Journal of the Neurological Sciences, 1979, 42: 407--4l6 © Elsevier/North-Holland Biomedical Press
T O P O G R A P H I C A L D I S T R I B U T I O N OF ARSENIC, SELENIUM IN T H E N O R M A L H U M A N BRAIN
407
MANGANESE,
AND
N1ELS A. LARSEN 1, HENNING PAKKENBERG x, ELSE DAMSGAARD ~and K. HEYDORN z 1 Department of Neurology, Hvidovre Hospital, 2650 Hvidovre, Copenhagen, and 2 Isotope Division, Risi~ National Laboratory, 5000 Roskilde (Denmark)
(Received 10 January, 1979) (Accepted 4 April, 1979)
SUMMARY The concentrations of arsenic, manganese and selenium per gram wet tissue weight were determined in samples from 24 areas of normal human brains from 5 persons with ages ranging from 15 to 81 years of age. The concentrations of the 3 elements were determined for each sample by means of neutron activation analysis with radiochemical separation. Distinct patterns of distribution were shown for each of the 3 elements. Variations between individuals were found for some but not all brain areas, resulting in coefficients of variation between individuals of about 30 ~o for arsenic, 10 % for manganese and 20 % for selenium. The results seem to indicate that arsenic is associated with the lipid phase, manganese with the dry matter and selenium with the aqueous phase of brain tissue.
INTRODUCTION Manganese and selenium are generally accepted as essential to mammals, while only slender evidence (Nielsen et al. 1975) points to arsenic as yet another essential element. When present in excess all 3 elements produce symptoms of toxicity from a number of organs including the central nervous system and the peripheral nerves. The importance of trace elements to enzyme activity and cell membrane function makes it relevant to map the topographical trace element chemistry of the body. The diversity of the human brain calls for an especially detailed investigation with samples This study was supported by grant No. 512-4164 from the Danish Medical Research Council and grant No. 15k-27c and d from the Danish Atomic Energy Commission. Correspondence to: Professor Henning Pakkenberg, Department of Neurology, Hvidovre Hospital, Kettegaard All6, DK-2650 Hvidovre, Denmark.
408 so small that only neutron activation analysis with elaborate chemical separation is sufficiently accurate when arsenic, manganese, and selenium are concerned. The advantages of unexcelled accuracy and the determination of concentrations of 3 elements from a single sample of less than 1 g, however, is counterbalanced by a very timeconsuming procedure which limits the size of the material investigated. The present material consists of samples from 24 areas of 5 normal human brains. MATERIAL AND METHODS Samples ideally weighing 500-1000 mg were taken from unfixated brain at autopsy. The areas sampled are listed in Table 1, Because of the gelatinous consistency of unfixated brain tissue, small areas such as the nucleus amygdalae and the hypothalamus could not always be accurately identified and were consequently not sampled. However, the loss of a certain number of samples was preferred to the risk of contamination of samples from various fixatives. For some areas the obtainable samples were smaller than desirable from an analytical point of view. The weights of the pineal bodies from two normal persons were 83 and 30 mg, which resulted in less accurate determinations. The group of normal persons comprised 2 women and 3 men with ages ranging from 15 to 81 years. Two of them died from localized brain damage. One had an epidural haematoma in the right parietal region, and the other had subdural and subarachnoid haemorrhages in the right occipital region. In both the injured and closely adjacent areas were avoided in sampling. Among the other 3 persons, one died from multiple traumata without brain lesions, one from cardiac insufficiency due to aortic stenosis, and one from an incarcerated abdominal hernia. None of the 5 persons had previous symptoms or signs of organic brain disease or physical illness apart from the cause of death.
Sampling The autopsies took place from 6 to 72 hr after death. The bodies were kept in refrigerated boxes when more than 6 hr elapsed. The skull was opened by ordinary procedures with metal tools and care was taken not to damage the dura mater which was cut with a polystyrene knife. Polystyrene tools were also used for the rest of the sampling procedure. Each sample was placed in its own polystyrene beaker, which was closed and kept at --15 to --20 °C until the samples could be weighed into halfdram polyvials (Olympic Plastics Corporation, Los Angeles). This took place a few hours to 3 days later. The polyvials were closed and stored at --15 to --20 °C until the time of analysis. To rule out significant contamination of the samples from the tools used, a knife and a beaker were analysed. Possible contaminants were extracted with 5 ml of redistilled water at 95 °C for 30 min, and the extract was subjected to neutron activation analysis for arsenic, manganese and selenium. No significant contribution from either knife or beaker was observed for arsenic and selenium. A possible contribution of manganese of 0.5 ng is below the detection limit for all samples analysed (Larsen et al. 1972).
1.5 3.2 2.0 4.7 1.7 4.7 2.0 5.2 1.3 1.5 1.9 3.4 1.2 2.0 1.8 2.2 2.5 1.0 2.6 1.6 4.3 2.4 3.5 4.4 (0 -2.5) (2.tk3.6) (0.8-2.8) (2.2-5.8) (1.63.7) (1.9-5.2) (2.5-9.1)
(0.9-2.2) (164.4) (0.9-3.2) (2.66.1) (l&2.4) (2.8-7.2) (1.5-2.5) (2.5-8.7) (0 -2.5) (0.5-1.9) (0.8-2.9) (2.14.3) (0.4-1.8) (1.5-2.4) (1.3-2.2) (1 sl-4.0)
5 4 5 5 5 5 5 5 5 4 4 5 4 4 4 5 1 2 3 5 5 4 5 4 162 244 179 268 133 260 204 279 165 149 218 194 333 395 449 270 281 268 271 236 245 169 191 208
mean
(211-324) (213-338) (182-298) (185-291) (13&200) (W-220) (54379)
(143-226) (101-397) (149-211) (181-371) (96-162) (163-310) (177-232) (233-355) (118-211) (1088185) (149-310) (146-235) (269-379) (241-555) (312-552) (199-328)
(range)
: 5 4
: 3 5
5 4 5 5 5 5 5 5 5 4 4 4 4 5 5 5
BRAIN
134 119 156 123 143 173 159 137 145 116 121 101 135 171 211 185 142 98 197 145 157 80 124 114
mean
Selenium
HUMAN
No. of brains sampled
IN 24 AREAS OF NORMAL
mean No. of brains sampled
AND SELENIUM
Manganese (range)
wet tissue weight.
