Article pubs.acs.org/est
Iron Kinetics and Evolution of Microbial Populations in Low-pH, Ferrous Iron-Oxidizing Bioreactors Rose M. Jones and D. Barrie Johnson* College of Natural Sciences, Bangor University, Deiniol Road, Bangor LL57 2UW, United Kingdom S Supporting Information *
ABSTRACT: Iron-rich, acidic wastewaters are commonplace pollutants associated with metal and coal mining. Continuous-flow bioreactors were commissioned and tested for their capacities to oxidize ferrous iron in synthetic and actual acid mine drainage waters using (initially) pure cultures of the recently described acidophilic, iron-oxidizing heterotrophic bacterium Acidithrix ferrooxidans grown in the presence of glucose and yeast extract. The bioreactors became rapidly colonized by this bacterium, which formed macroscopic streamer growths in the flowing waters. Over 97% of ferrous iron in pH 2.0−2.2 synthetic mine water was oxidized (at up to 225 mg L−1 h−1) at dilution rates (D) of 0.6 h−1. Rates of iron oxidation decreased with pH but were still significant, with influent liquors as low as pH 1.37. When fed with actual mine water, >90% of ferrous iron was oxidized at D values of 0.4 h−1, and microbial communities within the bioreactors changed over time, with Atx. ferrooxidans becoming increasingly displaced by the autotrophic iron-oxidizing acidophiles Ferrovum myxofaciens, Acidithiobacillus ferrivorans, and Leptospirillum ferrooxidans (which were all indigenous to the mine water), although this did not have a negative impact on net ferrous-iron oxidation. The results confirmed the potential of using a heterotrophic acidophile to facilitate the rapid commissioning of ironoxidizing bioreactors and illustrated how microbial communities within them can evolve without compromising the performances of the bioreactors.
■
INTRODUCTION Low-pH, ferruginous (iron-rich) waters are generated in many abandoned metal and coal mines, as well as in mineral wastes produced by mining (rock dumps and tailings), in many locations throughout the world. These waters, often referred to as acid mine drainage (AMD), are potentially highly polluting of other streams, rivers, and marine waters into which they flow. This is not only a consequence of their acidity (pH 2−4 is typical of AMD streams, though more acidic waters have been described)1 but also because they contain elevated concentrations of dissolved metals and metalloids, some of which (e.g., cadmium and arsenic) are extremely toxic to most life-forms. Mine-water streams may also have relatively high osmotic potentials due to the presence of large concentrations of dissolved inorganic solutes; sulfate is almost always the dominant anion in AMD. Various remediation strategies have been devised and implemented to remediate AMD.2 These use either chemical (e.g., lime addition) or biological (e.g., constructed wetlands) approaches to immobilize metals and to neutralize acidity, although both have major detractions in terms of costs and sustainabilities. In most AMD streams, the most abundant metal is iron, which, in low-pH waters (1.2 g L−1 copper, ∼26 g L−1 ferrous iron, 5.5 g L−1 manganese, ∼ 3 g L−1 nickel, and >3 g L−1 zinc, which is wellabove the concentrations of these metals present in most AMD waters.1 Atx. ferrooxidans biomass established rapidly in newly commissioned FOBs, and macroscopic streamer growths were observed shortly after the reactors were operated in continuousflow mode. The time taken to commission these bioreactors (4− 14 days) was considerably shorter than for similar systems that used the autotrophic iron-oxidizing bacterium Fv. myxofaciens (∼100 days),17 thereby supporting the first part of the hypothesis tested that, by using a heterotrophic iron-oxidizing acidophile as the initial microorganism, it would be possible to greatly reduce the time required to establish an effective iron-oxidizing bioreactor. The initial tests with the FOBs, carried out using synthetic acidic ferrous iron liquors, confirmed that these were effective at oxidizing ferrous