Search

An organomercurial phenylmercury activates AtPCS1, an enzyme known for detoxification of inorganic metal(loid) ions in Arabidopsis and the induced metal-chelating peptides phytochelatins are essential for detoxification of phenylmercury. Small thiol-rich peptides phytochelatins (PCs) and their synthases (PCSs) are crucial for plants to mitigate the stress derived from various metal(loid) ions in their inorganic form including inorganic mercury [Hg(II)]. However, the possible roles of the PC/PCS system in organic mercury detoxification in plants remain elusive. We found that an organomercury phenylmercury (PheHg) induced PC synthesis in Arabidopsis thaliana plants as Hg(II), whereas methylmercury did not. The analyses of AtPCS1 mutant plants and in vitro assays using the AtPCS1-recombinant protein demonstrated that AtPCS1, the major PCS in A. thaliana, was responsible for the PheHg-responsive PC synthesis. AtPCS1 mutants cad1-3 and cad1-6, and the double mutant of PC-metal(loid) complex transporters AtABCC1 and AtABCC2 showed enhanced sensitivity to PheHg as well as to Hg(II). The hypersensitivity of cad1-3 to PheHg stress was complemented by the own-promoter-driven expression of AtPCS1-GFP. The confocal microscopy of the complementation lines showed that the AtPCS1-GFP was preferentially expressed in epidermal cells of the mature and elongation zones, and the outer-most layer of the lateral root cap cells in the meristematic zone. Moreover, in vitro PC-metal binding assay demonstrated that binding affinity between PC and PheHg was comparable to Hg(II). However, plant ionomic profiles, as well as root morphology under PheHg and Hg(II) stress, were divergent. These results suggest that PheHg phytotoxicity is different from Hg(II), but AtPCS1-mediated PC synthesis, complex formation, and vacuolar sequestration by AtABCC1 and AtABCC2 are similarly functional for both PheHg and Hg(II) detoxification in root surficial cell types.

SQSTM1/p62, hereinafter referred to as p62, is a stress-induced cellular protein that interacts with various signaling proteins as well as ubiquitinated proteins to regulate a variety of cellular functions and cell survival. Methylmercury (MeHg) exposure increases the levels of p62, the latter playing a protective role in MeHg-induced toxicity. However, the underlying mechanism by which p62 alleviates MeHg toxicity remains poorly understood. Herein, we report the interaction of p62 with neural precursor cell expressed developmentally down-regulated protein 4 (NEDD4), a HECT E3 ubiquitin ligase. The region of p62 where NEDD4 binds is located at the proline- and arginine (PR)-rich region (amino acids: 102-119), C-terminal extension of the Phox and Bem1 (PB1) domain. To evaluate the importance of the p62-NEDD4 complex, we examined the compensation of deletion mutant (GFP-Δ102-119 p62) for the lack of endogenous p62 in MEFs. GFP-p62/p62KO cells exhibited significantly higher cell viability than GFP-Δ102-119 p62/p62KO cells after treatment with MeHg. Our findings suggest novel mechanisms to alleviate MeHg toxicity through p62-NEDD4 complex formation.

Gadolinium-based contrast agents (GBCAs) are widely used in clinical magnetic resonance imaging (MRI). Free gadolinium ions (Gd3+) released from GBCAs potentially increase the risk of GBCA-related toxicity. However, the cellular responses to Gd3+ and the underlying mechanisms responsible for protection against Gd3+ remain poorly understood. Recently, autophagy has been considered a cell survival mechanism against various toxic metals. Here, we investigated the relationship between Gd3+ and autophagy, as well as the effect of autophagy inhibition on the survival of cells exposed to Gd3+. We found that the increased expression of microtubule-associated protein 1 light chain 3 (LC3)-II, a marker protein of autophagy, in Gd3+-exposed human embryonic kidney 293 (HEK293) cells. Moreover, we found a greater accumulation of LC3-II after exposure to an autophagy inhibitor, chloroquine (CQ), combined with Gd3+ than that after exposure to CQ alone, suggesting that Gd3+ activated autophagy in HEK293 cells. Furthermore, we found that Gd3+ reduced cell viability, which was more pronounced after CQ treatment. Our findings indicated that autophagy exerted a cytoprotective effect against Gd3+ toxicity, suggesting a potential link between autophagy and GBCA-associated adverse events.

