• perennially ice covered lake;
  • Antarctica;
  • McMurdo Dry Valleys;
  • microbial assemblages;
  • phylogenetic diversity


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Lake Vida, located in the McMurdo Dry Valleys, Antarctica, is an ‘ice-sealed’ lake with ∼19 m of ice covering a highly saline water column (∼245 ppt). The lower portions of the ice cover and the lake beneath have been isolated from the atmosphere and land for circa 2800 years. Analysis of microbial assemblages within the perennial ice cover of the lake revealed a diverse array of bacteria and eukarya. Bacterial and eukaryal denaturing gradient gel electrophoresis phylotype profile similarities were low (<59%) between all of the depths compared (five depths spanning 11 m of the ice cover), with the greatest differences occurring between surface and deep ice. The majority of bacterial 16S rRNA gene sequences in the surface ice were related to Actinobacteria (42%) while Gammaproteobacteria (52%) dominated the deep ice community. Comparisons of assemblage composition suggest differences in ice habitability and organismal origin in the upper and lower portions of ice cover. Specifically, the upper ice cover microbiota likely reflect the modern day transport and colonization of biota from the terrestrial landscape, whereas assemblages in the deeper ice are more likely to be persistent remnant biota that originated from the ancient liquid water column of the lake that froze.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Many of the earth's perennially ice covered lakes are located in the McMurdo Dry Valleys, Antarctica (e.g. Lakes Bonney, Fryxell, Joyce, Miers, Vanda and Vida). These lakes are ice-covered (3–6 m on average) year round because of continuously cold air temperatures (annual average temperature −20°C, Wharton et al., 1993). The thickness of ice is typically controlled by sublimation off the top (∼30 cm year−1) and freezing on the bottom.

Several of the perennially ice covered lakes within the McMurdo Dry Valleys sustain biotic life. Microbiota have been described within the perennial ice covers of these lakes (Fritsen & Priscu, 1998; Priscu et al., 1998; Doran et al., 2003) and their water columns (Ward & Priscu, 1997; Voytek et al., 1999; Stackebrandt & Brambilla, 2002; Glatz et al., 2006). During warmer summers when air temperatures >0°C persist for several days, peripheral moats form around the edges of many of the ice covers (Wharton et al., 1993). The ice-free margins, often several meters wide, allow glacial melt water (formed during periodic melting of glacial ice) and associated biota to recharge the lakes. Aeolian transport deposits organisms directly onto the surface of the ice covers.

Lake Vida is an atypical perennially ice-covered lake in the McMurdo Dry Valleys with c. 19 m of ice covering a highly saline liquid brine (Doran et al., 2003). Organic matter locked within the Lake Vida ice at 12 m beneath the surface was radiocarbon dated at circa 2800 years old (Doran et al., 2003). Incoming glacial melt water during warm summer months periodically floods the frozen surface. This flooding was observed in the 1980s (C.P. McKay, pers. commun.) and again during the austral summer of 2001/2002 (J.C. Priscu, pers. commun.). The thickness of the Lake Vida ice cover presumably prevents the glacial melt water from flowing beneath the ice cover. Consequently, the periodic flooding and freezing of the Lake Vida ice cover increases the thickness of the ice from the top surface, unlike other perennially ice-covered lakes in the Dry Valleys. A net freezing of ∼7 cm year−1 also may occur on the bottom of the Vida ice cover (Doran et al., 2003). To date, microbial community composition within the Lake Vida ice cover or brine pocket had not been evaluated.

Biotic distribution and diversity were examined in the ice cover of Lake Vida to augment the paucity of data on microbial communities within perennial ice covers and to better understand the habitability of these ecosystems. Here, enumeration, diversity, and putative activity of microbial populations at five depths in the ice cover of Lake Vida are described. These results enable better resolution of the origin of microbiota in the Lake Vida ice cover and determination of whether the ice cover contains organisms indicative of the biota that may lie within the briny water column of the lake.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Study site

Lake Vida is located in Victoria Valley (77°23′S, 161°56′E) in the McMurdo Dry Valleys, Antarctica (Fig. 1). Annual precipitation in the McMurdo Dry Valleys ranges from 5 to 10 cm year−1, annual air temperature averages −20°C, and humidity is <50% (Matsumoto, 1993). Light levels fluctuate dramatically between the seasons, from continuous light in the summer to continuous darkness in the winter. The soils and many of the water bodies in the McMurdo Dry Valleys are saline due to concentration of marine salts derived from relict seawater in the region (Takamatsu et al., 1998), as well as evaporitic concentration of salts (Matsumoto, 1993).


Figure 1.  Location of Lake Vida, Antarctica. Landsat image of ice-covered Lake Vida located within the McMurdo Dry Valleys, courtesy of and adapted from NASA's Visible Earth (

Download figure to PowerPoint

Lake Vida is 3.5 km long and 1 km wide with a surface elevation of 390 m above sea level. The lake is covered year-round by ∼19 m of ice. Ice temperatures increase from c. −25°C at the surface of the ice cover to −10°C near the bottom (at 15.9 m, Doran et al., 2003). The brine pocket beneath the ice (detected by ground-penetrating radar) remains liquid at circa−10°C, which infers a salinity of >245 ppt (Doran et al., 2003). Chloride concentrations increase dramatically near the bottom of the ice (from 0.4 to 62.2 mg L−1 above 12.5 m to a striking 235.9 mg L−1 at 15.9 m), suggesting that this ice was originally derived from the brine layer (Doran et al., 2003). During warm summers, melt water from Victoria Upper and Lower Glaciers and smaller glaciers in the St Johns Range floods the frozen lake (Calkin & Bull, 1967). Doran et al. (2003) described five sediment layers, two sand layers and three microbial mat layers distributed within the ice cover above ∼12 m. Lake Vida has no surface outflows.

Sample collection and processing

Two ice cores (15.9 and 14 m) were collected from Lake Vida in 1996 using a Polar-Ice Coring Office 4-inch electromechanical ice-coring rig. Cores were kept frozen at −20°C until processing in 2003. Core segments containing enough volume of melt water for subsequent analyses were chosen to represent a range of depths within the portion of the ice cover that is frozen year-round (below ∼2 m). Five sections of the 15.9-m ice core were used for this study: 4.67–4.82, 6.7–6.86, 11.6–11.8, 14.47–14.65, and 15.5–15.85 m beneath the surface of the ice cover. Herein, these core segments are referred to by the bottom depth of each section: 4.8, 6.9, 11.8, 14.7, and 15.9 m.

Because of our intent to recover biota and DNA from the ice, core sections were handled aseptically in a UV and ethanol-sterilized laminar flow hood in a 4°C cold room. Core sections were cleaned by removing an average of 21.75% of the ice core mass by rinsing with sterile 3% hydrogen peroxide (H2O2). This was followed by a thorough rinse with sterile water to remove remnant H2O2. In an experiment designed to test the effectiveness of this cleaning method, pseudo-ice cores were intentionally contaminated with a bacterial isolate from a Lake Fryxell microbial mat and then cleaned with 3% H2O2. Most (99.2%) of the total contaminant cells (determined by microscopy) were removed from the cleaned pseudo-ice cores and 100% of the culturable cells (determined by viable cell counts) were removed. In a comparable experiment conducted with sterile 97% ethanol, cleaning removed 99.5% of the contaminant cells determined by total cell counts and only 75% those determined by viable cell counts (i.e. culturable cells).

Clean core sections were melted at 4°C in sterile Whirl-Pak® bags or HDPE containers. Ice-melt water was immediately subsampled for direct counts, chlorophyll a measurements, and biomass for nucleic acid extraction. Subsamples for direct cell counts were preserved with sterile-filtered glutaraldehyde (1% final concentration) and stored at 4°C until filtration within an average of 18 days (maximum of 47 days). Melt water (50–250 mL) was filtered onto 25 mm Whatman 0.4-μm pore size GF/F filters for chlorophyll a analysis and frozen at −80°C. Melt water (60–780 mL) for nucleic acid extraction was syringe-filtered through 0.2-μm pore size Supor (Gelman Scientific) sterivex filters. Filter cartridges were filled with 1.8 mL of sucrose lysis buffer (40 mM EDTA, 50 mM Tris-HCl, 0.75 M sucrose) and frozen at −80°C.

