iiiii!i!iiiiii!iiiiiiii!!!iiii!iiiii!ii i!ii!i!!i!iiiiiiiiiiiiiiiiiiiiii!!iiii

Is amyloidogenesis during Alzheimer's disease due to an IL-1-/IL-6-mediated 'acute

phase response' in the brain? Peter Vandenabeele and Walter Fiers

Alzheimer's disease is characterized by the presence of amyloid plaques in the brain, which comprise mainly f3-amyloid peptide and oq-antichymo- trypsin. Here, Peter Vandenabeele and Walter Fiefs advance the hypothesis that this amyloidogenesis results from an IL-1-/IL-6-mediated acute phase reaction in the brain. They propose possible intracerebral sources of cytokines

and acute phase proteins in microglia, astrocytes, neurons and cells of the choroid plexus.

Alzheimer's disease (AD) is the commonest form of dementia. The disease is characterized neuro- pathologically by cortical atrophy, neuronal loss and the presence of neurofibrillary tangles, senile plaques and vascular deposits of [3-amyloid peptide 1 in various re- gions of the cerebral cm'tex and in the hippocampus 2. Recently, it has been demonstrated that some families with early-onset AD show linkage to chromosome 21 markers3; some particular cases were even correlated with mutations in the amyloid protein precursor (APP) gene 4. However, most AD cases, which belong to the late- onset form, do not correlate with chromosome 21 markers 3. This strongly suggests that other etiological factors may contribute to ~3-amyloid peptide aggregation in late-onset AD.

Major components of the amyloid deposits are the 40 residue ~-amyloid peptide 1, and, to a lesser extent, cq- antichymotrypsin s. The ~-amyloid peptide is derived from the transmembrane portion of a ubiquitously ex- pressed family of membrane-bound APP 6. The three forms of APP, of 770, 751 and 695 amino acids, are generated from one gene by alternative mRNA splicing. It is probable that these APPs are similar to heparan sulphate proteoglycan core protein 7, which is involved in the anchorage of the cell in the extracellular matrix. The second constituent, %-antichymotrypsin s, is a protease inhibitor belonging to the group of acute phase proteins, and is specifically upregulated in ~-amyloid peptide amyloidoses 8.

Theories of pathogenesis The restricted expression of %-antichymotrypsin is

consistent with the hypothesis that an altered balance between proteases causes abnormal degradation of the ubiquitously expressed APP, which leads to the forma- tion of the self-aggregating 40 residue f3-amyloid pep- tides s. This idea is further supported by the observation that the 751 and 770 forms of APP contain a 56 residue domain that is highly homologous to the Kunitz serine proteinase inhibitors 9-11. Furthermore, the secreted forms of 751 and 770 APP are identical to protease nexin II (Refs 12,13) and to platelet coagulation factor XIa inhibitor 14. These findings led some researchers to inves-

tigate whether AD is accompanied by an altered balance between the 751 and 770 (Kunitz insert-containing) forms and the 695 form of APP. Conflicting results were reported: Johnson et al.lS observed an increased 751:695 mRNA ratio by quantitative in situ hybridization in the hippocampal pyramidal neurons of AD patients com- pared with 'normal' individuals; Palmert et al. 16 found a relative increase in APP mRNA lacking the Kunitz do- main, while Koo et al. 17, using the ribonuclease H-S1 nuclease protection assay, did not find evidence for a differential expression of the transcripts.

In this article, we discuss the possibility that the in- crease in %-antichymotrypsin expression in the brain (and, perhaps, the increase in APP) may belong to a cerebral 'acute phase response', an idea first proposed by Abraham et al. 8,18 The increase in these proteins may resemble cytokine-mediated acute phase protein induc- tion in the liver and the finding that AD spreads around areas in the cortex that are interconnected by well- defined sets of corticocortical connections 19 suggests that APP is a response protein. Since interleukin 1 (IL-1) and tumor necrosis factor (TNF) are strong inducers of IL-6, which in turn is the major inducer of acute phase proteins in the liver (reviewed in Ref. 20), it is possible that an analogous IL-1-/IL-6-mediated enhancement of %- antichymotrypsin and APP production occurs in the brain. This induction may take place, at a low level, either continuously or discontinuously, over a very long period. The concept of a cerebral acute phase response, augment- ing the amyloidogenesis during AD, is based on several postulates that can be tested: (1) f3-amyloid peptides and %-antichymotrypsin, which are important plaque con- stituents, are produced in the brain; (2) the APP and %-antichymotrypsin genes are inducible by cytokines, specifically by IL-6; (3) the APP- and %-antichymo- trypsin-producing cells express the relevant cytokine receptors; (4) cytokines are produced by astrocytes and microglial cells.

