Expression and distribution of amyloid precursor protein-like protein-2 ...

AmeriCnIourI / f1O/'la/holog', tolo 149 Ab. 4, Octohber 1996 Copyruqbght C Antcericani Sociel/ for Investigative Pathologyj

Short Communication Expression and Distribution of Amyloid Precursor Protein-Like Protein-2 in Alzheimer's Disease and in Normal Brain

Barbara J. Crain,* Weidong Hu,* Chun-l Sze,* Hilda H. Slunt,* Edward H. Koo,t Donald L. Price,* Gopal Thinakaran,* and Sangram S. Sisodia* From)} the I)epartment of Pathology, (Ncuropatholog}' Laborator)).* 71e johns Hopkins Uni 'ensity School o] MCdiv/Cil, IBaltmmnorc, rlyldandIciiicl the c-entCr/fir AClurolqOic DiseaseGst Brigham alnd Women's Hospital, Bostom/, Massachuselts

Amyloid precursor-like protein-2 (APLP-2) belongs to afamily of homologous amyloidprecursor-like proteins. In the present study we report on the expression and distribution of APLP-2 in fetal and adult human brain and in brains of patients with Alzheimer's disease. We demonstrate that APLP-2 mRNAs encoding isoforms predicted to undergo post-translational modification by chondroitin sulfate glycosaminoglycans are elevated in fetal and aging brains relative to the brains of young adults. Immunocytochemical labeling with APLP-2-specific antibodies demonstrates APLP-2 immunoreactivity in cytoplasmic compartments in neurons and astrocytes, in large part overlapping the distribution of the amyloidprecursorprotein. In Alzheimer's disease brain, APLP-2 antibodies also label a subset of neuritic plaques. APLP-2 immunoreactivity is particularly conspicuous in large dystrophic neurites that also label with antibodies specificfor APP and chromogranin A. In view of the age-dependent increase in levels of chondroitin sulfate glycosaminoglycan-modified forms of APLP-2 in aging brain and the accumulation of APLP-2 in dystrophic presynaptic elements, we suggest that APLP-2 may play roles in neuronal

sprouting or in the aggregation, deposition, and/or persistence of f3-amyloid deposits. (Am J Pathol 1996, 149:1087-1095)

Amyloid precursor-like protein-2 (APLP-2) belongs to a family of homologous amyloid precursor-like proteins (APLPs) which also includes APLP-11 and the amyloid precursor protein (APP) itself.2-7 Although the APLPs are highly homologous to APP in the amino- and carboxyl-terminal domains, the APLPs do not contain the ,B-amyloid peptide domain (AP3), which is deposited as extracellular amyloid in the brain parenchyma in Alzheimer's disease.89 At present, APLP-2 is the best characterized member of APLPs. Like APP, APLP-2 is encoded by several alternatively spliced mRNAs,10-12 and it matures through the same unusual secretory/cleavage pathway.11 Recently, we and others have demonstrated that unique APLP-2 and APP isoforms, encoded by alternatively spliced transcripts, are substrates for modification by a chondroitin sulfate glycosaminoglycan (CS GAG) chain.13-15 Both APLP-2 and APP are expressed in brain and peripheral tissues,10'11 and in situ hybridization studies disclose similarities in the distributions of APLP-2 and APP mRNAs in mouse11 and human10 central nervous system. We recently examined the distribution of APLP-2 in the mouse brain and found that APLP-2 is enriched in postsynaptic compartments in the cortex and hippocampus.16 In contrast, APLP-2 Supported by the National Institutes of Health (grants R01 AG09216, P50 AG05146, and NS20471) and the Adler Foundation. Accepted for publication June 11, 1996. Address reprint requests to Dr. Barbara J. Crain, Department of Pathology, The Johns Hopkins University School of Medicine, 519 Ross Building, 720 Rutland Avenue, Baltimore, MD 21205.

