Predominance of Roseobacter, Sulfitobacter, Glaciecola and Psychrobacter in seawater collected off Ushuaia, Argentina, Sub-Antarctica


  • Editor: Rosa Margesin

Correspondence: Sisinthy Shivaji, Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500 007, India. Tel.: +00 91 40 27192504; fax: +00 91 40 27160591; e-mail:


Bacterial diversity in sub-Antarctic seawater, collected off Ushuaia, Argentina, was examined using a culture independent approach. The composition of the 16S rRNA gene libraries from seawater and seawater contaminated with the water soluble fraction of crude oil was statistically different (P value 0.001). In both libraries, clones representing the Alphaproteobacteria, Gammaproteobacteria, the Cytophaga–Flavobacterium–Bacteroidetes group and unculturable bacteria were dominant. Clones associated with the genera Roseobacter, Sulfitobacter, Staleya, Glaciecola, Colwellia, Marinomonas, Cytophaga and Cellulophaga were common to both the libraries. However, clones associated with Psychrobacter, Arcobacter, Formosa algae, Polaribacter, Ulvibacter and Tenacibaculum were found only in seawater contaminated with hydrocarbons (Table 1). Further, the percentage of clones of Roseobacter, Sulfitobacter and Glaceicola was high in seawater (43%, 90% and 12% respectively) compared to seawater contaminated with hydrocarbons (35%, 4% and 9% respectively). One of the clones F2C63 showed 100% similarity with Marinomonas ushuaiensis a bacterium identified by us from the same site.

Table 1.   Comparison of 16S rRNA gene clones of seawater and seawater contaminated with the water-soluble fraction (WSF) of crude oil collected off Ushuaia, Argentina, Sub-Antarctica
Phylogenetic neighbour and nucleotide
accession number
Clone number and nucleotide accession number
Seawater (120)Seawater contaminated with
WSF of crude oil (104)
  1. The nucleotide accession numbers of the clones are given in round brackets. The percentage values within square brackets represent the percentage similarity of the 16S rRNA gene sequence of the clones with the nearest phylogenetic neighbour. The bold numericals in columns two and three indicate the number of clones in the particular class or genus. The clones which are in bold font represent unidentified and unculturable bacteria which have been assigned to the nearest phylogenetic neighbour based on their affiliations in the phylogenetic trees (Figs 1 and 2). The remaining clones which did not identify with any known phylogenetic neighbour are listed at the end of the table.