Chlorophyll a analysis, enumeration and sizing

Chlorophyll a from biomass collected on GF/F filters was extracted with cold 90% acetone overnight. Fluorescence (Fo) was determined with a Turner Designs 10AU Fluorometer calibrated with pure chlorophyll a (Sigma Corp.) (Welschmeyer, 1994). Direct bacterial counts were determined by epifluorescent microscopy (Hobbie et al., 1977) using the generalized nucleic acid DAPI (4′,6-diamidino-2-phenylindole) stain. Cell sizes (20 cells per depth) were determined with Image-Pro Plus (Media Cybernetics Inc.) image analysis software calibrated with fluorescent beads (Duke Scientific Corp.). Cell biovolume (μm3) was based on mean length and width of spheres (cocci) or cylinders with hemispherical caps (rods) (Simon & Azam, 1989).

Environmental genomic DNA extraction and PCR amplification

DNA extraction methods (Pospiech & Neumann, 1995; Massana et al., 1997; Sheridan et al., 2003) were tested for robustness by extracting nucleic acids from gram-positive cell cultures (Rhodococcus sp.). Additionally, the Massana et al. (1997) and Pospiech & Neumann (1995) methods were compared on a duplicate Vida ice section. The enzymatic and lysis method (Massana et al., 1997) had comparable nucleic acid recovery values, amplified better using PCR (brighter PCR bands), and produced more bacterial denaturing gradient gel electrophoresis (DGGE) bands on the duplicate ice sample. DNA from the Lake Vida ice sections was extracted using this method as previously described (Massana et al., 1997). Nucleic acids were quantified using the PicoGreen (Molecular Probes Inc.) dsDNA quantitation method.

PCR amplifications were carried out in an automated thermal cycler (Applied Biosystems). PCR mixtures contained 0.1–10 ng of template DNA, 1 × PCR reaction buffer, 0.2 mM dNTPs, 1 μM of each primer, 2.5–3.5 mM Mg2+ and 0.5 U Taq DNA polymerase. Initially, samples were screened with primers targeting the three domains of life using primers Bact27F (Lane, 1991) and Univ1492R (Kane et al., 1993) for bacteria, Arch20F (Massana et al., 1997) and Arch958R (DeLong, 1992) for archaea, and Euk1A (Diez et al., 2001) and Univ1492R for eukaryotes. Mixed template DNA was amplified for DGGE with bacterial (GC358F and 517R, Murray et al., 1996) and eukaryal (960FbGC and 1200R, Lin et al., 1996; Gast et al., 2004) primers with hot start conditions (Erlich et al., 1991). The thermal cycling program was similar to that described to Muyzer et al. (1993). In brief, the denaturing temperature began at 94°C and decreased 2°C every second cycle until annealing at 55°C for 20 cycles, followed by primer extension at 72°C for 7 min. Multiple PCR reactions were pooled (three reactions for eukarya and 8–11 reactions for bacteria) and purified with Microcon YM-100 concentrators (Millipore) for DGGE analysis. Bacterial clone libraries were generated by amplifying the 16S rRNA gene with 27F and 1492R primers. The conditions consisted of denaturation at 94°C for 5 min, 30 cycles of 94° for 1 min, 55°C for 1 min, and 72°C for 1 min, followed by extension at 72°C for 7 min.

Eukaryal and bacterial DGGE analysis

Eukaryal and bacterial DGGE analyses were performed using the BioRad D Code™ System and methods detailed by Murray et al. (1996). Equal amounts of PCR product (800 ng for bacteria and 300 ng for eukarya) were loaded onto 8% (w/v) polyacrylamide gels, 1 mm thick, and run in 1 × TAE (20 mM Tris acetate, 10 mM acetate, 0.5 mM disodium EDTA). The denaturing gradient contained 25–55% denaturant for the eukaryal DGGE and 30–60% denaturant for the bacterial DGGE (100% denaturant corresponds to 7 M urea and 40% v/v, formamide). Electrophoresis proceeded for 1000 V-h. Gels were stained with 1 × SYBR Gold (Molecular Probes) for 10 min, viewed on a UV-transilluminator, and photo-documented with a Kodak EDAS 290 System.

Gel images were analyzed using gelcompar II software (Applied Maths) in which bands were automatically detected and matched by the program. Band calling and matching was confirmed manually. In this study, DGGE was used to evaluate community changes between ice depths by analyzing shifts in banding patterns between samples. Pairwise comparisons of DGGE band similarities between samples were determined using Sorenson's index (Murray et al., 1996): Cs=2j/(a+b), where j is the number of bands common to two samples, and a and b are the number of bands in samples A and B, respectively.

Numerous eukaryal DGGE bands were selected for sequencing. Bands were excised from gels with a sterile surgical knife and placed in a 2-mL screw cap tube with 400 μL sterile distilled H2O and ∼200 mg sterile glass beads (Sigma). Acrylamide pieces were disrupted by vigorous shaking (5000 r.p.m. for 1 min) on a Mini Beadbeater and then the tubes were left at 4°C overnight. The gel-free DNA (5–10 μL) was reamplified as described previously with non-GC clamped primers (960Fb/1200R). The new PCR product was directly sequenced with the same forward and reverse primers on an Applied Biosystems ABI Prism 3730 DNA Analyzer (Nevada Genomics Center). Bands that resulted in high quality sequence data (15 in total) were matched via blast (Altschul et al., 1997) to the nearest identified sequence in the GenBank database.

16S rRNA gene clone library construction and sequencing

Clone libraries of nearly full-length bacterial 16S rRNA genes (27F and 1492R primers) were constructed for two depths in the ice cover: 4.8 and 15.9 m. PCR products were pooled from five individual reactions, purified, ligated into cloning vectors (TOPO TA cloning kit, Invitrogen), and transformed following the manufacturer's instructions. Clones containing inserts (as indicated by blue/white screening) were isolated (672 clones per depth).

From each library, 192 clones were sequenced unidirectionally. Twenty-three poor quality sequences (determined by ambiguous nucleotide identities) were removed from the analysis. All remaining sequences were checked for chimeras using the Bellerophon (Huber et al., 2004) and checkchimera programs (Cole et al., 2003): nine potential chimeras were identified and removed. An additional 37 short sequences (∼200 bp) were removed from the 4.8-m library. These sequences had the highest identities to cyanobacteria and were likely an effect of mispriming of the bacterial 27F primer. The short sequences aligned to the 3′ end of cyanobacterial 16S rRNA gene sequences. The remaining dataset contained 315 clone sequences ranging in length from 534 to 716 bp.

The frequency of unique phylotypes was determined by assembling sequences with >99% similarity using Sequencher (Gene Codes Corp.). Clone library phylotype richness (SChao1) and coverage (CACE) estimates were calculated using the web interface created by Kemp & Aller (2004a, b). SChao1 (Chao, 1984; Chao, 1987) is a nonparametric richness estimator and CACE (Chao et al., 1993; Lee & Chao, 1994) is a coverage estimate defined as the proportion of sequences from phylotypes with <10 clones that occur more than once in a library.

Near-full length, single-stranded 16S rRNA gene sequences were obtained for one representative clone from each unique phylotype. Phylotype sequences were matched via blast (Altschul et al., 1997) to the nearest identified sequence in the GenBank nr. and env_nt databases.

Phylogenetic analysis

Vida phylotypes with the highest blast identities to Actinobacteria and Gammaproteobacteria were analyzed for phylogenetic relatedness. Sequences were aligned (using Genetic Data Environment) to the secondary structure of neighboring 16S rRNA gene sequences downloaded from the Ribosomal Database Project. Aligned sequences were analyzed with the maximum likelihood method using fastDNAml (Felsenstein, 1981; Olsen et al., 1994) and a consensus tree was plotted. One hundred bootstraps were run using the fastDNAml_boot script. A mask excluded highly variable regions of the 16S rRNA gene from the analysis.

