- Email: [email protected]
P4-345 LXR agonist treatment affects APP processing in brain of APP X PS2 double transgenic mice
Poster Session P4: Therapeutics and Therapeutic Strategies - Therapeutic Strategies, Amyloid-Based the specific antibody response, toxicity, reduction of A[3 deposits in the brain, and to prevent the behavioral deficits shown by these transgenic mice. Methods: Six APP/PS1 mice, starting at 3 months of age, received monthly intramuscular injections of 10 t~g VLP-A[3. Six control mice received LI-L2 VLP mad one received PBS on the same schedule as the experimental group. Animals were enthanized at 12 months of age. Results: Neither the experimental nor the control mice showed signs of toxicity. A~3 antibody titers measured by ELISA were elevated in mice immunized with VLP-A[3 (IgG 1/5000 to 1/10000), but not in controls. The sera of immunized mice recognized amyloid plaques in postmortem tissue sections of Alzheimer's disease. In the open field test conducted at 12 months of age, mice immunized with VLP-A~ displayed a number of central crossings similar to non-transgenic mice, whereas mice injected with BPV-L1/L2 or PBS showed significant less number of central crossings, indicative of higher levels of anxiety. This behavioral effect may reflect the diminished burden of soluble and/or aggregated AI3 in the brain of mice immunized with VLP - AI3. The neuropathological assessment of experimental and control mice, including morphological and biochemical measurements of A~ burden and markers of inflammation, is in progress.
•
GLEEVEC AND GSKI3 INHIBITORS HAVE NO E F F E C T ON A B E T A P R O D U C T I O N IN APP-EXPRESSING CELL LINES
Scott J. Pollack .1 , Jonathan Wrigley 1, Earl Clarke i, John Trauger 2, Chris Winrow 2, Neil Wilkie 1. 1Merck Sharp & Dohme, Harlow, Essex,
United Kingdom; 2Merck Research Laboratories, San Diego, CA, USA. Contact e-mail: Scott [email protected] Background: Several recent reports have shown that inhibitors of various protein kinases reduce the generation of amyloid-~ (A[~) peptide from APP in vitro, representing a novel approach to treating Alzheimer's disease (AD). These include studies with the tyrosine kinase inhibitor Gleevec, rho kinase inhibitors, lithium chloride (LiC1) and other glycogen synthase kinase 3 (GSK3) inhibitors. Objective(s): We sought to evaluate Gleevec and a selection of GSK3 inhibitors (LiC1, kenpanllone and Compound 1, a potent mad selective GSK3 inhibitor) to investigate the reported effects on A[~ production. Methods: Assays were used that specifically measure A[31-40 and A[31-42 production from the conditioned media of a variety of APP-overexpressing cell lines. Inhibition of GSK3 activity was verified in cell-free kinase assays and in cell-based assays using GSK-3 and [3-catenin phosphorylation. Results: None of the compounds demonstrated significant inhibitory effects on either A131-40 or A131-42 production, compared to vehicle controls, that could be distinguished from effects on cytotoxicity, at all concentrations tested. Compound 1 did show inhibitory effects on both A~l-40 and A~1-42 production but these only occurred at concentrations where cell viability was adversely affected. The inhibitory effects of each of the GSK3 inhibitors on GSK3 activity were confirmed in both cell-free and cell-based assays and shown to occur at concentrations which had previously been shown to have no effect on A~ generation. Conclusions: It is unlikely that GSK3 or kinases targeted by Gleevec are important regulators of A[3 production and these kinases are therefore not good candidate drug targets for the treatment of AD via effects on lowering A~.
