Targeted Intracellular Demethylation of Methylmercury Enhances Elimination Kinetics and Reduces Developmental Toxicity in Transgenic Drosophila

Abstract

Methylmercury (MeHg) persists today as a priority public health concern. Mechanisms influencing MeHg metabolism, kinetics, and toxicity outcomes are therefore essential knowledge for informing exposure risks. Evidence points to different toxic potencies of MeHg and inorganic mercury (Hg²⁺), highlighting the role for biotransformation (demethylation) in regulating MeHg toxicokinetics/dynamics. Whereas microbial MeHg demethylation in the gut is seen to influence elimination kinetics, the potential for systemic demethylation in tissues and target organs to influence MeHg toxicity remains uncertain. To investigate the consequences of systemic MeHg demethylation across development, we engineered transgenic Drosophila to express the bacterial organomercurial lyase enzyme (merB) in a targeted and tissue-specific manner. With all combinations of merB-induced demethylation, ubiquitously (via an actin promoter) or in a tissue-specific manner (ie, gut, muscle, neurons), we observe a rescue of MeHg-induced eclosion failure at the pupal to adult transition. In MeHg-fed larvae with ubiquitous or targeted (gut and muscle) merB expression, we see a significant decrease in MeHg body burden at the pupal stage relative to control flies. We also observe a significant increase in the MeHg elimination rate with merB demethylation induced in adults (control, t_1/2 = 7.2 days; merB flies, t_1/2 = 3.1 days). With neuronal-specific merB expression, we observe a rescue of MeHg-induced eclosion failure without a decrease in Hg body burden, but a redistribution of Hg away from the brain. These results demonstrate the previously unidentified potential for intracellular MeHg demethylation to promote transport and elimination of Hg, and reduce developmental MeHg toxicity.

Impact Statement: These findings demonstrate the potential for MeHg demethylation in situ to contribute significantly to the MeHg elimination and distribution kinetics of whole animals and thereby affords a means of protection against the toxic insult of MeHg. Therefore, this study reveals important insight into processes that can determine an individual’s resistance or susceptibility to MeHg and provides rationale for therapies targeting a novel metabolism-based pathways to alleviate toxicity risk stemming from MeHg exposure.

Methylmercury (MeHg) continues to be a toxicant of concern due to its global release and well-established potency in disrupting organ development in mammalian and human systems, notably targeting the nervous system (Eto and Takeuchi, 1978; Holzman, 2010; Hou, 2022; Kazantzis, 1976; Lindberg, 2005; Liu et al., 2011; Takeuchi, 1989). Exposure to MeHg today occurs primarily through the consumption of contaminated fish and seafood, resulting in 10% of fish consumers globally having blood Hg levels above what is considered safe (Girard, 2018; Wells, 2020; Tollefson and Cordle, 1986). Although long studied, considerable uncertainty in the risk of MeHg exposure remains due to the intrinsic inter-individual variability in toxicokinetics seen across the human population (Caito, 2018; Jo, 2015; Rand, 2016; Rand and Caito, 2019). Among the factors contributing to variability in elimination from person to person is the metabolism of MeHg, ie, MeHg demethylation, and accumulation of Hg²⁺ at target organs over time (ie, brain, kidney, liver, gut) (Pamphlett, 2021). Several studies support a central role for MeHg biotransformation in overall Hg elimination from the body (Magos, 1971, 1984; Rowland, 1980, 1984; Takeuchi, 1989; Tan, 2022). In the gut specifically, it is thought that the microbiome mediates MeHg demethylation to produce Hg²⁺ which is rapidly excreted from the body via the feces (Caito, 2018; Rowland, 1980, 1984). Demethylation in this context is predicted to protect against MeHg toxicity as speeding up elimination would contribute to a lowering of overall Hg body burden. However, the microbial mechanisms for MeHg demethylation in the gut have yet to be elucidated.

On the other hand, systemic MeHg demethylation observed in host tissues and target organs, is less well understood. In fact, demethylation here has been proposed to increase the toxicity outcomes as intracellular accumulation of Hg²⁺ has been associated with increased levels of cytotoxicity and neurological outcomes (Björkman, 1995; Clarkson, 2002; Krout, 2022; Takanezawa et al., 2019). For example, studies in primates have reported that neurological decline due to MeHg exposure correlated with the accumulation of Hg²⁺ in the brain, attributed to the cumulative demethylation of MeHg in situ (Charleston, 1995; Vahter, 1994, 1995). More recently, with an approach of introducing the bacterial merB, enzyme, an organomercurial lyase, to mammalian cells in culture, Takanezawa et al. demonstrated an enhanced toxicity upon MeHg exposure, presumably due to intracellular demethylation yielding more toxic Hg²⁺ (Takanezawa et al., 2019).

MerB is the only known enzyme capable of MeHg demethylation and is part of a larger bacterial mercury resistance system, the mer operon. MerB is a well-described organomercurial lyase, capable of cleaving the carbon mercury bond of MeHg generating methane and Hg²⁺ (Barkay, 2003; Boyd and Barkay, 2012; Mathema, 2011). In most mer operons, which also contain merA—a mercuric reductase, Hg²⁺ is then reduced to generate volatile Hg⁰. Yet, genomes containing merB in the absence of merA have been recently identified (Christakis, 2021; Dash, 2017). In this context, we have recently demonstrated an enhanced MeHg cytotoxicity attributed to intracellular merB demethylation activity in the absence of merA reduction in a bacterial expression system (Krout, 2022). These findings and others predict that demethylation of MeHg in situ may enhance the toxic outcome of MeHg exposure. To date, controlled studies of MeHg demethylation have only been completed in vitro, and thus fail to recapitulate cellular function in the context of whole organ, and systemic handling of MeHg. However, controlled experiments to evaluate toxicity changes associated with MeHg demethylation in vivo have not been possible as an endogenous demethylating activity in eukaryotes remains uncharacterized.