Nucleotide sequence accession numbers

Sequences generated in this study can be found under accession numbers DQ521452DQ521569 in GenBank.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

The core sections chosen for this study spanned the length of a 15.9-m ice core (collected in 1996) and contained variable amounts of sediments: high amounts of sediments at 4.8 m, some sediment and flaky mat-like material at 6.9 m, fine-grained sand at 11.8 m, clear ice with faint flocculent-like matter at 14.8 m, and clear ice at 15.9 m. Chlorophyll a concentrations ranged from 0.3 μg (at 4.8 m) to 1.9 μg (at 11.8 m) per liter of ice core melt water (Table 1). Chlorophyll a was very low or nondetectable at 6.9 and 15.9 m (chlorophyll a was not measured in the 14.7-m ice section due to a limited amount of melt water). Total cell concentrations recovered from the ice core decreased an order of magnitude from 2.09 × 106 mL−1 at 4.8 m to 0.12 × 106 mL−1 at 15.9 m (Table 1) and exhibit the highest concentrations in the ice sections with sediments. The average coefficient of variation among the grids counted (5–20) was 44.95, which is high but not uncommon for samples with sedimentary material. The average cell volume across all five depths was 0.06 μm3 cell−1 (n=100, SE 0.01, Table 1), with an average length of 0.62 μm (SE 0.02) and width of 0.35 μm (SE 0.01). Several large diatom frustules were observed at 4.8 and 11.8 m.

Table 1.   Chlorophyll a values, total prokaryotic cell counts, and cell volume estimates for the Lake Vida ice cover microbial community
Sample depth (m)Chlorophyll a (μg L−1)*Prokaryote count (cells mL−1)*Avg. biovolume (μm3 cell−1)
  • *

    Volume of ice melt water.

  • Not measured due to insufficient melt water volume. × 1060.043 × 1060.052 × 1060.028
14.7n/a0.54 × 1060.044 × 1060.083

Nucleic acid extraction values ranged from 0.37 to 10.16 ng DNA mL−1 ice melt water filtered. PCR screening indicated that bacterial and eukaryal DNA was present at all depths studied, whereas no archaeal DNA was amplified from the ice sections studied.

Eukaryal and bacterial DGGE

Eukaryal DGGE illustrated distinct banding patterns between the depths analyzed (supplementary Fig. S1). One universal phylotype was evident across all of the depths (denoted by the white arrow in supplementary Fig. S1). Phylotype profile pairwise similarities (supplementary Table S1) were low between all depths (similarity values <48%). The 15.9-m phylotype profile was the least similar (20%) to 4.8 m and most similar (48%) to 14.7 m. The 4.8-m profile was unlike the other depths analyzed (pairwise similarities between 20% and 29%).

Eukaryal DGGE band sequences were predominantly green-algal taxa (Table 2); chlorophyte-related sequences represented one-third of those determined and were present at all depths. Three fungal sequences were identified at 11.8 m. One diatom sequence (98% identity to Navicula atomus) was present in the 4.8-m profile. Eight of the band sequences were more than 99% identical to sequences in GenBank. As anticipated, comigrating bands (e.g. PT_ANTLV_31 and PT_ANTLV_11) had identical sequences.

Table 2.   Affiliation of eukaryal denaturing gradient gel band sequences in the Lake Vida ice cover (see supplementary Fig. S1)
Taxonomic group18S rRNA gene identification (accession no.)*% identityBand name (accession no.)Band no.Sample depth (m)
  • *

    Closest identified neighbor in the NCBI nr database.

  • From direct PCR amplification of acrylamide bands.

ApicomplexaColpodella edax (AY234843)99PT_ANTLV_26 (DQ521454)266.9
BacillariophytaNavicula atomus var. permitis (AJ867024)98PT_ANTLV_33 (DQ521452)334.8
CercozoaUncultured rhizosphere cercozoan (AJ506034)94PT_ANTLV_27 (DQ521455)276.9
ChlorophytaChlamydomonas incerta (AY781664)100PT_ANTLV_25 (DQ521463)2511.8
Chlorella sp. (AF514413)100PT_ANTLV_37 (DQ521457)374.8
Chlorella sp. Pic 9/21 P-1w (AY197630)100PT_ANTLV_31 (DQ521464)316.9
 100PT_ANTLV_11 (DQ521459)1115.9
Uncultured Chlorophyta (AY180047)100PT_ANTLV_05 (DQ521458)515.9
ChrysophyceaeUncultured chrysophyte (DQ104087)100PT_ANTLV_15 (DQ521453)1514.7
 100PT_ANTLV_06 (DQ521456)615.9
ChytridiomycotaRhizophydium sp. UGA-F1 (AF164265)95PT_ANTLV_22 (DQ521461)2211.8
Uncultured chytridiomycete (AF372719)98PT_ANTLV_14 (DQ521460)1411.8
CiliophoraFrontonia sp. (AF255359)97PT_ANTLV_34 (DQ521465)344.8
Paramecium calkinsi (AF100310)96PT_ANTLV_36 (DQ521466)364.8
UnidentifiedUncultured fungus (AY821997)97PT_ANTLV_23 (DQ521462)2311.8

As seen in the eukaryal DGGE, the bacterial DGGE also exhibited differences in phylotype distribution at each of the depths surveyed in the ice cover (supplementary Fig. S2). Four universal phylotypes were evident across all samples (denoted by the white arrows in supplementary Fig. S2). Pairwise similarities (supplementary Table S1) ranged from 32% to 59% across all pairwise depth comparisons. The 15.9-m bacterial phylotype profile was the least similar (36%) to 4.8 m and the most similar (59%) to 14.7 m.

16S rRNA gene sequences

Analysis of the 16S rRNA gene clone libraries included 134 sequences from 4.8 m and 181 from 15.9 m (>530 bp in length). The 4.8-m library had 66 phylotypes (defined by 99% sequence identity), 52 of which occurred once or twice in the library. The 15.9-m library had 37 phylotypes with 22 representatives occurring once or twice. There were only two cases in which phylotypes from both libraries shared 99% sequence similarity. The consensus sequence of these phylotypes had the highest blast identities to Gillisia limnaea (affiliated with the Flavobacteria, 98% identity, AJ440991) isolated from Antarctic microbial mats and an alpine lake Betaproteobacterium (99% identity, AJ867919).

The number of predicted phylotypes (SChao1) in the 4.8-m ice habitat was 133 (supplementary Fig. S3). The complete sequenced library represented ∼50% of the phylotypes predicted to exist in the environment. Coverage estimates (CACE) indicated that the proportion of sequences from relatively rare phylotypes (with <10 clones) that occurred more than once in the library was c. 63%. The 15.9-m library had substantially lower richness estimates (predicted SChao1 was 48) and higher coverage estimates (CACE was c. 83%). The library represented ∼75% of the number of phylotypes predicted to exist in the environment.

Unique phylotypes (near-full length) were grouped by taxonomic distribution based on the closest identified neighbor in GenBank (Table 3). Bacteria grouped within at least 10 different phyla at 4.8 m. Actinobacteria dominated the clone library from 4.8 m (42% of clone sequences). The library also contained many Bacteroidetes (13%), Gammaproteobacteria (10%) and Cyanobacteria (8%) sequences. Four phylotypes were most closely related to eukaryal plastid sequences including three Bacillariophytes and one Chlorophyte. Approximately 36% of the sequences had the highest identities to phylotypes isolated from polar or glacial environments and 5% had the highest identities to phylotypes from marine environments (sea ice, sea water or marine sediments).