1087

1088

Crain et al

A/P October 1996, Vol. 149, No. 4

is present in both pre- and postsynaptic compartments in the olfactory bulb. Notably, mRNA encoding CS-GAG-modified forms of APLP-2 is enriched in the olfactory epithelium, and CS-modified APLP-2 forms accumulate in the olfactory bulb.16 In the present study, we report on the expression and distribution of APLP-2 in adult human brain and in brains of patients with Alzheimer's disease (AD). We demonstrate that APLP-2 mRNAs encoding isoforms predicted to undergo post-translational modification by CS GAG are elevated in aging brain relative to the brains of young adults. Immunocytochemical labeling with newly generated APLP-2specific antibodies demonstrates APLP-2 immunoreactivity in cytoplasmic compartments in neurons and astrocytes, in large part overlapping the cellular and subcellular distributions of APP.17-26 In AD brain, APLP-2 antibodies also label a subset of neuritic plaques throughout neocortex and hippocampal formation, particularly in the subiculum and CAl. APLP-2 immunoreactivity is particularly conspicuous in large, rounded dystrophic neurites that are also labeled with antibodies specific for APP and the dense-core vesicle protein chromogranin A. In view of the age-dependent increase in levels of CS-GAGmodified forms of APLP-2 in aging brain and its accumulation in dystrophic presynaptic elements, we suggest that APLP-2 may play roles in neuronal sprouting or in the aggregation, deposition, and/or persistence of A13 deposits.

Materials and Methods Cases Adult brain tissue samples were obtained from the Brain Resource Center at the Johns Hopkins University School of Medicine or the general autopsy service at the Johns Hopkins Hospital. The neuropathological diagnosis of AD was made according to standard criteria.27'28 Control patients had no history of neurological disease and no neuropathological lesions at autopsy. Permission to harvest fetal tissue was obtained from our institutional review board.

Polymerase Chain Reaction (PCR) Analysis of APLP-2 mRNA Total RNA was extracted from human brain tissue samples by the guanidine isothiocyanate-cesium chloride method.29 Postmortem intervals for the control (ages 66 and 75 years) and AD (age 72 years) cases were 5 and 7.5 hours, respectively. Two young

controls (ages 18 and 21 years) had postmortem delays of 8 and 7 hours, respectively. Tissues from two late second trimester fetuses were obtained within 2 hours of abortion. One fetus had Down's syndrome, with trisomy 21 established by karyotype

analysis. RNA isolated from human brain samples was reverse transcribed and a PCR was used to

amplify alternatively spliced APLP-2 mRNA. The primer pair SKPI, 5'-CGWGAYTACTACTATGACMCC-3', and AsKPI, 5'-GTCCATATCYGCWCGCTGYTC-3', was used to analyze the presence of alternatively spliced APLP-2 transcripts that encode the KPI domain; primer pair S12, 5'-CCGGGATCCGAGCAGCGTGCAGATATG-3', and As1 2, 5' -CCGGAATTCCACTGCTACTCAGACTG-3', was used to analyze alternatively spliced APLP-2 transcripts that encode a 12amino-acid peptide, DTQPELYHPMKK. The PCR products were separated on 2% agarose and visualized by ethidium bromide staining. No quantitative analysis was performed.

Western Blot Analysis of APLP-2 Expression APLP-2 expression was examined in brain extracts prepared from two controls and two cases of AD. The two controls were 91 years of age, with postmortem delays of 10 and 16 hours. The two cases of AD were 87 and 89 years of age, with postmortem delays of 11 and 5 hours, respectively. Membrane protein extracts were prepared from frozen brain samples as described elsewhere.30 Briefly, entorhinal cortex and hippocampus were homogenized with a Brinkmann Polytron in 10 volumes of ice-cold 10 mmol/L Tris/HCI (pH 7.4) containing 10% (w/v) sucrose and protease inhibitors (20 jig/ml each of leupeptin, pepstatin, antipain, and chymostatin, 0.2 U/ml aprotinin, 10 mmol/L benzamidine, 0.1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L EGTA, and 1 mmol/L EDTA). The homogenate was centrifuged at 1000 x g for 10 minutes to remove nuclei and cell debris, and the resulting supernatant fraction was centrifuged at 114,000 x g for 20 minutes at 40C. The pellet fraction (P2) containing membrane proteins was washed twice by resuspension in homogenization buffer (lacking sucrose), centrifuged at 114,000 x g for 20 minutes, and finally resuspended in homogenization buffer. The 10-,ug aliquots of the extracts were fractionated on sodium dodecyl sulfate (SDS)-polyacrylamide gels and analyzed by immunoblotting with a 1:2000 dilution of APLP-2 antibody

D211.16

Localization of APLP-2 and APP

1089

AJP October 1996, Vol. 149, No. 4

Antibodies The primary antibodies used in this study included D211, a polyclonal antibody to APLP-2 (1:100 to 1:2000) raised in rabbit,16 and CT15, a polyclonal antibody to the carboxy terminal of APP (1 :100 to 1:200) also raised in rabbit.31 The monoclonal antibody to chromogranin A was used without dilution (Zymed Laboratories, San Francisco, CA). The antibody to glial fibrillary acidic protein was diluted 1:200 (DAKO Corp., Carpinteria, CA). The antibody to the A,B-peptide (A,B) was diluted 1:250 for single labeling and 1:200 for double labeling (1OD5, residues 1 to 38 of A,B; gift from Athena Neurosciences, South San Francisco, CA).

