Miners, James Scott PhD; Baig, Shabnam PhD; Tayler, Hannah BSc; Kehoe, Patrick Gavin PhD; Love, Seth MBBCh, PhD, FRCP, FRCPath
The β-amyloid (Aβ) cascade hypothesis posits that the neurodegenerative abnormalities of Alzheimer disease (AD) are initiated by the accumulation of Aβ in the brain (1, 2). Production of Aβ is normally balanced by its removal through perivascular drainage (3), receptor-mediated transport across the blood-brain barrier (4), and enzymatic degradation (5, 6). In familial AD, increased amyloidogenic processing of amyloid precursor protein (APP), as a result of mutations in the APP or presenilin genes (7), causes Aβ to accumulate. In sporadic late-onset AD, in which no gene mutations seem to be causative, higher levels of β-secretase (the rate-limiting enzyme of APP amyloidogenesis) have been detected in human brain tissue (8). In mouse models of AD, however, abnormal accumulation of Aβ itself seems to induce a paradoxical upregulation of β-secretase activity (9). In sporadic AD, it remains unclear as to what extent Aβ accumulation results from the relative contributions of increased production and decreased removal.
Both neprilysin (NEP) and insulin-degrading enzyme (IDE) cleave Aβ in vitro (10). Disruption/deletion of the NEP (11-13) or IDE genes (14, 15) in mice caused Aβ to accumulate in the brain. In contrast, overexpression of human NEP or IDE in mice transgenic for human APP (hAPP) lowered Aβ levels in the brain, reduced plaque formation, and improved cognitive performance (16). The Aβ load was also lowered when human NEP was transferred into the brains of hAPP mice by lentiviral vector (12, 17) or implantation of genetically engineered fibroblasts (18).
Overexpression of NEP was reported to protect hippocampal neurons from Aβ-mediated toxicity in vitro (19), and upregulation of NEP or IDE has been proposed as a potential means of protecting the brain against Aβ accumulation and consequent cognitive decline (18, 20, 21).
We and others previously found that NEP was expressed by pyramidal neurons in the cerebral neocortex and hippocampus, and that neuronal immunopositivity for NEP was reduced in AD (22, 23). The IDE is also expressed by neurons and in AD; neuronal labeling for IDE is reduced in the CA2/3 and CA4 regions of the hippocampus (24). Reduced IDE protein levels or enzyme activity have been reported in the cytosolic (25) and membrane (26) fractions of postmortem brain tissue from patients with mild cognitive impairment or AD, whereas levels of both NEP and IDE have been found to be decreased with age in postmortem brain tissue from both AD patients and controls (27) and in hAPP transgenic mice (28, 29). APOE ϵ4, a genetic risk factor for AD, has been associated with reduced levels of NEP (23) and IDE (30) in AD.
These collective findings are consistent with the hypothesis that accumulation of Aβ in sporadic late-onset AD may result from reduced activity of NEP and/or IDE. The previous studies on human tissue have, however, mostly focused on relatively small cohorts of patients with end-stage disease and have rarely adjusted for reduction in NEP and IDE levels secondary to neuronal loss or damage. Furthermore, it has generally been assumed that measurements of NEP and IDE mRNA and protein levels provide a reliable indicator of the activity of these enzymes. This assumption may not be valid, however, particularly if there are disease-specific modifications to these proteins that might affect their activities.
Our aim in the present study was to determine whether sporadic AD results from decreased NEP and/or IDE activity or whether decreased NEP and/or IDE activity in sporadic AD is secondary to neurodegenerative changes. We examined NEP and IDE levels and their corresponding specific activities at different pathological stages of disease (as defined by Braak tangle stages) and in relation to APOE genotype in a large cohort of postmortem brains. We also measured the levels of neuron-specific enolase (NSE) as a marker of neuronal integrity (31) and used them to adjust measurements of NEP and IDE levels and enzymatic activities to account for neuronal loss or damage. Our data indicate that when adjusted for neuronal degeneration (as defined by reduction in NSE levels), NEP and IDE levels and activity in frontal cortex increase with progression of AD and are not influenced by APOE genotype.
MATERIALS AND METHODS
Brain tissue was obtained from the South West Dementia Brain Bank (Human Tissue Authority licence no. 12273), University of Bristol, United Kingdom, with the local research ethics committee approval. The brains had been subdivided midsagittally at autopsy; the right halves were fixed in formalin, and multiple blocks were taken for paraffin histology and detailed neuropathologic examination. The left halves were sliced, and the slices were frozen at −80°C for subsequent biochemical studies. To ensure that we examined the full spectrum of disease progression, cases of definite AD (assessed according to the criteria of the Consortium to Establish a Registry for Alzheimer's Disease [CERAD]) and controls (showing the absence of AD or other neuropathologic abnormalities) and cases with AD pathology of intermediate severity (CERAD possible and probable AD) were examined. All cases were categorized according to the Braak tangle stage and grouped for analysis into 3 categories: Braak Stages 0 to II, III to IV, and V to VI. For analyses in which the comparison was between AD and control brains, cases of possible, probable, and definite AD (as defined by CERAD) were included in the AD group. The AD and control groups were matched in terms of age, sex, and postmortem delay (Table).
