Validation of amyloid-beta peptides in CSF diagnosis of neurodegenerative dementias

Biomarkers for differential diagnosis of the three most frequent degenerative forms of dementia, Alzheimer's disease (AD), dementia with Lewy bodies (DLB) and frontotemporal dementias (FTD), are currently under intensive investigation, but disease-specific biomarkers for FTD and DLB are still lacking. We analyzed 303 cerebrospinal fluid (CSF) samples of 71 AD, 32 DLB and 36 FTD patients in comparison to 93 various other dementias (OD), 20 peripheral neurologic disease (PND) controls, 25 neurodegenerative disorders without dementia (ND) and 26 depressive cognitive complainers (DCC) for distinct CSF amyloid-beta (Abeta) peptide patterns, using the quantitative Abeta-SDS-PAGE/immunoblot. Additionally, the novel electrochemiluminescence technique (MSD) was used to validate the measures on Abeta1-38. The main outcome measures were a striking decrease of Abeta1-42 in AD (P=7.4 x 10(-19)), and most interestingly a pronounced decrease of Abeta1-38 in FTD (P=9.6 x 10(-7)). Moreover, a novel peptide that most probably represents an oxidized alpha-helical form of Abeta1-40 (Abeta1-40(ox)) displayed a highly significant increase in DLB (P=3.7 x 10(-3)) as compared to non-demented disease controls. The overall diagnostic accuracy of percentage Abeta peptide abundances (Abeta1-X%) was clearly superior to absolute CSF Abeta levels. Abeta1-42% and Abeta1-38% enabled contrasts of 85% or beyond to distinguish AD and FTD, respectively, from all other investigated subjects. Abeta1-40(ox)% yielded a diagnostic sensitivity and specificity of 88 and 73% for the detection of DLB among all other investigated patients. We found a strong correlation between Abeta1-38 levels as measured by the Abeta-SDS-PAGE/immunoblot and MSD, respectively. CSF Abeta peptides may reflect disease-specific impact of distinct neurodegenerative processes on Abeta peptide metabolism and represent a potential diagnostic biomarker for AD, FTD and DLB.

Proteomic analysis of the cerebrospinal fluid of patients with Creutzfeldt-Jakob disease

So far, only the detection of 14-3-3 proteins in cerebrospinal fluid (CSF) has been accepted as diagnostic criterion for Creutzfeldt-Jakob disease (CJD). However, this assay cannot be used for screening because of the high rate of false-positive results, whereas patients with variant CJD are often negative for 14-3-3 proteins. The aim of this study was to compare the spot patterns of CSF by 2-dimensional polyacrylamide gel electrophoresis (2D-PAGE) to search for a CJD-specific spot pattern. We analyzed the CSF of 28 patients [11 CJD, 9 Alzheimer's disease (AD), 8 nondemented controls (NDC)] employing 2D-PAGE which was optimized for minimal volumes of CSF (0.1 ml; 7-cm strips). All samples were run at least three times, gels were silver stained and analyzed by an analysis software and manually revised. We could consistently match 268 spots which were then compared between all groups. By the use of 5 spots, we were able to differentiate CJD from AD or NDC with a sensitivity of 100%. CJD could also be distinguished from both groups by using a heuristic clustering algorithm of 2 spots. We conclude that this proteomic approach can differentiate CJD from other diseases and may serve as a model for other neurodegenerative diseases.

Tauopathies and synucleinopathies: Do cerebrospinal fluid β-amyloid peptides reflect disease-specific pathogenesis?

To evaluate variations in amyloid beta (Abeta) peptide pattern in cerebrospinal fluid (CSF) in neurodegenerative disorders. A recently established quantitative urea-based Abeta-sodium-dodecylsulfate-polyacrylamide-gel-electrophoresis with western immunoblot (Abeta-SDS-PAGE/immunoblot) revealed a highly conserved Abeta peptide (Abeta1-37, 1-38, 1-39, 1-40, 1-42) pattern in CSF. We asked whether the variation might be useful to further elucidate the overlap between or distinctions among neurodegenerative diseases in Abeta-processing. We used the Abeta-SDS-PAGE/immunoblot to investigate CSF for disease-specific Abeta peptide patterns. CSF samples from 96 patients with mainly clinically diagnosed Alzheimer's disease (n = 15), progressive supranuclear palsy (n = 20), corticobasal degeneration (n = 12), Parkinson's disease (n = 11), multiple systems atrophy (n = 18), and dementia with Lewy-bodies (n = 20) were analysed as well a comparison group (n = 19). The Abeta peptide patterns varied between tauopathies and synucleinopathies and between all diseases and the comparison group, possibly due to the influence of tau and alpha-synuclein on Abeta-processing.

