2003 Kluwer Academic Publishers. Printed in the Netherlands. Molecular characterization of carotenogenic yeasts from aquatic environments in Patagonia, ...
Antonie van Leeuwenhoek 84: 313–322, 2003. 2003 Kluwer Academic Publishers. Printed in the Netherlands.
313
Molecular characterization of carotenogenic yeasts from aquatic environments in Patagonia, Argentina ´ Diego Libkind 1, *, Silvia Brizzio 1 , Alejandra Ruffini 1 , Mario Gadanho 2 , Maria van 1 2 Broock and Jose´ Paulo Sampaio 1
´ Aplicada y Biotecnologıa ´ , Universidad Nacional del Comahue, Centro Laboratorio de Microbiologıa ´ Regional Universitario Bariloche ( CRUB) – CONICET ( Consejo Nacional de Investigaciones Cientıficas y ´ ´ Negro, Argentina; 2 Centro de Recursos Microbiologicos ´ Tecnologicas ), Quintral 1250, Bariloche, Rıo , ˜ Autonoma ˆ ´ Secc¸ao de Biotecnologia, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, 2829 -516 Caparica, Portugal; * Author for correspondence (e-mail: libkind@ crub.uncoma.edu.ar) Received 18 October 2002; accepted in revised form 3 April 2003
Key words: Aquatic environments, Carotenogenic yeasts, MSP-PCR fingerprinting, Patagonia, 26S rDNA sequence
Abstract Fifteen aquatic environments (lakes, lagoons and rivers) of glacial origin in the northern Andean Patagonia (Argentina) were surveyed for the occurrence of red yeasts. Subsurface water samples were filtered and used for colony counting and yeast isolation. A preliminary quantitative analysis indicated that total yeast counts ranged between 0 and 250 cells l 21 . A polyphasic approach including physiological and molecular methods was used for the identification of 64 carotenogenic yeast strains. The molecular characterisation of the isolates was based on the mini / microsatellite-primed PCR technique (MSP-PCR) employing the (GTG) 5 and the M13 primers. Comparison of representative fingerprints of each group with those of the type strains of pigmented yeasts allowed the expeditious identification of 87.5% isolates. The sequence analysis of the D1 / D2 domains of the 26S rDNA was employed to confirm identifications and in the characterization of the unidentified MSP-PCR groups. Teleomorphic yeast species were detected by performing sexual compatibility assays. The isolates corresponded to 6 genera and 15 yeast species, including four new yeast species of the genera Cryptococcus (1), Rhodotorula (1) and Sporobolomyces (2). Rhodotorula mucilaginosa was found in the majority of the samples and represented ca. 50% of the total number of isolates. However, this yeast was not detected in aquatic environments with very low anthropic influence. Other frequent yeast isolates were teleomorphic yeast species of Rhodosporidium babjevae, R. kratochvilovae and Sporidiobolus salmonicolor. This study represents the first report on red yeast occurrence and biodiversity in northwestern Patagonia.
Introduction Yeasts are common inhabitants of aquatic environments and their density and species diversity depend on the water type and purity (Hagler and Mendonc¸aHagler 1981). Carotenogenic yeasts are present in marine or freshwater oligotrophic habitats (Hagler and Mendonc¸a-Hagler 1981; Hagler and Ahearn 1987; Brizzio and van Broock 1998). These yeasts develop pink, orange or reddish colonies due to their ability to produce carotenoids or carotenoid-like com-
pounds (Barnett et al. 2000), which are of interest as vitamin precursors and effective antioxidants (Nelis and De Leenheer 1991). The industrial application of red yeasts for the production of carotenoids through biotechnological processes is currently under study (Buzzini 2000; Bhosale and Gadre 2001). The northern Andean Patagonia (Argentina) offers a great variety of glacially formed water bodies. They cover an ultra to mesotrophic range of small and large lakes including small high elevation lakes, sometimes ´ and Drago 1985; surrounded by dense forest (Quiros
314 ´ et al. 2000). A research project aiming at the Dıaz characterization of the yeast community of ultra to oligotrophic lakes of glacial origin is presently being carried out, and a report dealing with killer factor distribution has been published (Brizzio and van Broock 1998). Yeast characterization based on physiological tests is normally employed for ecological studies ´ ´ et al. 1992; Rosa et al. 1995; Slavikova ´ ´ (Slavikova and Vadkertiova´ 1997; Boguslawska-Was and Dabrowski 2001), but this approach is often misleading due to inconclusive results. In addition, teleomorphic states normally are not investigated. Molecular biology techniques, especially DNA sequencing, give more accurate identification results but are expensive and difficult to implement for large sets of isolates. In this study, we used a previously tested molecular protocol employing the mini / microsatellite-primed PCR method (MSP-PCR) (Gadanho and Sampaio 2002). This approach allowed the rapid characterization of a considerable number of isolates since species-specific profiles were obtained and compared with those of reference strains. Subsequently, DNA sequence analysis was employed for confirmation of MSP-PCR identifications and for the elucidation of inconclusive molecular fingerprints. Thus, a polyphasic approach is proposed to take forward the identification of yeast isolates from northwestern Patagonian aquatic environments employing molecular techniques, conventional assimilation tests, and sexual compatibility studies. A preliminary quantitative estimation of the yeast community of Patagonian lakes is included.
