Chapter 13 Review Ions In Aqueous Solutions: Fill & Download for Free

Health Effects of Excess CopperTHIS chapter focuses on the health effects associated with acute and chronic exposure to excess copper. Information on those effects comes from human case-reports and population-based studies. The emphasis is placed on acute exposure effects on the gastrointestinal (GI) system. Effects on other target organs, such as the liver, in subjects following high-dose chronic exposure and in sensitive populations are considered. Toxicity data from animal studies are presented, and the use of animal models for studying the mechanism underlying the toxicity of copper in humans is discussed.ACUTE TOXICITYCase Reports and Population-Based StudiesHuman cases of acute copper toxicosis are presented in this section. The cases are cited in reports by the NRC (1977), EPA (1987), the U.S. Agency for Toxic Substances and Disease Registry (ATSDR 1990), and the World Health Organization's International Programme of Chemical Safety (IPCS 1998). Table 5-1 summarizes the reported health effects of ingested copper in humans. Most human data on high-dose acute poisoning are based on cases of suicidal intent with the ingestion of copper compounds or accidental consumption of copper-contaminated foods and beverages. In such cases, it is difficult to estimate the quantity of copper consumed, whether it was in solid form, aqueous suspension, or solution. It is alsoTABLE 5-1 Case Reports of Copper Toxicosis Following Oral Exposures of Humans to Copper SaltsReferenceSex, Age, NumberExposuresClinical Sequelae and CommentsPercival 1784F, 17 yr, 1Ingestion of ≈ 100 to 130 g of a pickled vegetable, shown by analysis to be contaminated with CuAbdominal pain, generalized rash, flatulence, vomiting, diarrhea, cardiac arrhythmia, death on the 8th day post-exposureGriffin 1951M, adults, 6Ingestion of coffee brewed with water from a Cu-contaminated, gas-heated boilerNausea, vomiting, malaiseRoss 1955M F, NS, 12Stewed apples, cooked in a Cu vessel, were served to 87 adults and 458 children; 3 adults and 9 children developed symptomsBitter taste, nausea, vomiting, and diarrheaRoberts 1956M, 24 yr, 1≈ 400 g of CuSO4 in water over a 4-mo period (≈ 2 g Cu per day)Abdominal pain, hemolytic anemiaSanghvi et al. 1957M, 18, 20 yr, 2Ingestion of CuSO4 (amounts unknown) with suicidal intentSulfhemoglobinemia, acute renal tubular necrosis, azotemia, anuria, coma, and deathWyllie 1957F, NR, 15Contaminated alcohol lemon cocktails from cocktail shakers containing Cu; estimated Cu ingestion, 5–32 mgWeakness, abdominal cramps, headache, nausea, dizziness, vomiting in 10/13 nurses; possible causes, other than Cu, not excludedSemple et al. 1960M, NR, 18Contaminated tea (≈ 250 mL) containing Cu at 44 mg/L; tea made with water from a Cu-lined, gas-heated boilerAbdominal pain, vomiting, diarrhea, dizziness, and headache developed in 18 of 150 workers within 10 min after drinking the teaLe Van and Perry 1961M, NR, NRUnspecified number of men on ship drank Cu-contaminated soft drink (Cu content not stated) from a vending machineSeveral men developed mild nauseaChowdhury et al. 1961M F, adults, 2012 women and 8 men attempted suicide by ingestion of unknown quantities of CuSO4Epigastric pain, nausea, jaundice, hemolysis, hemoglobinuria, proteinuria, glucosuria, hypotension, tachycardia, stupor; no deathsReferenceSex, Age, NumberExposuresClinical Sequelae and CommentsGhosh and Aggarwal 1962M F, children, 3231 children accidentally ingested unknown amounts of CuSO4; one boy (age 11) given CuSO4 for attempted homicideVomiting was chief complaint; the victim of attempted homicide developed severe hemolysisBohré et al. 1965M, adults, 7Naval officers consumed coffee prepared with water from a Cu-contaminated electrically heated boilerNausea, vomiting, malaise, collapse, no deathsChuttani et al. 1965M F, 14– 60 yr, 53Oral ingestion with suicidal intent; estimated 1–30 g of CuSO4Diarrhea (29%), hemoglobinuria and hematuria (29%), anuria (27%), jaundice (23%), hypotension (8%), coma (8%), death (15%)Fairbanks 1967F, 22 yr, 1Accidental ingestion of unknown amount of CuSO4 solutionNausea, vomiting, diarrhea, abdominal pain, icterus, melena, hemoglobinuria, proteinuria, mild anemiaWahal et al. 1965M F, 15– 58 yr, 100100 patients ingested CuSO4 (dose unknown, mostly suicidal)Hepatotoxicity (36%), renal toxicity (18%), cardiovascular toxicity (10%), neurotoxicity (10%), GI hemorrhage (6%), death (10%)Nicholas 1968M, NR, 20Contaminated tea (≈ 250 mL) containing Cu at >30 mg/L; tea made with water from a Cu-lined boilerDiarrhea, nausea, vomitingSingh and Singh 1968M F, 13– 85 yr, 4027 males and 13 females; accidental or suicidal ingestion of copper saltsVariable hypercupremia, transaminasemia, hyperbilirubinemia, and renal insufficiency; 4 deathsSalmon and Wright 1971M, 1.25 hr, 1Oral ingestion of drinking water (Cu at 0.35–0.79 mg/L) over a 3-mo periodProstration, vomiting, red extremities, hypotonia, photophobia, peripheral edema, hypercupremiaReferenceSex, Age, NumberExposuresClinical Sequelae and CommentsMcMullen 1971M, NR, ≥ 10Oral ingestion of Cu-contaminated orange- or lime-flavored soft drinks, dispensed through Cu-containing bottle pourers; drinks contained Cu at 190 and 220 mg/L.At least 10 members of a sports club vomited immediately after consuming the soft drinksMittal 1972M, 22 yr, 1Oral ingestion of ≈ 175 g of CuSO4 (≈ 70 g Cu)Severe abdominal pain, vomiting, diarrhea starting 1.5 hr post- ingestion; renal damage with hemoglobinuria; eventual recoveryChugh et al. 1975M, 27 yr, 1Oral ingestion of ≥ 50 g of CuSO4 (≈ 20 g Cu)Cyanosis, oliguria, hemolysis, methemoglobinemia, death 16 hr post-ingestionAgarwal et al. 1975F, 41 yr, 1Oral ingestion of CuSO4 with suicidal intent; patient treated by hemodialysis at 13 hr after ingestionVomiting, diarrhea, hepatorenal failure, coma; hemodialysis was ineffectual; death; autopsy showed Cu in brain, heart, liver, kidneyStein et al. 1976F, 44 yr, 1After the patient had ingested ethanol and diazepam, CuSO4 (796 mg Cu) was administered as an emeticRespiratory collapse, massive GI hemorrhage, hemolytic anemia, renal and hepatic failure, death; autopsy showed renal tubular necrosis and increased hepatic Cu contentChugh et al. 1977M F, 18– 35 yr, 11Report of 11 cases of acute renal failure in a series of 29 patients with acute CuSO4 poisoning with suicidal intent.Vomiting (100%), epigastric pain (45%), diarrhea (45%), oliguria or anuria (91%), jaundice (91%), coma (18%); death (45%); biopsy or autopsy showed renal tubular necrosis in 7 of 8 patients studiedWalsh et al. 1977M, 1.5 yr, 1Oral ingestion of 3 g of CuSO4 (≈ 1.2 g Cu)Hemolytic anemia was observed 2 days post-ingestion, with hematuria, glucosuria, proteinuria, cylindruria, hypercupremia, hypercupriuria; recovery period was ≈ 1 yrStenhammar 1979?, 1–2.5 yr, 3Oral ingestion for unknown duration of tap water containing Cu at 0.22–1 mg/LProlonged diarrhea with weight loss; symptoms disappeared with change of water sourceReferenceSex, Age, NumberExposuresClinical Sequelae and CommentsBerg and Lundh 1981M F, < 3 yr, NRChildren at 7 Swedish kindergartens with Cu content in first-draw water from 0.35 to 6.5 mg/LAssociation noted between Cu content of drinking water and diarrhea, but other possible causes of diarrhea were not studiedSpitalny et al. 1984M F, 5, 7, 32 yr, 3Oral ingestion of drinking water (Cu at 2–8 mg/L; median 3 mg/L) for 1.5 yr by a family living at the end of a copper water mainEpisodic nausea, vomiting, abdominal pain occurred 5–20 min after drinking tap water in morning in 3 of 4 family members; the symptoms ceased with change of water sourceJantsch et al. 1985M, 42 yr, 1Ingestion of ≈ 250 g of CuSO4 (≈ 100 g Cu) in attempted suicideProtracted vomiting, hepatic failure, response to chelation therapy, eventual recoveryMüller-Höcker et al. 1988M F, 1 yr, 2Two infant siblings consumed for >9 mo Cu-containing tap water (Cu at 2.2–3.4 mg/L); the Cu was derived from Cu pipesMicronodular cirrhosis with hepatic Cu storage, hepatosplenomegaly, jaundice, and hypertransaminasemia; 1 deathKnobeloch et al. 1994M F, NRFive studies of GI upsets associated with ingestion of Cu-containing drinking waterHigher incidence of symptoms following ingestion of first-draw water compared with flushed waterNR, not recordeddifficult to control for potential confounding factors, such as microbial agents and their toxins. Following acute ingestion of copper salts (e.g., copper sulfate) in amounts that exceed approximately 1 g, systemic effects are generally observed. The effects include GI mucosal ulcerations and bleeding, acute hemolysis and hemoglobinuria, hepatic necrosis with jaundice, nephropathy with azotemia and oliguria, cardiotoxicity with hypotension, tachyeardia and tachypnea, and central-nervous-system (CNS) manifestations, including dizziness, headache, convulsions, lethargy, stupor, and coma.The systemic sequelae of acute ingestion of copper salts are highly variable. Oral intake of ionic copper usually induces immediate emesis, which reduces the quantity of residual copper available for absorption from the GI tract. Oral intake of copper that is bound to particulates in water or to proteins, lipids, and other constituents of foods is less likely to cause emesis, since the bound forms of copper show reduced bioavailability compared with ionic copper. The principal targets for acute copper toxicosis are the GI, hepatic, renal, hematological, cardiovascular, and CNS systems. There are few, if any, reports of musculoskeletal, dermal, ocular, immunological, carcinogenic, reproductive, or developmental effects in humans following oral ingestion of copper salts, even at high exposure concentrations.Acute copper toxicosis, manifested by hemolysis, headache, febrile reactions, prostration, and GI symptoms, was observed in one child after a solution containing copper sulfate was applied to burned skin during a debridement procedure (Holtzman et al. 1966) and in numerous patients after inadvertent introduction of copper into the circulating blood during hemodialysis (Manzler and Schreiner 1970; CDC 1974; Lyle et al. 1976). In hemodialysis patients, ionic copper can be released from semipermeable membranes fabricated with copper or from copper tubing or heating coils of the dialysis equipment, especially when the dialysate has become acidic (Barbour et al. 1971; Blomfield et al. 1969, 1971; Klein et al. 1972). In two hemodialysis patients, copper intoxication was characterized by marked hemolysis, acidosis, methemoglobinemia, hypoglycemia, vomiting, epigastric pain, diarrhea, and headache, with fatal outcome (Matter et al. 1969). In one report, copper stopcocks in circuits used for exchange transfusions were identified as the source