We have previously demonstrated the utility of the Drosophila model for studying developmental MeHg toxicokinetics and toxicodynamics (Chifiriuc, 2016; Rand, 2010, 2014; Rand and Caito, 2019). Herein we use this fly-based system to evaluate the consequences of MeHg biotransformation in vivo through targeted MeHg demethylation via transgenic merB expression. In this study, we test the hypothesis that the process of MeHg demethylation in vivo results in modulated toxicity endpoints when compared with control, non-demethylating flies. We predict that this change in toxicity outcomes will be associated with a difference in kinetics between generated Hg²⁺ and the parent MeHg compound. Using the merB enzyme expressed under Gal4-UAS control in Drosophila, we evaluate MeHg demethylation via speciation of MeHg and Hg²⁺ in flies at both the pupal and adult stages after exposure during larval feeding. In parallel, we assess developmental toxicity outcomes through assays of eclosion, a neuromuscular endpoint. We correlate dose-dependent effects on development with total Hg (tHg) levels, and relative amounts of demethylation (MeHg: Hg²⁺). Unexpectedly, we find that demethylation intracellularly results in enhanced elimination and a protective redistribution of tHg, that overall spares the animal from developmental toxicity outcomes.

MATERIALS AND METHODS

 

 

Fly stocks

Fly stocks utilized for this project are described herein, with expression patterns of Gal4 drivers illustrated in Figure 1. These fly stocks were either purchased and validated from a major Drosophila supply center (ie, Vienna Drosophila Resource Center [VDRC] or the Bloomington Drosophila Stock Center), generated via services of Rainbow Transgenic Flies Inc. (Camarillo, California) utilizing plasmid DNA described below, or made available from colleagues at our University. Briefly, flies utilized for this project include the 86Fa Phi31C line as a genetic background control, as it is the fly line in which injections occurred for creating UAS-merB transformants (Bloomington, no. 24486, originally from Blasé and Bishcof). Gal4 drivers used include Actin (Bloomington, no. 4414), Mef2 (Bloomington, no. 27390), NP1 (a gift from Dr Benoit Biteau, University of Rochester), Elav(III) (Bloomington, no. 8760), and Mef-TubGal80-ts (named Mef2-ts, generated using the previously mentioned Mef2 line in the laboratory of Dr Dirk Bohman, University of Rochester). The UAS lines created for this project include the K62B-UAS (named merB-UAS, see below.).

Figure 1.

Experimental schematic, expression pattern of Gal4 drivers, and confirmation of merB protein expression. A, Experimental paradigm utilized for MeHg exposure and Hg assessments in the Drosophila model. MeHg exposure occurs over the course of larval development from the L1 to L3 stages. MeHg levels onboard over the course of pupal development, after pupal formation (APF), remain at steady state until time of eclosion (Rand and Caito, 2019). Hg levels in adult flies decrease over time as the adults are aged on MeHg-free food. B, Schematic depicting the Gal4-UAS system for transgene expression of a gene of interest (GOI, eg, merB) tissue-specific expression is achieved with promoter/enhancers indicated by the Gal4 drivers listed, and shown in (C). C, Expression patterns of various Gal4 drivers at both the larval and adult stages. Actin—ubiquitous, Elav(III)—neurons, Mef2—Muscle, NP1—Gut. Expression was revealed by crossing each of the drivers with a UAS-GFP line and imaging with epifluorescence microscopy. Br, Brain; VNC, ventral nerve cord; SMu, somatic muscle; SG, salivary gland. D, Expression of merB protein detected by Western blotting of protein extracts of Actin > merB and Actin > Cont. pupae. Blots are probed for myc-tag-merB (top panel) and actin (bottom panel) as a control for protein loading (NS, nonspecific bands).

Ectopic expression was carried out using the Gal4-UAS system (Brand, 1994; Brand and Perrimon, 1993), with crosses of virgin females of the GAL4 driver lines with males of the UAS responder lines or with the parent 86Fa strains (genetic background) (RTF, no. 24486, Rainbow Transgenic Flies, Inc.). Expression profiles for all drivers are depicted for the larval and adult stages crossed to an GFP reporter (Figure 1). Flies were maintained on a standard fly meal containing molasses, yeast, agar, and cornmeal. Fly meal was prepared in a centralized facility at the University of Rochester (Department of Molecular Genetics). Experiments utilizing MeHg treatments were prepared with Jazz Mix Drosophila food (Fisher Scientific, no.AS153) made daily as described previously (Rand, 2008). As a quality control measure, we routinely find Jazz Mix food contains a background level of <0.01 ppm Hg, which is considered inconsequential relative to the MeHg additions (eg, 5 µM [1 ppm] MeHg). Flies were all maintained in a humidified chamber at 25°C on a 12/12 h-light/dark cycle, unless otherwise noted (eg, Mef2TubGal80ts experiments).

MerB-UAS fly generation

In an effort to create 2 merB constructs, the coding region of merB gene sequences from the mer operons of the bacterial strains Pseudomonas sp. K-62 (K62B) and Staphylococcus aureus RN23 (RN23B) (Chien, 2010) were cloned downstream of the upstreaming activating sequence (UAS) in the pUAS-attB vector (Bischof, 2007) essentially as described previously (Vorojeikina, 2017). The demethylating activity of the K-62 and RN23 merB enzyme variants have been characterized previously in a bacterial system (Chien, 2010; Krout, 2022). Furthermore, the K-62-derived merB has previously been used for expression of MeHg demethylation in mammalian cells (Takanezawa et al., 2019). No changes in the bacterial merB coding sequence were made even though we targeted expression in a eukaryotic system. In short, the RN23B construct, despite generating several transformed Drosophila lines that produced mRNA transcript, failed to produce protein in downstream analyses (data not shown), whereas the K-62 construct proved to produce functional merB enzyme (shown in results below).