iron, with 97−98% of iron oxidized at dilution rates of between 0.05 and 0.56 h−1. Rates of ferrous iron oxidation (up to 0.23 g L−1 h−1) were are the lower end of the range reported elsewhere (0.13−1.60 g L−1 h−1),5−7,17 although the current tests were carried out at lower (and suboptimal) temperatures using bioreactors that were only partially colonized (in spatial terms) by iron-oxidizing acidophiles. These tests also confirmed that some ferrous iron was oxidized even when the pH of the influent liquor was as low as pH 1.37, although the rates and percentage of ferrous iron oxidation were much lower below pH 2. The oxidation of one mole of ferrous iron consumes one mole of protons (reaction 1), and the hydrolysis of one mole of ferric iron (to form the mineral schwertmannite) generates 2.75 mol of protons (reaction 2), so the combination of the two reactions is net-acid-producing. However, both rates and extents of ferrous-iron oxidation were much greater than those of ferriciron hydrolysis (e.g., while 98% of the ferrous iron present in the synthetic feed liquors was typically oxidized, only ∼3.6% of the ferric iron generated was precipitated inside the bioreactors). Because most of the ferric iron generated in the FOBs remained in solution in the freshly produced effluent liquors, this should have resulted in the net consumption of protons and, therefore, a higher pH in the effluent than the influent liquors, although in actuality, only minor changes in pH values were observed with both the synthetic liquors and actual AMD. The FOBs were also highly effective in oxidizing the ferrous iron present in AMD from the abandoned copper mine (∼200 L
were processed by each bioreactor during the course of the experiment). The main differences between the AMD and the synthetic test water were: (i) the AMD had a higher and nearconsistent pH of ∼2.6; (ii) the AMD contained elevated concentrations of several transition metals (copper, ferric iron, and zinc) that were not present in the synthetic test waters; and (iii) the AMD used, like most others, contained very little dissolved organic carbon (∼5 mg L−1), not all of which would be anticipated to be metabolized by heterotrophic acidophiles such as Atx. ferrooxidans. In contrast, the theoretical DOC of the synthetic test liquors was 410 mg L−1, most of which would be expected to be utilized by the bacteria. This means that conditions within the FOBs were far less conducive for the growth of heterotrophic iron oxidizers when AMD was used as the influent liquor. The T-RFLP profiles found in the Afon Goch AMD were quite different from those reported elsewhere.13 The reason for this is unclear but may reflect the fact that the samples for the present work were taken during the summer months (June and July). Although the temperature of the upper reaches of the Afon Goch shows little seasonal variation (as it is dictated by the nearconstant temperature of the underground lake from which it arises),13 the much-greater solar intensities in the summer months frequently correspond to a bloom of acidophilic microalgae (chiefly Euglena) in the stream. Exudates and lysates from photosynthetic primary producers could conceivably support populations of heterotrophic acidophiles that are lessabundant at other times of the year. Unfortunately, most of the T-RFs in T-RFLP profiles in the AMD were not identified, although one was confirmed to correspond to Metallibacterium scheffleri, a moderately acidophilic heterotrophic bacterium that had previously been reported in mine waters in the United Kingdom18 as well as Germany.15 As expected, many of the same T-RFs were also found in oxidized AMD generated by the FOBs, although only one of the bacteria represented by these (the T-RF of 75 nt) was established as part of the streamer communities within the bioreactors. The three bacteria that dominated the abundant streamer growths in the Afon Goch drainage channel13 (Fv. myxofaciens, At. ferrivorans, and Atx. ferrooxidans) were all detected in the AMD samples by T-RFLP