- Email: [email protected]
τ and ubiquitin in the human hypothalamus in aging and Alzheimer's disease
Brain Research, 590 (1992) 239-249 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
239
BRES 18064
z and ubiquitin in the human hypothalamus in aging and Alzheimer's disease D.F. Swaab a, I. G r u n d k e - I q b a l b, K. Iqbal b, H.P.H. K r e m e r c, R. Ravid a and J.A.P. van de Nes a,d a Netherlands Institute for Brain Research, Amsterdam (The Netherlands), b Institute for Basic Research in Decelopmoual Disabilities, Staten Island, NY 10314 (USA), c Department of Neurology, Leiden State Unicersity, Leiden (The Netherlands) and d Valerius Clinic, Free Unirersity, Amsterdam (The Netherlands)
(Accepted 7 April 1992)
Key words: Tau; Ubiquitin; Hypothalamus; Alzheimer; Aging; Cytoskeleton; AIz-50; Tangle
Immunocytochemical staining of hypothalamic cell groups with four antibodies to Alzheimer paired helical filaments (PHF) (i.e., anti-PtlF serum 60e and monoclonal antibody (mAb) AIz-50, both directed against normal and abnormally phosphorylated r; t a b r-l, which recognizes r; and mAb 3-39 to PHF, which recognizes the earboxy terminal domain of ubiquitin) revealed a clear distinction between 12 Alzheimer's disease (AD) patients and seven controls in the hypothalamus. Dystrophic neurites, which appeared to be the most specific components in AD, were most conspicuous after AIz-50 staining. However, AIz-50 also stained neuronal cytoplasm and normal, thin, beaded neurites in the paraventricular nucleus (PVN) of controls, even of young cases. This staining was clearly distinct from the staining of cytoplasm and dystrophic neurites in the PVN of Alzheimer patients. The abundant staining of dystrophic neurites and cell bodies in the nucleus tuberalis lateralis (NTL) in AD, in which no neuronal loss is observed, suggests that alterations in cytoskeletal markers do not necessarily indicate impending cell death. Moreover, the cytoskeletal changes in the NTL, sexually dimorphic and suprachiasmatic nuclei in AD indicate that this condition is not restricted to cortical areas or nuclei projecting to the cortex. Consequently, the pathophysiological implications of cytoskeletal staining in AD are at present far from clear, The human hypothalamus may not only provide a better insight into the pathogenesis of Alzheimer's disease, but could also be of help in the neuropathological diagnosis of this condition.
INTRODUCTION Alzheimer's disease is characterized histopathologically by the gradual accumulation in the brain of extracellular amyloid in the neuropil and vessel walls and by intraneuronal cytoskeletal changes. The latter changes may eventually lead to the formation of neurofibrillary tangles and dystrophic neurites. These neurites may be associated with amyloid in neuritic plaques, or they may occur independently of them, in which case they have also been called neuropil threads 7'9. In addition to the cytoskeletal pathology, loss of cells (see below) and synapses 35 is found in Alzheimer's disease. Silver staining methods probably visualize only the late stages of alterations 4. On the other hand, the diagnostic specificity of the more sensitive immunocytochemical staining of the cytoskeletal changes in Alzheimer's disease
may be restricted because some positively staining neurons are also found in the brains of non-demented young individuals "). The following hypotheses have been proposed in the literature to explain the cytoskeletal changes stained immunocytochemically: (i) it takes place preferentially in cortical neurons or neurons projecting to the cortex ms or, in terms of evolution, Alzheimer's disease would primarily affect those brain areas that have been acquired recently in phylogeny39; (ii) diffuse labeling of neuronal perikarya with antibodies recognizing paired helical filaments precedes the formation of tanglesh'27; (iii) cytoskeletal alterations are a sign of impending neuronal death, not only in Alzheimer's disease 4'55'62, but also in normal brain development 2'62. In the present study these possibilities have been examined in the hypothalamus. The hypothalamic area is a suitable region for test-
Correspondence: D.F. Swaab, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam-Zuidoost, Netherlands. Fax: (31) (20) 6918466.
240 TABLE ! Clinical information on controls and AIHwimer patients
Age is measured in years: bw, brain weight in g: pm = postmortem interval, in h, ft = fixation time, in days; AD = Alzheimer's disease. sex
age
bw
pm
ft
cause of death
d ? ,; d d ~ ~
47 59 74 83 85 88 91
1620 II10 1410 1280 1400 1030 1060
24 24 13 22 16 11 48
39 183 48 42 44 35 28
spinal muscular atrophy, embolies leukemia, subendocardial hemorrhage pulmonary artery thrombosis, pneumonia perforated colon diverticle leukemia, bronchopneumonia basalioma left ear. heart failure bronchopneumonia
Aizheimer's disease: 86364.52 d 85005.0 ~ 86196.6 d 87494.2 ~ 88.252 d 83170 ~ 8560 9 84283 ~ 8484 ~ 84050 9 86l~4 ~ 86001 d
45 56 61 64 66 70 70 86 87 90 90 97
!130 1180 1260 931 1250 780 1210 1150 1275 950 i 060 1200
4 24 84 62 3 14 18 34 42 2 3 5
!19 48 63 I11 30 34 34 49 292 33 38 40
AD. cachexia AD, bronchopneumonia AD AD AD, ischemic cerebellar stroke AD, status epilepticus AD, pulmonary embolism AD, bronchopneumonia AD, bronchopneumonia AD, pneumonia, adenocarcinoma of cecum AD, dehydration AD, ischemic cerebral strokes
Controls: 87271 800271 81032 81064 82175 81100 83179
ing the various hypotheses proposed because it contains both neurons that project to the cortex (e.g., the nucleus basa'is of Meynert and the tuberomammiilary nucleus) and neurons that do not so project (e.g., the supraoptic nucleus, the paraventricular nucleus, the suprachiasmatic nucleus and the sexually dimorphic nucleus), In addition, the human hypothalamus allows the study of the relation between cytoskeletal changes and cell death in aging and dementia because the hypothalamic nuclei can be clearly delineated, permitting accurate counting of their constituent cells. Earlier studies showed that the various nuclei appeared to follow a different pattern of cell loss in aging and Alzheimer's disease (see Table I1). The nucleus basalis of Meynert (NBM), along with the diagonal band of Broca, is the major source of cholinergic innervation of the neocortex. The number of large cholinergic neurons in the NBM is generally reported to decrease in Alzheimer's disease m2.34.59.Recent data show that neuronal loss in the NBM occurs during normal aging as well 13.3a. However, this de.. crease of cholinergic neurons might, at least partly, simply be due to cell shrinkage, which would increase the number of small neurons 3,4°.
The neurosecretory neurons of the supraoptic and paraventricular nuclei (SON and PVN) produce the neuropeptides vasopressin and oxytocin, which are released into the bloodstream. They form a population of extremely stable cells. No significant loss in neurons or total cell number has been observed, either in the course of normal aging or in AIzheimer's disease ~°. The sexually dimorphic nucleus (SDN) of the preoptic area or 'intermediate nucleus 't' in the male human brain contains twice as many cells as in the female human brain 4~. Cell numbers in the male SDN decrease sharply after 50 years of age, whereas in females, a phase of marked cell loss sets in around the age of 70 (refs. 25,51). Cell numbers in the SDN of Alzheimer's patients are within the normal range for age and sex 5~. The suprachiasmatic nucleus (SCN) is considered to be the major circadian pacemaker of the mammalian brain, coordinating hormonal and behavioral circadian rhythms 4m.Age-associated changes in circadian rhythms have been reported both in man and in animals 56. Among the most prominent changes is a fragmentation of sleep-wakefulness patterns in senescence, a phenomenon that is even more pronounced in Alzheimer's
Fig. 1. The nucleus basalis of Meynert (NBM) of an Alzheimer patient (No. 8560) stained by (A) 60e after formic acid pretreatment; (B) ~--1 after dephosphorylation; (C) 3-39; and (D) AIz-50. Note the cytoplasm staining and the dystrophic neurites, which are most conspicuous after AIz-50 staining (D). (E) shows that the NBM of a control patient (No, 82175) is negative after AIz-50 staining. The same holds for the supraoptic nucleus (SON) of an Alzheimer patient (No, 8560) stained by AIz-50 (F), Bar -- 200 p,m.
"ri
rll
,
•
4
- °
.
~.
. ; :
_
"
.
-
F
-
.
qb
" -
--.-
.
]
--
!
.
,
.;
,rl
.
•
.C)
-
-.
•
"-
•
o
o
" 4 .
IO
.
:
;-
.
"
•
-"
•
11
o °
j
..'t
o
. -#.
...
d ~
"-
-~'~
]~ "1 ,~
•"
• -
o
A
"'
.-,-
_m
~°
.
--
~
0
•
.
•
.
m
•
~ S
• ,,,.
i
"
.
}p
.
ql
•
•
~
"
f
,
D
~
•
"~ql
dP
•
•
,.
;
°
I
L
•
."
•
•
•
0
.
"
o
r
•
. -
-
'
,;
-
.
.
,
•
.
",~
,,'-
~ " 6 "
"~"
.
•
•
,.
-O
t
r
.
•
Oe
i,
.!
g
~
I
D
p
°
gp
J
#
S
.-
0
'/,
'
.m
"0
".
~
~"
.
•
:/.
.1
•
E
a
,f
a,
• .
h,
2
°
l",,..)
242
++ ,,
+
d'
t!
+
.I
+-
+
+ °"+
+
~+,, + ~ ++~+
+
, • ..,,. ,.,,
~l~,"
+ ,+
dl
•
.,
•
•
•.
""
~'
•
++
+
•
i
++:,++ " ~ . . I ' I P .
.
,de
.,
+.
+~II .
•
qalp
~II
".,..
+.I~.
,
,n.L +
8
.
:o"
,
+llp+~. •
•
. . . . . ' o . . ". ' - - " "
~
."
0
-
' + i,
•
-
.,
".
' a •k , .
+~
.-""
• d,j
,
w,++.~- +.
e'
,I
I qlllk
•
.
.
.
I°
" .~II. ~
."
°
q~
°
-'
' •
.......
,
U"
~,,, , . + 1 '
~+~
..
~'
\
i
0
#
i
+
+ o~,
,+."
• ..o;:'~F
,qp," j
•
"
~#"
~
•
-.
~
.
o
.~+
•
ii ....
+
~
,
A,,,
,all •
0
.
~"
. B
•
o.
O
O
,
O
" '
e
.;
lip
L ,it,
°
,
m.
e
II qlP'
O
D
,4pip
41
_
°
6
II''~
•
-,.
,++,.
~
, +
"
qj'.
tip
| +
Ip J
Dip
el,
41
"e
+
4
dip' ~1'4 '°ll
lP
,P
ql ,
+
II
~
° Oi+°
,Q
.
qp
o
•
Ii1%
tp
i+
-
o,
dl
,o
+.'+6 all
,,dll
.
o
,o
ii
in lJl
I
e I
,
lip
alp
dB
,.
i
.
Ii
.b
'
D ,
Ill
"
°.
'lb
ii
"-
~
-."
I
~
~"
-t,.
~ ,q
"~
+ , .
II
+,I
ip
6
•
"
p
•
•
•
. ":
• ,I
i1+
.~.~
o
. -
~o
.
.
,,b.
i
"
.
.,
"
.
..
. .4,,
tO
•
41
,+.
, •
tl
.
i,o
•
,.
•
~;
.,
..._
~
.
•
b
IQ+,L
•
~"
.
4
O+ iii
lip
IP
-,,,
'
.
,1~
,i
,.- q b
-,
~
,,
,~ •
"
"
,,lib'.
e
is 6
o a,
II O
•
• . •
,
ul ,i
•
•
•
.-,
":" •
~,
Q
,
"
+" ;
o,
dO
~. •
1
""
"
. ,
~
",
ql''
6.
.
. •
#
i
IB'II.
"o~q . t~ !
