©Copyright 1991 by The Humana Press, Inc. All rights of any nature, whatsoever, reserved. 1044-7393/91/1403-0213 $0280
Alzheimer s and Parkinson s Disease S
Brain Levels of Glutathione, Glutathione Disulfide, and Vitamin E JvEs D. ADAMS, JR.,* LORI K. K1.AIDMAN, IFEOMA N. ODUNZE, HOWARD C. SHEN, AND CAROL A MILLER School of Pharmacy, and Department of Pathology. School of Medicine, University of Southern California, and Los Angeles County/University of Southern California Medical Center, Los Angeles, CA 90033 USA Received August 21, 1990; Accepted February 11, 1991 Author to whom all correspondence and reprint requests should be addressed.
Human brain levels of glutathione (GSH), glutathione disulfide (GSSG), and vitamin E were measured in neurologically normal control patients and two groups of patients with neurodegeneration: those with Alzhejmer's disease (AD), and AD with some features of Parkinson's disease (AD-PD). Control brain samples contained GSH levels more than 50 times higher than GSSG. The levels of GSH were highest in the caudate nucleus and lowest in the medulla. In patients with AD or AD-PD, hippocampal levels of GSH were significantly higher than controls. Patients with AD also demonstrated high GSH levels in the midbrain compared to normal. In contrast, patients with AD-PD did not have significantly elevated GSH levels in this site. GSSG levels were not significantly different in any brain region between controls and diseased patients. In control brains, the medulla had higher levels of vitamin E than any other brain region. The caudate nucleus had the lowest levels, which were about half the levels in the medulla. Control levels of vitamin E in the midbrain were about 18.8 lig!g. In AD patients the midbrain levels of vitamin E doubled to 42.3 pg/g. This doubling also occurred in AD-PD patients where midbrain vitamin E levels increased to 44.0 pg/g. These results Molecula, and Chemical Neuropathology
213
Vol. 14, 1991
214
Adams et a!. may indicate that compensatory increases in GSH and vitamin E levels occur following damage to specific brain regions in patients with AD or AD-PD. Index Entries: Alzheimer's disease; Parkinson's disease; human brain; glutathione; glutathione disulfide; vitamin E.
INTRODUCTION A significant clinical and neuropathological overlap between AD and PD has emerged in the recent literature (Chin et al., 1986; Boiler et aL, 1980; Ditter and Mirra, 1987; Growdon and Corkin, 1986). These observations suggest similar and yet undetermined etiological factors may exist in AD and PD. Both AD and PD have clinical and neuropathological similarities that will be discussed later. It has been suggested that oxidative insults to the brain may play a pathogenetic role in PD and AD (Martins et al., 1986). AD brain tissues have been found to have high oxygen uptake, high glucose metabolism, and possibly uncoupled mitochondria (Sims et al., 1983; Sims et al., 1987). Enhanced oxidative metabolism and mitochondrial uncoupling could lead to oxygen radical formation. PD is associated with elevated brain levels of iron in the substantia nigra, which may accelerate the production of oxygen radicals and lead to lipid peroxidation (Dexter et al., 1987). Some AD patients have decreased plasma levels of vitamin E, which might be a consequence of altered vitamin E uptake or distribution into the brain (Jackson et al., 1988). In addition, a recent clinical trial found vitamin E useful in the treatment of PD (Factor et aL, 1990). Three reports found GSH levels in CNS tissues from PD patients were dramatically decreased in the substantia nigra (Perry et al., 1982; Perry and Yong, 1986; Riederer et al., 1989). Depletion of GSH, by an unknown mechanism, could make the substantia nigra more susceptible to oxidative challenges from toxins or endogenous oxygen radicals. GSH, a critical antioxidant, can detoxify toxins by conjugation and oxygen radicals by reduction as a cosubstrate for GSH peroxidase. To date, human brain GSH levels have not been reported in AD. The current study is not intended to be a reexamination of GSH levels in PD. In this study, GSH and GSSG levels are examined in brains of control and AD patients as well as patients with AD-PD. Vitamin E is a lipophilic vitamin that is found in all cellular membranes. The main purpose of vitamin E is to protect membranes from peroxidation as induced by oxygen radicals (Fariss et al., 1985). Vitamin E is transported in the blood in association with chylomicrons and lipoproteins just as are fats (Gallo-Torres, 1985). It has been postulated that low density lipoprotein transports vitamin E to a specific receptor on the capillary endothelium, which allows the uptake of the vitamin into the brain (Brown and Goldstein, 1986). in addition, a lipoprotein lipase is Molecular and Chemical Neuropathology
Vol. 14, 1991
GSH, GSSG, and Vitamin E in Brain Disease
215
relevant for the transfer of vitamin E from chylomicrons to cells (Traber et a!., 1985). GSH and vitamin E are related in their antioxidant capacities, in that both are critical for the protection of cells from oxidative stress. In membranes, vitamin P prevents lipid peroxidation during oxidative stress by donating a hydrogen and forming a stable radical, c-tocopheryl radical. In the cytosol, GSH detoxifies oxygen free radicals as a cosubstrate for GSH peroxidase. However, these antioxidants are also related in that GSH is involved in replenishing vitamin E from a-tocopheryl radical (McCay et a!., 1989).
Human brain tissues were obtained from the Alzheimer's Disease Research Center Consortium at the University of Southern California and the Division of Neurological Surgery, University of California, Irvine. Patients were clinically confirmed with a diagnosis of AD based on the NINCDS-ADRDA criteria (McKhann et a!., 1984). Neuropathological confirmation of AD was performed according to a modification of the diagnostic protocol of Khachaturian et al. (1985), as established by the Consortium to Establish a Registry in Alzheimer's Disease (CERAD). This protocol is based on age-related, neocortical plaque scores combined with a clinical history of dementia. Confirmation of PD was based on the presence of Lewy bodies in possible catecholaminergic neurons and on nerve cell loss in the zona compacta of the substantia nigra. In all cases the presence of PD or AD was defined by histopathological evaluation of brain material. Patients with both PD and AD comprised two groups: two patients who were initially diagnosed with PD and later developed dementia, and eight others who were diagnosed first with AD and later developed PD. These patients were grouped together for analysis of GSH and GSSG and are referred to as AD-PD. Brains from nine patients with AD, ten with AD-PD, and nine with no neurological disease were dissected at autopsy and cut into 3-cm blocks from specific brain regions. These regions were frontal cortex (Brodman's area 9), hippocampus, putamen, caudate nucleus, midbrain, and medulla. Tissue samples were flash frozen in liquid nitrogen chilled isopentane and stored at - 70°C. GSH and GSSG levels were assayed in brain samples according to the method of Adams et a!. (1983). Samples were cut into 50-mg pieces while on dry ice. The weighed pieces were immediately homogenized in solutions of bis(3-carboxy-4-nitrophenyl)disulfide (DTNB) or N-ethylmaleimide. This step forms GSH derivatives that cannot oxidize to form GSSG artefactually. The homogenates were centrifuged for 5 min at 15,000g to sediment opaque material. Samples for GSSG analysis were eluted through small columns of reverse phase HPLC column material to Molecular and Chemical Neuropathology
Vol. 14, 1991
216
Adams et al.
remove excess N-ethylmaleimide. The homogenates were then analyzed spectrophotometrically in the presence of GSSG reductase, NADPFL and DTNB for GSH or GSSG. For vitamin E (a-tocopherol) analysis, brain samples were homogenized in 09% saline. Ethanol containing ascorbic acid was added to stabilize vitamin E, according to the method of Vatassery and Hagen (1977). In addition, the internal standard, i-tocopherol, was added in the ethanollascorbate (Fariss et al., 1985). Lipids were extracted from the homogenates into hexane. Samples were reconstituted in methanol and analyzed by HPLC with fluorescence detection (Fariss et al., 1985; Vatassery and Hagen, 1977). The accuracy, reproducibility, and recovery of vitamin E with this procedure have been reported in rat brain (Vatassery and Hagen, 1977).