For a better understanding of metal–ligand interaction and its function in cells, we developed an easy, sensitive, and high-throughput method to quantify ligand–metal(loid) binding affinity under physiological conditions by combining ligand-attached affinity beads and inductively coupled plasma-optical emission spectrometry (ICP-OES). Glutathione (GSH) and two phytochelatins (PC2 and PC3, small peptides with different numbers of free thiols) were employed as model ligands and attached to hydrophilic beads. The principle of the assay resembles that of affinity purification of proteins in biochemistry: metals binding to the ligand on the beads and the rest in the buffer are separated by a spin column and quantified by ICP-OES. The binding assay using the GSH-attached beads and various metal(loid)s suggested the different affinity of the metal–GSH interactions, in accordance with the order of the Irving–Williams series and the reported stability constants. The binding assay using PC2 or PC3-attached beads suggested positive binding between PCs and Ni(II), Cu(II), Zn(II), Cd(II), and As(III) in accordance with the number of thiols in PC2 and PC3. We then conducted the competition assay using Cd(II), Mn(II), Fe(II), Cu(II), and Zn(II), and the results suggested a better binding affinity of PC2 with Cd(II) than with the essential metals. Another competition assay using PC2 and GSH suggested a robust binding affinity between PCs and Cd(II) compared to GSH and Cd(II). These results suggested the dominance of PC-Cd complex formation in vitro, supporting the physiological importance of PCs for the detoxification of cadmium in vivo. We also discuss the potential application of the assay.

Methylmercury (MeHg) is a highly toxic pollutant, and is considered hazardous to human health. In our previous study, we found that MeHg induces autophagy and that Atg5-dependent autophagy plays a protective role against MeHg toxicity. To further characterize the role of autophagy in MeHg-induced toxicity, we examined the impact of autophagy on microtubules and nuclei under MeHg exposure using Atg5KO mouse embryonic fibroblasts (MEFs). Low concentrations of MeHg induced a decrease in α-tubulin and acetylated-tubulin in both wild-type and Atg5KO cells. While α-tubulin acetylation was promoted by treatment with tubacin, a selective inhibitor of histone deacetylase 6, MeHg treatment inhibits the increase of tubacin-induced acetylated-tubulin. However, similar effects were observed for treatment with either tubacin or tubacin + MeHg in wild-type and Atg5KO cells. We also found a significant increase in the number of multinuclear cells upon MeHg exposure in Atg5KO MEFs compared to wild-type MEFs. In addition, DNA double strand breaks (DSBs), measured by phosphorylation of the core histone H2A variant (H2AX) on serine 139 (γH2AX), markedly increased in Atg5KO MEFs compared to wild-type MEFs. Our results therefore suggest that autophagy is not a simple elimination pathway of MeHg-induced damaged proteins, but that it also plays a protective role in the context of MeHg-associated DSBs.

Mercury superfamily proteins, i.e. inner membrane-spanning proteins (MerC, MerE, MerF and MerT) and a periplasmic mercury-binding protein (MerP), transport mercury into the cytoplasm. A previous study demonstrated that a Mer transporter homolog exhibits cadmium transport activity; based on this, the present study aimed to evaluate the cadmium transport activity of MerC, MerE, MerF and MerT and the effects of MerP co-expression in Escherichia coli. Bacteria expressing MerC, MerE, MerF or MerT without MerP were more sensitive to cadmium and significantly absorbed more cadmium than did the control strain. Expression of MerP in combination with MerC, MerE, MerF or MerT increased the bacterial sensitivity to cadmium and cadmium accumulation compared to a single expression of MerC, MerE, MerF or MerT. Cadmium uptake mediated by MerC, MerE, MerF or MerT was inhibited under cold or acidic conditions. These findings suggest that MerC, MerE, MerF and MerT are broad-spectrum heavy metal transporters that mediate both mercury and cadmium transport into cells and that MerP accelerates the cadmium transport ability of MerC, MerE, MerF and MerT.