I
E
:"
•
+1='
ql
243 disease 6°. Total SCN cell number and number of vasopressin neurons in control subjects of 80-100 years of age is lower than in younger subjects; cell loss in Alzheimer's patients is greater than in elderly control subjects s°.s2. The nucleus tuberalis lateralis (NTL) is a phylogenetically young structure that has been recognized only in man and higher primates. Its neuron number does not change in Aizheimer's disease 29. Dystrophic new rites in the NTL are observed in Alzheimer's disease 29. Since in various neurodegenerative diseases, i.e., Alzheimer's 17'29'46and Huntington's diseases 16,28.42and a non-Alzheimer type of dementias (H. Braak, personal communication) NTL pathology is associated with severe loss of weight despite normal or even increased food intake, this nucleus may play a role in feeding behavior and metabolism. The tuberomammillary nucleus (TM) is formed by large, irregularly bordered, dark staining neurons that surround the NTL, the fornix in its final descending course, and the mammillary body 14. Many of its neurons project extensively to the cortex 43. For example, from this group originates the major, if not the sole, histaminergic cortical innervation, in rodents 47'5s as well as in man L37. TM has since long been known to be affected by Alzheimer's disease: the occurrence of tangles and deposition of plaques can be found in this n u c l e u s 26'44'45'54. Morphometrics have only been applied to a few subjects and concern Galanin neurons. Their number did not change in Alzheimer's disease m'. No clear qualitative changes in the number of histamine neurons were observed between Aizheimer patients and controls'. The present study compares the staining patterns of various antibodies directed against cytoskeletai constituents in hypothalamic nuclei and relate them to the changes in cell numbers of the individual nuclei in Alzheimer's disease. MATERIALS AND METHODS For the present study, 6-,am-thick paraffin sections of formalinfixed hypothalami were used from 12 neuropathologically confirmed seriously affected AIzheimer patients in the age range of 45-97 years and seven controls between 47 and 91 years of age (Table I). Details of tissue processing have been described previously 4~'s°'51. The controis had neither clinical nor neuropathological features of AIzheimer's disease of any significance.
The following four antibodies were used. 60e , a polyclonal antiserum raised against isolated neurofibrillary tangles, recognizes both normally and abnormally phosphorylated ¢:2.23. The antibody was used both without pretreatment and after pretreatment of the sections with formic acid to uncover immunoreactive epitopes 22. The sections pretreated with formic acid were immunolabeled at 1:2,000 and the untreated ones at 1:1,000 dilutions. T-I, a monoclonal antibody (ascites) directed to the microtubule-associated protein ~-, was used at a dilution of 1:50,000. Dephosphorylation of the tissue sections was performed before immunocytochemistry, using 235 units alkaline phosphatase (Boehringer Mannheim, Indianapolis, IN) per mi, 0.1 M Tris-HCI pH 8.0, 0.01 M phenylmethylsulfonylfluoride at 37°C for 2.5 h; without dephosphorylation, minimal staining was obtained with ~'-1 in the brain of Alzheimer patients 2~. 3 - 3 9 , a monoclonal antibody raised against isolated neurofibrillary tangles and directed against ubiquitin 4'24.38.57, was used at a dilution of 1 : 10,000 (ascites). Tissue sections were incubated with the above three primary antibodies at 22"(2 for 18 h. Bound antibody was detected with biotinylated species-specific anti-rabbit (1:400) or anti-mouse (1 : 100) immunoglobulins (Amersham, Arlington Heights, IL) and avidin/horseradish peroxidase complex (1:400, Sigma, St. Louis, MO) followed by development (1-1.5 min) with 0.05% (w/v) 3,3'-diaminobenzidine and 0.01% (v/v) H20 2. Alz-50 , a monoclonai antibody that binds to normal ¢ and Alzheimer-specific protein A-68 (refs. 19,30,31,36,61); A-68 is an abnormally phosphorylated ¢ in PHF 32. Alz-50 (IgG) was used at a dilution of 1:100. The tissue sections were incubated with this antibody for one h at room temperature followed by overnight incubation at 4°C. The staining was based on Wolozin et al. 61, with a few modifications in the DAB step (0.05 M Tris-HCi and 0.15 M NaCI, pH 7.6, DAB 0.005%, 0.01% H20 2) and the use of 0.5% nickel ammonium sulfate as an enhancer (Van de Nes et al., Brain, in press.
RESULTS In general, all four antibodies clearly stained cell bodies to differing degrees in various hypothalamic nuclei of Alzheimer patients. Antibody 'r-1 without dephosphorylation did not stain the hypothalamic nuclei of either Alzheimer patients or controls. The nucleus basalis of Meynert (NBM) of controls showed no, or only rarely, a few immunoreactive cells or dystrophic neurites (i.e., thick, kinky, unbeaded fibers) with any of the four antibodies (e.g., Fig. 1E). In contrast, NBM neurons of all Alzheimer patients showed a clear staining of perikarya and dystrophic neurites with all four antibodies (Fig. 1A-D). The dystrophic neurites were most conspicuous after Alz-50 staining (Fig. 1D). None of the four antibodies revealed any staining in the supraoptic nucleus (SON), either in controls or in Alzheimer patients (e.g., Fig. IF).
Fig. 2. A: the paraventricular nucleus (PVN) of a 47-year-old control patient (No. 87271) stained by Aiz-50. Note the cytoplasmic staining and the thin beaded fibers. No staining was observed with 60e, ¢-1 and 3-39; B: the PVN of an AIzheimer patient (No. 8560) stained by AIz-50. Note the increase in cytoplasmic staining and the presence of dystrophic neurites; C-F: the nucleus tuberalis lateralis (NTL) of an AIzheimer patient (No. 84050) stained by (C) 60e, after formic acid pretreatment, (D) ~--1, after dephosphorylation, (E) 3-39, and (F) AIz-50. Note the cytoplasmic staining and the dense network of dystrophic neurites, especially after AIz-50 staining. Bar -- 200 ~m.
244 .~,-
•
". .,
."
|
." •
.
•
• i
IP a,~,
:
"8
"1
•
•
.
"
.
~'
"
.e
•
~.
o
~
.~".~¢
-
#
f
4''
Ad...
m
.,,.-2
-
•
_% r
)
.,
.,
..
"~.
-;~
-
:
... D
• d
..
~: ~
-~ e
.. •
,.
•
.,'.
•
.
..
•
] ~-
/r
B
'l
III
.
-
• oo i
~
.
•
~
P
•
e L.
i,
z!~ ' 't't ° •
m
o.
P e
!
.
.}
,
"., III t
• •
•
i~i
i
i ~ ~ ;*~
a
~ 0o~
~Q
i
eD
o.
•
i,i
alp.
E
.
~
~
F
245 TABLE 11 Cytoskeletal alterations in the hypothalamus in aging and AIzheimer's disease in relation to cell loss Staining: *, cytoplasmic staining and staining of normal, thin, beaded fibres; * *, decrease, but within the normal range for age and sex; - , no or negligible staining of Alzheimer changes (i.e., plaques, tangles, dystrophic neurites); +, only a very few Alzheimer changes; +, some Alzheimer changes; + +, moderate number of Alzheimer changes; + + +, large number of Alzheimer changes; + + + +, very large number of Alzheimer changes. The category in Table I! is based upon the median score of the individual cases. Cell loss: O, no cell loss; +, questionable/small cell loss; l , cell loss; J, ~,, strong cell loss; n.d., not determined. Antibodies: 60e, polycional against tangles; r-l, monoclonal against tau; 3-39, monoclonal against ubiquitin; AIz-50, monoclonai against abnormal tau (A68). Nuclei: NBM, nucleus basalis of Meynert; SON, supraoptic nucleus; PVN, paraventricular nucleus; NTL, nucleus tuberalis lateralis; SDN, sexually dimorphic nucleus; SCN, suprachiasmatic nucleus; TM, tuberomammillary nucleus. antibodies nuclei Control cases: NBM SON
PVN SDN SCN NTL TM
60e
60e (formic acid)
+
. . . . . .
~.-I (dephosph. )
5:
-
. .
. .
. . . .
.
.
. . . .
. . . .
3 - 30
A!z-50
cell loss in aging (refs.)
:t:
-
+ 0 0 J, ,I, 0 n.d.
. . . . .
*/-
(12, 13, 34) (20) (20) (25, 49, 51) (50, 52) (Kremer, unpubl.)
cell loss in Alzheimer (refs.) AIzheimer's disease cases: NBM + SON . PVN ±/SDN + SCN ± NTL +++ TM ++
+ +
.
+ +
. +/+ ± +++ ++
.
+
+ + +
+/± ± ++ +
*/+/+ + ++++ +++
. ~ + + ++ +
in the majority of the controls, and even in young cases, staining of some diffusely labeled cell bodies and of thin beaded fibers was observed in the paraventricular aucleus (PVN) with AIz-50 (Fig. 2A). In Alzhcimer patients, more neurons showed cell body staining, and in addition, thicker unbeaded dystrophic neurites were observed in the PVN (Fig. 2B). With the three other antibodies, no staining was observed in PVN neurons of controls, whereas in the Alzheimer patients, these three antibodies did reveal staining and dystrophic neurites in the PVN. The sexually dimorphic nucleus (SDN) of control subjects was not stained with the cytoskeletal markers, whereas in Alzheimer patients cytoplasmic staining and dystrophic neurites were observed. The most pronounced staining was obtained with Alz-50 (Fig. 3). In general, the suprachiasmatic nucleus (SCN) did not show any cytoskeletal staining in controls, with the
(12,34,59) (20) (20) **(51) ¢ (50, 52) 0 (29) 0 (11) 0 0
exception of sporadic dystrophic neurites in a few subjects (Nos. 81032 83179), whereas in Aizheimer patients, cell body staining and dystrophic neurites were clearly observed. However, in spite of the pronounced SCN changes in Alzheimer's disease, these alterations were only modest, except for one 64-year-old female Alzheimer patient (No. 87494.2), whose entire SCN was stained with ~'-1 and 3-39 but not with 60e or Alz-50 (Fig. 3). In the nucleus tuberalis lateralis (NTL) of controls, dystrophic neurites and a single stained cell body were observed only rarely. In contrast, in the NTL of Alzheimer patients, all four antibodies showed a large number of cells with cytoplasmic staining, although slightly fewer perikarya were stained with 3-39 (Fig. 2). The most remarkable feature was the extremely high density of dystrophic nearites, especially with Alz-50 staining (Fig. 2F). The latter antibody even enabled us
iii,-.
Fig. 3. A, B: the sexually dimorphic nucleus (SDN) of an Alzheimer patient (No. 8560) stained by (A) AIz-50 and (B) 3-39. Note the cytoplasmic staining and dystrophic neurites. For r-I staining of the SDN after dephosphorylation and 60e staining after formic acid treatment see 4s; C-F: the suprachiasmatic nucleus (SCN) of an Alzheimer patient (No. 87494.2), the borders of which are indicated by arrows. The SCN is weakly stained by (C) anti-vasopressin, for the localization of this nucleus (see reference 35; I!I, third ventricle), (D) r-l, (E) 3-39, and (F) AIz-50. Note the weak vasopressin staining (C) in the degenerating SCN of this Alzheimer patient and that the entire SCN can be delineated after r-I (D) or 3-39 (E) staining. However, the staining is much less pronounced in this case after AIz-50 staining (F). Bar -- 200/zm.
246 T A B L E 111
Cytoskeletai changes in NTL of indii,idual cases of controls and Aizheimer's disease patients in relation to age, post-mortem tissue and tissue ~¢ation inter~'als Age is measured in years, pro, postmortem interval, in h; ft, fixation time, in days; ND, not determined, in the tissue blocks of Nos. 87949.2 and 84283, no N T L was present. For an explanation o f the o t h e r abbreviations, see Table II.
age pm ft
60e
47 59 74 83 85 88 91
. . . . . . ±
60e ~'.1 3-39 (formic (dephosph.) acid)
Aiz-50
Controls." 87271 800271 81032 81064 82175 81100 83179
2g 24 13 22 16 11 48
39 183 48 42 44 35 28
. . . . . .
. . .
. .
. .
. .
. . .
. -
. .