Table 3.   Affiliation of 16S rRNA gene phylotypes at 4.8 and 15.9 m depths in the Lake Vida ice cover
Taxonomic phylumTaxonomic class16S rRNA gene identification (accession no.)*4.8 m15.9 m
Phylotype (clones)% identityPhylotype (clones)% identity
  • *

    Closest identified neighbor in either the NCBI nr or env_nt database.

  • Phylotype sequence lengths range from 1359 to 1556 bp and clone sequence lengths range from 534 to 716 bp.

  • Phylotype name (number of clone sequences in phylotype).

AcidobacteriaAcidobacterialesBacterium Ellin337 (AF498719)ANTLV1_C02 (1)96  
UnidentifiedUncultured Acidobacteria bacterium (AY211077)ANTLV1_D09 (1)94  
ActinobacteriaActinobacteridaeArctic sea ice bacterium ARK10173 (AF468440)  ANTLV7_C06 (1)99
Bacterium CS117 (AY124341)ANTLV1_H11 (3)99  
Cellulomonas sp. IFO16240 (AB023361)  ANTLV9_G12 (7)97
Cellulomonas sp. TR7-6 (AB166887)ANTLV2_A07 (3)98  
Cryobacterium psychrophilum (AJ544063)  ANTLV7_C10 (1)98
   ANTLV9_D07 (3)98
Curtobacterium sp. VKM Ac-2058 (AB042093)ANTLV1_D04 (1)96  
Frigoribacterium sp. 34/19 (AY571813)ANTLV2_A11 (2)97  
Janibacter-like sp. V4.BO.43 (AJ244674)ANTLV2_F12 (1)97  
 ANTLV2_G12 (4)97  
Leifsonia rubra (AJ438585)ANTLV1_B10 (1)96ANTLV7_E10 (3)99
 ANTLV1_H04 (22)97  
Luteococcus peritonei (AJ132334)ANTLV2_C06 (2)94  
UnidentifiedAntarctic bacterium R-8287 (AJ440992)ANTLV1_A09 (1)99ANTLV9_G02 (1)98
Uncultured actinobacterium (AF454303)ANTLV2_H06 (2)93ANTLV9_H09 (1)95
 ANTLV2_D07 (8)96  
Uncultured actinobacterium (AJ575506)ANTLV2_G03 (3)96  
Uncultured actinobacterium (AJ575519)  ANTLV7_H04 (2)99
Uncultured actinobacterium (AY743693)ANTLV2_H11 (1)96  
Uncultured bacterium ARFS-6 (AJ277690)ANTLV1_D11 (2)95  
Uncultured eubacterium WCHB1-18 (AF050574)  ANTLV9_C09 (1)96
BacteroidetesFlavobacteriaArctic sea ice bacterium ARK10063 (AF468410)  ANTLV7_G11 (7)99
Flavobacterium frigoris (AJ557887)ANTLV2_D10 (4)96  
Flavobacterium hibernum (L39067)ANTLV1_G11 (1)96  
Gillisia limnaea (AJ440991)ANTLV2_F09 (2)98ANTLV7_A10 (3)98
   ANTLV7_C07 (2)99
SphingobacteriaAlgoriphagus borotolerans (AB197852)ANTLV1_B03 (5)96  
Arctic sea ice bacterium ARK10280 (AF468433)  ANTLV9_G04 (14)98
Cyclobacterium amurskyense (AY960985)  ANTLV9_H08 (2)97
Glacier bacterium FJS5 (AY315161)ANTLV2_B07 (1)98  
UnidentifiedAntarctic bacterium R-7666 (AJ440984)ANTLV2_D11 (2)96  
Uncultured Bacteroidetes bacterium (AJ318142)ANTLV1_H05 (1)94  
Uncultured Bacteroidetes bacterium (AM114443)ANTLV2_H03 (1)97  
Uncultured Bacteroidetes bacterium (AY038797)ANTLV2_E05 (1)95  
Candidate Division OD1UnidentifiedUncultured candidate division OD1 bacterium (AY922093)  ANTLV9_G05 (1)82
Candidate Division TM7UnidentifiedUncultured candidate division TM7 bacterium (AF445701)  ANTLV9_C10 (6)93
   ANTLV9_F06 (2)93
   ANTLV9_G01 (1)92
   ANTLV9_G11 (9)93
ChloroflexiUnidentifiedUncultured Chloroflexi bacterium (AY921747)ANTLV1_C12 (1)88  
CyanobacteriaOscillatorialesLimnothrix redekei CCAP 1443/1 (AJ580007)ANTLV1_B04 (1)95  
 ANTLV1_H09 (1)98  
 ANTLV2_C09 (4)98  
Microcoleus vaginatus (AF284803)ANTLV1_A12 (1)97  
Tychonema bourrellyi (AB045897)ANTLV1_A05 (1)98  
UnidentifiedUncultured Antarctic cyanobacterium (AY151726)ANTLV2_A01 (2)97  
Uncultured Antarctic cyanobacterium (AY151728)ANTLV1_H06 (1)99  
FirmicutesBacillalesVirgibacillus necropolis (AJ315056)  ANTLV9_F07 (3)98
ClostridiaClostridium bowmanii (AJ506119)ANTLV1_E02 (1)99  
Uncultured Clostridiaceae bacterium (AY395437)ANTLV1_C11 (1)99  
GemmatimonadetesUnidentifiedUncultured Gemmatimonadetes bacterium (AY124345)ANTLV1_E09 (1)98  
PlanctomycetesPlanctomycetaciaGemmata-like str. CJuql4 (AF239693)ANTLV2_E09 (1)94  
Uncultured planctomycete (AJ616284)ANTLV2_C10 (1)98  
ProteobacteriaAlphaproteobacteriaAgrobacterium sanguineum (AB062105)ANTLV1_B09 (1)99  
Brevundimonas sp. FWC04 (AJ227793)ANTLV1_C06 (1)99  
 ANTLV1_E07 (2)99  
Catellibacterium nectariphilum (AB101543)ANTLV1_G08 (1)96  
Rhodobacter sp. 1-5 (AF513400)  ANTLV7_G07 (2)99
   ANTLV9_B12 (3)95
Sphingomonas sp. SIA181-1A1 (AF395032)ANTLV1_E12 (1)97  
 ANTLV1_H10 (2)97  
Uncultured Alphaproteobacterium (AY947933)ANTLV2_E06 (2)95  
BetaproteobacteriaAntarctic bacterium R-7724 (AJ440986)  ANTLV7_G03 (3)98
Betaproteobacterium Wuba72 (AF336361)ANTLV1_D07 (1)98ANTLV9_A06 (6)98
Leptothrix sp. S1.1 (DQ241397)ANTLV2_E04 (1)97  
Uncultured Betaproteobacterium (AJ867919)ANTLV1_B06 (1)99ANTLV7_H07 (1)99
Uncultured Betaproteobacterium (AY509483)ANTLV2_C11 (1)97  
Uncultured Betaproteobacterium (AY947997)ANTLV1_G06 (1)98  
EpsilonproteobacteriaUncultured Epsilonproteobacterium (AB015558)ANTLV2_G05 (1)89  
Uncultured Epsilonproteobacterium 1063 (AB030613)ANTLV2_G07 (1)97  
Unidentified Epsilonproteobacterium (AB015535)  ANTLV7_F08 (1)97
GammaproteobacteriaAntarctic bacterium R-9035 (AJ441002)  ANTLV9_C12 (4)99
Aquamonas fontana (AB120965)ANTLV1_A07 (2)97  
Arctic sea ice bacterium ARK10244 (AF468401)  ANTLV9_A11 (57)99
Gammaproteobacterium MN 154.3 (AJ313020)  ANTLV7_H08 (1)88
Gas vacuolate str. S36-W<gv>1 (U14584)  ANTLV7_A07 (23)98
   ANTLV9_B08 (2)98
Marinobacter maritimus (AJ704395)  ANTLV7_D09 (2)97
Pseudomonas sp. ARCTIC-P23 (AY573032)  ANTLV7_H01 (1)98
Pseudomonas sp. ARCTIC-P33 (AY573033)  ANTLV9_D02 (1)96
Stenotrophomonas maltophilia (AJ131114)ANTLV1_F11 (1)99  
Uncultured Gammaproteobacterium (AF445683)ANTLV2_E01 (3)98  
 ANTLV2_E03 (1)97  
 ANTLV2_F07 (3)94  
Uncultured Gammaproteobacterium (AJ704666)  ANTLV7_H05 (1)93
Uncultured Gammaproteobacterium(AY922114)ANTLV1_A02 (3)96  
Uncultured Gammaproteobacterium (DQ234120)  ANTLV9_C04 (1)97
Uncultured Gammaproteobacterium Sva0318 (AJ240989)  ANTLV9_H04 (2)95
Uncultured Xanthomonas sp. (AY571839)ANTLV1_C08 (1)99  
UnidentifiedUncultured proteobacterium (AF491673)ANTLV2_G10 (1)90  
VerrucomicrobiaUnidentifiedUncultured Verrucomicrobia bacterium (AF465651)ANTLV1_D01 (1)94  
Eukaryotic plastidsBacillariophytaThalassiosira pseudonana mitochondrion (DQ186202)ANTLV2_A04 (1)87  
Bacillaria paxillifer chloroplast 16S rRNA gene (AJ536452)ANTLV1_E03 (3)97  
 ANTLV2_F08 (3)97  
ChlorophytaChlorella saccharophila chloroplast 16S rRNA gene (D11348)ANTLV1_D06 (1)98  

