Biodegradation of thiocyanate by a native groundwater microbial consortium

Groundwater sampling and storage

Mining-contaminated groundwater was extracted from a monitoring well located adjacent to a TSF at an operational gold mine in central Victoria, using a low flow pump. The well was screened in weathered granodiorite and schist, at a depth of 55 m below the surface. A sample of approximately 15 L was extracted, sealed and stored on ice until it was returned to the lab the next day, where it was refrigerated at 4 °C until use. The groundwater was used for enrichments within 3 days of sampling, while the remaining groundwater (with no trace metals or vitamins added) was used as a filter-sterilised medium for further culturing transfers, having been stored for up to 10 weeks in the dark at 4 °C by the end of the experiments. The chemistry of the groundwater is monitored frequently at the site and is typically moderately saline (TDS 11,425 ± 457), has a temperature of 15.2 °C ± 1.6, pH of 6.6–6.8, SCN⁻ concentrations of 500–1,000 mg L⁻¹ and free CN⁻ concentrations of <0.03 mg L⁻¹ (Table A1). We thank Stawell Gold Mine for access to the field site and historical monitoring data (no permit number issued).

Groundwater geochemical analyses

At the time of sampling, a flow cell was used to determine the pH, E_H and DO measurements taken with a YSI Professional Plus™ multi-parameter meter with calibrated probes. The groundwater was also analysed by colorimetry for SCN⁻ Sörbo (1957) and NH₄⁺ upon return to the laboratory, using the ferric-nitrate method (Eaton & Franson, 2005) and the salicylate-nitroprusside method (Baethgen & Alley, 1989), respectively. During laboratory-based experiments, pH was measured using a Thermo Orion 5 Star Plus™ Electrolyte Analyser and calibrated probes. Growth of the culture was monitored by tracking optical density at 600 nm (OD₆₀₀). All colorimetric analyses were conducted using a Hach DR2800™ Portable Spectrophotometer with standard solutions.

Aerobic and anaerobic enrichment culturing experiments

Groundwater was incubated under oxic and anoxic conditions, with various nutrient amendments, to enrich an SCN⁻-degrading culture. The oxic replicates were made by decanting groundwater (100 mL) under sterile conditions into triplicate autoclaved 250 mL conical flasks, sealed with a cotton wool bung and foil. Anoxic cultures were prepared by adding 30 mL of groundwater to 50 mL serum bottles sealed with rubber stoppers and aluminium crimps, and degassed using pressurised nitrogen gas. Both the oxic and anoxic cultures were further amended with additions of either 5 g L⁻¹ DOC (as glucose) or 50 mg L⁻¹ PO₄³⁻ (as NaH₂PO₄), or both, alongside no-addition controls. All incubations were maintained in the dark on a rotary shaker at 30 °C and 120 rpm. The SCN⁻ concentration was monitored to determine if degradation was occurring; in cases where complete or near complete SCN⁻ removal was noted, this enrichment provided inoculum for further culturing using filter sterilised (0.22 μm filter) groundwater as the medium. The amendments that produced a stable SCN⁻ degrading culture upon further culturing were selected for further study.

Oxic cultures were sampled by extracting two mL with a sterile syringe in a biosafety hood to ensure sterility. Anoxic cultures were sampled by extracting two mL of the culture with a N₂-degassed sterile syringe and needle, sampled through the rubber stopper. Half of the sample was passed through a 0.22 μm filter, while the other half was used for OD₆₀₀ measurement prior to freezing at −20 °C. Samples for DNA sequencing were removed at late-log phase growth from microbial cultures stably degrading SCN⁻ after five transfers, and immediately frozen at −80 °C until thawing for DNA extraction.