±
-
. -
± ±
± ++ ++ +++ ++ ++ +++ ++ ++
+ + + ND +++ + +++ ++ +
+++ ++++ ++++ ++++ ++++ ++++ + ++++ ++++ ++++
Al:.heimer's disease: 86364.52 85005.0 86196.6 88.252 83170 8560 8484 84050 86004 86001
45 4 119 + ++ 56 24 48 + + + +++ 61 84 63 + + + +++ 66 3 30 + + + +++ 70 14 34 + + + +++ 70 18 34 + ++ 87 42 292 .-1: + 90 2 33 + + + +++ 90 3 38 + + + + + e7 5 40 + + ++
to distinguish the NTL in Alzheimer patients by the unaided eye. As in the NTL, in the tuberomammillary nucleus (TM) of controls immunostaining of perikarya and nzurites was rare. The AIzheimer brains, however, showed numerous staining cell bodies in all patients. The density and proportion of staining cell bodies appeared to be lower for the TM than for the NTL. Again, numerous Alz-50 staining neurites were observed. Neuritic staining with the other antibodies was markedly less. In contrast to the NTL, the TM did show neurofibriilary tangles in Palmgren's silver impregnation. Table Ill provides the individual data on the NTL the most intense staining structure of the hypothalamus of Alzheimer patients (cf. Table II). These data clearly indicate that a postmortem autolysis time of up to 84 h and a fixation time of up to 119 days are not critically important for staining with the four antibodies in the hypothalamus. Only in the case of the longest fixation time (292 days)was relatively little staining observed. It should be noted, however, that the classical neuropathological Alzheimer changes of this case (No. 8484) were also only moderate. The observation that Aiz-50 staining of the cortex of AIzheimer patients remained excellent for up to 471 days of formalin
fixation (Van de Nes, unpublished results), supports the idea that even longer fixation times than were used in the present study might not be critical. DISCUSSION
The four antibodies revealed basically similar patterns in staining of cell bodies and dystrophic neurites in the hypothalami of Alzheimer patients and hardly any staining in controls. The relative staining intensity of the four different antibodies varied per patient and per hypothalamic nucleus, but not in any systematic way. Dystrophic neurites were most conspicuous with Alz-50 staining, yet background staining was the lowest with this antibody. The specific pattern of cytoskeletal changes in nuclei such as the NBM and NTL might be of help in the neuropathological diagnosis of Alzheimer's disease. The NBM is the nucleus that has been studied most extensively so far in relation to its decreasing cholinergic output in Alzheimer's disease. There was full agreement between the presence of cytoplasmic staining and dystrophic neurites in the NBM of Alzheimer patients on the one hand and the clinical and neuropathological diagnosis on the other. In controls, hardly any NBM staining occurred. It would appear, therefore, that the NBM is a very suitable structure for the validation of the neuropathologicai diagnosis of AIzheimer's disease. Another such structure for diagnostic confirmation with these cytoskeletal markers seems to be the NTL, despite the fact that silver and thioflavine.S stains disclosed few neurofibrillary tangles or senile plaques 2'~, suggc~,tiug at first sight that the NTL is not involved in AIzheimer's disease. No neuronal loss was found in the NTL of AIzheimer patients 2~. Yet, immunocytochemistry with the four antibodies revealed a completely different picture. Alz50 showed remarkable staining of perikarya and an extremely dense network of dystrophic neurites in the Alzheimer patients and no such staining in controls. Alz-50 reactivity with dystrophic neurites was so abundant that the NTL of Alzheimer patients could be recognized by the unaided eye. The 60e and ~'-1 antibodies basically visualized the same number of positive neurons, but the staining of dystrophic neurites was less conspicuous and background activity was higher than with Alz-50. The distinction between Alzheimer patients and controls could easily be made on the basis of the NTL staining with each of the four antibodies. An exception was a 90-year-old female control (81.033) in another study, who showed considerably more Alz-50 staining in the NTL than in the NBM. The peculiar Alz-50-stained changes in the NTL that are seen in
247 Alzheimer's disease are also present in Down's syndrome, but are absent in a number of conditions, such as Huntington's disease, Parkinson's disease and AIDS dementia (unpublished results), suggesting the disease specificity of this staining. The pattern of neuronal alterations in the TM was essentially that of the NBM, with immunostaining of perikarya and neurites, but a smaller proportion of the neurons seem to be affected. Precisely because of this difference, the alterations in the immunostained sections were less spectacular than in the NTL. As in the NBM, this pattern of severe Alzheimer-related changes may be related to the existence of extensive cortical projections 43. In the NBM and SCN of Alzhein~er patients, structures known to suffer from cell loss in this disease 12'33'34'5°'52'59 cytoskeletal alterations have been observed invariably. In contrast, the SON does not lose neurons 2° and, apparently, no cytoskeletal staining takes place. Thus, it might be concluded that cytoskeletai staining is associated with imminent neuronal death. However, results from other hypothalamic nuclei make this hypothesis untenable. Contrary to what is generally reported in the literature t9'61 but in agreement with the findings by Byne et al. H~,Alz-50 did show activity in control brain sections, i.e., in the PVN. Although the PVN does not lose cells in aging or Alzheimer's disease 2°, some diffuse cytoplasmic staining and staining of thin beaded fibers were observed in this nucleus with Alz-50 in controls, and more cytoplasmic staining and dystrophic neurites were observed in the PVN of Alzheimer patients. Also, the abundant staining of all cytoskeletal markers in the NTL of Alzheimer patients related to its continuing normal complement of neurons and the absence of neurofibrillary tangles shows that such cytoskeletal staining does not necessarily indicate impending neuronal death or the imminent formation of tangles. Unlike in hypothalamic nuclei such as the NTL, in the neocortex and especially in the hippocampus of Alzheimer patients many neurons with neurofibrillary tangles seem to degenerate, leaving behind ghost tangles. It thus appears that neurons in hypothalamic nuclei might have a higher threshold and that they impede the progression of eytoskeletal changes to the formation of tangles and, ultimately, cell death. However, even these initial stages of cytoskeletal changes might have functional consequences in the affected individuals. The NTL cytoskeletal changes may correlate with disturbances in eating behavior and metabolism. In the PVN of Alzheimer patients, in addition to the cytoplasmic staining and the visualization of beaded
fibers as found by Aiz-50 staining in controls, thick unbeaded dystrophic neurites and tangles were observed. The difference in staining between controls and Alzheimer patients might be due to the presence of different antigens for the same antibody (cf. refs. 31,36). This idea is reinforced by the finding l° that Alz-50 immunoreactivity was associated with vesicles in controls and with filaments and vesicles in Alzheimer patients and by the observation that the PVN staining in controls was obtained only with Alz-50, and not with the other three antibodies, whereas all four antibodies showed cytoskeletal changes in Alzheimer's disease. As a similar PVN staining was not obtained with anti-~- in controls it seems unlikely that the two antibodies recognize the same type of native ~-. Whatever the chemical background might be, the two types of Alz-50 staining should be clearly distinguished if this marker is used for the histopathological diagnosis of Alzheimer's disease. Some positive cell bodies and dystrophic neurites were found with 60e, ~--1, 3-39 and Alz-50 in the SDN of Alzheimer patients, whereas controls were negative for all four antibodies. Because the SDN cell numbers decrease in a similar way in Alzheimer patients and controls, it should again be concluded that cytoskeletai staining in the hypothalamus cannot be related in a simple way to cell death in the SDN. A thiofiavine-S study has also revealed that the appearance of tangles does ilot coincide with neuronal death in the locus coeruleus and NBM 5. In contrast, in the cerebral cortex and hippocampus, a significant negative correlation was observed between tangle counts and neuronal counts 5. As that study was based on the measurement of neuronal densities and not on total cell counts, additional morphometry is needed to establish this difference firmly as one between cortical and subcortical structures in Alzheimer's disease (cf. refs. 12,53). Neuritic plaques were not generally observed in the hypothalamic nuclei of the present study. Nonetheless, the SCN loses 75% of its cells during the Alzheimer process. In the NTL, only a few non-neuritic plaques are present that are reactive with SP28 (a rabbit polyclonal antibody against the 28 residue synthetic peptide which is homologous to the NH2-terminal region of /3-protein) 29, whereas the distribution of Alz-50 positive dystrophic neurites is extremely dense. Apparently, the presence of plaques is not a prerequisite for cell death in Alzheimer's disease or in the initiation of the hypothalamic cytoskeletal changes. Cytoskeletal changes may also occur without the presence of diffuse plaques. Regarding the hypotheses mentioned in the intro-
248 duction on the distribution of affected neurons, several remarks can be made. The very strong AIz-50 staining in the NTL, from which no direct cortical connections are known, raises doubts about the idea that cytoskeletal changes would be restricted to cortical areas. The same can be said about the SCN and SDN staining observed in AIzheimer patients whereas, in addition, these nuclei are not considered to be directly connected to the cortex. In more general terms, the ~- and ubiquitin staining in various hypothalamic structures indicates that the hypothesis that Alzheimer's disease primarily affects areas that have been acquired recently in phylogenesis ~q is not tenable. The present study shows that the cytoskeletal changes in AIzheimer's disease does not necessarily precede tangle formation. Nor can it be regarded as a sign of impending cell death, in addition, a clear distinction should be made between cells and fibers that stain in controls and structures that stain in Alzheimer's disease. Moreover, it has been shown that Alzheimer's disease co-evolves with a specific pattern of hypothalamic r and ubiquitin staining. Neuropathological diagnosis of mildly affected Aizheimer patients is still a major problem t'~. It seems therefore worthwhile to investigate whether this pattern of hypothalamic staining might be of help in the diagnosis of such cases of Alzhcimer's disease. Acknowh,dgements. Brain material was obtained from the Netherlands Brain Bank (Amsterdam, Th~ Netherlands), Neuropatholosy was pt~rformed by Dr, W, Kamphorst (Fret; University, Amsterdam) and Dr. D. Troost (Academic Medical Center of the University of Amsterdam). Alz,5(I was kindly donated by Abbott Laboratories (Abbott Park, IL, USA). r-I was I~encrously supplied by Dr. L. Binder (University of Alabama, Birmingham, AL, USA). The authors wish to thank Mr, B. Fisser, Ms. J.P.H.M. PIoegmakers and Ms, Y.-C. Tung for their technical assistance, Mr. G, Van der Meulen for his photographical work, Ms. M, Stoddard Marlow for copy-editing this manuscript and Ms. O. Pach and Mr. A. Janssen for their secretarial assistance. The study was carried out as part of the AMSTEL project with support from the Stimuleringsprogramma Gezondheidsonderzoek (SGO) of the Ministry of Science and Education in The Netherlands, New York State Office of Mental Retardations and Developmental Disabilities, NIH Grants AG 05892, AG 08076, NS 18105 and AG 04220, and a grant from the American Health Assistance Foundation, Rockville, MD, USA. D.F, Swaab and I. Grundke-lqbal were recipients of Senior Travel Fellowships of AIzheimer Society of Canada and AIzheimer's Association, USA, respectively, to participate in the Second International Conference on Alzheimer's Disease. Toronto. Canada, 1990. REFERENCES I Airaksinen, M.S., Paer, A., Paljiirvi, L., Reinikanen, K., Riekkihen, P., Suomalainen, R. and Panula, P., Histamine neurons in human hypothalamus: anatomy in normal and Alzheimer diseased brains, Neuroscience, 44 ( 1991) 465-481. 2 AI-Ghoul, W.M, and Miller, M.W., Transient expression of AIz-50 immunt~reactivity in developing rat neocortex: a marker for naturally occurring neuronal death?, Brain Res,, 481 (1989) 361-367. 3 Allen, S.J., Dawbarn, D, and Wilcock, G.K., Morphometric ira-
