Reversibility of beta-amyloid self-assembly: effects of pH and added salts assessed by fluorescence photobleaching recovery

Abstract

The 40-residue peptide isoform beta-amyloid (Abeta(1-40)) is associated with Alzheimer's disease. Although found in the tangles and fibrous mats that characterize the brain in advanced stages of the disease, the toxic form of Abeta is believed to be oligomers or "protofibrils". Characterization of these fairly small structures in solution, especially in the presence of the much larger assemblies they also form, is a daunting task. Additionally, little is known about the rate of Abeta assembly or whether it can be triggered easily. Perhaps most importantly, the conditions for reversing assembly are not fully understood. Fluorescence photobleaching with modulation detection of the recovery profile is a sensitive and materials-efficient way to measure diffusers over a wide range of hydrodynamic sizes. The method does require attachment of a fluorescent label. Experiments to validate the use of 5-carboxyfluorescein-labeled Abeta(1-40) as a representative of the unlabeled, naturally occurring material included variation of photobleaching time and mixture of labeled and unlabeled materials. A dialysis cell facilitated rapid in situ changes in pH and salt conditions. Multiple steps and complex protocols can be explored with relative ease. Oligomeric aggregates were found by fluorescence photobleaching recovery to respond readily to pH and salt conditions. Changing these external cues leads to formation or disassembly of aggregates smaller than 100 nm within minutes.

Figure 1

Diffusion as a function of bleaching time for labeled Aβ(1–40) and a mixture of labeled and unlabeled Aβ(1–40). Filled squares: 100% 5-carboxyfluorescein Aβ(1–40), 50 μM peptide. Open circles: mixture of 25% 5-carboxyfluorescein-labeled Aβ(1–40) with 75% Aβ(1–40) expressed as percentage of 50 μM peptide. Conditions: 50 mM PBS, 150 mM NaCl, pH 7.4, 0.5 Watts laser power. Error bars represent standard deviations from triplicate runs.

Figure 2

Semilogarithmic FPR traces for three different Aβ solutions at different pH values, prior to substantial aggregation. Conditions: 100 μM 25% / 75% mixture of 5-carboxyfluorescein Aβ and unlabeled Aβ in 50 mM PBS, 150 mM NaCl at pH values ▷2.7, ○ 6.9, and □ 11. Data have been offset vertically for clarity.

Figure 3

Distributions of apparent diffusion coefficient from CONTIN analysis of FPR data for aggregated Aβ at three pH values. The samples were measured 14 days after preparation. Conditions: 100 μM 25% / 75% mixture of 5-carboxyfluorescein Aβ and unlabeled Aβ in 50 mM PBS, 150 mM NaCl at pH values 2.7, 6.9, and ■ 11. The vertical scales at pH 6.9 and pH 2.7 have been offset for clarity. Apparent hydrodynamic radii are indicated (does not imply spherical shape for the diffusers.)

Figure 4

Effect of temperature on the diffusion of Aβ peptide, corrected for solvent viscosity, η. Conditions: 100 μM Aβ in a 100 mM PBS, 150 mM NaCl, pH 7.4. The arrow indicates the monomeric Aβ_1–40 diffusion value (for 10 mM KOH).

Figure 5

One-pot dialysis of 50 μM 5-carboxyfluorescein labeled Aβ starting at pH 11, then dialyzing against 100 μM acetate buffer, pH 4, and increasing the concentration of CaCl₂.

Figure 6

Percent amplitude of fast and slow components of FPR recoveries corresponding to the aqueous conditions of Figure 5.

Figure 7

Diffusion from in situ FPR of 50 μM 5-carboxyfluorescein-Aβ starting at pH 11, then sequentially dialyzed to pH 2.7 and pH 7.2. Arrow indicates diffusion of monomeric Aβ(1–40) (in 10 mM KOH).

Figure 8

Diffusion from in situ FPR of 100 μM 5-carboxyfluorescein-Aβ (25% mixed with unlabeled 75% Aβ) starting at pH 11, then alternately dialyzed between pH 2.7 and pH 7.4 in 50 mM phosphate buffer.