- Email: [email protected]
227 Quantitative solution hybridisation assay of APP and APLP2 mRNA in Alzheimer's disease cerebral cortex
FIFTH INTERNATIONAL CONFERENCE ON ALZHEIMER'S DISEASE
225
227
Cloning and Characterization of the Human Amyloid Precursorlike Protein 1 (APLP1) K. Paliga*l, A. Weidemann t, U. Dtirrwang t, G.C. Peraus t, S. Kreger, C.L. Masters 2 and K. Beyreuther I. 1) ZMBH, Univ. Heidelberg, INF282, 69120 Heidelberg, Germany 2) Dept. Pathology, Univ. Melbourne, Parkville, Victoria, 3052 Australia. The family of APP-like proteins consists of three members: the amyloid precursor protein (APP), amyloid precursor-like protein. 1 (APLP1) and amyloid precursor-like protein 2 (APLP2). All three members represent type I transmembrane proteins with a large ectodomain and a short cytoplasmic tail. While many aspects of APP and APLP2 biology have been studied in great detail, relatively little attention has been paid to APLP1 since its cloning from a mouse cDNA library. Unlike APP and APLP2, APLP1 has been shown to be predominantly expressed in the nervous system and is supposed to be one of the highest expressed genes in the human brain. In phylogenetic terms APLP1 appears to be more closely related to the putative prototype of the APP-like molecules, since its homologues can be found in evolutionary such distant related organisms as D. melanogaster (APPL) and C. elegans (ap-l). To extend the knowledge about the biology of the APP-like molecules we have cloned and expressed the human APLP1 cDNA. Screening of human temporal cortex cDNA library resulted in the isolation of two cDNA clones covering the entire open reading frame and the 3"-untranslated region of huAPLPI. The full-length huAPLPI cDNA consists of 2356 bp and the longest ORF encodes a protein of 650 amino acids (aa), i.e. 3 aa less than the reported mouse APLPI sequence. At the amino acid level the huAPLPI displays 94% similarity and 89% identity to the murine APLP1 sequence. Expression of Nterminal c-myc tagged huAPLPI in COS and neuronal SH-SY5Y ceils revealed a doublet of about 85 kDa and 95 kDa within the cells and a single band around 90 kDa in the conditioned media of the cells. Similar to APP, the lower molecular weight band of 85 kDa represents an immature, N-glycosylated species of huAPLPI which becomes posttranslational modified including trimming of the N-linked carbohydrate moiety and additional O-glycosylation as exemplified by tunicamycin treatment of cells or carbohydrate analysis via lectins. The resulting mature forms of 95 kDa are proteolytically processed leading to the release of secretory, C-terminal truncated isoforms into the media of the cells. Analysis of the intracellular compartmentalization by light microscopy and confocal microscopy reveals partial colocalization of huAPLP1 and APP, with a pronounced vesicular staining for APLP1. In summary these data underscore the close relatedness of APLP1 to other APP-like proteins not only at the cDNA but also at the protein and cellular level.
Quantitative Solution Hybridisation Assay of APP and APLP2 mRNA in Alzheimer's Disease Cerebral Cortex J.A. Johnston*, S. Norgren, R. Ravid, L. Lannfelt and R.F. Cowburn.
$57
Department of Clinical Neuroscience and Family Medicine, Geriatric Section, Karolinska Institute, Novum KFC, 141 86 Huddinge, Sweden. Fax: 46 8 746 5235 Department of Clinical Genetics, Karolinska Hospital, 104 Ol Stockholm. Sweden. The Netherlands Brain Bank, Meibergdreef33, 1105 AZ Amsterdam ZO, The Netherlands.