Culturing of a SCN⁻-degrading microbial consortium from groundwater

Amendments that resulted in a groundwater culture capable of SCN⁻ degradation after repeated culturing were further tested. All culturing after the initial enrichment phase was performed in sterilised 250 mL conical flasks, containing 100 mL of filter-sterilised ground water, stoppered using a cotton wool bung and foil, and with previously used nutrient amendments. For transfer to fresh medium, 10% v/v of the inoculum culture was sampled in late log phase of growth and incubated on a rotary shaker (or in a static incubator, in the case of anoxic cultures) at 30 °C and 120 rpm in the dark. Before subsequent testing, the culture was routinely re-cultured a minimum of five times to ensure a stable microbial community had developed. To determine the behaviour of the end-product, NH₄⁺, an identical culturing experiment to those previously described was set-up and samples removed to monitor SCN⁻, OD₆₀₀ and NH₄⁺.

Biodegradation of SCN⁻ in the presence of ammonium

A further experiment was set up to determine the impact of NH₄⁺ on SCN⁻ biodegradation. This experiment used the re-cultured SCN⁻-degrading microbial community and was again performed using filter-sterilised (0.22 μm filter) groundwater from the same well. As with previous experiments, this was performed in triplicate 250 mL conical flasks, containing 100 mL of filter-sterilised groundwater, stoppered using a cotton wool bung and foil. The flasks were amended to low (no addition), moderate (10 mg L⁻¹) and high (40 mg L⁻¹) concentrations of NH₄⁺ using a filter-sterilised (0.22 μm) concentrated (NH₄)₂SO₄ solution, in addition to five g L⁻¹ glucose and 50 mg L⁻¹ PO₄³⁻. An inoculum of the late log-phase culture was then added at 10% v/v concentration and incubated on a rotary shaker at 30 °C and 120 rpm in the dark.

Whole community microbial DNA extraction and Illumina MiSeq 16S and 18S rRNA gene sequencing

The triplicate samples for microbial ecology analysis were firstly removed from the −80 °C freezer and thawed. The genomic DNA was then extracted with the PowerSoil DNA Isolation Kit (Mo Bio Laboratories, Inc. Carlsbad, CA, USA). The primers used for 16S and 18S rRNA gene sequencing consist of partial Illumina adapter at the 5′ end. The incorporation of the second-half of the Illumina adapter and dual-index barcode was performed in another round of PCR reaction (Illumina 16S Sequencing Protocol). The 16S and 18S rRNA gene amplicon sequencing (Caporaso et al., 2012) was performed using on Illumina MiSeq platform (Illumina, San Diego, CA, USA) located at the Monash University Malaysia Genomics Facility (2 × 250 bp run configuration).

The 16S rRNA gene was amplified by PCR using primers targeting the V3-V4 region of the 16S gene: Forward 5′-CCTACGGGNGGCWGCAG-3′ and Reverse 5′-GACTACHVGGGTATCTAATCC-3′ (Klindworth et al., 2013). High-fidelity PCR was performed on one μL of each DNA sample using 0.5 μL of Illumina-compatible universal primers, under the following thermal cycler conditions: initial denaturation step (98 °C for 30 s), followed by 25 cycles of denaturation (98 °C for 10 s), annealing (60 °C for 30 s) and extension (65 °C for 60 s), followed by a final extension step (65 °C for 120 s). The product was further purified with 20 μL of Ampure 0.8×, and washed with 200 μL of 80% ethanol and eluted in 50 μL for subsequent index ligation using Nextera XT Index primers i7 forward and i5 reverse Illumina adapters. The subsequent product was purified with 12 μL of Ampure 0.8×, washed with 200 μL of 80% ethanol, and eluted in 30 μL for sequencing.