Roseobacter sp. ANT909 (AY167254)F2C03A (AY794091), F2C05A (AY794093) F2C09A (AY794097), F2C14A (AY794099)
F2C01 (AY794100), F2C14 (AY794117),
F2C16 (AY794119), F2C17 (AY794120), F2C36 (AY794125), F2C31(AY794123), F2C37 (AY794126), F2C43 (AY697883), F2C67(AY794137), F2C74 (AY794144), F2C77 (AY794143), F2C81(AY794146), F2C86 (AY794147), F2C95 (AY697887), F2C97 (AY794148), F2C98 (AY697888), F4C21(AY794194), F4C14(AY697911), F4C28 (AY794199), F4C85 (AY697924), F4C31 (AY794202), F4C31S (AY936201), F4C31 (AY794202) and F4C70 (AY697920). [96–99%]
F1C14 (AY794061), F1C69T (AY794083), F1C27S (AY794068), F1C79 (AY794086), F1C98 (AY697874), F3C01A (AY794150), F3C05 (AY697890), F3C28 (AY697897), F3C44 (AY697902) and F3C50 (AY697904). [94–99%]
Roseobacter sp. ANT9274 (AY167261) F1C69T (AY794083). [97%]
Roseobacter sp. ANT9276a (AY167262) F1C75 (AY794085). [94%]
Roseobacter sp. ANT9270 (AY167260)F2C32 (AY694124). [99%] 
Roseobacter sp. ARCTIC-P4 (AY573044) F1C74S (AY936184). [96%]
Uncultured Roseobacter
Roseobacter sp. clone (AY627365)F4C01 (AY936200), F2C06A (AY79494) and F4C03 (AY697909). [98–99%] 
Roseobacter NAC11-6 (AF245634) F1C74T (AY936185). [97%]
Roseobacter NAC11-7 (AF245635)F2C02 (AY936186) and F4C13 (AY794188). [96–99%] 
Roseobacter sp. EB080-L11F12 (AY627365)F4C02D (AY794181), F2C04 (AY936187), F4C44 (AY697913) and F4C74 (AY697922). [98–99%] 
Roseobacter sp. (AY167254)F2C99 (AY794149). [99%] 
Roseobacter sp. DG1132 (AY258102) F1C10 (AY794059). [95%]
Unidentified and uncultured RoseobacterF2C08A (AY794096), F2C18 (AY697878), F2C35 (AY697881), F2C78 (AY794144), F2C79 (AY794145), F2C92 (AY697886), F2C100 (AY936189), F2C102 (AY794106).
F2C107 (AY794111), F2C110 (AY794114), F4C04 (AY697910) and F4C59 (AY697917). [92–99%].
F1C18 (AY697871), F1C28 (AY697872), F1C58M (AY794078), F1C60 (AY936180), F1C69S (AY794082), F1C69 (AY794084), F3C07 (AY697891), F3C17B (AY794155), F3C24 (AY794157), F3C48 (AY697903), F3C56 (AY794164), F3C59 (AY794165), F3C60 (AY794166), F3C65 (AY794167), F3C68 (AY697905), F3C73 (AY794169), F3C86 (AY794172), F3C91 (AY794174), F3C94 (AY794176), F3C96 (AY794177) and F3C99 (AY794179). [94–100%].
Sulfitobacter mediterraneus (Y17387)F2C25 (AY697879). [97%] 
Sulfitobacter sp. (AJ534236)F2C44 (AY794129). [94%] 
Sulfitobacter sp. GAI-21 (AF007257)F2C47 (AY697884), F2C84 (AY697885),
F4C57 (AY794211) and F4C56 (AY697915). [95–99%]
F3C23 (AY936196). [99%]
Sulfitobacter sp. DFL-41 (AJ534222)F2C07A (AY794095). [95%] 
Sulfitobacter sp. ANT9282b (AY167263)F2C10 (AY794105), F2C42 (AY794128) and F2C103 (AY794107). [96–97%]F1C67 (AY794080), F1C68D (AY794081) and F3C110 (AY697908). [96–98%]
Sulfitobacter sp. ANT9115 (AY167322)F2C52 (AY794132). [96%] 
Staleya guttiformis EL-38T (Y16427)F4C15 (AY697912) [98%].F1CA9 (AY697867). [98%]
Alphaproteobacterium Shippagan (AF100168) F1C27T (AY794069). [97–98%]
Alphaproteobacterium OM42 (U70680)F4C02E (AY794180). [98%] 
Marine Alphaproteobacterium (AJ391182)F4C42 (AY794207). [96%]F1C26D (AY794066). [97%]
Marine Alphaproteobacterium (AF365990) F1A08 (AY936173). [92%]
‘Candidatus pelagibacter’ ubiqueF2CB3 (AY697875) and F2CB7 (AY697876). [96%] 
Glaciecola mesophila (AJ548479) F3C55 (AY794163). [93%]
Glaciecola sp. EL-110 (AJ308105)F2C26 (AY794121). [96%]F1C08 (AY936188). [96%]
Glaciecola sp. HA02 (AB049729)F2C105 (AY794109). [95%] 
Glaciecola pallidula ACAM 615T (AJ308105) F1C16 (AY794062). [97%]
Glaciecola pallidula ACAM 615T (AJ308105)F2C04A (AY794092), F2C21 (AY794120), F2C28 (AY697880), F2C38 (AY697882), F2C53 (AY794133), F2C48 (AY794130), F2C60 (AY794134), F4C18 (AY794191), F4C19 (AY794192), F4C48 (AY794210), F4C63 (AY697918) and F4C71 (AY697921). [95–97%]F3C08 (AY697892), F3C14 (AY697894), F3C16 (AY697895), F3C20 (AY936194),
and F3C90 (AY794173). [95–97%]
Glaciecola punicea ACAM 611T (U85853) F3C95 (AY936198). [98%]
Psychromonasprofunda (AJ416756)F2C08 (AY936188) and F2C72 (AY794140). [97–99%] 
Colwellia piezophila (AB094412) F1C89T (AY794088). [98%]
Colwellia psychrerythraea IC064 (U85842)F4C86 (AY794216). [97%] 
Marinomonas ushuaiensis (AJ627909)F2C63 (AY794136) and F2C68 (AY794138). [97–100%]F3C93 (AY794175). [96%]
Marinomonas primoryensisAJ238597) F3C32 (AY697898). [99%]
Psychrobacter psychrophilus (AJ748270) F3C17A (AY794154). [98%]
Psychrobacter sp. DY9-2
 F3C12 (AY794152), F3C15 (AY794153), F3C39 (AY794160), F3C54 (AY794162) and F3C46 (AY936197). [98–99%}
Vibrio sp. LnSQ6 (AY158028)F4C73 (AY794214). [97%] 
Oleispira Antarctica RB-8T (AJ426420) F3C18 (AY697896). [93%]
Pseudomonas balearica SP1402T (U26418)F4C67 (AY697919). [99%] 
Uncultured and unidentified Gammaproteobacteria46
Gammaproteobacterium SUR (AF114620)F4C06 (AY794184). [97%] 
Gammaproteobacterium UMB8H (AF505737) F3C103B (AY936199) and F3C10 (AY697906). [98%]
Gammaproteobacterium Arctic96B (AF354595)F4C87 (AY794217). [96%] 
Gammaproteobacterium UMB10B (AF505738) F3C102 (AY936191) and F3C102B (AY936192). [98%]
Gammaproteobacterium UMB10B (AF469279) F3C21 (AY936195). [99%]
Uncultured and unidentified GammaproteobacteriaF4C93 (AY794218) and F4C87 (AY794217). [93–96%]F3C13 (AY697893). [97%]
Arcobacter sp. KT0913 (AF235110) F3C43 (AY697901). [97%]
Uncultured Proteobacteria33
Uncultured Proteobacterium (AY644707)F2C09 (AY794105), F2C104 (AY794107), and F2C108 (AY794112). [92–99%]F3C25 (AY794158), F3C78 (AY794107), F3C97 (AY794178). [94–100%]
Cytophaga–Flexibacter–Bacteriodetes group1520
Cytophaga (Stanierella)10
Cytophaga sp. (AB015532)F2C76 (AY794142). [96%] 
Cellulophaga lytica (AB032510)F4C20 (AY794193). [93%] 
Cellulophaga sp. (AF497997) F1C36 (AY794073). [92%]
Cellulophaga sp. (AY035869) F1C48 (AY936178). [98%]
Formosa algae02
Formosa algae KMM 3553 (AY228461) F3C41 (AY697899) and F3C42 (AY697900). [92%]
Maribacter orientalis (AY271624)F4C50 (AY697914). [96%] 
Flexibacter sp. IUB42 (AB058905)F4C94 (AY697925). [95%]F1C61 (AY697873). [96%]
Polaribacter franzmannii ANT9260 (AY167319) F1A13D (AY794060). [93%]
Polaribacter sp. HLE (AY198117) F3C72 (AY794168). [98%]
Ulvibacter litoralis (AY243096) F1C73 (AY936183). [94%]
Tenacibaculum maritum (92%) (AF359539) F1C24F (AY936177). [92%]
Uncultured Cytophaga–Flexibacter–Bacteroidetes Group1111
CFB bacterium MERTZ (AF424351)F4C47 (AY794209). [88%] 
CFB bacterium MERTZ (AF424352)F4C05 (AY794182). [90%]F1C38 (AY794075). [90%]
CFB bacterium MERTZ_0CM_184 (AF424353)F2C49 (AY794131). [98%] 
Bacteroidetes bacterium (AY274847) F1C29S (AY794070). [98%]
Bacteroidetes bacterium 1D10 (AY274838)F4C17 (AY794190) [96%]F3C79 (AY794171) and F1C08T (AY936176). [94–98%]
Bacteroidetes bacterium (AY580719)F2C106 (AY794110). [97%] 
Bacteroidetes bacterium (AY225660)F4C37 (AY794204). [89%] 
Flexibacterium (AJ297467) F3C16A (AY936193). [98%]
CFBF4C34 (AY794203), F4C41 (AY794206), F4C78 (AY697923) and F2C29 (AY794122). [90–99%]F1C37 (AY794074), F1C12 (AY697870), F1CA7 (AY697866), F1CA12 (AY697868) and F1C07 (AY697869). [92–98%]
Flavobacteriaceae bacterium G1112S4A (AY353814)F4C43 (AY794208). [94%] 
Flavobacteriaceae bacterium (AY285946) F1C17 (AY794063). [97%]
Unidentified and Uncultured organisms1011
Uncultured Antarctic sea ice bacterium (AY165570) F1C34 (AY794072) and F3C19 (AY794156). [97%]
Antarctic sea ice bacterium (AY165576)F2C12 (AY794116) and F4C29 (AY794200). [97–98%] 
Arctic sea ice bacterium ARK10076 (AF468411)F2C61 (AY794135). [95%] 
Uncultured marine bacterium ZD0409 (AJ400350) F1C49 (AY794076). [96%]
Uncultured marine bacterium (AJ400343) F1C30S (AY794071). [98%]
Uncultured marine bacterium (AJ298381) F1C66D (AY936181). [90%]
Uncultured marine bacterium ZD0207 (AJ400341)F2C06 (AY794103). [97%] 
Uncultured bacterium clone 223 (AY172247)F2C101 (AY936190), F4C61D (AY794212) and F4C61E (AY794213). [95–97%] 
Uncultured bacterium clone (AY133378) F1C53F (AY794077) and F1C53S (AY936179). [98%]
Uncultured bacterium GR-Sh1-209 (AJ296579)F2C11A (AY794098). [90%] 
Uncultured bacterium clone OB3 (AY133410)F4C23 (AY794196). [98%] 
Unidentified bacterium DNA (Z88574) F1C22 (AY794064) and F3C31 (AY794154). [96–98%]
Bacterium DG890 small subunit (AY258122) F1C66E (AY936182). [89%]
Gas vacuolate str. 206 (U73723) F1C83S (AY794087). [97%]
Ophiopholis aculeata symbiont (U63548)F2C41 (AY794127). [97%] 


