Altered levels of acetylcholinesterase in Alzheimer plasma

Abstract

Background: Many studies have been conducted in an extensive effort to identify alterations in blood cholinesterase levels as a consequence of disease, including the analysis of acetylcholinesterase (AChE) in plasma. Conventional assays using selective cholinesterase inhibitors have not been particularly successful as excess amounts of butyrylcholinesterase (BuChE) pose a major problem.

Principal findings: Here we have estimated the levels of AChE activity in human plasma by first immunoprecipitating BuChE and measuring AChE activity in the immunodepleted plasma. Human plasma AChE activity levels were approximately 20 nmol/min/mL, about 160 times lower than BuChE. The majority of AChE species are the light G(1)+G(2) forms and not G(4) tetramers. The levels and pattern of the molecular forms are similar to that observed in individuals with silent BuChE. We have also compared plasma AChE with the enzyme pattern obtained from human liver, red blood cells, cerebrospinal fluid (CSF) and brain, by sedimentation analysis, Western blotting and lectin-binding analysis. Finally, a selective increase of AChE activity was detected in plasma from Alzheimer's disease (AD) patients compared to age and gender-matched controls. This increase correlates with an increase in the G(1)+G(2) forms, the subset of AChE species which are increased in Alzheimer's brain. Western blot analysis demonstrated that a 78 kDa immunoreactive AChE protein band was also increased in Alzheimer's plasma, attributed in part to AChE-T subunits common in brain and CSF.

Conclusion: Plasma AChE might have potential as an indicator of disease progress and prognosis in AD and warrants further investigation.

Figure 1. Plasma AChE levels in healthy controls (wild-type) after BuChE immunodepletion and in BuChE-silent individuals.

(A) Control plasma was immunoprecipitated with anti-BuChE antibody and cholinesterase activity levels determined before^b and after^a immunoprecipitation (n = 6; 46±4 yrs). AChE activity level in plasma from BuChE-silent individuals is also shown (n = 3; 30±5 yrs). The anti-BuChE antibody does not immunoprecipitate AChE in BuChE-silent plasma (not shown). Values are means ± SEM. (B) Immunoprecipitation of control plasma with antibody, followed by immunoblotting with the anti-AChE antibody, N-19. The presence (+) or absence (−) of the anti-BuChE antibody linked to the resin is indicated in the top margin. Prior to electrophoretic analysis, proteins abundant in plasma were depleted by immunoaffinity-based protein subtraction chromatography with IgY microbeads (Seppro™). The anti-BuChE antibody does not immunoprecipitate AChE. Extracts incubated with protein A-Sepharose, in the absence of the antibody, were analyzed in parallel as negative controls. (C) Representative profiles of AChE and (D) BuChE molecular forms (G₄ = tetramers; G₁+G₂ = monomers and dimers) in control plasma samples before (•) and after (○) BuChE-immunoprecipitation, and in BuChE-silent plasma (▴). (E) Representative immunoblot of individual AChE G₄ and G₁+G₂ peak-fractions separated by sucrose gradient centrifugation from control plasma and detected with the N-19 antibody (a similar volume for both the G₄ and G₁+G₂ peaks was loaded in each lane). (F) Comparison of the AChE-banding pattern detected with the N-19 antibody, for fractions bound and unbound to the anti-AChE antibody HR2 and to the Fas2-Sepharose affinity matrix.

Figure 2. AChE molecular form and lectin-binding profile in human plasma, liver, RBCs, CSF and brain (frontal cortex).

(A) Representative profiles of AChE molecular forms (G₄ = tetramers; G₁+G₂ = monomers and dimers). (B) Comparison of Con A and LCA binding of AChE. Plasma, CSF and total extracts from liver, RBCs and brain (n = 6 for each group) were incubated with immobilized lectins, AChE activity was assayed in the supernatants and the percentage of (%) AChE activity unbound to lectins was calculated. For total CSF and brain extracts, both rich in tetramers, the % AChE unbound to Con A (%Unb Con A_CSF = 3±1; %Unb Con A_Brain = 4±1) and to LCA (%Unb LCA_CSF = 4±1; %Unb LCA_Brain = 17±2) were determined. Additionally, individual G₄ and G₁+G₂ fractions, separated by sucrose gradient centrifugation, from CSF and brain extracts (n = 5), were also pooled, dialyzed against Tris-saline-Triton X-100 buffer, and concentrated by ultrafiltration. AChE peaks were then assayed by incubation with immobilized lectins, and the percentages of unbound enzymatic activity were calculated. For CSF tetramers, the %Unb Con A was 0.9±0.5; and the %Unb LCA was 0.3±0.1. For brain tetramers, the %Unb Con A was 1.3±0.2; and the %Unb LCA 2.4±0.2. Please note differences in lectin binding for total AChE from brain or CSF, or its enriched G₄ fractions, when compared to the respective G₁+G₂ peaks (see Figure), revealing distinct glycosylation patterns for different AChE molecules. Values are means± SEM *p<0.05, significantly different from plasma samples, as assessed by one-way analysis of variance with Bonferroni posttest.

Figure 3. Immunodetection of AChE subunit variants in human plasma, liver, RBCs, CSF and brain (frontal cortex).

Samples were immunoblotted with three anti-AChE antibodies, (A) the N-terminal N-19, which recognizes all variants; (B) the C-terminal ab31276, which recognizes only AChE-T subunits; and (C) the anti- AChE-R antibody directed at the unique C-terminus of AChE-R. Representative immunoblot with the N-19 antibody of individual G₁+G₂ peak fractions separated by sucrose gradient centrifugation from CSF and brain are included (A, right panel).

Figure 4. AChE levels and molecular form pattern are altered in plasma samples from non-demented controls (ND) and Alzheimer's patients (AD).

(A) Box plot comparing total plasma-AChE activity from 15 ND and 14 AD cases, AChE activity was calculated after BuChE depletion. AChE molecular forms were separated, identified by sedimentation analysis (representative profiles; left panel), and a G₄/(G₁+G₂) ratio calculated (n = 6 for each group). * p<0.05 significantly different from NDs, as assessed by Student's t-test. (B) Total BuChE and molecular form pattern was also calculated, prior to immunoprecipitation, with no statistically significant differences between groups.

Figure 5. Altered AChE immunoreactivity in plasma of non-demented controls (ND) and Alzheimer's disease (AD) patients.

(A) Representative blot of plasma-AChE from AD and ND using the antibody N-19, and (B) densitometric quantification of the AChE-immunoreactive bands, expressed in arbitrary units (a u.), from 15 ND (▪; n = 15) and 14 AD (□; n = 14) subjects. Proteins abundant in plasma were depleted by immunoaffinity-based protein subtraction chromatography with IgY microbeads (Seppro™) and equivalent amounts of protein were loaded in each lane. Columns represent means ± SEM *p = 0.02 significantly different from NDs as assessed by Student's t test. (C) Representative blot of plasma AChE detected with the anti-AChE antibody to the C-terminal, ab31276.