Rates of bacteria-promoted solubilization of Fe from minerals: a review of problems and approaches

Abstract

Understanding the effects of bacteria on mineral weathering will require careful, controlled experiments in chemical reactors. Unfortunately, many of the details of protocol for such experiments have not yet been addressed. In particular, experimenters need to find ways to normalize experiments for the mass of bacteria involved in reaction and need to measure the rate of change of this mass. Furthermore, experimenters will need to define whether mineral dissolution occurs during the log or stationary phase of growth. Bacteria cell mass should be quantified both before and after dissolution. In some cases, flow bioreactors will be useful to understand dissolution under steady state conditions. Finally, even the choice of medium will involve trade-offs between encouraging growth of bacteria while mimicking natural solutions.

In the case study for mineral dissolution presented here, two bacterial species of the genera Streptomyces and Arthrobacter, each involved in the natural weathering of hornblende, were investigated in growth experiments in medium with hornblende. Experiments with and without desferrioxamine B (its mesylate salt deferoxamine mesylate, DFAM) were also completed. In the presence of bacteria or DFAM, Fe release from hornblende is accelerated by up to a factor of ∼20. Both bacteria produce catecholamide siderophores that are presumed responsible for the enhanced Fe release when bacteria are present. These results represent the first quantification of the rate of release of Fe from a mineral in the presence of the siderophore-producing bacteria, with subsequent identification of the siderophore. The rate of Fe release from hornblende increases non-linearly in bacteria-free experiments with increasing concentrations of DFAM. Such a rate–concentration function is consistent with formation of surface complexes on the hornblende surface. Surface complexation may also explain enhancement of Fe release in the presence of the catechol siderophores produced by the arthrobacter and streptomycete. The effects of siderophores on Fe transport could be significant in many soils.

Introduction

Few studies of microbe-influenced mineral dissolution have yielded quantitative information with respect to both rate and mechanism of dissolution (e.g., compare references such as Duff et al., 1963, Daragan, 1971, Kutuzova, 1973, Lyalikova and Petushkova, 1991, Sand and Bock, 1991, Grantham and Dove, 1996, Ullman et al., 1996, Barker et al., 1997, Barker et al., 1998, Forsythe et al., 1998, Kostka et al., 1999). In contrast, several investigators have quantitatively measured the effect of organic acids produced by bacteria, lichens and fungi on mineral weathering (e.g., Rozycki and Strzelczyk, 1986, Barman et al., 1992, Vandevivere et al., 1994, Watteau and Berthelin, 1994, Drever and Stillings, 1996, Holmén and Casey, 1996, Stone, 1997). Production of dissolved organic molecules can create surface complexes that promote dissolution (e.g., Holmén et al., 1997, Holmén et al., 1999, Roden and Urrutia, 1999, Kalinowski et al., 2000).

The relative lack of quantitative experiments dealing with mineral dissolution in the presence of bacteria is partially related to experimental difficulties. Given that measurements of abiotic mineral dissolution are difficult to replicate (White and Brantley, 1995), it is not surprising that many problems remain in development of appropriate techniques for measurement of dissolution under biotic conditions. It is the intent of this paper to summarize one case study investigating mineral dissolution Liermann et al., 2000, Kalinowski et al., 2000, and to discuss experimental details in this case study with an eye toward elucidating how such experiments might be improved in the future. The emphasis of our analysis is aimed at investigations of siderophore-promoted mineral dissolution in the presence of bacteria.

The case study we present in this paper focuses on microbial processes that contribute to the dissolution of hornblende, (Na,K)_0–1(Ca,Na)_2–3(Mg,Fe,Al)₅Si₆(Si,Al)₂O₂₂(OH)₂. Hornblende, a common soil mineral in granitic lithologies, is a major contributor of Fe and Mg to natural waters (Huang, 1977), and its dissolution plays a role in the long-term CO₂ cycle due to the release of Ca and Mg. We have investigated the mechanisms by which soil bacteria mobilize Fe, Al and Si from hornblende. Of the metal nutrients, Fe is important because it is essential to virtually all life forms (Madigan et al., 1997). Dissolution of hornblende investigated under three conditions as reported by Liermann et al. (2000) and Kalinowski et al. (2000) are summarized here: (1) abiotic without siderophore, (2) abiotic but in the presence of a commercially available siderophore, and (3) biotic (in the presence of bacteria). We also report new results from flow experiments, and from further analysis of the siderophore produced by one of the bacterial isolates.

