Vol. 179, No. 3, 1997

September 30, 1991

BIOCHEMKAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 1247-1254

PEPTIDES HOMOLOGOUS TO THE AMYLOID PROTEIN OF ALZHEIMER’S DISEASE CONTAINING A GLUTAMINE FOR GLUTAMIC ACID

SUBSTITUTION HAVE ACCELERATED AMYLOID FIBRIL FORMATION

Thomas Wisniewski,* Jorge Ghiso and Blas Frangionel

Departments of Pathology and *Neurology, NY University Medical Center, 550 First Avenue TH 427, New York, New York 10016

Received August 5, 1991

/+Amyloid (AD) deposition in fibril form is the central event in a number of dis- eases, including Alzheimer’s disease (AD) and hereditary cerebral hemorrhage with amy- loidosis - Dutch type (HCHWA-D). Ag is produced by degradation of a larger amyloid precursor protein (APP). Recently a mutation in the APP gene has been found in HCHWA-D causing a glutamine for glutamic acid substitution at residue 22 of A,% The influence of this mutation on tibrillogenesis is not known, although it is clear that affected patients have accelerated cerebrovascular amyloid deposition, with disease symptoms early in life. We report the in vitro demonstration of accelerated tibril formation in a 28 residue synthetic peptide homologous to the Dutch variant Ag. Furthermore, in eight residue peptides homologous to AS the presence of the mutation is necessary for fibril formation. These findings provide a mechanism for accelerated amyloid formation in the Dutch vari- ant of APP. Q 1991 Academic Press, Inc.

Alzheimer’s disease protein or p-amyloid (Ag) deposition in the form of fibrils is a central pathological event in a number of diseases that include: Alzheimer’s disease (AD), hereditary cerebral angiopathy with amyloidosis - Dutch type (HCHWA-D), Down’s syn- drome (DS), sporadic cerebral amyloid angiopathy (CAA) and normal aging (1). Together these diseases have been termed the p-amyloid related diseases (BAD) (2). As is a small polypeptide of 39 to 42 amino acids that is derived from a larger amyloid precursor pro- tein (APP) (3). S everal APPs of different length have been found: APP695, APP714, APP751 and APP770 (3-8); these are derived by alternative splicing from a single gene on chromosome 21(3-6). The amyloid isolated from vessels in HCHWA-D and AD is 39 amino acids long, whereas the amyloid of senile plaque cores contains a further 3 residues at the

1 To whom reprint requests should be addressed.

Abbreviations; AD: Alzheimer’s disease, A,% B-amyloid, BAD: B-amyloid related disease, APP: amyloid precursor protein, CCA: sporadic cerebral amyloid angiopathy, DS: Down’s Syndrome, FAD: familial Alzheimer’s disease, HCHWA-D: hereditary cerebral hemor- rhage with amyloidosis, Dutch type, HCHWA-I: hereditary cerebral hemorrhage with amyloidosis, Icelandic type, SAA: serum amyloid A.

0006-291X/91 $1.50

1247 Copright 0 1991 by Academic Press, Inc.

All rights of reproduction in any form reserved.

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

carboxyl terminal end (9-13). Recently a mutation in the APP gene has been found in HCHWA-D, consisting of a cytosine to guanine transversion at nucleotide 1852 or codon 618 (using APP695 numbering) (14). This corresponds to a substitution of glutamine for glutamic acid at position 22 of & (15). HCHWA-D or familial cerebral amyloid angiopa- thy is an autosomal dominant disease first described in four families from coastal villages in Holland, where over 500 individuals are at risk for developing the disease (16-18). Patho- logically there is extensive amyloid deposition limited to the small and medium sized ves- sels of the leptomeninges and cerebral cortex, leading possibly to increased vessel friability, with resultant recurrent hemorrhages beginning at the ages of 45 to 60 (12). Parenchymal amyloid accumulations are also present; these resemble preamyloid and primitive plaques of AD and DS (19-22). The neuritic component as seen in classical plaques is a less promi- nent feature and neurofibrillary tangles are absent. Whether the amyloid in the primitive plaques of HCHWA-D is 39 or 42 amino acids long is unknown. Recently a mutation in the APP gene has also been found in some familial AD (FAD) cases of early onset (23,24). In these patients a cytosine for thymine mutation at codon 642 of APP695 causes an Be for APP695 causes an Ile for Val substitution. Although this amino acid substitution is not contained within the As sequence amyloid fibrils from these patients have not been isolat- ed so such a possibility can not be ruled out.

