Distribution of Thiobacillus ferrooxidans and Leptospirillum ferrooxidans: Implications for Generation of Acid Mine Drainage

See allHide authors and affiliations

Science  06 Mar 1998:
Vol. 279, Issue 5356, pp. 1519-1522
DOI: 10.1126/science.279.5356.1519


Although Thiobacillus ferrooxidans andLeptospirillum ferrooxidans are widely considered to be the microorganisms that control the rate of generation of acid mine drainage, little is known about their natural distribution and abundance. Fluorescence in situ hybridization studies showed that at Iron Mountain, California, T. ferrooxidans occurs in peripheral slime-based communities (at pH over 1.3 and temperature under 30°C) but not in important subsurface acid-forming environments (pH 0.3 to 0.7, temperature 30° to 50°C). Leptospirillum ferrooxidans is abundant in slimes and as a planktonic organism in environments with lower pH. Thiobacillus ferrooxidansaffects the precipitation of ferric iron solids but plays a limited role in acid generation, and neither species controls direct catalysis at low pH at this site.

A fundamental component of the sulfur geochemical cycle is the release of sulfate into solution through oxidative dissolution of sulfide minerals. Because sulfides are at least a minor component of most rocks, this process is almost ubiquitous in chemical weathering. Weathering of sulfide-rich rocks with low neutralization capacity forms sulfuric acid–rich solutions that can carry high metal loads. When ore bodies are exposed by mining, this results in an environmental condition known as acid mine drainage (AMD).

Pyrite (FeS2) is the most abundant sulfide mineral in Earth's crust. Exposure of pyrite surfaces to oxygen and water results in the formation of sulfuric acid. Ferric iron, an abundant alternative electron acceptor in many AMD solutions, interacts effectively with surface sulfur species (1) and promotes pyrite dissolution by the following reaction: FeS2 + 14 Fe3+ + 8 H2O → 15 Fe2+ + 2 SO4 2- + 16 H+(2). Microorganisms greatly accelerate the rate of oxidation of Fe2+ to Fe3+, so that the rate of pyrite dissolution is generally controlled by microbial activity (2-4).

Numerous studies have measured and compared the abiotic, biotic, and Fe3+-induced rates of pyrite dissolution (2). Almost all experimental work has used Thiobacillus ferrooxidans, which is generally assumed to be the most important species accelerating the dissolution of metal sulfide (2).T. ferrooxidans is considered typical of AMD systems because it can be readily cultured from these environments. The importance of the iron-oxidizing species Leptospirillum ferrooxidans is now also widely accepted (5), and this species can outcompete T. ferrooxidans under certain conditions (6, 7). However, few studies have evaluated the potential geochemical impact of L. ferrooxidans in natural low-pH environments, and the distribution and abundance of these species have not been quantified.

We used molecular methods based on small-subunit ribosomal RNA (SSU rRNA) sequences (8) without prior cultivation (9) to study the role of microorganisms in an AMD environment at Iron Mountain, California, and to analyze the abundance and distribution of T. ferrooxidans and L. ferrooxidans as a function of geochemical and physical conditions. To determine the absolute contribution of these two bacterial species to the total microbial population, we evaluated the proportion of all cells in the domains Bacteria, Eukarya, and Archaea.

Iron Mountain is an in inoperative mine containing tens of kilometers of underground tunnels running through a sulfide ore body, as well as several runoff streams peripheral to the ore body. Pyrite-dominated sediments and solutions draining from the sulfide deposit were collected from the Richmond mine in January 1997. A few samples of seepage from tailings piles and storage tanks for AMD runoff from outside the mine (pH 2 to 4, temperature 10° to 25°C) were also collected and fixed for subsequent microbiological analysis (10).

The geochemical analyses from all sites fall into two clusters. Solutions with pH 1.5 to 2.5, temperature 17° to 30°C, and conductivity <30 mS/cm were confined to regions of the main tunnel and occasional pools. Solutions with pH 0.3 to 1.0, temperature 33° to 50°C, and conductivity >68 mS/cm were typical of most sites in contact with the ore body. Dissolved oxygen contents were higher in the higher temperature, lower pH regions, at ∼1.2 mg/liter at ∼20°C to ∼5.2 mg/liter. This is probably due to higher mixing that occurs at spillways.

Total cell counts were determined by DNA staining with 4′, 6-diamidino-2-phenylindole (DAPI) (7) (Fig.1). Cell numbers for rRNA probe–labeled samples relative to total cells were determined by dual counting of samples with differently labeled probes. Totals for the three domains should sum to the total detected with DAPI. In general, the total number of cells detected with rRNA probes was lower than that detected with DAPI, probably because some cells were dead or inactive.

Figure 1

Results from fluorescence in situ hybridization analysis (total cell counts determined by DNA staining with DAPI). The term “flow” refers to moving water, “pool” to standing water, “spill” to water actively flowing over a barrier in the Richmond ore body, “slime” to slime streamers on and in pools of standing water, and “sed.” to sediment consisting of pyrite accumulations on the floor of the drift. The term “matte” refers to a named body of standing water in a vertical shaft close to the mine entrance. Error bars were calculated from six to eight repetitions per sample (23) and ranged from ±2 to 15%, but were most commonly ±5 to 7%. Sediment numbers refer to cells attached to pyrite surfaces and in associated pore fluids. All Eukarya in these samples were in the solution fraction.

