THE JOURNAL 0 1993 by The American Society for Biochemistry and Molecular Biology, Inc.

OF BtOLOGICAL CHEMISTRY Vol. 268, No. 1, Issue of January 5, p : sO8-!jlZ, 1993 rrnted m U. S. A.

Identification of the Alzheimer @/A4 Amyloid Precursor Protein in Clathrin-coated Vesicles Purified from PC12 Cells*

(Received for publication, May 22, 1992)

Christer NordstedtSO, Gregg L. Caporam$, Johan Thybergll, Samuel E. GandySII, and Paul Greengard$** From the $Laboratory of Molecular and Cellular Neuroscience, The Rockefeller University, New York, New York 10021, the §Department of Geriatric Medicine, Huddinge University Hospital, Karolinskn Institute, S-141 86 Huddinge, Sweden, the (IDepartment of Cell Bwlogy, Medical Nobel Institute, Karolinskn Institute, Box 60 400, S-104 01 Stockholm, Sweden, and the 11 Department of Neurology and Neuroscience, The New York Hospital-Cornel1 Medical Center, New York, New York 10021

The Alzheimer D/A4 amyloid precursor protein (APP) can be proteolytically processed by at least two separate pathways in PC12 cells: chloroquine-insen- sitive secretory cleavage and chloroquine-sensitive in- tracellular degradation, presumably in the endosomal/ lysosomal system. To further investigate the possibility of APP processing in the endosomal/lysommal system, we have examined whether APP is present in clathrin- coated vesicles (CCVs), which mediate the transport of many proteins to the endosomal compartment. Using a procedure derived from established protocols for the purification of CCVs from mammalian organs, we ob- tained from PC12 cells highly purified CCVs that dis- played the same morphological features as described for CCVs purified from other sources. The CCVs were enriched in full-length mature (fully post-translation- ally modified) forms of APP, as well as in the carboxyl- terminal APP fragment produced by the secretory cleavage pathway. As CCVs are known to be involved in only two intracellular pathways (trafficking from the plasma membrane to early endosomes, and from the trans-Golgi network to late endosomes/prelyso- somes), these findings provide direct evidence that APP is transported to the endosomalflysosomal system. Furthermore, the presence in CCVs of the carboxyl- terminal fragment resulting from APP secretory cleavage suggests that APP secretory processing oc- curs in a pre-CCV compartment.

The amyloid deposits seen in the brain parenchyma and cerebromeningeal blood vessels of patients with Alzheimer disease are composed primarily of a -40-amino acid peptide termed PIA4 (Glenner and Wong, 1984; Masters et al., 1985). @/A4 is derived by proteolytic cleavage of the 695 to 770- amino acid amyloid precursor protein (APP),’ an integral membrane protein with a receptor-like structure (Goldgaber

* This research was supported by the Swedish Medical Research Council (to C. N. and J. T.), the Axel and Margret Ax:son Johnsons Foundation (to C. N.), and National Institutes of Health Grant GM- 07739 (to G. L. C.), Clinical Investigator Development Award NS- 01095 (to S. E. G.), and Grants AG-09464 and AG-10491 (to P. G.), The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

** To whom correspondence should be addressed. Tel.: 212-327- 8780; Fax: 212-327-7888.

‘The abbreviations used are: APP, amyloid precursor protein; CCV, clathrin-coated vesicle; Mes, 2-(N-morpholino)ethanesulfonic acid.

et al., 1987; Kang et al., 1987; Tanzi et d , 1987, 1988; Robakis et al., 1987; Ponte et al., 1988; Kitaguchi et al., 1988). The physiological function of APP in brain is not known, but a secreted APP fragment has been identified as protease nexin 11, a potent inhibitor of serine proteases (Oltersdorf et al., 1989; Van Nostrand et al., 1989). The proteolytic cleavage that generates protease nexin I1 from APP occurs within the PIA4 domain, and would therefore preclude amyloidogenesis (Sisodia et al., 1990; Esch et al., 1990). However, this secretory route might only represent a minor metabolic pathway, since in unstimulated rat neuroendocrine PC12 cells only a small fraction of APP molecules is targeted for secretion, whereas the bulk of APP is degraded in a separate chloroquine-sensi- tive compartment, presumably endosomes and/or lysosomes (Caporaso et al., 1992a). In addition, there is evidence that the endosomalllysosomal system might be the site for gener- ation of APP proteolytic fragments that preserve intact @/A4 and thus contribute to the pathology of Alzheimer disease (Golde et al., 1992). These studies, though, did not directly demonstrate the cellular trafficking of APP to the endosomal/ lysosomal system.