Reduced CSF carboxyterminally truncated Abeta peptides in frontotemporal lobe degenerations

Cerebrospinal fluid (CSF) carboxyterminally truncated amyloid-beta (Abeta) peptides, Abeta1-42 and tau protein were evaluated in 30 patients with frontotemporal lobe degenerations (FTLD), 30 Alzheimer's disease (AD) patients and 30 non-demented disease controls (NDC) by Abeta-SDS-PAGE/immunoblot as well as commercial ELISAs for Abeta1-42 and total tau. FTLD displayed a significant drop of Abeta1-37 (p = 2.7 x 10(-4)), Abeta1-38 (p = 4.2 x 10(-5)) and Abeta1-42 (p = 3.3 x 10(-4)). Abeta1-42 was selectively decreased in AD (p = 8.5 x 10(-10)). Decreased Abeta1-38 enabled contrasts of beyond 85% to distinguish FTLD from AD and NDC patients, alone or in combination. Accordingly, low CSF Abeta1-37 and Abeta1-38 represent a biomarker candidate for FTLD and may reflect disease-specific changes of APP metabolism. Further validation should be carried out on dementias other than AD, diagnostically relevant control groups without dementia and without any evident affection of the central nervous system and subgroups of FTLD. Moreover, independent methods of measurement should be applied to CSF Abeta1-38.

Amyloid beta peptide ratio 42/40 but not A beta 42 correlates with phospho-Tau in patients with low- and high-CSF A beta 40 load

Neurochemical dementia diagnostics (NDD) can significantly improve the clinically based categorization of patients with early dementia disorders, and the cerebrospinal fluid (CSF) concentrations of amyloid beta peptides ending at the amino acid position of 42 (A beta x-42 and A beta 1-42) are widely accepted biomarkers of Alzheimer's disease (AD). However, in subjects with constitutively high- or low-CSF concentrations of total A beta peptides (tA beta), the NDD interpretation might lead to erroneous conclusions as these biomarkers seem to correlate better with the total A beta load than with the pathological status of a given patient in such cases. In this multicenter study, we found significantly increased CSF concentrations of phosphorylated Tau (pTau181) and total Tau in the group of subjects with high CSF A beta x-40 concentrations and decreased A beta x-42/x-40 concentration ratio compared with the group of subjects with low CSF A beta x-40 and normal A beta ratio (p<0.001 in both cases). Furthermore, we observed significantly decreased A beta ratio (p<0.01) in the group of subjects with APOE epsilon 4 allele compared with the group of subjects without this allele. Surprisingly, patients with low-A beta x-40 and the decreased A beta ratio characterized with decreased pTau181 (p<0.05), and unaltered total Tau compared with the subjects with high A beta x-40 and the A beta ratio in the normal range. We conclude that the amyloid beta concentration ratio should replace the 'raw' concentrations of corresponding A beta peptides to improve reliability of the neurochemical dementia diagnosis.

Multiplexed quantification of dementia biomarkers in the CSF of patients with early dementias and MCI: a multicenter study

In this report we evaluated the clinical performance of APOE genotyping and three protein biomarkers (total tau, beta-amyloid(1-42), and tau phosphorylated at threonine 181) in a prospective multicenter study using the INNO-BIA AlzBio3 assay applied on Luminex platform. Concentration of biomarkers of Alzheimer's disease in cerebrospinal fluid (CSF) was measured with multiplexing technology (n=223), and compared to the results of ELISA assays in patients with early dementias or mild cognitive impairment (MCI) collected at 12 gerontopsychiatric university departments, and APOE genotyping was performed. Concentrations of Abeta(1-42) were statistically significantly lower in MCI-AD subjects compared to MCI-O, and significantly lower in D-AD patients compared to MCI-O. P-tau(181P) concentrations were significantly higher in MCI-AD patients compared to MCI-O, and significantly higher in D-AD patients compared to MCI-O. The total tau concentrations in MCI-AD patients were significantly higher compared to MCI-O, and higher in D-AD compared to MCI-O, moreover, the concentration of total tau was significantly higher in D-AD compared to MCI-AD patients. For the differential diagnosis between D-AD and D-O, the optimal cutoff concentration of Abeta(1-42) was 197.7 pg/mL, and that for P-tau(181P) was 47.9 pg/mL. These cutoff values were also applied to discriminate between MCI-AD and MCI-O subjects. Simultaneous measurement of the biomarkers significantly improves management of the samples and quality control of the assays' performance.