of potentially hazardous infusions of copper in neonates (Blomfield 1969). Administration of copper sulfate as an emetic was identified as another iatrogenic cause of acute copper toxicosis (Holtzman and Haslam 1968).Wyllie (1957) describe one episode of acute GI symptoms associated with a presumed exposure to copper as a result of mixing alcoholic drinks in a copper-contaminated cocktail shaker. In reconstructing the exposure, the author concluded that the lowest adverse effect level was approximately 5.3 mg in 3/4 fluid ounces or 10.65 mg of copper. However, signifi-cant questions have been raised about the suitability of those data for estimating toxic doses of copper (Donohue 1997).Hopper and Adams (1958) presented five instances where faulty check valves in vending machines were responsible for carbon dioxide back flow and subsequent build-up of copper in vending machine water lines. The first drink in the morning can have a metallic taste, and cause salivation, nausea, vomiting, epigastric burning, or diarrhea (Hopper and Adams 1958).Semple et al. (1960) reported an outbreak of copper poisoning from ingestion of tea that was contaminated with copper sulfate scale deposited in the water used to make the tea. The authors estimated that the total copper in the suspension was 44 mg/L. That estimate is unreliable, however, because exposure likely occurred after a large portion of the scale was dislodged in the vessel, and the water used to make the tea was not available for analysis.The Centers for Disease Control and Prevention (CDC) reported multiple outbreaks of copper poisoning from ingestion of contaminated beverages (CDC 1974, 1975, 1977, 1996). In most cases, the copper concentration associated with illness was in excess of 30 mg/L. A major incident occurred in 1993–1994, where 43 individuals became ill from a single point source in a hotel. Exposures were estimated to range from 4.0 to 70 mg/L.Recurrent GI illness, including nausea and vomiting, occurred in a Vermont family. Exposure was traced to a build-up of copper in the water overnight. Copper concentrations reached 7.8 mg/L (with a range of 2.8 to 7.8 mg/L) (Spitalny et al. 1984). Family members had increased copper concentrations in hair, but not blood. Relief of symptoms occurred when their drinking water was replaced by bottled water.Knobeloch et al. (1994) investigated five individuals who ingested water above EPA's MCLG of 1.3 mg/L and reported abdominal symptoms. The authors suggested that increased copper in tap water can be a relatively common cause of GI symptoms.Using survey methods to gather data on copper-induced illness from soft drinks, Low et al. (1996) queried 2,100 state and local departments of health and agriculture and water utilities. Although the response rate was only 40%, they found 70 incidents of copper poisonings affecting 462 people. Copper concentration data were available for 24 cases and were less than 10 mg/L in 6 of those 24 cases (range of 3.5 to <10 mg/L. Dose estimation from those data is difficult however, because little information is available on the amount of beverage consumed and on whether the drink was consumed with food.Larger epidemiological studies have also investigated the relationship between exposure to excess copper environmentally and occupationally and adverse health effects. Recently, Buchanan et al. (1998) studiedhouseholds from lists supplied by the Nebraska Department of Health. Households were chosen on the basis of their drinking-water copper concentrations, which were above 3 mg/L in 60 households, between 2 and 3 mg/L in 60 households, and below 1.3 mg/L in 62 households. A telephone interview was conducted with one adult from each household, addressing the occurrence of GI illness. A nested study was also performed where in-person interviews were conducted and copper water concentrations were measured. The investigators found no association between copper concentration and GI illness. The copper water concentrations ranged from 0.06 to about 5 mg of copper/L in the first draw water.Roberts et al. (1996) examined communities made up of more than 100 people in Delaware. They considered communities in which 10% of the water samples, measured during a statewide survey in 1995, had copper concentrations greater than 5 mg/L. Four communities that met that criteria and one trailer park with older homes and acidic water were studied. First morning tap water was collected, and the households with concentrations in excess of 5 mg/L were revisited for study. Residents were interviewed once per week for 12 weeks and asked about GI symptoms. Although people with high concentrations of copper in their drinking water were slightly more likely to report becoming ill at some point during the study, there was no significant association of tap-water copper concentration and GI symptoms.The committee had difficulty in determining how to use the available epidemiological data. Although copper concentrations were used to stratify exposures in communities, it is difficult to link copper concentrations to individual exposures. Not everyone in a household can drink first-draw water; therefore, the high exposure would be encountered by one person in a household. The epidemiological data provide some assurance that copper concentrations in first-draw water above the current MCLG do not produce a high frequency of adverse GI effects within a community. However, the data do not allow the conclusion that water consumed at those concentrations would not give rise to GI-related symptoms in individuals. As a consequence, the committee relied on experimental studies that involved administration of water containing identified concentrations of copper.In one of the few experimental studies on copper in drinking water, Pizarro et al. (1999) designed a study to determine a threshold concentration for acute GI effects from copper in tap water. Sixty healthy adult women of low socioeconomic status from Santiago, Chile were randomly assigned to groups receiving 1, 3, or 5 mg of copper/L for 2 weeks, followed by 1 week of standard tap water. The women prepared the test water every morning and recorded their water intake and GI symptoms. Daily does of copper were not reported. Nausea, vomiting, abdominal pain orcramps, diarrhea, and food intolerance were recorded. Of the 60 participants, one woman recorded nausea only and two abdominal pain with 0 mg of copper/L of water. With 1 mg of copper/L, one woman recorded abdominal pain only, two recorded diarrhea only, one recorded diarrhea and vomiting, and one recorded all three symptoms. With 3 mg of copper/L, six women recorded nausea only, three recorded abdominal pain only, and one recorded vomiting only. Four women experienced diarrhea and abdominal pain. At 5 mg of copper/L, nausea only, abdominal pain only, diarrhea only, and vomiting only were recorded by five, two, three, and two women, respectively. One experienced both diarrhea and vomiting at 5 mg of copper/L. Therefore, a total of 21 subjects reported GI problems at some time during the study. Nausea, abdominal pain, or vomiting occurred 5%, 2%, 17%, and 15% of the time at 0, 1, 3 and 5 mg of copper/L, respectively. The data suggest that at or greater than 3 mg of copper/L can be associated with GI effects. The data also indicate a range of sensitivity in the population: 17% of the subjects reported symptoms at 3 mg of copper/L, while 85% did not report symptoms at 5 mg/L.In follow-up to Pizarro et al. (1999), a larger experimental study involving controlled randomized trials (sponsored by the International Copper Association) is being conducted to determine the dose-response relationship more precisely for the GI effects of copper in drinking water (ICA, unpublished material, Oct. 13 1999). The results of that study are nearing completion as this report goes to press. The experimental protocol, methods, and individual data were made available to the committee. The study involved 60 adult volunteers in Ireland (Colerain), Chile (Santiago), and North Dakota (Grand Forks). Each subject drank 200 mL of water containing 0, 2, 4, 6, or 8 mg of copper/L. Symptoms were noted at various times up to 60 rain at each testing center, and a 24-hr follow-up was made for any other symptoms. Each subject received each of the doses in a randomized order once a week.A dose-response for nausea was noted, although a lack of masking for taste might have affected the relationship. Not all individuals noted nausea even at higher doses, and a large variation in sensitivity among subjects is apparent. The fractions of those reporting nausea out of 180 subjects were 8, 7, 11, 25, and 44 at 0, 2, 4, 6, and 8 mg of copper/L of solution, respectively. Those results appear to be consistent with Pizarro et al. (1999), although they were considered too preliminary for further conclusions.An additional study is under way to investigate the acute effects of copper in drinking water. No data are available from that study (L. Chaffin, League of Nebraska Municipalities, personal commun., Feb. 9, 2000).In summary, there are inconsistencies among the data and they suffer from several limitations. There is little information on possible confound-ers and biases in the studies (including microbial water contamination and water averting behavior of residents), and there is an uneven distribution of risk in multi-member families. Copper concentrations often are not reliable because of a wide variability in first-draw concentrations and samples typically were not collected during the actual study period. Possible acclimatization is also not typically considered. In the experimental and epidemiological studies, sample sizes are small; therefore, the committee is concerned that effects on a sensitive population could have been missed. In addition, little work has addressed the health effects associated with exposure of infants and children to increased concentrations of copper in water. The only experimental study published in the literature indicates that GI symptoms arise from exposure at approximately 3 mg of copper/L (Pizarro et al. 1999).It is important to stress the point that all of the above cases represent acute toxicity as a consequence of the consumption of water or beverages that contain high levels of copper.For systemic effects, doses of copper in water associated with health effects differ from toxic doses in environmental media because of differences in bioavailability among water, food, and other environmental media. Copper ions in water have the highest bioavailability. The bioavailability of copper in the diet is a function of its solubility and also the types of complexes in which it is present. For example, complexes of copper with some amino acids and organic acids result