The Three Modern Faces of Mercury€¦ · Environmental Health Perspectives • VOLUME 110 |...

Environmental Health Perspectives • VOLUME 110 | SUPPLEMENT 1 | February 2002 11

Several reviews of the toxicology of mercuryhave appeared recently (1–4). In this presentreview I do not repeat all the material pre-sented in these extensive reviews. Instead Ifocus on three chemical species of mercurythat are currently the source of intensepublic health interest.

Public health concerns about methylmercury in edible tissue of fish suddenlyerupted in 1969 when fish from LakeSt. Clair bordering Michigan were found tohave high levels. This and other findings dis-cussed in this review have maintained publichealth concerns over this form of mercury.In 1997, the U.S. Environmental ProtectionAgency (U.S. EPA) reduced recommendedsafe intakes of methyl mercury by a factor ofabout five (1), which brought public appre-hension to new heights.

The U.S. EPA-recommended safe intakelevel is referred to as the reference dose andis defined as that dose that can be absorbeddaily for a lifetime without a significant riskof adverse effects. The new reference dosewas estimated in 1997 to be 0.1 µg methylmercury/kg body weight/day. This doseimplies that the amount of methyl mercuryingested in just one 7-oz can of tuna fish perweek would equal or even slightly exceed thenew limit, depending on the consumer’sbody weight. Other federal regulatoryagency guidelines allow higher levels (4):The U.S. Food and Drug Administration(FDA) guideline is equivalent to 0.5 µgHg/kg/day, and that of the Agency for ToxicSubstances and Disease Registry (ATSDR) is0.3 µg Hg/kg/day.

Mercury amalgam tooth fillings havebeen used since the early nineteenth century.Periodically, debates have arisen about thepotential danger from mercury. Thesedebates are sometimes referred to as the“amalgam wars.” The most recent began withan observation in the 1970s that mercuryvapor was released from amalgam, especiallyduring the process of chewing, and that thisvapor could be inhaled. Concentrations of

mercury vapor measured in the air of the oralcavity approached and even exceeded occupa-tional health limits. The debate has nowreached new heights (or lows, depending onthe side of the argument) with claims thatchronic degenerative diseases of the nervoussystem such as Parkinson’s disease andAlzheimer’s disease are caused or exacerbatedby mercury released from amalgam.

In the late summer of 1999, concern wasexpressed by the American Academy ofPediatrics and by the U.S. Public HealthService about the safety of a mercury preserva-tive used in many vaccine preparations rou-tinely administered to infants (5). Withinabout 18 months, the mercury preservativewas removed by the manufacturers from allvaccines destined for use in the United States.

The mercury preservative has the molec-ular formula CH3CH2–Hg–S–C6H4–COOH. This preservative was introducedinto vaccines in the early 1930s and has beenused ever since (6). It was given a clean billof health by the FDA in 1976 (7). However,the U.S. EPA later lowered its allowable safelong-term daily intake for mercury, as dis-cussed above. As a result, a more recentreview of thimerosal by the FDA raisedquestions about possible health risks.

My objective in this review, therefore, isto give the toxicologic background for thesethree species of mercury and the public healthissues surrounding them. In each case, Iaddress human exposure, disposition in thebody, and adverse effects. Where possible, Idiscuss the underlying mechanisms. Emphasisis on the human target. I also discuss ecologicaspects only inasmuch as they may play a rolein human exposure.

Quantitative estimates of human healthrisks are not made in this review. Such a taskis left to “expert committees” covering arange of disciplines that cannot be masteredby one individual. Nevertheless, the toxico-logic background presented here should giveat least a qualitative idea of the type of healthrisks we face from these forms of mercurys.

Methyl Mercury in Fish

History of Human Exposure

The first methyl mercury compounds weresynthesized in a chemical laboratory inLondon in the 1860s (8). Two of the labora-tory technicians died of methyl mercury poi-soning. This so shocked the chemicalcommunity that methyl mercury com-pounds were given a wide berth for the restof the century. However, early in the twenti-eth century the potent antifungal propertiesof the short-chain alkyl mercury compoundswere discovered, leading to application toseed grains, especially for cereal crops. Thewidespread global use of these mercury com-pounds was found to be highly protective ofseed grain from what otherwise would bedevastating fungal infections and the loss ofthe grain harvest.

Despite this widespread use, few cases ofpoisoning were reported for the first half ofthe twentieth century. However, in the late1950s and early 1960s serious outbreaks ofalkyl mercury poisoning erupted in severaldeveloping countries (9). The largest, mostrecent outbreak occurred in rural Iraq in thewinter of 1971–1972 (10). Some 6,000 caseswere admitted to hospitals. An epidemiologicfollow-up suggested that as many as 40,000individuals may have been poisoned.

These outbreaks were caused by prepar-ing homemade bread directly from thetreated seed grain. Several factors con-tributed to these mass health disasters. Thewarning labels were not written in the locallanguage. Well-known symbols for poisonsin the Western world, such as the skull andcrossbones, have no meaning to rural Arabsunfamiliar with stories of “pirates on theSpanish Main.” Typically, a red dye is addedto the treated grain to indicate the presenceof a fungicide. This was counterproductive,as the victims washed away the dye, thinkingthey had also removed the poison. Theinsidious properties of methyl mercury wereanother important factor, as there is a long

Address correspondence to T.W. Clarkson, Dept. ofEnvironmental Medicine, University of RochesterSchool of Medicine, 601 Elmwood Ave., Rochester,NY 14642 USA. Telephone: (716) 275-3911. Fax:(716) 256-2591. E-mail: [email protected]

My thanks to B. Weiss for his review of the man-uscript. I acknowledge support from the NationalInstitute of Environmental Health Sciences, grantsES01247, ES06484, and ES010219.

Received 17 August 2001; accepted 1 November2001.

Reviews, 2002

The three modern “faces” of mercury are our perceptions of risk from the exposure of billions ofpeople to methyl mercury in fish, mercury vapor from amalgam tooth fillings, and ethyl mercury inthe form of thimerosal added as an antiseptic to widely used vaccines. In this article I review humanexposure to and the toxicology of each of these three species of mercury. Mechanisms of action arediscussed where possible. Key gaps in our current knowledge are identified from the points of viewboth of risk assessment and of mechanisms of action. Key words: amalgam, ethyl mercury, mercury,methyl mercury, thimerosal. Environ Health Perspect 110(suppl 1):11–23 (2002).http://ehpnet1.niehs.nih.gov/docs/2002/suppl-1/11-23clarkson/abstract.html

The Three Modern Faces of Mercury

Thomas W. Clarkson

Department of Environmental Medicine, University of Rochester School of Medicine, Rochester, New York, USA

latent period between ingestion and the firstappearance of symptoms (10).

Also in the late 1950s, evidence emergedof environmental damage from treated grain(11). It was observed in Sweden that preda-tory birds were developing neurologic disor-ders. These birds were at the top of a foodchain that began with small mammals con-suming the treated grain freshly planted inthe fields. Analysis of feathers from museum-preserved birds indicated a sharp rise in mer-cury levels at the time when mercurialcompounds were introduced as agriculturalfungicides. Because some of these birds weremigratory, it was possible to show that ele-vated mercury levels were found only inthose feathers that grew when the birds werein Sweden.

As a control measure, the Swedish inves-tigators decided to check mercury levels inthe feathers of fish-eating birds, where mer-cury levels were assumed to be low. To theirastonishment, mercury levels were elevateddespite that these birds had no dietary con-nection with the treated grain. Eventuallythis finding led to a landmark discovery onthe environmental fate of mercury, namelythat microorganisms in the aquatic environ-ment are capable of converting inorganicmercury to methyl mercury. This is the firststep in the aquatic food chains, where methylmercury bioaccumulates in higher organismsto plankton, herbivorous, and finally in thetop fish predators such as sharks and fish-eat-ing marine mammals. A similar food chainexists in bodies of freshwater, with suchspecies as pike and bass having some of thehighest levels of methyl mercury.

The potential for bioaccumulation inaquatic food chains was demonstrated dra-matically in two outbreaks of human poison-ing in Japan at about this time. The Japanesehealth authorities in Minamata had beenaware for some time that fishermen and theirfamilies were suffering from a neurologic dis-ease, exhibiting signs of incoordination, con-stricted visual fields, and numbness in theextremities. The cause was elusive until a vis-iting physician from Scotland recognized theneurologic signs and symptoms from cases ofoccupational methyl mercury poisoning hehad seen in England in 1939 (8). Eventuallythe source in Japan was traced to a factorymanufacturing acetaldehyde, where inor-ganic compounds of mercury were used as acatalyst. The producers were unaware thatthe synthetic process converted some of themercury to methyl mercury, which was dis-charged into Minamata Bay. It was difficultto believe that methyl mercury released intoa large ocean bay could be bioaccumulatedto such an extent that the fish carried levelsof methyl mercury that would prove lethalwhen consumed by humans.

Global Cycling of Mercury

The twin discoveries of biomethylation andbioaccumulation aroused intense interest inthe environmental fate of mercury and inpathways to human exposure. Methyl mer-cury was soon detected in all species of fishand in fish-consuming animals. The sourceappeared to be inorganic mercury biomethy-lated by microorganisms in sediments ofboth fresh and ocean water.

Many anthropogenic sources were iden-tified. Chloralkali plants discharged inor-ganic mercury as waste into rivers, lakes, andocean bays. Paper pulp factories likewise dis-charged a variety of mercury compoundsused a slimicides. These practices now havebeen eliminated, but contamination ofaquatic sediments now occurs worldwidebecause of extensive goldmining operations,for example, in the Amazon basin (12).

Large quantities of liquid mercury areused to extract the sedimentary gold foundin river beds. Pure gold is recovered whenthe mercury is evaporated from the amalgamby heating. It has been estimated that over130 tons of mercury have been released eachyear into the Amazon basin alone (13).

The global cycling of mercury beginswith the evaporation of mercury vapor fromland and sea surfaces. Volcanoes can be animportant natural source (14). The burningof fossil fuel, especially coal and municipalwaste incineration, is a major anthropogenicsource to the atmosphere. Mercury vapor is achemically stable monatomic gas. Its resi-dence time in the general atmosphere is esti-mated to be about 1 year. Thus, mercuryvapor is globally distributed even from pointsources. By processes not yet fully under-stood, the vapor is oxidized in the upperatmosphere to a water-soluble ionic mercury,which is returned to the earth’s surface inrainwater. This global cycling of mercuryresults in the distribution of mercury to themost remote regions of the planet. Forexample, environmental mercury levels evenin the arctic water may not differ greatlyfrom levels in more southern latitudes.

The global cycling of mercury, alongwith the processes of biomethylation andbioaccumulation, implies that humans musthave consumed methyl mercury in fish dat-ing back to times before Homo sapiensevolved (15). It could be argued that environ-mental levels of mercury vapor were muchhigher in an earlier period of the earth’s his-tory when oxygen had not yet appeared inthe atmosphere. As levels of oxygen began torise, increasing amounts of the vapor wouldbe converted to the ionic form. Life forms atthose Archean times had to protect them-selves not only from this new toxic gas, oxy-gen, but also from ionic mercury pouring

down in rainwater. Perhaps it is no coinci-dence that those proteins and antioxidantmolecules present in today’s cellular machin-ery to protect against oxygen also are ourmain line of defense against mercury.

Disposition in the BodyThe U.S. EPA (1,2) and ATSDR (3) inrecent reviews give extensive details on thedisposition of methyl mercury in the body.A brief review and update are provided here.

About 95% of methyl mercury ingestedin fish is absorbed in the gastrointestinaltract, although the exact site of absorption isnot known. It is distributed to all tissues in aprocess completed in about 30 hr. About 5%is found in the blood compartment andabout 10% in brain. The concentration inred blood cells is about 20 times the concen-tration in plasma. Methyl mercury crossesthe placental barrier. Levels in cord blood areproportional to but slightly higher than levelsin maternal blood. Levels in the fetal brainare about 5–7 times that in maternal blood(16). Brain-to-blood ratios in adult humansand other primates are approximately in thesame range.

Methyl mercury avidly accumulates ingrowing scalp hair. Concentrations in hairare proportional to simultaneous concentra-tions in blood but are about 250 timeshigher. They are also proportional to con-centrations in the target tissue, the brain(16). Longitudinal analysis of strands ofscalp hair can recapitulate past blood andbrain levels (17). Hair and blood are used asbiologic indicator media for methyl mercuryin both the adult and fetal brain (in the lat-ter case, maternal hair or cord blood).