MANGANESE
Arsenic
are nanogram/gram
OF ARSENIC,
Frontal lobe, grey matter Frontal lobe, white matter Parietal lobe, grey matter Parietal lobe, white matter Temporal lobe, grey matter Temporal lobe, white matter Occipital lobe, grey matter Occipital lobe, white matter Insula Hippocampal gyrus Hippocampus Corpus callosum Nucleus caudatus Globus pallidus Putamen Thalamus Hypothalamus Pineal body Substantia nigra Cerebellar cortex Cerebellum, white matter Optic chiasma Pons Medulla oblongata
Area
All concentrations
CONCENTRATIONS
TABLE 1
(57-138) (139-262) (96-209) (91-238) (57-119) (80-155) (78-158)
(101-209) (67-149) (129-212) (84-160) (111-188) (121-313) (127-202) (105-167) (104228) (8&150) (10&150) (89-123) (102-155) (147-196) (160-275) (146-237)
(range)
: 3 5 5 4 5 4
5 4 5 5 5 5 5 4 5 4 4 5 5 5 5 5
No. of brains sampled
410
Analysis technique The samples were irradiated at a thermal neutron flux density of 7 × 1012 n cm -2 s -1 for one hour in the Danish reactor D R 2. After decomposition in a sulphuricnitric acid mixture, chemical separation of aim Se, 76As and 5aMn took place before measurement by gamma-ray spectroscopy. Chemical yields were determined by means of added 54Mn tracer and by re-irradiation of the separated arsenic and selenium samples. Details of the analytical procedure have been published separately (Heydorn and Damsgaard 1973). RESULTS For each of the 24 brain areas the mean values and ranges of arsenic, manganese and selenium concentrations with the number of brains sampled are shown in Table 1. In order to assess the variation between individuals, a two-way variance analysis was carried out for each of the 3 elements, including all brain areas in which samples had been obtained from all 5 persons. The coefficient of variation between individuals was about 30 ~ for arsenic, about 10 % for manganese, and about 20 % for selenium. The coefficients of variation for arsenic and manganese are lower than those found for other organs of normal man, while the figure for selenium corresponds to those found for other organs (Larsen et al. 1972). Assuming that grey and white matter are present in equal amounts within the central nervous system, and employing the mean brain weights determined by Pakkenberg and Voigt (1964), we estimated the total brain contents of the 3 elements. Since accurate estimations or determinations of the total body contents of these elements are not available, estimates of the amounts found in the liver and in total muscle mass, based on our own previous investigations (Larsen et al. 1972) are shown for comparison (Table 2).
TABLE 2 ESTIMATED CONTENT OF ARSENIC, MANGANESE AND SELENIUM IN HUMAN BRAIN, LIVER AND TOTAL MUSCLE MASS The ranges shown represent mean :k 1 SD in rounded figures. The figures for liver and muscle are based on results published by Larsen et al. (1972). Organ (weight) Brain ( 1400 g) Liver (1500 g) Muscle mass (28 kg)
Arsenic (/~g)
Manganese (/zg)
Selenium (pg)
3-5
310--380
160-240
7-26
1370-1930
400-770
65-160
1380-2200
3700-5800
411
Arsenic Highly significant differences in arsenic concentrations (P < 0.001) were found between cerebral white matter and grey matter of the cerebral cortex with a white to grey ratio of 2.70 ± 0.25 (SEM). A similar ratio was found between cerebellar white matter and cerebellar cortex. Neither white nor grey matter of the cerebral cortex showed any topographical differences in concentrations between the samples from different lobes of the same individual. In this respect the corpus callosum follows the white matter of the cerebral hemisphere. While the differences in absolute concentrations of arsenic between individuals amount to about 30 ~o, the white to grey ratio of cerebral cortex and cerebellum shows no significant variation between individuals. Nor do the absolute concentrations of arsenic in cerebellar white matter and cerebellar cortex show significant variations between individuals. A number of areas including the insula, the hippocampal gyrus, the hippocampus, the basal ganglia, substantia nigra, and the hypothalamus show no significant differences in concentration within the same individual, but show variation between individuals. Large and unsystematic differencee in concentrations found in the pons and medulla oblongata are probably due to the inhomogeneous composition of samples from these tissues. Similar differences in the well defined and homogeneous optic chiasma were observed but remain unexplained. The arsenic concentration in peripheral nerves (Larsen et al. 1972) does not differ significantly from that of white matter of the central nervous system.
Manganese The manganese concentration ratio of cerebral white matter to grey matter of the cerebral cortex is 1.44 ~ 0.12 (SEM). About the same ratio obtains for dry substance ratio of white to grey matter (Tower 1969). Thus, related to dry weight, equal concentrations of manganese are found in white and grey matter of the cerebral cortex. In the cerebellum no significant difference is found between the manganese concentrations of the white matter and the cortex. This part of the brain, however, shows bigger individual differences in manganese concentrations than any other brain area sampled. In order to deduce from the results a possible division of the brain into manganese compartments, the manganese concentrations of the other brain areas were related to the individual concentrations of the cerebellum. The cerebral cortex and the basal ganglia both show individual variations significantly different from those of the cerebellum and thus belong to different compartments. Identical areas of the grey matter of cerebral cortex show no significant differences between individuals, whereas the differences in concentration between the different lobes of the same individual are considerably greater than the differences between individuals and show a consistent pattern. The globus pallidus and the putamen show no significant difference between individuals, nor do the insula, gyrus hippocampi and hippocampus. The corpus callosum, pons, medulla oblongata, and optic chiasma show no significant differences between either individuals or areas. The distribution of manganese in the basal ganglia with higher concentrations in the nucleus caudatus, globus pallidus and putamen than in the thalamus and substantia nigra is consistently found although the absolute concentrations vary.
