Analysis of the sulfate-reducing bacterial and methanogenic archaeal populations in contrasting Antarctic sediments

Abstract

The distribution and activity of communities of sulfate-reducing bacteria (SRB) and methanogenic archaea in two contrasting Antarctic sediments were investigated. Methanogenesis dominated in freshwater Lake Heywood, while sulfate reduction dominated in marine Shallow Bay. Slurry experiments indicated that 90% of the methanogenesis in Lake Heywood was acetoclastic. This finding was supported by the limited diversity of clones detected in a Lake Heywood archaeal clone library, in which most clones were closely related to the obligate acetate-utilizing Methanosaeta concilii. The Shallow Bay archaeal clone library contained clones related to the C(1)-utilizing Methanolobus and Methanococcoides and the H(2)-utilizing Methanogenium: Oligonucleotide probing of RNA extracted directly from sediment indicated that archaea represented 34% of the total prokaryotic signal in Lake Heywood and that Methanosaeta was a major component (13.2%) of this signal. Archaea represented only 0.2% of the total prokaryotic signal in RNA extracted from Shallow Bay sediments. In the Shallow Bay bacterial clone library, 10.3% of the clones were SRB-like, related to Desulfotalea/Desulforhopalus, Desulfofaba, Desulfosarcina, and Desulfobacter as well as to the sulfur and metal oxidizers comprising the Desulfuromonas cluster. Oligonucleotide probes for specific SRB clusters indicated that SRB represented 14.7% of the total prokaryotic signal, with Desulfotalea/Desulforhopalus being the dominant SRB group (10.7% of the total prokaryotic signal) in the Shallow Bay sediments; these results support previous results obtained for Arctic sediments. Methanosaeta and Desulfotalea/Desulforhopalus appear to be important in Lake Heywood and Shallow Bay, respectively, and may be globally important in permanently low-temperature sediments.

FIG. 1.

Map of Signy Island, South Orkney Islands, Antarctica (60°43′S, 45°36′W), showing the positions of the two sampling sites, Lake Heywood and Shallow Bay, and the British Antarctic Survey research base.

FIG. 2.

Headspace accumulation of methane in sediment slurries from Lake Heywood (a) (methane production in AS_LH and AM_LH slurries was very similar to that in A_LH slurries, and so these data are not shown) and Shallow Bay (b) (methane production in S_SB, A_SB, AS_SB, and AF_SB slurries was very similar to that in control slurries, and so these data are not shown). Symbols: •, control; ○, acetate; □, sulfate; ▿, acetate plus chloroform; ▵, acetate plus fluoroacetate; ⋄, acetate plus molybdate. The results are the means for triplicate slurries with each treatment; error bars represent 1 SEM. Note the smaller scale on the y axis of the Shallow Bay plot (b).

FIG. 3.

Inferred phylogenetic relationships among archaeal environmental sequences, reference taxa, and other environmental clones and relative abundance of MA in RNAs extracted from the two Antarctic sites (EHB, 1MT*, 2MT*, and 2C* are clones from a United Kingdom estuary [39, 49]). (a) Logdet/paralinear distances tree based on 44% (the estimated number of variable sites) of 940 nucleotides that could be aligned. Bootstrap (1,000 replicates) values of >50% are shown. (b) Relative abundance of MA based on hybridization of MA-targeted oligonucleotide probes (Table 2) to rRNAs extracted from sediments from the two sites. Data are expressed relative to the combined signals for general bacterial and archaeal probes. Signal level means were determined after arcsine transformation of the data and then back-transformed to give the results shown (71). Results are the means of 18 extractions; error bars represent 1 SEM. M', Methano; LH, Lake Heywood; SB, Shallow Bay.

FIG. 4.

Inferred phylogenetic relationships among SRB-related environmental sequences from the two Antarctic sites, reference taxa, and other environmental clones and relative abundance of SRB in RNAs extracted from the two Antarctic sites (Sva clones were from Arctic sediments ([54]). (a) Logdet/paralinear distances tree based on 61% (the estimated number of variable sites) of 1,139 nucleotides that could be aligned. Bootstrap (1,000 replicates) values of >50% are shown. Known psychrophilic and psychrotolerant organisms are underlined. (b) Relative abundance of SRB based on hybridization of SRB-targeted oligonucleotide probes (Table 2) to rRNAs extracted from sediments from the two sites. Data are expressed relative to the combined signals for general bacterial and archaeal probes. Signal level means were determined after arcsine transformation of the data and then back-transformed to give the results shown (71). Results are the means of 18 extractions; error bars represent 1 SEM. D', Desulfo; LH, Lake Heywood; SB, Shallow Bay.