Diagnostic Approaches to Trace Element Deficiencies

Home , Chromium deficiency, Glycine reductase

Ananda S. Prasad

Department of Medicine, Wayne State University School of Medicine, Harper-Grace Hospital, Detroit, Michigan 48202; and Veterans Administration Medical Center, Allen Park, Michigan 48101

Although several trace elements, such as iron, copper, zinc, chromium, selenium, manganese, cobalt, iodine, and fluorine, are considered to be essential for human health, only a few have clinical relevance. Manganese deficiency in human subjects has not been established beyond doubt. The only role of cobalt appears to be that related to the vitamin-B12 molecule. Although important for dental health, fluorine is not essential for life. The essentiality of iron, which is needed for heme synthesis, has been known for more than 100 years, and several reviews on this subject are available; therefore this topic is not considered here. The role of iodine in thyroid metabolism has also been well known for more than a century, and inasmuch as many books deal with this subject iodine is not covered in this review. Deficiencies of zinc and copper in human subjects have been recognized only during the past two decades (1-4), but a great deal of progress has been made in understanding their important roles in biochemical functions. Clinical problems of their deficiencies are being discovered with increasing frequency in association with several disease states. It is therefore important to deal with the diagnostic approaches of the deficiencies of zinc and copper in human subjects in detail. Although clinical deficiencies of chromium and selenium are not well established thus far, it is important to recognize their potentially useful roles in clinical medicine; as such, these topics are discussed in this chapter.

ZINC DEFICIENCY

Etiology It is becoming more apparent that deficiency of zinc is very prevalent, accom- panying inadequate protein intake, such as is seen in cases of protein—calorie malnutrition in populations subsisting on low incomes and in geriatric cases. Predominant use of cereal proteins by the majority of the world population is a very important predisposing factor for zinc deficiency. The availability of zinc in such diets is very poor because of high phosphate and phytate content. 17 18 DIAGNOSTIC APPROACHES

Zinc deficiency in human subjects has been reported to occur in conditions where there is an increased requirement of zinc. Included here are infants and children during the rapid-growth-age period and women who are pregnant and lactating. The recommended dietary allowances for zinc in infants, adolescents, and pregnant women are relatively high and are unlikely to be met on an ordinary diet. Thus physicians have to be aware of this possibility and take proper preventive measures. Zinc deficiency has been reported to occur in patients with malabsorption syndrome. In such cases, zinc deficiency is likely to occur in association with other deficiencies. Hyperzincuria over an extended period may lead to zinc depletion. Urinary excretion of zinc appears to vary with the amount of element bound to the plasma amino acid pool. Thus, in those conditions where histidine and cysteine increase in the plasma, one would expect to find hyperzincuria. Conditions associated with a hypercatabolic state, such as surgery, burns, multiple injuries, major fractures, diabetes mellitus, protein deprivation, and starvation, usually exhibit hyperzincuria (5). Proteinuria and use of chelating agents such as penicillamine also result in excessive zinc loss in the urine. Severe deficiency of zinc resulting from penicillamine therapy in a patient with Wilson's disease was reported by Klingberg et al. (6). Hyperzincuria is likely to occur after chlorothiazide administration. Thus, hypertensive patients on long-term therapy with chlorothiazide should be monitored for zinc deficiency. Glucagon is also known to cause hyperzincuria (7). Patients with chronic liver disease, nephrotic syndrome, and sickle cell disease are known to have hyperzincuria (8-10). It has been shown that several of the clinical manifestations of sickle cell disease and chronic liver disease are indeed due to zinc deficiency and that zinc supplementation corrects these manifestations (11,12). At present, two genetic disorders, acrodermatitis enteropathica and sickle cell disease, are known to be associated with zinc deficiency. All the clinical manifestations of acrodermatitis enteropathica seem to be reversible with zinc supplementation (13). In sickle cell disease, clinical manifestations, such as growth retardation, hypogonadism in males, abnormal dark adaptation, hyperammonemia, general lethargy, and poor appetite, appear to be related to a deficiency of zinc (14). Severe zinc deficiency in patients receiving long-term total parenteral nutrition without zinc supplementation has been reported by several investigators (15). Symptoms are similar to those seen in acrodermatitis enteropathica and include skin rashes, alopecia, diarrhea, and depression. The onset of symptoms usually 5 to 10 weeks after the start of total parenteral nutrition is related to the severity of zinc depletion in the patient.

Clinical Manifestations Growth retardation, hypogonadism in males, poor appetite, mental lethargy, and skin changes were the typical clinical features of chronically zinc-deficient subjects DIAGNOSTIC APPROACHES 19 from the Middle East, as reported by the author during the early 1960s (1-3). These features were corrected by zinc supplementation. The mechanism of the characteristic enlargement of spleen and liver in this syndrome is not well understood at present. Abnormal dark adaptation in alcoholic cirrhotics has been related to a deficiency of zinc (16). Zinc administration to these patients corrected the abnormal dark adaptation. Similar clinical observations have been made in zinc-deficient sickle cell anemia patients (17). It has been proposed that the effect of zinc on the retina may be mediated by an enzyme, retinene reductase, which is known to be zinc- dependent. In sickle cell disease, delayed onset of puberty and hypogonadism in the male, characterized by decreased facial, pubic, and axillary hair, short stature, low body weight, rough skin, and poor appetite have been noted and related to a secondary zinc-deficient state (14). Many patients with sickle cell disease develop chronic leg ulcers that do not heal, and a beneficial effect of zinc supplementation in such cases has been reported. Some patients with celiac disease who failed to respond to diet, steroids, or nutritional supplements made remarkable recoveries when zinc was administered. They gained weight, the D-xylose-absorption test improved, and the steatorrhea was alleviated after zinc therapy (18). Zinc supplementation in a few subjects with malabsorption syndrome (other than celiac disease) has produced beneficial results with respect to growth retardation, hypogonadism in males, mental lethargy, skin changes, and loss of hair (19). One should therefore be aware of the occurrence of zinc deficiency as a possible complication of malabsorption syndrome. The conclusion that zinc can promote the healing of cutaneous sores and wounds has been controversial for several years. Most studies now provide evidence that zinc supplementation promotes wound healing in zinc-deficient patients and that zinc therapy in zinc-sufficient subjects is not an effective therapy for wound healing. Abnormalities of taste have been related to a deficiency of zinc in many clinical conditions by some investigators (20). Decreased taste acuity (hypogeusia) has been observed in zinc-deficient subjects with liver disease, malabsorption syndrome, thermal burns, or chronic uremia and in subjects following administration of penicillamine or histidine. Although a double-blind study failed to show the effectiveness of zinc in the treatment of hypogeusia in various diseases (21), another double-blind study indicated that zinc was effective in improving taste acuity in subjects with chronic uremia (22). This discrepancy suggests that depletion of zinc may lead to decreased taste acuity, but not all cases of hypogeusia are due to zinc deficiency. The role of zinc in hypogeusia needs to be delineated further. The dermatological manifestations of severe zinc deficiency include progressive bullous-pustular dermatitis of the extremities and the oral, anal, and genital areas, combined with paronychia and generalized alopecia, such as seen in acrodermatitis enteropathica. Infection with Candida albicans is a frequent complication. These manifestations are seen in patients with severe zinc deficiency. 20 DIAGNOSTIC APPROACHES

Neuropsychiatric signs include irritability, emotional disorders, tremors, and occasional cerebellar ataxia. The patients generally have retarded growth, and males exhibit hypogonadism. Zinc therapy has been shown to produce remarkable improvements and is considered to be a life-saving measure in these subjects. A similar clinical picture has been reported in a patient receiving penicillamine therapy for Wilson's disease (6). After total parenteral nutrition and excessive ingestion of alcohol, clinical manifestations of zinc deficiency resemble acrodermatitis enteropathica. Once a deficiency of zinc is recognized, zinc therapy be- comes imperative in such cases. According to Jameson (23), zinc-deficiency syndrome during pregnancy is characterized by increased maternal morbidity, abnormal taste sensations, prolonged gestation, inefficient labor, atonic bleeding, and increased risk to the fetus, especially postmaturity. A possible correlation between maternal zinc deficiency and congenital malformations, especially of the central nervous system, has been postulated. A high incidence of congenital malformations has been observed in fetuses born of adult women suffering from acrodermatitis enteropathica. In children recovering from severe malnutrition, limitation of lean-tissue synthesis with resultant obesity and a propensity to infection are the major features of a mild zinc deficiency (24).