after collection) and sterile distilled water was used as control. Isolation and quantitative analysis Filter membranes (Millipore – 0.45 m m) were placed on the surface of YM agar plates (g l 21 , yeast extract 3; malt extract 3; peptone 5; dextrose 10; agar 20), pH 4, containing chloramphenicol 100 mg l 21 , and incubated at room temperature (20–22 8C) until mesophilic yeast colony emergence. CFU (colony forming units) of pigmented and non-pigmented yeasts, and yeast-like fungi was registered for preliminary quantitative analysis of yeast occurrence, and yeast cell counts mean values and standard errors were calculated. Red colonies were transferred to YM agar plates, pH 5.5, without antibiotic and preserved on YM slants at 4 8C. Physiological characterization and sexual compatibility A selection of physiological tests (assimilation of inositol and D-glucuronate as sole carbon sources; assimilation of nitrate as sole nitrogen source, and production of amyloid compounds – PAC-) was performed according to Yarrow (1998). For sexual compatibility studies, pairs of 2–4 days old cultures were crossed on SG agar (g l 21 , soytone 2; glucose 2; agar 15). After 1-week incubation at room temperature, the plates were examined upside down under the optical microscope at low magnification for production of mycelium and teliospores.
Methods PCR fingerprinting Area description and sampling Andean lakes are oligotrophic to ultra-oligotrophic ´ et al. temperate water bodies of glacial origin (Dıaz 2000) and have been classified as warm monomictic ´ and with a period of summer stratification (Quiros Drago 1985). These lakes are frequently surrounded by dense native forests of Nothofagus spp. and Austrocedrus chilensis trees. Water samples were collected from 15 aquatic reservoirs during 1997–2001, summer campaigns as described by Brizzio and van Broock (1998). In those water bodies for which quantitative analysis is reported, an average of three independent samples were taken per sampling site; they were kept at 4 8C until processed (within 48 h
DNA extraction was done according to Sampaio et al. (2001a) with slight modifications in order to reduce procedure times. Two loopfuls of MYP agar (g l 21 , malt extract 7; yeast extract 0.5; soytone 2.5 agar 1.5) grown cultures were suspended in Eppendorf tubes containing 500 m l of lysing buffer (50 mmol Tris l 21 ; 250 mmol NaCl l 21 ; 50 mmol EDTA l 21 ; 0.3%, w / v SDS; pH 8) and a volume of 200 m l of 425–600 m m glass beads. After vortexing for 3 min, the tubes were incubated for 1 h at 65 8C. After vortexing again for 3 min, the suspensions were centrifuged for 15 min at 4 8C. Finally, the collected supernatant was diluted 1:750, and 5 m l were used straight for PCR studies. The nucleic acids in the remaining supernatant
315 were precipitated with 1 / 10 volume of sodium acetate (3 M) and 2 volumes of absolute ethanol at 220 8C for 24 h. The DNA was washed twice with ethanol (70%, v / v), dried under vacuum, resuspended in 50 m l of TE (10 mmol Tris / HCl l 21 , pH 8, and 1 mmol EDTA l 21 ) and immediately placed at 220 8C. The synthetic oligonucleotide (GTG) 5 was used in microsatellite-primed PCR (MSP-PCR) experiments (Meyer et al. 1993). In some cases, the core sequence of the phage M13 (GAGGGTGGCGGTTCT) was used in additional MSP-PCR assays. PCR reactions were performed in 25 m l reaction volumes containing 1X PCR buffer (Pharmacia, Biotech), 2 mmol l 21 of each of the four dNTPs (Promega), 0.8 m mol l 21 primer, 10–15 ng genomic DNA and 1 U taq DNA polymerase (Pharmacia, Biotech). DNA amplification was performed in a Uno II Thermal Cycler (Biometra), consisting of an initial denaturing step at 95 8C for 5 min, followed by 40 cycles of 45 s at 93 8C, 60 s at 50 8C and 60 s at 72 8C, and a final extension step of 6 min at 72 8C. A negative control containing sterile distilled water instead of DNA, was also included in all PCR reactions. Amplified DNA fragments were separated by electrophoresis of 1.4% (w / v) agarose gels (GibcoBRL) in 0.5 X TBE (Tris-borate-EDTA) buffer at 90 V for 3.5 h, and stained with ethidium bromide. A molecular size marker was added on each gel as reference ( l DNA cleaved with HindIII and FX174 DNA cleaved with HaeIII – Pharmacia, Biotech). DNA banding patterns were visualised under a UV transilluminator and images were acquired with a Kodak Digital Science EDA 120 System and Kodak Digital Science 1D Image Analysis Software. DNA banding patterns were analysed using the GelCompar software package, version 4.1 (Applied Maths, Kortrijk). rDNA sequence analysis DNA was extracted using the methods described for PCR fingerprinting and amplified using primers ITS5 (59-GGA AGT AAA AGT CGT AAC AAG G) and LR6 (59-CGC CAG TTC TGC TTA CC-39). Cycle sequencing of the 600–650 bp region at the 59-end of the 26S rDNA D1 / D2 domain employed forward primer NL1 (59-GCA TAT CAA TAA GCG GAG GAA AAG) and reverse primer NL4 (59-GGT CCG TGT TTC AAG ACG G). Sequences were obtained with an Amersham Pharmacia ALF express II automated sequencer using standard protocols.