412 The manganese concentrations found in white matter of the central nervous system are significantly higher than the concentrations found in peripheral nerves (Larsen et al. 1972). Selenium Significant differences between individuals are found for grey matter of cerebral cortex, pons, medulla oblongata and the optic chiasma. These differences disappear if the absolute concentrations are transformed to concentrations relative to the individual means for cerebral cortex. The white matter of the cerebral hemisphere, corpus callosum, insula, gyrus hippocampi and hippocampus show no significant differences in concentration either between areas or between individuals. Within the same individual the samples representing the 4 lobes of cerebral cortex show no significant differences. However, while the ratio between the mean concentrations of selenium in cerebral white matter and grey matter of cerebral cortex is close to unity, the variances for white and grey matter show a highly significant difference. The ratio of water content between white and grey matter is quite close to unity. This similarity in distribution might be due to water solubility of the majority of the selenium found in the brain. In the cerebellum no significant differences are found between either individuals or between cortex and white matter. Neither are individual differences found for the basal ganglia and the substantia nigra. However, the selenium concentrations of the lentiform nucleus, thalamus and substantia nigra are significantly higher than those of the caudate nucleus and the hypothalamus. The selenium concentration of white matter of the central nervous system is significantly higher than in peripheral nerves (Larsen et al. 1972). DISCUSSION The advance of analytical technique has led to a stepwise decrease in the concentrations of trace elements in biological material accepted as the physiological levels. The concentration of arsenic in normal human brain has attracted little interest. Using nuclear activation analysis with radiochemical separation Smith (1967) found a mean value of 16 ng arsenic per gram dry tissue with a range of one to 36 ng per gram dry tissue. These results are based on 19 samples from undefined brain areas of several individuals. If the dry tissue weight is assumed to be 20-30 ~ of the wet tissue weight, the concentrations are in the same range as ours. While the concentration of arsenic in grey matter of the central nervous system is significantly lower than in either white matter or peripheral nerves, it is higher than that found in normal human serum (Damsgaard et al. 1973). This difference in concentrations might represent a binding or affinity to substances or structures present in varying quantity in different tissues. The arsenic concentration ratio between white and grey matter of the cerebral and cerebellar cortex is very close to the ratio for total lipids, phospholipids and phosphatides. This accords well with the proposition of
413 Schroeder and Balassa (1966) based on the distribution of arsenic in marine fish, that arsenic has a predilection for fat. Although manganese has received more attention than arsenic in the trace element literature, little information is available on its topographical distribution in the human brain. By pooling samples from several brains, Tingey (1937) succeeded in determining manganese concentrations in cerebral cortex, cerebral white matter, corpus striatum, and cerebellar cortex practically identical to our results. With emission spectroscopy Alexander and Myerson (1938) analysed single samples from normal cerebral grey and white matter and found about twice the levels demonstrated in this investigation. In an investigation of trace element levels in subjects from Africa, the Near and Far East, Europe and America, also employing emission spectroscopy (Tipton et al. 1965; Schroeder et al. 1966), manganese levels in human brain similar to our results were found except for some extremely high values probably due to contamination of occasional samples. With atomic absorption spectroscopy Banta et al. (1975) found manganese concentrations of 250 and 340 ng per gram wet tissue in brain biopsies from the frontal cortex of two demented patients. Since such biopsies contain both grey and white matter, these concentrations are in accordance with ours. On the other hand Lechner et al. (1966) and Lehmann et al. (1971), both using spectrophotometric analysis, found manganese concentrations about twice as high as ours and with a different topographical pattern of concentrations. Similar levels were found by Leu et al. (1971) using neutron activation analysis with radiochemical separation for 4 brain samples from two normal persons, and by Yase et al. (1973) using non-destructive neutron activation analysis of samples from one normal brain fixed for 8 days in 80 ~ ethanol. While Yase et al. found higher concentrations than we did with the more accurate and sensitive analysis procedure of radiochemical separation, the relative concentrations show similarities. Thus, both materials show higher than average levels in the basal ganglia, especially in the caudate nucleus and the putamen. No difference in concentration between white and grey matter of cerebral cortex is apparent in the results of Yase et al. Since their results are related to dry tissue weight, this substantiates our hypothesis that the manganese content is related to the amount of dry substance in the brain samples. Determination of manganese concentrations in defined areas or whole brains of rhesus monkeys (Bird et al. 1967), cows and pigs (Wong and Fritze 1969), and Swiss albino mice (Cotzias et al. 1974) by means of neutron activation analysis with radiochemical separation show manganese concentrations similar to our results for human brains. Thus, with comparable analytical procedures little difference is found between different mammals including man and between different geographical areas. The study with S4Mn in rhesus monkeys of Dastur et al. (1971) supports our proposition of several manganese compartments within the brain. In their study the relatively highest manganese concentration was found in the cerebellum which shows the greatest individual variations in our investigation. The basal ganglia showed an equally high 5aMn activity and the highest concentrations of stable manganese in the brain. In both studies a rating of the nuclei of the basal ganglia according to activity/ concentration places the lentiform nucleus (putamen q- globus pallidus) higher than
414 the thalamus and the caudate nucleus. This distribution of manganese between the different parts of the basal ganglia suggests a functional significance, although the absolute levels of concentration do not seem to be critical. Similarly the systematic differences in manganese concentration found between the various parts of the grey matter of the cerebral cortex and the absence of significant individual differences suggest a specific function for manganese. In their investigation spanning 278 days Dastur et al. found that allthe components of the nervous system seemed to retain rather than to discharge manganese. This, however, is not consistent with the absence of correlation of manganese concentration with age found in this as well as in earlier investigations (Tingey 1937). Selenium concentrations in the normal human brain have been investigated by means of neutron activation analysis by Dickson and Tomlinson (1967), Henke et al. (1971), Schicha et al. (1974), and H6ck et al. (1975). Henke et al. record selenium concentrations in 14 different brain areas ranging from 370 to 13,550 ng per gram wet tissue weight, while our results show a total range of 57-313 ng per gram wet tissue weight. Selenium concentrations of cerebral cortex reported by the other groups and by Schroeder et al. (1970) are similar to our results. The existence of several selenium compartments within the brain is suggested by the differences in individual variations between different brain areas described in the Results section. Thus, grey matter of the cerebral cortex seems to belong to a different compartment from underlying white matter. From a study of selenium-deficient rats injected with Na,~ 75SeOa, Trapp and Millam (1975) report significant difference in 75Se uptake between the cerebral hemispheres and the cerebellum. These authors also point out that much of the 75Se was soluble in an aqueous solution. The selenium concentrations related to dry tissue weight published by Schicha et al. (1974) show a higher selenium level in grey than in white matter. If the different water content of grey and white matter is taken into account, this difference in selenium concentration is eliminated. Thus, it is likely that the majority of selenium in the brain is in the aqueous phase. The investigation confirms that arsenic, manganese and selenium are not randomly distributed throughout the central nervous system. The results are compatible with an association between arsenic and the lipid phase, between manganese and the dry matter, and between selenium and the aqueous phase. Manganese and selenium are probably distributed in several different compartments within the brain. Apart from this a specific distribution of manganese in the grey matter of cerebral cortex and in the basal ganglia points towards a special role for this element. On the other hand, unless frank disturbances in trace element metabolism are discovered - as in Menkes' disease (Heydorn et al. 1976) - the mapping of trace element concentrations in the human brain has no immediate consequences. However, with the accumulation of similar topographical investigations of enzyme activities, membrane physiology, synaptic transmission, and the composition of subcellular fractions, information on trace element concentrations will be necessary to provide a comprehensive understanding of the biochemistry of the brain.