The merB coding sequence used as a template for PCR was obtained through restriction digestion (Xba1 and BamHI) and the DNA sequences specific for merB were excised out of the previously described pET9a bacterial vector construct (Chien, 2010; Krout, 2022). The following PCR primers were utilized to add on the BglII and KpnI restriction sites (N and C terminus, respectively) as well as an N-terminal myc-tag, used for downstream immuno-detection of K62B merB protein.

K62B (restriction sites in bold; Myc-tag sequence in italics):

F: (5′-CGCAGATCTATGGAACAAAAACTCATCTCAGAAGAGGATCTGATGGACAAGACTATTTAT-3′)

R: (5′-CGCGGTACCTCATACTGGGCTTTCCTC-3′)

PCR product was then cleaved and ligated into BglII and KpnI sites on the pUAS-attB vector and sequenced verified to confirm proper orientation. This DNA construct was sent to Rainbow Transgenic Flies Inc. for injection and PhiC31 integrase transformation into fly embryos carrying acceptor sites at 86Fa on the third chromosome (genetic parent line, 86Fa strain RTF, no. 24486, Rainbow Transgenic Flies, Inc.) Genomic insertion of the merB construct into the fly genome was confirmed with PCR using DNA isolated from 15 adult UAS-K62-merB flies homogenized in 250 µl of 0.1 M Tris-HCl, 0.1 M EDTA, and 1% SDS, extracted with phenol-chloroform, isopropanol precipitated and rinsed in Tris-EDTA (TE) buffer (data not shown).

Transcription of the merB enzyme coding sequence was confirmed via qPCR (data not shown) using RNA from Actin Gal4 > UAS-K62-merB flies. Total RNA was isolated from late-stage pupae (pooled samples of n = 10) using a 5:1 TRIzol reagent: chloroform solution (Fischer, no. 15596018). MerB RNA presence was verified and compared with the housekeeping gene RP49 by qPCR using the iTaq Universal SYBR Green One-Step Kit (Fisher, no. 1725151). The following primer sequences were used.

RP49:

F: (5′-AGTATCTGATGCCCAACATCG-3′)

R: (5′-TTCCGACCAGGTTACAAGAAC-3′)

K62B:

F: (5′-ATGGAACAAAAACTCATCTCAGAAGAGGAT-3′)

R: (5′-GGTGAGAGTGATCGGCCTAC-3′)

Immunoblotting

Protein was harvested from 15 late-stage pupae of each genotype by homogenization in 300 µl of Protein Lysis Buffer (50 mM Tris, 150 mM NaCl, 1% IGEPAL, 1% Halt protease inhibitor cocktail) (Thermo Fisher Scientific, no.87785). After centrifugation at 10 000rpm for 10 min at 4°C, supernatant was collected and 100 µl of each sample was placed into 100 µl of 5% BME and 95% Laemmli Buffer and boiled for 5 min at 100°C. Proteins were separated on a 12% SDS-PAGE gel and subsequently electrotransferred to polyvinylidene difluoride (PVDF) membranes (Millipore Sigma, Vancouver, Canada). Membranes were probed with either an anti-actin JLA-20, (Developmental Studies Hybridoma Bank) or anti-myc tag antibody (9E10, Abcam) at a 1:2500 or 1:2000 dilution, respectively. Blots were incubated with goat anti-mouse conjugated with horseradish peroxidaseat a 1:2500 dilution and developed via chemiluminescence using Clarity Western enhanced chemiluminescence assay (Bio-Rad, Hercules, California) and imaging the gel according to the manufacturers recommendations on a ChemiDoc MP imaging system. Actin was used as a control to confirm equal protein loading between genotypes.

Developmental toxicity assay: Eclosion

Eclosion assays were performed as described previously (Rand, 2014). Briefly, first instar larvae were loaded on vials of fly food (Jazz Mix, Fisher Scientific, no. AS153) containing MeHg at concentrations ranging from 0 to 20 µM. Replicates of 3–4 vials (n = 150–200 larvae/flies) were assessed at each concentration of MeHg exposure. Emergence of the adult fly from the pupal casing was determined 13 days post loading larvae on food. Rates of eclosion for each of the indicated strains at each MeHg concentration were measured as a proportion of the total number of larvae loaded that exited the pupae casing and expressed as a percentage. Technical variation is reflected by the standard deviation of the 3–4 replicates and is represented by error bars for each data point.

Mercury analysis: Total Mercury

The total mercury (tHg) (tHg = MeHg + Hg²⁺) was determined in flies at the pupal and adult stages to determine the body burden reached relative to MeHg dose. The distribution of tHg between the head and body regions was also determined in late-stage pupae. tHg was determined on a wet weight basis µg/g (ppm), or as a total ng per number of flies or body part, via thermal degradation and amalgamation methods coupled with atomic adsorption using a Direct Mercury Analyzer (DMA-80) (Milestone SRL). For each measurement, 10–20 pupae or adult whole flies were weighed as a pooled sample on a Mettler MX5 microbalance (Mettler, Columbus, Ohio) and loaded directly on the DMA-80. The direct mercury analyzer was calibrated prior to analysis using reference materials (SRM, no. IAEA-086 and IAEA-085 Certified Reference Hair, IAEA, Vienna, Austria).

