APP gene family: unique age-associated changes in splicing of Alzheimer's βA4-amyloid protein precursor

Neurobiology of Disease 1994, 1, 13–24

Rupert Sandbrink 1 Colin L. Masters 2 Konrad Beyreuther 1

APP gene family: unique age-associated changes in splicing of Alzheimer’s βA4-amyloid protein precursor 1

Zentrum für Molekulare Biologie Heidelberg (ZMBH), University of Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany 2 Department of Pathology, University of Melbourne, Parkville,Victoria 3052, Australia

Summary

The βA4-amyloid protein precursor (APP) is a transmembrane glycoprotein that is the source of the characteristic βA4-amyloid deposits of Alzheimer brains. It exists in eight isoforms generated by alternative splicing of exons 7, 8 and 15, of which the LAPP mRNAs lacking exon 15 are significantly expressed in non-neuronal cells and tissues, but not in neurones. Recently, it was shown that APP is a member of a multigene family of which the amyloid precursor-like protein 2 (APLP2) is the nearest relative. Analysis of APLP2 expression revealed regulated alternative splicing of the Kunitz protease inhibitor domain (KPI, homologous to exon 7 of APP) and a nonhomologous insert of 12 amino acids on the NH2-terminal side of the transmembrane domain.While expression of the KPI encoding exon of APLP2 is abundant in neurones and thus differs from APP, L-APLP2 mRNA isoforms lacking the latter, non-homologous insert show a tissue-specific expression pattern similar to that of exon 15 of APP. Comparison of alternatively spliced APP and APLP2 mRNA isoforms in rat brain regions from early post-natal and adult rats revealed significantly higher relative amounts of KPI-encoding APP isoforms in the adult rat brain and an even more pronounced augmentation of L-APP mRNAs. Both effects were not observed for APLP2.This indicates an APP-specific age-associated regulation pattern within the APP gene family which has intriguing implications for the development of Alzheimer’s disease in humans.

Key words

Alzheimer’s disease, APP gene family, βA4-amyloid protein precursor, amyloid precursor-like protein, alternative splicing, age-association of isoform expression, amyloidogenicity Received 7 April 1994, accepted for publication 20 June 1994.

Introduction Alzheimer’s disease (AD) is a neurodegenerative disorder histopathologically characterized by the abundance of proteinaceous deposits in the cerebral cortex and hippocampus of affected individuals, extracellularly as senile plaques and intracellularly as neurofibrillary tangles (Müller-Hill & Beyreuther 1989). The major component in plaques is the βA4-amyloid protein, a ~4 kD amyloidogenic peptide (Glenner & Wong 1984, Masters et al. 1985), derived from a larger transmembrane glycoprotein termed the βA4-amyloid protein precursor (APP) (Kang et al. 1987). APP constitutes a family of alternatively spliced isoforms with 770,

Correspondence: Rupert Sandbrink, Zentrum für Molekulare Biologie Heidelberg (ZMBH),University of Heidelberg,Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany. Copyright © 1994 Academic Press, Inc. All rights of reproduction in any form reserved.

752, 751,733, 714, 696, 695 and 677 amino acids as a result of alternative splicing of exons 7, 8 and 15 (Sandbrink et al. 1994a). The four largest isoforms contain a domain structurally and functionally homologous to the Kunitz family of protease inhibitors (KPI; Kitaguchi et al. 1987, Ponte et al. 1988, Tanzi et al. 1988) that is encoded by exon 7 of the APP gene (Yoshikai et al. 1990). Exon 8 encoding a domain with homology to the MRC OX-2 antigen (Clark et al. 1985) is also alternatively spliced: this domain is present in the two largest APP isoforms and in the two only weakly expressed APP isoforms with 714 and 696 amino acids (Golde et al. 1990, Kang & Müller-Hill 1990). Exon 15 was then identified as the third alternatively spliced exon in APP transcripts encoding for transmembrane APP isoforms. This exon encodes a domain preceding the NH2-terminus of the βA4-region of APP, encoded 13

14

R. Sandbrink et al.

within the adjacent exons 16 and 17, by only 16 amino acids.The APP transcripts excluding exon 15 were first discovered in peripheral leukocytes and immunocompetent cells of the brain, therefore denoted leukocyte-derived APP (L-APP) mRNAs (König et al. 1992, Mönning et al. 1992). It was then shown that all four possible APP mRNA isoforms without exon 15 are indeed expressed in vivo, i.e. the L-APP752, L-APP733, L-APP696, and L-APP677 mRNA isoforms (Sandbrink et al. 1994a). In peripheral rat tissues, these L-APP mRNA isoforms represent between 25% (skeletal muscle) and approx. 70% (aorta, pancreas) of total APP transcripts. In the rat central nervous system, where LAPP expression accounts for about 4% of total APP mRNA, non-neuronal tissues like meninges, plexus choroideus and brain vessels also significantly express LAPP transcripts. While primary cultured astroghal cells equally contain high portions of L-APP mRNA, primary cultured septal neurones were shown not to express detectable levels of L-APP transcripts (Sandbrink et al. 1994a). In cultured cells, APP matures through the constitutive secretory pathway and is modified by the addition of Nand O-linked oligosaccharides and tyrosine sulfation (Weidemann et al. 1989). Full length APP appears on the plasma membrane and, concomitantly, soluble C-terminal truncated derivates can be detected in the conditioned medium generated by proteolytic cleavage within the βA4peptide region (Weidemann et al. 1989, Esch et al. 1990, Sisodia et al. 1990). Additionally, βA4-protein is also released into the medium, although the biochemical mechanism(s) involved in its generation are presently unknown (Haass et al. 1992; Seubert et al. 1992; Shoji et al. 1992, Dyrks et al. 1993). A distinct physiological role for the ubiquitously distributed APP has not yet been determined. Several putative functions have been ascribed for the transmembrane and/or secreted APPs (discussed in Mönning et al 1992). Apart from the putative function of the KPI insert (exon 7) as protease inhibitor, however, the functional significance of alternative splicing of exon 8 and 15 has not been elucidated yet. Recently, cDNAs were isolated coding for proteins related to APP, indicating that there is a multigene family of APP-like proteins in mammals. However, the βA4 domain is unique to APP and absent in all other members of the superfamily, including APP-related proteins of dipteras (Rosen et al. 1989) and nematodes (Daigle & Li 1993). From a mouse brain library, a cDNA was cloned that encodes a protein, the predicted amino acid sequence of which is 42% identical to that of APP (Wasco et al. 1992). This 653-amino acid protein, similar to APP in overall structure, is now denoted as amyloid precursor-like protein 1 (APLP1). More recently, a human full-length cDNA was identified encoding for another protein, which is the nearest APP-relative detected so far.This molecule, which has a very similar overall domain organization to APP including a KPI

