Stable Isotope Fractionation Reveals Similar Atomic-Level Controls during Aerobic and Anaerobic Microbial Hg Transformation Pathways

ABSTRACT

Mercury (Hg) is a global pollutant and potent neurotoxin that bioaccumulates in food webs as monomethylmercury (MeHg). The production of MeHg is driven by anaerobic and Hg redox cycling pathways, such as Hg reduction, which control the availability of Hg to methylators. Anaerobes play an important role in Hg reduction in methylation hot spots, yet their contributions remain underappreciated due to how challenging these pathways are to study in the absence of dedicated genetic targets and low levels of Hg⁰ in anoxic environments. In this study, we used Hg stable isotope fractionation to explore Hg reduction during anoxygenic photosynthesis and fermentation in the model anaerobe Heliobacterium modesticaldum Ice1. We show that cells preferentially reduce lighter Hg isotopes in both metabolisms, leading to mass-dependent fractionation, but mass-independent fractionation commonly induced by UV-visible light is absent. Due to the variability associated with replicate experiments, we could not discern whether dedicated physiological processes drive Hg reduction during photosynthesis and fermentation. However, we demonstrate that fractionation is affected by the interplay between pathways controlling Hg recruitment, accessibility, and availability alongside metabolic redox reactions. The combined contributions of these processes lead to isotopic enrichment during anoxygenic photosynthesis that is in between the values reported for anaerobic respiratory microbial Hg reduction and abiotic photoreduction. Isotope enrichment during fermentation is closer to what has been observed in aerobic bacteria that reduce Hg through dedicated detoxification pathways. Our work suggests that similar controls likely underpin diverse microbe-mediated Hg transformations that affect Hg’s fate in oxic and anoxic habitats.

IMPORTANCE Anaerobic and photosynthetic bacteria that reduce mercury affect mercury delivery to microbes in methylation sites that drive bioaccumulation in food webs. Anaerobic mercury reduction pathways remain underappreciated in the current view of the global mercury cycle because they are challenging to study, bearing no dedicated genetic targets to establish physiological mechanisms. In this study, we used stable isotopes to characterize the physiological processes that control mercury reduction during photosynthesis and fermentation in the model anaerobe Heliobacterium modesticaldum Ice1. The sensitivity of isotope analyses highlighted the subtle contribution of mercury uptake to the isotope signature associated with anaerobic mercury reduction. When considered alongside the isotope signatures associated with microbial pathways for which genetic determinants have been identified, our findings underscore the narrow range of isotope enrichment that is characteristic of microbial mercury transformations. This suggests that there are common atomic-level controls for biological mercury transformations across a broad range of geochemical conditions.

INTRODUCTION

Mercury (Hg) is a global pollutant and potent neurotoxin that bioaccumulates in aquatic and terrestrial food webs as monomethylmercury (MeHg) (1). Anaerobic microbes with the hgcAB gene cluster, which encodes the metabolic machinery responsible for Hg methylation (2), contribute to MeHg production in habitats such as aquatic sediments, water columns, and flooded soils (3–6). Hg methylation is thought to occur intracellularly, and as such, it is ultimately controlled by the bioavailability of Hg to methylators in anoxic habitats (7).

Hg redox cycling plays a key role in determining the inorganic Hg substrates available to methylators. Its role is 2-fold: anaerobic Hg⁰ oxidation can supply dissolved Hg^II required to generate MeHg, and evasion of Hg⁰ due to reduction can limit MeHg production by removing inorganic Hg substrates. Although it is well established that abiotic photochemical Hg reduction dominates Hg redox cycling in oxic surface systems where light is present (8–16), Hg reduction pathways in anoxic zones where Hg methylation occurs have mostly been characterized under laboratory conditions and remain challenging to assess in the field.

Anoxic Hg reduction can occur via abiotic redox reactions with dissolved organic carbon (17) and magnetite (18, 19) but also through biotic pathways mediated by chemotrophic and phototrophic microorganisms (20–24). Unlike many aerobes that reduce Hg using dedicated enzymatic machinery encoded by the mer operon (25, 26), Hg reduction in anaerobes occurs through cometabolic pathways tied to anaerobic respiration, fermentation, and anoxygenic photosynthesis (20–24). In contrast to mer-mediated Hg reduction and demethylation, dedicated genetic determinants have yet to be identified for anaerobic Hg reduction, making these pathways challenging to study from a mechanistic standpoint. Furthermore, while some studies have shown that Hg⁰ can significantly contribute to Hg speciation in anoxic habitats (27, 28), it is possible that low levels of Hg⁰ are maintained by an active cycle of reduction and oxidation (29). In the absence of genetic targets and measurable Hg⁰ accumulation, the contributions of anaerobic Hg⁰ production to global Hg cycling remain cryptic and difficult to assess.

The use of stable Hg isotope fractionation is a powerful tool for providing biogeochemical proxies for different abiotic and biotic Hg transformations (30, 31). All abiotic and biotic Hg transformation pathways studied to date demonstrate kinetic mass-dependent fractionation (MDF), in which the product pool becomes isotopically lighter (enrichment of light isotopes) as the reaction proceeds (32). Some pathways, such as abiotic photochemical reduction and demethylation, also demonstrate mass-independent fractionation (MIF), which is defined as fractionation that occurs outside the predicted MDF pattern (30). MIF is most commonly observed due to the magnetic isotope effect for the odd-Hg isotopes ¹⁹⁹Hg and ²⁰¹Hg in the presence of sunlight (33), and the extent to which MIF occurs is controlled in part by water chemistry and the presence of photosynthetic organisms (34).