RESULTS Histopathological findings are presented for all patients in Table 1. Control patients had no changes in the midbrain indicative of PD. Some of the older control patients had mild AD changes that were not consistent with a diagnosis of AD given the age of the patient. In other words, some neuritic plaques are expected in control patients of sufficient age. AD-PD patients all had histopathological changes consistent with PD and AD. AD patients all had severe neuropathological changes consistent with AD. Three AD patients had mild degeneration of the midbrain as denoted by the presence of extracellular pigment, but did not exhibit the formation of Lewy bodies. These AD patients were, therefore, not found to have PD. Drug therapy, especially levodopa therapy, has been suggested to have an effect on the induction of oxidative stress in the brain. Dementia in AD patients was treated with haloperidol in seven of nine patients. Of the seven patients given haloperidol, six developed movement problems (slowed gait), which may have been due to transient extrapyramidal side effects. None of the AD patients were given levodopa. Most of the AD-PD patients were treated as AD patients and, in five cases, were given haloperidol. Only three of the AD-PD patients were treated with levodopa, which might stimulate dopamine turnover in neurons, increase oxygen radical formation, and lead to GSH oxidation in the brain. Patients with AD-PD differed clinically from those with AD alone, in that the latter group survived somewhat longer (about 1 y), after diagnosis with AD, than did the patients with AD-PD. Control patients succumbed to cancer or cardiovascular disease, whereas AD and AD-PD patients had a number of causes of death including pneumonia, cardiovascular disease, cancer, and gangrene. None of them had a history of dementia or movement disorders. In addition, none were given levodopa or antipsychotic drugs (such as haloperidol). Molecular and Chemical Neuropathology
Vol. 14. 1991
GSH, GSSG, and Vitamin E in Brain Disease
217
Table 1 Neuropathological Findings in All Patients' Age (y)
Illness duration (y)
Midbrain changes
C
73.2 ± 4.3
0
0
AD-PD
81.7 ± 2.0
1-7 AD 0-14 PD
AD
81.9 ± 2.1
1-9
AD changes Hippocampus
Cortex
none/8 mild/i
none!7 mild/2
10
none/1 mild/2 modera te/3 severe/4
none/1 mild/1 moderate/3 severe/5
3
moderate/4 severe/5
severe/9
'Patients are: C, control (ii = 9); AD, Alzheimer's disease patients (n = 9); AD-PD, patients with both Parkinson's disease and AD (11 = 10). Midbrain changes in AD-PD patients were neuron loss and Lewy body formation, and in AD patients were the presence of extraneuronal pigment and are expressed as the number of patients exhibiting changes. AD changes were the presence of neuritic plaques as discussed in the text, and are expressed as a relative degree of involvement per number of patients. Age is not different between controls and AD-PD or AD patients.
The data in Table 2 show that control levels of GSH were 50-90 times higher than levels of GSSG. The levels of GSH were significantly lowest in the medulla and significantly highest in the caudate nucleus (p < 0.05, ANOVA with Newman-Keuls test). There were no differences among control GSSG levels in the various brain regions. All groups of diseased patients exhibited enhanced levels of GSH in the hippocampus (Table 2). Patients with AD had elevated GSH levels in the midbrain, whereas in patients with AD-PD, midbrain GSH levels were comparable to controls. GSSG levels did not differ in any brain region among controls and diseased patients. Brain vitamin E levels are presented in Table 3. In control brain, the medulla contained significantly higher levels of vitamin E than the other regions examined. The caudate nucleus had the lowest vitamin E levels, which were half the levels in the medulla. Vitamin E levels in PD and AD changed only in the midbrain and both groups of patients (AD and AD-PD) had the same change, which was a doubling of vitamin F levels. There were no significant differences between patients with AD and patients with AD-PD in terms of levels of vitamin F in any brain region examined.
DISCUSSION Control GSH levels reported here are similar to the human brain GSH levels reported by Slivka et al. (1987b), and higher than those reported by Perry et al. (1982) or Riederer et al. (1989). Control GSSG levels are similar to those of Slivka et a!. (1987b) and much lower than Molecular and Chemical Neuropathology
Vol. 14, 1991
*
~ ....
,-I:..
-
z;: ,...
~
I
~
"if
~ n' !:L
£2
2.
~
s,
;;; ,.,
5.4 ± 1.9
3.1 ± 0.4
AD-
AD
1.13 ± 0.07* 18 ± 4 (8)
0.95 ± 0.06 26 ± 5 (10)
0.81 ± 0.09 16 ± 5 (6)
Midbrain
1.35 ± 0.07 15 ± 4 (9)
0.72 ± 0.06 13 ± 4 (9)
0.73 ± 0.10 23 ± 4 (10)
(7)
(7)
1.17 ± 0.09 25 ± 3 (10)
0.63 ± 0.10 10 ± 4
Medulla
1.15 ± 0.11 13 ± 3
Frontal cortex
1.50 ± 0.10 33 ± 5 (9)
1.43 ± 0.13 41 ± 6 (9)
1.20 ± 0.14 24 ± 7 (7)
Putamen
1.88 ± 0.11 32 ± 7 (9)
1.56 ± 0.19 36 ± 5 (9)
17 ± 3 (6)
1.25 ± 0.18
Caudate nucleus
1.51 ± 0.09* 33 ± 8 (9)
1.39 ± 0.09* 33 ± 9 (7)
(6)
0.96 ± 0.04 13 ± 4
Hippocampus
"Patients are: C, control (n = 9); AD, Alzheimer's disease (n = 9); and AD-PD, Parkinson's disease and AD (n = 10). Top figure for each group is glutathione (IJ-mol/g tissue). Lower figure is glutathione disulfide (nmollg tissue). Means ± S.E.M. Figures in brackets are the numbers of patients examined. *p < 0.05, as compared to controls (ANOVA with Newman-Keuls test). The postmortem interval is not significantly different between controls and AD-PD or AD patients. The n values do not always match the number of patient tissues examined because tissues from each region were not always available for each patient.
PO
7.4 ± 1.5
C
(h)
Post mortem interval
Table 2 Glutathione and Glutathione Disulfide Levels in Human Brain Samples"
~
I1l .....
~
):.
~
ex>
N ......
~
~
-....
........
~
~
~
~
.g
E;
~ ~
9 ~.
~
2..,is>
~
(9)
36.9 ± 3.6
(10)
34.0 ± 4.6
(7)
34.6 ± 4.4
Medulla
(9)
23.6 ± 1.9
(9)
23.5 ± 2.6
(7)
21.7 ± 2.8
Putamen
(9)
21.8 ± 2.7
(9)
15.1 ± 1.7
(6)
17.3 ± 2.4
Caudate nucleus
(9)
24.3 ±2.1
(7)
27.1 ±1.1
Hippocampa 19.9 ±2.2 (6)
"Patients are: C, control (11 "" 9);AD, Alzheimer's disease (11 "" 9); and AD-PD, Parkinson's disease and AD (11 = 10). Top figure for each group is vitamin E (J.Lg/g tissue). Lower figure (in brackets) is number of patients for each region analyzed. Means ± S.E.M. "p < 0.05, as compared to controls (ANOVA with Newman-Keuls test). The 11values do not always match the number of patient tissues examined because tissues from each region were not always available for each patient.