APP is an abundant, widely distributed protein in the body (25-30). The mechanism for fibril processing from APP in vessels walls and in the form of senile plaques remains to be determined. The APP molecule on hydrophobicity plots is strikingly hydrophilic, con- taining several alpha helical sequences in the first 595 residues (3), whereas the As 1-42 has an approximately 90% p-sheet structure and is therefore highly insoluble (9). Synthet- ic peptides B( l-28), B( l-42), fi( l-39), @( 12-28) and a( 14-28) have all been shown to form fibrils in solution that have the same properties and electron microscopic (EM) appear- ance as amyloid (31-35). However ~(21-28) (SP8) did not form recognizable structures by EM (33). We have extended prior observations by investigating the influence of the amino acid substitution found in HCHWA-D (glutamine for glutamic acid at position 22) on fibril formation. In addition we have labeled these fibrils with antibodies to specific por- tions of Afi using immunoelectron microscopic techniques.

METHODS

Synthetic Peptides: Synthetic peptides SP28, SP28Q, SP41, SP8 and SP8Q (Table 1) were synthesized

at the Center for the Analysis and Synthesis of Macromolecules (SUNY, Stony Brook) by solid phase techniques. Crude peptides were purified via HPLC using a p-Bondapak Cl8 column (0.78 x 30 cm) (Waters) and a linear gradient of 0 to 66% acetonitrile in 0.1% tri- fluoacetic acid at a flow rate of 2.0 ml/min. The column effluent was monitored by absorb- ance at 214 nm. Peptide sequences were corroborated by amino acid composition and sequence analysis.

Electron Microscopy: Synthetic peptides were dissolved in 100~1 of 15OmM NaCl at a concentration rang-

ing from 0.5 mg/ml to 10 mg/ml, pH 7.3, and incubated at room temperature from 1 hour to two weeks. The suspensions were prepared for electron microscopy as previously de- scribed (31).

1248

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

TABLE 1

SYNTHETIC PEPTIDES USED FOR FIBRIL FORMATION

SYNTHETIC AMINO ACID SEQUENCE PEPTIOE

597* 624

lv 2:

SP2B ~EAFRHDSGYEVHAOKLVFFAEDVG~NL

SP2BCI DEAFRHDSGYEVHAKLVFFAOQDVGSNK

SP8 AEDVGSNK

SP80 AQDVGSNK

* - APPsg5 Numbering.

Y - Ap Numbering.

Italic - Glutamine for glutamic acid substitution.

Immunoelectron Microscopy: This was done in a modification of prior published procedures (36). Fresh glow-

discharged nickel carbon coated grids were placed carbon coat side down onto 15~1 of the fibril solution for five minutes, followed by phosphate buffer saline (PBS) with 1:lO normal goat serum. The primary antibody was applied for 5 minutes at a dilution of 1:5. The following polyclonal antibodies were used: anti-SP28 (raised against the synthetic peptide SP28, homologous to residues 597-624 of APP695) and anti-SP41 (raised against the synthetic peptide SP41, homologous to residues 584-624 of APP695). Anti-SP28 recognizes the p-turn located at the C-terminus of SP28 (contained in segment 617-624 of APP695) and anti-SP41 reacts with the hydrophilic stretch 602-612 of APP695 (37). Monoclonal antibody 4G8 recognizes amino acid residues 17-24 of As (38). Incubations with normal rabbit serum or PBS alone were used as controls. The secondary antibody (Auroprobe EM 1OnM antibody colloid gold, Amersham Co.), either anti-mouse or anti-rabbit IgG were used at l:lO, with an incubation of 5 minutes. The grids were stained in 7% uranyl acetate for 5 minutes. Between each step the grids were washed in PBS.

RESULTS

Peptides SP28, SP28Q, and SP8Q all formed fibrils with the typical 8-10nm diameter of amyloid fibrils by EM (Figure 1). Fibril lengths varied from 100 to 2OOOmn. SP28 formed fibrils under these conditions after a 24 hour incubation, whereas SP28Q and SP8Q formed fibrils within one hour. SP8 did not form fibrils even after a two week incubation. Fibril formation was not dependent on concentration in the range used.