Fig. 1. The structure of APP and APLP2.The domain structures of full length rat APP (APP770) and rat APLP2 (rAPLP2 765) are illustrated. Regions of high homology are indicated by diagonal hatching or dotted. Alternatively spliced exons are also marked. The cDNA fragments amplified by APP and APLP2 RT-PCR analysis are indicated, and for APP, the Bgl I site used for digestion of PCR products is shown. SP denotes the presumed signal peptide (crosshatched); Cys, the conserved cysteine-rich region; E/D, the less conserved acidic domain; KPI, the Kunitz protease inhibitor domain; OX-2, the MRC OX-2 domain of APP, CT, cytoplasmatic domain, βA4, the βA4-amyloid region of APP (black bar), while Tm and dark striped bars indicate the putative transmembrane domains. For APP, boundaries of regions encoded by individual exons are shown.The region with the highest divergence is indicated, white in APP, and grey in APLP2. Asterisks mark the positions where rat APLP2 differs from its human and murine homologues by a deletion of two amino acids and an insertion of four amino acids.

domain, was termed amyloid precursor protein homologue (APPH) (Sprecher et al. 1993) or amyloid precursor-like protein 2 (APLP2) (Wasco et al 1993) (Fig. 1). As identified by Sprecher et al. (1993) it consists of 763 amino acids. However, partial cDNAs of a murine CDE1 binding protein (Vidal et al 1992) and a rat sperm membrane protein YWK-II (Yan et al 1990) had been described before this, and were then shown to represent 5'-truncated versions of the rodent APLP2 homologues (Sandbrink et al. 1994b). Meanwhile, the complete coding sequences of murine APLP2 (Hanes et al. 1993, Slunt et al. 1994) and rat APLP2 (Sandbrink et al. 1994c) have also been established. Unique to the rat APLP2 sequence is a deletion of two codons and an insertion of four codons within the acidic domain as compared with its human and murine homologues (Fig. 1). Therefore, the total length of full length rAPLP2 is 765 amino acids, two amino acids longer than that of hAPLP2 and mAPLP2. On the protein level, there is an overall identity of APLP2 with APP of ~50%. The highest homologies are found within two regions of the extracellular part of APLP2 as well as the transmembrane and intracellular domain (each with ~70% amino acid identity). However, directly adjacent to the NH2-terminus of the presumed transmembrane domain, there is stretch of 113 amino acids highly divergent to APP and to the other members of the gene family Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Age-associated splicing of APP and APLP2

(Fig. 1). As a consequence, a βA4-protein-like sequence is also not present within APLP2. Very recently, alternative splicing of APLP2 was characterized in detail by analysis of perfused rat tissues (Sandbrink et al. 1994b). We were able to show that there are two alternatively spliced inserts, i.e. the Kunitz protease inhibitor domain and a 12 amino acids encoding region on the NH2-terminal side of the transmembrane domain, which is part of the region of highest divergence between APP and APLP2. Analysis of the tissue-specific differential expression of the resulting four APLP2 mRNA isoforms revealed that isoforms lacking the latter, non-homologous insert are highly expressed in non-neuronal tissues and in primary cultured astrocytes, but only weakly in primary cultured septal neurones. This resembles the tissue specific alternative splicing of exon 15 of APP. Therefore, APLP2 isoforms lacking this 12 ammo acids region were denoted as L-APLP2 isoforms (Sandbrink et al. 1994b). In contrast to this, expression of the KPI encoding exon of APLP2 is abundant in both neuronal and non-neuronal tissues and thus different from APP mRNA expression patterns. In this study, we confirm and extend our previous results regarding the differential expression of APLP2 and APP transcripts by analysing different brain regions from early post-natal and adult rats. We detected a similar distribution of total APLP2 and total APP mRNA and subsequently analysed the relative contribution of the different APLP2 and APP mRNA isoforms. Apart from the similar brainspecific low frequency of L-APP and L-APLP2 transcripts, we observed a subtle but specific age-associated increase in the relative amount of L-APP mRNA in brain which was not seen with L-APLP2. Because age is the major risk factor for AD, this finding might be of potential significance for understanding the pathogenesis of the disease at the molecular level.

Material and methods Tissue preparation and primary cell culture Male Wistar rats, 7 months old, were killed by CO2-inhalation and immediately transcardially perfused with ice-cold phosphate-buffered saline. P6 rats (day of birth counted as day 0) were killed by decapitation and not perfused.Analysis of expression patterns for both APP and APLP2 had previously shown that the results are virtually independent of perfusion (Sandbrink et al 1994 a, b). Different brain regions were dissected immediately and frozen quickly in liquid nitrogen. Primary neuronal cultures were prepared from enzymatically dissociated hippocampal or septal cells of E17 rat brain or from cortical cells of E14 brain (Sandbrink et al. 1994a). Preparation of astrocyte-enriched cultures and primary culture conditions for microglial cells were performed as described (Banati et al. 1993, Sandbrink et al. 1994a). Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

15

Isolation of RNA, reverse transcription and polymerase chain reaction (RT-PCR) Total RNA extraction, preparation of cDNA using oligo(dT)17 as primer and polymerase chain reaction (PCR) amplification were performed exactly as described (Sandbrink et al. 1994a). As APP-specific PCR primers were used (Sandbrink et al. 1994a): sr841: 5'-GAGTCTGTGGAGGAGGTAGTCC-3'; and αr1970: 5'-CTTCTGTCTTGATGTTTGTCAACC3'. The lengths of the amplified cDNA fragments corresponding to the eight alternatively spliced isoforms are: 1153 bp for APP770; 1099 bp for L-APP752; 1096 bp for APP751; 1042 bp for L-APP733, 985 bp for APP714; 931 bp for L-APP696; 928 bp for APP695; 874 bp for LAPP677. The APLP2-specific primer pair was (Sandbrink et al. 1994b): sml838: 5'-GGAGACGATTACAATGAGGAGAATCC-3'; and αrl1877: 5'-CCATCTTGCTCTGCCATTCCAG-3'. The lengths of the amplified cDNA fragments corresponding to the four alternatively spliced isoforms are: 1061 bp for rAPLP2 765; 1015 bp for rL-APLP2 753; 893 bp for rAPLP2 709; 857 bp for rL-APLP2 697. 5'-Endlabelling of αr1970 and αrl1877 was performed with polynucleotide kinase and β-[33P]-ATP. Undigested PCR products and Bgl I digested amplified APP cDNA fragments were analysed by denaturing polyacrylamide gel electrophoresis (4.0%) and autoradiography. For quantification of individual bands, a Phosphorlmager was employed.