We have now investigated whether APP is present in clath- rin-coated vesicles (CCVs). CCVs are responsible for the trafficking of many proteins to the endosomal compartment, including the transport of plasma membrane receptors to early endosomes and of proteins destined for lysosomes from the trans-Golgi network to late endosomes/prelysosomes (for re- views, see Goldstein et al. (1985), Brodsky (19881, and Korn- feld and Mellman (1989)). While little is known about the mechanism(s) whereby integral membrane proteins are tar- geted to the population of CCVs that exit the tram-Golgi network (e.g. see Harter and Mellman (1992)), specific signals have been identified that are responsible for targeting cell- surface proteins to the CCVs that bud off the plasma mem- brane. Both the influenza virus hemagglutinin molecule and the cation-independent mannose 6-phosphate receptor are efficiently endocytosed only if a tyrosine residue is present at specific locations in their cytoplasmic domains (Lazarovits and Roth, 1988; Lobe1 et al., 1989). Endocytosis of the low- density lipoprotein receptor requires an asparagine-proline- X-tyrosine (NPXY, where X represents any amino acid) motif in its cytoplasmic domain (Chen et al., 19901, perhaps enabling its interaction with the assembly/adaptor proteins that are believed to mediate the interactions between internalized proteins and the clathrin cage of the CCV (Pearse, 1988). APP is among the numerous cell-surface proteins possessing an NPXY motif in their cytoplasmic domains, suggesting that it too might be targeted to and internalized via clathrin- coated pits (Chen et al., 1990; Gandy et aL, 1991). The iden-

608

APP in Clathrin-coated Vesicles 609

tification of APP in CCVs would provide direct evidence for the trafficking of APP to the endosomal/lysosomal system.

In this report, we describe the purification of CCVs from PC12 cells using a combination of protocols developed for the isolation of CCVs from mammalian organs. The CCV prepa- ration was characterized by SDS-polyacrylamide gel electro- phoresis, electron microscopy, and immunoblot analysis using as a marker the transferrin receptor, an internalized protein known to be present in CCVs (Turkewitz and Harrison, 1989). We show that full-length mature (fully post-translationally modified) APP and the carboxyl-terminal fragment resulting from APP secretory cleavage are enriched in CCVs.

EXPERIMENTAL PROCEDURES

Purification of CCVs-CCVs were prepared from undifferentiated PC12 cells grown in suspension using a combination of protocols described earlier (Pearse, 1976; Nandi et al., 1982; Campbell et al., 1984). Pelleted cells were stored at -70 “C prior to use. The cells (approximately 20 ml of packed cell volume) were homogenized at room temperature in 10 volumes of buffer A (0.1 M Mes, pH 6.5, 1 mM EGTA, 0.5 mM MgClZ, 0.02% NaN3, 7 mM 2-mercaptoethanol, 6 units/ml aprotinin, 10 pg/ml leupeptin, 1 pg/ml antipain, 1 pg/ml pepstatin) with a steel Dounce-type homogenizer. The homogenate was centrifuged at 20,000 X g for 30 min, and the supernatant was further centrifuged at 55,000 X g for 1 h. The pellet was then homogenized with a loose-fitting glass-Teflon homogenizer in 10 volumes of buffer A and mixed with an equal volume of buffer A containing 12.5% Ficoll400 (Pharmacia, Uppsala, Sweden) and 12.5% sucrose. This mixture was centrifuged at 38,000 X g for 40 min. The supernatant was diluted with 5 volumes of buffer A and centrifuged at 100,000 X g for 1 h. The pellet was homogenized in approximately 30 volumes of buffer A and centrifuged at 20,000 X g for 20 min. The supernatant was layered over a cushion of buffer A made with 100% DzO and containing 8% sucrose. The CCVs were pelleted by centrif- ugation at 80,000 X g for 2 h. This centrifugation was performed at 20 “C, whereas all other procedures were performed at 4 “C. The purified CCVs were further fractionated either by density-equilibrium centrifugation on 10-ml linear 2/9%-20/90% Ficoll400/Dz0 gradients in buffer A (80,000 X g for 15 h) as described (Turkewitz and Harrison, 1989) or by gel-filtration chromatography on a 1 X 50-cm Sephacryl S-1000 column (Pharmacia) as described (Campbell et aL, 1984).