Sequences in the 15.9-m library grouped within six different bacterial phyla. The library was dominated by Gammaproteobacteria (52% of clone sequences). The library also contained many Bacteroidetes (15%), Actinobacteria (11%), and sequences from the candidate division TM7 (10%). Approximately 72% of the sequences from the 15.9-m library had the highest identities to phylotypes isolated from polar or glacial environments, including Arctic and Antarctic sea ice (Gosink & Staley, 1995; Brinkmeyer et al., 2003), Antarctic microbial mats (Van Trappen et al., 2002), and an Antarctic pond (Reddy et al., 2003). Roughly 65% of the sequences had the highest identities to phylotypes from marine environments.

Gammaproteobacteria and Actinobacteria phylogeny

Phylogenetic trees were constructed from Lake Vida ice-associated Gammaproteobacteria and Actinobacteria 16S rRNA gene sequences based on maximum likelihood analysis. Many 15.9-m ice Gammaproteobacteria phylogenetically clustered with sequences from Antarctic and Arctic environments (Fig. 2), including an Antarctic Marinobacter isolate from Lake Bonney that is psychrophilic, moderately halophilic, and capable of denitrification (Marinobacter sp. ELB17, Ward & Priscu, 1997). Sequences from the 4.8-m library were phylogenetically distinct from the 15.9-m sequences and were more closely related to Xanthomonas and Stenotrophomonas sequences.


Figure 2.  Maximum likelihood phylogenetic tree showing the affiliation of Lake Vida Gammaproteobacteria 16S rRNA gene sequences. Phylotypes from Lake Vida are indicated with open circles for the 4.8-m library and closed circles for the 15.9-m library. Bootstrap values greater than 50 are shown.

Download figure to PowerPoint

The Lake Vida ice-associated actinobacterial 16S rRNA gene sequences grouped with a broad range of distantly related clusters (Fig. 3). Phylotypes from the 4.8- and 15.9-m libraries clustered closely with each other in several cases. Many of the Actinobacteria from Lake Vida ice shared close relationships with other Arctic and Antarctic organisms, including organisms associated with Antarctic microbial mats (from Lake Fryxell and Grace Lake) and polar sea ice. Several phylotypes were phylogenetically related to the Antarctic psychrophile Leifsonia rubra (AJ438585, Reddy et al., 2003), including the three phylotypes (representing 26 clones) with L. rubra as their nearest identified neighbor in GenBank. This group also contained the marine actinobacterium PHSC20C (Murray et al., 2006) and represents a cold-adapted lineage found in both marine and freshwater environments.


Figure 3.  Maximum likelihood phylogenetic tree showing the affiliation of Lake Vida Actinobacteria 16S rRNA gene sequences. Phylotypes from Lake Vida are indicated with open circles for the 4.8-m library and closed circles for the 15.9-m library. Bootstrap values greater than 50 are shown.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Distribution of Lake Vida biota

Bacterial numbers in the Lake Vida ice were comparable to data obtained from other studies of microorganisms in ice. Total bacterial cell counts in the Vida ice core (0.12 × 106–2 × 106 cells mL−1) are comparable to those found in many polar environments, including cryoconite holes (C.H. Fritsen, pers. commun.), Arctic sea ice (e.g. Junge et al., 2002; Brinkmeyer et al., 2003), Antarctic sea ice (e.g. Helmke & Weyland, 1995; Gowing et al., 2004), and the ice covers and water columns (Ward & Priscu, 1997; Priscu et al., 1998, respectively) of perennially ice-covered lakes. The Lake Vida total bacterial cell counts were orders of magnitude higher than bacterial numbers (102–103) in Lake Vostok accretion ice (Karl et al., 1999; Priscu et al., 1999), which is covered by several thousand meters of glacial ice. Lake Vida ice and Greenland ice (Sheridan et al., 2003; Miteva & Brenchley, 2005) cells were similar in size (<0.1 μm3). These small cell sizes may be an adaptation to stresses induced within the ice (Fogg, 1998), including nutrient limitation via reductive cell division or adaptation to oligotrophy.

Putative activity

The 16S rRNA gene similarity of many Vida sequences (178) and isolates (e.g. seven that are related to Cryobacterium psychrophilum; data not shown) to organisms isolated from polar or glacial environments is not direct evidence for in situ activity. However, a previous study reported detectable primary production and thymidine incorporation in ice melt water throughout a 15-m Lake Vida ice core with the highest rates in the upper 4 m (Doran et al., 2003). The abundance of chlorophyll a at 4.8 m and presence of cyanobacteria at 4.8 and 6.9 m (as evidenced by representation in the 4.8-m clone library and culturing; data not shown) supports the potential for photoautotrophic primary production above 6.9 m in the ice cover. In ice covers, photosynthetically available radiation is attenuated by high gas bubble density and entrained sediment (McKay et al., 1994; Fritsen & Priscu, 1999). This does not necessarily mean that the phototrophic cells are inactive. Light saturation of photosynthesis occurs at low irradiances (30 μmol photons m−2 s−3) within Lake Bonney (Neale & Priscu, 1998) and phytoplankton exhibit mixotrophy in Lake Fryxell and Lake Hoare (Roberts & Laybourn-Parry, 1999; Marshall & Laybourn-Parry, 2002). Thus, phototrophs in Lake Vida ice may photosynthesize below 6 m in low light conditions or may exhibit heterotrophy.

Bacteroidetes was one of the most abundant groups in both of the Lake Vida ice clone libraries and may be major contributors of heterotrophic activity in the ice cover. Bacteroidetes play an important role in remineralization processes in aquatic systems (Kirchman, 2002) and may also be important in ice systems. Some Bacteroidetes can utilize cellulose, chitin, DNA, lipids, or proteins that would be trapped within the ice after cell decay. Many of the Vida sequences had nearest identified neighbors to Bacteroidetes from polar environments, including the psychrotroph Flavobacterium hibernum isolated from a freshwater lake of the Vestfold Hills region of Antarctica (Crooked Lake; McCammon et al., 1998). Flavobacterium hibernum may have a bacteriovorous role in the environment because of its ability to glide on surfaces and to lyse gram-negative cells (McCammon et al., 1998).