Methyl mercury is slowly metabolized toinorganic mercury mainly by microflora inthe intestines, probably at a rate of about 1%of the body burden per day. Some demethy-lation also occurs in phagocytic cells The bio-chemical mechanism is unknown. Althoughmethyl mercury is the predominant form ofmercury during exposure, inorganic mercuryslowly accumulates and resides for long peri-ods in the central nervous system. It isbelieved to be in an inert form, probablyinsoluble mercury selenide (18).

Urinary excretion is negligible, of theorder of 10% or less of total elimination fromthe body. Methyl mercury undergoes exten-sive enterohepatic cycling. It is secreted intobile and partly reabsorbed into the portal cir-culation and thereby returned to the liver. Afraction of the biliary mercury is converted bymicroflora to inorganic mercury. The latter isreabsorbed only to a small extent. Thus, mostof the methyl mercury is eliminated from thebody by demethylation and excretion of theinorganic form in the feces. The processes ofbiliary secretion and demethylation by

Reviews, 2002 • Clarkson

12 VOLUME 110 | SUPPLEMENT 1 | February 2002 • Environmental Health Perspectives

microflora do not occur in suckling animals.The role of these two processes in sucklinghuman infants is unknown.

The high mobility of methyl mercury inthe body is not due to lipid solubility, asclaimed in some textbooks. Methyl mercuryis present in the body as water-soluble com-plexes mainly if not exclusively attached tothe sulfur atom of thiol ligands. It enters theendothelial cells of the blood–brain barrieras a complex with L-cysteine. The process isso specific that the complex with the opticalisomer D-cysteine is not transported.Structurally, the L-complex is similar to thelarge neutral amino acid L-methionine and iscarried across the cell membrane on the largeneutral amino acid carrier (19).

Methyl mercury is pumped out of mam-malian cells as a complex with reduced glu-tathione. For example, it is secreted into bileas a glutathione complex. The glutathionemoiety is degraded in the bile duct and gallbladder to a dipeptide and finally to theL-cysteine complex. Presumably, in this formit is reabsorbed into the bloodstream to bereturned to the liver, thereby completing theenterohepatic cycle (20–22).

The elimination of methyl mercuryfrom the body approximates first-orderkinetics. Half-times vary from one tissue toanother but generally fall in the range of45–70 days. Thus, individuals with long-term regular exposure to methyl mercuryattain a steady-state body burden in about1 year (five half-times).

Several thiol-containing complexingagents have been successfully used to removemethyl mercury from the body [e.g., in theIraq outbreak; see Clarkson et al. (23)]. Aninteresting example is a thiol-containingresin that, when given by mouth, traps themethyl mercury secreted in bile and carries itinto the feces. Perhaps the most promisingcomplexing agent is N-acetylcysteine (24). Itenhances methyl mercury excretion whengiven orally, has a low toxicity, and is widelyavailable in the clinical setting.

Adverse EffectsThe major toxic effects of methyl mercuryare on the central nervous system. Its toxicaction on the developing brain differs inboth mechanism and outcome from itsaction on the mature organ, so the twoactions are treated separately here [fordetailed reviews, see U.S. EPA and ATSDR(2,3)]. However, recent reports have raisedthe possibility that methyl mercury mayhave adverse effects on other targets inthe body.

The mature central nervous system. Theaction of methyl mercury on adults is char-acterized by a latent period between expo-sure and onset of symptoms. The period can

be several weeks or even months, dependingof the dose and exposure period. Perhaps themost dramatic example of latency was in thecase of severe, ultimately fatal poisoning of achemistry professor from exposure todimethyl mercury (25). A single exposurefrom a spill of liquid dimethyl mercury tookplace in August. The professor continuedher normal professional work without anyapparent ill effects. In November she pre-sented a paper at an overseas conference. Itwas not until late December that the firstsymptoms appeared. Within a few weeks thefull syndrome of severe methyl mercury poi-soning became manifest. Despite manydecades of research on methyl mercury toxi-cology, the mechanism underlying this longlatent period is still unknown.

Paresthesia, a numbness or a “pins andneedles” sensation, is the first symptom toappear at the lowest dose (10). This mayprogress to cerebellar ataxia, dysarthria, con-striction of the visual fields, and loss of hear-ing. These signs and symptoms are causedby the loss of neuronal cells in specificanatomical regions of the brain. For exam-ple, ataxia results from the loss of the granulecells in the cerebellum. The neighboringPurkinje cells are relatively unaffected.

The mechanism underlying the focaldamage to the adult brain is still not estab-lished with any certainty. Syversen (26)examined the effect on protein synthesis invarious areas of brain of rats poisoned withmethyl mercury. Protein synthesis wasinhibited in all three areas studied—thegranule and Purkinje cells of the cerebellum,and the cells from the cortical areas of thebrain. Protein synthesis recovered in twotypes of neurons but not in the granule cells.These data suggest that the focal damage tothe brain is not due to the initial insult butdepends on the capacity of neuronal cells forrepair, as suggested by Jacobs et al. (27).Apparently the small granule cells lack therepair systems present in the other largercells. Sarafian et al. (28) have suggested thatthe selective vulnerability of cells in the ner-vous system may arise from a “criticalabsence of inherent protective mechanisms.”

Cellular defenses may be decisive indetermining the toxic outcome and deservefurther study. If we understand the defensemechanism, we may be able to predict whichindividuals are most susceptible. Thiol com-pounds probably play a key role (29).Resistant cells have higher levels of the thiol-containing peptide glutathione (30).Glutathione also plays a key role in theexcretion of methyl mercury [for furtherdiscussion, see Sarafian et al. (28)].

Selenium is a dietary component thatmay affect the disposition and toxicity ofmethyl mercury. Ganther et al. (31) were the

first to observe that selenium compoundscould delay the onset of toxic effects in ani-mals fed methyl mercury in tuna. This gaverise to a series of studies by his group andothers. However, despite promising indica-tions from animal studies, no definite studieshave yet been carried out on human popula-tions co-exposed to methyl mercury andselenium [for a recent review, see NationalResearch Council (4)].

Methyl mercury is converted to inor-ganic mercury in the brain. It is possible thatthe inorganic ion is the proximate toxicagent responsible for the brain damage.However, experiments on rats comparingmethyl and ethyl mercury compounds sug-gest that the intact methyl mercury radical isthe toxic agent (32). Ethyl mercury convertsto inorganic mercury more rapidly thanmethyl mercury, but the latter producesmore severe brain damage.

Autopsy samples taken years after expo-sure to methyl mercury reveal that inorganicspecies account for most if not all of theremaining mercury in the brain (33). It hasbeen suggested that the long residence timeis due to inorganic mercury forming aninsoluble complex with selenium (18).However, Charleston et al. (34) have chal-lenged this view, suggesting that inorganicmercury released in brain tissue from methylmercury may be the proximate toxic agent.The toxicologic role of inorganic mercuryremains a matter of debate.

Other adverse effects in adults. Most epi-demiologic studies and clinical reports onadults [for review, see WHO (18)] haveidentified neurologic signs and symptoms ofpoisoning associated mainly with the centralnervous system. An important exception isan extensive study on the relationshipbetween fish consumption, levels of mercuryin urine and scalp hair, and risk of cardiovas-cular disease in adult male residents living ineastern Finland (35). A statistically signifi-cant correlation was found between mercurylevels and cardiovascular disease even aftercorrection for numerous cardiovascular riskfactors. A subsequent study by the samegroup found a correlation between mercuryaccumulation and accelerated progression ofcarotid atherosclerosis (36).

However, it is difficult to draw firmconclusions. Stress, believed to be a majorrisk factor (37), was not directly measured.The highest recorded hair level of 15.7 ppmwas more than six standard deviations fromthe mean. A histogram of hair levels was notpresented, but these statistics imply that asmall percentage of the study group hadhigh mercury levels. Outlying and “influen-tial points” may play a major role in studiesof this type [e.g., Myers et al. (38)]. Itwould have been of interest to see if these

Reviews, 2002 • Three faces of mercury


correlations persisted when the very highmercury levels were excluded.

Given the serious health implications, arepeat of this study in another population isneeded. If these findings are confirmed, twolong-held dogmas may have to be aban-doned, namely, that methyl mercury primar-ily affects the central nervous system andthat the prenatal period (see below) is themost susceptible part of the life cycle.

Effects on the developing brain. The firstindication of the special susceptibility of thedeveloping brain to prenatal exposure tomethyl mercury came from anecdotal reportsfrom Minamata that mothers with mildsymptoms gave birth to offspring with severebrain damage. The Iraq outbreak confirmedthat severe brain damage can occur fromhigh prenatal exposure. A milder syndromewas also identified in the Iraq outbreak (39).Children apparently normal nevertheless hada history of delayed achievement of develop-mental milestones and, on examination,exhibited neurologic abnormalities such asbrisk tendon reflexes. When the prenatalexposure was determined from mercury lev-els in maternal hair samples, it was possibleto construct a dose–response relationshipbetween peak hair mercury levels in preg-nancy versus number of abnormal offspringshowing developmental delays and abnormalneurologic findings (39,40).

This study was of interest for two rea-sons. First, a dose–response relationship hasbeen established for prenatal exposures to atoxicant; that is, a dose to the mother pre-dicts the probability of effects in her off-spring. This discovery laid the groundworkfor further quantitative estimate of prenatalrisks from methyl mercury. No doubt thisrelationship was made possible by the paral-lel between levels of mercury in maternaland fetal tissues. Indeed, it was later demon-strated in another study that maternal hairlevels of mercury were proportional to levelsin autopsy samples of brain tissue frominfants who died shortly after birth (16).Ernhart et al. (41), at virtually the sametime, published a dose–response curve forprenatal exposures to ethanol. Probably aswith methyl mercury, the high mobility ofethanol ensures that maternal levels predictthose in the fetus.

A second unique aspect of the Iraqistudy (40) was the application of continuoussingle-strand hair analysis to determine peaklevels during pregnancy. By the use of X-rayfluorescence analysis, it was possible to mea-sure the concentration of mercury in con-tiguous 2-mm segments of a single strand ofmaternal hair, thus giving a complete pictureof mercury levels in pregnancy. Moreover,because exposure in Iraq took place over asingle period of time, it was possible to fit

the hair data with a single compartmentmodel covering both the rising levels duringintake and the exponential fall afterward.This allowed the true peak value to be calcu-lated from the curve by fitting all the datapoints, as opposed to taking the single high-est value, which would be more prone toerror. It is unfortunate that this method ofanalysis was not used in subsequent studiesof prenatal exposure.

The studies of the Iraq outbreak con-firmed what had been suspected from theoutbreak in Japan, that the fetal brain wasmore sensitive than the mature organ. ASwedish expert group (11) had estimated athreshold level for neurologic effects inadults at about 50 ppm in hair, an estimateconfirmed by the findings in Iraq (10). Thislevel may be compared with an estimatedthreshold as low as 10 ppm for prenataleffects (milestones of development and neu-rologic change) in Iraq (40). As these studieswere being conducted and early findings pre-sented at scientific meetings, concern arosethat methyl mercury in fish normally con-sumed in our diet might present risks of pre-natal damage. Several large epidemiologicstudies were conducted in people consumingfreshwater fish (42) and ocean water fish(43), and large-scale studies are continuingeven to this day focusing on neuropsycho-logic development [e.g., (44,45)]. Thesestudies have not yet provided a consistentpicture of the lowest prenatal levels that offera measurable risk of damage to the develop-ing brain. However, at this time it can besaid that these studies on fish-eating popula-tions taken as a whole are consistent with theoriginal findings in Iraq that effects can bedetected in the range of 10 ppm in maternalhair. Indeed, a U.S. EPA reference dose pub-lished recently (2) is identical to the previousestimate from the Iraq data (1).

Mechanism of prenatal damage. Severalstudies have given some insight into themechanism underlying prenatal brain dam-age. Autopsy brain samples from theMinamata outbreak indicated widespreaddamage to all areas of the fetal brain, asopposed to the focal lesions seen in adult tis-sue. Microcephaly was also observed (18).Autopsy tissue from Iraq also gave invaluableclues to the nature of prenatal brain damage(46). The normally ordered parallel arrays ofneuronal cells in the cortex were found to bedisrupted, which is indicative of a generaldisturbance in the developmental growth ofthe brain. Moreover, neurons were presentsuch as Purkinje cells that had failed tomigrate to the cerebellum. These findingsfrom both Japan and Iraq indicated that themost basic processes in brain developmentwere affected, namely, neuronal cell divisionand migration.