Alignments were made with MegAlign (DNAStar) and visually corrected. PAUP*4.0, version b8 (Swofford 2000), was used to perform phylogenetic analyses using the maximum parsimony method. Bootstrap analyses were based on 1000 random resamplings (Felsenstein 1985). GenBank accession numbers of the sequences determined in this study are given in Table 1–4.
Results and discussion Preliminary quantitative studies Water samples from seven aquatic reservoirs were submitted to a preliminary quantitative estimation of yeast densities and the total number of viable yeasts was registered as described above. As expected for oligotrophic freshwaters, cell counts were low, usually ranging from 0–150 cells l 21 and rarely exceeding 200 cells l 21 . The seven aquatic environments surveyed were preliminarily grouped into three classes (Figure 1). The first one corresponds to Nahuel Huapi Lake, the largest lake in the region, with the city of San Carlos de Bariloche located on its southeast coast. Twelve sites were sampled in an area of approximately 185 km 2 (33% of the area of the lake, Figure 1) and yielded variable yeast counts (5–250 cells l 21 ). The second class corresponds to the Manso glacial lagoon (reservoir formed after ice melting), the Mascardi lake and the Manso river which connects both water bodies (Figure 1). High numbers of viable yeast cells were observed in the water samples from Manso glacial lagoon and Manso river (81617 cells l 21 and 162.5617.5 cells l 21 respectively), but samples from the Mascardi lake showed the lowest cell counts (ca. 10.664.2 cells l 21 ), probably due to a dilution effect. The third group includes aquatic environments of relatively low anthropic influence.Viable yeast counts in Fonck, Hess, and Ortiz Basualdo lakes resulted in 166.6610.6 cells l 21 , 245.3635.9 cells l 21 , and2463.8 cells l 21 respectively. We believe that the high yeast counts could be due to a significant phylloplane run-off. Considering the studied sites, the Ortiz Basualdo lake had the lowest anthropic influence, due to difficult access. At this lake, low cell count values were observed and Sporobolomyces species seem to prevail. Rhodotorula mucilaginosa was not found in group 3 lakes, contrary to what was observed in groups 1 and 2. Many yeast biodiversity studies in aquatic environ-
316 Table 1. Combined MSP-PCR / sequence analysis grouping and identification of inositol negative, nitrate negative and amyloid compounds negative strains. Strain a
MSP-PCRb
GenBank c
(GTG) 5
M13
PYCC 5166 T CRUB 0027 CRUB 0032 CRUB 0047 CRUB 0048 CRUB 0049 CRUB 1026 CRUB 1030 CRUB 0120 CRUB 0124 CRUB 0142 CRUB 0134 CRUB 0154 CRUB 0527 CRUB 0023 CRUB 0072 CRUB 0018 CRUB 0064 CRUB 0069 CRUB 0056 CRUB 0057 CRUB 0058 CRUB 0022 CRUB 0045 CRUB 0046 CRUB 0028 CRUB 1049 CRUB 0034 CRUB 1031 CRUB 1021 CRUB 1022 CRUB 0138 CRUB 0526 CRUB 0029 CRUB 0149 CRUB 0150 CRUB 0151 CRUB 1027 PYCC 4709 T CRUB 0025 CRUB 0076 CRUB 1029 CRUB 1028 CRUB 1032
MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI *A *A *A *A *A *A *A *A *A *A *A MINU MINU MINU *B *C *D
MUCI ND ND ND MUCI ND MUCI ND ND ND ND MUCI MUCI ND ND ND ND ND ND ND ND ND ND MUCI ND ND ND MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MUCI MINU MINU MINU *B *C *D
AY158650
AY158651
AY158653 AY158652 AY158654
Species
Source
Rhodotorula mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. mucilaginosa Rh. minuta Rh. minuta Rh. minuta Rh. slooffiae Rh. pinicola Rhodotorula sp.