Siderophores are chelating agents secreted by bacteria and fungi with formation constants for ferric iron in the range 10²⁵ to 10³⁵, and in exceptional cases as high as 10⁵¹Hider, 1984, Hughes and Poole, 1989, Winkelmann, 1991, Hersman et al., 1995. The Fe complex formation constant for the trihydroxamate desferrioxamine B (Fig. 1) used in the case study presented here is 10^30.6(Winkelmann, 1991). Siderophores bind Fe(III) in soils more effectively than low molecular weight organic acids. For example, the association constants of oxalic and citric acids with Fe³⁺ are 10^7.6 and 10^17.3, respectively (Perrin, 1979). Siderophores generally form 1:1 complexes with Fe³⁺Jalal and vander Helm, 1991, Matzanke, 1991.

The siderophore–Fe(III) complexes are typically taken up by the cell membrane of bacteria, where the Fe is reduced and released from the siderophore into the cell (Fig. 2). In some cases, the siderophore is destroyed during this reduction and in other cases the molecule is recycled (Madigan et al., 1997). The two most common groups of siderophores are the hydroxamates and the catecholamides, which both can form hexadentate siderophore–metal complexes (Fig. 2). The functional group in hydroxamates is hydroxamic acid, which is a carbonyl oxygen combined with an amino group. The catecholamide ligands have adjacent hydroxyl oxygens on an aromatic ring (=catechol).

Previously, it was believed that only fungi produced hydroxamate siderophores and that bacteria produced catecholate siderophores. However, this statement has been revised (Winkelmann, 1991), since many bacterial species have been found to produce hydroxamates. Siderophores also bind ions other than Fe³⁺ such as Al³⁺, Ga³⁺, Cu²⁺ and some actinides, but with lower affinity Anderegg et al., 1963, Hernlem et al., 1996.

Siderophore production by microorganisms is highly regulated by iron availability in the surrounding environment. Under conditions in which siderophore is available, further production of siderophore should be unnecessary. For example, isotopic labeling and Fe uptake studies have shown that species of Streptomyces can incorporate siderophores produced by other streptomycetes (Imbert et al., 1995). In addition, some species of Arthrobacter (e.g., Arthrobacter flavescens) do not produce siderophores of their own but rather rely completely on other siderophore-producing bacteria (Winkelmann, 1991). Organisms are also known to regulate siderophore release in response to changes in the extracellular medium (Stone, 1997). Bergeron and McManis (1991) suggested that because catecholate siderophores have higher Fe complex formation constants than hydroxamate siderophores they may work as a “back-up” system for retrieving Fe under low Fe conditions.

In a typical soil, pore water might contain 1000 μM oxalic acid (Fox, 1990), 174 μM acetic acid (Fox, 1990), and 240 μM siderophores (Hersman et al., 1995), as well as trace concentrations of other chelators such as malic, citric and succinic acid. Hydroxamate siderophores function from pH 2 to 11 (Borgias et al., 1989) and catecholate siderophores function from neutral to pH 12 (Winkelmann, 1991). In contrast, neither oxalic nor citric acid chelates effectively at neutral to alkaline conditions. Thus, when present, siderophores must play an important role compared to organic acids in the mobilization of Fe in soils. Powell et al. (1980) presented evidence that large reservoirs of siderophores are adsorbed to soil organic matter and they suggested that hydroxamate siderophores in soils perform many chelate functions previously attributed to organic acids and polyphenolic humic material.

Despite the importance of siderophores in weathering and soil formation, siderophore-promoted dissolution of iron-containing minerals has only been investigated by a few workers (Watteau and Berthelin, 1994, Hersman et al., 1995, Holmén and Casey, 1996, Kraemer et al., 1999, Liermann et al., 2000, Kalinowski et al., 2000; see also Stone, 1997).