Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

mercury have different metabolic and biotrans-

formation properties which influence their overall

toxicity.

0010 Historically, mercury poisoning has been a well-

documented occupational hazard. However, it was

only well into the twentieth century that specific out-

breaks of environmental mercury poisoning demon-

strated that specific mercury species could exert

serious health effects on exposed human populations.

Examples of major outbreaks are: (1) Japan (Mina-

mata, Niigata), from the mid-1950s to the early

1960s from ingestion of contaminated fish and shell-

fish from waters polluted by industrial discharges; (2)

Sweden and Canada, in the mid- to late 1960s, from

similar but milder cases of industrial pollution; and

(3) Iraq in the early 1970s, from ingestion of grain

and grain products originating from seed treated with

a mercury fungicide. In each case, methylmercury was

the causative agent.

0011 Environmental mercury – whether from natural or

human activities – will continue to enter the human

food chain. Presently, the aquatic food chain is more

susceptible than the terrestrial food chain. However,

the ability of mercury to undergo biotransformation

to more toxic organic species in various environmen-

tal compartments (air, soil, aquatic sediment, water,

animal tissue) justifies the essential need to evaluate

mercury speciation when assessing mercury contam-

ination of foodstuffs. Therefore, any analytical meth-

odology for mercury must be capable of qualitative

and quantitative determination of specific mercury

species found in food.

Properties and Effects of Food

Preparation

Introduction

0012Environmental mercury species that eventually

become incorporated in food may be classified as

inorganic or organic. This section considers the

specific physical and chemical properties of the vari-

ous forms that would also allow the element to

enter the food chain. The effect of food-processing

methods on mercury content and speciation will also

be discussed.

Properties of Mercury Species

0013Inorganic The two inorganic forms are elemental

and ionic, the latter commonly occurring as the diva-

lent species Hg

2þ

. The high vapor pressure of elem-

ental mercury renders it quite volatile from water and

soil surfaces. However, it is readily soluble in these

media in the whole range. It is readily oxidized

to Hg

2þ

by bacterial action or chemical oxidizing

agents. Divalent inorganic mercury can form insol-

uble salts with the ions O

2

,Cl



, and S

2

rendering it less mobile and bioavailable to food-

chain organisms. However, its solubility and bioavail-

ability can be enhanced through complex formation

with Cl



,CO

2

, and OH



ions, with ammonia

and chelation with humic substances. An acidic

pH environment will also enhance its solubility.

Divalent mercury can be reduced to elemental mer-

cury by interaction with bacteria, specific inorganic

reducing agents, and sunlight. It also binds strongly

to sulfhydryl (SH) ligands on plant and animal

tissue proteins. Monovalent mercury (Hg

2þ

) can

originate in the environment from Hg

2þ

dispropor-

tionation, but the predominance of oxidizing condi-

tions in water and soil insures its transformation to

2þ

species.

0014Organic Of all the possible organomercurials,

the monomethylmercury ion (CH

) is the most

common environmental species. It originates from

2þ

through bacterial synthesis, organic oxidants,

and chelation. Its formation from Hg

2þ

in natural

waters is optimized at low salinity. It also has a

high affinity for sulfhydryl ligands on tissue proteins,

but these protein complexes rapidly break down

at very low pH. Its solubility – and bioavailability –

in soil and water is enhanced by its ability to

tbl0003 Table 3 The distribution of mercury in various foods

Food Totalmercury

concentration

(mgkg

1

freshweight)

Methylmercury

content (%)

Vegetables and fruits

Cabbage 4.1–9.0 25–29

Beans 0.4–1.0 ND

Tomatoes 0.1–8.0 13

Peppers 1.7 13

Potatoes 0.1–1.7 ND

Onion 0.5–5.1 2.4

Lettuce 0.1–12.2 9–17

Squash 0.7–1.6 1–7

Cucumber 0.1–2.0 2

Broccoli 1.8–5.1 6–8

Carrot 0.6–5.0 15

Garlic 2.1–4.8 24

Spinach 0.5–20.0 7–11

Grains

Wheat <0.1–46 NA

Oats 1.0–30 NA

Other

Mushrooms 1–16 <1–8

Edible parts.

ND, not determined; NA, not available.

3854 MERCURY/Properties and Determination

complex with anions (mostly OH



,Cl



) and humic

materials. Under conditions of very low (< 2.0) or

very high (> 11.0) pH, the methylmercury ion is

converted to inorganic mercury via bacteria or by

sunlight. Dimethylmercury is the only divalent

organic form commonly reported. Its higher volatility

allows it to exist mostly in air. However, acidic

conditions cause it to convert readily to monomethyl-

mercury.

Impact of Food Preparation

0015 Little is known about the influence of food processing

(dehydration, steaming, preservation, and other addi-

tives), storage (packing medium and container mater-

ial, temperature, time period), and cooking (backing,

freezing, smoking) techniques on mercury content

and speciation in food. Infact, little is known about

the exact chemical nature of inorganic mercury and

methylmercury in food that may influence their rela-

tive bioavailability.