in bioavailability similar to that of soluble copper sulfate (Wapnir 1998), whereas other dietary elements and certain amino acids can inhibit copper absorption.CHRONIC TOXICITYA major target of chronic copper toxicity is the liver. Liver toxicity is usually seen in specific populations, such as individuals with Wilson disease and children with various cirrhosis syndromes (see Chapter 4 for descriptions). However, there has been a case report of chronic ingestion of a high-dose copper supplement (30 mg per day for 2 years followed by 60 mg per day for 1 year) resulting in liver disease (O'Donohue et al. 1993). In that case, the pathological picture is similar to that seen in Wilson disease or the various childhood cirrhosis syndromes associated with excess copper exposure. Experimental animal studies have also demonstrated that ingestion of high amounts of copper in feed can lead to hepatic and renal disease (see section in this chapter on animal studies).A paper by Scheinberg and Sternlieb (1996) is frequently cited as evidence that high concentrations of copper do not cause liver toxicity. That article reported on a study of deaths from cirrhosis among children under6 years of age1 exposed to tap water containing copper at 8.5 to 8.8 mg/L in three towns in Massachusetts. The authors reported that no deaths from cirrhosis occurred among those children. However, the crudeness of the end point measured (death), the potentially small number of children actually at risk, and the variability in exposure depending upon whether the child was breast fed or formula fed puts that conclusion into question.The CNS can also be a target of chronic copper toxicity. Generally speaking, reports of neurotoxicity from chronic copper exposure have been limited to humans with Wilson disease. The CNS effects of copper in Wilson individuals are discussed in Chapter 4. Typically, neurological abnormalities have only been reported in animals administered very high doses of copper. Genetic animal models, such as the LEC rat (Mori et al. 1994; Kitaura et al. 1999), and Bedlington terriers with canine copper toxicosis do not have an increased susceptibility to the neurotoxic effects of copper (Owen et al. 1980, Hultgren et al. 1986).Hemolytic anemia due to high concentrations of circulating copper can also occur. Anemia has been seen occasionally in Wilson-disease patients (Scheinberg and Sternlieb 1984; Brewer and Yuzbasiyan-Gurkan 1992) and with copper poisoning in sheep (Gooneratne et al. 1981). In both situations, the anemia is due to a concomitant acute hepatic necrosis (Scheinberg and Sternlieb 1984; Brewer and Yuzbasiyan-Gurkan 1992). That breakdown of the liver cells releases a very large amount of copper into the circulation, damaging red blood cells and causing the acute hemolytic anemia (Scheinberg and Sternlieb 1984; Brewer and Yuzbasiyan-Gurkan 1992). Cessation of menstruation, an increased incidence of gall stones and renal stones, a form of osteoarthritis, and some kidney-function abnormalities can also occur in acute, untreated Wilson patients (Scheinberg and Sternlieb, 1984; Brewer and Yuzbasiyan-Gurkan, 1992). After exposure to exogenous copper in large amounts, acute renal failure can occur and result in permanent renal damage (Chugh et al. 1975; Holtzman et al. 1966).Reproductive and Developmental ToxicitySmall amounts of copper from intrauterine devices can prevent embryogenesis by blocking implantation and blastocyst development (Hurley and Keen 1979; Keen 1996; Hanna et al. 1997). In women with untreated Wilson disease, pregnancy is rare and often ends in spontaneous abortion.1A total of 64,124 child years based on the average 0–5-year-old population from 1969 to 1991 multiplied by the 23 years in that period.There are no reports in the literature of teratogenic effects associated with excess copper in humans. In a number of cases in which pregnancy continued despite the presence of a copper-containing intrauterine device, teratogenic effects of copper were not noted (Barash et al. 1990). In addition, there are no reports of abnormalities in the offspring of untreated Wilson patients.The reproductive and developmental effects of excess copper in animals are discussed later in this chapter.Genotoxicity, Mutagenicity, and Carcinogenicity of CopperThere have been numerous epidemiological studies looking at the relationship between exposure to copper, copper intake, serum copper, and various cancers. Those studies are complicated by the fact that serum copper concentrations can be increased as a result of cancer. Ceruloplasmin increases at times of stress (i.e., it is an acute-phase reactant) and 90% of serum copper is associated with ceruloplasmin. Wilson patients do have an increased risk of liver cancer, but only secondary to the liver cirrhosis associated with Wilson disease (Scheinberg and Sternlieb, 1984).In a large prospective cohort study (more than 10,000 Dutch individuals), Kok et al. (1988) found no association between serum copper concentrations and overall cancer deaths. However, when the group was stratified by copper concentration, there was a significant increase in risk (odds ratio (OR) of 3.7; confidence limits, 1.5–9.1) for the subjects with the highest serum copper concentrations. Cetinkaya et al. (1988) conducted a relatively small study comparing mean copper and zinc concentrations in 20 healthy women and 100 women with gynecological malignancies. They reported that mean serum copper concentrations were highest in the patients with malignant tumors. It is important to note that the serum concentrations of copper were measured after diagnosis and might have been due to increased ceruloplasmin. In addition, the analysis did not control for other possible confounders. Coates et al. (1989) used a case-control design among 5,000 Washington state employees and compared serum copper concentrations in 133 cancer cases and 241 controls among the employees. When the data were adjusted for potential confounders, there was no significant risk.A case-control study by Cavallo et al. (1991) compared the mean dietary intake of copper in controls (noncancer patients) and patients diagnosed with primary breast cancer. There was no significant association between dietary intake of the copper and copper blood concentrations. Further, there was no significant association between copper intake and breast-cancer occurrence. In a subset of cases and controls in this study, there was asignificantly increased serum copper concentration in cases compared with controls; however, there was no evidence of any association with serum copper concentration when they were examined by quintiles. Prasad et al. (1992) conducted a case-control study of 35 esophageal cancer patients and 35 controls in India and found no difference in serum copper concentrations between the cases and the controls.Overvad et al. (1993) studied plasma copper concentrations in women who developed breast cancer between 1968 and 1985. In this prospective study, copper concentrations in 46 women with breast cancer were compared with a stratified random sample of 38 women. The adjusted odds ratios did not suggest an increased risk for breast cancer due to copper exposure, although when copper levels were stratified, the highest exposure group did have a significantly increased relative risk.Dabek et al. (1992) studied 13 premenopausal and postmenopausal breast-cancer cases and compared them with 14 premenopausal and 11 postmenopausal noncancer controls. This prospective study was conducted to determine whether serum copper concentrations are altered significantly over a period of 1 year. Dabek reported a significantly higher serum copper concentration in the premenopausal breast cancer patients, but not in the postmenopausal patients, compared with the controls. In fact, postmenopausal breast cancer patients had significantly lower ceruloplasmin concentrations compared with their controls. Their lower concentrations might be due to lower estrogen concentrations. Dabek concluded that the copper-to-ceruloplasmin ratio in breast-cancer patients might reflect disordered copper metabolism in the disease.Occupational studies have been difficult to interpret, because most exposures involve mixtures of metals. However, Logue et al. (1982) found no evidence of any increase in all cancers in 3,550 men who were employed in the tank house of copper refineries. No other studies have adequately addressed the question of copper carcinogenicity in the workplace.In general, reviews of the epidemiological literature have highlighted the fact that most studies have determined serum copper concentrations only after diagnosis. Therefore, increased serum copper concentrations might be an effect of the cancer rather than a cause. Proper analysis of confounding variables is extremely rare, and any meaningful dose-response analysis is essentially absent from most studies. However, the few prospective studies that have been done provide little evidence for an association between copper and malignant disease. Therefore, there is inadequate evidence that copper plays any direct role in the development of cancer in humans.Copper generates oxygen radicals via a Fenton-type reaction (Goldstein and Czapski 1986), and many investigators have hypothesized that excess copper might cause cellular injury via an oxidative pathway, giving rise to enhanced lipid peroxidation, thiol oxidation, and, ultimately, DNA dam-age. Such damage can occur when copper is used in combination with lipids, hydroquinone, and aliphatic and aromatic aldehydes (Li and Trush 1993; Becker et al. 1996; Glass and Stark 1997). Therefore, copper could enhance endogenous oxidative reactions that cause DNA damage.Von Rosen (1964) reported that copper induces chromosomal breakage in plant cells in vitro. Other investigators have confirmed that copper (primarily tested as copper sulfate) is clastogenic in a number of different test systems, including the mouse bone-marrow system (Bhunya and Pati 1987). A dose-dependent increase in unscheduled DNA synthesis is also associated with in vitro exposure of rat hepatocytes to copper (Denizeau and Marion 1989). Most researchers report the chromosomal damage to be dose dependent. In contrast, copper sulfate does not induce DNA damage in prokaryotes (Matsui 1980), and copper chloride does not induce DNA strand breaks in phage PM2 in vitro (Becker et al. 1996).Results from mutagenic studies are conflicting. Copper (II) sulfate is negative in the Ames Salmonella reversion assay (Moriya et al. 1983; Marzin and Phi 1985), the SOS Chromotest with E. coil. PQ 37 (Olivier and Marzin 1987), and several other bacterial mutagenicity assays (Iyer and Szybalski 1958; Matsui 1980; Clark 1953). However, Hansen and Stefan (1984) tested