Laboratory Diagnosis Measurement of the zinc level in plasma is useful, provided that the sample is not hemolyzed or contaminated. In conditions of acute stress, following myocardial infarction or acute infection, zinc from the plasma compartment may redistribute to other tissues, thus making assessment of zinc status in the body difficult (25,26). Intravascular hemolysis would also increase the plasma zinc level inasmuch as the content of red blood cell zinc is much higher than that in plasma. Many investigators have utilized the plasma copper/zinc ratio for clinical assessment in certain diseases (27,28). It has been suggested that an increase in this ratio in patients with malignancy may indicate activity of the disease and poor prognosis. A plasma copper/zinc ratio greater than 2.0 in chronic alcoholics has been associated with a greater incidence of alcoholic hepatitis and cirrhosis of the liver. In another study, a plasma copper/zinc ratio of less than 2.0 was seen in patients who had uncomplicated alcohol withdrawal. Those subjects who had a ratio greater than 2.0 showed delirium tremens and a prolonged, severe hallucinatory state. The plasma copper/zinc ratio also has been utilized to monitor therapeutic response in pulmonary tuberculosis. Although this is an interesting observation, the mechanism of the alteration in the plasma copper/zinc ratio in various disease states is not clear. Further investigations are needed to properly understand and use this ratio as a diagnostic clinical tool. Zinc in the red blood cells and in hair also may be used to assess body zinc status. Because these tissues "turn over" slowly, the zinc levels do not reflect recent changes with respect to body zinc stores. Neutrophil zinc determination, on the DIAGNOSTIC APPROACHES 21 other hand, appears to reflect the body zinc status more accurately and is thus a useful parameter (29). Quantitative assay of alkaline phosphatase activity in the neutrophils is also a useful tool in our experience. Urinary excretion of zinc is decreased as a result of zinc deficiency. Thus, determination of zinc in a 24-hr urine specimen may be of additional help in diagnosing zinc deficiency, provided that the possibility of cirrhosis of the liver, sickle cell disease, chronic renal disease, and other conditions known to cause hyperzincuria is eliminated. Hyperzincuria may be associated with zinc deficiency in the above-mentioned disorders. A metabolic balance study may clearly distinguish zinc-deficient from zinc- sufficient subjects (30). An oral zinc-tolerance test has now been utilized for diagnostic purposes (31,32). In one study, an increased plasma response to a zinc load and a decrease in salivary sediment zinc level were noted after institution of a vegetarian diet, suggesting that this diet may have affected the body zinc status adversely (33). This test may provide useful information with respect to zinc status, but further studies are needed to document its usefulness in clinical medicine. The activities of many zinc-dependent enzymes have been shown to be affected adversely in zinc-deficient tissues. Three enzymes, alkaline phosphatase, carboxy- peptidase, and thymidine kinase, appear to be most sensitive to zinc restriction in that their activities are affected adversely within 3 to 6 days of institution of a zinc- deficient diet to experimental animals. In human studies, the activities of deoxythymidine kinase in proliferating skin collagen and alkaline phosphatase activity in neutrophils were shown to be sensitive to dietary zinc intake. As a practical test, quantitative measurement of alkaline phosphatase activity in neutrophils may be a useful adjunct to the neutrophil zinc-level determination in order to assess body zinc status in man. After supplementation with zinc to deficient subjects, a prompt response in the activities of sensitive enzymes is observed. In one study, the ratio of the maximal posttreatment to pretreatment serum alkaline phosphatase activity correlated inversely with the pretreatment serum zinc level in subjects who received a zinc-restricted diet initially and later were given supplements of zinc (34). The authors suggested that a serial determination of serum alkaline phosphatase and calculation of the alkaline phosphatase ratio during a trial of zinc therapy may provide biochemical confirmation of the adequacy of zinc replacement and may be useful in the detection of mild zinc deficiency.

Biochemistry Studies by Vallee (35) have shown that zinc is a constituent of several metalloenzymes. Keilin and Mann (36) first showed that carbonic anhydrase was a zinc metalloenzyme. It was shown later that there is a single tightly bound zinc atom at the active site of this enzyme. During the next 20 years, only five additional zinc metalloenzymes were identified, and during the last 15 years, the total number has risen to 24. If related enzymes from different species are included, more than 200 zinc-dependent enzymes are on record. Zinc enzymes are known to participate in 22 DIAGNOSTIC APPROACHES many metabolic processes, including carbohydrate, lipid, protein, and nucleic acid synthesis or degradation. The metal is present in several dehydrogenases, aldolases, peptidases, and phosphatases. Zinc is required for both deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) synthesis (25). Many studies show that zinc deficiency in animals impairs the incorporation of labeled thymidine into DNA. This effect has been detected within a few days after the zinc-deficiency diet was begun. Prasad and Oberleas (37) provided evidence that decreased activity of deoxythymidine kinase may be responsible for this early reduction in DNA synthesis and may ultimately relate to growth retardation. As early as 6 days after the animals were placed on the dietary treatment, the activity of deoxythymidine kinase was reduced in rapidly regener- ating connective tissue of zinc-deficient rats, compared to pair-fed controls. These results have been confirmed by other investigators (38). Zinc may also play a role in the maintenance of polynucleotide conformation. An abnormal polysome profile in the liver of zinc-deficient rats and mice has been observed (39). Acute administration of zinc appeared to stimulate polysome formation in vivo and in vitro. The data of Fernandez-Madrid et al. (40), who noted a decrease in the polyribosome content of zinc-deficient connective tissue from rats and concomitant increase in inactive monosomes, support this finding. The role of zinc in gonadal function has been investigated in rats (41). The increase in luteinizing hormone (LH), follicle-stimulating hormone (FSH), and testosterone were assayed after intravenous administration of synthetic LH-releasing hormone (LH-RH) to zinc-deficient and restricted-diet control rats. Their body- weight gain, testicular zinc content, and weight were significantly lower in the zinc- deficient rats than in the controls. The serum LH and FSH responses to LH-RH administration were higher in the zinc-deficient rats, but the serum testosterone response was lower in comparison to the restricted-diet controls. These data indicate a specific effect of zinc on the testes. Similar results have now been reported in experimentally induced zinc-deficient human subjects, sickle cell anemia patients, and chronic uremics who are also zinc-deficient (25). A decrease in sperm count and serum testosterone level was related to testicular failure due to zinc deficiency in these human subjects. Supplementation with zinc resulted in reversal of testicular failure in such cases. Zinc has been used to isolate intact cell membranes and neurotubules from rat brain, suggesting that it may have a role in the stabilization of plasma membranes (42). Zinc prevents induced histamine release from mast cells. This effect of zinc may be due to action on the cell membrane. Platelets are also affected by zinc ions. Collagen-induced aggregation of dog platelets and collagen or epinephrine- induced release of [14C]serotonin were significantly inhibited by zinc. Supplemen- tation of zinc to dogs effectively decreased aggregability of platelets as well as the magnitude of [14C]serotonin release. Several enzymes attached to the plasma membranes control the structures and functions of the membranes, and the activities of these enzymes may be controlled by zinc. Adenosine triphosphatase (ATPase) and phospholipase A2 are inhibited DIAGNOSTIC APPROACHES 23 by zinc, which may explain immobilization of energy-dependent activity of plasma membrane or increased integrity of the membrane structure. In the erythrocyte, excessive intracellular calcium causes hemoglobin retention by erythrocyte membranes as well as erythrocyte membrane shrinkage. Both of these events may occur in sickle erythrocytes because of sickling-induced calcium accumulation. Zinc appears to inhibit both of these events in sickle erythrocytes. Many cell types are activated by calcium and inhibited by zinc (43). Examples include release of histamine by stimulated mast cells, platelet aggregation, and phagocytosis by neutrophils. Zinc has been shown to inhibit the calcium ATPase of the erythrocyte membrane. This enzyme serves as the calcium pump and is stimulated by calcium-activated calmodulin. The exact molecular mechanisms by which zinc inhibits calmodulin are not well understood. Studies clearly indicate that zinc is required for lymphocyte transformation. Since Alford (44) first discussed the essentiality of zinc for phytohemagglutinin- induced transformation of human peripheral blood lymphocytes, several other laboratories have confirmed this observation. The effect of zinc appears to be that of a mitogen, and the kinetics of these influences most closely approximate the effects of antigen stimulation on lymphocyte culture. Currently available data suggest a direct stimulatory influence of zinc on DNA metabolism, either by enzyme stimulation or by altering the binding of F, and F3 histones to DNA so as to affect RNA synthesis. A direct cell surface effect of zinc, however, cannot be ruled out. Zinc could be operating at several levels to influence lymphocyte monoclonal proliferation. Assessment of the role of zinc in the development and functions of different lymphoid cell populations strongly indicates that this element has an effect predom- inantly on T-lymphocytes (45). A monocyte factor may be required for the zinc- induced mitogenic response of T-lymphocytes (46). It has been known for many years that zinc deficiency in experimental animals results in atrophy of thymic and lymphoid tissues and lymphopenia. Fraker et al. (47) first showed that severely and marginally zinc-deficient young adult A/Jax mice have involuted thymuses and reduced ability to form antibodies to sheep red blood cells. Iwata et al. (48) have shown that one obvious hormonal effect on thymocytes of zinc deficiency in mice and humans is reduction of thymic hormone levels, which would clearly be another discrete local effect on thymocyte maturation. Frost et al. (49) have confirmed Fraker's observation of the reduction by zinc deficiency of T-cell-mediated antibody production but have also noted that the antibody production that occurs in deficient BALB/c mice is prolonged compared to controls. This phenomenon may be explained by inhibition of T-cell-suppressive function due to zinc deficiency. Deficiency of the purine enzyme nucleoside phosphorylase is associated with a severe T-cell-immune deficiency. Our studies in experimental animals indicate that nucleoside phosphorylase may be zinc-dependent. Thus, a decreased activity of nucleoside phosphorylase may additionally account for T-cell dysfunction in zinc 24 DIAGNOSTIC APPROACHES

