1Medical Hypotheses MadiaJ Hy+hem a992,38,7585 0 LongmnCmupUKLtd1992

Deficiency of Copper Can Cause Neuronal Degeneration H.A. HARTMANN and M.A. EVENSON Department

of Pathology,

University of Wisconsin,

1300 University Avenue, Madison,

WI 53705,

USA

Abstract - The aim of this article is to emphasize the important role that copper plays in the function of nerve cells. We are reporting preliminary data which suggest that the swelling of axons which we produce in rats by iminodipropionitrile, IDPN, is due to its chelating action on copper, and how conversely supplementation with copper abolishes both symptoms and lesions. The copper values we obtained by atomic absorption spectrophotometry of the spinal cord and brain from the animals fully support this contention. In comparing these results with the diseases that are known to be due to copper deficiency, namely Menkes disease in man, swayback in lambs and several neurological mutant mice, we find not only similar axonal swellings, but also amelioration of symptoms and lesions by early administration of copper. Considering the main forms in which copper is present, we discuss the cuproproteins, i.e. ceruloplasmin and metallothionein, and their role in transport and delivery of copper to various organs. Further, the many cuproenzymes i.e. superoxide dismutase, tryptophan-2,3dioxygenase, lysine oxidase, cytochrome oxidase, monoamine oxidases, tyrosinase, dopamine-&hydroxylase and d-amino levulinate dehydratase are noted for their roles in the nervous system. Finally, we suggest that neuronal copper deficiency should be more fully investigated as a possible etiological factor in the more common neurodegenerative diseases, such as Alzheimer’s disease and amyotrophic lateral sclerosis, ALS.

Introduction and Hypothesis There are several reasons why deficiency of copper has received relatively little attention as a cause of disease in the nervous system One reason is that only recently has the sufficiently sensitive and accurate technique of atomic absorption spectraphotometry become available. Another mason is Date received 20 June 1991 Date accepted 3 October 1991

that investigators have directed their attention towards more easily accessible tissues, e.g. liver, kidney and blood, rather than towards the complex nervous system A third reason is the long time it takes to produce deficiency of copper in experimental animals requiring meticulous care of several generations of animals fed a copper deficient diet. In our judgement a quicker and easier way to 75

76 produce copper deficiency may be to give a compound which chelates copper. The structure of the neurotoxic chemical iminodipropionitrile, IDPN, suggests that it should chelate copper. To prove this point, we have performed some preliminary experiments. Preliminary studies We have observed an exothermic chemical reaction, a color change aud precipitation when copper was mixed with IDPN in test tubes. This prompted us to conduct a preliminary feasibility study employing one rat in each of three groups. Rat l(400 g) received 2.0 g of IDPN per kg in 10 equal doses of 80 mg of IDPN given intraperitoneaily in 0.5 ml of isotonic saline buffered with 0.01 M TRIS (and the pH adjusted to 7.0) on Mondays and Thursdays over a 5-week period. Rat 2 (400 g) received the same amount of IDPN as rat 1 in the same TRIS buffered saline solution but it also contained 0.6 M copper chloride. Each dose contained 80 mg of IDPN and 1.9 mg of copper in 0.5 ml of the buffered saline. As before 10 doses were given over a 5-week period so the animal received 800 mg of IDPN (corresponding to 2 g per kg) and the IDPN plus copper rat received 19 mg of copper from CuC12 (corresponding to 47.5 mg per kg). Rat 3 (400 g) received 0.5 ml of a TRIS buffered saline containing only copper chloride and was given tbe same amount of copper as was given to rat 2. Over the 5-week injection period, clinical observations of dysfunctions due to IDPN alone were apparent. Rat 1 (IDPN only) and rat 3 (copper only) lost weight down to 275 g and 250 g, respectively. Rat 2 (IDPN and copper) only lost weight to 325 g. Rat 1 (IDPN only) showed circling hyperexcitability, retropulsion and choreoathetotic head movements and also did somersaults. Rat 2 (IDPN plus copper) had a tendency to show similar symptoms to rat 1. namely circling and choreoathetotic head movements, but it was much less noticeable and less severe. Rat 3 (copper only) developed hematomas where the copper was injected into the abdomen, hut showed no neurological symptoms. It was clinically obvious in rat 2 that the IDPN was protecting the animal against copper toxicity (there were no hematomas at the injection sites) and that the copper was protecting the animal against IDPN (essentially no spinal cord lesions).

MEDICAL HYFoTHJr.sJrs Histotogicaiprotocols Each rat was given 0.5 ml of ketamine (87 mg/kg) in an intramuscular injection followed by 0.25 ml of Zylazin (13 mg/kg). Perfusion was done with 10% formalin. About half of each brain was placed in a 10% formalin solution and the remaining half placed in Saran wrap plastic. A 1 cm long portion of the lumbar spinal cord (about 4-5 cm from the brain stem) and a 1 cm portion of the cervical cord (about 1 cm from the brain stem) were also taken for metal analysis. The tissues in formalin were dehydrated in increasing concentrations of ethanol, cleared in xylene, infiltrated and embedded in paraffin. Sections were cut at 8 microns. The areas of the spinal cord were stained with Nissl stain. Axonal swellings were counted in all sections following the procedure described by Hartmann et al. (1). Nerve cell bodies were counted separately from the axonal swellings in the same section. Copper analysis These methods am only minor modifications of the liver analysis methods as published by Evenson and Anderson (2). National Bureau of Standard and Technology‘s-Standard Reference Material (NISTSRM) Number 1577-Bovine Liver was brought to constant dry weight by lyophilization (0.5560 g per liter), digested with 2 ml of Ultrex concentrated nitric acid at 60°C for 15 h and diluted to 1.OOOlitre with cold distilled 1 mM nitric acid. All glassware and laboratory ware was soaked in 20% nitric acid. rinsed a minimum of 8 times with Mill&Q 18 megohm water and air and dried beforn using. The stock standard was purchased as a 10 000 ppm atomic absorption standard from Fisher Scientific Company. Working standards were prepared by serial dilutions to 0.1 and 1.0 ppm copper with 1 mM nitric acid. Three different portions of the bovine liver standard were prepared and each measured in triplicate on three different days. The range of the 9 measurements of the NIST-SRM number1577wasfrom88-1139bwithameanvalue of 102% of the expected value. This result indicates that the copper method used for these analyses is accurate and precise. The brain and spinal cord tissues were weighed wet, transferred into acid washed polyethylene test tubes and brought to constant weight in 15-18 h on the lyophilizer. The samples were reweighed to