Immunogold labeling of SP28 and SP28Q fibrils was obtained with the anti-SP28, anti-SP41, and 4G8 antibodies. SP8Q fibrils were labeled with both the anti-SP28 and 4G8 antibodies (Figure lA, insert), but not the anti-SP41.

DISCUSSION

The recent finding of a point mutation within the APP gene in HCHWA-D (14), causing a single amino acid substitution of glutamine for glutamic acid at position 22 of

1249

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

i

._ 1,.

.~

b .

Figure 1. Electron micrographs showing: a) SP8Q fibrils (x22.500); insert (bottom right hand comer): Immunogold labeling of SP8Q fibrils with anti-W28 antibodies ( x~.OOO), b) SP8, which fails to form fibrils #22,500).

1250

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

As (15), has raised the question of how such a change may promote fibrillogenesis. Of interest, in another form of amyloidosis, hereditary cerebral hemorrhage with amyloidosis- Icelandic type (HCHWA-I) there is also a point mutation in the amyloid precursor protein resulting in the substitution of a single amino acid for glutamine (39-41). The histopatho- logical and clinical picture in HCHWA-I is similar to HCHWA-D (42), although the pre- cursor molecule in the former is a variant of cystatin C, an inhibitor of cysteine proteinase (39-41). Our findings suggest that this amino acid substitution has significant effects on the fibrillogenesis of A& The peptide SP8, which spans residues 21-28 of A& has in our, as in prior studies failed to make fibrils (33). On hydrophobicity plots, this portion of AS is hydrophilic and therefore is highly soluble in water or saline solutions. The single amino acid substitution at position 22 of Ag renders it able to easily form amyloid like fibrils. Furthermore the presence of glutamine at position 22 in SP28 greatly speeds the incuba- tion time for fibril formation.

As is a small internal aberrant degradation product of APP. Since an intron inter- rupts the DNA sequence of APP at position 613 (APP695 numbering), it is believed that As arises by abnormal proteolysis of its precursor and not by aberrant splicing (3-8,43). APP is an abundant protein; soluble derivatives have been found in the platelets, brain cytosol and cerebrospinal fluid (2530). Ag deposition can be driven by two alternative mechanisms: Proteolytic processing of the APP may be altered causing the production of an intrinsically insoluble self-aggregating B-peptide. Alternatively extrinsic features specific to the BAD may lead to amyloid deposition. In vitro studies using human embryonic kidney 293 cells transfected with cDNA encoding APP751 and APP695 suggest that normal APP cleavage occurs within the ,9-peptide sequence at residue 16 (44). However, the presence of other alternative but normal, minor pathways is also possible. The latter is supported by the recent finding of a 16kDa fragment of APP that contains the entire Ap sequence from leptomeningeal vessels of both normal and AD individuals (45). The presence of an amino acid substitution at position 22 of Ag could cause the APP to be processed abnormally, although this is unlikely as no such substitution has been found in AD. Alternatively the increased fibrillogenic nature of the peptide, as shown by our findings, may lead directly to amyloid formation. The presence of a mutation may be more important in determining the rate, site and age at which amyloid fibrils are deposited in a patient. It is likely that the mutation promotes accelerated catabolism of APP and that the levels of the mutation containing APP will be lower. This is known to be the case for mutation containing cystatin C in HCHWA-I (46), where a single amino acid is also replaced by glutamine. Recently a mutation in the APP gene of some early onset familial AD (23,24) has been described. The mutation occurs outside the Afi sequence and it remains to be determined whether this has an effect on the processing of APP or its inherent fibrillogenic properties. Clearly the dif- ferent sites of the mutation in the APP gene influences the clinicopathological manifesta- tions of the resulting disease since in HCHWA-D the amyloid is deposited mainly in blood vessels, causing strokes, and in FAD senile plaque formation causing dementia is more prominent.

It is likely that there will be multiple factors that lead to A#r deposition such as a mutation in the APP gene, post-translational modification and/or other extrinsic factors. Each of these may be associated with a particular clinical setting. Such a situation is seen in other amyloidosis, such as those associated with serum amyloid A (SAA) (47). In a minority

1251

Vol. 179, No. 3, 1991 BiOCHEMlCAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

of patients with chronic infections or inflammatory disease SAA is deposited as amyloid due to abnormal processing by the reticuloendothelial system (48). Alternatively in the experimental model of SAA amyloidosis, murine-experimental induced amyloidosis, an amyloidogenic isoform (SAA2) is found (49). The increased fibrillogenic properties of As with glutamine instead of glutamic acid is likely an important factor in the localization and early age at which HCHWA-D patients get amyloid deposition. Whether this will be an important factor in the other FAD remains to be determined.