Amyloid precursor protein (APP) is metabolised to produce AI~, a peptide found aggregated in Alzheimer's disease neuritic plaques. APP is a member of a multigene protein family which includes amyloid precursor-like protein 2 (APLP2). Since AI3 accumulation is triggered by over-expression of APP in Down's syndrome, we investigated whether APP mRNA expression was altered in Alzheimer's disease post mortem cerebral cortex. In addition, we characterised cortical APLP2 mRNA levels in the disease. A new RNA-RNA solution hybridisation-RNase protection assay was used to determine the levels of total APP, APP containing the KPI insert and APLP2 mRNA. This assay is quantitative, due to the inclusion of a standard curve generated by hybridisation with in vitro transcribed sense RNA, sensitive, and reproducible. Total APP, APP KPI and APLP2 mRNA was quantified in midtemporal and superior frontal cortices from 14 Alzheimer's disease, 9 nondemented, and 5 positive disease control subjects (Netherlands Brain Bank). All subjects were Apolipoprotein E-genotyped. Approximately three times more APP than APLP2 mRNA was detected in both mid-temporal and superior frontal cortices, and about 70% of total APP mRNA contained the KPI insert in the control cases. Total APP and APLP2 mRNA levels were significantly reduced in Alzheimer's disease mid-temporal, but not superior frontal cortex, suggesting that regional reductions in these mRNAs correlate with severity of disease pathology. A small, significant increase in the proportion of APP KPI mRNA was seen in Alzheimer's disease mid-temporal and superior frontal cortices. Apolipoprotein E genotype did not influence cortical levels of total APP, APP KPI or APLP2 mRNA, suggesting that it's role in the aetiology of Alzheimer's disease is not exerted at this level. In summary, these results indicate that Alzheimer's disease, unlike Down's syndrome, is not associated with over-expression of APP. The levels of APP and also APLP2 mRNA appear instead to be reduced in cortical areas affected by the disease. Our findings also revealed a disease-associated increase in APP KPI-containing isoforms, expressed as a proportion of total APP mRNA. Further investigation is required to clarify whether these changes occur early in the disease, predisposing affected individuals to AI3 production and aggregation, or reflect later events such as gliosis and neuronal cell death.
228 226 Effect of Fibroblast Growth Factor and Insulin on B-Amyloid Production in Normal and Familial AIzheimer's Disease Fibroblasts C. Vigo-Pelfrey*, A. Cedazo-Minguez, R. F. Cowburn, M. Jensen, L. Lannfelt and J. A. Johnston NeuroCal International, Mountain View, California, USA. Karolinska Institute, Department of Geriatric Medicine, Huddinge, Sweden Strong evidence that B-amyloid (AB) deposition plays an important role in Alzheimer's disease (AD) has come from the identification of familial AD (FAD) kindreds in which the AD phenotype cosegregates with mutations in the AB precursor protein (APP) located close to the AB domain. The gene encoding APP on chromosome 21 is known to be the site of at least six different missence mutations that cause an early onset of FAD. Each mutation has shown to change the length or quantity of the critical AB. In the present work we investigated the effect of fibroblast growth factor (FGF) and insulin on AB release in fibroblasts obtained from the upper arm skin biopsy of patients carrying the Swedish (APP 670/671), chromosome 14 (S182) mutation and from normal subjects genetically tested negative for these mutations. Two cell lines fox"each type were used. Fibroblasts were grown in culture to confluency with fibroblast growth media (Clonetics/supplemented with l n J m l FGF and 5ughnl insuliu and without supplememation. After reaching conflnency the cells were incubated for nine days changing the media every 72 hours. The last media was assayed for total AB and A/?,42 using two enzyme linked immunosorbent sandwich assays with antibodies specific to these peptides. Fibroblasts carrying the Swedish and chromosome 14 mutations produce 2- to 10-fold more AB 42 than control fibroblasts when incubated in the presence of FGF and insulin. Only small or no difference in the levels of AS 42 were found in fibroblasts grown in the absence of FGF and insulin. The total levels of AB were also increased approximately 2-fold by FGF and insulin in fibroblasts with or without these mutations. It has been shown before that epidermal, fibroblast and nerve growth factors can induce APP expression in cultured rat PC 12 cells and in developing hamster brain. The effect of FGF and insulin on APP expression and AB production in nornlal and FAD fibroblasts will be discussed.
