Standard Article

You have free access to this content

Mercury and inorganic mercury compounds [MAK Value Documentation, 2001]

Documentations and Methods

Published Online: 31 JAN 2012

DOI: 10.1002/3527600418.mb743997anoe0015

The MAK Collection for Occupational Health and Safety

The MAK Collection for Occupational Health and Safety

How to Cite

2012. Mercury and inorganic mercury compounds [MAK Value Documentation, 2001]. The MAK Collection for Occupational Health and Safety. 82–122.

Publication History

  1. Published Online: 31 JAN 2012
MAK value (1970)0.1 mg/m3 (as Hg)
Peak limitation (1983)Category III
Absorption through the skin-
Sensitization (1999)Sh
Carcinogenicity (1999)Category 3
Prenatal toxicity-
Germ cell mutagenicity-
BAT value (1998)100 µg mercury per l urine
Synonymsmetallic mercury colloidal mercury
Chemical name (CAS)mercury
CAS number7439–97–6
Structural formulaHg
Molecular weight200.59
Melting point−38.9°C
Boiling point356.6°C
Density at 20°C13.5 g/cm3
Vapour pressure at 20°C0.0016 hPa
log Pow*5.95

The present documentation is based on reviews of the toxicological data for mercury and its inorganic compounds (ATSDR 1997, EPA 1997, WHO 1991). There are practically no data available for the toxicity of univalent mercury compounds (most important representative: calomel Hg2Cl2). However, because of their poor solubility in water, it may be assumed that these compounds are of relatively low toxicity. Moreover, the toxicological profile of univalent mercury compounds is likely to be at least qualitatively like that of the divalent inorganic compounds, which are readily soluble in water, because in the presence of thiol groups, univalent mercury disproportionates to yield elemental and divalent mercury. An equilibrium between the three forms which is subject to the redox potential of the medium is established.

1 Toxic Effects and Mode of Action

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

At the workplace, exposure to mercury vapour is most common. Exposure to dust containing inorganic mercury compounds is very rare.

Mercury vapour is very readily absorbed in the lungs; metallic mercury is absorbed practically not at all through the gastrointestinal tract. About 10% of an ingested dose of inorganic mercury(II) compounds is taken up by the organism. In the blood, dissolved metallic mercury is oxidized to the divalent form by a metabolic pathway which is saturable. As the dose increases, the proportion of metallic mercury in the blood increases; the mercury is distributed in the organism according to its highly lipophilic properties so that it also enters the brain where it accumulates in the oxidized form. Divalent mercury binds to sulfhydryl groups and accumulates especially in the kidneys. Elemental mercury is exhaled; divalent mercury is excreted with urine and faeces.

The main effect of brief exposure to high concentrations of mercury vapour is lung damage, that of ingestion of mercury(II) compounds is damage in the gastrointestinal tract and the kidneys. In experimental animals, the acute toxicity of inhaled mercury and ingested inorganic mercury(II) compounds is high. Inorganic mercury(II) compounds are sensitizing on the skin. In persons who have inhaled mercury vapour, generalized hae-matogenous eczema sometimes develops.

After long term exposure of man to mercury vapour, the target organ is the central nervous system. Characteristic kinds of tremor are followed by psychic and neurological changes (mercurial erethism). In experimental animals which have ingested mercury(II) chloride, the most sensitive target organ is the kidney.

Inorganic mercury(II) compounds are clearly clastogenic in vitro; this effect has also been confirmed in an animal study. To date, mutagenic effects on germ cells have not been demonstrated.

Whether or not mercury is a human carcinogen cannot be decided on the basis of the data available from epidemiological studies. Treatment with mercury(II) chloride produces squamous cell papillomas in the forestomach of the male rat and adenomas and carcinomas of the renal tubules in the male mouse.

2 Mechanism of Action

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

The biological activity of bivalent mercury ions is determined essentially by their high affinity for thiol groups. Binding of the ions to the thiol groups of proteins can result in impairment of a multiplicity of functions. All the biological and toxicological effects of exposure to elemental mercury are put down to the action of mercury ions. The chains of reactions which begin with the binding to thiol groups and result in cell damage or organic disease have been described in numerous hypotheses based on more or less adequate experimental data (reviewed in ATSDR 1997).

3 Toxicokinetics and Metabolism

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

The toxicokinetics and metabolism of metallic mercury and inorganic mercury compounds have been described in detail elsewhere (ATSDR 1997, EPA 1997, IARC 1993, WHO 1991). The results are summarized below.

3.1 Absorption and distribution

3.1.1 Elemental mercury

About 80% of an inhaled dose of mercury vapour is absorbed via the lungs. Being readily soluble in lipids, it rapidly penetrates the alveolar membranes. Via the gastrointestinal tract, metallic mercury is absorbed practically not at all. In the rat less than 0.01 % of the ingested dose was absorbed. During exposure to mercury vapour the amount absorbed through the skin is small in comparison with that absorbed via the lungs. In man, less than 3 % of the total body burden of mercury results from dermal uptake.

In the blood, mercury is largely bound to the erythrocytes where, in the presence of hydrogen peroxide, it may be oxidized by the action of catalase to divalent mercury and bound, for example, to intracellular or extracellular sulfhydryl groups. The fate of the oxidized mercury is like that of an ingested mercury salt (see below). As the enzymatic oxidation of mercury is saturable, the proportion of the mercury dissolved in plasma which is in the metallic form increases with increasing dose. Because metallic mercury is lipophilic and readily diffusible, it is distributed in body fluids, cells and organs and also crosses the blood-brain barrier to enter the brain where it can also be oxidized. As divalent mercury does not readily cross the blood-brain barrier, it accumulates in the brain after exposure to elemental mercury, presumably in complexes with thiol groups or other ligands.

3.1.2 Inorganic mercury compounds

The few available studies indicate that inorganic mercury compounds inhaled in the form of aerosols or dusts are absorbed via the lungs; there are, however, no quantitative data for this absorption. In the gastrointestinal tract, 7–15% of an ingested dose of mercury (II) compounds is absorbed; the amount absorbed correlates with the solubility of the compound in water. After application of mercury(II) chloride to the skin of guinea pigs, 2–3 % of the dose is absorbed within 5 hours.

In the blood, the divalent mercury ion binds to the sulfhydryl groups of plasma constituents and erythrocyte proteins. The ion accumulates in the liver and kidney, predominantly in the latter. After exposure for longer periods, up to 90 % of the total mercury in the organism is found in the kidneys, mostly in the proximal tubules. The mercury ions are probably mostly bound to metallothionein, a heavy-metal-binding protein which can be induced by mercury. Unlike metallic mercury, the divalent mercury ion cannot readily cross the blood-brain barrier. Animal studies have demonstrated that the mercury levels in the brains of animals exposed to elemental mercury are about 10 times those found in animals treated with similar body burdens of divalent mercury.

3.2 Metabolism and excretion

3.2.1 Elemental mercury

In the blood and other tissues, elemental mercury is oxidized almost quantitatively to the divalent form which is excreted (see below). The elimination of elemental mercury dissolved in the blood takes place mainly by exhalation of mercury vapour with an elimination half-time of about 18 hours. During and immediately after brief exposures to mercury vapour, metallic mercury can also be detected in the urine, faeces, saliva and perspiration.

3.2.2 Inorganic mercury compounds

After oral administration of mercury(II) salts to the rat and mouse, small amounts of mercury can be detected in the exhaled air; this suggests that divalent mercury can be reduced to metallic mercury in the organism. Methylation has not been demonstrated in mammalian cells. In the rat, however, divalent mercury has been shown to be methylated by intestinal bacteria.

Divalent mercury is excreted with the urine and faeces. Volunteers given mercury nitrate by intravenous injection excreted between 6.3 % and 35 % of the dose with the urine and between 17.9 % and 38.1 % with the faeces within 70 days. The elimination half-time for the total mercury in the organism was estimated to be about 58 days. For individual compartments, the values deviated markedly from this estimate. Thus the biological half-life of mercury in the brain can be several years.

4 Effects in Man

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

4.1 Single exposures

There are no data available for the effects of inhalation of inorganic mercury compounds by man.

4.1.1 Inhalation of elemental mercury

The main effect of short-term exposure to high concentrations of mercury vapour is lung damage which can eventually be fatal. In persons who survived such an exposure for longer periods, bronchitis and bronchiolitis with interstitial pneumonia and pulmonary emphysema developed and were sometimes accompanied by CNS disorders such as tremor and increased excitability. As the disorder progresses, symptoms like those seen after ingestion of mercury salts can develop: gastro-enteritis and functional renal disorders or even kidney failure. When large amounts of mercury which can no longer be eliminated via the kidneys are excreted via the intestines, colitis mucomembranacea accompanied by severe colic may develop after a latency period of several days (Kark 1979). The literature contains more than 10 reports of cases which ended in death; the high concentrations of mercury in the air to which the persons had been exposed were mostly produced by heating the metal (ATSDR 1997, EPA 1997, WHO 1991).

4.1.2 Ingestion of elemental mercury

Because metallic mercury is hardly absorbed from the gastrointestinal tract, ingestion of single doses of the metal produces only little toxicity. In one case, ingestion of 204 g of the liquid metal did not produce signs of systemic toxicity (Schäfer et al. 1994).

4.1.3 Ingestion of inorganic mercury compounds

Ingestion of mercury(II) chloride causes irritation of the gastrointestinal mucosa and, depending on the dose, nausea, vomiting, abdominal pain and diarrhoea. The estimated lethal dose for an adult has been given as 10–42 mg/kg body weight (expressed as Hg). The causes of death after ingestion of a single large dose of an inorganic mercury compound have been given as cardiac arrest, gastrointestinal damage and acute kidney failure, whereby the kidney is considered to be the critical target organ (ATSDR 1997, EPA 1997, WHO 1991).

4.2 Repeated exposures

4.2.1 Inhalation of elemental mercury

Occupational medical data suggest that after long-term exposure to mercury vapour the most sensitive target is the central nervous system. Therefore a threshold level for occupational exposure to elemental mercury must represent the threshold level for neurotoxic effects in man; only this end point is discussed below. Detailed descriptions of the systemic toxic effects of exposure to high concentrations may be found elsewhere (ATSDR 1997, EPA 1997, WHO 1991). The repeatedly described functional renal disorders play the main role in these reports, even though practically all organ systems can be affected by exposure to high levels of mercury. The effects of long-term exposure to mercury on the immune system which have been seen in animal studies (see Section 5.8) have not yet been shown convincingly to occur in man (e.g. Barregard et al. 1997, Langworth et al. 1992b, Moszczynski 1999, Soleo et al. 1997). In any case, the significance of changes in single immunological parameters for human health cannot at present be assessed.

In most publications about the consequences of repeated exposure to mercury (reviewed by Ratcliffe et al. 1996), the mercury levels to which the workers were exposed were not determined by analysis of the concentrations in the air but were estimated from analyses of the mercury levels in urine or blood. The amounts of mercury excreted with the urine were expressed in terms of urine volume in some publications and in others in terms of the creatinine level. As an approximation, the units “µg mercury per litre urine” and “µg mercury per gram creatinine” may be considered to be equivalent. Therefore the available data is suitable for direct derivation only of a threshold value for the body burden of the substance. From the known relationship between the level of exposure to mercury in the inhaled air and the resulting body burden, a threshold level for elemental mercury in the workplace air can be derived (see Section 6).

Chronic mercury poisoning is almost always caused by inhalation of mercury vapour at the workplace, although there are also descriptions of exposure to a combination of mercury vapour and dust of inorganic mercury salts such as mercuric chloride, found in aerosols in choralkali plants where chlorine and mercury are present (Skerfving and Vostal 1972). The classical signs and symptoms of chronic mercury poisoning caused by inhalation of mercury vapour as described by these authors are listed below:

  • salivation with a strong metallic taste

  • gingivitis and stomatitis with swelling and inflammation of the mucosa

  • kidney damage in the form of nephrosis preceded by albuminuria, hypoproteinaemia and oedema

  • CNS damage

    • tremor bilateral, fine central intention tremor, beginning peripherally in the fingers, eyelids and lips and progressing in severe cases to a generalized tremor with chronic painful spasms in the extremities; regression after the end of exposure is possible

    • mercurial erethism

      This syndrome, which is typical of chronic mercury poisoning, is characterized by changes in behaviour and personality. The persons suffer from increased excitability and irritability, decreased ability to remember, sleep disorders, loss of self-control and self-confidence, headaches and depression.

    Non-specific symptoms which may also be found in persons suffering from chronic mercury poisoning include general weakness, tiredness, anorexia, weight loss and gastrointestinal disorders.

The occupational medical data described in detail in the 1981 MAK documentation indicated that no objective neurotoxic symptoms are to be expected in persons exposed to mercury concentrations in the air up to 0.1 mg/m3, which was considered to produce mercury concentrations in urine up to a maximum of 200 µg/l (Henschler 1981).

The symptoms caused by repeated high level exposure resulting in mercury concentrations in the urine above 200 µg/l have also been confirmed more recently (Albers et al. 1982, 1988, Levine et al. 1982). In addition, in numerous cross-sectional studies apparent neurotoxic effects have been reported in persons with urine mercury levels at or even below 50 µg/l urine or 50 µg/g creatinine; these are discussed below. Studies in which the persons were also exposed to organic mercury compounds (Schiele et al. 1979, Triebig et al. 1981, 1984), those in which the exposure was insufficiently characterized (Williamson et al. 1982) and those in which the exposure of the collectives had taken place many years previously (Ellingsen et al. 1993, Kishi et al. 1993, 1994) are not included here.

From a group of 43 male workers exposed to mercury, on average for 4 years (range 0.5–16 years), either in the electrolysis of alkali metal chlorides (chloralkali workers) or in a factory producing mercury-zinc alloys, single urine samples were analysed and found to contain mercury levels between 9.9 and 286 µg/g creatinine (mean value 95.5, median 71). Two different systems were used in an attempt to detect preclinical signs of tremor. The reader is referred to a detailed discussion of the difficulties encountered in determining this parameter (Beuter and De Geoffroy 1996). The control group comprised 47 workers carefully matched for age and habits who were not occupationally exposed to mercury. In an orthokinesimeter test (aiming test) the persons had to touch marks at the corners of a rectangular plate with a pointer as exactly and as quickly as possible. The parameters recorded for the exposed and control groups were not significantly different although there was a tendency to slightly better results in the control group. In tests with the hole-tremometer, which proved to be the more sensitive method in this study, a peg had to be placed in 8 holes of decreasing diameter without touching the sides. The frequency of touching the sides of the holes was increased significantly in the exposed group only for one hole. Comparison of the total numbers of touches for all holes did not reveal a significant difference although, again, the results for the control group tended to be better. When the persons were compared in groups (control, urine mercury level < 50, 50–99.9 and > 100 µg/g creatinine), no dose-response relationship for the prevalence of test results which lay outside the 95-percentile for the control group could be demonstrated. Although the results for the groups with medium and high urine mercury levels could be shown to be significantly different from the control values, they were much higher in the group with the medium body burden than in that with the higher levels of mercury (Roels et al. 1982). Because of the absence of a dose-response relationship, the results of this study cannot be used for the derivation of a threshold level for neurotoxic effects.