13-Amyloid peptides and %-antichymotrypsin are of cerebral origin

Immunodetection methods have shown that APP is synthesized in neurons and is axonally transported to the

© 1991, Elsevier Science Publishers Ltd, UK. 0167-4919/91/$02.00

Immunology Today 2 1 7 Vol 12 No. 7 1991

ii!iiiiiiiiiiiiilji iiiiii iiiiii !iiii j iiiiii iiiiiiiiiiii!iiiiiiiiiiSiiill

neuronal endings 21 and in s i tu hybridization experiments have shown that neurons of the areas most heavily affected in the course of AD display enhanced local expression of APP 22-24. Similarly, the expression of oq- antichymotrypsin mRNA is increased in astrocytes situ- ated in the affected areas 8. Thus, the two major constitu- ents of the neuritic plaque appear to be produced locally and do not necessarily originate in an extracerebral source (this would also be unlikely in view of the strict blood-brain and blood-cerebrospinal fluid barriers). It is also possible that the choroid plexus is a source of intracerebral oq-antichymotrypsin 8 and APP. The choroid plexus is part of the blood-cerebrospinal fluid barrier in the ventricular cavities and maintains homeo- stasis of the cerebrospinal fluid. In fact, it has already been shown that this structure can secrete acute phase proteins and related polypeptides 2s.

IL-6 induces cq-antichymotrypsin production and augments APP synthesis

The synthesis of cxl-antichymotrvpsin by liver cells and hepatomas is highly potentiated by IL-6 and, to a lesser extent, by IL-1 (reviewed in Ref. 20). The possibility that these cytokines may also modulate the cq- antichymotrypsin or APP gene expression in the brain.is supported by the finding that IL-1 can increase the ex- pression of APP mRNA in cultured human endothelial cells 26. An additional argument suggesting that in- ducibility is IL-6 mediated is the fact that the promoter sequence of both APP and oq-antichymotrypsin genes contains a stretch of nucleotides that corresponds to the consensus sequence CTGGGAA/T (Ref. 27) found in the promoter region of most acute phase proteins that can be induced by IL-6 (reviewed in Ref. 20) (CTGGGAT in APP at position 1896 (Ref. 28) and CTGGGTA in otl-antichymotrypsin at position 327 (Ref. 29)). The presence of this consensus sequence is consistent with IL-6-mediated regulation of the APP and the cx 1- antichymotrypsin genes in neurons and astrocytes, re- spectively, or, as already mentioned above, in the choroid plexus.

The presence of cytokine receptors on astrocytes, neurons and cells from the choroid plexus

The expression of IL-1 receptors on astrocytes is shown by the IL-l-driven astroghosis 3° and by an IL-1- mediated induction of IL-6 secretion 31. The expression of IL-6 receptors on astrocytes is illustrated by the capacity of IL-6 to elicit the production of nerve growth factor (NGF) in these cells 31.

To our knowledge, there is no information on the expression of IL-6 receptors on neurons. However, the observation that IL-6 induces the formation of neurites in PC12 (Ref. 32), a pheochromocytoma cell line, indirectly suggests a potential role for IL-6 during neuronal differ- entiation and suggests the presence of IL-6 receptors on neurons. To date, there are no reports of the presence of IL-6 receptors on cells constituting the choroid plexus.

Intracerebral production of cytokines: astrocytes and microglial cells produce IL-1 and IL-6

In view of the concept of a strict barrier between blood and the brain, it seems appropriate to assume an intra-

cerebral origin of cytokines. Microglial cells 33 and astro- cytes 31 can secrete IL-1 and IL-6, respectively, and Griffin et al. 34 observed an increased expression of IL-1 in micro- glial cells in AD. IL-1 augments and stabilizes neuronal levels of NGF mRNA 3s but whether this effect is exerted directly or indirectly via the induction of IL-6 is not yet clear. IL-6, as previously mentioned, can also enhance the spontaneous secretion of NGF by astrocytes 31.

The combined actions of IL-1, IL-6 and NGF may be important in the increased amyloidogenesis in AD - IL-1 through the induction of APP and oq-antichymotrypsin synthesis and through its capacity to elicit IL-6 secretion in astrocytes and IL-6 as an inducer of APP, (x 1- antichymotrypsin and NGF production. NGF poten- tiates the neurotoxicity of [3-amyloid peptide 36 and may contribute to the appearance of so-called neuritic plaques. The fact that, in many cases of AD, basal forebrain cholinergic neurons are primarily affected 2 and that NGF is trophically active on these neurons 37 suggests an involvement of NGF in neuritic sprouting around the amyloid plaques.

What induces IL-1 or IL-6 secretion? An important etiological question remains: what

events in the brain elicit enhanced cytokine secretion by astrocytes or microglial cells? Several possibilities can be considered.