TABLE. Summary of Pa...Image Tools
Brain tissue homogenates and subcellular fractions were prepared from dissected samples of unfixed frozen cortex from the left midfrontal region (Brodmann area 6). Protein concentrations were determined using Total Protein Kit (Sigma Aldrich, Dorset, UK) according to manufacturer's instructions. Tissue (200 mg) allowed to thaw to 4°C was homogenized for 30 seconds in a Precellys 24 automated tissue homogenizer (Stretton Scientific, Derbyshire, UK) with 2.3-mm silica beads (Biospec, Thistle Scientific, Glasgow, UK). For ELISA, the tissue was homogenized in 1% sodium dodecyl sulfate buffer containing 5 mol/L NaCl, 1 mol/L Tris pH 7.6. For enzyme activity assays, the tissue was homogenized in 0.5% triton X-100, 20 mmol/L Tris pH 7.4, 10% sucrose (wt/vol) (all reagents from Sigma Aldrich). Both sets of lysis buffers contained aprotinin (1 μg/mL; Sigma Aldrich) and phenylmethane sulfonyl fluoride (10 μmol/L; Sigma Aldrich). The homogenates were centrifuged at 13,000 rpm for 15 minutes at 4°C, and the supernatants were aliquoted and stored at −80°C until used.
For preparation of tissue for separate analysis of plasma membrane and cytosolic fractions, 400-mg samples of cortex were homogenized in 1 mL ice-cold PBS with a handheld homogenizer. Fractionation was then performed using the Fraction-Prep system (BioVision, Cambridge Biosciences, Cambridge, UK) according to the manufacturer's guidelines. All fractions were stored at −80°C until used.
The Neprilysin Duoset ELISA kit (R&D Systems Europe, Oxford, UK) was used according to the manufacturer's guidelines with minor modifications. Goat anti-human NEP (1.6 μg/mL) diluted in PBS (pH 7.4) was coated on a high-binding Costar 96-well plate (R&D Systems Europe) for 18 hours at room temperature (RT). The plates were washed 6 times in PBS containing 0.5% Tween 20 (Sigma Aldrich). Nonspecific binding of antibody was blocked by addition of 1% PBS-bovine serum albumin (BSA) (Sigma Aldrich) for 3 hours at RT, after which the plates were washed a further 6 times. Serial dilutions of recombinant human NEP or crude homogenates (250 μg total protein) diluted in 1% PBS-BSA, or 1% PBS-BSA alone as a control, were incubated for 2 hours with continuous shaking at RT. After a further 6 washes, biotinylated anti-NEP (1.6 μg/mL) was added for 2 hours before another wash and incubation with streptavidin-peroxidase (1:100) for 20 minutes in the dark. Substrate solution (tetramethyl benzidine; R&D Systems Europe) was added for 30 minutes, and the optical density for each well was read at 450 nm in a FLUOstar plate reader (BMG Labtech, Aylesbury, UK). The NEP protein levels were interpolated from the standard curve generated from serial dilutions of recombinant human NEP (R&D Systems Europe). Each measurement was repeated on 2 separate occasions, and the average value was calculated. According to the manufacturers, the ELISA demonstrated minimal cross reactivity with recombinant human NEP-2, endothelin-converting enzyme 1, or endothelin-converting enzyme 2.
An indirect sandwich ELISA was developed and optimized to measure IDE levels in crude tissue homogenates. Anti-human IDE (2.5 μg/mL) (Abcam, Cambridge, UK) diluted in PBS was coated on high-binding Costar 96-well plates and incubated overnight at RT. The plates were washed 6 times in PBS containing 0.5% Tween 20, and nonspecific binding of antibody was blocked by addition of 1% PBS-BSA (Sigma Aldrich) for 3 hours at RT, after which the plates were washed a further 6 times. Recombinant human IDE (R&D Systems Europe) and crude tissue homogenates (total protein, 100 μg) were incubated for 3 hours at RT with shaking. After 6 washes, anti-IDE ([2 μg/mL] R&D Systems) was added to the plate for 2 hours at RT. After a further wash, biotinylated anti-mouse antibody (1:500) (Vector Labs, Peterborough, UK) was added for 20 minutes, followed by a wash step and incubation with streptavidin-peroxidase ([1:100] R&D Systems Europe) for 20 minutes in the dark. Tetramethyl benzidine was added for 30 minutes and the optical density for each well was read at 450 nm. The IDE protein levels were interpolated from a standard curve generated by serial dilution of recombinant human IDE (R&D Systems Europe). The assay was linear in range 0 to 1,000 ng/mL.
