232 Neuroscience Letters, 97 (1989) . . . . . . . Elsevier Scientific Publishers Ireland Ltd.

NSL 05881

Down patients: extracellular preamyloid deposits precede neuritic degeneration and senile plaques*

Giorgio Giaccone t , Fabrizio Tagliavini l, Giovanni Linoli l, Constantin Bouras 2, Luciano Frigerio l, Bias Frangione 3 and Orso Bugiani l

Ilstituto neurologico Carlo Besta, Milan (Italy), :Institutions Universitaires de Psychiatrie, Geneva (Switzerland) and SNew York University Medical Center, New York, N Y (U.S.A.)

(Received 31 August 1988: Revised version received 11 October 1988; Accepted 17 October 1988)

Ke.v word~." Amyloid; Senile plaque; Alzheimer's disease; Down's syndrome; Immunocytochemistry

Using anti-SP28 (a polyclonal antibody to a 28 residue synthetic peptide homologous to the NH2-termi- nal region of the Alzheimer amyloid fl-protein) to investigate the cerebral cortex of 6 Down patients aged 6 55 y, we found that, besides senile plaques and congophilic vessels, extracellular deposits unrelated to degenerating neurites, tangle-bearing neurons or congophilic vessels were labelled. These deposits were similar to the extracellular deposits previously observed in the cerebral cortex of Alzheimer patients and non-demented individuals. The material accumulated in the deposits did not react with Congo red, thiofla- vine S or, on some occasions, silver salts and therefore might have been constituted by fl-protein precursors lacking the molecular conformation of amyloid fibrils. Age-related analysis of the cortical lesions in Down patients suggested that such extracellutar deposits precede degenerating neurites and evolve into senile plaques.

Knowledge of the chemical properties of the cerebral amyloid of Alzheimer's dis- ease (AD) has progressed since the amino acidic sequences of the amyloid protein, referred to as amyloid fl-protein [7, l 1], and of its putative precursor [9] were estab- lished. Accordingly, relevant fragments of the fl-protein molecule were synthetized and related antibodies were raised. Such antibodies recognized the amyloid of senile (neuritic, mature and burned-out) plaques and congophilic angiopathy in AD and Down's syndrome [ 1, 2, 13, 21], and the amyloid of the cerebral angiopathy of heredi- tary (Dutch-type) and sporadic cerebral hemorrhages [4, 16]. Using anti-SP28 anti- body (a rabbit polyclonal antibody to the 28 residue synthetic peptide homologous to the NH2-terminal region of amyloid fl-protein [3]) in Alzheimer patients, we

*Preliminary reports were presented at the Fifth Paulo Foundation International Symposium on Patho- biology of Alzheimer's Disease (Hanaasari, Espoo, Finland) June 17-19, 1988 and the First International Conference on Alzheimer's Disease and Related Disorders (Las Vegas, Nevada, USA), September 6-9, 1988. Correspondence: G. Giaccone, lstituto neurologico Carlo Besta, 11 via Celoria, 20133 Milan, Italy.

0304-3940/89/$ 03.50 © 1989 Elsevier Scientific Publishers Ireland Ltd.

233

observed the labelling in the cerebral cortex not only of senile plaques (SP) and con- gophilic vessels, but also of extracellular deposits that differed from SP since they were neither congophilic or thioflavine S positive, nor related to degenerating neu- rites, tangle-bearing neurons or congophilic vessels [14]. We did not identify the ma- terial that was accumulated in the deposits; however, since it was recognized by the antibody raised to one fragment of amyloid fl-protein, but did not react with Congo red or thioflavine S, we speculated that it contained fl-protein precursors lacking the molecular conformation responsible for the tinctorial and optical properties of amyloid fibrils [6]. We suggested that such extracellular material could be tentatively named preamyloid and that preamyloid deposits could anticipate SP [14]. These con- clusions were speculative, however, since each Alzheimer patient was seen at only one point in time. The problem is now addressed by using Down's syndrome as a model for plaque morphogenesis, as from their 20s onward Down patients develop the full spectrum of Alzheimer changes, i.e. neurofibrillary tangles (NFT), congophilic angio- pathy and SP [10, 17, 19]. Assuming that Down patients of different ages provide the age-related sequence of morphological events leading to SP, we have designed the present study in order to find out whether extracellular preamyloid deposits pre- vail over SP in young adult Down patients and the opposite occurs in the aged.

