Mercuric chloride damages cellular DNA by a non-apoptotic mechanism

Mutation Research 470 (2000) 19–27

Mercuric chloride damages cellular DNA by a non-apoptotic mechanism E.Y. Ben-Ozer a , A.J. Rosenspire a , M.J. McCabe Jr. b , R.G. Worth a , A.L. Kindzelskii a , N.S. Warra a , H.R. Petty a,∗ a

Department of Biological Sciences, Wayne State University, Detroit, MI 48202, USA Institute of Chemical Toxicology, Wayne State University, Detroit, MI 48202, USA

b

Received 28 December 1999; received in revised form 7 June 2000; accepted 7 June 2000

Abstract Mercury is a xenobiotic metal that is well known to adversely affect the immune system, however, little is known as to the molecular mechanism. Recently, it has been suggested that mercury may induce immune dysfunction by triggering apoptosis in immune cells. Here, we studied the effects of Hg2+ (HgCl2 ) on U-937 cells, a human cell line with monocytic characteristics. We found that these cells continued to proliferate when exposed to low doses of mercury between 1 and 5 ␮M. Using the single cell gel electrophoresis (SCGE) or ‘comet’ assay, we found that mercury damaged DNA at these levels. Between 1 and 50 ␮M Hg2+ , comet formation was concentration-dependent with the greatest number of comets formed at 5 ␮M mercury. However, the appearance of mercury-induced comets was qualitatively different from those of control cells treated with anti-fas antibody, suggesting that although mercury might damage DNA, apoptosis was not involved. This was confirmed by the finding that cells treated with 5 ␮M mercury were negative for annexin-V binding, an independent assay for apoptosis. These data support the notion that DNA damage in surviving cells is a more sensitive indicator of environmental insult than is apoptosis, and suggests that low-concentrations of ionic mercury may be mutagenic. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Mercury; DNA; Apoptosis

1. Introduction Over the past decade, concern has escalated regarding human exposure to mercury [1]. Mercuric chloride has been shown to be mutagenic in Chinese hamster ovary cells, while several studies have indicated that mercuric chloride or organic mercury can

∗ Corresponding author. Tel.: +1-313-577-2896; fax: +1-313-577-9008. E-mail address: [email protected] (H.R. Petty).

cause tumors in rodents [2,3]. Recent results show that occupationally-exposed chloralkali workers can have urine mercury levels as high as 2921 nmol/l per year [4]. However, even for non-industrially exposed individuals, studies have consistently shown that mercury levels found in body fluids and tissues, while quite variable, can be surprisingly high [5]. For approximately 3% of the population blood levels of mercury exceed 0.25 ␮M and 4% of the population exceed 0.125 ␮M mercury in urine, with the highest concentrations reported in individuals with no known special risk factors to be about 1 ␮M [6].

1383-5718/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 3 - 5 7 1 8 ( 0 0 ) 0 0 0 8 3 - 8

20

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

Mercury is a xenobiotic metal that has been reported to promote immune system dysfunction and autoimmune abnormalities [7–14]. Such immunotoxic effects could potentially lead to infection, malaria susceptibility [15], or immunologically-mediated disease [16]. To date, most studies of mercury and the immune system have been done on animal cells and extrapolated to the human immune system [1]. For example, many researchers have demonstrated that mercury inhibits rodent lymphocyte proliferation and function in vitro [14,17–20]. Moreover, mercury can lower basal oxygen consumption by alveolar macrophages [1,21]. Nevertheless, despite the abundance of data linking mercury to harmful consequences to the immune system, the mechanism by which the metal affects the immune system is poorly understood. Although the mechanism linking mercury to immune system dysfunction is not fully understood, several suggestions, which are not mutually exclusive, have been proposed to explain the toxic effects of mercury on a cellular level. They include: altered membrane potassium conductance, modulation of chlorine channels, destruction of sodium-potassium-ATPase activity, inhibition of phospholipid turnover, DNA strand breakage and activation of phospholipase C [1,22–30]. Recently, there has been much interest in apoptosis, or programmed cell death, in lymphocytes and monocytes as the pathway leading to immune system dysfunction [1,31]. To fully understand the effects of mercury in the form of HgCl2 on monocytes, we studied mercuryinduced DNA damage in U-937 human monocyte-like cells using the comet assay in conjunction with apoptosis assessments. We demonstrate that in the presence of low concentrations of mercury, U-937 cells can undergo DNA damage without triggering apoptosis.