Protein and Zmmunoblot Analyses-The protein content of CCV fractions obtained by gel-filtration chromatography was determined by spectrophotometric absorbance at a wavelength of 280 nm. The protein concentration of cell homogenates and of CCV fractions obtained by density-equilibrium centrifugation was determined by the method of Bradford (1976).

Proteins from cell homogenates and CCVs were separated on 4- 15% SDS-polyacrylamide gradient gels under reducing conditions (Laemmli, 1970) and transferred to nitrocellulose membranes (Tow- bin et al., 1979). The nitrocellulose blots were first blocked with Tris- buffered saline (150 mM NaC1, 50 mM Tris-HC1, pH 7.3) containing 0.05% Tween 20 (TBST) and then with TBST containing 5% nonfat milk (Blotto). Primary antibodies were made up in Blotto with 0.02% NaN3 and stored at 4 “C between experiments. Antibody 369A is an affinity-purified rabbit antibody prepared against a synthetic peptide corresponding to the cytoplasmic domain of human APP (residues 645-694 of APP-6) (Buxbaum et al., 1990). Antibody 22Cll is a murine monoclonal antibody prepared against the amino terminus of human APP (Weidemann et al., 1989). A murine monoclonal antibody against rat transferrin receptor was purchased from Chemicon Inter- national (Temecula, CA). After a 3-h incubation at room temperature with primary antibody, blots were washed several times with TBST. Blots were then incubated for 1 h with either donkey anti-rabbit or sheep anti-mouse secondary antibody coupled to horseradish peroxi- dase (Amersham Corp.) made up in Blotto. Following extensive washing with TBST, immunoreactivity was visualized with the en- hanced chemiluminescence system (Amersham) according to the manufacturer’s instructions, and blots were exposed to x-ray film. Total protein was visualized by staining the nitrocellulose blots with 0.1% Amido Black.

Electron Microscopy-Peak fractions of APP immunoreactivity eluted from the S-1000 column were examined by electron micros- copy. Small droplets of sample were placed on grids coated with a carbon-stabilized Formvar film and left for 30-60 s. After removal of

excess liquid, the adsorbed material was negatively stained with unbuffered 1% uranyl acetate for 1 min. Imaging was performed with a Jeol lOOCX electron microscope.

RESULTS

Highly purified CCVs were prepared from PC12 cells by differential centrifugation according to established methods (see “Experimental Procedures”). The CCVs were then frac- tionated according to differences in density or size, and ex- amined for possible co-distribution of clathrin and APP. This procedure allowed for the separation of minor contaminating structures from CCVs.

In these experiments, CCVs were separated according to density on a 10-ml linear density-equilibrium gradient con- sisting of 2/9%-20/90% Ficoll 400/D20. Twelve fractions of approximately 0.8 ml each were collected from the gradient. Equal volumes of these fractions were separated on 4-15% SDS-polyacrylamide gradient gels followed by transfer of the proteins to nitrocellulose membranes. The total protein was visualized by Amido Black staining of the nitrocellulose mem- branes (Fig. 1). Fractions 8-12 contained a protein pattern typical of CCV preparations: a large quantity of clathrin heavy chain (MI - 180,000) and smaller quantities of clathrin light chains (MI - 32,000) and assembly/adaptor proteins (MI - 100,000, 50,000, and 17,000) (Ahle et al., 1988).