munochemical analysis of neurons in the nucleus basalis of Meynert in Alzheimer's disease, Brain lies,, 454 (1988) 275-281. 4 Bancher, C., Brunner, C., Lassmann, H., Budka, H., Jellinger, K., Wiche, G., Seitelberger, F., Grundke-lqbal, I., lqbal, K. and Wisniewski, H.M., Accumulation of abnormally phosphorylated ~" precedes the formation of neurofibrillary tangles in Alzheimer's disease, Brain Res., 477 (1989)90-99, 5 Bondareff, W., Mountjoy, C.Q., Roth, M. and Hauser, D.L., Neurofibriilary degeneration and neuronal loss in Alzheimer's disease, Neurobiol. Aging, 10 (1989) 709-715. 6 Braak, H. and Braak, E., The hypothalamus of the human adult: chiasmatic region, Anat. Embryol., 176 (1987)315-330. 7 Braak, H. and Braak, E., Neuropii threads occur in dendrites of tangle-bearing nerve cell, Neuropathol. Appl. Neurobiol., 14 (1988) 39 -44. 8 Braak, H. and Braak, E., Cortical and subcortical argyrophilic grains characterize a disease associated with adult onset dementia, Neuropathol. App! NeurobioL, 15 (1989) 13-26. 9 Braak, H., Braak, E., Grundke-lqbal, I. and lqbal, K., Occurrence of neuropil threads in the senile human brain and in AIzheimer's disease: a third location of paired helical filaments outside of neurofibrillary tangles and neuritic plaques, Neurosci. Lett., 65 (1986) 351-355. 10 Byne, W., Mattiace, L., Kress, Y. and Davies, P., Aiz-50 immunoreactivity in the hypothalamus of the normal and Alzheimer human and the rat, J. Comp. Neurol., 306 (1991) 602-612. I1 Chan-Palay, V. and Jentsch, B., Galinin tuberomammi!lary neurons in the hypothalamus in AIzheimer's and Parkinson's disease. In D.F. Swaab, M. Mirmiran, R. Ravid and F.W. Van Leeuwen (Eds.), The Human Hypothalamus in Health and Disease, Progress in Brain Research, Vol. 93, Elsevier, Amsterdam, in press. 12 Coleman, P.D. and Flood, D.G., Neuron numbers and dendritic extent in normal aging and Alzheimer's disease, Neurobiol. Aghzg, 8 (1987) 521-545. 13 De Lacalle, S., iraizoz, I. and Ma Gonzalo, L., Differential changes in cell size and number in topographic subdivisions of human basal nuclet.s in normal aging, Neuroscience, 43 (1991) 445 -456. 14 Diepen, R., Der Hypothalamus. in W. Bargmann (Ed.), Handbach der mikro,~'kopische Anatomie des Mens('hen. IV~ 7, Springer, Berlin, 1962, pp, 1-181. 15 Duykaerts, C., Dela~re, P., Hauw, J.-L, Abbamondi-Pinto, A.L., Sorbi, S., Allen, !,, Brion, J,P,, Flament-Durand, J., Duchen, L., Kauss, J,, Schlote, W., Lowe, J., Probst, A., Ravid, R., Swaab. D.F., Renkawek, K, and Tomlison, B., Rating of the lesions in senile dementia of the Alzheimer type: concordance between laboratories, J. Neurol, Sci., 97 (1990)295-323. 16 Farrer, L.A., Yu, P.-L., Anthropometric discrimination among affected, at risk, and not-at-risk individuals in families with Huntington's disease, Am. J. Med. Gen., 21 (1985) 307-316, 17 Franklin, C.A. and Karkeck, J., Weight loss and senile dementia in an institutionalized elderly population, J. Am. Diet Assoc, 89 (1989) 7Q0-792, 18 German, D,C., White Ill, C,L, and Sparkman, D.R., Alzheimer's disease: neurofibrillary tangles in nuclei that project to the cerebral cortex, Neuroscienee, 21 (1987)305-312. 19 Goedert, M., Spillantini, M.G. and Jakes, R,, Localization of the AIz-50 epitope in recombinant human microtubule-associated protein r, Neurosci, Lett,, 126 (1991) 149-154. 20 Goudsmit, E,, Herman, M,A,, Fliers, E, and Swaab, D.F., The supraoptic and paraventricular nuclei of the human hypothalamus in relation to sex, age and AIzheimer's disease, Neurobiol. Aging, I1 (1990)529-536, 21 Grundke-lqbal, i,, lqbal, K,, Tung, Y.-C., Quinlan, M., Wisniewski, H,M, and Binder, L.I., Abnormal phosphorylation of the microtubule-associated protein r (r) in AIzheimer cytoskeletal pathology, Prec. Natl, Acad. Sci. USA, 83 (1986) 4913-4917. 22 Grundke-lqbal, I., Iqbal, K., Tung, Y,-C, and Wisniewski H.M., AIzheimer paired helical filaments: immunochemical identification of polypeptides, Acta Neuropathoi. (Bed.), 62 (1984) 259-267. 23 Grundke-lqbal, l,, Vorbrodt, A,W., Iqbal, K., Tung, Y.-C., Wang, G,P. and Wisniewski, H,M,, Microtubule-associated polypeptides
249 r are altered in Alzheimer paired helical filaments, Mol. Brain Res., 4 (1988) 43-52. 24 Grundke-lqbal, !., Wang, G.P., lqbal, K., Tung, Y.-C. and Wisniewski, H.M., Alzheimer paired helical filaments: identification of polypeptides of monoclonal antibodies, Acta Neuropathol. (Beri.), 68 (1985) 279-283. 25 Hofman, M.A. and Swaab, D.F., The sexually dimorphic nucleus of the preoptic area in the human brain: a comparative morphometric study, J. Anat., 164 (1989) 55-72. 26 Ishii, T., Distribution of Alzheimer's neurofibrillary changes in the brain stem and hypothalamus of senile dementia, Acta Neuropathot, 6 (1966) 181-187. 27 Joachim, C.L., Morris, J.H., Selkoe, D.J. and Kosik, K.S., r epitopes are incorporated into a range of lesions in Alzheimer's disease, J. Neuropathol. Exp. Neurol., 6 (1987) 611-622. 28 Kremer, H.P.H., Roos, R.A.C., Dingian, G., Marani, E. and Bots, G. Th. A.M., Atrophy of the hypothalamic lateral tuberal nucleus in Huntington's disease, J. Neuropathol. Exp. Neurol., 49 (1990) 371-382. 29 Kremer, B., Swaab, D.F., Bots, G., Fisser, B., Ravid, R. and Roos, R., The hypothalamic lateral tuberal nucleus in Alzheimer's disease, Ann. Neurol., 29 (1991) 279-284. 30 Ksiezak-Reding, H., Binder, L.I. and Yen, S.-H., Immunochemical and biochemical characterization of r proteins in normal and Alzheimer's disease brains with Alz-50 and r-l, J. Biol. Chem., 263 (1988) 7948-7953. 3-I Ksicsak-Reding, H., Davies, P. and Yen, S.-H., AIz 50, a monoclonal antibody to Alzheimer's disease antigen, cross-reacts with r proteins from bovine a," ! normal human brain, J. Biol. Chem., 263 (1988) 7943-7947. 32 Lee, V.M.-Y., Balin, B.J., Otvos Jr., L. and Trojanowski, J.Q., A68: a major subunit of paired helical filaments and derivatized forms of normal r, Science, 251 (1991) 675-678. 33 Lowes-Hummel, P., Gertz, H.-J., Ferszt, R. and Cervos-Navarro, J., The basal nucleus of Meynert revised: the nerve cell number decreases with age, Arch. Gerontol. Geriatr., 8 (1989) 21-27. 34 Mann, D.M.A., Yates, P.O. and Marcyniuk, B., Changes in nerve cells of the nucleus basalis of Meynert in Alzheimer's disease and their relationship to ageing and to the accumulation of lipofuscin pigment, Mech. Ageing Dr,r,., 25 (1984) 189-204. 35 Masliah, E,, Terry, R.D., DeTeresa, R.M. and Hansen, L.A., lmmunohistochemical quantification of the synapse-related protein synaptophysin in AIzheimer disease, Neurosci. Lett., 103 (1989) 234-239, 36 Nukina, N,, Kosik, K.S. and Selkoe, D.J., The monoclonal antibody, AIz 50, recognizes r proteins in Alzheimer's disease brain, Neurosci. Lett., 87 (1988) 240-246. 37 Panda, P,, Airaksinen, M.S., Pirvola, U. and Kotilainen, E., Histamine containing neuronal systt,m in human brain, Neuro. science, 34 (1990) 129-132, 38 Perry, G., Mulvihill, P., Fried, V.A., Smith, H.T., Grundke-lqbal, !. and lqbal, K., lmmunochemical properties of ubiquitin conjugates in the paired helical filaments of Alzheimer disease, J. Neurochem., 52 (I 989) 1523-1528. 39 Rapoport, S.I., Hypothesis: Alzheimer's disease is a phylogenetic disease, Med. Hypotheses, 29 (1989) 147-150. 40 Rinne, J.O., Paljiirvi, L. and Rinne U.K., Neuronal size and density in the nucleus basalis of Meynert in Alzheimer's disease, J. Neurol. ScL, 79 (1987) 67-76. 41 Rusak, B. and Zucker, I., Neural regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526. 42 Sanberg, P.R., Fibiger, H.C. and Mark, R.F., Body weight and dietary factors in Huntington's disease patients compared with matched controls, Med. J. Austr., 1 (1981) 407-409. 43 Saper, C.B., Organization of cerebral cortical afferent systems in the rat. II. Hypothalamocortieal projections, J. Comp. NeuroL, 237 (1985) 21-46. 44 Saper, C.B. and German, D.C., Hypothalamic pathology in Alzheimer's disease, Neurosci. Lett., 74 (1987) 364-370. 45 Simpson, W.A., Yates, C.M., Watts, A.G. and Fink, G., Congo red birefringent structures in the hypothalamus in senile demen-
46 47
48
49 50 51 52
53
54
55
56
57
58
59
60
61
62
tia of the Alzheimer type, Neuropathol. Appl. NeurobioL, 14 (1988) 381-393. Singh, S,, Mulley, G.P. and Losowsky, M.S. Why are AIzheimer patients thin?, Age Aging, 17 (1988) 21-28. Steinbusch, H.W.M. and Mulder, A.H., Localization and projections of histamine immunoreactive neurons in the central nervous system of the rat. in A. Bjfrklund, T. H6kfelt and M.J. Kuhar (Eds.), Handbook of Chemical Neuroanatomy, Vol. 3, Elsevier, Amsterdam, 1984, pp. 126-140. Swaab, D.F., Eikelenboom, P., Grundke-lqbal, I., lqbal, K., Kremer, H.P.H., Ravid, R. and Van de Nes, J.A.P., Cytoskeletal alterations in the hypothalamus during aging and in AIzheimer's disease are not necessarily a marker for impending cell death. In K. lqbal, D.R.C. McLachlan, B. Winblad and H.M. Wisniewski (Eds.), Alzheimer's Disease: Basic Mechanisms, Diagnosis and Therapeutic Strategies, John Wiley, New York, 1991, pp. 181-190. Swaab, D.F. and Fliers, E., A sexually dimorphic nucleus in the human brain, Science, 228 (1985) 1112-1115. Swaab, D.F., Fliers, E. and Partiman, T.S., The suprachiasmatic nucleus of the human brain in relation to sex, age and dementia, Brain Res., 342 (1985) 37-44. Swaab, D.F. and Hofman, M.A., Sexual differentiation of the human hypothalamus: ontogeny of the sexually dimorphic nucleus of the preoptic area, Dec. Brain. Res., 44 (1988) 314-318. Swaab, D.F., Roozendaal, B., Ravid, R., Velis, D.N., Gooren, L. and Williams, R.S., Suprachiasmatic nucleus in aging, AIzheimer's disease, transsexuality and Prader-Willi syndrome. In R. de Kloet, V.M. Wiegant and D. de Wied (Eds.), Neuropeptides and Brain Function. Progress in Brahl Research, Vol. 72, Elsevier, Amsterdam, 1987, pp. 301-310. Swaab, D.F. and Uylings, H.B.M., Comments on review by Coleman and Flood 'Neurons numbers and dendritic extent in normal aging and Alzheirner's disease', Density measures: parameters to avoid, Neurobiol. Aging, 8 (1987) 574-575. Ulfig, N. and Braak, H., Amyloid deposits and neurofibrillary changes in the hypothalamic tuberomammillary nucleus, J. Neural Transm. (P.-D. Sect.), ! (1984) 143. Ulrich, J., Probst, A., Anderton, B.H., Brion, J.P., Cytoskeletal immunohistochemistry of Alzheimer dementia and related diseases. In H.P. yon Hahn (Ed.), h~terdisciplinary Topics in Gerontology, VoL 2£ !!1. The Neuronal Cytoskeleton in Senile Dementia of AIzheimer Type and Other Types of Dementia, S. Karger, Basel, 1988° pp. 101 - 105. Van Gool, W.A. and Mirmiran, M., Aging and circadian rhythms. In D.F. Swaab, E. Fliers, M. Mirmiran, W.A. van Gool and F. van Haaren (Eds.), Aging of the Brain and Alzheimer's Disease. Progress in Brahz Re~earch, VoL 70, Elsevier, Amsterdam, 198f~, 255 -279. Wang, G.P., Grundke-lqbal, I., Kascsak, R.J., lqbal, K., Wisniewski, H.M., Alzheimer neurofibrillary tangles: monoclonal antibodies to inherent antigen(s), Acta NeuropathoL (Bet'/.), 62 (1984) 268-275. Watanabe, T., Taguchi, Y., Shiosaka, S., Tanaka, J., Kubota, H., Terano, Y., Tohyama, M. and Wada, H., Distribution of the histaminergic neuron system in the central nervous system of rats: a fluorescent immunohistochemical analysis with histidine decarbox'ylase as a marker, Brain Res., 295 (1984) 13-25. Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A,W., Coyle, J.T. and DeLong, M.R., Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain, Sci¢,nce, 215 (1982) 12371239. Witting, W., Kwa, I.H., Eikelenboom, P., Mirmiran, M. and Swaab, D.F., Alterations in the circadian rest-activity rhythm in aging and Alzheimer's disease, Biol. P~ychiatry, 27 (1990) 563-572. Wolozin, B.L., Pruchnicki, A., Dickson, D.W. and Davies, P., A neuronal antigen in the brains of Alzheimer patients, Science, 232 (1986) 648-650. Wolozin, B., Scicutella, A. and Davies, P., Reexpression of a developmentally regulated antigen in Down syndrome and Alzheimer disease, Proc. Natl. Acad. Sci. USA, 85 (1988) 62026206.
