Environ. Sci. Technol. 2007, 41, 5323-5329
Biogeochemistry of Metalliferous Peats: Sulfur Speciation and Depth Distributions of dsrAB Genes and Cd, Fe, Mn, S, and Zn in Soil Cores C A R M E N E N I D M A R T IÄ N E Z , * C A R O L I N A Y AÄ N ˜ EZ, SOH-JOUNG YOON, AND MARY ANN BRUNS Department of Crop and Soil Sciences, The Pennsylvania State University, University Park, Pennsylvania 16802
Spatial relationships between concentrations of Cd, Fe, Mn, S, and Zn and bacterial genes for dissimilatory sulfate reduction were studied in soils of the Manning peatland region in western New York. Peat cores were collected within a field exhibiting areas of Zn phytotoxicity, and pH and elemental concentrations were determined with depth. The oxidation states of S were estimated using S-XANES spectroscopy. Soil microbial community DNA was extracted from peat soils for ribosomal RNA intergenic spacer analysis (RISA) of diversity profiles with depth. To assess the presence of sulfate-reducing microorganisms (SRM), DNA extracts were also used as templates for PCR detection of dsrAB genes coding for dissimilatory (bi)sulfite reductase. Elemental distributions, S redox speciation, and detection of dsrAB genes varied with depth and water content. The pH of peat soils increased with depth. The highest concentrations of Zn, Cd, and S occurred at intermediate depths, whereas Mn concentrations were highest in the topmost peat layers. Iron showed a relatively uniform distribution with depth. Concentrations of redox sensitive elements, S and Mn, but not Fe, seemed to respond to variations in water content and indicated vertical redox stratification in peat cores where topmost peats were typically acidic and oxidizing and deeper peats were typically circumneutral and reducing. Even then, S-XANES analyses showed that surface peats contained >50% of the total S in reduced forms while deep peats contained generally 50% reduced S but no ZnS). Most importantly, linear combination fit analyses (Table S3) of the S-XANES spectra (Figure S2) revealed that organic forms of reduced S (cysteine, cystine) are major constituents of surface peats while inorganic forms of reduced S (ZnS) dominate in deep peats. RISA and DsrAB Gene Detection. RISA fingerprints of peat community DNA were used to assess gross, qualitative differences in bacterial populations with depth and season. Visible PCR products in RISA gels ranged from 500 to 1300 bp in length, but the primers’ broad-specificity and high soil community complexity resulted in many fingerprints having few to no sharp DNA bands (Figure 3, Panel A). Each RISA fingerprint was unique, however, indicating high spatial heterogeneity and seasonal differences in bacterial community composition. In PCR-based fingerprinting that employs broad-specificity primers, a bacterial population must comprise at least 1-5% of the total bacterial DNA before it will yield a discernible DNA band (29). Even though SRM have been cultured from oxic, surface soils, SRM numbers were expected to be too low to yield DNA bands in RISA fingerprints. To assess SRM presence, therefore, we used the same community DNA extracts in PCR with primers specific for dsrAB (8), a gene shared by all recognized groups of SRM (7, 30). All six bottom samples and five deep samples yielded dsrAB-PCR products of the expected size (1.9 kb), providing evidence for the presence of SRM at depths greater than 35 cm (Figure 3, Panel B). However, bands ranging from 600 to 1500 bp (shorter than the expected size of 1900 bp) can be seen in 1 of 9 dry season cores and 8 of 9 wet season cores. These are nonspecific amplification products, which commonly occur when using PCR conditions of only moderate stringency VOL. 41, NO. 15, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5327
to improve the probability of amplifying dsrAB genes from uncharacterized organisms. We ignore these nonspecific amplification products, and consider only DNA bands of the expected lengths to indicate potential presence of dsrAB genes. Because mere presence of DNA bands of appropriate length is not confirmatory evidence, the bands were excised for DNA sequencing. We obtained confirmatory evidence for the presence of dsrAB genes from DNA sequences of the amplified products (accession numbers DQ855242 to DQ855261), which were homologous and similar to dsrAB sequences in GenBank. In contrast, only one of six surface samples yielded amplicons from primers for dsrAB. Oxic conditions created by agricultural drainage, crop uptake of water, and evapotranspiration would prohibit anaerobic bacterial activity in the upper portions of the peat profiles and likely favor the persistence of sporeforming (e.g., Desulfotomaculum spp.) over non-sporeforming groups of SRM (e.g., Desulfovibrio spp.). In comparing RISA patterns with the presence or absence of dsrAB-PCR products for the same sample, no consistent associations could be discerned (Figure 3, Panels A and B). DNA sequencing of RISA bands from dsrAB-positive samples, however, could yield additional evidence of SRM or shed more light on bacterial populations possibly associated with SRM. Further studies are needed to elucidate phylogenetic affiliations and functions of bacterial populations in these peat soils. Biogeochemical Cycles in the Manning Peatlands. Our chemical data (pH, Cd, Fe, Mn, S, and Zn profiles) represented the cumulative legacy of fluctuating oxidation-reduction processes over time. These data indicated acidic and oxidizing conditions in the upper portions of peat cores and circumneutral and reducing conditions in the lower portions. All peat cores were characterized by the highest S, Zn, and Cd concentrations occurring at depths below ∼35 cm, while the highest Mn concentrations occurred in topmost peats (