analysis, albeit each in small relative abundance. Although Atx. ferrooxidans was the only planktonic bacterium detected in liquors draining the FOBs prior to the passage of AMD, its relative abundance decreased in T-RFLP profiles with time, and it was not detected in those oxidized AMD after passage of collection batch 2 in FOB#1. However, analysis of the streamer communities in both FOBs at the end of the experiment confirmed that this acidophile was still present, and presumably active, in significant relative abundance in both bioreactors. Results from T-RFLP analysis confirmed that the indigenous bacterial communities in both bioreactors had undergone considerable, although different, transitions as a consequence of being exposed to, and oxidizing, unfiltered AMD (i.e., containing indigenous microflora). The presence of known autotrophic iron-oxidizers in the streamer communities in both bioreactors by the end of the experiment was particularly noteworthy. These were: (i) Fv. myxofaciens, present as the dominant bacterium in FOB#1 and which has also been used elsewhere as an effective bacterium for mine-water treatment;9 (ii) At. ferrivorans (in much smaller relative abundance) in FOB#2, and (iii) L. ferrooxidans, which was present in both. Although L. ferrooxidans was not detected by the T-RFLP F
DOI: 10.1021/acs.est.6b02141 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
(2) Johnson, D. B.; Hallberg, K. B. Acid mine drainage: remediation options. Sci. Total Environ. 2005, 338, 3−14. (3) Dopson, M. Physiological and phylogenetic diversity of acidophilic bacteria. In Acidophiles: life in Extremely Acidic Environments; Quatrini, R., Johnson, D. B., Eds.; Caister Academic Press: Norfolk, England, 2016; pp 79 − 92. (4) Kelly, D. P. Bioenergetics of chemolithotrophic bacteria. In Companion to Microbiology; Selected Topics for Further Discussion; Bull, A. T., Meadow, P. M., Eds.; Longman: London, 1978; pp 363−386. (5) Armentia, H.; Webb, C. Ferrous sulphate oxidation using Thiobacillus ferrooxidans cells immobilised in polyurethane foam support particles. Appl. Microbiol. Biotechnol. 1992, 36, 697−700. (6) Breed, A. W.; Hansford, G. S. Effect of pH on ferrous-iron oxidation kinetics of Leptospirillum ferrooxidans in continuous culture. Biochem. Eng. J. 1999, 3, 193−201. (7) Rowe, O. F.; Johnson, D. B. Comparison of ferric iron generation by different species of acidophilic bacteria immobilised in packed-bed reactors. Syst. Appl. Microbiol. 2008, 31, 68−77. (8) Johnson, D. B.; Hallberg, K. B.; Hedrich, S. Uncovering a microbial enigma: isolation and characterization of the streamer-generating, ironoxidizing acidophilic bacterium “Ferrovum myxofaciens”. Appl. Environ. Microbiol. 2014, 80, 672−680. (9) Hedrich, S.; Johnson, D. B. Remediation and selective recovery of metals from acidic mine waters using novel modular bioreactors. Environ. Sci. Technol. 2014, 48, 12206−12212. (10) Jones, R. M.; Johnson, D. B. Acidithrix ferrooxidans gen. nov., sp. nov.; a filamentous and obligately heterotrophic, acidophilic member of the Actinobacteria that catalyzes the dissimilatory oxido-reduction of iron. Res. Microbiol. 2015, 166, 111−120. (11) Ň ancucheo, I.; Rowe, O. F.; Hedrich, S.; Johnson, D. B. Solid and liquid media for isolating and cultivating acidophilic and acid-tolerant sulfate-reducing bacteria. FEMS Microbiol. Lett. 2016, 363, fnw083. (12) Stookey, L. Ferrozine − a new spectrophotometric reagent for iron. Anal. Chem. 1970, 42, 779−781. (13) Kay, C. M.; Rowe, O. F.; Rocchetti, L.; Coupland, K.; Hallberg, K. B.; Johnson, D. B. Evolution of microbial “streamer” growths in an acidic, metal-contaminated stream draining an abandoned underground copper mine. Life 2013, 3, 189−210. (14) Johnson, D. B.; Hallberg, K. B. Techniques for detecting and identifying acidophilic mineral-oxidising microorganisms. In Biomining, Rawlings, D. E, Johnson, D. B., Eds.; Springer-Verlag: Heidelberg, Germany, 2007; pp 237−262. (15) Ziegler, S.; Waidner, B.; Itoh, T.; Schumann, P.; Spring, S.; Gescher, J. Metallibacterium schef f leri gen. nov., sp. nov., an alkalinizing gammaproteobacterium isolated from an acidic biofilm. Int. J. Syst. Evol. Microbiol. 