Seawater microbial communities from temperate and polar regions are known to influence global energy, atmospheric-oceanic interactions and oceanic food web (Legendre et al., 1992; Brown & Bowman, 2001). Therefore, there is a need to build an inventory of these microbial communities so as to establish baseline data on the structure of these communities prior to human intervention, by way of research, tourism and natural resource exploration (Cripps & Shears, 1977). Such interventions may lead to pollution of the environment by petroleum products used for transport and energy production. As a consequence many novel species may be lost or reduced. Oil spills have occurred in Antarctica (Aislabie et al., 2004) and sub-Antarctic regions (Coulon & Delille, 2003). Several novel hydrocarbon degrading bacteria belonging to the genera Rhodococcus, Acinetobacter, Pseudomonas, Sphingomonas etc. have been isolated from Antarctica and other cold regions (Leahy & Colwell, 1990; Aislabie et al., 2000; Bej et al., 2000; Sepulveda-Torres et al., 2001; Baraniecki et al., 2002).

Biodiversity studies on psychrophilic microorganisms have been concentrated to regions of Antarctica, Arctic, alpine regions and the Himalayan mountain ranges (Brinkmeyer et al., 2003; Margesin et al., 2003b, 2004; Sheridan et al., 2003; Sjoling & Cowan, 2003; Bowman, 2004; Shivaji et al., 2004, 2005; Stackebrandt et al., 2004; Chaturvedi et al., 2005), and these studies have led to the identification of several novel microorganisms. In November 2001 a multi-disciplinary research programme was undertaken on the seawater collected off Ushuaia, Argentina, a sub-Antarctic region, to study the effects of toxic chemicals resulting from human activities in a sub-Antarctic region. For this purpose, the diversity of bacteria from seawater collected off Ushuaia, Argentina, was compared with that of a seawater sample supplemented with the water-soluble fraction (WSF) of a crude oil, from the same site. The results are discussed with respect to earlier studies related to bacterial populations in cold regions exposed to hydrocarbon spills using the culture-independent approaches.