deficiency (50). It is clear that zinc deficiency exerts a profound and apparently specific effect on the thymus, thymocytes, and cellular immune functions that is reversible with zinc repletion. The serum of animals or humans undergoing bacterial infection contains a substance called leukocyte endogenous mediator (LEM), which causes a flux of serum zinc, amino acids, and iron into the liver within a matter of hours after activation of phagocytic cells (51,52). LEM is a low-molecular-weight protein produced by activated macrophages and granulocytes and may be identical to endogenous pyrogen (EP). The decline of plasma zinc is due to redistribution of zinc from the plasma pool to the liver. Hepatic uptake of zinc is accompanied by LEM-mediated synthesis by the liver of a2-acute-phase reactant proteins, such as ceruloplasmin, fibrinogen, arantitrypsin, and haptoglobin. Serum copper increases because of hepatic synthesis and release of ceruloplasmin. LEM also appears to mediate the release of neutrophils from the bone marrow. An increased number of blood neutrophils can be detected within 8 hr after parenteral injection of LEM. These alterations of zinc in acute stress may be significant inasmuch as the reduced level of plasma zinc may increase the nonspecific activity of local phagocytosis. Briggs et al. (53) showed that granulocytes from chronic uremics who were zinc- deficient showed significantly impaired mobility, both chemotactic and kinetic, in comparison to those subjects who were supplemented with zinc. Furthermore, a significant correlation between granulocyte chemotaxis and plasma and granulocyte zinc concentrations among all patients supports a pathophysiological relationship between the severity of impaired granulocytic chemotactic response and zinc deficiency in these patients. Abnormal granulocyte chemotaxis, corrected by zinc supplementation, has been observed by others in nonuremic patients with acrodermatitis enteropathica. It appears, therefore, that although the two neutrophil functions, phagocytosis and chemotaxis, are zinc-dependent, their requirements for zinc are different. Zinc may also intervene in nonenzymic, free-radical reactions. In particular, zinc protects against iron-catalyzed free-radical damage. The free-radical oxidation (auto-oxidation) of polyunsaturated lipids is most effectively induced by the interaction of inorganic iron, oxygen, and various redox couples. Other work suggests that this interaction underlies the pathological changes and clinical manifestations of iron toxicity. Zinc, ceruloplasmin, metalloenzymes (catalase, peroxidases, and superoxide dismutase) and free-radical scavenging antioxidant vitamin E inhibit iron-catalyzed, free-radical oxidation. Bettger and O'Dell (54) observed that chicks fed a low-zinc diet developed severe skin lesions on the toes and footpads as well as gross joint abnormalities. Supple- mentation of the zinc-deficient diet with vitamin E significantly reduced the severity of the skin and joint pathology but had no effect on the decreased rate of growth in the zinc-deficient chicks. Their observations support the hypothesis that zinc plays a biochemical role analogous to that of vitamin E by stabilizing membrane structure and thereby reducing peroxidative damage to the cell. Further studies in the red blood cell suggested that zinc was not acting as an antioxidant but as a DIAGNOSTIC APPROACHES 25 stabilizer of red blood cell membrane against damaging events that occur after peroxidation. Carbon tetrachloride-induced liver injury is another animal model for studying free-radical injury to tissues. Animals maintained on a high zinc regimen are resistant to this type of biochemical injury, suggesting that zinc may be protective against free-radical injury. Other studies have shown that zinc also inhibits the analogous metramidazole-dependent, free-radical sequence. Some of the observed physiological effects of zinc may be related to competition between zinc and several cations such as cadmium, lead, calcium, and copper in vitro and in vivo. Beneficial effects of zinc on ameliorating toxicities of cadmium and lead, accentuation of zinc deficiency by administration of calcium and phytate, and production of hypocupremia by excessive zinc intake in human and animal models are some of the examples of competition phenomena.

COPPER DEFICIENCY

Etiology Nutritional copper deficiency was first described by Cordano et al. (55) in malnourished Peruvian children who were being fed a formula composed of diluted cow's milk, sucrose, and cottonseed oil. These infants developed neutropenia as well as anemia which was unresponsive to iron supplementation. Serum copper and ceruloplasmin were decreased, and correction of anemia was observed when patients were supplemented with copper. Subsequently, copper deficiency was described in a premature infant fed a high- calorie, low-copper diet for 6 weeks (56). During the past decade other examples of copper deficiency in human subjects in both infants and adults receiving total parenteral nutrition without copper supplementation have been recorded in the literature (57). Hypoceruloplasminemia, hypocupremia, neutropenia, and microcytosis have been observed in patients with sickle cell disease who were treated with an intensive regimen of zinc acetate (58). Copper supplementation corrected the above manifestations. Menkes' disease, an X-linked disorder, was first described in 1962 (59), and the importance of copper in its pathogenesis was demonstrated in 1972 (60). The disease is characterized by growth retardation, white hair with peculiar twisting, brittleness, lack of pigmentation, seborrheic dermatitis, scorbutic bone changes, arterial tortuosity and aneurysms, bladder diverticula, pulmonary emphysema, and neurological abnormalities which include mental retardation, seizures, hypothermia, intracranial hemorrhage, and optic nerve atrophy. Excretion of copper in the urine is decreased, intestinal absorption of copper is impaired, and hepatic copper content is decreased in Menkes' disease. Red blood cell copper is normal, and intravenous copper is handled normally. The use of a wide range of copper preparations, however, did not result in clinical improvement, although an increase in serum copper and ceruloplasmin was observed. 26 DIAGNOSTIC APPROACHES

Lott et al. (61) employed high oral doses of copper histidine and observed that although the serum copper was increased serum ceruloplasmin did not change, suggesting that the abnormality of Menkes' disease may reside in a defective transport protein, making copper unavailable for holoceruloplasmin biosynthesis. The primary problem in absorption relates to the transport of copper across the gut mucosal cell rather than failure of uptake from the intestinal lumen (57). Menkes' fibroblasts take up MCu for a longer period of time and accumulate three times as much radioactivity as normal. In other experiments it was shown that the cells released ^Cu less rapidly. The radioactive copper was bound to a small- molecular-weight protein (10,000 daltons). It has been postulated that the basic metabolic defect in the genetic copper deficiency disease relates to the rate of synthesis, and/or degradation of a thionein-like protein in some cells (62). A complete understanding of the metabolic defect in Menkes' disease patients may provide the key to therapy of this disease.