(8)
(9)
24.9 ± 2.8
42.3 ± 3.0*
(10)
AD
(10)
23.0 ± 3.3
(7)
44.0 ± 7.0*
(6)
24.5 ± 4.4
18.8 ± 3.3
AD-PD
c
Frontal cortex
Midbrain
Table 3 Vitamin E Levels in Human Brain Samples"
\0
tv -'
~
0;'
Ql S· b
t1:l
s·
rn
~s·
$:
~
Vi f)
a ~ @
220
Adams et aL
those of Perry et a!. (1982). The analysis of brain GSH and GSSG by Slivka et al. was done partly in humans and in anesthetized animals, to inhibit agonal responses. Slivka et a!. also flash froze all brain tissues to liquid nitrogen temperatures to inhibit postmortem oxidation of GSH. Slivka et a!. found that GSH and GSSG brain levels are about the same in animals and humans, 1-3 pmol/g and about 10 nmol/g, respectively, which is identical to the current report. In addition, the human levels reported here are very similar to previous animal work from this laboratory (Adams et al., 1989; Odunze et al., 1990a). However, Perry et a!. (1982) and Riederer et a!. (1989) did not prevent the artefactual oxidation of GSH in their samples (Adams et a!., 1983). Thus, their GSH levels may be too low and GSSG levels too high. This is a critical issue when examining brains from PD patients, since their midbrains have elevated levels of iron (Riederer et al., 1989) that might greatly accelerate the artefactual oxidation of GSH. Brain vitamin E levels reported here are very similar to previously reported human brain and liver levels but lower than levels in adipose tissue (Gallo-Torres, 1985; Metcalfe et al., 1989). In addition, mouse brain vitamin E levels reported by this laboratory (Odunze et al., 1990b) are substantially lower than human brain levels. The mouse brain analysis was done with flash frozen extracts of fresh brain tissue to inhibit postmortem changes in vitamin E. Our previously reported mouse brain vitamin E levels are similar to reported rat brain vitamin E levels, where rats were killed under anesthesia to inhibit agonal changes (Vatassery and Hagen, 1977; Vatassery et al., 1984a). It has been reported that the rat medulla accumulates vitamin E with age such that the medulla contains high levels of vitamin E when rats are very old (Vatassery et at., 1984a). This may be true in humans also, in that the medulla contains higher levels of vitamin F than other brain regions in the 70- to 80-y-old patients studied. The postmortem interval, which is the time elapsed after death until the samples were frozen in liquid nitrogen, was not significantly different between controls and neurologically diseased patients (Table 2). It is important to keep this interval as short as possible to minimize artefactual oxidation of GSH. The sampling interval here was shorter than that of Riederer et at. (1989), which was 12-14 h and about the same as Perry et al. (1982), which was 4 h. Our results suggest that damage to the brain occurs in AD and ADPD, causing a compensatory increase in brain GSH or vitamin E levels. One possible etiological factor could be a decrease in blood flow through selected areas of the brain (Hoyer, 1986) with concomitant hypoxia. Vascular changes associated with amyloid angiopathy in AD may be of importance. GSH is required to detoxify the oxygen radicals formed during hypoxic stress, which results in the oxidation of GSH by GSH peroxidase. Dopaminergic neurons of the midbrain are particularly sensi-
Molecular and Chemical Neuropathology
Vol. 14, 7991
GSH, GSSG, and Vitamin E in Brain Disease
221
tive to hypoxia (Phebus et a!., 1986). Other, nonoxidative mechanisms might also explain the increases in brain GSH, including induction of GSH synthetase activity. Increases in brain GSH are common during brain oxidative stress as induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (Adams et al., 1989) and 6-hydroxydopamine (Perumal et al., 1989). This may be indicative of the ability of the brain to defend itself from oxidative stress. The midbrain has a unique ability to concentrate vitamin E during these disease processes. It is not clear why increased vitamin E levels were not seen in the cerebral cortex or hippocampus in AD, since these regions suffer large amounts of damage in the disease process, especially given the increase in GSH levels in AD and AD-PD patients in the hippocampus. These findings suggest biochemical differences between vitamin E and GSH in terms of responses to damage in the hippocampus and midbrain. 1-Methyl-4-phenyl-1, 2,3,6-tetrahydropyridine induced oxidative stress in the brain (Adams et al., 1989) is associated with increases in vitamin E in the midbrain and other regions (Odunze et at., 1990b). The mechanism by which vitamin E levels might increase in the brain is not known. The brain does not possess a cytosolic binding protein for vitamin E such as is important in the transport of vitamin E in the liver (Murphy and Mavis, 1981a). The peripheral nervous system and other organs may take up vitamin E according to the amount of lipid present in the tissue (Clement and Bourre, 1990). However, the brain affinity for vitamin E may not correlate with vitamin E levels or total fatty acid content (Murphy and Mavis, 1981b). The brain is very different from other organs in that vitamin E does not readily penetrate into the brain in large amounts (Vatassery et al., 1984b), nor does vitamin E readily leave the brain (Goss-Sampson et a!., 1988). The midbrain does not appear to have a significantly greater intrinsic uptake of vitamin E than other regions (Goss-Sampson and Muller, 1987). Lipid levels are known to decrease in the brain with age (Soderberg et a!., 1990) and more specifically decrease in the midbrain in PD (Dexter et al., 1989). Using the lipid levels reported by Soderberg et al. (1990), control hippocampal vitamin E levels appear to be about 765 gig lipid. Drug therapy was probably not involved in alteration of antioxidant levels. Haloperidol was given to most of the AD and AD-PD patients. However, GSH levels in the two patient groups were not affected identically in the midbrain. Haloperidol does not produce permanent parkirtsonian side effects, nor does it induce Lewy body formation. Therefore, AD patients treated with haloperidol do not automatically develop PD. In addition, levodopa was clearly not a factor in this study since only three of the AD-PD patients received the drug. Levodopa therapy may be a confounding influence in published accounts of GSH levels in PD, since patients with PD alone are routinely treated with levodopa.
Molecular and Chemical Neuropathology
Vol 14, 1991
222
Adams et aL
It could be that increased brain GSH levels are a result of reactive gliosis or nerve terminal proliferation in response to neuronal death. Glial cells and nerve terminals contain higher levels of GSH than neuronal somas, which are destroyed both in PD and AD (Slivka et al., 1987a). Therefore, destruction of dopaminergic neurons, which is followed by gliosis, in the midbrain could actually result in an increase in GSH levels rather than a decrease. However, if reactive gliosis were an explanation for increased GSH levels in all brain regions, then GSH levels should also be increased in the frontal cortex, a major site of damage in AD. Since the cortex exhibited no increase in GSH levels, these results suggest the hippocampus and midbrain respond in a different manner or are more susceptible than the cortex and other sites to damage that alters GSH levels. The overlapping clinical and pathological profiles of AD and PD suggest similarities in the pathogenesis of the diseases. Boiler et al. (1980) reviewing 36 PD patients, demonstrated 55% with clinical dementia, and 42% of these with the neuropathological changes of AD, such as neuritic plaques and neurofibrillary tangles. Over 20% of the patients had these histological changes in the absence of dementia. The prevalence of pathologically established AD changes plus dementia was 33%. Loss of neurons in the nucleus basalis of Meynert and the locus ceruleus, the principal cholinergic and noradrenergic nuclei projecting to the cortex, respectively, have been observed in both PD and AD. Quantimetric studies have also shown a decreased number of neurons in the zona compacta of the substantia nigra of AD patients (Chui et al., 1986). Neurons are lost in the midbrain even in AD patients with no obvious Lewy bodies, in other words, patients that have AD and not AD-PD (Bergeron and Pollanen, 1989; Chui et al., 1986). In addition, neurofibrillary tangles are present in dopaminergic neurons of the substantia nigra in AD (Nakashima and Ikuta, 1985). However, the pathology of PD and AD may differ in the oculomotor neurons of the rostral midbrain where PD produces a loss of neurons and Lewy body formation, not seen in AD (Hunter, 1985). Hunter did find neurofibrillary tangles and gliosis in the rostral midbrain in AD. The Lewy body, intracytoplasmic inclusions in neurons of the substantia nigra, locus ceruleus, and other catecholaminergic neurons found in PD, have also been noted in AD in small and medium cortical neurons, particularly in layers 5 and 6 (Kosaka et al., 1976). They are especially numerous in cortical Lewy body disease, a separate neuropathological entity (Kosaka, 1978). Parkinsonian changes in AD patients are even more prevalent than the inverse. Ditter and Mirra (1987) demonstrated histological change of PD, including Lewy bodies, neuronal loss, and gliosis of pigmented nuclei in 11 of 20 AD patients. Clinically, rigidity was noted in 80% of these patients with parkinsonian pathology, but far fewer (14%) of the patients with no pathological changes of PD. No tremor was observed in either patient group. /sloieajlar and Chemical Neuropathology
Vol. 14, 1991
GSH, GSSG, and Vitamin E in Brain Disease
223
It has been previously suggested that the midbrain might be very susceptible to oxidative challenges that might alter GSH levels (Dexter et al., 1987, Kish et al., 1985). This might be due to the low levels of protective substances such as GSH and GSH peroxidase, and the high levels of free radical generating neuromelanin in dopaminergic neurons (Kish et a!., 1985, Slivka et al., 1987a). It may now be apparent that the midbrain, unlike other brain regions, also has the ability to concentrate vitamin E in its membranes in response to disease processes. This may mean that raising vitamin E levels in plasma by the administration of the vitamin to patients could provide for even higher vitamin E concentrations in the midbrain. This may provide a mechanism for the protective effects of vitamin E in PD (Factor et al., 1990) and may suggest that vitamin F therapy could be of benefit in AD and AD-PD.
ACKNOWLEDGMENTS This work was supported by NINCDS grant no. NS23515 (JA), NIA grant no. AG05142 (CAM), NIMH grant no. MH39145 (CAM), and a grant from the Alzheimer's Disease Association (CAM). The authors are grateful to Dr. James W. Geddes of the Division of Neurological Surgery at the University of California, Irvine College of Medicine for supplying some of the brain material.