Cancer cells enhance autophagic activity as a survival measure against metabolic and therapeutic stresses. The inhibition of autophagy may represent a valuable sensitizing target for cancer treatment. Recently, we examined the ability of various cytochalasins to inhibit autophagy and demonstrated the potent inhibitory effect of cytochalasin E (CE) on autophagic flux. The present study was conducted to investigate whether CE inhibited autophagosome-lysosome fusion, and to determine whether CE enhanced chemotherapy-induced cell death. Cell exposure to CE led to the accumulation of microtubule-associated protein light chain 3-II (LC3-II) and sequestosome-1/ubiquitin-binding protein p62 (SQSTM1/p62) in a dose- and time-dependent manner. Cells treated with CE exhibited distinct formation of p62-positive structures on lysosome-associated membrane protein 2 (LAMP2)-positive lysosomal vesicles. CE treatment following serum starvation robustly reduced cell viability and increased expression levels of LC3-II and p62, in comparison to those of cells treated with CE alone. Furthermore, combination treatment with CE and bortezomib, an inhibitor of the 26S proteasome, showed a synergistic effect in targeting human lung cancer A549 cells. Altogether, our results demonstrated that CE treatment inhibited autophagosome-lysosome fusion, and this activity, in part, augmented bortezomib-induced cell death. Therefore, we concluded that CE may be a potentially effective therapeutic agent against lung cancer, especially in a combination therapy with proteasome inhibitors.

Bacterial resistance to mercury compounds (mercurials) is mediated by proteins encoded by mercury resistance (mer) operons. Six merE variants with site-directed mutations were constructed to investigate the roles of the cysteine and histidine residues in MerE protein during mercurial transport. By comparison of mercurial uptake by the cell with intact and/or variant MerE, we showed that the cysteine pair in the first transmembrane domain was critical for the transport of both Hg(II) and CH 3Hg(I). Also, the histidine residue located near to the cysteine pair was critical for Hg(II) transport, whereas the histidine residue located on the periplasmic side was critical for CH 3Hg(I) transport. Thus, enhanced mercurial uptake mediated by MerE may be a promising strategy for the design of new biomass for use in the bioremediation of mercurials in the environment.

Some methylmercury (MeHg) is converted to inorganic mercury (Hg2+) after incorporation into human and animal tissues, where it can remain for a long time. To determine the overall toxicity of MeHg in tissues, studies should evaluate low concentrations of Hg2+. Although demethylation is involved, the participating enzymes or underlying mechanisms are unknown; in addition, the low cell membrane permeability of Hg2+ makes these analyses challenging. We established model cell lines to assess toxicities of low concentrations of Hg2+ using bacterial organomercury lyase (MerB). We engineered MerB-expressing HEK293 and HeLa cell lines that catalyze MeHg demethylation. These cells were significantly more sensitive to MeHg exposure compared to the parental cells. MeHg treatment remarkably induced metallothioneins (MTs) and hemeoxygenase-1 (HMOX-1) mRNAs and modest expression of superoxide dismutase 1, whereas catalase and glutathione peroxidase 1 mRNAs were not up-regulated. merB knockdown using small interfering RNA supported the induction of MT and HMOX-1 mRNA by MerB enzymatic activity. Pretreatment with Trolox, a water-soluble vitamin E analog, did not inhibit MeHg-induced elevation of MT-Ix and HMOX-1 mRNAs in MerB-expressing cells, suggesting that Hg2+ works independently of reactive oxygen species generation. Similar results were obtained in cells expressing MerB, suggesting that high MTs and HMOX-1 induction and cytotoxicity are common cellular responses to low intracellular Hg2+ concentrations. This is the first study to establish cell lines that demethylate intracellular MeHg to Hg2+ using bacterial MerB for overcoming the low membrane permeability of Hg2+ and exploring the intracellular responses and toxicities of low Hg2+ concentrations.