Northern blot analysis Ten micrograms of total RNA from rat brain regions were separated by denaturing formaldehyde agarose gel electrophoresis (1%) and transfered to a nylon membrane using standard protocols (Sambrook et al. 1989). Hybridization was performed by use of [32P] cDNA fragments obtained by random primed labelling. Hybridization conditions were 0.5M sodium phosphate pH 7.2, 7% sodium dodecylsulfate and 1% BSA for 16 h at 65°C, washing conditions were 40 mM sodium phosphate pH 7.2 and 1% sodium dodecylsulfate at 65°C for a total period of 1–2 h during which the washing solution was changed five times. All hybridizations were performed using the same filter which was stripped in between by incubation in 1% sodium dodecylsulfate and 1 mM EDTA, pH 8.0 for 1 h at 80°C. For detection of APP mRNA, the 1.05 kb Eco R1 fragment of human APP was used. APLP2 mRNA was detected using a Ball-Fspl fragment of 542 bp obtained by digestion of PCR-amplified rAPLP2 cDNA fragments. This probe is virtually identical to the non-homologous region of APLP2 on the NH2-terminal side of the transmembrane domain. For standardization, a hybridiziation experiment using a human glyeraldehyde- 3-phosphate dehydrogenase (GAPDH) cDNA fragment was also performed. Quantification of hybridization signals was carried out by PhosphorImaging.

16

R. Sandbrink et al.

Results Neurone-specific alternative splicing patterns of APP and APLP2 mRNA isoforms To illustrate and to expand our previous results of tissuespecific APP and APLP2 mRNA isoform expression, we performed a series of polymerase chain reactions from reverse transcribed RNA (RT-PCR) of perfused rat tissues and primary cultured rat brain cells. For this, 2 µg of total RNA preparations were reverse transcribed using oligo(dT)17 as primer, and aliquots were amplified using a set of either APP-specific or APLP2-specific PCR primers (Fig. 1). For later quantification, the antisense primers were labelled radioactively with γ-[33P]-ATP (Sandbrink et al. 1994 a, b). Amplified cDNA fragments were then separated electrophoretically and analysed by autoradiography and PhosphorImaging.Quantitative data obtained by this method have previously been shown to be highly reproducible and accurate (Sandbrink et al. 1994a, b). In Fig. 2 (top), the distribution of APP mRNA isoforms is shown. The eight possible APP mRNA isoforms are detected in this assay as six different bands, since APP751 and L-APP752 as well as APP695 and L-APP696 cDNA fragments remain unresolved on the gel. The upper three bands represent the KPI encoding APP mRNA isoforms. For assessment of total L-APP mRNA content, a digestion of the PCR products with Bgl I was performed cutting in between the two alternatively spliced regions (Fig. 1). Subsequently, only the 3'-fragments are detected, representing the exon 15 encoding and the L-APP transcripts (Fig. 2, middle) Individual APLP2 PCR products are detected as four different bands (Fig. 2, bottom). Of these, rAPLP2 765 (equivalent to human APLP2 763) and rLAPLP2 753 (equivalent to human L-APLP2 751) are the KPI encoding isoforms, while the lower two bands represent the isoforms without the KPI exon (Sandbrink et al. 1994b, c). APLP2 transcripts lacking the second alternatively spliced exon within the non-homologous part of APLP2 are indicated by the prefix ‘L-’ in analogy to L-APP denotation. With cDNA preparations from liver, kidney, and lung, the typical APP mRNA expression pattern for non-neuronal tissues is observed: KPI encoding isoforms are predominantly expressed (~90% of total APP mRNA), and LAPP transcripts are significantly expressed, contributing between 35 and 63%. The APLP2 expression patterns in these tissues show an expression of the KPI encoding exon of 68–84%, while the L-APLP2 isoforms are even more abundant than L-APPs with contributions from 63–84%. In skeletal muscle, high portions of KPIencoding transcripts of both APP and APLP2 were found (each >90%), while both L-APP (25%) and L-APLP2 (35%) were lowest in this tissue among all peripheral tissues analysed (Sandbrink et al. 1994 b, c). In adult rat brain which is represented in Fig. 2 by analysis of a cortex-cDNA preparation, significantly different

Fig. 2. RT-PCR analysis of APP and APLP2 mRNA isoform expression in rat tissues and primary cultured brain cells 33Pradiolabelled RT-PCR products were analysed by denaturing polyacrylamide gel electrophoresis and analysed by autoradiography and PhosphorImaging: top, APP; bottom, APLP2. For APP, PCR products were also analysed after Bgl I digestion (middle). Detected APP and APLP2 mRNA isoforms are indicated, whereby (L-)APP751/752 refers to unresolved APP751+L-APP752 and (L-)APP695/696 to unresolved APP695+L-APP696 mRNAs Hc, hippocampus; Sp, septum; Cx, cortex; and Sk, skeletal.

expression patterns of APP and APLP2 are observed. For APP, only small amounts of KPI encoding isoforms are detected (11%), but 82% of APLP2 mRNA are expressed as KPI encoding transcripts. In regard to L-APP and L-APLP2, however, a similar, low expression is observed (4 and 11%). The brain specific splice patterns of APP and APLP2 are due to the typical neuronal expression patterns, as shown in Fig. 2. In primary cultured neuronal cells from embryonic rat brain, cultured in vitro for 8 days, virtually undetectable amounts of L-APP mRNA (<1%) and only small amounts of L-APLP2 mRNA (9–13%) are observed.The results are the same for neuronal cultures from cortex, hippocampus and septum. In all three preparations, only low amounts of KPI encoding APP isoforms are observed (~6%), while KPI encoding APLP2 transcripts are expressed to ~85%. The contribution by non-neuronal cells was characterized by analysis of primary cultured astrocytes and microglial cells. For both cell types, a very similar expression pattern was observed, both for APP and APLP2 expression. KPI encoding isoforms of APP were expressed to ~95%, those of APLP2 were assessed to ~70%. L-APP transcripts were present in 40–50% of total APP mRNA, and L-APLP2 mRNAs were even more abundant (~70%). This very much resembles the expression pattern of peripheral tissues. Hence, primary cultured glial cells and neurones Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Age-associated splicing of APP and APLP2

17

same blot, P6 and 7M data can be compared directly with each other. The levels of APP were highest in hippocampus, not only in adult brain but also in P6 brain. In adult brain, high levels of APP expression were also observed in cerebellum and in cortex, low levels were found especially for the olfactory bulb and the septum. In the latter brain regions of early postnatal rats, however, relatively higher levels of APP

Fig. 3. Northern blot analysis of APP and APLP2 mRNA expression in brain regions from P6 (postnatal day 6) and 7M (7 months old) rats: Autoradiograms Cx, cortex; Cb, cerebellum; Hc, hippocampus; BO, olfactory bulb; Sp, septum; Str, striatum; Th, thalamus; Po, pons; MO, Medulla oblongata. Molecular weight standards arc indicated, with the 28 S rRNA shown as 4.7 kb and the 18 S rRNA as 1.8 kb standard. Shown beneath is a control hybridization with a glyeraldehyde-3-phosphate dehydrogenase cDNA (GAPDH).The autoradiographs are from independent hybridizations to the same filter.

are similar in regard to the predominance of KPI-encoding APLP2 transcripts, but differ in alternative splicing of the more downstream, 12 amino acid encoding region.