Experimental work in animals andin vitro has provided a mechanism explainingwhy methyl mercury inhibits both cell divi-sion and migration (47–49). These studiesshow that the cytoarchitecture first affected atthe lowest levels of methyl mercury is themicrotubular system. Intact microtubules arerequired for both cell division and migration.Microtubules are formed by a treadmillingprocess whereby assembly from α-and β-tubulin monomers occurs at one end and dis-assembly at the other. Apparently, methylmercury binds to thiol ligands (–SH) groupson the tubulin monomers and blocks theassembly process. The disassembly continuesunchanged, thus leading to the complete lossof the tubule.

Other adverse effects of prenatalexposures. Studies in 7-year-old childrenrevealed an elevation in both systolic anddiastolic blood pressure that correlated withprenatal exposure to methyl mercury (50).The study was conducted in the FaroeIslands on a large cohort of children whosemothers had ingested methyl mercurymainly from whale meat but also from fish.This effect is seen only at the lower range ofblood levels from about 1 to 10 µg Hg/L.Above this range no further increase is seen,even at blood levels in the mother ranging ashigh as 250 µg Hg/L.

As elevated blood pressure in childrenmay be indicative of later cardiovascular prob-lems, this finding is of public health concern.Further work is needed to confirm thisfinding and to understand its mechanism.

Thimerosal in Vaccines

Mercury in the thimerosal molecule is inthe form of ethyl mercury (CH3CH2-Hg+),for which there is limited toxicologic infor-mation. Thus, estimates of health risksfrom thimerosal in vaccines (7) were basedon the assumption that ethyl mercury istoxicologically similar to its close chemicalrelative, methyl mercury (CH3–Hg+),about which much is known. However, asdiscussed below, there are reasons to believethat this assumption is not necessarily cor-rect for all aspects of the disposition andtoxicity of ethyl mercury compounds,including thimerosal.

History of Human ExposureEthyl mercury compounds were first synthe-sized in the nineteenth century in a chemi-cal laboratory in London (8). In the late1880s diethyl mercury was first used in thetreatment of syphilis, a practice soon aban-doned because of the toxic properties of thisagent. However, early in the twentieth cen-tury, the fungicidal properties of the short-chain alkyl mercury compounds led tocommercial applications in agriculture. For



example, they are especially effective in aplant root disease in wheat caused byTelletia triticia. In fact, many differentorganic mercury compounds were beingused to prevent seed-borne diseases of cerealby 1914 (51,52).

Generally speaking, the ethyl mercuryfungicides were used effectively and safely.However, a number of outbreaks of poison-ing occurred in some developing countries(8). For example, two outbreaks occurred inrural Iraq in 1956 and 1960 from the misuseuse of the fungicide ethyl mercury toluenesulfonilamide (51). The farmers’ families pre-pared homemade bread directly from thetreated grain instead of planting it. Hundredsof cases of severe poisoning occurred, manyof which had a fatal outcome. Cases of ethylmercury poisoning have occurred in China asrecently as the 1970s. The exposure pathwaywas the same as in Iraq: The farmersconsumed rice treated with ethyl mercurychloride (53).

Ethyl mercury in the form of thimerosalhas found wide application in medicine as adisinfectant. Axton (54) reported case histo-ries of four children and two adults severelypoisoned by accidental exposure. Five of thesix cases died. Rohyans et al. (55) reported acase of severe poisoning from treatment ofan infected ear. Pfab et al. (56) reported onan attempted suicide from drinking asolution of thimerosal, resulting in severepoisoning. Treatment of infants withomphaloceles resulted in high levels of mer-cury in autopsy tissues (57). Cases of humanpoisoning have also occurred from infusionof large volumes of plasma containingthimerosal as a preservative (58,59).

Disposition in the BodyIf, after injection of the vaccine, thethimerosal molecules were to remain intactfor a period sufficient to allow diffusion tothe bloodstream and thence to the kidneys,rapid excretion might take place. The car-boxyl group of thiosalicylic acid might allowthimerosal to be a substrate for the systemresponsible for the tubular secretion of weakacids. Rapid urinary excretion of thimerosalwould then be possible.

This possibility seems unlikely.Thimerosal contains the ethyl mercury radi-cal attached to the sulfur atom of the thiolgroup of salicylic acid. Generally, mercuricions bind tightly but reversibly to thiol lig-ands (60). It is likely, therefore, that the ethylmercury cation will dissociate from the thios-alicylic acid moiety immediately after injec-tion to bind to the surrounding thiol ligandspresent in great excess in tissue proteins.

Thimerosal is used as a thiol titrationreagent in numerous experimental studies[e.g., Elferink (61)]. This application would

be possible only if rapid dissociation of ethylmercury took place in the presence ofendogenous thiol groups in the cells and tis-sues under study. In vitro, the thiosalicylatemoiety is degraded by oxidation to dithio-salicylic acid followed by further oxidationto 2-sulfinobenzoic acid (62).

Ulfvarson (63) demonstrated that thetype of anion attached to the alkyl mercuryradical made little difference to the ultimatedisposition in the body. These include suchanions as hydroxyl, cyanide, and even thethiol-containing propane diolmercaptide.These findings suggest that the mercuryradical rapidly dissociates from the anion inthe parent compound to attach to ligandsin tissues.

Therefore, it is assumed that administra-tion of thimerosal results in the immediaterelease of the ethyl mercury to the surround-ing tissues. Toxicologically, ethyl mercury inthimerosal is assumed to follow the samepathways of disposition as ethyl mercuryabsorbed into the body from other ethylmercury compounds.

Patterns of tissue disposition andexcretion. Little is known about mercury lev-els in human tissue after administration ofthimerosal. Suzuki et al. (58) reported levelsof total and inorganic mercury in the tissuesof a 13-year-old boy who had died 5 daysafter receiving infusion of artificial humanplasma containing thimerosal as a preserva-tive. The infusion of plasma had taken placeover a period of 6 months, with a total esti-mated dose of 284–450 mg Hg. The levelsof total mercury from high to low were inthe following order: liver, kidneys, skin,brain, spleen, and lowest in plasma. The redcell levels were at least 10-fold higher thanplasma. The distribution pattern is generallysimilar to that seen for methyl mercury.These findings are supported by studies inprimates dosed with thimerosal (64).

It is interesting that hair levels werehigh. The section proximal to the scalp hada level of 187 µg Hg/g, whereas the level inblood was approximately 7 µg Hg/mL, giv-ing a hair-to-blood ratio of 27:1. This islower than the commonly assumed ratio formethyl mercury of 250:1, but possibleredistribution of mercury in autopsy bloodsamples and uncertainty in the length andexact position of the proximal segmentmake the estimates of the hair-to-bloodratio uncertain. However, it does indicatethat ethyl mercury, like methyl mercury, isaccumulated in scalp hair.

Matheson et al. (59) reported on bloodand urine levels in one patient exposed tothimerosal in long-term injections of gammaglobulin. Specifically, they reported on levelsof total and inorganic mercury before andafter one injection of gamma globulin. The

data allow a rough calculation of how similarthe observed increase in blood level is to thatexpected from methyl mercury. The injecteddose was 0.6 mL/kg containing 50.3 µgHg/mL, to give a total mercury dose of 30 µgHg/kg. The disposition parameters formethyl mercury in adult humans (18) predictthat 5% of the dose—1.5 µg Hg—isdeposited in the blood compartment. Thevolume of the latter is 70 mL, assuming theblood compartment is 7% of the bodyweight. Thus, 1.5 µg Hg would be depositedin 70 mL of blood to give an increase onconcentration of 1500/70 µg Hg/L = 21 µgHg/L. The observed increase was 18 µgHg/L. This calculation suggests that the dis-position of mercury after thimerosal is notvery different from that expected frommethyl mercury.

The pattern of urinary excretion alsoindicates similarities to that with methylmercury. Matheson et al. (59) do not quotea specific figure for the change in urinaryexcretion after injection of thimerosal, butthe graph published in their article indicateslittle change. They state that 90% of thetotal mercury in urine was in the inorganicform. Adult humans exposed to methyl mer-cury excrete little mercury in urine and all inthe inorganic form (10).

Conversion to inorganic mercury. Thereis, however, one important difference frommethyl mercury illustrated in the reportfrom Matheson et al. (59). Inorganic mer-cury accounted for about 50% of the totalmercury in blood samples collected from thispatient. This is in marked distinction frommethyl mercury, where inorganic mercuryaccounts for only about 10% of total mercuryin blood (10).

Similar findings were made in the casedescribed by Suzuki et al. (58). A significantfraction of the total mercury in both grayand white matter of the brain was in theform of inorganic mercury of the order of30–40%. The kidney cortex had the highestpercentage. These findings are confirmed bystudies on experimental animals (32). Bloodand tissue levels, including the brain, werehigher in animals dosed with ethyl mercurycompared with an equivalent dose of amethyl mercury compound. The high tissuelevels of inorganic mercury seen in bothhumans and animals indicate that ethyl mer-cury breaks down to inorganic mercurymore rapidly than methyl mercury.

Blood levels from thimerosal in vaccines.Stajich et al. (65) are the first and only investi-gators to measure disposition of mercurybefore and after administration of vaccinescontaining thimerosal. They reported onblood levels of mercury before and 48–72 hrafter administration of a single dose of hepati-tis vaccine in the first week after birth. Seven



were preterm infants (average birth weight,748 g) and five were term infants (averagebirth weight, 3,588 g). Prevaccination bloodlevels were 0.04–0.5 µg Hg/L. The preterminfant levels rose to an average value of 7.4 µgHg/L, whereas the levels in term infants were2.2 µg Hg/L.

These levels are similar to those expectedfrom methyl mercury. The dose was thesame for all infants, 12.5 µg Hg. Five per-cent, or 0.625 µg Hg, should be deposited inthe blood compartment, which is assumedto be 8% of the infant’s body weight. Thus,for the preterm infants, 0.625 µg Hg wouldbe deposited in a blood volume of 0.08 ×748 = 60 mL to give a predicted concentra-tion of 0.625 × 1,000/60 = 10.4 µg Hg/L.This compares to an observed increase of 6.8µg Hg/L. The predicted increase for theterm infants based on methyl mercury is0.625 × 1,000/287 = 2.2. The observedincrease was identical, 2.2 µg Hg/L.

These estimates also suggest that the dis-position of mercury after a dose of thimerosalis similar to that expected from methyl mer-cury. However, these estimates can beregarded as approximate at best. Individualvalues for each infant were not reported. Theblood levels in samples collected between48 and 72 hr may not have been the truemaximum levels after distribution of theinjected dose.

A preliminary report by Pichichero et al.(66) indicated blood levels of mercury ininfants lower than what would be expectedfrom methyl mercury. These infants, ≤6months of age, had received some vaccinescontaining thimerosal. Most blood sampleswere collected 1 week or more after the lastvaccination. The highest recorded level was4.1 µg Hg/L, and many were below the detec-tion limit of about 0.5 µg Hg/L. When theauthors performed calculations similar to thosedescribed above, they found that the methylmercury dispositional parameters predictedsignificantly higher levels than those observed.

The main difference in design of thePichichero et al. (66) study compared withthat of the Stajich et al. (65) study is that inthe former, samples were collected much laterafter the last dose of thimerosal. Both studiescould be consistent if the half-time for ethylmercury in blood is shorter than that formethyl mercury. The time after collection of48–72 hr is too short for a measurable declinein blood levels in the Stajich et al. study. Theurine levels in the Pichichero et al. study werelow, which is consistent with other observa-tions on thimerosal discussed previously.Significant amounts of mercury were foundin fecal samples that might account for thelower blood levels.

In conclusion, both animal and humanstudies indicated that the pattern of tissue

disposition of ethyl mercury was qualita-tively similar to that of methyl mercury, withbrain levels of the intact mercury beingslightly higher for methyl than for ethyl. Theconversion in body tissues to inorganic mer-cury appears to be substantially faster fromethyl than from methyl. We know littleabout the kinetics of elimination of mercuryfrom the body when dosed with ethyl mer-cury compounds. The feces is the mainpathway of elimination. The residence timein the body is probably shorter for ethyl, butquantitative data are lacking.