The A and C drifts are two of four horizontal tunnels that diverge from the horizontal Richmond entrance tunnel about 450 m into the mine. Solutions draining from, or collecting in, the A drift had temperatures between 42° and 45°C and pH values of 0.5 to 0.7. In the C drift, the temperature was 47° to 48°C and the pH was 0.4 to 0.6. Samples from all environments were found to contain abundant microbial life. In the A and C drifts, typical direct cell counts were 2.5 × 105/ml in solutions, 1.6 × 109/ml in slime streamers, and 4.2 × 106/ml in pyrite sediment.

Cells in sediment, water, or slime that hybridized with the T. ferrooxidans probe were completely absent. The probe effectively hybridized with cultured T. ferrooxidans cells [American Type Culture Collection (ATCC) number 19859], both in solution and on pyrite surfaces. The conclusion that T. ferrooxidans is not important in environments typified by the A and C drifts is supported by the absence of this species in enrichment cultures that used samples from the A drift as inoculum in standard T. ferrooxidansmedia (4). We found that in all cases, bacteria were the predominant form of microbial life (at least 75% of cells). Eukarya were minor constituents of many assemblages but ranged up to 25% of cells in some slimes. Archaea were a minor component in solutions.

The abundance of L. ferrooxidans in the A and C drifts varied with microenvironment (Fig. 1). This species accounted for almost the entire bacterial component of some slimes and was present in flowing and stagnant water (Fig. 2).Leptospirillum ferrooxidans has been cultured as a planktonic organism from these sites (11). Although only a few of the bacteria in sediments are L. ferrooxidans, this species occurs in relatively high numbers associated with, but unattached to, the sediments (∼105 cells/ml). Enrichment cultures also contained bacillus-shaped cells that colonized pyrite surfaces and hybridized with the bacterial probe (Fig.3) but not with the L. ferrooxidans or T. ferrooxidans probe. We have shown that the acidophilic mesothermophilic bacteria in these cultures are chemolithotrophic, metabolize ferrous iron, and accelerate pyrite dissolution rates (∼10 5 μmol of Fe per cell per day at pH 0.7 and temperature 42°C) (11).

Figure 2

Probe results for slime from the A drift. (A) Slime stained with DAPI. (B) Slime stained with LC206 (probe for L. ferrooxidans). The filaments on the right-hand side are Eukarya. Scale bar, 5 μm.

Figure 3

Bac338 (bacterial probe) to pyrite sediment. The image shows the colonized surface of a pyrite grain from sediment.

Bacteria were also the predominant form of microbial life (<<5% Archaea or Eukarya) in less extreme environments along the horizontal tunnel into the mine (∼20°C, pH 1.3 to 2.4; Fig. 1). However, in contrast to the situation in the pH < 1.0 environments, T. ferrooxidans was an important constituent and accounted for about one-third of the total population of pH > 1.0 slime communities. In addition, we successfully cultured T. ferrooxidans from these sites with the same standard culture medium used to test for this species in the A drift.

Because T. ferrooxidans is (i) not directly associated with the main ore body where primary oxidative dissolution is taking place and (ii) is a common inhabitant only of the more accessible, cooler, higher pH regions, we infer that the impact of this species on pyrite oxidation reactions in the mine is restricted.Thiobacillus ferrooxidans may be essentially an opportunist, deriving metabolic energy from dissolved Fe2+ but contributing little to acid generation at this site. This conclusion is consistent with the observation that conditions associated with the ore body are below the normal pH and above the normal temperature range for T. ferrooxidans(12). Thiobacillus ferrooxidans still has an important geochemical impact at this site because the oxidation of Fe2+ leads to precipitation of ferric iron solids, reducing the metal load in solutions. This potentially beneficial role differs considerably from the negative role often assigned to this species.

Leptospirillum ferrooxidans is extant over most of the range of conditions sampled. Although its distribution suggests that it plays an important ecological role in the microbial community by catalyzing sulfide mineral dissolution, its relative importance in the generation of AMD is not yet known. Our evidence suggests that this species is a dominant planktonic microorganism associated with the ore body, where conditions are generally >40°C and pH is 0.7 to 1.0. Leptospirillum ferrooxidans may be the species primarily responsible for catalysis of sulfide oxidation by aqueous ferric iron.

We have sampled the Iron Mountain site throughout the year. Our results show that substantial fluctuations in geochemical conditions are accompanied by variability in microbial population statistics. However, the key conclusions relating to the distribution of T. ferrooxidans and L. ferrooxidans are valid (13).

Although solutions draining most AMD sites have pHs of 2 to 4 (2), conditions may typically be more extreme close to reaction sites, as we have observed at Iron Mountain. Sulfuric acid–forming reactions are quite exothermic (14), and pHs in proximity to pyrite surfaces are likely much lower than those measured in bulk solution (2). Consequently, the organisms that are most important to sulfide dissolution may frequently encounter conditions similar to those found in the tunnels associated with the Iron Mountain ore body. Current models based on T. ferrooxidans should be reevaluated to reflect the involvement of different species promoting sulfide weathering by different mechanisms and at different rates.

  • * To whom correspondence should be addressed. E-mail: katrina{at}


View Abstract

Stay Connected to Science

Navigate This Article