Unknown Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lagoon Senguer, Argentina Lake Mascardi, Argentina Lake Mascardi, Argentina Lake Escondido, Argentina Lake Escondido, Argentina Lake Escondido, Argentina Lake Gutierrez, Argentina Lake Gutierrez, Argentina Lake Gutierrez, Argentina ´ Lake Cantaros, Argentina ´ Lake Cantaros, Argentina ´ Lake Cantaros, Argentina River Valcheta, Argentina River Agrio, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Lake Traful, Argentina Lake Traful, Argentina Lake Toncek, Argentina Lagoon Senguer, Argentina River Valcheta, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lake Toncek, Argentina Lake Nahuel Huapi, Argentina Atmosphere, Japan Lake Mascardi, Argentina Lake Mascardi, Argentina Lake Nahuel Huapi, Argentina Lake Nahuel Huapi, Argentina Manso Glacial Lagoon, Argentina
a
PYCC, Portuguese Yeast Culture Collection, FCT-UNL, Portugal; CBS, Centraalbureau voor Schimmelcultures, Yeast Division, Utrecht, The Netherlands; CRUB, Centro Regional Universitario Bariloche, Bariloche, Argentina; T , type strain. b The designations ‘MUCI’ and ‘MINU’ indicate MSP-PCR fingerprints similar to those of the type strains of Rhodotorula mucilaginosa and Rh. minuta, respectively; ‘*A’ to ‘*D’ correspond to fingerprints that did not match those of the type strains available in our database at the time this study was made and therefore were not useful for identification; ‘ND’ not determined. c GenBank accession numbers (D1 / D2 domains of the 26S rDNA) are indicated for the strains that were investigated by sequence analysis.
ments reported red yeasts as normal components of the yeast community and at high numbers (.50% of the total yeast community) (Simard and Blackwood
´ ´ et 1971a, 1971b; Hagler and Ahearn 1987; Slavikova al. 1992). In this study, carotenogenic yeasts were present in almost all water samples, but their numbers
317 Table 2. Combined MSP-PCR / sequence analysis grouping and identification of inositol negative, nitrate positive and amyloid compounds negative strains. Strain a
PYCC 5168 T CRUB 1033 CRUB 1034 CRUB 1036 CRUB 1037 CRUB 1025 CRUB 1023 PYCC 4583 T CRUB 0121 CRUB 0127 CRUB 1035
MSP-PCRb
GenBank c
(GTG) 5
M13
BAB *E *E *E *E *E *E KRA *F *F *F
BAB BAB BAB BAB BAB BAB BAB KRA KRA KRA KRA
AY158645 AY158646 AY158647 AY158644
AY158649 AY158648
Species
Sex d
Species
Rhodosporidium babjevae R. babjevae R. babjevae R. babjevae R. babjevae R. babjevae R. babjevae R. kratochvilovae R. kratochvilovae R. kratochvilovae R. kratochvilovae
MT A1 SF SF ANA ANA ANA ANA SF MT A2 MT A2 MT A2
Herbaceous plant, Russia Verde Lagoon, Argentina Verde Lagoon, Argentina Lake Hess, Argentina Lake Fonck, Argentina Lake Nahuel Huapi, Arg. Lake Traful, Arg. Unknown Lake Toncek, Argentina Lake Toncek, Argentina Lake Escondido, Arg.
a PYCC, Portuguese Yeast Culture Collection, FCT-UNL, Portugal; CRUB, Centro Regional Universitario Bariloche, Bariloche, Argentina; T , type strain. b The designations ‘BAB’ and ‘KRA’ indicate MSP-PCR fingerprints similar to those of the type strains of Rhodosporidium babjevae and Rhodosporidium kratochvilovae, respectively; ‘*E’ and ‘*F’ correspond to fingerprints that did not match those of the type strains available in our database at the time this study was made and therefore were not useful for identification. c GenBank accession numbers (D1 / D2 domains of the 26S rDNA) are indicated for the strains that were investigated by sequence analysis. d Sex, sexuality; MT, mating type; SF, self-fertile; ANA, anamorphic.