Adverse EffectsBall et al. (7) have reviewed the animal litera-ture on ethyl mercury toxicity. Toxicity testsconducted before marketing thimerosal in1931 in several animal species involved high(≤45 mg Hg/kg) acute doses with only ashort follow-up period (67). The studies havelittle relevance to today’s concern over risksfrom low doses from vaccines. Chronic car-cinogenicity studies were conducted on ratswith twice weekly doses ranging from 30 to1,000 µg Hg/kg. Weight loss was observed atthe highest-dose group. Unfortunately, nobrain histopathology was reported, makingthese studies difficult to extrapolate to cur-rent human exposure.

Magos et al. (32) compared the targetorgan toxicity of ethyl and methyl mercuryin rats. Five daily doses of 8 mg Hg/kg weregiven by gavage for 5 consecutive days. Brainand kidney histopathology was examined 3and 10 days after the last dose. In general,kidney damage was more severe after ethylmercury and brain damage more severe aftermethyl mercury. However, when the dose ofethyl mercury was increased by only 20%,the brain damage was similar or slightlymore severe than that seen from the lowerdose of methyl mercury.

Magos (68) has reviewed the publishedcases of human poisoning resulting fromexposure to thimerosal. Severe cases of poi-soning can result in the same neurologicsigns and symptoms associated with methylmercury poisoning, for example, constrictionof the visual fields. Ethyl mercury poisoningwas characterized by a latent period of sev-eral weeks between first exposure and onsetof the first symptom of poisoning, as hasbeen observed for methyl mercury. In dis-tinction from methyl mercury, signs of renaldamage are found in severe cases.

A detailed review of case histories onexposure to ethyl mercury includingthimerosal allowed Magos (68) to constructa table comparing blood levels at the time ofonset of symptoms. To estimate such bloodlevels from samples collected at a later date,he assumed a half-time in blood of 50 days.Severe intoxication was associated with

blood levels in excess of 2,000 µg Hg/L,with milder intoxication at 1,000 µg Hg/L.Five cases with blood levels of 140–650 µgHg/L had no reported adverse effects. Only18 cases were involved, with ages rangingfrom infants to 79 years. Because of thesmall number of individuals, no statisticalevaluation is possible in terms ofdose–response relationships. However, thedata suggest that ethyl mercury is somewhatless potent in producing neurologic signsand symptoms than methyl mercury, wherethe threshold for neurologic effects has beenestimated at about 200 µg Hg/L (18).

Allergic reactions. Allergic response, usu-ally by skin application, is well known tooccur from organomercurial compounds,including thimerosal (69). Santucci et al.(70) have demonstrated that contact allergyto thimerosal is due to the ethyl mercury rad-ical and that it is indistinguishable in its aller-gic action from methyl mercury. Goncalo etal. (71) also noted that allergy to thimerosalwas mainly related to the mercurial compo-nent, but some allergic reactions may be dueto the thiosalicylic acid component.

Allergy to thimerosal and related mer-cury compounds is a rare event. There is evi-dence that individuals with certainpolymorphisms in glutathione transferasegenes may be susceptible to allergic reactionsto thimerosal (72). Glutathione is necessaryfor the biliary excretion of methyl and inor-ganic mercury (20), and intracellular glu-tathione is protective against the toxicity ofmethyl mercury (29).

Other effects. Only one case of acrodyniahas been reported from exposure tothimerosal (59). This occurred in a 20-year-old man receiving regular gamma globulininfusions containing thimerosal as a pre-servative. The total dose was estimated as40–50 mg Hg.

Acrodynia is now a rare disease. It waswell known to pediatricians when childrenwere exposed to mercury compounds inteething powders, vermifuge preparations,and diaper disinfectants (73). The childrencharacteristically have pink hands and feet(hence the alternative name “pink’s dis-ease”). They are photophobic and sufferfrom joint pains. A typical picture is that ofa child with head buried in a pillow and con-tinually crying. The distraught parentsinvariably take the child for medical atten-tion. Thus, it is unlikely that cases of acrody-nia would escape attention. It is interestingthat not a single case of acrodynia has beenreported from exposure to vaccines despitethe propensity of thimerosal to produce thissyndrome when given in sufficient amounts.

An important characteristic of the dis-ease is that only 1 child in 500 exposed chil-dren develops this disease. The reason for



this susceptibility is not known but presum-ably has a genetic basis. It has made theidentification of mercury as the causal agentvery difficult (74).

Although it is unlikely that thimerosal invaccines has ever caused acrodynia, the clini-cal history of this disease gave rise to the ideathat a genetic susceptibility to mercury mayunderlie other rare childhood diseases. Forexample, Bernard et al. (75) claim thatautism is a “novel form of mercury poison-ing” that occurs in rare infants who aregenetically susceptible.

In attempts to estimate health risks fromthimerosal in vaccines, a key gap in ourknowledge on the human toxicology of mer-cury has become apparent. Little is knownabout the tissue disposition and toxicity ofmercury in human infants, or in animals forthat matter.

Current health risks from thimerosal invaccines depend on the assumption thatethyl mercury is equally toxic to the nervoussystem as methyl mercury.

For example, Ball et al. (7) quote the U.S.EPA reference dose as a guide for safe intakeof ethyl mercury. However, this referencedose is based on data from prenatal exposuresto methyl mercury. In the case of vaccines, weare dealing with postnatal exposures. The pre-natal stage is believed to be the window ofhighest susceptibility to methyl mercury (18).Also, evidence reviewed above suggests thatthe systemic toxicity of thimerosal is less thanthat of methyl mercury compounds.

Dental Amalgam and Mercury Vapor

History of Human Exposure

Dental amalgam. Dental amalgam was intro-duced more than 150 years ago as a tooth fill-ing restoration. Today it is still the mostpopular restorative despite the introductionof new types of fillings. It is an amalgam ofseveral metals, but mercury is the principalcomponent, usually accounting for about50% by weight. Other metals include silverand copper. Periodically throughout the his-tory of dental amalgam, concern has beenexpressed about health risks because of thehigh content of mercury. These recurrentconcerns have sometimes been referred to asthe “amalgam wars,” reflecting the argumentsbetween the proponents and opponents ofits use. Today we are in the third amalgamwar, which started in the early 1970s andcontinues today unabated.

This present war was started by reportsthat amalgams released mercury vapor thatcould be inhaled. Concentrations of mercuryvapor in the air in the oral cavity were shownto exceed occupational health standards.This finding provoked further investigations

and a series of reviews of potential healthrisks from amalgam [e.g., World HealthOrganization (76)]. It was soon realized thatcomparison with occupational health stan-dards gave misleadingly high estimates ofhealth risks. The concentration of mercuryvapor in the oral cavity could indeed reachoccupational health danger levels, but thequantity of vapor was small because the vol-ume of the cavity was small. Eventuallymore meaningful data have been obtainedindicating that the retained vapor is muchless than that inhaled under conditions ofoccupational exposures, except for an inter-esting exception to be discussed below.

Levels of mercury vapor in the ambientatmosphere are so low that intake from thissource is negligible. Thus, with the excep-tion of certain occupational exposures, den-tal amalgam is the main source of humanexposure to mercury vapor. As discussed fur-ther below, other forms of mercury releasedfrom amalgam do not appear to be impor-tant. Thus, a consideration of health risksfrom amalgam depends on our knowledge ofthe toxicology of inhaled mercury vapor andthe quantities released and inhaled fromamalgam restorations.

Mercury vapor. An important sourcebook on this topic is Leonard Goldwater’sMercury: A History of Quicksilver (77).Ramazinni’s Diseases of Workers (78), one ofthe first books on occupational disease, con-tains fascinating historical details of occupa-tional exposures to this metal, as doesDonald Hunter’s masterpiece Diseases ofOccupations, last printed some 30 years ago(8). The most important recent source booksare by the World Health Organization (76),ATSDR (3), and the U.S. EPA (1,2).

Mercury vapor is a monatomic gas thatevaporates from liquid metallic mercury or isproduced by chemical or physical processesfrom chemical compounds of mercury. Theprincipal ore is cinnabar, a brilliant crimsoncrystalline form of mercuric sulfide. Thelargest and oldest mine is located in Almaden,Spain, which has production records datingback many centuries. These records showspurts in production as new uses of mercurywere discovered. Perhaps its oldest applicationwas in the form of cinnabar first used by theChinese to make red ink for official docu-ments many centuries before the modern era.Thus, mercury has the dubious distinction asthe founder of bureaucracy.

In its liquid metallic form, mercury hasfound innumerable applications. Spread as athin film over a sheet of glass, mercury makesan excellent reflecting surface. An island inthe vicinity of Venice, Italy, is famous for itsmirror makers dating back to the middleages. Ramazinni (78) describes the “mirrormakers of Venice” in these terms:

At Venice on the Island called Murano wherehuge mirrors are made, you may see these work-ers gazing with reluctance and scowling at thereflection of their own sufferings in their mirrorsand cursing the trade they have adopted.

Besides mirror making and gilding, it wasalso used in the extraction of gold and silver.Enormous quantities were shipped for extrac-tion of gold and silver from Almaden, as wellas from mines in Peru during the Spanishoccupation of Central and South America.Today, mercury continues to be used in thelargest gold rush of the twentieth century inthe Amazon basin (13). Exposure to mercuryvapor and contamination of local fish alsoappears to be occurring at other gold miningoperations around the world (79–81).

Liquid mercury has found importantapplications in scientific instruments andmeasuring devices. It has found its way intomany homes in thermostats, barometers, andthermometers. It has been contained inhousehold gas regulators. Recent attempts bypower companies to replace such meters hasled to spills in the homes because of the care-less nature by which the meters wereremoved. As many as 200,000 homes in theChicago, Illinois, area may have been conta-minated in this way (82).

Perhaps the earliest medical applicationmay have been in ancient Egypt, where mer-cury compounds in ointments were used totreat skin infections. The skin sores fromsyphilis may have prompted the early appli-cation of mercury to combat this disease as itswept across Europe soon after the return ofChristopher Columbus. Treatment includednot only the application of mercury com-pounds but also the exposure of the person’sskin surfaces to mercury vapor. Paracelsuswas one of the first advocates for the mer-cury treatment, which included skin expo-sure to the vapor. He soon realized, however,that a little too much mercury might kill thepatient, hence his famous dictum “Dosemakes the poison.” So it may be argued thatmercury played a key role in establishing thebasic guiding principle in modern toxicologyand risk assessment.

Thus, the signs and symptoms of poison-ing from inhalation of mercury vapor, atleast in its severe form, have been known forcenturies if not millennia. Severe damage tothe brain, kidneys, and lungs may result,depending on the length and intensity ofexposure. As discussed below, today’s con-cerns are with subtle changes in brain andkidney function associated with occupationalexposure and possibly with amalgam undercertain circumstances. Speculations havebeen put forward that inhalation of mercuryvapor from amalgam may be a causative fac-tor in chronic degenerative diseases of thebrain such as Alzheimer’s disease.



Disposition in the Body

Mercury vapor. Several recent reviews havediscussed in detail the uptake, distribution,excretion, metabolism, and kinetics ofinhaled mercury vapor (1,3,76). A briefsummary is presented here with an updatefrom recent reports.

About 80% of inhaled mercury vapor isretained in the body. However, approxi-mately 7–14% is exhaled within a week afterexposure. The half-time of the process isabout 2 days. The dissolved vapor accumu-lates in red blood cells and is carried to all tis-sues in the body. It crosses the blood–brainand placental barriers. The half-time of dis-tribution to the plasma compartment isapproximately 5 hr (83). The amount of timeto reach a peak value is 9 hr, with a range of7–24 hr in nine adult subjects. The amountof mercury in plasma at the time of the peakconcentration was 4% of the inhaled dose(95% confidence limit, 3–5%).

Approximately 7% is deposited in thecranial region after a single exposure to non-toxic levels of the vapor. The kidney is themain depository.

Once the vapor has entered the cell, it issubject to oxidation to divalent inorganicmercury. The oxidation step is catalyzedspecifically by the enzyme catalase, withendogenously produced hydrogen peroxideas the other substrate. The process is inhib-ited by ethanol. As a result, workers imbib-ing a moderate amount of an alcoholic drinkretain less of the inhaled vapor. The findingthat the half-time for exhalation from thelung is about 2 days suggests that the half-time for the oxidation in body tissues isabout the same.