Accordingly, we have examined the frontal and temporal cortex of 6 Down patients aged 6, 12, 31, 39, 46 and 55 y. Frontal specimens included Brodmann's areas 9-12 and 32, and temporal samples included areas 20, 21, 36, parahippocampal gyrus and hippocampus. Serial 7-/1m-thick sections from formalin-fixed, paraplast- embedded blocks were stained with haematoxylin-eosin, silver salts (Bodian's and Gallyas' methods), thioflavine S, Congo red, anti-PHF (paired helical filaments of AD), immunoperoxidase (IP) [8] and anti-SP28 IP. Additional series of 20-¢tm-thick frozen sections were cut from the same cortical areas for Globus' and Bielschowsky's silver and for anti-SP28 IP. Immunostaining was carried out as reported previously [14]. Finally, a quantitative evaluation of cortical amyloid changes was made on two adjacent paraplast-embedded sections stained with anti-SP28 IP and thioflavine S, respectively, and on two adjacent frozen sections stained with anti-SP28 IP and GIo- bus', respectively.

Results were as follows. No lesions were observed in Patients 1 and 2, aged 6 and 12, respectively. In Pat. 3, aged 31, anti-SP28 IP revealed amounts of extracellular preamyloid deposits distributed throughout the cortical layers (Fig. la). In frozen sections, a substantial quantity of such deposits appeared to be decorated with silver grains (Globus', Bielschowsky's) (Fig. lb). SP, NFT and congophilic angiopathy were absent. In frozen sections, Pats. 4 and 5, aged 39 and 46, respectively, had more preamyloid deposits decorated with silver. Pat. 5 had, in addition, some senile (neuri- tic, mature and burned-out) plaques (Fig. lc), but no tangle-bearing neurons or con- gophilic angiopathy, in frozen and embedded sections stained for amyloid fibrils (anti-SP28 IP, Globus', Bielschowsky's, thioflavine S, Congo red, Bodian's) or for degenerating neurites (anti-PHF IP, Globus', Bielschowsky's, Bodian's, Gallyas' and thioflavine S). Besides extracellular preamyloid deposits, the full spectrum of Alz- heimer changes was found in the last patient of the series, aged 55 (Fig. ld). Table

234

Fig. 1. a: extracellular preamyloid deposit (Pat. 3, age 31, anti-SP28 immunoperoxidase). This large depo- sit is in contact with cell bodies and small vessels, b: extracellular preamyloid deposit delicately decorated with silver grains (Pat. 4, age 39, Globus' silver impregnation), c: senile plaque, mature type (Pat. 5, age 46, anti-SP28 immunoperoxidase), d: neurofibrillary tangle-bearing neuron and the neuritic component of a senile plaque (Pat. 6, age 55, Gallyas' silver impregnation) (Normarsky optics, x 490).

I supplies the morphometric data showing that, as to the labelling of amyloid fl-pro- tein, anti-SP28 IP was more active than silver salts on frozen sections or thioflavine S on embedded sections. In the oldest patient, for example, anti-SP28 IP revealed 40 percent more extracellular preamyloid deposits and SP than silver salts in frozen sections, and 80 percent more than thioflavine S in embedded sections. Further, thioflavine S was more active than Congo red and Bodian's. Fig. 2 matches anti-SP28 IP against thioflavine S and illustrates the difference in quantity and intracortical dis- tribution between extracellular preamyloid deposits and SP. Fig. 3 illustrates the per- cent distribution, in each patient, of SP and extracellular preamyloid deposits decor- ated or not with silver. A time-related analysis of these results (see Fig. 3) showed an age-dependent, inverse correlation between extraceltular preamyloid deposits and SP in Down's patients. This correlation suggests that extraceltular preamyioid de- posits evolve into, rather than develop together with, SP. In our case series, in fact, extracellular preamyloid deposits were first detected in Pat. 3, aged 31, and found