Materials and methods

Sampling site and sample collection

Ushuaia is located in the southern most tip of Tierra del Fuego in South America located at 54°80′S and 68°31′W. The average minimum and maximum air temperatures at Ushuaia ranged from −1 to 5°C and 4 to 14°C, respectively, and the mean seawater temperature was 7.5±0.7°C. Thus, Ushuaia could be a habitat where cold adapted microorganisms thrive. The present study was undertaken in Ushuaia to study the effects of toxic chemicals resulting from human activities in these regions. For this purpose, seawater from a depth of 2 m was pumped, using a submersible electric pump, into two large clear polyethylene bags (∼1900 L). Into these polyethylene bags, two microcosms (cylindrical Teflon® bags, resistant to UV degradation) containing 9 L of seawater were immersed. One of the immersed Teflon bags was supplemented with the WSF of a Patagonia Crude Oil (obtained from an Argentinean company Comodoro Rivadavia) to a final concentration of ∼5 mg L−1. The other microcosm to which WSF was not added served as a control. Both microorganisms were incubated for 5 days in situ. Due to logistic difficulties, only two microorganisms were set up representing the control and the hydro-carbon added microcosm, respectively.

The WSF was obtained as described earlier (Gearing & Gearing, 1982; Heitkamp & Cerniglia, 1987; Siron et al., 1987; Yamada et al., 2003). The procedure involved layering of 40 mL of Pantagonian crude oil over 4 L of 0.45 μm filtered seawater in a sterile glass jar. The jar was sealed and the mixture was stirred slowly for 60 h at 18°C. Subsequently, without disturbing the oil/water interface, the WSF was removed by siphoning.

The hydrocarbons present in the WSF typically include 50% monoaromatics (benzene, toluene, xylene and other alkylbenzenes), 30% diaromatics (naphthalene and alkylnaphthylenes), 15% other polyaromatic hydrocarbons (including phenanthrene, anthracene and alkylated 3- and 4-ring aromatic hydrocarbons) and <5% polar compounds (Gearing & Gearing, 1982; Heitkamp & Cerniglia, 1987; Siron et al., 1987; Yamada et al., 2003). The WSF composition is known to change very rapidly. For instance, it has been observed that as soon as the fraction is added to the mesocosm lighter hydrocarbons such as benzenes and napthalenes are lost due to evaporation at the air/water interface. Further, at water temperatures of about 20°C compounds such as benzene, naphthalenes and phenanthrenes disappear from the mesocosms within 2 days after introduction (Yamada et al., 2003). However, the heavier compounds such as pyrene and chrysene persist for many days. Similar losses also occurred in Ushuaia but the process was slightly slower due to lower temperatures in sub-Antarctica.

For microbiological analysis, 250 mL of the seawater was collected in sterile Teflon bags on day 5, and filtered using a 0.45 μM filter. The filter was then used for the extraction of DNA, amplification of the rRNA genes, construction of 16S rRNA gene library and subsequent identification of the bacteria by the culture independent rRNA approach. Inoculated plates (six replicates) were incubated for 10 days at 18°C. A total of four libraries were constructed using two DNA samples each from the control microcosm and the microcosm to which WSF of crude oil was added. Each DNA sample was prepared from 250 mL of the seawater. The number of viable heterotrophic bacteria in each sample was estimated using the spread plate technique on nutrient agar (Difco) plates which were incubated at 18°C for 10 days.

Extraction of DNA from filters

DNA was extracted from the filter samples according to the procedure described by Tsai & Olson (1991, 1992) and modified by Shivaji et al. (2004). The filters were cut into small pieces and were suspended in 500 μL of lysis buffer (0.15 M NaCl, 0.1 M EDTA, pH 8) containing 30 mg mL−1 lysozyme and incubated at 37°C for 2 h in a shaking water bath. Subsequently, 200 μL of 0.1 M NaCl prepared in 0.5 M Tris-HCl (pH 8) buffer and 10% SDS were added such that the final concentration of SDS was 1%. The contents were then subjected to five freeze–thaw cycles by freezing the samples in liquid nitrogen and thawing at 65°C, so as to release the DNA from bacterial cells in the filters. De-proteinization was done once with tris-saturated phenol and then with chloroform: isoamyl alcohol mixture (24 : 1, v/v). The aqueous phase was collected and DNA was precipitated with 2% sodium acetate and one-third volume of isopropanol. The DNA pellet was then recovered by centrifugation at 15 000 g for 20 min at room temperature, washed with 70% alcohol, vacuum dried and dissolved in 50 mL of TE buffer (10 mM Tris-HCl and 1 mM EDTA, pH 8) or sterile water.

Amplification and cloning of 16S rRNA gene

The 16S rRNA gene was amplified by PCR, in a total volume of 50 μL, using the conditions described earlier (Shivaji et al., 2004) and the two primers used were 16S1 (5′-GAG TTT GAT CCT GGC TCA -3′) and 16S2 (5′-ACG GCT ACC TTG TTA CGA CTT -3′), complementary to nucleotide positions 9–27 (forward) and 1498–1477 (reverse) respectively of the E. coli 16S rRNA gene (Lane, 1991). In all the amplification reactions, water was used in the place of DNA as a negative control. The amplified DNA was purified using Qiagen gel columns and cloned into pGEM-T Vector system II (Promega, Heidelberg, Germany) and used for transformation of E. coli DH5α as suggested by the manufacturer.