When the fractions were analyzed by immunoblotting using an antibody against the carboxyl terminus of APP, three distinct APP species with MI -140,000, -120,000, and -14,000-15,000 were shown to co-purify with the CCVs (Fig. 1). An antibody against the amino-terminal end of APP recognized only the two higher molecular weight APP species (Fig. 1). Based on comparison of the APP immunoblot pattern seen with cell homogenates (see Fig. 4 and below) and previ- ously reported identification of the APP species present in PC12 cells (Weidemann et al., 1989; Buxbaum et al., 1990; Caporaso et al., 1992b), we assign the M, - 140,000 and 120,000 species as full-length mature (fully post-translation- ally modified) APP,sl/,,O and APPags, respectively, and the M, - 14,000-15,000 species as the carboxyl-terminal APP frag- ment resulting from normal APP secretory cleavage (Sisodia et al., 1990; Esch et al., 1990; Caporaso et al., 1992a).

As a marker for CCVs, we also examined the distribution in the fractions of transferrin receptor, a protein known to be present and concentrated in CCVs (Turkewitz and Harrison, 1989). An antibody against the rat transferrin receptor rec- ognized two proteins with M, - 190,000 and 95,000 (Fig. l ) , representing the dimeric and monomeric forms of transferrin receptor, respectively. Both transferrin receptor species had distribution patterns in the density gradient virtually identi- cal to those of clathrin, clathrin-associated proteins, and APP immunoreactivity.

In order to demonstrate that the APP immunoreactivity was indeed present in CCVs, and not in contaminating struc- tures with a density similar to that of CCVs, we also subjected the CCVs to gel-filtration chromatography on an S-1000 column, in which structures were separated according to size rather than density (Fig. 2). Measurement of protein content by spectrophotometric absorbance of the column eluate at a wavelength of 280 nm produced a broad peak centered at fractions 26 and 28, with a “shoulder” at fraction 32. The protein pattern in the fractions was virtually identical to the protein pattern obtained when CCVs were separated accord- ing to density (compare Fig. 2 with Fig. 1). However, a protein with MI - 150,000 and some smaller proteins that peaked in fraction 32 were clearly separated from the clathrin peak, and probably represent a smaller sized contaminating structure (Fig. 2). The APP species with M, - 140,000, 120,000, and

610 APP in Clathrin-coated Vesicles

0 . 1 2 4 ' ' ' ' ' ' ' ' ' 200-

97.4-

6 9 -

4 6 -

30- 21.5- 14.3-

200-

97.4-

21.5- 14.3-

200-

97.4-

200-

97.4-

to ta l Drotein

0.10 - j 0.08 - - 0 0.06 - 2 0.04 -

0.02 - 0 -

._ 0 n.

0 10 20 30 40

Fraction

200-

""I"

anti-APP C-terminus

- ""- 97.4-

69-

46-

30-

21.5- 14.3-

200-

anti-transferrin receptor

6 9 -

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 Fraction

FIG. 1. Co-purification of APP and transferrin receptor with CCVs fractionated by density-equilibrium centrifuga- tion. CCVs prepared from PC12 cells by differential centrifugation (see "Experimental Procedures") were loaded on a linear 2/9%-20/ 90% Ficoll 400/D20 gradient and centrifuged at 80,000 X g for 15 h. Proteins from equal volumes of each fraction ( I = top, I2 = bottom of gradient) were separated on 4-15% SDS-polyacrylamide gels and transferred to nitrocellulose membranes. Total protein was visualized by Amido Black staining, and revealed a large quantity of clathrin heavy chain (M, - 180,000) and smaller quantities of clathrin light chains (M, - 32,000) and assembly/adaptor proteins (M, - 100,000, 50,000, and 17,000). Immunoblotting with an antibody against the carboxyl terminus of APP revealed proteins of M, - 140,000,120,000, and 14,000-15,000 corresponding to full-length mature APP711/770 and APPsss, and the carboxyl-terminal APP fragment resulting from secretory processing, respectively (see "Results"). Antibody against the amino terminus of APP recognized only the M, - 140,000 and 120,000 proteins. Antibody against rat transferrin receptor revealed proteins of M, - 190,000 and 95,000, corresponding to the dimeric and monomeric forms of the molecule, respectively (the weaker protein band below the dimeric transferrin receptor might represent antibody cross-reactivity with clathrin heavy chain). Approximate M, (~10') values are indicated.