The combination of MDF and MIF tracers and the rate of isotopic enrichment, referred to as the fractionation factor, can be used to outline the underlying components of different microbe-mediated Hg transformation pathways. Over the last decade, Hg isotope fractionation has been studied in microbes capable of mer-mediated reduction (35) and MeHg demethylation (36), anaerobic non-mer-mediated Hg reduction (37, 38), Hg methylation (38–41), and oxygenic phototrophic MeHg degradation and Hg reduction (34). These studies have yielded valuable insights and mechanistic details that explain how microbes interact with Hg at the atomic level, making them a promising option for studying mechanisms for anaerobic Hg reduction.

Our objective in this study was to characterize Hg isotope fractionation patterns during anoxygenic phototrophic and fermentative Hg^II reduction, for which no data exist. By providing isotopic characterization of these pathways, we aim to better understand the underlying physiological pathways involved in anaerobic Hg reduction and their role in controlling Hg availability to methylators in anoxic habitats.

In this work, we test whether Hg reduction by the model anaerobe Heliobacterium modesticaldum Ice1 during anoxygenic photosynthesis and fermentation leads to distinct isotope enrichment signatures, which would suggest that the different metabolic Hg reduction pathways are controlled by separate processes. We chose to work with this model heliobacterium because its metabolic versatility is representative of the strategies that allow this family to thrive in the same niche as methylators in habitats such as rice paddies where MeHg accumulation is a major health concern (42–45). Furthermore, our previous work showed that H. modesticaldum has the capacity to reduce high proportions of Hg^II, making this strain well suited to producing the high levels of Hg⁰ required for measuring isotope fractionation (23). In addition to comparing isotope fractionation associated with different metabolic pathways in H. modesticaldum, we compare the isotope enrichment observed in this study to previously reported isotope signatures for biotic and abiotic Hg transformations to critically evaluate whether common controls exist for microbe-mediated Hg transformations. Finally, we discuss whether the current analytical capacity for isotope fractionation can be used to track the contributions of such pathways in the environment.

RESULTS AND DISCUSSION

Hg⁰ production during anoxygenic photosynthesis and fermentation.

In our experiments, Hg⁰ production relied on the presence of live cells under both fermentative and phototrophic growth conditions. Cumulative Hg⁰ production was an order of magnitude higher for live-cell treatments than for abiotic controls (>0.60 versus <0.05 nmol) (Fig. 1). Fermentative cells produced double the amount of Hg⁰ compared to phototrophic cells (1.33 to 1.64 nmol versus 0.60 to 0.81 nmol, respectively) despite having been supplied with slightly less Hg^II (Fig. 1B and E). In relative terms, phototrophic cells reduced 12 to 15% of the initial Hg^II supplied, whereas fermentative cells reduced 31 to 40% (see Fig. S2 in the supplemental material).

FIG 1.

Total Hg and cumulative Hg⁰ production by H. modesticaldum grown phototrophically and fermentatively. (A and D) THg for live cells and sterile controls without cells. (B and E) Cumulative Hg⁰ production for live cells and sterile controls without cells. (C and F) Microbial growth as measured by the OD at 600 nm for live cells. A total of 10 nM Hg was supplied in all experiments from the same stock (NIST 3133).

Relative Hg⁰ production in these experiments was lower yet consistent with our previous work, which showed that fermentatively grown cells reduced 15% more Hg^II than phototrophically grown cells, reducing between 60 and 75% of the initial Hg^II supplied (23). The relative amount of Hg⁰ production did not increase with higher Hg^II exposure (10 nM) versus previous experiments (250 pM) (23). It is unlikely that the lower reduction rate observed in the experiments presented here was due to the toxicity of 10 nM Hg supplied, given that H. modesticaldum grows well under the same conditions at 50-times-higher Hg concentrations (23). Instead, we suspect that the lack of constant bubbling of the reactor may be responsible for the observed results. It is possible that the accumulation of CO₂ in the bioreactor following organic carbon oxidation may compete with Hg^II as an electron sink, decreasing the amount of Hg⁰ produced. Inorganic carbon is required for anaplerotic reactions that fulfill the biosynthetic needs of H. modesticaldum, and CO₂ could have competed with Hg^II for reducing power, as previously shown (46). These results could also be due to Hg⁰ oxidation allowed by the increased residence time of Hg⁰ in the reactor afforded by the intermittent sparging, as has been observed in other anaerobes (47, 48).

Despite carefully controlling the growth conditions in our experiments, fermentative cells from replicate 1 exhibited a lag phase of 12 h (Fig. 1F). Although this lag did not affect the final cell density, the slow-growing culture from replicate 1 initiated Hg⁰ production earlier than the culture in replicate 2 (Fig. 1E and F). These results may be attributable to cells in replicate 1 maintaining a higher Hg-to-cell ratio in the first 12 h of the experiment, which suggests that Hg is being reduced intracellularly, in line with our previous work on heliobacteria (23, 49). Indeed, even at low initial densities, cells can maintain micromolar levels of intracellular reducing power, which could be used to reduce nanomolar levels of Hg^II (24).