Studies with radioactive tracers indicatethat the rate of overall excretion of mercuryfrom the body can be described by a singlehalf-time of about 58 days, corresponding toan excretion rate of slightly more than 1% ofthe body burden per day. Most tissues havethe same or shorter half-times.

The decline in plasma levels of mercuryconsists of at least two components: a shorthalf-time of less than 1 day and a longer oneof about 10 days. Blood levels thereforereflect recent exposure.

Excretion takes place via both urine andfeces. Urinary mercury originates mainlyfrom mercury in kidney tissue. Urine is thecommonly used biologic marker, as itreflects the cumulative dose to one of themain target organs, the kidney. The relation-ship between urinary excretion and levels inthe other target tissue, the nervous system, isnot well established. As discussed below, uri-nary mercury levels have been found to showa rough correlation with signs and symptomsof damage to the nervous system.

Dental amalgam. Several studies over thepast 30 years or so have demonstrated thatamalgam filling releases mercury vapor intothe oral cavity. Mouth breathing carries thevapor to the lung, where it is absorbed anddistributed to tissues, as discussed above.Mercury levels in autopsy tissue samples,including the brain, have been shown to cor-relate with the total number of surfaces ofamalgam restorations. The estimate for therate of release in people with amalgamrestoration is 2–17 µg Hg/day (18). Themost recent estimate based on applying phar-macokinetic parameters to steady-stateplasma levels in people with amalgam sug-gests an average intake between 5 and 9 µgHg/day (83). Kingman et al. (84), in a studycorrelating urinary excretion of mercury withamalgam surfaces, estimated that 10 amal-gam surfaces would raise urinary levels by 1µg Hg/L. As discussed below, these are farbelow toxic levels. However, excessive chew-ing, such as occurs when smokers try to stopsmoking by using nicotine-containing chew-ing gum, may lead to urine levels in excess of20 µg Hg/g creatinine, thereby approachingoccupational health safe limits (85).

Increased amounts of mercury areexcreted in feces in individuals with amalgamfillings. Engqvist et al. (86) found that only25% of the total mercury in fecal sampleswas in the form of amalgam particles in sam-ples taken from six adults with a moderateload of amalgam fillings. About 80% of anoral dose of amalgam particles or mercuricmercury attached to sulfhydryl groups wasexcreted in the feces. Interestingly, 60% of anoral dose of vapor dissolved in water wasretained. Previously it had been assumed thatintake of vapor was due solely to inhalation.

Adverse EffectsMercury vapor. ACUTE TOXICITY. Cases con-tinue to occur of severe poisoning and evenfatalities from acute exposure to high levelsof mercury vapor [e.g., see Solis et al. (87)].Severe lung damage can lead to death fromhypoxia. The poisoning appears to occur inthree phases. The initial phase is character-ized by flulike symptoms lasting 1–3 days.The intermediate phase is dominated bysigns and symptoms of severe pulmonarytoxicity. The victim in the final phase willexperience gingivostomatitis, tremor, anderethism (memory loss, emotional lability,depression, insomnia, and shyness).

The signs and symptoms of the finalphase are identical to those seen in workerschronically exposed to mercury levels.Generally speaking, such cases are rarely seen,at least in developed countries, where indus-trial hygiene measures are strictly enforced.

THE NERVOUS SYSTEM. Today, healthconcerns are directed toward the risk from

lower levels of exposures. In general, air con-centrations above 50 µg Hg/m3 in the work-place, corresponding to steady-state urinaryexcretion rates of 60 µg Hg/g creatinine, areassociated with fine tremors in the extremi-ties that frequently are not noticed by theworker (76). Slowed nerve conduction veloc-ity is another preclinical effect found at theselower levels.

Studies on dentists have suggestedadverse effects at air concentrations lowerthat 50 µg Hg/m3 [for review, see Langworthet al. (88)]. Average air concentrations as lowas 14 µg Hg/m3 were associated withdecreased performance on psychomotor tests.Changes in mood and behavior have alsobeen noted, such as emotional lability,somatosensory irritation, and alterations inmood scores. As noted by Langworth et al.(88), such effects may be due to mercuryexposure. An alternative explanation for theobserved correlations is that “dentists withspecial personality traits are less careful in thehandling of mercury spills etc. and thus aremore exposed to mercury vapor.” If indeedthese effects result from exposure to mercury,one should bear in mind that the average lev-els reported in these studies could be substan-tially less than peak values that may occurduring installation of the amalgam fillings.

Follow-up studies of workers exposed tohigh levels of mercury vapor and no longerexposed during 10 or more years before beingexamined have revealed that adverse effectsmay persist on the nervous system.Mathiesen et al. (89) examined 70 previouslyexposed workers (time from last exposure, 1to 35 years, average 12.7 years). The averageyearly exposure was 8–584 µg Hg/m3. Peakexposures during any specific year could havebeen much higher than these average levels.Decreased performance on a number of neu-ropsychologic tests was found, comparedwith a control group of 52 workers. Despitethese high exposure levels, no residual effectswere observed on general intellectual abilityor ability to reason logically.

Workers exposed to high levels of vapor atsome time during 1953–1966 in a nuclearweapons facility have been the subject of fol-low-up studies (90,91). Columns of liquidmercury were used in the separation oflithium isotopes. The exposure was expressedas “cumulative average quarterly urine mer-cury measurements” in units of microgramsof mercury per liter, from which informationone cannot determine the actual urinaryexcretion rate (90). However, according tocomments in the text of this paper (90), mer-cury workers had urine levels in excess of 600µg Hg/L. In the more recent study (91), 104of the surviving workers were compared withan unexposed group of 201. Residual adverseeffects were found primarily on the peripheral



nervous system. Such long-term adverseeffects, as quoted from the authors, “were notobserved for a measure of dementia or othermeasures of cognitive function.”

As discussed in the following section onamalgam, a suggestion has been made thatinhaled vapor originating from amalgam fil-ings is a cause or predisposing factor toAlzheimer’s disease (92). The fact that boththese studies (90,91) were unable to detectany signs or symptoms remotely related tothis disease years after heavy exposure tomercury vapor argues strongly against thissuggestion. Many of these workers wereexposed for many years to intakes of vapormore than 100-fold higher than that experi-enced from amalgam fillings.

No new information on adverse effectsfrom prenatal exposure has emerged sinceprevious reviews. However, one study (93)reported that female squirrel monkeysexposed during their pregnancy to air con-centations of 500–1,000 µg Hg/m3 hadblood levels ranging from 25 to 180 µgHg/L. No difference was observed betweenexposed and nonexposed offspring in variousschedules of reinforcement in terms of leverpressing and other behavioral measures. Theexposed offspring, however, appeared to varymore in the test performance. Given thehigh levels of prenatal exposure and the min-imal effects found in the offspring, thesedata would suggest that the prenatal periodmay not be especially sensitive to the effectsof vapor inhaled by the mother. This is con-sistent with what is known of the dispositionof inhaled vapor in the maternal–fetal unit.Although vapor passes across the placenta,much less accumulates in the fetal brain thanin that of the mother. The fetal liver appearscapable of oxidizing the vapor in its first passthrough this organ. The product of oxida-tion, divalent inorganic mercury, passes theblood–brain barrier far more slowly thanthe vapor.

KIDNEYS. Distinct from the action ofinorganic mercuric compounds, exposure tomercury vapor does not produce severe kid-ney damage. However, low-level chronicexposures at air concentrations above 50 gHg/m3 do have adverse effects on the kidney(94). Decreased selectivity of the glomerularfilter is evidenced by increased excretion ofalbumin. Tubular reabsorptive function isslightly diminished, leading to increasedexcretion of low-molecular-weight proteinssuch as retinol-binding protein. Damage tothe brush border of the tubular cells is indi-cated by increased urinary excretion of brushborder antigens. Interstitial effects of mer-cury result in loss of prostaglandins into theurine. These biochemical markers detecteffects of mercury well before kidney functionis significantly compromised.

Taylor et al. (94) reviewed results from awide variety of urinary markers. The resultssuggest that mercury, lead, and cadmiummay produce different patterns of changes inthese markers. The most sensitive tests forthe action of mercury are the tubular andinterstitial markers.

MECHANISMS OF TOXICITY. The mecha-nism of action of inhaled mercury vapor onbrain function is not known. It is assumedthat the vapor is first oxidized to inorganicdivalent mercury that functions as the proxi-mate toxic agent. The latter can attach tothiol groups present in most proteins. Thus,almost any enzyme or structural protein is apotential target.

As discussed previously, it appears thatthe intact mercurial and not its metabolicproduct, inorganic mercury, is the proximatetoxic agent in the neurotoxic action ofmethyl mercury. Conversely, we assumedivalent inorganic mercury is the proximatetoxic agent after exposure to mercury vapor.The underlying reason for this apparent con-flict is not known. Most likely it is because ofthe differences in transport and distributionwithin the brain. Methyl mercury is trans-ported as a water-soluble complex that ismetabolized slowly to inorganic mercuryonly in phagocytic cells not in neuronal cells.Mercury vapor diffuses to all parts of thebrain as a lipid-soluble monatomic gas that israpidly oxidized to inorganic mercury by thecatalase–hydrogen peroxide pathway presentin all cells.

Pendergrass et al. (92) have presented evi-dence that inhaled vapor may damage themicrotubular system in brain cells in a man-ner somewhat similar to that seen for methylmercury. They reported that inhaled vaporcan inhibit the binding of guanosine triphos-phate (GTP) to a β subunit of the tubulindimer. The microtubules of neuronal andother cells are formed from the polymeriza-tion of tubulin protein subunits in a tread-milling process such that as one end of themicrotubules is formed, the other end isbeing depolymerized (95). GTP binding isessential for the polymerization step. Thus, ifthe formation step is inhibited, the micro-tubule will disappear as the depolymerizationcontinues. Microtubules are key cytoskeletalstructures involved in axonal transport, celldivision, and cell migration. It will be inter-esting to see if exposure to mercury vaporleads to disappearance of the microtubules, ashas been demonstrated for methyl mercury.

Consistent with action on microtubularstructures, Leong et al. (96) observed thatmercuric ions added in vitro to cultured neu-rons inhibited outgrowth and disrupted-membrane structure. Tests with antibodiesfor tubulin and actin indicated that themicrotubular structure had disintegrated.

The inhibitory action of methyl mercuryon the assembly of microtubules is well doc-umented. If further investigation shows thatthe microtubule assembly is a common bio-chemical target for both forms of mercury,then we face the problem of explainingwhy the pathology and clinical signs andsymptoms differ so much.

Understanding the mechanisms of cellu-lar defenses is just as important as under-standing the mechanisms of damage.Thiol-containing molecules probably play arole in defense as well as being targets fortoxicity. Glutathione complexes with inor-ganic mercury in liver cells are secreted inbile and ultimately in the feces. Intracellularlevels of glutathione probably divert mercuryfrom sensitive sites. The thiol-rich family ofproteins known generically as metalloth-ioneins also plays a protective role. Forexample, metallothionein has been shown toprotect against kidney damage from inor-ganic mercury (97). More recently, it wasshown that lung damage was more severe inmetallothionein-null mice than in normalmice after exposure to mercury vapor (98).

Amalgam. Contact hypersensitivity tomercury is a well-established adverse effect ofamalgam fillings [e.g., see Camisa et al. (99)].According to these authors, a completeremission may be expected about 3 monthsafter the last amalgam filling is removed.

The existence of other adverse healtheffects due to amalgam is presently unknownbut is becoming an area of intensive specula-tion and controversy. This is partly becauseof the limited amount of research on thesafety of amalgam fillings and partly becauseof the increased visibility of mercury as ahealth risk and stringent regulatory actionsconcerning this metal.

Ahlqwist et al. (100) reported on the lat-est findings of a long-standing study of acohort of 1,462 Swedish women establishedin 1968–1969. Follow-up studies were con-ducted in 1974 and 1975, 1980 and 1981,and in 1992 and 1993. Serum mercury lev-els correlated with the number of amalgamfillings. Different clusters of symptoms wererecorded as well as the incidence of diabetes,myocardial infarction, stroke, and cancer.No association could be found betweenserum mercury levels and disease in this pop-ulation of middle-aged and older women.