77

COPPER DEFICIENCY CAN CAUSE NFiURONAL DEGENERATION

obtain the constant dry weight, then digested with 1 ml of ultra pure, 14 M nitric acid (Ultrex by Fisher) at conditions as described above for the bovine liver standard. After lsl8 h of digestion, the solutions may be slightly turbid but will be mostly a clear yellow solution. ‘Ihe nitric acid was evaporated just to dryness in a vacuum film evaporator in the original digestion tube. Next, 100 pl of I mMmetal free nitric acid was added per mg of dry weight tissue and the tube mixed vigorously. Then, l-5 pl of the sample was accurately injected into the graphite furnace atomic absorption spectrophotometer and then peak heights of the samples compared to standards. In addition, a standard addition study was conducted by adding a known amount of copper to the starting tissue sampbs to see if the amount recovered was near 100% of added standard. This test helped show that there ate no significant interferences in the metal measurements. The instrument used was a model 503, Perkin-Elmer instrument with a model 2100 graphite furnace. The wavelength used for the analysis was 324.7 ML Either helium or argon gas was used to prevent oxidation of the pyrolytic graphite tube and to reduce the oxygen reaction with the metals in the sample. Each sample and standard were analyzed in triplicateandthemean values used to calculate the tissue concentrations as well as the NBST-SRM bovine liver samples. Histology preliminary results

Wedecidedtotryathree(statisticalarm)group,one rat per group pilot study to test our hypothesis that copper would be chelated in vivo by IDPN and the low concentration of copper in the neuronal tissue was involved in the formation of the lesion. We measured the tissue concentrationsof copper and did histological examinations of the spioal cords for the lesions caused by IDPN. We compared the histology and tissue copper concentrations of the IDPN only, the IDPN plus copper and the copper only rats. The rat which received 2 g IDPN alone, displayed all the characteristic symptoms described above. Notice in Figure A the many light stainiog spherical lesions in the anterior (ventral) gray horns of the spinal cord. They are several times larger than the darker staining nerve cell bodies, and have been called axooal balloons or spheroids, since in the silver stained sections or electron micrographs they are seen to be enlarged axons (3.4). Approximately

40 balloons can be counted in the two ventral horns. These lesions are the reason why IDPN rats have been called the best animal model for amyotrophic lateral sclerosis, ALS. In Figure B the same section is taken from a rat which was given the same amount of IDPN plus copper. Only two slightly swollen axons, one in each ventral horn, are seen in this section, and many of the other sections were indistinguishable from normal sections. The histological examination of the spinal cord from the rat which only received copper was normal. We conclude from the histological examination that copper protects rats from developing the lesions that are caused by IDPN alone. Furthermote, the rat that received copper supplement and IDPN showed only mild neurological symptoms.

Copper analysispreliminaryresults The Table presents the data of the tissue copper analysis of our preliminary pilot study with three rats and some of the literature values for both rat and human tissues. If one examines the IDPN only column of the Table, one notices that the whole brain value of 8.7 ppm dry weight of the IDPN only treated rat is only 60% of the no& value in the literature. If one looks at the copper concentrations of the brains of copper supplemented rats, one notices the concentrations of 13.7 and 13.4 ppm respectively; therefore, we conclude that, even with the in vitro reactions between copper and IDPN, the copper and probably the IDPN teach the spinal cord and brain. These values found in the IDPN plus copper and the copper only rat brain are consistent with the hypothesis that normal concentrations of copper in neuronal tissue appear to protect against the IDPN caused lesions. We am unable to find oormal values in the literature for copper in rat spinal cord. Our measured value in the copper supplemented rat is similar to the normal brain values found in the literatnre. We noted with interest that the normal copper concentrations in whole human brain is about three times more concentrated than that found in rat brain. We tentatively conclude from the above values that IDPN chelates the copper from nerve tissue, that added copper increases the neuronal tissue copper concentrations both with and without IDPN, and that this increased copper in the spinal cord protects the motor neurons ftom IDPN.