Hand tremor frequency was recorded with an accelerometer while the hand was at rest and while it was supporting a weight of 1.25 kg; the results obtained for a group of persons exposed to mercury (n = 26) were significantly different from those obtained for the control group (n = 25). The mercury concentrations in the urine of the exposed persons were determined once in single evening urine samples; the average concentration was about 20 ± 2 µg/g creatinine. The average exposure duration was 15.3 ± 2.6 years. The authors pointed out that practically no information was available about the concentrations to which the individuals had been exposed in the past, but that these were presumably higher. They were also of the opinion that the clinical relevance of their findings is difficult to assess (Fawer et al. 1983), for which reason these results cannot be used in the establishment of a threshold level.

Roels et al. 1985 studied in detail another collective of male (n = 131) and female (n = 54) workers who were exposed to mercury in the electrolysis of alkali metal chlorides or in factories producing mercury-zinc alloys or electronic components, and compared them with a carefully selected control group. Analysis of the mercury concentration in single urine samples revealed an average value of 52 ± 43 µg/g creatinine (median 37.1, range 7.3–272) in the male workers and of 36 ± 16 µg/g creatinine (median 35, range 7.3–89.4) in the female workers. Some of the subjective symptoms recorded in a questionnaire were significantly more common in the persons exposed to mercury than in the controls but there was no relationship between the urine mercury level or the exposure duration and the symptoms described. Tests for simple reaction time, short-term memory, colour discrimination and the critical flicker fusion frequency revealed no significant differences between the control and the exposed groups. Hand tremor was measured with an accelerometer at rest and while supporting a load of 2 kg, and while aiming for a point with and without load (2 kg). All the results obtained with the exposed women were normal, but in the men preclinical signs of increased hand tremor were found. The severity of the symptoms, however, was not related to the current mercury levels in the urine. The prevalence of altered parameters in these tests for hand tremor was increased only in the group of individuals exposed for more than 10 years. The authors concluded from these results that the neurotoxic symptoms of mercury exposure correlate better with the product of dose and time than with the current body burden (Roels et al. 1985). Thus it is clear that a single determination of systemic mercury levels and a single examination of the exposed persons is hardly suitable for the derivation of a threshold for long-term exposure.

21 fluorescent lamp factory workers with urine mercury concentrations between 16 and 94 µg/g creatinine were examined with a hole-tremometer and a balance-tremometer for preclinical signs of tremor. The authors reported that the test results deteriorated with increasing body burden. The supposed dose-response relationship is, however, not apparent in the published data. Comparison with a control group was not carried out (Verberk et al. 1986). Therefore, and because of the small number of persons examined, this study is not relevant for the present evaluation.

The research group of Roels has published another study of 54 male workers exposed to mercury in the electrolysis of alkali metal chlorides or in factories producing mercury-zinc alloys, in which the methods used in the earlier studies (orthokinesimeter, hole-tremometer and accelerometer) were applied and compared. This group of workers had been exposed on average for 7.7 years (range 1–20 years); the mercury concentrations in the single urine samples analysed were in the range between 11.3 and 224 µg/g creatinine (mean value 75 ± 46). For some of the workers, the mercury levels from previous years were known and so the product of body burden and time could be estimated. The control group comprised 48 carefully matched workers who were not occupationally exposed to mercury. With the accelerometer no significant differences between the two groups could be detected. With both of the other test systems, evidence of preclinical intention tremor and postural tremor was detected significantly more frequently in the exposed workers than in the controls. However, analysis of the results did not reveal a correlation between the prevalence of abnormal test results and years of exposure, current body burden, or the product of body burden and time. The results of the complex statistical analysis suggested that younger workers were more sensitive to mercury. An explanation for this phenomenon or its confirmation is not yet available (Roels et al. 1989). As a whole, and especially because of the absence of a dose-response relationship, these data are not suitable for the derivation of a threshold value for mercury.

An extensive battery of psychological tests revealed no differences between the results obtained for a control group and those for 60 persons who had been exposed as chloralkali workers for periods between 5 and 28 years (mean ± SD: 13.7 ± 5.5 years) to average mercury concentrations of 25 µg/m3 (estimated from the blood mercury levels). The individual symptoms which were reported by the exposed persons significantly more frequently than by the controls (sleep and memory disorders, tiredness, confusion, excitability) could not be associated unambiguously with the mercury exposure; thus no adverse neurological effects could be detected. The mercury levels determined in morning urine samples once before carrying out the psychological tests were between 3 and 52 µg/l, on average 17 ± 11 µg/l (Piikivi and Hänninen 1989). This research group also recorded electroencephalograms for a similarly exposed collective (n = 41) and compared them with those for a control group of persons not exposed to mercury. The differences between the groups were interpreted by the authors as first preclinical effects of mercury (Piikivi and Tolonen 1989). However, such changes are also seen in 15 % of the “normal population” and so cannot be considered to represent adverse effects. Until these findings have been confirmed, they cannot be used for the establishment of a threshold.

The WHO test battery for the detection of preclinical signs of central nervous performance deficits was applied to 28 persons employed in a factory making fluorescent lamps and exposed to relatively low mercury concentrations. The average mercury levels in urine, which had been monitored twice yearly for several years in 24-hour urine samples, varied between 7 and 28 µg/l for 8 persons who were exposed regularly and between 8 and 20 µg/l for 20 occasionally exposed persons. The maximum mercury levels detected in urine were below 140 µg/l. Comparison of the test results with those obtained with the 22 control persons revealed significant impairment only of the acoustic short-term memory performance of the regularly exposed workers. In addition, the members of this group seemed to suffer more from depression and to have experienced marked personality changes. However, it is conceivable that these behavioural changes were caused by earlier exposure to higher levels of mercury at the workplace (Soleo et al. 1990). Thus the suggestion that even very low body burdens of mercury can cause adverse effects is not sufficiently documented. Dose-effect relationships were not investigated.

The tremor frequency in a group of 18 exposed persons was increased above that in a group of 18 control persons. However, the individual results were scattered very widely and there was no clear relationship between the measured result for the parameter of tremor and the mercury level in urine or the duration of exposure. There was no difference between the two groups in the average amplitude of tremor. The tremor parameters were measured on the stretched out forefinger of the dominant hand. The results of detailed physical and neurological examinations of the exposed persons were not different from those for the 18 control persons. The urine mercury levels were determined only once in 24-hour pooled urine samples and were 23 ± 28 µg/l (range 10–121) in the exposed workers (Chapman et al. 1990). Because of the absence of a dose-response relationship and the large inter-individual differences in the tremometer results, these findings cannot be clearly associated with the exposure to mercury.

Abnormal neurological test results (tremor, deviant Romberg test, dysdiadocho-kinesia) were obtained more frequently from 84 employees in a thermometer factory than from the control persons. The urine mercury concentrations were in the range from 1.3 to 345 µg/g creatinine (mean ± SD: 73 ± 70 µg/g creatinine). There was no significant difference between the mean urine mercury levels in employees with abnormal and in those with normal neurological parameters. However, the workers with abnormal test results also had significantly higher values of the “exposure index”, an estimate of the cumulative exposure (Ehrenberg et al. 1991). These results confirm those described above for the study by Roels et al. 1985.

Disorders of well-being (poor concentration, gastrointestinal complaints, nervousness, sleep and memory disorders, tiredness) were reported more frequently by 89 chloralkali workers whose average urine mercury levels were 25 µg/g creatinine (range 0.5–75) than by 75 control persons. On the other hand, no differences between the groups were detected in the psychomotor tests (forearm tremor measured with the accelerometer, hand-eye coordination, finger tapping, simple reaction time, number-symbol test, memory span for numbers and Sternberg test). The symptoms described correlated weakly with the current mercury levels in the urine but not with the duration of exposure (Langworth et al. 1992a). On its own, an increase in the prevalence of self-reported symptoms, especially in the light of the normal results of the psychomotor test battery, is not considered to represent a significantly adverse effect.

Adverse effects on several psychomotor parameters (e.g. reaction time) and on the sense of well-being (self-reported symptoms) compared with the same parameters from a group of 70 control persons were found for a group of 88 workers who were exposed to mercury. The mercury concentration in a 24-hour pooled urine sample was determined once for each worker (mean ± SD: 25 ± 59 µg/l). Dose-response relationships were not mentioned. However, the workers who had been exposed for longer periods did produce poorer results in the neurological tests; an effect of age was excluded. The authors point out that more studies are necessary to confirm their results (Liang et al. 1993).

Longer latency periods for evoked potentials were diagnosed in the 10 chloralkali workers with the highest body burdens of mercury, with average urine mercury levels of 59 ± 28 µg/g creatinine. Data for earlier exposure levels were not available. Data recorded previously from apparently healthy persons were used as the control; potential confounders were taken into account. The findings were interpreted as first signs of neurotoxicity. Their clinical relevance is, however, questionable (Chang et al. 1995). Because of the small size of the group and other inadequacies in the methods, also this study cannot be used for the establishment of a threshold.

More recently, in a collective of persons exposed to mercury in the production of thermometers and precision instruments, disorders of colour vision – detected with the Lanthony colour test – were reported. In comparison with a control group, an increased incidence of subclinical loss of colour vision especially in the blue-yellow range was detected in persons exposed to mercury who excreted more than 50 µg/g creatinine. Only 2 of the 33 persons produced correct results; in the control group it was 10 of 33. Both the test results and the urine mercury levels (range 28–287 µg/g creatinine) were scattered over a wide range. There was no apparent dose-response relationship. A threshold level for adverse effects cannot be deduced from these results (Cavalleri et al. 1995). One year after the successful introduction of improved industrial hygiene to reduce the exposure to mercury, 21 workers were examined again. The urine mercury levels were reduced to 10.0 ± 7.6 µg/g creatinine (mean ± SD, range 1.8–25.7) and the colour vision disorders were no longer more frequent than in the control group. Thus it may be assumed that the changes are reversible (Cavalleri and Gobba 1998).

Nowadays, dental personnel are hardly ever exposed to mercury and the urine mercury levels in this collective are generally markedly below 50 µg/l. Accordingly, in various studies of persons employed in the dental sector (e.g. Langworth et al. 1997, Nilsson et al. 1990), no abnormal neurological parameters were detected. In contrast, other studies report neurological disorders and conclude that they are caused by low level mercury exposure, but various inadequacies in the study methods make this questionable. The neuropsychological tests carried out with 98 dentists and 54 control persons revealed several significant differences between the groups, but the alpha-correction made necessary by the multiple test procedures was not carried out. In addition, a dose-response relationship was established between mercury exposure levels and psychomotor performance deficits. The authors do point out, however, that it may be assumed that the persons were exposed to higher, no longer quantifiable levels of mercury in the past (Ngim et al. 1992). The results obtained for systemic mercury levels are not consistent with those from other studies; therefore it is not possible to make a conclusive assessment of these results.

Psychological tests were carried out with 19 dentists and the results compared with those from a control group of 20 persons. The average urine mercury concentration was determined from single urine samples and found to be 36 µg/l for the dentists; no mercury was detected in the control samples. There was a statistically significant association between the urine mercury concentration and various disorders of well-being such as emotional lability, tiredness and tension. There were no significant deficits in cognitive or motor functions. The authors described their study as hypothesis-generating and suggested testing the hypothesis with a larger collective (Echeverria et al. 1995). The significance of these results is limited in the same way as that of the results of Langworth et al. 1992a described above. For 47 dentists and dental assistants with urinary mercury concentrations in the range of the background levels in the normal population (< 4 µg/l), various results of neurological tests were reported to be associated with the body burden of mercury (Echeverria et al. 1998). These results cannot be evaluated because they were not compared with results from a control group.

It is clear from the account given above that it is not possible to derive a scientifically documented threshold for the neurotoxic effects of mercury from the available cross-sectional studies. This is so especially because there is no correlation between the mercury concentrations in single urine samples and the symptoms described by persons who have been exposed to mercury for long periods. The results suggest rather that the product of concentration and time correlates more closely with the effects.

Therefore a decisive role in the establishment of the threshold is played by the few longitudinal studies in which internal or external exposure levels were determined repeatedly, and which demonstrated constant exposure levels over time.

This criterion is met by a study of 39 chloralkali workers from two plants in which for several years extensive exposure data were recorded. The workers had been employed on average for 15.3 years and for at least 7 years. The mercury concentration found in the air at static sampling sites in the years 1974 to 1976 was 75 µg/m3, averaged from about 2800 single samples. Of the assayed concentration values, 66 % were between 30 and 90 µg/m3 and 7 peaks between 300 and 500 µg/m3 were reported. In the year 1977, the measurements were carried out with personal air samplers, and the average levels in the two factories were 48 and 35 µg/m3. During the three years, the average mercury concentrations found in urine were 124, 73 and 87 µg/m3 (dithizone method). Of the individual values, 8 % were between 200 and 300 µg/l, another 8 % above 300 µg/l. In 1977 the urine concentrations were determined with an atomic absorption spectrometer and averaged 108 µg/l. Various psychomotor functions such as tremor, coordination of the two hands, colour discrimination and reaction time were studied in the exposed workers and the results compared with those from a control group. No significant differences were found (Schuckmann 1979).

The negative findings of Schuckmann 1979 were confirmed in another report of long-term (maximum 21 years) occupational monitoring of 110 chloralkali workers in a factory in which the mercury concentrations in the air, determined daily by the UV method after static sampling, were between 50 and 100 µg/m3 with occasional peaks up to 130 µg/m3. The workers were divided into three groups according to the time spent daily in the atmosphere containing mercury. The urine mercury concentration in the group with the longest exposure time (n = 51) averaged 133 µg/l (range 36–365). For the medium (n = 50) and low level groups (n = 9) the average levels were 69 (range 12–267) and 31 µg/l (range 6–44). Extensive questioning as to subjective symptoms, and medical examinations in which all kinds of tremor, coordination (finger to nose, finger to finger), reflexes, sensitivity to vibrations and muscle function were included revealed no significant differences between the groups. This result was confirmed with another collective of chloralkali workers (n = 126) with comparable exposure conditions who were repeatedly examined during a period of 3.5 years. Thus, at least with the occupational medical examinations which were in general use until the end of the 1970s, no adverse effects were detectable in workers exposed repeatedly to mercury levels up to 100 µg/m3 or who had mean urine mercury concentrations of over 100 µg/l (Bunn et al. 1986).