Mercury analysis: Speciation of MeHg and Hg²⁺

Determination of MeHg and Hg²⁺ isoforms in flies of the various genotypes was assessed using high-performance liquid chromatography coupled with inductively coupled plasma mass spectroscopy (HPLC ICP-MS) similar to the methods described previously (Krout, 2022). Late-stage pupae (10–20 pooled) were collected, weighed, and homogenized in 300 µl of a digestion buffer solution containing 3.25 mM glutathione (GSH), 32.3 mM HCl, and 20 mM KCl and mixed overnight on a platform shaker at room temperature. Digested samples were subsequently filtered through a 0.45-μm cellulose acetate filter, and 50 μl of filtrate was subjected to analysis via HPLC ICP-MS using a Perkin Elmer NexION 2000 ICP-MS with a 150-mm Agilent Eclipse XBD C18 HPLC column and a pore size of 5 μm. The mobile phase used for all experiments was 0.1% cysteine and 0.1% HCl run isocratically with a flow rate of 1 ml/min. Instrument calibrations were performed prior to analysis using standards for both MeHg and Hg²⁺ at 500, 250, 125, 62.50, 31.25, 15.62, 7.81, and 3.90 ng/ml. Empower software (Perkin Elmer) was used to generate the LC output, whereby peak volumes generated from all samples were compared against those of the calibration curve to determine Hg concentration for each Hg species. Instrument conditions for all runs were 1400 W power in standard mode, measuring the ²⁰²Hg isotope. The limit of detection (LOD) for both MeHg and Hg²⁺ in LC ICP-MS experiments was 0.1 ng/ml (0.1 ppb), and the limit of quantification is 10× the LOD, 1 ng/ml (1 ppb). Total MeHg and Hg²⁺ was determined and expressed on a wet weight bases of µg/g (ppm).

Elimination rate

The elimination rate of tHg was determined with adult flies. In this system, the animal accumulates MeHg during the larval feeding stages (L1–L3), with MeHg then being retained without excretion through the pupal stages. Thus, a steady state of tHg is achieved which then declines once the adult emerges (ecloses) and begins excretion of Hg via the feces. Elimination rate could thereby be determined in a population of MeHg exposed larvae by making tHg measurements in 10–15 pooled adult flies collected from the population at various days post eclosion. The elimination rate constant, k_el (days⁻¹), was determined from the slope of a linear regression line fit to the plot of ln(Hg) versus time, created in Excel.

Statistical analysis

Statistics were performed using JMP Pro 16.0 software (Cary, North Carolina). Eclosion assay statistics were performed using similar methods to those reported previously (Prince, 2014; Vorojeikina, 2017). Briefly, for eclosion assays, a 2-tailed z test was performed using percent of flies successfully eclosed at each MeHg concentration as a noncontinuous value between 0% and 100%. A z test was chosen because eclosion rate is determined as a proportion value (number of flies eclosed out of n = 150–200, reported as percentage) and proportion values are restricted at the edges (ie, near 100%). Categorical treatment of MeHg concentrations is seen as more straightforward and interpretable for statistical purposes due to the narrow concentration range of the dose response in this model system. Statistics were performed by comparing respective genotype controls to the Gal4-UAS cross. All values are represented as the average of 3–4 replicates of 50 animals/vial ± the standard deviation (total n = 150–200). Statistical significance was considered reached if the p value was ≤.05.

For Hg body burden and distribution, 3–4 replicate Hg determinations were done for each MeHg concentration of each strain, using pooled samples of pupae, adult flies, or body regions (head or abdomen/thorax). Two-tailed pairwise t-tests were used to determine significant difference between genotypes at equal concentrations of MeHe exposure (ie, Actin > 86Fa vs Actin > K62B). Statistical significance was considered reached if the p values was ≤.05.

RESULTS

 

Expression and Activity of Bacterial merB Enzyme in Drosophila

To examine the effects of targeted intracellular MeHg demethylation, we first sought to produce functional bacterial merB under the spatiotemporal control of the Drosophila Gal4-UAS system. We successfully cloned the coding region of K-62 merB into a UAS expression construct that produced positive transformant flies as confirmed PCR of genomic DNA and RT-qPCR of merB mRNA transcripts in Actin Gal4 > UAS-K62B flies (data not shown). Furthermore, protein product of the correct size for merB + the myc-tag was detected by Western blotting of Actin Gal4 > UAS-K62B pupal extracts (Figure 1D). The UAS-K62B responder line was therefore utilized for the following analyses and referred to as UAS-merB from here forward.

We next sought to confirm that the merB was enzymatically active and capable of demethylating MeHg in this eukaryotic expression system. The overall scheme for expression and analysis is illustrated in Figures 1A–C. MeHg exposures were implemented with feeding at the larval stages (Figure 1A), which is known to produce the accumulation of MeHg that reaches a steady-state level and persists through pupal development (Rand, 2014; Rand and Caito, 2019). Levels of tHg in the whole animal (body burden) and speciation of Hg (MeHg vs Hg²⁺) were then determined at pupal and adult stages (Figure 1A). Targeted expression of merB was implemented with various Gal4 “drivers” using the Gal4 > UAS expression system to achieve ubiquitous or targeted expression to specific tissues including muscle, gut, and nervous system (Figs. 1B and 1C). MeHg demethylation was then determined in each Gal4-UAS cross by speciating and quantifying Hg²⁺ and MeHg in MeHg exposed flies expressing merB and controls. For example, Hg²⁺:MeHg proportions were compared between flies where merB was expressed under the control of the Actin driver (Actin > merB) and control flies exposed to various MeHg concentrations and treated identically (Figure 2A). Control flies consisted of the Gal4 driver crossed with the parent 86Fa strain which was used to generate the UAS transformants (referred to as “cont.” hereafter) (see Materials and Methods). In Actin > Cont. flies, the majority of tHg was found in the MeHg form at all concentrations of exposure (92%–97% MeHg) (Figure 2A.). This finding confirms previous observations of a lack of MeHg demethylation in Drosophila as the remaining Hg²⁺ (3%–8%) is derived from low levels of Hg²⁺ in the stock solution (Rand and Caito, 2019). In contrast, merB expressing flies showed a significant proportion of the tHg in the Hg²⁺ form (46%–56%) at all levels of MeHg exposure (Figure 2A). This finding confirms the bacterial-derived merB enzyme functions properly in vivo to demethylate MeHg and generate Hg²⁺.