415 REFERENCES Alexander, L. and A. Myerson (1938) Minerals in normal and in pathologic brain tissue, studied by microincineration and spectroscopy, Arch. NeuroL Psychiat. (Chic.), 39: 131-149. Banta, R. G., D. Clark and W. Markesberry (1975) Elevated manganese levels associated with Alzheimer's disease and extrapyramidal signs, Neurology (Minneap.), 25: 354. Bird, E. D., L. Grant and W. Ellis (1967) Measurement of the effect of phenothiazine on the manganese concentration in the basal ganglia of sub-human primates by activation analysis, ln: Nuclear Activation Techniques in the Life Sciences, IAEA, Vienna, pp. 491-499. Cotzias, G. C., P. S. Papavasiliou, I. Mena, L. C. Tang and S. T. Millner (1974) Manganese and catecholamines. In: Advances in Neurology, Vol. 5, Raven Press, New York, pp. 235-243. Damsgaard, E., K. Heydorn, N. A. Larsen and B. Nielsen (1973) Simultaneous determination of arsenic, manganese and selenium in human serum by neutron activation analysis, Ris~ Report No. 271, pp. 1-35. Dastur, D. K., D. K. Mangani and K. V. Raghavendran (1971) Distribution and fate of 54Mn in the monkey - - Studies of different parts of the central nervous system and other organs, J. clin. Inve.~t., 50: 9-20. Dickson, R. C. and R. H. Tomlinson (1967) Selenium in blood and human tissues, Clin. chim. Acta, 16: 311-321. Henke, G., H. Mollmann and H. Alfes (1971) Vergleichende Untersuchungen fiber die Konzentration einiger Spurenelemente in menschlichen Hirnarealen durch Neutronaktivierungsanalyse. Z. Neurol., 199: 283-294. Heydorn, K. and E. Damsgaard (1973) Simultaneous determination of arsenic, manganese and selenium in biological materials by neutron activation analysis, Talanta, 20:1-11. Heydorn, K., E. Damsgaard, N. Horn, M. Mikkelsen, I. Tygstrup, S. Vestermark and J. Weber (1975) Extrahepatic storage of copper - - A male foetus suspected of Menkes' disease, Humangenetik, Vol. 29, pp. 171-175. H6ck, A., U. Demmel, H. Schicha, K. Kasparek and L. E. Feinendegen (1975) Trace element concentration in human brain - - Activation analysis of cobalt, iron, rubidium, selenium, zinc, chromium, silver, cesium, antimony and scandium, Brain, 98 : 49-64. Horn, N. (1976) Copper incorporation studies on cultured cells for prenatal diagnosis of Menkes' disease, Lancet, 1 : 1156-1158. Larsen, N. A., B. Nielsen, H. Pakkenberg, P. Christoffersen, E. Damsgaard and K. Heydorn (1972) Neutron activation analysis of arsenic, manganese and selenium concentrations in organs of uraemic and normal persons. In: Nuclear Activation Techniques in the Life Sciences, IAEA, Vienna, pp. 561-568. Lechner, H., W. Beyer, O. Wawschinek, H. Wielinger and H. H. Tagger (1966) Quantitative Untersuchungen fiber die Verteilung von Mangan im menschlichen Gehirn, Wien. klin. Wschr., 78: 328-329. Lehmann, B. H., J. D. L. Hansen and P. J. Warren (1971) The distribution of copl~er, zinc and manganese in various regions of the brain and in other tissues of children with protein-calorie malnutrition, Brit. J. Nutr., 26: 197-202. Leu, M. L., G. T. Strickland and S. J. Yeh (1971) Tissue copper, zinc and manganese levels in Wilson's disease - - Studies with the use of neutron activation analysis, J. Lab. clin. Med., 77:438 44~. Nielsen, F. H., S. H. Givand and D. R. Myron (1975) Evidence of a possible requirement for arsenic by the rat, Fed. Proc., 34 (No. 3987): 923. Pakkenberg, H. and J. Voigt (1964) Brain weight of the Danes - - A forensic material, Acta anat. (Basel), 56: 297-307. Schicha, H., W. Mtiller, K. Kasparek and R. Schr~Sder (1974) Neutronaktivierungsanalytische Bestimmung der Spurenelemente Kobalt, Eisen, Rubidium, Selen, Zink, Chrom, Silber, Caesium, Antimon und Scandium in operativ entnommen Hirntumoren des Menschen (1. Mitteilung), Beitr. Path. Anat., 151: 281-296. Schroeder, H. A. and J. J. Balassa (1966) Abnormal trace metals in man - - Arsenic, J. chron. Dis., 19: 85-106. Schroeder, H. A., J. J. Balassa and I. H. Tipton (1966) Essential trace metals in man - - Manganese. A study in homeostasis, J. chron. Dis., 19: 545-571. Schroeder, H. A., D. V. Frost and J. J. Balassa (1970) Essential trace metals in man - - Selenium, J. chron. Dis., 23 : 227-243.
416 Smith, H. (1967) The distribution of antimony, arsenic, copper and zinc in human tissue, J. forens. Sci. Soc., 7: 97-102. Tingey, A. H. (1937) The iron, copper and manganese content of the human brain, J. ment. Sci., 83 : 452-460. Tipton, I. H., H. A. Schroeder, H. M. Perry, Jr. and M. J. Cook (1965) Trace elements in human tissue, Part 3 (Subjects from Africa, the Near and Far East and Europe), Health Physics, 11 : 403-451. Tower, D. B. (1969) Inorganic constituents. In A. Lajtha (Ed.), Handbook ofNeurochemistry, Vol. 1, Plenum Press, New York, pp. 1-19. Trapp, G. A. and J. Millam (1975) The distribution of 7~Se in brains of selenium-deficient rats, J. Neurochem., 24: 593-596. Wong, P. Y. and K. Fritze (1969) Determination by neutron activation of copper, manganese and zinc in the pineal body and other areas of brain tissue, J. Neurochem., 16: 1231-1234. Yase, Y., Y. Shinjo, F. Yoshimasu and T. Kumamoto (1973) Neutron activation analysis studies in degenerative diseases of the nervous system, with special reference to divalent cation metals. In: Proc. 2nd Internat. Congress of Muscle Diseases, Perth, 1971 (International Congress Series, No. 295), Excerpta Medica, Amsterdam, pp. 226-234.