78

haDICJu -Es

Review of literature

IDPN impairs the slow transport of neurotilament proteins, as well as actin and tub&. Historyof IDPN Papasozomenos et al (9) when giving IDPN to rats noted displacement of microtubules to the Delay et al (5) gave the fmt description of the center and neurofilaments to the periphery of the spectacular neurological symptoms seen in mice axons, and found that the fast axonal transport of after injection iminodipropionitrile, IDPN. Be- proteins, as measured by qualitative electron cause the rapid circling it induces resembles a microscopic autoradiography,was not altered. genetic mutation in mice, it has been called the Hirokawa et al (10) studied the microtubular ‘Waltzing Syndrome’ or the ‘ECC syndrome’, associatedproteins MAP 1A,MAP 1B , MAP 2 and abbreviations for excitement, circling and choreo- tubulin in the same model by quick-freeze athetotic head movements, Selye (6). deep-etch electron microscopy. They found that Hartmann et al (1) found that bis-fl-cyano- anti-MAP 1A and anti-MAP 1B stained the ethylamine, synonymous with IDPN, induces IDPN-treated axons brightly, with the staining extensive lesions in the motor neurons of the spinal exclusivelylimited to the microtubule domains, and cord. Later Chou and Hartmann (3) noted that it concluded that the rapid transport is affected by was the axons that were enlarged in volume to many IDPN. Eyer et al (11) found that the autophosphoryltimes that of the cell body. They suggested that these swollen axons, so called balloons, were filled ation of neurofdament subunits was significantly with large bundles of neurofibrils and vacuolated increased by IDPN. They suggested that IDPN interferes with the neurofilament-associated protmitochondria (4). Slagel and Hartmann (7) found that mice injected eins resulting in an increased interaction between with IDPN had not only swollen axons in the motor the polymers. It is our opinion that an investigation of whether neurons but also in the reticular formation and in the the axonal enlargement caused by IDPN can be vestibular system suggesting a functional prevented by copper or other ions, would be helpful relationship between behavi ur and lesions. Griffin et al (8) injected [9HI-leuciue and [35S]- in explaining normal and abnormal axonal flow. methionine into the lumbar ventral horns of rats which had received IDPN. They analyzed the proteins in segments of the sciatic nerve from l-8 Diseases due to copper deficiency weeks later. All of the triplet neurofilament proteins were present in control and IDPN nerves. Within the last decades, several neurological Since, however, the major slow component peak diseases in man and animals have been found to be was retained in the IDPN rats, they concluded that due to copper deficiency. These diseases are often Table Tissue copper concentrations* NOrmCrl

rat Whole brain Spinal cord Liver

‘14.7 g9

IDPN ~reai& ral 8.7 9.9 17.9

Cu + IDPNA rat

CS rat

13.7 13.3 16.3

13.4 16.6

NOr??ll.Zl human ‘35:42,444 216.6 ‘26

Wilson’s disease human 6215.2 5239.5 2over 500

*dry weight basis, pg/g - some values converted fmm wet weight to dry weight using the assumption that the wet weight is 75% water 1. Evensonetal.ClinChem21:219,1975 2. Scheinberg et al. Copper toxicity and Wilson’s disease (Tables 2 and 3). In: Drasad and Obeleas eda. Trace Elements in Human Health and Disease. Academic Press 1: 1976 3. Ibid p 422. mean of 17 values in Table 3 4. Goldberg et al. Clin C&m 27: 562,198l ‘5. Op tit Sckinbcrg et al. p. 422, mean of 2 values in Table 3 6. Ibid, mean of 19 values of Table 3, p.422 7. Prohaska Physiol Rev 67 (3): 861,1987 - means of values in Table 1 8. Lorenzen, Smith J Nutr 33: 143.1947 A. Our measured values

B

Fig. Transvm se&ionsof the spinal cord from rats given: A. IDPN only about 40 light staining ‘balbms’ or sphemids. a, larger than dark staining xrrve cell bodies, XI,in ante&r burns. B. ILPN + CuClz two small light staining sph~A&,s, about the same size as dark staining nelye eel1 bodies, II,in anterior horns. Sections: 8 pm thick, Nisd stain, magnification x 35

so severe at birth that life will terminate early. However, if copper is administered soon atIer birth or preferably during pregnancy, considerable

clinical improvement has resulted. Menkes syndrome (12) is a condition characterized as a sex-linked recessive mental

80 retardation in man with cerebral degeneration. Danks et al (13) recognized that it was due to a defect in copper metabolism. Menkes et al (12) demonstrated dramatic abnormalities of the dendritic arborization of the Purkinje cells, and felt that these were abortive attempts at regeneration following axonal injury as suggested by Cajal(14). Aguilar et al (15) reported 9 siblings with Met&es syndrome, and believed that not only axonal torpedoes and retraction bulbs, but also somal sprouts, represented efforts at regeneration. Williams et al (16) recognized axonal degeneration even during fetal life, but believed that while copper therapy may restore Cu levels, the disordered neuronal development which had already happened in utero, could not be completely reversed after birth. Recently isolated cases have been reported which survived for more than 8 years after injection of Cu-histidine (17) and another group of six mildly affected patients have been noted to survive for more than 13 years (18). Several mutations in mice are considered to be models of Menkes syndrome. The brindled mouse was first described by Fraser et al (19). Later Hunt (20) showed that copper in this mouse is low in brain and liver due to a defect in copper transport Some of these x-linked mutants, which are designated MO for their mottled fur, are either lethal in utero, like MO or MO dp for dappled which dies at birth. Others, like MO blo for blotchy, are viable and fertile, or MO br for brindled dies at 14 days of age. Mann et al (21) studied the MO brand found that a single injection of 50 pg CuCh at 7 days of age would prevent tremors and spasms, raise the activities of copper containing ceruloplasmin and lysyl oxidase, increase pigmentation of the skin and fur and prolong life to 10 months or older. Fujii et al (22) recently have shown that the activity of cytochrome c oxidase of the brain mitochondria increases significantly and neurological symptoms are prevented if the MO br mice are injected with 10 pg copper/g on day 4 or 7, but not on day 12. Yajima and Suzuki (23) found widespread neuronal degeneration in the cerebral cortex and thalamic nuclei of MO br mice and were impressed with the extensive axonal degeneration in white matter (24). Later they were able to prevent the symptoms and lesions by injection of 10 pg cupric chloride (25). In Japan, Nishimura (26) described the macular mouse Ml/y, another mutant with symptoms resembling Menkes syndrome and later with Kawasaki et al (27,28) concluded thatGolgi stained preparations showed delayed arborization of