NEP Enzyme Activity Assay
The enzyme activity assay has been described in detail (32). Goat anti-human NEP (1.6 μg/mL) diluted in PBS (pH 7.4) was coated on Nunc maxi-sorp 96-well plates (Fisher Scientific, Loughborough, UK). Nonspecific binding of antibody was blocked by addition of 1% PBS-BSA. After washing, crude brain tissue homogenate or subcellular fractions (250 μg total protein), diluted in PBS (pH 7.4), were incubated together with serial dilutions of recombinant human NEP (R&D Systems Europe) at RT for 2 hours with continuous shaking. After washing, thiorphan (200 nmol/L) (Sigma Aldrich) prepared in 100 mmol/L Tris-HCl pH 7.5, 50 mmol/L NaCl, 10 μmol/L ZnCl2, or buffer alone, was added to the wells for 10 minutes at RT. The fluorogenic peptide Mca-RPPGFSAFK(Dnp)-OH (10 μmol/L) (R&D Systems Europe) diluted in 100 mmol/L Tris-HCl pH 7.5, 50 mmol/L NaCl, and 10 μmol/L ZnCl2 was added to each well and incubated at 37°C for 3 hours in the dark. Fluorescence was measured with excitation at 320 nm and emission at 405 nm.
Each homogenate and subcellular fraction were assayed in duplicate in the presence and absence of the NEP inhibitor, thiorphan. Specific NEP activity levels were calculated after subtraction of fluorescence in the presence of thiorphan. The NEP levels were interpolated from a standard curve generated from serial dilutions of recombinant full-length NEP. Average NEP enzyme was calculated for each homogenate from measurements made on 2 separate occasions.
IDE Activity Assay
The assay method was similar to that for NEP and has been described in detail (32, 33). Briefly, anti-IDE antibodies raised against the inactive domain of IDE (2.5 μg/mL) (Abcam) were used to capture IDE in crude brain tissue homogenates or subcellular fractions (100 μg total protein). Enzyme activity was calculated by measuring fluorescence after the addition of the fluorogenic peptide substrate Mca-RPPGFSAFK(Dnp)-OH (10 μmol/L) (R&D Systems Europe); the assay was calibrated against serial dilutions of full-length rat IDE (Calbiochem, Nottingham, UK).
Measurement of NSE
Neuron-specific enolase levels in crude brain homogenates were measured by a direct ELISA. Costar EIA microplate wells were coated with 10 μL of brain tissue homogenate in 90 µL PBS for 2 hours, washed 5 times with PBS containing 0.5% Tween 20 and incubated with anti-NSE (1:1000) for 2 hours. After further washes, mouse-horseradish peroxidase (1:100) was added for 20 minutes in the dark, washed, and tetramethyl benzidine was added for 10 minutes. Serial dilutions of recombinant human NSE (Biomol, Exeter, UK) were used to construct a best-fit curve, and NSE concentrations were calculated by interpolation. Each sample was run in duplicate, and the mean was determined. The NSE concentration was used to provide a proxy measurement of the number of neurons in tissue samples. Measurements of NEP and IDE levels and activity were adjusted for neuronal loss or damage according to the formula: NEP/IDEadjusted = NEP/IDEsample × NSEmean control/NSEsample, where NEP/IDEadjusted corresponds to the level or activity of NEP or IDE after correcting for neuronal loss in the sample, NEP/IDEsample was the unadjusted level or activity of NEP or IDE in the sample, NSEmean control was the mean level of NSE in the control cohort, and NSEsample was the measured level of NSE in individual test samples.
NEP and IDE Activities and Postmortem Stability
To test the postmortem stability of NEP and IDE activity, we used 1) several immediately adjacent samples of frontal cortex from 2 control brains with relatively short postmortem delays (6-10 hours) that were incubated for 6, 12, 18, 24, 48, and 72 hours at RT, and 24, 48, and 72 hours at 4°C before homogenization; 2) cold (4°C) saline-perfused mouse brains (n = 3) that were finely chopped, mixed and incubated for 10, 30, 60, 120, and 360 minutes before homogenization; and 3) recombinant human standards (500 ng/mL) that were diluted in PBS and incubated at 4°C or RT for 10, 30, 60, and 120 minutes before measurement. Enzyme activities were measured as previously described.
Data were analyzed using an independent-samples t-test, 1-way analysis of variance (ANOVA) with Bonferroni posttesting, and Spearman or Pearson correlation analysis, as appropriate, using the Statistical Package for Social Science software (12.0.1). Values of p < 0.05 were considered statistically significant. To normalize the data (which were right-skewed in both the AD and control groups), all NSE-adjusted NEP and IDE protein and activity measurements were logarithmically transformed for statistical analysis.