Apolipoprotein E Genotype Determines Survival in the Oldest-old Age 85+ Who Have Good Cognition H. Basun*, E. Corder, L. Larmfelt, M. Viitanen, L. Corder, K. Maroon, B. Winblad Department of Clinical Neuroscience, Huddinge University Hospital, S-141 86 HUDDINGE, Sweden (+46 8 746 3690 FAX) and the Center for Demographic Studies, Duke University, Durham, North Carolina 27708-0408, USA We investigated the influence of apolipoprotein E (APOE) polymorphism on cognition and survival in a population sample age 75+ in Stockholm (the Kungsholmen Project). The 1077 of 1124 genotyped subjects carrying the common e2/3, e3/3 or e3/4 genotypes, out of 1810 subjects in the cohort, are included in the analysis. Information on cognition at cohort inception during 1987-89 and on subsequent mortality to January 1, 1995 were available. The odds of cognitive impairment between the e3/4 and e3/3 genotypes declined with age from 4.8 at ages 75-79 to 1.7 at ages 80-84; and to 1.0 (i.e., no association) in subjects age 85+. Despite this age-pattern of associations, APOE polymorphism dM not significantly predict survival prior to age 85; nor did it predict survival in subjects aged 85+ who were eognitively impaired. Survival did vary 4-fold with respect to APOE polymorphism in Cox proportional hazard models anmng those aged 85+ who had good cognition: Mortality in subjects with e2/3 was half that in those with e2/3 (hazard ratio (HR)=0.5, 95% CI 0.20.9); mortality in subjects with e3/4 was twice that in those with e3/3 (HR=2.0, 95% CI 1.1-3.5). This 4-fold variation resulted in a 2-year difference in survival. Although the apolipoprotein E isoforms differ by only one or two amino acids, the physiologic effects of these differences produces significant variation in survival among the oldest-old age 85+ with good cognition. The pattern of risk for death parallels the pattern of risk found for Alzheimer disease (AD) in subjects aged 60 to 75, i.e., E4 confers risk, and E2 confers protection, compared to E3. However, the lack of association of e4 with cognitive impairment in subjects age 85+ and the weak association of c4 with AD in the Kungsholmen Project cohort (Lannfelt et al., 1994) indicate that the observed survival differences are not accounted for by the risk of AD. Future studies are needed to identify the disorders which account for survival differentials.
225
227
Cloning and Characterization of the Human Amyloid Precursorlike Protein 1 (APLP1) K. Paliga*l, A. Weidemann t, U. Dtirrwang t, G.C. Peraus t, S. Kreger, C.L. Masters 2 and K. Beyreuther I. 1) ZMBH, Univ. Heidelberg, INF282, 69120 Heidelberg, Germany 2) Dept. Pathology, Univ. Melbourne, Parkville, Victoria, 3052 Australia. The family of APP-like proteins consists of three members: the amyloid precursor protein (APP), amyloid precursor-like protein. 1 (APLP1) and amyloid precursor-like protein 2 (APLP2). All three members represent type I transmembrane proteins with a large ectodomain and a short cytoplasmic tail. While many aspects of APP and APLP2 biology have been studied in great detail, relatively little attention has been paid to APLP1 since its cloning from a mouse cDNA library. Unlike APP and APLP2, APLP1 has been shown to be predominantly expressed in the nervous system and is supposed to be one of the highest expressed genes in the human brain. In phylogenetic terms APLP1 appears to be more closely related to the putative prototype of the APP-like molecules, since its homologues can be found in evolutionary such distant related organisms as D. melanogaster (APPL) and C. elegans (ap-l). To extend the knowledge about the biology of the APP-like molecules we have cloned and expressed the human APLP1 cDNA. Screening of human temporal cortex cDNA library resulted in the isolation of two cDNA clones covering the entire open reading frame and the 3"-untranslated region of huAPLPI. The full-length huAPLPI cDNA consists of 2356 bp and the longest ORF encodes a protein of 650 amino acids (aa), i.e. 3 aa less than the reported mouse APLPI sequence. At the amino acid level the huAPLPI displays 94% similarity and 89% identity to the murine APLP1 sequence. Expression of Nterminal c-myc tagged huAPLPI in COS and neuronal SH-SY5Y ceils revealed a doublet of about 85 kDa and 95 kDa within the cells and a single band around 90 kDa in the conditioned media of the cells. Similar to APP, the lower molecular weight band of 85 kDa represents an immature, N-glycosylated species of huAPLPI which becomes posttranslational modified including trimming of the N-linked carbohydrate moiety and additional O-glycosylation as exemplified by tunicamycin treatment of cells or carbohydrate analysis via lectins. The resulting mature forms of 95 kDa are proteolytically processed leading to the release of secretory, C-terminal truncated isoforms into the media of the cells. Analysis of the intracellular compartmentalization by light microscopy and confocal microscopy reveals partial colocalization of huAPLP1 and APP, with a pronounced vesicular staining for APLP1. In summary these data underscore the close relatedness of APLP1 to other APP-like proteins not only at the cDNA but also at the protein and cellular level.
