Limits...
Improved learning and memory in aged mice deficient in amyloid beta-degrading neutral endopeptidase.

Walther T, Albrecht D, Becker M, Schubert M, Kouznetsova E, Wiesner B, Maul B, Schliebs R, Grecksch G, Furkert J, Sterner-Kock A, Schultheiss HP, Becker A, Siems WE - PLoS ONE (2009)

Bottom Line: Neutral endopeptidase, also known as neprilysin and abbreviated NEP, is considered to be one of the key enzymes in initial human amyloid-beta (Abeta) degradation.We found that while endogenous Abeta concentrations were elevated in the brains of NEP-knockout mice at all investigated age groups, immunohistochemical analysis using monoclonal antibodies did not detect any Abeta deposits even in old NEP knockout mice.Surprisingly, tests of learning and memory revealed that the ability to learn was not reduced in old NEP-deficient mice but instead had significantly improved, and sustained learning and memory in the aged mice was congruent with improved long-term potentiation (LTP) in brain slices of the hippocampus and lateral amygdala.

View Article: PubMed Central - PubMed

Affiliation: Department of Cardiology, Charité-University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany. thomas.walther@hyms.ac.uk

ABSTRACT

Background: Neutral endopeptidase, also known as neprilysin and abbreviated NEP, is considered to be one of the key enzymes in initial human amyloid-beta (Abeta) degradation. The aim of our study was to explore the impact of NEP deficiency on the initial development of dementia-like symptoms in mice.

Methodology/principal findings: We found that while endogenous Abeta concentrations were elevated in the brains of NEP-knockout mice at all investigated age groups, immunohistochemical analysis using monoclonal antibodies did not detect any Abeta deposits even in old NEP knockout mice. Surprisingly, tests of learning and memory revealed that the ability to learn was not reduced in old NEP-deficient mice but instead had significantly improved, and sustained learning and memory in the aged mice was congruent with improved long-term potentiation (LTP) in brain slices of the hippocampus and lateral amygdala. Our data suggests a beneficial effect of pharmacological inhibition of cerebral NEP on learning and memory in mice due to the accumulation of peptides other than Abeta degradable by NEP. By conducting degradation studies and peptide measurements in the brain of both genotypes, we identified two neuropeptide candidates, glucagon-like peptide 1 and galanin, as first potential candidates to be involved in the improved learning in aged NEP-deficient mice.

Conclusions/significance: Thus, the existence of peptides targeted by NEP that improve learning and memory in older individuals may represent a promising avenue for the treatment of neurodegenerative diseases.

Show MeSH

Related in: MedlinePlus

NEP-dependent neuropeptide degradation.(a) HPLC-monitored degradation (recovery in %) of GLP-1 and galanin (each 5 µM) over 120 min using recombinant (rc) NEP (20 ng; R&D Systems, Wiesbaden, Germany); mean values with s.e.m.; n≥3. The reactions were stopped by adding 0.35 M perchloric acid. In parallel assays, heat-inactivated probes (5 min at 90°C) were used as a control. After centrifugation of sedimented proteins, HPLC analyses were performed by isocratic elution as described by Siems et al. [45]. (b) HPLC-monitored peptide degradation (in %) of GLP-1 and galanin (5 µM) in brain membranes (0.5 mg protein/ml) of wild-type (+/+) and knockout mice (−/−) over 30 min. The reactions were stopped by addition of 0.35 M perchloric acid. In parallel assays, heat-inactivated NEP (5 min at 90°C) was used as a control. HPLC analysis was performed by isocratic elution as described [45]. The significance of differences was calculated by two-sided t test; n≥4; **P<0.01. (c) For galanin and GLP-1 immunohistochemistry, slices were incubated with the specific antibodies, followed by incubation with biotinylated anti-rabbit IgG. Immunoreactive products were visualized by the nickel ammonium sulfate-intensified diaminobenzidine reaction. Unequivocally stained regions in the murine cortex were localized and defined as ROIs (regions of interest), and luminometrically compared with adjacent regions without detectable staining (reference ROIs) (for both regions n = 40). The calculated differences were shown as mean±s.e.m.). Significant differences were calculated by Student's t test, indicated by **P<0.01, ***P<0.001.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2643003&req=5