NSE Level in Relation to Braak Stage
The NSE level was significantly reduced in AD compared with controls (p = 0.0018) (Fig. 1A). The NSE levels differed significantly between the Braak Stages 0 to II, III to IV, and V to VI subgroups (p = 0.003, ANOVA). Post hoc comparisons showed that the level in Braak Stages V to VI was significantly lower than in Braak Stages 0 to II (p = 0.020) and Stages III to IV (p = 0.013) (Fig. 1B).
NEP Level in Relation to AD Diagnosis and Braak Stage
Unadjusted NEP level did not significantly differ with respect to diagnosis (Fig. 2A) or Braak stage (Fig. 2B), although there was a trend toward increased NEP level in AD and with Braak stage. After adjustment for NSE, NEP level was significantly higher in AD cases than controls (p = 0.007) (Fig. 2C) and showed a highly significant correlation with Braak stage (p = 0.001, rs = 0.301) (Fig. 2D). Adjusted NEP level significantly differed between the Braak Stages 0 to II, III to IV, and V to VI subgroups (p = 0.004, ANOVA); post hoc comparison between subgroups indicated that the level was significantly higher in Braak Stages V to VI than Braak Stages 0 to II brains (p = 0.006) (Fig. 2D).
NEP Level in Relation to Age
In a combined cohort including all cases, there was a trend toward an inverse correlation between NEP level and age, but this was not significant for either unadjusted or adjusted levels. In the AD cohort alone, unadjusted (p = 0.066, Pearson r = −0.194) and adjusted (p = 0.023, r = −0.238) NEP level showed an inverse correlation with age (Figs. 2E, F). In the control cohort alone, unadjusted and adjusted NEP level did not correlate with age (Figs. 2E, F). When analyzed according to Braak tangle stage, unadjusted and adjusted NEP levels did not correlate with age in the Braak Stages 0 to II or III to IV subgroups. In Braak Stages V to VI, however, both unadjusted (p = 0.036, r = −0.259) and adjusted (p = 0.027, r = −0.272) NEP levels negatively correlated with age.
NEP Activity in Relation to AD Diagnosis and Braak Stage
Unadjusted NEP activity was significantly higher in the AD cohort (p = 0.035) (Fig. 3A) but did not correlate significantly with Braak stage (Fig. 3B). Adjusted NEP activity was significantly higher in the AD cohort (p = 0.001) (Fig. 3C), positively correlated with Braak stage (p = 0.008, rs = 0.232) (Fig. 3D), and significantly differed between the Braak Stages 0 to II, III to IV, and V to VI subgroups (p = 0.009, ANOVA) (Fig. 3D). Further post hoc testing showed the increased activity in Braak Stages V to VI to approach significance compared with Stages 0 to II (p = 0.051) and to be significantly greater than in the Stages III to IV subgroup (p = 0.027).
NEP Activity in Relation to Age
In the combined cohort, unadjusted NEP activity inversely correlated with age (p = 0.021, r = −0.188), but the correlation disappeared after correction for NSE. Within the AD cohort alone, both unadjusted (p = 0.004, r = −0.291) and adjusted (p = 0.001, r = −0.337) NEP activity showed a significant inverse correlation with age (Figs. 3E, F). In the control group, neither unadjusted nor adjusted NEP levels correlated with age. When stratified according to Braak stage subgroups, unadjusted and adjusted NEP activity did not correlate with age in the Stages 0 to II or III to IV subgroups. In Braak Stages V to VI, both unadjusted (p = 0.001, r = −0.388) and adjusted (p = 0.0007, r = −0.398) NEP activity showed significant inverse correlations with age.
IDE Level in Relation to AD Diagnosis and Braak Stage
Unadjusted IDE level was significantly decreased in the AD cohort compared with controls (p = 0.003) (Fig. 4A) and showed a significant inverse correlation with Braak stage (p = 0.028, rs = −0.197) (Fig. 4B). Analysis of variance revealed significantly different IDE level in Braak Stages 0 to II, III to IV, and V to VI subgroups (p = 0.024), and post hoc testing indicated that the level was significantly lower in the Braak Stages V to VI than the 0 to II subgroup (p = 0.032) (Fig. 4B). After adjustment for NSE, however, neuronal IDE level did not correlate with AD or Braak stage or differ significantly between Braak subgroups (Figs. 4C, D).
IDE Level in Relation to Age
In the combined cohort, unadjusted and adjusted IDE protein did not correlate with age. In the AD group, adjusted (p = 0.004, r = −0.332) but not unadjusted IDE level inversely correlated with age (Figs. 4E, F). There was no correlation of either adjusted or unadjusted IDE level with age in the control group. In Braak Stages 0 to II and II to IV, IDE level did not correlate with age. In Braak Stages V to VI, both unadjusted (p = 0.032, r = −0.259) and adjusted (p = 0.0061, r = −0.332) IDE levels were significantly inversely correlated with age.