Clinical Manifestations In animals, manifestations of copper deficiency include anemia, neutropenia, hypopigmentation, defective wool keratinization, ataxia with defective myelination, abnormal bone formation with spontaneous fractures, reproductive failure, heart failure, and arterial and cardiac aneurysms (57,62). In humans, hypochromic anemia and neutropenia have been observed. Although neutropenia remains poorly explained, the anemia appears to be partly related to defects of iron mobilization due to impaired ceruloplasmin ferroxidase activity, as well as defective intracellular iron utilization. In patients with Menkes' disease, hypopigmentation of hair, most likely due to decreased activity of tyrosinase, a copper-dependent enzyme, has been noted. Copper supplementation is effective therapy for hypopigmentation. The abnormality of hair observed in Menkes' disease and the uncrimped wool noted in copper- deficient sheep do not appear to be due to decreased lysyl oxidase activity but, rather, to defective formation of disulfide bonds. In infants with dietary copper deficiency and in patients with Menkes' disease, osteoporosis and pathological bone fractures as well as scurvy-like abnormalities of the bones have been noted. Vascular abnormalities have also been observed in patients with Menkes' disease. These findings are similar to the connective tissue and vascular abnormalities observed in copper-deficient animals. In experimental animals, bone defects are associated with a higher proportion of soluble collagen within the organic matrix. Neurological manifestations have not been described as a feature of uncomplicated human copper deficiency. However, a number of neurological symptoms have been observed in patients with Menkes' disease, including hypotonia, hypothermia, episodic apnea, seizures, and mental retardation. These manifestations may be related to decreased cytochrome oxidase activity within the central nervous system, impaired catecholamine synthesis, or alterations of lipid composition, or to some alternate mechanism. DIAGNOSTIC APPROACHES 27

Hypercholesterolemia in copper-deficient animals has been reported. The pathophysiological basis for these biochemical changes is unclear, and the clinical asso- ciations remain to be demonstrated.

Diagnosis Anemia in copper deficiency is hypochromic and microcytic and does not respond to iron administration. Neutropenia is also seen frequently with copper deficiency. As mentioned earlier these hematological disorders are not seen in Menkes' disease. Copper supplementation results in correction of both anemia and neutropenia. Decreased levels of serum copper and ceruloplasmin, urinary copper, and hepatic copper are seen in copper-deficient human subjects. Although levels of serum copper and ceruloplasmin are also decreased in patients with Wilson's disease, a condition characterized by excessive copper storage, hepatic and urinary copper levels are increased. Inasmuch as several enzymes are known to be copper-dependent, measurement of their activities in suitable tissues may be diagnostic of copper deficiency in human subjects. Measurement of cytochrome oxidase in hematopoietic cells and/ or the liver, lysyl oxidase in the collagen connective tissue, and superoxide dismutase in the neutrophils may be diagnostic of copper deficiency in human subjects. Further studies are needed to document their usefulness in clinical diagnosis. Characteristics of MCu uptake by cultured fibroblasts may be useful in distinguish- ing nutritional copper deficiency from Menkes' disease and copper storage disorders. Such data are not available in the literature, and future studies are required.

Biochemistry Lack of myelination has been observed in nutritional copper deficiency and Menkes' disease (62). A marked decrease in the activity of the myelin marker 2', 3'-cyclic nucleotide-3'-phosphohydrolase in weanling rats whose mothers were deprived of copper has been reported. Whether the reduction in myelin is the result of a specific metabolic defect or simply failure of nerve cell production and survival remains unknown. Nutritional copper deficiency results in decreased norepinephrine levels in brain except the hypothalamus in experimental animals. Dopamine, a direct precursor of norepinephrine, is unchanged in the brains of the MObr mouse (an animal model of Menkes' disease), although a decrease has been observed in the brains of ataxic copper-deficient lambs and second-generation copper-deficient rats. The low norepinephrine level is probably due to impaired activity of the copper- dependent dopamine-(3-hydroxylase (DBH) enzyme. The norepinephrine level is restored to normal levels in vivo by copper administration; on the other hand, depressed dopamine levels are not readily reversible, suggesting a structural rather than an enzymatic defect due to copper deficiency. 28 DIAGNOSTIC APPROACHES

A decreased dopamine level may be due to decreased activity of the enzyme tyrosine hydroxylase. Its activity in the copper-deficient rat brain is decreased in proportion to the lowered dopamine level, but in the MObr mouse it is significantly lowered. Although it is not known if tyrosine hydroxylase is copper-dependent, tyrosinase, which performs a similar function, is copper-dependent. Tyrosinase is not present in brain. Lack of tyrosinase probably accounts for the lack of pigmentation in Menkes' disease and nutritionally copper-deficient animals. Another enzyme deficiency which may contribute to the low norepinephrine level is cytosolic superoxide dismutase, a copper- and zinc-dependent enzyme which promotes au- toxidation of catecholamines and related compounds. The activity of cytochrome oxidase, a copper-dependent enzyme, is decreased in most tissues of copper-deficient animals. Clearly a severe deficiency of cytochrome oxidase, which is of key importance in cell energetics, could cause cell death or impair cell function in general. However, the normal excess of cytochromes a and a3 over other components of the electron transport chain argues against their becoming limiting in the brain under usual conditions of copper deficiency. The metabolic defect responsible for dissecting aneurysm and angiorrhexis is failure of cross-link formation in the collagen and elastin due to decreased activity of lysyloxidase, a copper enzyme. The emphysema-like lung that occurs during development under conditions of copper deficiency appears to also result from failure of cross-link formation.

CHROMIUM DEFICIENCY

Etiology Nutritional chromium deficiency has been suspected to occur in children living in refugee camps in Jordan and in malnourished children from Nigeria and Turkey (63). A few cases of chromium deficiency have been reported in subjects receiving long-term total parenteral nutrition (64). Multiparous women appear to have lower chromium stores than nulliparae. Inasmuch as in many countries nutritional intake of chromium may not be adequate, repeated pregnancies may further stress the nutritional chromium status of pregnant women. Whether deterioration of glucose tolerance in pregnancy is related to chromium deficiency is not known. Body store of chromium is known to decline with aging. This may be an expression of suboptimal chromium intake, but a metabolic defect in chromium metabolism due to aging has not been ruled out. Further studies are required to relate chromium status in older subjects with abnormal oral glucose tolerance seen so frequently in older-age-onset diabetics. Chromium deficiency should be suspected in diabetic cases where unexplained insulin resistance develops (63). It has been reported that insulin (or stimuli that induce secretion of insulin) may mobilize chromium from unidentifiable tissue stores. This leads to acute increases in plasma chromium levels and increased DIAGNOSTIC APPROACHES 29 excretion of chromium in the urine. Thus in cases of hyperinsulinemia the chromium requirement is increased.

Clinical Manifestations Impaired glucose tolerance is the major clinical manifestation of chromium deficiency. In a controlled study of 14 malnourished children in Turkey, the oral administration of 250 (xg of chromium chloride, corresponding to approximately 50 (xg of chromium, resulted in striking improvement of the impaired glucose tolerance in nine subjects (63). It has been hypothesized that chromium deficiency may be a risk factor in atherosclerotic heart disease. Chromium supplementation in an uncontrolled fashion has shown a decrease in serum cholesterol and a significant increase in HDL cholesterol level in adult subjects. Two additional epidemiological studies lend support to this hypothesis. One study carried out in Finland found a strong negative correlation between cardiovascular morbidity and mortality and chromium. In another study low serum chromium levels correlated with an increased incidence of coronary artery disease. Further studies are required to understand the biochemical mechanism of the role of chromium in lipid metabolism and its significance in ischemic heart disease.

Diagnosis Chromium determination in serum and other biological samples is not a readily available laboratory procedure. The results of such assays are not comparable from one laboratory to the other, and assays are not well standardized. No single test is currently available to assess the chromium status of individuals. Chromium status can be determined only retrospectively depending on the clinical response of prolonged chromium supplementation in human subjects. Dietary analysis of chromium in a given population may provide some knowledge with respect to adequacy of chromium nutrition. Normal intake for adults is 50 to 200 jxg/day. A consistent intake below this range in any population increases the risk of chromium deficiency. Analyses of hair, serum, liver, and urine for chromium have been utilized for assessment of chromium status. However, because of uncertainty of the techniques used for assay, it is not possible to relate the levels of chromium in these samples to body status of chromium.