REFERENCES Adams J. D., Lauterburg B. H., and Mitchellj. R. (1983) Plasma glutathione and glutathione disulfide in the rats Regulation and response to oxidative stress. J. Pharmacol, Exp. Ther. 227, 749-754. Adams J. D., Klaidman L. K., and Odunze 1. N. (1989) Oxidative effects of MPTP in the midbrain. Res. Commun. Subst. Abuse 10, 169-180. Bergeron C. and Pollanen M. (1989) Lewy bodies in Alzheimer disease—one or two diseases? Alzheimer Dis. Assoc. Disorders 3, 197-204. Boiler F., Mizutani T., Roessmann U., and Gambetti, P. (1980) Parkinson disease, dementia and Alzheimer disease: Clinicopathological correlations. Ann. Neurol. 7, 329-335. Brown M. S. and Goldstein J. L. (1986) A receptor-mediated pathway for cholesterol homeostasis. Science 232, 34-47, Chui H. C., Mortimer J. A., Slager U., Zarow C., Bondareff W., and Webster D. D. (1986) Pathologic correlates of dementia in Parkinson's disease. Arch. Neurol. 43, 991-995. Clement M., and BourreJ. M. (1990) Alteration of a-tocopherol content in the developing and aging peripheral nervous system: Persistence of high correlations with total and specific (n-6) polyunsaturated fatty acids. J. Neurochem. 54, 2110-2117. Molecular and Chemical Neuropathology
VoL 14. 1991
224
Adams et aL
Dexter D. T., Carter C. J., Wells F. R., Javoy-Agid F., Agid Y., Lees A., Jenner P., and Marsden C. D. (1989) Basal lipid peroxidation in substantia nigra is increased in Parkinson's disease. J. Neurochem. 52, 381-389. Dexter D. T., Wells F. R., Agid F., Agid Y., Lees A. J., Jenner P., and Marsden C. D. (1987) Increased nigral iron content in postmortem parkinsonian brain. Lancet 2, 1219-1220. Ditter S. M. and Mirra S. S. (1987) Neuropathologic and clinical features of Parkinson's disease in Alzheimer's disease patients. Neurology 37, 754-760. Factor S. A., Sanchez-Ramos J. R., and Weiner W. J. (1990) Vitamin E therapy in Parkinson's disease. Adv. Neural. 53, 457-461. Fariss M. W., Pascoe C. A., and Reed D. J. (1985) Vitamin E reversal of the effect of extracellular calcium on chemically induced toxicity in hepatocytes. Science 227, 751-754. Gallo-Torres H. E. (1985) Transport and metabolism, in Fat Soluble Vitamins: Their Biochemistry and Applications (Diplock A., ed.), pp 193-267, Technomic, Lancaster, Great Britain. Goss-Sampson M. A., MacEvilly C. J., and Muller D. P. R. (1988) Longitudinal studies of the neurobiology of vitamin E and other antioxidant systems, and neurological function in the vitamin E deficient rat. I. Neural. Sci. 87, 25-35. Goss-Sampson M. A. and Muller D. P. R. (1987) Studies on the neurobiology of vitamin E and some other antioxidant systems in the rat. Neuroputh. Appi. Neurobiol. 13, 289-296. Growdon J. H. and Corkin S. (1986) Cognitive impairments in Parkinson's disease. Adv. Neural. 45, 383-392. Hoyer S. (1986) Senile dementia and Alzheimer's disease, brain blood flow and metabolism. Prog. Neuro-Psychopharmacol. Bid. Psychiat. 10, 447-478. Hunter S. (1985) The rostra] mesencephalon in Parkinson's disease and Alzheimer's disease. Acta Neuropathol. 68, 53-58. Jackson C. V. E., Holland A. J., Williams C. A., and Dickerson J. W. T. (1988) Vitamin E and Alzheimer's disease in subjects with Down's syndrome. I. Mental Defic. Res. 32, 479-484. Khachaturian Z. S. (1985) Diagnosis of Alzheimer's disease. Arch. Neural. 42, 1097-1105. Kish S. J., Morito C., and Hornykiewicz, 0. (1985) Glutathione peroxidase activity in Parkinson's disease brain. Neurosci. Lett. 58, 343-346. Kosaka K. (1978) Lewy bodies in cerebral cortex, report of three cases. Acta Neuro path. 42, 127-134, Kosaka K., Oyanagi S., Matsushita M., Hon A., and Iwase S. (1976) Presenile dementia with Alzheimer-, Pick- and Lewy-body changes. Acta Neuropath. 36, 221-233. Martins R. N., Harper C. G., Stokes G. 13., and Masters C. L. (1986) Increased cerebral glucose-6-phosphate dehydrogenase activity in Alzheimer's disease may reflect oxidative stress. J. Neurochem. 46, 1042-1045. McCay P. B., Brueggemann C., Lai E. K., and Powell S. R. (1989) Evidence that a-tocopherol functions cyclically to quench free radicals in hepatic microsomes. Ann. New York Acad. Sci. 570, 32-45. McKhann C., Drachrnan D., Foistein M., Katzman R., Price D., and Stadlan E. M. (1984) Clinical diagnosis of Alzheimer's disease: Report of the NINCDSADRDA work group under the auspices of the Department of Health and Human Services task force of Alzheimer's disease. Neurology 34, 939-944. Molecular and Chemical Neuropathology
Vol. 74, 1991
GSH, GSSG, and Vitamin E in Brain Disease
225
Metcalfe T., Bowen D. M., and Muller D. P. R. (1989) Vitamin E concentrations in human brain of patients with Alzheimer's disease, fetuses with Down's syndrome, centenarians, and controls. Neurochem. Res. 14, 1209-1212. Murphy D. J. and Mavis R. D. (1981a) A comparison of the in vitro binding of c-tocopherol to microsomes of lung, liver, heart and brain of the rat. Biochim. Biophys. Ada 663, 390-400. Murphy D. J. and Mavis R. D. (1981b) Membrane transfer of a-tocopherol. J. Biol. Chem. 256, 10464-10468. Nakashima S. and Ikuta F. (1985) Catecholamine neurons with Alzheimer's neurofibriliary changes and alteration of tyrosine hydroxylase. Ada Neuropatiwl. 66, 37-41. Odunze 1. N., Klaidman L. K., and Adams, J. D. (1990a) MPTP toxicity: Differential effects in the striatum, cerebral cortex and midbrain on glutathione, glutathione disulfide and protein sulfhydryl levels. Res. Cornmun. Subst. Abuse. ii, 123-134, Odunze I. N., Klaidman L. K., and Adams, J. D. (1990b) MPTP toxicity in the mouse brain and vitamin E. Neurosci. Lett. 108, 346-349. Perry T. L., Godin D. V., and Hansen S. (1982) Parkinson's disease: A disorder due to nigral glutathione deficiency. Neurosci. Lett. 33, 305-310, Perry T. L. and Yong V. W. (1986) Idiopathic Parkinson's disease, progressive supranuclear palsy and glutathione metabolism in the substantia nigra of patients. Neurosci. Lett. 67, 269-274. Perumal A. S., Tordzro W. K., Katz M., Jackson-Lewis V., Cooper T. B., Fahn S., and Cadet J. L. (1989) Regional effects of 6-hydroxydopamine on free radical scavengers in rat brain. Brain Res. 504, 139-141. Phebus L. A., Perry K. W., Clemens J. A., and Fuller R. W. (1986) Brain anoxia releases striatal dopamine in rats. Life Sci. 38, 2447-2453. Riederer P., Sofic E., Rausch W. D., Schmidt B., Reynolds G. P., Jellinger K., and Youdim M. B. H. (1989) Transition metals, ferritin, glutathione, and ascorbic acid in parkinsonian brains. 1. Neurochem. 52, 515-520. Sims N. R., Bowen D. M., Neary D., and Davison A. N. (1983) Metabolic processes in Alzheimer's Disease: Adenine nucleotide content and production of 14 CO 2 from [U- 14 C]gluc6se in vitro in human neocortex. I. Neurochem. 41, 1329-1334, Sims N. R., Finegan J. M., Blass J. P., Bowen D. M., and Neary D. (1987) Mitochondrial function in brain tissue in primary degenerative dementia. Brain Res. 436, 30-38. Slivka A., Mytilineou C., and Cohen G. (1987a) Histochemical evaluation of glutathione in brain. Brain Res. 409, 275-284. Slivka A., Spina M. B., and Cohen C. (1987b) Reduced and oxidized glutathione in human and monkey brain. Neurosci. Lett. 74, 112-118. Soderberg M., Edlund C., Kristensson K., and Daliner C. (1990) Lipid compositions of different regions of the human brain during aging. J. Neurochem. 54, 415-423. Traber M. C., Olivecrona T., and Kayden H. J. (1985) Bovine milk lipoprotein lipase transfers tocopherol to human fibroblasts during triglyceride hydrolysis in vitro. J. Clin. Invest. 75, 1729-1734. Vatassery G. 1., Angerhofer C. K., and Knox, C. A. (1984a) Effect of age on vitamin E concentrations in various regions of the brain and a few selected Molecular and Chemical Neuropathology
Vol. 14. 1991
Adams et al.
226
peripheral tissues of the rat, and on the uptake of radioactive vitamin E by various regions Of rat brain. J. Neurochern. 43, 409-412. Vatassery C. T., Angerhofer C. K., Knox C. A., and Desmukh D. S. (1984b) Concentrations of vitamin E in various neuroanatomical regions and subcellular fractions, and the uptake of vitamin E by specific areas, of rat brain.
Biochim. Biophys. Acta 792, 118-422.
Vatassery C. T. and Hagen D. F. (1977) A liquid chromatographic method for quantitative determination of u-tocopherol in rat brain. Anal. Biochern. 79, 129-434.