T O P O G R A P H I C A L D I S T R I B U T I O N OF ARSENIC, SELENIUM IN T H E N O R M A L H U M A N BRAIN
407
MANGANESE,
AND
N1ELS A. LARSEN 1, HENNING PAKKENBERG x, ELSE DAMSGAARD ~and K. HEYDORN z 1 Department of Neurology, Hvidovre Hospital, 2650 Hvidovre, Copenhagen, and 2 Isotope Division, Risi~ National Laboratory, 5000 Roskilde (Denmark)
(Received 10 January, 1979) (Accepted 4 April, 1979)
SUMMARY The concentrations of arsenic, manganese and selenium per gram wet tissue weight were determined in samples from 24 areas of normal human brains from 5 persons with ages ranging from 15 to 81 years of age. The concentrations of the 3 elements were determined for each sample by means of neutron activation analysis with radiochemical separation. Distinct patterns of distribution were shown for each of the 3 elements. Variations between individuals were found for some but not all brain areas, resulting in coefficients of variation between individuals of about 30 ~o for arsenic, 10 % for manganese and 20 % for selenium. The results seem to indicate that arsenic is associated with the lipid phase, manganese with the dry matter and selenium with the aqueous phase of brain tissue.
INTRODUCTION Manganese and selenium are generally accepted as essential to mammals, while only slender evidence (Nielsen et al. 1975) points to arsenic as yet another essential element. When present in excess all 3 elements produce symptoms of toxicity from a number of organs including the central nervous system and the peripheral nerves. The importance of trace elements to enzyme activity and cell membrane function makes it relevant to map the topographical trace element chemistry of the body. The diversity of the human brain calls for an especially detailed investigation with samples This study was supported by grant No. 512-4164 from the Danish Medical Research Council and grant No. 15k-27c and d from the Danish Atomic Energy Commission. Correspondence to: Professor Henning Pakkenberg, Department of Neurology, Hvidovre Hospital, Kettegaard All6, DK-2650 Hvidovre, Denmark.
408 so small that only neutron activation analysis with elaborate chemical separation is sufficiently accurate when arsenic, manganese, and selenium are concerned. The advantages of unexcelled accuracy and the determination of concentrations of 3 elements from a single sample of less than 1 g, however, is counterbalanced by a very timeconsuming procedure which limits the size of the material investigated. The present material consists of samples from 24 areas of 5 normal human brains. MATERIAL AND METHODS Samples ideally weighing 500-1000 mg were taken from unfixated brain at autopsy. The areas sampled are listed in Table 1, Because of the gelatinous consistency of unfixated brain tissue, small areas such as the nucleus amygdalae and the hypothalamus could not always be accurately identified and were consequently not sampled. However, the loss of a certain number of samples was preferred to the risk of contamination of samples from various fixatives. For some areas the obtainable samples were smaller than desirable from an analytical point of view. The weights of the pineal bodies from two normal persons were 83 and 30 mg, which resulted in less accurate determinations. The group of normal persons comprised 2 women and 3 men with ages ranging from 15 to 81 years. Two of them died from localized brain damage. One had an epidural haematoma in the right parietal region, and the other had subdural and subarachnoid haemorrhages in the right occipital region. In both the injured and closely adjacent areas were avoided in sampling. Among the other 3 persons, one died from multiple traumata without brain lesions, one from cardiac insufficiency due to aortic stenosis, and one from an incarcerated abdominal hernia. None of the 5 persons had previous symptoms or signs of organic brain disease or physical illness apart from the cause of death.
Sampling The autopsies took place from 6 to 72 hr after death. The bodies were kept in refrigerated boxes when more than 6 hr elapsed. The skull was opened by ordinary procedures with metal tools and care was taken not to damage the dura mater which was cut with a polystyrene knife. Polystyrene tools were also used for the rest of the sampling procedure. Each sample was placed in its own polystyrene beaker, which was closed and kept at --15 to --20 °C until the samples could be weighed into halfdram polyvials (Olympic Plastics Corporation, Los Angeles). This took place a few hours to 3 days later. The polyvials were closed and stored at --15 to --20 °C until the time of analysis. To rule out significant contamination of the samples from the tools used, a knife and a beaker were analysed. Possible contaminants were extracted with 5 ml of redistilled water at 95 °C for 30 min, and the extract was subjected to neutron activation analysis for arsenic, manganese and selenium. No significant contribution from either knife or beaker was observed for arsenic and selenium. A possible contribution of manganese of 0.5 ng is below the detection limit for all samples analysed (Larsen et al. 1972).
1.5 3.2 2.0 4.7 1.7 4.7 2.0 5.2 1.3 1.5 1.9 3.4 1.2 2.0 1.8 2.2 2.5 1.0 2.6 1.6 4.3 2.4 3.5 4.4 (0 -2.5) (2.tk3.6) (0.8-2.8) (2.2-5.8) (1.63.7) (1.9-5.2) (2.5-9.1)
(0.9-2.2) (164.4) (0.9-3.2) (2.66.1) (l&2.4) (2.8-7.2) (1.5-2.5) (2.5-8.7) (0 -2.5) (0.5-1.9) (0.8-2.9) (2.14.3) (0.4-1.8) (1.5-2.4) (1.3-2.2) (1 sl-4.0)
5 4 5 5 5 5 5 5 5 4 4 5 4 4 4 5 1 2 3 5 5 4 5 4 162 244 179 268 133 260 204 279 165 149 218 194 333 395 449 270 281 268 271 236 245 169 191 208
mean
(211-324) (213-338) (182-298) (185-291) (13&200) (W-220) (54379)
(143-226) (101-397) (149-211) (181-371) (96-162) (163-310) (177-232) (233-355) (118-211) (1088185) (149-310) (146-235) (269-379) (241-555) (312-552) (199-328)
(range)
: 5 4
: 3 5
5 4 5 5 5 5 5 5 5 4 4 4 4 5 5 5
BRAIN
134 119 156 123 143 173 159 137 145 116 121 101 135 171 211 185 142 98 197 145 157 80 124 114
mean
Selenium
HUMAN
No. of brains sampled
IN 24 AREAS OF NORMAL
mean No. of brains sampled
AND SELENIUM
Manganese (range)
wet tissue weight.