~ L X R
A G O N I S T T R E A T M E N T AFFECTS APP P R O C E S S I N G IN B R A I N OF APP X PS2 DOUBLE TRANSGENIC MICE
Christian Czech*, Matthew Wright, Fabienne Goepfert, Denise Blum, Markus H~inggi, Patrick Biry, Helmut Jacobsen, Dominik Hainzl, Dieter Reinhardt, Laurence Ozmen, Hansruedi Loetscher. EHoffmann-La
Roche, Basel, Switzerland. Contact e-mail: [email protected] Background: The nuclear oxysterol receptors, liver X receptor-alpha (LXRalpha) and LXRbeta, coordinately regulate genes involved in cholesterol homeostasis. Both LXR subtypes are expressed in the brain; LXRbeta is even expressed more abundantly in CNS than in liver. LXR agonists (T0901317 and LG100268) have marked effects on gene expression in
$573
murine brain tissue both in vitro and in vivo. In primary astrocyte cultures, LXR agonists regulate several established LXR target genes, including ATP binding cassette transporter A1, and enhance cholesterol efflux. In contrast, little or no effect on gene expression or cholesterol efflux was detected in primary neuronal cultures. Treatment of mice with a selective LXR agonist resulted in the induction of several LXR target genes related to cholesterol homeostasis in the cerebellum and hippocampus. These data provide the first evidence that the LXRs regulate cholesterol homeostasis in the central nervous system. Objective(s): There are several lines of evidence which suggest that cholesterol content of neuronal membranes influences APP processing. Pharmacological modulation of brain cholesterol content by induction of LXR receptor could therefore be a target for APP processing, thereby reducing the amount of Abeta peptide. Methods: We have treated double transgenic mice expressing human mutant Amyloid Precusor Protein (APPsw) and human mutant Presenilin 2 (PS2 N141I) with the LXR agonist TO901317. Results: Cholesterol profile in the serum of treated mice shows an increase of HDL and a decrease of total Cholesterol. Quantitative RT-PCR analysis showed a dose dependent induction of LXR target genes in the brain of the treated animals indicating that the compound has functional activity in the brain. Analysis of APP metabolism revealed that the treatment of the transgenic mice with TO901317 affects APP processing in the brain. We could demonstrate only a small effect on total Abeta levels brain of the mice. However, analysis of APP C-terminal fragments showed that low doses (10mg/kg) lead to an increase of Abeta/C99 ratio (= increase of gamma-secretase cleavage) whereas high dose (50mg/kg) decrease Abeta/C99 ratio (= decrease gamma-secretase cleavage). Conclusions: These results suggest two contrary mechanisms on APP processing of high and low doses of TO901317.
•
A N T I - A ~ H 6 , A~35-40AND A~3s-42 S I N G L E - C H A I N V A R I A B L E R E G I O N F R A G M E N T S AS P O T E N T I A L T H E R A P E U T I C A G E N T S FOR AD
Yona Levites*, Pritam Das, Todd E. Golde. Mayo Clinic College of
Medicine, Jacksonville, FL, USA. Contact e-mail: [email protected] Background: One of the potential therapeutic strategies in AD is prevention of A[~ aggregation or enhancing its clearance. This approach is exemplified by use of active or passive immunization. Objectives: Agents with highaffinity for A[5, such as Fab or single chain variable region fragment (scFv) of anti-A[3 antibodies might offer a safe and effective alternative to current forms of anti-A[3 immunotherapy. We are currently developing anti-A~ scFv derived from antibodies that show efficacy in passive immunization paradigms. These will be tested in Tg2576 mice for their effect on A[3 deposition. Methods a n d Results: Initially, we conducted a passive immunization studies using a recently generated monoclonal antibody 2.1.3 raised against A~35-42 and specific for A1~1_42, as well as monoclonal antibody 32.1.3 raised against A[~1-16 and a Fab fragment derived from this antibody. Tg2576 mice were injected with 500 Ixg of Mab or Fab weekly for 3-4 months. A~l-40 and A~1_42 levels in brain tissue were measured by capture ELISA. A[3 levels were significantly reduced (>75% vs. control) in the 2.1.3 injected 7-month-old mice. Immunization with 2.1.3 in the older 11-monthold Tg2576 mice had no effect on amyloid loads/cognitive function, despite the sequestration/large increase in plasma A[~l_42/Ab complex. Both 32.1.3 and its Fab caused a significant reduction of the A~ levels in 11-month-old Tg2576 mice. In addition, we have cloned variable heavy and light chains of monoclonal antibodies raised against A~I-16, as A1335.40 and A~35-42. Expression plasmid encoding scFv and a leader sequence to determine the delivery of scFv protein via the secretory pathway was constructed using c-myc tag to detect the expression of scFv protein. The binding affinity and specificity of the scFv was examined by ELISA and by immnnohistochemical staining. Conclusions: Anti-A[3 2.1,3, 32.1.3 and 32.1.3-Fab antibodies efficiently reduced amyloid burden in TG2576 mice brain in a passive immunization paradigm. ScFv derived from anti-A[3 antibodies is expressed and secreted to the conditioned media of mammalian cells. It binds aggregated A[~1-42 and stains amyloid plaques in TG2576 mice and human AD brains. Studies are underway to determine the in vivo efficacy of adenoviral vector mediated expression of anti-A~3 scFv in mouse.
