal and pelagic sites in NH Lake
Diversity indexesSampling sitesNahuel Huapi
Coast sitesPelagic sitesMean±SE
  1. Coastal sites: NH0 and NH6; Pelagic Sites: NH1, NH2, NH3, NH4, NH5. The numbers in parentheses represent the lower and upper diversity values, respectively, with 95% confidence and bootstrap values calculated from 1000 iterations.

Shannon H2.22.82.5±0.2
(2.1/2.8)(2.6/3.0) 
Simpson's index0.80.90.85±0.6
(0.8/0.9)(0.8/0.9) 
Dominance D0.10.080.09±0.06
(0.06/0.1)(0.06/0.1) 

Table 4 shows the distribution among the NH sampling sites of the various yeast species in terms of their production of PPC (mycosporines and carotenoids). Interesting cases include the yeast-like fungus A. pullulans, which has higher counts in the anthropogenically impacted coast and gradually reduces its numbers in sites away from that coast. This is in agreement with the fact that A. pullulans is a ubiquitous organism and that it is closely related to human activities (Zalar et al., 2008). A similar case is that of R. mucilaginosa, a pigmented but mycosporine-negative yeast, also considered ubiquitous (Libkind & Sampaio, 2009) and that appears to be introduced into the lake from the city coast. This is in agreement with the fact that R. mucilaginosa was a minor component of the yeast community of Nothofagus phylloplane (Muñoz, 2010). Contrary to A. pullulans, R. mucilaginosa is apparently well-adapted to survive in NH waters given that it occurred in all sampling sites. This might be partially related to the high tolerance of R. mucilaginosa to UVR (Molinéet al., 2010b; Libkind et al., 2011). The category including yeasts producing only mycosporines comprised mostly Cryptococcus species, and these were mainly found in pelagic sites. Yeasts not producing any of the two PPCs included most ascomycetous yeasts and also several Cryptococcus species.

Table 4.   Identification, distribution and occurrence of yeasts isolated from NH Lake
Yeast speciesSampled points
NH0NH1NH2NH3NH4NH5NH6
  • *

    Media and SD of CFU L−1 of yeast in the water samples.

  • Yeast-like fungus.

  • MYC, Mycosporines; PPC, Photoprotective compounds.

Pigmented and MYC-positive species
 Aureobasidium pullulans4.4±3.7*3.3±2.32.6±2.5 1.8±2.8  
 Delphinella strobiligena   0.6±1.4   
 Dioszegia hungarica  0.6±1.4    
 Dioszegia sp.1 1.3±1.8     
 R. laryngis 9.7±13.717.5±39.116.8±29.94±8.95.9±11.6 
 R. pinicola  13.5±30    
 R. slooffiae    43.5±318.5±12.4 
Pigmented and MYC-negative species
 C. victoriae 6.6±1311±10.36.6±11.717.5±21.93.1±4 
 Cystofilobasidium capitatum1.8±2.62±4.49±12.44±8.9   
 C. Infirmominiatum1.3±2.90.6±1.4     
 R. colostri2.6±5.9      
 Rhodosporidium diobovatum  0.5±1.1    
 Rhodotorula mucilaginosa22.9±25.66.6±6.61±2.20.5±1.10.6±1.41±1.32±2.8
 Rhodotorula sp.1 0.6±1.4     
Nonpigmented and MYC-positive species
 Bullera dendrophila     1.3±1.8 
 Cryptococcus adeliensis3.3±7.4 7.8±8.42.6±5.9 0.5±1.1 
 C. diffluens 0.6±1.40.5±1.1    
 C. festucosus 1.3±2.9     
 C. heveanensis      4±5.6
 C. magnus1.3±1.80.6±1.40.5±1.1  0.6±1.4 
 C. saitoi 10.5±22.7 1.5±2.21±2.2  
 C. stepposus   0.6±1.4   
 C. wieringae 3.3±7.4     
 Guehomyces pullulans    0.5±1.1 0.8±1.7
Species without PPC
 Candida parapsilosis   1.5±3.3   
 C. railenensis12.6±28.35.9±11.60.5±1.1    
 C. sake0.6±1.45.3±7.6  0.6±1.4  
 Candida sp. 1      4±5.6
 C. carnescens     1.3±2.9 
 Cryptococcus sp.1   2±4.4   
 C. tephrensis    2.5±5.5 4±5.6
 Debaryomyces hansenii 6.6±14.8     
 G. pullulans0.6±1.4      
 Hanseniaspora uvarum 1.3±1.8 0.6±1.40.6±1.4  
 Pichia fermentans2±4.4      