Figure 2.

Effects of ubiquitous merB expression. A, Speciation of Hg in Actin > merB and Acin > Cont. pupae after larval MeHg exposure at the indicated concentrations. The Hg²⁺ and MeHg was evaluated via ICP-MS and is expressed as a percent of the tHg as either species. B, The corresponding body burden (in ppm) of Actin > merB and Actin > Cont. pupae as treated in (A). Mean body burden and standard deviation is expressed. (n = 3–4 determinations of 10–20 pupae. A t test was used to compare means with significance set at p < .05. *p < .05, * *p < .005, ***p < .0005). C, Eclosion behavior of Actin > merB and Actin > Cont. treated with MeHg exposures ranging from 0 to 20 µM. (n = 150–200 flies [3–4 replicates of 50 larvae loaded]. Mean and standard deviation is indicated. z test statistic: *p < .05, **p < .005, ***p < .0005).

Effects of Ubiquitous Expression of merB on Body Burden and Development

We next investigated the effects of MeHg demethylation in situ on kinetic and toxicity outcomes in developing flies. To broadly assess net effects of demethylation on uptake and elimination of Hg over the course of larval development, we measured tHg in pupae in order to determine Hg body burden. In both merB expressing and control flies, a dose-dependent increase in body burden is seen with larval feeding exposures of 5–20µM MeHg in the food (Figure 2B). However, with ubiquitous merB expression using the Actin driver (Actin > merB) the tHg body burden was significantly lower at all exposure concentrations, with levels reduced to half or less than levels of tHg in control flies (Actin > Cont.), particularly at the higher MeHg exposure levels (Figure 2B).

In parallel, we assessed the developmental toxicity experienced by flies reared on the same concentrations of MeHg with and without merB expression using an eclosion assay. Eclosion, the act of the adult fly exiting from the pupae casing through coordinated pulsatile contractions, is the first neuromuscular movement the adult fly makes at the completion of metamorphosis (Denlinger, 1994; Kimura and Truman, 1990; Rivlin, 2004). Inhibition of eclosion can be attributed to a failure in the proper development of the neuromuscular system (Montgomery, 2014; Rivlin, 2004). In the absence of MeHg exposure, Actin > merB and Actin > Cont. flies eclose at identical rates (Figure 2C). However, we see a significant rescue of the MeHg-induced inhibition of eclosion with Actin > merB flies at all MeHg concentrations, whereby at 15 and 20 µM exposures 71% and 73% of the flies successfully eclose compared with 2% and 0% of the Actin > Cont. flies, respectively (Figure 2C). Together, these data demonstrate that ubiquitous upregulation of MeHg demethylation not only results in a reduction in tHg body burden, but also a corresponding rescue to MeHg-induced developmental toxicity.

Effects of Tissue-Specific Expression of merB on Body Burden and Development

We next sought to investigate the sensitivity of individual target organs to the toxicity of propagating Hg²⁺ from MeHg in situ. Therefore, we targeted merB activity specifically to the gut (NP1 Gal4), muscle (Mef2 Gal4), and neurons (Elav(III) Gal4) (see Figure 1C). Expression of merB in the gut (NP1 > merB) and muscle (Mef2 > merB) gave a significant reduction in tHg body burden compared with NP1 > Cont. and Mef2 > Cont. flies, respectively, with an overall similar profile as that seen with Actin > merB expression (compare Figs. 3A and 4A with Figure 2B). In parallel, both gut and muscle expression of merB resulted in a robust rescue of MeHg inhibition of eclosion, similar to that seen with Actin > merB flies (Figs. 3B and 4B with Figure 2C). In contrast, when merB expression is restricted specifically to neurons (Elav(III) > merB) there was no significant reduction in tHg body burden compared with Elav(III) > Cont. (Figure 5A). Remarkably, a significant rescue of MeHg inhibition of eclosion was observed with merB expression specifically localized in the neurons (Figure 5B).

Figure 3.

Effects on MeHg kinetics and developmental toxicity associated with muscle-specific expression of merB (A). Body burdens of Mef2 > merB and Mef2 > Cont. pupae after larval MeHg exposure at the indicated concentrations. Mean body burden and standard deviation is indicated. (n = 3 determinations of 10–20 pupae). A t test was used to compare means with significance set at p < .05. *p < .05, **p < .005, ***p < .0005 (B). Eclosion behavior of Mef2 > merB and Mef2 > Cont. flies treated as in A. (n = 150–200 flies were assessed [3–4 replicates of 50 larvae loaded). Mean and standard deviation are indicated. z test statistic: *p < .05, **p < .005, ***p < .0005).

Figure 4.

Effects on MeHg kinetics and developmental toxicity associated with gut-specific expression of merB. A, Body burdens of NP1 > merB and NP1 > Cont. pupae after larval MeHg exposure at the indicated concentrations. Mean body burden and standard deviation is indicated. (n = 3 determinations of 10–20 pupae). A t test was used to compare means with significance set at p < .05. *p < .05, **p < .005, ***p < .0005. B, Eclosion behavior of NP1 > merB and NP1 > Cont. flies treated as in A. (n = 150–200 flies were assessed [3–4 replicates of 50 larvae loaded]. Mean and standard deviation are indicated. z test statistic: *p < .05, **p < .005, ***p < .0005).

Figure 5.