The finding that dental amalgam doesnot affect mental health is from two well-conducted epidemiologic studies—one ontwins in Sweden (101) and the other onolder women, the so-called Nun Study inthe United States (102). The Swedish studyinvolved approximately 587 subjects from anon-going Swedish Adoption/Twin Study ofAging (103). The twin study allowed controlfor genetic predisposition to the toxic effects



of mercury when evaluating the role of amal-gam fillings. No negative effects on physicalor mental health were found. The mean ageof the study group was 66 years.

The study on 129 Catholic nuns aged75–102 years took advantage of a populationwith homogeneous adult life styles and envi-ronment. No effect of amalgam status(determined by the number and surface areaof the occlusal surfaces) could be found oneight different tests of cognitive function.

Cederbrant et al. (104) attempted toaddress the possibility that a susceptibleimmune system might explain why someindividuals with amalgam fillings claimed tohave psychologic, sensory, or neurologicsymptoms from exposure to mercury. Theyused an in vitro lymphocyte proliferationassay to test for immune sensitivity to inor-ganic mercury on 23 amalgam patients, 30healthy blood donors with amalgam, 10healthy subjects without amalgam, and 9patients with oral lichen planus (OLP) adja-cent to the amalgam. In addition to the lym-phocyte proliferation assay, a wide range ofimmune parameters was measured. None ofthese end points revealed any significant dif-ference between amalgam patients and con-trols despite the fact the in vitro assay wassensitive to the positive control group (OLP).

Because the inhaled mercury vapor istoxic to the central nervous system,researchers are now speculating that vaporfrom amalgams may be a cause of or an exac-erbating factor in some well-known degener-ative diseases such as amyotrophic lateralsclerosis, Alzheimer’s disease (AD), multiplesclerosis, and Parkinson’s disease. Speculationhas been most intense concerning AD after areport that mercury levels were higher inautopsy brains of AD patients than in brainsof members of a control group (105).

However, subsequent reports have pre-sented an equivocal picture concerning cor-relations between tissue levels of mercuryand AD. Fung et al. (106) found no differ-ence between blood mercury levels or mer-cury to selenium ratios in AD patients andcontrols. All subjects resided in a nursinghome, thus ensuring that environmental anddietary exposures were similar. Conversely,Hock et al. (107) found that blood levels inAD patients were higher than in controls. Inearly-onset AD patients, blood levels were 3times higher than in controls. Blood mer-cury correlated with concentrations of amy-loid β peptide on the cerebrospinal fluid in asubset of these patients. Interestingly, theincreases in blood mercury levels were unre-lated to the status of dental amalgam. Thereason for the difference in the outcome ofthe two studies is not clear, although thestudy by Fung et al. (106) may have bettercontrol over exposure to mercury.

Subsequent studies on brain and relatedtissues have also discounted a connectionbetween mercury levels and AD. Fung et al.(108) found mercury levels were the same invarious anatomical regions of the brain inAD patients and matched controls. Cornettet al. (109) found elevated brain levels inmost regions of the brain that were mea-sured, but no statistical difference could beestablished from corresponding mercury lev-els in control subjects. Mercury levels inpituitary glands of AD patients were foundto be similar to those of controls (110). In astudy of 56 AD patients and 21 controls,Saxe et al. (111) found no significant associ-ation of AD with number, surface area, orhistory of having dental amalgam restora-tions. Mercury levels in the brain were thesame in AD and control patients.

Overall studies relating tissue levels ofmercury to AD have not produced a con-vincing picture of any kind of correlationwith this disease. Even if one were estab-lished, the “chicken and the egg” issuewould arise: Is mercury the cause of AD ordoes AD tissue accumulate mercury morethan normal tissue?

Nevertheless, biochemical studies ofin vitro preparations of nerve cells and of ADtissue continue to raise a question at least ona mechanistic basis that the levels of mercurymay in some way be connected with AD.The brain pathology of AD is characterizedby plaques of amyloid protein and neurofib-rillary tangles (112). The latter consist ofaltered microtubules and microtubule-associated proteins, especially tau (113) andare needed for assembly of microtubules fromthe tubulin monomers. Phosphorylation ofthis protein blocks its ability to promotemicrotubule assembly. Mercury can interferewith the complex process of the treadmillingof microtubules.

The study by Leong et al. (96) on theeffect of mercury on neurite growth alsonoted the appearance of structures resem-bling neurofibrillary tangles. The study byPendergrass et al. (92), noted that mercurycan block the binding of GTP to tubulin,thus interfering with microtubule assembly.Other studies have indicated that mercurycan cause hyperphosphorylation of the tauprotein (114).

Oxidative stress has been invoked as acause of AD (115,116). Mercury is wellknown to cause biochemical changes in cellsis consistent with oxidative stress (114).Indeed, it has been argued that the sameenzyme system that protects against oxygenattack also protects against mercury (15).

Such biochemical observations offer tan-talizing possibilities that mercury can beinvolved in a mechanism of AD. The processof microtubular treadmilling is controlled

largely by thiol-containing proteins. Perhapsmercury is simply acting as a thiol reagentand any other thiol-reactive chemical wouldproduce the same effects. For example, thelipid peroxidation product 4-hydroxynone-nal inhibits neurite outgrowth, disrupts neu-ronal microtubules, and modifies cellulartubulin. Is this also acting by oxidizing thiolgroups? As yet we do not have a completeplausible biochemical mechanism for thegenesis of AD, nor do we know how mer-cury interferes with this process in vitro orwhether or not mercury acts in vivo.

Conclusions and Research NeedsThe three modern faces of mercury—methylmercury in fish, mercury vapor from amal-gam tooth fillings, and ethyl mercury in vac-cines—represent our most recent encounterwith this ancient metal. Despite thousands ofyears of history of human exposure andintense research activity in our lifetime, manyof its toxic actions remain unexplained. Thisreview reveals key gaps in our knowledge,gaps that highlight important research needs.

The main features of the disposition ofmethyl mercury in the body are well known.Nevertheless, some key gaps remain both inpharmacokinetics and in the mechanisms oftransport and metabolism.

Fecal excretion is the main pathway ofexcretion in adults. Animal data indicatethat this process does not start until the endof the suckling period. However, we have asyet no confirmation in human infants. Thus,we are unable to estimate the cumulativebody burden from methyl mercury knownto be secreted in breast milk. This gap in ourknowledge is especially critical for risk esti-mates from thimerosal in vaccines.

Demethylation of methyl mercury bymicroflora in the gut is a key, probably rate-determining, process in the removal ofmethyl mercury from the body. Themicrobes involved have not been identifiednor have the biochemical mechanisms ofcleavage of the carbon–mercury bond. Thedemethylation process in the gut might wellconstitute an important site for interactionbetween diet and methyl mercury accumula-tion in the body. The fiber content of thediet has already been shown to affect theexcretion rate of mercury (117). The dietchange at the time of weaning may alsoaffect the activity and composition of themicroflora. Further studies in this area mightshed light on why there is such a broad rangeof biologic half-times reported for adultsexposed to methyl mercury.

Molecular mechanisms of transport ofmercury across cell membranes have beenidentified, indicating that specific thiol com-plexes of methyl mercury can enter cells via



the large neutral amino acid carrier and exiton carriers for glutathione. However, nostudies have reported to date on how methylmercury gains entry into the hair follicle andthen concentrates over a hundredfold com-pared to its concentration in whole blood.This is an important research priority, ashead hair is the most widely used biologicindicator for this form of mercury. If thesame entry mechanism operates for hair fol-licular cells as has been shown for endothe-lial cells of the blood–brain barrier, thenmercury in hair would represent the speciesof mercury in blood that enters the brain.This would explain why levels in hair havebeen shown to parallel levels in brain.

The long latent period between the endof exposure and the sudden appearance ofsymptoms and signs of neurologic damage isboth a fascinating and an insidious propertyof the action of methyl mercury on themature central nervous system. A slow releaseof inorganic mercury might explain thisproperty if the inorganic form were the prox-imate toxic species. However, animal experi-ments indicate that this role is played by theintact organomercurial moiety. Because thelength of the latent period appears to beindependent of the dose, it is also intriguingand argues against the accumulation of atoxic metabolite. It we could determine themechanisms underlying the latent period, wewould learn much more about the toxicaction of the “element of mystery.”

Two studies have indicted the possibilityof adverse effects on the cardiovascular sys-tem both in adults and in prenatallyexposed children. Such effects appear to beoccurring at methyl mercury levels in thebody comparable to those associated withthe lowest levels affecting the central ner-vous system. There is an urgent need toconfirm these findings in other populations,preferably where no co-exposure is occur-ring to other persistent organic pollutantssuch as polychlorinated biphenyls.

Generally, a broad research agenda isneeded to develop the toxicology ofthimerosal, given the paucity of our currentinformation. Studies should be directed to testthe assumption that the toxicology ofthimerosal is similar to that of methyl mer-cury, given the fact the current estimates ofhuman health risks, in particular in infantsreceiving vaccines, are based on this assump-tion. The immediate tissue disposition of mer-cury following a dose of thimerosal appears tobe both qualitatively and quantitatively similarto that of methyl mercury, as discussed in thisreview. However, such limited evidence asnow exists suggests that the rate of conversionto inorganic and, subsequently, the rate ofexcretion are more rapid, perhaps substantiallyso, compared with methyl mercury. Data on

the biologic half-time of the ethyl mercuryradical in body tissues, especially the brain, areessential for estimates of tissues burdens andhealth risk from cumulative exposure fromrepeated doses of thimerosal in vaccines givento infants. Such information needs to be gath-ered both during and after the suckling period.

Thimerosal also differs from methyl mer-cury in that it causes kidney damage at aboutthe same doses that damage the nervous sys-tem. Experimental evidence indicated thatdamage to the nervous system is caused bythe intact organomercurial radical, whethermethyl or ethyl. However, inorganic mer-cury released from ethyl mercury may be theproximate toxic agent for kidney damage.Indeed, the suspected greater rate of releasefrom ethyl mercury may explain why kidneydamage, if any, occurs only at the later stagesof intoxication from methyl mercury. Thus,comparative tests of methyl and ethyl mer-cury should include the renal–cardiovascularsystem as well as the nervous system indeveloping animals.

It is almost 30 years ago that mercuryvapor was shown to be emitted from dentalamalgam fillings. This led to an outpouringof numerous articles attempting to measurethe precise amounts of vapor released and offactors affecting the release rate. In general,levels of inorganic mercury in tissue causedby release of vapor from amalgam are wellbelow those associated with overt toxic effectsor even with subtler neurobehavioral andrenal affects. However, excessive chewing canraise urine levels close to the lowest safetylimits for occupational exposure to mercuryvapor. Interest is now focused on possibleindirect effects of vapor released from amal-gam. Despite the fact that several well-con-ducted epidemiologic studies have indicatedno relationship between dental amalgam andAlzheimer’s disease, speculation continuesthat the small amounts of vapor inhaled fromamalgam may in some as yet unknown wayexacerbate the progress and severity of thisdisease. Biochemical studies reviewed in thisarticle raise intriguing possibilities.

Mechanisms of cellular resistance towardand defense against these three faces of mer-cury have received some research attention. Itis suggested that the focal lesions produced inthe adult brain by methyl mercury are theresult not of selective toxic action but ofselective resistance. Those cells having inade-quate defense mechanisms succumb to theinitial insult. It is likely that intracellular glu-tathione plays a protective role both indeflecting methyl mercury from sensitive sitesin the cell and by enhancing its exit from thecell. Other thiol compounds, such as themetallothioneins, may also play a defensiverole for both inorganic and organic forms ofmercury (118). More detailed biochemical

information of these defense processes shouldlead to the identification of genes controllingcellular resistance and thereby give somegenetic insight into host susceptibility.

As we gaze at these three modern faces ofmercury and reflect upon the extensiveresearch conducted in our lifetime, we mustreluctantly agree with the title of a BBC docu-mentary broadcast over 25 years ago that thismetal still remains “an element of mystery.”

REFERENCES AND NOTES

1. U.S. EPA. Mercury Report to Congress Office of Air Qualityand Standards. Washington, DC:U.S. EnvironmentalProtection Agency, 1997.

2. U.S. EPA. Water Quality Criterion for the Protection ofHuman Health: Methyl Mercury. EPA 0823-R-01-001.Washington, DC:U.S. Environmental Protection Agency,2001.