Chloralkali workers and appropriate control persons (n = 37–43) were examined 4 times in all during a period of 7 years. The potentially confounding effects of age, sex and verbal intelligence were allowed for statistically. The urine mercury concentrations in the low level exposure group (n = 34–50) remained constant during this period at 25 ± 13µg/l, but in the high level exposure group (n = 14–21) there was more fluctuation and the highest level was 152 ± 197 µg/l. At the start of the study the persons had been exposed on average for 12 years. Records of mercury concentrations in the air were available for the previous 20 years and revealed an average concentration of 85 µg/m3. Over the years there were no constant differences between the groups in the results for reported symptoms and the sense of well-being, or for the memory and concentration tests. Only in the test for finger dexterity were the results for the high level exposure group repeatedly different from those for the control group. The exposed group required longer times to complete the tests for aiming with a stylus, which had to be placed in one hole or in several holes in sequence, than did the control group on 3 and 2 of the 4 test days, respectively. A dose-response relationship could not be demonstrated. The authors concluded from these results that deficits in fine muscle movement can be caused by body burdens of mercury producing urine levels in the range of 150 µg/l (Günther et al. 1996).

16 redevelopment workers who were exposed to mercury at work were subjected 4 times in the course of 2 years to extensive occupational medical examinations including neurophysiological and neuropsychological tests. At blood mercury concentrations of 0.5–22 µg/l and urine mercury concentrations of 1.1–80 µg/l, neurotoxic effects were not detectable. The conduction velocity of peripheral nerves was determined three times during the study period; the age-adjusted values differed slightly from the reference values only in occasional tests. A relationship with the body burden of mercury could not be demonstrated so that the authors concluded that urine mercury concentrations up to 80 µg/l do not cause neurotoxic effects (Dietz et al. 1997).

4.2.2 Inhalation of inorganic mercury compounds

Inhalation of dust containing inorganic mercury compounds is uncommon and is known to date only simultaneously with inhalation of mercury vapour. Therefore there are no data for concentrations of inorganic mercury compounds in the air and associated adverse effects on health.

4.2.3 Ingestion of inorganic mercury compounds

There are no data available for the effects on man of ingestion of elemental mercury.

The kidney is considered to be the target organ of the adverse effects of repeated ingestion of mercury(II) salts. It is not possible to establish a dose-response relationship from the little available data, which have been discussed extensively elsewhere (ATSDR 1997, EPA 1997, WHO 1991).

4.3 Local effects on skin and mucous membranes

Especially in children exposed repeatedly to mercury vapour or various mercury compounds, acrodynia may develop. This syndrome is characterized by the desquamation of large scales of skin and scarlet skin discoloration (“pink disease”). The other reported symptoms include sleeplessness, irritability, perspiration, increased blood pressure and tachycardia. The pathogenesis of this disorder is not yet fully understood. Nor are any data available for a dose-response relationship (ATSDR 1997, EPA 1997, WHO 1991).

4.4 Allergenic effects

4.4.1 Elemental mercury and mercury alloys

The clinical symptoms of an allergic reaction to elemental mercury—which, because of its volatility, is generally absorbed by inhalation—are (generalized) exanthem, more rarely urticarial reactions and very rarely erythema multiforme-like reactions (see e.g. Nakada et al. 1997, Nakayama et al. 1983).

There is little information in the literature as to the prevalence of allergic reactions to elemental mercury or amalgam (see Table 1). Occupational allergies to elemental mercury (see e.g. Schrallhammer-Benkler et al. 1992) are nowadays very rare, being observed occasionally in persons who have been in contact with broken thermometers in nursing or dental occupations (see e.g. Faria and de Freitas 1992) or in persons preparing amalgam manually (see e.g. Goh and Ng 1988, Kanerva et al. 1994).

Table 1. Prevalence of positive results in patch tests with mercury (Hg), mercury(II) chloride (HgCl2) or mercury(II) amide chloride (Hg(NH2)Cl)
Persons testedTest substance, concentration, vehicle (unless petrolatum)Proportion of positive resultsCommentsReferences
372 patients (ages 6–15 years)Hg(NH2)Cl,1 %6.5 % Brasch and Geier 1997
2592 patientsHg, 0.5 %3.7 %in 62 cases, clinical relevance assumed and in 44 of those because of contact with merbrominDe la Cuadra 1993
3389 consecutively tested patientsHg(NH2)Cl, 1 %3.2 %according to authors, assessment of relevance frequently not possibleFrosch 1990
11962 consecutively tested patientsHgCl2, 0.1 % (water)1.3 % Gollhausen et al. 1988
329 patients (ages up to 14 years)Hg(NH2)Cl, 1 %6.1 % Goncalo et al. 1992
441 consecutively tested patientsHg(NH2)Cl, 1 % Hg, 1% HgCl2, 0.1 % (water)0.9 % 1.1 % 0.7 %only + + reactions counted; in only 2 cases clinical relevance of reactions assumedHandley et al. 1993
11544 consecutively tested patientsHg(NH2)Cl, 1 %1.6 % Kränke et al. 1995
539 consecutively tested patientsHgCl2, 0.1 %2.2 % Lindemayr and Becerano 1985
1538 consecutively tested patientsHg(NH2)Cl, 1 % HgCl2, 0.03 % (water)9 % 6 %clinical relevance assumed in 23 cases; on re-testing, reaction to Hg(NH2)Cl in 55/56 patientsNebenführer et al. 1984
141 patients (ages up to 14 years)Hg, 0.5 %6.4 %in 7 cases current or recent clinical relevanceRomaguera and Vilaplana 1998
3206 patientsHg(NH2)Cl, 1 %4.6 % Rudner et al. 1975
593 persons with healthy skinHg(NH2)Cl, 1 %1.2 % Seidenari et al. 1990
660 patientsHg(NH2)Cl, 1 %3 %in 4 cases assumed clinical relevance of the reactionsStorrs et al. 1989
42839 consecutively tested patientsHg(NH2)Cl, 1 %1.2 % Sertoli et al. 1999
507 patients with facial dermatitisHg(NH2)Cl, 1 %6.1 % Sun 1987

In the equally rare cases of (allergic) reactions of the oral mucosa to amalgam fillings, the clinical symptoms included lichen planus or other lichenoid changes, recurring aph-tha or leukoplakia in the vicinity of the fillings (see e.g. Alanko et al. 1996, Ibbotson et al. 1996, Koch and Bahmer 1995, von Mayenburg et al. 1996, Pang and Freeman 1995; for reviews of early reports see e.g. Götz and Fortmann 1959, Holmstrup 1991, Thomson and Russell 1970). It is still unknown whether the observed (lichenoid) changes are a result of sensitization to elemental mercury or a predisposing factor for such sensitization and to what extent other factors may also play a role, for example, rough surfaces of the fillings or previous contact with inorganic or organic mercury compounds leading to an initially asymptomatic sensitization.

4.4.2 Inorganic mercury compounds

Irritative and allergic reactions to mercury compounds, which were once widely used in medicine, used to be very common (Bonnevie 1939). In recent times too, individual cases of sensitization caused by topical mercurials have been described, for example, after medication of psoriasis (Kanerva et al. 1993), pediculosis of the eyebrows (Anonide and Massone 1996) or phthiriasis (Vena et al. 1994). Other, nowadays very rare possibilities for contact include mercury-containing cosmetics, for example, ammoniated mercury-containing skin-lightening creams (Aberer et al. 1990, Kawai et al. 1994, Sun 1987), tattooing pigments such as cinnabar (Juhlin and Öhman 1968, Sulzberger and Tolmach 1959), mercury compounds in chemical laboratories (Kanerva et al. 1994, Thomas et al. 1994), mercury fulminate (Bonnevie 1939), or mercury oxycyanide as a preservative (van Ketel and Roeleveld 1977). Allergic reactions to inorganic mercury compounds are manifested mostly as contact dermatitis of the hands or face. Under certain conditions, for example, after internal uptake of calomel, generalized eczema or erythroderma can also develop. Occasionally the eczematous reaction is preceded by urticaria (3 cases, test with 0.03 % mercury(II) chloride solution in water) (Temesvari and Daroczy 1989).

Although in serial patch testing of patients, reactions to mercury compounds are still observed relatively frequently, their clinical relevance is often questionable or not demonstrable. In some cases, however, the causal relationship between the skin changes and contact with a product containing mercury is clear and then avoidance of this product generally results in regression of the symptoms.

The prevalence of positive test results for inorganic mercury compounds published in the literature varies widely from region to region. It is also noteworthy that when mercury(II) chloride or mercury(II) amide chloride is tested in aqueous solution, chemical reactions with the aluminium chambers (e.g. Finn chambers) frequently occur and result in false positive irritant reactions (Kalveram et al. 1980, Lachapelle and Douka 1985). Both in early studies and in more recent ones with large collectives (see Table 1), positive results with mercury(II) amide chloride or mercury(II) chloride were obtained with a significant proportion of the test persons. It is conspicuous that the prevalences found in tests with children are similar to those found when testing adults.

Reactions to inorganic mercury compounds are often seen in patients who also produce positive results in the patch test with merbromin. In contrast, only a small proportion of persons sensitized primarily to thimerosal appear to show reactions to inorganic mercury compounds (see also MAK documentation for “Merbromin” and “Thimerosal”, this volume). In most cases, such reactions are the result of simultaneous sensitization (Fisher 1976). Almost all patients for whom positive results were obtained in a patch test with 5 % amalgam in petrolatum also produced reactions to 1 % mercury(II) amide chloride in petrolatum. In contrast, of 49 patients for whom positive results were obtained with mercury(II) amide chloride, only 27 produced reactions to 5 % amalgam (von Mayenburg et al. 1996). Still unclear is the reason for the reactions to mercury(II) chloride seen in some studies in more than half of the patients with an allergy to gold. Thus 17/25 and 21/35 patients who produced positive results in patch tests with 0.2 % gold tetrachloride or 0.2 % gold trichloride, respectively, also produced reactions to 0.05 % mercury(II) chloride in petrolatum, whereas only 2/14 and 2/19 produced reactions to 0.5 % elemental mercury in petrolatum. Anamnesis of these patients revealed no evidence of any sources of sensitization to mercury (Nakada et al. 1993, 1997).

Also in experimental studies with volunteers, inorganic mercury compounds were shown to cause sensitization of the skin. After 5 applications of 2 % mercury(II) chloride in petrolatum (occlusively for 48 hours) in a maximization test and provocation with 0.05 % mercury(II) chloride in petrolatum, positive results were obtained in 23/25 test persons (Kligman 1966). In a modified Draize test in which 0.5 g of a formulation containing 2 % mercury(II) chloride in petrolatum was applied 10 times every 2 to 3 days, allergies were induced in 2/24 and 2/18 persons. The provocation with 0.05 % mercury (II) chloride in petrolatum was carried out on the arm or dorsal skin of the test persons (Marzulli and Maibach 1973). The maximization test also yielded positive results with mercury(II) amide chloride. Induction consisted in 5 epicutaneous applications for 48 hours of 25 % mercury(II) amide chloride in petrolatum subsequent to a 24-hour occlusive treatment with 5 % sodium dodecylsulfate. The provocation treatment with 10 % mercury(II) amide chloride in petrolatum after pretreatment for 1 hour with 10 % sodium dodecylsulfate yielded positive results in 13/25 test persons. Similar results were obtained in 2 repeat provocations in which 11/24 and 15/25 test persons produced reactions. Details of control studies are, however, not given (Kligman 1966).

The literature contains no information as to the potential of elemental mercury or inorganic mercury compounds to cause airway sensitization.

4.5 Reproductive and developmental toxicity

The reproductive and developmental toxicity of mercury and its inorganic compounds is to be reviewed later in a separate document.

4.6 Genotoxicity

4.6.1 Elemental mercury

Apart from two studies (Hansteen et al. 1993, Shamy et al. 1995) the publications describing cytogenetic studies of lymphocytes from workers exposed to mercury vapour have been reviewed previously (ATSDR 1997, De Flora et al. 1994, EPA 1997, IARC 1993). Both negative and positive results (acentric fragments, aneuploidy, SCE, micronuclei) were reported. However, the significance of these results is limited because of inadequate methods, because confounders were not taken into account or because there was no dose-response relationship between the urine mercury levels and the observed effects. On this basis, the genotoxic potential of elemental mercury for man cannot be evaluated conclusively.

4.6.2 Inorganic mercury compounds

In the lymphocytes of 18 workers with high body burdens of mercury (890 µg/l urine) who were exposed to mercury(II) chloride and organic mercury compounds, evidence was found that the number of chromosomal aberrations was increased relative to that in a control group. However, the smoking habits of the control group did not correspond with those of the exposed collective and the control group contained only 10 persons and included women (Popescu et al. 1979). In the lymphocytes of 29 workers who were exposed to mercury fulminate [Hg(CNO)2] and had urine mercury concentrations of 123.2 ± 54.1 µg/l, significantly more chromosomal aberrations (gaps, breaks and fragments) and micronuclei were found than in those from a control group with an average urine mercury level of 39.2 ± 11.2 µg/l. However, the effects correlated with neither the urine mercury level nor with the exposure duration (Anwar and Gabal 1991).

4.7 Carcinogenicity

4.7.1 Elemental mercury

The published case-control studies and studies of cancer incidence or mortality in collectives working in the electrolysis of alkali metal chlorides, in dental practice or in the production of nuclear weapons and mostly exposed to relatively low levels of mercury provide no clear evidence of carcinogenic effects of mercury vapour. Cancer incidence or mortality has not yet been adequately studied in collectives exposed in earlier years to much higher levels of mercury (e.g. in mines). Therefore the available data are inadequate for a conclusive evaluation and are presented here only in summary. Detailed reviews have been published (Bofetta et al. 1993, EPA 1997, IARC 1993).

Evidence for increased risks of lung and kidney cancer and of tumours in the central nervous system were found in only some of the studies of the above-mentioned cohorts, but the findings could not always be ascribed unambiguously to the mercury exposure and therefore require confirmation.

Of the 3 available case-control studies, 2 yielded weak evidence of carcinogenic effects of mercury. Of 340 male and 36 female lung cancer patients from Florence (Italy), 6 of the women had once been milliners; thus a significantly increased risk of lung cancer was calculated for persons in this occupation who were exposed to high levels of mercury and arsenic. Among the male patients and in the control group, there were no milliners. In Montreal (Canada) a collective of 4576 patients with tumours at a wide variety of locations were questioned in detail as to their occupations, especially with respect to exposure to any of 293 individual substances. The control group consisted of 740 citizens chosen at random and 2 collectives of patients without cancer. The response rate was over 70 %. The prevalence of exposure to elemental mercury was 0.6 %, for exposure to any form of mercury compound 2 %, and the odds ratios for some tumour locations were sometimes significantly increased. Thus among 449 persons with prostate carcinoma, 5 had been exposed to elemental mercury and 14 to other mercury compounds as well; the calculated OR values (odds ratios) were 6.2 (90 % confidence interval (CI) 1.2–33.2) and 1.7 (90 % CI 1.0–3.0). Of the 857 patients with lung cancer, 4 had been exposed to elemental mercury, which yielded an OR of 4.0 (90 % CI 1.2–13). Exposure to mercury compounds was recorded by 14 of the 484 patients with bladder cancer. The corresponding OR was 1.5 (90 % CI 0.9–2.6). Not all the conceivable confounders were adequately taken into account in the data evaluation.

