Combination of specific monoclonal antibodies allow identification of soluble aggregates of by sandwich ELISA ()
1. INTRODUCTION
Amyloid beta protein is 1 of the proteins that cause Alzheimer’s disease. Aβ is produced in the human brain and is a major constituent of senile plaques [1,2]. There are mainly 2 types of Aβ, viz. Aβ1-40 and Aβ1-42. Aβ1-42 shows greater toxicity against neuronal cells than do Aβ1-40 [3]. Aβ1-42 represents the monomeric form of the protein; it can, however, also form oligomers from these monomers [4].
In solution, Aβ1-42 is aggregated by self-assembly, and various sizes of Aβ1-42 aggregate have been reported. For example, Aβ-derived diffusible ligands (ADDLs) occur as 17 kDa and 42 kDa aggregates [5]. Low molecular weight (LMW) aggregates are oligomers of Aβ1-42 of approximately 14 kDa [6]. The globulomer in which methionine at the N-terminus of Aβ1-42 is displaced, is approximately 64 kDa, while non-displaced globulomers are 38 - 48 kDa [7,8]. Medium-sized aggregates of Aβ1-42, ASPDs, are 158 - 669 kDa, and aggregated Aβ1-42 of 80 - 500 kDa have also been reported [9,10]. In the class of large Aβ1-42 aggregates, there are large oval aggregates (LOA) that contain an added Aβ16-20 (KLVFF) peptide, as well as aggregates of Aβ1-42 that contain an added Fc-KLVFF [11,12]. The Aβ16-20 (KLVFF) peptide binds to the Aβ1-42 β-sheet and inhibits Aβ1-42 fibril formation. Recently, soluble Aβ has been reported to occur during the process of Aβ aggregation, which is more toxic than the fibril and monomer forms [13,14].
Immunoassays are an effective way to capture the causative agent of a disease, and can be used to recognize the structure of Aβ1-42 by capturing the aggregates of different sizes. It is important that the size of the aggregated Aβ1-42 is identified in this way [15]. Antibodies against various Aβ1-42 aggregates have been produced to date. Among the small Aβ1-42 aggregates, the antibody 6E10 has been used to capture ADDLs and N-Met-Aβ1-42 globulomers, which form comparatively small oligomers. In the context of medium-sized Aβ1-42 aggregates, some antibodies have been reported, but it is difficult to produce a specific antibody against aggregates in this sizerange.
Here, we prepared soluble aggregates and amorphous aggregates of Aβ1-42. Soluble aggregates smaller than LOA were recognized by a combination of antibodies against soluble aggregates, those against soluble aggregates and amorphous aggregates, and those against LOA, by sandwich ELISA. This approach allows structural analysis of Aβ1-42 aggregates and facilitates diagnosis of AD.
2. MATERIALS AND METHODS
2.1. Preparation of Aggregated Aβ1-42
Aβ1-42 was purchased from AnyGen Co. Ltd., Korea. Amorphous aggregates were prepared as follows. A suspension of 0.44 mM Aβ1-42 was prepared in Milli Q water and incubated for 30 min at 4˚C. The solution was prepared in Dulbecco’s PBS without Ca2+ or Mg2+ (Wako, Osaka, Japan) 0.22 mM. To this solution, 2.2 mM of Aβ16-20 (AnyGen Co. Ltd.) was added and the mixture stirred at 7 rpm for 16 h, at 37˚C [11].
For soluble aggregates, Aβ1-42 (100 μg) was dissolved in 200 μl of 1,1,1,3,3,3-hexafluoro-2-propanol. This solution was incubated for 16 h at 4˚C and then incubated for 3 h at 37˚C [16]. Thereafter, the solution was lyophilized; these steps were repeated twice. After the final lyophilization, a solution of Aβ1-42 was prepared at a concentration of 1 mg/ml in Milli Q water.
2.2. Reactivity of Monoclonal Antibodies during the Preparation of Aggregated and Fibrillar Aβ1-42
The 79-3 monoclonal antibody against soluble aggregates and the 78-6 monoclonal antibody against soluble aggregate and amorphous aggregate were used. Reactivity of these monoclonal antibodies were assessed during the preparation of aggregated and fibrillar Aβ1-42. The measurements were carried out every 4 h, up to 24 h.
2.3. ELISA
As primary antibodies, 8 monoclonal antibodies against soluble aggregates and amorphous aggregates were used. These were used to coat 96-well plates (F96 MAXISORP NUNC-IMMUNO PLATE; Thermo Fisher SCIENTIFIC Inc. Rochester NY) (100 μl/well, at a concentration of 2 μg/ml). The plates were incubated for 60 min at 37˚C and washed with PBS containing 0.05% Tween. Blocking buffer (Immunoblock; DS Pharma Biomedical) were used to coat the 96 well plates (200 μl/well) and incubated for 60 min at 37˚C. The plates were washed with PBS containing 0.05% Tween 5 times. Soluble aggregates and aggregates were then used to coated these prepared 96-well plates (100 μl/well at a concentration of 1 μg/ml). The plates were then incubated for 60 min at 37˚C, and washed with PBS containing 0.05% Tween 5 times.
