The amyloid beta-protein precursor and its mammalian homologues ...

7 downloads 40 Views 2MB Size Report
Tanzi, R. E., Gusella, J. F., Watkins, P. C., Bruns, G. A. P., St. George-Hyslop, ... Bush, A. I., Multhaup, G., Moir, R. D., Williamson, T. G., Small, D. H., Rumble,.
Communication

THE JOURNAL OF BIOWGICAL CHEMISTRY Vol. 269, No.43, Issue of October 28, pp. 26618-26621, 1994 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

The Amyloid P-Protein Precursor and Its Mammalian Homologues

zinc metabolism has recently been implicatedin the etiopathogenesis of Alzheimer AP amyloid (10, 11); hence, characterization of the interaction of zinc withAPP a n d its homologues may yield further insights intothe pathophysiology of the disorder. The zinc binding site EVIDENCE FOR A ZINC-MODULATED in the APP ectodomain appears to HEPARIN-BINDING SUPERFAMILY* modulate an increase in the affinity of APP for heparin(9). In addition, a zinc-induced increase in APP affinity for laminin (Received for publication, July 21, 1994) (121, as well as a zinc-induced increase in serine protease inhiAshley I. Bush, Warren H. Pettingell, Jr., bition by formsof APP containing a Kunitz protease-inhibitory Marc de Paradis, Rudolph E. Tanzi, and domain (131, could also be mediated through the sameectodoWilma WascoS main zinc bindingsite. The obligatory motif for the ectodomain From the Laboratory of Genetics a n d Aging, zinc binding region onAPP is a novel sequence for a zinc bindMassachusetts General Hospital, Harvard Medical ing site, GVEFVCCP, that is highly conserved i n A P L P l and School, Boston, Massachusetts 02129 APLP2, as well as in the Drosophila and Caenorhabditis elTheAlzheimerP-amyloidprecursorprotein (APP) eguns APP-like proteins (14,15) (Fig. l). Such profound evolucontains an ectodomain zinc binding site that has been tionary conservation indicates that this zinc binding site may reported to modulate the heparin affinity and protease-be critical to the functionality of these proteins. To determine inhibitory properties of the molecule. This motif, whether the function of this domain is conserved inthe mamGVEFVCCP, is highly conserved in amyloid precursor- malian genefamily, the interactionof zinc with these proteins like proteins 1 and 2 (APLP1 and APLPI), as well as in was investigated.

the Drosophila and Caenorhabditiselegans APP-like proteins (APPL and APL-1). To determine whether the EXPERIMENTALPROCEDURES function ofthis domain is preserved in the human APPReugents-Analytical-grade reagents, electrophoresis-grade Trislike proteins, the effect of zinc in modulating theelution HC1 (Bio-Rad),and highly deionized water were used. Chelating Sephaprofile of these proteins upon heparin-Sepharose chro- rose and heparin-Sepharose were from Pharmacia Biotech Inc.r3,S1Mematography was studied. Both APLPl andAPLP2bound thionine and Rainbow molecular size markers were from Amersham heparin-Sepharose and hadNaCl elution profiles simi- Corp. lar to that of APP. As previously reported forAPP, zinc In Vitro IkansZation (AT)-Protein synthesis was performedin vitro increased the recovery of APLPl and APLP2 upon hep- using both polymerase-coupled wheat germ extract and rabbit reticuarin-Sepharose chromatography. APP, APLP1, and locyte lysate kits according to manufacturer’s instructions (TNT Systems, Promega, Madison, WI). APP (16),APLPl,,, (7), and APLP2753( 8 ) APLP2 all bindzinc-chelatingSepharose,indicating that the zinc binding motif may be functionally con- cDNAs have been described previously. APLPl and APLP2 constructs served in these proteins. Additionally,APP, APLP1, and were created in pSK Bluescript (Stratagene, La Jolla, CAI, and APP constructs were created in pGEM (Promega, Madison, WI).APLPl and APLP2 migrate at higher molecular sizes ( 4 0 kDa) on APLP2 transcriptions were by T3 (high concentration) polymerase. SDS-polyacrylamidegel electrophoresis than their pre- APP,, and APP,,, transcriptions wereby SP6 (high concentration) dicted molecularsizes. We report data that compare the polymerase. physicochemical properties of APP to its novel APLP Protein Binding to Heparin-Sepharose-Heparin-Sepharose columns homologues and indicate that these molecules behaveas (250-pl bed volume) were preequilibrated with 3.25 ml of buffer 1(150 a family of zinc-modulated, heparin-binding proteins. mM NaCl, 20 mM Tris-HC1,l m~ phenylmethylsulfonyl fluoride,pH 7.4,