IDE Activity in Relation to AD Diagnosis and Braak Stage
Unadjusted IDE activity did not vary according to diagnosis (Fig. 5A) or Braak tangle stage (Fig. 5B). Adjusted IDE activity was higher in AD than control brains (p = 0.016) (Fig. 5C) and positively correlated with Braak stage (p = 0.031, rs = 0.191) (Fig. 5D), ANOVA did not reveal significant differences between the Braak Stages 0 to II, III to IV, and V to VI subgroups.
IDE Activity in Relation to Age
In the combined cohort, neither unadjusted nor adjusted IDE activity correlated with age. No effect of age was seen on comparison of IDE activity when examined in either the AD or control groups (Figs. 5E, F) or when stratifying according to Braak stage.
NEP and IDE Levels and Activity in Relation to APOE
In the cohort as a whole, there was no significant association between APOE and NEP level or activity (either unadjusted or adjusted). Similarly, no significant association was demonstrated between APOE and IDE levels or activity.
NEP and IDE Activity in Subcellular Fractions
Cytosolic NEP activity was significantly increased in AD (p = 0.008) (Fig. 6A) and significantly correlated with Braak stage (p = 0.029, rs = 0.301) (Fig. 6B). The NEP activity in the plasma membrane fraction did not significantly differ with diagnosis or Braak stage (Figs. 6C, D). The IDE activity did not differ with either diagnosis or correlate with Braak stage in either the cytosolic (Figs. 6E, F) or the plasma membrane (Figs. 6F, G) fraction.
NEP and IDE Activity in Relation to Postmortem Delay and Tissue Storage
The NEP and IDE levels and activities did not correlate with sex, age, or postmortem delay. Incubation of human brain tissue at RT or 4°C for up to 72 hours did not result in significant changes in NEP or IDE activity (Figs. 7A-D). Incubation of recombinant human standards (Figs. 7E, F) and mouse brain tissue homogenates (Figs. 7G, H) at RT or 4°C for shorter incubation times (10, 30, 60, 120, and 360 minutes) did not result in an appreciable loss of enzyme activity.
We have performed the most comprehensive analysis to date of NEP and IDE levels and activity in AD and control brain tissue. Our findings do not support the hypothesis that a reduction in the activity of either enzyme is a primary cause of Aβ accumulation in AD. After adjustment for NSE levels, the amount and activity of NEP were found to rise with progression of disease, as indicated by Braak tangle stage. Although unadjusted IDE levels fell with progression of disease, adjusted IDE levels did not significantly change, and IDE activity increased in AD. NEP levels and activity declined more steeply in relation to age at death in AD than control brains, but this reflected elevated NEP levels in younger AD patients rather than reduction below control levels in the older patients.
In previous studies, we (23) and others (22, 34) showed reduced immunolabeling of neuronal NEP in AD. NEP mRNA levels have also been reported to be reduced in AD (34, 35), particularly within regions vulnerable to neurofibrillary tangles and neuritic plaque formation. Western blot analysis of NEP in human brain tissue homogenates has yielded inconsistent findings; some studies showed reduced levels of NEP in AD (35, 36), and others showed no significant changes (23, 27, 37). We also noted that hippocampal IDE (24) labeling was reduced in AD; others have reported reduced IDE mRNA and protein as well as reduced IDE enzyme activity in AD (25, 26, 30).
With very few exceptions (36), previous studies of NEP and IDE expression have been based on small sample sizes, a general lack of adjustment of measurements for the effects of neuronal loss or damage (e.g. as a result of tau hyperphosphorylation) and an almost exclusive focus on end-stage AD (i.e. on patients with Braak Tangle Stage V or VI disease and a diagnosis of "definite AD" according to CERAD criteria). In contrast to immunolabeling of NEP in human sections or semiquantitative Western blot analysis, we used indirect sandwich ELISAs and immunocapture-based fluorogenic activity assays to measure protein levels and enzyme activity, respectively, in brain tissue homogenates. These methodological differences, the size and disease characteristics of selected cohorts, the adjustment for neuronal loss, the preparation of tissue samples in the presence of a nonionic detergent, and the area studied (Brodmann Area 6) are likely to account for differences between the present and previous findings. We also note that although NEP and IDE are predominantly neuronal (23, 24), expression in other cell types such as astrocytes and microglia and the presence of soluble forms of the enzymes may have contributed to measurements made on brain tissue homogenates.
There was considerable case-by-case variation in our cohorts. One potential cause and limitation of the present study might be the effect of postmortem delay on enzyme stability. We did not find any association between postmortem delay and NEP or IDE protein level or activity, however. Furthermore, we assessed this by simulating postmortem conditions in human brain tissue during a long period (up to 72 hours) and in mouse brain tissue during a shorter period (10-120 minutes) at both RT and 4°C, as well as by measuring the activity of recombinant human protein. These results suggest that postmortem delay was not a significant contributor to case-by-case variation.