Yeast identification and ecology

We obtained 94 pigmented yeasts (including the yeast-like fungi A. pullulans and Delphinella strobiligena) and 55 nonpigmented yeast isolates. The isolates with white to cream colonies were grouped as nonpigmented yeasts. All yeasts were preliminarily grouped based on their cultural and physiological characteristics, and the groups with similar or identical morphological and physiological characteristics were subsequently subjected to MSP-PCR fingerprinting. Characterization by MSP-PCR fingerprinting allowed the formation of seven groups of identical DNA banding patterns among the pigmented yeasts and five patterns among nonpigmented yeasts (data not shown). One representative strain from each group was selected for sequence analysis of the D1/D2 domains of the 26S rRNA gene. The pigmented species of these MSP-PCR fingerprinting groups were represented by A. pullulans, Cystofilobasidum infirmominiatum, Cystofilobasidum capitatum, R. mucilaginosa, Rhodotorula laryngis, Rhodotorula slooffiae and two isolates of the genus Dioszegia, identified as Dioszegia sp. 1. The nonpigmented yeast groups were identified as Candida railenensis, Cryptococcus adeliensis, Cryptococcus magnus, Cryptococcus saitoi and Cryptococcus victoriae. Among the pigmented yeasts, five isolates showed unique MSP-PCR fingerprinting patterns and were identified by sequencing as Dioszegia hungarica, Rhodosporidiumdiobovatum, Rhodosporidium colostri, Rhodosporidium pinicola and Rhodotorula sp. 1. Ten yeast isolates showed unique physiological characteristics and carbon assimilation patterns and were identified as C. parapsilosis, Candida sp. 1, Cryptococcus sp. 1, Cryptococcus carnescens, Cryptococcus festucosus, Cryptococcus haveanensis, Cryptococcus stepposus, Cryptococcus weringae, Debaryomyces hansenii and the yeast-like fungus D. strobiligena. The remaining isolates were grouped based on morphological similarity and identical results for physiological tests. These yeasts were identified as Bullera dendrophila, Candida sake, Cryptococcus diffluens, Cryptococcus tephrensis, Guehomyces pullulans, Hanseniaspora uvarum and Pichia fermentans (Table 4).

Some yeast isolates showed five or more nucleotide differences in D1/D2 domains of the rRNA gene when compared with the most closely related known species; therefore, they could represent novel yeast species. According to Kurtzman & Robenett (1998), isolates of the same species usually have only zero to two nucleotide differences in the D1/D2 region of the large subunit of the rRNA gene. One isolate of Cryptococcus (strain identified as Cryptococcus sp. 1, GenBank accession number HM990974) differed by nine nucleotide substitutions in the D1/D2 region of the rRNA gene from the closest known species, Cryptococcus spencermartinsiae, a yeast isolated from glacial melting waters in Patagonia by de Garcia et al. (2010). Our isolate probably represents a new Cryptococcus species that is related to C. spencermartinsiae. Another isolate, identified as Rhodotorula sp. 1, showed five nucleotide substitutions in the D1/D2 region of the rRNA gene compared with the Rhodotorula sp. NBRC 105035 (Sporidiobolales, GenBank accession number AB462346) and 24 substitutions from the closest known relative species Sporobolomyces inositophilus (GenBank accession number AF189987). This isolate probably also represents a new yeast species of the genus Rhodotorula because it does not produce ballistoconidia. The isolate identified as Candida sp.1 presented 100% sequence similarity to Candida sp. SDY 211 (GenBank accession number AY731817) and Candida sp. AS 2.3084 (GenBank accession number DQ451013), which were isolated from acidic aquatic environments in Portugal and from an unidentified flower in Tibet, respectively. These three isolates represent new yeast species that are most closely related to several Candida species but have >70 indels of difference in their D1/D2 domains. The isolate identified as Dioszegia sp. 1 presented one substitution in the D1/D2 domain sequence compared with the strain Dioszegia sp. CRUB 1147 (GenBank accession number EF595753), which was isolated from altitudinal lakes in Argentina. These isolates probably represent a new species that is closely related to D. hungarica.