amphotericin B (Gibco, Grand Island, NY). Cells were incubated at 37◦ C with 5% CO2 . 2.2. Mercury treatment Cells were treated with various concentrations of mercuric chloride (HgCl2 ) (ACS reagent, Aldrich Chem. Co., Milwaukee, WI) in 96 well cell culture clusters (Costar, Cambridge, MA) for 24, 48, and 72 h. 2.3. Cell viability Cell viability was tested using trypan-blue exclusion. Cells were treated with 0.025% trypan-blue (Life Technologies, Grand Island, NY). The percentage of stained cells in each sample was determined using a hemocytometer (American Optical, Buffalo, NY) and brightfield microscopy (Leitz, Germany). 2.4. Comet assay Was performed as previously described [32]. Briefly, glass microscope slides were first covered with a 200 ␮l layer of normal melting temperature 0.5% agarose gel (Sigma Chem. Co., St. Louis, MO). Control or mercury-treated cells were resuspended in 100 ␮l low melting temperature 0.5% agarose gel (Sigma) then placed on the first layer. After agarose solidification, the slide was treated with alkali lysing solution, and electrophoresis of the sample accomplished at 125 V, 300 mA using a Grass SD9 power supply unit (Quincy, MA) for 20 min. The slides were then washed with Hanks Balanced Salt Solution (Life Technologies, Grand Island, NY), and the sample stained with ethidium bromide (20 ␮g/ml) (Sigma). 2.5. Microscopy

2. Materials and methods 2.1. Cell preparation U-937 cells were obtained from the ATCC (Bethesda, MD). Cells were grown in RPMI 1640 media (Gibco, Grand Island, NY) with 10% FCS (Hyclone, Logan, UT) and 100 U/ml penicillin G sodium, 100 ␮g/ml streptomycin sulfate, and 0.25 ␮g/ml

Cells were observed using an axiovert inverted fluorescence microscope (Carl Zeiss, New York, NY) with a mercury lamp interfaced to a Scion (Frederick, MD) image processing system. Narrow bandpass discriminating filter sets (Omega Optical, Brattleboro, VT) were used with excitation at 540/20 nm emission at 590/30 nm for ethidium bromide. A long pass dichroic mirror at 560 nm was used for ethidium bromide. Fluorescence images were collected with

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

an ICCD camera (Model XC-77; Hamamatsu, Japan) and stored as digital TIFF files.

21

3. Results 3.1. Mercury toxicity

2.6. Annexin binding and flow cytometry Cells were treated for 4 h with 5 ␮M of mercuric chloride, or as a positive control, anti-fas antibody (250 ng/ml CH-11, Upstate Biotechnology, Lake Placid, NY). Incubations were done at 37◦ C in a 5% CO2 atmosphere in RPMI 1640 media (Gibco, Grand Island, NY) with 10% FCS (Hyclone, Logan, UT) and 100 U/ml penicillin G sodium, 100 ␮g/ml streptomycin sulfate, and 0.25 ␮g/ml amphotericin B (Gibco, Grand Island, NY). After incubation, cells were collected, washed twice in PBS, and resuspended in PBS. Next, cells were stained in the dark at room temperature with annexin-V-FITC (PharMingen’s, San Diego, CA) and propidium iodide (5 ␮g/ml) for 15 min, according to manufacture’s protocol. Green fluorescence (530/30 nm) and red fluorescence (585/42 nm) were indicative of annexin-V-FITC binding and propidium iodide, respectively. Data were collected using logarithmic amplification of the fluorescence intensities with electronic compensation for spectral overlap using a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA). 2.7. Comet analysis Comet tail lengths were measured by displaying the stored images on a video monitor that was calibrated for distance. Comet tail lengths did not include the size of the nucleus. Comparison between groups were performed with Dunnett’s t-test. A p-value<0.01 was considered statistically significant. Each mercury treatment was replicated nine times with at least 150 cells being examined for each concentration. The positive controls were the cells exposed to anti-fas antibody; the negative controls were those cells allowed to proliferate for 24, 48, and 72 h in the absence of mercuric chloride or anti-fas antibody.