Total APP and APLP2 mRNA expression in different brain regions We now analysed and compared the expression of APP and APLP2 mRNA in different brain regions of early post-natal and adult rats. First, we determined the distribution of total APP and APLP2 transcripts in brains of 6-day-old rats (P6), in which extensive synaptogenesis occurs or begins, and of adult, 7-month-old rats (7M) by Northern bloc analysis on total RNA from nine different brain regions.The same blot was hybridized to cDNA fragments of APP and APLP2 thus allowing a direct comparison of expression of the two genes. Additionally, a control hybridization to a human GAPDH cDNA fragment was performed. Direct inspection of the autoradiograms (Fig. 3) reveals a very similar distribution of APP and APLP2 within the brain regions analysed. APP mRNA was detected as a band at ~3.2 kilobases (kb) while the APLP2 message is slightly larger and ~3.8–4 kb in size.We then quantified the hybridization signal by means of a PhosphorImager and standardized on the GAPDH signal, the results are given in Table 1 and illustrated in Fig. 4. For comparison of APP with APLP2 data, all figures were normalized with the average of each hybridization series being set to 100%. Since P6 and 7M mRNA preparations were analysed on the Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Fig. 4. Results from Northern Blot analysis of APP and APLP2 mRNA expression in P6 and 7M brain regions. Quantitative data from PhosphorImaging analysis of the Northern blots shown in Fig. 3 were expressed as APP/GAPDH ratio (a) and APLP2/GAPDH ratio (b). All values were normalized with the mean value of each series set as 100%. (c) Relative change from P6 to 7M,a calculated in %. All abbreviations as in Fig. 3.

Table 1. Differential expression of APP and APLP2 mRNA in brain regions from P6 (post-natal day 6) and 7M (7 months old) rats. RT-PCR: Radiolabelled and electrophoretically separated RT-PCR products (Fig. 5) were analysed on a PhosporImager and relative amounts calculated as a percentage (% of total). KPI-APP was calculated as the sum of APP770, APP751+L-APP752 and L-APP733 bands, while L-APP values were obtained from analysis of the Bgl I digested APP PCR products. APP751 est. and L-APP752 est. are estimations for the relative amounts of these isoforms (Sandbrink et al. 1994a). L-APLP2 denotes the sum of rL-APLP2 753 and rLAPLP2 697, while KPI-APLP2 represents the KPI domain encoding APLP2 mRNA isoforms (sum of rAPLP2 765 and rL-APLP2 753). Also given in the table are the quantification results from Northern blot analysis (Fig. 3). *Total APP and APLP2 values were normalized with the average of each hybridization series set to 100%, denoted as, % of av. Cx, cortex; Cb, cerebellum; Hc, hippocampus; BO, olfactory bulb; Sp, septum; Str, striatum;Th, thalamus; Po, pons; MO, Medulla oblongata. P6-Brain Region

Cx

Cb

Hc

BO

Sp

Str

Th

Po

MO

Northern Blot Total APP (% of av.) 74 Total APLP2 (% of av.) 85 APLP2/APP ratio 1.15

94 147 1.56

145 117 0.81

127 135 1.06

108 120 1.11

75 64 0.85

100 68 0.68

112 96 0.86

58 65 1.12

RT-PCR (% of total) APP770 APP751+L-APP752 L-APP733 APP714 APP695+L-APP696 L-APP677: appr.

3.0 3.1 0.4 0.7 92.8 0.1

10.1 12.2 0.5 1.0 76.0 0.1

3.8 4.5 0.4 0.6 90.6 0.1

7.0 7.5 1.0 0.5 83.7 0.3

6.1 6.9 0.9 0.3 85.7 0.1

5.5 5.6 0.8 0.4 87.5 0.1

3.4 3.7 0.5 0.4 91.8 0.2

4.4 4.7 0.5 0.5 89.8 0.1

7.0 8.4 0.7 0.5 83.2 0.2

Sum KPI-APP Total L-APP

6.4 1.5

22.9 1.9

8.7 1.5

15.6 3.0

13.8 2.2

11.9 2.1

7.6 1.4

9.6 1.3

16.2 2.0

L-APP752 est. APP751 est.

0.8 2.2

1.1 11.0

0.9 3.6

1.6 5.9

1.2 5.7

1.0 4.5

0.6 3.1

0.6 4.1

1.0 7.4

APLP2 765 L-APLP2 753 APLP2 709 L-APLP2 697

79.2 11.6 6.8 2.4

74.4 9.6 13.5 2.5

69.2 14.4 11.7 4.7

70.3 17.1 8.3 4.3

68.4 11.2 16.3 4.0

77.8 12.2 7.7 2.3

67.2 12.4 16.8 3.5

54.2 10.5 30.8 4.6

48.0 9.3 37.2 5.5

Sum KPI-APLP2 Sum L-APLP2

90.8 14.0

84.0 12.2

83.6 19.1

87.4 21.4

79.6 15.3

90.9 14.5

79.6 15.9

64.6 15.1

57.3 14.8

Northern Blot Total APP (% of av.) 136 Total APLP2 (% of av.) 127 APLP2/APP ratio 0.93

162 198 1.22

164 136 0.83

50 69 1.38

45 60 1.33

73 78 1.07

93 83 0.89

78 60 0.77

108 93 0.86

RT-PCR (% of total) APP770 APP751+L-APP752 L-APP733 APP714 APP695+L-APP696 L-APP677: appr.

2.5 7.3 1.7 0.3 88.0 0.3

2.4 5.9 0.9 0.4 90.2 0.2

2.1 7.5 1.6 0.2 88.4 0.2

5.1 12.2 4.3 0.2 77.5 0.7

5.5 15.7 4.0 0.2 73.7 0.8

4.1 11.5 2.9 0.3 80.6 0.5

2.8 9.5 2.1 0.4 84.8 0.4

3.7 16.4 2.3 0.5 76.5 0.6

3.8 15.3 2.2 0.5 77.3 0.8

Sum KPI-APP Total L-APP

11.4 4.1

9.2 2.6

11.2 4.1

21.6 11.2

25.3 9.2

18.5 7.0

14.5 5.2

22.4 6.1

21.4 5.8

L-APP752 est. APP751 est.