MANGANESE
Arsenic
are nanogram/gram
OF ARSENIC,
Frontal lobe, grey matter Frontal lobe, white matter Parietal lobe, grey matter Parietal lobe, white matter Temporal lobe, grey matter Temporal lobe, white matter Occipital lobe, grey matter Occipital lobe, white matter Insula Hippocampal gyrus Hippocampus Corpus callosum Nucleus caudatus Globus pallidus Putamen Thalamus Hypothalamus Pineal body Substantia nigra Cerebellar cortex Cerebellum, white matter Optic chiasma Pons Medulla oblongata
Area
All concentrations
CONCENTRATIONS
TABLE 1
(57-138) (139-262) (96-209) (91-238) (57-119) (80-155) (78-158)
(101-209) (67-149) (129-212) (84-160) (111-188) (121-313) (127-202) (105-167) (104228) (8&150) (10&150) (89-123) (102-155) (147-196) (160-275) (146-237)
(range)
: 3 5 5 4 5 4
5 4 5 5 5 5 5 4 5 4 4 5 5 5 5 5
No. of brains sampled
410
Analysis technique The samples were irradiated at a thermal neutron flux density of 7 × 1012 n cm -2 s -1 for one hour in the Danish reactor D R 2. After decomposition in a sulphuricnitric acid mixture, chemical separation of aim Se, 76As and 5aMn took place before measurement by gamma-ray spectroscopy. Chemical yields were determined by means of added 54Mn tracer and by re-irradiation of the separated arsenic and selenium samples. Details of the analytical procedure have been published separately (Heydorn and Damsgaard 1973). RESULTS For each of the 24 brain areas the mean values and ranges of arsenic, manganese and selenium concentrations with the number of brains sampled are shown in Table 1. In order to assess the variation between individuals, a two-way variance analysis was carried out for each of the 3 elements, including all brain areas in which samples had been obtained from all 5 persons. The coefficient of variation between individuals was about 30 ~ for arsenic, about 10 % for manganese, and about 20 % for selenium. The coefficients of variation for arsenic and manganese are lower than those found for other organs of normal man, while the figure for selenium corresponds to those found for other organs (Larsen et al. 1972). Assuming that grey and white matter are present in equal amounts within the central nervous system, and employing the mean brain weights determined by Pakkenberg and Voigt (1964), we estimated the total brain contents of the 3 elements. Since accurate estimations or determinations of the total body contents of these elements are not available, estimates of the amounts found in the liver and in total muscle mass, based on our own previous investigations (Larsen et al. 1972) are shown for comparison (Table 2).
TABLE 2 ESTIMATED CONTENT OF ARSENIC, MANGANESE AND SELENIUM IN HUMAN BRAIN, LIVER AND TOTAL MUSCLE MASS The ranges shown represent mean :k 1 SD in rounded figures. The figures for liver and muscle are based on results published by Larsen et al. (1972). Organ (weight) Brain ( 1400 g) Liver (1500 g) Muscle mass (28 kg)
Arsenic (/~g)
Manganese (/zg)
Selenium (pg)
3-5
310--380
160-240
7-26
1370-1930
400-770
65-160
1380-2200
3700-5800
411
Arsenic Highly significant differences in arsenic concentrations (P < 0.001) were found between cerebral white matter and grey matter of the cerebral cortex with a white to grey ratio of 2.70 ± 0.25 (SEM). A similar ratio was found between cerebellar white matter and cerebellar cortex. Neither white nor grey matter of the cerebral cortex showed any topographical differences in concentrations between the samples from different lobes of the same individual. In this respect the corpus callosum follows the white matter of the cerebral hemisphere. While the differences in absolute concentrations of arsenic between individuals amount to about 30 ~o, the white to grey ratio of cerebral cortex and cerebellum shows no significant variation between individuals. Nor do the absolute concentrations of arsenic in cerebellar white matter and cerebellar cortex show significant variations between individuals. A number of areas including the insula, the hippocampal gyrus, the hippocampus, the basal ganglia, substantia nigra, and the hypothalamus show no significant differences in concentration within the same individual, but show variation between individuals. Large and unsystematic differencee in concentrations found in the pons and medulla oblongata are probably due to the inhomogeneous composition of samples from these tissues. Similar differences in the well defined and homogeneous optic chiasma were observed but remain unexplained. The arsenic concentration in peripheral nerves (Larsen et al. 1972) does not differ significantly from that of white matter of the central nervous system.
Manganese The manganese concentration ratio of cerebral white matter to grey matter of the cerebral cortex is 1.44 ~ 0.12 (SEM). About the same ratio obtains for dry substance ratio of white to grey matter (Tower 1969). Thus, related to dry weight, equal concentrations of manganese are found in white and grey matter of the cerebral cortex. In the cerebellum no significant difference is found between the manganese concentrations of the white matter and the cortex. This part of the brain, however, shows bigger individual differences in manganese concentrations than any other brain area sampled. In order to deduce from the results a possible division of the brain into manganese compartments, the manganese concentrations of the other brain areas were related to the individual concentrations of the cerebellum. The cerebral cortex and the basal ganglia both show individual variations significantly different from those of the cerebellum and thus belong to different compartments. Identical areas of the grey matter of cerebral cortex show no significant differences between individuals, whereas the differences in concentration between the different lobes of the same individual are considerably greater than the differences between individuals and show a consistent pattern. The globus pallidus and the putamen show no significant difference between individuals, nor do the insula, gyrus hippocampi and hippocampus. The corpus callosum, pons, medulla oblongata, and optic chiasma show no significant differences between either individuals or areas. The distribution of manganese in the basal ganglia with higher concentrations in the nucleus caudatus, globus pallidus and putamen than in the thalamus and substantia nigra is consistently found although the absolute concentrations vary.