235

TABLE l

DENSITY (number /mm 2) OF C O R T I C A L A M Y L O I D C H A N G E S ( P R E A M Y L O I D DEPOSITS A N D SENILE PLAQUES) D E T E C T E D BY ANTI-SP28, SILVER SALTS OR T H I O F L A V I N E S IN

AREA 9

Morphometry (number and diameter) of amyloid changes was carried out on adjacent cortical columns

using a semiautomatic image analyzer (Mop Videoplan Kont ron Zeiss) ( × 580, final magnification on the

monitor screen). In each patient, two 20-/~m-thick sections were cut from frozen blocks for anti-SP28 vs Globus" comparison, and lwo 7-pm-thick sections were cut from paraplast-embedded blocks for anti-SP28

vs thioflavine S comparison. In each section, a total cortical area of 12.2 to 13.0 mm 2 was scanned. Dia-

meter values (no! reported in the table) were employed to correct cortical amyloid change in densily

according to the size (Abercrombie's formula). In sections stained with anti-SP28, Globus ' and thioflavine S, diameter mean values ( _+ S.D.) were 31.4_+ 19.7, 132.0_+ 18.2 and 26.3 + 12.1 pm, and median values were

26.7, 26.9 and 25.9 l~m, respectively. Since these values were not significantly different, corrected densily values and raw density values were found to be similar.

Patient Age (y) anti-SP28 vs Globus ' anti-SP28 vs Thioflavine S

1 6 0 0 0 0

2 12 0 0 0 0

3 31 50.l 23.3 24.3 0

4 39 57.0 26.7 26.5 0

5 46 49.7 35.4 17.3 0.9

6 55 91.6 64.0 46.5 25.6

V -

Patient 1 : 6 yrs Patient 3:31 ¥rs Patient 4 :39 ¥rs Patient 5 : 4 6 yrs Patient 6 :55 yrs Patient 2:12 vrs

Fig. 2. Cerebral cortex, area 9. Intracortical distribution of extracellular preamyloid deposits and senile plaques. Compar ison between anti-SP28 immunoperoxidase (white bars) and thioflavine S (black bars)

illustrates the difference in quanti ty and distribution between extracellular preamyloid deposits and senile plaques labelled by the antibody, and senile plaques fluorescent after thioflavine S treatment. From Pat. 3, aged 31, onward, immunolabelled lesions appeared to be homogeneously distributed throughout the cortical layers (l VI), whereas thioflavine S-positive lesions first developed in Pat. 5, aged 46, and prevailed in the upper half of cortex (Pat. 6).

236

(O

2"~

~ . e -

o_ =

, w

'10

OJ O.

100

50

(

:i:!:i:ii

. . . . . . . j

, . . . . . . . - .

. . . . .

, . , . . .

]]!lllllllll:

!!!!!!!!!!!! 1 (6) 3 (31) 4 (39) 5 (46) 6 (55)

2 (12) Patient (yrs)

[ ] no extracellular preamyloid deposits or senile plaques [ ] extracellular preamyloid deposits, anti-SP28 positive [ ] extracellular preamyloid deposits, anti-SP28 positive and decorated with silver • senile plaques, anti-SP28 positive, argyrophilic, thioflavine S positive and congophlllc

Fig. 3. Percent distribution, in each patient, of extracellular preamyloid deposits, decorated or not with silver grains, and senile plaques. Counts were carried out in area 9.