hxEDICAL HYPoTEmEs

dendrites both in cerebral and cerebellar neurons. The Purkinje cells also showed thickening of the primary dendrites and axonal swelling, similar but not identical to the somatic sprouts and grotesque dendritic arborizations described in Menkes syudromebyPurpuraetal(29)andbyHiranoetal(30). Yamano et al (3 1) abolished the seizures, the ataxia and the ultrastructural findings of abnormal mitochondria, neurofilamentous inclusions and swelling of dentrites by injecting the Ml/y mice with cupric chloride. Shiraishi et al (32) found that the copper content was low in brain and liver of Ml/y but high in kidney and small intestines. There was no difference in zinc levels. Cultures of fibroblasts frompatients with Met&es syndrome accumulate excess Cu which chromatographs both with high molecular weight protein and with a metallothionein-like protein according to Bonewitz and Howell (33). In Western Australia, lambs had suffered from ataxia for many years before Bennetts and Chapman, basedon their investigations, reached the conclusion that it was due to copper deficiency (34). Later, Bennetts and Beck were able to cure the ataxia by administration of copper (35). This was more than 20 years before Menkes et al (12) described the above mentioned neurological syndrome in children and 30 years before Danks et al (13), also in Australia, concluded that Menkes syndrome was due to copper deficiency. The early opinion that ataxia in lambs, also known as swayback, was primarily a demyelinating disease, was doubted by Schulz and Behrens (36) who found demyelination to be preceded by swelling of the axons and break- down of the blood brain barrier. Cancilla and Barlow (37). in an ultrastructural study of swayback lambs, found a vast increase of neurofibrils and suggested that it was similar to the axonal aggregates induced in rats with imiuodipropionitrile, IDPN, by Chou and Harlmann (3,4). Induced copper deficiency in rats Other investigators, Carltou and Kelly (38), fed a copper-deficient diet to female rats from 3 weeks of age, the food containing 0.9 ppm copper as determined by atomic absorption spectrophotometry. After mating, the females were returned to the copper-deficient diet. Two breedings were completed and when their pups were born, they were also weaned on the same diet as the dams. Two to five weeks later signs of severe neurologic

COFTW

DEFICIENCY

CAN CAUSE NEURONAL

81

DFX3ENJZRATION

disturbance were observed. Provoked by noise, the rats were rushing wildly about in their cages, trembling and attempting to climb the walls. Between the ‘running fits’, the rats seemed catatonic with limbs in fixed position. Some animals displayed convulsive seizures while others rose on their rear legs and fell over backwards. These symptoms are similar to those described after IDPN administration by Chou and Hartmann (3). Carlton and Kelly described widespread lesions both in cerebral cortex and in corpus striatum in these Cu deficient rats, varying from small focal areas of necrosis to extensive devastation of the neural tissue, which they attributed to a significantly lower cytochrome-oxidase activity (38). Other metals in CNS It is interesting to note that while the effect of excess metal ions, e.g. aluminum, on the CNS has received so much attention, less attention has been given to the effect of deficiency of metal ions on the CNS. In the study of Alzheimer’s disease there are numerous reports (39,40) dealing with Al. Also the literature abounds with reports about experimental injection of aluminum salts producing neurofibrillary lesions (41, 42). The precise mechanism by which this happens is unknown. Pierson and Evenson (43) recently showed that the 200 kD human and bovine neurofilament proteins bind at least one mole of Al, one mole of Cu and four moles of Zn. On theoretical grounds, Glick (44) suggests that deficiency ofmagnesium is an importantetiological factor in the pathogenesis of Alzheimer’s disease. Also on theoretical grounds, Bumet (45) suggests that a genetically based progressive inability of neurons to incorporate zinc ions into DNA and RNA enzymes may play a role in Alzheimer’s disease. Gajdusek (46). after studying the environmental factors in the Western Pacific where ALS and Parkinson dementia is endemic, suggested that Ca deficiency may be an etiological factor. Subsequently his group (47) fed monkeys a low calcium diet for more than 40 months. The motor neurons of the spinal cord showed chromatolysis, accumulations of phosphorylated neurofiiaments, axonal spheroids and inclusions, all of which are compatible with early ALS . Khare et al (48) analyzed trace elements from 7 ALS patients comparing them to 9 control patients. The neutron activation analysis of 15 elements

revealed widespread changes in Hg and Se levels in ALS patients. They did not include copper. Copper proteins

Metals occur as components of metalloproteins which serve regulatory purposes according to Kagi and SchrUfer (49). As far as transport of copper is concerned, radioactive tracer studies have shown that, after it is absorbed by the intestines, most of the copper is found in the portal blood attached to albumin, and according to some investigators, 15% is bound to a protein called transcuprein (50). In 2 h the liver starts to synthesize cemloplasmin, a blue cuproprotein, first found in the plasma by Holmberg and Laurell(51). Cemloplasmin transports copper to other organs, and according to Broman (52). incorporation only takes place when copper is presented as cemloplasmin. Later investigators, however, have shown that this may depend on specific surface receptors for ceruloplasmin (53). The mammalian brain is one of the richest copper containing organs in the body (54), and the concentration in the brain is several fold greater than in blood. There is also a marked variation between different regions in the brain, and a copper concentrating mechanism has been proposed for areas such as the locus cemleus and basal ganglia. Precise knowledge of how copper is taken up specifically into the brain is not available. Data from other systems, however, such as copper transport into K562 cells by Harris and Percival (55) suggest that cemloplasmin binds to the cell membranes, and that only the copper transfers into the cytosol. The protein moiety does not enter the cells. Cytosolic radioactive 67Cu was bound to superoxide dismutase, the major cytosolic copper protein in these cells. Two isoforms of metallothionein MT have been found in the bovine and murine hippocampus by Paliwal and Ebadi (56) who suggest that they play an essential role in regulating the transport of zinc. They did not however measure zinc, copper or any other metal directly. Recent work has established that the structure of the MT protein and gene has been highly conserved between birds and other auimals. suggesting a functionally important role for this protein (57). According to Kagi (58) up until now 42 different types of MT have been isolated It has been suggested that one function of MT might be to regulate metalloenzyme activity by controlling the supply of metals (59).