2.0 5.2

1.4 4.5

2.2 5.3

6.1 6.1

4.3 11.4

3.5 8.0

2.6 6.9

3.2 13.2

2.7 12.6

APLP2 765 L-APLP2 753 APLP2 709 L-APLP2 697

74.3 7.7 13.9 4.1

74.7 6.3 16.6 2.4

79.6 6.3 10.9 3.1

70.4 15.0 8.4 6.2

71.1 8.9 15.2 4.7

77.7 8.0 9.0 5.3

63.9 9.9 21.3 4.9

31.2 7.5 49.8 11.5

30.3 7.1 52.9 9.6

Sum KPI-APLP2 Sum L-APLP2

82.0 11.8

81.0 8.7

86.0 9.5

85.4 21.2

80.0 13.7

85.7 13.3

73.8 14.8

38.7 19.0

37.4 16.7

7M-Brain Region

Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Age-associated splicing of APP and APLP2

19

remarkedly resembles that of APP, and the changes in total APP and APLP2 expression from 6 days to 7 months are very similar for both genes (Fig. 4c). Differential expression of APP and APLP2 mRNA isoforms in different brain regions

Fig. 5. RT-PCR analysis of APP and APLP2 mRNA isoform expression in brain regions from (a) P6 and (b) 7M rats. 33 P-radiolabelled RT-PCR products were analysed by denaturing polyacrylamide gel electrophoresis and analysed by autoradiography and PhosphorImaging: top, APP; bottom, APLP2. For APP, PCR products were also analysed after Bgl I digestion (middle). Detected APP and APLP2 mRNA isoforms are indicated. Abbreviations as in Fig.3.

mRNA expression were detected, while in P6 cortex and cerebellum APP expression was much lower. In contrast to APP, APLP2 is maximally expressed in cerebellum, already in P6 brain, but especially among the brain regions of the 7 month-old rat Strong hybridization was also observed in P6 hippocampus, olfactory bulb and septum. Similar to APP, the latter two regions of 7M brain show only weak hybridization to the APLP2 cDNA fragment, while at this age APLP2 expression is increased in cortex. Overall, APLP2 reveals a pattern of expression that Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

The described neurone-specific expression pattern comprising almost undetectable amounts of L-APP transcripts and only small amounts of L-APLP2 mRNA determines the alternative splice pattern of APP and APLP2 in brain that is unique among all tissues analysed (Sandbrink et al. 1994 a, b). The distribution of individual APP and APLP2 mRNA isoforms in different brain regions was analysed by RT-PCR. In Fig. 5, autoradiograms of the PCR products are shown after electrophoretic separation in a denaturing urea-polyacrylamide gel. All four APLP2 mRNA isoforms were directly quantified from the gel by use of a PhosphorImager, while APP-PCR products were analysed both undigested and after Bgl I digestion for assessment of total L-APP mRNA as described above. From this, an estimation of the contributions of APP751 and L-APP752 among the unresolved APP751+L-APP752 band was also performed as previously described (Sandbrink et al. 1994a). Quantification results are given in Table 1. In adult rat brain, APP695 is the APP transcript most strongly expressed. ~90% in cerebellum, but only ~74% in septum. As next, APP7751 mRNA contributes between ~4% (cerebellum) and ~13% (brain stem regions), in most areas about twice as much as APP770. APP714 is expressed in all 7M brain regions analysed to no more than 0.5%, which is slightly weaker than in P6 brain (for instance, in P6 cerebellum, APP714 contributes to ~1%). The moderately stronger expression of exon 8 containing APP transcripts in early postnatal rat brain is also reflected by the relative amounts of APP770, which are relatively higher in all P6 brain regions than in adult rat brain and about as strong as the APP751 mRNA isoform. Among the L-APP transcripts L-APP752 and L-APP733 isoforms are expressed in adult rat brain weakly, although the levels are detectable. They are most strongly expressed in the olfactory bulb and in the septum (each ~4% or more), while the weakest expression was observed in cerebellum (both ~1%). In contrast to APP695, the corresponding APP mRNA isoform lacking exon 15, i.e. L-APP677 mRNA, is expressed in adult rat brain only very weakly, with a similar distribution as the other L-APP mRNA isoforms. Among the APLP2 mRNA isoforms, the full length rAPLP2 765 transcripts are predominent (~75%) in all brain regions except for brain stem areas. In the latter, the corresponding APLP2 isoform lacking the KPI-encoding exon, rAPLP2 709, is expressed strongly, with ~50% of total APLP2 message. This is in excellent agreement with the previously described low portion of KPI-encoding APLP2 transcripts in the spinal cord (Sandbrink et al. 1994b). The two L-APLP2 transcripts are both moderately expressed,

20

R. Sandbrink et al.

the weakest expression was observed in cerebellum similar to the described L-APP distribution. Again, the KPIencoding L-APLP2 transcript, rL-APLP2 753, was more strongly expressed in most brain regions than the isoform lacking the KPI exon, but not in brain stem areas. The strongest expression of L-APLP2 transcripts was detected in olfactory bulb, both for early post-natal and adult rat brain, which again is very similar to the regional distribution of LAPP mRNA.

Different regulation of the expression of KPI-encoding Exons in APP and APLP2 For easier comparison between APP and APLP2 as well as between P6 and 7M brain regions, the total contributions of all KPI encoding and all `L-’ isoforms were calculated for both APP and APLP2 (Table 1). In Fig. 6, these data are visualized, top, APP pattern (Fig. 6 a, b), bottom, APLP2 (Fig. 6 c, d), for each, KPI isoforms (Fig. 6 a, c), and L-isoforms (Fig. 6 b, d). High levels of total KPI encoding APP isoforms in adult rat brain (all more than 20%) are detected in the olfactory bulb, septum and in the brain stem areas (pons and medulla oblongata), the lowest level was detected in cerebellum (below 10%). For each of the corresponding brain regions in early post-natal rat brain, a significantly lower portion of KPI-APP transcripts was observed. There is one exception, cerebellum: For this, in P6 brain, the highest level of KPI-encoding APP transcripts was detected: more than twice as high as in adult brain.This high amount of KPI-encoding APP transcripts in P6 cerebellum coincides with the late differentiation of granule cells and cerebellar synaptogenesis (Leclerc et al. 1989). The consistent increase of KPI-encoding APP transcripts from P6 to 7M brain in all other regions than cerebellum is associated with a rather conserved regional distribution pattern among the different brain areas (Fig. 6a). For APLP2, a completely different picture is observed for KPI-encoding isoforms (Fig. 6c). Apart from brain stem regions like pons and medulla oblongata, KPI-APLP2 transcripts contribute to ~80% of all APLP2 mRNA.This portion remains virtually unchanged in P6 and 7M brain regions. However, in the two brain stem regions of P6 brain the levels of KPI-APLP2 are not yet as low as in the adult brain. For the cerebellum of P6 brain, no selective increase in KPI-encoding APLP2 message was observed in contrast to KPI-APP isoforms.