copper sulfate in its pesticide form and reported it to be mutagenic. Copper chloride has been uniformly negative in bacterial mutagenicity tests (Wong 1988; Nishioka 1975; Kanematsu et al. 1980).Sensitive PopulationsInfants appear to be more sensitive to both low and excessive dietary copper intake than adults. With respect to excessive intakes, infants are sensitive to elevated copper in water for both exposure and physiological reasons. Infants fed formula reconstituted with tap water would consume a high amount of tap water, particularly on a per body weight basis. They also have higher absorption and reduced capacity to excrete copper at higher doses relative to older ages (see Chapter 2). Similar to infants, children may also be more sensitive to copper than adults, although there are less data available on children. Cases of liver cirrhosis in infants resulting from elevated copper in formula or milk, however, appear to also have a genetic basis (see below). Genetic defects in copper metabolism might confer sensitivity to excess copper exposure (also reviewed in Chapter 4).Carriers of Genetic Defects in Copper HomeostasisAs previously discussed in the Chapter 4 section Heterozygotes for Wilson Disease, there is evidence that presumed heterozygotes carriers of mu-tations in the Wilson gene can have subclinical abnormalities in copper metabolism at typical levels of dietary copper intake. Therefore, if ingestion of copper is substantially increased, heterozygotes might develop copper-induced liver disease. If current case-frequency estimates are correct, 1% of the U.S. population is at risk. If the frequency is severely underestimated, as much as 2% of the U.S. population is at risk (see calculations in the Chapter 4 section Heterozygotes for Wilson Disease).In Chapter 4, Tyrolean infantile cirrhosis (TIC), Indian childhood cirrhosis (ICC), and idiopathic copper toxicosis (ICT) are reviewed. All three diseases occur in infants or children; are familial, with an autosomal recessive genetic pattern; and often are associated with a documented or inferred increased ingestion of copper. Based on an experimental reconstruction, the concentration of copper was 10–63 mg/L (Müller et al. 1996). Those data support the hypothesis that there is a genetic susceptibility in many populations to disease from increased ingestion of copper affecting infants and young children. Although the gene mutations that underlie those diseases remain unknown, at least three genes might be hypothesized.The first hypothesis is that the canine copper toxicosis gene underlies the genetic susceptibility in the childhood cirrhosis syndromes. That gene, which is different from the Wilson (Dagenais et al. 1999), is present at high frequency in Bedlington terriers (Yuzbasiyan-Gurkan et al. 1997) and could be present in the human population and be associated with cirrhosis in young children exposed to increased concentrations of copper in milk and water. However, the canine toxicosis gene defect in homozygous form produces a severe, lifelong disease that is not dependent upon high levels of copper ingestion. In contrast, TIC, ICC, and ICT are restricted to early life, appear to require increased copper ingestion, and do not require treatment in survivors. Therefore, with the different clinical course, homozygocity for the defective canine toxicosis gene is unlikely to be involved in the childhood cirrhosis syndromes. Heterozygosity remains another possibility.A second hypothesis is that heterozygosity for Wilson gene defect combined with abnormal copper handling and increased copper exposure (Brewer in press) results in an increased sensitivity to copper in early childhood and underlies TIC, ICC, and many of the ICT cases. Assuming that a common use of copper pots caused ubiquitous exposure, the number of cases in the TIC epidemic is compatible with the hypothesis that the affected children were carriers of the Wilson gene. The population in the Tyrolean area was approximately 45,000 individuals. At the peak of the epidemic, there were about 3.8 infant cases per year. The estimated frequency of heterozygotes for the Wilson gene defect, 1%, could easily ac-count for the incidence of TIC based on the number of infants in a population of that size. Furthermore, there appeared to be a genetic component to the disease, as it occurred in some families but not in others, and the component appeared to be recessive, because in many families multiple, but not all, siblings were affected. That pattern of disease frequency would be expected if heterozygosity for the Wilson gene resulted in susceptibility to the disease and only infants fed milk that contained the high concentrations of copper. Therefore, sensitivity to increased copper concentrations would act as a dominant genetic trait that manifests only in infants because of their exposure to copper. The gradual disappearance of the disease might have been due to a decrease in the use of copper or brass containers for boiling. Similarly, in ICC, a high intake of copper from milk or water seems to affect certain infants or very young children in a familial fashion (Tanner et al. 1983). Eleven of 132 cases reported by Tanner et al. (1983) were associated with increased copper in water.The ICT data again suggest that a defective gene is present that makes infants and children unusually susceptible to high copper intake. Although the majority of TIC and ICC cases appear to be associated with high concentrations of copper in milk, nearly all of the ICT cases associated with increased copper exposure appear to be associated with increased copper concentrations in water rather than in milk. Müller et al. (1998) reviewed 15 studies of ICT. All the patients had increased hepatic copper concentrations, even those in whom drinking-water copper concentrations were not increased, suggesting that other copper sources must also be involved in ICT.A hypothesis of Horslen et al. (1994) is that the ICT cases can be categorized into two disease subgroups based on age of manifestation, clinical course of disease, ultrastructure of the liver, and copper exposure. With the first subgroup, ICT manifests before age 4 and is generally associated with increased copper exposure (as in ICC and TIC), whereas with the second subgroup, ICT appears after about age 4 without identifiable sources of copper excess.Of the cases reviewed, some ICT cases indicate a genetic predisposition (i.e., effects in siblings and consanguineous parents). In others, that information is not available. Among those age 4 and older, none of the cases reviewed had evidence of excess copper exposure (DuBois et al. 1971; Lim and Choo 1979; Lefkowitch et al. 1982; Maggiore et al. 1987; Horslen et al. 1994; Ludwig et al. 1996). Of the 10 cases reviewed involving bottle-fed children under age 4, copper in tap water was moderately to highly increased (above 1.3 mg/L) in seven cases (two are siblings; see Table 6-3 for copper concentrations) (Müller-Höcker et al. 1987, 1988; Baker et al. 1995; Bent and Bohm 1995; Walker-Smith and Blomfield 1973; Adamson et al.1992; Aljajeh et al. 1994). The other three cases involve a breast-fed infant, a previously bottle-fed 29-month-old child with copper at less than 0.1 mg/L of drinking water, and a 2-year-old who had consanguineous parents and who at 2 months of age received a formula prepared with tap water containing copper at 1.13 mg/L of water for 3–4 weeks.It should be noted that other possible sources of copper were not always investigated, and the actual exposure concentration in water is uncertain. Thus, it is difficult to determine the exact concentration of copper in drinking water needed to precipitate disease in susceptible populations of children. Nevertheless, some of the cases might be explained by the number of individuals who were heterozygous for Wilson disease and thereby more sensitive to copper exposure.A third hypothesis is that an undiscovered copper susceptibility gene, in combination with the high concentrations of copper, is responsible for the childhood cirrhosis epidemics. That gene could be either recessive or dominant and result in susceptibility to increased copper concentrations in infants. For example, genes involved in copper homeostasis have been recently discovered in yeast and animals. Many of those genes probably have homologous human genes. Mutations in such genes could underlie copper sensitivity.Regardless of whether heterozygosity for the canine copper toxicosis, Wilson disease, or an undiscovered copper homeostasis gene underlies childhood cirrhosis, the risk of increased copper intake remains if a genetic susceptibility to copper is involved in these syndromes.Glucose-6-Phosphate Dehydrogenase Deficiency (G6PD)It has been hypothesized that individuals with G6PD deficiency are at increased risk to copper exposure, because, in vitro, G6PD-deficient red blood cells are more susceptible to hemolysis and damage from copper than non-G6PD-deficient cells (Moore and Calabrese 1980). However, in vivo, about 90% of copper is covalently bound to ceruloplasmin and is not likely to cause red-blood-cell toxicity. With the exception of suicide attempts with copper salts, there is little evidence that low molecular weight copper increases significantly as a consequence of high dietary copper intake. Indeed, even in Wilson patients who have increased copper concentrations, hemolysis does not occur in the absence of hepatic necrosis. Therefore, the relatively small change in free copper in plasma that might result from a change of copper concentrations in typical diet or water is not likely be sufficient to alter the survival of G6PD-deficient red blood cells. Thus, individuals with GOD deficiency would not be expected to have increased sensitivity to high dietary copper intake.ANIMAL STUDIESIn this section, data on the toxicological effects of excess copper in experimental animals and the mechanism of copper toxicosis are reviewed. The appropriateness of animal models for assessing human toxicity to excess copper is also addressed. It should be noted that typical daily intake of copper in adult nonoccupationally exposed humans ranges between 0.9 and 2.2 mg (11.9 and 67.2 gg of copper/kg per day for the 10th and 90th percentile for typical intake of copper). The major route is oral, and the variability in copper intake reflects differences in dietary habits and agricultural and food processing practices. Levels of intake can be exceeded, particularly in cases where the drinking water contains high copper concentrations. Those factors should be kept in mind when human studies are contrasted with experimental data in animal studies of chronic exposure to copper (see Table 5-2). Unless otherwise stated, doses are reported as