14,000-15,000 that co-purified with CCVs in the density gra- dient also co-purified with CCVs on the S-1000 column (Fig. Z), supporting the conclusion that APP was present in CCVs.

To determine the purity of the CCV preparation, CCV fractions eluted from the S-1000 column were examined by electron microscopy (Fig. 3). The morphology of the purified structures was identical to that of CCVs prepared from other sources (e.g. see Brodsky (1988)). The preparation was found almost exclusively to contain CCVs with well-defined mem- brane vesicles (Fig. 3, closed arrows) and vesicle-free clathrin cages (Fig. 3, open arrows). Very few contaminating structures were seen, with approximately 99% of the structures (as determined by manual counting) representing CCVs or clath- rin cages. The fine granular pattern seen in the background of the electron micrograph in Fig. 3 is a staining artifact that

total protein

97.4-

anti-APP C-terminus 21.5-

14.3-

H 18 20 22 24 26 28 30 32 34 36 38 4 0 Fraction

FIG. 2. Co-purification of APP with CCVs fractionated by gel-filtration chromatography. CCVs prepared from PC12 cells by differential centrifugation were eluted from an S-1000 column and the protein content of each fraction was monitored by spectrophoto- metric absorbance a t a wavelength of 280 nm. Fractions correspond- ing to the absorbance peak (fractions 18-40) were concentrated by centrifugation at 100,000 x g for 90 min. Proteins from a total cell homogenate ( l a n e H) or from equal volumes of each fraction were separated on 4-15% SDS-polyacrylamide gels and transferred to nitrocellulose membranes. Total protein was visualized by Amido Black staining, and revealed a protein pattern very similar to that seen in CCVs separated by density-equilibrium centrifugation (see Fig. 1) with CCVs being most concentrated in fractions 26 and 28. However, a protein of M, - 150,000 (along with several lower molec- ular mass proteins) that peaked in fraction 32 probably represents a smaller sized contaminating structure. Immunoblotting with an an- tibody against the carboxyl terminus of APP indicated that APP immunoreactivity was concentrated in the structures centered at fractions 24-28. Approximate M, (x103) values are indicated.

also is obtained with buffer alone (data not shown). Due to the presence of Ficoll 400 in the fractions from the density- equilibrium gradient, these CCVs could not be examined by electron microscopy.

To assess whether APP was enriched in CCVs compared to the total cell homogenates, we compared the APP content in equal amounts of protein from cell homogenates and CCVs (Fig. 4). Equal amounts of protein from a total cell homoge- nate and from the peak fractions of CCVs purified on density- equilibrium gradients were separated on 4-15% SDS-poly- acrylamide gradient gels, transferred to nitrocellulose mem- branes, and analyzed by Amido Black total protein staining and by immunoblotting with antibodies against the carboxyl or amino terminus of APP and antibody against the transfer- rin receptor. Immunoblot analysis with the antibody against the carboxyl terminus of APP revealed in cell homogenates immunoreactive protein bands of M, - 140,000, 120,000, 115,000, 105,000, and 14,000-15,000 corresponding to mature APP751,770, mature APPeg5, immature APP751,770, immature

APP in Clathrin-coated Vesicles 611

FIG. 3. Electron micrograph of CCVs fractionated by gel- filtration chromatography. CCVs eluted in fraction 25 from an S- 1000 column (see Fig. 2) were negatively stained and viewed by electron microscopy. Both CCVs with membrane vesicles (closed arrows) and vesicle-free clathrin cages (open arrows) can be seen. Scale bar represents 100 nm.