82 Cuproenzymes Evidence that copper is an important part of many enzymes, i.e. cuproenzymes, has been steadily accumulating (60). Already in 1928, Hartet al. (61) showed that copper plays a key physiological role in preventing anemia. Soon thereafter the same laboratory (62) reportedthat copper is essential for synthesis of heme A, a component of cytochrome oxidase. Cytochrome oxidase catalyzes the final step in the mitochondria-electron-transport chain. The enzyme contains two atoms of copper. It is strongly inhibited by cyanide, which will kill an animal even in a small dose. An important cuproprotein was fmt isolated from bovine erythrocytes by Mann and Keilin (63) who later found that it functions as an enzyme, which subsequently was named superoxide dismutase, SOD, by McCord and Fridovich (64). In the human liver and brain it was found to be high, about 3 pm01 per kg wet tissue (65). The superoxide dismutase copper protein contains two copper and two zinc ions per enzyme molecule, and the molecular weight is 34 000 (66). Tryptophan-2,3-dioxygenase, containing 2g atoms of copper, has been purified from rat liver (67). Lack of some of these enzymes have been correlated with pathological findings. Cardiac enlargement with aneurysms at the apex were described in rats fed a copper deficient diet (68) and dissecting aneurysms as well as hemopericardium due to rupture of the heart were described in pigs reared from birth on a copper deficient diet by Shields et al. (69). Tropoelastin is a soluble elastin-like protein isolated from aortas of copper deficient pigs (70, 71). It lacks the cross-linking compounds desmosine and isodesmosine, and contains 38 residues of lysine as compared to 8 residues of lysine in insoluble elastin (72). The cross-links in desmosines are derived fromiysine or hydrolysine and include the formation of aIdols and allysine. Lysine oxidase is an enzyme which catalyzes some of these reactions and is classified as a copper metalloenzyme since it contains 0.14% of copper, i.e. 22 nmoles/mg. Monoamine oxidase catalyzes the oxidative deamination of a variety of monoamine& many of which are found in the brain. Lysyl oxidase has already been described in elastic tissue, but in the brain the important amine oxidases will inactivate the neurotransmitters such as adrenalin and noradrenaline. Deficiency of monoamine oxidase

MEDICAL

HYPOTHESES

in the brain may presumably lead to different diseases such as migraine and schizophrenia, which is associated with the low monoamine oxidase levels in blood platelets (73). Tyrosinase catalyzes the initial reaction of melanin pigment from tyrosine. It consists of 4 similar subunits, each of which contains 1 copper (74). The genetic absence of tyrosinase leads to albinism. Dopamine-l!l-hydroxylase is an enzyme related to tyrosine and is the last step in the formation of noradrenalin. This large (mol wt 290 000) glycoprotein tetramer contains 8 mol copper (75) correlates in distribution with noradrenergic neumns (76). Dahlstrom and Fuxe (77) found that it was high in neurons of the brain stem, while Coyle and Axelrod (78) found a 2300 fold increase in the DBH-activity in the nerve terminals in the rostra1 part of the rat brain between 17 days gestation and adulthood. d-Aminolevulinate dehydratase is an enzyme which condenses two molecules of d-aminolevulinate into porphobilinogen, one of the key compounds necessary for hemoglobin formation. This enzyme is a copper enzyme. Acute intermittent Porphyria is sometimes characterized by acute psychosis and polyneuropathy and is due to a block in porphyrin synthesis (79). Administration of copper sulfate has been found to reverse the deleterious effect of D-penicillamine on collagen synthesis in rats, and to increase the monoamine oxidase activity. It also prevents paralysis of hind legs, although that was only mentioned briefly (80). iPace elements A review of essential trace elements has been given by Mertz (81) who gives the following definition: ‘An element is essential when a deficient intake consistently results in an impairment of a function and when supplementation of this element cures this impairment’. According to Bertrand, the degree of deficiency as well as the degree of toxicity may be formulated mathematically for each nutrient (82). During the last decade the following trace elements have become recognized as essential in animals: silicon, vanadium, chromium, manganese, iron, cobalt, nickel, copper, zinc, arsenic, selenium, molybdenum and iodine. The basic mechanisms of action have yet to be defined, their deficiencies are difficult to Induce and their role in human nutrition are unknown.