Mass-dependent fractionation during photosynthetic and fermentative growth.

Hg reduction in phototrophically grown H. modesticaldum cultures resulted in consistent positive MDF and negligible Δ¹⁹⁹Hg values (−0.01‰ ± 0.05‰) (n = 28) in the reactant pool, indicating an absence of MIF (Fig. 2A and B and Tables S1 and S2). Abiotic controls where light was supplied showed minimal change in Hg isotope values related to both MDF and MIF, confirming that Hg reduction and subsequent isotopic fractionation were predominantly driven by cellular processes (Fig. 2A and Tables S1 and S2).

FIG 2.

Mass-dependent fractionation of ²⁰²Hg in Hg^II (A and C) and Hg⁰ (B and D) pools during phototrophic and fermentative growth of H. modesticaldum and no-cell controls. δ²⁰²Hg values for Hg^II and Hg⁰ are plotted with respect to the fraction of remaining total inorganic Hg^II in the bioreactor. Insufficient Hg⁰ was recovered from the no-cell controls, and thus, no results are displayed. A total of 10 nM Hg was used in all experiments, and Hg was supplied from the same stock (NIST 3133).

We attribute the lack of photochemical fractionation in our experiments to the low levels of visible light and anoxic conditions employed in our setup. These conditions were designed to minimize the amount of abiotic photoreduction observed compared to high-energy UV light sources (natural or controlled), such as those used to study Hg stable isotope fractionation in the oxygenic phototrophic green alga Isochrysis galbana (34). In the study on I. galbana, the authors suggested that UV-mediated production of free radical pairs, such as reactive oxygen species, during photosynthetic electron transport contributed to MIF during Hg reduction (34). Such reactions would likely be inhibited in our experimental setup because of the low-energy wavelengths that are characteristic of the irradiance that we provided and the active removal of oxygen required for the anoxygenic phototrophic growth of H. modesticaldum.

For phototrophically grown cultures, the reactant pool δ²⁰²Hg^II increased steadily over time, suggesting that cells preferentially reduced lighter Hg^II (Fig. 2A). A mirror trend was observed for δ²⁰²Hg⁰, which was depleted in ²⁰²Hg⁰ at the beginning of the experiment, became enriched with heavier isotopes as the reaction proceeded, and eventually approached the initial isotopic composition of the National Institute of Standards and Technology (NIST) 3133 standard (approximately 0‰) (Fig. 2B).

Curiously, the δ²⁰²Hg⁰ for an f_r (i.e., Hg^II in the reactor at a given time/Hg^II in the reactor at the start of the experiment) of 0.94 at 3 h in the two phototrophic live-cell replicates showed that the product pool was initially enriched with the ²⁰²Hg⁰ isotope (Fig. 2B). A similar pattern has been observed in iron-reducing bacteria that preferentially reduced heavier Fe^III (50) and the model anaerobic Hg methylator and reducer Geobacter sulfurreducens PCA (38, 40). The most recent work on G. sulfurreducens has shown that the uptake of Hg selects for lighter Hg isotopes, but cells can also access an isotopically heavier pool of Hg^II from the dissolved phase (38). It is possible that a similar process is occurring for phototrophically grown H. modesticaldum wherein cells access an isotopically heavier pool of Hg during equilibrium binding of Hg^II to the outside of the cell, followed by an alternative uptake process that selects for lighter pools of Hg. Based on these results, we included all time points in isotopic enrichment calculations for Hg^II but omitted the 3-h time point in calculations for Hg⁰ given that other cellular fractionation processes could be occurring (Fig. S3 and S4 and Table S3).

Replicate experiments with fermentative cultures showed more variability with respect to their fractionation patterns than phototrophic cultures; however, positive MDF was discernible (Fig. 2C and D). Replicate 1 showed an initial depletion in reactant pool δ²⁰²Hg^II and became progressively enriched with ²⁰²Hg^II relative to the NIST 3133 standard, suggesting that lighter isotopes were still preferentially reduced (Fig. 2C). Note that the second δ²⁰²Hg^II data point available for replicate 1 occurs at an f_r of 0.72 due to the high rate of Hg^II reduction leading to substantially less Hg remaining in the reactor (Fig. 1E and Fig. 2C). The results for replicate 2 further suggested that lighter Hg^II isotopes were preferentially reduced, with δ²⁰²Hg^II slightly increasing at the beginning of the experiment before rising to a maximum of 0.81‰ between an f_r of 0.85 at 12 h and an f_r of 0.48 at 48 h (Fig. 2C).