The findings presented here are consistent with those of other studies. In human CSF, NEP increased with progression from mild cognitive impairment to AD (38). The levels of both NEP and IDE were found to be greater in aging hAPP mice than in wild-type littermates, the increase predominantly occurring after the appearance of plaques (39-41). Furthermore, IDE mRNA levels strongly correlated with those of Aβ1-40 and Aβ1-42 in hAPP transgenic mice (41). These data suggest that upregulation of NEP and IDE may be induced by Aβ; in keeping with this is the sharp rise in numbers of NEP- and IDE-positive astrocytes in close proximity to Aβ plaques in hAPP mice at 16 months of age (39, 40).
Other supportive work comes from a variety of in vitro studies. Incubation of neuronal, glial, or smooth muscle cells with fibrillar Aβ1-40 or Aβ1-42 induced expression of a number of Aβ-degrading enzymes, including IDE (40), matrix metalloproteinase 2, 3, and 9 (42-44), and plasminogen activators (45). Enhancement of NEP was found to be neuroprotective in vitro (19) and also ameliorated extracellular amyloid pathology, synaptic dysfunction, and memory defects in both mouse models of Aβ amyloidosis (20) and in transgenic Drosophila (46). It seems likely that any perturbation in Aβ production or removal that results in Aβ accumulation is likely to induce expression of Aβ-degrading enzymes and is a physiological response that minimizes further Aβ accumulation.
The possible transcriptional regulation of NEP expression by APP-FL or its derivatives such as APP intracellular domain has been explored. Pardossi-Piquard et al (47) demonstrated presenilin-dependant transcriptional control of NEP by APP intracellular domain and Aβ precursor-like protein, but another recent study did not find evidence of APP intracellular domain-mediated regulation (48). Posttranslational modification may provide another means of increasing enzyme activity. NEP undergoes posttranslational modification by N-glycosylation and has also been shown to be phosphorylated by casein kinase 2 (49). The activity of several kinases that can potentially phosphorylate NEP and IDE are increased in AD, such as glycogen synthase kinase 3 (50). The effects of phosphorylation and glycosylation on NEP and IDE enzyme activities are currently unknown.
Several studies have found that brain NEP protein and mRNA levels decline with age in humans (27, 37), transgenic hAPP and wild-type mice, (28, 29, 41) and in Drosophila (46). It may be of relevance that the level of somatostatin, a neuropeptide that has been shown to stimulate NEP activity, also falls with age (51) and is lower in people positive for APOE ϵ4 (52). We found that the decline in NEP activity with age at death was significant only in AD, reflecting relative elevation of NEP activity in younger AD patients and a convergence of NEP activity to control levels with increasing age. In contrast, IDE activity did not change with age at death. This is in keeping with studies in hAPP mice (40, 41).
In summary, our findings suggest that reduction in NEP and IDE activity is not the primary cause of Aβ accumulation in AD, but rather a late-stage phenomenon secondary to neurodegeneration.
1. Hardy JA, Higgins GA. Alzheimer's disease: The amyloid cascade hypothesis. Science 1992;256:184-85
2. Selkoe DJ. The molecular pathology of Alzheimer's disease. Neuron 1991;6:487-98
3. Weller RO, Subash M, Preston SD, et al. Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer's disease. Brain Pathol 2008;18:253-66
4. Deane R, Sagare A, Zlokovic BV. The role of the cell surface LRP and soluble LRP in blood-brain barrier Abeta clearance in Alzheimer's disease. Curr Pharm Des 2008;14:1601-5
5. Carson JA, Turner AJ. Beta-amyloid catabolism: Roles for neprilysin (NEP) and other metallopeptidases? J Neurochem 2002;81:1-8
6. Turner AJ, Fisk L, Nalivaeva NN. Targeting amyloid-degrading enzymes as therapeutic strategies in neurodegeneration. Ann N Y Acad Sci 2004;1035:1-20
7. Hardy J. Amyloid, the presenilins and Alzheimer's disease. Trends Neurosci 1997;20:154-59
8. Tyler SJ, Dawbarn D, Wilcock GK, et al. alpha- and beta-secretase: Profound changes in Alzheimer's disease. Biochem Biophys Res Commun 2002;299:373-76