0016 Processing Research from the author’s laboratory

concerning the distribution of inorganic mercury

and methylmercury in processed and nonprocessed

fish and seafood revealed different mercury distri-

bution patterns. Freshwater and older processed

(canned, > 5 years’ storage) marine fish had the ma-

jority (55–90%) of total mercury tissue content in a

water-extractable form. Only 22–47% was extract-

able for nonprocessed and newly processed (< 2 years’

storage) samples. Water-extractable mercury is of

interest because it represents mercury bound to

lower-molecular-weight tissue proteins or unbound

(neutral, ionic) species. Such species may be more

readily absorbed from the intestine into the blood

stream, thus having higher bioavailability. These spe-

cific mercury distribution patterns possibly result

from differences in the mercury source and form in

marine and fresh-water food chains.

0017 Storage The above marine fish mercury distribution

pattern suggests that food storage time may have a

greater impact on tissue mercury form and distribu-

tion than food processing. Although processing

may initiate a change in mercury distribution, further

changes can occur over an extended storage period.

Samples of canned tuna and seafood showed no ap-

parent effect of packing medium (water, oil) on the

percentage of water-extractable mercury. Both types

of packing media contained mostly inorganic mer-

cury, but this represented a very small fraction (gen-

erally < 2%) of the corresponding total mercury

sample content.

0018Cooking Broiling, baking, or smoking fish meat

does not alter the total and methylmercury content.

However, smoking increased the methylmercury per-

centage in brown trout fillets. (See Smoked Foods:

Principles.)

Analysis: Sample Preparation

Introduction

0019Sample preparation is a key component in any

analytical procedure, yet it is often the most time-

consuming to conduct and optimize. Ideally, it should

be simple, rapid, reproducible, provide quantitative

analyte recovery, and effectively remove all interfer-

ing substances prior to instrumental assay. The spe-

cific sample preparation procedure used depends on:

(1) complexity of the sample matrix; (2) physical and

chemical properties of the desired mercury species;

and (3) the instrumental technique. For food analysis,

the sample matrix may be relatively simple (water,

beverage) or quite complex (vegetation, soft tissue).

For total mercury analysis of food, sample prepar-

ation is greatly simplified by the use of mineral acid

(nitric, perchloric, sulfuric). However, mercury speci-

ation usually necessitates more elaborate solvent

extraction, clean-up, and analyte isolation steps.

Sample Preservation and Storage

0020Preservation of total mercury concentration in aque-

ous samples is critical, especially at microgram per

kilogram levels. The major mercury loss occurs by

conversion of Hg

2þ

(and Hg

2þ

) to elemental vapor

and subsequent loss to air. Another source of loss is

from adsorption of mercury on the container walls.

Several factors affect mercury preservation, and in-

clude mercury concentration, container type, method

of container cleaning, and types of oxidizing/com-

plexing agents used as preservatives. Generally, low

pH, high ionic strength, acidic (nitric or sulfuric acid)

dichromate as an oxidizer/preservative insures opti-

mum mercury preservation. Acid-washed containers

of Teflon, polypropylene, polyethylene, Pyrex, and

soft glass are satisfactory for mercury storage and

preservation. Methylmercury preservation in aque-

ous solutions requires storage in acid-washed poly-

tetrafluoroethylene (PTFE) containers under slightly

acidic pH as optimum conditions for long-term

preservation.

0021The need for mercury preservation in solid food

samples is less demanding since nearly all of the

mercury – organic and inorganic – is tightly bound

to tissue proteins, which minimizes absorption and

MERCURY/Properties and Determination 3855

volatilization losses. However, to minimize mercury

biotransformation – especially organomercury break-

down – by bacterial activity, all food and beverage

samples should be kept frozen (20



C) until analysis

preparation.

Sampling

0022 Sample homogenization is a convenient way to insure

that a more representative subsample will be obtained

for analysis. This is commonly achieved using a

blender and/or homogenizer, after which the sample

is sieved using 30–40 mesh. Deionized, organic-free

water is then added and the homogenate is further

mixed to a more uniform texture. However, care must

be taken not to contaminate the food sample through

contact with the container walls. A more recent sam-

pling procedure which further enhances sample

homogeneity is digestion with a mixture of 40–45%

sodium hydroxide (50%), 1% cysteine, and 1%

sodium chloride. This is the basis for preliminary

sample preparation in several (vapor atomic absorp-

tion spectroscopy (VAAS) and gas–liquid chromatog-

raphy (GLC)) extraction procedures. Alternatively,

for samples having mercury levels approaching the

lower detection limit of a given method, lyophiliza-

tion provides a dry powder more concentrated in

mercury.

Sample Extraction

0023 Total mercury Most food sample preparation pro-

cedures for total mercury involve various degrees of

chemically treating the sample (i.e., wet ashing). This

is mandatory since inorganic and organic mercury are

essentially volatilized by dry ashing. Wet ashing with

mineral acids usually destroys organic matter, which

can sometimes produce interferences, depending on

the instrumental technique being used. Acids and acid

mixtures that have been used successfully and pro-

duce negligible mercury loss are: perchloric, nitric,

sulfuric; nitric/sulfuric/perchloric, nitric/perchloric,

sulfuric/nitric. With the acid mixtures oxidants such

as hydrogen peroxide, permanganate, and vanadium

oxide have also been used. Loss of mercury can be

minimized by refluxing/condensing the acidic digest

vapors.