Beta-amlyoid 1-42 and tau-protein in cerebrospinal fluid of patients with Parkinson's disease dementia

Measurement of tau-protein and beta-amyloid(1-42 )(Abeta42) in cerebrospinal fluid (CSF) has gained increasing acceptance in the differential diagnosis of Alzheimer's disease. We investigated CSF tau-protein and Abeta42 concentrations in 73 patients with advanced idiopathic Parkinson's disease with dementia (PDD) and 23 patients with idiopathic Parkinson's disease without dementia (PD) and in a comparison group of 41 non-demented neurological patients (CG) using commercially available enzyme-linked-immunoabsorbant-assay (ELISA). tau-Protein levels were statistically significantly higher and Abeta42 lower in the PDD patients compared to PD patients and the CG. This observation was most marked (p < 0.05) in a subgroup of patients with PDD carrying the apolipoprotein genotype epsilon3/epsilon3. The distribution of the apolipoprotein genotypes in PDD and PD patients was similar to that of the CG. Although a significant difference in tau-protein values was observed between PDD and CG, no diagnostic cut-off value was established. These findings suggest that such protein CSF changes may help to support the clinical diagnosis of cognitive decline in PD and that there may be apolipoprotein-E-isoform-specific differences in CSF protein regulation in advanced PDD.

International quality control survey of neurochemical dementia diagnostics

Currently, neurochemical dementia diagnostics (NDD) are increasingly entering routine clinical neurochemistry, offering improved early and differential diagnosis of dementias. However, there is an obvious lack of standardization in pre-analytical sample handling and systematic quality surveys. Therefore, in this study, 14 laboratories in Germany, Austria, and Switzerland were given aliquots of a human cerebrospinal fluid (CSF) sample, and were asked to measure Alzheimer's disease (AD) biomarkers (amyloid beta (Abeta) peptides, total Tau protein, and phosphorylated Tau protein (P-tau(181P))) according to their routine protocols.

RESULTS

The inter-laboratory coefficients of variation of the results obtained by the laboratories participating in this study were in the range of 20-30%. Although the results of this quality control survey are promising, the quality of measurements has to be further optimized.

CSF diagnosis of Alzheimer's disease and dementia with Lewy bodies

Differential diagnosis of Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) is often crucial. CSF Tau protein and Amyloid-beta (A beta) peptides have shown diagnostic value for the diagnosis of AD, but discrimination from DLB was poor.Herein, we investigate CSF of 18 patients with probable AD, 25 with probable DLB and 14 non-demented disease controls (NDC) by A beta-SDS-PAGE/immunoblot and commercially available ELISAs for A beta1-42 and tau. CSF A beta peptide patterns and tau exhibited disease specific alterations among AD and DLB. The ratio of A beta1-42 to A beta1-38 and A beta1-42 to A beta1-37, respectively, in combination with absolute tau, yielded a sensitivity and specificity of 100 and 92%, respectively. We conclude that CSF A beta peptide patterns and tau levels reflect disease-specific pathophysiological pathways of these dementias as distinct neurochemical phenotypes. Combined evaluation of these biomarkers provides a reasonable accuracy for differential diagnosis of AD and DLB.