Biochemistry Chromium is believed to be a cofactor for insulin. Chromium in the form of the naturally occurring dinicotinic acid glutathione complex (glucose tolerance factor) increased the effect of exogenous insulin on glucose metabolism of epididymal adipose tissue in vitro. The effect of chromium is lacking in the absence of insulin. 30 DIAGNOSTIC APPROACHES

SELENIUM DEFICIENCY

Etiology and Clinical Manifestations New Zealand residents have naturally lower blood selenium levels and glutathione peroxidase activities in erythrocytes and whole blood than do residents in most other countries at all stages of life (65). However, the general good health of New Zealanders makes it hardly justifiable to regard them as clinically deficient despite their low selenium status. The first evidence of a possible human need for selenium was noted in children with protein-energy malnutrition. A growth response to selenium supplementation was reported in two children suffering from kwashiorkor who had previously failed to respond to the usual dietary treatment. It was later noted that these children were being treated with New Zealand skim milk powder, and therefore they were probably low in selenium. A growth response was also observed in two children from Jordan. Unfortunately, neither group had a nontreated control group. Keshan disease from China has been described, and selenium deficiency has been implicated in its pathogenesis (65,66). In this condition, the heart is severely fibrosed after changes in the mitochondria leading to necrosis of myocardial cells, heart failure, and death. Sodium selenite supplementation was reported to be effective in preventing Keshan disease. Low selenium in hair and blood, decreased glutathione peroxidase in erythrocytes, and decreased selenium intake were seen in these subjects. Lombeck et al. (67) from West Germany found no evidence of cardiomyopathy in children consuming selenium 1 to 5 jig/day (normal intake approximately 100 jig/day), despite a low selenium status as judged by the selenium content of blood. Thus in view of the fact that there is no correlation between low selenium intake and cardiomyopathy in areas such as New Zealand and Finland, one must presume that factors in addition to low selenium intake are responsible for cardiomyopathy in China. In one subject on total parenteral nutrition, blood selenium decreased to low levels and the patient developed pain in the thigh muscles (65). Muscular tenderness was also noted. These symptoms improved after selenium supplementation. These features, however, have not been observed by other investigators in subjects on total parenteral nutrition despite low selenium status. Thus, at present, it is not possible to relate these manifestations to selenium deficiency. Westermack and Sandhold (68) reported a transitory clinical response in patients with neuronal ceroid lipofuscinosis. These findings have not been confirmed by other investigators. Low selenium status has been observed in many chronic diseases, including malignancy, hypertension, artherosclerosis, alcoholic cirrhosis, arthritis, muscular dystrophy, cystic fibrosis, infertility, aging, macular degeneration, cataract, and diabetic retinopathy. No causal relationship between selenium and the above-mentioned disorders has been established as yet. The clinical relevance of low selenium status in chronic diseases needs further study. DIAGNOSTIC APPROACHES 31

Diagnosis The best method to measure selenium in biological samples appears to be spectrofluorimetry. Neutron activation analysis also appears to be very accurate, but the equipment is not readily available. Blood selenium levels reliably reflect dietary intake of selenium. Selenium status has also been assessed by measuring selenium in serum, erythrocytes, hair, toenails, fingernails, and other tissues at autopsy. Selenium in hair, nails, and erythrocytes reflects the chronic status of selenium. Glutathione peroxidase activity has been used as a measure of biological availability of selenium, and its activity in erythrocytes is assayed routinely by many laboratories. Some investigators prefer to assay glutathione peroxidase activity in platelets.

Biochemistry Glutathione peroxidase is a selenium-containing enzyme. The main metabolic role of selenium is via glutathione peroxidase to help protect the cell against oxidant damage, along with other cellular antioxidant defense mechanisms such as vitamin E, catalase, and superoxide dismutase. The activity of glutathione peroxidase is consistently decreased in various tissues of selenium-deficient animals and man. Selenium may have a role in hepatic heme metabolism that is unrelated to glutathione peroxidase. Induction of hepatic cytochrome P450 by phenobarbital treatment was impaired in chronically selenium-deficient rats. Because phenobarbital stimulates heme oxygenase in deficient but not supplemented rats, impaired cytochrome P450 induction in the deficient rats appeared to be related to increased catabolism rather than decreased synthesis of heme (66). Several selenium-containing enzymes such as glycine reductase, formic dehydrogenase, and possibly nicotinic acid hydroxylase, xanthine dehydrogenase, and thio- lase have been found in microorganisms. The identification of so many microbial selenoproteins suggests that more than one selenoprotein of biological importance (glutathione peroxidase) may exist in mammals.

ACKNOWLEDGMENTS Supported in part by grants from the National Heart, Lung, and Blood Institute; the Sickle Cell Center; the National Institute of Arthritis, Metabolic, and Digestive Diseases, NIH; and the Veterans Administration Research Service.

REFERENCES 1. Prasad AS, Halsted JA, Nadimi M. Syndrome of iron deficiency anemia, hepatosplenomegaly, hypogonadism, dwarfism, and geophagia. Am J Med 1961;31:532. 2. Prasad AS, Miale A, Farid Z, Sandstead HH, Schulert AR, Darby WJ. Biochemical studies on dwarfism, hypogonadism, and anemia. Arch Intern Med 1963;lll:407. 3. Prasad AS, Miale A, Farid Z, Schulert A, Sandstead HH. Zinc metabolism in patients with the syndrome of iron deficiency anemia, dwarfism, and hypogonadism. J Lab Clin Med 1963;61:537. 32 DIAGNOSTIC APPROACHES

4. Prasad AS. Trace elements and iron in human metabolism. New York: Plenum, 1978:251. 5. Casey CE, Hambidge KM. Epidemiological aspects of human zinc deficiency. In: Nriagu JO, ed. Zinc in the environment. New York: Wiley, 1980:1. 6. Klingberg WG, Prasad AS, Oberleas D. Zinc deficiency following penicillamine therapy. In: Prasad AS, ed. Trace elements in human health and disease. New York: Academic Press, 1976;51. 7. Victery W, Levenson R, Vander AJ. Effect of glucagon on zinc excretion in anesthetized dogs. Am J Physiol 1981;240:F299. 8. Prasad AS. Deficiency of zinc in man and its toxicity. In: Prasad AS ed. Trace elements in human health and disease. New York: Academic Press, 1976; 1. 9. Gudbjarnason A, Prasad AS. Cardiac metabolism in experimental alcoholism. In: Sardesai V, ed. Biochemical and clinical aspects of alcohol metabolism. Springfield, 111: Charles C Thomas, 1969:266. 10. Allan JG, Fell GS, Russell RI. Urinary zinc in hepatic cirrhosis. Scott Med J 1975;20:109. 11. Prasad AS, Schoomaker EB, Ortega J, Brewer GJ, Oberleas D, Oelshlegel FJ. Zinc deficiency in sickle cell disease. Clin Chem 1975;21:582. 12. Prasad AS, Abbasi A, Ortega J. Zinc deficiency in man: studies in sickle cell disease. In: Brewer G, Prasad AS, eds. Zinc metabolism: current aspects in health and disease. New York: Alan R. Liss, 1977:211. 13. Barnes PM, Moynahan EJ. Zinc deficiency in acrodermatitis enteropathica: multiple dietary intolerance treated with synthetic diet. Proc R Soc Med 1973;66:327. 14. Prasad AS. Zinc deficiency and effects of zinc supplementation on sickle cell anemia subjects. In: Brewer GJ, ed. The red cell: fifth Ann Arbor conference. New York: Alan R. Liss, 1981:99. 15. Greene HL. Trace metals in parenteral nutrition. In: Brewer GJ, Prasad AS, eds. Zinc metabolism: current aspects in health and disease. New York: Alan R. Liss, 1977:87. 16. Morrison SA, Russell RM, Carney EA, Oaks EV Zinc deficiency: a cause of abnormal dark adaptation in cirrhotics. Am J Clin Nutr 1978;31:276. 17. Warth JA, Prasad AS, Zwas F, Frank RN. Abnormal dark adaptation in sickle cell anemia. J Lab Clin Med 1981 ;98:189. 18. Sandstead HH, Vo-Khactu KP, Solomons N. Conditioned zinc deficiencies. In: Prasad AS, ed. Trace elements in human health and disease, Vol. 1. New York: Academic Press, 1976:3. 19. McClain C, Soutor C, Zierge L. Zinc deficiency: a complication of Crohn's disease. Gastroen- terology 1980;78:272. 20. Henkin RI, Bradley DF. Regulation of taste acuity by thiols and metal ions. Proc Natl Acad Sci USA 1969;62:30. 21. Henkin RI, Scheichter PJ, Friedewald WT, Demets DL, Raff M. A double-blind study of the effects of zinc sulfate on taste and smell function. Am J Med Sci 1976;272:285. 22. Mahajan SK, Prasad AS, Lambujon J, Abbasi AA, Briggs WA, McDonald FD. Improvement of uremic hypogeusia by zinc: a double-blind study. Am J Clin Nutr 1980;33:1517. 23. Jameson S. Zinc and pregnancy. In: Nriagu JO, ed. Zinc in the environment. Part II. Health effects. New York: Wiley, 1980:183. 24. Golden MHN, Golden BE. Effect of zinc supplementation on the dietary intake, rate of weight gain, and energy cost of tissue deposition in children recovering from severe malnutrition. Am J Clin Nutr 1981;34:9O0. 25. Prasad AS. Clinical and biochemical spectrum of zinc deficiency in human subjects. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982:3. 26. Prasad AS. Clinical disorders of zinc deficiency. New York: Alan R. Liss, 1982:89. 27. Bogden JD. Blood zinc in health and disease. In: Nriagu JO, ed. Zinc in the environment. New York: Wiley, 1980:138. 28. Karcioglu Z, Karcioglu G, Sarper RM, Hargovcic M. Zinc and copper in neoplastic diseases. In: Karcioglu Z, Sarper RM, eds. Zinc and copper in medicine. Springfield, 111: Charles C Thomas, 1980:464. 29. Prasad AS, Cossack ZT. Neutrophil zinc: an indicator of zinc status in man. Trans Assoc Am Physician 1982;95:165. 30. Prasad AS, Rabbani P, Abbasi A, Bowersox E, Fox MRS. Experimental zinc deficiency in humans. Ann Intern Med 1978;89:483. 31. Oelshlegel FJ Jr, Brewer GJ. Absorption of pharmacologic doses of zinc. In: Brewer GJ, Prasad DIAGNOSTIC APPROACHES 33