The trends for δ²⁰²Hg⁰ in fermentative cultures varied between the two replicates but mostly followed the MDF patterns observed in the Hg^II pool. Under fermentative conditions, four of the seven Hg⁰ samples obtained from the product pool exceeded the initial composition of the NIST 3133 standard (0‰), which was not observed for phototrophic cultures (Fig. 2B and D). Our observations align with previously reported results for anaerobic cultures of G. sulfurreducens and Desulfovibrio desulfuricans ND132, which produced product pool δ²⁰²Hg values that exceeded the initial composition of the Hg^II supplied during Hg methylation (40). These results were not predicted based on the starting isotope composition of Hg^II provided to the organisms and suggested that the enrichment in the product pool could be driven by differences in the bioaccessible pools of Hg (40). Indeed, more recent work with G. sulfurreducens has reinforced the view that subcellular partitioning processes can contribute to Hg isotope fractionation, although such changes were not detected at the level of the total Hg (THg) pool (38). It is possible that similar mechanisms are driving the Hg⁰ isotope enrichment observed in fermentative cultures of H. modesticaldum and that bioaccessible Hg^II pools may change as a function of the growth conditions.

In our work, Hg reduction provided such a strong isotopic shift that we could detect fractionation in the THg pool (Fig. 2). However, we cannot discount the contributions of lower-magnitude (<0.10‰) subcellular partitioning processes or the preferential utilization of different intracellular Hg pools during Hg^II reduction that may more drastically affect the isotope values of the Hg⁰ product pool. Such processes could be driving the slight depletion of ²⁰²Hg^II (−0.29 to −0.18‰) early on for fermentative cells in replicate 1, where low cell densities may have amplified uptake and adsorption-driven fractionation in comparison to experiments where the cell density increased by an order of magnitude over the same time frame (Fig. 1F and Fig. 2C). It is more challenging to discuss the contributions of such processes for replicate 2, where the low level of cumulative Hg⁰ production at f_r values of 0.94 and 0.88 (3-h and 6-h time points, respectively) precluded our ability to measure Hg⁰ isotope fractionation (Fig. 1E and Table S2). Despite this limitation, the high δ²⁰²Hg⁰ value obtained for the final time point for replicate 2 supports that additional processes are contributing to fractionation during fermentative growth, which warrants further exploration.

Our experimental results suggest that Hg⁰ oxidation is not a major contributor at earlier stages of reduction (i.e., 3 to 12 h) but may be contributing to isotope fractionation at later time points associated with longer Hg residence times in the reactor. Under phototrophic conditions, the measured values of δ²⁰²Hg^II at 48 h were isotopically heavier than the predicted values, which suggests that Hg⁰ oxidation may be occurring (Table S4). Under fermentative conditions, the potential contribution of Hg oxidation is unclear since consistent enrichment was not observed between replicates (Table S4). The variability in isotope results for later time points in both fermentative replicates suggests that processes other than reduction may be occurring.

Although Hg⁰ oxidation has never been observed in anoxygenic phototrophs, it has been observed in other model chemotrophic anaerobes (48). Recent work showed that thiol-bearing molecules preferentially oxidize heavy ²⁰²Hg⁰, leading to negative MDF, in addition to a negative MIF signal (51). Although we did not observe any negative MDF or MIF in our experiments (Tables S1 and S2) to the extent seen in previous work (51), we acknowledge that such processes have the potential to contribute to net fractionation alongside internal partitioning related to uptake or adsorption processes.

The results presented in this work are an important step toward resolving the physiological processes that drive anaerobic Hg transformations in H. modesticaldum. We demonstrate that while the biotic reduction of Hg during anoxygenic photosynthesis and fermentation drives Hg isotope fractionation, other processes such as adsorption, cellular uptake, and equilibrium effects have an important impact on the net isotope values observed. Our results suggest that the extent to which these additional processes affect isotope fractionation is tightly coupled to cellular Hg availability and the physiological state of cell cultures. These biological controls and overlapping Hg reactions are key to elucidating the Hg isotope signature associated with environmental processes such as reduction. Bearing these nuances in mind, we aim to carry out additional experiments that assess the contributions of uptake, adsorption, and redox transformations to the net Hg isotope fractionation associated with anaerobic Hg reduction.

Comparing enrichment factors for abiotic and biotic Hg transformations.

Our initial motivation for this study was to test if phototrophic and fermentative Hg reduction led to different isotope enrichment factors and establish whether different underlying processes supported different metabolic Hg reduction pathways. When comparing ε values, which are used as an index to distinguish between different isotope fractionation processes, we obtained average enrichment factors using ratios of product to reactant (ε_p/r) ±2 standard deviations (2SD) for MDF in the reactant pool Hg^II of −0.80‰ ± 0.56‰ for anoxygenic photosynthesis (ε_AP) and −1.15‰ ± 0.48‰ for fermentation (ε_FM) (Fig. 3, Fig. S3, and Tables S3 and S5). The enrichment factor could not be calculated using the product Hg⁰ pool for fermentative cultures, and ε was overestimated in phototrophic cultures (−2.86‰ ± 1.32‰), likely due to challenges in obtaining analytical replicates of gaseous Hg⁰ and quantitatively collecting this pool (52) (Fig. S4 and Table S3). The small difference between ε_AP and ε_FM (i.e., 0.35‰) could suggest common controls for Hg reduction during photosynthesis and fermentation. However, we acknowledge that we could not effectively test our hypothesis due to the variability observed in replicated fermentative experiments where the cell populations used were in different physiological states, as supported by the growth curves (Fig. 1F). Despite this limitation, our work demonstrates that net isotopic enrichment stems from a combination of multiple processes most likely affected by Hg transport (e.g., adsorption, uptake, and efflux) that would occur regardless of the intracellular transformations in question.