In a population-based case-control study of 110 patients with glioma and 60 with meningioma and 417 control persons from Adelaide (Australia) no association between the number of amalgam fillings and the disease was detected.

4.7.2 Inorganic mercury compounds

With the exception of the population-based case-control study described above (Montreal) in which exposure to elemental mercury and to mercury compounds were recorded separately, there are no available studies of the potential carcinogenic effects of inorganic mercury compounds in man.

5 Animal Experiments and in vitro Studies

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

5.1 Acute toxicity

5.1.1 Inhalation

Exposure to mercury at a concentration of 29 mg/m3 for 2 hours was lethal for 20 of 32 rats. A concentration of 27 mg/m3 for 20 hours was survived by rabbits. One of two animals exposed for 30 hours died. The cause of death was acute lung failure (ATSDR 1997, EPA 1997).

5.1.2 Ingestion

The LD50 of mercury(II) chloride for the rat is in the range between 26 and 78 mg/kg body weight (ATSDR 1997, EPA 1997).

5.2 Subacute, subchronic and chronic toxicity

Because the data from animal studies play only a subordinate role in the establishment of a workplace threshold concentration for mercury, they are not described in detail here. The reader is referred to other reviews (ATSDR 1997, EPA 1997, Friberg and Nordberg 1972). Only evidence of effect thresholds is presented below.

There are no data available for the local effects of mercury or inorganic mercury compounds on the skin or mucous membranes of experimental animals.

5.2.1 Inhalation

In the few available animal studies of the toxicity caused by repeated exposure to mercury vapour, the target organs proved to be the same as in man, the nervous system and the kidneys. The concentrations tested were mostly very high (at least 0.8 mg/m3) so that other organ systems were also affected and a concentration without effect was not found. According to one study carried out in the 1950s, exposure of rats, rabbits and dogs to a mercury concentration of 0.1 mg/m3 (7 hours daily, 5 days per week for up to 83 weeks) did not result in histological changes in the brain, kidneys, liver or lungs. Kidney function parameters and neurological parameters were not recorded.

5.2.2 Ingestion

In experimental animals which had ingested mercury(II) chloride, the most sensitive target organ was the kidney. Oral administration of the substance on 5 days weekly for 6 months to F344 rats resulted in increases in relative and absolute kidney weights in males from the lowest tested dose of 0.31 mg/kg body weight and in females from 0.63 mg/kg body weight. The histological examination revealed an increase in the severity of nephropathy. The kidney damage increased in severity with the dose. In female mice the no observed effect level (NOEL) was 5 mg/kg body weight, in male mice 2.5 mg/kg (Dieter et al. 1992, NTP 1993). In numerous other studies with various species, the doses administered were much higher and all in the active range so that no NOEL can be derived.

5.3 Allergenic effects

There are no published animal studies of the allergenicity of elemental mercury or mercury alloys.

Inorganic mercury compounds

In various studies with guinea pigs, sensitizing effects of mercury(II) chloride have been demonstrated. Negative results were obtained in a Bühler test with 10 % mercury(II) chloride and in a Draize test with 0.1 % mercury(II) chloride (see Table 2). With 5 % mercury(II) chloride in a Bühler test, it was not possible to carry out repeated epicutaneous induction because of the irritant effects (Bühler 1965).

Table 2. Studies of the sensitizing effects of mercury(II) chloride (HgCl2) and mercury(II) amide chloride (Hg(NH)2Cl) in guinea pigs
Test substance method, strain, sexIntradermal/intramuscular induction: injection volume, concentration (vehicle); dose/animalEpicutaneous induction: treatment, concentration (vehicle)Provocation: treatment, concentration (vehicle)Animals with reactionsCommentsReferences
  1. a

    n.s. not specified

HgCl2, maximization test, Hartley, ♀i.d. pairwise 100 µl FCA, 100 µl substance and 100 µl substance in FCA, 0.1 % (water); 0.4 mgday 7: 10 % sodium dodecylsulfate non-occlusive in petrolatum; day 8: 48 h occlusive, 1 % (petrolatum)day 21: 24 h occlusive, 0.1 % (water)8/25controls not specifiedMagnusson and Kligman 1969
HgCl2, modified Polak test, albino, ♂♀i.m. 100 µl substance and 100 µl FCA, 0.1 % (water) in legs and neck; 1 mg days 16, 23, 30, 37: i.d. 100 µl, 0.1 % (water) in the neck; 0.1 mgdays 15, 22, 29, 36: 10 % sodium dodecylsulfate non- occlusive in petrolatum or 70 % DMSO; days 16, 23, 30, 37: 24 h occlusive, 1 % (water)day 44: 24 h occlusive, 0.1 % (water) reactions recorded after 24, 48, 72 and 96 h3/1510 control animals treated with solvent and adjuvant or vehicleZiegler and Süss 1985
HgCl2, Draize test, Hartley, ♀i.d. a total of 10×100 µl (on alternate days), 0.1 % (physiological saline); 10 mg-2 weeks after the last injection on previously untreated skin: 50 µl i.d., 0.1 % (physiological saline)0/25 Magnusson and Kligman 1969
HgCl2, Polak test, Hartley, inbred strains II and XIII, sex n.s.-days 1, 2, 3 and 14: non- occlusive, each dose 0.05 ml on the ear, 30 % (ethanol)day 28: non-occlusive on the flanks, 10 % (ethanol)7/15 (Hartley)no reaction in not induced controls (no other details)Polak et al. 1973
 8/10 (II)
 0/8 (XIII)
Hg(NH2)Cl, Bühler test, Hartley, ♂♀-6×6 h occlusive, 10 % (cream base); 10 mg applied to 5.65 cm2 0/10negative also with 50 % nickel sulfate and 50 % cobalt chlorideBühler 1965

In addition to the classical sensitization studies with guinea pigs, the literature contains reports of a series of other studies with more or less well-validated alternative models. In a modified mouse ear swelling test, groups of 7 to 15 animals from each of 23 strains of mice were treated for 7 days with 5 mg mercury(II) chloride in 100 µl physiological saline (5 % test solution) applied under a plastic bandage to the shaved flank; 7 days later, after measuring the thickness of the ear, the provocation treatment involved application of 500 µg mercury(II) chloride in 25 µl physiological saline (2 %); the ear thickness was measured again after 24 hours. The animals of 14 of the strains including BALB/c and C57B1/6 produced a marked reaction, 7 strains a weak reaction and 2 strains including C3H/He no reaction relative to that of the untreated (no induction) control animals. After injection of lymph node cells from sensitized C57B1/6 mice into non-sensitized (not induced) animals, provocation with mercury(II) chloride yielded positive results in these animals too. If only the serum from sensitized mice was injected, the reaction after provocation was weak (Ishii et al. 1991).

Groups of 6 female BALB/c mice were treated occlusively on the abdominal skin with 5 %, 10 % or 15 % mercury(II) chloride in petrolatum, which was left undisturbed for 10 days. Four days later provocation was carried out by non-occlusive application to the ear of 20 µl of a 5 % solution of mercury(II) chloride in 97 % ethanol. In 5/6, 5/6 and 6/6 animals, respectively, this treatment increased the ear thickness relative to that of the controls (number not specified). On the other hand, sensitization was not produced by a single subcutaneous induction with 0.5 or 1 mg mercury(II) chloride per animal in 0.5 ml of a mixture containing Freund's complete adjuvant (FCA) and water (Vreeburg et al. 1991). Subcutaneous injection of 3, 10, 30 or 60 µg mercury (as mercury(II) chloride) into the paws of female mice of 22 inbred strains resulted in a significant, mostly dose-dependent increase in the size of the popliteal lymph nodes in the animals of most of the strains. In only one strain (DBA/2) was no clear increase observed, three strains reacted only to the two highest doses or the highest dose, and from two strains no useful result was obtained. The authors accounted for the differences in terms of the different genetic make-up of the animals (Stiller-Winkler et al. 1988). Likewise in the local lymph node test (LLNA) with CBA/Ca mice, a clearly positive result was obtained with 5 % and 10 % mercury(II) chloride in acetone/olive oil, 4:1 (Basketter et al. 1994).

Hartley guinea pigs which had been successfully sensitized to merbromin in a maximization test, unlike the control animals, also produced a reaction when the provocation treatment was carried out by non-occlusive application of 0.01 ml 2 % mercury(II) chloride (Osawa et al. 1994), so that the simultaneous reaction to these two substances which has been frequently seen in man can also be reproduced in experimental animals.

5.4 Reproductive and developmental toxicity

The reproductive and developmental toxicity of mercury and its inorganic compounds is to be reviewed later in a separate document.

5.5 Genotoxicity

There are no data available from genotoxicity studies with elemental mercury.

5.5.1 In vitro studies with inorganic mercury compounds

Mercury(II) chloride was not mutagenic in the Salmonella typhimurium strains TA1535, TA1537, TA98 or TA102 either with or without addition of a metabolic activation system (ATSDR 1997, Codina et al. 1995, De Flora et al. 1994, EPA 1997, IARC 1993, Schoeny 1996). Likewise in the strain TA100 and in an Escherichia coli WP2 strain, the substance produced no mutagenic effects. A positive result was obtained in a commercially available bacterial test system for mutagenicity (“Mutatox Test”) (Codina et al. 1995) but, because this test has not been validated, conclusions cannot be drawn from this result. In a test for gene mutations in L5178Y mouse lymphoma cells, mercury(II) chloride yielded weak positive results only after metabolic activation with S9 mix from the livers of rats treated with Aroclor. The mutation frequency was 2.1 times the control value at 6 µg/l and 3.5 times at 8 µg/l. Survival of the cells under these conditions was 56 % and 24 %, respectively, clearly more than the minimum survival of 10 % required for this test (Oberly et al. 1982). Another test without metabolic activation and with sufficient vitality of the cells also yielded clearly positive results. The highest test concentration was 1 µg/l and the highest positive result was three times the control number of mutated colonies (McGregor et al. 1988, NTP 1993). In neither study were details given of the size distribution of the mutated colonies from which conclusions as to the mechanism of mutation could be drawn. In transgenic hprt/gpt+ V79 cells (G10 and G12) which have been described as a sensitive test system for the demonstration of the mutagenic effects of known clastogens such as bleomycin and X-rays, induction of thioguanine-resistant mutants with mercury(II) chloride was not detected (Klein et al. 1994). In transgenic CHO-AS52 cells which, instead of the normal hprt gene, have a single copy of the bacterial equivalent, the gpt gene, with concentrations of 0.1 µM and more, concentration-dependent mutagenic effects of mercury(II) acetate were detected as an increase in thioguanine-resistant mutants. The number of mutants was between two and four times the control value. The colony-forming ability was reduced at concentrations of 0.4 µM or more (Ariza and Williams 1996, Ariza et al. 1994). It was also demonstrated that mercury(II) acetate increased the concentration of hydrogen peroxide in these cells, probably by two different mechanisms. In the low concentration range (< 1 µM) the intracellular concentration of hydrogen peroxide was increased one hour after treatment of the cells, at higher concentrations however the increase did not take place until 24 hours after treatment. Mercury(II) acetate inhibited various purified enzymes from the antioxidative defence of the cell (catalase, glutathione peroxidase and reductase) but increased the activity of Cu-Zn superoxide dismutase and xanthine oxidase. This imbalance could lead to an increase in hydrogen peroxide formation. When the xanthine oxidase was inhibited by addition of allopurinol, both the hydrogen peroxide formation induced by higher mercury concentrations and the number of thioguanine resistant mutants were reduced. When the mercury(II) acetate concentrations were under 1 µM, allopurinol did not affect the number of mutations or the formation of hydrogen peroxide (Ariza et al. 1998); thus the mechanism of hydrogen peroxide formation at low mercury concentrations is unclear. Analysis of the mutations in the gpt gene revealed that, after incubation with mercury(II) acetate at concentrations up to 0.4 µM, mainly point mutations were formed but at higher concentrations also partial or complete deletions. From these results as well, the authors concluded that there are two different mechanisms of mutagenicity (Ariza and Williams 1999).

Thus inorganic mercury compounds are not mutagenic in bacterial test systems and produce only weak effects in mammalian cells, an effect profile typical of inorganic metal compounds (Beyersmann and Hartwig 1993, Hartwig 1995).

In some tests for effects in prokaryotes which can be associated with mutation, mainly positive results were obtained with mercury(II) chloride. Thus 50 µl of a 50 mM solution of mercury(II) chloride applied on a piece of filter paper in a test for differential killing of DNA repair proficient and repair deficient strains of Bacillus subtilis had more growth-inhibiting effect on the repair deficient strain (Kanematsu et al. 1980). In the SOS chromotest with Escherichia coli UA-4567, positive results were obtained with mercury(II) chloride doses of 0.07 µmol or more. The minimum toxic dose in this test system was 0.007 µmol (no other details) (Codina et al. 1995); the corresponding concentration in the incubation mixture cannot be determined from the data given in this publication. In a lysogenic E. coli strain, no induction of lytic phages was detectable after incubation with 2.5 µM mercury(II) chloride; in the DNA repair deficient RecA mutant, a very weak induction was seen. The concentration used inhibited cell growth (Brandi et al. 1990).