AS, eds. Zinc metabolism: Current aspects in health and disease. New York: Alan R. Liss, 1977:299. 32. Sullivan JF, Jetton MM, Burch RE. A zinc tolerance test. J Lab Clin Med 1979;93:485. 33. Freeland-Graves JH, Ebaugh L, Hendrikson PJ. Alterations in zinc absorption and salivary sediment zinc after a lacto-vegetarian diet. Am J Clin Nutr 1980;33:1757. 34. Kasarskis FJ, Schune A. Serum alkaline phosphatase after treatment of zinc deficiency in humans. Am J Clin Nutr 1980;33:2609. 35. Vallee BL. Biochemistry, physiology and pathology of zinc. Physiol Rev 1959;39:443. 36. Keilin D, Mann J. Carbonic anhydrase: purification and nature of the enzyme. Biochem J 1940;34:1163. 37. Prasad AS, Oberleas D. Thymidine kinase activity and incorporation of thymidine into DNA in zinc-deficient tissue. J Lab Clin Med 1974;83:634. 38. Dreosti IE, Hurley LS. Depressed thymidine kinase activity in zinc-deficient rat embryos. Proc Soc Exp Biol Med 1975; 150:161. 39. Sandstead HH, Hollaway WL, Baum V Zinc deficiency: effect on polysomes. Fed Proc 1971;30:517. 40. Fernandez-Madrid F, Prasad AS, Oberleas D. Effect of zinc deficiency on nucleic acids, collagen, and noncollagenous protein of the connective tissue. J Lab Clin Med 1973;82:951. 41. Lei KY, Abbasi A, Prasad AS. Function of the pituitary-gonadal axis in zinc-deficient rats. Am J Physiol 1976;23O:173O. 42. Chvapil M. Effect of zinc on cells and biomembranes. Med Clin North Am 1976;60:799. 43. Brewer GJ. Calmodulin, zinc and calcium in cellular and membrane regulation: an interpretive review. Am J Hematol 1980;8:231. 44. Alford RM. Metal cation requirements for phytohemagglutinin-induced transformation of human peripheral blood leukocytes. J Immunol 1970;104:698. 45. Good RA, Fernandes G. Nutrition, immunity, and cancer—a review. Clin Bull 1979;9:3. 46. Ruhl H, Kirchner H. Monocyte-dependent stimulation of human T cells by zinc. Clin Exp Immunol 1978;32:484. 47. Fraker PJ, Haas S, Luecke RW Effect of zinc deficiency on the immune response of the young adult A/Jax mouse. J Nutr 1977;107:1889. 48. Iwata T, Incefy G, Tanaka T, et al. Circulatory thymic hormone levels in zinc-deficiency. Cell Immunol 1979;47:100. 49. Frost P, Chen JC, Rabbani I, Smith J, Prasad AS. The effects of zinc deficiency on the immune response. In: Brewer GJ, Prasad AS, eds. Zinc metabolism: current aspects in health and disease. New York: Alan R. Liss, 1977;143. 50. Prasad AS, Rabbani P Nucleoside phosphorylase in zinc deficiency. Trans Assoc Am Physician 198I;94:314. 51. Beisel WR, Pekarek RS, Wannemacher RW Jr. Homeostatic mechanisms affecting plasma zinc levels in acute stress. In: Prasad AS, ed. Trace elements in human health and disease. Vol. I. New York: Academic Press, 1976:87. 52. Wannemacher RW, DuPont ML, Pekarek RS. An endogenous mediator of depression of amino acids and trace elements during typhoid fever. J Infect Disl972; 126:77. 53. Briggs WA, Pedersen MM, Mahajan SK, Sillix DH, Prasad AS, McDonald FD. Lymphocyte and granulocyte function in zinc-treated and zinc-deficient hemodialysis patients. Kidney Int 1982;21:827. 54. Bettger W, O'Dell BL. A critical physiological role of zinc in the structure and function of biomembranes. Life Sci 1981 ;28:1425. 55. Cordano A, Baertl JM, Graham GG. Copper deficiency in infancy. Pediatrics 1966;26:326. 56. Al-Rashid RA, Spangler J. Neonatal copper deficiency. N Engl J Med 1971;285:841. 57. Williams DM. Clinical significance of copper deficiency and toxicity in the world population. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982:277. 58. Prasad AS, Brewer GJ, Schoomaker EB, Rabbani R Hypocupremia induced by zinc therapy in adults. JAMA 1978;240:2166. 59. Menkes JH, Alter M, Steigleder GK, Weakley DR, Sung JH. A sex linked recessive disorder with retardation of growth, peculiar hair and focal cerebral and cerebellar degeneration. Pediatrics 1962;29:764. 60. Danks DM, Campbell PE, Stevens BJ, Mayne V, Cartwright E. Menkes' kinky hair syndrome: an inherited defect in copper absorption with widespread effects. Pediatrics 1972:50:188. 61. Lott IT, DiPaolo R, Schwartz D, Janowska S, Kanfer JN. Copper metabolism in the steely hair syndrome. N Engl J Med 1975;292:197. 34 DIAGNOSTIC APPROACHES

62. O'Dell BL. Biochemical basis of the clinical effects of copper deficiency. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982:301. 63. Mertz W. Clinical and public health significance of chromium. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982:315. 64. fcejeebhoy KN, Cher RG, Marliss EB, Greenberg GR, Bruce-Robertson A. Chromium deficiency, glucose intolerance, and neuropathy reversed by chromium supplementation in a patient receiving long-term parenteral nutrition. Am J Clin Nutr 1977;30:531. 65. Robinson ME Clinical effects of selenium deficiency and excess. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982:325. 66. Levander OA. Selenium: biochemical actions, interactions, and some human health implications. In: Prasad AS, ed. Clinical, biochemical, and nutritional aspects of trace elements. New York: Alan R. Liss, 1982; 345. 67. Lombeck I, Kasperek K, Feinendegen LE, Bremer HJ. Low selenium state in children. In: Spallholz JE, Martin JL, Ganther HE, eds. Selenium in biology and medicine. Westport: AVI Publishing, 1981:269. 68. Westermack T, Sandhold M. Decreased erythrocyte glutathione peroxidase activity in neuronal ceroid lipofuscinosis corrected with selenium supplementation. Acta Pharmacol Toxicol (Copenh) 1977;40:70.