APPsg5, and the APP carboxyl-terminal fragment as earlier identified (Weidemann et al., 1989; Buxbaum et al., 1990; Caporaso et al., 1992b). Both mature APP holoprotein and the APP carboxyl-terminal fragment were enriched in the CCVs compared to the total cell homogenate (Fig. 4). Imma- ture APP holoprotein was not detected in CCVs (Figs. 2 and 4) as might be expected, since CCVs are formed only at post- Golgi locations, distal to the site where APP maturation is believed to occur (Caporaso et al., 1992a). Immunoblotting with the antibody against the amino terminus of APP con- firmed that mature APP holoprotein was enriched in CCVs. However, small amounts of proteins with M, - 125,000 and 110,000, corresponding to secreted APP751/770 (protease nexin 11) and APPeg5 (Caporaso et al., 1992b), were also detectable in CCVs. The much greater quantity of transferrin receptor in the CCVs compared to the total cell homogenate was consistent with enrichment of transferrin receptor in CCVs as previously reported (Turkewitz and Harrison, 1989).

DISCUSSION

In this report, we have shown that CCVs prepared from PC12 cells are enriched in both full-length mature APP and the carboxyl-terminal fragment generated by APP secretory cleavage, indicating that these proteins are targeted to the endosomal compartment. These data are consistent with our earlier findings in PC12 cells that degradation of mature full- length APP molecules and the carboxyl-terminal fragment can be inhibited by chloroquine, which implies that acidic organelles such as endosomes or lysosomes are involved in their metabolism (Caporaso et al., 1992a). In addition, other groups have found evidence suggesting that APP is present in lysosomes (Benowitz et al., 1989), that APP metabolic processing occurs in endosomes or lysosomes (Cole et al., 1989), and that generation of amyloidogenic APP fragments might occur in endosomes or lysosomes (Golde et al., 1992). The present study provides direct evidence that APP utilizes a specific intracellular route that targets proteins to the endosomal/lysosomal system.

We were not able, with the procedures used, to distinguish between the two known populations of CCVs, i.e. those that transport proteins to early endosomes from the cell surface and those that transport proteins to late endosomes/prelyso- somes from the trans-Golgi network. Therefore, it was not possible to determine whether APP is targeted to the endo- soma1 system from the plasma membrane or from the trans- Golgi network. A plasma membrane origin for APP in CCVs

2 00-

97.4-

69-

46-

3 0-

21.5- 14.3-

v) = C .- E

v) 3 K .- E

H V H V H V H V FIG. 4. Comparison of total protein, APP immunoreactiv-

ity, and transferrin receptor immunoreactivity in cell homog- enates (H) and CCVs fractionated by density-equilibrium cen- trifugation ( V ) . Equal amounts of protein (20 pgllane) were sepa- rated by 4-15% SDS-polyacrylamide gradient gels, transferred to nitrocellulose membranes, and either stained with Amido Black (total protein) or immunoblotted with antibodies against the carboxyl ter- minus of APP, the amino terminus of APP, or the rat transferrin receptor. The predominant APP species in cell homogenates are the immature and mature APP7al/770 isoforms (M, - 115,000 and 140,000, respectively) as well as the carboxyl-terminal APP fragment resulting from secretory cleavage (M, - 14,000-15,000). Immature APPma is barely visible (M, - 105,000) in this film exposure, but mature APPm5 can be seen as a faintly staining band (M, - 120,000) above immature APP751/,70. CCVs are enriched in mature APP isoforms and the carboxyl-terminal APP fragment, with mature APP751/,7~ being the major species. Due to the large amount of clathrin heavy chains present, the APP bands have been pushed slightly downward. None- theless, it can be seen that both of the full-length APP species in CCVs migrate more slowly than either immature APP species in cell homogenates. The M, - 40,000 protein seen by immunoblotting with antibody against the carboxyl terminus of APP is a nonspecific cross- reacting protein. The M, - 97,000 protein seen with antibody against the amino terminus of APP probably represents a truncated form of APP,51/,70, as it is also recognized with an antibody against a domain specific for APPT,,~~~. Approximate M, (X103) values are indicated.

seems more likely at present, although, since some APP has been shown to be present on the cell surface (Weidemann et al., 1989) and since APP contains in its cytoplasmic domain an NPXY motif, suggesting that it may be targeted to clath- rin-coated pits (Chen et al., 1990).