83

COPPER DEmcIENCY CAN CAUSE NEURONAL. DEGENERATION

Conclusions and recommendations studies

for future

Based on the evidence presented, we conclude that copper can prevent axonal swelling and degeneration of nerve cells. This is true both when the azonal lesions are produced experimentally by injection of iminodipropionitrile IDPN, and equally true when they occur in certain neurological diseases in man and animal. Regarding the action of IDPN, many studies have demonstrated profound cytoskeletal changes in the neurofibrillary and microtubular network, and slowing of axonal flow have been carefully measured, yet the results are conflicting with respect to its mode of action. Menkes’ disease in children is ameliorated by injection of copper to the pregnant woman. Swayback in lambs is prevented by feeding of copper. The neurological mutant mice such as MO, brand Ml/y are cured by injection of copper. All of these conditions axe pathologically expressed by axonal swelling, which responds favorably to injection of copper. Which chemical species of copper is most critical for the axonal flow is not known since there is a paucity of accurate measurements correlating regional and cellular types of copper with the morphological findings. Whether lack of specific cuproproteins of cuproenzymes is critical for production of axonal lesions, remains to be established. Future studies need to consider the role of Cu in the nervous system by careful measurements of Cu by atomic absorption spectrophotometq, correlating the regional and cellular types of copper concentrations and copper speciation with the morphological findings of axonal swelling. In cases of proven copper deficiency it has been shown in sheep that CuSO4 would be less toxic than CuCl2 as substitution therapy, but that even more copper may pass through cell membranes if the Cu is complexed with histidine and transported into the cell as such. Eventually these studies need to be expanded to neurodegenerative disease such as Amyotrophic Lateral Sclerosis, ALS, and to Alzheimer’s disease, which Gajducek (83) has postulated may be due to interference with axonal transport. References 1. Hartmann HA, Lalich JJ. Akert K. Lesions in the anterior motor horn cells of rats after administration of bis-beta cyanoethylamine. J Neuropath Exp Neural 17: 298-303, 1958.

2. Evenson MA. Anderson CT. Ultramicroansl~sis for coouer, absorption epectrophotometry and the heated graphite&be atomizer. Clin Chem 21: 537-543.1975. 3. Chou SM. Ha&mum HA. Axanal lesions and waltzing syndrome afler I.D.P.N. administration in mts. Acta Neuropatho13: 428-450,1964. 4. Chou SM. Hartmann HA. Blectmn microscopy of focal neuroaxonallesions produced by B-p-imina@opionitrile (IDPN) in rats. Acta Neuropathol4: 590-603,1965. 5. Delay J. Pichot P, Thuillier J. Maquiaet JP. Action de l’imino-dipropionite sur le corn-t moteur de la souris b&he. CooptRend Sot Biol146: 533-534,1952. 6. Selye H. Lathyrism. Revue Canadienne de Biologie 16: l-82.1957. 7. Slagal DE, Hartmann HA. The distribution of neuroaxonal lesions in mice injected with iminodipmpionitrile with special reference to the vestibular system. J Neumpathol Exp Neural 24: 599-620,1965. 8. Griffin JW, Hoffmann PN, Clark AW, Carmll Pr, Price DL. Slow axonal ?xansPortof neurofilament proteins: impairment by BP-iminodipmpionitrile administration. Science 202: 633-635.1978. 9. Papasozomenos SCH. Yoon M, Crane R, Autilio-Gambelti L, Gambetti P. Redistribution of pmteins of fast axonal tnmsportfollowingadministrationof~~iminodipropionitrile: a quantitative autoradiographic study. J Cell Bio195: 672-675,1982. 10. Hirokawa N, Bloom GS, Vallee RB. Cytoskeletal architecture and immunccytochemical localization of microtubuleassociated proteins in regions of axons associated with lapid axonal trampoh The BP-iminodipropionitrile intoxicatedaxonasamodelsystem. J CellBiol 101: 227-239, 1985. 11. Eyer J, Graham-Mclean W, Leterrier JF. meet of a single dose of fI-fl’-iminodipropionitle in vivo on the propertics of neumfilaments in vitro. J Neurochem 52: 1759-1765, 1989.

12. Menkes JH, Alter M, Steigleder GK. Weakley DR, Sung JH. A sex-Ii&d recessive disorder with retardation ofgmwth, peculiar hair and focal cerebral degeneration. Pediatrics 29: 764-779,1962. 13. Danks DM, Campbell PE, Stevens BJ, Mayne V. Cartwright E. Menkes’s kinky hair syndrome. An inherited defect in copper absorption with widespread effects. Pediatrics 50: 188-201.1972. 14. Cajal RY. Degeneration and regeneration of the nervous system, vol II, (translated and edited by RM May), Hafner, New York, pp 597-616.1959. 15. Aguilar MJ. Chadwick DL, Okuyama K, Kamoshita S. Kinky hair disease I. Clinical and pathological features. J Neuropathol Exp Neurol25: 507-616,1966. 16. Williams RS, Ma&all PC, Lott IT, Caviness VS. The cellular pathology of Menkes steely hair syndrome. Neurology 28: 575-583,1978. 17. Gerdes AM, Tonnesen T, Pergament E, Sander C. Baerlo&r KE. Guttler F, Horn N Variability in clinical expression of Menkes syndrome. Eur J Pediatr 148: 132- 135. 1988. 18. Sander C, Niederhoff H, Horn N. Life-span and Menkes kinky hair syndrome: report of a 13-year course of this disease. Clin Genet 33: 228-233.1988. 19. Fraser AS, Sobey S, Spicer CC. Mottled, a sex-modified lethal in the house mouse. J Genet 51: 217-221, 1953. 20. Hunt DM. primary defect in copper transport underlies mottled mutants in the mouse. Nature 249: 852-853,1974.