9. Zhao J, Fu Y, Yasvoina M, et al. Beta-site amyloid precursor protein cleaving enzyme 1 levels become elevated in neurons around amyloid plaques: Implications for Alzheimer's disease pathogenesis. J Neurosci 2007;27:3639-49
10. Wang DS, Dickson DW, Malter JS. Beta-amyloid degradation and Alzheimer's disease. J Biomed Biotechnol 2006;2006:58406
11. Eckman EA, Adams SK, Troendle FJ, et al. Regulation of steady-state beta-amyloid levels in the brain by neprilysin and endothelin-converting enzyme but not angiotensin-converting enzyme. J Biol Chem 2006;281:30471-78
12. Iwata N, Mizukami H, Shirotani K, et al. Presynaptic localization of neprilysin contributes to efficient clearance of amyloid-beta peptide in mouse brain. J Neurosci 2004;24:991-98
13. Iwata N, Tsubuki S, Takaki Y, et al. Metabolic regulation of brain Abeta by neprilysin. Science 2001;292:1550-52
14. Farris W, Mansourian S, Chang Y, et al. Insulin-degrading enzyme regulates the levels of insulin, amyloid beta-protein, and the beta-amyloid precursor protein intracellular domain in vivo. Proc Natl Acad Sci U S A 2003;100:4162-67
15. Miller BC, Eckman EA, Sambamurti K, et al. Amyloid-beta peptide levels in brain are inversely correlated with insulysin activity levels in vivo. Proc Natl Acad Sci U S A 2003;100:6221-26
16. Leissring MA, Farris W, Chang AY, et al. Enhanced proteolysis of beta-amyloid in APP transgenic mice prevents plaque formation, secondary pathology, and premature death. Neuron 2003;40:1087-93
17. Marr RA, Rockenstein E, Mukherjee A, et al. Neprilysin gene transfer reduces human amyloid pathology in transgenic mice. J Neurosci 2003;23:1992-96
18. Hemming ML, Patterson M, Reske-Nielsen C, et al. Reducing amyloid plaque burden via ex vivo gene delivery of an Abeta-degrading protease: A novel therapeutic approach to Alzheimer disease. PLoS Med 2007;4:e362
19. El-Amouri SS, Zhu H, Yu J, et al. Neprilysin protects neurons against Abeta peptide toxicity. Brain Res 2007;1152:191-200
20. El-Amouri SS, Zhu H, Yu J, et al. Neprilysin: An enzyme candidate to slow the progression of Alzheimer's disease. Am J Pathol 2008;172:1342-54
21. Poirier R, Wolfer DP, Welzl H, et al. Neuronal neprilysin overexpression is associated with attenuation of A(-related spatial memory deficit. Neurobiol Dis 2006;24:475-83
22. Carpentier M, Robitaille Y, DesGroseillers L, et al. Declining expression of neprilysin in Alzheimer disease vasculature: Possible involvement in cerebral amyloid angiopathy. J Neuropathol Exp Neurol 2002;61:849-56
23. Miners JS, Van Helmond Z, Chalmers K, et al. Decreased expression and activity of neprilysin in Alzheimer disease are associated with cerebral amyloid angiopathy. J Neuropathol Exp Neurol 2006;65:1012-21
24. Miners JS, Baig S, Palmer J, et al. Abeta-degrading enzymes in Alzheimer's disease. Brain Pathol 2008;18:240-52
25. Perez A, Morelli L, Cresto JC, et al. Degradation of soluble amyloid beta-peptides 1-40, 1-42, and the Dutch variant 1-40Q by insulin degrading enzyme from Alzheimer disease and control brains. Neurochem Res 2000;25:247-55
26. Zhao Z, Xiang Z, Haroutunian V, et al. Insulin degrading enzyme activity selectively decreases in the hippocampal formation of cases at high risk to develop Alzheimer's disease. Neurobiol Aging 2007;28:824-30
27. Hellstrom-Lindahl E, Ravid R, Nordberg A. Age-dependent decline of neprilysin in Alzheimer's disease and normal brain: Inverse correlation with Abeta levels. Neurobiol Aging 2008;29:210-21
28. Caccamo A, Oddo S, Sugarman MC, et al. Age- and region-dependent alterations in Abeta-degrading enzymes: Implications for Abeta-induced disorders. Neurobiol Aging 2005;26:645-54
29. Carter TL, Pedrini S, Ghiso J, et al. Brain neprilysin activity and susceptibility to transgene-induced Alzheimer amyloidosis. Neurosci Lett 2006;392:235-39
30. Cook DG, Leverenz JB, McMillan PJ, et al. Reduced hippocampal insulin-degrading enzyme in late-onset Alzheimer's disease is associated with the apolipoprotein E-epsilon4 allele. Am J Pathol 2003;162:313-19
31. Harada H, Tamaoka A, Ishii K, et al. Beta-site APP cleaving enzyme 1 (BACE1) is increased in remaining neurons in Alzheimer's disease brains. Neurosci Res 2006;54:24-29