DISCUSSION Dr. Chandra: Many "stresses" affect trace element levels. In most patients suspected of trace element deficiencies, there is usually some other primary disease, e.g., infection or malabsorption, cancer, renal disease, and so on. Are there any precautions that can be taken when interpreting these results if such an additional stress is present? Dr. Prasad: It is my understanding that in acute stress there is a transfer of zinc from the plasma pool to the liver, and it is related to liberation of lymphokines. This is not the same in patients with chronic infection and debilitation. Thus the significance of low plasma zinc in many patients must be interpreted cautiously. Dr. Zlotkin: Dr. Prasad, would you discuss in greater detail how clinicians should cope with the diagnosis and treatment of trace metal deficiencies? For example, having been tantalized by the prospect of using neutrophil trace metal content for the diagnosis of trace metal deficiencies, I have been trying to develop a micromethod for neutrophil zinc analysis. Unfortunately, my efforts have been totally unsuccessful. It is my impression that in order to use this test in pediatrics one must have access to a minimum of 10 ml of blood and preferably 20 to 50 ml. Quite obviously, obtaining this amount of blood is totally impossible. I would appreciate your comments on the practicality of using this rather new diagnostic tool in pediatrics, and especially in small infants who are potentially at high risk of developing zinc deficiency. Unfortunately Dr. Prasad did not mention in his verbal presentation the use of hair analysis for the determination of trace metal status. In North America, hair trace element analysis has become a popular business for a great number of nonprofessionals interested more in making money than in diagnosing trace metal deficiencies. One of the major advantages of hair analysis is that it is easy to obtain from the patient, easy to transfer from the patient to the laboratory, and easy to work with. One usually does not even need human experimen- tation committee approval to take hair samples from most subjects. There are, however, a number of major problems especially in the commercial use of multielement analysis. For example, contamination can occur from any one of a number of sources. I must emphasize that in performing hair analysis we are interested in the endogenous content of the trace metal in the hair as opposed to the content of the trace metal that is found on the surface of the hair. The first problem encountered therefore would be DIAGNOSTIC APPROACHES 35 exposure of the sample to and contamination from the external environment. Even if the hair is sampled from as close as 1 to 2 cm from the scalp, because hair normally grows on the adult patient at about 1 cm per month the hair will have been exposed to the external environment for a fairly long period of time prior to sampling. The choice and the effectiveness of sample washing procedures also have a major effect on environmental contamination. Whether one choses a non-ionic wash or an organic wash, all have a major effect on the amount of trace metal actually washed from the hair prior to analysis. Many shampoos contain selenium and/or zinc. This also must be controlled for when interpreting analytical results. Variations with hair color, location, diameter, sex, or season all seem to affect what we call normal levels of trace metals within the hair. Variations with age are certainly very important. If one is willing to look at an age-corrected normal distribution of hair trace metal values and all the other variables are taken into consideration, then one might have some faith in the results. Certainly as a research tool, hair trace metal analysis may be well used. Rates of hair growth are of major importance, as we know that in protein-calorie malnutrition, or in zinc deficiency itself, hair may in fact stop growing. Under these circumstances the results of hair analyses may be impossible to interpret. There are certain problems associated with analytical techniques. In North America when one pays $200 to obtain hair analysis, one receives an analysis of anywhere up to 25 or 30 elements. The technique is called multielement analysis. I know how difficult it is in my own laboratory to do an analysis of one element at a time, so I have great misgivings about claims that 25 or 30 elements can be measured on one strand of hair at one time. The correlation of hair trace element concentrations with other body tissues or fluid levels is also worth considering. We know from research in animals and humans that there are correlations between hair zinc and copper and the content of zinc and copper in various body tissues. However, I think it is fair to say that for most of the other trace elements we have no data on correlations between hair values and concentrations in other organs within the body. I do think, though, that hair analysis can be used as a good research tool and for public health studies where one is interested in the detection of toxic levels of a couple of minerals such as lead, cadmium, or mercury and population studies where one is not particularly interested in individual values but trends; in that case, I think hair analysis can in fact be reasonably used. Dr. Prasad: I did not include hair in my presentation because I am not sure what it really means, particularly if you are trying to assess acute status with respect to zinc nutrition. Now, with respect to what can we do in an infant where you cannot get enough neutrophils because you cannot draw 20 ml of blood for assay, you should determine plasma zinc and copper and perhaps use the reciprocal relationship between copper and zinc for assessment of zinc status. Urinary zinc levels could be another parameter. If the urinary zinc excretion is decreased, one might use this finding as additional evidence for deficiency. With respect to the enzymes, it may be possible to assay the neutrophil alkaline phosphatase activity in the infant despite the small blood sample size. I recommend that the activity of this enzyme be assayed quantitatively, not qualitatively. The reason I emphasize neutrophil alkaline phosphatase is because this isoenzyme has been shown to be the zinc metalloenzyme. In erythrocytes there are two enzymes that may be assessed: One is nucleoside phosphorylase, and the other is 8-aminolevulinic acid dehydratase. Dr. Thompson: Sometimes we talk about deficiency when we mean whole body depletion. For instance, loss of weight reduces body zinc or, as in liver disease, the liver is depleted of zinc but the body is not necessarily zinc-depleted. Perhaps "depletion" should be defined 36 DIAGNOSTIC APPROACHES as an appropriate reduction of zinc content, and "deficiency" defined where some physiological function such as dark adaptation or an enzyme activity might improve after physiological supplementation. Secondly, I support Dr. Prasad's strictures against the plasma zinc measurement because I believe that in clinical practice we are moving away from simple plasma levels. We would not, for instance, detect osteomalacia from plasma calcium levels but would look at the histology of bone. Similarly, plasma potassium levels are a poor measure of body potassium levels. I think this is true for zinc, a major part of it being within the tissues. I support Dr. Prasad's emphasis on white cells. For some years now, in parallel with his laboratory, we have been looking at leukocytes. We have not been able to use the small samples that pediatricians wish, and our pediatric work has used cord blood where large quantities are available, but I suggest that, using mixed leukocytes, a smaller quantity of blood can be used. We have found this to be a useful indicator of what we think is zinc depletion, although I realize that there is no ultimate standard. However, I am worried about the amount of zinc in the medium used for separating fractions of white cells. There is a high concentration of zinc in mononuclear cells, and they may be a better measure of zinc depletion than the neutrophils, even though there are many more neutrophils. I do not know why the white cells decrease or increase their zinc depending on the zinc status. I wonder if they take up and lose zinc in blood, or if they come out of the marrow depleted and remain depleted, perhaps because their zinc-binding proteins are low. Finally, we have looked at the neutrophils in Crohn's disease and found only a few patients to be zinc-depleted. They have, of course, low albumin levels. I think that it is not malabsorption that causes the few to become depleted, but loss of appetite and intake. The patients with short bowels eat little, rather like jujunoileal bypass patients. Dr. Prasad: These are difficult questions to answer. If you have a clinical parameter such as dark adaptation abnormality correctible with zinc supplementation, then it is possible to distinguish between depletion and deficiency. On the other hand, if the deficiency is mild there may not be any clinical parameter available to diagnose zinc deficiency and only abnormal laboratory tests are there to indicate abnormal zinc status, then one cannot be certain if indeed there is deficiency. With respect to the lymphocytes and monocytes in our neutrophil preparation, we have no more than 4 to 5% monocytes and 10 to 15% lymphocytes. I therefore do not think that these cells are contributing very much zinc to our neutrophil assay. In the lymphocyte preparation, however, we have up to 20% neutrophils, but fortunately the levels of zinc in both neutrophils and lymphocytes are about the same. The real problem, however, is with respect to erythrocyte contamination, which is difficult to eliminate. With respect to your last question, I suspect that most of the zinc is being incorporated in the bone marrow during the synthesis of cells. Dr. Chandra: We have some recent data which may be pertinent to Dr. Thompson's question about where zinc is incorporated into the leukocytes. The administration of adren- aline releases the marginal pool of granulocytes, whereas Pseudotnonas polysaccharide releases the bone marrow pool of granulocytes. These challenges result in a decrease in peripheral blood leukocyte zinc, which may indicate that the content of zinc in these two pools, the marginal and the bone marrow, is probably slightly less compared with the peripheral blood circulating leukocytes. These pools are often released in infection and in time of other stresses in response to endogenous mediators, and this leukocyte migration may therefore influence the level of zinc and perhaps other trace elements during acute infections. DIAGNOSTIC APPROACHES 37