The presence in CCVs of the carboxyl-terminal APP frag- ment produced in the secretory pathway suggests that secre- tory processing occurs prior to the arrival of APP at the endosomal/lysosomal system. APP secretory cleavage might occur within constitutive secretory vesicles en route to the cell surface or while resident at the plasma membrane. Alter- natively, such cleavage might occur during transit from the trans-Golgi network to late endosomes/prelysosomes. Al- though we cannot exclude secretory processing within the

612 APP in Clathrin-coated Vesicles

trans-Golgi network, processing in the Golgi cisternae seems unlikely since secretory cleavage of APP is completely inhib- ited by treatment of PC12 cells with brefeldin A, a drug that prevents transport of proteins from the trans-Golgi to the trans-Golgi network (Caporaso et aL, 1992a). Finally, if secre- tory cleavage occurred within CCVs, one might expect to find an enrichment of amino-terminal APP fragments in CCVs. However, only a negligible amount of amino-terminal APP secretory products was detected in CCVs, and it is possible that these fragments actually represent secreted APP frag- ments that were endocytosed from the medium. While our results limit the number of cellular sites where secretory cleavage could occur, precise localization of the site of APP secretory processing awaits further study.

Acknowledgments-We thank David Wunderlich and the Institute for Molecular Biologicals at Miles Research Center, West Haven, CT, for large scale production of cultured PC12 cells. We thank Konrad Beyreuther for the gift of antibody 22Cll and Helen Shio for assistance with the electron microscopy.

Addendum-The presence of full-length APP and carboxyl-ter- minal APP fragments in a subcellular fraction enriched in lysosomes has recently been reported (Haass et al., 1992). Cleavage of some APP molecules at the cell surface has also recently been reported (Sisodia, 1992).

REFERENCES

Ahle, S.. Mann, A., Eichelsbacher, U., and Ungewickell, E. (1988) EMBO J. 7 , 919-929

Benowitz, L. I., Rodriguez, W., Paskevich, P., Mufson, E. J., Schenk, D., and Neve, R. L. (1989) Exp. Neurol. 106,237-250

Bradford, M. M. (1976) Anal. Biochem. 72,248-254 Brodsky, F. M. (1988) Science 242,1396-1402 Buxbaum, J. D., Gandy, S. E., Cicchetti, P., Ehrlich, M. E,, Czernik, A. J.,

Fracasso, R. P., Ramabhadran, T. V., Unterbeck, A. J., and Greengard, P. (1990) Proc. Natl. Acad.,Sci. U. S. A . 8 7 , 6003-6006

Campbell, C., Squicciarinl, J., Shia, M., Pilch, P. F., and Fine, R. E. (1984) Biochemistry 23,4420-4426

Caporaso, G. L., Gand , S. E., Buxbaum, J. D., and Greengard, P. (1992a) Proc. Natl. Acad. Sci. U. 3 A . 89,2252-2256

Ca oraso, G. L., Gandy, S. E., Buxbaum, J. D., Ramabhadran, T. V., and ereengard, P. (1992b) Proc. Natl. Acad. Sei. U. S. A . 89,3055-3059

Chen, W.-J., Goldstein, J. L., and Brown, M. S. (1990) J. Bml. Chem. 2 6 6 , 3116-3123

Cole, G. M., Huynh, T. V., and Saitoh, T. (1989) Neurochem. Res. 14,933-939 Esch, F. S., Keim, P. S., Beattie, E. C., Blacher, R. W., Culwell, A. R., Oltersdorf,

Gandy, S. E., Buxbaum, J. D., and Greengard, P. (1991) in Alzheimer's Disease: T., McClure, D., and Ward, P. J. (1990) Science 248,1122-1124

Basic Mechanisms, Diagnosis and Therapeutic Strategies (Iqbal, K., Crapper-

Wiley & Sons, New York McLachlan, D., Winblad, B., and Wisniewski, H., e&) pp. 155-172, John

Glenner, G. G., and Wong, C. W. (1984) Biochem. Biophys. Res. Commun. 120 , 885-890

Golde, T. E., Estus, S., Younkin, L. H., Selkoe, D. J., and Younkin, S. G. (1992)

Goldgaber, D., Lerman, M. I., McBride, 0. W., Saffiotti, U., and Gajdusek, D. Science 266,728-730