412 The manganese concentrations found in white matter of the central nervous system are significantly higher than the concentrations found in peripheral nerves (Larsen et al. 1972). Selenium Significant differences between individuals are found for grey matter of cerebral cortex, pons, medulla oblongata and the optic chiasma. These differences disappear if the absolute concentrations are transformed to concentrations relative to the individual means for cerebral cortex. The white matter of the cerebral hemisphere, corpus callosum, insula, gyrus hippocampi and hippocampus show no significant differences in concentration either between areas or between individuals. Within the same individual the samples representing the 4 lobes of cerebral cortex show no significant differences. However, while the ratio between the mean concentrations of selenium in cerebral white matter and grey matter of cerebral cortex is close to unity, the variances for white and grey matter show a highly significant difference. The ratio of water content between white and grey matter is quite close to unity. This similarity in distribution might be due to water solubility of the majority of the selenium found in the brain. In the cerebellum no significant differences are found between either individuals or between cortex and white matter. Neither are individual differences found for the basal ganglia and the substantia nigra. However, the selenium concentrations of the lentiform nucleus, thalamus and substantia nigra are significantly higher than those of the caudate nucleus and the hypothalamus. The selenium concentration of white matter of the central nervous system is significantly higher than in peripheral nerves (Larsen et al. 1972). DISCUSSION The advance of analytical technique has led to a stepwise decrease in the concentrations of trace elements in biological material accepted as the physiological levels. The concentration of arsenic in normal human brain has attracted little interest. Using nuclear activation analysis with radiochemical separation Smith (1967) found a mean value of 16 ng arsenic per gram dry tissue with a range of one to 36 ng per gram dry tissue. These results are based on 19 samples from undefined brain areas of several individuals. If the dry tissue weight is assumed to be 20-30 ~ of the wet tissue weight, the concentrations are in the same range as ours. While the concentration of arsenic in grey matter of the central nervous system is significantly lower than in either white matter or peripheral nerves, it is higher than that found in normal human serum (Damsgaard et al. 1973). This difference in concentrations might represent a binding or affinity to substances or structures present in varying quantity in different tissues. The arsenic concentration ratio between white and grey matter of the cerebral and cerebellar cortex is very close to the ratio for total lipids, phospholipids and phosphatides. This accords well with the proposition of
413 Schroeder and Balassa (1966) based on the distribution of arsenic in marine fish, that arsenic has a predilection for fat. Although manganese has received more attention than arsenic in the trace element literature, little information is available on its topographical distribution in the human brain. By pooling samples from several brains, Tingey (1937) succeeded in determining manganese concentrations in cerebral cortex, cerebral white matter, corpus striatum, and cerebellar cortex practically identical to our results. With emission spectroscopy Alexander and Myerson (1938) analysed single samples from normal cerebral grey and white matter and found about twice the levels demonstrated in this investigation. In an investigation of trace element levels in subjects from Africa, the Near and Far East, Europe and America, also employing emission spectroscopy (Tipton et al. 1965; Schroeder et al. 1966), manganese levels in human brain similar to our results were found except for some extremely high values probably due to contamination of occasional samples. With atomic absorption spectroscopy Banta et al. (1975) found manganese concentrations of 250 and 340 ng per gram wet tissue in brain biopsies from the frontal cortex of two demented patients. Since such biopsies contain both grey and white matter, these concentrations are in accordance with ours. On the other hand Lechner et al. (1966) and Lehmann et al. (1971), both using spectrophotometric analysis, found manganese concentrations about twice as high as ours and with a different topographical pattern of concentrations. Similar levels were found by Leu et al. (1971) using neutron activation analysis with radiochemical separation for 4 brain samples from two normal persons, and by Yase et al. (1973) using non-destructive neutron activation analysis of samples from one normal brain fixed for 8 days in 80 ~ ethanol. While Yase et al. found higher concentrations than we did with the more accurate and sensitive analysis procedure of radiochemical separation, the relative concentrations show similarities. Thus, both materials show higher than average levels in the basal ganglia, especially in the caudate nucleus and the putamen. No difference in concentration between white and grey matter of cerebral cortex is apparent in the results of Yase et al. Since their results are related to dry tissue weight, this substantiates our hypothesis that the manganese content is related to the amount of dry substance in the brain samples. Determination of manganese concentrations in defined areas or whole brains of rhesus monkeys (Bird et al. 1967), cows and pigs (Wong and Fritze 1969), and Swiss albino mice (Cotzias et al. 1974) by means of neutron activation analysis with radiochemical separation show manganese concentrations similar to our results for human brains. Thus, with comparable analytical procedures little difference is found between different mammals including man and between different geographical areas. The study with S4Mn in rhesus monkeys of Dastur et al. (1971) supports our proposition of several manganese compartments within the brain. In their study the relatively highest manganese concentration was found in the cerebellum which shows the greatest individual variations in our investigation. The basal ganglia showed an equally high 5aMn activity and the highest concentrations of stable manganese in the brain. In both studies a rating of the nuclei of the basal ganglia according to activity/ concentration places the lentiform nucleus (putamen q- globus pallidus) higher than
414 the thalamus and the caudate nucleus. This distribution of manganese between the different parts of the basal ganglia suggests a functional significance, although the absolute levels of concentration do not seem to be critical. Similarly the systematic differences in manganese concentration found between the various parts of the grey matter of the cerebral cortex and the absence of significant individual differences suggest a specific function for manganese. In their investigation spanning 278 days Dastur et al. found that allthe components of the nervous system seemed to retain rather than to discharge manganese. This, however, is not consistent with the absence of correlation of manganese concentration with age found in this as well as in earlier investigations (Tingey 1937). Selenium concentrations in the normal human brain have been investigated by means of neutron activation analysis by Dickson and Tomlinson (1967), Henke et al. (1971), Schicha et al. (1974), and H6ck et al. (1975). Henke et al. record selenium concentrations in 14 different brain areas ranging from 370 to 13,550 ng per gram wet tissue weight, while our results show a total range of 57-313 ng per gram wet tissue weight. Selenium concentrations of cerebral cortex reported by the other groups and by Schroeder et al. (1970) are similar to our results. The existence of several selenium compartments within the brain is suggested by the differences in individual variations between different brain areas described in the Results section. Thus, grey matter of the cerebral cortex seems to belong to a different compartment from underlying white matter. From a study of selenium-deficient rats injected with Na,~ 75SeOa, Trapp and Millam (1975) report significant difference in 75Se uptake between the cerebral hemispheres and the cerebellum. These authors also point out that much of the 75Se was soluble in an aqueous solution. The selenium concentrations related to dry tissue weight published by Schicha et al. (1974) show a higher selenium level in grey than in white matter. If the different water content of grey and white matter is taken into account, this difference in selenium concentration is eliminated. Thus, it is likely that the majority of selenium in the brain is in the aqueous phase. The investigation confirms that arsenic, manganese and selenium are not randomly distributed throughout the central nervous system. The results are compatible with an association between arsenic and the lipid phase, between manganese and the dry matter, and between selenium and the aqueous phase. Manganese and selenium are probably distributed in several different compartments within the brain. Apart from this a specific distribution of manganese in the grey matter of cerebral cortex and in the basal ganglia points towards a special role for this element. On the other hand, unless frank disturbances in trace element metabolism are discovered - as in Menkes' disease (Heydorn et al. 1976) - the mapping of trace element concentrations in the human brain has no immediate consequences. However, with the accumulation of similar topographical investigations of enzyme activities, membrane physiology, synaptic transmission, and the composition of subcellular fractions, information on trace element concentrations will be necessary to provide a comprehensive understanding of the biochemistry of the brain.