To study the effects of mercury on U-937 cells, we first examined the effects of mercury on U-937 cell viability and proliferation. Cells were initially cultured at 2×105 cells/ml. To examine viability, U-937 cells were exposed to various concentrations of mercury for 24, 48, and 72 h incubation periods, then tested for viability using trypan-blue exclusion. Control samples showed viabilities of ∼89% for the 24, 48, and 72 h incubation periods. As shown in Fig. 1A, at all time points, concentrations of mercury 0.5 ␮M and below had essentially no effect on cell viability. At 1 ␮M mercury, there was a slight trend to decreased viability as exposure time increased, but it was not statistically significant. However, at all time points there was a statistically significant decrease in cell viability (p≤0.01) between controls and cells exposed to mercury, at mercury concentrations above 1 ␮M. Viability went from ∼90% when the cells were exposed to no mercury, to less than 5% when exposed to 50 and 100 ␮M mercuric chloride, with the LD50 at about 5 ␮M mercury. Fig. 1B shows that the number of cells recovered from control populations, as well as from populations of cells exposed to low concentrations of mercury (i.e. ≤5 ␮M), increased with incubation time. For example, at 0.1 ␮M HgCl2 3.70×105 viable cells/ml were observed after 24 h, 9.45×105 viable cells/ml after 48 h, and 1.41×105 viable cells/ml after 72 h. In contrast, mercury concentrations of 10 ␮M and above led to a dramatic decrease in cell number. For instance, cells exposed to 10 ␮M mercury showed approximately 3.45×104 viable cells/ml after 24 h, 1.00×104 viable cells/ml after 48 h, and 5.00×103 viable cells/ml after 72 h. After 24 h, cells exposed to 50 and 100 ␮M concentrations were all dead. In general, HgCl2 concentrations ≤1 ␮M could not be distinguished from controls, whereas ≥10 ␮M HgCl2 concentration led to a substantial reduction in the number of viable cells relative to controls for cells exposed for 24, 48, or 72 h.

2.8. Statistical analysis 3.2. Comet assay Statistical comparisons were performed using one- or two-way analysis of variance (ANOVA) and Dunnett’s t-test.

We used the comet assay to detect DNA damage in individual U-937 cells. Cells were exposed to 0,

22

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

Fig. 1. U-937 viability after mercury exposure. The U-937 cells were treated with concentrations of HgCl2 from 0.1 to 100 ␮M for 24, 48, and 72 h at 37◦ C. (A) Viability was determined via trypan-blue exclusion at 24 h (solid), 48 h (stripes), and 72 h (cross-hatched). (B) Viable cell density as a function of time for 0 ␮M (䉬), 0.1 ␮M (䊏), 0.5 ␮M (䉱), 1.0 ␮M (×), 5 ␮M (∗), 10 ␮M (䊉), 50 ␮M (|) and 100 ␮M (–).

2.5, 5, 10, 25, 50, or 100 ␮M of mercuric chloride for 24, 48, and 72 h. DNA damage increased as the exposure time increased. Fig. 2B and C show comet assays of cells treated with 5 ␮M of mercury for incubation periods of 24 and 72 h, respectively. DNA damage is apparent by the formation of a comet tail, presumably due to fragmented and/or unwound DNA [33]. Cells incubated in the absence of mercury served as a negative control; no DNA damage was observed (Fig. 2A). In contrast, cells treated with anti-fas for 24 h, the positive control for apoptosis, showed distinct tails that were longer than the mercury-induced comets (Fig. 2D). Thus, cells treated with mercury or anti-fas underwent DNA damage as judged by the comet assay. 3.3. Quantitative results with the comet assay Cells were exposed to various concentrations of mercury for 24, 48, and 72 h. Cells forming comets was then assessed as described above. In an individual experimental setup, for each concentration of mercury at least 150 cells were analyzed at each time point.