The yeast isolates from NH Lake were identified as belonging to 13 genera and 34 species (Table 4). Basidiomycetous yeasts were represented by 73.8% of the isolates. In general, these yeasts are more nutritionally versatile and more tolerant of extreme environmental conditions than ascomycetous yeasts (Sampaio, 2004). In addition, basidiomycetous yeasts are often found in association with the phyllosphere of terrestrial plants (Fonseca & Inácio, 2006); their occurrence in aquatic environments could be considered the result of a run-off from this substrate (Hagler & Ahearn, 1987; Lachance & Starmer, 1998).

Species of Cryptococcus were common to all of the sites sampled in NH Lake. These yeasts, which represented 34.8% of the total isolates, were the most frequent and diverse group, followed by Rhodotorula, which represented 26.7% of the total isolates. These genera have been reported in other studies from Patagonian aquatic environments (Libkind et al., 2003, 2009; de García et al., 2007; Russo et al., 2008), suggesting that these yeasts occur frequently in such environments. The wide nutritional plasticity and the adaptability to harsh environmental conditions of many yeasts species of these genera explain their high frequencies of isolation in NH Lake.

Rhodotorula. mucilaginosa was the most frequently isolated yeast (21 isolates; 28.7% of total pigmented strains) and was present at all of the sampled points. This species is ubiquitous and has been isolated in all kinds of natural substrates (Gadanho et al., 2006). Libkind et al. (2009) reported that 87.5% of the pigmented yeasts occurring in Negra Lake, an ultra-oligotrophic freshwater from Patagonia, Argentina, were R. mucilaginosa. According to these authors, an increase in the population of this species may be related to a temporary increase in the organic matter in the lakes. Molinéet al. (2010b) suggest that R. mucilaginosa enhances UVB survival by producing the carotenoid pigment torularhodin; however, it does not produce mycosporines. This yeast species appeared in the NH Lake at a lower frequency than in the highly transparent Patagonian mountain lakes (>1400 m.a.s.l.), where it was prevalent (Libkind et al., 2009). Other pigmented species such as D. hungarica, Dioszegia sp. 1, R. diobovatum, R. colostri, R. pinicola, R. slooffiae and Rhodotorula sp. 1 were less frequently isolated. Species of the pigmented mycosporine-negative yeasts C. infirmominiatum, C. capitatum, R. diobovatum and Rhodotorula sp. 1 were poorly represented, suggesting that they have a low resistance to the UVR conditions found in NH Lake. In a previous work, we reported the low tolerance of Cystofilobasidium species to UVB (Libkind et al., 2009). Dioszegia strains have been found recently in glacial meltwaters and mountain lakes in Patagonia (de García et al., 2007; Libkind et al., 2009). In this work, two Dioszegia species were observed at low frequencies in NH Lake. These yeasts are pigmented and able to produce mycosporine and show high tolerance to UVB (Libkind et al., 2009). However, our results suggest that Dioszegia species are minor components of the yeast communities in Andean aquatic environments. This yeast genus is frequently found in association with plants and terrestrial substrates (Fonseca & Inácio, 2006).

The pigmented species C. victoriae occurred frequently in NH Lake, and this is the first report on its occurrence in Patagonian lakes. This yeast was originally isolated from soil, moss, lichen, soil, Granite Harbour soil, Lichen Valley and Vestfold Hills in Antarctica (Thomas-Hall et al., 2002). However, the habitat of this yeast is wider than previously thought because it was isolated from glaciers in Italy and from indoor dust in central Finland (Pitkäranta et al., 2008; Branda et al., 2010). Other Cryptococcus species occurred in minor frequencies, including C. adeliensis, C. carnescens, C. diffluens, C. magnus and C. saitoi, which are frequently reported in cold habitats (Vishniac, 2006; de García et al., 2007; Russo et al., 2008; Libkind et al., 2009) and therefore appear to be autochthonous in cold ecosystems.