First, IL-6 may be produced locally after unspecified brain injuries and elicits a type of acute phase response in the choroid plexus to maintain the intracerebral homeo- stasis. Note that the systemic administration of inducers of acute phase response in the liver cannot elicit acute phase protein synthesis in the choroid plexus 2s.

Second, the IL-6 may result from immunological pro- cesses. For example, a blood-brain or a blood- cerebrospinal fluid barrier compromised by aging may, increasingly, allow access of the immune system to the central nervous system. Participation of the immune system in AD has been extensively discussed by McGeer et al. 38 Since microglial cells express major histocom- patibility complex (MHC) class II antigens and are antigen-presenting 39, and since activated astrocytes 4° and cells of the choroid plexus 41 also exhibit these fea- tures, it is possible that, during restricted recognition of foreign antigen by invading T cells, antigen-presenting microglial cells, astrocytes and cells of the choroid plexus are stimulated to produce IL-1 and IL-6.

Third, a variation on the above is that a specific autoimmune response, for example against [3-amyloid peptide plaque components, may initiate such a cerebral immune response. The autoreactive T cells may be gener- ated distally, enter the brain through a deteriorated blood-brain barrier or by antigen-specific extravasation, and cause restricted cytokine production. In this context it should be noted that Joachim et al. 42 reported, in AD patients and even in some healthy age-matched individ- uals, the formation of [3-amyloid peptide plaques in the skin.

Fourth, persistent or repeated (multiple) viral infections are possible causes of constitutive cerebral IL-6 pro- duction leading to amyloidogenesis.

It is important to note that f3-amyloid plaque depo- sition also occurs in nondiseased aged persons. It should

Immunology Today 218 Vol. 12 No. 7 1991

iiiiiiiiiiiiiii!iiiiiiiiiiii!iiiiiiii iiiiiiiiiiii!iiiiiiiiiiiiiiiiiii

therefore be stressed that the causal link between amyloidogenesis and the development of dementia is not at all clear, f3-Amyloid plaque formation, per se, may be insufficient to cause AD or perhaps the site or extent of plaque formation may be crucial.

Testing the hypothesis How can the hypothesis of an IL-6-mediated acute

phase response in the brain be verified? IL-6 concen- trations in the cerebrospinal fluid (CSF) of AD patients should be measured. Serum levels of IL-6 are not elevated in AD 43, but this does not exclude the possibility that elevated levels might be present in the CSF. Indeed, during acute viral or bacterial infections of the central nervous system, the high cerebral IL-6 levels are, in most cases, not reflected by increased levels in the serum 44.

Owing to the high incidence of late-onset AD, obvious correlations have not yet been found between the occur- rence of AD and other pathologies, even when ac- companied by temporarily high cerebral IL-6 levels (such as in viral meningitis and encephalitis). It is possible that a continuous expression of low, perhaps barely detect- able, IL-6 levels in the brain may be more harmful than an acute peak of IL-6 production. Since AD affects well- defined hippocampal and cortical regions 2,19, IL-6 pro- duction, and consequently APP secretion, may also ap- pear very localized, allowing no detection in the CSF. In situ hybridization or a polymerase chain reaction (PCR) amplification approach may clarify this. Transgenic mice bearing the IL-6 gene under control of a constitutive, brain-specific promoter may also be used to test the hypothesis of IL-6-mediated cerebral acute phase protein production and amyloidogenesis.

Research in the authors' laboratory was supported by the Belgian FGWO, IUAP, ASLK and the National Lottery. We thank E. Van Mechelen for fruitful discussion and M. Van- decasteele for editorial assistance.

Peter Vandenabeele and Walter Fiers are at the Laboratory o f Molecular Biology, State University, Ledeganckstraat 35, B-9000 Gent, Belgium.

References 1 Wong, C.W., Quaranta, V. and Glenner, G.G. (1985) Proc. Natl Acad. Sci. USA 82, 8729-8732 2 Hyman, B.T., Van Hoesen, G.W., Damasio, A.R. and Barnes, C.L. (1984) Science 225, 1168-1170 3 St George-Hyslop, P.H., Haines, J.L, Farrer, L.A. et al. (1990) Nature 347, 194-197 4 Goate, A., Chartier-Harlin, M-C., Mullan, M. et al. (1991) Nature 349, 704-706 5 Abraham, C.R., Selkoe, D.J. and Potter, H. (1988) Cell 52, 487-501 6 Kang, J., Lemaire, H-G., Unterbeck, A. et al. (1987) Nature 325, 733-736 7 Schubert, D., Schroeder, R., LaCorbiere, M. et al. (1988) Science 241,223-226 8 Abraham, C.R., Shirahama, T. and Potter, H. (1990) Neurobiol. Aging 11,123-129 9 Ponte, P., Gonzalez-DeWhitt, P., Schilling, J. et al. (1988) Nature 331,525-527 10 Tanzi, R.E., McClatchey, A.I., Lamperti, E.D. et al. (1988) Nature 331,528-530