As secondary antibodies, the 37-11 and 79-3 antibodies labeled with horse-radish peroxidase (HRP) were used to coat the 96-well plates (100 μl/well at a concentration of 0.144 μg/ml). The plates were incubated for 60 min at 37˚C and were washed with PBS containing 0.05% Tween 5 times. Citrate buffer solution (25 ml), containing 10 mg O-phenylalanine dichloride (SigmaAldrich, St. Louis, MO) and 30% hydrogen peroxide, was then used to coat the wells (100 µl/well) and the plates were incubated for 20 min at room temperature. Absorbance was then measured using a microplate reader at 492 nm.
3. RESULTS AND DISCUSSION
3.1. Reactivity of the 79-3 and 78-6 Monoclonal Antibodies during Preparation of Aggregated Aβ1-42
The size of the amorphous aggregates of Aβ1-42 in the presence of Aβ16-20-containing LOA is 50 - 400 nm. The size of the soluble aggregates is nearly all approximately 50 nm, although some are 300 nm. The reactivity of the 79-3 and 78-6 monoclonal antibodies decreased during preparation of the amorphous aggregate of Aβ1-42 in the presence of Aβ16-20-containing LOA (Figure 1). After 16 h, the reactivity of the 79-3 monoclonal antibody was lower than that of the 78-6 antibody. This result indicated that the 79-3 monoclonal antibody is more specific for the soluble aggregate than the 78-6 monoclonal antibody.
The 78-6 antibody reacts not only with soluble aggregates, but also with amorphous aggregates.
3.2. Measurement of Soluble Aggregates by Sandwich ELISA
To confirm the specificity of the 79-3 monoclonal antibody for soluble aggregates of Aβ1-42, we performed an evaluation of the specificity of the 79-3 monoclonal an-
Figure 1. Reactivity of the antibodies 79-3 (○), against soluble aggregates, and 78-6 (●), against soluble aggregates and amorphous aggregates, in the presence of 2.2 mM Aβ16-20.
tibody labeled with HRP against soluble aggregates of Aβ1-42 with that of the 78-6 monoclonal antibody, using sandwich ELISA. As antigen, soluble aggregates and amorphous aggregates were used. As primary antibodies, 72-10, 73-1, 76-3, 80-4, 75-2, 77-3, and 83-3, which react against both soluble aggregates and amorphous aggregates, as does the 78-6 antibody, were used. The reactivity of the 79-3 monoclonal antibody against soluble aggregates was high (Figure 2); on the other hand, the reactivity of the 79-3 monoclonal antibody against amorphous aggregates was low. By combining the different monoclonal antibodies, it was possible to detect a combination of soluble Aβ1-42 aggregates. These results indicated that these 2 types of antibodies have different epitope sites.
The structure and size of soluble aggregates are different from those of amorphous aggregates. 4G8 and 82E1 are specific monoclonal antibodies against the N-terminus of Aβ [17]. 11A1 is a specific monoclonal antibody directed against positions 22 and 23 of Aβ1-42 oligomers [18]. Thus, the 8 monoclonal antibodies used as primary antibodies reacted with Aβ1-42 of various dimensions. The 79-3 monoclonal antibody recognized only soluble aggregates. Therefore, the 79-3 monoclonal antibody recognizes an epitope on the surface of Aβ1-42.
Furthermore, we have attempted to construct a similar sandwich ELISA using the monoclonal antibodies that reacts against LOA other than 79-3, such as 37-11.
HRP was used to label the 37-11 monoclonal antibody, as for 79-3, and the reactivity of the 37-11 antibody was evaluated against those of the 9 monoclonal antibodies that react against soluble aggregates and amorphous aggregates, using sandwich ELISA. These results are shown in Figure 3. The results indicated that a combination of the 37-11 and 77-3 antibodies yielded the highest reactivity, whereas those of the other were low. The 37-11 monoclonal antibody is specific for LOA. How-
Figure 2. Reactivity of the 79-3 monoclonal antibody against soluble aggregates (□) and amorphous aggregates (■) by sandwich ELISA.
Figure 3. Reactivity of the 37-11 monoclonal antibody against soluble aggregates by sandwich ELISA.
ever it is possible that the 37-11 antibody could recognize soluble aggregates different from LOA by changing the measurement system of the sandwich ELISA. These results suggested that the epitopes of the 37-11 and 77-3 monoclonal antibodies differ.
In this study, we were able to recognize soluble aggregates by sandwich ELISA. Further refinement of the size of aggregated Aβ1-42 would be possible and may facilitate the analysis of the structure of Aβ1-42 and the diagnosis of AD.
4. ACKNOWLEDGEMENTS
We thank the president of Immuno Probe Co. Ltd., Mr. Hiroshi Nomura, for technical assistance in producing the monoclonal antibodies.
NOTES