Fig. 2. Qualitative micrographs of the comet tail lengths after mercury exposure. (A) Control; (B) 24 h incubation with 5 ␮M of mercury; (C) 72 h incubation with 5 ␮M of mercury and (D) anti-fas treatment after 24 h. Pictures are representatives of the various comet tail lengths observed at different incubation times. Magnification: 870×.

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

Fig. 3. Non-cytotoxic concentrations of mercury induce DNA damage. Cells were treated with various concentrations of mercury for 24, 48, and 72 h incubation periods and analyzed for percent comet formation. Incubation times were 24 h (䊏), 48 h (䊉), and 72 h (䉱). The two error bars (S.E.M.) shown are greater than or equal to the S.E.M. of all data points in the figure.

The basic setup was replicated nine times, so that for each time point and concentration, a total of at least 1350 cells were analyzed. Controls always exhibited comet tails ≤1 ␮m in length (not including cell nucleus). For mercury-treated cells, comet tails >1 ␮m were considered positive for mercury-induced comet formation. As shown in Fig. 3, from 0 to 5 ␮M of mercury, the percentage of cells forming comets increased as the concentration of mercury increased. At 5 ␮M HgCl2 , a significant increase (p<0.01) in the fraction of cells forming comet tails was observed. Interestingly, higher concentrations of mercury (>5 ␮M) produced less comet formation. We next quantitated the comet tail length for the cells treated with 5 ␮M mercury. Not unexpectedly as evident in Fig. 4, comet tail length increased in a statistically significantly manner (p<0.01) with duration of mercury exposure. For those cells forming comets, the mean tail length at 24 h was 5.5±0.06 ␮m; at 48 h, 7.2±0.06 ␮m; and at 72 h, 8.9±0.04 ␮m. For comparative purposes, a separate experiment was conducted where cells were treated with anti-fas for 24 h to induce apoptosis [34]. Apoptotic cells had comet tail lengths that were 42±2.4 ␮m (n=150, data not shown). Quantitative histograms of the comet tail lengths are shown in Fig. 5. As shown in Fig. 5A, when cells were exposed to HgCl2 for 24 h, comet tail lengths for ∼64% of the cells centered graphically about a narrow peak of 3 ␮m length. The remaining cells were in a

23

Fig. 4. Mercury-induced DNA damage increases as a function of time. Cells were treated with 5 ␮M of mercury for 24, 48, and 72 h and analyzed for the extent to which DNA damage occurred by measuring average comet tail lengths. Results are given as a mean S.E.M.

more broadly distributed peak averaging ∼14 ␮m. As seen in Fig. 5B, after exposure for 48 h, the amount of cells contained in the sharp peak was ∼60%. After 72 h, the narrow peak broadened slightly and contained 47% of the comets, while the remaining comet tail lengths were broadly distributed averaging ∼12 ␮m. In general, although the longer the cells were exposed to mercury, the longer the comet tails, tail length never exceeded 19 ␮m. Increased comet tail length is associated with greater DNA fragmentation [32,33]. Therefore, these data suggest that upon increased exposure time to mercury, the cellular DNA became more heavily damaged, but also the fragments were larger than those found in apoptotic cells. Thus, it is unlikely that Hg2+ at these concentrations induces apoptosis in U-937 cells. 3.4. Apoptosis detection via annexin binding To confirm that mercury was not inducing apoptosis, annexin-V binding as an independent assay for apoptosis was employed. Accordingly, cells were treated with anti-fas (as a positive control) or with various mercury concentrations for 4 h. Next, cells were stained with annexin-V to detect apoptotic cells. Externalized phosphatidylserine is an early event in apoptosis, and annexin-V detects apoptotic cells by binding to phosphatidylserine expressed on the cell surface. Thus, annexin-V binding was assessed at an earlier point (4 h) than DNA damage (24 h). As apparent in Fig. 6, various mercury concentrations below 10 ␮M demon-

24

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

Fig. 6. Detection of apoptosis via annexin binding. Cells were treated with anti-fas as well as various mercury concentrations for 4 h (solid). Then, they were labeled with annexin-V to detect apoptosis by binding to phosphatidylserine. Data are representative of three repetitions. The mercury concentrations below 10 ␮M show no apoptosis. Cell viability was confirmed by propidium iodide staining (cross-hatched).