Methylmercury (MeHg) is a widespread environmental pollutant and causes a serious hazard to health worldwide. However, molecular mechanisms underlying MeHg toxicity remain elusive. We show that MeHg reduced mouse embryonic fibroblast (MEF) viability in a dose-dependent manner. Furthermore, MeHg treatment increased levels of autophagy markers LC3-II and p62, possibly by acting on the MAPKs signaling pathway in several cell types. MeHg exposure elevated the number of LC3 puncta in stable GFP-LC3 MEFs and the number of autophagic vacuoles. The accumulation of LC3-II and p62 increased further when complementing MeHg with autophagy inhibitor, chloroquine. Moreover, we found that autophagy-related gene 5-deficient (Atg5-/-) MEFs exhibited higher sensitivity and higher levels of p62 compared to their wild-type counterparts following MeHg exposure. This suggested that p62 was upregulated at the transcription level by MeHg and degraded by Atg5-dependent autophagy. Our data demonstrate that MeHg exposure promotes autophagy, and Atg5-dependent autophagy serves to protect cells from MeHg cytotoxicity.

Autophagy is a cell survival process that represents a therapeutic target in cancer treatment. Many types of cytochalasins have been identified and some of them have been reported to interfere with the formation of the autophagosome, although only limited data are available to assess their potential effects. Therefore, in this study, we examined the effects of cytochalasins and structurally related compounds on cell survival and the regulation of autophagy in human lung A549 adenocarcinoma cells. Cytochalasin D (CD) and cytochalasin E (CE) prominently inhibited the growth of A549 cells in a dose-dependent manner. Following treatment with CE, F-actin filaments were disrupted, and the proportion of binucleated cells increased, whereas no such effects were observed with the seven other cytochalasins tested. We found that cytochalasin H (CH), CD, and especially CE could induce the up-regulation of autophagy-related protein (LC3-II) and SQSTM1/p62. Using bafilomycin A1, we demonstrated that CD, CE, and CH inhibited autophagosome turnover, resulting in a dysfunctional autophagic process. The results of this study reveal that CE is the most potent cytochalasin in terms of its ability to induce cell death and inhibit autophagy. CE may therefore be an effective therapeutic agent against lung cancer.

The characteristics of bacteria take up mercury into cells via membrane potential-dependent sequence-divergent members of the mercuric ion (Mer) superfamily, i.e., a periplasmic mercuric ion scavenging protein (MerP) and one or more inner membrane-spanning proteins (MerC, MerE, MerF, and MerT), which transport mercuric ions into the cytoplasm, have been applied in engineering of bioreactor used for mercurial bioremediation. We engineered bacteria to express MerC, MerE, MerF, or MerT with or without MerP to clarify their individual role and potential in transport of mercurial. By immunoblot analysis using specific polyclonal antibody, the proteins encoded by merC, merE, merF, merT or merP, were certainly expressed and identified in the membrane fraction. Bacteria expressing MerC, MerE, MerF or MerT in the absence of MerP transported significantly more C6H5Hg(I) and Hg(II) across bacterial membrane than their isogenic strain. In vivo expression of MerP in the presence of all the transporters did not cause apparent difference to the C6H5Hg(I) transport, but gives an apparently higher Hg(II) transport than that did by MerE, MerF or MerT but not by MerC. Among the four transporters studied, MerC showed more potential to transport Hg(II) across bacterial membrane than MerE, MerF and MerT. Together these findings, we demonstrated for the first time that in addition to MerE and MerT, MerF and MerC are broad-spectrum mercury transporters that mediate both Hg(II) and phenylmercury transport into cells. Our results suggested that MerC is the most efficient tool for designing mercurial bioremediation systems, because MerC is sufficient for mercurial transport into cells.