0024 Mercury speciation Since organomercury species

are readily degraded to inorganic mercury under

the drastic acidic and oxidizing conditions for total

mercury, much milder digestion conditions are re-

quired. In food samples, any protein-bound inorganic

mercury and organomercury requires an acidic

medium (usually hydrochloric, hydrobromic, and

sulfuric acids are employed) for liberation prior to

solvent extraction. Depending on the complexity of

the original sample, one or more organic solvent

cleaning steps are usually required. Aqueous or

ethanolic thiosulfate is the most widely used clean-

up reagent because of its rapid quantitative extraction

of organomercurials. The organic mercury species are

usually isolated either as a volatile halide derivative

(chloride, bromide, iodide for GLC assay) or as a

water-soluble thiol complex (for high-performance

liquid chromatography (HPLC) assay). Typical over-

all organic mercury recoveries are 70–100%.

0025Inorganic mercury assay by GLC requires prior

derivatization of the nonvolatile inorganic species

into a volatile organic form. Reagents that have

been used for this purpose are listed in Table 4.

Cold VAAS (CVAAS) employs specific reagents – tin

chloride and cadmium chloride – to convert separ-

ately inorganic mercury and both inorganic and

organic mercury, respectively, to elemental mercury.

0026All of the above wet-ashing procedures can be

partially or fully automated. This has been success-

fully implemented for CVAAS. Semiautomated head-

space analysis has been used for GLC analysis of

methylmercury in biological samples.

Analysis: Instrumental Techniques

0027Because of environmental and toxicological consider-

ations, most modern mercury analysis methods for

food emphasize speciation instead of total element

assay. Therefore, this review will highlight modern

mercury speciation techniques and summarize

classical – total mercury – techniques.

Total Mercury Techniques

0028Numerous analytical methods for total mercury

in biological media are summarized in Table 5.

tbl0004Table 4 Modern mercury speciation techniques

Te c hni q u e MDL

Gas-liquid chromatography (GLC)

Detector

Electron capture (ECD) 5–200 pg

Atomic absorption (AA) 0.1 ng

Microwave-induced plasma (MIP) 0.2–0.6 pg

Direct-current plasma (DCP) 0.5 ng

High-performance liquid chromatography (HPLC)

Detector

Electrochemical (EC) 40 pg

Atomic absorption (AA) 0.1 ng

Inductively coupled plasma (ICP) 0.3–0.7 ng

Ultraviolet (UV) 0.5 ng

Atomic absorption spectroscopy (AAS)

Mode

Cold vapor 0.02 ng

Graphite furnace 0.05 ng

MDL, minimum detection limit.

3856 MERCURY/Properties and Determination

While many of these methods are still reliable,

they suffer from major limitations which render

them unsuitable for routine analysis of food samples:

(1) they cannot accurately detect or measure mercury

at low levels (micrograms per kilogram), which are

normally background levels in foods; (2) they cannot

be adopted for speciation; and (3) most are time-

consuming and are subject to numerous inorganic

and organic interferences from the sample matrix. In

addition, some techniques – X-ray fluorescence

(XRF), neutron activation – require the use of expen-

sive specialized equipment and skilled operators.

Mercury Speciation Techniques

0029 Chromatography Chromatographic methods pro-

vide specific identification of the form in which

mercury is present. Early methodology in this area

was primarily intended for sample purification

prior to mercury determination by other methods.

These early techniques are listed in Table 5. The

microgram detection limits and time requirements

have greatly restricted their routine application.

Modern chromatographic mercury speciation tech-

niques for food analysis are GLC and HPLC.

Both are highly powerful and sophisticated tech-

niques which demand more technical attention

and skill for optimization and routine operation.

However, they can be highly automated in terms of

sample preparation and introduction, operation,

and data processing. Each technique is summarized

in Table 4.

0030 GLC has been the most widely used chromato-

graphic method for routine mercury speciation. It

was initially used for this purpose in the mid-1960s,

being applied to water, fish, and related seafood

samples. The major focus was on methylmercury,

the most toxic chemical form. Eventually, ethyl-,

phenyl-, and inorganic mercury were also included.

Currently, conventional (i.e., packed) column

chromatography coupled with electron capture detec-

tion (ECD) is still the most commonly used GLC

system. However, recent significant improvements

in GLC mercury speciation involve the use of

capillarycolumns and element-specific detectors. Ca-

pillary columns exhibit enhanced performance over

packed columns – greater separation efficiency and

shorter analysis times. Wide-bore (0.53 mm internal

diameter) based silica capillary columns are the most

popular compared to narrower-bore and borosilicate

glass columns. During the 1980s atomic emission and

absorption spectroscopic detectors began to be used

more for GLC mercury speciation. These detectors

provide better elemental selectivity and sensitivity

compared to ECD, along with freedom from con-

taminant interferences. (See Chromatography: Gas

Chromatography.)