Immunohistochemistry Paraffin-embedded blocks of hippocampal formation and/or frontal neocortex were examined from six cases of AD (ages 69 to 91 years; postmortem intervals, 4 to 24 hours) and four control patients (ages 62 to 85 years; postmortem intervals, 14 to 31 hours). Sections were cut at 10 ,um, deparaffinized, and rehydrated. They were treated with 1.5% hydrogen peroxide in methanol, microwaved in water, and placed in 5% milk, 0.1% Triton X-1 00, and 0.1 mol/L phosphate-buffered saline, pH 7.4. The sections were then placed in the first primary antibody (diluted with 5% milk, 0.1% Triton X-100, and phosphate buffer) and incubated overnight at 4°C. The following day, this antibody was visualized using an avidin-biotin complex (ABC) kit (Vector Laboratories, Burlingame, CA) with biotinylated secondary antibody diluted 1:20032 and diaminobenzidine tetrachloride with hydrogen peroxide. Selected sections were counterstained with hematoxylin and coverslipped. Other sections were incubated overnight at 4°C with a second primary antibody. These antibodies were also visualized with the ABC method, using benzidine dihydrochloride as the chromogen (1 mg of benzidine dihydrochloride, 2 mg of sodium nitroferricyanide in 9.5 ml of 0.01 mol/L phosphate buffer, pH 6.8).21 Sections labeled with anti-A,B were pretreated with 99% formic acid for 8 minutes before administration of the primary antibody. Control studies were performed by omission of either the first or second primary antibody or the first or second secondary antibody.

Results

encode APLP-2 polypeptides of 694, 706, 751, or 763 amino acids. The 751- and 763-amino-acid isoforms contain a domain homologous to the Kunitz protease inhibitors, whereas the 694- and 706-amino-acid forms lack this region. Moreover, the 706and 763-amino-acid forms contain an additional 12amino-acid sequence encoded by an alternatively spliced exon. APLP-2-751 undergoes modification by the addition of CS GAG at a single serine at position 614,13 and insertion of the 12-amino-acid exon immediately upstream of the Ser-614 inhibits CS GAG attachment. 14 To determine whether the APLP-2 mRNA expressed in human brain contains sequences encoding the Kunitz protease inhibitor domain, we performed reverse transcriptase (RT)-PCR analysis with primers that flank that sequence. Samples of mRNA from each tissue were reverse transcribed, and the resulting cDNA was subjected to PCR for an identical number of cycles. We found that equivalent proportions of APLP-2 mRNA encode, or lack, the KPI domain in normal fetal, young adult, and aged brain as well as in AD brain and fetal Down's syndrome brain (Figure 1A, upper panel). Although the total levels of APLP-2 appear lower in the aged individuals, this result may reflect post- or antemortem degradation of RNA in their brain tissue. To assess the levels of APLP-2 mRNAs that contain or lack the 12-amino-acid peptide, we performed RT-PCR analysis with primers flanking this sequence (Figure 1A, lower panel). In contrast to the relatively constant ratios of KPI-positive and KPInegative transcripts, the levels of APLP-2 mRNA encoding the 12-amino-acid exon varied dramatically in the different groups of subjects. The vast majority of APLP-2 mRNA in fetal brain does not encode the 12-amino-acid insert and hence would be predicted to encode APLP-2 isoforms that are modified by CS GAG. This pattern is reversed in brains of young adults, who express elevated levels of APLP-2 mRNA containing the 36-nucleotide exon. Essentially identical results have been reported in quantitative RT-PCR studies of mRNA from adult rodent brain.12 Interestingly, the aged control and AD brain samples showed the same pattern as the fetal brains. Hence, it is likely that the levels of CS-GAG-modified APLP-2 are increased relative to CS-GAG-free forms in the elderly compared with young adults.

Expression of APLP-2 mRNA in Human

Expression of APLP-2 in Human Brain

Brain

To assess the presence of APLP-2 in adult human brain, we used D211 antiserum in Western blot analysis of membrane-associated protein preparations

The primary APLP-2 transcript undergoes alternate splicing10-12.33; the alternatively spliced transcripts

1090

Crain et al

AJP October 1996, l(kl. 149, No. 4

Ent. Cortex

A

Hippocampus

B 0. .0 -

m

IL

m_LL C\j

0)

00(

CD

U*rcr.