11 Kitaguchi, N., Takahashi, Y., Tokushima, Y. et al. (1988) Nature 331,530-532 12 Oltersdorf, T., Fritz, L.C., Schenk, D.B. et al. (1989) Nature 341,144-147 13 Van Nostrand, W.E., Wagner, S.L., Suzuki, M. et al. (1989) Nature 341,546-549 14 Smith, R.P., Higuchi, D.A. and Broze, G.J., Jr (1990) Science 248, 1126-1128 15 Johnson, S.A., McNeill, T., Cordell, B. and Finch, C.E. (1990) Science 248, 854-857 16 Palmert, M.R., Golde, T.E., Cohen, M.L. et al. (1988) Science 241, 1080-1084 17 Koo, E.H., Sisodia, S.S., Cork, L.C. et al. (1990) Neuron 4, 97-104 18 Abraham, C.R. and Potter, H. (1989) Bio/Technology 7, 147-153 19 Pearson, R.C.A., Esiri, M.M., Hiorns, R.W. et al. (1985) Proc. Natl Acad. Sci. USA 82, 4531-4534 20 Heinrich, P.C., Castell, J.V. and Andus, T. (1990) Biochem. J. 265, 621-636 21 Koo, E.H., Sisodia, S.S., Archer, D.R. et al. (1990) Proc. Natl Acad. Sci. USA 87, 1561-1565 22 Cohen, M.L., Golde, T.E., Usiak, M.F. et al. (1988) Proc. Natl Acad. Sci. USA 85, 1227-1231 23 Higgins, G.A., Lewis, D.A., Bahmanyar, S. et al. (1988) Proc. Natl Acad. Sci. USA 85, 1297-1301 24 Martin, L.J., Sisodia, S.S., Koo, E.H. et al. (1991) Proc. Natl Acad. Sci. USA 88, 1461-1465 25 Schreiber, G., Tsykin, A., Aldfred, A.R. et al. (1989) Ann. N Y Acad. Sci. 557, 61-85 26 Goldgaber, D., Harris, H.W., Hla, T. et al. (1989) Proc. Natl Acad. Sci. USA 86, 7606-7610 27 Tsuchiya, Y., Hattori, M., Hayashida, K. et al. (1987) Gene 57, 73-80 28 Salbaum, J.M., Weidemann, A., Lemaire, H-G. et al. (1988) EMBO J. 7, 2807-2813 29 Bao, J-J., Sifers, R.N., Kidd, V.J. et al. (1987) Biochemistry 26, 7755-7759 30 Giulian, D. and Lachman, L.B. (1985) Science 228, 497-499 31 Frei, K., Malipiero, U.V., Leist, T.P. et al. (1989) Eur. J. Immunol. 19, 689-694 32 Satoh, T., Nakamura, S., Taga, T. et al. (1988) Mol. Cell. Biol. 8, 3546-3549 33 Giulian, D., Baker, T.J., Shih, L-C.N. and Lachman, L.B. (1986) J. Exp. Med. 164, 594-604 34 Griffin, W.S.T., Stanley, L.C., Ling, C. et al. (1989) Proc. NatI Acad. Sci. USA 86, 7611-7615 35 Heumann, R., Lindholm, D., Ban&low, C. et al. (1987) Proc. Natl Acad. Sci. USA 84, 8735-8739 36 Yankner, B.A., Caceres, A. and Duffy, L.K. (1990) Proc. Natl Acad. Sci. USA 87, 9020-9023 37 Whittemore, S.R. and Seiger, A. (1987) Brain Res. Rev. 434, 439-464 38 McGeer, P.L., Akiyama, H., Itagaki, S. et al. (1989) Can. J. Neurol. Sci. 16, 516-527 39 Frei, K. and Fontana, A. (1989) in Neuroimmune Networks: Physiology and Diseases (Goetzl, E.J. and Spector, N.H., eds), p. 127, Liss 40 Fontana, A., Fiers, W. and Wekerle, H. (1984) Nature 307, 273-276 41 Nathanson, J.A. and Chun, L.L.Y. (1989) Proc. Natl Acad. Sci. USA 86, 1684-1688 42 Joachim, C.L., Mori, H. and Selkoe, D. (1989) Nature 341,226-230 43 Van Duijn, C.M., Hofman, A. and Nagelkerken, L. (1990) Neurosci. Lett. 108, 350-354 44 Houssiau, F.A., Bukasa, K., Sindic, C.J. et al. (1988) Clin. Exp. Immunol. 71,320-323

~mmuno~ogy Today 219 ro t 12 No. 7 1991