•
GLEEVEC AND GSKI3 INHIBITORS HAVE NO E F F E C T ON A B E T A P R O D U C T I O N IN APP-EXPRESSING CELL LINES
Scott J. Pollack .1 , Jonathan Wrigley 1, Earl Clarke i, John Trauger 2, Chris Winrow 2, Neil Wilkie 1. 1Merck Sharp & Dohme, Harlow, Essex,
United Kingdom; 2Merck Research Laboratories, San Diego, CA, USA. Contact e-mail: Scott [email protected] Background: Several recent reports have shown that inhibitors of various protein kinases reduce the generation of amyloid-~ (A[~) peptide from APP in vitro, representing a novel approach to treating Alzheimer's disease (AD). These include studies with the tyrosine kinase inhibitor Gleevec, rho kinase inhibitors, lithium chloride (LiC1) and other glycogen synthase kinase 3 (GSK3) inhibitors. Objective(s): We sought to evaluate Gleevec and a selection of GSK3 inhibitors (LiC1, kenpanllone and Compound 1, a potent mad selective GSK3 inhibitor) to investigate the reported effects on A[~ production. Methods: Assays were used that specifically measure A[31-40 and A[31-42 production from the conditioned media of a variety of APP-overexpressing cell lines. Inhibition of GSK3 activity was verified in cell-free kinase assays and in cell-based assays using GSK-3 and [3-catenin phosphorylation. Results: None of the compounds demonstrated significant inhibitory effects on either A131-40 or A131-42 production, compared to vehicle controls, that could be distinguished from effects on cytotoxicity, at all concentrations tested. Compound 1 did show inhibitory effects on both A~l-40 and A~1-42 production but these only occurred at concentrations where cell viability was adversely affected. The inhibitory effects of each of the GSK3 inhibitors on GSK3 activity were confirmed in both cell-free and cell-based assays and shown to occur at concentrations which had previously been shown to have no effect on A~ generation. Conclusions: It is unlikely that GSK3 or kinases targeted by Gleevec are important regulators of A[3 production and these kinases are therefore not good candidate drug targets for the treatment of AD via effects on lowering A~.
~ L X R
A G O N I S T T R E A T M E N T AFFECTS APP P R O C E S S I N G IN B R A I N OF APP X PS2 DOUBLE TRANSGENIC MICE
Christian Czech*, Matthew Wright, Fabienne Goepfert, Denise Blum, Markus H~inggi, Patrick Biry, Helmut Jacobsen, Dominik Hainzl, Dieter Reinhardt, Laurence Ozmen, Hansruedi Loetscher. EHoffmann-La
Roche, Basel, Switzerland. Contact e-mail: [email protected] Background: The nuclear oxysterol receptors, liver X receptor-alpha (LXRalpha) and LXRbeta, coordinately regulate genes involved in cholesterol homeostasis. Both LXR subtypes are expressed in the brain; LXRbeta is even expressed more abundantly in CNS than in liver. LXR agonists (T0901317 and LG100268) have marked effects on gene expression in
$573
murine brain tissue both in vitro and in vivo. In primary astrocyte cultures, LXR agonists regulate several established LXR target genes, including ATP binding cassette transporter A1, and enhance cholesterol efflux. In contrast, little or no effect on gene expression or cholesterol efflux was detected in primary neuronal cultures. Treatment of mice with a selective LXR agonist resulted in the induction of several LXR target genes related to cholesterol homeostasis in the cerebellum and hippocampus. These data provide the first evidence that the LXRs regulate cholesterol homeostasis in the central nervous system. Objective(s): There are several lines of evidence which suggest that cholesterol content of neuronal membranes influences APP processing. Pharmacological modulation of brain cholesterol content by induction of LXR receptor could therefore be a target for APP processing, thereby