1-1

_r CM CO ll-

0

0

m

r"

co

-

CD

cn co

0. -

0

.0

tco

_T

CM

c co

KPI

+12aa

Western blot

RT-PCR

Figure 1. Expression of APIP-2 in hbiuman brain. A: Anialysis of APLP-2 miRAA isqofnms. PCR products neregeneratedcifiom reierse-transcribed R'A thce 12-c auiiio-acid inisert (lower panel) and seIl)arated on1 agarosegels anid v'isumalized using primcr pairs that flank the KPI domain (upper panel) ethiCdiumo brommidcle staining. B: Western l)lot analysis of' APlLP-2 exp)ression 7emi micrograms of the P2 fraction isolatedi fromi hneian braimi homi,ogenates uasfiractiomnatel by SDS-PAGF candl siibjectecl tci immunostaininmg nith APIP-2-specific anitibodyl D2-II.1llolecular mnass standards are or

listed ini kilodaltomns

from the entorhinal cortex or hippocampus of two aged control individuals and two individuals with autopsy-confirmed AD. The D211 antibody is specific for APLP-2.16 Qualitative examination of the data showed no indication of any differences among the groups in total APLP-2 concentrations, so quantitative studies were not pursued. Interestingly, however, immunoblot analysis of membrane fractions from both brain regions did reveal the presence of discrete APLP-2-related species of -95, -100, and -115 kd and a heterogeneous pattern of polypeptides with apparent molecular mass of -120 to 160 kd (Figure 1B). The latter may represent accumulated CS-GAG-modified species, similar to those reported to accumulate in the rodent olfactory bulb.16 Quantitative studies were not pursued in the absence of any qualitative evidence of a large differ-

ence in total APLP-2 concentrations in control versus AD cases.

Distribution of APLP-2 in Human Brain The distribution of APLP-2 immunoreactivity in normal brain was very similar to that of APP immunoreactivity (Figure 2, A and B). There was prominent granular cytoplasmic staining of subpopulations of neurons, particularly larger neurons, in the neocortex and hippocampal formation, including both granule cells and pyramidal cells. Large myelinated axons were frequently labeled. Granular cytoplasmic staining was also associated with astrocytes. In one control case, a focal area of reactive astrocytosis was identified around a blood vessel as the paraffin block was resectioned for the present study. The gemisto-

Figure 2. Iinniiiiniobhistocheniical localization of APLP-2 and APP. A: Sonic' nen-obs an1d castrocytes sbhow grannlatr cy) toplasiniic inmmunioreCactiv'ityJoi anltibody APLP-2. 1)2II 1:800, lith hbematoxylin coiuniterstaini; mnagnification. X 600. B: Si1nnilcar Ilnoronal an1d astrocytic stainingl is evident nith to APP. (TI5, 1200, nit/7 hemiatoxyiclcoiuiterstaini; nagnification. X600. C: (emiistoc )tic astroc ytes display miolre unllfo)rm c ytopla.smiiic APIP-2 witb antibody to AlPP. imioiinmoreactivit. 1)2II. 1:800, nit/i heinatoxylni colunlterstain; miagnfication, X200 D: Siuniilar,lastro(c)ytic staii1itig is CTI 5, 1:200, with hebematoxylin coijoterstaiii; mag7nification. X 200. E: APIP-2 inininniiiohistocbe?ii.stry demnonstrates laige nenrites nithin plaques in c itb the oiolcilu/ar lala )r of the clenitate grus. D2II. 1: 500. with hbematoxilin counter5taini; mnagnification. X 400. F: ihe samie plaques can be labeled an

seen

an

anlti-AP9P an1tibody in

nearby,

section7. Antoe the

poartictidar11b

lauge labeled niecrites ont the right. C7i5. 1:200; miagjnficatioun. X400.

Localization of APLP-2 and APP AJP October 1996,

1091

Vol. 149, No. 4

1092

Crain et al


Figure 3. Double-label immunohistocbemical studies. A: Large neurites containing APLP-2 (brown) do not contain the AP3 peptide (blue-black). D2MI, 1:2000; anti-Af3, 1:200; magnification, x 160. B: APLP-2 (brown) and chromogranin A (blue-black) are found in the same structures in neuritic plaques. D2MI, 1:2000; anti-cbromogranin A, undiluted; magnification, X 240.