milligrams of copper per kilogram per day regardless of the form of copper administered.Toxicity in AnimalsThe literature on copper toxicity in animals is replete with studies in a variety of species. It is beyond the scope of this review to summarize those studies and the reader is referred to previous reviews (NRC 1977; ATSDR 1990; IPCS 1998; Barceloux 1999). Key studies that illustrate copper toxicosis are reviewed to provide background.Hepatic and occasionally renal changes are the most common effects found in animals that are fed high concentrations of copper. However, sensitivity to copper toxicosis is highly species dependent. In general, poultry appear to resist chronic copper toxicosis better than most mammals (NRC 1977). Sheep display toxicity even with relatively low copper concentrations in the diet, especially when the content of dietary molybdenum is low (0.5 mg/kg). Repeated doses of 1.5–7.5 mg of copper/kg of body weight per day as copper(II) sulfate or copper(II) acetate are associated with progressive liver damage, hemolytic crisis and ultimately death (IPCS 1998). Conversely, high levels of molybdenum in the diet of sheep have been associated with copper deficiency due to the formation of thiomolybdates in the rumen, which are potent anticopper agents (Auza 1983; NRC 1977; Aaseth and Norseth 1986; Dick et al. 1975; Mason 1990).In nonruminant mammalian species, such as rats, mice, rabbits, pigs and dogs, significant toxic effects of copper are associated with long-term ingestion of high doses of copper, well beyond those tolerated by humans. Increased mortality and growth retardation have been noted in rats subse-TABLE 5-2 Select Copper Effect Levels Observed in Long-Term Animal StudiesSpeciesCopper FormEffectsLOAEL or NOAEL (mg Cu/kg/day)aExposure DurationExposure RouteReferenceMouseCopper sulfateDevelopmental malformations53 (NOAEL)80 (LOAEL)30 days before mating; in females exposure continued until sacrifice on day 19Oral (feed)Lecyk 1980MouseCopper sulfateReproductive399, M (NOAEL)537, F (NOAEL)92 daysOral (feed)NTP 1993MouseCopper sulfateSurvival, body weight, general toxicity44, M; 126, F (NOAEL)92 daysOral (feed)NTP 1993Forestomach hyperplasia97 (NOAEL)187, M (LOAEL)267, F (LOAEL)Hepatic1,060 (NOAEL)Renal1,060 (NOAEL)MouseCopper gluconateReduction in life span42.5 (LOAEL)Throughout life spanOral (drinking water)Massie and Aiello 1984RatCopper sulfateNeoplastic27, M (LOAEL)40, F (LOAEL)280–308 daysOral (feed)Harrison et al. 1954RatCopper acetateIncrease in hepatic SGOT activity7.9 (LOAEL)90 day studyOral (drinking water)Epstein et al. 1982RatCopper sulfatePoor growth, hepatic hypertrophy, necrosis, inflammation270 (LOAEL)105 daysOral (feed)Haywood 1985; Haywood and Loughran 1985RatCopper acetateBody weight, organ weight, and bone size130 (LOAEL)126 daysOral (feed)Llewellyn et al. 1985SpeciesCopper FormEffectsLOAEL or NOAEL (mg Cu/kg/day)aExposure DurationExposure RouteReferenceRatCopper sulfateSurvival, body weight, general toxicity16 (NOAEL)92 daysOral (feed)NTP 1993Forestomach hyperplasia16 (NOAEL)33 (LOAEL)Hematological55 (NOAEL)113 (LOAEL)Hepatic16 (NOAEL)33 (LOAEL)Renal16 (NOAEL)Reproductive66 (NOAEL)MinkCopper sulfateReproductive6 (NOAEL)12 (LOAEL)9 months before, and for 3 months after matingOral (feed)Aulerich et al. 1982PigCopper sulfateHematological (decreased hemoglobin and hematocrit)8.8 (NOAEL)14.6 (LOAEL)54 daysOral (feed)Kline et al. 1971Other (decreased body weight)8.8 (NOAEL)14.6 (LOAEL)PigCopper carbonateHematological36 (LOAEL)48 daysOral (feed)Suttle and Mills 1966Copper hydroxideHepatic, increased SGOT activity36 (LOAEL)DogCopper gluconateElevated SGPT 8.4(NOAEL)6–12 monthsOral (feed)Shanaman 1972RabbitCopper sulfateMarked hepatic toxicity10 (LOAEL)Up to 400 days and overOral (drinking water)Tachibana 1952aConcentrations are expressed as mg of copper/kg of body weight per day regardless of the form of copper exposure.Abbreviations: NOAEL, no-observed-adverse-effect level; LOAEL, lowest-observed-adverse-effect level; M, males; F, females; SGOT, serum glutamic oxaloacetate transaminase; SGPT, serum glutamic pyruvic transaminase.quent to chronic ingestion of 27–300 mg of copper/kg of body weight (IPCS 1998; Boyden et al. 1938; Haywood 1985). Deaths have been commonly attributed to anorexia and extensive hepatic centrilobular necrosis. In mice exposed to copper guconate equivalent to 42.5 mg of copper/kg per day in the drinking water throughout life, a 12.8% reduction in maximal life span was noted (Massie and Aiello 1984).The highest reported no-observed-adverse-effect level (NOAEL) in the rat following chronic exposure in feed is approximately 130 mg of copper/ kg per day (administered as copper acetate) for 18 weeks (Llewellyn et al. 1985). The lowest-observed-adverse-effect levels (LOAEL) reported in the rat, based on hepatic effects, is 7.9 mg of copper/kg per day in feed (administered as copper acetate) for 90 days (increased serum glutamic oxaloacetate transaminase (SGOT) activity; Epstein et al. 1982). Pigs and rats appear equally sensitive to hepatic injury, and mice appear least sensitive (ATSDR 1990), having NOAELs of 1,060 and 97 mg of copper/kg per day (13 weeks of exposure in feed) for hepatic and gastric effects, respectively. In rabbits, chronic exposure to copper in drinking water (10 mg of copper/kg of body weight) was associated with marked hepatic toxicity, whereas dogs exposed to copper gluconate equivalent to 8.4 mg of copper/kg of body weight in feed had no toxic effects (Shanaman 1972; Shanaman et al. 1972; see IPCS 1998). The wide range of LOAELs and NOAELs in response to copper exposure reflects the sensitivity of the different animals and strains to copper, the exposure paradigm (duration of exposure), and the test performed (see Table 5-2). The LOAEL (and by inference the NOAEL) is also likely affected by the route of oral exposure, because copper that is administered in drinking water and in the absence of other food is much more readily absorbed across the GI tract.Hepatotoxicity is the most prominent and characteristic systemic effect following chronic exposure to copper. The hepatic lesions seen in copper overload appear to vary from species to species. In humans, marked mitochondrial abnormalities are seen in Wilson disease, and diet overloaded rats show nuclear destruction and various membrane abnormalities (Alt et al. 1990). Copper overload affects a variety of functions by cytosolic proteins, membranes, and subcellular organelles. High concentrations of copper can result in increased rates of lipid peroxidation as the intracellular concentrations of reactive oxygen species can be increased via the HaberWeiss reaction (Lindquist 1968; Miller et al. 1990; Halliwell 1989). Protein kinase C (PKC) activation, secondary to copper-mediated reactive-oxygen species generation, has also been seen in copper-induced cell death (Mudassar et al. 1992). Alternatively, Csermely and colleagues (1988) suggested that copper might directly activate PKC. Impairment of bile secretion, the predominant route for copper elimination, also results in excessive lysosomal copper accumulation, and, in turn, decreased lyso-somal membrane fluidity, increased lysosomal pH, breakdown of membranes, and the leakage of lysosomal enzymes, such as phosphatases, into the cytosol (Lindquist 1967; McNatt et al. 1971; Myers et al. 1993; Nieminen and Lemasters 1996).In rats fed high copper in the diet as copper sulfate, serum glutamic pyruvic transaminase (SGPT) activity was increased after 1 week of exposure at 100 mg of copper/kg per day (intake not calculated) (Haywood and Comerford 1980). Livers from male rats fed a high copper diet (1500 mg of copper/kg feed; intake not provided) for 16 weeks showed an increase in the number and diversity of lysosomes; swelling of smooth endoplasmic reticulum, mitochondria, and canalicular microvilli; and fragmentation of rough endoplasmic reticulum. Nuclear degeneration occurred early, culminating in lysis (Fuentealba and Haywood 1988). In rats exposed to copper sulfate equivalent to 300 mg of copper/kg per day for 1 week, centrilobular necrosis was noted (Haywood 1985). Centrilobular necrosis and inflammatory foci were also noted in the rat after 2–3 weeks of exposure to copper sulfate equivalent to 40–250 mg of copper/kg per day (Haywood 1980; 1985; Haywood and Comerford 1980; Haywood and Loughran 1985; Haywood et al. 1985a, 1985b; Rana and Kumar 1980).An extensive study (Shanaman 1972; Shanaman et al. 1972; see IPCS 1998) was conducted on the effects of copper in the beagle dog. That study was not available to the committee for review. Copper administered as copper gluconate for 6–12 months in the diet (equivalent to 0–8.4 mg of copper/kg of body weight per day) did not show significant toxic effects. A reversible increase in SGPT activity was noted in 2 of the 12 experimental dogs at the highest dose of copper (8.4 mg of copper/kg per day), an effect that was considered insignificant by a task group (IPCS 1998). There were no apparent gross or microscopic pathological lesions or changes in organ weight associated with this exposure paradigm. The insensitivity of the dog to copper toxicosis relative to the rat might be related to the dog's ability to rely more on transcuprein and low-molecular-weight complexes and less on albumin and ceruloplasmin for transport of copper to cells (Montaser et al. 1992). Total serum copper concentrations in the dogs are one-third those in the rat, and plasma ceruloplasmin concentrations are 8-fold less in the dog than in the rat (Montaser et al. 1992). Another speculation that might account for the lessened sensitivity of dogs to copper is that they might have a particularly effective copper transporter.Renal effects have also been observed in some studies. In rats fed diets containing 270–540 mg of copper/kg per day for 15 weeks (administered as copper sulfate), copper in the kidney rose to a plateau concentration in 4 weeks (Haywood 1980). Copper protein in the cells of the proximal tubules could be detected after 2 weeks(Haywood 1980; Haywood et al. 1985a, 1985b). Widespread sloughing of necrotic copper-containing tubule cellsbecame marked after 5 weeks of exposure to copper sulfate equivalent to 270 mg of copper/kg per day but declined subsequently as regeneration occurred (Haywood 1980; Haywood et al. 1985a; Haywood et al. 1985b). In the group receiving copper at 540 mg/kg per day, the toxicosis was prolonged (Haywood 1980). However, no renal changes were noted in rats exposed to copper sulfate equivalent to 100 mg of copper/kg per day in the diet for 2 weeks (Haywood 1980).A comprehensive study was conducted by the National Toxicology Program (NTP 1993) looking at the effects of cupric sulfate in drinking water (2 weeks) and feed (2 and 13 weeks) in B6CF1 mice and F344/N rats. Based on water intake and weight measurements, mice or rats consumed equivalent to between 0 and 368 mg of copper/kg per day or 0 and 97 mg of copper/kg per day, respectively. Consumption by mice or rats in the feeding study was between 0 and 783 mg of copper/kg per day or 0 and 325 mg of copper/kg per day (as cupric sulfate pentahydrate), respectively. In a 13-week feeding study, rats were given copper at 0 to 141 mg of copper/kg per day and mice were given 0 to 1,061 mg of copper/kg per day. Hematology, clinical chemistry, urinalysis, reproductive toxicity, and histopathology were evaluated. The reproductive end points are discussed in the reproductive section later in this chapter.In the 2-week drinking-water studies, water consumption in the three highest dose groups of both species was reduced by more than 65%. Some animals in the third highest dose group and all animals in the two highest dose groups died. The authors attributed the deaths to dehydration. The only gross or microscopic change specifically related to cupric sulfate toxicity was an increase in the size and number of cytoplasmic protein droplets in the epithelium of the renal proximal convoluted tubule in male rats consuming excess copper at 45 mg of copper/kg per day (NTP 1993).In the 2-week feed studies, rats and mice in the two highest dose groups had reduced body-weight gains compared with controls. The reduction was attributed to decreased feed consumption. Hyperplasia with hyperkeratosis of the squamous epithelium on the limiting ridge of the forestomach was seen in rats and mice; the more severe lesions were in rats. Inflammation of the liver, periportal to midzonal in distribution, occurred in rats consuming excess copper at 179 mg of copper/kg per day. Depletion of hematopoietic cells in the rat was evident in the bone marrow and spleen. Kidneys of rats consuming excess copper at 92 mg of copper/kg per day had an increased number and size of protein droplets in the epithelial of the renal cortical tubules (NTP 1993). Comparison of toxicity following exposure in drinking water and feed is a difficult task given the differences in consumption of copper in the two exposure paradigms. The studies, nevertheless substantiate the insensitivity of both species to copper toxicosis when contrasted with human exposure paradigms.In the 13-week feed study, there were no chemically related deaths in rats or mice, and no clinical signs of cupric sulfate toxicity were recorded. At 13 weeks, mean body weight was lower in copper-exposed animals than in controls for both rats and mice receiving excess copper at 64 mg of copper/kg per day and 188 mg of copper/kg per day, respectively. In the rat, hepatocellular damage was apparent, with increases in serum SGPT and sorbitol dehydrogenase activities, as well as increases in 5'-nucleotidase and bile salts (restricted to males). Renal tubule epithelial damage was suggested due to increases in urinary glucose, N-acetyl-β-D-glucosaminidase (a lysosomal enzyme), and SGOT (a cytosolic enzyme). Development of a microcytic anemia (i.e., decreases in mean cell volume, hematocrit, and hemoglobin) was noted, and increases in reticulocyte numbers suggested a compensatory response to the anemia by the bone marrow. Rats in the highest dose groups (more than 33 mg of copper/kg per day) had hyperplasia and hyperkeratosis of the forestomach, inflammation of the liver, and increases in the number and size of protein droplets in the epithelial cytoplasm and the lumina of the proximal convoluted tubules (NTP 1993). Those droplets stained strongly positive for protein but were negative for iron, PAS, and acid-fast (lipofuscin) staining methods. Copper was present in most of the protein droplets. Transmission electron microscopy of the livers of rats revealed increases in the number of secondary lysosomes in hepatocytes in the periportal area (NTP 1993).Mice in the 13-week feed study appeared to be much more resistant to the toxic effects of cupric sulfate than rats, and no effects on kidney function or histology were noted. There was a dose-related decrease in liver weights (NTP 1993). In mice receiving excess copper at 188 mg of copper/ kg per day, there was a dose-related increase in hyperplasia, with hyperkeratosis of the squamous mucosa on the limiting ridge of the forestomach. Minimal positive staining for copper was present in the liver and was limited to mice given high doses (more than 766 mg of copper/kg per day).Reproductive and Developmental ToxicityAn EC50 of 0.15 mg of copper/L (in the aqueous medium) has been reported to be teratogenic to frog embryos. The effects seen were defects of the eye, gut, notochord, and heart (Luo et al. 1993). The teratogenicity of excess copper in mammals has not been established. However, no defects are seen in newborn rats, hamsters, rabbits, sheep, or guinea pigs that were experimentally exposed to a uterine environment of high copper (Keen et al. 1982).Keen et al. (1982) studied the effect of feeding pregnant mice diets high in copper under varying conditions. Mice fed diets containing up to 500mg of copper/kg had normal litters. However, mice fed diets containing 2,000 mg of copper/kg during pregnancy did not carry their pregnancies to term. Mice exposed to that diet for only 5 days of pregnancy (days 7 to 12 of gestation) had resorption frequencies of more than 50%. The diet fed before and after the 5-day period contained 250 mg of copper/kg. Surviving fetuses were not visibly malformed, and their copper content was not appreciably higher than that of fetuses from dams fed diets containing 250 mg of copper/kg body weight throughout pregnancy. The high-copper diet caused a severe reduction in maternal food intake and a reduction of maternal body weight. It is thought that this caloric deprivation, rather than the excess of copper per se, was the cause of resorption. In pregnant mice, short periods of fasting (40 hr) can result in total litter resorption (Runner and Miller 1956). The importance of monitoring food intake in studies assessing the teratogenic potential of a nutrient is emphasized by the above results.Ferm and Hanlon (1974) have reported that copper (10 mg of copper/kg) injected intraperitoneally (i.p.) on day 8 of gestation in the hamster is teratogenic. Fetal resorption, kinked-tail, thoracic and ventral hernias, microphthalmia, cleft lip, and ectopic cordis were among the abnormalities found. In rats, i.p. injection of copper from day 7 through day 10 of gestation resulted in a resorption frequency of 50% (Marois and Bovet 1972).NTP (1993) examined vaginal cytology and sperm morphology, motility, and density, following a 13-week exposure to copper in feed, as previously described. No adverse effects on any of the reproductive characteristics measured in rats or mice of either sex were reported (NTP 1993). Reduced neonatal body and organ weights have been seen in the offspring of rats at doses of copper in excess of 30 mg of copper/kg of body weight per day over extended time periods; similarly, fetotoxic effects and malformations are seen with high doses (more than 80 mg of copper/kg of body weight per day) (IPCS 1998). In those cases, food intake was reduced, and that could account for the negative reproductive effects.In summary, injected copper can be teratogenic. When administered in the diet, copper can be teratogenic if the amounts are high enough to cause marked inanition. Periods of starvation can be teratogenic or embryo lethal in many species.NeurotoxicityGenerally speaking, neurotoxicity from copper seems to occur only in humans with Wilson disease; however, in various animal models in which copper ingestion is increased, some neurological effects have been observed. In rats maintained on a 10% casein diet, daily administration ofmanganese chloride (1 mg of magnesium/mL of drinking water) and copper sulfate (equivalent to 250 mg of copper/kg of diet; equivalent to about 20 mg of copper/kg of body weight per day) for 30 days resulted in learning and memory impairment (Murthy et al. 1981). The behavioral changes were associated with a marked accumulation of copper in the brain. Combined exposure to copper and manganese also produced increases in the dopamine (DA) and norepinephrine (NE), and a depression in 5-hydroxytryptamine (5-HT). In contrast, daily administration of 1,250 mg of copper sulfate/L of drinking water (0.125%; equivalent to about 46 mg of copper/ kg of body weight per day) for a period of 11 months in weaning rats was not associated with changes in concentrations of dopamine in the brain (de Vries et al. 1986). However, the concentration of 3,4-dihydroxyphenylacetic acid in the corpus striatum was lowered (25% decrease) in the copper-exposed group. Saturation studies of the striatal D-2 dopamine receptors indicated that copper increased receptor affinity, with a trend for decreased receptor number (de Vries et al. 1986). In genetic animal models, such as the Long-Evans cinnamon rat (LEC), and in the canine copper toxicosis common in Bedlington terriers, no neurological abnormalities have been reported (Kodama 1996; Brewer et al. 1992).Taken together, the chronic toxicity information indicates that, in the absence of genetic abnormalities, animals (with the exception of sheep) are not very sensitive to copper. Furthermore, given the massive doses (described above) required to induce chronic toxicity in mammalian animal species that have balanced mineral intake, animal studies provide little information on the mechanisms that underlie copper toxicity relevant to human dietary concentrations of copper. It is true that copper toxicity can occur in animals if they are given very high concentrations of copper in their food or water; however, interpretation of this finding must be bounded by the essentiality of copper at lower concentrations. The requirement for copper in tissues is under tight homeostatic control mechanisms (see Chapter 2). Cellular copper transport processes are required by an organisms for the optimal utilization of copper and the avoidance of toxicity due to excess copper. Chronic toxicity is, therefore, likely to occur when such control mechanisms are impaired or overcome by massive intakes of copper.CarcinogenicitySeveral animal studies have investigated the carcinogenic potential of various copper compounds. When judged against current methods, the data are not optimal, because most of the studies are dated and are single oral-dose studies (Tachibana 1952; Harrisson et al. 1954; Howell 1958;Carlton and Price 1973). Few studies used parenteral exposure (Stoner et al. 1976). However, the data clearly do not provide any suggestion that copper is carcinogenic in animals. Data from the LEC rat, which has