0031HPLC mercury speciation application to food

analysis has been more recent. However, its use

has increased significantly since the late 1970s. Key

reasons for this are due to the advantages of HPLC

over GLC: (1) no analytical restrictions on sample

volatility and thermal stability; (2) more versatility

in optimizing instrument operation parameters be-

cause of specific separation modes (normal-phase,

reversed-phase, ion exchange), wide selection of

mobile phases, and use of solvent and flow gradient

programming; (3) sample derivatization can be per-

formed using precolumn, on-column (in situ) or post-

column techniques. This greatly simplifies sample

preparation; and (4) larger sample extract volumes

can be injected, typically, up to 250 ml against 10 ml

and 1 ml, respectively, for packed and capillary

column GLC. Recently, the availability and applica-

tion of more sensitive and element-specific detectors

have allowed HPLC to be a more viable mercury

speciation alternative to GLC. These include electro-

chemical (EC) and atomic spectroscopic (AAS,

inductively coupled plasma (ICP), and microwave-

induced plasma (MIP) ) detection. Because of cost,

simplicity, and commercial availability, the EC

detector is the most widely used. Currently,

reversed-phase chromatography (RPC) using conven-

tional analytical columns with 5 or 10 mmC

pack-

ings is the most commonly used HPLC separation

mode for mercury speciation. The polar aqueous-

based mobile phase is compatible with the newer

selective detectors and offers greater flexibility in

optimizing analyte resolution. (See Chromatography:

High-performance Liquid Chromatography.)

Atomic Spectroscopy

0032Two modes of this technique – CVAAS and graphite

furnace (GFAAS) – have been successfully applied to

tbl0005 Table 5 Early mercury analytical techniques

Totalmercury Chromatographic

speciation

Titrimetric (indicator, thermometric,

amperometric, potentiometric)

Ion exchange

Neutron activation analysis Thin-layer

Spectrophotometry Paper

Catalysis Column

X-ray fluorescence Radiochemical

Polarography (alternate current, anodic

stripping voltametry)

Atomic spectroscopy

Absorption (flame, furnace, cold vapor)

Emission (inductively coupled plasma,

fluorescence)

MERCURY/Properties and Determination 3857

routine assay of biological materials containing low

levels (microgram per kilogram) of organic and inor-

ganic mercury. While neither mode can identify the

exact form of organic mercury, they are still useful if

one considers that methylmercury is by far the pre-

dominant organomercury species in foods and most

other biological materials. CVAAS is the more com-

monly used technique. It selectively reduces inorganic

mercury to mercury vapor via tin chloride, while both

inorganic and organic mercury (i.e., total mercury)

are reduced via cadmium chloride with tin chloride.

Organic mercury is calculated by difference. The total

sample aliquot as its acid or alkaline digest is nor-

mally used. In GFAAS, mercury (organic/inorganic) is

released as mercury vapor during absorption or ther-

mal decomposition of the total sample aliquot. How-

ever, it requires prior separation of both mercury

forms using extraction procedures similar to those

of GLC, and the use of special matrix modifiers

to facilitate volatilization removal of chemical

interferences.

0033 Both techniques have distinct advantages over

GLC and HPLC in the area of mercury speciation:

(1) the entire sample/aliquot can generally be used;

(2) they are less subject to matrix interferences, thus

simplifying sample handling and eliminating the need

for solvent clean-up. They can also be more exten-

sively automated regarding sample handling, mixing,

and injection. Increased sample throughput is also a

major advantage.

0034 Recently, isotope dilution inductively coupled

plasma–mass spectroscopy (ICPMS) using flow injec-

tion analysis has been successfully applied to deter-

mining organic mercury in seafood. This technique

offers lower detection limits, freedom from spectral

interferences, and rapid sample throughput. As with

GFAAS, it does require prior isolation of RHgCl from

the sample. (See Mass Spectrometry: Principles and

Instrumentation.)

0035 Mercury’s environmental ubiquity and specific

chemical properties allow the element ease of entry

to the human food chain in several chemical forms.

The resulting long-term health and economic impact

man-dates the use for mercury analysis of modern

instrumental analysis techniques which are capable

of qualitative and quantitative mercury speciation at

submicrogram per kilogram levels. Currently, such

methods include GLC, HPLC, and flameless AAS

(CVAAS, GFAAS).

See also: Chromatography: High-performance Liquid

Chromatography; Gas Chromatography; Mass

Spectrometry: Principles and Instrumentation; Shellfish:

Contamination and Spoilage of Molluscs and

Crustaceans; Smoked Foods: Principles

Further Reading

Cappon CJ (1987) GLC speciation of selected trace

elements. LC-GC 5: 400–418.

Cappon CJ (1988) HPLC speciation of selected trace

elements. LG-GC 6: 584–599.

Cappon CJ (1990) Speciation of selected trace elements in

edible seafood. In: Hriagu JO and Simmons MS (eds)

Food Contamination from Environmental Sources,

pp. 145–193. New York: Wiley.

Fishbein L (1971) Chromatographic and biological aspects

of inorganic mercury. Chromatographic Reviews 15:

195–238.

Krenkel PA (1974) Mercury: environmental considerations.

Part II. CRC Critical Reviews in Environmental Control

4: 252–314.

Shrivastava AK and Tandon SG (1982) The determination

of mercury: a mini review. Toxicological and Environ-

mental Chemistry Reviews 5: 311–329.