varied significantly (1.0–60%). In order to assess the influence of biotic and abiotic features, as well as the impact of human settlings, in the total yeast counts and / or the prevalence of certain red yeast species, data concerning these parameters are currently being collected by us. Polyphasic characterization of the isolates A total of 64 red yeast isolates from 15 Patagonian aquatic environments were studied. Among the carotenogenic yeasts, the cultural and physiological characteristics of the isolates allowed the formation of
four groups: inositol negative (Ino 2), nitrate negative (Nit 2) and PAC negative (PAC 2) strains (Table 1, 65% of the total number of isolates); Ino 2 , Nit 1 , PAC 2 strains (Table 2, 14% of the total number of isolates); Ino 1 , Nit 1 , PAC 1 strains (Table 3, 5% of the total number of isolates) and ballistoconidiaproducing isolates (Table 4, 16% of the total number of isolates). Subsequently, MSP-PCR assays employing the microsatellite primer (GTG) 5 allowed a more detailed grouping and the identification of 50% of the total number of isolates under study. The taxa that were detected by direct comparison of DNA banding pat-
Table 3. Combined MSP-PCR / sequence analysis grouping and identification of inositol positive, nitrate positive and amyloid compounds positive strains. Strain a
PYCC 4418 T CRUB 1047 CRUB 1045 CRUB 1046 a
MSP-PCRb (GTG) 5
M13
CAP CAP *G *H
ND ND *G *H
GenBank c
Species
Sex d
Source
AY158643 AY158642
Cystofilobasidium capitatum Cyst. capitatum Cyst. infirmominiatum Cryptococcus sp.
SF SF MT A1 ANA
Zooplankton, Antarctic Lake Nahuel Huapi, Arg. Lake Fonck, Arg. Lake Mascardi, Arg.
PYCC, Portuguese Yeast Culture Collection, FCT-UNL, Portugal; CRUB, Centro Regional Universitario Bariloche, Bariloche, Argentina; T , type strain. b The designation ‘CAP’ indicates MSP-PCR fingerprints similar to those of the type strain of Cystofilobasidium capitatum; ‘*G’ and ‘*H’ correspond to fingerprints that did not match those of the type strains available in our database at the time this study was made and therefore were not useful for identification; ‘ND’ not determined. c GenBank accession numbers (D1 / D2 domains of the 26S rDNA) are indicated for the strains that were investigated by sequence analysis. d Sex, sexuality; MT, mating type; SF, self-fertile; ANA, anamorphic.
318 Table 4. Combined MSP-PCR / sequence analysis grouping and identification of ballistoconidia forming strains. Strain a
MSP-PCRb
T
PYCC 4111 CRUB 1039 CRUB 1051 CRUB 1052 CRUB 1053 PYCC 5678 T CRUB 1040 CRUB 1041 PYCC 4463 T CRUB 1042 CRUB 1043 CRUB 1038 CRUB 1044
(GTG) 5
M13
SAL *I *I *I *I RUB RUB RUB ROS ROS *J *J *K
SAL SAL SAL SAL SAL RUB RUB RUB ND ND *J *J *K
GenBank c
Species
Sex d
Source
AY158656 AY158655 AY158657
Sporidiobolus salmonicolor S. salmonicolor S. salmonicolor S. salmonicolor S. salmonicolor Sporobolomyces ruberrimus Sp. ruberrimus Sp. ruberrimus Sp. roseus Sp. roseus Sporobolomyces sp. 1043 Sporobolomyces sp. 1038 Sporobolomyces sp. 1044
MT A1 MT A1 ANA ANA ANA ANA ANA ANA ANA ANA ANA ANA ANA
Culture contaminant Lake Nahuel Huapi, Arg. River Agrio, Argentina River Agrio, Argentina River Agrio, Argentina Air, Japan Lake Ortiz Basualdo, Arg. Lake Ortiz Basualdo, Arg. Unknown Lake Ortiz Basualdo, Arg. Lake Hess, Argentina Lake Fonck, Argentina Lake Fonck, Argentina
a
PYCC, Portuguese Yeast Culture Collection, FCT-UNL, Portugal; CRUB, Centro Regional Universitario Bariloche, Bariloche, Argentina; T , type strain. b The designations ‘SAL’, ‘RUB’ and ‘ROS’ indicate MSP-PCR fingerprints similar to those of the type strain of Sporidiobolus salmonicolor, Sporobolomyces ruberrimus and Sp. roseus, respectively; ‘*I’ to ‘*K’ correspond to fingerprints that did not match those of the type strains available in our database at the time this study was made and therefore were not useful for identification; ‘ND’ not determined. c GenBank accession numbers (D1 / D2 domains of the 26S rDNA) are indicated for the strains that were investigated by sequence analysis. d Sex, sexuality; MT, mating type; SF, self-fertile; ANA, anamorphic.