pone-0004590-g005: NEP-dependent neuropeptide degradation.(a) HPLC-monitored degradation (recovery in %) of GLP-1 and galanin (each 5 µM) over 120 min using recombinant (rc) NEP (20 ng; R&D Systems, Wiesbaden, Germany); mean values with s.e.m.; n≥3. The reactions were stopped by adding 0.35 M perchloric acid. In parallel assays, heat-inactivated probes (5 min at 90°C) were used as a control. After centrifugation of sedimented proteins, HPLC analyses were performed by isocratic elution as described by Siems et al. [45]. (b) HPLC-monitored peptide degradation (in %) of GLP-1 and galanin (5 µM) in brain membranes (0.5 mg protein/ml) of wild-type (+/+) and knockout mice (−/−) over 30 min. The reactions were stopped by addition of 0.35 M perchloric acid. In parallel assays, heat-inactivated NEP (5 min at 90°C) was used as a control. HPLC analysis was performed by isocratic elution as described [45]. The significance of differences was calculated by two-sided t test; n≥4; **P<0.01. (c) For galanin and GLP-1 immunohistochemistry, slices were incubated with the specific antibodies, followed by incubation with biotinylated anti-rabbit IgG. Immunoreactive products were visualized by the nickel ammonium sulfate-intensified diaminobenzidine reaction. Unequivocally stained regions in the murine cortex were localized and defined as ROIs (regions of interest), and luminometrically compared with adjacent regions without detectable staining (reference ROIs) (for both regions n = 40). The calculated differences were shown as mean±s.e.m.). Significant differences were calculated by Student's t test, indicated by **P<0.01, ***P<0.001.

Mentions: It has been shown that NEP targets a variety of peptides involved in learning and memory such as oxytocin [36], NPY [37] or CCK [38]. Thus, marked improvement in learning and memory in very old NEP-deficient mice if compared with their aged-matched controls appears to be explained best by these NEP-degradable neuropeptides. Nevertheless, the known catalytic properties of NEP - it preferentially hydrolyses oligopeptides by cleaving on the N-terminal side of hydrophobic amino acid residues - and the knowledge of further potential learning-associated peptides [39]–[42] motivated us to test several other neuropeptides for being potential NEP substrates. Consequently, we initiated our own studies on the postulated NEP-dependent metabolism of GLP-1 and galanin. First results underscore the likelihood of our assumption. We observed in vitro a significant GLP-1 degradation by rcNEP (Figure 5a) and a reduced rate of GLP-1 degradation in brains lacking NEP (Figure 5b) comparable to that in wild-type membranes pretreated with NEP inhibitor candoxatrilat (data not shown). This is in agreement with a study of Hupe-Sodmann et al. (1995) postulating NEP peptidolytic activity for GLP-1 [39]. Our finding is all the more important, since GLP-1 has been found to improve associative and spatial learning if injected intracerebroventricularly [40]. Learning-associated properties are also known for galanin [41], [42], and due to its chemical structure, galanin is a prominent candidate for degradation by NEP. As shown in Figures 5a and b, we also identified galanin as a NEP substrate. Since, as we showed, GLP-1 and galanin are substrates for NEP, NEP deficiency may lead to an elevation of both peptides, thus improving learning and memory. To also test this hypothesis, we measured the concentration of both peptides in the cortex of NEP-deficient mice and their wild-type controls, and checked for possible age-dependency. While differences did not occur for galanin or have been minor for GLP-1 in 6-month-old mice that could not be discriminated in their learning capacity, significantly higher levels of both peptides were measured in aged NEP knockouts that are characterized by sustained learning capacity (Figure 5c). These findings further strengthen the hypothesis that both peptides could be involved in the improved learning in mice lacking enzymatic NEP activity.