CSF amyloid-beta-peptides in Alzheimer's disease, dementia with Lewy bodies and Parkinson's disease dementia

As the differential diagnosis of dementias based on established clinical criteria is often difficult, biomarkers for applicable diagnostic testing are currently under intensive investigation. Amyloid plaques deposited in the brain of patients suffering from Alzheimer's disease, dementia with Lewy bodies (DLB) and Parkinson's disease dementia (PDD) mainly consist of carboxy-terminally elongated forms of amyloid-beta (Aβ) peptides, such as Aβ1–42. Absolute Aβ1–42 levels in CSF have shown diagnostic value for the diagnosis of Alzheimer's disease, but the discrimination among Alzheimer's disease, DLB and PDD was poor. A recently established quantitative urea-based Aβ-sodium-dodecylsulphate–polyacrylamide-gel-electrophoresis with Western immunoblot (Aβ-SDS–PAGE/immunoblot) revealed a highly conserved Aβ peptide pattern of the carboxy-terminally truncated Aβ peptides 1–37, 1–38, 1–39 in addition to 1–40 and 1–42 in human CSF. We used the Aβ-SDS–PAGE/immunoblot to investigate the CSF of 23 patients with Alzheimer's disease, 21 with DLB, 21 with PDD and 23 non-demented disease controls (NDC) for disease-specific alterations of the Aβ peptide patterns in its absolute and relative quantities. The diagnostic groups were matched for age and severity of dementia. The present study is the first attempt to evaluate the meaning of Aβ peptide patterns in CSF for differential diagnosis of the three neurodegenerative diseases—Alzheimer's disease, DLB and PDD. The Aβ peptide patterns displayed disease-specific variations and the ratio of the differentially altered Aβ1–42 to the Aβ1–37 levels subsequently discriminated all diagnostic groups from each other at a highly significant level, except DLB from PDD. Additionally, a novel peptide with Aβ-like immunoreactivity was observed constantly in the CSF of all 88 investigated patients. The pronounced percentage increase of this peptide in DLB allowed a highly significant discrimination from PDD. Using a cut-off point of 0.954%, this marker yielded a diagnostic sensitivity and specificity of 81 and 71%, respectively. From several lines of indication, we consider this peptide to represent an oxidized α-helical form of Aβ1–40 (Aβ1–40^*). The increased abundance of Aβ1–40^* probably reflects a disease-specific alteration of the Aβ1–40 metabolism in DLB. We conclude that Aβ peptide patterns reflect disease-specific pathophysiological pathways of different dementia syndromes as distinct neurochemical phenotypes. Although Aβ peptide patterns failed to fulfil the requirements for a sole biomarker, their combined evaluation with other biomarkers is promising in neurochemical dementia diagnosis. It is noteworthy that DLB and PDD exhibit distinct clinical temporal courses, despite their similar neuropathological appearance. Their distinct molecular phenotypes support the view of different pathophysiological pathways for each of these neurodegenerative diseases.

Total tau protein, phosphorylated tau (181p) protein, β-amyloid1–42, and β-amyloid1–40 in cerebrospinal fluid of patients with dementia with Lewy bodies

The intra vitam diagnosis of different dementias is still based on clinical grounds. So far, no technical investigations have been available to support these diagnoses. For tau protein and β-amyloid

Follow-up investigations of tau protein and S-100B levels in cerebrospinal fluid of patients with Creutzfeldt-Jakob disease

BACKGROUND

S-100B and tau protein have a high differential diagnostic potential for the diagnosis of Creutzfeldt-Jakob disease (CJD). So far there has been only limited information available about the dynamics of these parameters in the cerebrospinal fluid (CSF). However, there is a special interest in finding biochemical markers to monitor disease progression for differential diagnosis and treatment.

PATIENTS AND METHODS

We analyzed CSF of 45 patients with CJD and of 45 patients with other neurological diseases for tau protein and S-100B in a follow-up setting. All diagnoses of CJD were later neuropathologically verified. A ratio between tau protein differences and the time between lumbar puncture was calculated. The same was done for S-100B.

RESULTS

Tau protein levels of 34 cases were above the cut-off level for CJD (>1,300 pg/ml) in the first CSF sample. In 7 of 11 patients with lower tau levels in the first CSF sample, tau levels rose. The above-mentioned ratio was significantly higher in the CJD group than in the group with other neurological diseases. Similar results were obtained for S-100B.

CONCLUSION

We conclude that follow-up investigations and calculation of ratios is a useful tool in the differential diagnosis of CJD. Variations in this pattern were observed in single cases.