Cells treated with anti-fas serving as the positive control, also displayed substantial levels of annexin-V binding.

4. Discussion

Fig. 5. Histograms of DNA migration lengths for U-937 cells exposed to HgCl2 . Cells were treated with 5 ␮M of mercury for 24, 48, and 72 h. Next, comet tail lengths were measured under fluorescence microscopy. (A) The tail lengths after 24 h showed a sharp peak at approximately 3 ␮m and a broader peak between 9 and 15 ␮m. (B) After 48 h, a sharp peak was displayed between 3 and 4 ␮m; a broader peak was present between 8 and 17 ␮m. (C) After 72 h, two broad peaks were seen between 3 and 7 ␮m as well as between 8 and 19 ␮m.

strate no evidence for apoptosis. In particular, there was no evidence of apoptosis at 5 ␮M mercury, the peak dose associated with DNA damage as judged by the comet assay. However, substantial annexin-V binding indicative of apoptosis was observed at 10 ␮M.

Recently, it has been proposed that mercury may compromise immune function by initiating apoptosis in T cells and monocytes [1,31]. In this study, we have shown that monocytic cells exposed to low concentrations of ionic mercury undergo cellular DNA damage, but by a mechanism independent of apoptosis. As a model system, we investigated the effects of ionic mercury on U-937 cells. The U-937 cell line, originally established from a diffuse histiocytic lymphoma of human origin, is well characterized and expresses many of the monocyte-like characteristics exhibited by cells of histiocytic origin [35]. We found that while the LD50 for ionic mercury was approximately 5 ␮M, below 1 ␮M cell viability and proliferation was essentially unaffected by mercury. However, even at 5 ␮M at least some cells proliferated, as viable cell numbers continued to increase (Fig. 1). Generally speaking, while concentrations of mercury 10 ␮M and above was toxic after 24 h, concentrations of mercury ≈5 ␮M were not cytotoxic or cytostatic. However, this is not to say that mercury

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

does not also have a deleterious effect at these concentrations. It appears that the predominant mechanism(s) whereby mercury affects U-937 cells may be dose-dependent. As has been suggested, at high and toxic doses mercury may destroy mitochondria, leading to rapid cell death [36]. At lower doses, the mechanism may be quite different. There is evidence that low-dose exposure (0.5 ␮M) to mercury(II) may lead to DNA damage in fibroblasts and Chinese hamster ovary cells [2,37]. In order to assess mercury-dependent cellular DNA damage in mononuclear cells, we employed the comet (SCGE) assay. The comet assay is a sensitive single cell assay, which detects many DNA changes, including the formation of strand breaks and unwinding [32,33]. DNA damage is visualized when a ‘comet tail’ (electrophoresed DNA) is observed under fluorescence microscopy. Comet tails are not completely uniform, but rather can have a granular and fragmented structure [32,33]. Although there are many potential origins of comet tail non-uniformity, at least one important factor is diversity in the size of DNA fragments. All other things being equal, DNA fragments of smaller size would be expected to migrate farther than larger fragments during electrophoresis. As shown by Figs. 4 and 5, the more time a cell spent exposed to 5 ␮M mercury prior to SCGE analysis, the longer the comet tail, suggesting smaller DNA fragments, and hence greater DNA damage. It has been recently shown that in a similar manner 0.5–5 ␮M Hg2+ leads to single strand breaks in a human fetal hepatic cell line [38]. DNA could be fragmented as a result of mercurydependent induction of apoptosis as has been suggested [1,31], or perhaps by a different mechanism. To compare mercury-induced comets with comets known to result from apoptotic DNA damage, control cells were exposed to anti-fas. Fas, membrane receptor that controls apoptosis in many cells, normally binds to fas ligand to trigger apoptosis [38]. However, when fas is cross-linked by anti-fas antibodies, apoptosis is also initiated. As seen in Fig. 2D, U-937 cells exposed to anti-fas antibodies displayed comet tails after SCGE analysis that were 42±2.4 ␮m long. Those cells exposed to mercury showed tail lengths that averaged 7.2 ␮m for the three incubation periods (Fig. 4). Since the anti-fas induced comet tail lengths were approximately four times longer than those of the mercury-exposed cells, it appears that anti-fas treated