C. (1987) Science 236.877-880 ~~~~ , ~~~... . . Goldstein, J. L., Brown, M. S., Anderson, R. G. W., Russell, D. W., and

Haass, C., Koo, E. H., Mellon, A., Hung, A. Y., and Selkoe, D. J. (1992) Nature

- - -

Schneider, W. J. (1985) Annu. Reu. Cell BioL 1, l -39

357~ m u o x Harter, C., and Mellman, I. (1992) J. Cell Biol. 117,311-325 Kang, J., Lemaire, H. G., Unterbeck, A., Salbaum, J. M., Masters, C. L.,

Grzeschik, K. H., Multhaup, G., Beyreuther, K., and Mdler-Hill, B. (1987)

Kitagucbi, N., Takahashi, Y., Tokushima, Y., Shiojiri, S., and Ito, H. (1988) Nature 326 , 733-736

Nature 331,530-532 Kornfeld, S., and Mellman, I. (1989) Annu. Reu. Cell Biol. 6, 483-525 Laemmli, U. K. (1970) Nature 227,680-685

Lobel, P., Fuimoto, K., Ye, R. D., Griffiths, G., and Kornfeld, S. (1989) Cell Lazarovits, J. and Roth, M. (1988) Cell 63,743-752

67,787-796 Masters, C. L., Simms, G., Weinmann, N. A., Multhau G , McDonald, B. L.,

and Beyreuther, K. (1985) Proc. Natl. Acad. Sci. U. d?4.82,4245-4249 Nandi, P. K., Irace, G., Van Jaarsveld, P. P., Lippoldt, R. E., and Edelhoch, H.

(1982) Proc. Natl. Acad. Sci. U. S. A. 79,5881-5885 Oltersdorf, T., Fritz, L. C., Schenk, D. B., Lieberburg, I., Johnson-Wood, K. L.,

Beattie. E. C.. Ward. P. J.. Blacher. R. W.. Dovev. H. F.. and Sinha, S. (1989)

- - - I --- ---

Nature'341,'144-147 . - .

Pearse, B. M. F. (1976) Proc. Natl. Acad. Sci. U. S. A. 73,1255-1259 Pearse, B. M. F. (1988) EMBO J. 7,3331-3336 Ponte, P., Gonzalez-DeWhitt, P., Schillin J., Miller, J., Hsu, D., Greenberg,

B., Davis, K., Wallace, W., Lieberburg,?:, Fuller, F., and Cordell, B. (1988)

Robakis, N. K., Ramakrishna, N., Wolfe, G., and Wisniewski, H. M. (1987) Nature 331,525-527

Sisodia, S. S. (1992) Proc. Natl. Acad. Sci. U.S.A. 89,6075-6079 Proc. Natl. Acad. Sci. U. S. A. 8 4 , 4190-4194

Sisodia, S. S., Koo, E. H., Beyreuther, K., Unterbeck, A., and Price, D. L. (1990) Science 248,492-495

Tanzi, R. E., Gusella, J. F., Watkins, P. C., Bruns, G. A. P., St. George-Hyslop, P., van Keuren, M. L., Patterson, D., Pagan, S., Kurnit, D. M., and Neve, R.

Tanzi, R. E., McClatche , A. I., Lamperti, E. D., Villa-Komaroff, L., Gusella, L. (1987) Science 236,880-883

Towbin. H.. Staehelin. T.. and Gordon. J. (1979) Proc. Natl. Acad. Sci. U. S. A. J. F., and Neve, R. L. 6988) Nature 331 , 528-530

~ ~ ~~~~, ~~, ~ ~~ . . .~ I , , . .

76,4350-4354 Turkewitz, A. P. and Harrison, S. C. (1989) J. Cell Biol. 108,2127-2135 Van Nostrand, W . E., Wa er, S. L., Suzuki, M., Choi, B. H., Farrow, J. S.,

. Geddes, J. W., Cotman, f? W., and Cunningham, D. D. (1989) Nature 341 ,

Weidemann, A., Konig, G., Bunke, D., Fischer, P.,.Salbaum, J. M., Masters, C. 546-549

L., and Beyreuther, K. (1989) Cell 67,115-126