Among the ascomycetous yeasts isolated in our study, the yeast-like fungi A. pullulans and D. strobiligena were able to produce mycosporines. This is the first report of the presence of mycosporine-producing ascomycetous yeasts in the lakes of Patagonian Argentina. Aureobasidium pullulans was the most frequent ascomycetous species. This yeast-like species is often isolated from many different types of water (Sláviková & Vadkertiová, 1997a). Debaryomyces hansenii, a ubiquitous yeast species found in aquatic environments (Nagahama, 2006), was found at site NH1, which was relatively highly impacted by human activities. The Candida species C. parapsilosis, C. sake, C. railenensis and Candida sp.1 represented 8% of the total yeast isolates. A single isolate of C. parapsilosis was observed at site NH3 in the middle of the lake. This species is often reported in aquatic environments that have high levels of organic matter from industrial and domestic wastes (Hagler, 2006; Nagahama, 2006; Medeiros et al., 2008). However, our results indicated that the NH Lake has low levels of anthropogenic organic pollution because only one opportunistic yeast species of faecal origin was obtained. Most isolates of C. sake were obtained at point NH1. Candida sake occurs in different aquatic environments, including lagoons (Boguslawska-Was & Dabrowski, 2001), algae, shore soil, lakes and penguin dung from Antarctic environments (Goto et al., 1969). This species is able to grow in habitats ranging from c. 5 to 30 °C (Vishniac, 1996). C. railenensis occurred in higher amounts at points NH0 and NH1, which are near the city coast. The species was described based on isolates from a rotten trunk of N. dombeyi and Nothofagus obliqua (Ramírez & González, 1984) and is probably associated with the forest vegetation found in Patagonia.

When the yeast diversity of NH Lake was compared with that of other cold and tropical aquatic environments (Table 5), a remarkable resemblance to Antarctic habitats and other oligotrophic aquatic environments in Patagonia was observed. Thirty-five per cent of the species isolated in our study are present in Antarctic habitats (Vishniac, 2006), and 54% of the species are present in other water bodies in Patagonia (Libkind et al., 2003, 2009; de García et al., 2007; Russo et al., 2008). In this work, only five ubiquitous yeast species (A. pullulans, C. parapsilosis, C. infirmominiatum, D. hansenii and R. mucilaginosa) were also found in tropical aquatic habitats (Table 5).

Table 5.   Comparison of yeast species isolated from NH Lake and from other cold and tropical aquatic environments
Species from Nahuel
Huapi Lake
Tropical aquatic
environments*
Alpine
glaciers
Arctic
environments
Antarctic
environments§
Oligo to ultra-oligotrophic aquatic
environments from Patagonia,
Argentina
Aureobasidium pullulans+++
Bullera dendrophila
Candida sake++
C.railenensis
C. parapsilosis+++
Cryptococcus adeliensis+++
C. carnescens
C. diffluens+
C. festucosus+
C. heveanensis
C. magnus+
C.saitoi++++
C. stepposus+
C. tephrensis
C. victoriae+++
C. wieringae
Cystofilobasidium capitatum+
C. infirmominiatum++
Debaryomyces hansenii+++
Dioszegia hungarica++
Delphinella strobiligena
Guehomyces pullulans+
Hanseniaspora uvarum
Pichia fermentans
Rhodotorula mucilaginosa++++
R. colostri+
R. laryngis++++
R. pinicola+
R. sloofiae+
Rhodosporidium diobovatum++