been used as a model for Wilson disease, support the hypothesis that the cirrhotic effects of copper can play a role in hepatic cancer.The ability of metal ions to damage DNA and cause mutagenesis has also been analyzed with reversion and forward mutation assays using single-stranded DNA templates. Incubation of phi X174 am3 DNA with Fe2+ in vitro leads to mutagenesis when the treated DNA is transfected into Escherichia coli spheroplasts (Loeb et al., 1988). In the same assay, the frequency of mutants produced by copper ions was shown to be greater than that by Fe2+ (Tkeshelashvili et al., 1991). Although there are some data to suggest that oxidative DNA damage is important in the etiology of breast cancer (Malins et al. 1996), few human data are available to directly address that hypotheses.Mechanisms and Animal Models for Copper ToxicityMechanism of Acute Copper ToxicityAlthough the symptoms associated with acute and chronic copper toxicity have been well defined in humans, the mechanisms of copper toxicosis and the concentration at which toxicosis occurs remain poorly understood. A number of investigators have studied the mechanism of the gastric response to excess copper in an effort understand the implications of acute exposure to excess copper in drinking water. Wang and Borison (1951) noted that the latency for the gastric response is usually short (less than 1 hr) and that upper GI effects predominate. They severed the gastric nerves in dogs to determine if neuronal stimulation was necessary for the emetic response to copper exposure. In their work, when both the vagus nerve and sympathetic nerves were severed, the emetic dose increased approximately 8–10-fold, and the latency response rose by a factor of 10. Subsequent studies, most recently using the ferret as a model, suggest that stomach infusions are necessary to produce the emetic response (Makale and King 1992). Recent research also suggests that the emetic response can be blunted at the CNS level through the use of neurokinin-1 (NK-1) receptor antagonists (Saito et al. 1998). Therefore, the mechanism of the GI response to acute copper toxicity appears to be direct irritation of the stomach by copper ions. The emesis is primarily mediated neuronally but is affected by individual sensitivity, the volume of copper-containing material ingested, the state of the copper (free versus bound), and the presence or absence of gastric contents. There has also been some suggestion that adaptation might occur, making repeated exposure more tolerable.Mechanism of Chronic Copper ToxicityGenetic variants in animals provide a unique opportunity to study copper toxicosis. The LEC rat is an inbred mutant strain, isolated from Long-Evans rats, with spontaneous hepatitis. Approximately 40% of LEC rats die from fulminant hepatitis. The remaining 60% survive to develop chronic hepatitis and, subsequently, liver and kidney tumors (Mori et al. 1994; Kitaura et al. 1999). Therefore, the LEC rat has been used as an animal model for studying the role of chronic hepatitis in the development of liver cancer.The LEC rat manifests increased hepatic copper, defective incorporation of copper into ceruloplasmin, and reduced biliary excretion of copper. Therefore, it serves as an important animal model for studying the etiology of Wilson disease (Li et al. 1991; Wu et al. 1994; Mori et al. 1994; Cuthbert 1995; Vulpe and Packman 1995; Harris and Gitlin 1996; Koizumi et al. 1998). Recent studies on the structure and expression of the ATP7B gene protein (ATP7B) support its role as a copper transporter involved in the intracellular trafficking of copper in hepatocytes. In LEC rats and Wilson disease, copper accumulation is associated with a defective copper-transporting P-type ATPase, resulting in reduced biliary excretion of copper (Muramatsu et al. 1998; Bingham et al. 1998; Terada et al. 1998; Terada et al. 1999). Wu et al. (1994) cloned cDNAs from the rat gene homologous to the human Wilson gene (ATP7B) and identified a partial deletion in that gene in the LEC rat. At least 900 bp of the coding region at the 3' end, including the crucial ATP binding domain, is missing from the gene (Wu et al. 1994).Comparative dose-response relationships and hepatic or renal changes between the LEC rat and other rat strains are generally not available, because the LEC rat spontaneously develops tissue damage in the absence of copper fortified diets. Although LEC rats develop copper toxicosis after consumption of regular rodent diets (6 mg/kg),2 it is necessary to fortify the diet or drinking water with massive amounts of copper for other strains to see copper toxicosis (see the above section Toxicity in Animals).To investigate the role of the gene product (ATP7B) in hepatocytes, an expression plasmid carrying full-length complementary DNA for the human Wilson gene was constructed and expressed in hepatocytes of LEC rats (Nagano et al. 1998). ATP7B localized to the membrane with a molecular weight of 155 kilodaltons (kD). Upon expression of ATP7B in hepatocytes from LEC rats, the protein was present in the trans-Golgi network and at the plasma membrane, a distribution pattern similar to that of2Typical Analysis of Diet #F3156 Rodent Diet, American Institute of Nutrition-93G.Menkes-disease protein (ATP7A). Cotransfection and coexpression of the human Wilson gene and ceruloplasmin gene in cultured hepatocytes indicated that ATP7B always accompanied the distribution of ceruloplasmin at the perinuclear region but that part of ATP7B localized irrespective of the distribution of ceruloplasmin. Accordingly, ATP7B was proposed to localize to the trans-Golgi network and to transport copper into this compartment, ensuring optimal delivery of copper to the apoceruloplasmin (Nagano et al. 1998). In contrast, a fraction of ATP7B that was not accompanied by ceruloplasmin in the perinuclear region and at the plasma membrane was proposed to contribute to efflux of copper from the hepatocytes. The distribution patterns of ATP7B in hepatocytes might, therefore, explain the dual roles of this P-type ATPase in hepatocytes (Nagano et al. 1998).Chronic copper toxicity in LEC rats is associated with the uptake of copper-loaded metallothionein (MT) into lysosomes, where it is incompletely degraded and polymerized to an insoluble material containing reactive copper (Koizumi et al. 1998). This copper, together with iron, has been postulated to catalyze Fenton-type reactions and lysosomal lipid peroxidation, leading to hepatocyte necrosis (Ma et al. 1997; Koizumi et al. 1998; Klein et al. 1998). Subsequent to phagocytosis by Kupffer cells, the reactive copper might amplify liver damage either directly or through stimulation of those cells (Klein et al. 1998). There is no general consensus regarding the role of MT in lipid peroxidation as the primary mechanism mediating copper toxicity. In vitro studies suggest that other thiolrich cellular proteins might represent the primary site of copper-induced injury (Sokol et al. 1989). In that model, disturbances in the normal function of glutathione (GSH) in complexing copper soon after its uptake into the cell and the subsequent transfer of the complexed metal to MT where it is normally stored are cited as a potential mechanism of copper toxicosis (Freedman et al. 1989; Steinebach and Wolterbeek 1994).Copper toxicosis in the LEC rat has been postulated to result from liberation of massive concentrations of copper from MT when the intracellular capacity of MT synthesis was bypassed (Suzuki 1995; Rui and Suzuki 1997). Copper containing MT exhibited antioxidant properties in the presence of zinc. When zinc was not present, however, MT liberated cuprous ions, exhibiting prooxidant activity (Suzuki et al. 1996). The removal of copper complexed to molybdenum into the bloodstream was postulated to depend upon the amount of copper accumulating in MT (Ogra and Suzuki 1998).The fact that expression of heme oxygenase-1 (HO-1), an inducible isoform of heme oxygenase, and HO-2, a constitutive isoform of heme oxygenase, was enhanced in the LEC rat also supports that conclusion (Matsumoto et al. 1998). The high expression of HO-1 in the LEC rat waspostulated to arise from the oxidative stress caused by the accumulation of free copper, free iron, and free heme concentrations, thus representing an adaptive response to oxidative stress. Depletion of hepatic selenium in the LEC rat and reduced capacity to protect cells from copper-induced free-radical damage have also been suggested as potential mechanisms for enhanced oxidative stress (Downey et al. 1998). Ma et al. (1997) speculated that generation of free radicals by copper precipitates the induction of hepatic DNA damage and oncogenesis.Cellular copper homeostasis was recently studied in vitro in hepatic cell lines from the liver of LEC rats (Nakamura et al. 1995). Cells from the LEC rats accumulated larger amounts of copper than did control cells when the concentrations of copper in the culture medium exceeded 5 µM. However, the secretion of ceruloplasmin from the cultured cells was not reduced in hepatocytes from LEC cells compared with controls. Additional studies confirmed that the genetic defect in LEC rats did not alter the biosynthetic and secretory pathways of ceruloplasmin, and that the intracellular copper concentration did not regulate the synthesis and processing of ceruloplasmin in the cultured hepatocytes. Accordingly, the copper transporting ATPase encoded in the Wilson disease gene might not be an integral part of the biochemical mechanism of copper incorporation into apo-protein (Nakamura et al. 1995). However, given that Nakamura et al. (1995) used high concentrations of copper histidine, in which copper is in the cuprous form and can bypass normal transport mechanisms, the conclusion regarding the processing of ceruloplasmin in the LEC rat does not represent a relevant physiological situation.The available data on the mechanisms of carcinogenicity of copper in the LEC rat are sparse. The DNA binding activity of the serum response factor (SRF) was reported to be higher in the liver of LEC rats (approximately 2-fold) than in that of Wistar rats (Maeda et al. 1997). There was a close correlation between the intensity of the activity and the concentrations of copper in the nuclear protein. The DNA binding activity of Sp 1, on the other hand, showed similar levels in both LEC and Wistar rats. It has been postulated that SRF might play an important role in the development of hepatocellular carcinoma in LEC rats by mediating induction of the protooncogene c-fos and that the copper in nuclear protein might be involved in the activation of SRF (Maeda et al. 1997).In addition, LEC rats exhibit abnormal expression of γ-aminobutyric acidA (GABAA) receptor subunit genes in