25

(apoptotic) cells have DNA fragments that are quantitatively different than those found in mercury-treated cells. This suggests that although in cells treated with low-concentrations of mercury DNA is fragmented, fragmentation is not likely due to apoptosis. One of the earliest changes in apoptotic cells, that take place well before DNA fragmentation, is the externalization of phosphatidylserine on the plasma membrane. As a further control to determine if low ionic mercury concentration induced apoptosis, we utilized the annexin-V binding assay to visualize externalized phosphatidylserine. We found that cells exposed to those low concentrations of ionic mercury ≤5 ␮M, which led to DNA damage, did not externalize phosphatidylserine, thus confirming DNA damage was not a consequence of apoptosis. Other reports have shown that lymphocytes and monocytes exposed to 0.6–5 ␮M methylmercuric chloride (MeHgCl) underwent apoptosis [1,31]. The mechanism of organic mercury damage may differ from that of ionic mercury used in this study. Our results support the view that cellular DNA damage as a result of ionic mercury can occur by a mechanism independent of apoptosis. As Shenker suggests, mercury may indeed target mitochondria and lead to oxidative stress [1]. However, in U-937 cells, as opposed to T cells that were previously studied, mercury induced oxidative stress need not trigger apoptosis, but may still lead to DNA damage. For example, it has been shown that mercury-induced stress may induce astrocytes to release oxygen free radicals without necessarily leading to cell death [39]. Others have shown that peripheral blood lymphocytes exposed to mercuric compounds underwent genotoxic effects due to an elevated level of 8-hydroxydeoxyguanosine brought by the generation of reactive oxygen species [40]. Similarly, we propose that U-937 cells exposed to ionic mercury, undergo DNA damage, perhaps occurring from an increase in reactive oxygen metabolites (ROM), directly attacking DNA [40]. The effect of mercury on DNA damage can be assessed by measuring the fraction of cells undergoing comet formation. At 5 ␮M, the percentage of comets peaks. At lower concentrations, mercury causes less DNA damage. However, at much higher mercury concentrations, the percentage of comet formation is also reduced. This likely can be attributed to the fact that high mercury concentrations kill cells by an

26

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27

independent mechanism before DNA damage can occur [40–42]. Thus, genotoxic effects of mercury are perhaps dose-dependent. At extremely low doses, cells may be unaffected by mercury, while at extremely high doses, cells are likely rapidly killed. We suspect there may be a ‘window’ between the very low and very high mercury concentrations where DNA is damaged, but cells are not killed. In the case of U-937 cells, this appears to be between 1 and 5 ␮M. This would support the view that under the proper circumstances, mercury could be mutagenic to immune cells.

Acknowledgements This work was supported in part by an Environmental Health Sciences Center Grant (ES 06639) through pilot project support and the use of the Imaging and Cytometry Facility Core, as well as NIH Grants AI 27409 and ES 10351. References [1] B.J. Shenker, T.L. Guo, I.M. Shapiro, Low-level methylmercury exposure causes human T-Cells to undergo apoptosis: evidence of mitochondrial dysfunction, Environ. Res. 77 (1997) 149–159. [2] M.E. Ariza, J. Holliday, M.V. Williams, Mutagenic effect of mercury(II) in eukaryotic cells, In Vivo 8 (1994) 559–563. [3] Research Triangle Institute, Toxicological profile for mercury, Contract no. 205-93-0606, prepared for US Department of Health and Human Services, March 1999. [4] T. Mathiesen, D.G. Ellingsen, H. Kjuus, Neurophysical effects associated with exposure to mercury vapor among former chloralkali workers, Scand. J. Work Environ. Health 25 (1999) 342–350. [5] R. Bluhm, R.A. Branch, Clinical problems interpreting mercury levels, Int. Arch. Occup. Environ. Health 68 (1996) 421–424. [6] L.J. Goldwater, Normal mercury in man, in: Mercury: A History of Quicksilver, York Press, Baltimore, MD, 1972, pp. 135–149. [7] D.C.M. Dantas, M.L.S. Queiroz, Immunoglobulin E and autoantibodies in mercury-exposed workers, Immunopharmacol. Immunotoxicol. 19 (1997) 383–392. [8] R.D. Barr, The mercurial nephrotic syndrome, East Afr. Med. J. 67 (1990) 381–386. [9] J. Roger, D. Zillikens, G. Burg, E. Gleichmann, Systematic autoimmunity in a patient with longstanding exposure of mercury, Eur. J. Dermatol. 2 (1992) 168–170. [10] K. Schrallhammer-Benkler, J. Ring, B. Przybila, M. Meurer, M. Landthaler, Acute mercury intoxication with lichenoid