3. ATSDR. Toxicological Profile for Mercury. Atlanta,GA:Agency for Toxic Substances and Disease Registry,1999.

4. National Research Council. Toxicological Effects ofMethylmercury. Washington, DC:National AcademyPress, 2000.

5. AAP and USPHS. Joint Statement of the AmericanAcademy of Pediatrics (AAP) and the United StatesPublic Health Service (USPHS). Pediatrics 104:568–569(1999).

6. Pless R, Risher JF. Mercury, infant neurodevelopment,and vaccination. J Pediatr 136:571–573 (2000).

7. Ball LK, Ball R, Pratt RD. An assessment of thimerosaluse in childhood vaccines. Pediatrics 107:1147–1154(2001).

8. Hunter D. Diseases of Occupations. Boston:Little Brown ,1969;314–328.

9. WHO. Environmental Health Criteria 1. Mercury.International Program on Chemical Safety. Geneva:World Health Organization, 1976.

10. Bakir F, Damluji SF, Amin-Zaki L, Murtadha M, Khalidi A,al-Rawi NY, Tikriti S, Dhahir HI, Clarkson TW, Smith JC,et al. Methylmercury poisoning in Iraq. Science181:230–241 (1973).

11. Swedish Expert Group. Methylmercury in fish. A toxico-logical epidemiological evaluation of risks. Nord HygTidskr 4(suppl):19–364 (1971).

12. Cleary D. Anatomy of the Amazon Gold Rush. Iowa City,IA:University of Iowa Press, 1990.

13. Pfeiffer WC, Lacerda LD. Mercury inputs in the Amazonregion. Environ Technol Lett 9:325–330 (1988).

14. Fitzgerald WF, Clarkson TW. Mercury and monomethylmercury: present and future concerns. Environ HealthPerspect 96:159–166 (1991).

15. Clarkson TW. Health effects of metals: a role of evolu-tion. Environ Health Perspect 103(suppl 1):9–12 (1995).

16. Cernichiari E, Brewer R, Myers GJ, Marsh DO, LaphamLW, Cox C, Shamlaye CF, Berlin M, Davidson PW,Clarkson TW. Monitoring methylmercury during preg-nancy: maternal hair predicts fetal brain exposure.Neurotoxicology 16(4):705–710 (1995).

17. Amin-Zaki L, Elhassani S, Majeed MA, Clarkson TW,Doherty RA, Greenwood M. Intra-uterine methyl mercurypoisoning in Iraq. Pediatrics 54(5):587–595 (1974).

18. WHO. Environmental Health Criteria 101: Methylmercury.International Program on Chemical Safety.Geneva:World Health Organization, 1990.

19. Kerper LE, Ballatori N, Clarkson TW. Methylmercurytransport across the blood-brain barrier by an aminoacid carrier. Am J Physiol 267:R761–R765 (1992).

20. Ballatori N, Clarkson TW. Biliary secretion of glutathioneand glutathione-metal complexes. Fundam Appl Toxicol5:816–831 (1985).

21. Dutczak WJ, Ballatori N. γ-Glutamyl transferase depen-dent biliary-hepatic recycling of methyl mercury in theguinea pig. J Pharmacol Exp Ther 262:619–623 (1992).

22. Dutczak WJ, Ballatori N. Transport of the glutathione-methyl mercury complex across liver canalicular mem-branes on reduced glutathione carriers. J Biol Chem269:9746–9751 (1994).



23. Clarkson TW, Magos L, Cox C, Greenwood MR, MajeedMA, Damluji SF. Tests of efficacy of antidotes forremoval of methylmercury in human poisoning during theIraq outbreak. J Pharmacol Exp Ther 218:74–83 (1981).

24. Ballatori N, Lieberman NMW, Wang W.N-Acetylcysteine as an antidote in methylmercury poi-soning. Environ Health Perspect 106:267–271 (1998).

25. Nierenberg DW, Nordgren RE, Chang MB, Siegler W,Blayney MG, Hochberg F, Toribara TY, Cernichiari E,Clarkson T. Delayed cerebellar disease and death afteraccidental exposure to dimethyl mercury. N Engl J Med338:1672–1675 (1998).

26. Syversen T. Effects of methyl mercury on in vivo isolatedcerebral and cerebellar neurons. Neuropathol ApplNeurobiol 3:225–236 (1977).

27. Jacobs JM, Carmichael N, Cavanagh JB. Ultrastructuralchanges in the nervous system of rabbits poisoned withmethyl mercury. Toxicol Appl Pharmacol 39:249–261 (1977).

28. Sarafian TA, Bredesen DE, Verity MA. Cellular resistanceto methyl mecury. Neurotoxicology 17(1):27–36 (1996).

29. Miura K, Ikeda K, Naganuma A, Imura N. Important roleof glutathione in susceptibility of mammalian cells tomethylmecury. In Vitro Technol 7:59–64 (1994).

30. Miura K, Clarkson TW. Reduced methylmecury accumula-tion in a methylmercury-resistant rat pheochromocytomaPC12 cell line. Toxicol Appl Pharmacol 118:39–45 (1993).

31. Ganther HE, Goudie C, Sunde ML, Kopecky MJ, WagnerP, Oh S-H, Hoekstra WG. Selenium: relation to decreasedtoxicity of methylmercury added to diets containing tuna.Science 17:1122–1124 (1972).

32. Magos L, Brown AW, Sparrow S, Bailey E, Snowden RE,Skipp WR. The comparative toxicology of ethyl- andmethyl mercury. Arch Toxicol 57:260–267 (1985).

33. Davis LE, Kornfeld M, Mooney HS, Fiedler KJ, Haaland KY,Orrison WW, Cernichiari E, Clarkson W. Methyl mercurypoisoning: long term clinical, radiological, toxicologicaland pathological studies. Ann Neurol 35:680–688 (1994).

34. Charleston JS, Body RL, Bolender RP, Mottet NK, VahterME, Burbacher TM. Changes in the number of astrocytesand microglia in the thalamus of the monkey Macacafascicularis following long-term subclinical methylmer-cury exposure. Neurotoxicology 17(1):127–138 (1996).

35. Salonen JT, Seppanen K, Nyyssonen K, Korpela H,Kauhanen J, Kantola M, Tuomilehto J, Esterbauer H,Tatzber F, Salonen R. Intake of mercury from fish, lipidperoxidation and risk of myocardial infarction and coro-nary cardiovascular and any death in Eastern Finnishmen. Circulation 91:645–655 (1995).

36. Salonen JT, Seppanen K, Lakka TA, Salonen R, KaplanGA. Mercury accumulation and accelerated progressionof carotid atherosclerosis: a population–based prospec-tive 4-year follow-up study in men in eastern Finland.Atherosclerosis 148(2):265–273 (2000).

37. Bosma H, Peter R, Siegrist J, Marmot M. Two alternativejob stress models and the risk of coronary heart disease.Am J Public Health 88:68–74 (1998).

38. Myers GJ, Davidson PW, Cox C, Shamlaye CF, TannerMA, Choisy O, Sloane-Reeves J, Marsh DO, CernichiariE, Cox A, et al. Neurodevelopmental outcomes ofSeychellois children sixty-six months after in utero expo-sure to methylmercury from a maternal fish diet: pilotstudy. Neurotoxicology 16(4):639–652 (1995).

39. Marsh DO, Clarkson TW, Cox C, Myers GJ, Amin-Zaki L,Al-Tikriti S. Fetal methylmercury poisoning: relationshipbetween concentration in single strands of maternal hairand child effects. Arch Neurol 44:1017–1022 (1987).

40. Cox C. Clarkson TW, Marsh DO, Amin-Zaki L, Al-Tikriti S,Myers G. Dose-response analysis of infants prenatallyexposed to methyl mercury: an application of a singlecompartment model to single-strand hair analysis.Environ Res 49:318–332 (1989).

41. Ernhart CB, Sokol RJ, Matier S, Moron P, Nadler D, AgerJW, Wolf A. Alcohol teratogenicity in the human: adetailed assessment of specificity, critical period andthreshold. Am J Obstet Gynecol 156:33–39 (1987).

42. McKoewn-Eyssen GE, Ruedy J, Neims A. Methylmercuryexposure in northern Quebec. II: Neurological findings inchildren. Am J Epidemiol 118:470–479 (1983).

43. Kjellstrom T, Kennedy P, Wallis S, Stewart A, Friberg L,Lind B, Weatherspoon P, Mantell C. Physical and mentaldevelopment of children with prenatal exposure to mer-cury from fish. Stage 2. Interviews and psychologicaltests at age 6. Report No. 3642. Solna, Sweden:SolnaNational Swedish Environmental Board, 1989.

44. Steuerwald U, Weihe P, Joeensen PJ, Bjerve K, Brock J,Heinzow B, Budtz-Jorgensen E, Grandjean P. Maternalseafood diet, methylmercury exposure, and neonatalneurologic function J Pediatr 136(5):599–605 (2000).

45. Davidson PW, Kost J, Myers GJ, Cox C, Clarkson TW,Shamlaye CF. Methylmercury and neurodevelopment:reanalysis of the Seychelles Child Development Studyoutcomes at 66 months of age. JAMA 285(10):1291–1293(2001).

46. Choi BH, Lapham LW, Amin-Zaki L, Saleem T. Abnormalneuronal migration, deranged cerebellar cortical organi-zation, and diffuse white matter astrocytosis of humanfetal brain. A major effect of methyl mercury poisoning inutero. J Neuropathol Neurol 37:719–733 (1978).

47. Miura K, Imura N. Mechanism of methylmercury cytotox-icity. Crit Rev Toxicol 18(3):161–168 (1987).

48. Rodier PM, Aschner M, Sager PR. Mitotic arest in thedeveloping CNS after prenatal exposure to methyl mer-cury. Neurobehav Toxicol Teratol 6:379–385 (1984).

49. Philbert MA, Billiingsley ML, Reuhl KR. Mechanisms ofinjury in the central nervous system. Toxicol Pathol28(1):43–53 (2000).

50. Sorensen N, Murata K, Budtz-Jorgensen E, Weihe P,Grandjean P. Prenatal methylmercury exposure as a car-diovascular risk factor at seven years of age.Epidemiology 10(4):370–375 (1999).

51. Jalili MA, Abbasi AH. Poisoning by ethyl mercury toluenesulfonanilide. Br J Ind Med 18:303–308 (1961).

52. Al-Damluji S. Mercurial poisoning with the fungicide gra-nosan M. J Faculty Med (Baghdad) 4:82–103 (1962).

53. Zhang J. Clinical observations in ethyl mercury chloridepoisoning Am J Ind Med 5:251–258 (1984).

54. Axton JHM. Six cases of poisoning after a parenteralorganic mercurial compound (Merthiolate). PostgradMed J 48:417–421 (1972).

55. Rohyans J, Walson PD, Wood GA. MacDonald WA.Mercury toxicity following Merthiolate ear irrigations.J Pediatr 104:311–313 (1984).

56. Pfab R, Muckter H, Roider G, Zilker T. Clinical course ofsevere poisoning with thimerosal. Clin Toxicol 34:453–460(1996).

57. Fagan DG, Pritchard JS, Clarkson TW, Greenwood MR.Organ mercury levels in infants with omphalocelestreated with organic mercurial antiseptic. Arch Dis Child52:962–964 (1977).

58. Suzuki T, Takemoto T-I, Kashiwazaki H, Miyama T.Metabolic fate of ethylmercury salts in man and animal.In: Mercury, Mercurials and Mercaptans (Miller MW,Clarkson TW, eds). Springfield, IL:Charles C. Thomas,1973;209–232.

59. Matheson DS, Clarkson TW, Gelfand EW. Mercury toxic-ity (acrodynia) induced by long-term injection of gammaglobulin. J Pediatr 97:153–155 (1980).

60. Carty AJ, Malone SF. The chemistry of mercury in biolog-ical systems. In: The Biogeochemistry of Mercury in theEnvironment (Nriagu JO, ed). New York:Elsevier/NorthHolland, 1979;433–459.

61. Elferink JG. Thimerosal: a versatile sulfhydryl reagent,calcium mobilizer and cell function-modulating agent.Gen Pharmacol 33(1):1–6 (1999).