Molecular and Chemical Neuropathology
VoL 14, 1991
Alzheimer s and Parkinson s Disease S
Brain Levels of Glutathione, Glutathione Disulfide, and Vitamin E JvEs D. ADAMS, JR.,* LORI K. K1.AIDMAN, IFEOMA N. ODUNZE, HOWARD C. SHEN, AND CAROL A MILLER School of Pharmacy, and Department of Pathology. School of Medicine, University of Southern California, and Los Angeles County/University of Southern California Medical Center, Los Angeles, CA 90033 USA Received August 21, 1990; Accepted February 11, 1991 Author to whom all correspondence and reprint requests should be addressed.
Human brain levels of glutathione (GSH), glutathione disulfide (GSSG), and vitamin E were measured in neurologically normal control patients and two groups of patients with neurodegeneration: those with Alzhejmer's disease (AD), and AD with some features of Parkinson's disease (AD-PD). Control brain samples contained GSH levels more than 50 times higher than GSSG. The levels of GSH were highest in the caudate nucleus and lowest in the medulla. In patients with AD or AD-PD, hippocampal levels of GSH were significantly higher than controls. Patients with AD also demonstrated high GSH levels in the midbrain compared to normal. In contrast, patients with AD-PD did not have significantly elevated GSH levels in this site. GSSG levels were not significantly different in any brain region between controls and diseased patients. In control brains, the medulla had higher levels of vitamin E than any other brain region. The caudate nucleus had the lowest levels, which were about half the levels in the medulla. Control levels of vitamin E in the midbrain were about 18.8 lig!g. In AD patients the midbrain levels of vitamin E doubled to 42.3 pg/g. This doubling also occurred in AD-PD patients where midbrain vitamin E levels increased to 44.0 pg/g. These results Molecula, and Chemical Neuropathology
213
Vol. 14, 1991
214
Adams et a!. may indicate that compensatory increases in GSH and vitamin E levels occur following damage to specific brain regions in patients with AD or AD-PD. Index Entries: Alzheimer's disease; Parkinson's disease; human brain; glutathione; glutathione disulfide; vitamin E.
INTRODUCTION A significant clinical and neuropathological overlap between AD and PD has emerged in the recent literature (Chin et al., 1986; Boiler et aL, 1980; Ditter and Mirra, 1987; Growdon and Corkin, 1986). These observations suggest similar and yet undetermined etiological factors may exist in AD and PD. Both AD and PD have clinical and neuropathological similarities that will be discussed later. It has been suggested that oxidative insults to the brain may play a pathogenetic role in PD and AD (Martins et al., 1986). AD brain tissues have been found to have high oxygen uptake, high glucose metabolism, and possibly uncoupled mitochondria (Sims et al., 1983; Sims et al., 1987). Enhanced oxidative metabolism and mitochondrial uncoupling could lead to oxygen radical formation. PD is associated with elevated brain levels of iron in the substantia nigra, which may accelerate the production of oxygen radicals and lead to lipid peroxidation (Dexter et al., 1987). Some AD patients have decreased plasma levels of vitamin E, which might be a consequence of altered vitamin E uptake or distribution into the brain (Jackson et al., 1988). In addition, a recent clinical trial found vitamin E useful in the treatment of PD (Factor et aL, 1990). Three reports found GSH levels in CNS tissues from PD patients were dramatically decreased in the substantia nigra (Perry et al., 1982; Perry and Yong, 1986; Riederer et al., 1989). Depletion of GSH, by an unknown mechanism, could make the substantia nigra more susceptible to oxidative challenges from toxins or endogenous oxygen radicals. GSH, a critical antioxidant, can detoxify toxins by conjugation and oxygen radicals by reduction as a cosubstrate for GSH peroxidase. To date, human brain GSH levels have not been reported in AD. The current study is not intended to be a reexamination of GSH levels in PD. In this study, GSH and GSSG levels are examined in brains of control and AD patients as well as patients with AD-PD. Vitamin E is a lipophilic vitamin that is found in all cellular membranes. The main purpose of vitamin E is to protect membranes from peroxidation as induced by oxygen radicals (Fariss et al., 1985). Vitamin E is transported in the blood in association with chylomicrons and lipoproteins just as are fats (Gallo-Torres, 1985). It has been postulated that low density lipoprotein transports vitamin E to a specific receptor on the capillary endothelium, which allows the uptake of the vitamin into the brain (Brown and Goldstein, 1986). in addition, a lipoprotein lipase is Molecular and Chemical Neuropathology
Vol. 14, 1991
GSH, GSSG, and Vitamin E in Brain Disease
215
relevant for the transfer of vitamin E from chylomicrons to cells (Traber et a!., 1985). GSH and vitamin E are related in their antioxidant capacities, in that both are critical for the protection of cells from oxidative stress. In membranes, vitamin P prevents lipid peroxidation during oxidative stress by donating a hydrogen and forming a stable radical, c-tocopheryl radical. In the cytosol, GSH detoxifies oxygen free radicals as a cosubstrate for GSH peroxidase. However, these antioxidants are also related in that GSH is involved in replenishing vitamin E from a-tocopheryl radical (McCay et a!., 1989).
Human brain tissues were obtained from the Alzheimer's Disease Research Center Consortium at the University of Southern California and the Division of Neurological Surgery, University of California, Irvine. Patients were clinically confirmed with a diagnosis of AD based on the NINCDS-ADRDA criteria (McKhann et a!., 1984). Neuropathological confirmation of AD was performed according to a modification of the diagnostic protocol of Khachaturian et al. (1985), as established by the Consortium to Establish a Registry in Alzheimer's Disease (CERAD). This protocol is based on age-related, neocortical plaque scores combined with a clinical history of dementia. Confirmation of PD was based on the presence of Lewy bodies in possible catecholaminergic neurons and on nerve cell loss in the zona compacta of the substantia nigra. In all cases the presence of PD or AD was defined by histopathological evaluation of brain material. Patients with both PD and AD comprised two groups: two patients who were initially diagnosed with PD and later developed dementia, and eight others who were diagnosed first with AD and later developed PD. These patients were grouped together for analysis of GSH and GSSG and are referred to as AD-PD. Brains from nine patients with AD, ten with AD-PD, and nine with no neurological disease were dissected at autopsy and cut into 3-cm blocks from specific brain regions. These regions were frontal cortex (Brodman's area 9), hippocampus, putamen, caudate nucleus, midbrain, and medulla. Tissue samples were flash frozen in liquid nitrogen chilled isopentane and stored at - 70°C. GSH and GSSG levels were assayed in brain samples according to the method of Adams et a!. (1983). Samples were cut into 50-mg pieces while on dry ice. The weighed pieces were immediately homogenized in solutions of bis(3-carboxy-4-nitrophenyl)disulfide (DTNB) or N-ethylmaleimide. This step forms GSH derivatives that cannot oxidize to form GSSG artefactually. The homogenates were centrifuged for 5 min at 15,000g to sediment opaque material. Samples for GSSG analysis were eluted through small columns of reverse phase HPLC column material to Molecular and Chemical Neuropathology
Vol. 14, 1991
216
Adams et al.
remove excess N-ethylmaleimide. The homogenates were then analyzed spectrophotometrically in the presence of GSSG reductase, NADPFL and DTNB for GSH or GSSG. For vitamin E (a-tocopherol) analysis, brain samples were homogenized in 09% saline. Ethanol containing ascorbic acid was added to stabilize vitamin E, according to the method of Vatassery and Hagen (1977). In addition, the internal standard, i-tocopherol, was added in the ethanollascorbate (Fariss et al., 1985). Lipids were extracted from the homogenates into hexane. Samples were reconstituted in methanol and analyzed by HPLC with fluorescence detection (Fariss et al., 1985; Vatassery and Hagen, 1977). The accuracy, reproducibility, and recovery of vitamin E with this procedure have been reported in rat brain (Vatassery and Hagen, 1977).