Figure 1. Map showing the location of the three classes of aquatic environments studied. 1, Nahuel Huapi lake (the rectangle indicates the studied area); 2, Manso Glacial group; 3, Ortiz Basualdo, Fonck and Hess lakes.
terns of native yeast isolates with those of several type strains were Rhodotorula mucilaginosa, Rh. minuta, Sporobolomyces ruberrimus, Sp. roseus, and Cystofilobasidium capitatum (Table 1–4). Fingerprinting using primer M13 was subsequently performed and accounted for the identification of 37.5% of the yeast isolates (Rhodotorula mucilaginosa, Rhodosporidium babjevae, R. kratochvilovae and Sporidiobolus salmonicolor) (Table 1,2,4, Figure 2– 4), Therefore, MSP-PCR fingerprinting allowed the identification of ca. 90% of the total number of isolates. Recently, Herzberg et al. (2002) evaluated the RAPD-PCR approach for the characterization of yeasts isolated from the nectar of various plant species. However, due to the low reproducibility of this fingerprinting method and to the absence of a suitable reference database they were unable to employ this technique for yeast identification. Our present results, as well as previously published reports (Sampaio et al. 2001a, 2001b; Gadanho et al. 2001; Gadanho and Almeida J.M.G.C.F. Sampaio 2003; Gadanho and Sampaio 2002), clearly demonstrate the advantages of MSP-PCR over RAPD-PCR for that purpose. MSP-PCR analysis based on primer (GTG) 5 allowed the identification of the majority (70%) of the Rh. mucilaginosa isolates (Table 1, Figure 3A). The remaining isolates had (GTG) 5 profiles that could not
319
Figure 2. Yeast species identified with MSP-PCR. (A) Fingerprints obtained with the (GTG) 5 primer; (B) fingerprints obtained with the M13 primer. 1, Cystofilobasidium capitatum; 2, Rhodotorula minuta; 3, Sporobolomyces ruberriums; 4, Sporobolomyces roseus; 5, Rhodosporidium kratochvilovae; 6, Rhodosporidium babjevae; and 7, Sporidiobolus salmonicolor. M1, molecular size marker ( l DNA cleaved with HindIII and FX174 DNA cleaved with HaeIII); M2, molecular size marker 1 Kb Plus DNA ladder (GIBCO, BRL).
Figure 3. Selected MSP-PCR fingerprintings of Rhodotorula mucilaginosa. (A) Fingerprints obtained with the (GTG) 5 primer; (B) fingerprints obtained with the M13 primer. In (A), the arrowheads indicate fingerprints that did not match those of the type strain. M, molecular size marker ( l DNA cleaved with HindIII and FX174 DNA cleaved with HaeIII).
be associated with the type strain of Rh. mucilaginosa. However, they were assigned to this species based on the fingerprints obtained with primer M13 (Table 1, Figure 3B). The polymorphisms observed for Rh. mucilaginosa with the (GTG) 5 primer have to be investigated further to clarify their significance. As is typical for Rh. mucilaginosa, the assimilation of Dglucuronate was not observed in the MUCI group (Table 1). The identities of strains CRUB 1026 and CRUB 1027, the latter identified only with primer M13 (Figure 3), were confirmed by sequence analysis of the D1 / D2 domains of the 26S rDNA (Figure 4). Sequence analysis of the same domains was also applied when inconclusive results were obtained. This approach allowed the identification of Rhodotorula slooffiae (one substitution towards the type strain), Rh. pinicola, Cystofilobasidium infirmominiatum and the detection of four novel red yeast species, two in the genera Sporobolomyces (CRUB 1038 / CRUB 1043 and CRUB 1044), one in the genus Rhodotorula (CRUB 1032), and one in the genus Cryptococcus (CRUB 1046) (Figure 4). The two Sporobolomyces strains, CRUB 1038 and CRUB 1043, had four different nucleotides towards Sp. marcillae, their closest relative; Sporobolomyces CRUB 1044 showed 11 nucleotide differences in the D1 / D2 domains compared to its closest relative, Sp. blumae; Rhodotorula CRUB 1032 had 6 substitutions when compared with the type strain of Rh. lamellibrachii; and Cryptococcus CRUB 1046 had 8 different nucleotides in relation to the type strain of Cr. macerans.
320
Figure 4. Phylogenetic placement of selected Patagonian isolates obtained by maximum parsimony analysis of the D1 / D2 domains of the 26S rDNA. Bootstrap values higher than 50% are shown (1000 replicates). (A) Rhodosporidium /Rhodotorula glutinis clade (outgroup Rh. sonckii; the sequences of the type strains of Sporobolomyces japonicus and Sp. carnicolor were determined in this study and their GenBank accession numbers are AY158640 and AY158641, respectively); (B) Rhodotorula minuta clade (outgroup Naohidea sebacea; ‘Rh. cassicola’, ‘Rh. nymphaeae’ and ‘Rh. samaneae’ have not been validly described); (C) Cystofilobasidiales (outgroup Filobasidium floriforme); *, sequences determined in this study; abbreviations: Cr., Cryptococcus; Cyst., Cystofilobasidium; R., Rhodosporidium; Rh., Rhodotorula; S., Sporidiobolus; Sp., Sporobolomyces.