Uthe JF and Armstrong FAJ (1974) The microdetermination

of mercury and organomercury compounds in environ-

mental materials. Toxicological and Environmental

Chemistry Reviews 2: 45–77.

VanLoon JC (1985) Selected methods of trace analysis:

biological and environmental samples. In: Elving PJ,

Wine-fordner JD and Kothoff IM (eds) Chemical

Analysis, vol. 80. New York: Wiley.

Toxicology

G L Klein and W R Snodgrass, The University of

Texas Medical Branch, Galveston, TX, USA

Introduction

0001This section will cover the absorption and bioavail-

ability of mercury, its distribution within and excre-

tion from the body, the mechanisms involved in its

harmful effects, toxic levels in humans, and accepted

treatment.

Absorption and Bioavailability

0002Since mercury is a toxic heavy metal with no known

role in normal metabolism, absorption must be inter-

preted in the broadest possible terms. Vapor inhal-

ation, enteric ingestion, and skin contact are the chief

routes by which mercury enters the body. A more

recent concern has been mercury contamination of

hepatitis B immune globulin as well as the adjuvant

for the hepatitis B vaccine. Inhalation of mercury

3858 MERCURY/Toxicology

vapor may come from its release from ‘silver’ dental

amalgams, which are approximately 50% mercury,

and from which vapors are released into the mouth by

chewing or brushing the teeth. Humans have been

documented to absorb a mean of 10 mgday

1

from

this source, with a range of 1.2–27 mg day

1

. Metallic

mercury, as a contaminant of latex paints, batteries,

or electrical industrial equipment, vaporizes and,

either in its metallic or inorganic form, is almost

totally absorbed across the pulmonary alveolar

membrane, perhaps owing to its high lipid solubility.

(See Heavy Metal Toxicology.)

Toxicology

0003 The mechanism of mercury absorption across the

skin from topical application of various fungicidal

preparations is uncertain. Absolute absorption

rate increases directly with increasing topical mer-

cury concentration, until a plateau is reached. (See

Fungicides.)

0004 A major organic salt of mercury, methylmercury, is

approximately 100% absorbed in the intestines of

humans and mice, while inorganic mercury salts

(mercurous and/or mercuric) are only about 15%

absorbed in human volunteers. Mercury is primarily

concentrated in the aquatic food chain as a result of

release of inorganic mercurials as industrial contam-

inants into either salt or fresh water. The inorganic

mercury compounds are acted upon by methanogenic

bacteria, producing methyl mercury. As such, it is

taken up by various forms of sea life and enters the

human food chain by direct consumption of fish and

other seafood. It has been suggested that high selen-

ium concentrations in fish can be protective against

methyl mercury toxicity. There is intracellular co-

localization of mercury and selenium and both have

been reported to be bound in a complex with a

plasma protein. Methylmercury can also enter the

human food chain indirectly by means of human

consumption of poultry and hens’ eggs when the

poultry have been raised on mercury-contaminated

fish meal. (See Fish Meal.)

Distribution and Excretion

0005 It is thought that the broadest exposure to mercury

now comes from mercury vapors from dental amal-

gams. Studies in sheep and cynomolgus monkeys have

demonstrated large concentrations of mercury in the

mandible, kidneys, and liver. Significant concentra-

tions have also been found in brain tissue. In pregnant

sheep, fetal tissues begin to accumulate mercury

within days of amalgam placement in the mothers

and fetal tissue accumulation increased progressively

during gestation. In monkeys a significant quantity of

mercury has also been found in bile and in the colon.

With methylmercury the distribution is similar but

more uniform. There is a significant difference be-

tween methylmercury and inorganic mercury with

regard to the red cell-to-plasma ratio (i.e., the distri-

bution between red blood cells and plasma), with

methylmercury in erythrocytes approximately 10-

fold higher than inorganic mercury. This difference

in concentration may affect the interpretation of urin-

ary mercury excretion since it may affect the amount

of mercury available for filtering through the kidneys.

Furthermore, there is a great deal of interspecies vari-

ability in the red cell-to-plasma concentration of

methylmercury, with the rat having the greatest and

humans the least. Interestingly, the rat has the highest

blood–brain mercury ratio, while humans have the

lowest.

0006While mercury is excreted primarily by the urine,

its appearance in the bile and colon after ingestion of

dental amalgam vapor by monkeys suggests that the

intestine is another source of mercury elimination.

The mercury body burden varies directly with the

excretory half-time. Thus, humans, with an excretory

half-time of 70 days, develop steady-state tissue levels

of mercury 10 times that of the mouse, which has an

excretory half-life of only 7 days. However, this is

merely an illustrative point because, based on autopsy

data of brain mercury concentrations in two human

beings after 15–25 years of Minamata disease, the

true mercury excretory half-life in humans may be

much longer than 70 days.

0007The excretion of mercury has been studied in rats,

and three steps have been identified. The initial rapid

phase involves approximately one-third of the total

dose, lasts a few days, and is associated with high

concentrations of mercury in liver and feces. The

two slower phases involve the remaining two-thirds

of the mercury intake and have a combined excretory

half-life of approximately 130 days. These slower

excretory phases are primarily associated with urin-

ary excretion of mercury. Mercury is probably also

excreted by the renal tubules and not the glomeruli.