RESULTS Histopathological findings are presented for all patients in Table 1. Control patients had no changes in the midbrain indicative of PD. Some of the older control patients had mild AD changes that were not consistent with a diagnosis of AD given the age of the patient. In other words, some neuritic plaques are expected in control patients of sufficient age. AD-PD patients all had histopathological changes consistent with PD and AD. AD patients all had severe neuropathological changes consistent with AD. Three AD patients had mild degeneration of the midbrain as denoted by the presence of extracellular pigment, but did not exhibit the formation of Lewy bodies. These AD patients were, therefore, not found to have PD. Drug therapy, especially levodopa therapy, has been suggested to have an effect on the induction of oxidative stress in the brain. Dementia in AD patients was treated with haloperidol in seven of nine patients. Of the seven patients given haloperidol, six developed movement problems (slowed gait), which may have been due to transient extrapyramidal side effects. None of the AD patients were given levodopa. Most of the AD-PD patients were treated as AD patients and, in five cases, were given haloperidol. Only three of the AD-PD patients were treated with levodopa, which might stimulate dopamine turnover in neurons, increase oxygen radical formation, and lead to GSH oxidation in the brain. Patients with AD-PD differed clinically from those with AD alone, in that the latter group survived somewhat longer (about 1 y), after diagnosis with AD, than did the patients with AD-PD. Control patients succumbed to cancer or cardiovascular disease, whereas AD and AD-PD patients had a number of causes of death including pneumonia, cardiovascular disease, cancer, and gangrene. None of them had a history of dementia or movement disorders. In addition, none were given levodopa or antipsychotic drugs (such as haloperidol). Molecular and Chemical Neuropathology
Vol. 14. 1991
GSH, GSSG, and Vitamin E in Brain Disease
217
Table 1 Neuropathological Findings in All Patients' Age (y)
Illness duration (y)
Midbrain changes
C
73.2 ± 4.3
0
0
AD-PD
81.7 ± 2.0
1-7 AD 0-14 PD
AD
81.9 ± 2.1
1-9
AD changes Hippocampus
Cortex
none/8 mild/i
none!7 mild/2
10
none/1 mild/2 modera te/3 severe/4
none/1 mild/1 moderate/3 severe/5
3
moderate/4 severe/5
severe/9
'Patients are: C, control (ii = 9); AD, Alzheimer's disease patients (n = 9); AD-PD, patients with both Parkinson's disease and AD (11 = 10). Midbrain changes in AD-PD patients were neuron loss and Lewy body formation, and in AD patients were the presence of extraneuronal pigment and are expressed as the number of patients exhibiting changes. AD changes were the presence of neuritic plaques as discussed in the text, and are expressed as a relative degree of involvement per number of patients. Age is not different between controls and AD-PD or AD patients.
The data in Table 2 show that control levels of GSH were 50-90 times higher than levels of GSSG. The levels of GSH were significantly lowest in the medulla and significantly highest in the caudate nucleus (p < 0.05, ANOVA with Newman-Keuls test). There were no differences among control GSSG levels in the various brain regions. All groups of diseased patients exhibited enhanced levels of GSH in the hippocampus (Table 2). Patients with AD had elevated GSH levels in the midbrain, whereas in patients with AD-PD, midbrain GSH levels were comparable to controls. GSSG levels did not differ in any brain region among controls and diseased patients. Brain vitamin E levels are presented in Table 3. In control brain, the medulla contained significantly higher levels of vitamin E than the other regions examined. The caudate nucleus had the lowest vitamin E levels, which were half the levels in the medulla. Vitamin E levels in PD and AD changed only in the midbrain and both groups of patients (AD and AD-PD) had the same change, which was a doubling of vitamin F levels. There were no significant differences between patients with AD and patients with AD-PD in terms of levels of vitamin F in any brain region examined.
DISCUSSION Control GSH levels reported here are similar to the human brain GSH levels reported by Slivka et al. (1987b), and higher than those reported by Perry et al. (1982) or Riederer et al. (1989). Control GSSG levels are similar to those of Slivka et a!. (1987b) and much lower than Molecular and Chemical Neuropathology
Vol. 14, 1991
*
~ ....
,-I:..
-
z;: ,...
~
I
~
"if
~ n' !:L
£2
2.
~
s,
;;; ,.,
5.4 ± 1.9
3.1 ± 0.4
AD-
AD
1.13 ± 0.07* 18 ± 4 (8)
0.95 ± 0.06 26 ± 5 (10)
0.81 ± 0.09 16 ± 5 (6)
Midbrain
1.35 ± 0.07 15 ± 4 (9)
0.72 ± 0.06 13 ± 4 (9)
0.73 ± 0.10 23 ± 4 (10)
(7)
(7)
1.17 ± 0.09 25 ± 3 (10)
0.63 ± 0.10 10 ± 4
Medulla
1.15 ± 0.11 13 ± 3
Frontal cortex
1.50 ± 0.10 33 ± 5 (9)
1.43 ± 0.13 41 ± 6 (9)
1.20 ± 0.14 24 ± 7 (7)
Putamen
1.88 ± 0.11 32 ± 7 (9)
1.56 ± 0.19 36 ± 5 (9)
17 ± 3 (6)
1.25 ± 0.18
Caudate nucleus
1.51 ± 0.09* 33 ± 8 (9)
1.39 ± 0.09* 33 ± 9 (7)
(6)
0.96 ± 0.04 13 ± 4
Hippocampus
"Patients are: C, control (n = 9); AD, Alzheimer's disease (n = 9); and AD-PD, Parkinson's disease and AD (n = 10). Top figure for each group is glutathione (IJ-mol/g tissue). Lower figure is glutathione disulfide (nmollg tissue). Means ± S.E.M. Figures in brackets are the numbers of patients examined. *p < 0.05, as compared to controls (ANOVA with Newman-Keuls test). The postmortem interval is not significantly different between controls and AD-PD or AD patients. The n values do not always match the number of patient tissues examined because tissues from each region were not always available for each patient.
PO
7.4 ± 1.5
C
(h)
Post mortem interval
Table 2 Glutathione and Glutathione Disulfide Levels in Human Brain Samples"
~
I1l .....
~
):.
~
ex>
N ......
~
~
-....
........
~
~
~
~
.g
E;
~ ~
9 ~.
~
2..,is>
~
(9)
36.9 ± 3.6
(10)
34.0 ± 4.6
(7)
34.6 ± 4.4
Medulla
(9)
23.6 ± 1.9
(9)
23.5 ± 2.6
(7)
21.7 ± 2.8
Putamen
(9)
21.8 ± 2.7
(9)
15.1 ± 1.7
(6)
17.3 ± 2.4
Caudate nucleus
(9)
24.3 ±2.1
(7)
27.1 ±1.1
Hippocampa 19.9 ±2.2 (6)
"Patients are: C, control (11 "" 9);AD, Alzheimer's disease (11 "" 9); and AD-PD, Parkinson's disease and AD (11 = 10). Top figure for each group is vitamin E (J.Lg/g tissue). Lower figure (in brackets) is number of patients for each region analyzed. Means ± S.E.M. "p < 0.05, as compared to controls (ANOVA with Newman-Keuls test). The 11values do not always match the number of patient tissues examined because tissues from each region were not always available for each patient.