The treatment of various kinds of mammalian cells with mercury(II) chloride resulted in DNA damage which has been demonstrated in several different test systems. An increase in the incidence of chromosomal aberrations (CA) was detected after incubation of CHO cells (a cell line from Chinese hamster ovary) with mercury(II) chloride at a concentration of 1 µM or more (14 % of the cells with CA, control value 6 %). At 10 µM the incidence was increased to 20 %; 100 µM proved to be cytotoxic (no other details). In each case 100 cells arrested in metaphase were evaluated. The most common aberrations were polycentric chromosomes, ring chromosomes, chromatid breaks and sister chromatid exchange (SCE). The incidence of gaps was not increased. The incidence of SCE was increased by incubation of the cells with 10 µM mercury(II) chloride only slightly, to 1.5 times the control value. A metabolic activation system was not used (Howard et al. 1991). Another study with CHO cells confirmed the induction of CA by mercury(II) chloride but only in the absence of a metabolic activation system. When the cells were harvested later than usual, the effect was more pronounced. This was put down to a lengthening of the cell cycle by mercury(II) chloride. In many of the cells there were complex aberrations such as chromosome rearrangement and translocations and the number of aberrations was greater than the number of damaged cells. Without metabolic activation, SCE was not increased; in the presence of S9 mix the incidence of SCE was at least 20 % above the control value (NTP 1993). Incubation of a culture of whole human blood for 72 hours with mercury(II) chloride at concentrations of 0.4 µM or more resulted in a concentration-dependent increase in SCE in the lymphocytes to a maximum of 16.5 events per cell at 50 µM (control value 8.9). The mitosis index decreased with increasing concentration from 3.1 % to 0.5 % (Morimoto et al. 1982). With mercury(II) nitrate in non-toxic or only very slightly toxic concentrations of 1 to 30 µM, no increase in SCE was detected in human lymphocytes. The highest concentration caused a significant increase in the incidence of endoreduplication, possibly caused by disturbance of the spindle apparatus (Lee et al. 1997). Another study with cultured whole human blood incubated with mercury(II) chloride revealed significant increases in CA (not including gaps) and in micronuclei in the lymphocytes at concentrations of 10 and 20 µM, respectively. The incidence of gaps was increased significantly at concentrations as low as 5 µM. The observed chromosomal aberrations included chromatid breaks, asymmetric exchange and dicentric or ring-shaped chromosomes. In addition, the level of 8-hydroxydeoxyguanosine was increased significantly at concentrations of 10 µM or more, evidence of the production of reactive oxygen species. Vitality and mitosis index decreased linearly in the concentration range tested (from 2 to 20 µM) to about 40 % and 20 % of the control values, respectively (Ogura et al. 1996). In general, the appearance of complex aberrations is considered to be evidence of substance-related effects because the aberrations which result from cytotoxicity are mostly restricted to simple breaks (NTP 1993). Clastogenic effects were not seen in a mouse mammary carcinoma cell line (FM3A cells), in Don cells, P388D cells or in the two diploid human cell lines WI38 and MRC5 (ATSDR 1997, De Flora et al. 1994, EPA 1997, IARC 1993, Schoeny 1996).

The causes of the cytogenetic effects of mercury(II) chloride in CHO cells have been investigated extensively by one research group. By centrifugation of isolated DNA in an alkaline sucrose gradient, it was demonstrated that mercury(II) chloride and other metal compounds such as nickel chloride and chromate induce DNA strand breaks. The cells were first incubated for 16 hours with 100 µM mercury(II) chloride. When the incubation time was increased to 24 hours, the effect was much more pronounced; incubation for 3 hours with 10 µM had no effect. The DNA damage was accompanied by reduction in the growth rate of the cells (Robison et al. 1982). Likewise, incubation of CHO cells with 25 to 100 µM mercury(II) chloride produced a concentration-dependent increase in DNA strand breaks (determined with the alkaline elution method). DNA-protein or DNA-DNA crosslinks could not be detected. Mercury(II) chloride concentrations above 25 µM proved to be highly cytotoxic (colony-forming ability less than 20 % of the control value), which the authors ascribed to the severe DNA damage (Cantoni et al. 1982). Cell growth under these conditions was completely inhibited at concentrations of 10 µM or more (Cantoni et al. 1984a). Incubation with 25 to 100 µM mercury(II) chloride also resulted in a concentration-dependent decrease in the intracellular level of glutathione. Increasing the glutathione level in the incubation medium prevented the formation of strand breaks (Cantoni et al. 1982). More evidence for an involvement of reactive oxygen species in the effects of mercury(II) chloride was obtained with the demonstration of increased levels of superoxide radicals in the cell incubation medium (Cantoni et al. 1984b). There was a linear relationship between the induction of DNA strand breaks and the loss of colony-forming ability, a measure of cytotoxicity. The level of DNA damage caused by incubation of CHO cells for one hour with mercury(II) chloride was increased by subsequent incubation for one hour in a mercury-free medium. In contrast, strand breaks induced at similar levels by X-ray irradiation were practically completely repaired during such an incubation period (Cantoni and Costa 1983). The authors concluded from these data that mercury(II) chloride specifically inhibits one of the DNA repair mechanisms, a conclusion which was confirmed in subsequent studies. It was demonstrated that the repair of single strand breaks induced by X-rays is inhibited by mercury(II) chloride even in concentrations which are not cytotoxic and do not cause DNA damage. The DNA repair induced by UV irradiation, on the other hand, was not inhibited (Cantoni and Costa 1983, Christie et al. 1986). The level of DNA repair synthesis in CHO cells and SHE cells (Syrian hamster embryo cells), determined by centrifugation of the DNA to equilibrium in density gradients of caesium salts, was only slightly increased at low mercury(II) concentrations which caused no or only very little toxicity or DNA damage and decreased with increasing mercury concentration (Cantoni et al. 1984a, Robison et al. 1984). In contrast, other metal compounds such as nickel chloride and chromate and X-rays, which also induce DNA strand breaks in this test system, caused a significant increase in DNA repair synthesis (Cantoni and Costa 1983, Robison et al. 1984).

A more detailed characterization of the DNA damage induced by mercury(II) chloride revealed that single strand breaks are the primary event. The primary damage was shown not to involve alkali-labile sites or double strand breaks. The formation of single strand breaks was also confirmed by demonstration of reduced sedimentation rates of DNA fragments in sucrose density gradients (“nucleoid gradient sedimentation”). This method also makes it possible to demonstrate such effects after incubation of a substance with isolated DNA. With mercury(II) chloride, clearly positive results were obtained; the authors concluded that the strand breaks resulted at least partially from direct effects of the mercury on the DNA. With the concentration-dependent increase in strand breaks, the binding of mercury to the DNA also increased. This was determined by incubation of the cells with 203mercury chloride and subsequent assay of the activity contained in the isolated DNA (Cantoni et al. 1984a). DNA binding was detected at concentrations lower than those causing DNA damage (Christie et al. 1984). Other authors were also able to demonstrate concentration-dependent binding of mercury to DNA after incubation of cells of a human KB cell line or of CHO AS52 cells with non-cytotoxic concentrations of mercury(II) acetate (Ariza et al. 1994). More detailed analysis of the binding of the mercury to the DNA demonstrated, however, that, although the bonds were stable, they were not covalent. On breakdown of the DNA into single nucleotides, the metal was released. This suggests that mercury can bind only to polynucleotides. Thymidine has been suggested as the primary binding site (Cantoni et al. 1984b).

In all, the results from this research group show that mercury(II) chloride induces single strand breaks in CHO cells very effectively. Other authors have demonstrated similar effects in V79 cells and in muscle cells from chicken embryos (Burkart and Ogorek 1986) and in a human liver cell line (WRL-68 cells) (Bucio et al. 1999). The mechanism of this effect is likely to involve, on the one hand, intracellular formation of reactive oxygen species and, on the other, also direct interaction of mercury with the DNA. Mercury(II) chloride, even at very low concentrations, also inhibits DNA repair and therefore the DNA lesions must be considered to be particularly critical for the cells. This could also provide an explanation for the high cytotoxicity of mercury(II) chloride.

In another study, the induction of single strand breaks by mercury(II) chloride could not be confirmed. The results of a nick translation assay and of a nucleoid sedimentation assay indicated that the DNA damage demonstrated in CHO cells by means of alkaline elution after incubation with mercury(II) chloride (15 µM, 3 hours or 50 µM, 16 hours) are alkali-labile sites and not DNA strand breaks. Human fibroblasts were much less sensitive to DNA damage than CHO cells but more sensitive with respect to impairment of colony-forming ability (IC50 3 µM compared with 6 µM for CHO cells). An explanation as to why the results differed from those of other authors was not provided (Hamilton-Koch et al. 1986).

Other studies, which, however, do not belong to the repertoire of methods usually used to investigate genotoxic effects, also yield evidence of a genotoxic potential of mercury (II) chloride. With cultured human lymphocytes, evidence was obtained of adverse effects of mercury(II) chloride on the mitotic spindle like those long known in plant cells (reviewed by Ramel 1972). The cells were incubated for 4 hours with 10 µM mercury(II) chloride and in the last hour also with colchicine to arrest the chromosomes in metaphase. The lengths of the chromosomes (here only chromosome 1 was evaluated) served as a measure of spindle function. After treatment with mercury(II) chloride, chromosome 1 was significantly shorter. Incubation for 48 hours with 1 µM mercury(II) chloride had no effects, a result which was explained by the authors in terms of induction of the heavy metal-binding protein metallothionein and consequent detoxification (Andersen et al. 1983). Oocytes from Swiss-Webster mice were incubated for 5 or 16 hours with mercury acetate at a mercury concentration of 1–100 µg/ml, and then the first or second metaphase of meiosis was analysed cytogenetically. At concentrations of 25 µg/ml and more, abnormal distribution of the chromosomes in the second metaphase was seen, from 35 µg/ml also in the first metaphase; the cause was suggested to be adverse effects on the spindle. Division of the oocytes was completely inhibited at concentrations of 50 µg/ml or more (Jagiello and Lin 1973). Survival, DNA replication and cell division of CHO cells were impaired in a concentration-dependent manner by mercury(II) chloride (10- 50 µM). The sensitivity of the cells differed markedly in different phases of the cell cycle. The authors concluded from their results that the effects of mercury could involve either binding to proteins or binding to DNA precursors or binding to the DNA itself (Kasschau et al. 1977). Incubation of human lymphocytes for up to 28 hours with mercury(II) chloride in concentrations up to 100 µM either before the phase of DNA synthesis or from this point until mitosis resulted in chromosomal non-disjunction (Verschaeve et al. 1984). CA were also detected, but only at concentrations which caused segregation disorders. The authors suggested that the effects might result not only from disturbance of spindle function but also from inhibition of RNA polymerase I (Verschaeve et al. 1985).

At concentrations of 50 µM or more, mercury(II) chloride increased the rate of transformation of primary SHE cells by SA7 viruses. In all, the authors tested 83 different metal salts; all those metal salts known to be carcinogenic in animal studies or to have genotoxic activity in vitro also produced positive results in this transformation test (Casto et al. 1979).

5.6 In vivo studies with inorganic mercury compounds

In bone marrow cells isolated 24 hours after administration of single oral doses of 2.2, 4.4 or 8.9 mg of mercury(II) chloride to groups of 5 Swiss albino mice, a dose-dependent increase in CA, mainly chromatid breaks, was detected. Per animal, 50 cells arrested in metaphase were evaluated. The proportion of cells with CA (not including gaps) increased from 0.8 % in the negative control to 4.0 %, 5.2 % and 7.2 % with increasing doses of mercury. In the positive control group treated with cyclophosphamide doses of 25 mg/kg body weight, the result was 43.6 %. The mitosis index sank from 4.3 % in the control to 3.8 %, 2.5 % and 1.9 % in the groups treated with mercury and was 1.8 % in the positive control (Ghosh et al. 1991). In contrast, after single intraperitoneal injections of the substance into Swiss albino mice in doses (as Hg) up to 4.4 mg/kg body weight, no CAs were detectable in the bone marrow isolated 12, 24, 36 or 48 hours post applicationem . For each assay time and dose, 375–500 cells arrested in metaphase were evaluated (125 cells/animal). A positive control was not included (Poma et al. 1981). Single subcutaneous injections of mercury(II) chloride in doses (as Hg) of 6.4 and 12.8 mg/kg body weight induced neither CA nor aneuploidy in the bone marrow cells of Syrian hamsters (Watanabe et al. 1982). As the bone marrow cells were first harvested 5 days after the injection, the significance of these results is very limited.

The available studies have not yielded any good evidence for mutagenic effects of inorganic mercury compounds in germ cells. No clastogenic effects were detectable in the spermatogonia of Swiss albino mice given mercury(II) chloride in single intraperitoneal doses (as Hg) up to 4.4 mg/kg body weight. The spermatogonia were harvested 12, 24, 36 and 48 hours after the injections, and between 116 and 282 cells arrested in metaphase were evaluated per harvest time and dose (Poma et al. 1981). The incidence of CA or of aneuploidy was also not increased in the oocytes of Syrian hamsters given a mercury dose of up to 12.8 mg/kg body weight as a subcutaneous injection of mercury(II) chloride on day one of oestrus. The oocytes were harvested for cytogenetic analysis on day 2 of the current or the subsequent oestrus (metaphase II oocytes). The proportion of degenerate oocytes was significantly increased above the control values (Watanabe et al. 1982). In the mouse, mercury acetate did not cause an increase in CA in the oocytes after either intravenous or subcutaneous injection of doses up to 2 and 10 mg/kg body weight, respectively (ATSDR 1997, De Flora et al. 1994, EPA 1997, IARC 1993, Schoeny 1996).

In female (101×C3H)F1 mice (n = 93) given one dose of mercury of 1.5 mg/kg body weight as an intraperitoneal injection of mercury(II) chloride before mating, the proportion of dead implants (10.3 %) was increased slightly but not significantly above the control value (6.4 %). At the same time, the number of living implants per female animal was significantly reduced. In another study in which groups of 32 to 38 female (101×C3H)F1 mice were given a single dose of mercury(II) chloride of 2 mg/kg body weight by intraperitoneal injection, the reproductive capacity of the treated mice was slightly reduced below that of the controls. This effect was not observed in two hybrid varieties of mice, (SEC×C57Bl)F1. However, the possibility that the effects were a result of the maternal toxicity of the substance could not be excluded. The author was of the opinion that the results do not demonstrate germ cell mutagenicity of mercury(II) chloride (Suter 1975). A slight but dose-dependent and significant increase in dominant lethal mutations has been described in rats treated with mercury(II) chloride. The male rats were given daily oral doses of mercury(II) chloride of 0.025, 0.25 or 2.5 mg/kg body weight for a period of one year. Then they were mated with untreated females. The uteruses of 10 pregnant animals from each group were examined on day 20 of gestation. The number of corpora lutea and implantation sites in the various groups did not differ significantly. The number of live embryos per dam was reduced significantly to 7.6 ± 0.28 (control 9.11 ± 0.54) only in the middle dose group. The number of resorptions per dam increased with increasing dose (0.2 ± 0.22, 0.78 ± 0.28, 1.0 ± 0.24) and was significantly different from the control value (0.11 ± 0.11) in the highest dose group. The total embryo mortality in the 3 treated groups was 3.9 ± 2.9 %, 15.4 ± 2.7 % and 17.2 ± 3.1 % and was thus significantly different from the control value of 3.9 ± 2.9 % in the two higher dose groups (Zasukhina et al. 1983). Because of various inadequacies in the documentation such as the unspecified strain of animal, number of treated males, ratio of males to females for mating, and duration of mating period, these results cannot be evaluated. In addition, the number of 10 pregnant animals per group is insufficient. It is also questionable whether the exceedingly low doses of mercury(II) chloride used can have any biological effects at all. Without further confirmation, therefore, this study cannot be considered to provide evidence of a mutagenic effect of the substance in germ cells.