Mercury may bind to sulfhydryl groups within the

tubular cells and, during excretion, damage to

the tubular cells could trigger the exfoliation of mer-

cury-containing tubular cells with a resultant increase

in urinary mercury excretion.

0008Binding to metallothionein may play a protective

role in both liver and kidney. In liver, biliary mercury

excretion is reduced with a high-zinc diet, suggesting

that zinc induces metallothionein to bind mercury

and the biliary excretion of mercury is primarily

related to nonmetallothionein-bound mercury. In

kidney, administration of mercury to rats is

MERCURY/Toxicology 3859

associated with an increase in urinary metallothio-

nein. (See Zinc: Physiology.)

Mechanisms of Toxic Effects

0009 Once absorbed from the digestive and respiratory

tracts, or absorbed transcutaneously, mercury pro-

duces a variety of symptoms affecting the marrow,

kidneys, skin, and blood cells. In children it produces

a syndrome complex called acrodynia, or ‘pink’ dis-

ease. Acrodynia is characterized by redness of the lips

and pharynx, loss of teeth, a strawberry tongue,

swelling, redness, and desquamation of the skin,

with pink or red fingertips, palms, and soles. Redness

of the conjunctival membranes, photophobia, cer-

vical lymph node enlargement, joint pain, loss of

appetite, and vascular thromboses are also observed.

Neurological manifestations include irritability,

apathy, withdrawn behavior, proximal muscle weak-

ness and hypotonia, and diminished reflexes.

0010 Methylmercury ingestion, which has occurred

twice in Japan and in Iraq on epidemic scales, pro-

duces acute and sustained neurotoxicity. In exposed

infants, cerebral palsy-like symptoms developed, in-

cluding psychomotor retardation, microcephaly,

spastic or flaccid paralysis, ataxia, athetotic move-

ments of the hands, constriction of the visual fields,

and generalized tonic convulsions. In utero exposure

can result in distinct toxicity to the embryo, including

low birth weight, dysplasias of the cerebral and

cerebellar cortices, and an abnormal migratory

pattern of neurons. Gross findings, such as severe

mental retardation, have been reported in these

epidemics, as have more subtle findings, such as

failure to achieve developmental milestones, observed

on follow-up studies.

0011 Occupational exposure to mercury vapour has led

to acute cases of respiratory distress, renal failure

requiring dialysis, and severe oropharyngeal in-

flammation and a ‘flu-like’ syndrome. With more

prolonged exposure, proteinuria develops. The hy-

pothesized mechanism for proteinuria is stimulation

by mercury of T lymphocytes to produce an antibody

to the glomerular basement membrane. This may be

mediated in part by mercury binding to nucleoprotein

of T lymphocytes. However, it has recently been

shown that methylmercury induces T-cell apoptosis

(programmed cell death), possibly by damaging mi-

tochondria and inducing oxidative stress that acti-

vates pathways for apoptosis.

0012 Neurological symptoms in workers exposed to mer-

cury 20–35 years previously have included reduced

muscle strength and coordination, increased tremor,

decreased sensation, and increased prevalence of

abnormal reflexes. It is hypothesized that progressive

neuron loss with aging may unmask previous expos-

ure to mercury. Furthermore, mercury has been dem-

onstrated to persist in neurons for many years after

exposure has ceased. A survey of workers previously

exposed to mercury failed to reveal any impairment of

reproductive function.

0013Also of interest, although not directly observed in

humans, in vitro experiments with human platelets

have shown that mercury can produce platelet aggre-

gation and increase production of eicosanoids by

platelets. Increased platelet aggregation may contrib-

ute to the thrombus formation seen in acrodynia as

well as some of the neurological structural abnormal-

ities produced by hemorrhages in the brain secondary

to mercury exposure of animals. Despite the consider-

able exposure to mercury from dental amalgams, no

toxicity has yet been reported in humans resulting

from this exposure.

0014In animals, toxicity is of a similar nature to that

observed in humans. Rats given mercuric chloride

intravenously develop complete proximal tubular ne-

crosis. If given a high-protein diet 48 h prior to mer-

cury ingestion, they develop only partial proximal

tubular necrosis, and blood urea nitrogen and creati-

nine concentration are normal. Mechanisms for renal

injury are, at least in part, felt to be autoimmune.

Mercuric chloride administration to the brown

Norway rat not only resulted in production of anti-

glomerular basement membrane antibodies but also

induced a rise in CD4þ splenocytes, CD4þ and

CD8þ (helper and suppressor) T cells, and B lympho-

cytes in peripheral lymphoid organs. In glomeruli

there was an increase in CD8þ (suppressor) T

lymphocytes. These findings suggested that the mech-

anism for autoimmunity is T lymphocyte-dependent.

Glomerular basement membrane autoantibodies are

both immunoglobulin G (IgG) and IgM, detectable 7

days after mercury injection. The degree of protei-

nuria seen in this group of rats did not correlate

with autoantibody levels. Although glomerular hist-

ology was unremarkable in these rats, by day 18 after

injection there were vascular changes, including

endothelial cell swelling in arteries and arterioles.