75 p~ ZnC1,) at 20 “C. [35SlMethionine-labeled IVT synthesis product (20 pl) was added to buffer1(500 pl), applied to the column, which was then washed with buffer 1(3.25 ml). Elution of bound protein proceeded AP, the principal constituentof the cerebral amyloid deposits with a 50 mM NaCl step gradient (in 20 mM Tris-HC1, 1 mM phenylof Alzheimer’s disease (1, 21, is derived from the much larger methylsulfonyl fluoride,pH 7.4, ? 75 1” ZnCl,, 600 p1) from 200 to 1000 amyloid protein precursor(APP)’ (3-6), which is a member of a mM, followed by a final NaCl pulse a t 1500 mM (600 pl). The columns mammalian gene family that includes APLPl andAPLP2 (7,B). were regenerated with 2 M NaCl(3.25 ml). Protein Binding to CheZuting Sepharose-Chelating Sepharose colAlthough the physiological function of these proteins is still umns (250-pl bed volume) werecharged with 2.5 mM ZnC1, (3.25 ml)and unclear, the strong homologies of the structural domains within preequilibrated with 3.25 mlof buffer 2 (500 mM NaCI, 50mM Tris-HC1, them indicate purposive conservation. One such domain is the pH 7.4) at 20 “C. [35SIMethionine-labeled IVT synthesis product (20pl) zinc binding site in t h e APP ectodomain (9).Abnormal cerebral was added to buffer 2 (500 pl), applied to the column, which was then washed with buffer 2 (3.25 ml). The bound protein was eluted by low* This work was supported by National Institutes of Health Grants ering the pH in pulsed steps (0.5 pH units) from 7.0to 3.5 (600 pl of 500 R 0 1 NS30428-03 and R01 AG11899-01, funds from the American mM NaCl, 50 mM buffer; MOPS (6.5-7.0), MES (5.0-6.0), acetate (3.5Health Assistance Foundation, a Harkness Fellowship from the Com- 4.5)). The columns were regenerated with EDTA (50 mM, 3.25 ml). monwealth Fund of New York (to A. I. B.), a French Foundation FelSDS-PAGE and Densitometry-Samples were protein-precipitated lowship (to R. E. T.), and a Pew Scholarship in Biomedical Sciences(to (17), dissolved in 2% SDSsample buffer containing p-mercaptoethanol R. E. T.). The costs of publication of this article were defrayed in part by (5% v/v), and applied to 9% SDS-PAGE. The amounts of [35Slmethithe payment of page charges. This article must therefore be hereby onine-labeled IVT products were quantified by autoradiography using a marked “advertisement”in accordance with 18 U.S.C.Section1734 laser densitometer (TJltroScan XL, LKB, Bromma, Sweden). solely to indicate this fact. j: To whom correspondence should be addressed: Laboratory of GeRESULTS netics and Aging, Massachusetts General Hospital, Harvard Medical School, Bldg.149,13th St., Boston, M A , 02129. Tel.: 617-726-5746;Fax: Comparing the electrophoretic patterns of IVT-synthesized 617-726-5736. The abbreviations used are: APP, p-amyloid precursor protein; IVT, APP,,,, APP,,,, APLP1, and APLP2 on SDS-PAGE indicates that these proteins migrate at apparent molecular sizes that in vitro translation; PAGE, polyacrylamide gel electrophoresis. f

26618

Zinc-and Heparin-bindingAPP Superfamily 183 185 152 169 167 Consensus APLP2 APLPl APPL APL-1 APP

TKDY PATP HFKT DQTN AEES

. . . .L.P C . . . . r, . .:.:.. . V C C I . . . . .