Fig. 6. Results from RT-PCR analysis of APP and APLP2 mRNA isoform expression in P6 and 7M brain regions. KPI-APP (a), KPI-APLP2 (c), and L-APLP2 (d) were calculated as sum of the corresponding isoforms as described in Table 1, L-APP (b) values were obtained from analysis of the Bgl 1 digested APP PCR products. Abbreviations as in Fig 3. Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Age-associated splicing of APP and APLP2

Age-associated increase in the relative amount of L-APP mRNA is not observed with L-APLP2 Both L-APP and L-APLP2 transcripts are only weakly to moderately expressed in all brain regions analysed (Fig.6 b,d). The strongest regional expression of `L-isoforms’ of both genes is seen in the olfactory bulb of both early post-natal and adult rat brain. For APP, however, there is a drastic increase in the relative amount of L-APP message during development from post-natal day 6 to 7 months. Increased amounts of L-APP transcripts are observed for all brain regions analysed:The mean value in the adult rat brain was ~6%, while in P6 brain only 1.9% of total APP mRNA were expressed in average as L-APP. Nevertheless, the regional pattern was rather conserved (Fig. 6 b). For L-APLP2, no such increase from P6 to 7M brain was noted. The observed age-associated changes in the relative amounts of L-APLP2 mRNA during this developmental period are not consistent and generally small, apart from a significant decrease in hippocampus.This indicates a different regulation of the alternative splicing of exon 15 of APP and the related, non-homologous exon of APLP2 with aging.

Discussion L-APP transcripts and alternative splicing of APLP2 were only recently identified, and very little is known about the regulation and the functional significance of alternative splicing of exon 15 of APP and its APLP2 equivalent (Konig et al. 1992, Sandbrink et al. 1994a, b).The neuronespecific low frequency of both L-APP and L-APLP2 transcripts might well be related closely to typical neuronal function, while abundant expression of these species in non-neuronal cells as demonstrated here for primary cultured astroglial and microglial cells could be associated with characteristic non-neuronal features including proliferation capability. In this regard, it will be interesting to evaluate whether the high levels of both L-APP and L-APLP2 observed in the olfactory bulb occur because of the presence of continuously dividing neuronal progenitor cells typical for this brain region: in adult brain, the olfactory bulb is the only region where new synapses are continuously formed by newly formed sensory cells (Graziadei & Monti Graziadei 1979). Because of the similarities of tissue-specific regulation of APP and APLP2 alternative splicing, a related function of the corresponding non-homologous domains has been suggested. This notion is supported by the finding of similar structural features of the protein sequences and the predicted secondary structures of both domains (Sandbrink et al. 1994b).But not only the functional significance of alternative splicing of exon 15 of APP and its APLP2 equivalent remains to be elucidated: generally, a distinct physiological role for the ubiquitously distributed APP has not yet been determined, and the function of the only recently identified APP-like proteins including APLP2 has not been shown as Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

21

well. For the KPI-domains of APP and APLP2, however, a functional activity has been described involving efficient inhibition of serine-proteases.Therefore, KPI-encoding isoforms were postulated to take part in controlling the activity of degradatory enzymes during development, regeneration and degeneration of the nervous system (Müller-Hill & Beyreuther 1989). Because long-term potentiation and seizure kindling is associated with an increase in the expression of tissue-plasminogen activator (Qian et al. 1986), it was suggested that locally released or activated proteases are involved in the mechanisms of long-term potentiation. APP which is found at synaptic sites (Schubert et al. 1991) may regulate this proteolysis by its KPI isoforms. In brain, we observed two interesting features of developmental and age-associated regulation of KPI-expression in APP mRNA. One is defined by the uniquely high level of KPItranscripts in differentiating cerebellum which may indicate an association of KPI-APP isoforms with neuronal differentiation and synaptogenesis. We are currently analysing this in more detail by further experiments. The other result is the increase in KPI-encoding APP mRNA isoforms observed from early postnatal to adult brain which was also seen in several other aged rat brains that were analysed. For human APP, a similar but much more pronounced increase in KPI-encoding APP mRNA isoforms in the aged brain has been well documented by Northern blot hybridization and S1 Nuclease protection assays (Koo et al. 1990, König et al. 1991) This increase may partly reflect changes in abundance of brain cell types (gliosis), but more probably changes in neuronal isoform expression patterns are involved. As was shown for APLP2, neurones are well capable of expressing the majority of their isoforms to include the KPI-encoding exon,and so a regulated alternative splicing of exon 7 of APP is well conceivable. Additionally, our in vitro results clearly demonstrate significant amounts of KPI-APP transcripts in all neuronal cell types analysed by primary cell culturing. This is also supported by in situ hybridization and immunohistochemistry observations of significant human KPI-APP expression in vivo predominantly in neurones, but only very weakly in glial cells, for both AD patients and control cases (Tanzi et al. 1993). Since altered APP processing required for amyloid deposition in AD could very well be influenced by altered expression of the various APP transcripts, the alternative splicing of APP in AD has been studied by many groups (discussed in Tanzi et al. 1993) All these studies have been focused solely on expression of the KPI encoding exon 7 and exon 8 of APP. Unfortunately, the results remained inconsistent, primarily because of a variety of differences including methods of RNA quantification, and the specific brain regions being assessed, as well as sampling and interindividual variabilities. Although a final consensus has not been reached, most researchers believe today that cellular specificity and regional distribution of APP transcripts alternatively spliced in exon 7 and 8 are unaltered in AD as compared with age-matched controls (Tanzi et al. 1993).

22

R. Sandbrink et al.

Similar experiments have not yet been performed for LAPPs, which were only recently identified as being ubiquitously expressed (Sandbrink et al. 1994a).Whether the ageassociated increase that we have observed for all brain regions analysed is caused by altered neuronal splice patterns or simply reflects more prominent glial L-APP mRNA expression in the aged brain, can not yet be answered from our experiments. Virtually undectectable levels of L-APP mRNA in primary cultured neuronal cells from embryonic rat brain do not exclude increasing levels of L-APP expression in the mature and aged neurones under in vivo conditions. The related L-APLP2 isoforms, which are generally expressed to higher relative amounts than the corresponding L-APPs, were clearly demonstrated to be expressed not only in glial cells, but also in neurones. On the other hand, in most brain regions only about one tenth of total LAPP mRNA results from L-APP677 expression. Most LAPP mRNAs encode the KPI-domain of APP. This could be because of the glial origin of the major part of these transcripts (Table 1). Alternatively, a neuronal, coupled regulation of alternative splicing of exons 7 and 15 could explain these observations.The latter interpretation is supported by the finding that an age-associated increase in L-APLP2 mRNA expression was not observed which would be expected if glial cells account for the age-dependent L-APP increase. If it is shown in future experiments that virtually all L-APP mRNAs observed in vivo are expressed by glial cells and not by neurones, this will provide a sensitive biochemical marker for the abundance and activity of glial cells in brain It remains to be shown whether such an increase in LAPP mRNA expression is also observed in the aged human brain. Regarding pathogenesis and progression of AD, it is an important open question whether there is any alteration of L-APP expression in affected individuals Enhanced levels of L-APP expression might well represent a risk factor for development of the disease, since exon 15 of APP is separated from the NH2-terminal cleavage site of the βA4protein by only 16 amino acids.Thus, alternative splicing of exon 15 may influence binding and recognition of this cleavage site by secretase β and alter the amounts of soluble βA4-protein synthesized in vivo. This notion is supported from recently published in vitro results of increased levels of βA4-protein due to a double mutation at the NH2-terminus but outside of the βA4-protein sequence that was originally identified in individuals of two Swedish families of familial AD (Citron et al. 1992,Cai et al. 1993,Felsenstein et al. 1994). The age-associated increase in L-APP levels might be significant especially since L-APLP2 transcripts were not enhanced.As described above, there are extensive similarities in domain structure and alternative splicing between APP and APLP2.Thus, comparison of expression and processing of APP and APLP2 might provide a clue to the yet unknown function(s) of the members of theAPP gene family and the specific role of APP in the generation of Alzheimer’s disease. Recently performed studies revealed that murine