FIG 3.

Compilation of isotopic enrichment factors (ε) associated with mass-dependent fractionation for abiotic and biotic Hg transformations. Hg transformation pathways are denoted by colors, and the detection of mass-independent fractionation (MIF) is denoted by different shapes. Note that the data originating from this work are presented as separate replicates to emphasize the range of ε values that the observations fall within. For all other data, error bars denote 2 standard deviations for studies where these data were available. In the event that those data were presented as a standard error between replicate experiments, the standard deviation was calculated based on the sample size indicated in each study. Abbreviations: DOC, dissolved organic carbon; DOM, dissolved organic matter; AP, anoxygenic photosynthesis; FM, fermentation. The curves represent gradients for light and oxygen (O₂). The values for the following factors were obtained from previous studies or the present study: abiotic—Hg^II photoreduction (32, 34), abiotic—Hg^II reduction (54), abiotic—thiol Hg⁰ oxidation (51), aerobic respiration—MerA Hg^II reduction (35, 37), anaerobic respiration—Hg^II methylation (40), anaerobic respiration—Hg^II reduction (37, 38), anoxygenic photosynthesis—Hg^II reduction (this study), fermentation—Hg^II reduction (this study), and oxygenic photosynthesis—Hg^II reduction (34). The raw data can be found in Table S5 in the supplemental material.

The isotope enrichment signatures observed for Hg reduction during anoxygenic photosynthesis and fermentation can be compared to previously published data to better understand the physiological processes that control microbial Hg transformations and advance Hg isotope applications. For convenience, we present the enrichment values for our replicate experiments separately in Fig. 3 and provide data compiled from the literature alongside any manipulations carried out to facilitate comparisons in Table S5.

The isotopic enrichment observed for both metabolisms tested in H. modesticaldum falls between the values observed for abiotic and microbial Hg reduction pathways (Fig. 3 and Table S5). The isotopic enrichment for MDF observed for replicate experiments with phototrophic H. modesticaldum is slightly lower than or equal to photoreduction in the presence of dissolved organic carbon (−1.00 to −0.60‰ versus −0.60‰ ± 0.28‰, respectively) (32) but higher than photoreduction in the presence of marine exudates and UVB light (−1.00 to −0.60‰ versus −1.47‰, respectively) (34) (Fig. 3 and Table S5). When comparing Hg reduction in the anoxygenic phototroph H. modesticaldum to that in the oxygenic phototrophic green alga I. galbana, the isotopic enrichment observed is slightly lower or equal, depending on the treatment in question (−1.00 to −0.60‰ versus −0.14‰ for high UV and −0.61‰ for low UV, respectively) (34) (Fig. 3 and Table S5). Although the similar ε value of the low-UV treatment suggests that anoxygenic and oxygenic photosynthetic forms of Hg reduction share a physiological pathway, the presence of MIF in I. galbana due to free radical generation within the cell rules out this possibility (34) (Table S1). Thus, Hg reduction during anoxygenic photosynthetic Hg reduction could be distinguished from other photoinduced reduction pathways, which display MIF in the reactant Hg^II pool (53).

Compared with previous work on microbial Hg reduction in pure cultures, the isotopic enrichment observed for phototrophically grown H. modesticaldum was higher than what has previously been reported for aerobic mer-mediated reduction and Hg reduction during anaerobic respiration (−1.00 to −0.60‰ versus −1.80 to −1.20‰, respectively) (35, 37, 38) (Fig. 3 and Table S5). Interestingly, the isotopic enrichment observed in fermentatively grown H. modesticaldum (−1.32 to −0.98‰) was closer to aerobic mer-mediated reduction (−1.40 to −1.20‰) (35) than to Hg reduction during anaerobic respiration in Shewanella oneidensis MR-1 and a modified strain of G. sulfurreducens PCA incapable of Hg methylation (−1.80 to −1.60‰) (37, 38) (Fig. 3 and Table S5). The enrichment values reported for dark abiotic Hg^II reduction catalyzed by dissolved organic matter (DOM) are slightly lower than those for fermentative Hg^II reduction (−1.32 to −0.98‰ versus −1.52‰) (Fig. 3 and Table S5). Similar to what was mentioned for photoreduction previously, DOM-mediated transformations display various degrees of MIF such that fermentative Hg^II reduction could be distinguished from DOM-mediated Hg reduction in the dark (54) (Fig. 3 and Table S5). Additional lines of evidence supporting that Hg⁰ oxidation is unlikely to be contributing to the isotope signature observed in our system are the ε values for thiol-mediated Hg⁰ oxidation in the dark. These values are considerably larger than what we observed for H. modesticaldum (1.51‰ ± 0.20‰ for cysteine and 1.47‰ ± 0.24‰ for glutathione) (51) (Fig. 3 and Table S5).

These comparisons illustrate that isotopic enrichments for microbial Hg reduction pathways that display only MDF fall within a narrow range (i.e., −0.60 to −1.80‰) (Fig. 3 and Table S5). We find this to be a strikingly small difference given the ecological diversity of the model organisms and the variety of growth conditions used to study microbial Hg reduction. The similar ε values reported for fermentation in this study and the mer operon are also noteworthy, as they suggest that nearly identical isotope enrichment can occur in aerobes harboring dedicated enzymatic machinery to reduce Hg and anaerobes where mer-based strategies are largely absent.

