The role of unchelated Fe in the iron nutrition of phytoplankton

The important question of iron bioavailability in the sea has become complicated by the discovery that marine phytoplankton can take up Fe bound in very stable chelates via reductive processes, and some particular Fe species through specialized transport mechanisms. As a result there is some question of whether the small fraction of Fe that is ‘‘free’’ or unchelated in seawater is important in the nutrition of natural phytoplankton assemblages. A careful examination of published laboratory studies on Fe uptake by model organisms all support the idea that unchelated Fe(III) is highly available for uptake and that it is an important source of the Fe taken up by phytoplankton under a variety of experimental conditions. Comparing these results with field data on Fe speciation shows that unchelated Fe can be an important source of Fe to the phytoplankton in the sea: it is likely sufficient to contribute the bulk of the Fe supporting primary production in regions that are not limited by Fe and a significant fraction everywhere, including high-nutrient low-chlorophyll areas. The revolution in our understanding of ocean productivity started by Martin 20 yr ago (Martin and Fitzwater 1988) is continuing. In addition to high-nutrient lowchlorophyll (HNLC) oceanic areas, some coastal ecosystems (Hutchins et al. 1998) and deep chlorophyll a maxima of oligotrophic regions (Hopkinson and Barbeau pers. comm.) have now been shown to be limited or colimited by iron. To go beyond observations or phenomenological experiments and truly understand the link between Fe inputs and productivity, we need to elucidate the relation between the concentration of various chemical forms of Fe in seawater and its uptake by phytoplankton; that is, we need to define chemically the ‘‘availability’’ of Fe species to marine phytoplankton. For the purpose of our discussion, despite their variety and complexity, we can classify the Fe uptake systems of microorganisms into three categories (Fig. 1A): (1) transport systems that are specific for particular Fe compounds or families of compounds, such as Fe citrate, Fe siderophores, or hemes; (2) Fe(II) transporters of various specificities, including divalent metal ion transporters and oxidase–permease complexes that oxidize Fe(II) while transporting it across the external membrane; (3) transport systems that include reductases able to reduce various Fe(III) species at the cell surface and deliver Fe(II) to (2). Notably absent from our list are transporters of unchelated Fe(III), Fe(III)9, which have been reported to exist at the inner but not the outer membranes of gram negative bacteria. The existence of such transporters in the outer membrane of some phytoplankter would, of course, not affect our argument regarding the importance of unchelated Fe(III). We have at present no firm information regarding either the existence in phytoplankton of transport systems that belong to category (1) or the concentration of the corresponding compounds in seawater. As a result, the question of Fe bioavailability in seawater is focused on uptake via systems (2) and (3), which are supported by genomic and transcriptomic data in two marine diatoms (Kustka et al. 2007). Early studies demonstrated a correspondence between the concentration of unchelated Fe in the medium and Fe uptake by phytoplankton (the Fe9 model; Hudson and Morel 1990). Following the demonstration that Fe bound in some strong complexes can be taken up by some species of phytoplankton and that Fe(III) must be reduced for uptake (Soria-Dengg and Horstmann 1995; Maldonado and Price 2001, Shaked et al. 2005), newer studies have focused on the role of Fe(II) in uptake. Recently two models have been proposed to describe the kinetics of Fe uptake by phytoplankton and effectively quantify Fe availability in seawater: (1) the Fe(II)s model (Fig. 1B; Shaked et al. 2005) uses the surface concentration of reduced iron, Fe(II)s, as the parameter controlling uptake; it is based on experimental data with diatoms and explicitly incorporates the previous Fe9 model by making unchelated Fe(III), an important source of reduced Fe at the cell surface. (2) The FeL model (where L represents an Fe-chelating ligand; Fig. 1C; Salmon et al. 2006) makes uptake dependent on the concentration of Fe(II) in the bulk medium and considers chelated Fe(III) to be the only source of reduced Fe; it is based on data with natural samples of the cyanobacterium Lyngbya majuscula (but is presented as applicable to other phytoplankton including diatoms) and emphasizes the reoxidation of Fe(II), particularly that of the chelated form, Fe(II)L, as a key process competing with uptake. An examination of the similarities and differences between these two models and a comparison of their predictions with available experimental data provide a good basis for exploring the question of Fe availability to phytoplankton. The most obvious difference between the Fe(II)s and FeL models is that one considers the surface concentration of Fe(II) and the other its bulk concentration as the controlling parameter. This difference stems from the difference in the principal reduction mechanisms assumed to generate Fe(II) as a substrate for uptake: cell surface reductases in one case, bulk reduction by O 2 in the other. This difference is immaterial in the mathematical formulation of the models since they both ignore the diffusion of Fe(II) species between the cell surface and the bulk solution. But the question of what processes are responsible for Fe reduction is in fact important. Because of the necessary diffusion to the cell surface and of the fast reoxidation of the unchelated Fe(II), Fe(II)9, it matters greatly if reduction occurs at the surface of cells or in the bulk solution and by what mechanism. For example, the photoreduction of an Fe(III)chelate (such as Fe(III)ethylenediaminetetraacetate [EDTA] in culture medium) results 400 Notes