APP and APLP2 are expressed in similar, if not identical, neuronal populations and at similar levels. Additionally, APLP2 appears to mature though the same unusual secretory/cleavage pathway as APP (Slunt et al. 1994). In our experiments, we have also observed a similar regional distribution of APP and APLP2 in rat brain confirming these results. However, by analysis of the age-associated alternative splicing of both the KPI-encoding exon and the second exon missing in the `L-isoforms’, we identified significant differences between APP and APLP2, and we propose that these might be of relevance for understanding the special role of APP in the pathogenesis of age-associated AD.

Acknowledgements We gratefully acknowledge the donation of primary cultured rat microglial cells by Richard Banati.This work was supported by grants from the Deutsche Forschungsgemeinschaft through SFB 317 and 258, the Bundesminister für Forschung und Technologie of Germany, the Metropolitan Life Foundation, the Fonds der Chemischen Industrie, the Forschungsschwerpunkt Baden-Württemberg (to K.B.), and the National Health and Medical Research Council of Australia, the Victorian Health Promotion Foundation, and the Aluminium Development Corporation of Australia (to C. L. M.).

References Banati RB., Gehrmann J., Czech C., Mönning U., Jones L., König G. Beyreuther K. & Kreutzberg G.W. (1993) Early and rapid upregulation of amyloid precursor protein (APP) in activated microglia, Cha 9. 199–210. Cai X.-D., Golde T.E. & Younkin S G. (1993) Release of excess amyloid β protein from a mutant amyloid β protein precursor. Science 259, 514–516. Clark M J, Gagnon J. Williams A.F, & Barcley A.N (1985) MRC OX-2 antigen a lymphoid/neuronal membrane glycoprotein with a structure like a single immunoglobulin light chain. EMBOJ 4, 113–118. Citron M, Oltersdorf T, Haass C., McConlogue L., Hung A.Y., Seubert P, Vigo-Pelfrey C., Lieberburg I. & Selkoe D.J. (1992) Mutation of the β-amyloid precursor protein in familial Alzheimer’s disease increases β-protein production. Nature 360, 672–674. Daigle I & Li C (1993) Apl-1, a Caenorhabditis elegans gene encoding a protein related to the human β-amyloid protein precursor. Proc. Natl Acad. Sci. USA 90, 12045–12049 Dyrks T., Dyrks E., Mönning U., Urmoneite B.,Turner J. & Beyreuther K. (1993) Generation of βA4 from the amyloid precusor protein and fragments thereof. FEBS Lett. 335, 89–93. Esch F.S., Kelm P.S, Beattie E.C., Blacher R.W., Culwell A.R., Oltersdorf T., McClure D. & Ward P.J. (1990) Cleavage of amyloid β peptide during constitutive processing of its precursor. Science 248, 1122–1124 Felsenstein K.M., Hunihan L.W. & Roberts S.B. (1994) Altered cleavage and secretion of a recombinant β-APP bearing the Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

Age-associated splicing of APP and APLP2 Swedisch familial Alzheimer’s disease mutation. Nature Genet. 6, 251–256. Glenner G.G. & Wong C.W. (1984) Alzheimer’s disease and Down’s syndrome Sharing of a unique cerebrovascular amyloid fibril protein. Biochem. Biophys. Res. Commun. 122, 1131–1135. Golde T.E., Estus S., Usiak M., Younkin L.H. & Younkin S.G. (1990) Expression of β amyloid protein precursor mRNAs: Recognition of a novel alternatively spliced form and quantification in Alzheimer’s disease using PCR. Neuron 4, 253–267. Graziadei P.P.C. & Monti Graziadei G.A. (1979) Neurogenesis and neuron regeneration in the olfactory system of mammals. I. Morphological aspects of differentiation and structural organization of the olfactory sensory neurones. J Neurocytol 8, 1–18. Haass C, Schlossmacher M.G., Hung A.Y., Vigo-Pelfrey C., Mellon A., Ostaszewski B.L., Lieberburg I., Koo E.H., Schenk D., Teplow D.B. & Selkoe D.J. (1992b) Amyloid βpeptide is produced by cultured cells during normal metabolism. Nature 359, 322–325. Hanes J., von der Kammer H., Kristjansson G. I. & Scheit K.H. (1993) The complete cDNA coding sequence for the mouse CDEI binding protein Biochim. Biophys. Acta 1216, 154–156. Kang J, Lemaire H.G., Unterbeck A., Salbaum J.M., Masters, C.L., Grzeschik K.H., Multhaup G., Beyreuther K. & Müller-Hill B. (1987) The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell-surface receptor. Nature 325, 733–736. Kang J, & Müller-Hill B. (1990) Differential splicing of Alzheimer’s disease amyloid A4 precursor RNA in rat tissues: PreA4(695) mRNA is predominantly produced in rat and human brain. Biochem. Biophys. Res. Commun. 166, 1192–1200. Kitaguchi N, Takahashi Y, Tokushima Y., Shiojiri S. & Ito H. (1988) Novel precursor of Alzheimer’s disease amyloid protein shows protease inhibitor activity. Nature 331, 530–532. König G., Salbaum J.M., Wiestler O., Lang W., Schmitt H.P., Masters C.L. & Beyreuther K. (1991) Alternative splicing of the βA4 amyloid gene of Alzheimer’s disease in cortex of control and Alzheimer’s disease patients. Molec. Brain Res. 9, 259–262. König G., Mönning U., Czech C., Prior R., Banati R.B., Schreiter-Gasser U., Bauer J., Masters C. L., & Beyreuther K. (1992) Identification and differential expression of a novel alternative splice isoform of the βA4 amyloid precusor protein (APP) mRNA in leukocytes and brain microglial cells. J. Biol. Chem. 267, 10804–10809. Koo E.H., Sisodia S.S., Cork L C., Unterbeck A., Bayney R.M. & Price D.L. (1990). Differential expression of amyloid precursor protein mRNAs in cases of Alzheimer’s disease and in aged nonhuman primates Neuron 4, 97–104. Leclerc N., Beesley P W., Brown I., Colonnier M., Gurd J.W., Paladino T. & Hawkes R. (1989) Synaptophysin expression during synaptogenesis in the rat cerebellar cortex. J. Compr. Neurol. 280, 197–212 Masters C.L., Simms G., Weinmann N.A., Multhaup G., McDonald B.L. & Beyreuther K. (1985) Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl. Acad. Sci. USA 82, 4245–4249. Mönning U., König G., Banati R.B., Mechler H., Czech C., Gehrmann J., Schreiter-Gasser U., Masters C.L. & Beyreuther Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.