Sequencing and phylogenetic analysis of the 16S rRNA gene clones

A single colony was inoculated into 3 mL of LB broth containing 50 μg mL−1 ampicillin and incubated overnight at 37oC. The pellet was recovered by centrifugation at 2800 g for 5 min at 4°C and used for plasmid preparation according to the method of Birnboim & Doly (1979), as described in Sambrook et al. (1989). The plasmid was then stored in 50 μL of TE buffer in a freezer at −20°C. The partial sequence of 16S rRNA gene was obtained using one of the two primers, namely the T7 universal primer (5′-GTAATACGACTCACTATAGGGC-3′) or the SP6 universal primer (5′-ATTTAGGTGACACTATAG-3′). Where necessary, internal sequencing primers pD (5′-CAGCAGCCGCGGTAATAC-3′) and pE (5′-CCGTCAATTCCTTTGAGTTT-3′), complementary to the nucleotide positions 518–536 and 928-908, respectively, of the E. coli 16S rRNA gene, were used and sequenced as described earlier (Lane, 1991; Shivaji et al., 2004).

The sequences obtained were annotated and analyzed using the autoassembler software (Perkin Elmer, Applied Biosystems) and blast ( for sequence similarity search and identification of chimeras. Later these were aligned with closely related 16S rRNA gene sequences using clustal w (Thompson et al., 1994). The pair-wise evolutionary distances were computed using the dnadist program with the Kimura two-parameter model (Kimura, 1980). Phylogenetic trees were constructed using two tree-making algorithms (Neighbour-Joining and dnapars) of the phylip software package (Felsenstein, 1993). The stability among the clades of the phylogenetic tree was assessed by taking 1000 replicates and analysing the dataset using the programs seqboot, dnadist, neighbor and consense of the phylip package (Felsenstein, 1993). Libshuff statistical protocol (Version 1.22) was used to ascertain differences in diversity of clones between seawater and seawater to which WSF of crude oil was added (Schloss et al., 2004).

Nucleotide sequence accession numbers

All the sequences of the 16S rRNA gene clones were deposited in Genbank with accession numbers AY644707, AY697866AY6977925, AY794059AY794218 and AY936173AY936201.


16rRNA gene libraries

Heterotrophic bacterial abundance both in seawater and seawater contaminated with WSF of crude oil increased from 103 to about 105 CFU mL−1 by day 5 of the experiment. These seawater samples were used for bacterial diversity determination based on the composition of the 16S rRNA gene libraries. Filtration of 250 mL of seawater and seawater contaminated with WSF of crude oil yielded about 50 μg of DNA per filter. About 200 ng of genomic DNA from each sample was used for PCR amplification of rRNA genes and the amplicons were used to construct a 16S rRNA gene library. A total of four libraries were constructed using DNA from two filters through which seawater was filtered and two filters through which seawater contaminated with WSF of crude oil was filtered. The number of clones in each of the four libraries ranged from 100 to 125. All the clones from each library were sequenced. After eliminating chimeras generated during PCR and chloroplast sequences, the number of clones available for analysis was about 50–60 for each of the filters. Only those clones which were >800 nt were employed for phylogenetic analysis. But, the nearest phylogenetic neighbour was determined for all clones >400 nt (Table 1). The duplicate libraries of seawater were very similar based on the phylogenetic identity of the clones and the frequency of their distribution, which varied by <5%. This observation was also true for the seawater contaminated with WSF of crude oil. Therefore, the data for the two seawater (120 clones) and the two seawater samples to which WSF of crude oil (104 clones) was added were pooled (Table 1) and they did not include the chloroplast clones. Using Libshuff statistical protocol, the diversity between seawater and seawater contaminated with WSF of crude oil appeared to be significantly different (P value of 0.001).

Clones from seawater

Out of a total of 120 clones from seawater 67 of the clones (56%) belonged to the class Alphaproteobacteria represented predominantly by 51 clones of Roseobacter, 11 clones of Sulfitobacter, one clone of Staleya and four clones of uncultured Alphaproteobacteria (Table 1). However, the clones grouping with Roseobacter were not identical and sequence similarity between them ranged from 92% to 99%. The next abundant group, represented by 25 clones (21%) included representatives of Gammaproteobacteria, belonging to genera Glaciecola, Psychromonas, Colwellia, Marinomonas, Pseudomonas, Vibrio and unculturable Gammaproteobacteria. Within the Gammaproteobacteria 14 out of the 21 clones belonged to the genus Glaciecola and 12 of these were phylogenetically related to G. pallidula ranging from 95% to 97% (Table 1). The seawater also included 15 clones (12.5%) belonging to Cytophaga–Flexibacter–Bacteroidetes group bacteria showing <97% 16S rRNA gene sequence similarity with species of Stanierella, Cellulophaga, Maribacter, Flexibacter and Flavobacteria. A good number (11 out of 15 clones) were grouped as unculturable Cytophaga–Flexibacter–Bacteriodetes group. The remaining 10 clones (8%) did not identify with any of the known culturable bacteria. Results in Table 1 are based on the blast analysis of the clone sequences and also on the phylogenetic affiliation of the clones as depicted in Figs 1 and 2. Phylogenetic affiliation was used to assign the large number of clones that were identified by blast as unidentified and unculturable bacteria with their nearest phylogenetic neighbour. Two clones viz., F2CB3 and F2CB7 clustered with candidatus ‘Pelagibacter ubique’ under SAR11 cluster of Rickettsiales.

Figure 1.

 Neighbour joining tree showing the relationship of 16S rRNA gene clones from seawater collected off Ushuaia, Argentina, Sub-Antarctica. Clones in bold indicate those identified as uncultured organisms.

Figure 2.

 Neighbour joining tree showing the relationship of 16S rRNA gene clones from seawater collected off Ushuaia, Argentina, Sub-Antarctica and contaminated with the water soluble fraction of crude oil. Clones in bold indicate those identified as uncultured organisms.