415 REFERENCES Alexander, L. and A. Myerson (1938) Minerals in normal and in pathologic brain tissue, studied by microincineration and spectroscopy, Arch. NeuroL Psychiat. (Chic.), 39: 131-149. Banta, R. G., D. Clark and W. Markesberry (1975) Elevated manganese levels associated with Alzheimer's disease and extrapyramidal signs, Neurology (Minneap.), 25: 354. Bird, E. D., L. Grant and W. Ellis (1967) Measurement of the effect of phenothiazine on the manganese concentration in the basal ganglia of sub-human primates by activation analysis, ln: Nuclear Activation Techniques in the Life Sciences, IAEA, Vienna, pp. 491-499. Cotzias, G. C., P. S. Papavasiliou, I. Mena, L. C. Tang and S. T. Millner (1974) Manganese and catecholamines. In: Advances in Neurology, Vol. 5, Raven Press, New York, pp. 235-243. Damsgaard, E., K. Heydorn, N. A. Larsen and B. Nielsen (1973) Simultaneous determination of arsenic, manganese and selenium in human serum by neutron activation analysis, Ris~ Report No. 271, pp. 1-35. Dastur, D. K., D. K. Mangani and K. V. Raghavendran (1971) Distribution and fate of 54Mn in the monkey - - Studies of different parts of the central nervous system and other organs, J. clin. Inve.~t., 50: 9-20. Dickson, R. C. and R. H. Tomlinson (1967) Selenium in blood and human tissues, Clin. chim. Acta, 16: 311-321. Henke, G., H. Mollmann and H. Alfes (1971) Vergleichende Untersuchungen fiber die Konzentration einiger Spurenelemente in menschlichen Hirnarealen durch Neutronaktivierungsanalyse. Z. Neurol., 199: 283-294. Heydorn, K. and E. Damsgaard (1973) Simultaneous determination of arsenic, manganese and selenium in biological materials by neutron activation analysis, Talanta, 20:1-11. Heydorn, K., E. Damsgaard, N. Horn, M. Mikkelsen, I. Tygstrup, S. Vestermark and J. Weber (1975) Extrahepatic storage of copper - - A male foetus suspected of Menkes' disease, Humangenetik, Vol. 29, pp. 171-175. H6ck, A., U. Demmel, H. Schicha, K. Kasparek and L. E. Feinendegen (1975) Trace element concentration in human brain - - Activation analysis of cobalt, iron, rubidium, selenium, zinc, chromium, silver, cesium, antimony and scandium, Brain, 98 : 49-64. Horn, N. (1976) Copper incorporation studies on cultured cells for prenatal diagnosis of Menkes' disease, Lancet, 1 : 1156-1158. Larsen, N. A., B. Nielsen, H. Pakkenberg, P. Christoffersen, E. Damsgaard and K. Heydorn (1972) Neutron activation analysis of arsenic, manganese and selenium concentrations in organs of uraemic and normal persons. In: Nuclear Activation Techniques in the Life Sciences, IAEA, Vienna, pp. 561-568. Lechner, H., W. Beyer, O. Wawschinek, H. Wielinger and H. H. Tagger (1966) Quantitative Untersuchungen fiber die Verteilung von Mangan im menschlichen Gehirn, Wien. klin. Wschr., 78: 328-329. Lehmann, B. H., J. D. L. Hansen and P. J. Warren (1971) The distribution of copl~er, zinc and manganese in various regions of the brain and in other tissues of children with protein-calorie malnutrition, Brit. J. Nutr., 26: 197-202. Leu, M. L., G. T. Strickland and S. J. Yeh (1971) Tissue copper, zinc and manganese levels in Wilson's disease - - Studies with the use of neutron activation analysis, J. Lab. clin. Med., 77:438 44~. Nielsen, F. H., S. H. Givand and D. R. Myron (1975) Evidence of a possible requirement for arsenic by the rat, Fed. Proc., 34 (No. 3987): 923. Pakkenberg, H. and J. Voigt (1964) Brain weight of the Danes - - A forensic material, Acta anat. (Basel), 56: 297-307. Schicha, H., W. Mtiller, K. Kasparek and R. Schr~Sder (1974) Neutronaktivierungsanalytische Bestimmung der Spurenelemente Kobalt, Eisen, Rubidium, Selen, Zink, Chrom, Silber, Caesium, Antimon und Scandium in operativ entnommen Hirntumoren des Menschen (1. Mitteilung), Beitr. Path. Anat., 151: 281-296. Schroeder, H. A. and J. J. Balassa (1966) Abnormal trace metals in man - - Arsenic, J. chron. Dis., 19: 85-106. Schroeder, H. A., J. J. Balassa and I. H. Tipton (1966) Essential trace metals in man - - Manganese. A study in homeostasis, J. chron. Dis., 19: 545-571. Schroeder, H. A., D. V. Frost and J. J. Balassa (1970) Essential trace metals in man - - Selenium, J. chron. Dis., 23 : 227-243.
416 Smith, H. (1967) The distribution of antimony, arsenic, copper and zinc in human tissue, J. forens. Sci. Soc., 7: 97-102. Tingey, A. H. (1937) The iron, copper and manganese content of the human brain, J. ment. Sci., 83 : 452-460. Tipton, I. H., H. A. Schroeder, H. M. Perry, Jr. and M. J. Cook (1965) Trace elements in human tissue, Part 3 (Subjects from Africa, the Near and Far East and Europe), Health Physics, 11 : 403-451. Tower, D. B. (1969) Inorganic constituents. In A. Lajtha (Ed.), Handbook ofNeurochemistry, Vol. 1, Plenum Press, New York, pp. 1-19. Trapp, G. A. and J. Millam (1975) The distribution of 7~Se in brains of selenium-deficient rats, J. Neurochem., 24: 593-596. Wong, P. Y. and K. Fritze (1969) Determination by neutron activation of copper, manganese and zinc in the pineal body and other areas of brain tissue, J. Neurochem., 16: 1231-1234. Yase, Y., Y. Shinjo, F. Yoshimasu and T. Kumamoto (1973) Neutron activation analysis studies in degenerative diseases of the nervous system, with special reference to divalent cation metals. In: Proc. 2nd Internat. Congress of Muscle Diseases, Perth, 1971 (International Congress Series, No. 295), Excerpta Medica, Amsterdam, pp. 226-234.