Our study also shows that similar isotopic enrichments can be observed for different anaerobic Hg transformations. The anaerobic Hg-methylating chemotrophs G. sulfurreducens PCA and D. desulfuricans ND132 displayed ε signatures (i.e., −0.92 and −1.10‰, respectively) that fall within the range of ε values that we report in our study (i.e., −1.32 to −0.60‰) (Fig. 3 and Table S5) (40). The similarity in enrichment signatures suggests that common processes may affect the delivery of inorganic Hg substrates to intracellular sites independently of the Hg transformations in question. Although it is outside the scope of the current study, identifying the nature of these processes may provide insights into similar controls that exist for Hg uptake and transformation in anaerobes.

Our work showcases the strengths but also the challenges of using isotope fractionation-based approaches to decipher cryptic Hg cycling pathways. To illustrate the challenges of using isotopes to track Hg cycling pathways in the environment, we visualized our comparison of ε values in the context of a vertically redox-stratified habitat (e.g., a water column) (Fig. 3). Through Hg stable isotope fractionation, we can distinguish between abiotic and biotic redox reaction pathways that control Hg’s availability in methylation hot spots, such as the limit of the photic zone, on the bases of MIF signatures associated with photochemical and DOM-mediated reactions. In contrast, teasing apart biological pathways that contribute to the net isotope signature of Hg^II in the environment is more challenging because competing MDF processes occur in both dark and photo-mediated pathways where Hg^II is a reactant (e.g., reduction and methylation) and a product (e.g., demethylation and oxidation). This overlap makes it difficult to translate δ²⁰²Hg values observed under controlled conditions that are sensitive to changes in cellular physiology to environments where pools of bioavailable Hg are not homogeneous due to variable mixing.

Conclusion.

In this study, we demonstrate that Hg reduction during anoxygenic photosynthesis and fermentation leads to MDF of stable Hg isotopes. We show that isotopic enrichment during anoxygenic phototrophic Hg reduction leads to an intermediate fractionation process that lies between photochemical reactions and dark microbial Hg reduction. We also show that fermentative Hg reduction leads to isotopic enrichment that is similar to that of aerobic mer-mediated reduction. Although we could not discern whether dedicated physiological processes support Hg reduction during anoxygenic photosynthesis and fermentation, our study shows that isotopes are well suited to teasing apart the complex interwoven processes that drive microbe-mediated Hg transformations and highlighting common processes that control Hg bioavailability to microbes.

In future work, it will be important to isolate the different steps involved in Hg uptake and transformation by rigorously controlling Hg availability and cellular physiology. Leveraging molecular tools alongside isotopes will be a useful strategy to address these knowledge gaps, and the recent availability of genetically tractable H. modesticaldum deletion mutants offers a promising means to do so (55). In that regard, gene deletion approaches targeting redox-active enzymes could be used to characterize the physiological components contributing directly to Hg^II reduction, whereas targeting homologues for proteins involved in Hg uptake in other model anaerobes could be the key to identifying routes of entry for Hg into H. modesticaldum (56, 57). By combining molecular tools with isotope-based approaches to study anaerobic Hg reduction, we may finally be able to characterize the contributions of such pathways to the global Hg cycle in present-day environments and historical archives (e.g., sediment records). Such studies would help us better understand the ecological conditions that gave rise to the evolution of anaerobic Hg redox cycling pathways prior to the acquisition of dedicated Hg detoxification strategies such as the mer operon.

MATERIALS AND METHODS

Cell growth conditions and bioreactor setup.

All experiments with live cells employed the strain Heliobacterium modesticaldum Ice1 grown in PYE medium. PYE medium was prepared by dissolving the following ingredients in 1 liter of ultrafiltered high-purity (18.2 MΩ) water: dipotassium phosphate (K₂HPO₄) (1 g liter⁻¹), magnesium sulfate heptahydrate (MgSO₄·7H₂O) (0.2 g liter⁻¹), calcium chloride dihydrate (CaCl₂·2H₂O) (0.02 g liter⁻¹), sodium thiosulphate pentahydrate (Na₂S₂O₃·5H₂O) (0.2 g liter⁻¹), sodium pyruvate (2.20 g liter⁻¹), yeast extract (YE) (4.00 g liter⁻¹), and ammonium sulfate [(NH₄)₂SO₄] (1.00 g liter⁻¹). The pH of the medium was adjusted to 6.8, after which 1 ml of a filter-sterilized (0.2-μm pore size) trace element solution was added to the medium as a 1,000× dilution. The trace element stock was prepared by adding the following ingredients to 1 liter of ultrafiltered high-purity water: zinc sulfate heptahydrate (ZnSO₄·7H₂O) (0.10 g liter⁻¹), manganese chloride tetrahydrate (MnCl₂·4H₂O) (0.03 g liter⁻¹), boric acid (H₃BO₃) (0.30 g liter⁻¹), cobalt chloride hexahydrate (CoCl₂·6H₂O) (0.20 g liter⁻¹), cupric chloride dihydrate (CuCl₂·2H₂O) (0.01 g liter⁻¹), nickel chloride hexahydrate (NiCl₂·6H₂O) (0.02 g liter⁻¹), and sodium molybdate dihydrate (Na₂MoO₄·2H₂O) (0.03 g liter⁻¹). The medium was then boiled for 15 min under a stream of sterile nitrogen (N₂) gas before being aliquoted into serum bottles or Balch tubes that were crimped shut, flushed with N₂ gas, and autoclaved at 121°C for 30 min.