Diversity of biota found within ice

The variety of eukaryal phylotypes in the bottom portions of the ice cover was somewhat unanticipated. Microscopic enumeration of eukaryotes was not performed due to the low availability of ice core melt water, though a few eukaryal cells were observed during prokaryotic cell counts. It is uncertain whether the amplified eukaryal DNA originated from intact cells or was free within the ice. Samples for nucleic acid extraction were collected on 0.2-μm filters, suggesting that the cellular, and not dissolved, fraction was analyzed. If the recovered eukaryal DNA originated from intact cells, these organisms may have survival mechanisms for withstanding cryogeny. Some fungi, for instance, are stress tolerant and have lower minimal water activities for growth than prokaryotes (Onofri et al., 2004). Many Antarctic fungi have shown resistance to low temperatures, high UV, high ionizing radiation, and desiccation (Onofri et al., 2004). Alternatively, eukaryal DNA may have been recovered from intact cells that were not alive but instead lifeless and frozen within the ice matrix. Endogenous nucleases that normally degrade DNA after a cell dies can be destroyed or inactivated by low temperatures, high salinity, and rapid desiccation (Hofreiter et al., 2001), thereby slowing the rate of DNA decay within ice.

Although bacteria and eukarya are prevalent, archaea are not commonly detected in ice environments. Archaeal 16S rRNA gene primers did not amplify rRNA gene from Vostok accretion ice (Lyons et al., 1998; Christner et al., 2001), ancient glacial ice (Christner et al., 2003b), Arctic and Antarctic sea ice (Brown & Bowman, 2001), and Vida ice. However, archaea have been documented in young, seasonal Antarctic and Arctic sea ice (Stewart et al., 2003; Junge et al., 2004). There are several suppositions explaining the apparent absence of archaea in older ice, including the possibility that archaea are out-competed by other prokaryotes in the ice or that the archaea are more prone to death and decay once entombed within the ice.

Our results indicate that the 4.8-m bacterial assemblage was more diverse than the 15.9-m assemblage with no 100% similar phylotypes found at both depths. We found that the sensitivity of detection with DGGE is somewhat limited in rich assemblages with many rare phylotypes (e.g. the 4.8-m ice) and is more robust in a less rich and more evenly distributed assemblage such as that seen in the 15.9-m ice. To the best of our knowledge, richness has not been calculated from other ice environment's including cryoconites, glacial ice, permanent lake ice, and sea ice. Richness estimates calculated for the water column of Lake Bonney based on unique restriction fragment length polymorphism (RFLP) patterns (20–392 phylotypes; Glatz et al., 2006) were within the range seen in the Vida ice libraries. Richness estimates were also calculated for low temperature marine sediments: 541-1128 phylotypes in Antarctic sediments from the continental shelf (Bowman & McCuaig, 2003) and 2716 phylotypes in Arctic marine sediments (Kemp & Aller, 2004a, b). As expected, richness for the Vida libraries is low in comparison with sediment environments.

PCR biases (Reysenbach et al., 1992; Suzuki & Giovannoni, 1996; Suzuki et al., 1998) and multiple rRNA gene copies per cell (Klappenbach et al., 2000) can affect the distribution and relative frequency of phylotypes within clone libraries. Therefore, the abundance of Gammaproteobacteria (c. 50% of the sequences) in the 15.9-m Vida clone library elicits guarded interpretation (as does the relative abundance data for both libraries). It is also important to note that the Vida 15.9-m library was estimated to have a high percent coverage of the natural community – roughly 70% of the community was described with only 50 clone sequences. Though the Vida 16S rRNA gene clone libraries are potentially subject to biases, Gammaproteobacteria were also abundant in several other polar studies, including Arctic (Junge et al., 2002; Brinkmeyer et al., 2003) and Antarctic sea ice (Bowman et al., 1997; Brinkmeyer et al., 2003) and the water column of Lake Bonney (Glatz et al., 2006). These data suggest that certain Gammaproteobacteria (e.g. Colwellia and Marinobacter) are common among these habitats.

Bacteria isolated from ice cores have likely endured a number of environmental stresses (including freezing, desiccation, and starvation). Therefore, spore-forming bacteria are expected to be present in large numbers in ice environments. Interestingly, the Vida ice clone libraries had only a few representatives of known endospore-forming genera (two phylotypes at 4.8 m and one at 15.9 m). However, both of the libraries contained many Actinobacteria-related (high G+C, gram positives) sequences (∼25% of unique phylotypes detected in each library). Similar trends were also seen in a Greenland glacier (Sheridan et al., 2003; Miteva et al., 2004) and an Antarctic cryoconite hole (Christner et al., 2003a). Willerslev et al. (2004) showed that Actinobacteria outlast low G+C gram-positive and gram-negative cells in old permafrost, shifting from c. 30% to 100% of the taxonomic distribution of clones after 300 000–600 000 years. They concluded that this group has survival mechanisms that are not fully understood at this point but which may include transition to resting forms with low level metabolic activity (i.e. anabiosis; Mulyukin et al., 2001). The 15.9-m library also had many representatives of candidate division TM7, which contain gram-positive cell envelopes (Hugenholtz et al., 2001).

Source and origin

While a few taxa may be ubiquitous in the Lake Vida ice cover (e.g. Chlorella sp. as evident by the universal eukaryal DGGE band), the low DGGE pairwise similarities between the depths and the depth-specific phylotypes are indicative of differences in ice habitability (i.e. the ability of the physical-chemical environment within the ice to sustain life) or organismal origin in the upper and lower portions of ice cover. The source of the organisms in the upper ice is thought to be a combination of modern aeolian deposition and glacial melt water that flows through streams and sediments before flooding the ice cover. Thus, it is not surprising that the biotic community in the 4.8-m ice contains many bacteria and eukarya that have high similarities to organisms from soil and freshwater habitats.

The bottom of the Vida ice cover is likely to have fluctuated through periods of melting and freezing over the 2800-year period of its presumed existence. Cycling of the thickness and mass of the ice cover is likely controlled by winter and summer fluctuations that determine the net annual mean temperature at the bottom of the ice cover. Despite the present-day seasonal fluctuations, thermodynamic modeling suggests a net annual accretion of ice (Doran et al., 2003). Net accretions at the bottom further imply that ice near the bottom is expected to be of lake water origin. The significant increase in chloride concentrations in the 15.9-m ice (Doran et al., 2003) supports the notion that this ice formed at or in the transition from a freshwater water column to one with higher dissolved salts (Doran et al., 2003). If the 15.9-m ice did indeed form from the water column of the lake, the biota found in the deep ice may also be present in the lake itself, including a diverse array of bacteria and eukaryotes. The high percentage (∼65%) of clone sequences in the 15.9-m library with the highest identities to sequences from marine environments (compared to only ∼5% at 4.8 m) suggests that these organisms may be able to sustain activity or viability within the high salinity of the water column.

Christner et al. (2001) suggest that organisms found in the Lake Vostok (Antarctica) accretion ice regularly seed the water column of the lake. Similarly, the organisms found at the bottom of the Vida ice cover may seed the water column during periods of ice melt. The population of microorganisms existing within the deep ice has the potential to seed the underlying brine pocket of Lake Vida during periods of bottom-ice ablation. Thus, a dynamic exchange of biota may occur between the water column and deep ice during melting and freezing cycles at the bottom of the ice cover, where the lake water and ice are both seeds and sinks.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

A diverse range of bacteria and eukarya are present at all depths in the Lake Vida ice cover. The conundrum of whether these organisms are active within the ice has not been solved. The data suggest that there are differences in ice habitability or organismal origin in the upper and lower portions of the ice cover. We suggest that the 4.8-m ice community has a terrestrial origin (e.g. McMurdo Dry Valleys soil and freshwater) and that the 15.9-m community has likely been derived from the water column. A dynamic exchange of biota may occur between the water column and deep ice when the bottom of the ice cover melts or freezes. The Lake Vida water column may be a refuge from the harsh environmental conditions within the McMurdo Dry Valleys.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