Improved learning and memory in aged mice deficient in amyloid beta-degrading neutral endopeptidase.

Walther T, Albrecht D, Becker M, Schubert M, Kouznetsova E, Wiesner B, Maul B, Schliebs R, Grecksch G, Furkert J, Sterner-Kock A, Schultheiss HP, Becker A, Siems WE - PLoS ONE (2009)

NEP-dependent neuropeptide degradation.(a) HPLC-monitored degradation (recovery in %) of GLP-1 and galanin (each 5 µM) over 120 min using recombinant (rc) NEP (20 ng; R&D Systems, Wiesbaden, Germany); mean values with s.e.m.; n≥3. The reactions were stopped by adding 0.35 M perchloric acid. In parallel assays, heat-inactivated probes (5 min at 90°C) were used as a control. After centrifugation of sedimented proteins, HPLC analyses were performed by isocratic elution as described by Siems et al. [45]. (b) HPLC-monitored peptide degradation (in %) of GLP-1 and galanin (5 µM) in brain membranes (0.5 mg protein/ml) of wild-type (+/+) and knockout mice (−/−) over 30 min. The reactions were stopped by addition of 0.35 M perchloric acid. In parallel assays, heat-inactivated NEP (5 min at 90°C) was used as a control. HPLC analysis was performed by isocratic elution as described [45]. The significance of differences was calculated by two-sided t test; n≥4; **P<0.01. (c) For galanin and GLP-1 immunohistochemistry, slices were incubated with the specific antibodies, followed by incubation with biotinylated anti-rabbit IgG. Immunoreactive products were visualized by the nickel ammonium sulfate-intensified diaminobenzidine reaction. Unequivocally stained regions in the murine cortex were localized and defined as ROIs (regions of interest), and luminometrically compared with adjacent regions without detectable staining (reference ROIs) (for both regions n = 40). The calculated differences were shown as mean±s.e.m.). Significant differences were calculated by Student's t test, indicated by **P<0.01, ***P<0.001.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2643003&req=5

pone-0004590-g005: NEP-dependent neuropeptide degradation.(a) HPLC-monitored degradation (recovery in %) of GLP-1 and galanin (each 5 µM) over 120 min using recombinant (rc) NEP (20 ng; R&D Systems, Wiesbaden, Germany); mean values with s.e.m.; n≥3. The reactions were stopped by adding 0.35 M perchloric acid. In parallel assays, heat-inactivated probes (5 min at 90°C) were used as a control. After centrifugation of sedimented proteins, HPLC analyses were performed by isocratic elution as described by Siems et al. [45]. (b) HPLC-monitored peptide degradation (in %) of GLP-1 and galanin (5 µM) in brain membranes (0.5 mg protein/ml) of wild-type (+/+) and knockout mice (−/−) over 30 min. The reactions were stopped by addition of 0.35 M perchloric acid. In parallel assays, heat-inactivated NEP (5 min at 90°C) was used as a control. HPLC analysis was performed by isocratic elution as described [45]. The significance of differences was calculated by two-sided t test; n≥4; **P<0.01. (c) For galanin and GLP-1 immunohistochemistry, slices were incubated with the specific antibodies, followed by incubation with biotinylated anti-rabbit IgG. Immunoreactive products were visualized by the nickel ammonium sulfate-intensified diaminobenzidine reaction. Unequivocally stained regions in the murine cortex were localized and defined as ROIs (regions of interest), and luminometrically compared with adjacent regions without detectable staining (reference ROIs) (for both regions n = 40). The calculated differences were shown as mean±s.e.m.). Significant differences were calculated by Student's t test, indicated by **P<0.01, ***P<0.001.
Mentions: It has been shown that NEP targets a variety of peptides involved in learning and memory such as oxytocin [36], NPY [37] or CCK [38]. Thus, marked improvement in learning and memory in very old NEP-deficient mice if compared with their aged-matched controls appears to be explained best by these NEP-degradable neuropeptides. Nevertheless, the known catalytic properties of NEP - it preferentially hydrolyses oligopeptides by cleaving on the N-terminal side of hydrophobic amino acid residues - and the knowledge of further potential learning-associated peptides [39]–[42] motivated us to test several other neuropeptides for being potential NEP substrates. Consequently, we initiated our own studies on the postulated NEP-dependent metabolism of GLP-1 and galanin. First results underscore the likelihood of our assumption. We observed in vitro a significant GLP-1 degradation by rcNEP (Figure 5a) and a reduced rate of GLP-1 degradation in brains lacking NEP (Figure 5b) comparable to that in wild-type membranes pretreated with NEP inhibitor candoxatrilat (data not shown). This is in agreement with a study of Hupe-Sodmann et al. (1995) postulating NEP peptidolytic activity for GLP-1 [39]. Our finding is all the more important, since GLP-1 has been found to improve associative and spatial learning if injected intracerebroventricularly [40]. Learning-associated properties are also known for galanin [41], [42], and due to its chemical structure, galanin is a prominent candidate for degradation by NEP. As shown in Figures 5a and b, we also identified galanin as a NEP substrate. Since, as we showed, GLP-1 and galanin are substrates for NEP, NEP deficiency may lead to an elevation of both peptides, thus improving learning and memory. To also test this hypothesis, we measured the concentration of both peptides in the cortex of NEP-deficient mice and their wild-type controls, and checked for possible age-dependency. While differences did not occur for galanin or have been minor for GLP-1 in 6-month-old mice that could not be discriminated in their learning capacity, significantly higher levels of both peptides were measured in aged NEP knockouts that are characterized by sustained learning capacity (Figure 5c). These findings further strengthen the hypothesis that both peptides could be involved in the improved learning in mice lacking enzymatic NEP activity.