In Drosophila melanogaster, sex-linked recessive lethal mutations could not be induced either by feeding or by injection of mercury(II) chloride (ATSDR 1997, De Flora et al. 1994, EPA 1997, IARC 1993, Schoeny 1996).

5.7 Carcinogenicity

5.7.1 Elemental mercury

A total of 39 three-month-old rats (including rats of the strains BD-III and BD-IV of both sexes) were given two intraperitoneal injections of 0.05 ml metallic mercury within 14 days and were then observed for the rest of their lives. A control group was not included in the study. The average lifetime of the animals (580 days) was significantly shorter than the historical control value of 780 days. The typical symptoms of mercurial erethism were seen transiently in most of the animals. Autopsy revealed the typical kidney damage caused by mercury poisoning in all animals. Tiny droplets of mercury were visible in the abdominal cavity, in some animals also in the abdominal wall, muscles or connective tissue, in males also in the scrotum. The first tumour, a haemorrhagic growth in the small pelvis, appeared 22 months after the start of the study in a female rat. At this time 12 animals were still alive. Subsequently peritoneal sarcomas were found in 2 females and 2 males. All the tumours were classified histologically as spindle cell sarcomas and contained droplets of mercury. In spite of the severe kidney damage, kidney tumours did not develop. The historical incidence of spontaneous benign and malignant tumours at all locations is 3 % for these strains. In the opinion of the authors, the observed carcinogenic effects are not to be seen as unspecific (e.g. a result of the physical properties of the material) but rather as substance-specific effects (Druckrey et al. 1957). The chosen method does not, however, make it possible to draw conclusions as to the carcinogenic potential of mercury vapour for man.

5.7.2 Inorganic mercury compounds

Groups of 60 B6C3F1 mice per dose and sex were given oral doses of an aqueous mercury (II) chloride solution of 5 and 10 mg/kg body weight and day by gavage on 5 days per week for 2 years. An interim killing of 10 animals per group was carried out after 15 months. At this time the relative kidney weights of the treated female animals were increased. Histopathological examination revealed conspicuously more vacuolation in the renal tubulus epithelial cells of the treated male mice than in the control animals; pathological changes were not seen in the females. In the high dose group, the incidence of inflammatory changes in the olfactory epithelium of the nasal cavity was increased. Survival at the end of the study was slightly reduced only in the female mice of the 10 mg/kg group; body weight gains were unchanged. The neoplastic and non-neoplastic alterations are listed in Table 3. From these results it may be seen that the incidence of tubulus cell adenomas or adenocarcinomas in the male mice of the 10 mg/kg group was 3/49 and was statistically not significantly different from the control value (0/50). However, as this kind of tumour is very rare in the mouse, the result was considered to provide equivocal evidence of carcinogenic activity (NTP 1993).

Table 3. Studies of the carcinogenicity of ingested mercury(II) chloride (NTP 1993)
  • 1

    p = 0.107; trend test: p = 0.032

  • 1

    significantly different from control value (p < 0.01, Fisher exact test)

  • 2

    significantly different from control value (p < 0.001, Fisher exact test)

  • 3

    p = 0.044 (logistic regression)

  • 4

    incidence based on evaluation of an increased number of sections

Species:mouse (B6C3F1), 60 ♂, 60 ♀, interim killing of 10 animals per group after 15 months
Administration route:gavage, aqueous solution of mercury(II) chloride
Dose:5 and 10 mg/kg body weight and day
Duration:104 weeks
Toxicity:increased incidence and severity of age-related nephropathy; incidence of nasal olfactory epithelial metaplasia increased in a dose- dependent manner
 mercury(II) chloride (mg/kg body weight and day)
survivors (after 104 weeks)36/5036/5031/50
hyperplasia and tumours:    
tubule cell hyperplasia (kidney)2/500/502/50
tubule cell adenoma (kidney)0/500/502/49
tubule cell adenocarcinoma (kidney)0/500/501/49
tubule cell adenoma or adenocarcinoma (kidney)0/500/503/491
Species:rat (F344); 60 ♂, 60 ♀; interim killing of 10 animals per group after 15 months
Administration route:gavage, aqueous solution of mercury(II) chloride
Dose:2.5 and 5 mg/kg body weight and day
Duration:104 weeks
Toxicity:: increased incidence and severity of age-related nephropathy (suggested as reason for increased mortality); various secondary effects of impaired kidney function (osteodystrophy, mineralization disorders in various tissues, hyperplasia of the parathyroid)
 5 mg/kg group ♂♀: increased incidence of nasal mucosal inflammation
 mercury(II) chloride (mg/kg body weight and day)
survivors (after 104 weeks)26/5010/505/50
hyperplasia and tumours:
squamous cell hyperplasia (forestomach)3/4916/50135/501
squamous cell papilloma (forestomach)0/503/5012/492
follicle cell hyperplasia (thyroid)2/504/502/50
follicle cell adenoma (thyroid)1/504/500/50
follicle cell carcinoma (thyroid)1/502/506/503
tubule cell hyperplasia (kidney)43/501/5010/50
tubule cell adenoma (kidney)44/502/505/50

F344 rats were given oral doses of mercury(II) chloride of 2.5 and 5 mg/kg body weight and day for a period of 2 years. At the interim killing after 15 months, the relative kidney weights were increased in all treated groups, and the incidence and severity of age-related nephropathy was increased only in the male animals. In the high dose group, slight hyperplasia of the epithelial basal cell layer was observed in the forestomach. At the end of the study, survival of the treated male animals was markedly reduced. Nephrotoxicity was suggested to be the cause. Body weights were at least 11 % lower than the control values in the high dose group in both sexes, and in the low dose group only in the males (Dieter et al. 1992, NTP 1993). The incidences of neoplastic and non-neoplastic alterations are listed in Table 3. The increased incidence of squamous cell papillomas in the forestomach of the male rat was perhaps a result of the mucosal damage known to be caused by mercury(II) chloride and also reflected in the dose-dependent increase in hyperplasia. Malignant neoplasms were not observed. The authors considered the results to provide some evidence of carcinogenic activity. In the female rats, 2 squamous cell papillomas of the forestomach were observed; this incidence was statistically not significantly different from the control value (0/50). However, as this kind of tumour is very rare in the rat and, in addition, the incidence of the corresponding hyperplasia was significantly increased, the tumours were considered to be perhaps substance-related and to provide equivocal evidence of carcinogenic activity. Whether the follicle cell carcinomas seen in the thyroid glands of the male rats in the 5 mg/kg group can be considered to be effects of the treatment with mercury(II) chloride is questionable. The incidence of hyperplasia and adenomas in this organ was not increased. In addition, substances with carcinogenic effects on the thyroid are mostly active in both sexes and in both rodent species. In the kidney, the target organ of chronic mercury(II) chloride exposure, the incidence of hyperplasia but not that of tumours was increased. It is conceivable that the average survival period was not long enough for the development of such tumours (NTP 1993).

The life-long administration of mercury(II) chloride in the drinking water at concentrations of mercury of 5 mg/l (doses of about 0.4 mg/kg body weight and day) to 54 male and 54 female Swiss mice had no significant effect on the body weight development or survival of the animals. In the treated animals, tumours were observed at various locations in 21/41 female animals (51.2 %) and in 21/48 males (43.8 %). In the control group, there were tumours in 14/47 females (29.8 %) and in 11/38 males (28.9 %). The incidences of individual kinds of tumours were given only for lymphatic leukaemia (treated females 13/41, males 5/48; control females 3/47, males 3/38) and for lung tumours (7/41, 9/48; 9/47, 5/38). According to the authors, the incidences in the treated groups were statistically not significantly different from the corresponding control values. Apart from the tumours, the autopsy revealed no other abnormal findings. This was also the case for the histopathological examination which, however, was limited to the heart, lungs, liver, kidney and spleen (Schroeder and Mitchener 1975).

5.8 Other effects

In numerous studies, an effect of mercury or inorganic mercury compounds on the immune system has been described. Depending on the dose, duration, administration route and species, immunodepression or autoimmunity were observed. The phenomenon of autoimmunity is limited to certain genetically susceptible species (e.g. the brown Norway rat) in which, after exposure of the animals to mercury, antibodies are formed against various proteins of the glomerular and tubular basal membranes (laminin, type IV collagen, entactin) or nuclear proteins (fibrillarin). The attachment of IgG antibodies to the basal membrane eventually causes glomerulonephritis which can progress to a nephrotic syndrome (ATSDR 1997, Druet 1995, Moszczynski 1997, Pollard and Hultman 1997, Warfvinge et al. 1995, WHO 1991).

6 Manifesto (MAK value/classification)

  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References

Whether or not elemental mercury is a human carcinogen cannot be decided on the basis of the data available from epidemiological studies. For inorganic mercury compounds, there are no data available for man. The only animal study with elemental mercury is not relevant for the present assessment because of the administration route used. Mercury(II) chloride induced squamous cell papillomas in the forestomach of the male rat, but only when administered in doses above the maximum tolerated dose (MTD). In the male mouse, the number of adenomas and carcinomas in the renal tubules was increased. Because the mechanism of these carcinogenic effects is unknown, their relevance for man is still unclear. Therefore mercury(II) chloride is classified in Carcinogen category 3. Because mercury in both the elemental and univalent forms is oxidized in the organism practically completely to divalent mercury, which appears to be responsible for all biological effects (see Section 3), the classification also applies to elemental mercury and other inorganic mercury compounds.

Inorganic mercury(II) compounds are unambiguously clastogenic in vitro; this effect has also been confirmed in an animal study. The clastogenicity results from various mechanisms (inhibition of DNA repair or of other enzymes or protein systems involved in DNA replication, production of reactive oxygen species, direct non-covalent interaction with the DNA), whereby it is still not clear which qualitative or quantitative role is played by the individual processes. However, they all have in common a non-linear dose-response relationship. The dose of mercury shown to have genotoxic effects in animal studies was 2 mg/kg body weight, at least one hundred times the dose taken up during exposures to concentrations equivalent to the MAK value for mercury of 0.1 mg/m3 (about 15 µg/kg and day, assuming a respiratory volume of 10 m3 per shift, 100 % pulmonary absorption and a body weight of 70 kg). Because of this large difference, the MAK value may be retained provisionally in spite of evidence of genotoxic effects in the high dose range and in spite of the fact that the NOEL for genotoxicity is unknown.

The effects of chronic exposure to mercury vapour are manifested mainly in the central nervous system; the relevant threshold concentration is a matter of controversy. Numerous cross-sectional studies suggest that urine mercury levels as low as 50 µg/l or less are associated with first signs of adverse effects on the central nervous system. However, this suggestion does not stand up to critical scientific analysis. The results of the longitudinal studies are of more significance; they suggest that even many years of exposure to mercury concentrations which result in urinary mercury levels of 100 µg/l or even more do not cause objective adverse effects. Therefore a BAT value of 100 µg/l urine has been established (see Greim and Lehnert 1999). Recent studies of the correlation between mercury concentrations in the air and those excreted in the urine have demonstrated that a urine mercury level of 100 µg/l is attained at a concentration of 100 µg/m3 in the air (reviewed by Alessio et al. 1993). This is confirmed in the longitudinal studies in which mercury concentrations in the air are reported.

Therefore the MAK value for mercury and its inorganic compounds of 0.1 mg/m3 (as Hg) and Category III for the limitation of exposure peaks have been retained.

The potential of metallic mercury to cause germ cell mutations cannot be assessed because of lack of data. The studies available do not yield evidence that inorganic mercury compounds have such effects. Therefore mercury is not classified as a germ cell mutagen.

Because mercury and inorganic mercury compounds do not readily penetrate the skin, they are not designated with an “H”.

In the early literature, eczematous reactions to mercury compounds were listed as common occupational diseases. Although it may be assumed that (skin) contact with mercury and mercury compounds is avoided nowadays as far as possible both at work and in the private sphere, the recent literature often contains case reports of allergic reactions and relatively frequent positive results in patch tests carried out with patients at clinics. Likewise, in many animal studies, inorganic mercury compounds have been shown to have sensitizing effects. Therefore metallic mercury and inorganic mercury compounds are designated with “Sh”. To date there are no reports of airway sensitization caused by these substances.