62. Tan M, Parkin JE. Route of decomposition of thiomerosal(thimerosal). Int J Pharm 208(1–2):23–34 (2000).

63. Ulfvarson U. Distribution and excretion of some mercurycompounds after long term exposure. Int ArchGewerbepath Gewerbehyg 19:412–422 (1962).

64. Blair AMJN, Clark B, Clarke AJ, Wood P. Tissue concen-trations of mercury after chronic dosing of squirrel mon-keys with thimerosal. Toxicology 3:171–176 (1975).

65. Stajich GV, Lopez GP, Harry SW, Sexson WR. Iatrogenicexposure to mercury after hepatitis B vaccination in pre-term infants. J Pediatr 136:679–681 (2000).

66. Pichichero ME, Clarkson T, Lepricato J, Cernichiari E,Treanor J. Blood mercury levels to infants receivingthimerosal-containing vaccines. Abstract 1385. PediatrRes 49(4):243A (2001).

67. Powell HM, Jamieson WA. Merthiolate as a germicideAm J Hyg 13:296–310 (1931).

68. Magos L. Review on the toxicity of ethyl mercury includ-ing its presence as a preservative in biological and phar-maceutical preparations. J Appl Toxicol 21:1–5 (2001).

69. Wantke F, Hemmer W, Jarisch R. Thimerosal inducestoxic reactions. Int Arch Allergy Immunol 105:408–407(1994).

70. Santucci B, Cannistraci C, Cristaudo A, Camera E, Picardo

M. Thimerosal positives: the role of organomercury alkylcompounds. Contact Dermatitis 38:325–328 (1998).

71. Goncalo M, Figueiredo A, Goncalo S. Hypersensitivity tothimerosal: the sensitizing moiety. Contact Dermatitis34:201–203 (1996).

72. Westphal GA, Schnuch A, Sculz TG, Reich K, Aberer W,Brasch J, Koch P, Wessbecher R, Szliska C, Bauer A, et al.Homozygous gene deletions of the glutathione S-trans-ferases MI and TI are associated with thimerosal sensiti-zation. Int Arch Occup Environ Health 73:384–388 (2000).

73. Black J. The puzzle of pink disease. J R Soc Med92:478–491 (1999).

74. Warkany J, Hubbard DM. Acrodynia and mecury.J Pediatr 42:365–386 (1953).

75. Bernard S, Enayati A, Redwood L, Roger H, Binstock T.Autism: A Novel Form of Mercury Poisoning. Cranford,NJ:ARC Research, 2000;1–11.

76. WHO. Environmental Health Criteria 118. InorganicMecury. International Program on Chemical Safety.Geneva:World Health Organization, 1991.

77. Goldwater, LJ. Mercury: A History of Quicksilver.Baltimore, MD:York Press, 1972.

78. Ramazinni B. Disease of Workers. 1713. Reprint (WrightWC, transl). New York:Hafner Publishing Co, 1964.

79. Drasch G, Bose-O’Reilly B, Beinhof C, Roider G, Maydl S.The Mt Diwata study on the Philippines 1999—assessingmercury intoxication of the population by small scale godmining. Sci Total Environ 267:151–168 (2001).

80. Akagi H, Castillo E, Corest-Maramba N, Francisco-RiveraAT, Timan TD. Health asessment for mercury exposureamong school children residing near a gold processingand refining plant in Apokon, Tagum, Davoa del Norte,Philippines. Sci Total Environ 259:31–41 (2000).

81. Van Straaten P. Human exposure to mercury due tosmall scale gold mining in northern Tanzania. Sci TotalEnviron 259:45–53.

82. Gibson R, Taylor TS. Nicor says mercury spilled at moresites. Contamination found at 6 new locations, companytells state. Chicago Tribune (Chicago Sports FinalEdition) News Page, 1 Zone, 14 September 2000.

83. Sandborgh-Englund G, Elinder C-G, Johanson G, Lind B,Skare I, Ekstrand J. The absorption, blood levels andexcretion of mercury after a single dose of mercury vaporin humans. Toxicol Appl Pharmacol 150:46–153 (1998).

84. Kingman A, Albertini T, Brown LJ. Mercury concentra-tions in urine and whole blood associated with amalgamexposure in a US military population. J Dent Res77:461–467 (1998).

85. Sallsten G, Thoren J, Barregard L, Schutz A, Skarping G.Long-term use of nicotine chewing gum and mercuryexposure from dental amalgam fillings. J Dent Res75:594–598 (1996).

86. Engqvist A, Colmsjo A, Skare I. Speciation of mercury infeces from individuals with amalgam fillings. ArchEnviron Health 53:205–213 (1998).

87. Solis MT, Yuen E, Cortez PS, Goebel PJ. Family poisonedby mercury vapor inhalation. Am J Emerg Med18:599–602 (2000).

88. Langworth S, Sallsten G, Barregard L, Cynkier L, Lind M-L, Soderman E. Exposure to mercury vapor and impacton health on the dental profession in Sweden. Dent Res76:1397–1404 (1997).

89. Mathiesen T, Ellingsen DG, Kjuus H. Neuropsychologicaleffects associated with exposure to mercury vaporamong former chloralkali workers. Scand J EnvironHealth 25:342–350 (1999).

90. Albers JW, Kalenbach LR, Fine LJ, Langulf GD, Wolfe RA,Donofrio PD, Alessi AG, Stolp-Smith KA, Bromberg MB.Neurological abnormalities associated with remoteoccupational elementary mercury exposure. Anal Neurol24:651–659 (1988).

91. Letz R, Gerr F, Cracle D, Green RC, Watkins J, Fidler AT.Residual neurological deficits 30 years after occupa-tional exposure to elemental mercury. Neurotoxicology21:459–474 (2000).

92. Pendergrass JC, Haley BE, Vimy MJ, Winfield SA,Lorscheider FL. Mercury vapor inhalation inhibits bindingof GTP to tubulin in rat brain: similarity to a molecularlesion in Alzheimer diseased brain. Neurotoxicology18:315–324 (1997).

93. Newland MC, Warfinge K, Berlin M. Behavioral conse-quences of in utero exposure to mercury vapor: alter-ations in lever-press durations and learning in squirrelmonkeys. Toxicol Appl Pharmacol 139:374–386 (1996).



94. Taylor SA, Chiver ID, Price RG, Arce-Tomas M, MilliganP, Fels LM. The assessment of biomarkers to detectnephrotoxicity using an integrated database. EnvironRes 75:23–33 (1997).

95. Panda D, Miller HP, Wilson L. Rapid treadmilling of brainmicrotubules free of microtubule-associated proteins invitro and its suppression by tau. Proc Natl Acad Sci U SA 96(22):12459–12464 (1999).

96. Leong CC, Syed NI, Lorscheider FL. Retrograde degener-ation of neurite membrane structural integrity of nervegrowth cones following in vitro exposure to mercury.Neuroreport 12:733–737 (2001).

97. Satoh M, Nishimura N, Kanayama Y, Naganuma A,Suzuki T, Tohyama C. Enhanced renal toxicity by inor-ganic mercury in metallothionein-null mice. J PharmacolExp Ther 283:1529–1533 (1997).

98. Yoshida M, Satoh M, Shimada A, Yasutake A, Sumi Y,Tohyama C. Pulmonary toxicity caused by acute expo-sure to mercury vapor enhanced in metallothionein-nullmice. Life Sci 64:1861–1867 (1999).

99. Camisa CC, Taylor JS, Bernat JR, Helm TN. Contacthypersensitivity to mercury in amalgam restorations maymimic oral lichen planus. Cutis 63:189–192 (1999).

100. Ahlqwist M, Bengtsson C, Lapidus L, Gergdahl IA, SchutzA. Serum mercury concentration in relation to survival,symptoms and disease: results from a prospective popu-lation study of women in Gothenburg, Sweden. ActaOdontol Scand 57:168–174 (1999).

101. Bjorkman L, Pedersen NNL, Lichtenstein P. Physical andmental health related to dental amalgam fillings inSwedish twins. Comm Dent Oral Epidemiol 24:260–267(1996).

102. Saxe SR, Snowden DA, Wekstein MW, Henry RG, Grant

FT, Donegan SJ, Wekstein DR. Dental amalgam and cog-nitive function in older women: findings from the NunStudy. J Am Dent Assoc 126:1495–1501 (1995).

103. Pedersen NL, McClearn GE, Plomin R, Nesselroade JR,Berg S, DeFaire U. The Swedish Adoption Twin Study ofAging: an update. Acta Genet Med Gemellol (Roma)40:7–20 (1991).

104. Cederbrant K, Gunnarsson L-G, Hultman P, Norda R,Tibbling-Grahn L. In vitro lymphoproliferative assays withHgCl2 cannot identify patients with systemic symptomsattributed to dental amalgam. J Dent Res 78:1450–1458(1999).

105. Thompson CM, Markesbery WR, Ehmann WD, Mao YX,Vance DE. Regional brain trace-element studies inAlzheimer’s disease. Neurotoxicology 9(1):1–7 (1988).

106. Fung YK, Meade AG, Rack EP, Blotcky AJ, Claassen JP,Beatty MW, Durham T. Determination of blood mercuryconcentrations in Alzheimer’s patients. J Toxicol ClinToxicol 33(3):243–247 (1995).

107. Hock C, Drasch G, Golombowski S, Muller-Spahn F,Willershausen-Zonnchen B, Schwarz P, Hock U,Growdon JH, Nitsch RM. Increased blood mercury levelsin patients with Alzheimer’s disease. J Neural Transm105(1):59–68 (1998).

108. Fung YK, Meade AG, Rack EP, Blotcky AJ. Brain mercuryin neurodegenerative disorders. J Toxicol Clin Toxicol35(1):49–54 (1997).

109. Cornett CR, Markesbery WR, Ehmann WD. Imbalances oftrace elements related to oxidative damage in Alzheimer’sdisease brain. Neurotoxicology 19(3):339–345 (1998).

110. Cornett CR, Ehmann WD, Wekstein DR, Markesbery WR.Trace elements in Alzheimer’s disease pituitary glands.Biol Trace Elem Res 62(1–2):107–114 (1998).

111. Saxe SR, Wekstein MW, Kryscio RJ, Henry RG, CornettCR, Snowdon DA, Grant FT, Schmitt FA, Donegan SJ,Wekstein DR, et al. Alzheimer’s disease, dental amalgamand mercury. J Am Dent Assoc 130(2):191–199 (1999).

112. Iqbal K, Alonso AC, Gong CX, Khatoon S, Pei JJ, WangJZ, Grundke-Iqbal I. Mechanisms of neurofibrillarydegeneration and the formation of neurofibrillary tan-gles. J Neural Transm 53(suppl):169–180 (1998).

113. Buee L, Bussiere T, Buee-Scherrer V, Delacourte A, HofPR. Tau protein isoforms, phosphorylation and role inneurodegenerative disorders. Brain Res Rev 33(1):95–130(2000).

114. Olivieri G, Brack C, Muller-Spahn F, Stahelin HB,Herrmann M, Renard P, Brockhaus M, Hock C. Mercuryinduces cell cytotoxicity and oxidative stress andincreases beta-amyloid secretion and tau phosphoryla-tion in SHSY5Y neuroblastoma cells. J Neurochem74(1):231–236 (2000).

115. Neely MD, Sidell KR, Graham DG, Montine TJ. The lipidperoxidation product 4-hydroxynonenal inhibits neuriteoutgrowth, disrupts neuronal microtubules, and modifiescellular tubulin. J Neurochem 72(6):2323–2333 (1999).

116. Aksenov MY, Aksenova MV, Butterfield DA, Geddes JW,Markesbery WR. Protein oxidation in the brain inAlzheimer’s disease. Neuroscience 103(2):373–383 (2001).

117. Rowland IR, Mallett AK, Flynn J, Hargreaves RJ. Theeffect of various dietary fibres on tissue concentrationand chemical form of mercury after methylmercuryexposure in mice. Arch Toxicol 59(2):94–98 (1986).

118. Yao CP, Allen JW, Aschner M. Metallothioneins attenu-ate methylmercury-induced neurotoxicity in culturedastrocytes and astrocytoma cells. Ann N Y Acad Sci890:223–226 (1999).



The Three Modern Faces of Mercury€¦ · Environmental Health Perspectives • VOLUME 110 |...

Documents

Transcript of The Three Modern Faces of Mercury€¦ · Environmental Health Perspectives • VOLUME 110 |...