Phototrophic cultures were grown at a constant visible-light intensity of 80 μmol photons m⁻² s⁻¹ with a peak irradiance in the near-red-to-red spectrum (600 nm to 700 nm) at 50°C, whereas fermentative cultures were grown in the dark at 50°C, in line with our previous work (23). H. modesticaldum Ice1 was obtained from the Leibniz Institute Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ) culture collection (catalogue number DSM-792). The bioreactor methodology employed in these experiments was similar to the one used in previous work (23), with the following modifications: the bioreactor was periodically supplied with sterile N₂ gas that passed through an activated carbon filter and a 0.2-μm-pore-size air filter at a flow rate of 0.25 liter min⁻¹, and two gold traps were connected in series at the reactor outlet to capture Hg⁰ (see Fig. S1 in the supplemental material).

The bioreactor was kept in an incubator set to 50°C. Phototrophic experiments were performed using a 60-W incandescent light bulb with a visible light intensity being maintained at 20 μmol photons m⁻² s⁻¹ and mirroring the irradiance of the growth conditions. Fermentative experiments were performed in the dark. Background Hg⁰ was purged and removed from reactors prior to phototrophically or fermentatively grown H. modesticaldum cells being added as a 10% (vol/vol) inoculum, after which an additional subsample was withdrawn to verify the initial cell density via the optical density (OD) at 600 nm. For sterile treatments, no cells were added to the bioreactor, and the volume of the growth medium was adjusted to ensure that the experiments were carried out with a working volume of 500 ml. We chose to use sterile medium rather than autoclaved cultures based on our previous work, which showed that autoclaved cultures of H. modesticaldum supplied with Hg^II under the same conditions reduced negligible amounts of Hg^II (23). In all experiments, National Institute of Standards and Technology (NIST) 3133, a standard with certified Hg isotope values, was added to the bioreactor to a final concentration of approximately 10 nM Hg. Additional details on the bioreactor setup are provided in the supporting methods in the supplemental material.

Hg⁰ measurements.

The bioreactor was bubbled periodically rather than continuously, in contrast to previous work, because continuous bubbling led to fouling of gold traps and poor total Hg (THg) and isotope recoveries. Subsamples were collected at designated time points by connecting dual gold traps to the reactor and purging the reactor for 30 min to collect Hg⁰. In the final 5 min, the reactor liquid was sampled for Hg^II concentrations, isotopic composition, and cell growth, after which the gas flow was stopped. All gold traps were capped to prevent Hg⁰ loss, and aqueous samples were conserved with 1% (vol/vol) trace element-grade hydrochloric acid (HCl) and stored at 4°C in the dark. Reactor washes were performed using trace element-grade HCl according to previously established protocols (23).

Total Hg analyses and sample preparation for stable isotope measurements.

All samples for aqueous Hg analyses were oxidized with 10% (vol/vol) bromine chloride (BrCl), in line with previous work (40). THg concentrations were determined by stannous chloride reduction coupled to cold-vapor atomic fluorescence spectrometry at the U.S. Geological Survey Mercury Research Lab (USGS MRL) (Middleton, WI) (58). During THg analyses, duplicates and matrix standard spikes were analyzed for every 10 samples; analyses showed a relative percent difference of <10% between duplicates and spike recoveries of 90 to 110%. The detection limit was 0.2 pM.

Total Hg mass balances were calculated in line with previous work (23, 24), with corrections to account for the mass of Hg removed for each subsample. For Hg⁰ collected onto gold traps, a thermal desorption system was used to liberate Hg⁰ over a 40-min desorption cycle, which was then captured in a 40% nitric acid (HNO₃)–BrCl (3:1) oxidizing solution. An aliquot of the trapping solution was subsequently analyzed for THg concentrations according to a protocol established in previous work (59). The desorption efficiency during preconcentration was tested by trapping an NIST 3133 standard with each batch of samples, and recoveries for standard trapping were 101.27% ± 4.26% (n = 15). THg recoveries from bioreactor experiments were 94.90% ± 9.50% (n = 7) (Fig. S2).

Hg stable isotope analyses.

Using the previously determined THg concentration, an aliquot of the trapping solution was diluted with ultrahigh-purity water to an acid content of <10% H⁺ and a THg concentration of 3.75 to 5.00 nM (60). A Neptune Plus multicollector-inductively coupled plasma mass spectrometer (MC-ICP-MS) (Thermo Scientific) was used for Hg stable isotope ratio measurements at the USGS MRL. A concentration- and matrix-matched NIST 3133 Hg standard was used for sample bracketing (61). To achieve detection at low Hg concentrations, Hg was reduced with stannous chloride in line and then introduced continuously into a custom-designed gas-liquid separator along with a thallium standard (NIST SRM 997) (195 nM in 3% HCl) for mass bias correction (60, 62). MC-ICP-MS instrument conditions and analytical expectations were derived from previously published work (60). All quality control metrics for Hg stable isotope analysis and raw data are provided in the supporting methods and Tables S1 and S2 in the supplemental material. Note that very low levels of Hg⁰ were produced in the abiotic controls for both phototrophic and fermentative experiments, and as such, there was insufficient Hg⁰ to measure isotope fractionation in the product pool (Table S2).

