these into the cell via various transporters (Liu et al. 2018; Bennett et al. 2018). Interestingly, small alpha‐hydroxy acids such as lactate could help to transport iron(III) into MR‐1 via undefined porins in a fashion similar to siderophores (Bennett et al. 2018). Once in the cell, Fe3+ could be reduced to Fe2+ and, if concentrations of Fe2+ are high, imported from the periplasm into the cytoplasm with the help of Feo and FicI protein importers (Bennett et al. 2018). Otherwise, if concentrations are low, Feo alone could suffice for importing iron.
Similar to MR‐1, in G. sulfurreducens cytoplasmic transporters FeoB and FeoB2 transport Fe2+ from the periplasm into the cytoplasm (Cartron et al. 2006, Embree et al. 2014). However, siderophore uptake mechanisms are not yet as well understood as in MR‐1. Furthermore, microbes can also take up heme groups from outside the cell, produced by other bacteria in the community, and they can use special ATP‐binding cassette (ABC) transporters to bring iron into the cell (Li et al. 2014, Toulza et al. 2012).
The strategy that siderophores are produced if iron becomes limiting, may in fact be related to the importance of iron as a catalyst – as a cofactor in many enzymes – rather than an electron acceptor (Andrews et al. 2003). If these siderophores are capable of accessing iron from more insoluble silicates, which are prevalent in most ocean margin sediments, such strategies may become more efficient (perhaps even outcompeting the reaction with sulfide). As reactive iron centers, such as in heme groups or iron‐sulfur clusters, play an important role in all organisms of the phylogenetic tree, it is understandable that such pathways of iron uptake must have occurred in the earliest organisms on Earth, which may indeed have occurred in marine or brackish environments on the margins of early continents (cf. Sleep 2018), and which would have encountered some challenges in accessing insoluble iron after the onset of an oxic atmosphere and a complete sulfur cycle.
3.4. DIVERSITY OF POTENTIAL IRON‐REDUCING AND IRON‐OXIDIZING ORGANISMS
3.4.1. Correlation of Phylogenetic Abundances with Porewater Chemistry Data
Before discussing correlations of known or potential iron reducers with geochemical data, it is necessary to raise a note of caution. It has become a widely used practice to correlate phylogenetic abundance with all kinds of geochemical parameters, in particular with porewater concentration of potential metabolites, using multivariate statistics (principal component analysis, PCA; Ramette 2007). However, such correlations can be misleading, as the highest solute concentrations do not necessarily equate to highest activity with respect to a particular metabolite. Ramette (2007) specifically warns not to overinterpret the correlations by relying on unjustified causality. Depth profiles in porewater concentrations are mainly subject to diffusive transport along concentration gradients. A linearly increasing or decreasing gradient with depth essentially means that no consumption or production occurs. Instead the second derivative of the concentration after depth, which represents the curvature of the profile according to Fick’s second law of diffusion (Eq. 1; Fick 1855), is an indication of the rate of production or consumption of a particular solute.
c = concentration of a solute (mmol/l)
t = time
D = diffusion coefficient (m2/s)
z = sediment depth
s = source or sink of solute
According to Eq. 1, the source or sink of a solute is equal to the second derivative after depth, if the profile is at steady state, i.e. ∂c/∂t = 0. In theory, the second derivative of concentration after depth could be used in PCA analyses instead, which could potentially provide a more meaningful outcome. However, such an approach would have to be used with caution, as nonsteady state conditions may result in incorrect rate determinations. Ideally, phylogenetic data should be compared with geochemistry using depth profiles. In this review, we do not consider interpretations that are based on the correlation of solute concentrations in PCA analyses alone.
3.4.2. Diversity of Iron Reducers in Suboxic Zones
Several studies have attempted to find and cultivate iron reducers from the iron‐reduction zone in the uppermost centimetres of marine sediments. Stapleton et al. (2005) isolated seven bacteria closely related to different Shewanella genera (91‐98% similarity) from the iron‐reduction zone (0‐10 cm) of sediments collected from a transect off the coast of Washington State (USA) and Puget Sound (USA).
In the Scheldt Estuary (Netherlands, Lin et al. 2007), DNA was extracted from five layers of a 15‐cm long sediment core (Site Waard), and the 16S rRNA genes of Geobacteraceae, Shewanella, Anaeromyxobacter, and Geothrix were amplified and quantified using a most probable number polymerase chain reaction (MPN‐PCR) method. Shewanella and Geobacteraceae as well as the dissimilatory iron reducers Anaeromyxobacter and Geothrix were detected at all depths. Both Shewanella and Geobacteraceae were relatively abundant (between 0.1 and 1%) compared with the other two dissimilatory iron reducers (0.001%). Lin et al. (2007) also made enrichment cultures using the 5–15 cm depth interval of the sediment supplemented with various iron oxides and acetate and lactate as substrates. Microorganisms in the enrichments were identified by first amplifying the 16S rRNA gene followed by denaturing gradient gel electrophoresis (DGGE). In these enrichments, the fermenters Ralstonia and Clostridium, which are both capable of reducing iron, were the most abundant community members, while Shewanella was not as dominant (≤1%; see Figure 5 in Lin et al. 2007). Overall, Lin et al. (2007) found that the classical iron reducers and other dissimilatory iron reducers were not the predominant iron reducers in these suboxic sediments. Instead, fermenters that are known to donate their electrons to Fe(III) (Hamman and Ottow 1974; Dobbin et al. 1999; Park et al. 2001) were the most abundant.
Several other bacteria and archaea that have been isolated from marine sediment environments and can reduce iron are listed in Table 3.1 and include: Shewanella loihica, isolated from iron‐rich mats near hydrothermal vents (Stapleton et al. 2005; Gao et al. 2006), and Geothermobacter sp. str. HR‐1, isolated from Lō‘ihi Seamount, Hawai’i (Smith et al. 2018).
In‐situ diversity studies using 16S rRNA sequencing to detect potential iron reducers in marine surface sediments have been performed at several sites, such as Brown Bay in the Antarctic (Powell et al. 2003), the Baltic Sea at Askö (Sweden; Edlund et al. 2008), and the Archipelago Sea located at the intersection of the Gulf of Bothnia and the Gulf of Finland with the Baltic Sea (Sinnko et al. 2011, see Table S5 in Reyes et al. 2016). Although in some of these studies multivariate statistics were used to link microbial communities to zones of iron reduction, it is worth mentioning the fact that, independent of statistical analysis, these groups were identified as being closely related to known iron reducers.
Using next generation sequencing techniques, potential iron reducers were detected within the zone of iron reduction in the top few centimetres of the sediment of Skagerrak and Bothnian Bay (Figure 3.5; Reyes et al. 2016). These communities were later shown by quantitative PCR (qPCR) to be potentially active in situ and in incubations with marine sediments (Reyes et al. 2017). Microorganisms