This work was supported by: Nevada Space Grant Consortium, Desert Research Institute Research Foundation, NASA ASTEP (Grant NAG5-12889), NASA EPSCoR (Grant NCC5-583), NSF LExEN (Grant OPP-9814972), and NSF Office of Polar Programs. We would like to thank B. Carter and J. Grzymski for their assistance during stages of this research as well as P. Doran and J. Priscu for providing the initial efforts in obtaining the Lake Vida ice core.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller W & Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402.
  • Bowman JP & McCuaig RD (2003) Biodiversity, community structural shifts, and biogeography of prokaryotes within Antarctic continental shelf sediment. Appl Environ Microbiol 69: 24632483.
  • Bowman JP, McCammon SA, Brown MV, Nichols DS & McMeekin TA (1997) Diversity and association of psychrophilic bacteria in Antarctic sea ice. Appl Environ Microbiol 63: 30683078.
  • Brinkmeyer R, Knittel K, Jurgens J, Weyland H, Amann R & Helmke E (2003) Diversity and structure of bacterial communities in arctic versus antarctic pack ice. Appl Environ Microbiol 69: 66106619.
  • Brown MV & Bowman JP (2001) A molecular phylogenetic survey of sea-ice microbial communities (SIMCO). FEMS Microbiol Ecol 35: 267275.
  • Calkin PE & Bull C (1967) Lake Vida, Victoria Valley Antarctica. J Glaciol 6: 833836.
  • Chao A (1984) Nonparametric-estimation of the number of classes in a population. Scand J Stat 11: 265270.
  • Chao A (1987) Estimating the population-size for capture recapture data with unequal catchability. Biometrics 43: 783791.
  • Chao A, Ma MC & Yang MCK (1993) Stopping rules and estimation for recapture debugging with unequal failure rates. Biometrika 80: 193201.
  • Christner BC, Mosley-Thompson E, Thompson LG & Reeve JN (2001) Isolation of bacteria and 16S rDNA's from Lake Vostok accretion ice. Environ Microbiol 3: 570577.
  • Christner BC, Kvitko BH & Reeve JN (2003a) Molecular identification of Bacteria and Eukarya inhabiting an Antarctic cryoconite hole. Extremophiles 7: 117183.
  • Christner BC, Mosley-Thompson E, Thompson LG & Reeve JN (2003b) Bacterial recovery from ancient glacial ice. Environ Microbiol 5: 433436.
  • Cole JR, Chai B, Marsh TL et al. (2003) The Ribosomal Database Project (RDP-II): previewing a new autoaligner that allows regular updates and the new prokaryotic taxonomy. Nucleic Acids Res 31: 442443.
  • DeLong EF (1992) Archaea in coastal marine environments. Proc Natl Acad Sci 89: 56855689.
  • Diez B, Pedros-Alio C, Marsh T & Massana R (2001) Application of denaturing gradient gel electrophoresis (DGGE) to study the diversity of marine picoeukaryotic assemblages and comparison of DGGE with other molecular techniques. Appl Environ Microbiol 67: 29422951.
  • Doran PT, Fritsen CF, McKay CP, Priscu JC & Adams EE (2003) Formation and character of an ancient 19-m ice cover and underlying trapped brine in an “ice-sealed” east Antarctic lake. Proc Natl Acad Sci 100: 2631.
  • Erlich HA, Gelfand D & Sninsky JJ (1991) Recent advances in the polymerase chain-reaction. Science 252: 16431651.
  • Felsenstein J (1981) Evolutionary trees from DNA sequences: a maximum likelihood approach. J Mol Evol 17: 368376.
  • Fogg GE (1998) The Biology of Polar Habitats. Oxford University Press, New York.
  • Fritsen CH & Priscu JC (1998) Cyanobacterial assemblages in permanent ice covers on Antarctic lakes: distribution, growth rate, and temperature response of photosynthesis. J Phycol 34: 587597.
  • Fritsen CH & Priscu JC (1999) Seasonal change in the optical properties of the permanent ice cover on Lake Bonney, Antarctica: consequences for lake productivity and phytoplankton dynamics. Limnol Oceanogr 44: 447454.
  • Gast RJ, Dennett MR & Caron DA (2004) Characterization of protistan assemblages in the Ross Sea, Antarctica, by denaturing gradient gel electrophoresis. Appl Environ Microbiol 70: 20282037.
  • Glatz RE, Lepp PW, Ward BB & Francis CA (2006) Planktonic microbial community composition across steep physical/chemical gradients in permanently ice-coverd Lake Bonney, Antarctica. Geobiology 4: 5367.
  • Gosink JJ & Staley JT (1995) Biodiversity of gas vacuolate bacteria from Antarctic sea-ice and water. Appl Environ Microbiol 61: 34863489.
  • Gowing MM, Garrison DL, Gibson AH, Krupp JM, Jeffries MO & Fritsen CH (2004) Bacterial and viral abundance in Ross Sea summer pack ice communities. Mar Ecol Prog Ser 279: 312.
  • Helmke E & Weyland H (1995) Bacteria in sea-ice and underlying water of the eastern Weddell Sea in midwinter. Mar Ecol Prog Ser 117: 269287.
  • Hobbie JE, Daley RJ & Jasper S (1977) Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl Environ Microbiol 33: 12251228.
  • Hofreiter M, Serre D, Poinar HN, Kuch M & Paabo S (2001) Ancient DNA. Nat Rev Genet 2: 353359.
  • Huber T, Faulkner G & Hugenholtz P (2004) Bellerophon: a program to detect chimeric sequences in multiple sequence alignments. Bioinformatics 20: 23172319.
  • Hugenholtz P, Tyson GW, Webb RI, Wagner AM & Blackall LL (2001) Investigation of candidate division TM7, a recently recognized major lineage of the domain bacteria with no known pure-culture representatives. Appl Environ Microbiol 67: 411419.
  • Junge K, Imhoff F, Staley T & Deming JW (2002) Phylogenetic diversity of numerically important Arctic sea-ice bacteria at subzero temperature. Microb Ecol 43: 315328.
  • Junge K, Eicken H & Deming JW (2004) Bacterial activity at −2 to −20 degrees C in Arctic wintertime sea ice. Appl Environ Microbiol 70: 550557.
  • Kane MD, Poulsen LK & Stahl DA (1993) Monitoring the enrichment and isolation of sulfate-reducing bacteria by using oligonucleotide hybridization probes designed from environmentally derived 16S rRNA sequences. Appl Environ Microbiol 59: 682686.
  • Karl DM, Bird DF, Bjorkman K, Houlihan T, Shackelford R & Tupas L (1999) Microorganisms in the accreted ice of Lake Vostok, Antarctica. Science 286: 21442147.
  • Kemp PF & Aller JY (2004a) Bacterial diversity in aquatic and other environments: what 16S rDNA libraries can tell us. FEMS Microbiol Ecol 47: 161177.
  • Kemp PF & Aller JY (2004b) Estimating prokaryotic diversity: When are 16S rDNA libraries large enough? Limnol Oceanogr: Methods 2: 114125.
  • Kirchman DL (2002) The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol Ecol 39: 91.
  • Klappenbach JA, Dunbar JM & Schmidt TM (2000) rRNA operon copy number reflects ecological strategies of bacteria. Appl Environ Microbiol 66: 13281333.
  • Lane DJ (1991) 16S/23S rRNA Sequencing (StackebrandtE & GoodfellowM, eds), pp. 130141. Wiley Press, Chichester.
  • Lee SM & Chao A (1994) Estimating population-size via sample coverage for closed capture-recapture models. Biometrics 50: 8897.
  • Lin E, Caron D & DeLong E (1996) Development and field application of a quantitative method for examining natural assemblages of protists with oligonucleotide probes. Appl Environ Microbiol 62: 14161423.
  • Lyons WB, Tyler SW, Wharton RA, McKnight DM & Vaughn BH (1998) A late Holocene desiccation of Lake Hoare and Lake Fryxell, McMurdo Dry Valleys, Antarctica. Antarct Sci 10: 247256.