Clones from hydrocarbon-contaminated seawater

The 104 clones sequenced from seawater contaminated with WSF of crude oil (Table 1) were phylogenetically related to the class Alphaproteobacteria (44 clones, 42%), Gammaproteobacteria (25 clones, 24%), Epsilonproteobacteria (one clone, 1%) and Cytophaga–Flexibacter–Bacteroidetes group (20 clones, 19%). Though Roseobacter (36 clones, 35%) and Sulfitobacter (four clones, 4%) were predominant, a clone F1CA9 was found to cluster with Staleya guttiformis. Gammaproteobacteria was represented by species belonging to the genus Glaciecola (nine clones), Colwellia (one clone), Marinomonas (two clones), Psychrobacter (six clones) and Oleispira (one clone). Six more clones were grouped as unculturable Gammaproteobacteria. One clone F3C43 was affiliated to Arcobacter sp. KT0913. Among the Cytophaga–Flexibacter–Bacteroidetes group (20 clones) Cellulophaga, Formosa algae, Flexibacter, Ulvibacter, Polaribacter and Tenacibaculum were represented and majority of the clones (11 out of 20) were identified as unculturable Cytophaga–Flexibacter–Bacteroidetes group. The remaining 11 clones (11%) did not identify with any of the known bacterial groups. Phylogenetic analysis of the 104 clones (Fig. 2) indicated the formation of three distinct clusters represented by Alphaproteobacteria, Gammaproteobacteria and Cytophaga–Flexibacter–Bacteriodetes group (Fig. 2).

Comparison of clones from seawater and hydrocarbon contaminated seawater

Bacterial diversity of seawater and seawater contaminated with WSF of crude oil based on phylogenetic analysis of the 16S rRNA gene clones indicated that clones identifying with the genera Roseobacter, Sulfitobacter, Staleya, Glaciecola, Colwellia, Marinomonas, Cytophaga and Cellulophaga were common to both. But clones identifying with Psychrobacter, Arcobacter, Formosa algae, Polaribacter, Ulvibacter and Tenacibaculum were found only in seawater contaminated with WSF of crude oil (Table 1). It is possible that these organisms were present in very low numbers in seawater but following hydrocarbon addition the numbers increased. In addition, clones related to Psychromonas, Vibrio, Pseudomonas, Cytophaga and Maribacter, which were represented by one or two clones, in seawater were not present in the seawater contaminated with WSF of crude oil. Further, the percentage of clones of Roseobacter, Sulfitobacter and Glaceicola was high in the seawater (43%, 90% and 12% respectively) compared to seawater contaminated with WSF of crude oil (35%, 4% and 9% respectively). However, independent of the presence of hydrocarbons, the uncultured bacteria were the most predominant and these were related to Proteobacteria, Flavobacteriaceae, and Cytophaga–Flexibacter–Bacteriodetes group. Among the genera listed above, Arcobacter is reported to withstand hydrocarbon contamination and some species are sulfide-oxidizing (Wirsen et al., 2002).

Phylogenetic similarity of 16S rRNA gene clones of seawater with SAR11 and RCA clusters

Recent studies have indicated that the SAR11 clade and Roseobacter clade affiliated (RCA) clusters are predominant in tropical/temperate and polar seawater samples (Morris et al., 2002; Selje et al., 2004). Phylogenetic studies of our unidentified clones with SAR11 clade showed the affiliation of only two clones (F2CB3 and F2CB7) with ‘Candidatus Pelagibacter ubique’ (Fig. 1), a taxon of SAR11 (Morris et al., 2002; Rappe et al., 2002). A few of the unidentified clones were affiliated with unidentified Roseobacter clade (Figs 1 and 2) and phylogenetic analysis revealed that they are related to bacteria belonging to genera Antarctobacter (F3C86 and F2C108), Roseobacter (F1C10), Staleya (F4C15), Loktanella (F1C28 and F1C58), Sulfitobacter (F2C47, F2C84, F4C56, F4C57) and Agrobacterium (F1CA12, F1CA7 and F4C06). These genera are phylogenetically closely related to the Roseobacter clade (Selje et al., 2004).


Studies on the bacterial diversity of seawater from cold regions such as Arctic, Antarctic and temperate regions based on the rRNA approach are limited (Eilers et al., 2000; Brown & Bowman, 2001; Brinkmeyer et al., 2003; Bowman, 2004; Shivaji et al., 2004; Stackebrandt et al., 2004; Suzuki et al., 2004). To the best of our knowledge this is the first study on the bacterial diversity of seawater from a sub-Antarctic region. A number of clones (23%) were not analysed in this study since these clones represented chloroplast 16S rRNA gene clones. This is not unexpected and is in accordance with the observations of Brown & Bowman (2001) who reported that >65% of the clones from certain Antarctic sea ice samples are phylogenetically similar to chloroplast 16S rRNA gene clones. A closer evaluation of the phylogenetic relationship of the clones from seawater indicated that most of the 16S rRNA gene clones differed from the nearest phylogenetic neighbor by 0–12% at the rRNA gene level. A few of the clones were very closely related (>98%) to reported species such as F2C08 (Psychromonas profunda), F2C32 (Marinomonas primoryensis), F1C89T (Colwellia piezophila) and F1CA9 (Staleya guttiformis). Interestingly clone F2C63 showed 100% similarity with Marinomonas ushuaiensis, a species described by us from the same sample (Prabagaran et al., 2005).