cytic astrocytes in this restricted area showed prominent diffuse cytoplasmic staining with antibodies to APLP-2, APP, and glial fibrillary acidic protein (Figure 2, C and D). APLP-2 labeling of neurons and astrocytes was evident over a wide range of dilutions. However, neuritic plaques were usually best visualized with dilutions greater than 1:400 or 1:800. At these higher dilutions, APLP-2 labeled a subset of neuritic plaques throughout the neocortex and hippocampal formation, particularly in the subiculum and area CAl but also in area CA4 and the molecular layer of the dentate gyrus. Although APLP-2 immunoreactivity could sometimes be seen throughout the plaque, the most striking features in labeled plaques were large, rounded dystrophic neurites 2 to 10 gm or larger in diameter (Figures 2E and 3, A and B). These neurites could be seen either at the edge of the plaque or within it, and they sometimes appeared in the neuropil in the absence of any obvious associated AP deposition. This pattern of APLP-2 immunoreactivity was virtually identical to that seen with the antibodies to APP (Figure 2F) and chromogranin A, a protein found in dense-core vesicles within presynaptic processes. Double-label immunohistochemistry confirmed the co-localization of APLP-2 with chromogranin A (Figure 3B). As expected, AP immunoreactivity was absent from the large APLP-2-positive neurites, as seen both in single-labeled adjacent sections and in double-labeled sections (Figure 3A); the glial fibrillary acidic protein antibody also failed to

label these structures. There was no evidence of APLP-2 immunoreactivity within amyloid cores. Within meningeal vessels, there was immunoreactivity for APP, APLP-2, and A,B. There was also variable immunoreactivity for each of these markers in the ependyma and choroid plexus. There was no evidence of APLP-2 immunostaining within diffuse plaques, diffuse subpial amyloid deposits, or neurofibrillary tangles. No APLP-2 was identified within parenchymal blood vessels, although only one of the AD cases used in this study had extensive amyloid angiopathy visible in the AP3 immunostain.

Discussion A,B deposition in brain parenchyma is a prominent pathological feature of AD.8'9 The -4-kd AP peptide is derived by proteolytic cleavage of APP,27 a member of a family of homologous APLPs also including APLP-1 and APLP-2.1 10'1 133 Our earlier efforts demonstrated that APLP-2 matures through the same unusual secretory/cleavage pathway as APP11 and that unique APLP-2 isoforms, encoded by specific alternatively spliced APLP-2 mRNA, are substrates for modification by CS GAG.13'14 We have mapped the distributions of APLP-2 in the rodent central nervous system using a series of APLP-2-specific antibodies in immunocytochemical and Western blot studies16 and documented that APLP-2 is highly expressed in somatodendritic compartments of neu-

Localization of APLP-2 and APP 1093 A/P October 1996, Vol. 149, No. 4

rons in the cortex, hippocampus, and olfactory system. Moreover, APLP-2 immunoreactivity is enriched in sensory axons and sensory axon terminals in the olfactory bulb. Biochemical analysis revealed that APLP-2 isoforms transported to, and accumulated in, the olfactory bulb are modified by CS GAG, leading to the suggestion that APLP-2 may be involved in axogenesis, path finding, or synaptogenesis in this system. 16 In the present report, we extend our analysis of APLP-2 expression in rodents by examining in a qualitative fashion the expression and distribution of APLP-2 in fetal and adult human brain and in brains of patients with AD and Down's syndrome. We demonstrate that the ratio between APLP-2 mRNA encoding isoforms predicted to be modified by CS GAG and total APLP-2 mRNA is increased in normal aged and AD brains compared with brains of young adults and more closely resembles that found in fetal brains. We further show that APLP-2 is present in the cytoplasm of neurons and astrocytes, as well as within large blood vessels in the subarachnoid space, and that large APLP-2-immunoreactive neurites are found within plaques in AD. This study is the first to document the distribution of a specific member of the APP/APLP-2 gene family in human central nervous system, an effort we deemed necessary in view of our demonstration that many of the antibodies raised against independent epitopes of APP cross-react with APLP-2.11 Thus, it is quite likely that several reports on APP distribution, including our own,21'22 actually reflect combined immunoreactivity to APP and APLP-2, and perhaps to APLP-1 as well. In the present study we therefore used a highly specific APLP-2 antibody, D211,16 along with an APP antibody, CT15,31 to compare the distributions of APLP-2 and APP in normal human brain and in AD brain. We documented that APLP-2 immunoreactivity in normal adult brain appears as granular cytoplasmic staining in neurons and normal and reactive astrocytes. The same general pattern of APP staining has been reported previously in light microscopic studies of both human19'20'25 and animal21'22'26 tissue. Ultrastructurally, neuronal APP appears to be localized to lysosomes in human material17 but not in rodents18 823,24,34 or primates,212 in which APP immunoreactivity appears to be associated with the plasmalemma, Golgi membranes, and outer membranes of mitochondria.21 In view of the cross-reactivity of APP antibodies to APLPs, the differences observed in these localization studies may reflect inherent affinities of the antibodies used to each study to APP,