  • Marshall W & Laybourn-Parry J (2002) The balance between photosynthesis and grazing in Antarctic mixotrophic cryptophytes during summer. Freshwater Biol 47: 20602070.
  • Massana R, Murray AE, Preston CM & DeLong EF (1997) Vertical distribution and phylogenetic characterization of marine planktonic Archaea in the Santa Barbara Channel. Appl Environ Microbiol 63: 5056.
  • Matsumoto GI (1993) Geochemical Features of the McMurdo Dry Valley Lakes, Antarctica, Vol. 59 (GreenWJ & FriedmannEI, eds), pp. 95118. American Geophysical Union, Washington, DC.
  • McCammon SA, Innes BH, Bowman JP, Franzmann PD, Dobson SJ, Holloway PE, Skerratt JH, Nichols PD & Rankin LM (1998) Flavobacterium hibernum sp. nov., a lactose-utilizing bacterium from a freshwater Antarctic lake. Int J Syst Bacteriol 48: 1405.
  • McKay CP, Clow GD, Andersen DT & Wharton RAJ (1994) Light transmission and reflection in perennially ice-covered Lake Hoare, Antarctica. J Geophys Res 99: 2042720444.
  • Miteva VI & Brenchley JE (2005) Detection and isolation of ultrasmall microorganisms from a 120,000-year-old Greenland glacier ice core. Appl Environ Microbiol 71: 78067818.
  • Miteva VI, Sheridan PP & Brenchley JE (2004) Phylogenetic and physiological diversity of microorganisms isolated from a deep Greenland glacier ice core. Appl Environ Microbiol 70: 202213.
  • Mulyukin AL, Demkina EV, Kozlova AN, Soina VS & El'-Registan GI (2001) Synthesis of anabiosis autoinducers by non-sporeforming bacteria as a mechanism regulating their activity in soil and subsoil sedimentary rocks. Microbiology 70: 535541.
  • Murray AE, Hollibaugh JT & Orrego C (1996) Phylogenetic compositions of bacterioplankton from two California estuaries compared by denaturing gradient gel electrophoresis of 16S rDNA fragments. Appl Environ Microbiol 26762680.
  • Murray A, Ferriera S, Johnson J et al. (2006) Marine actinobacterium PHSC20C1 genome, direct GenBank submission.
  • Muyzer G, De Waal EC & Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59: 695700.
  • Neale PJ & Priscu JC (1998) Fluorescence Quenching in Phytoplankton of the McMurdo Dry Valley Lakes (Antarctica): Implications for the Structure and Function of the Photosynthetic Apparatus, Vol. 72 (PriscuJ, ed), pp. 241253. American Geophysical Union, Washington, DC.
  • Olsen GJ, Matsuda H, Hagstrom R & Overbeek R (1994) fastDNAml: a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput Appl Biosci 10: 4148.
  • Onofri S, Selbmann L, Zucconi L & Pagano S (2004) Antarctic microfungi as models for exobiology. Planet Space Sci 52: 229237.
  • Pospiech A & Neumann B (1995) A Versatile Quick-Prep of Genomic DNA from Gram-Positive Bacteria. Trends Genet 11: 217218.
  • Priscu JC, Fritsen CH, Adams EE, Giovannoni SJ, Paerl HW, McKay CP, Doran PT, Gordon DA, Lanoil BD & Pinckney JL (1998) Perennial Antarctic lake ice: an oasis for life in a polar desert. Science 280: 20952098.
  • Priscu JC, Adams EE, Lyons WB et al. (1999) Geomicrobiology of subglacial ice above Lake Vostok, Antarctica. Science 286: 21412144.
  • Reddy GSN, Prakash JSS, Srinivas R, Matsumoto GI & Shivaji S (2003) Leifsonia rubra sp. nov. and Leifsonia aurea sp. nov., psychrophiles from a pond in Antarctica. Int J Syst Evol Microbiol 53: 977984.
  • Reysenbach AL, Giver LJ, Wickham GS & Pace NR (1992) Differential Amplification of Ribosomal-RNA Genes by Polymerase Chain-Reaction. Appl Environ Microbiol 58: 34173418.
  • Roberts EC & Laybourn-Parry J (1999) Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwater Biol 41: 737746.
  • Sheridan PP, Miteva VI & Brenchley JE (2003) Phylogenetic analysis of anaerobic psychrophilic enrichment cultures obtained from a Greenland glacier ice core. Appl Environ Microbiol 69: 21532160.
  • Simon M & Azam F (1989) Protein content and protein synthesis rates of planktonic marine bacteria. Mar Ecol Prog Ser 51: 201213.
  • Stackebrandt E & Brambilla E (2002) Life in Cold Lakes (HorneckG & Baumstark-KhanC, eds), pp. 161168. Springer, Berlin.
  • Stewart FJ, Fritsen CH, Murray AE & Carter BJ (2003) Prokaryotic Diversity in Antarctic Sea Ice: Inter-Habitat Variability. American Society of Limnology and Oceanography 2003 Aquatic Sciences Meeting: The Earth's Eyes: Aquatic Sciences Through Space and Time. Salt Lake City, Utah.
  • Suzuki MT & Giovannoni SJ (1996) Bias caused by template annealing in the amplification of mixtures of 16S rRNA genes by PCR. Appl Environ Microbiol 62: 625630.
  • Suzuki M, Rappe MS & Giovannoni SJ (1998) Kinetic bias in estimates of coastal picoplankton community structure obtained by measurements of small-subunit rRNA gene PCR amplicon length heterogeneity. Appl Environ Microbiol 64: 45224529.
  • Takamatsu N, Kato N, Matsumoto GI & Torii T (1998) The origins of salts in water bodies of the McMurdo Dry Valleys. Antarct Sci 10: 439448.
  • Van Trappen S, Mergaert J, Van Eygen S, Dawyndt P, Cnockaert MC & Swings J (2002) Diversity of 746 heterotrophic bacteria isolated from microbial mats from ten Antarctic lakes. Syst Appl Microbiol 25: 603610.
  • Voytek MA, Priscu JC & Ward BB (1999) The distribution and relative abundance of ammonia-oxidizing bacteria in lakes of the McMurdo Dry Valley, Antarctica. Hydrobiologia 401: 113130.
  • Ward BB & Priscu JC (1997) Detection and characterization of denitrifying bacteria from a permanently ice-covered Antarctic lake. Hydrobiologia 347: 5768.
  • Welschmeyer N (1994) Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and pheopigments. Limnol Oceanogr 38: 19851992.
  • Wharton RAJ, McKay CP, Clow GD & Andersen DT (1993) Perennial Ice Covers and their Influence on Antarctic Lake Ecosystems, Vol. 59 (GreenWJ & FriedmannEI, eds), pp. 5370. American Geophysical Union, Washington, DC.
  • Willerslev E, Hansen AJ, Ronn R, Brand TB, Barnes I, Wiuf C, Gilichinsky D, Mitchell D & Cooper A (2004) Long-term persistence of bacterial DNA. Curr Biol 14: R9R10.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Figure S1. Eukaryal 18S rRNA gene denaturing gradient electrophoresis gel showing distribution of phylotypes at different depths in the ice cover. Total number of bands per sample (as determined by GelComparII and visually confirmed) listed at the bottom of each lane. Numbers to the left of individual bands indicate bands that were excised and sequenced. Arrow pointing to the only universal phylotype (note low intensity in some samples). Figure S2. Bacterial 16S rRNA gene denaturing gradient electrophoresis gel showing distribution of phylotypes at different depths in the ice cover. Total number of bands per sample (as determined by GelComparII and visually confirmed) listed at the bottom of each lane. Arrows pointing to universal phylotypes (note low intensity in some samples). Figure S3. 16S rRNA gene clone library richness and coverage estimates. Predicted SChao1 values (panel A) and predicted CACE values (panel B) for the Lake Vida 4.8 m (filled circles) and 15.9 m (open circles) clone libraries. Table S1. Eukaryal and bacterial DGGE profile pairwise similarities (Sorenson’s index).

FEM+220_FigsS1-S3.pdf576KSupporting info item
FEM+220_Table+S1.pdf15KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.