Bottom Line: Neutral endopeptidase, also known as neprilysin and abbreviated NEP, is considered to be one of the key enzymes in initial human amyloid-beta (Abeta) degradation.We found that while endogenous Abeta concentrations were elevated in the brains of NEP-knockout mice at all investigated age groups, immunohistochemical analysis using monoclonal antibodies did not detect any Abeta deposits even in old NEP knockout mice.Surprisingly, tests of learning and memory revealed that the ability to learn was not reduced in old NEP-deficient mice but instead had significantly improved, and sustained learning and memory in the aged mice was congruent with improved long-term potentiation (LTP) in brain slices of the hippocampus and lateral amygdala.

View Article: PubMed Central - PubMed

Affiliation: Department of Cardiology, Charité-University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany. thomas.walther@hyms.ac.uk

ABSTRACT

Background: Neutral endopeptidase, also known as neprilysin and abbreviated NEP, is considered to be one of the key enzymes in initial human amyloid-beta (Abeta) degradation. The aim of our study was to explore the impact of NEP deficiency on the initial development of dementia-like symptoms in mice.

Methodology/principal findings: We found that while endogenous Abeta concentrations were elevated in the brains of NEP-knockout mice at all investigated age groups, immunohistochemical analysis using monoclonal antibodies did not detect any Abeta deposits even in old NEP knockout mice. Surprisingly, tests of learning and memory revealed that the ability to learn was not reduced in old NEP-deficient mice but instead had significantly improved, and sustained learning and memory in the aged mice was congruent with improved long-term potentiation (LTP) in brain slices of the hippocampus and lateral amygdala. Our data suggests a beneficial effect of pharmacological inhibition of cerebral NEP on learning and memory in mice due to the accumulation of peptides other than Abeta degradable by NEP. By conducting degradation studies and peptide measurements in the brain of both genotypes, we identified two neuropeptide candidates, glucagon-like peptide 1 and galanin, as first potential candidates to be involved in the improved learning in aged NEP-deficient mice.

Conclusions/significance: Thus, the existence of peptides targeted by NEP that improve learning and memory in older individuals may represent a promising avenue for the treatment of neurodegenerative diseases.

Show MeSH
Related in: MedlinePlus