The 18S rRNA gene from the genomic DNA samples was amplified using the forward primer 1391f 5′-GTACACACCGCCCGTC-3′, and the reverse primer EukBr 5′-AGACAGTGATCCTTCTGCAGGTTCACCTAC-3′ (Amaral-Zettler et al., 2009). PCR was performed with one μL of the DNA extract in the presence of 10 μM Illumina-compatible primer, under the following thermal cycler conditions; initial denaturation (98 °C for 30 s), followed by 25 repetitions of denaturation (98 °C for 10 s), annealing (65 °C for 60 s) and extension (65 °C for 120 s) and a final extension (65 °C for 120 s). This PCR product was purified with 25 μL of Ampure 1× and washed with 200 μL of 80% ethanol, then eluted in 40 μL in preparation for index ligation, using Nextera XT Index i7 forward primer Nextera XT Index i5 reverse primer. The product was purified with 10 μL of Ampure 1, washed with 200 μL of 80% ethanol and eluted in 30 μL for 18S sequencing.

16S and 18S rRNA gene sequence analysis

Prior to any bioinformatic processing, the raw 18S and 16S rRNA gene sequences were uploaded to the National Centre for Biotechnology Information’s (NCBI’s) sequence read archive, with the BioProject accession number PRJNA356784. Analysis of the 18S and 16S rRNA gene sequencing data was performed using the QIIME software package in order to determine the phylogenetic structure of the microbial community (Caporaso et al., 2010a). Bases with a phred quality score of <20, and reads that retained less than 75% of their original sequence length, were discarded. Forward and reverse reads of the 16S and 18S rRNA genes were joined at paired ends and aligned. The sequences were demultiplexed, filtered and processed through the QIIME software package (Caporaso et al., 2010a).

The 16S rRNA gene sequences were compared to those in the GreenGenes Bacterial and Archaeal 16S rRNA gene database (DeSantis et al., 2006) using BLAST (Altschul et al., 1990), picking operational taxonomic units (OTUs) at a 97% similarity cut-off. Representative sequences from each OTU were aligned using the PyNAST tool (Caporaso et al., 2010b), and chimeric sequences were identified and removed using ChimeraSlayer (Haas et al., 2011). The 18S rRNA gene sequences were assigned to OTUs via a de novo approach using USEARCH v5.2.236 (Edgar, 2010). Representative sequences from each OTU were checked for chimeric sequences, and these were removed using UCHIME v6.1.544 (Edgar et al., 2011). The resulting OTUs were assigned taxonomy by comparison to the SILVA 16S/18S rRNA gene database (SILVA 119, Quast et al., 2013) using Blastall v2.2.22. All OTUs that were assigned to prokaryotic taxa were then removed from the 18S rRNA gene dataset.

Operational taxonomic units representing >1% abundance were processed through the NCBI BLAST program and assigned taxonomies according to highest sequence similarity. The resulting BLAST assigned identities were compared to the taxonomic identities assigned by the GreenGenes and SILVA databases for the 16S rRNA (Table A2) and 18S rRNA sequencing (Table A3), respectively.

Groundwater chemistry

The geochemical conditions of the groundwater at the time of sampling are presented in Table 1. The groundwater pH was slightly acidic and contained SCN⁻ in addition to a small concentration of NH₄⁺. The prevailing redox conditions in the groundwater were reducing, with low oxygen levels.

Groundwater enrichment culturing experiments

During the initial enrichment experiment, no SCN⁻ removal was noted in the absence of oxygen, regardless of nutrient amendment (see Table 2 for initial and final SCN⁻ concentrations). In the oxic enrichment experiments, SCN⁻ removal was observed in the absence of any nutrient amendment; however, upon inoculation of this culture into filter-sterilised groundwater, no SCN⁻ removal was observed. The sole addition of DOC or PO₄³⁻ also resulted in SCN⁻ biodegradation in the initial enrichment, but when transferred into filter-sterilised groundwater, SCN⁻ degradation did not occur. The only condition to result in a SCN⁻ degrading culture, which was culturable in filter-sterilised groundwater, was the addition of DOC and PO₄³⁻ in the presence of air. Initially, complete removal of SCN⁻ (from approximately 130 mg L⁻¹) was achieved within 4 days, through combined addition of DOC and PO₄³⁻. This culture was used for subsequent experiments.