Effects on MeHg kinetics and developmental toxicity associated with neuron-specific expression of merB. A, Body burdens of Elav(III) > merB and Elav(III) > Cont. pupae after larval MeHg exposure at the indicated concentrations. Mean body burden and standard deviation is indicated. (n = 3 determinations of 10–20 pupae). A t test was used to compare means with significance set at p < .05. *p < .05, **p < .005, ***p < .0005. B, Eclosion behavior of Elav(III) > merB and Elav(III) > Cont. flies treated as in A. (n = 150–200 flies were assessed [3–4 replicates of 50 larvae loaded]. Mean and standard deviation are indicated. z test statistic: *p < .05, **p < .005, ***p < .0005).

Effects of Tissue-Specific Expression of merB on Elimination and Distribution of tHg

The stark reduction in tHg body burden in pupae with ubiquitous, muscle-, or gut-localized merB expression indicated either a change in MeHg uptake or elimination occurred during the larval feeding stage compared with controls. Because uptake and elimination kinetics in larvae are complicated by the rapid growth at this stage (Church and Robertson, 1966; Rand and Caito, 2019), we turned to the adult stage to assess the kinetics of elimination specifically. To perform a comparative analysis of elimination at the adult stage, we needed to restrict expression of merB during the larval and pupal stages in order to obtain an equivalent body burden in newly eclosed adults of control and merB expressing flies, at which time merB expression could be turned on. To achieve this, we used a Mef2 Gal4 driver line containing a temperature responsive (ts) expression system that utilized a TubGal80 repressor of Gal4 (Mef2-ts >). With this construct, the merB expression is repressed at 25°C and induced when flies are subject to 29°C exposure. By keeping larvae and pupae at 25°C, then moving the 1-day old adult flies to 29°C, we could induce merB expression only at the adult stage and evaluate Hg body burden in Mef2-ts > Cont. flies and Mef2-ts > merB flies in parallel over time (Figure 6). The loss in tHg over time for both fly strains showed a first-order decay allowing us to determine elimination rates from linear fits to the semilog transformed data (Figure 6A). The elimination rate for tHg in merB expressing flies was seen to be much faster than that of control flies (t_1/2 = 3.1 days vs 7.2 days, Figure 6B, inset). We next evaluated the adult flies for their MeHg demethylation activity. Speciation of tHg in the adult Mef2-ts > merB flies reared on 5 µM MeHg food showed 18% Hg²⁺ compared with the 2% Hg²⁺ seen in control flies (Figure 6C), confirming that merB demethylation was occurring in conjunction with faster elimination. This result suggests that the reduced body burden observed at the pupal stage is most likely a result of a faster elimination of the generated Hg²⁺ product of demethylation of MeHg.

Figure 6.

Elimination rate changes associated with merB expression during the adult life-stage. A, Total Hg (tHg) in pooled samples of 10–15 adult Mef2-ts > merB and Mef2-ts > Cont. flies collected at various time points post eclosion from day 1 to day 11. Arrow depicts the day at which Hg was speciated in adult flies, shown in (C). B, Elimination rates for tHg in Mef2-ts > merB and Mef2-ts > Cont. adult flies. C, Speciation of Hg in adult flies at day 7 post eclosion as measure by LC-ICP-MS.

We next reexamined the flies expressing merB in neurons (Elav(III) > merB), where a rescue of MeHg-inhibited eclosion is seen despite no change in body burden. A kinetic factor not accounted for in the overall tHg body burden determinations is distribution of tHg among tissues. Therefore, we examined if Elav(III) > merB demethylation in neurons cause a redistribution of Hg away from bulk of the neural tissue, which is contained in the brain and eyes of the head. To test this, we measured tHg in isolated heads, thoraces, and abdomens of Elav(III) > merB and Elav(III) > Cont. flies reared on MeHg food. First, we found no significant difference in tHg body burden when summing the tHg in the head and body regions of Elav(III) > merB compared with Elav(III) > Cont. flies (Figure 7A), consistent with our prior determination in intact flies (see Figure 5A). Speciation of Hg in these tissues verified that demethylation was occurring in the heads of merB expressing flies, showing an increase in the proportion of Hg²⁺ in Elav(III) > merB flies (21%–26% of tHg) relative to Elav(III) > Cont. flies (5%–6% of tHg) (Figure 7B). Assessing the concentration of tHg in the head and body regions separately, we find there is a trend toward less Hg in the heads as a percent of the total Hg (Figure 6C). Evaluating Hg amounts with respect to concentration on a wet weight basis, significantly less tHg is seen in the heads of Elav(III) > merB flies when compared with controls (Elav(III) > Cont. = 152 ppm, Elav(III) > merB = 101 ppm) (Figs. 7C–D). This 33% decrease in the head did not appear to cause a difference across the whole body tHg amounts because the head constitutes only approximately 10%–15% of body mass (Rand and Caito, 2019). This result is consistent with demethylation, and the formation of Hg²⁺, causing a redistribution of tHg away from locations where MeHg demethylation occurs, in this case, sensitive sites such as the neurons of the central brain. Nonetheless, the composition of Hg remaining in the heads is comprised of substantially more Hg²⁺ than controls, still pointing to a potential toxicodynamic difference between the Hg species.

Figure 7.