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

drug eruption followed by mercury contact allergy and development of antinuclear antibodies, Acta DermatoVenereol. 72 (1992) 294–296. R.L. Siblerud, E. Kienholz, Evidence that mercury from silver dental fillings may be an etiological factor in multiple sclerosis, Sci. Total Environ. 142 (1994) 191–205. G. Warfvinge, M. Hellman, M. Maroti, U. Ahlstrom, O. Larsson, Hg-provocation of oral mucosa in patients with oral lichenoid lesions, Scand. J. Dent. Res. 102 (1994) 34–40. M.D. Stonard, B.V. Charter, D.P. Duffield, J.J. O’Sullivan, C.M. Lockwood, Circulating immune complexes in individuals occupationally exposed to mercury vapor, in: S.S. Brown, J. Savory (Eds.), Chemical Toxicology and Clinical Chemistry of Metals, Academic Press, New York, 1983, p. 135. M.J. McCabe, A.J. Rosenspire, Mercury represses membrane Ig-mediated signal transduction in the WEHI-231 B cell Lymphoma, Int. J. Immunopharmacol. 20 (1998) 697–707. E.K. Silbergeld, C. Trevani, G.T. Strickland, G. Liggans, S. Woodruff, J. Sacci, A. Azad, K. McKenna, Effects of mercury on resistance to P. yoelli malaria in a mouse model, Toxicol. Sci. 42 (1998) 205–205. M.J. McCabe Jr., D.A. Lawrence, The effects of metals on the development of the immune system, in: L.B. Schook, D.L. Laskin (Eds.), Xenobiotics and Inflammation, Academic Press, New York, 1994, pp. 193–216. M.P. Dieter, M.I. Luster, G.A. Boorman, C.W. Jameson, J.H. Dean, J.W. Cox, Immunological and biochemical responses in mice treated with mercuric chloride, Toxicol. Appl. Pharmacol. 68 (1983) 218–228. S. Nakatsuru, J. Oohashi, H. Nakada, N. Imura, Effect of mercurials on lymphocyte function, Toxicology 157 (1985) 4830–4836. K. Nordlind, Inhibition of lymphoid cell DNA synthesis by metal allergens at various concentrations, Int. Arch. Allergy Appl. Immunol. 70 (1983) 191–192. L. Pelletier, R. Pasquier, J. Rossert, P. Druet, HgCl2 induces nonspecific immunosuppression in Lewis rats, Eur. J. Immunol. 17 (1987) 49–54. V. Castranova, L. Bowman, M. Reasor, P. Miles, Effects of heavy metal ions on selected oxidative metabolic processes in rat alveolar macrophages, Toxicol. Appl. Pharmacol. 53 (1980) 13–23. O. Arakawa, M.K. Nakahiro, T. Narahasi, Mercury modulation of GAGA-activated chloride channels and nonspecific cation channels in rat dorsal root ganglion neurons, Brain Res. 551 (1991) 58–63. J. Contoni, R. Ribaudo, P. Bigazzi, D. Kreutzer, Characterization of DNA lesions produced by HgCl2 in cell culture systems, Chemico-Biol. Interact. 49 (1988) 209–224. N. Imura, K. Miura, M. Inokawa, S. Nakada, Mechanism of methylmercury cytotoxicity: by biochemical and morphological experiments using cultured cells, Toxicology 17 (1980) 241–254. K. Jung, H. Endou, Mercury chloride as a possible phospholipase C activator: effect on angiotensin II-induced Ca2+ transient in the rat early proximal tube, Biochem. Biophys. Res. Commun. 173 (1990) 606–613.