APLP-2, or APLP-1, true differences across species, or both. In AD brain, the APLP-2 antibody also labeled a subset of neuritic plaques throughout the neocortex and hippocampal formation, particularly in the subiculum and area CAl. APLP-2 immunoreactivity was most conspicuous in large, rounded dystrophic neurites that were also visualized with antibodies specific for APP and the dense-core vesicle protein chromogranin A, characteristic of presynaptic processes. Such large neurites have been reported previously, both as components of neuritic plaques19'35'36 and as components of tangle-associated neuritic clusters.37 They may also contain synaptophysin35 and GAP-4338 as well as a variety of peptides or transmitters36.37 and markers associated with neurofibrillary pathology.19'39 Interestingly, they do not contain AP3, either in our material or in that of others.37 The appearance of APLP-2 in presynaptic elements of neuritic plaques is consistent with our demonstration that APLP-2 undergoes rapid anterograde transport in sensory axons of dorsal root ganglia neurons (Thinakaran and Sisodia, unpublished observations). Interestingly, the vast majority of transported APLP-2 in sensory axons is modified by CS GAG. The biology of APLP-2 in the olfactory system16 and the predominance of the CS-GAG-modified form of APLP-2 in the developing brain suggest that the CS GAG forms of APLP-2 may normally be involved in an intimate way with some aspect of axogenesis, synaptic remodeling, or sprouting. Furthermore, the relative predominance of CS GAG forms of APLP-2 in AD and the presence of both GAP-4338 and APLP-2 in large dystrophic neurites in AD suggest that this molecule may perform similar functions in this disease state. It remains to be established whether the APLP-2 in plaques is truly of the CS-GAG-modified form, a point that could be settled with antibodies that detect epitopes including the stub remaining after chondroitinase treatment and adjacent amino acid residues of APLP-2. However, both heparan sulfate and CS proteoglycans have been localized within senile plaques.40 43 The cells of origin of the APLP-2 in plaques also remain to be determined. In this report, we demonstrate that neurons are not unique in expressing APLP-2. Indeed, microglia and reactive astrocytes also express APLP-2, and earlier studies demonstrated that these cells express high levels of CS-GAG-modified APLP2.12

Accumulation of the CS-GAG-modified form of APLP-2 may also be related to the aggregation and/or persistence of AP in the adjacent neuropil.

1094

Crain et al


Canning et a144 have provided strong support for the notion that AP is capable of activating microglia and astrocytes, which in turn deposit CS proteoglycan. Moreover, it is clear that AP3 binds with high affinity to GAGs,45 an interaction that results in the formation of fibrillar AP3-proteoglycan complexes that are resistant to both removal and/or degradation by cultured microglial cells.46'47 These latter findings are consistent with a role for proteoglycans in A,B aggregation, fibril formation, and persistence of amyloid deposits. The possibility that the CS-GAG-modified form of APLP-2 may be fulfilling such functions warrants additional investigation.

References 1. Wasco W, Bupp K, Magendantz M, Gusella JF, Tanzi RE, Solomon F: Identification of a mouse brain cDNA that encodes a protein related to the Alzheimer disease-associated amyloid-f-protein precursor. Proc Natl Acad Sci USA 1992, 89:10758-10762 2. Kang J, Lemaire H-G, Unterbeck A, Salbaum JM, Masters CL, Grzeschik K-H, Multhaup G, Beyreuther K, MOller-Hill B: The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor.

Nature 1987, 325:733-736 3. Kitaguchi N, Takahashi Y, Tokushima Y, Shiojiri S, Ito H: Novel precursor of Alzheimer's disease amyloid protein shows protease inhibitory activity. Nature 1988, 331: 530-532 4. Ponte P, Gonzalez-DeWhitt P, Schilling J, Miller J, Hsu D, Greenberg B, Davis K, Wallace W, Lieberburg I, Fuller F, Cordell B: A new A4 amyloid mRNA contains a domain homologous to serine proteinase inhibitors. Nature 1988, 331:525-532 5. Tanzi RE, McClatchey Al, Lampert ED, Villa-Komaroff L, Gusella JF, Neve RL: Protease inhibitor domain encoded by an amyloid protein precursor mRNA associated with Alzheimer's disease. Nature 1988, 331:528530 6. Golde TE, Estus S, Usiak M, Younkin LH, Younkin SG: Expression of : amyloid protein precursor mRNAs: recognition of a novel alternatively spliced form and quantitation in Alzheimer's disease using PCR. Neuron 1990, 4:253-267 7. Kbnig G, Monning U, Czech C, Prior R, Banati R, Schreiter-Gasser U, Bauer J, Masters CL, Beyreuther K: Identification and differential expression of a novel alternative splice isoform of the 14 amyloid precursor protein (APP) mRNA in leukocytes and brain microglial cells. J Biol Chem 1992, 267:10804-10809 8. Glenner GG, Wong CW: Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun 1984, 120:885-890 9. Masters CL, Simms G, Weinman NA, Multhaup G, Mc-