239
BRES 18064
z and ubiquitin in the human hypothalamus in aging and Alzheimer's disease D.F. Swaab a, I. G r u n d k e - I q b a l b, K. Iqbal b, H.P.H. K r e m e r c, R. Ravid a and J.A.P. van de Nes a,d a Netherlands Institute for Brain Research, Amsterdam (The Netherlands), b Institute for Basic Research in Decelopmoual Disabilities, Staten Island, NY 10314 (USA), c Department of Neurology, Leiden State Unicersity, Leiden (The Netherlands) and d Valerius Clinic, Free Unirersity, Amsterdam (The Netherlands)
(Accepted 7 April 1992)
Key words: Tau; Ubiquitin; Hypothalamus; Alzheimer; Aging; Cytoskeleton; AIz-50; Tangle
Immunocytochemical staining of hypothalamic cell groups with four antibodies to Alzheimer paired helical filaments (PHF) (i.e., anti-PtlF serum 60e and monoclonal antibody (mAb) AIz-50, both directed against normal and abnormally phosphorylated r; t a b r-l, which recognizes r; and mAb 3-39 to PHF, which recognizes the earboxy terminal domain of ubiquitin) revealed a clear distinction between 12 Alzheimer's disease (AD) patients and seven controls in the hypothalamus. Dystrophic neurites, which appeared to be the most specific components in AD, were most conspicuous after AIz-50 staining. However, AIz-50 also stained neuronal cytoplasm and normal, thin, beaded neurites in the paraventricular nucleus (PVN) of controls, even of young cases. This staining was clearly distinct from the staining of cytoplasm and dystrophic neurites in the PVN of Alzheimer patients. The abundant staining of dystrophic neurites and cell bodies in the nucleus tuberalis lateralis (NTL) in AD, in which no neuronal loss is observed, suggests that alterations in cytoskeletal markers do not necessarily indicate impending cell death. Moreover, the cytoskeletal changes in the NTL, sexually dimorphic and suprachiasmatic nuclei in AD indicate that this condition is not restricted to cortical areas or nuclei projecting to the cortex. Consequently, the pathophysiological implications of cytoskeletal staining in AD are at present far from clear, The human hypothalamus may not only provide a better insight into the pathogenesis of Alzheimer's disease, but could also be of help in the neuropathological diagnosis of this condition.
INTRODUCTION Alzheimer's disease is characterized histopathologically by the gradual accumulation in the brain of extracellular amyloid in the neuropil and vessel walls and by intraneuronal cytoskeletal changes. The latter changes may eventually lead to the formation of neurofibrillary tangles and dystrophic neurites. These neurites may be associated with amyloid in neuritic plaques, or they may occur independently of them, in which case they have also been called neuropil threads 7'9. In addition to the cytoskeletal pathology, loss of cells (see below) and synapses 35 is found in Alzheimer's disease. Silver staining methods probably visualize only the late stages of alterations 4. On the other hand, the diagnostic specificity of the more sensitive immunocytochemical staining of the cytoskeletal changes in Alzheimer's disease
may be restricted because some positively staining neurons are also found in the brains of non-demented young individuals "). The following hypotheses have been proposed in the literature to explain the cytoskeletal changes stained immunocytochemically: (i) it takes place preferentially in cortical neurons or neurons projecting to the cortex ms or, in terms of evolution, Alzheimer's disease would primarily affect those brain areas that have been acquired recently in phylogeny39; (ii) diffuse labeling of neuronal perikarya with antibodies recognizing paired helical filaments precedes the formation of tanglesh'27; (iii) cytoskeletal alterations are a sign of impending neuronal death, not only in Alzheimer's disease 4'55'62, but also in normal brain development 2'62. In the present study these possibilities have been examined in the hypothalamus. The hypothalamic area is a suitable region for test-
Correspondence: D.F. Swaab, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam-Zuidoost, Netherlands. Fax: (31) (20) 6918466.
240 TABLE ! Clinical information on controls and AIHwimer patients
Age is measured in years: bw, brain weight in g: pm = postmortem interval, in h, ft = fixation time, in days; AD = Alzheimer's disease. sex
age
bw
pm
ft
cause of death
d ? ,; d d ~ ~
47 59 74 83 85 88 91
1620 II10 1410 1280 1400 1030 1060
24 24 13 22 16 11 48
39 183 48 42 44 35 28
spinal muscular atrophy, embolies leukemia, subendocardial hemorrhage pulmonary artery thrombosis, pneumonia perforated colon diverticle leukemia, bronchopneumonia basalioma left ear. heart failure bronchopneumonia
Aizheimer's disease: 86364.52 d 85005.0 ~ 86196.6 d 87494.2 ~ 88.252 d 83170 ~ 8560 9 84283 ~ 8484 ~ 84050 9 86l~4 ~ 86001 d
45 56 61 64 66 70 70 86 87 90 90 97
!130 1180 1260 931 1250 780 1210 1150 1275 950 i 060 1200
4 24 84 62 3 14 18 34 42 2 3 5
!19 48 63 I11 30 34 34 49 292 33 38 40
AD. cachexia AD, bronchopneumonia AD AD AD, ischemic cerebellar stroke AD, status epilepticus AD, pulmonary embolism AD, bronchopneumonia AD, bronchopneumonia AD, pneumonia, adenocarcinoma of cecum AD, dehydration AD, ischemic cerebral strokes
Controls: 87271 800271 81032 81064 82175 81100 83179
ing the various hypotheses proposed because it contains both neurons that project to the cortex (e.g., the nucleus basa'is of Meynert and the tuberomammiilary nucleus) and neurons that do not so project (e.g., the supraoptic nucleus, the paraventricular nucleus, the suprachiasmatic nucleus and the sexually dimorphic nucleus), In addition, the human hypothalamus allows the study of the relation between cytoskeletal changes and cell death in aging and dementia because the hypothalamic nuclei can be clearly delineated, permitting accurate counting of their constituent cells. Earlier studies showed that the various nuclei appeared to follow a different pattern of cell loss in aging and Alzheimer's disease (see Table I1). The nucleus basalis of Meynert (NBM), along with the diagonal band of Broca, is the major source of cholinergic innervation of the neocortex. The number of large cholinergic neurons in the NBM is generally reported to decrease in Alzheimer's disease m2.34.59.Recent data show that neuronal loss in the NBM occurs during normal aging as well 13.3a. However, this de.. crease of cholinergic neurons might, at least partly, simply be due to cell shrinkage, which would increase the number of small neurons 3,4°.
The neurosecretory neurons of the supraoptic and paraventricular nuclei (SON and PVN) produce the neuropeptides vasopressin and oxytocin, which are released into the bloodstream. They form a population of extremely stable cells. No significant loss in neurons or total cell number has been observed, either in the course of normal aging or in AIzheimer's disease ~°. The sexually dimorphic nucleus (SDN) of the preoptic area or 'intermediate nucleus 't' in the male human brain contains twice as many cells as in the female human brain 4~. Cell numbers in the male SDN decrease sharply after 50 years of age, whereas in females, a phase of marked cell loss sets in around the age of 70 (refs. 25,51). Cell numbers in the SDN of Alzheimer's patients are within the normal range for age and sex 5~. The suprachiasmatic nucleus (SCN) is considered to be the major circadian pacemaker of the mammalian brain, coordinating hormonal and behavioral circadian rhythms 4m.Age-associated changes in circadian rhythms have been reported both in man and in animals 56. Among the most prominent changes is a fragmentation of sleep-wakefulness patterns in senescence, a phenomenon that is even more pronounced in Alzheimer's
Fig. 1. The nucleus basalis of Meynert (NBM) of an Alzheimer patient (No. 8560) stained by (A) 60e after formic acid pretreatment; (B) ~--1 after dephosphorylation; (C) 3-39; and (D) AIz-50. Note the cytoplasm staining and the dystrophic neurites, which are most conspicuous after AIz-50 staining (D). (E) shows that the NBM of a control patient (No, 82175) is negative after AIz-50 staining. The same holds for the supraoptic nucleus (SON) of an Alzheimer patient (No, 8560) stained by AIz-50 (F), Bar -- 200 p,m.
"ri
rll
,
•
4
- °
.
~.
. ; :
_
"
.
-
F
-
.
qb
" -
--.-
.
]
--
!
.
,
.;
,rl
.
•
.C)
-
-.
•
"-
•
o
o
" 4 .
IO
.
:
;-
.
"
•
-"
•
11
o °
j
..'t
o
. -#.
...
d ~
"-
-~'~
]~ "1 ,~
•"
• -
o
A
"'
.-,-
_m
~°
.
--
~
0
•
.
•
.
m
•
~ S
• ,,,.
i
"
.
}p
.
ql
•
•
~
"
f
,
D
~
•
"~ql
dP
•
•
,.
;
°
I
L
•
."
•
•
•
0
.
"
o
r
•
. -
-
'
,;
-
.
.
,
•
.
",~
,,'-
~ " 6 "
"~"
.
•
•
,.
-O
t
r
.
•
Oe
i,
.!
g
~
I
D
p
°
gp
J
#
S
.-
0
'/,
'
.m
"0
".
~
~"
.
•
:/.
.1
•
E
a
,f
a,
• .
h,
2
°
l",,..)
242
++ ,,
+
d'
t!
+
.I
+-
+
+ °"+
+
~+,, + ~ ++~+
+
, • ..,,. ,.,,
~l~,"
+ ,+
dl
•
.,
•
•
•.
""
~'
•
++
+
•
i
++:,++ " ~ . . I ' I P .
.
,de
.,
+.
+~II .
•
qalp
~II
".,..
+.I~.
,
,n.L +
8
.
:o"
,
+llp+~. •
•
. . . . . ' o . . ". ' - - " "
~
."
0
-
' + i,
•
-
.,
".
' a •k , .
+~
.-""
• d,j
,
w,++.~- +.
e'
,I
I qlllk
•
.
.
.
I°
" .~II. ~
."
°
q~
°
-'
' •
.......
,
U"
~,,, , . + 1 '
~+~
..
~'
\
i
0
#
i
+
+ o~,
,+."
• ..o;:'~F
,qp," j
•
"
~#"
~
•
-.
~
.
o
.~+
•
ii ....
+
~
,
A,,,
,all •
0
.
~"
. B
•
o.
O
O
,
O
" '
e
.;
lip
L ,it,
°
,
m.
e
II qlP'
O
D
,4pip
41
_
°
6
II''~
•
-,.
,++,.
~
, +
"
qj'.
tip
| +
Ip J
Dip
el,
41
"e
+
4
dip' ~1'4 '°ll
lP
,P
ql ,
+
II
~
° Oi+°
,Q
.
qp
o
•
Ii1%
tp
i+
-
o,
dl
,o
+.'+6 all
,,dll
.
o
,o
ii
in lJl
I
e I
,
lip
alp
dB
,.
i
.
Ii
.b
'
D ,
Ill
"
°.
'lb
ii
"-
~
-."
I
~
~"
-t,.
~ ,q
"~
+ , .
II
+,I
ip
6
•
"
p
•
•
•
. ":
• ,I
i1+
.~.~
o
. -
~o
.
.
,,b.
i
"
.
.,
"
.
..
. .4,,
tO
•
41
,+.
, •
tl
.
i,o
•
,.
•
~;
.,
..._
~
.
•
b
IQ+,L
•
~"
.
4
O+ iii
lip
IP
-,,,
'
.
,1~
,i
,.- q b
-,
~
,,
,~ •
"
"
,,lib'.
e
is 6
o a,
II O
•
• . •
,
ul ,i
•
•
•
.-,
":" •
~,
Q
,
"
+" ;
o,
dO
~. •
1
""
"
. ,
~
",
ql''
6.
.
. •
#
i
IB'II.
"o~q . t~ !