To develop the potential of plants to sequester and accumulate mercurials from the contaminated sites, we engineered a tobacco (Nicotiana tabacum) plant to express a bacterial ppk gene, encoding polyphosphate kinase (PPK), under control of a plant promoter. The designated plant expression plasmid pPKT116 that contains the entire coding region of ppk was used for Agrobacterium-mediated gene transfer into tobacco plants. A large number of independent transgenic tobacco plants were obtained, in some of which the ppk gene was stably integrated in the plant genome and substantially translated to the expected PPK protein in the transgenic tobacco. The presence of Hg2+ did not cause considerable morphological abnormalities in the transgenic tobacco, which grew, flowered, and set seed similarly to the wild-type tobacco on the medium containing normally toxic levels of Hg2+. The ppk-transgenic tobacco showed more resistance to Hg2+ and accumulated more mercury than its wild-type progenitors. These results suggest that ppk-specified polyphosphate has abilities to reduce mercury toxicity, probably via chelation mechanism, and also to accumulate mercury in the transgenic tobacco. Based on the results obtained in the present study, the expression of ppk gene in transgenic tobacco plants might provide a means for phytoremediation of mercury pollution.

The feasibility of transgenic tobacco, engineered to express bacterial polyphosphate (polyP), for phytoremediation of mercury pollution was evaluated. T3 progeny of the transgenic tobacco produced a large amount of polyP in leaves and showed a relatively high resistant phenotype to Hg2+ than its wild-type progenitors. These results suggest that the integrated ppk gene, encoding polyphosphate kinase (PPK), a key enzyme for polyP biosynthesis, is stably conserved in tobacco genome, and translated to active PPK which catalyzed biosynthesis of polyP, and suggest that polyP is capable of reducing the cytotoxicity of Hg2+, probably via chelation formation with polyP. The transgenic tobacco expressing polyP accumulated significantly more mercury than its wild-type progenitors from Hg2+-containing agar medium and simulated soils without taxing the tobacco plants suggesting that the transported Hg2+ was accumulated as a less toxic Hg-polyP complex in the tobacco tissues. Based on the results obtained in the present study, the polyP-mediated accumulation of mercury from mercurial-contaminated soils may provide an ecologically compatible approach for phytoremediation of mercury pollution.

To investigate the individual role of MerT and MerP encoded by Pseudomonas K-62 pMR26 in the transport of phenylmercury, a series of mutants with a specific point mutation in merT and/or genetic deletion in merP were constructed and transformed into Escherichia coil XL-1-Blue. Transport of phenylmercury across the cytoplasmic membrane of E. coli mediated by MerT and MerP was inhibited by NaCN and by cold temperatures. Deletion of merP reduced, but did not completely abolish the C6H5Hg+-hyperuptake and -hypersensitive phenotypes suggesting that transport of phenylmercury into the cytoplasm of E. coli is still occurring. Mutations of the vicinal cysteine residues (Cys24 and Cys25) in the first transmembrane region of MerT to serine caused complete loss of Hg2+-hyperuptake and -hypersensitivity, whereas the mutations did not affect the C6H5Hg+-hyperuptake and -hypersensitive phenotypes. In addition, no additive effect on the C6H5Hg+-hyperuptake and -hypersensitive phenotypes was found, when mutations of the two cysteines in MerT to serine were further introduced in the merP-deleted mutants. These results clearly demonstrated that the vicinal cysteine residues of MerT are not involved in the transport of C6H5Hg+, but indeed are involved in the transport of Hg2+ as previously reported.