Donald BL, Beyreuther K: Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci USA 1985, 82:4245-4249 10. Wasco W, Gurubhagavatula S, Paradis MD, Romano DM, Sisodia SS, Hyman BT, Neve RL, Tanzi RE: Isolation and characterization of APLP2 encoding a homologue of the Alzheimer's associated amyloid f protein precursor. Nat Genet 1993, 5:95-99 11. Slunt HH, Thinakaran G, von Koch C, Lo ACY, Tanzi RE, Sisodia SS: Expression of a ubiquitous, cross-reactive homologue of the mouse ,B-amyloid precursor protein (APP). J Biol Chem 1994, 269:2637-2644 12. Sandbrink R, Masters CL, Beyreuther K: Similar alternative splicing of a non-homologous domain in ,BA4amyloid protein precursor-like proteins. J Biol Chem 1994, 269:14227-14234 13. Thinakaran G, Sisodia S: Amyloid precursor-like protein 2 (APLP2) is modified by the addition of chondroitin sulfate glycosaminoglycan at a single site. J Biol Chem 1994, 269:22099-22104 14. Thinakaran G, Slunt HH, Sisodia SS: Novel regulation of chondroitin sulfate glycosaminoglycan modification of amyloid precursor protein and its homologue, APLP2. J Biol Chem 1995, 270:16522-16525 15. Pangelos MN, Efthimiopoulos S, Shioi J, Robakis NK: The chondroitin sulfate attachment site of appican is formed by splicing out exon 15 of the amyloid precursor gene. J Biol Chem 1995, 270:10388-10391 16. Thinakaran G, Kitt CA, Roskams AJI, Slunt HH, Masliah E, von Koch C, Ginsberg SD, Ronnett GV, Reed RR, Price DL, Sisodia SS: Distribution of an APP homolog, APLP2, in the mouse olfactory system: a potential role for APLP2 in axogenesis. J Neurosci 1995, 15:63146326 17. Benowitz Li, Rodriguez W, Paskevich P, Mufson EJ, Schenk D, Neve RL: The amyloid precursor protein is concentrated in neuronal lysosomes in normal and Alzheimer disease subjects. Exp Neurol 1989, 106:237250 18. Card JP, Meade RP, Davis LG: Immunocytochemical localization of the precursor protein for 3-amyloid in the rat central nervous system. Neuron 1988, 1:835-8846 19. Cras P, Kawai M, Siedlak S, Mulvihill P, Gambetti P, Lowery D, Gonzalez-DeWhitt P, Greenberg B, Perry G: Neuronal and microglial involvement in ,3-amyloid protein deposition in Alzheimer's disease. Am J Pathol 1990, 137:241-246 20. Hyman BT, Tanzi RE, Marzloff K, Barbour R, Schenk D: Kunitz protease inhibitor-containing amyloid f protein precursor immunoreactivity in Alzheimer's disease. J Neuropathol Exp Neurol 1992, 51:76-83 21. Martin LJ, Pardo CA, Cork LC, Price DL: Synaptic pathology and glial responses to neuronal injury precede the formation of senile plaques and amyloid deposits in the aging cerebral cortex. Am J Pathol 1994, 145: 1358-1381 22. Martin LJ, Sisodia SS, Koo EH, Cork LC, Dellovade TL, Weidemann A, Beyreuther K, Masters C, Price DL:

Localization of APLP-2 and APP

1095


23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.

34.