2013, 63, 1499−1504. (16) Kinnunen, P. H-M.; Puhakka, J. A. High-rate ferric sulfate generation by a Leptospirillum ferriphilum-dominated biofilm and the role of jarosite in biomass retainment in a fluidized-bed reactor. Biotechnol. Bioeng. 2004, 85, 697−705. (17) Hedrich, S.; Johnson, D. B. A modular continuous flow reactor system for the selective bio-oxidation and precipitation of iron in mineimpacted waters. Bioresour. Technol. 2012, 106, 44−49. (18) Hallberg, K. B.; Johnson, D. B. Novel acidophiles isolated from moderately acidic mine drainage waters. Hydrometallurgy 2003, 71, 139−148. (19) Heinzel, E.; Janneck, E.; Glombitza, F.; Schlömann, M.; Seifert, J. Population dynamics of iron-oxidizing communities in pilot plants for the treatment of acid mine waters. Environ. Sci. Technol. 2009, 43, 6138− 6144. (20) Heinzel, E.; Hedrich, S.; Janneck, E.; Glombitza, F.; Seifert, J.; Schlomann, M. Bacterial diversity in a mine water treatment plant. Appl. Environ. Microbiol. 2009, 75, 858−861.
analysis of Mynydd Parys AMD in the present study, it has been in others.13 Data from these experiments also supported the second part of the hypothesis tested, which was that the microbial communities within the bioreactors would evolve from predominantly heterotrophic to autotrophic iron oxidizers with time due to the fact that the AMD used contained little DOC. It was reassuring to note that this transition appeared to be gradual and did not at any time result in a downturn in the performances of the FOBs. Once established, the autotrophic populations would be expected to persist as dominant microflora, barring any incident like a periodic influx of AMD with an elevated DOC content. The microbial communities in other bioreactors used to catalyze the oxidation of ferrous iron have also been reported to change with time. For example, the pilot-scale systems used to oxidize ferrous iron in moderately acidic groundwater from a lignite mine at Nochten, Germany that were originally inoculated with Acidithiobacillus and Leptospirillum spp. were found within a relatively short time to be dominated by Fv. myxofaciens and bacteria related to Gallionella.19,20 The oxidation of ferrous iron is, however, only the first step in remediation of ferruginous mine waters. Hedrich and Johnson 17 described a sequential remediation process that involved three modules: (i) a ferrousiron-oxidizing bioreactor; (ii) an inorganic reactor where NaOH was added to the oxidized mine water at a controlled pH of 3.0− 3.5, to precipitate schwertmannite; and (iii) a “polishing” biofilm reactor, which removed residual soluble iron to concentrations below regulatory discharge consent levels. The FOBs described in the current work would readily fit into such a composite system and allow this module to be rapidly commissioned and installed in a mine-water-impacted environment.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b02141. Tables showing hydraulic and chemical parameters of FOB #1 and FOB #2 operated at different flow rates and chemical and hydraulic data from the Atx. ferrooxidans bioreactors operated at different flow rates, FOB #1, and FOB #2. Images showing the bioreactor, feed and drain vessels, and pumps used to input and to drain the reactor vessel; FOBs after processing; and T-RFLP profiles of bacterial 16S rRNA genes. (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*Telephone: +44 1248 382358; fax: +44 1248 370731; e-mail: d.
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS R.M.J. is grateful to the European Union and Nelsons UK for the provision of a research studentship under the Knowledge Economy Skills Scholarship (KESS) scheme.
■
REFERENCES
(1) Nordstrom, D. K.; Blowes, D. W.; Ptacek, C. J. Hydrogeochemistry and microbiology of mine drainage: an update. Appl. Geochem. 2015, 57, 3−16. G
DOI: 10.1021/acs.est.6b02141 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