to decrease progressively in older patients, whereas SP were first found in Pat. 5, aged 46, and seen at their highest density in the oldest patient. Further, our analysis sug- gested that the ability of extracellular preamyloid deposits to react with silver pre- cedes SP formation. In fact, a substantial quantity of preamyloid deposits was found to be decorated with silver grains. These deposits differed from SP in that they were undetectable with Congo red or thioflavine S and were not associated with degenerat- ing neurites. Extracellular preamyloid deposits decorated with silver grains could be identical with focal argyrophilic, non-congophilic lesions of the neuropil described in Alzheimer patients and regarded as putative SP antecedents [12]. Finally, our ana- lysis showed that degenerating neurites were strictly concomitant with thioflavine S- positive and congophilic amyloid fibrils. Degenerating neurites, in fact, were present in senile (neuritic and mature) plaques, but not in extracellular preamyloid deposits decorated or not with silver. These findings suggest that a relationship exists between amyloid fibrils and neurite degeneration, and that fl-protein, unless provided with the conformation responsible for Congo red and thioflavine S affinity, is unable to in- volve neurites. It is noteworthy, in this regard, that in SP most degenerating neurites, which develop after amyloid deposition [5], contain paired helical filaments [15], and that such filaments are responsible for perikaryonal tangles [18]. Since in Down patients SP appear much, up to 20 years, earlier than perikaryonal tangles [10, 20], neurofibrillary degeneration due to paired helical filaments might originate in periph- eral neurites that are close to amyloid fibrils and later reach the nerve cell bodies.

In conclusion, our analysis suggests that SP morphogenesis might be related to a sequence of structural modifications of the amyloid fl-protein precursors. Extracellu-

237

lar deposits of/#protein precursors lacking the molecular conformation of amyloid fibrils might develop first. Subsequently, fl-protein might become argyrophilic and finally acquire the conformation responsible for the tinctorial and optical properties of amyloid fibrils. Extracellular amyloid fibrils might cause neuritic degeneration.

Prof. G. Pilleri, Hirnanatomisches Institut, Universitfit Bern (CH) and Prof. A. Rovescalli, La Sacra Famiglia, Cesano Boscone (Italy) kindly provided brain tissue samples. Supported by the Italian Ministry of Health, Department of Social Services, and the U.S. National Institutes of Health (Grants AR01431 and AG05891 to B.F.).

I Allsop, D., Landon, M., Kidd, M., Lowe, J.S., Reynolds, G.P. and Gardner, A., Monoclonal antibo- dies raised against a subsequence of senile plaque core protein react with plaque cores, plaque peri- phery and cerebrovascular amyloid in Alzheimer's disease, Neurosci. Lett., 68 (1986) 252 256.

2 BoNn, S.A., Currie, J.R., Merz, P.A., Miller, D.L., Styles, J., Walker, W.A., Wen, G.Y. and Wis- niewski, H.M., The comparative immunoreactivities of brain amyloids in Alzheimer's disease and scra- pie, Acta Neuropathol., 74 (1987) 313 323.

3 Castafio, E.N., Ghiso, J., Prelli, F., Gorevic, P.D., Migheli, A. and Frangione, B., In vitro formation of amyloid fibrils from two synthetic pcptides of different lengths homologous to Alzheimer's disease //-protein, Biochem. Biophys. Res. Commun., 141 (1986) 782 789.

4 Coria, F., Castafio, E.N. and Frangione, B., Brain amyloid in normal aging and cerebral amyloid angiopathy is antigenically related to Alzheimer's disease/,C-protein, Am. J. Pathol., 129 (1987) 422 428.

5 Duyckaerts, C., Dela6re, P., Poulain, V., Brion, J.-P., and Hauw, J.-J., Does amyloid precede paired helical filaments in the senile plaques'? A study of 15 cases with graded intelleclual status in aging and Alzheimer's disease, Neurosci. Lelt., 91 (1988) 354~359.

6 Glenner, G.G., Amyloid deposits and amyloidosis, N. Engl. J. Med., 302 (1980) 1283 1292. 7 Glenner, G.G. and Wong, C.W., Alzheimer's disease: initial report of the purification and characteriza-

tion of a novel cerebrovascular amyloid protein, Biochem. Biophys. Res. Commun., 120 (1984) 885 890.