I
E
:"
•
+1='
ql
243 disease 6°. Total SCN cell number and number of vasopressin neurons in control subjects of 80-100 years of age is lower than in younger subjects; cell loss in Alzheimer's patients is greater than in elderly control subjects s°.s2. The nucleus tuberalis lateralis (NTL) is a phylogenetically young structure that has been recognized only in man and higher primates. Its neuron number does not change in Aizheimer's disease 29. Dystrophic new rites in the NTL are observed in Alzheimer's disease 29. Since in various neurodegenerative diseases, i.e., Alzheimer's 17'29'46and Huntington's diseases 16,28.42and a non-Alzheimer type of dementias (H. Braak, personal communication) NTL pathology is associated with severe loss of weight despite normal or even increased food intake, this nucleus may play a role in feeding behavior and metabolism. The tuberomammillary nucleus (TM) is formed by large, irregularly bordered, dark staining neurons that surround the NTL, the fornix in its final descending course, and the mammillary body 14. Many of its neurons project extensively to the cortex 43. For example, from this group originates the major, if not the sole, histaminergic cortical innervation, in rodents 47'5s as well as in man L37. TM has since long been known to be affected by Alzheimer's disease: the occurrence of tangles and deposition of plaques can be found in this n u c l e u s 26'44'45'54. Morphometrics have only been applied to a few subjects and concern Galanin neurons. Their number did not change in Alzheimer's disease m'. No clear qualitative changes in the number of histamine neurons were observed between Aizheimer patients and controls'. The present study compares the staining patterns of various antibodies directed against cytoskeletai constituents in hypothalamic nuclei and relate them to the changes in cell numbers of the individual nuclei in Alzheimer's disease. MATERIALS AND METHODS For the present study, 6-,am-thick paraffin sections of formalinfixed hypothalami were used from 12 neuropathologically confirmed seriously affected AIzheimer patients in the age range of 45-97 years and seven controls between 47 and 91 years of age (Table I). Details of tissue processing have been described previously 4~'s°'51. The controis had neither clinical nor neuropathological features of AIzheimer's disease of any significance.
The following four antibodies were used. 60e , a polyclonal antiserum raised against isolated neurofibrillary tangles, recognizes both normally and abnormally phosphorylated ¢:2.23. The antibody was used both without pretreatment and after pretreatment of the sections with formic acid to uncover immunoreactive epitopes 22. The sections pretreated with formic acid were immunolabeled at 1:2,000 and the untreated ones at 1:1,000 dilutions. T-I, a monoclonal antibody (ascites) directed to the microtubule-associated protein ~-, was used at a dilution of 1:50,000. Dephosphorylation of the tissue sections was performed before immunocytochemistry, using 235 units alkaline phosphatase (Boehringer Mannheim, Indianapolis, IN) per mi, 0.1 M Tris-HCI pH 8.0, 0.01 M phenylmethylsulfonylfluoride at 37°C for 2.5 h; without dephosphorylation, minimal staining was obtained with ~'-1 in the brain of Alzheimer patients 2~. 3 - 3 9 , a monoclonal antibody raised against isolated neurofibrillary tangles and directed against ubiquitin 4'24.38.57, was used at a dilution of 1 : 10,000 (ascites). Tissue sections were incubated with the above three primary antibodies at 22"(2 for 18 h. Bound antibody was detected with biotinylated species-specific anti-rabbit (1:400) or anti-mouse (1 : 100) immunoglobulins (Amersham, Arlington Heights, IL) and avidin/horseradish peroxidase complex (1:400, Sigma, St. Louis, MO) followed by development (1-1.5 min) with 0.05% (w/v) 3,3'-diaminobenzidine and 0.01% (v/v) H20 2. Alz-50 , a monoclonai antibody that binds to normal ¢ and Alzheimer-specific protein A-68 (refs. 19,30,31,36,61); A-68 is an abnormally phosphorylated ¢ in PHF 32. Alz-50 (IgG) was used at a dilution of 1:100. The tissue sections were incubated with this antibody for one h at room temperature followed by overnight incubation at 4°C. The staining was based on Wolozin et al. 61, with a few modifications in the DAB step (0.05 M Tris-HCi and 0.15 M NaCI, pH 7.6, DAB 0.005%, 0.01% H20 2) and the use of 0.5% nickel ammonium sulfate as an enhancer (Van de Nes et al., Brain, in press.
RESULTS In general, all four antibodies clearly stained cell bodies to differing degrees in various hypothalamic nuclei of Alzheimer patients. Antibody 'r-1 without dephosphorylation did not stain the hypothalamic nuclei of either Alzheimer patients or controls. The nucleus basalis of Meynert (NBM) of controls showed no, or only rarely, a few immunoreactive cells or dystrophic neurites (i.e., thick, kinky, unbeaded fibers) with any of the four antibodies (e.g., Fig. 1E). In contrast, NBM neurons of all Alzheimer patients showed a clear staining of perikarya and dystrophic neurites with all four antibodies (Fig. 1A-D). The dystrophic neurites were most conspicuous after Alz-50 staining (Fig. 1D). None of the four antibodies revealed any staining in the supraoptic nucleus (SON), either in controls or in Alzheimer patients (e.g., Fig. IF).
Fig. 2. A: the paraventricular nucleus (PVN) of a 47-year-old control patient (No. 87271) stained by Aiz-50. Note the cytoplasmic staining and the thin beaded fibers. No staining was observed with 60e, ¢-1 and 3-39; B: the PVN of an AIzheimer patient (No. 8560) stained by AIz-50. Note the increase in cytoplasmic staining and the presence of dystrophic neurites; C-F: the nucleus tuberalis lateralis (NTL) of an AIzheimer patient (No. 84050) stained by (C) 60e, after formic acid pretreatment, (D) ~--1, after dephosphorylation, (E) 3-39, and (F) AIz-50. Note the cytoplasmic staining and the dense network of dystrophic neurites, especially after AIz-50 staining. Bar -- 200 ~m.
244 .~,-
•
". .,
."
|
." •
.
•
• i
IP a,~,
:
"8
"1
•
•
.
"
.
~'
"
.e
•
~.
o
~
.~".~¢
-
#
f
4''
Ad...
m
.,,.-2
-
•
_% r
)
.,
.,
..
"~.
-;~
-
:
... D
• d
..
~: ~
-~ e
.. •
,.
•
.,'.
•
.
..
•
] ~-
/r
B
'l
III
.
-
• oo i
~
.
•
~
P
•
e L.
i,
z!~ ' 't't ° •
m
o.
P e
!
.
.}
,
"., III t
• •
•
i~i
i
i ~ ~ ;*~
a
~ 0o~
~Q
i
eD
o.
•
i,i
alp.
E
.
~
~
F
245 TABLE 11 Cytoskeletal alterations in the hypothalamus in aging and AIzheimer's disease in relation to cell loss Staining: *, cytoplasmic staining and staining of normal, thin, beaded fibres; * *, decrease, but within the normal range for age and sex; - , no or negligible staining of Alzheimer changes (i.e., plaques, tangles, dystrophic neurites); +, only a very few Alzheimer changes; +, some Alzheimer changes; + +, moderate number of Alzheimer changes; + + +, large number of Alzheimer changes; + + + +, very large number of Alzheimer changes. The category in Table I! is based upon the median score of the individual cases. Cell loss: O, no cell loss; +, questionable/small cell loss; l , cell loss; J, ~,, strong cell loss; n.d., not determined. Antibodies: 60e, polycional against tangles; r-l, monoclonal against tau; 3-39, monoclonal against ubiquitin; AIz-50, monoclonai against abnormal tau (A68). Nuclei: NBM, nucleus basalis of Meynert; SON, supraoptic nucleus; PVN, paraventricular nucleus; NTL, nucleus tuberalis lateralis; SDN, sexually dimorphic nucleus; SCN, suprachiasmatic nucleus; TM, tuberomammillary nucleus. antibodies nuclei Control cases: NBM SON
PVN SDN SCN NTL TM
60e
60e (formic acid)
+
. . . . . .
~.-I (dephosph. )
5:
-
. .
. .
. . . .
.
.
. . . .
. . . .
3 - 30
A!z-50
cell loss in aging (refs.)
:t:
-
+ 0 0 J, ,I, 0 n.d.
. . . . .
*/-
(12, 13, 34) (20) (20) (25, 49, 51) (50, 52) (Kremer, unpubl.)
cell loss in Alzheimer (refs.) AIzheimer's disease cases: NBM + SON . PVN ±/SDN + SCN ± NTL +++ TM ++
+ +
.
+ +
. +/+ ± +++ ++
.
+
+ + +
+/± ± ++ +
*/+/+ + ++++ +++
. ~ + + ++ +
in the majority of the controls, and even in young cases, staining of some diffusely labeled cell bodies and of thin beaded fibers was observed in the paraventricular aucleus (PVN) with AIz-50 (Fig. 2A). In Alzhcimer patients, more neurons showed cell body staining, and in addition, thicker unbeaded dystrophic neurites were observed in the PVN (Fig. 2B). With the three other antibodies, no staining was observed in PVN neurons of controls, whereas in the Alzheimer patients, these three antibodies did reveal staining and dystrophic neurites in the PVN. The sexually dimorphic nucleus (SDN) of control subjects was not stained with the cytoskeletal markers, whereas in Alzheimer patients cytoplasmic staining and dystrophic neurites were observed. The most pronounced staining was obtained with Alz-50 (Fig. 3). In general, the suprachiasmatic nucleus (SCN) did not show any cytoskeletal staining in controls, with the
(12,34,59) (20) (20) **(51) ¢ (50, 52) 0 (29) 0 (11) 0 0
exception of sporadic dystrophic neurites in a few subjects (Nos. 81032 83179), whereas in Aizheimer patients, cell body staining and dystrophic neurites were clearly observed. However, in spite of the pronounced SCN changes in Alzheimer's disease, these alterations were only modest, except for one 64-year-old female Alzheimer patient (No. 87494.2), whose entire SCN was stained with ~'-1 and 3-39 but not with 60e or Alz-50 (Fig. 3). In the nucleus tuberalis lateralis (NTL) of controls, dystrophic neurites and a single stained cell body were observed only rarely. In contrast, in the NTL of Alzheimer patients, all four antibodies showed a large number of cells with cytoplasmic staining, although slightly fewer perikarya were stained with 3-39 (Fig. 2). The most remarkable feature was the extremely high density of dystrophic nearites, especially with Alz-50 staining (Fig. 2F). The latter antibody even enabled us
iii,-.
Fig. 3. A, B: the sexually dimorphic nucleus (SDN) of an Alzheimer patient (No. 8560) stained by (A) AIz-50 and (B) 3-39. Note the cytoplasmic staining and dystrophic neurites. For r-I staining of the SDN after dephosphorylation and 60e staining after formic acid treatment see 4s; C-F: the suprachiasmatic nucleus (SCN) of an Alzheimer patient (No. 87494.2), the borders of which are indicated by arrows. The SCN is weakly stained by (C) anti-vasopressin, for the localization of this nucleus (see reference 35; I!I, third ventricle), (D) r-l, (E) 3-39, and (F) AIz-50. Note the weak vasopressin staining (C) in the degenerating SCN of this Alzheimer patient and that the entire SCN can be delineated after r-I (D) or 3-39 (E) staining. However, the staining is much less pronounced in this case after AIz-50 staining (F). Bar -- 200/zm.
246 T A B L E 111
Cytoskeletai changes in NTL of indii,idual cases of controls and Aizheimer's disease patients in relation to age, post-mortem tissue and tissue ~¢ation inter~'als Age is measured in years, pro, postmortem interval, in h; ft, fixation time, in days; ND, not determined, in the tissue blocks of Nos. 87949.2 and 84283, no N T L was present. For an explanation o f the o t h e r abbreviations, see Table II.
age pm ft
60e
47 59 74 83 85 88 91
. . . . . . ±
60e ~'.1 3-39 (formic (dephosph.) acid)
Aiz-50
Controls." 87271 800271 81032 81064 82175 81100 83179
2g 24 13 22 16 11 48
39 183 48 42 44 35 28
. . . . . .
. . .
. .
. .
. .
. . .
. -
. .