To investigate the role of aromatic amino acids of MerT protein in phenylmercury transport, two merT variants (pMRTm1P and pMRTm2P) with specific site-directed mutations were constructed and examined their effects on C6H5Hg+-transport. Substitution of Phe-36 and Trp-40 residues located on the periplasmic loop of MerT in turn with Ala and Val, respectively, did not affect the Hg2+-uptake, but caused a significant reduction in the C6H5Hg+-uptake by bacterial cells with intact merT gene. Introduction of specific mutations changing Phe-108, 114, 115 to Ala, and Tyr-110, 116 to Ser in the C-terminal region of the third transmembrane of MerT also caused large reduction in the uptake of C6H5Hg+, but had no effect on the Hg2+-uptake. In addition, both mutations caused a significant reduction in the hypersensitivity to C6H5Hg+, but without affecting the Hg2+-hypersensitive phenotype. Together these results suggest that the aromatic amino acids of MerT protein may play an important role in the transport of C6H5Hg+ across the cell membrane.

To investigate the biochemical mechanism responsible for NO-induced neurotoxicity, the effect of sodium nitroprusside(SNP), a NO-generating agent, on PC12 cells was studied. The cell density was dose-dependently inhibited by SNP. Neuronally differentiated PC12 cells showed a higher resistance to SNP than the undifferentiated cells. The inhibitory effect was enhanced by 8-Br-cGMP, and reduced by methylene blue. However, 8-Br-cGMP alone had no significant cytotoxicity. SNP also inhibited [3H]-thymidine incorporation into the cells in a dose-dependent manner. The dose response curves for reducing cell density and for inhibiting thymidine incorporation; were found to be virtually superimposable. These results suggested that cytotoxieity elicited by NO seemed to be due to inhibition of DNA synthesis in PC12 cells.

To study the physiological role of the red pigments in soil strain Pseudomonas K-62, we isolated a red pigment-deficient white mutant from the soil strain by treatment with mitomycin C and compared the phenotypic properties of the mutant and parent strain. The red pigments, which were classified as one of carotenoids based on their physicochemical properties, were separated into two groups, designated pigment A and B respectively on NH-Chromatorex HPLC.The crude pigments and pigment B which could react with Hg2+ in the wild-type Pseudomonas K-62 and its mercury-resistant plasmid-deficient strain were enhanced by the addition of Hg2+. The white mutant thus obtained showed a greater sensitivity to Hg2+ than the wild-type reddish strain despite containing the resistant plasmids. The major component in pigment B was identified by mass spectrometric analysis as 1-hydroxy-1-methoxy-1,2, 1′,2′,7′,8′-hexahydro-ψ,ψ-caroten-4-one, a carotenoid monoketone. These results suggested that red pigments, especially pigment B, may account, at least partially, for defense against Hg2+ in the bacterial environments.

To clarify whether the merT and merP genes play a role in the transport of methylmercury, we constructed a deletion plasmid, pMRD141 which lacked the genes conferring the organomercurial lyase and the mercuric reductase from plasmid, pMRA17 containing the entire broad-spectrum mercury resistance determinants of the 26 kb-plasmid from Pseudomonas K-62. Plasmid, pMRD141 showed hypersensitivity to Hg2+ but still expressed a normal sensitivity to methylmercury. The mercury-induced hypersensitive cells carrying pMRD141 took up appreciably more 203Hg2+ than the induced resistant cells with pMRA17 and the sensitive cells with cloning vector, Bluescript II SK(+) but no difference in the uptake of CH3(203)Hg2+ among these three strains was found. These results suggested that the merT and merP are only involved in the Hg2+ transport but do not participate in the transport of methylmercury.