84 21. Mann JR, Camakaris J, Danks DM, Wallicxck EG. Copper metabolism in mottled mouse mutants. Biochem J 180: 605-612,1979. 22. Fujii T, Ito M, Tsuda H, Mikawa H. Biochemical study on the critical period for treatment of the mottled brindled mouse. J Neumchem 55: 885-889.1990. 23. Yajima K, Suzuki K. Neumnal degeneration in the brain of brindled mouse. A light microscopic study. J Neumpathol Exp Neurol38: 3546.1979. 24. Yajima K, Suzuki K. Neumnal degeneration in the brain of brindled mouse 1. Chronological studies on the long-surviving group. Acta Neumpathol48: 127-132,1979. 25. Nagam H. Yajima K. Suzuki K. The effect of copper supplementation on the brindled mouse. J Neumpathol Exp Nemo1 40: 428446.1981. 26. Nishimum M. A new mutant mouse, Macular (Ml). Exp A&n (Tokyo) 24: 185,197s (in Japanese). 27. Kawasaki H, Onaga A, Yamano T. Shimada M. Nishimura M. Golgi study on brain of macular mutant mouse as a model of Menkes kinky hair disease. Acta Neuropathol72: 349-354.1987. 28. Kawasaki H, Yamano T, Iwane S, Shimada M. Golgi study on macular mutant mouse after copper therapy. Acta Neuropatbol76: 606-612.1976. 29. Purpum DP, Himno A, French JH. Polydendritic Purkinje cells in X-chromosomes linked copper malabsorption: a Golgi study. Brain Res 117: 125-129.1976. 30. Hitano A, Llena JF, French JH, Gatak NR. Fine structure of the ccmbellar cortex in Menkes kinky hair disease. Arch Neurol34: 52-56.1977. 31. Yamano T, Shimada M, Onaga A, Kawasaki H, Iwane S, Ono K, Nishimum M. Electron microscopic study on brain of macular mutant mouse after copper therapy. Acta Neuropathol76: 574580,1988. 32. Shitaishi N, Aono K, Tagochi T. Copper metabolism in the macular mutant mouse: an animal model of Menkes kinky hair disease. Biol Neonate 54: 173-180,1988. 33. Bonewitz RF, Howell RR. Synthesis of a metallothioneinlike protein in cultumd human skin fibroblasts: relation to abnormal copper distribution in Menkes disease. J Cell Phvsiollo6: 339-348.1981. 34. Bet&& HW, Chapman FE. Copper deficiency in sheep in Western Australia. Aust Vet J 13: 138-149,1937. 35. Bennetts HW. Beck AB. Enxootic ataxia and copper de& ciency of sheep in Western Australia. Commonwealth Australia Council Sci Ind Res Bull 147: 4-55 1942. 36. Schulz LC, Behmns H. Beitmg zur Pathogeneses der Kupfermangelencephalopathie des Schatlammes (Swayback). Beitr z Pathologischen Anatomie u z Allgemeinen Patho logie 122: 282-288,196O. 37. Cancilla PA, Barlow RM. Structural changes of the central nervous system in Swayback (enxootic ataxia) of lambs. Acta Neuropathol6: 251-259.1966. 38. Carhon WW, Kelly WA. Neural lesions in the offspring of female rats fed a copper-deficient diet. J Nutr 97: 42-52, 1969. 39. Crapper DR, Karlik S. DeBoni U. Ahuninium and other metals in senile (Alxheimer) dementia. In: R. Katxman, R.D. Terry. K.L. Bick eds. Alxheimer’s Disease: Senile Dementia and Related Disorders. Raven Press, New York, pp 471-485 1978. 40. Per1DP. Ah&inter’s disease: X-ray spectmmettic evidence of aluminium accumulation in neurofibrillary tangle-bearing neurons. Science 208: 297-299.1980. 41. Klatxo I, Wisniewski H, Streicher E. Experimental produc-

MEDICALBYPOTHESBS tion of neuroflbrillary degeneration. J Neumpathol Exp Neurol24: 187-199.1965. 42. Thomas E, Wisniewska K. Histochemische Unterauchungen an Nervenxellen mit experimentell erxeugten fibrillaren Ablagerungen dumb Ahuniniumphosphat. Acta Neuropathol9: 335-345.1967. 43. Pierson KB. Evenson MA. 200 KD neumfilament protein binds Al. Cu and Zn. Biochem Biophys Res Commun 152: 598604.1988. 44. Glick JL. Dementias: the role of magnesium deficiency and an hypothesis concerning the pathogenesis of Alxheimer’s disease. Med Hypoth 31: 221-225,199O. 45. Bumet FM. A possible role of zinc in the pathology of dementia. Lancet 8213: 186-188.1981. 46. Gajdusek DC. Environmental factors provoking physiological changes which induce motor neuron disease and early neuronal aging in high-incidence foci in the Western Pacific. In: F.C. Rose ed. Ptoaress In Motor Neuron Disease. Pitman Books, Tunbridge Wells. pp 44-69 1984. 47. Garruto RM, ShankarSK, Yanagihara R, Salazar AM, Amyx H, Gajdusek DC. Low-calcium, high-aluminium diet-mduced motor neuron pathology in cynomolgus monkeys. ActaNeuropathol78: 210-219.1989. 48. Khare SS, Ehmann WD, Kasatakis EJ, Markesbery WR. Trace element imbalances in amyotmphic lateral sclerosis. Neurol Toxic01 11: 521-532.1990. 49. Kigi JHR, Schiiffer A. Biochemistry of metallothionein. Biochemistry27: 8509-8515.1988. 50. Goode CA, Dinh CT, Limier MC. Mechanism of copper transport and delivery in mammals. Itx C. Kies ed Copper Bioavailability and Metabolism. Advances Exp Med Biol. PlenumPtess.New York258: 131-144,1989. S1. Holmberg CG, Laurel1 CB. Investigations in serum copper 11. Isolation of the copper-comaining protein and description of some of its properties. Acta Chem Stand 2: 550, 1948. 52. Bmman L, Harris ED, Percival SS. Coppertransport: insight into a ccruloplasmin-based delivery system. In: C. Kies ed. Copper Bioavailability and Metabolism. Advances Exp Med Biol. Plenum Press, New York 258: 95102,1989. 53. Stevens MD, DiSilvestro RA. Harris ED. Snecitlc receotots for cemloplasmin in membrane fragments-from aortic and heart tissues. Biochemistry 23: 261-266.1984. 54. Barnea A, Hatter DE, Cho G, Bhasker KR, Katz BM, Edwards MD. Further characterization of in vitm uptake of radiolabeled copper by the rat brain. J Inorg B&hem 40: 103-l 10.1990. 55. Harris ED, Percival SS. Copper transport: insight into a ceruloplasnin-based delivery system. In: C. Kies ed. Copper Bioavailability and Metabolism. Advances Exp Med Biol, Plenum Press, New York, 258: 95-102.1989. 56. Paliwal VK, Bbadi M. Biochemical properties of metallothionein isoforms from bovine hippocampus. Exp Brain Res 75: 477-482.1989. 57. Fernando LP, Wei D, Andmws GK. Structure and expmssionofchickenmetallothionein. J Nutr 119: 309-318,1989. 58. Kiigi JHR. Personal Communication. 59. Richards MP. Role of metallothionein in copper and zinc metabolism. J Nutr 119: 1062-1070.1989. 60. O’Dell BL. Biochemistry and physiology of copper in vertebrates. In AS. Pmsad, D.Ober1ea.seds. Trace Elements In Human Health and Disease. Academic Press, New York, 1: 391-413,1976. 61. Hart EB. Steenbock H. Wade11J, Elvehjem CA. iron in nutrition VII. Copper as a supplement to iron for hemo-