Screening for extracellular enzymatic activity

The extracellular enzymatic activity profiles of the yeast and yeast-like isolates are shown in Table 6. Of the 148 tested strains, 82% showed at least one extracellular enzymatic activity at 5 and/or 20 °C. These yeasts were represented by 72.4% of basidiomycetous isolates and 24.8% of ascomycetous isolates. The percentage of yeasts producing extracellular enzymes was slightly higher in the pelagic sites of the lake (56.5%). Esterasic activity (hydrolysis of Tween-80) was the most widely expressed extracellular enzyme activity (positive for 71.8% of the total isolates), followed by the degradation of carboxymethyl-cellulose (cellulase activity, 53.0%), pectinase (42.9%), amylase (26.8%) and protease (22.1%) activities. Among the ascomycetous yeasts, 19 isolates of A. pullulans, four isolates each of D. strobiligena and H. uvarum, three isolates each of C. parapsilosis and C. railenensis, two isolates of C. sake and one isolate of D. hansenii showed esterasic activity. Cellulase was the second most prevalent enzymatic activity observed among the ascomycetous yeast isolates; 29 isolates were positive for this trait. Almost all of the strains of A. pullulans showed the ability to produce all of the tested enzymes. Among the basidiomycetous species of the genus Cryptococcus, 50 isolates showed at least one extracellular enzymatic activity at 5 and/or 20 °C. Cryptococcus adeliensis, C. magnus, C. saitoi and C. victoriae exhibited activity for all enzymes tested. Lipolytic and cellulolytic activities were expressed mainly by species of Cryptococcus and Rhodotorula. Candida sp. 1, Cryptococcus heveanensis, P. fermentans, R. colostri and Rhodotorula sp. 1 were not able to hydrolyse any of compounds tested. de García et al. (2007) also showed that a significant proportion of yeast isolates from glacial meltwater rivers in Patagonia, Argentina, were capable of degrading natural compounds. The fact that a significant proportion of yeasts are able to hydrolyse natural compounds such as lipids, starch, protein and pectin suggests that these strains are metabolically adapted to cold environments and have a significant ecological role in organic matter decomposition and nutrient cycling.

Table 6.   Extracellular enzymatic activities of yeasts from NH Lake
SpeciesNumber strainsAmA*PrAPecACelALpA
5°C20°C5°C20°C5°C20°C5°C20°C5°C20°C
  • *

    AmA, amylolytic activity; PrA, proteolytic activity; PcA, pectinolytic activity; CelA, cellulolytic (degradation of carboxymethyl-celullose); LpA, lipidic (esterasic) activity.

Aureobasidium pullulans2016171719181820201919
Bullera dendrophila2    22  11
Candida sake411 1  2222
C. railenensis61   221121
C. parapsilosis1  11  1111
Cryptococcus adeliensis103111553366
C. carnescens1      1111
C. diffluens2    12  22
C. festucosus1      1111
C. heveanensis1          
C. magnus64333226666
C. Saitoi63112554456
Cryptococcus sp. 11    1111  
C. stepposus1      1111
C. tephrensis2      2212
C. victoriae2043223316171818
C. wieringae1   1111111
Cystofilobasidium capitatum71   13 244
C. infirmominiatum2    22    
Debaryomyces hansenii1      1111
Dioszegia sp. 12    112122
Dioszegia hungarica1      1 11
Delphinella strobiligena1        1 
Guehomyces pullulans321  223322
Hanseniaspora uvarum4   1114444
Pichia fermentans2          
Rhodotorula colostri1          
Rhodotorula mucilaginosa21  115111377
R. laryngis9        88
R. pinicola1       111
R. sloofiae611    1 33
Rhodotorula sp. 11          
Rhodosporidium diobovatum1  1 11111 
Total number of strains148          
Total number of positive strains for the tested enzymatic activities 3628273253627477102101

Conclusion

The occurrence and distribution of yeasts along a transect of NH Lake showed peculiar distributional patterns probably influenced by inputs of allochthonous organic matter from the borders of the lake and by abiotic factors such as UVR. PPC-producing yeasts were mainly found in pelagic points of the lake, suggesting that both mycosporine and carotenoid production capacities are important for yeast survival under high UVR conditions expected in the upper layer of NH Lake. The significant relationship between the ability of the yeasts to produce PPC and their distance from the lake's shore also indicates that yeasts entering NH Lake are subjected to extreme conditions that imply a significant force towards the selection of UV-tolerant yeasts. Most of the species isolated in our work were typical of cold aquatic environments. In addition, most of the yeast isolates presented at least one active extracellular enzyme, which suggests that these microorganisms are metabolically active in the lake and could contribute to organic matter recycling in this cold freshwater environment. This work represents the first comprehensive survey of the cultivable ascomycetous and basidiomycetous yeast community of an oligotrophic lake in the Patagonia region.

Acknowledgements

This work was accomplished with financial aid from the Universidad Nacional del Comahue (Project B143) and ANPCYT (Project PICT-1176). This work was also supported by the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) of Brazil. We would like to thank the authorities of Parques Nacionales (Argentina) for providing permission to collect water samples within the NHNP. Special thanks are given to V.d.G. for assistance in enzymes assays and A. Denegri for map design.

Ancillary