Tau protein, Abeta42 and S-100B protein in cerebrospinal fluid of patients with dementia with Lewy bodies

The intra vitam diagnosis of dementia with Lewy bodies (DLB) is still based on clinical grounds. So far no technical investigations have been available to support this diagnosis. As for tau protein and beta-amyloid(1-42) (Abeta42), promising results for the diagnosis of Alzheimer's disease (AD) have been reported; we evaluated these markers and S-100B protein in cerebrospinal fluid (CSF), using a set of commercially available assays, of 71 patients with DLB, 67 patients with AD and 41 nondemented controls (NDC) for their differential diagnostic relevance. Patients with DLB showed significantly lower tau protein values compared to AD but with a high overlap of values. More prominent differences were observed in the comparison of DLB patients with all three clinical core features and AD patients. Abeta42 levels were decreased in the DLB and AD groups versus NDC, without significant subgroup differences. S-100B levels were not significantly different between the groups. Tau protein levels in CSF may contribute to the clinical distinction between DLB and AD, but the value of the markers is still limited especially due to mixed pathology. We conclude that more specific markers have to be established for the differentiation of these diseases.

Electrophoretic separation of amyloid beta peptides in plasma

In this prospective study, for the first time we have separated and quantified amyloid beta (Abeta) peptides in the plasma of patients with Alzheimer's disease (AD, n = 8) and age- and environment-matched healthy controls (n = 9) with urea-based Abeta-sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)/immunoblot. In addition to the Abeta peptides 1-37/38/39/40/42, which we recently identified as regular constituents of human cerebrospinal fluid (CSF), we have observed a novel electrophoretic band migrating slightly cathodically to Abeta1-42. Since a standard peptide with the amino acid sequence Abeta2-40 migrates in the same position, we hypothesize that this plasma-specific band may correspond to Abeta2-40. The concentration of Abeta peptides in the plasma has been approximately 100-fold lower compared to the CSF. Interestingly, the concentration of the two shortest peptides and the longest one of these considered here (i.e., Abeta1-37/38/42) have increased significantly when the samples have been frozen at -80 degrees C before immunoprecipitation, while the 'middle-length' peptides (i.e., Abeta1-39/40) have not been affected by this procedure. We have not observed significant differences of the Abeta peptides concentrations between AD and control subjects. Our method can be used to investigate the significance of plasma Abeta peptides in neurodegenerative disorders, and to monitor the efficiency of drugs with beta/gamma-secretase inhibitory potency.

Cerebrospinal fluid amyloid beta peptide patterns in Alzheimer's disease patients and nondemented controls depend on sample pretreatment: indication of carrier-mediated epitope masking of amyloid beta peptides

A quantitative urea-based amyloid beta (Abeta)-sodium dodecyl sulfate-polyacrylamide gel electrophoresis with Western immunoblot (Abeta-SDS-PAGE/immunoblot) reveals highly conserved and disease-specific Abeta peptide patterns (Abeta 1-37, 1-38, 1-39, 1-40, 1-42) in Alzheimer's disease (AD) patients and nondemented controls. For further standardization of this method, we analyzed cerebrospinal fluid (CSF) of eight probable AD patients and seven nondemented controls using different preanalytical procedures for Abeta-SDS-PAGE/immunoblot and Abeta1-42-enzyme linked immunosorbent assay (ELISA). Both diagnostic groups were discriminated significantly by absolute levels of Abeta1-42 and ratios of Abeta1-42/40, 1-42/38, 1-42/39. Preanalytical freezing of CSF led to a highly significant loss of all Abeta peptide species. This effect was most pronounced for Abeta1-42 and completely prevented by SDS-heat denaturation prior to freezing. Prolonged storage of SDS-heat denatured CSF led to a selective loss of Abeta1-42 and impaired the discrimination of diagnostic groups as measured by Abeta-SDS-PAGE/immunoblot. Neither freezing nor storage significantly affected absolute Abeta1-42 levels as determined by Abeta1-42-ELISA, but both impaired the discrimination of diagnostic groups. Hence, we suggest immediate analysis of samples for Abeta1-42-ELISA, analysis after a short freezing interval for Abeta-SDS-PAGE/immunoblot, and avoidance of prolonged storage intervals. Remarkably, Abeta-SDS-PAGE/immunoblot measured threefold higher levels of Abeta1-42 in CSF than Abeta1-42-ELISA. In summary, our results indicate carrier-mediated epitope masking of Abeta1-42.