±
-
. -
± ±
± ++ ++ +++ ++ ++ +++ ++ ++
+ + + ND +++ + +++ ++ +
+++ ++++ ++++ ++++ ++++ ++++ + ++++ ++++ ++++
Al:.heimer's disease: 86364.52 85005.0 86196.6 88.252 83170 8560 8484 84050 86004 86001
45 4 119 + ++ 56 24 48 + + + +++ 61 84 63 + + + +++ 66 3 30 + + + +++ 70 14 34 + + + +++ 70 18 34 + ++ 87 42 292 .-1: + 90 2 33 + + + +++ 90 3 38 + + + + + e7 5 40 + + ++
to distinguish the NTL in Alzheimer patients by the unaided eye. As in the NTL, in the tuberomammillary nucleus (TM) of controls immunostaining of perikarya and nzurites was rare. The AIzheimer brains, however, showed numerous staining cell bodies in all patients. The density and proportion of staining cell bodies appeared to be lower for the TM than for the NTL. Again, numerous Alz-50 staining neurites were observed. Neuritic staining with the other antibodies was markedly less. In contrast to the NTL, the TM did show neurofibriilary tangles in Palmgren's silver impregnation. Table Ill provides the individual data on the NTL the most intense staining structure of the hypothalamus of Alzheimer patients (cf. Table II). These data clearly indicate that a postmortem autolysis time of up to 84 h and a fixation time of up to 119 days are not critically important for staining with the four antibodies in the hypothalamus. Only in the case of the longest fixation time (292 days)was relatively little staining observed. It should be noted, however, that the classical neuropathological Alzheimer changes of this case (No. 8484) were also only moderate. The observation that Aiz-50 staining of the cortex of AIzheimer patients remained excellent for up to 471 days of formalin
fixation (Van de Nes, unpublished results), supports the idea that even longer fixation times than were used in the present study might not be critical. DISCUSSION
The four antibodies revealed basically similar patterns in staining of cell bodies and dystrophic neurites in the hypothalami of Alzheimer patients and hardly any staining in controls. The relative staining intensity of the four different antibodies varied per patient and per hypothalamic nucleus, but not in any systematic way. Dystrophic neurites were most conspicuous with Alz-50 staining, yet background staining was the lowest with this antibody. The specific pattern of cytoskeletal changes in nuclei such as the NBM and NTL might be of help in the neuropathological diagnosis of Alzheimer's disease. The NBM is the nucleus that has been studied most extensively so far in relation to its decreasing cholinergic output in Alzheimer's disease. There was full agreement between the presence of cytoplasmic staining and dystrophic neurites in the NBM of Alzheimer patients on the one hand and the clinical and neuropathological diagnosis on the other. In controls, hardly any NBM staining occurred. It would appear, therefore, that the NBM is a very suitable structure for the validation of the neuropathologicai diagnosis of AIzheimer's disease. Another such structure for diagnostic confirmation with these cytoskeletal markers seems to be the NTL, despite the fact that silver and thioflavine.S stains disclosed few neurofibrillary tangles or senile plaques 2'~, suggc~,tiug at first sight that the NTL is not involved in AIzheimer's disease. No neuronal loss was found in the NTL of AIzheimer patients 2~. Yet, immunocytochemistry with the four antibodies revealed a completely different picture. Alz50 showed remarkable staining of perikarya and an extremely dense network of dystrophic neurites in the Alzheimer patients and no such staining in controls. Alz-50 reactivity with dystrophic neurites was so abundant that the NTL of Alzheimer patients could be recognized by the unaided eye. The 60e and ~'-1 antibodies basically visualized the same number of positive neurons, but the staining of dystrophic neurites was less conspicuous and background activity was higher than with Alz-50. The distinction between Alzheimer patients and controls could easily be made on the basis of the NTL staining with each of the four antibodies. An exception was a 90-year-old female control (81.033) in another study, who showed considerably more Alz-50 staining in the NTL than in the NBM. The peculiar Alz-50-stained changes in the NTL that are seen in
247 Alzheimer's disease are also present in Down's syndrome, but are absent in a number of conditions, such as Huntington's disease, Parkinson's disease and AIDS dementia (unpublished results), suggesting the disease specificity of this staining. The pattern of neuronal alterations in the TM was essentially that of the NBM, with immunostaining of perikarya and neurites, but a smaller proportion of the neurons seem to be affected. Precisely because of this difference, the alterations in the immunostained sections were less spectacular than in the NTL. As in the NBM, this pattern of severe Alzheimer-related changes may be related to the existence of extensive cortical projections 43. In the NBM and SCN of Alzhein~er patients, structures known to suffer from cell loss in this disease 12'33'34'5°'52'59 cytoskeletal alterations have been observed invariably. In contrast, the SON does not lose neurons 2° and, apparently, no cytoskeletal staining takes place. Thus, it might be concluded that cytoskeletai staining is associated with imminent neuronal death. However, results from other hypothalamic nuclei make this hypothesis untenable. Contrary to what is generally reported in the literature t9'61 but in agreement with the findings by Byne et al. H~,Alz-50 did show activity in control brain sections, i.e., in the PVN. Although the PVN does not lose cells in aging or Alzheimer's disease 2°, some diffuse cytoplasmic staining and staining of thin beaded fibers were observed in this nucleus with Alz-50 in controls, and more cytoplasmic staining and dystrophic neurites were observed in the PVN of Alzheimer patients. Also, the abundant staining of all cytoskeletal markers in the NTL of Alzheimer patients related to its continuing normal complement of neurons and the absence of neurofibrillary tangles shows that such cytoskeletal staining does not necessarily indicate impending neuronal death or the imminent formation of tangles. Unlike in hypothalamic nuclei such as the NTL, in the neocortex and especially in the hippocampus of Alzheimer patients many neurons with neurofibrillary tangles seem to degenerate, leaving behind ghost tangles. It thus appears that neurons in hypothalamic nuclei might have a higher threshold and that they impede the progression of eytoskeletal changes to the formation of tangles and, ultimately, cell death. However, even these initial stages of cytoskeletal changes might have functional consequences in the affected individuals. The NTL cytoskeletal changes may correlate with disturbances in eating behavior and metabolism. In the PVN of Alzheimer patients, in addition to the cytoplasmic staining and the visualization of beaded
fibers as found by Aiz-50 staining in controls, thick unbeaded dystrophic neurites and tangles were observed. The difference in staining between controls and Alzheimer patients might be due to the presence of different antigens for the same antibody (cf. refs. 31,36). This idea is reinforced by the finding l° that Alz-50 immunoreactivity was associated with vesicles in controls and with filaments and vesicles in Alzheimer patients and by the observation that the PVN staining in controls was obtained only with Alz-50, and not with the other three antibodies, whereas all four antibodies showed cytoskeletal changes in Alzheimer's disease. As a similar PVN staining was not obtained with anti-~- in controls it seems unlikely that the two antibodies recognize the same type of native ~-. Whatever the chemical background might be, the two types of Alz-50 staining should be clearly distinguished if this marker is used for the histopathological diagnosis of Alzheimer's disease. Some positive cell bodies and dystrophic neurites were found with 60e, ~--1, 3-39 and Alz-50 in the SDN of Alzheimer patients, whereas controls were negative for all four antibodies. Because the SDN cell numbers decrease in a similar way in Alzheimer patients and controls, it should again be concluded that cytoskeletai staining in the hypothalamus cannot be related in a simple way to cell death in the SDN. A thiofiavine-S study has also revealed that the appearance of tangles does ilot coincide with neuronal death in the locus coeruleus and NBM 5. In contrast, in the cerebral cortex and hippocampus, a significant negative correlation was observed between tangle counts and neuronal counts 5. As that study was based on the measurement of neuronal densities and not on total cell counts, additional morphometry is needed to establish this difference firmly as one between cortical and subcortical structures in Alzheimer's disease (cf. refs. 12,53). Neuritic plaques were not generally observed in the hypothalamic nuclei of the present study. Nonetheless, the SCN loses 75% of its cells during the Alzheimer process. In the NTL, only a few non-neuritic plaques are present that are reactive with SP28 (a rabbit polyclonal antibody against the 28 residue synthetic peptide which is homologous to the NH2-terminal region of /3-protein) 29, whereas the distribution of Alz-50 positive dystrophic neurites is extremely dense. Apparently, the presence of plaques is not a prerequisite for cell death in Alzheimer's disease or in the initiation of the hypothalamic cytoskeletal changes. Cytoskeletal changes may also occur without the presence of diffuse plaques. Regarding the hypotheses mentioned in the intro-
248 duction on the distribution of affected neurons, several remarks can be made. The very strong AIz-50 staining in the NTL, from which no direct cortical connections are known, raises doubts about the idea that cytoskeletal changes would be restricted to cortical areas. The same can be said about the SCN and SDN staining observed in AIzheimer patients whereas, in addition, these nuclei are not considered to be directly connected to the cortex. In more general terms, the ~- and ubiquitin staining in various hypothalamic structures indicates that the hypothesis that Alzheimer's disease primarily affects areas that have been acquired recently in phylogenesis ~q is not tenable. The present study shows that the cytoskeletal changes in AIzheimer's disease does not necessarily precede tangle formation. Nor can it be regarded as a sign of impending cell death, in addition, a clear distinction should be made between cells and fibers that stain in controls and structures that stain in Alzheimer's disease. Moreover, it has been shown that Alzheimer's disease co-evolves with a specific pattern of hypothalamic r and ubiquitin staining. Neuropathological diagnosis of mildly affected Aizheimer patients is still a major problem t'~. It seems therefore worthwhile to investigate whether this pattern of hypothalamic staining might be of help in the diagnosis of such cases of Alzhcimer's disease. Acknowh,dgements. Brain material was obtained from the Netherlands Brain Bank (Amsterdam, Th~ Netherlands), Neuropatholosy was pt~rformed by Dr, W, Kamphorst (Fret; University, Amsterdam) and Dr. D. Troost (Academic Medical Center of the University of Amsterdam). Alz,5(I was kindly donated by Abbott Laboratories (Abbott Park, IL, USA). r-I was I~encrously supplied by Dr. L. Binder (University of Alabama, Birmingham, AL, USA). The authors wish to thank Mr, B. Fisser, Ms. J.P.H.M. PIoegmakers and Ms, Y.-C. Tung for their technical assistance, Mr. G, Van der Meulen for his photographical work, Ms. M, Stoddard Marlow for copy-editing this manuscript and Ms. O. Pach and Mr. A. Janssen for their secretarial assistance. The study was carried out as part of the AMSTEL project with support from the Stimuleringsprogramma Gezondheidsonderzoek (SGO) of the Ministry of Science and Education in The Netherlands, New York State Office of Mental Retardations and Developmental Disabilities, NIH Grants AG 05892, AG 08076, NS 18105 and AG 04220, and a grant from the American Health Assistance Foundation, Rockville, MD, USA. D.F, Swaab and I. Grundke-lqbal were recipients of Senior Travel Fellowships of AIzheimer Society of Canada and AIzheimer's Association, USA, respectively, to participate in the Second International Conference on Alzheimer's Disease. Toronto. Canada, 1990. REFERENCES I Airaksinen, M.S., Paer, A., Paljiirvi, L., Reinikanen, K., Riekkihen, P., Suomalainen, R. and Panula, P., Histamine neurons in human hypothalamus: anatomy in normal and Alzheimer diseased brains, Neuroscience, 44 ( 1991) 465-481. 2 AI-Ghoul, W.M, and Miller, M.W., Transient expression of AIz-50 immunt~reactivity in developing rat neocortex: a marker for naturally occurring neuronal death?, Brain Res,, 481 (1989) 361-367. 3 Allen, S.J., Dawbarn, D, and Wilcock, G.K., Morphometric ira-