Isotope calculations.

Delta calculations followed the conventions set forth by others (61). This convention calls for mass-dependent fractionation (MDF) to be expressed in terms of δ^xxxHg, where graphically, δ²⁰²Hg is used to signify MDF. δ^xxxHg is calculated as

$${\text{δ}}^{\text{xxx}}\text{Hg}(‰)\hspace{0.17em}=\hspace{0.17em}\left[\left({}^{\text{xxx}}\text{Hg}/{}^{198}\text{H}{\text{g}}_{\text{sample}}\right)/\left({}^{\text{xxx}}\text{Hg}/{}^{198}\text{H}{\text{g}}_{\text{NIST\hspace{0.17em}3133}}\right)-1\right]\times \text{1,000}$$ (1)

xxx is used to signify the isotope of interest. Hg also undergoes MIF of both even and odd isotopes. Here, odd MIF is described by Δ¹⁹⁹Hg, and even MIF is described by Δ²⁰⁰Hg. MIF is calculated as

$${\Delta }^{\text{xxx}}\text{Hg\hspace{0.17em}}\approx {\text{\hspace{0.17em}δ}}^{\text{xxx}}\text{Hg}-\left({\text{δ}}^{202}\text{Hg}\times \text{β}\right)$$ (2)

β denotes the mass-dependent scaling constant, which is determined by the laws of mass dependence (61). Isotope enrichment effects based on the ratio of products to reactants (i.e., ε_p/r, referred to here as ε_AP for anoxygenic phototrophic reduction and ε_FM for fermentative reduction) were calculated using Rayleigh fractionation models to account for the Hg⁰ that was removed according to methods from previous work on microbial Hg reduction (35). This method involved fitting a linear regression (shown in Fig. S3 and S4 and Table S3) based on the change in relative isotope ratios, ln(R/R₀), as a function of ln(f_r), where

$$R=\left({\text{δ}}^{202}\text{Hg}/\text{1,000}\right)+1{\text{\hspace{0.17em}for\hspace{0.17em}Hg}}^{\text{II}}\left(\text{reactant}\right){\hspace{0.17em}\text{or Hg}}^0\left(\text{product}\right)\hspace{0.17em}\text{at\hspace{0.17em}a\hspace{0.17em}given\hspace{0.17em}time\hspace{0.17em}point\hspace{0.17em}and\hspace{0.17em}}\hspace{0.17em}\hspace{0.17em}{R}_0=\left({\text{δ}}^{\text{2}0\text{2}}\text{Hg}/\text{1,000}\right)+\text{1\hspace{0.17em}at\hspace{0.17em}the\hspace{0.17em}beginning\hspace{0.17em}of\hspace{0.17em}the\hspace{0.17em}experiment}$$ (3)

$$f_r={\text{\hspace{0.17em}Hg}}^{\text{II}}\text{\hspace{0.17em}in\hspace{0.17em}the\hspace{0.17em}reactor\hspace{0.17em}at\hspace{0.17em}a\hspace{0.17em}given\hspace{0.17em}time}/{\text{Hg}}^{\text{II}}\text{\hspace{0.17em}in\hspace{0.17em}the\hspace{0.17em}reactor\hspace{0.17em}at\hspace{0.17em}the\hspace{0.17em}start\hspace{0.17em}of\hspace{0.17em}the\hspace{0.17em}experiment\hspace{0.17em}}$$ (4)

and the slope of the linear regression, which is equivalent to ε_AP or ε_FM:

$$\text{ε}=\text{ln}\left(R/R_0\right)/\text{ln}\left(f_r\right)$$ (5)

We chose to use the total Hg analyses from the bioreactor to establish f_r as this is a true representation of instantaneous isotope fractionation rather than a time-integrated sample according to the methods outlined in previous work (35). Note that in our slope calculations, we do not have data for ²⁰²Hg⁰ when f_r is 1 (the 0-h sampling point) because no isotope fractionation was taking place prior to the start of Hg reduction (Tables S2 and S3). Additionally, in experiments where negligible amounts of Hg^II were reduced to Hg⁰, data points for δ²⁰²Hg in the Rayleigh fractionation plots tend to cluster around an f_r of 1 since ∼100% of the original Hg supplied remained in the reactor after 48 h. Additional calculations for isotope mass balances were carried out to test for deviation from the Rayleigh model and can be found in Table S4.

ε values for MDF were compiled from previous studies on pure microbial cultures and abiotic Hg transformations for comparison purposes. Currently, there is no standard guideline for how to calculate ε, and studies vary widely in how they present isotope enrichment data and their associated variability. We provide details on how we normalized ε values from different studies in Table S5. In this work, we compared only enrichment factors from studies that used MC-ICP-MS techniques and slope calculations in line with our analyses; however, we have included all of the ε values for microbe-driven Hg transformations published to date in Table S5.