Acknowledgments. This work was supported by the National Institutes of Health, Wash- ington, D.C., Grants AGO5891 and AG08721. We thank Fran Hitchcock for manuscript preparation.

REFERENCES

1. 2. 3.

4.

5.

6.

7.

8.

9.

10.

II.

12.

13.

14.

15.

Frangione B. (1989) Ann. Medicine 21,69-72. Castano E.M. and Frangione B. (1988) Lab. Invest. 58,122-132. Kang J, Lemaire H.G., Unterbeck A., Salbaum J.M., Masters C.L., Grzeschik K.H., Malthaup G., Beyreuther K, and Muller-Hill B. 1987) Nature 325,733-736. Robakis N.K., Ramakrishna N., Wolfe G., and Wisniewski H.M. (1987) Proc. Natl. Acad. Sci. (USA) 84,4190-4194. Goldgaber D., Lerman M.J., McBride O.W., Saffiotti U., and Gajdusek D.C. (1987) Science 235,877-880. Tanzi R.E., Gusella J.F., Watkins P.C., Bruns G.A.P., St. George-Hyslop P., Van Keuren M.C., Patterson D., Pajan S., Kurnit D.M., and Neve R.L. (1987) Science 235,880-884. Tanzi R.E., McClatchy A.I., Lamperti E.D., Villa-Kornaroff L., Gusella J.F., and Neve R.L. (1988) Nature 331,528-530. Golde T.E., Estuo S., Usiak M., Younkin L.H., and Yom&in S.G. (1990) Neuron 4, 253-267. Masters C.L., Simms G., Weinman N.A., Multhaup G., McDonald B.L., and Bey- reuther K. (1985) Proc. Natl. Acad. Sci. (USA) 82,4245-4249. Glenner G.G. and Wong C.W. (1984) Biochem. Biophys. Res. Commun. 122, 1131- 1135. Prelli F., Castano E.M., Glenner G.G. and Frangione B. (1988) J. Neurochem. 51, 648-65 1. Van Duinen S.G., Castano E.M., Prelli F., Bots G.T.AM., Luyendijk W., and Fran- gione B. (1987) Proc. Natl. Acad. Sci. (USA) 84,5991-5994. PrelIi F., Castano E.M., Van Duinen S.G., Bots G.T.AM., Luyendijk W., and Fran- gione B. (1988) Biochem. Biophys. Res. Commun. 151, 1150-l 155. Levy E., Carman M.D., Fernandez-Madrid I., Lieberburg I., Power M.D., Van Duinen S.G., Bots G.T.A.M., Luyendijk W., and Frangione B. (1990) Science 248, 1124-l 126. Prelli F., Levy E., Van Duinen S.G., Bots G.T.A.M., Luyendijk W., and Frangione B. (1990) Biochem. Biophys. Res. Commun. 170,301-307.

1252

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

16.

17.

18. 19.

20.

21.

22. 23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35. 36.

37. 38.

39.