the brain, leading to an increase in GABAergic tone (Follesa et al. 1999). Neurochemical disturbances involved in abnormal catecholamine metabolism in the cerebral cortex of the LEC rat before excessive copper accumulation have also been described (Saito et al. 1996). It is unclear whether brain defects in the LEC rat are characteristic of human Wilson disease (Kodama 1996). The latter arecaused primarily by neuronal damage due to copper deposition in the CNS, and at times they are caused by encephalopathy secondary to hepatic dysfunction. It is noteworthy that deposition of copper in the CNS of LEC rats has not been reported (Kodama 1996).The molecular defect in ATPase of LEC rat alludes to its validity as a model of Wilson disease. Despite the abundance of clinical and biochemical similarities, however, the differences between the ATPase defect in LEC rats and Wilson disease must be recognized (see Table 5-3). In LEC rats, serum ceruloplasmin concentrations are almost normal, and the oxidase activity of ceruloplasmin, which represents holoceruloplasmin, is low in Wilson disease. H epatocellular carcinoma occurs frequently in the LEC rat, and cirrhosis does not. It is unclear whether these difference are species specific (rat vs. human) or due to the primary defect (Kodama 1996). It is clear that the effects of dietary copper loads in rats that do not carry genetic defects in ATP7B do not mimic those of Wilson disease. Therefore, the LEC rat model offers a unique opportunity to study the etiology of this disease.Toxic milk (tz), an autosomal recessive mutation in mice and a similar mutation in LEC rats in P-type ATPase, has also been described (Rauch, 1983). Mutant females produce offspring that exhibit poor growth, hypopigmentation, tremors, and ultimately die at 2 weeks of age. The mutants themselves accumulate large concentrations of copper in the liver and ultimately develop hepatic disease. Recent studies in the tx mouse have demonstrated that MT binds to copper in the liver and that high copper and MT concentrations are present. MT was postulated to enhance lipid peroxidation and account for genotoxicity and apoptosis (Deng et al., 1998).A third inherited form of copper toxicosis has been characterized as Indian childhood cirrhosis (ICC). A homozygous defect in the ATP7B gene has been excluded as a cause of human copper overload (van de Sluis et al. 1999), but a 50-kD purified major copper-binding protein (MCuBP) was recently identified as a potential contributor to the total copper-binding activity in the ICC liver (Prasad et al. 1998). Because copper administration alone at relevant dietary concentrations has not been shown to induce cirrhosis in animals, synergy between copper and a second hepatotoxin has been suggested (Tanner and Mattocks 1987; Aston et al. 1998; Tanner 1998) in the etiology of ICC. The hypothesis that ICC results not only from copper overload but also a from a second hepatotoxin has been recently tested. Rats chronically fed with excess copper in combination with a pyrrohzidine alkaloid (PA), retrorsine, were evaluated for morphological damage in a number of tissues. Increased plasma bilirubin, falling plasma albumin, histological changes, and massive accumulation of copper in the liver occurred. When retrorsine was given alone to lactating rats, no adverse effects were seen, although the suckling newborns did showTABLE 5-3 Comparison of LEC Rats with Wilson DiseaseWilson DiseaseLEC RatsInheritanceAutosomal recessiveAutosomal recessiveCopper concentrationLiverSeverely highSeverely highBiliaryLowLowSerumLowLowBrainHighNormalKidneyHighHighUrinaryHighHighSerum ceruloplasminOxidase activityLowLowAmount of proteinLowAlmost normalSerum SGOT, SGPT levelsIncreaseIncreaseSerum immunoglobulin G levelsHighLowHepatic pathologySimilar to Wilson diseaseOutcomeLiver cirrhosisHepatocellular carcinomaTherapy with chelatorsEffectiveEffectiveAbbreviations: SGOT, serum glutamic oxaloacetate transaminase; SGPT, serum glutamic pyruvic transaminase.Source: Modified from Kodama 1996.liver accumulation of copper. When copper and retrorsine were subsequently administered, the rats developed severe liver damage with retention of copper. Retrorsine passing to rat neonates via breast milk led to an accumulation of hepatic copper and an impairment in the rise in serum ceruloplasmin, suggesting either a decline in synthesis of ceruloplasmin, or a failure of copper to incorporate into the apo-protein (Aston et al. 1996). Retrorsine also caused a decrease in hepatic MT and serum albumin levels (also indicative of decreased protein synthesis) and reduced hepatic DNA levels (indicative of decreased cell number but increased cell size). An accumulation of copper in liver, with a reduction in copper-binding proteins could result in an increase in the pool of copper which is available to generate reactive oxygen species. This could explain the synergistic hepatotoxicity of copper and retrorsine (Aston et al. 1996). While the above model produced cirrhosis, significant differences from the histology seen in thehuman disorder were noted (Aston et al. 1998). The lack of an animal model, the inconsistent relationship between liver copper concentrations and liver damage, and the rarity of liver disease in adults suggest that other factors contribute to the etiology of ICC. Therefore, the hypothesis that ICC results from copper and a second hepatotoxin requires additional testing.Finally, a form of chronic copper toxicosis has been reported in certain Bedlington terriers (Owen et al. 1980). The condition is inherited and leads to progressive chronic hepatic degeneration (Hultgren et al. 1986). The DNA microsatellite marker C04107 has been linked to the copper toxicosis locus in Bedlington terriers, and it is used diagnostically to detect the disease allele (Yuzbasiyan-Gurkan et al. 1997; Holmes et al. 1998). The copper toxicosis locus in Bedlington terriers is located on dog chromosome 10 in a region syntenic to human chromosome region 2p13-p16 (van de Sluis et al. 1999; Dagenais et al 1999). Hepatic concentrations of copper in genetically afflicted Bedlington terriers are extremely high, and plasma ceruloplasmin concentrations are normal (Owen et al. 1980). Despite histological evidence of hepatitis in young dogs and cirrhosis in older ones, plasma levels of hemophilic factors VIII, IX and XI were observed to be above normal and were more closely related to the age of the dog than to hepatic copper concentrations (Owen et al. 1980). At the morphological level, ultrastructural and microanalytical techniques identified copper peaks in lysosomes, the nucleus, and the cytoplasm in descending order and profound cellular changes (Haywood et al. 1996). Hepatocytes generally appeared shrunk with compacted electron dense organelles; nuclei were contracted, misshapen with chromatin condensation and fragmentation; and apoptotic bodies were identified in sinusoids. Excess copper in the Bedlington terriers was initially sequestered in lysosomes, but following increasing saturation of this compartment, nuclear copper accumulation and DNA damage occurred. Apoptosis followed, probably mediated by induction of p53 protein (Haywood et al. 1996). Hepatic copper concentrations in copper toxicosis-afflicted Bedlington terriers range from 1,000 to 10,000 µg/g of dry weight (Owen et al. 1980), compared with 200 to 3,000 µg/g of dry weight in patients with Wilson disease and below 50 µg/g of dry weight in normal humans. Furthermore, the plasma ceruloplasmin of genetically afflicted Bedlington terriers is not altered (Kodama 1996); blood 64Cu concentrations 24 hr after administration of 64Cu shows no difference between affected and unaffected Bedlington terriers (Brewer et al. 1992); and Kayser-Fleischer rings (characteristic of copper accumulation at the periphery of the cornea) and neurological dysfunction are always absent in afflicted Bedlington terriers. Thus, the Bedlington terrier model must have a new and as yet unidentified gene that is important for copper homeostasis.CONCLUSIONSAcute GI effects of copper, including nausea and vomiting, have been seen in case reports and epidemiological studies. Dose-response information is difficult to determine from those studies.Recent controlled human experimental studies have demonstrated a dose-response relationship for the acute GI effects of copper.Acute copper toxicity does not seem to pose a significant reproductive risk for humans. In experimental animals, high concentrations of dietary copper do not pose a reproductive risk unless food intake is reduced. High concentrations of copper given by injection can be teratogenic, but the significance of that finding in humans is unclear.Copper metal is inactive in most assays of mutagenicity, although it can induce chromosomal and DNA damage via a free-radical-mediated mechanism under the appropriate conditions.There is inadequate evidence that copper plays a direct role in the development of cancer in humans.In sensitive human populations, the major target of chronic copper toxicity is the liver. In Wilson disease, neurological toxicity also occurs.Based on cases of TIC, ICC, and ICT, there appears to be a subset of the population sensitive to hepatic copper toxicity. Cases generally occur in infants or young children, have a familial pattern suggestive of recessive inheritance, and usually involve increased ingestion of copper in milk or water. The data suggest that a gene causes a predisposition to copper-induced liver cirrhosis. Given that some heterozygous carriers of the Wilson gene accumulate abnormal concentrations of copper, a reasonable hypothesis is that carriers of mutations of this gene comprise the infants with copper toxicosis disorders. An alternative hypothesis is that an unknown copper-susceptibility gene is present in many populations. Irrespective of which hypothesis is correct, increase in the ingestion of copper should be cautioned against until the hepatic susceptibility is clearly identified.In general, studies on the toxicity of copper in animals provide little information except for some data on physiological, biochemical, and pathological aspects of copper metabolism or chronic toxicity relevant to human dietary concentrations of copper.Although animal models provide some qualitative insight into the toxicology of copper, they are of limited value for establishing dose-response relationships in humans.The LEC rat, an inbred mutant strain isolated from the Long-Evans rat, is prone to copper toxicosis because of a defective ATP7B copper transporter. This illustrates how genetic errors in experimental animal models can result in syndromes that more closely represent the human situation. Therefore, such models are useful for studying human genetic defects.