23

K. (1992) Alzheimer βA4-amyloid protein precursor in immunecompetent cells. J. Biol. Chem. 267, 23950–23956. Müller-Hill B. & Beyreuther K. (1989) Molecular biology of Alzheimer’s disease. Annu. Rev Biochem. 58, 287–307. Ponte P., Gonzalez-DeWhitt P., Schilling J., Miller J., Hsu D., Greenberg B., Davis K, Wallace W., Lieberburg I., Fuller F. & Cordell B. (1988) A new A4 amyloid mRNA contains a domain homologous to serine protease inhibitors. Nature 361, 453–457. Qian Z., Gilbert M.E., Colicos M.A., Kandel E.R. & Kuhl D. (1986) Tissue-plasminogen activator is induced as an immediate-early gene during seizure kindling and long-term potentiation. Nature 331, 525–527. Rosen D. R., Martin-Morris L., Luo L. & White K. (1989) A Drosophila gene encoding a protein resembling the human β-amyloid protein precursor. Proc. Natl. Acad. Sci. USA 86, 2478–2482. Sambrook J., Fritsch E.F., & Maniatis T. (1989) Molecular Cloning:A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sandbrink R., Masters C.L. & Beyreuther K. (1994a) βA4-amyloid protein precursor mRNA isoforms without exon 15 are ubiquitously expressed in rat tissues including brain, but not in neurones. J. Biol. Chem. 269, 1510–1517. Sandbrink R., Masters C.L. & Beyreuther K. (1994b) Similar alternative splicing of a non-homologous domain in βA4amyloid protein precursor-like proteins. J. Biol. Chem. 269, 14227–14234. Sandbrink R., Masters C.L. & Beyreuther K. (1994c) The complete nucleotide and deduced amino acid sequence of rat amyloid precursor-like protein 2: Two amino acids length difference to human and murine homologues. Biochim Biophys. Acta (in press). Schubert W., Prior R, Weidemann A., Dircksen H., Multhaup G., Masters C.L. & Beyreuther K. (1991) Localization of Alzheimer βA4-amyloid precursor protein at central and peripheral synaptic sites. Brain Res. 563, 184–194. Seubert P., Vigo-Pelfrey C., Esch F, Lee M., Dovey H., Davis D., Sinha S., Schlossmacher M., Whaley J., Swindelhurst C., McCormack R., Wolfert R., Selkoe DJ., Lieberburg I. & Schenk D. (1992) Isolation and quantification of soluble Alzheimer’s β-peptide from biological fluids. Nature 359, 325–327. Shoji M., Golde T.E., Cheung T.T., Ghiso J., Estus S., Shaffer L.M., Cai X.D., McKay D.M., Tintner R., Frangione B. & Younkin S.G. (1992) Production of the Alzheimer amyloid β protein by normal proteolytic processing. Science 258, 126–129. Sisodia S.S., Koo E.H., Beyreuther K., Unterbeck A. & Price D. (1990) Evidence that β-amyloid protein in Alzheimer’s disease is not derived by normal processing. Science 248, 492–495. Slunt H.H., Thinakaran G.,Von Koch C., Lo A.C.Y., Tanzi R.E. & Sisodia S.S. (1994) Expression of a ubiquitous, cross-reactive homologue of the mouse β-amyloid precursor protein (APP). J. Biol. Chem. 269, 2637–2644 Sprecher C.A., Grant F.J., Grimm G., O’Hara P.J., Norris, F., Norris K. & Foster D.C. (1993) Molecular cloning of the cDNA for a human amyloid precursor protein homolog: Evidence for a multigene family. Biochemistry 32, 4481–4486. Tanzi R.E., McClatchey A.I., Lamperti E.D., Villa-Komaroff L., Gusella J.F. & Neve R.L. (1988) Protease inhibitor domain

24

R. Sandbrink et al.

encoded by an amyloid protein precursor mRNA associated with Alzheimer’s disease. Nature 331, 528–530. Tanzi R.E., Wenniger J.J & Hyman B.T (1993) Cellular specificity and regional distribution of amyloid beta protein precursor alternative transcripts are unaltered in Alzheimer’s disease. Molec. Brain Res 18, 246–252. Vidal F., Blangy A., Rassoulzadegan M. & Cuzin F. (1992) A murine sequence-specific DNA binding protein shows extensive local similarities to the amyloid precursor protein. Biochem. Biophys. Res Commun. 189, 1336–1341. Wasco W., Bupp K., Magendantz M., Gusella J.F., Tanzi R.E. & Solomon F. (1992) Identification of a mouse brain cDNA that encodes a protein related to the Alzheimer disease-associated amyloid β protein precursor. Proc. Natl. Acad Sci. USA 89, 10758–10762. Wasco W., Gurubhagavatula S., Paradis M.D., Romano D.M.,

Sisodia S.S., Hyman B.T., Neve R.L. & Tanzi R.E (1993) Isolation and characterization of APLP2 encoding a homologue of the Alzheimer’s associated amyloid β protein precursor. Nature Genet. 5, 95–100. Weidemann A., König G., Bunke D., Fischer P., Salbaum J.M., Masters C.L. & Beyreuther K. (1989) Identification, biogenesis, and localization of precursors of Alzheimer’s disease A4 amyloid protein. Cell 57, 115–126. Yan Y C., Bai Y., Wang L., Miao S. & Koide S.S (1990) Characterization of cDNA encoding a human sperm membrane protein related to A4 amyloid protein. Proc. Natl Acad. Sci. USA 87, 2405–2408. Yoshikai S., Sasaki H., Dohura K., Furuya H. & Sakaki Y. (1990) Genomic organization of the human amyloid β-protein precursor gene. Gene 87, 257–263.

Copyright © 1994 Academic Press, Inc., Neurobiology of Disease, 1, 13–24 All rights of reproduction in any form reserved.