Wattendorff A.R., Bots G.T.A.M., Went L.N., and Endtz L.J. (1982) J. Neurol. Sci. 55, 121-135. Luyendijk W., Bots G.T.A.M., Vegter-vander Vlis M., and Went L.N. (1986) Ned. Tijdschr. Geneeskd. 130, 1935-1940. Haan J., Algra P.R., and Roos R.A.C. (199) Arch. Neurol. 47,649-653. Giaccone G., Tagliavini F., Linoli G., Bourase C., Frigerio L., Frangione B., and Bugiani 0. (1989) Neurosci. Lett. 97,232-238. Tagliavini F., Giaccone G., Frangione B., and Bugiani 0. (1988) Neurosci. Lett. 93, 191-196. Yamaguchi H., Hirai S., Morimatsu M., Shoji M., and Ihara Y. (1988) Acta Neuro- path01 76,541-549. Joachim C.L., Morris J.H., and Selkoe D.J. (1989) Am J. Pathol. 135,309-319. Goate A., Chartierk-Harlin M.C., Brown J., Crawford F., Fidani L., Giuffra L., Haynes A., Irving N., James L., Mant R., Newton P., Rooke K., Roques P., Talbot C., Pericak-Vance M., Roses A., Williamson R., Rossor M., Gwen M., and Hardy J. (1991) Nature 349,704-706. Naruse S., Igarashi S., Aoki K., Kaneko K., IIhara K., Miyatake T., Kobayashi H., Inuquka T., Shimizu T., Kojima T., and Tsuji S. (1991) Lancet 337, 978-979. Ghiso J., Tagliavini F., Timmers W.F., and Frangione B. (1989) Biochem. Biophys. Res. Commun. X3,430-437. Palmert M.R., Podlisny M.B., Witker D.S., Oltersdorf T., Younkin L.H., Selkoe D.J., and You&in SG. (1989) Proc. Natl. Acad. Sci. (USA), 86: 6338-6342. Weidemann A., Konig G., Bunke D., Fischer P., Salbaum J.M., Masters C.L., and Beyreuther K. (1989) Cell 57, 115-126. Selkoe D.J., Mammen A., Podlisny M., Palmert M., Younkin S., and Schenk D. (1989) J. Neuropathol. Exp. Neurol. 48,377A. Van Nostrand W.E., Schmaier A.H., Farrow J.S., Cines D.B., and Cunningham D.D. (1991) Biochem. Biophys. Res. Commun. 175, 15-21. Miller D.L., Currie J.R., Iqbal K., Potempska A., and Styles J. (1989) Ann Med 21, 83-87. Castano E.M., Ghiso J., Prelli F., Gorevic P.D., Mighelli A., and Fragione B. (1986) Biochem. Biophys. Res. Commun. 141,782-789. Kirschner D.A., Inouye H., Duffy L.K., Sinclair A., Lind M., and Selkoe D.J. (1987) Proc. Natl. Acad. Sci. (USA) 84,6953-6957. Gorevic P.D., Castano E.M., Saema R., and Frangione B. (1987) Biochem. Biophys. Res. Commun. 147,854-862. Halverson K., Fraser P.E., Kirschner D.A., and Lansbury P.T. (1990) Biochem. 29,2639-2644. Barrow C.J. and Zagorski M.G. (1991) Science 253,179-182. Geuze H.J., Slot J.W., Van der Ley P.A., Scheffer R.C.T., and Griffith J.M. (1981) J. Cell. Biol. 89,653-665. Ghiso J., Wisniewski T., Vidai R., Rostagno A., and Frangione B. Submitted. Kim K.S., Wen G.Y., Bancher C., Chen C.M.J., Sapienza V.J., Hong H., and Wis- niewski H.M. (1990) Neurosci. Res. Commun. 7,113-122. Ghiso J., Jensson O., and Frangione B. (1986) Proc. Natl. Acad. Sci. (USA) 83, 2974-2978.

1253

Vol. 179, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

40.

41.

42.

43.

44.

45. 46.

47.

48.

49.

Ghiso J., Pons-Estel B., and Frangione B. (1986) Biochem. Biophys. Res. Commun. 136,548-554. Levy E., Lopez-Otin C., Ghiso J., Geltner D., and Frangione B. (1989) J. Exp. Med. 169, 1771-1778. Gudmundsson G., Hallgrimsson J., Jonasson T.A., and Bjarnason 0. (1972) Brain 95,387-4&t. Lemaire H.G., Salbaum J.M., Muithaup G., Kanz J., Bayney R.M., Unterbeck A., Beyreuther K., and Muller-Hill B. (1989) Nucleic Acid Res 17,517-522. Esch F.S., Keim P.S., Beattie E.C., Blather R.W., Culwell A.R., Oltersdorf T., McClure D., and Ward P.J. (1990) Science 248,1122-1124. Ghiso J. and Frangione B. Submitted. Grubb A., Jensson O., Gugmundsson G., Arnason A., Loefberg H., Malm J. (1989) N.Eng.J.Med.311: 1547-1549. Westermark G.T., Sletten K., and Westermark P. (1989) Stand J Immunol 30, 605- 613. Lavie G., Zucker-Franklin D., and Franklin E.C. (1978) J. Exp. Med. 148, 1020- 1031. Shiro M., Kawahara E., Nakanishi I., and Migita S. (1987) Scan. J. Immunol. 26, 709-716.

1254