E.Y. Ben-Ozer et al. / Mutation Research 470 (2000) 19–27 [26] A. Jungwirth, M. Ritter, M. Paulmichl, F. Lang, Activation of cell membrane potassium conductance by mercury in cell cultured renal epitheloid (MDCK) cells, J. Cell. Physiol. 146 (1991) 25–33. [27] H. Kramer, H. Gonick, E. Lu, In vitro inhibition of Na-KATPase by trace metals: relation to renal and cardiovascular damage, Nephron 44 (1986) 329–336. [28] W. Shier, D. DuBourdiew, Stimulation of phospholipid hydrolysis and cell death by mercuric chloride: evidence for mercuric ion acting as a calcium-mimetic agent, Biophys. Res. Commun. 110 (1983) 758–765. [29] B.M. Anner, M. Moosmayer, Mercury inhibits Na-K-ATPase primarily at the cytoplasmic side, Am. J. Physiol. 262 (1992) 843–848. [30] J.D. Gallagher, R.J. Noelle, F.V. McCann, Mercury suppression of a potassium current in human B lymphocytes, Cell. Signal 7 (1995) 31–38. [31] O. InSug, S. Datar, C. Koch, I. Shapiro, B. Shenker, Mercuric compounds inhibit human monocyte function by inducing apoptosis: evidence for formation of reactive oxygen species, development of mitochondrial membrane permeability and loss of reductive reserve, J. Toxicol. 124 (1997) 211–224. [32] A. Kindzelskii, H.R. Petty, Ultrasensitive detection of hydrogen-peroxide mediated DNA damage after alkaline single cell gel electrophoresis using occulation microscopy and TUNEL labeling, Mutat. Res. 426 (1999) 11–22. [33] V. McKelvey-Martin, E.T.S. Ho, S.R. McKeown, S.R. McCarthy, P.J. Johnston, N.F. Rajab, C.S. Downes, Emerging applications of the single cell gel electrophoresis (comet) assay. I. Management of invasive transitional cell human bladder carcinoma; II. Fluorescent in situ hybridization comets for the identification of damaged and repaired DNA sequences in individual cells, Mutagenesis 13 (1998) 1–8.

27

[34] A. Ashkenazi, V.M. Dixit, Death receptors: signaling and modulation, Science 281 (1998) 1305–1308. [35] C. Sundstrom, K. Nilsson, Establishment and characterization of a human histiocytic lymphoma cell line (U-937), Int. J. Cancer 17 (1976) 565–577. [36] B.J. Shenker, T.L. Guo, I. O, I.M. Shapiro, Induction of apoptosis in human T-cells by methyl mercury: temporal relationship between mitochondrial dysfunction and loss reductive reserve, Toxicol. Appl. Pharmacol. 157 (1999) 23– 35. [37] W. Hamilton-Koch, R.D. Snyder, J.M. Lavelle, Metal-induced DNA damage and repair in human diploid fibroblasts and Chinese hamster ovary cells, Chem. Biol. Interact. 59 (1986) 17–28. [38] L. Bucio, C. Garcia, V. Souza, E. Hernandez, C. Gonzalez, M. Betancourt, M.C. Gutierrez-Ruiz, Uptake, cellular distribution and DNA damage produced by mercuric chloride in human fetal hepatic cell line, Mutat. Res. 423 (1999) 65–72. [39] J.R. Brawer, G.F. McCarthy, M. Gornitsky, D. Frankel, K. Mehindate, H.M. Schipper, Mercuric chloride induces a stress response in cultured astrocytes characterized by mitochondrial uptake of iron, Neurotoxicology 19 (1998) 767–776. [40] H. Ogura, T. Takeuchi, K. Morimoto, A comparison of the 8-hydroxydeoxyguanosine, chromosome aberrations and micronucleus techniques for the assessment of the genotoxicity of mercury compounds in human blood lymphocytes, Mutat. Res. 340 (1996) 175–182. [41] R. Schoeny, Use of genetic toxicology data in US EPA risk assessment: the mercury study report example, Environ. Health Perspect. 104 (1994) 663–673. [42] O. Cantoni, N.T. Christie, S.H. Robison, M. Costa, Characterization of DNA lesions produced by HgCl2 in cell culture systems, Chem. Biol. Interact. 49 (1984) 209–224.