(8)
(9)
24.9 ± 2.8
42.3 ± 3.0*
(10)
AD
(10)
23.0 ± 3.3
(7)
44.0 ± 7.0*
(6)
24.5 ± 4.4
18.8 ± 3.3
AD-PD
c
Frontal cortex
Midbrain
Table 3 Vitamin E Levels in Human Brain Samples"
\0
tv -'
~
0;'
Ql S· b
t1:l
s·
rn
~s·
$:
~
Vi f)
a ~ @
220
Adams et aL
those of Perry et a!. (1982). The analysis of brain GSH and GSSG by Slivka et al. was done partly in humans and in anesthetized animals, to inhibit agonal responses. Slivka et a!. also flash froze all brain tissues to liquid nitrogen temperatures to inhibit postmortem oxidation of GSH. Slivka et a!. found that GSH and GSSG brain levels are about the same in animals and humans, 1-3 pmol/g and about 10 nmol/g, respectively, which is identical to the current report. In addition, the human levels reported here are very similar to previous animal work from this laboratory (Adams et al., 1989; Odunze et al., 1990a). However, Perry et a!. (1982) and Riederer et a!. (1989) did not prevent the artefactual oxidation of GSH in their samples (Adams et a!., 1983). Thus, their GSH levels may be too low and GSSG levels too high. This is a critical issue when examining brains from PD patients, since their midbrains have elevated levels of iron (Riederer et al., 1989) that might greatly accelerate the artefactual oxidation of GSH. Brain vitamin E levels reported here are very similar to previously reported human brain and liver levels but lower than levels in adipose tissue (Gallo-Torres, 1985; Metcalfe et al., 1989). In addition, mouse brain vitamin E levels reported by this laboratory (Odunze et al., 1990b) are substantially lower than human brain levels. The mouse brain analysis was done with flash frozen extracts of fresh brain tissue to inhibit postmortem changes in vitamin E. Our previously reported mouse brain vitamin E levels are similar to reported rat brain vitamin E levels, where rats were killed under anesthesia to inhibit agonal changes (Vatassery and Hagen, 1977; Vatassery et al., 1984a). It has been reported that the rat medulla accumulates vitamin E with age such that the medulla contains high levels of vitamin E when rats are very old (Vatassery et at., 1984a). This may be true in humans also, in that the medulla contains higher levels of vitamin F than other brain regions in the 70- to 80-y-old patients studied. The postmortem interval, which is the time elapsed after death until the samples were frozen in liquid nitrogen, was not significantly different between controls and neurologically diseased patients (Table 2). It is important to keep this interval as short as possible to minimize artefactual oxidation of GSH. The sampling interval here was shorter than that of Riederer et at. (1989), which was 12-14 h and about the same as Perry et al. (1982), which was 4 h. Our results suggest that damage to the brain occurs in AD and ADPD, causing a compensatory increase in brain GSH or vitamin E levels. One possible etiological factor could be a decrease in blood flow through selected areas of the brain (Hoyer, 1986) with concomitant hypoxia. Vascular changes associated with amyloid angiopathy in AD may be of importance. GSH is required to detoxify the oxygen radicals formed during hypoxic stress, which results in the oxidation of GSH by GSH peroxidase. Dopaminergic neurons of the midbrain are particularly sensi-
Molecular and Chemical Neuropathology
Vol. 14, 7991
GSH, GSSG, and Vitamin E in Brain Disease
221
tive to hypoxia (Phebus et a!., 1986). Other, nonoxidative mechanisms might also explain the increases in brain GSH, including induction of GSH synthetase activity. Increases in brain GSH are common during brain oxidative stress as induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (Adams et al., 1989) and 6-hydroxydopamine (Perumal et al., 1989). This may be indicative of the ability of the brain to defend itself from oxidative stress. The midbrain has a unique ability to concentrate vitamin E during these disease processes. It is not clear why increased vitamin E levels were not seen in the cerebral cortex or hippocampus in AD, since these regions suffer large amounts of damage in the disease process, especially given the increase in GSH levels in AD and AD-PD patients in the hippocampus. These findings suggest biochemical differences between vitamin E and GSH in terms of responses to damage in the hippocampus and midbrain. 1-Methyl-4-phenyl-1, 2,3,6-tetrahydropyridine induced oxidative stress in the brain (Adams et al., 1989) is associated with increases in vitamin E in the midbrain and other regions (Odunze et at., 1990b). The mechanism by which vitamin E levels might increase in the brain is not known. The brain does not possess a cytosolic binding protein for vitamin E such as is important in the transport of vitamin E in the liver (Murphy and Mavis, 1981a). The peripheral nervous system and other organs may take up vitamin E according to the amount of lipid present in the tissue (Clement and Bourre, 1990). However, the brain affinity for vitamin E may not correlate with vitamin E levels or total fatty acid content (Murphy and Mavis, 1981b). The brain is very different from other organs in that vitamin E does not readily penetrate into the brain in large amounts (Vatassery et al., 1984b), nor does vitamin E readily leave the brain (Goss-Sampson et a!., 1988). The midbrain does not appear to have a significantly greater intrinsic uptake of vitamin E than other regions (Goss-Sampson and Muller, 1987). Lipid levels are known to decrease in the brain with age (Soderberg et a!., 1990) and more specifically decrease in the midbrain in PD (Dexter et al., 1989). Using the lipid levels reported by Soderberg et al. (1990), control hippocampal vitamin E levels appear to be about 765 gig lipid. Drug therapy was probably not involved in alteration of antioxidant levels. Haloperidol was given to most of the AD and AD-PD patients. However, GSH levels in the two patient groups were not affected identically in the midbrain. Haloperidol does not produce permanent parkirtsonian side effects, nor does it induce Lewy body formation. Therefore, AD patients treated with haloperidol do not automatically develop PD. In addition, levodopa was clearly not a factor in this study since only three of the AD-PD patients received the drug. Levodopa therapy may be a confounding influence in published accounts of GSH levels in PD, since patients with PD alone are routinely treated with levodopa.
Molecular and Chemical Neuropathology
Vol 14, 1991
222
Adams et aL
It could be that increased brain GSH levels are a result of reactive gliosis or nerve terminal proliferation in response to neuronal death. Glial cells and nerve terminals contain higher levels of GSH than neuronal somas, which are destroyed both in PD and AD (Slivka et al., 1987a). Therefore, destruction of dopaminergic neurons, which is followed by gliosis, in the midbrain could actually result in an increase in GSH levels rather than a decrease. However, if reactive gliosis were an explanation for increased GSH levels in all brain regions, then GSH levels should also be increased in the frontal cortex, a major site of damage in AD. Since the cortex exhibited no increase in GSH levels, these results suggest the hippocampus and midbrain respond in a different manner or are more susceptible than the cortex and other sites to damage that alters GSH levels. The overlapping clinical and pathological profiles of AD and PD suggest similarities in the pathogenesis of the diseases. Boiler et al. (1980) reviewing 36 PD patients, demonstrated 55% with clinical dementia, and 42% of these with the neuropathological changes of AD, such as neuritic plaques and neurofibrillary tangles. Over 20% of the patients had these histological changes in the absence of dementia. The prevalence of pathologically established AD changes plus dementia was 33%. Loss of neurons in the nucleus basalis of Meynert and the locus ceruleus, the principal cholinergic and noradrenergic nuclei projecting to the cortex, respectively, have been observed in both PD and AD. Quantimetric studies have also shown a decreased number of neurons in the zona compacta of the substantia nigra of AD patients (Chui et al., 1986). Neurons are lost in the midbrain even in AD patients with no obvious Lewy bodies, in other words, patients that have AD and not AD-PD (Bergeron and Pollanen, 1989; Chui et al., 1986). In addition, neurofibrillary tangles are present in dopaminergic neurons of the substantia nigra in AD (Nakashima and Ikuta, 1985). However, the pathology of PD and AD may differ in the oculomotor neurons of the rostral midbrain where PD produces a loss of neurons and Lewy body formation, not seen in AD (Hunter, 1985). Hunter did find neurofibrillary tangles and gliosis in the rostral midbrain in AD. The Lewy body, intracytoplasmic inclusions in neurons of the substantia nigra, locus ceruleus, and other catecholaminergic neurons found in PD, have also been noted in AD in small and medium cortical neurons, particularly in layers 5 and 6 (Kosaka et al., 1976). They are especially numerous in cortical Lewy body disease, a separate neuropathological entity (Kosaka, 1978). Parkinsonian changes in AD patients are even more prevalent than the inverse. Ditter and Mirra (1987) demonstrated histological change of PD, including Lewy bodies, neuronal loss, and gliosis of pigmented nuclei in 11 of 20 AD patients. Clinically, rigidity was noted in 80% of these patients with parkinsonian pathology, but far fewer (14%) of the patients with no pathological changes of PD. No tremor was observed in either patient group. /sloieajlar and Chemical Neuropathology
Vol. 14, 1991
GSH, GSSG, and Vitamin E in Brain Disease
223
It has been previously suggested that the midbrain might be very susceptible to oxidative challenges that might alter GSH levels (Dexter et al., 1987, Kish et al., 1985). This might be due to the low levels of protective substances such as GSH and GSH peroxidase, and the high levels of free radical generating neuromelanin in dopaminergic neurons (Kish et a!., 1985, Slivka et al., 1987a). It may now be apparent that the midbrain, unlike other brain regions, also has the ability to concentrate vitamin E in its membranes in response to disease processes. This may mean that raising vitamin E levels in plasma by the administration of the vitamin to patients could provide for even higher vitamin E concentrations in the midbrain. This may provide a mechanism for the protective effects of vitamin E in PD (Factor et al., 1990) and may suggest that vitamin F therapy could be of benefit in AD and AD-PD.
ACKNOWLEDGMENTS This work was supported by NINCDS grant no. NS23515 (JA), NIA grant no. AG05142 (CAM), NIMH grant no. MH39145 (CAM), and a grant from the Alzheimer's Disease Association (CAM). The authors are grateful to Dr. James W. Geddes of the Division of Neurological Surgery at the University of California, Irvine College of Medicine for supplying some of the brain material.