Follow-up investigations in cerebrospinal fluid of patients with dementia with Lewy bodies and Alzheimer's disease

Measuring proteins in cerebrospinal fluid (CSF) has gained wide acceptance for the differential diagnosis of dementia. Some groups have already extended these investigations in Alzheimer's disease (AD) by asking how stable these markers are in follow-up analysis, if they depend on the stage of disease and whether they can be used to monitor the progression and biological effects of treatment. We evaluated 21 patients with dementia with Lewy bodies (DLB) and 19 patients with AD, on two occasions, with regard to levels of tau protein, tau protein phosphorylated at threonine 181 (p-tau), Abeta42, Abeta40 and S-100B protein, using a set of commercially available assays. Tau protein levels were lower in DLB in first and second LP compared to AD and decreased during course of both groups. P-tau levels were increased in AD and DLB and decreased during follow-up. Abeta42 and Abeta40 remained relatively stable during follow-up but we found a slight increase of the median Abeta42 level in DLB, whereas in AD, Abeta42 tends to decrease during follow-up. S-100B protein increased during follow-up in both diseases. The protein dynamics in DLB and AD are relatively similar. S-100B protein may be a useful marker for follow-up in neurodegenerative diseases but has to be analysed in longer follow-up periods. Tau protein may be used to differentiate between DLB and AD. Follow-up CSF analyses are of limited value for the differentiation of AD and DLB. We conclude that more specific markers have to be established for the differentiation and follow-up of these diseases.

Heart fatty acid binding protein as a potential diagnostic marker for neurodegenerative diseases

The diagnosis of neurodegenerative diseases with dementias requires several different test approaches and often remains uncertain. Using a proteomic approach it was shown in nine patients that heart fatty acid binding protein (H-FABP) might be a biomarker for Creutzfeldt-Jakob disease (CJD). The aim of our study was to evaluate whether H-FABP is a biomarker for the differential diagnosis of dementias. Therefore we measured H-FABP in cerebrospinal fluid (CSF) and serum of patients having CJD, dementia with Lewy-bodies (DLB), Alzheimer's disease (AD) and in non-demented control (NDC) patients. H-FABP levels in CSF and serum of CJD patients are increased compared to non-demented controls. Levels of H-FABP were significantly higher in CJD patients compared to AD and DLB in CSF. However, discrimination between CJD and AD was not possible in serum. Interestingly, highest levels of H-FABP were found in serum of DLB patients. Our results suggest that H-FABP might be a useful biomarker for the differentiation between the dementias examined if levels in CSF and serum are determined in parallel.

Amyloid beta peptides in cerebrospinal fluid as profiled with surface enhanced laser desorption/ionization time-of-flight mass spectrometry: evidence of novel biomarkers in Alzheimer's disease

BACKGROUND

The advent of new therapeutic avenues for Alzheimer's disease (AD) calls for an improved early and differential diagnosis.

METHODS

With surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS), cerebrospinal fluid from patients with AD (n = 10) and nondemented control subjects (n = 9) was studied.

RESULTS

Molecular mass signals were observed corresponding to three novel amyloid beta (Abeta) peptides that have not previously been described, in addition to those previously known, with molecular masses of 4525.1 d, 4846.8 d, and 7755.8 d. The signal-to-noise ratios (S/NR) of Abeta(4525.1) and Abeta(7758.8+2H) were significantly decreased in AD [Abeta(4525.1): median 2.2 and 4.3 in AD and control subjects, respectively, p <.01; Abeta(7758.8+2H): median 1.0 and 14.0 in AD and control subjects, respectively, p <.01], whereas the S/NR of Abeta(4846.8) was significantly increased in AD (median 3.6 and 2.5 in AD and control subjects, respectively, p <.05). The S/NR of two known AD biomarkers, Abeta1-42 and Abeta1-40, expectedly turned out to be significantly decreased (p <.01) and unaltered in AD, respectively. A moderate and highly significant correlation was observed between S/NR of Abeta1-42 and Abeta42 concentration as measured with enzyme-linked immunosorbent assay (R =.67, p <.01).

CONCLUSIONS

We report evidence of three novel amyloid beta peptides that might play an important role in the diagnosis and pathophysiology of Alzheimer's disease.