Antiglomerular basement membrane IgG appeared

by day 9 after injection, peaked by day 15, and then

declined. Proximal tubular basement membrane IgG

appeared on day 9 and peritubular capillary basement

membrane IgG appeared on day 15. These antibodies

persisted for at least 30 days. Urinary metallothionein

increased after mercury administration, probably re-

flecting mercury induction of tissue metallothionein

as a means of self-protection.

0015Significant neuropathology has also been produced

by mercury in vivo, both in the fetus and in more mature

animals. In pregnant rats given an intraperitoneal

3860 MERCURY/Toxicology

injection of methylmercuric chloride effects were

seen in the fetal nervous system within 2 h. These

early changes included mitochondrial degeneration

of the endothelium of cerebral capillaries leading

to subsequent hemorrhage, which interferes with

normal neuronal migration and normal development

of cortical cytoarchitecture. It is of interest that alter-

ations in vascularity leading to thrombus forma-

tion are seen in the kidney, as mentioned above.

Methylmercury also disrupts neuronal microtubular

assembly. Mercury is thought to reduce glutathione

peroxidase activity in cells, to interact with the

sulfhydryl groups of natural tubular and membrane

proteins, and to cause peroxidative injury. Mercury

has been found in organelles, primarily in lysosomes,

bound to selenium. When injected intraperitoneally

into the rat, methylmercuric chloride distribution was

uniform in cerebellar Purkinje cells and Golgi cells, as

well as in cerebral cortical layers III, IV, and VI. In

rats exposed to inorganic mercury, only scattered

mercury was isolated in level VI. Once in the brain,

methylmercury is slowly converted into inorganic

mercury. Both methylmercury and inorganic mercury

accumulated in the anterior horn motor neurons of

the spinal cord.

0016 Functionally, subcutaneous administration of

methyl mercuric chloride to neonatal rats resulted

in movement and postural disorders developing

during the fourth week of life. These occurred in

association with a selective degeneration of cortical

interneurons, which use g-aminobutyric acid (GABA)

as a neurotransmitter. In the putamen and caudate

nucleus, GABAergic and somatostatin immunoreac-

tive interneurons showed degenerative changes.

Sodium-dependent high-affinity uptake of tritiated

choline in cerebral cortex was also reduced, suggest-

ing that both cholinergic and GABAergic neuronal

function may be disturbed. Finally, in vitro incuba-

tion of rat cerebellar granular cells with methylmer-

cury caused a delayed but increased phosphorylation

of selected cellular proteins. The time course from

exposure to phosphorylation was 12–24 h, a time

period consistent with mercury alteration of gene

expression.

0017 Long-term neurotoxic effects from methylmercury

exposure have also been demonstrated in cynomolgus

monkeys. In one group of monkeys given methyl-

mercury orally from birth to 7 years of age, whole

blood concentrations of mercury fell almost immedi-

ately after the cessation of oral administration.

However, monkeys assessed neurologically 6 years

later exhibited insensitivity to touch and pinprick,

clumsiness, and slowness. Other monkeys were

exposed to methylmercury in utero by administra-

tion of methylmercury to their pregnant mothers

and continuing postpartum administration to the

infants for 4–5 years. There were more profound

visual defects in the monkeys with prenatal and post-

natal exposure compared to those exposed after

birth only.

Toxic Levels in Humans

0018The most accessible sampling sites in humans are

urine, hair, and whole blood. Urinary concentration

of mercury is the monitoring test which is most

often reported and is considered to be the analysis

of choice for chronic mercury exposure. Normal

values for urinary mercury concentration are less

than 100 nmol l

1

(20 mgl

1

). These reference values

were obtained as a result of a World Health Organ-

ization (WHO) study in 1961. Urinary mercury con-

centration was determined in urine samples from

1107 adults in 15 countries, and 85% of the subjects

had urine mercury concentrations within the above-

mentioned range. No reference values exist for chil-

dren, although acrodynia has been reported in

children with urinary mercury concentrations as low

as 249 nmol l

1

(50 mgl

1

). Urinary mercury concen-

trations do not correlate with ambient mercury levels

due to the long half-life of mercury in the body (ap-

proximately 40–70 days). In addition, mercury is

eliminated by other routes, such as sweat and bile.

Young children generally have higher urinary mer-

cury concentrations than adults because they crawl

on floors, where the highest concentrations of mer-

cury vapor tend to accumulate; toddlers also tend to

become dirty and may mouth objects that have been

lying in mercury-contaminated soil.

0019Workers occupationally exposed to mercury 20–35

years prior to being tested had neurological symp-

toms, including weakness, tremors, decreased sensa-

tion, and abnormal Babinski reflex (upgoing toes),

with peak urinary mercury levels of 0.6 mg l

1

(3.0 mmol l

1

). Initial serum mercury concentration

in two patients with acute intoxication from occupa-

tional mercury exposure were 550 and 490 mgdl

1

(27 and 24 mmol l

1

). Serum or whole blood mercury

concentration is a sensitive indicator of acute mercury

loading, especially with inorganic mercury, owing to

the near-equal distribution between red blood cells

and plasma, with normal serum mercury concentra-

tion being < 50 nmol l

1

(1 mgdl

1

). Hair mercury

content is reported to be a reliable indicator of

methylmercury load. When methylmercury has

undergone distribution to various body parts, the

blood and hair are in equilibrium, with hair concen-

tration being approximately 250 times that of whole

blood. Normal hair content of mercury is less than

7.5 mgg

1

hair.