munochemical analysis of neurons in the nucleus basalis of Meynert in Alzheimer's disease, Brain lies,, 454 (1988) 275-281. 4 Bancher, C., Brunner, C., Lassmann, H., Budka, H., Jellinger, K., Wiche, G., Seitelberger, F., Grundke-lqbal, I., lqbal, K. and Wisniewski, H.M., Accumulation of abnormally phosphorylated ~" precedes the formation of neurofibrillary tangles in Alzheimer's disease, Brain Res., 477 (1989)90-99, 5 Bondareff, W., Mountjoy, C.Q., Roth, M. and Hauser, D.L., Neurofibriilary degeneration and neuronal loss in Alzheimer's disease, Neurobiol. Aging, 10 (1989) 709-715. 6 Braak, H. and Braak, E., The hypothalamus of the human adult: chiasmatic region, Anat. Embryol., 176 (1987)315-330. 7 Braak, H. and Braak, E., Neuropii threads occur in dendrites of tangle-bearing nerve cell, Neuropathol. Appl. Neurobiol., 14 (1988) 39 -44. 8 Braak, H. and Braak, E., Cortical and subcortical argyrophilic grains characterize a disease associated with adult onset dementia, Neuropathol. App! NeurobioL, 15 (1989) 13-26. 9 Braak, H., Braak, E., Grundke-lqbal, I. and lqbal, K., Occurrence of neuropil threads in the senile human brain and in AIzheimer's disease: a third location of paired helical filaments outside of neurofibrillary tangles and neuritic plaques, Neurosci. Lett., 65 (1986) 351-355. 10 Byne, W., Mattiace, L., Kress, Y. and Davies, P., Aiz-50 immunoreactivity in the hypothalamus of the normal and Alzheimer human and the rat, J. Comp. Neurol., 306 (1991) 602-612. I1 Chan-Palay, V. and Jentsch, B., Galinin tuberomammi!lary neurons in the hypothalamus in AIzheimer's and Parkinson's disease. In D.F. Swaab, M. Mirmiran, R. Ravid and F.W. Van Leeuwen (Eds.), The Human Hypothalamus in Health and Disease, Progress in Brain Research, Vol. 93, Elsevier, Amsterdam, in press. 12 Coleman, P.D. and Flood, D.G., Neuron numbers and dendritic extent in normal aging and Alzheimer's disease, Neurobiol. Aghzg, 8 (1987) 521-545. 13 De Lacalle, S., iraizoz, I. and Ma Gonzalo, L., Differential changes in cell size and number in topographic subdivisions of human basal nuclet.s in normal aging, Neuroscience, 43 (1991) 445 -456. 14 Diepen, R., Der Hypothalamus. in W. Bargmann (Ed.), Handbach der mikro,~'kopische Anatomie des Mens('hen. IV~ 7, Springer, Berlin, 1962, pp, 1-181. 15 Duykaerts, C., Dela~re, P., Hauw, J.-L, Abbamondi-Pinto, A.L., Sorbi, S., Allen, !,, Brion, J,P,, Flament-Durand, J., Duchen, L., Kauss, J,, Schlote, W., Lowe, J., Probst, A., Ravid, R., Swaab. D.F., Renkawek, K, and Tomlison, B., Rating of the lesions in senile dementia of the Alzheimer type: concordance between laboratories, J. Neurol, Sci., 97 (1990)295-323. 16 Farrer, L.A., Yu, P.-L., Anthropometric discrimination among affected, at risk, and not-at-risk individuals in families with Huntington's disease, Am. J. Med. Gen., 21 (1985) 307-316, 17 Franklin, C.A. and Karkeck, J., Weight loss and senile dementia in an institutionalized elderly population, J. Am. Diet Assoc, 89 (1989) 7Q0-792, 18 German, D,C., White Ill, C,L, and Sparkman, D.R., Alzheimer's disease: neurofibrillary tangles in nuclei that project to the cerebral cortex, Neuroscienee, 21 (1987)305-312. 19 Goedert, M., Spillantini, M.G. and Jakes, R,, Localization of the AIz-50 epitope in recombinant human microtubule-associated protein r, Neurosci, Lett,, 126 (1991) 149-154. 20 Goudsmit, E,, Herman, M,A,, Fliers, E, and Swaab, D.F., The supraoptic and paraventricular nuclei of the human hypothalamus in relation to sex, age and AIzheimer's disease, Neurobiol. Aging, I1 (1990)529-536, 21 Grundke-lqbal, i,, lqbal, K,, Tung, Y.-C., Quinlan, M., Wisniewski, H,M, and Binder, L.I., Abnormal phosphorylation of the microtubule-associated protein r (r) in AIzheimer cytoskeletal pathology, Prec. Natl, Acad. Sci. USA, 83 (1986) 4913-4917. 22 Grundke-lqbal, I., Iqbal, K., Tung, Y,-C, and Wisniewski H.M., AIzheimer paired helical filaments: immunochemical identification of polypeptides, Acta Neuropathoi. (Bed.), 62 (1984) 259-267. 23 Grundke-lqbal, l,, Vorbrodt, A,W., Iqbal, K., Tung, Y.-C., Wang, G,P. and Wisniewski, H,M,, Microtubule-associated polypeptides
249 r are altered in Alzheimer paired helical filaments, Mol. Brain Res., 4 (1988) 43-52. 24 Grundke-lqbal, !., Wang, G.P., lqbal, K., Tung, Y.-C. and Wisniewski, H.M., Alzheimer paired helical filaments: identification of polypeptides of monoclonal antibodies, Acta Neuropathol. (Beri.), 68 (1985) 279-283. 25 Hofman, M.A. and Swaab, D.F., The sexually dimorphic nucleus of the preoptic area in the human brain: a comparative morphometric study, J. Anat., 164 (1989) 55-72. 26 Ishii, T., Distribution of Alzheimer's neurofibrillary changes in the brain stem and hypothalamus of senile dementia, Acta Neuropathot, 6 (1966) 181-187. 27 Joachim, C.L., Morris, J.H., Selkoe, D.J. and Kosik, K.S., r epitopes are incorporated into a range of lesions in Alzheimer's disease, J. Neuropathol. Exp. Neurol., 6 (1987) 611-622. 28 Kremer, H.P.H., Roos, R.A.C., Dingian, G., Marani, E. and Bots, G. Th. A.M., Atrophy of the hypothalamic lateral tuberal nucleus in Huntington's disease, J. Neuropathol. Exp. Neurol., 49 (1990) 371-382. 29 Kremer, B., Swaab, D.F., Bots, G., Fisser, B., Ravid, R. and Roos, R., The hypothalamic lateral tuberal nucleus in Alzheimer's disease, Ann. Neurol., 29 (1991) 279-284. 30 Ksiezak-Reding, H., Binder, L.I. and Yen, S.-H., Immunochemical and biochemical characterization of r proteins in normal and Alzheimer's disease brains with Alz-50 and r-l, J. Biol. Chem., 263 (1988) 7948-7953. 3-I Ksicsak-Reding, H., Davies, P. and Yen, S.-H., AIz 50, a monoclonal antibody to Alzheimer's disease antigen, cross-reacts with r proteins from bovine a," ! normal human brain, J. Biol. Chem., 263 (1988) 7943-7947. 32 Lee, V.M.-Y., Balin, B.J., Otvos Jr., L. and Trojanowski, J.Q., A68: a major subunit of paired helical filaments and derivatized forms of normal r, Science, 251 (1991) 675-678. 33 Lowes-Hummel, P., Gertz, H.-J., Ferszt, R. and Cervos-Navarro, J., The basal nucleus of Meynert revised: the nerve cell number decreases with age, Arch. Gerontol. Geriatr., 8 (1989) 21-27. 34 Mann, D.M.A., Yates, P.O. and Marcyniuk, B., Changes in nerve cells of the nucleus basalis of Meynert in Alzheimer's disease and their relationship to ageing and to the accumulation of lipofuscin pigment, Mech. Ageing Dr,r,., 25 (1984) 189-204. 35 Masliah, E,, Terry, R.D., DeTeresa, R.M. and Hansen, L.A., lmmunohistochemical quantification of the synapse-related protein synaptophysin in AIzheimer disease, Neurosci. Lett., 103 (1989) 234-239, 36 Nukina, N,, Kosik, K.S. and Selkoe, D.J., The monoclonal antibody, AIz 50, recognizes r proteins in Alzheimer's disease brain, Neurosci. Lett., 87 (1988) 240-246. 37 Panda, P,, Airaksinen, M.S., Pirvola, U. and Kotilainen, E., Histamine containing neuronal systt,m in human brain, Neuro. science, 34 (1990) 129-132, 38 Perry, G., Mulvihill, P., Fried, V.A., Smith, H.T., Grundke-lqbal, !. and lqbal, K., lmmunochemical properties of ubiquitin conjugates in the paired helical filaments of Alzheimer disease, J. Neurochem., 52 (I 989) 1523-1528. 39 Rapoport, S.I., Hypothesis: Alzheimer's disease is a phylogenetic disease, Med. Hypotheses, 29 (1989) 147-150. 40 Rinne, J.O., Paljiirvi, L. and Rinne U.K., Neuronal size and density in the nucleus basalis of Meynert in Alzheimer's disease, J. Neurol. ScL, 79 (1987) 67-76. 41 Rusak, B. and Zucker, I., Neural regulation of circadian rhythms, Physiol. Rev., 59 (1979) 449-526. 42 Sanberg, P.R., Fibiger, H.C. and Mark, R.F., Body weight and dietary factors in Huntington's disease patients compared with matched controls, Med. J. Austr., 1 (1981) 407-409. 43 Saper, C.B., Organization of cerebral cortical afferent systems in the rat. II. Hypothalamocortieal projections, J. Comp. NeuroL, 237 (1985) 21-46. 44 Saper, C.B. and German, D.C., Hypothalamic pathology in Alzheimer's disease, Neurosci. Lett., 74 (1987) 364-370. 45 Simpson, W.A., Yates, C.M., Watts, A.G. and Fink, G., Congo red birefringent structures in the hypothalamus in senile demen-
46 47
48
49 50 51 52
53
54
55
56
57
58
59
60
61
62
tia of the Alzheimer type, Neuropathol. Appl. NeurobioL, 14 (1988) 381-393. Singh, S,, Mulley, G.P. and Losowsky, M.S. Why are AIzheimer patients thin?, Age Aging, 17 (1988) 21-28. Steinbusch, H.W.M. and Mulder, A.H., Localization and projections of histamine immunoreactive neurons in the central nervous system of the rat. in A. Bjfrklund, T. H6kfelt and M.J. Kuhar (Eds.), Handbook of Chemical Neuroanatomy, Vol. 3, Elsevier, Amsterdam, 1984, pp. 126-140. Swaab, D.F., Eikelenboom, P., Grundke-lqbal, I., lqbal, K., Kremer, H.P.H., Ravid, R. and Van de Nes, J.A.P., Cytoskeletal alterations in the hypothalamus during aging and in AIzheimer's disease are not necessarily a marker for impending cell death. In K. lqbal, D.R.C. McLachlan, B. Winblad and H.M. Wisniewski (Eds.), Alzheimer's Disease: Basic Mechanisms, Diagnosis and Therapeutic Strategies, John Wiley, New York, 1991, pp. 181-190. Swaab, D.F. and Fliers, E., A sexually dimorphic nucleus in the human brain, Science, 228 (1985) 1112-1115. Swaab, D.F., Fliers, E. and Partiman, T.S., The suprachiasmatic nucleus of the human brain in relation to sex, age and dementia, Brain Res., 342 (1985) 37-44. Swaab, D.F. and Hofman, M.A., Sexual differentiation of the human hypothalamus: ontogeny of the sexually dimorphic nucleus of the preoptic area, Dec. Brain. Res., 44 (1988) 314-318. Swaab, D.F., Roozendaal, B., Ravid, R., Velis, D.N., Gooren, L. and Williams, R.S., Suprachiasmatic nucleus in aging, AIzheimer's disease, transsexuality and Prader-Willi syndrome. In R. de Kloet, V.M. Wiegant and D. de Wied (Eds.), Neuropeptides and Brain Function. Progress in Brahl Research, Vol. 72, Elsevier, Amsterdam, 1987, pp. 301-310. Swaab, D.F. and Uylings, H.B.M., Comments on review by Coleman and Flood 'Neurons numbers and dendritic extent in normal aging and Alzheirner's disease', Density measures: parameters to avoid, Neurobiol. Aging, 8 (1987) 574-575. Ulfig, N. and Braak, H., Amyloid deposits and neurofibrillary changes in the hypothalamic tuberomammillary nucleus, J. Neural Transm. (P.-D. Sect.), ! (1984) 143. Ulrich, J., Probst, A., Anderton, B.H., Brion, J.P., Cytoskeletal immunohistochemistry of Alzheimer dementia and related diseases. In H.P. yon Hahn (Ed.), h~terdisciplinary Topics in Gerontology, VoL 2£ !!1. The Neuronal Cytoskeleton in Senile Dementia of AIzheimer Type and Other Types of Dementia, S. Karger, Basel, 1988° pp. 101 - 105. Van Gool, W.A. and Mirmiran, M., Aging and circadian rhythms. In D.F. Swaab, E. Fliers, M. Mirmiran, W.A. van Gool and F. van Haaren (Eds.), Aging of the Brain and Alzheimer's Disease. Progress in Brahz Re~earch, VoL 70, Elsevier, Amsterdam, 198f~, 255 -279. Wang, G.P., Grundke-lqbal, I., Kascsak, R.J., lqbal, K., Wisniewski, H.M., Alzheimer neurofibrillary tangles: monoclonal antibodies to inherent antigen(s), Acta NeuropathoL (Bet'/.), 62 (1984) 268-275. Watanabe, T., Taguchi, Y., Shiosaka, S., Tanaka, J., Kubota, H., Terano, Y., Tohyama, M. and Wada, H., Distribution of the histaminergic neuron system in the central nervous system of rats: a fluorescent immunohistochemical analysis with histidine decarbox'ylase as a marker, Brain Res., 295 (1984) 13-25. Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A,W., Coyle, J.T. and DeLong, M.R., Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain, Sci¢,nce, 215 (1982) 12371239. Witting, W., Kwa, I.H., Eikelenboom, P., Mirmiran, M. and Swaab, D.F., Alterations in the circadian rest-activity rhythm in aging and Alzheimer's disease, Biol. P~ychiatry, 27 (1990) 563-572. Wolozin, B.L., Pruchnicki, A., Dickson, D.W. and Davies, P., A neuronal antigen in the brains of Alzheimer patients, Science, 232 (1986) 648-650. Wolozin, B., Scicutella, A. and Davies, P., Reexpression of a developmentally regulated antigen in Down syndrome and Alzheimer disease, Proc. Natl. Acad. Sci. USA, 85 (1988) 62026206.