Quantitative Solution Hybridisation Assay of APP and APLP2 mRNA in Alzheimer's Disease Cerebral Cortex J.A. Johnston*, S. Norgren, R. Ravid, L. Lannfelt and R.F. Cowburn.
$57
Department of Clinical Neuroscience and Family Medicine, Geriatric Section, Karolinska Institute, Novum KFC, 141 86 Huddinge, Sweden. Fax: 46 8 746 5235 Department of Clinical Genetics, Karolinska Hospital, 104 Ol Stockholm. Sweden. The Netherlands Brain Bank, Meibergdreef33, 1105 AZ Amsterdam ZO, The Netherlands.
Amyloid precursor protein (APP) is metabolised to produce AI~, a peptide found aggregated in Alzheimer's disease neuritic plaques. APP is a member of a multigene protein family which includes amyloid precursor-like protein 2 (APLP2). Since AI3 accumulation is triggered by over-expression of APP in Down's syndrome, we investigated whether APP mRNA expression was altered in Alzheimer's disease post mortem cerebral cortex. In addition, we characterised cortical APLP2 mRNA levels in the disease. A new RNA-RNA solution hybridisation-RNase protection assay was used to determine the levels of total APP, APP containing the KPI insert and APLP2 mRNA. This assay is quantitative, due to the inclusion of a standard curve generated by hybridisation with in vitro transcribed sense RNA, sensitive, and reproducible. Total APP, APP KPI and APLP2 mRNA was quantified in midtemporal and superior frontal cortices from 14 Alzheimer's disease, 9 nondemented, and 5 positive disease control subjects (Netherlands Brain Bank). All subjects were Apolipoprotein E-genotyped. Approximately three times more APP than APLP2 mRNA was detected in both mid-temporal and superior frontal cortices, and about 70% of total APP mRNA contained the KPI insert in the control cases. Total APP and APLP2 mRNA levels were significantly reduced in Alzheimer's disease mid-temporal, but not superior frontal cortex, suggesting that regional reductions in these mRNAs correlate with severity of disease pathology. A small, significant increase in the proportion of APP KPI mRNA was seen in Alzheimer's disease mid-temporal and superior frontal cortices. Apolipoprotein E genotype did not influence cortical levels of total APP, APP KPI or APLP2 mRNA, suggesting that it's role in the aetiology of Alzheimer's disease is not exerted at this level. In summary, these results indicate that Alzheimer's disease, unlike Down's syndrome, is not associated with over-expression of APP. The levels of APP and also APLP2 mRNA appear instead to be reduced in cortical areas affected by the disease. Our findings also revealed a disease-associated increase in APP KPI-containing isoforms, expressed as a proportion of total APP mRNA. Further investigation is required to clarify whether these changes occur early in the disease, predisposing affected individuals to AI3 production and aggregation, or reflect later events such as gliosis and neuronal cell death.