End Notes
  • *

    n-octanol/water distribution coefficient


  1. Top of page
  2. Toxic Effects and Mode of Action
  3. Mechanism of Action
  4. Toxicokinetics and Metabolism
  5. Effects in Man
  6. Animal Experiments and in vitro Studies
  7. Manifesto (MAK value/classification)
  8. References
  • Aberer W, Gerstner G, Pehamberger H (1990) Ammoniated mercury ointment: outdated but still in use. Contact Dermatitis 23: 168171
  • Alanko K, Kanerva L, Jolanki R, Kannas L, Estlander T (1996) Oral mucosal diseases investigated by patch testing with a dental screening series. Contact Dermatitis 34: 263267
  • Albers JW, Cavender GD, Levine SP, Langolf GD (1982) A symptomatic sensorimotor poly-neuropathy in workers exposed to elemental mercury. Neurology 32: 11681174
  • Albers JW, Kallenbach LR, Fine LJ, Langolf GD, Wolfe RA, Donofrio PD, Alessi AG, Stolp-Smith KA, Bromberg MB (1988) Neurological abnormalities associated with remote occupational elemental mercury exposure. Ann Neurol 24: 651659
  • Alessio L, Crippa M, Lucchini R, Roi R (1993) Inorganic mercury. in: Roi R, Sabbioni E (Eds) CEC criteria documents for occupational exposure limit values. Joint Res Centre (CEC-Ispra)
  • Andersen O, Rønne M, Nordberg GF (1983) Effects of inorganic metal salts on chromosome length in human lymphocytes. Hereditas 98: 6570
  • Anonide A, Massone L (1996) Periorbital contact dermatitis due to yellow mercuric oxide. Contact Dermatitis 35: 61
  • Anwar WA, Gabal MS (1991) Cytogenetic study in workers occupationally exposed to mercury fulminate. Mutagenesis 6: 18992
  • Ariza ME, Williams MV (1996) Mutagenesis of AS52 cells by low concentrations of lead(II) and mercury(II). Environ Mol Mutagen 27: 3033
  • Ariza ME, Williams MV (1999) Lead and mercury mutagenesis: type of mutation dependent upon metal concentration. J Biochem Toxicol 13: 107112
  • Ariza ME, Bijur GN, Williams MV (1998) Lead and mercury mutagenesis: role of H2O2, superoxide dismutase, and xanthine oxidase. Environ Mol Mutagen 31: 352361
  • Ariza ME, Holliday J, Williams MV (1994) Mutagenic effect of mercury (II) in eukaryotic cells. In Vivo 8 : 559564
  • ATSDR (Agency for Toxic Substances and Disease Registry) (1997) Toxicological profile for mercury. Draft for public comment (Update). US Department of Health & Human Services
  • Barregard L, Enestrom S, Ljunghusen O, Wieslander J, Hultman P (1997) A study of autoanti-bodies and circulating immune complexes in mercury-exposed chloralkali workers. Int Arch Occup Environ Health 70: 101106
  • Basketter DA, Scholes EW, Kimber I (1994) The performance of the local lymph node assay with chemicals identified as contact allergens in the human maximization test. Food Chem Toxicol 32: 543547
  • Beuter A, De Geoffroy A (1996) Can tremor be used to measure the effects of chronic mercury exposure in human subjects? Neurotoxicology 17: 213228
  • Beyersmann D, Hartwig A (1993) Genotoxic effects of metal compounds. in: Bolt HM, Hellman B, Dencker L (Eds) Use of mechanistic information in risk assessment, Springer, Berlin, 192198
  • Bofetta P, Merler E, Vainio H (1993) Carcinogenicity of mercury and mercury compounds. Scand J Environ Health 19: 17
  • Bonnevie P (1939) Aetiologie und Pathogenese der Ekzemkrankheiten. J A Barth, Leipzig, 275 and 308–321
  • Brandi G, Schiavano GF, Albano A, Cattabeni F, Cantoni O (1990) Growth delay and filamen-tation of Escherichia coli wild-type and rec A cells in response to hexavalent chromium and other metal compounds. Mutat Res 245: 201204
  • Brasch J, Geier J (1997) Patch test results in schoolchildren. Results from the Information Network of Departments of Dermatology (IVDK) and the German Contact Dermatitis Research Group (DKG). Contact Dermatitis 37: 286293
  • Bucio L, Garcia C, Souza V, Hernandez E, Gonzalez C, Betancourt M, Gutierrez-Ruiz MC (1999) Uptake, cellular distribution and DNA damage produced by mercuric chloride in a human fetal hepatic cell line. Mutat Res 25: 6572
  • Bühler EV (1965) Delayed contact hypersensitivity in the guinea pig. Arch Dermatol 91 : 171175
  • Bunn III WB, McGill CM, Barber TE, Cromer Jr JW, Goldwater LJ (1986) Mercury exposure in chloralkali plants. Am Ind Hyg Assoc J 47: 249254
  • Burkart W, Ogorek B (1986) Genotoxic action of cadmium and mercury in cell cultures and modulation of radiation effects. Toxicol Environ Chem 12: 173183
  • Cantoni O, Christie NT, Robison SH, Costa M (1984a) Characterization of DNA lesions produced by HgCl2 in cell culture systems. Chem Biol Interact 49: 209224
  • Cantoni O, Christie NT, Swann A, Drath DB, Costa M (1984b) Mechanism of HgCl2 cytotoxicity in cultured mammalian cells. Mol Pharmacol 26: 360368
  • Cantoni O, Costa M (1983) Correlations of DNA strand breaks and their repair with cell survival following acute exposure to mercury(II) and X-rays. Mol Pharmacol 24: 8489
  • Cantoni O, Evans RM, Costa M (1982) Similarity in the acute cytotoxic response of mammalian cells to mercury(II) and X-rays: DNA damage and glutathione depletion. Biochem Biophys Res Commun 108: 614619
  • Casto BC, Meyers J, DiPaolo JA (1979) Enhancement of viral transformation for evaluation of the carcinogenic or mutagenic potential of inorganic metal salts. Cancer Res 39: 193198
  • Cavalleri A, Belotti L, Gobba F, Luzzana G, Rosa P, Seghizzi P (1995) Colour vision loss in workers exposed to elemental mercury vapour. Toxicol Lett 77: 351356
  • Cavalleri A, Gobba F (1998) Reversible color vision loss in occupational exposure to metallic mercury. Environ Res 77: 173177
  • Chang YC, Yeh CY, Wang JD (1995) Subclinical neurotoxicity of mercury vapor revealed by a multimodality evoked potential study of chloralkali worker. Am J Ind Med 27: 271279
  • Chapman LJ, Sauter SL, Henning RA, Dodson VN (1990) Differences in frequency of finger tremor in otherwise asymptomatic mercury workers. Br J Ind Med 47: 838843
  • Christie NT, Cantoni O, Evans RM, Meyn RE, Costa M (1984) Use of mammalian DNA repair-deficient mutants to assess the effects of toxic metal compounds on DNA. Biochem Pharmacol 33: 16611670
  • Christie NT, Cantoni O, Sugiyama M, Cattabeni F, Costa M (1986) Differences in the effects of Hg(II) on DNA repair induced in Chinese hamster ovary cells by ultarviolet or X-rays. Mol Pharmacol 29: 173178
  • Codina JC, Pérez-Torrente C, Pérez-García A, Cazorla FM, de Vicente A (1995) Comparison of microbial tests for the detection of heavy metal genotoxicity. Arch Environ Contam Toxicol 29: 260265
  • De La Cuadra J (1993) Sensibilisation cutanée au mercure et à ses composés. Ann Dermatol Venereol 120: 3742
  • De Flora S, Bennicellsi C, Bagnasco M (1994) Genotoxicity of mercury compounds. A review. Mutat Res 317: 5779
  • Dieter MP, Boorman GA, Jameson CW, Eustis SL, Uraih LC (1992) Development of renal toxicity in F344 rats gavaged with mercuric chloride for 2 weeks, or 2, 4, 6, 15, and 24 months. J Toxicol Environ Health 36: 319340
  • Dietz MC, Ihrig A, Lichtnecker H, Sladaczek C, Sladaczek F, Triebig G (1997) Längsschnittuntersuchung zur Quecksilberbelastung und zu neurotoxischen Beanspruchungs-reaktionen bei Sanierungsarbeiten. Arbeitsmed Sozialmed Umweltmed 32: 342348
  • Druckrey H, Hamperl H, Schmähl D (1957) Cancerogene Wirkung von metallischem Quecksilber nach intraperitonealer Gabe bei Ratten. Z Krebsforsch 61: 511519
  • Druet P (1995) Metal-induced autoimmunity. Hum Exp Toxicol 14: 120121
  • Echeverria D, Aposhian HV, Woods JS, Heyer NJ, Aposhian MM, Bittner Jr AC, Mahurin RK, Cianciola M (1998) Neurobehavioral effects from exposure to dental amalgam Hg0: new distinctions between recent exposure and Hg body burden. FASEB J 12: 971980
  • Echeverria D, Heyer NJ, Martin MD, Naleway CA, Woods JS, Bittner A (1995) Behavioral effects of low-exposure to Hg0 among dentists. Neurotoxicol Teratol 17: 161168
  • Ehrenberg RL, Vogt RL, Smith AB, Brondum J (1991) Effects of elemental mercury exposure at a thermometer plant. Am J Ind Med 19: 495507
  • Ellingsen DG, Mørland T, Andersen A, Kjuus H (1993) Relation between exposure related indices and neurological and neurophysiological effects in workers previously exposed to mercury vapour. Br J Ind Med 50: 736744
  • EPA (Environmental Protection Agency) (1997) Mercury study report to congress, December 1997. EPA-452/R-97–003, EPA, Washington, DC, USA
  • Faria A, de Freitas C (1992) Systemic contact dermatitis due to mercury. Contact Dermatitis 27: 110111
  • Fawer RF, De Ribaupierre Y, Guillemin MP (1983) Measurement of hand tremor induced by industrial exposure to metallic mercury. Br J Ind Med 40: 204208
  • Fernandez L, Maquiera E, Garcia-Abujeta JL, Yanez S, Rodriguez F, Martin-Gil D, Jerez J (1995) Baboon syndrome due to mercury sensitivity. Contact Dermatitis 33: 5657
  • Fisher AA (1976) Allergic dermatitis medicamentosa: the “Systemic Contact-Type Variety”. Cutis 18: 637642
  • Friberg L, Nordberg GF (1972) Inorganic mercury—relation between exposure and effects. in: Friberg L, Vostal D (Eds) Mercury in the environment, CRC Press, Cleveland, 113139
  • Frosch PJ (1990) Aktuelle Kontaktallergene. Hautarzt 41, Suppl 10: 129133
  • Ghosh AK, Sen S, Sharma A, Talukder G (1991) Effect of chlorophyllin on mercuric chloride-induced clastogenicity in mice. Food Chem Toxicol 29: 777779
  • Goh CL, Ng SK (1988) Occupational allergic contact dermatitis from metallic mercury. Contact Dermatitis 19: 232233
  • Gollhausen R, Enders F, Przybilla B, Burg G, Ring J (1988) Trends in allergic contact sensitization. Contact Dermatitis 18: 147154
  • Goncalo S, Goncalo M, Azenha A, Barros MA, Sousa Bastos A, Brandao FM, Faria A, Marques MSJ, Pecegueiro M, Rodrigues JB, Salgueiro E, Torres V (1992) Allergic contact dermatitis in children. Contact Dermatitis 26: 112115
  • Götz H, Fortmann I (1959) Bewirken Amalgamfüllungen der Zähne eine Quecksilber-sensibilisierung der Haut? Haut Geschlechtskrh 26: 3436
  • Greim H, Lehnert G (Eds) (1999) Mercury, metallic mercury and inorganic mercury compounds. in: Biological exposure values for occupational toxicants and carcinogens, Volume 3. Wiley-VCH, Weinheim, 123142
  • Günther W, Sietman B, Seeber A (1996) Repeated neurobehavioral investigations in workers exposed to mercury in a chloralkali plant. Neurotoxicology 17: 605614
  • Hamilton-Koch W, Snyder RD, Lavelle JM (1986) Metal-induced DNA and repair in human diploid fibroblasts and Chinese hamster ovary cells. Chem Biol Interact 59: 1728
  • Handley J, Todd D, Burrows D (1993) Mercury allergy in a contact dermatitis clinic in Northern Ireland. Contact Dermatitis 29: 258261
  • Hansteen IL, Ellingsen DG, Clausen KO, Kjuus H (1993) Chromosome aberrations in chloralkali workers previously exposed to mercury vapor. Scand J Work Environ Health 19: 375381
  • Hartwig A (1995) Current aspects in metal genotoxicity. BioMetals 8: 311
  • Henschler D (Ed.) (1981) Quecksilber. in: Toxikologisch-arbeitsmedizinische Begründungen von MAK-Werten, 8th issue, VCH-Verlagsgesellschaft, Weinheim
  • Holmstrup P (1991) Reactions of the oral mucosa related to silver amalgam: a review. J Oral Pathol Med 20: 17
  • Howard W, Leonard B, Moody W, Kochhar TS (1991) Induction of chromosome changes by metal compounds in cultured CHO cells. Toxicol Lett 56: 179186
  • IARC (International Agency for Research on Cancer) (1993) Beryllium, cadmium, mercury and exposures in the glass manufacturing industry. IARC monographs on the evaluation of carcinogenic risks to humans, Vol 58, IARC, Lyon, 239347
  • Ibbotson SH, Speight EL, Macleod RI, Smart ER, Lawrence CM (1996) The relevance and effect of amalgam replacement in subjects with oral lichenoid reactions. Br J Dermatol 134: 420423
  • Ishii N, Ishii H, Nakajima H, Aoki I, Shimoda N (1991) Contact sensitivity to mercuric chloride is associated with I-A region in mice. J Dermatol Sci 2: 268273
  • Jagiello G, Lin JS (1973) An assessment of the effects of mercury on the meiosis of mouse ova. Mutat Res 17: 9399
  • Juhlin L, Öhman S (1968) Allergic reactions to mercury in red tattoos and in mucosa adjacent to amalgam fillings. Acta Derm Venereol 48: 103105
  • Kalveram KJ, Rapp-Frick C, Sorck G (1980) Misleading patch test results with aluminium Finn chambers and mercury salts. Contact Dermatitis 6: 507508
  • Kanematsu N, Hara M, Kada T (1980) Rec assay and mutagenicity studies on metal compounds. Mutat Res 77: 109116
  • Kanerva L, Komulainen M, Estlander T, Jolanki R (1993) Occupational allergic contact dermatitis from mercury. Contact Dermatitis 28: 2628
  • Kanerva L, Tarvainen K, Estlander T, Jolanki R (1994) Occupational allergic contact dermatitis caused by mercury and benzoyl peroxide. Eur J Dermatol 4: 359361
  • Kark P (1979) Clinical and neurochemical aspects of inorganic mercury intoxication. in: Vinken PJ, Bruyn GW (Eds) Handbook of clinical neurology, Elsevier, Amsterdam, 147197
  • Kasschau MR, Elder J, Meyn RE (1977) Effects of mercuric chloride on synchronized Chinese hamster ovary cells: survival and DNA replication. Chem Biol Interact 18: 205215
  • Kawai K, Zhang XM, Nakagawa M, Kawai J, Okada T, Kawai K (1994) Allergic contact dermatitis due to mercury in a wedding ring and a cosmetic. Contact Dermatitis 31: 330331
  • van Ketel WG, Roeleveld CG (1977) A curious case of allergy to mercuric compounds. Contact Dermatitis 3: 106
    Direct Link:
  • Kishi R, Doi R, Fukuchi Y, Satoh H, Satoh T (1993) Subjective symptoms and neurobehavioral performance of ex-mercury miners at an average at 18 years after the cessation of chronic exposure to mercury vapor. Environ Res 62: 289302
  • Kishi R, Doi R, Fukuchi Y, Satoh H, Satoh T, Ono A, Moriwaka F, Tashiro K, Takahata N, Sasatani H (1994) Residual neurobehavioural effects associated with chronic exposure to mercury vapour. Occup Environ Med 51: 3541