Another nine strains were selected for sequence analysis in order to confirm the MSP-PCR identifications. Their phylogenetic placement is also shown in Figure 4. The strain CRUB 1045 was identified by sequence analysis as Cystofilobasidium infirmominiatum. Since this strain produced true mycelium and teliospores when mated with strains PYCC 4413 and PYCC 4414 (mating types A3 and A2, respectively), but not when crossed with strain PYCC 3955 T (mating type A1), we concluded that strain CRUB 1045 belongs to mating type A1 of Cyst. infirmominiatum. We confirmed the occurrence of other teleomorphic species namely, R. kratochvilovae, R. babjevae, S. salmonicolor and Cyst. capitatum after sexual compatibility studies and microscopic investigations. Since some isolates of these species
failed to produce true mycelium and teliospores, they were regarded as anamorphic strains (Table 2–4). The Patagonian isolates were classified in 6 genera and 15 species. Thirty seven (58%) belong to the ubiquitous yeast species Rhodotorula mucilaginosa. Less frequent species were R. babjevae, R. kratochvilovae, S. salmonicolor, Sp. ruberrimus, Sp. roseus, Rh. minuta, Rh. slooffiae, Rh. pinicola, Cyst. capitatum, and Cyst. infirmominiatum. An observation made during this study, is that in lakes with low anthropic influence (Ortiz Basualdo, Fonck and Hess) ballistoconidial yeasts of the genus Sporobolomyces were present but Rh. mucilaginosa was absent. Rhodotorula glutinis is frequently considered a common species in freshwater environments (Simard and Blackwood 1971a, 1971b; Rosa et al. 1995;
321 ´ ´ and Vadkertiova´ 1997; Boguslawska-Was Slavikova and Dabrowski 2001). However, no isolates of this species were detected in our samples. The reported ubiquity of Rh. glutinis is probably due to incorrect identifications and to confusion with R. babjevae, as recently reported in Gadanho and Sampaio (2002). Another interesting remark corresponds to isolate CRUB 1028 identified as Rhodotorula pinicola on the basis of identical 26S rDNA D1 / D2 domain nucleotide sequences (Figure 4). Rhodotorula pinicola was originally found as the sole endophytic yeast in the xylem of the Chinese hard pine (Pinus tabulaeformis), in Beijing, China (Zhao et al. 2002). Patagonian native forests are mainly composed of Nothofagus spp. and Austrocedrus chilensis. Although some exotic pine species have been introduced, in the Nahuel Huapi area none has been recorded as Chinese hard pine (Dimitri 1982; Simberloff et al. 2002). In this study we report the isolation of one strain of Rh. pinicola from water samples from the Nahuel Huapi lake; its allochthonous or autochthonous nature remains unknown. Polyphasic approaches combining molecular and conventional methods are seldom used in yeast biodiversity studies. In this investigation, the MSPPCR technique proved to be a rapid, low cost and reliable method that allowed the accurate identification and characterization of ca. 90% of our yeast isolates. The 15 species found are distributed along the three main clades of pigmented species, the Rhodosporidium /Rhodotorula glutinis, Rhodotorula minuta and Cystofilobasidium lineages, indicating that oligotrophic Patagonian aquatic reservoirs with low human impact are attractive for wider yeast diversity studies. In addition, four non-described yeast species representing the three clades mentioned above were found. The formal description of these new species is under preparation.
Acknowledgements This work was accomplished with financial aid from the Universidad Nacional del Comahue and Consejo ´ Nacional de Investigaciones Cientıficas y Tecnolo´ gicas (CONICET) Argentina. M. Gadanho was supported by a grant SFRH / BD/ 1170 / 2000.