32. Miners JS, Verbeek MM, Rikkert MO, et al. Immunocapture-based fluorometric assay for the measurement of neprilysin-specific enzyme activity in brain tissue homogenates and cerebrospinal fluid. J Neurosci Methods 2008;167:229-36
33. Miners JS, Kehoe PG, Love S. Immunocapture-based fluorometric assay for the measurement of insulin-degrading enzyme activity in brain tissue homogenates. J Neurosci Methods 2008;169:177-81
34. Akiyama H, Kondo H, Ikeda K, et al. Immunohistochemical localization of neprilysin in the human cerebral cortex: Inverse association with vulnerability to amyloid beta-protein (Abeta) deposition. Brain Res 2001;902:277-81
35. Yasojima K, Akiyama H, McGeer EG, et al. Reduced neprilysin in high plaque areas of Alzheimer brain: A possible relationship to deficient degradation of beta-amyloid peptide. Neurosci Lett 2001;297:97-100
36. Wang DS, Lipton RB, Katz MJ, et al. Decreased neprilysin immunoreactivity in Alzheimer disease, but not in pathological aging. J Neuropathol Exp Neurol 2005;64:378-85
37. Russo R, Borghi R, Markesbery W, et al. Neprylisin decreases uniformly in Alzheimer's disease and in normal aging. FEBS Lett 2005;579:6027-30
38. Maruyama M, Higuchi M, Takaki Y, et al. Cerebrospinal fluid neprilysin is reduced in prodromal Alzheimer's disease. Ann Neurol 2005;57:832-42
39. Apelt J, Ach K, Schliebs R. Aging-related down-regulation of neprilysin, a putative beta-amyloid-degrading enzyme, in transgenic Tg2576 Alzheimer-like mouse brain is accompanied by an astroglial upregulation in the vicinity of beta-amyloid plaques. Neurosci Lett 2003;339:183-86
40. Leal MC, Dorfman VB, Gamba AF, et al. Plaque-associated overexpression of insulin-degrading enzyme in the cerebral cortex of aged transgenic tg2576 mice with Alzheimer pathology. J Neuropathol Exp Neurol 2006;65:976-87
41. Vepsalainen S, Hiltunen M, Helisalmi S, et al. Increased expression of Abeta degrading enzyme IDE in the cortex of transgenic mice with Alzheimer's disease-like neuropathology. Neurosci Lett 2008;438:216-20
42. Deb S, Zhang JW, Gottschall PE. Activated isoforms of MMP-2 are induced in U87 human glioma cells in response to beta-amyloid peptide. J Neurosci Res 1999;55:44-53
43. Jung SS, Zhang W, Van Nostrand WE. Pathogenic A beta induces the expression and activation of matrix metalloproteinase-2 in human cerebrovascular smooth muscle cells. J Neurochem 2003;85:1208-15
44. Lee JM, Yin KJ, Hsin I, et al. Matrix metalloproteinase-9 and spontaneous hemorrhage in an animal model of cerebral amyloid angiopathy. Ann Neurol 2003;54:379-82
45. Tucker HM, Kihiko-Ehmann M, Wright S, et al. Tissue plasminogen activator requires plasminogen to modulate amyloid-beta neurotoxicity and deposition. J Neurochem 2000;75:2172-77
46. Iijima-Ando K, Hearn SA, Granger L, et al. Overexpression of neprilysin reduces Alzheimer amyloid-beta42 (Abeta42)-induced neuron loss and intraneuronal Abeta42 deposits but causes a reduction in cAMP-responsive element-binding protein-mediated transcription, age-dependent axon pathology, and premature death in Drosophila
. J Biol Chem 2008;283:19066-76
47. Pardossi-Piquard R, Petit A, Kawarai T, et al. Presenilin-dependent transcriptional control of the Abeta-degrading enzyme neprilysin by intracellular domains of betaAPP and APLP. Neuron 2005;46:541-54
48. Chen AC, Selkoe DJ. Response to: Pardossi-Piquard et al. "Presenilin-Dependent Transcriptional Control of the Abeta-Degrading Enzyme Neprilysin by Intracellular Domains of betaAPP and APLP." Neuron 46, 541-554. Neuron 2007;53:479-83
49. Ganju RK, Shpektor RG, Brenner DG, et al. CD10/neutral endopeptidase 24.11 is phosphorylated by casein kinase II and coassociates with other phosphoproteins including the lyn src-related kinase. Blood 1996;88:4159-65
50. Hooper C, Killick R, Lovestone S. The GSK3 hypothesis of Alzheimer's disease. J Neurochem 2008;104:1433-39
51. Saito T, Iwata N, Tsubuki S, et al. Somatostatin regulates brain amyloid beta peptide Abeta42 through modulation of proteolytic degradation. Nat Med 2005;11:434-39
52. Grouselle D, Winsky-Sommerer R, David JP, et al. Loss of somatostatin-like immunoreactivity in the frontal cortex of Alzheimer patients carrying the apolipoprotein epsilon 4 allele. Neurosci Lett 1998;255:21-24