Dr. Cheek: I would like to comment on the problem regarding the need for large volumes of blood for white blood cell (WBC) zinc and for the separation of species of WBCs. For some years we have been working with the technique of proton-induced x-ray emission (Cheek, Hay, Newton. Aust Phys Sci Med, 1979;2:85-95), which we refer to as PIXE. One uses a 5-mm disc of foil (Hostophan) or membrane (Mylar), and only a single layer of lymphocytes or granulocytes is required in a 2- or 3-mm ring (diameter). The protons bombard the membrane (using a Van der Graaff accelerator), and each element gives off x- ray energy, according to atomic number. Thus zinc in WBCs can be determined—and provided one separates cells appropriately, taking great care to avoid contamination—in only small volumes of blood (e.g., 0.5 ml). In our work with the aboriginal children, lymphocytes and granulocytes were layered onto Mylar discs. The "a" cells were from aboriginal children and the "c" cells from Caucasian children. There was no overlap between "a" and "c" cells. The "a" cells were low in zinc. Use of PIXE has advantages, as picograms (10~n g) of zinc can be detected. The results of this work are published elsewhere (Aust NZ J Med, in press). There is one more important point to be made. No attention has been drawn to the fact that different types of WBC contain different amounts of zinc. For example, we have known for years and from the early work of Vallee (J Biol Chem 1949;445) that eosinophilic cells contain very large amounts of zinc (x 10) which appears to be bound to arginine-rich peptides (Olsson et al. Lab Invest 1977;36:493-500). By the same token, Nishi et al. in Japan (Hiroshima J Med Sci 1981;30:65-9) claimed that monocytes contain twice as much zinc as granulocytes. Admittedly, when one separates lymphocytes and granulocytes, they appear to give the same value for zinc content, but the important issue is that the cells should be inspected with the aid of a cytospin and microscopy, especially as parasitic infection causes eosinophilia. Dr. Gebre Mehdin: Dr. Prasad, regarding the choice of a model for the study of zinc, you mentioned chronic renal disease and sickle cell anemia. Would you comment on the interrelation between zinc, retinol, and retinol-binding protein, as the metabolism of these three nutrients is greatly affected in both chronic renal disease and, in our experience though limited, sickle cell anemia? As far as zinc deficiency, circulating zinc, and zinc depots are concerned, the few studies we have concluded on liver concentrations of zinc show that patients with chronic liver and kidney disease have massive amounts of zinc in their livers, suggesting an inability to mobilize this element for target cell consumption. Concerning the comparison of plasma and cell studies, Dr. Cheek mentioned the PIXE method. Our experience in Sweden is that PIXE probably does not give much more information than does the plasma assay. Dr. Prasad: With respect to the relationship between zinc and retinol-binding protein and vitamin A metabolism: Retinol-binding protein is zinc-dependent. The other is that retinene reductase is a zinc-dependent enzyme. Thus both zinc and vitamin A have an effect on night blindness. However, I have observed that the retinol-binding protein in our mildly zinc-deficient human volunteers did not change as a result of dietary zinc restriction. Therefore I presume that the deficiency of zinc has to be more severe in order to see some effect on the retinol-binding protein. With respect to the liver disease and zinc concentration of the liver, I am surprised to hear your comment. Dr. Vallee reported many years ago that liver zinc is decreased in alcoholic cirrhosis of the liver. What kind of liver disease did your patients have, and how high was the liver zinc? Dr. Gebre Mehdin: These were liver diseases of various etiologies but mainly, of course, chronic alcoholism. I cannot give you the exact concentrations, but they were not lower, and in fact in some instances they were higher, than the concentrations found in apparently 38 DIAGNOSTIC APPROACHES well-nourished Swedish subjects killed in car accidents. The same thing applies to the subjects with kidney disease. They also seem to have adequate amounts of zinc in the liver. Dr. Hambidge: One thing that particularly puzzles me about neutrophil zinc concentrations is that it has been reported that the zinc concentration correlates closely with muscle zinc, and yet it has been shown experimentally that, at least in rat muscle, zinc does not decrease in a zinc-deficiency state. The combination of these observations seems to suggest that neutrophil zinc is not going to be as promising as reported here. The other point I would like to make concerns normal levels; our normal levels are about 45 u,g/1010 cells, or about half of the values reported here and a lot less than concentrations of mononuclear zinc. Dr. Aggett: We investigated the value of mixed leukocyte zinc determinations in groups of zinc-deficient, pair-fed, and ad libitum-fed pigs. Although we had severely zinc-deficient pigs, the mixed leukocyte zinc content expressed in dry weight, DNA, and protein was the same in all three groups, as was the muscle zinc content. The zinc content of the pancreas, liver, and plasma was reduced, however. A point I want to raise is how people should express leukocyte zinc content. Should one use net weights, dry weights, or cell numbers, or should a more physiological index be used such as DNA, protein, or other intracellular marker? Dr. Thompson: I agree with Dr. Prasad, but our experience is that the concentration of zinc in the liver falls dramatically in cirrhosis, although we have some evidence that in acute and chronic hepatitis the hepatic uptake and concentration of zinc per dry weight are increased. Regarding Dr. Aggett's pigs, I suggest that the white cells are showing that the animals were not short of zinc, but that if the diet were continued the level would eventually fall. Then the animals might be truly deficient. The plasma zinc level may fall, but the animal will still have sufficient zinc and not be whole-body-depleted. Regarding the reference for white cell zinc, we initially used dry weight because it is easier and also because this removes the complication that shortage of zinc impairs protein and nucleic acid synthesis; therefore the weight of the tissue might be a better "parameter," but this is debatable. We are now looking at the amount of zinc per cell. Dr. Aggett: I want to make a quick point that our pigs were zinc-deficient; they were not zinc-depleted. They were growing either very slowly or not at all. They had typical skin rashes, and they were showing all other classic features of zinc deficiency. Dr. Prasad: I want to make two comments. One concerns the expression of neutrophil zinc. We expressed zinc in terms of micrograms per 1010 cells. However, theoretically I think the best expression would be in terms of DNA. The second comment is if you have a problem with aggregation of neutrophils during separation then your results are going to be very high because you have a smaller number of cells to be counted. Dr. Bergmann: In our experience with hair zinc determinations and successive measure- ments of growth rate in infants and children, we found that with values below 60 ppm the age-standardized growth rate correlates significantly with hair zinc concentration. We concluded that hair values below 60 ppm are suggestive of zinc status. At which level of total body depletion of zinc would you talk of zinc deficiency? Dr. Golden: I am amazed that if you take a pig, put it on a zinc-deficient ration, the animal gets skin lesions, stops growing, and demonstrates all the classical features of zinc deficiency—and you find that the zinc concentration in a tissue has not dropped—that you can say that this animal is not zinc-deficient! If you put an animal on a zinc-deficient diet and it suffers all the clinical manifestations, one must accept that these are due to zinc deficiency no matter what the tissue concentration of zinc is. Certainly in malnutrition the DIAGNOSTIC APPROACHES 39 disease itself affects the metabolism of many constituents of the body, particularly the proteins and the enzymes which hold and lock zinc into tissues. My own view is that it is much more likely in many illnesses that the fall in the zinc content of the tissue is a secondary phenomenon—secondary to the adaptation of the tissue to the disease, so that the zinc metalloenzyme levels that are synthesized and maintained in the tissues are quite responsive to metabolic state irrespective of the zinc status. Dr. Hurley: I would like to comment on the differentiation between depletion and deficiency in zinc. I agree with what Dr. Golden has just said, and I think it is important to remember that depletion is not the same as lack of incorporation. When we say depletion, we mean that a certain amount of zinc was present in a tissue and then left that tissue; however, in the case of a growing animal, if there are low zinc levels, they are due to lack of incorporation. My basic point is that we do not have evidence of any kind that zinc can be released specifically from tissues. In most cases we have studied, when there is a loss of zinc from a tissue there has been a breakdown of that tissue. When a cell is catabolized, the zinc is released but in most cases of zinc deficiency there is not zinc depletion.