Frc. 1. Homology of the amino-terminal APP zinc binding domain amongAPP homologous proteins. The obligate APP zinc binding domain (in bold, residues 181-188),a s well a s a n adjacent region whose participation in zinc coordination cannot yet he excluded (91, is compared for sequence homology with mouseAPLPl(7), human APLP2 (8), Drosoph.ila APPL (14),and C. elegans APL-l(l5). Residues sharing identity with the APP sequence are boxed. The consensus residues between all five proteins are indicated. Sequence numbering of the amino terminus is indicated.

are 20-50 kDa greater than their predicted molecular sizes (Fig. 2). Theapparent molecularsizes for APP,,,, APP7,,, APLPl and APLP2 are 110, 130,95, and 135 kDa, respectively, a s compared to the predicted molecular sizes of 78,85, 73, and 87 kDa, respectively. This artifact could not be explained by oxidative cross-linking caused by the presence of hemoglobin (18), since identical migration patterns wereachievedwith IVT systems, or when both reticulocyte lysate and wheat germ performed in the presence of ascorbate (1mM) as an anti-oxidant, according to the method of Dyrks et al. (18) (data not shown). These data may indicate that this highly homologous family of proteins possess common physicochemical properties such as charge distribution, which cause aberrantly retarded migration on SDS-PAGE. To determine whetherAPLPl and APLP2 bind heparin, the NaCl elution profiles of these IVT proteins when subjected to heparin-Sepharose chromatographywere compared to thoseof IVT-synthesized APP,,, and APP,,,. Each of the synthesized proteins bound heparin-Sepharose with similar NaCl elution profiles, eluting between 200 and 700 mM NaCl (Fig. 3). The presence of zinc (75 PM)during heparin-Sepharose chromatography increases the recoveries of APP,,, and APP,,, 100% and 50% respectively (Fig. 3, 6 and c ) , in agreement with thefindings of Bush et al. (1993). The presence of zinc had a more profound effect upon APLPl and APLP2 behaviors in this system, increasing recoveries of both proteins by 230% (Fig. 3, d and e ) . Two heparin binding domains have been proposed for APP: one within the amino terminus (APP,,, exon 3, residues 99110; Ref. 191, and another encoded by exon 9 (APP,,,, residues 318-330) (31).Both domains possess the heparin-binding consensus sequence BBXB (20), where B is a basic residue. This consensus sequence is found in both APLPl and APLP2 in regions homologous to theAPP exon 9 heparin bindingdomain (Fig. 4a). Although a heparin-binding consensus sequence is found in theregion ofAPLP2 that ishomologous to the putative APP exon 3 heparin binding domain, the homologous region in APLPl lacks this motif (Fig. 46). Finally, to determine whether the interaction with zinc is conserved among the mammalian proteins that possess the homologous APP zinc binding site, we assayed the binding of UP,,,, APP751,APLP1, and APLP2 to zinc-chelating Sepharose. These proteins allbind zinc-chelating Sepharose in a pHdependent manner (Fig. 5 ) . The elution profiles of the proteins upon step-gradient elution (lowering pH from 7.0 to 4.0)are similar, with peak elutions a t pH 5.0-5.5, and 85-95% of the bound protein eluting from the column by pH 4.5. DISCUSSION

Our data indicate that APP,,,, APP,,,, APLP1, and APLP2 migrate on SDS-PAGE a t -1.5-fold their predicted molecular sizes. APP species in brain, platelets, and plasma havemolecular sizes on SDS-PAGE of 130 and 110 kDa (21-231, which are similar to those that we have observed for the APPssS and

26619

g7.jq, E 69

l

r

3

0

46

FIG.2. SDS-PAGE of IVT products. Themigration of APP,9,, APP,,,, APLP1, and APLP2 [R5S1methionine-labeledIVT products upon 9% SDS-PAGE, run under reducing conditions, is shown. The migration of high molecular mass markers (Amersham) are indicated on the left (in kDa). The relative molecular mass assignments of the major IVT protein species are indicated on the right.