The relative dominance of Alphaproteobacteria in the clone libraries of seawater (56% of the clones) as well as in seawater contaminated with WSF of crude oil (42%) observed by us has been previously observed (Cottrell & Kirchman, 2000; McCaig et al., 2001; Hagström et al., 2002; Bowman & McCuaig, 2003; Ellis et al., 2003; Shivaji et al., 2004) in various marine regions, including cold habitats. Most of the Alphaproteobacteria were members of the family Rhodobacteriaceae and included clones related to Roseobacter and Sulfitobacter genus. These clones of Roseobacter and Sulfitobacter were closely related to one another at the rRNA gene level (97–99%) and constituted a separate cluster. Abundance of Roseobacter in temperate and polar regions has been reported earlier (Cottrell & Kirchman, 2000; Adachi et al., 2003; Buchan et al., 2005). Selje et al. (2004) studied the global distribution of Roseobacter and established that the RCA cluster, comprising mainly of as yet uncultured phylotypes occurs from temperate to polar regions; this phylotype is highly abundant in the Southern Ocean and its distribution has not yet been studied from the tropical and subtropical regions. This study confirms the presence of Roseobacter phylotype as a dominant component of seawater off Ushuaia, a sub-Antarctic region and also demonstrates that the frequency of the Roseobacter clones (43%) was higher compared to its distribution in Baltic sea (0.5%), Arctic ocean (5%), German Bight (10%) and Southern Ocean (20%) (Selje et al., 2004). The Roseobacter clones from Ushuaia are not phylogenetically close to any of the reported species of Roseobacter viz., R. denitrificans, R. galleciensis and R. litoralis. But, these clones associate with the Roseobacter clade, which includes the genera, Roseobacter, Sulfitobacter, Staleya, Roseovarius, Loktanella etc. (Selje et al., 2004). It may be of interest to note that our clones were >1300 bp and the RCA reference sequences in the database were shorter (∼400 bp) and this discrepancy in size may not be the optimal method to determine similarity with high confidence.

Sulfitobacter, which belongs to the class Alphaproteobacteria, was present both in seawater (11 clones) and as well as in seawater contaminated with WSF of crude oil (four clones). But, except F2C25, which was similar to S. mediterraneus (97% rRNA gene similarity), the remaining clones did not identify with any of the other known species of Sulfitobacter viz., S. dubius, S. delicatus, S. brevis and S. pontiacus though they did identify with a number of isolates of Sulfitobacter sp. with a similarity of 95–99% at the rRNA gene level. Members of this genus (except S. brevis) are sulfite oxidizers (1) and have been isolated from various habitats like the Mediterranean sea (Pukall et al., 1999), hypersaline lake (Labrenz et al., 2000), starfish and sea-grass (Ivanova et al., 2004). The genus Staleya has only one valid species (S. guttiformis) and it is closely related to Roseobacter and Sulfitobacter. Clones F4C15 and F1CA9 are similar to S. guttiformis (>98%). So far there is no report on the identification of Staleya from seawater.

About 21–24% of the clones in both the libraries represented genera from the class Gammaproteobacteria. Earlier reports from Arctic and Antarctic regions had indicated this class to be dominant (Bowman et al., 1997; Brinkmeyer et al., 2003). In this class, clones with similarity to the genus Glaciecola are predominant in both the cases and majority of them were phylogenetically similar to G. pallidula (97% at the rRNA gene level). One of the clones F3C55 was most similar to G. mesophila (93%). However, none of the clones exhibited >97% similarity with any of the known species of Glaciecola viz. G. pallidula, G. mesophila and G. polaris implying that these clones may represent new species of Glaciecola. Gammaproteobacteria is also represented by clones similar to strains of the genera Colwellia and Marinomonas. However, a few clones similar to Vibrio and Psychromonas were detected only in seawater, and six clones similar to strains of Psychrobacter were detected only in seawater contaminated with WSF of crude oil. The possible explanation for the absence of Psychrobacter strains in seawater could be that these strains existed in very small numbers beyond the level of detection by PCR but get enriched due to hydrocarbon contamination. Further, the presence of single clones of Colwellia and Vibrio may be indicative of a library that has not been fully sampled as suggested by Schmidt et al. (1991). It is of interest that many of the species of the above genera are psychrophilic and halophilic, features, which would be characteristic of isolates from sub-Antarctic seawater. Thus, the present study is in accordance with earlier studies that have indicated dominance of Alphaproteobacteria and Gammaproteobacteria in sub-tropic and polar regions (Gosink & Staley, 1995; Bowman et al., 1997; Cottrell & Kirchman, 2000; Adachi et al., 2003; Brinkmeyer et al., 2003; Sheridan et al., 2003; Selje et al., 2004). But, the diversity of Gammaproteobacteria recorded in the earlier studies is high compared to the present study.

Betaproteobacteria in very low frequency has been reported from the polar/temperate regions (Gosink & Staley, 1995; Brinkmeyer et al., 2003; Sheridan et al., 2003). However, in the present study no clones were identified. Interestingly, clone F3C43 was identified as phylogenetically related to Arcobacter of class Epsilonproteobacteria, which has so far not been identified in sub-tropic and polar regions though reports have indicated their presence in marine environments.

Clones associated with the Cytophaga–Flexibacter–Bacteriodetes group were present both in seawater (12.5%) and in seawater contaminated with WSF of crude oil (19%). Phylogenetically they were affiliated to Stanierella, Cellulophaga, Formosa algae, Maribacter, Flexibacter, Polaribacter, Ulvibacter and Tenacibaculum. Cytophaga–Flexibacter–Bacteriodetes clones have been reported earlier to be present in low frequency in ice and seawater of temperate regions (Bowman et al., 1997; Cottrell & Kirchman, 2000; Adachi et al., 2003; Selje et al., 2004) though they are known to be predominant in seawater containing phytoplankton.