85

COPPER DEFICIENCY CAN CAUSE NEURONAL DEGENERA-ITON

globin building in the rat. J Biol Chem 77: 797-812,1928. 62. Cohen E, Elvehjem CA. The relation of iron and copper to the cytochrome and oxidase content of animal tissue. J Biol Chem 107: 97-105,1934. 63. Mann T, Keilin D. Haemocuptein and hepatocupmin, copper protein compounds of blood and liver in m-ah. ProcRoySoc,SerB 126: 303-315,1938. 64. McCord JM, Fridovich I. Supcmxide dismutase. An enxymic function for erythmcuprcin (hemocuprcin). J Biol Chem 244.6049-6055.1969. 65. Hartz JW, Funakoshi S, Deutsch HF. The levels of supcroxide dismutase and catalase in human tissues as determined immunochemically. Clinica Chim Acta 46: 125-132,1973. 66. Ostcrberg R. Physiology and pharmacology of copper. In: G. Webbced. PharmacologyandTherapcutics 9: 121-146, 1980. 67. Brady FO, Monaco ME. Forman HJ, Schultz G, Feigclson P. On the role of copper in activation of and catalysis by tryptophan-2,-3-dioxygenase. J Biol Chem 247: 7915 7922,1972. 68. Kelly WA, Kesterson JW, Carlton WW. Myocardial lesions in the offspring of female rats fed a copper deficient diet. Exp Mol Pathol20: 40-56,1974. 69. Shields GS, Coulson WF, Kimball DA, Cames WI-I, Cartwright GE, Wintmbe M. Studies oncopper deficient swine. Am J Path01 41: 603-621,1962. 70. Smith DW, Weissman N, Cames WH. Cardiovascular studies on copper deficient swine XII. Pattial puritication of a soluble protein resembling elastin . Biochem Biophys Res Commun 31: 309-315,1968. 71. Sandberg LB, Weissman N, Smith DW. The puriBcation and partial characterization of a soluble elastin-like protein

72.

73. 74. 75.

76. 77.

78. 79. 80.

81. 82. 83.

from copper-deficient porcine aorta. Biochemistry 8: 2940-2945.1969. Smith DW, Brown DM, Carnes WH. Pmpamtion and ptopcrties of salt soluble elastin. J Biol Chem 247: 24272432.1972. Axelmd 1. Neumtransmitters. Sci Am 59-71, June 1974. Pot CT. The role of copper in the mechanism of tyrosinase action. Pmc XVIth ICCC Dublin 1: 24,1974. Ash DE, Papadopoulos J, Colombo G, Villafranca JJ. Kinetic and spectroscopic studies on the interaction of copper with dopamine-l3-hydmxylase. J Biol Chem 259. 3395-3398,1984. Pmhaska JR. Function of trace elements in brain metabo lism. Physiol Rev 67(3): 859-901, 1987. Dahlstmm A, Fuxe K. Evidence for the existence of monoamine-containing neuron in the central nervous system. Acta Physiol Stand 62, suppl232: l-55, 1964. Coyle JT, Axehod J. Dopaminc-B-hydmxylase in the rat brain. J Neumchem 19: 449-459,1972. Sweeney VP, Pathak MA, Asbury AK. Acute intermittent porphyria. Brain 93: 369-380, 1970. Keiser HR. Henkin RI, Kate M. Reversal by copper of the Proc Sot Exp Biol lathymgenic action of D-penicilhunine. Med 129: 516-522,1968. Mertx W. The essential trace elements. Science 213: 13321338,1981. Bettrand G. Eighth International Congress Applied Chemistry, New York 28: 30 1912. Gadjusek DC. Hypothesis: interference with axonal transport of neurofilamcnt as a common pathogenetic mcchanism in certain diseases of the central nervous system. N Engl J Med 312: 714-719,1985.