REFERENCES Adams J. D., Lauterburg B. H., and Mitchellj. R. (1983) Plasma glutathione and glutathione disulfide in the rats Regulation and response to oxidative stress. J. Pharmacol, Exp. Ther. 227, 749-754. Adams J. D., Klaidman L. K., and Odunze 1. N. (1989) Oxidative effects of MPTP in the midbrain. Res. Commun. Subst. Abuse 10, 169-180. Bergeron C. and Pollanen M. (1989) Lewy bodies in Alzheimer disease—one or two diseases? Alzheimer Dis. Assoc. Disorders 3, 197-204. Boiler F., Mizutani T., Roessmann U., and Gambetti, P. (1980) Parkinson disease, dementia and Alzheimer disease: Clinicopathological correlations. Ann. Neurol. 7, 329-335. Brown M. S. and Goldstein J. L. (1986) A receptor-mediated pathway for cholesterol homeostasis. Science 232, 34-47, Chui H. C., Mortimer J. A., Slager U., Zarow C., Bondareff W., and Webster D. D. (1986) Pathologic correlates of dementia in Parkinson's disease. Arch. Neurol. 43, 991-995. Clement M., and BourreJ. M. (1990) Alteration of a-tocopherol content in the developing and aging peripheral nervous system: Persistence of high correlations with total and specific (n-6) polyunsaturated fatty acids. J. Neurochem. 54, 2110-2117. Molecular and Chemical Neuropathology
VoL 14. 1991
224
Adams et aL
Dexter D. T., Carter C. J., Wells F. R., Javoy-Agid F., Agid Y., Lees A., Jenner P., and Marsden C. D. (1989) Basal lipid peroxidation in substantia nigra is increased in Parkinson's disease. J. Neurochem. 52, 381-389. Dexter D. T., Wells F. R., Agid F., Agid Y., Lees A. J., Jenner P., and Marsden C. D. (1987) Increased nigral iron content in postmortem parkinsonian brain. Lancet 2, 1219-1220. Ditter S. M. and Mirra S. S. (1987) Neuropathologic and clinical features of Parkinson's disease in Alzheimer's disease patients. Neurology 37, 754-760. Factor S. A., Sanchez-Ramos J. R., and Weiner W. J. (1990) Vitamin E therapy in Parkinson's disease. Adv. Neural. 53, 457-461. Fariss M. W., Pascoe C. A., and Reed D. J. (1985) Vitamin E reversal of the effect of extracellular calcium on chemically induced toxicity in hepatocytes. Science 227, 751-754. Gallo-Torres H. E. (1985) Transport and metabolism, in Fat Soluble Vitamins: Their Biochemistry and Applications (Diplock A., ed.), pp 193-267, Technomic, Lancaster, Great Britain. Goss-Sampson M. A., MacEvilly C. J., and Muller D. P. R. (1988) Longitudinal studies of the neurobiology of vitamin E and other antioxidant systems, and neurological function in the vitamin E deficient rat. I. Neural. Sci. 87, 25-35. Goss-Sampson M. A. and Muller D. P. R. (1987) Studies on the neurobiology of vitamin E and some other antioxidant systems in the rat. Neuroputh. Appi. Neurobiol. 13, 289-296. Growdon J. H. and Corkin S. (1986) Cognitive impairments in Parkinson's disease. Adv. Neural. 45, 383-392. Hoyer S. (1986) Senile dementia and Alzheimer's disease, brain blood flow and metabolism. Prog. Neuro-Psychopharmacol. Bid. Psychiat. 10, 447-478. Hunter S. (1985) The rostra] mesencephalon in Parkinson's disease and Alzheimer's disease. Acta Neuropathol. 68, 53-58. Jackson C. V. E., Holland A. J., Williams C. A., and Dickerson J. W. T. (1988) Vitamin E and Alzheimer's disease in subjects with Down's syndrome. I. Mental Defic. Res. 32, 479-484. Khachaturian Z. S. (1985) Diagnosis of Alzheimer's disease. Arch. Neural. 42, 1097-1105. Kish S. J., Morito C., and Hornykiewicz, 0. (1985) Glutathione peroxidase activity in Parkinson's disease brain. Neurosci. Lett. 58, 343-346. Kosaka K. (1978) Lewy bodies in cerebral cortex, report of three cases. Acta Neuro path. 42, 127-134, Kosaka K., Oyanagi S., Matsushita M., Hon A., and Iwase S. (1976) Presenile dementia with Alzheimer-, Pick- and Lewy-body changes. Acta Neuropath. 36, 221-233. Martins R. N., Harper C. G., Stokes G. 13., and Masters C. L. (1986) Increased cerebral glucose-6-phosphate dehydrogenase activity in Alzheimer's disease may reflect oxidative stress. J. Neurochem. 46, 1042-1045. McCay P. B., Brueggemann C., Lai E. K., and Powell S. R. (1989) Evidence that a-tocopherol functions cyclically to quench free radicals in hepatic microsomes. Ann. New York Acad. Sci. 570, 32-45. McKhann C., Drachrnan D., Foistein M., Katzman R., Price D., and Stadlan E. M. (1984) Clinical diagnosis of Alzheimer's disease: Report of the NINCDSADRDA work group under the auspices of the Department of Health and Human Services task force of Alzheimer's disease. Neurology 34, 939-944. Molecular and Chemical Neuropathology
Vol. 74, 1991
GSH, GSSG, and Vitamin E in Brain Disease
225
Metcalfe T., Bowen D. M., and Muller D. P. R. (1989) Vitamin E concentrations in human brain of patients with Alzheimer's disease, fetuses with Down's syndrome, centenarians, and controls. Neurochem. Res. 14, 1209-1212. Murphy D. J. and Mavis R. D. (1981a) A comparison of the in vitro binding of c-tocopherol to microsomes of lung, liver, heart and brain of the rat. Biochim. Biophys. Ada 663, 390-400. Murphy D. J. and Mavis R. D. (1981b) Membrane transfer of a-tocopherol. J. Biol. Chem. 256, 10464-10468. Nakashima S. and Ikuta F. (1985) Catecholamine neurons with Alzheimer's neurofibriliary changes and alteration of tyrosine hydroxylase. Ada Neuropatiwl. 66, 37-41. Odunze 1. N., Klaidman L. K., and Adams, J. D. (1990a) MPTP toxicity: Differential effects in the striatum, cerebral cortex and midbrain on glutathione, glutathione disulfide and protein sulfhydryl levels. Res. Cornmun. Subst. Abuse. ii, 123-134, Odunze I. N., Klaidman L. K., and Adams, J. D. (1990b) MPTP toxicity in the mouse brain and vitamin E. Neurosci. Lett. 108, 346-349. Perry T. L., Godin D. V., and Hansen S. (1982) Parkinson's disease: A disorder due to nigral glutathione deficiency. Neurosci. Lett. 33, 305-310, Perry T. L. and Yong V. W. (1986) Idiopathic Parkinson's disease, progressive supranuclear palsy and glutathione metabolism in the substantia nigra of patients. Neurosci. Lett. 67, 269-274. Perumal A. S., Tordzro W. K., Katz M., Jackson-Lewis V., Cooper T. B., Fahn S., and Cadet J. L. (1989) Regional effects of 6-hydroxydopamine on free radical scavengers in rat brain. Brain Res. 504, 139-141. Phebus L. A., Perry K. W., Clemens J. A., and Fuller R. W. (1986) Brain anoxia releases striatal dopamine in rats. Life Sci. 38, 2447-2453. Riederer P., Sofic E., Rausch W. D., Schmidt B., Reynolds G. P., Jellinger K., and Youdim M. B. H. (1989) Transition metals, ferritin, glutathione, and ascorbic acid in parkinsonian brains. 1. Neurochem. 52, 515-520. Sims N. R., Bowen D. M., Neary D., and Davison A. N. (1983) Metabolic processes in Alzheimer's Disease: Adenine nucleotide content and production of 14 CO 2 from [U- 14 C]gluc6se in vitro in human neocortex. I. Neurochem. 41, 1329-1334, Sims N. R., Finegan J. M., Blass J. P., Bowen D. M., and Neary D. (1987) Mitochondrial function in brain tissue in primary degenerative dementia. Brain Res. 436, 30-38. Slivka A., Mytilineou C., and Cohen G. (1987a) Histochemical evaluation of glutathione in brain. Brain Res. 409, 275-284. Slivka A., Spina M. B., and Cohen C. (1987b) Reduced and oxidized glutathione in human and monkey brain. Neurosci. Lett. 74, 112-118. Soderberg M., Edlund C., Kristensson K., and Daliner C. (1990) Lipid compositions of different regions of the human brain during aging. J. Neurochem. 54, 415-423. Traber M. C., Olivecrona T., and Kayden H. J. (1985) Bovine milk lipoprotein lipase transfers tocopherol to human fibroblasts during triglyceride hydrolysis in vitro. J. Clin. Invest. 75, 1729-1734. Vatassery G. 1., Angerhofer C. K., and Knox, C. A. (1984a) Effect of age on vitamin E concentrations in various regions of the brain and a few selected Molecular and Chemical Neuropathology
Vol. 14. 1991
Adams et al.
226
peripheral tissues of the rat, and on the uptake of radioactive vitamin E by various regions Of rat brain. J. Neurochern. 43, 409-412. Vatassery C. T., Angerhofer C. K., Knox C. A., and Desmukh D. S. (1984b) Concentrations of vitamin E in various neuroanatomical regions and subcellular fractions, and the uptake of vitamin E by specific areas, of rat brain.
Biochim. Biophys. Acta 792, 118-422.
Vatassery C. T. and Hagen D. F. (1977) A liquid chromatographic method for quantitative determination of u-tocopherol in rat brain. Anal. Biochern. 79, 129-434.
Molecular and Chemical Neuropathology
VoL 14, 1991