APP,,, IVT products in the current studies.Although the IVTsynthesized proteins retain an intact signal sequence, this domain is predicted to contribute only -2 kDa to molecular size. N - and 0-glycosylation has been shown to contribute up to 10 kDa to the molecular mass ofAPP in humancerebrospinal fluid (24). Our observations indicate that post-translational modification may contribute little to the apparent mass of APP and indicate that its migration on SDS-PAGE is artifactually retarded, possibly as a result of charge properties. APP,,, and APLP2 nearly comigrate upon SDS-PAGE. Since both proteins possess Kunitz protease-inhibitory domains, an alternatively spliced form of APLP2 lacking the Kunitz protease-inhibitory domain may be expected to comigrate with APP,,,. Both proteins possess considerable homology and cross-react with sevAPP (25). Therefore, differentiation of eral antibodies raised to native APP from APLP2 species by procedures suchas Western blot may be unreliable. We have describedrelatively straightforward chromatographic strategies for the purification of native APLPl and APLP2. Furthermore, chromatography by zinc-chelating Sepharose offers a novel purification strategy for all members of the APP superfamily of proteins. The heparin-Sepharose NaCl elution profiles of IVT APP,,, and APP,,, are in agreement with previous reports of native APP heparin-Sepharose chromatography (26, 27), indicating that the heparin binding properties of these proteins are not dependent upon post-translational modification. Both APLPl and APLP2 bound t o heparin-Sepharose, exhibitingNaCl elution profiles similar to thatof APP, indicating that these proteins may sharea conserved heparin bindingdomain withAPP. Because the affinity for heparin is so similar for APLPl and APLP2 compared to APP, and since APLPl appears to lack one of the two heparin-binding consensus sequences found in APP and APLP2 (Fig. 4b), the putative heparin bindingdomain in the APLPs would more likely appear to be that homologous to the domain encoded by exon 9 of APP (Fig. 4a 1. The consensus heparin bindingmotif in thisdomain overlaps withthe RERMS motif, which was reported to mediate neurotrophic properties of APP in cell culture (12). We have demonstrated that the effect of Zn2+(75 PM) upon the heparin-Sepharosechromatography of IVT APP,,, was similar to that reported earlier for the nativeprotein (91, promoting both protein recovery and affinity for heparin. Interestingly, although the presence of ZnZ+had a similar effect on the binding of heparin toAPP,,,, it hada significantly more pronounced effect upon APLPl and APLP2. APP,,,,APP,,,, APLP1, and APLP2 all bound zinc-chelating Sepharose, probably mediated by the highly homologous amino-terminal zinc binding motif. Although AP also manifests specific, high affinity binding of zinc (lo),this domain in APP is unlikely to modulate the interaction between these proteins and zinc-chelating Sepharose, since APLPl and APLP2 lack homology to AP (7, 8). The evo-

Zinc- and Heparin-bindingAPP Superfamily

26620

a APPgg5

APLPl 313 FLRAKMDLEERRMRQINEVMRE I :II I I :: ::Ill1 372 FQKAKERLEAKARERMSOVMRE IIIIII:II:II/:iI I :I A P L P ~3 9 0 F Q K A K E Q L E I R A R N R M D R V K K E

316/i1PP~~~

b ;;?LFl 117

RWCGGTRSGRCAHPH I1 I Ill APP 96 NWCKRGRKQCKTHPH

III:I::IIII

:

APLP2 114 NWCRRDKKQCK..SR

FIG.4. Homologies of the putative human APP heparin binding sitesbetween APLPl and APLPB. APP sequences were aligned with mouse APLPl and human APLPZ protein sequences using the UWGCG GAP analysis (30). I, identical amino acids; :, similar amino acids. The heparin binding consensusmotif BBXB is indicated by bold characters. The amino-terminal sequence number is indicated. Punel a , the putative heparin binding domain in APP exon 9 is compared for homology to similar domains in APLPl and APLP2. Panel b, the putative heparin binding domain in APP exon 3 is compared for homology to similar domains in APLPl andAPLP2.

20 15 10

5 0 200

400

Bw

Bw

1oW

Hac1(mH)

1200

1400

30 25

4

20

10

5 400

Bw

800 loo0 1200 1400 conc. (mu)

WI

200

400

600

800

1000

1200

1400

Hac1 (d)

e

APLP?(Arbitrary

5

515 pH

6

65

7

FIG.5.Elution profileof pH-dependentbinding ofAPP.APLP1, and APLPB to zinc-chelating Sepharose. pH elution fractions of zinc-chelatingSepharosechromatography of ["'Slmethionine-labeled IVT products were analyzed by 9 6 SDS-PAGE, autoradiography, and densitometry. Densitometry values (in arbitrary units) were measured a s a proportionof the totalIVT product elutedfrom the chromatography column.