SCN⁻ and NH₄⁺ biodegradation by a consortium of groundwater microorganisms

The enriched microbial consortium, amended with DOC and PO₄³⁻, was further investigated to determine the fate of the NH₄⁺ released by SCN⁻ degradation. The consortium completely degraded SCN⁻ in the filter-sterilised groundwater within a period of 50 h (Fig. 1). The initial NH₄⁺ present in the groundwater was consumed prior to any SCN⁻ removal. After this, an increase in OD₆₀₀ was noted, in tandem with the consumption of SCN⁻ and formation of NH₄⁺. The concentration of NH₄⁺ decreased to below detection after all SCN⁻ had been consumed.

Inhibition of SCN⁻ biodegradation by NH₄⁺ addition

Further experimentation was conducted to determine the effect NH₄⁺ had upon SCN⁻ biodegradation. This work revealed that low to moderate concentrations of NH₄⁺ did not inhibit biodegradation of SCN⁻ (Fig. 2). The highest NH₄⁺ concentration however, completely inhibited SCN⁻ biodegradation. SCN⁻ degradation occurring at lower NH₄⁺ concentrations only proceeded after complete NH₄⁺ removal (Fig. 3). Potentially, SCN⁻ biodegradation could have occurred more efficiently after repeated transfers to NH₄⁺-containing media, but this adaptation hypothesis was not tested in this study.

Microbial community characterisation by 16S rRNA gene sequencing

The taxonomic assignments for the 16S rRNA gene sequences, from the enriched groundwater community are given in Fig. 4A. At the phylum level, the microbial community enriched through DOC and PO₄³⁻ addition and exposure to air in the groundwater was dominated by Proteobacteria (72.8%) and Bacteroidetes (25.8%), with a minor proportion of Actinobacteria (1.3%). The dominant families within the Proteobacteria were Phyllobacteriaceae (27.8%), Rhodobacteriaceae (12.8%) and Sphingomonadaceae (12.5%). The latter was entirely assigned to the Novosphingobium genus, and the dominant OTU found to be most closely related to Novosphingobium panipatense strain UMTKB-4 (99%), by comparison to NCBI and Greengenes databases (Tables A4 and A5). The Phyllobacteriaceae family was dominated by a single unclassified OTU (27.8%), most closely related to an uncultured Mesorhizobium sp. clone S3_F08 (99% similarity). The two dominant OTUs for the Rhodobacteraceae family were most closely related to an uncultured bacterium clone MAL_E01 (12.8% abundance, 99% similarity), the higher abundance of the two OTUs had equal sequence similarity to an environmental sample, Thioclava pacifica (98% similarity) known to be capable of sulphur oxidation and consumption of simple organics (Sorokin et al., 2005). In addition to these high abundance members, sequences assigned to the genera Martelella sp. (4.0%) and Xanthobacter (5.4%) made up significant minority taxa.

The Bacteroidetes phylum was largely dominated by a single OTU unassigned using the Greengenes database below family level, but most closely related to an uncultured Owenweeksia sp. Clone (99% similarity). Other Bacteroidetes sequences assigned to the Flammeovirgaceae family belong to the Roseivirga genus.

Microbial community characterisation by 18S rRNA gene sequencing

The 18S rRNA gene analysis showed a simple eukaryotic distribution (Fig. 4B), with only two unique OTUs identified: Tremella indecorata (98.5%), a fungus of the Basidiomycota phylum and Jakoba libera (1.5%), a trophic flagellate from the Loukozoa phylum. The SILVA assigned taxonomies for these OTUs were compared to the identities assigned by the NCBI database by BLAST (Table A5). The dominant OTU, Tremella indecorata was most closely related as Tremellales sp. LM630 (99% similarity), while the less abundant OTU was most closely related to J. libera at 99% gene sequence similarity, which is consistent with the SILVA identity.