8 Gorcvic, P.D., Gofii, F., Pons-Estel, B., Alvarez, F., Peress, N.S. and Frangione, B., Isolation and partial characterization of neurofibrillary tangles and amyloid plaque core in Alzheimer's disease: immunohistological studies, J. Neuropathol. Exp. Neurol., 45 (1986) 647 664.

9 Kang, J., Lemaire, H.G., Unterbeck, A., Salbaum, J.N., Masters, C.L., Grzeschik, KH., Multhaup, G., Beyreuther, K. and M/iller-Hill, B., The precursor of Alzheimer's disease amyloid A4 protein re- sembles a cell-surface receptor, Nature (Lond.), 325 (1987) 733 736.

I0 Mann, D.M.A., Yates, P.O., Marcyniuk, B. and Ravindra, C.R., The topography of plaques and tan- gles in Down's syndrome patients of different ages, Neuropathol. Appl. Neurobiol., 12 (1986) 447 457.

I 1 Masters, C.L., Simms, G., Weinman, N.A., Mullhaup, G., McDonald, B.L. and Beyreuther, K., Amy- loid plaque core protein in Alzheimer disease and Down syndrome, Proc. Natl. Acad. Sci. U.S.A., S2 (1985) 4245 4249.

12 Probst, A., Brunnschweiler, H., Lautenschlager, C. and Ulrich, J., A special type of senile plaque, pos- sibly an initial stage, Acta Neuropathol., 74 (1987) 133 141.

13 Sclkoe, D.J., Bell, D.S., Podlisny, M.B., Price, D.L. and Cork, L.C., Conservation of brain amyloid proteins in aged mammals and humans with Alzheimer's disease, Science, 235 (1987) 873 877.

14 Tagliavini, F., Giaccone, G., Frangione, B. and Bugiani, O., Preamyloid deposits in the cerebral cortex of patients with Alzheimer's disease and nondemented individuals, Neurosci. Lett., 93 (1988) 191 196.

15 Terry, R.D. and Winsniewski, H.M., The ultrastructure of the neurofibrillary tangles and the senile plaques. In W. Wolstenholme and M. O'Connor (Eds.), CIBA Foundation Symposium on Alzheimer's Disease and Related Conditions, Churchill, London, 1970, pp. 145 168.

16 Van Duinen, S.G., Castafio, E.M, Prelli, F., Bots, G.T.A.B., Luyendijk, W. and Frangione, B., Heredi-

238

tary cerebral hemorrhage with amyloidosis in patients of Dutch origin is related to Alzheimer disease, Proc. Natl. Acad. Sci. U.S.A., 84 (1987) 5991 5994.

17 Whalley, L.J., The dementia of Down's syndrome and its relevance to aetiological studies of Alz- heimer's disease. In F.M. Sinex and C.R. Merril (Eds.), Down's syndrome and aging, Ann. N.Y. Acad Sci., 396 (1982) 39-53.

18 Wisniewski, H., Terry, R.D., and Hirano, A., Neurofibrillary pathology, J. Neuropathol. [:~xp. Neurol., 29 (1970) 163 176.

19 Wisniewski, K.E., Wisniewski, H.M. and Wen, G.Y., Occurrence of nearopathological changes and dementia of Alzheimer's disease in Down's syndrome, Ann. Neurol., 17 (1985) 278--282.

20 Wisniewski, K.E., Wisniewski, H.M., Rabe, A. and Wen, G.Y., Early formation of neuritic plaques in Down syndrome. Fifth Paulo Foundation International Symposium: Pathobiology of Alzheimer's disease, Hanaasari, Espoo, Finland, June 17 19, 1988, Abstracts: 49.

21 Wong, C.W., Quaranta, V. and Glenner, G.G., Neuritic plaques and cerebrovascular amyloid in Alz- heimer disease are antigenically related, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 8729-8732.