An unique feature of this study is the preponderance of clones associated with unculturable bacteria belonging to Alpha and Gammaproteobacteria, uncultured Cytophaga–Flexibacter–Bacteriodetes group bacteria, Arctic sea ice bacterium, Antarctic sea ice bacterium, etc. (Table 1). These results apart from reflecting the high degree of diversity also hint at our inability to yet establish the identity of many bacteria. One of the unique clones in this category is clone F1C83S, which shows 97% similarity with a gas vacuolate strain grouping in the genus Glaciecola. Gosink & Staley (1995) isolated various gas vacuolated bacteria from Antarctic sea ice and water and such bacteria are assumed to produce gas vesicles as a strategy to regulate their position in vertically stratified water columns (Walsby, 1994). Some clones clustered with Loktanella, Oleispira, Ulvibacter, Tenacibaculum, Pelagibacter, Roseovarius etc. (Figs 1 and 2). Clones similar to Tenacibaculum and Roseovarius were identified earlier in German Wadden Sea (Brinkhoff et al. 2004). Bacteria belonging to the other genera listed above have been reported from marine or polar habitats.

Recently, Morris et al. (2002) studied globally the abundance of SAR11, an Alpha-proteobacterial clade and established that SAR11 clade dominates ocean surface bacterioplakton and can represent up to 50% of the total surface microbial community. Thus, SAR11 is considered to be the most abundant and most successful bacteria on earth. It is possible that the unidentified and unculturable bacteria of seawater of Ushuaia are affiliated to SAR11 clade. Phylogenetic analysis of all the clones, which were identified as ‘uncultured’ in the present study indicated that except two clones, namely F2CB3 and F2CB7, which formed a clade with Pelagibacter ubique, a SAR11 affiliate, the remaining clones did not cluster with SAR11 clade. Suzuki et al. (2004) also detected many Pelagibacter SAR11 members while studying the diversity of Monterey Bay. Rappe et al. (2002) cultivated a few organisms belonging to this clade. It is envisaged that further studies on this culturable bacteria from SAR11 clade would provide valuable information on the population dynamics, metabolic potential and biogeochemical relevance to several habitats (Suzuki et al., 2004).

Research related to microbial degradation of hydrocarbons in polar regions (Antarctica and Arctic) received impetus mainly due to increase in human activities in polar regions and the urge to maintain the poles free of contaminations. Thus far, it has been established that hydrocarbon degradation in cold regions is mainly brought about by bacteria (Aislabie et al., 2006) and yeasts (Margesin et al., 2003a). The bacteria are affiliated to the genera Rhodococcus and Pseudomonas (Aislabie et al., 2000, 2006) and the yeasts to the genera Rhodotorula, Candida and Cryptococcus (Margesin & Schinner, 1997; Margesin et al., 2003a). However, in the cold regions, the rates of degradation are slow and it is dependent on temperature (Atlas, 1986; Margesin & Schinner, 1999; Rike et al., 2003, 2005), nutrient levels (Whyte et al., 2001), moisture and pH. Low temperatures, low nutrient levels, low moisture content and low pH reduce the rate of hydrocarbon degradation in polar soils (Aislabie et al., 2006). In the present study, the rate of degradation was not monitored but changes in bacterial populations were studied in seawater to which WSF of crude oil was added. Increase in bacterial density from 103 to 105 CFU mL−1 was observed when WSF of crude oil was added. The observed increase in bacterial numbers is in accordance with earlier studies which demonstrated that in polar regions numbers increase following hydrocarbon spillage (Atlas, 1986; Aislabie et al., 2004).

Earlier studies had indicated distinct and prominent changes in the bacterial community following biostimulation and/or bioaugmentation by hydrocarbons. But, in these earlier studies the exposure to hydrocarbon was for longer periods ranging from a few weeks to a year (Aislabie et al., 2000; Mohn et al., 2001; Thomassin-Lacroix et al., 2002; Coulon & Delille, 2003). In contrast, in the present study bacterial diversity following addition of WSF of crude oil to seawater showed significant changes though the treatment was only for 5 days. For instance, clones associated with Psychrobacter, Arcobacter, Formosa algae, Polaribacter, Ulvibacter and Tenacibaculum were found only in WSF added seawater. Increase in their numbers may imply that WSF addition enriched these clones but whether they are involved in biodegradation needs to be demonstrated. It was also observed that clones related to Psychromonas, Vibrio, Cytophaga and Maribacter which were present in seawater were no longer detectable in seawater with WSF of crude oil, probably because of toxic effects of the hydrocarbons present. However, a common feature was the predominance of unculturable bacteria in seawater independent of the presence of hydrocarbons. Some of these unculturable bacteria were phylogenetically related to Antarctobacter, Roseobacter, Staleya, Loktanella, Sulfitobacter and Agrobacterium. So far, none of these bacteria have been implicated in hydrocarbon biodegradation. Therefore, one of the challenge would be to cultivate these organisms so as to be able to understand better the role of these organisms in the ecosystem.

In summary, attempts have been made in this study to enumerate the biodiversity of Ushuaia off Argentina, a sub-Antarctic region. A preponderance of clones representing Proteobacteria (Alpha and Gamma) and Cytophaga–Flexibacter–Bacteriodetes group were observed. Clones with affiliation to Roseobacter and Sulfitobacter dominated both the libraries. Some unique clones identified with genera such as Loktanella, Oleispira, Ulvibacter, Tenacibaculum, Pelagibacter, Roseovarius etc., which were found to be present, as reported earlier, in other cold habitats. Few genera such as Arcobacter (Epsilonproteobacteria) were minimally represented and a few of the clones were affiliated to SAR11 clade.


We would like to thank The Indo-French Centre for the Promotion of Advanced Research, New Delhi, for funding this work.