Distribution of Hg species with neuronal-specific expression of merB. A, Total Hg (ng) in combined head and body regions of late-stage Elav(III) > merB and Elav(III) > Cont. pupae after larval exposure at 10 µM MeHg. (n = 45–60 body regions) (3 replicates of 15–20 head or body regions combined). Mean and standard deviation is indicated. A t test was used with statistical significance set at p < .05. B, Speciation of Hg in heads of late stage pupae at indicated MeHg exposure concentrations. C, Hg in the various body regions of Elav(III) > merB and Elav(III) > Cont. late stage pupae expressed as a percentage of total Hg. D, Hg in the various body regions of Elav(III) > merB and Elav(III) > Cont. late stage pupae expressed as a concentration of Hg (ppm) (n = 3 pooled samples of 15–20 pupae. Mean and standard deviation are indicated. A t test was used to compare means with statistical significance set at p < .05. *p < .05, **p < .005, ***p < .0005). E, Hg in the various body regions of Mef2 > merB and Mef2 > Cont. late-stage pupae expressed as a percentage of total Hg. (n = 3 pooled samples of 15–20 pupae. Mean and standard deviation is indicated. A t test was used to compare means with statistical significance set at p < .05).

To further evaluate the influence of demethylation on redistribution, we also assessed the distribution of tHg in Mef2 > merB flies compared with Mef2 > Cont. flies exposed to 20 µM MeHg. The distribution of tHg between the head, abdomen, and thorax, expressed as a percent of total Hg, we demonstrated that Mef2 > merB flies have a reduced level of tHg in the thorax with a corresponding increased level of tHg in the abdomen (Figure 7E). Because the bulk of the tissue in the thorax is the indirect flight muscles, this profile is consistent with a substantial redistribution of tHg away from the thorax, presumably due to the generation and transport of the Hg²⁺ species.

Generation of Hg²⁺ Reduces Developmental Toxicity

With the observation of a rescue to toxicity, independent of a body burden reduction, as seen with the neuron specific merB activity, we wanted to further assess if reduced toxicity stemmed entirely from changes in kinetics (transport/elimination/distribution) or due to a change in the specific activity (toxicodynamics) of Hg²⁺ compared with MeHg.

Therefore, we attempted to remove the elimination rate difference seen with the muscle-specific drive using Mef2-ts > merB flies and inducing expression only at the pupal stage, prior to eclosion and where elimination (excretion) does not occur. We reared Mef2-ts > merB and Mef2-ts > Cont. larvae on MeHg (15 µM) food from larval to pre-pupal stage at 25°C. At day 1 of pupal development, we then moved the flies to 29°C to induce expression of merB. Late-stage pupae from both Mef2-ts > merB and control lines showed similar amounts of tHg on board post temperature induction (Figure 8A). However, the proportion of MeHg to Hg²⁺ was substantially different with Mef2-ts > merB flies showing a demethylation of approximately 50% of the MeHg in pupae at the late stage of development compared with controls where approximately 1% of tHg is Hg²⁺ (Figure 8B). Assessing eclosion ability in these same fly lines, we see a moderate rescue of MeHg-inhibited eclosion with pupal-induced merB expression at 10 and 15 µM MeHg exposures (Figs. 8C–D). Together, these data indicate that conversion of MeHg to Hg²⁺, where elimination is restricted, does not result in more toxicity, supporting the notion that Hg²⁺ is not a substantially more toxic species of Hg compared with MeHg in this developmental context.

Figure 8.

Changes in MeHg developmental toxicity associated with muscle-specific expression of merB during the pupal stage. A, Body burden of 15 pooled late-stage Mef2-ts > merB and Mef2-ts > Cont pupae post heat shock induction (HS) of merB expression. B, Speciation of Hg in Mef2-ts > merB and Mef2-ts > Cont pupae at various stages pre- and post-heat shock induction. C and D, Eclosion ability of Mef2-ts > merB and Mef2-ts > Cont flies without (C) and with (D) heat shock induction of merB at the pupal stage. Mean and standard deviation is shown. (n = 150 [3 replicates of 50 loaded larvae]. A t test was used with statistically significance set at *p < .05, **p < .005, ***p < .0005).

DISCUSSION AND CONCLUSION

Despite the longstanding recognition of the toxicity associated with methyl- and inorganic mercury (MeHg and Hg²⁺), the relative toxicity of these 2 different Hg species in animal tissues remains incompletely understood. To our knowledge, this study is the first to assess the kinetics and developmental toxicity consequences of experimentally controlled demethylation of MeHg in situ of tissues and organs in a whole animal. These experiments were made possible due to the fact that the bacterial merB enzyme is capable of demethylation without the need for subsequent Hg²⁺ reduction (Krout, 2022), and our findings here that bacterial merB expressed in Drosophila can introduce MeHg demethylation to the system. Overall, we find that demethylation of MeHg, systemically, when targeted to all tissues, or limited to discrete organs of the muscle, gut, and the nervous system, results in a reduction of the toxicity experienced upon MeHg exposure. Our data shows that this reduced toxicity likely stems from a modulation in Hg kinetics causing a reduction in accumulation of MeHg (body burden) and/or redistribution of tHg. The fact that adult flies expressing merB eliminate tHg twice as fast as control flies leads to the notion that the production of Hg²⁺ is supporting faster elimination kinetics. We find that demethylation activity confined to the nervous system is protective against MeHg developmental toxicity, which is accompanied by a redistribution of tHg away from the brain, again implicating enhanced transport kinetics with Hg²⁺ production. In addition to these kinetic attributes, our data show that MeHg demethylation induced in the pupal stage, where whole body elimination does not occur, results in a reduced developmental toxicity (ie, rescue of eclosion), which is counter to the notion that Hg²⁺ carries a greater toxic potency. These findings expand our understanding that demethylation of MeHg in situ in host tissues has the potential to contribute significantly to reducing the toxic outcomes of MeHg exposure. The implications of this novel finding are discussed below:

We observe a stark upregulation of Hg²⁺ in MeHg exposed flies where merB is expressed under the control of the Gal4-UAS system, affording us a model to generate Hg²⁺ intracellularly and bypass absorption and uptake differences between Hg species. This observation mirrors that of previous in vitro cell culture work using the merB enzyme for demethylation (Chien, 2010; Krout, 2022; Schottel, 1978; Takanezawa et al., 2019), but extends this model to grant us the novel ability to examine this process in vivo in a whole animal system. In contrast to our findings here, prior in vitro experiments characterizing the effects of merB mediated demethylation in bacterial (Krout, 2022; Schottel, 1978) and mammalian cells (Takanezawa et al., 2019) have found that an increased toxicity results when Hg²⁺ is generated intracellularly. However, in these prior studies, the systems level handling of Hg²⁺ and MeHg that occurs in vivo is not account for. For example, the transporters expressed in isolated cell and bacterial cultures likely differ from those expressed in the context of a whole-body system, and may not achieve transport and elimination or distribution processes seen in a multi-organ paradigm. In this regard, our whole animal Drosophila model is more likely to express the full complement of relevant transporters which could prove more efficient in removing the Hg²⁺ product from sensitive target sites and the system. Whereas in vitro the cultured cells may be less likely to overcome persistent accumulation of intracellular Hg²⁺ and inadvertently reflect a greater toxicity.

Interestingly, and again in contrast to our findings here, seminal work on long term, subclinical, MeHg exposure in vivo in Macaca fascicularis (Crab-eating macaques) reported MeHg demethylation in the brain, where the Hg²⁺ generated here showed much slower elimination than MeHg (Vahter, 1995). Most brain regions (such as thalamus and pituitary) show a Hg²⁺  t_1/2 of approximately 230–540 days compared with a t_1/2 of approximately 37 days for MeHg (Vahter, 1995). Nonetheless, the Hg²⁺ constituted only 9%–12% of the tHg in most brain region in these analyses. Interestingly, the pituitary demonstrated accumulation of up to 46% of Hg²⁺, suggesting that endogenous demethylation activity does exist, and can possibly vary even among brain regions (Vahter, 1994, 1995). Although this previous study finds a slower half-life associated with Hg²⁺ the context is primarily in specific regions of the brain and the exposure paradigm involves fully mature animals. Our model and analyses encompass the potential influence of target organs across the whole body, and furthermore, investigates a developmental exposure paradigm. Overall, the mode of action associated with enhanced tHg elimination and redistribution in our model remains to be resolved. One possible mechanism could be a difference in transporters utilized for Hg²⁺ but not MeHg expressed during development. There may be a higher abundance of Hg²⁺ specific transporters in the tissues we assessed. In this way, the generation of Hg²⁺ from MeHg would result in a form of Hg more easily excreted out of target sites and eventually the whole body. In this regard, the Drosophila model is ideally poised for future experiments to address the candidate transporters in a forward genetics approach, as has previously been shown (Prince, 2014).

Although we predict that the majority of toxicity rescue observed in our experiments is due to the kinetic changes and decreased body burden associated with the production of Hg²⁺, there is evidence that points to an inherent lower toxicity associated with Hg²⁺ compared with MeHg. In a recent study by Beamish (2021), it was observed that with respect to internal dose, Hg²⁺ showed starkly less toxic potency compared with MeHg as measured by the same developmental assessment of eclosion as is used here (Beamish, 2021). Interestingly, other developmental endpoints, such as pupariation do not show any toxic potency difference between MeHg and Hg²⁺ at equal internal doses (Beamish, 2021). Note that this rescue in eclosion could actually be a result of changes to internal distribution seen with Hg²⁺, as is observed in our current work when demethylation is upregulated in the neural tissue (Elav(III) > merB). Similarly, our later experiments with the Mef2-ts driver, expressing only during pupariation, may reflect a rescue due to Hg redistribution. Alternatively, the enhanced elimination and redistribution of Hg²⁺ compared with MeHg may still be consistent with Hg²⁺ having an inherent higher toxic potency, and reflect a greater need for kinetic and dynamic mechanisms to be in place to specifically handle this Hg species.

One limitation to this study is the use of the merB enzyme to upregulate the process of MeHg demethylation. This artificially induced mechanism may not be representative of the endogenous demethylation occurring in mammalian and human systems. However, with a lack of understanding of the endogenous mechanisms for MeHg demethylation, merB becomes a unique tool capable of assessing the effects of MeHg demethylation in vivo. Another limitation is that our assessment relies primarily on developmental toxicity outcomes. Although MeHg is most potent during the developmental stages, generation and accumulation of Hg²⁺ have been observed most commonly to occur over longer periods of time such as into the adult life stages (Pamphlett, 2021; Vahter, 1995). Therefore, the rescue to toxicity seen in the developmental stage may exaggerate the phenotype that would be observed in the adult stage. Future studies examining toxicity during the adult life stages will be necessary for a more complete picture of relevant MeHg exposure paradigms. Lastly, although rescue to toxicity is observed in all cases accompanying Hg²⁺ production by merB, we cannot say for certain that there is no side effects of exogenous production of merB playing a role in this rescue. For example, one possible way in which exogenous merB production could result in a rescue, outside of direct production of Hg²⁺, would be through a bulk sequestration of Hg. Sequestration of MeHg or Hg²⁺ could tie up the highly toxic metal away from sensitive sites, thus allowing for a rescue to toxicity. However, a sequestration of either Hg species is highly unlikely to be occurring because the overall Hg elimination from the system being sped up with merB expression.

In summary, our results demonstrate that the process of MeHg demethylation, and subsequent generation of Hg²⁺, can modulate kinetic parameters of elimination and distribution of Hg, lowering the body burden experienced by flies post MeHg exposure. This change in kinetics is accompanied by a rescue to developmental toxicity outcomes, highlighting a potential protection associated with the systemic demethylation of MeHg. Overall, these results indicate that the rate/efficiency/variability of MeHg demethylation, which may be regulated by host genetics, can play a role in susceptibility or resistance to MeHg exposure for an individual due to its effects on overall kinetics and toxicity.