References Barnett J.A., Payne R.W. and Yarrow D. 2000. Yeasts: Characteris-
tics and Identification. 3rd edn. Cambridge University Press, Cambridge. Bhosale P. and Gadre R.V. 2001. Optimization of carotenoid production from hyper-producing Rhodotorula glutinis mutant 32 by a factorial approach. Lett. Appl. Microbiol. 33: 12–16. Boguslawska-Was E. and Dabrowski W. 2001. The seasonal variability of yeasts and yeast-like organisms in water and bottom sediment of the Szczecin Lagoon. Int. J. Hyg. Environ. Health 203: 451–458. Brizzio S. and van Broock M. 1998. Characterization of wild yeast killer from Nahuel Huapi National Park (Patagonia, Argentina. J. Food Technol. Biotechnol. 4: 273–278. Buzzini P. 2000. An optimization study of carotenoid production by Rhodotorula glutinis DBVPG 3853 from substrates containing concentrated rectified grape must as the sole carbohydrate source. J. Ind. Microbiol. Biotechnol. 24: 41–45. ´ M., Pedrozo F. and Baccala N. 2000. Summer classification of Dıaz Southern Hemisphere temperate lakes (Patagonia, Argentina. Lakes Reservoirs 5: 213–229. ´ ´ Dimitri M.J. 1982. Flora Dendrologica y Cultivada. In: La region ´ de los bosques andino patagonicos. Tomo II. Secretaria de ´ de la Nacion. ´ Instituto Nacional de Agricultura y Ganaderıa ´ Agropecuaria, Bs. As., Argentina. Tecnologıa Felsenstein J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783–791. Gadanho M. and Sampaio J.P. 2002. Polyphasic taxonomy of the basidiomycetous yeast genus Rhodotorula: Rh. glutinis sensu stricto and Rh. dairenensis comb. nov. FEMS Yeast Res. 2: 47–58. Gadanho M. and Almeida J.M.G.C.F. Sampaio J.P. 2003. Assessment of yeast diversity in a marine environment in the south of Portugal by microsatellite-primed PCR. Anton. Leeuw. Int. J. G. (in press). Gadanho M., Sampaio J.P. and Spencer-Martins I. 2001. Polyphasic taxonomy of the basidiomycetous yeast genus Rhodosporidium: R. azoricum sp. nov. Can. J. Microbiol. 47: 213–221. Hagler A.N. and Ahearn D.G. 1987. Ecology of aquatic yeasts. In: Rose A.H. and Harrison J.S. (eds), The Yeasts Vol. vol. 1. Academic Press, London, UK, pp. 181–205. Hagler A.N. and Mendonc¸a-Hagler L.C. 1981. Yeasts from marine and estuarine waters with different levels of pollution in the state of Rio de Janeiro, Brazil. Appl. Environ. Microbiol. 41: 173– 178. Herzberg M., Fischer R. and Titze A. 2002. Conflicting results obtained by RAPD-PCR and large-subunit rDNA sequences in determining and comparing yeast strains isolated from flowers: a comparison of two methods. Int. J. Syst. Evol. Microbiol. 52: 1423–1433. Meyer W., Mitchell T.G., Freedman E.Z. and Vilgalys R. 1993. Hybridization probes for conventional DNA fingerprinting used as single primers in the polymerase chain reaction to distinguish strains of Cryptococcus neoformans. J. Clin. Microbiol. 31: 2274–2280. Nelis H.J. and De Leenheer A.P. 1991. Microbial sources of carotenoid pigments used in foods and feeds. J. Appl. Bacteriol. 70: 181–191. ´ R. and Drago E. 1985. Relaciones entre variables fısicas, ´ Quiros ´ ´ ´ morfometricas y climaticas en lagos patagonicos. Rev. Asoc. Cs. Nat. Litoral 16: 181–199. Rosa C.A., Resende M.A., Barbosa F.A.R., Morais P.B. and Franzot S.R. 1995. Yeast diversity in a mesotrophic lake on the karstic
322 plateau of Lagoa Santa, MG–Brazil. Hydrobiologia 308: 103– 108. Sampaio J.P., Gadanho M., Santos S., Duarte F., Pais C., Fonseca A. et al. 2001a. Polyphasic taxonomy of the genus Rhodosporidium: R. kratochvilovae and related anamorphic species. Int. J. Syst. Evol. Microbiol. 51: 687–697. Sampaio J.P., Gadanho M. and Bauer R. 2001b. Taxonomic studies on the genus Cystofilobasidium: description of Cystofilobasidium ferigula sp. nov. and clarification of the status of Cystofilobasidium lari-marini. Int. J. Syst. Evol. Microbiol. 51: 221–229. Simard R.E. and Blackwood A.C. 1971a. Ecological studies on yeasts in the St. Lawrence River. Can. J. Microbiol. 17: 353– 357. Simard R.E. and Blackwood A.C. 1971b. Yeasts from the St. Lawrence River. Can. J. Microbiol. 17: 197–203. ˜ M. 2002. Gringos en el Simberloff D., Relva M.A. and Nunez bosque: introduced tree invasion in a natural Nothofagus /Austrocedrus forest. Biol. Invasions 4: 35–53.
´ ´ E. and Vadkertiova´ R. 1997. Seasonal occurrence of Slavikova yeasts and yeast-like organisms in the river Danube. Anton. Leeuw. Int. J. G. 72: 77–80. ´ ´ E., Vadkertiova´ R. and Kockova-Kratochvılova ´ ´ ´ A. 1992. Slavikova Yeasts isolated from artificial lake waters. Can. J. Microbiol. 38: 1206–1209. Swofford D.L. 2000. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Sinauer Associates, Sunderland, Massachusetts. Yarrow D. 1998. Methods for the isolation, maintenance and identification of yeasts. In: Kurtzman C.P. and Fell J.W. (eds), The Yeasts: A Taxonomic Study. Elsevier Science Publishers, Amsterdam, pp. 77–100. Zhao J.-H., Bai F.-Y., Guo L.-D. and Jia J.-H. 2002. Rhodotorula pinicola sp. nov., a basidiomycetous yeast species isolated from xylem of pine twigs. FEMS Yeast Res. 2: 159–163.