15

0 2cm

4:5

unns)

lutionary conservation of the amino-terminal zinc binding site in APPdown to the level of C. elegans indicates that thefunction of this domain may be critical to thephysiological purpose of the APP/APLP superfamily. Our data indicate that by virtue of their similar physicochemical properties, APPBS5, APP,,,, APLP1, and APLP2 constitute a family of zinc-modulated, heparin-binding proteins. Extracellular zinc may play a role in modulating the physiologic function of APP by modifying its adhesiveness to extracellular matrix elements. This is particularly important since APP may play a role in cell adhesiveness (28) and neurite outgrowth (19, 29). The physiological function of the APP/ APLP-zinc interaction remains unclear; however, increased adhesiveness of these proteinsat the synaptic terminal or neuritic growth cone may promote the binding of these proteins to extracellular matrix elements like heparan sulfate proteoglycans and laminin (9, 12). The possibility that zinc is a local environmental cofactor modulating this interaction and in shaping synaptic contact and maintenance merits further investigation. REFERENCES

Hac1 (W) FIG.3. Zinc-modulatedbinding of APP, APLPl, and APLPB to heparin-Sepharose. Panel a , NaCl elutionfractions of heparinSepharosechromatography of [J"Slmethionine-labeledAPLP2were analyzed by 9%SDS-PAGE and autoradiography. The chromatography was performed in the presence of either Zn2+ (75 PM)or EDTA(1 mM) a s indicated. The molecular mass assignment of APLP2 a t 135 kDa is indicated on the left. The migration of high molecular mass markers is

1. Masters, C. L., Simms,G., Weinman, N. A,, Multhaup, G., McDonald, B. L., and Beyreuther, K. (1985) Proc. Nutl. Acad. Sci. U.S. A. 82, 4245-4249

indicated (in kDa) on the right. Panels b-e, heparin-Sepharose NaCl elution profiles in the presence of Zn2+(75 PM)or EDTA (1 mM), for APP,,,,APP,,,,APLPl, andAPLP2, respectively, measured by densitometry. Densitometryvalues(inarbitraryunits) were measuredasa proportion of the total IVT product loaded upon the chromatography column.