228 226 Effect of Fibroblast Growth Factor and Insulin on B-Amyloid Production in Normal and Familial AIzheimer's Disease Fibroblasts C. Vigo-Pelfrey*, A. Cedazo-Minguez, R. F. Cowburn, M. Jensen, L. Lannfelt and J. A. Johnston NeuroCal International, Mountain View, California, USA. Karolinska Institute, Department of Geriatric Medicine, Huddinge, Sweden Strong evidence that B-amyloid (AB) deposition plays an important role in Alzheimer's disease (AD) has come from the identification of familial AD (FAD) kindreds in which the AD phenotype cosegregates with mutations in the AB precursor protein (APP) located close to the AB domain. The gene encoding APP on chromosome 21 is known to be the site of at least six different missence mutations that cause an early onset of FAD. Each mutation has shown to change the length or quantity of the critical AB. In the present work we investigated the effect of fibroblast growth factor (FGF) and insulin on AB release in fibroblasts obtained from the upper arm skin biopsy of patients carrying the Swedish (APP 670/671), chromosome 14 (S182) mutation and from normal subjects genetically tested negative for these mutations. Two cell lines fox"each type were used. Fibroblasts were grown in culture to confluency with fibroblast growth media (Clonetics/supplemented with l n J m l FGF and 5ughnl insuliu and without supplememation. After reaching conflnency the cells were incubated for nine days changing the media every 72 hours. The last media was assayed for total AB and A/?,42 using two enzyme linked immunosorbent sandwich assays with antibodies specific to these peptides. Fibroblasts carrying the Swedish and chromosome 14 mutations produce 2- to 10-fold more AB 42 than control fibroblasts when incubated in the presence of FGF and insulin. Only small or no difference in the levels of AS 42 were found in fibroblasts grown in the absence of FGF and insulin. The total levels of AB were also increased approximately 2-fold by FGF and insulin in fibroblasts with or without these mutations. It has been shown before that epidermal, fibroblast and nerve growth factors can induce APP expression in cultured rat PC 12 cells and in developing hamster brain. The effect of FGF and insulin on APP expression and AB production in nornlal and FAD fibroblasts will be discussed.
Apolipoprotein E Genotype Determines Survival in the Oldest-old Age 85+ Who Have Good Cognition H. Basun*, E. Corder, L. Larmfelt, M. Viitanen, L. Corder, K. Maroon, B. Winblad Department of Clinical Neuroscience, Huddinge University Hospital, S-141 86 HUDDINGE, Sweden (+46 8 746 3690 FAX) and the Center for Demographic Studies, Duke University, Durham, North Carolina 27708-0408, USA We investigated the influence of apolipoprotein E (APOE) polymorphism on cognition and survival in a population sample age 75+ in Stockholm (the Kungsholmen Project). The 1077 of 1124 genotyped subjects carrying the common e2/3, e3/3 or e3/4 genotypes, out of 1810 subjects in the cohort, are included in the analysis. Information on cognition at cohort inception during 1987-89 and on subsequent mortality to January 1, 1995 were available. The odds of cognitive impairment between the e3/4 and e3/3 genotypes declined with age from 4.8 at ages 75-79 to 1.7 at ages 80-84; and to 1.0 (i.e., no association) in subjects age 85+. Despite this age-pattern of associations, APOE polymorphism dM not significantly predict survival prior to age 85; nor did it predict survival in subjects aged 85+ who were eognitively impaired. Survival did vary 4-fold with respect to APOE polymorphism in Cox proportional hazard models anmng those aged 85+ who had good cognition: Mortality in subjects with e2/3 was half that in those with e2/3 (hazard ratio (HR)=0.5, 95% CI 0.20.9); mortality in subjects with e3/4 was twice that in those with e3/3 (HR=2.0, 95% CI 1.1-3.5). This 4-fold variation resulted in a 2-year difference in survival. Although the apolipoprotein E isoforms differ by only one or two amino acids, the physiologic effects of these differences produces significant variation in survival among the oldest-old age 85+ with good cognition. The pattern of risk for death parallels the pattern of risk found for Alzheimer disease (AD) in subjects aged 60 to 75, i.e., E4 confers risk, and E2 confers protection, compared to E3. However, the lack of association of e4 with cognitive impairment in subjects age 85+ and the weak association of c4 with AD in the Kungsholmen Project cohort (Lannfelt et al., 1994) indicate that the observed survival differences are not accounted for by the risk of AD. Future studies are needed to identify the disorders which account for survival differentials.