reducing the amount of Abeta peptide. Methods: We have treated double transgenic mice expressing human mutant Amyloid Precusor Protein (APPsw) and human mutant Presenilin 2 (PS2 N141I) with the LXR agonist TO901317. Results: Cholesterol profile in the serum of treated mice shows an increase of HDL and a decrease of total Cholesterol. Quantitative RT-PCR analysis showed a dose dependent induction of LXR target genes in the brain of the treated animals indicating that the compound has functional activity in the brain. Analysis of APP metabolism revealed that the treatment of the transgenic mice with TO901317 affects APP processing in the brain. We could demonstrate only a small effect on total Abeta levels brain of the mice. However, analysis of APP C-terminal fragments showed that low doses (10mg/kg) lead to an increase of Abeta/C99 ratio (= increase of gamma-secretase cleavage) whereas high dose (50mg/kg) decrease Abeta/C99 ratio (= decrease gamma-secretase cleavage). Conclusions: These results suggest two contrary mechanisms on APP processing of high and low doses of TO901317.
•
A N T I - A ~ H 6 , A~35-40AND A~3s-42 S I N G L E - C H A I N V A R I A B L E R E G I O N F R A G M E N T S AS P O T E N T I A L T H E R A P E U T I C A G E N T S FOR AD
Yona Levites*, Pritam Das, Todd E. Golde. Mayo Clinic College of
Medicine, Jacksonville, FL, USA. Contact e-mail: [email protected] Background: One of the potential therapeutic strategies in AD is prevention of A[~ aggregation or enhancing its clearance. This approach is exemplified by use of active or passive immunization. Objectives: Agents with highaffinity for A[5, such as Fab or single chain variable region fragment (scFv) of anti-A[3 antibodies might offer a safe and effective alternative to current forms of anti-A[3 immunotherapy. We are currently developing anti-A~ scFv derived from antibodies that show efficacy in passive immunization paradigms. These will be tested in Tg2576 mice for their effect on A[3 deposition. Methods a n d Results: Initially, we conducted a passive immunization studies using a recently generated monoclonal antibody 2.1.3 raised against A~35-42 and specific for A1~1_42, as well as monoclonal antibody 32.1.3 raised against A[~1-16 and a Fab fragment derived from this antibody. Tg2576 mice were injected with 500 Ixg of Mab or Fab weekly for 3-4 months. A~l-40 and A~1_42 levels in brain tissue were measured by capture ELISA. A[3 levels were significantly reduced (>75% vs. control) in the 2.1.3 injected 7-month-old mice. Immunization with 2.1.3 in the older 11-monthold Tg2576 mice had no effect on amyloid loads/cognitive function, despite the sequestration/large increase in plasma A[~l_42/Ab complex. Both 32.1.3 and its Fab caused a significant reduction of the A~ levels in 11-month-old Tg2576 mice. In addition, we have cloned variable heavy and light chains of monoclonal antibodies raised against A~I-16, as A1335.40 and A~35-42. Expression plasmid encoding scFv and a leader sequence to determine the delivery of scFv protein via the secretory pathway was constructed using c-myc tag to detect the expression of scFv protein. The binding affinity and specificity of the scFv was examined by ELISA and by immnnohistochemical staining. Conclusions: Anti-A[3 2.1,3, 32.1.3 and 32.1.3-Fab antibodies efficiently reduced amyloid burden in TG2576 mice brain in a passive immunization paradigm. ScFv derived from anti-A[3 antibodies is expressed and secreted to the conditioned media of mammalian cells. It binds aggregated A[~1-42 and stains amyloid plaques in TG2576 mice and human AD brains. Studies are underway to determine the in vivo efficacy of adenoviral vector mediated expression of anti-A~3 scFv in mouse.