MERCURY/Toxicology 3861

Ranges of Dietary Intakes

0020 Mercury levels in the diet, especially of methylmer-

cury, are of great importance, but are often poorly

documented. Recently, data have been reported on

the mercury content of poultry and dairy products

from hens and broilers raised on fish meal from a

number of countries in Europe and North America.

Table 1 lists these values and, notably, most products

fall within the acceptable tolerance limits of 50 p.p.b.

wet weight set by the Food and Agriculture Organiza-

tion (FAO) and the WHO of the United Nations.

However, these FAO/WHO limits were established

for foods other than fish and other seafood. Acidifi-

cation of lakes by acid rain has increased the levels of

mercury in the edible tissues of fish and has resulted in

the prohibition of sport fishing in many American

and Canadian lakes and rivers. In addition, the sale

of shark and swordfish has been restricted, owing to

their high mercury content.

0021 Inhalation of mercury vapor is also a significant

source of mercury intake as inorganic mercury is

volatile at room temperature, and absorption through

the respiratory tract occurs readily. Mercury can

be detected in human volunteers within 30 min

following exposure. Current recommendations by

the American Conference of Governmental Industrial

Hygienists are for a threshold limit value of mercury

of 25 mgm

3

(125 nmol m

3

) for a 40-h work week,

while the US National Institute of Occupational

Safety and Health (NIOSH) for workplace mercury

vapor exposure for 10 h day

1

is 250 nmol m

3

(50 mg

3

). For mercury concentration in household air, the

US Agency for Toxic Substances and Disease Registry

mandates an acceptable level of only 2.5 nmol m

3

(0.5 mgm

3

). In the ambient outdoor air in a major

US city such as Detroit, mercury concentration is

reported to be as low as 0.015 nmol m

3

(0.003 mg

3

). In contrast, indoor mercury concentrations in a

house just painted with mercury-containing latex

paint can be as high as 847 nmol m

3

(170 mgm

3

falling to 8 nmol m

3

by 7 days after the painting is

completed. However, this latter amount is still three

times greater than the recommended indoor ambient

concentration.

0022Mercury tends to accumulate near the floor. Thus,

in the room of a child in the UK who developed

acrodynia from mercury accumulation in a vacuum

cleaner, ambient mercury concentration was 1.0–

1.2 mgm

3

(5–6 nmol m

3

), while at the floor level it

was 30.0 mgm

3

(150 nmol m

3

). Similar findings

have been reported in the USA, with a range of ambi-

ent mercury reported from 0.5 to 50.0 nmol m

3

(0.1–10 mgm

3

Incidences of Mercury Toxicity and

Scares

0023The most notable occurrences of epidemic mercury

exposure were described in a group of cities bordering

Minamata Bay in Japan in 1956 and 1965, and in

Iraq in 1971–72. In the Japanese epidemics, an acet-

aldehyde plant utilizing mercury released the mercury

in its industrial waste into Minamata Bay, thus

entering the aquatic food chain. In the case of

Iraq, wheat contaminated with a mercury-containing

fungicide was used to make bread. Follow-up

studies in 1986–87 of 35 victims of Minamata dis-

ease, diagnosed between 1968 and 1978, revealed

not only persistent and progressive neurological

dysfunction but newly appeared lesions of spon-

taneous nystagmus in nine cases (25%). Fourteen

affected Japanese patients from the Niigata area

demonstrated temporal lobe gyrus degeneration, loss

of small myelinated fibers, and a reduction in large

neurons in the cochlear nerve. These findings sug-

gested that hearing impairment in these victims may

have been caused by progressive or irreversible degen-

eration of neurons or nerve fibers. It is from these

epidemics that we have learned most about methyl-

mercury poisoning.

Treatment

0024Both dimercaprol (BAL) and d-penicillamine have

been used to treat victims of mercury poisoning. Use

of BAL has been complicated by an increase in brain

concentration of mercury. Dimercaptosuccinic acid

(DMSA or succimer – Chemet

), is currently con-

sidered to be the chelator of choice to reduce the

body burden of mercury.

tbl0001 Table 1 Mercury content (methylmercury and total mercury) of

poultry products and eggs reported from Europe and North

America

Country Henmeat

(p.p.b.)

Broilermeat

(p.p.b.)

Eggs

(p.p.b.)

Methylmercury

Greece 0.5–3.1

0.1–3.0

0.4–28.9

0.4–2.9

0–2.7

0.1–0.6

1.7–16.5

0.5–7.1

USA 40

Sweden 17–37

6–22

1–2

Total mercury

Germany <1–141

Italy <1–15

Canada 1–23

USA 20–40

Breast;

leg;

liver;

free-range;

commercial range;

after prohibition of

methylmercury in agriculture;

egg white;

egg yolk.

3862 MERCURY/Toxicology

See also: Fish Meal; Fungicides; Heavy Metal

Toxicology; Immunology of Food; Renal Function and

Disorders: Kidney: Structure and Function; Selenium:

Physiology; Zinc: Physiology