Zinc-and Heparin-bindingAPP Superfamily 2. Glenner, G. G., and Wong, C. W. (1984) Biochem. Biophys. Res. Commun. 120, 885-890 3. Kang, J., Lemaire, H., Unterbeck, A., Salbaum, J. M., Masters, C. L., Grzeschik, K., Multhaup, G., Beyreuther, K., and Miiller-Hill, B. (1987) Nature 328, 733-736 4. 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. L. (1987) Science 236, 880-884 5. Robakis, N. K , Ramakrishna, N., Wolfe, G., and Wisniewski, H. M. (1987) Proc. Natl. Acad. Sci. U.S. A. 64,4190-4194 6. Goldgaber, D., Lerman, M. I., McBride, 0. W., Saffiotti, U., and Gajdusek, D. C. (1987) Science 235, 877-880 7. Wasco, W., Bupp, K., Magendantz, M., Gusella, J. F., Tanzi, R. E., and Solomon, F. (1992) Proc. Natl. Acad. Sci. U.S. A. 89, 10758-10762 8. Wasco, W., Gurubhagavatula, S., de Paradis, M., Romano, D., Sisodia, S. S., Hyman, B. T., Neve, R. L., and Tanzi, R. E. (1993) Nature Genet. 6,95-100 9. Bush, A. I., Multhaup, G., Moir, R.D., Williamson, T.G., Small, D. H., Rumble, B., Pollwein, P.,Beyreuther, K., and Masters, C. L. (1993) J. Biol. Chem. 268,16109-16112 10. Bush, A. I., Pettingell, W. H., Jr., de Paradis, M., and Tanzi, R. E.(1994) J. Biol. Chem. 269.12152-12158 , ~ ~ - " 11. Bush, A. I., Pettingell, W. H., Multhaup, G., de Paradis, M., Vonsattel, J.-P., Gusella, J. F., Beyreuther, K., Masters, C. L., and Tanzi,R. E. (1994) Science 266, 1464-1467 12. Roch, J.-M., Shapiro, I. P., Sundsmo, M. P., Otero, D. A. C.,Refolo,L. M., Robakis, N. IC, and Saitoh, T. (1992) J. Biol. Chem. 267, 2214-2221 13. Komiyama, Y.,Murakami, T., Egawa, H., Okubo, S., Yasunaga, K., and Murata, K. (1992) Thramb. Res. 66,397-408 14. Rosen, D. R., Martin-Morris, L., Luo, L., and White, K.(1989)Proc. Natl. Acad. Sci. U.S. A . 86,2478-2482 15. Daigle, I., and Li, C. (1993) Proc. Natl. Acad. Sci. U.S. A . 90, 12045-12049 16. Tanzi, R. E.,McClatchey,A. I., Lamperti, E. D., Villa-Komaroff,L., Gusella, J. F., and Neve, R. L. (1988) Nature 311,528-530 ~

~~~

~

26621

17. Wessel, D., and Fliigge, U. 1. (1984) Anal. Biochem. 138, 141-143 18. Dyrks, T.,Dyrks, E., Hartmann, T., Masters, C., and Beyreuther, K. (1992)J. Biol. Chem. 267, 18210-18217 19. Small, D. H., Nurcombe, V., Reed, G., Clams, H., Moir, R., Beyreuther, K., and Masters, C. L. (1994) J. Nenrosci. 14, 2117-2127 20. Cardin, A. D., and Weintraub, H. J. R. (1989) Arteriosclerosis 9, 21-32 21. Selkoe, D. J., Podlisny, M. B., Joachim, C. L., Vlckers, E. A,, Lee, G., Fritz, L. C., and Oltersdorf, T. (1988) Proc. Natl. Acad. Sci. U.S . A. 85, 7341-7345 22. Bush, A. I., Martins, R. N., Rumble, B., Moir,R., Fuller, S., Milward, E., Currie, J., Ames, D., Weidemann, A., Fischer, P., Multhaup, G., Beyreuther, K., and Masters, C. L. (1990) J. B i d . Chem. 266, 15977-15983 23. Bush, A. I., Whyte, S., Thomas, L. D., Williamson, T. G., Van Tiggelen, C. J., Currie, J., Small, D. H., Moir, R. D., Li, 9.-X., Rumble, B., Monning, U., Beyreuther, K., and Masters,C. L. (1992) Ann. Neurol. 3 2 , 5 7 4 5 24. Saito, F., Yanagisawa, K., and Miyatake,T. (1993)Mol. BrainRes. 19,171-174 25. Slunt, H. H., Thinakaran, G., Von Koch, C.,Lo, A. C., Tanzi, R. E., and Sisodia, S. S. (1994) J. B i d . Chem. 269,2637-2644 26. Schubert, D., LaCorbiere, M., Saitoh, T., and Cole, G. (1989) Proc. Natl. Acad. Sci. U.S. A . 86, 2066-2069 27. Moir, R. D., Martins, R. N., Bush,A. I., Small, D. H., Milward, E. A,, Rumble, E. A., Multhaup, G., Beyreuther, K., and Masters, C. L. (1992) J . Neurochem. 69, 1490-1498 28. Shivers, B. D., Hilbich, C., Multbaup, G., Salbaum, M., Beyreuther, K., and Seeburg, P. H. (1988) EMBO J. 7, 1365-1370 29. Milward, E. A., Papadopoulos, R., Fuller, S. J., Moir,R.D., Small, D., Beyreuther, K., and Masters, C. L. (1992) Neuron 9, 129-137 30. Genetics Computer Gmup (1991) Program Manual for the GCG Package, Version 7,Genetics Computer Group, University of Wisconsin, Madison, WI 31. Multhaup, G. (1994) in Amyloid Protein Precursor in Deuelopment, Aging and Alzheimer's Disease (Masters, C. L., Beyreuther, K., Trillet, M., and Christen, Y.,eds) pp. 7G89, Springer-Verlag, Heidelberg