Amyloid precursor protein in aged nonhuman primates. Proc NatI Acad Sci USA 1991, 88:1461-1465 Schubert W, Prior R, Weidemann A, Dircksen H, Multhaup G, Masters CL, Beyreuther K: Localization of Alzheimer ,3A4 amyloid precursor protein at central and peripheral synaptic sites. Brain Res 1991, 563: 184-194 Shigematsu K, McGeer PL, McGeer EG: Localization of amyloid precursor protein in selective postsynaptic densities of rat cortical neurons. Brain Res 1992, 593: 353-357 Tate-Ostroff B, Majocha RE, Marotta CA: Identification of cellular and extracellular sites of amyloid precursor protein extracytoplasmic domain in normal and Alzheimer disease brains. Proc Natl Acad Sci USA 1989, 86:745-749 Topper R, Gehrmann J, Banati R, Schwarz M, Block F, Noth J, Kreutzberg GW: Rapid appearance of p3-amyloid precursor protein immunoreactivity in glial cells following excitotoxic brain injury. Acta Neuropathol (Berl) 1995, 89:23-28 Khatchaturian ZA: Diagnosis of Alzheimer's disease. Arch Neurol 1985, 42:1097-1105 Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee L, Vogel FS, Hughes JP, van Belle G, Berg L: The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). II. Standardization of the neuropathological assessment of Alzheimer's disease. Neurology 1991, 41:479-486 Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979, 18:5294-5299 Blackstone CD, Moss SJ, Martin LJ, Levey Al, Price DL, Huganir RL: Biochemical characterization and localization of a non-N-methyl-D-aspartate glutamate receptor in rat brain. J Neurochem 1992, 58:1118-1126 Sisodia SS, Koo EH, Hoffman PN, Perry G, Price DL: Identification and transport of full-length amyloid precursor proteins in rat peripheral nervous system. J Neurosci 1993, 13:3136-3142 Hsu SM, Raine L, Fanger H: Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase technique: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem 1981, 29:577-580 Sprecher CA, Grant FJ, Grimm G, O'Hara PJ, Norris F, Norris K, Foster DC: Molecular cloning of the cDNA for a human amyloid precursor protein homolog: evidence for a multigene family. Biochemistry 1993, 32:44814486 Yamazaki T, Yamaguchi H, Kawarabayashi T, Hirai S: Ultrastructural localization of amyloid f3/A4 protein pre-

cursor in the normal rat brain. Virchows Arch B Cell Pathol 1993, 63:173-180 35. Brion JP, Couck AM, Bruce M, Anderton B, FlamentDurand J: Synaptophysin and chromogranin A immunoreactivities in senile plaques of Alzheimer's disease. Brain Res 1991, 539:143-150 36. Munoz DG: Chromogranin A-like immunoreactive neurites are major constituents of senile plaques. Lab Invest 1991, 64:826-832 37. Munoz DG, Wang D: Tangle-associated neuritic clusters: a new lesion in Alzheimer's disease and aging suggests that aggregates of dystrophic neurites are not necessarily associated with ,B/A4. Am J Pathol 1992, 140:1167-1178 38. Masliah E, Mallory M, Hanesen L, Alford M, Albright T, DeTeresa R, Terry R, Baudier J, Saitoh T: Patterns of aberrant sprouting in Alzheimer's disease. Neuron 1991, 6:729-739 39. Lenders MB, Peers MC, Tramu G, Delacourte A, Defossez A, Petit H, Mazzuca M: Dystrophic peptidergic neurites in senile plaques of Alzheimer's disease hippocampus precede formation of paired helical filaments. Brain Res 1989, 481:344-349 40. Snow AD, Mar H, Nochlin D, Kimata K, Kato M, Suzuki S, Hassell J, Wight TN: The presence of heparan sulfate proteoglycans in the neuritic plaques and congophilic angiopathy in Alzheimer's disease. Am J Pathol 1988, 133:456-463 41. Snow AD, Mar H, Nochlin D, Sekiguchi RT, Kimata K, Koike Y, Wight TN: Early accumulation of heparan sulfate in neurons and in the f-amyloid protein-containing lesions of Alzheimer's disease and Down's syndrome. Am J Pathol 1990, 137:1253-1270 42. Su JH, Cummings BJ, Cotman CW: Localization of heparan sulfate glycosaminoglycan and proteoglycan core protein in aged brain and Alzheimer's disease. Neuroscience 1992, 51:801-813 43. DeWitt DA, Silver J, Canning DR, Perry G: Chondroitin sulfate proteoglycans are associated with the lesions of Alzheimer's disease. Exp Neurol 1993, 121:149-152 44. Canning DR, McKeon RJ, DeWitt DA, Perry G, Wujek JR, Frederickson RCA, Silver J: /3-amyloid of Alzheimer's disease induces reactive gliosis that inhibits axonal outgrowth. Exp Neurol 1993, 124:289-298 45. Brunden KR, Richter-Cook NJ, Chaturvedi N, Frederickson RCA: pH-dependent binding of synthetic 3-amyloid peptides to glycosaminoglycans. J Neurochem 1993, 61:2147-2154 46. Shaffer LM, Dority MD, Guptabansal R, Frederickson RCA, Younkin SG, Brunden KR: Neurobiol Aging 1995, 16:737-745 47. Gupta-Bansal R, Frederickson RCA, Brunden KR: J Biol Chem 1995, 270:18666-18671