Pseudomonas strain K-62 was found to contain six plasmids. A mutant derivative cured of the 26-kb plasmid showed a higher sensitivity to mercurials; however, the strain was still able to volatilize them. Loss of the 68-kb plasmid in addition to the 26-kb plasmid abolished the ability of mercury volatilization in this strain and led to a further decrease in the level of mercurial resistance. These results are the first to demonstrate that the organomercurial resistance of Pseudomonas strain K-62 is plasmid-based, and that both the 26- and 68-kb plasmids are required for full expression of the mercurial resistance. Probes specific for the mer genes merA, merB, and merR strongly hybridized with the 26-kb plasmid, but not with the 68-kb plasmid. Two fragments of the 26-kb plasmid that hybridized with the mer genes were cloned and expressed in Escherichia coli. One recombinant plasmid (pMRA17) inducibly encoded a typical broad-spectrum mercurial resistance, whereas the other recombinant plasmid (pMRB01) constitutively conferred hypersensitivity to phenylmercury in the absence of mercuric reductase activity. The results suggest that the two organomercurial lyases in the cells are transcribed from different operator-promoters.

An Escherichia coli strain was generated by fusion of a merA-deleted broad-spectrum mer operon from Pseudomonas K-62 with a bacterial polyphosphate kinase gene (ppk) from Klebsiella aerogenes in vector pUC119. A large amount of the ppk-specified polyphosphate was identified in the mercury-induced bacterium with the fusion plasmid designated pMKB18 but not in the cells without mercury induction. These results suggest that the synthesis of polyphosphate as well as the expression of the mer genes is mercury-inducible and regulated by merR. The E. coli strain with pMKB18 was more resistant to both Hg2+ and C6H5Hg+ than its isogenic strain with cloning vector pUC119. The recombinant strain accumulated more mercury from Hg2+- and C6H5Hg+-contaminated medium. Hg2+ transported into the cytoplasm appeared to be bound by chelation with the polyphosphate produced by the recombinant cells. The transported phenylmercury was degraded to Hg2+ before the chelation since polyphosphate did not directly chelate with C6H5Hg+. These results indicate that polyphosphate is capable of reducing the cytotoxicity of the transported Hg2+ probably via chelation between polyphosphate and Hg2+.

To evaluate the utility of polyphosphate kinase gene (ppk)-specified polyphosphate in mercury remediation, a fusion plasmid, pMK27, with ppk from Klebsiella aerogenes and mercury transport genes, merT and merP, from Pseudomonas K-62, was constructed. The transcription and translation of ppk, merT and merP were found to be mercury-inducible. The ppk-specified polyphosphate was identified in cells preinduced by Hg2+, but not in cells without mercury induction, suggesting that the synthesis of polyphosphate is regulated by merR. The hyper-sensitive phenotype to Hg2+, shown by bacteria with pMRD141, which contains merT and merP, was almost completely restored to its original levels when the ppk was introduced into the plasmid, suggesting that the Hg2+-toxicity was reduced by the polyphosphate, probably via chelation formation. Bacteria with pMK27 accumulated approximately 6-fold more mercury than the bacteria with cloning vector, pUC119. These results clearly demonstrate that the polyphosphate is capable of retaining mercury in the cells without taxing the cells. Based on the results obtained in the present study, the fusion plasmid pMK27 may serve as a strategy for mercury remediation.

In order to clarify the physiological role of the merE gene of transposon Tn21, a pE4 plasmid that contained the merR gene of plasmid pMR26 from Pseudomonas strain K-62, and the merE gene of Tn21 from the Shigella flexneri plasmid NR1 (R100) was constructed. Bacteria with plasmid pE4 (merR-o/p-merE) were more hypersensitive to CH3Hg(I) and Hg(II), and took up significantly more CH3Hg(I) and Hg(II), than the isogenic strain. The MerE protein encoded by pE4 was localized in the membrane cell fraction, but not in the soluble fraction. Based on these experimental results, we suggest for the first time that the merE gene is a broad mercury transporter mediating the transport of both CH3Hg(I) and Hg(II) across the bacterial membrane.