  • Klein CB, Kargacin B, Su L, Cosentino S, Snow ET, Costa M (1994) Metal mutagenesis in transgenic Chinese hamster cell lines. Environ Health Perspect 102: 6367
  • Kligman AM (1966) The identification of contact allergens by human assay. J Invest Dermatol 47: 393409
  • Koch P, Bahmer FA (1995) Oral lichenoid lesions, mercury hypersensitivity and combined hypersensitivity to mercury and other metals: histologically-proven reproduction of the reaction by patch testing with metal salts. Contact Dermatitis 33: 323328
  • Kränke B, Binder M, Derhaschnig J, Komericki P, Pirkhammer D, Ziegler V, Aberer W (1995) Epikutantestung mit der “Standardreihe Österreich”—Testepidemiologische Kenngrößen und Ergebnisse. Wien Klin Wochenschr 107: 323330
  • Lachapelle JM, Douka MA (1985) An evaluation of the compatibility between aluminium Finn chambers and various mercurials dissolved in water or dispersed in petrolatum. Dermatosen Beruf Umwelt 33: 1214
  • Langworth S, Almkvist O, Söderman E, Wikstrom BO (1992a) Effects of occupational exposure to mercury vapour on the central nervous system. Br J Ind Med 49: 545555
  • Langworth S, Elinder CG, Sundquist KG, Vesterberg O (1992b) Renal and immunological effects of occupational exposure to inorganic mercury. Br J Ind Med 49: 394401
  • Langworth S, Sällsten G, Barregard L, Cynkier I, Lind ML, Söderman E (1997) Exposure to mercury vapor and impact on health in the dental profession in Sweden. J Dent Res 76: 13971404
  • Lee CH, Lin RH, Liu SH, Lin-Shiau SY (1997) Distinct genotoxicity of phenylmercury acetate in human lymphocytes as compared with other mercury compounds. Mutat Res 392: 269276
  • Levine SP, Cavender GD, Langolf GD, Albers JW (1982) Elemental mercury exposure: peripheral neurotoxicity. Br J Ind Med 39: 136139
  • Levy A, Hanau D, Fousserau J (1980) Contact dermatitis in children. Contact Dermatitis 6: 260262
  • Liang YX, Sun RK, Sun Y, Chen ZQ, Li LH (1993) Psychological effects of low exposure to mercury vapor: Application of a computer-administered neurobehavioral evaluation system. Environ Res 60: 320327
  • Lindemayr H, Becerano S (1985) Interaction of mercury compounds and aluminium. Contact Dermatitis 13: 274
  • Magnusson B, Kligman AM (1969) The identification of contact allergens by animal assay. The guinea pig maximization test. J Invest Dermatol 52: 268276
  • Marzulli FN, Maibach HI (1973) Antimicrobials: experimental contact sensitization in man. J Soc Cosmet Chem 24: 399421
  • von Mayenburg J, Frosch PJ, Fuchs T, Aberer W, Bäurle G, Brehler R, Busch R, Gaber G, Hensel O, Koch P, Peters KP, Rakoski J, Rueff F, Szliska C (1996) Mercury and amalgam sensitivity. Dermatosen Beruf Umwelt 44: 213221
  • McGregor DB, Brown A, Cattanach P, Edwards I, McBride D, Riach C, Caspary WJ (1988) Response of the L5178Y tk+/tk- mouse lymphoma cell forward mutation assay: III 72 coded chemicals. Environ Mol Mutagen 12: 85154
  • Morimoto K, Iijima S, Koizumi A (1982) Selenite prevents the induction of sister-chromatid exchanges by methyl mercury and mercuric chloride in human whole-blood cultures. Mutat Res 102: 183192
  • Moszczynski P (1997) Mercury compounds and the immune system: a review. Int J Occup Med Environ Health 10: 247258
  • Moszczynski P (1999) Immunological disorders in men exposed to metallic mercury vapour. A review. Centr Eur J Publ Health 7: 1014
  • Nakada T, Higo N, Iijima M, Nakayama H, Maibach HI (1997) Patch test materials for mercury allergic contact dermatitis. Contact Dermatitis 36: 237239
  • Nakada T, Iijima M, Fujisawa R, Nakayama H (1993) Cross hypersensitivity to ionized mercury and not to non-ionized mercury with gold dermatitis patients caused by ear piercing. J Invest Dermatol 101: 473
  • Nakayama H, Niki F, Shono M, Hada S (1983) Mercury exanthem. Contact Dermatitis 9: 411417
  • Nebenführer L, Korossy S, Vincze E, Gózony M (1984) Mercury allergy in Budapest. Contact Dermatitis 10: 121122
  • Ngim CH, Foo SC, Boey KW, Jeyaratnam J (1992) Chronic neurobehavioural effects of elemental mercury in dentists. Br J Ind Med 49: 782790
  • Nilsson B, Gerhardsson L, Nordberg GF (1990) Urine mercury levels and associated symptoms in dental personnel. Sci Total Environ 94: 179185
  • NTP (National Toxicology Program) (1993) Toxicology and carcinogenesis studies of mercuric chloride (CAS No. 7487-94-7) in F344 rats and B6C3F1 mice (gavage studies), Technical report No. 408, NIH Publication No. 91–3139, National Institutes of Health, Bethesda
  • Oberly TJ, Piper CE, McDonald DA (1982) Mutagenicity of metal salts in the L5178Y mouse lymphoma assay. J Toxicol Environ Health 9: 367376
  • Ogura H, Takeuchi T, Morimoto K (1996) 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: 175182
  • Osawa J, Kitamura K, Ikezawa Z, Hariya T, Nakajima H (1994) Gold dermatitis due to ear piercing: correlations between gold and mercury hypersensitivities. Contact Dermatitis 31: 8991
  • Pang BK, Freeman S (1995) Oral lichenoid lesions caused by allergy to mercury in amalgam fillings. Contact Dermatitis 33: 423427
  • Piikivi L, Hänninen H (1989) Subjective symptoms and psychological performance of chlorine-alkali workers. Scand J Work Environ Health 15 : 6974
  • Piikivi L, Tolonen U (1989) EEG findings in chlor-alkali workers subjected to low long term exposure to mercury vapour. Br J Ind Med 46: 370375
  • Polak L, Turk JL, Frey JR (1973) Studies on contact hypersensitivity to chromium compounds. in: Kallos P, Waksman BH, de Weck A (Eds) Progress in Allergy, S. Karger, Basel, 17: 145226
  • Pollard KM, Hultman P (1997) Effects of mercury on the immune system. Metal Ions Biol Sys 34: 421440
  • Poma K, Kirsch-Volders M, Susanne C (1981) Mutagenicity study on mice given mercuric chloride. J Appl Toxicol 1: 314316
  • Popescu HI, Negru L, Lancranjan I (1979) Chromosome aberrations induced by occupational exposure to mercury. Arch Environ Health 34: 461463
  • Ramel C (1972) Genetic effects. in: Fridberg L, Vostal D (Eds) Mercury in the environment. An epidemiological and toxicological appraisal, CRC Press, Cleveland, 169181
  • Ratcliffe HE, Swanson GM, Fischer LJ (1996) Human exposure to mercury: A critical assessment of the evidence of adverse health effects. J Toxicol Environ Health 49: 221270
  • Robison SH, Cantoni O, Costa M (1982) Strand breakage and decreased molecular weight of DNA induced by specific metal compounds. Carcinogenesis 3: 657662
  • Robison SH, Cantoni O, Costa M (1984) Analysis of metal-induced DNA lesions and DNA-repair replication on mammalian cells. Mutat Res 131: 173181
  • Roels H, Abdeladim S, Braun M, Malchair J, Lauwerys R (1989) Detection of hand tremor in workers exposed to mercury vapor: A comparative study of three methods. Environ Res 49: 152165
  • Roels H, Gennart JP, Lauwerys R, Buchet JP, Malchaire J, Bernard A (1985) Surveillance of workers exposed to mercury vapour: validation of a previously proposed biological threshold limit value for mercury concentration in urine. Am J Ind Med 7: 4571
  • Roels H, Lauwerys R, Buchet JP, Bernard A, Barthels A, Oversteyns M, Gaussin J (1982) Comparison of renal function and psychomotor performance in workers exposed to elemental mercury. Int Arch Occup Environ Health 50: 7793
  • Romaguera C, Vilaplana J (1998) Contact dermatitis in children: 6 years experience (1992–1997). Contact Dermatitis 39: 277280
  • Rudner EJ, Clendenning WE, Epstein E, Fisher AA, Jillson OF, Jordan WP, Kanof N, Larsen W, Maibach HI, Mitchell JC, O'Quinn SE, Schorr WF, Sulzberger MB (1975) The frequency of contact sensitivity in North America 1972–74. Contact Dermatitis 1 : 277280
  • Schäfer SG, Elsenhans B, Forth W, Schümann K (1994) Metalle. in: Marquardt H, Schäfer SG (Eds) Lehrbuch der Toxikologie, Spektrum Akademischer Verlag, Heidelberg Berlin, 504549
  • Schiele R, Schaller KH, Grobe T (1979) Untersuchungen an beruflich Quecksilber-exponierten Personen. Arbeitsmed Sozialmed Präventivmed 14 : 226229
  • Schoeny R (1996) Use of genetic toxicology data in U.S. EPA risk assessment: the mercury study report as an example. Environ Health Perspect 104, Suppl 3: 663673
  • Schrallhammer-Benkler K, Ring J, Przybilla B, Meurer M, Landthaler M (1992) Acute mercury intoxication with lichenoid drug eruption followed by mercury contact allergy and development of antinuclear antibodies. Acta Derm Venereol (Stockh) 72: 294296
  • Schroeder HA, Mitchener M (1975) Life-term effects of mercury, methyl mercury, and nine other trace metals on mice. J Nutrition 105: 452458
  • Schuckmann F (1979) Study of preclinical changes in workers exposed to inorganic mercury in chloralkali plants. Int Arch Occup Environ Health 44: 193200
  • Seidenari S, Manzini BM, Danese P, Motolese A (1990) Patch and prick test study of 593 healthy subjects. Contact Dermatitis 23: 162167
  • Sertoli A, Francalanci S, Acciai MA, Gola M (1999) Epidemiological survey of contact dermatitis in Italy (1984–1993) by GIRDCA (Gruppo Italiano Ricerca Dermatiti da Contatto e Ambientali). Am J Contact Dermatitis 10: 1830
  • Shamy MY, El-Gazzar RM, Abou Taleb AN, Christie NT, El Said KF (1995) Somatic cell mutation in workers occupationally exposed to mercury vapors. J Environ Pathol Toxicol Oncol 14: 165171
  • Skerfving S, Vostal J (1972) Symptoms and signs of intoxication. in: Friberg L, Vostal J (Eds) Mercury in the environment. CRC Press, Cleveland, Ohio, USA, p 93107
  • Soleo L, Urbano ML, Petrera V, Ambrosi L (1990) Effects of low exposure to inorganic mercury on psychological performance. Br J Ind Med 47: 105109
  • Soleo L, Vacca A, Vimercati L, Bruno S, Di Loreto M, Zocchetti C, Di Stefano R, Candilio G, Lasorsa G, Franco G, Fao V (1997) Minimal immunological effects on workers with prolonged low exposure to inorganic mercury. Occup Environ Med 54: 437442
  • Stiller-Winkler R, Radaskiewicz T, Gleichmann E (1988) Immunopathological signs in mice treated with mercury compounds. I. Identification by the popliteal lymph node assay of responder and nonresponder strains. Int J Immunopharmacol 10: 475484
  • Storrs FJ, Rosenthal LE, Adams RM, Clendenning W, Emmett EE, Fisher AA, Larsen WG, Maibach HI, Rietschel RL, Schorr WF, Taylor JS (1989) Prevalence and relevance of allergic reactions in patients patch tested in North America − 1984 to 1985. J Am Acad Dermatol 20: 10381045
  • Sulzberger MB, Tolmach JA (1959) Allergische Aufflammungs-Reaktionen in roten Täto-wierungen (Beobachtungen über Entstehen und Abklingen einer Quecksilberüberempfind-lichkeit und über allergische granulomatöse und sarkoide Reaktionen). Hautarzt 10: 110114
  • Sun CC (1987) Allergic contact dermatitis of the face from contact with nickel and ammoniated mercury in spectacle frames and skin-lightening creams. Contact Dermatitis 17: 306309
  • Suter KE (1975) Studies on the dominant-lethal and fertility effects of the heavy metal compounds methylmercuric hydroxide, mercuric chloride, and cadmium chloride in male and female mice. Mutat Res 30: 365374
  • Temesvari E, Daroczy J (1989) Histological examination of immediate and delayed contact allergy provoked by mercuric chloride. Contact Dermatitis 21 : 271272
  • Thomas P, Rueff F, Przybilla B (1994) Aerogene allergische Kontaktdermatitis durch Quecksilberverbindungen bei einer Chemiestudentin. Allergologie 17: 557558
  • Thomson J, Russell JA (1970) Dermatitis due to mercury following amalgam dental restorations. Br J Dermatol 82: 292297
  • Triebig G, Grobe T, Saure E, Schaller KH, Weltle D, Valtentin H (1984) Untersuchungen zur Neurotoxizität von Arbeitsstoffen. VI. Längsschnittstudie bei beruflich Quecksilber-belasteten Personen. Int Arch Occup Environ Health 55: 1931
  • Triebig G, Schaller KH, Valentin H (1981) Untersuchungen zur Neurotoxizität von Arbeitsstoffen. I. Messung der motorischen und sensorischen Nervenleitgeschwindigkeit bei beruflich Quecksilber-belasteten Personen. Int Arch Occup Health 48: 119129
  • Vena GA, Foti C, Grandolfo M, Angelini G (1994) Mercury exanthem. Contact Dermatitis 31: 214216
  • Verberk MM, Salle HJ, Kemper CH (1986) Tremor in workers with low exposure to metallic mercury. Am Ind Hyg Assoc J 47: 559562
  • Verschaeve L, Kirsch-Volders M, Hens L, Susanne C (1985) Comparative in vitro cytogenetic studies in mercury-exposed human lymphocytes. Mutat Res 157: 221226
  • Verschaeve L, Kirsch-Volders M, Susanne C (1984) Mercury-induced segregational errors of chromosomes in human lymphocytes and in Indian muntjac cells. Toxicol Lett 21 : 247253
  • Vreeburg KJJ, de Groot K, Hoogstraten IMW, von Blomberg BME, Scheper RJ (1991) Successful induction of allergic contact dermatitis to mercury and chromium in mice. Int Arch Allergy Appl Immunol 96: 179183
  • Warfvinge K, Hansson H, Hultman P (1995) Systemic autoimmunity due to mercury vapor exposure in genetically susceptible mice: dose-response studies. Toxicol Appl Pharmacol 132: 299309
  • Watanabe T, Shimada T, Endo A (1982) Effects of mercury compounds on ovulation and meiotic and mitotic chromosomes in female golden hamsters. Teratology 25: 381384
  • WHO (World Health Organization) (1991) Inorganic mercury. IPCS-Environmental Health Criteria 118, WHO, Genf
  • Williamson AM, Teo RK, Sanderson J (1982) Occupational mercury exposure and its consequences for behaviour. Int Arch Occup Environ Health 50: 273286
  • Zasukhina GD, Vasilyeva IM, Sdirkova NI, Krasovsky GN, Vasyukovich LY, Kenesariev UI, Butenko PG (1983) Mutagenic effect of thallium and mercury salts on rodent cells with different repair activities. Mutat Res 124: 163173
  • Ziegler V, Süss E (1985) The TINA test. in: Andersen KE, Maibach HI (Eds) Contact allergy predictive tests in guinea pigs, Curr Probl Dermatol 14: 172192