Bone Morphogenetic Protein 2 (BMP-2) Aggregates Can be Solubilized by Albumin-Investigation of BMP-2 Aggregation by Light Scattering and Electrophoresis

Abstract

Bone morphogenetic protein 2 (BMP-2) has a high tendency to aggregate at physiological pH and physiological ionic strength, which can complicate the development of growth factor delivery systems. The aggregation behavior in differently concentrated BMP-2 solutions was investigated using dynamic and static light scattering. It was found that at higher concentrations larger aggregates are formed, whose size decreases again with increasing dilution. A solubilizing effect and therefore less aggregation was observed upon the addition of albumin. Imaged capillary isoelectric focusing and the simulation of the surface charges of BMP-2 were used to find a possible explanation for the unusually low solubility of BMP-2 at physiological pH. In addition to hydrophobic interactions, attractive electrostatic interactions might be decisive in the aggregation of BMP-2 due to the particular distribution of surface charges. These results help to better understand the solubility behavior of BMP-2 and thus support future pharmaceutical research and the development of new strategies for the augmentation of bone healing.

Keywords: BMP-2; albumin; protein aggregation; protein solubilization; protein-protein interactions.

Figure 1

Illustration of the principle of a 2D Blue Native/SDS-PAGE showing the separation of BMP-2 aggregates and the separation of BSA (serum fraction V).

Figure 2

Representative electropherogram of BMP-2 (two peaks at 8.2 and 8.5) and adjacent isoelectric point (pI) markers (7.05 and 9.5).

Figure 3

Electrostatic surface potential of BMP-2 at pH 5 as determined by continuum electrostatic calculations. (A): BMP-2 displayed from all sides and (B): view on the central cavities of two BMP-2 molecules showing the possible interaction patches frontally. The dotted line indicates the axis at which negatively charged (red) areas are reflected in complementary positively charged (blue) areas and at which hydrophobic (white) areas are reflected in other hydrophobic areas. (C): Possible interaction patches facing each other. Compared to (B), the two BMP-2 dimers were rotated 90° towards each other. io = possible electrostatic attraction, hy = possible hydrophobic interaction.

Figure 4

Representative Coomassie brilliant blue stained SDS-PAGE. I = 125 ng BMP-2, II = I + 30 µg BSA, III = Ladder, IV = 500 ng BMP-2.

Figure 5

(A,B) silver-stain and (C) Western blot of blue native PAGEs. (D) Silver-stain of three second dimension gels of 2D Blue Native/SDS-PAGEs. Darker grey value corresponds to higher degree of light attenuation during transillumination (A,B,D) or to higher relative luminescence (C). Blue boxes indicate where the lanes were cut out of the blue native PAGE and how they were applied on the second dimension SDS-PAGEs. White labels indicate approximate molecular mass (kDa). Red arrows highlight distribution of BMP-2. I = 30 µg BSA, II = Ladder, III = 2 µg BMP-2, IIIs = 0.5 µg BMP-2, IV = I + III, IVs = I + IIIs, V = 16 ng BMP-2, VI = V + 10 µg BSA.

Figure 6

(A) Silver staining and overlay of luminescence intensity after Western blot of identical gel, after anodic native PAGE at pH 8.3, (A_S): explanatory illustration of (A), (B) silver staining after cathodic native PAGE at pH 7.6, (B_S): explanatory illustration of (B). Red arrows illustrate the apparent direction of movement.

Figure 7

Silver staining after anodic native PAGE at pH 8.3 with proteins starting from drilled holes in the middle of the polyacrylamide gel.

Figure 8

HSA and BMP-2 recovery (dRI detector of AF4 system) for different injected masses.

Figure 9

(A) Weight-average molecular weights (M_w) of BMP-2 injection series a, b and c into MES buffer, determined by MALS/dRI. The error bars indicate standard deviation of the determined molecular weights given by Debye fitting. (B) M_w of the injection series in correlation to the dilution factor (concentration determined by dRI/concentration before injection). Dotted line indicates an exponential decrease fit of b and c. The two very first injections are marked in red.

Figure 10

Molar masses of two sets of measurements with a one-day interval in PBS buffer pH 7.4. (A) Diluted from a 1.15 mg/mL HSA solution. (B) Diluted from a mixture of 0.2 mg/mL BMP-2 and 1.15 mg/mL HSA. Each symbol represents a single measurement. Error bars indicate standard deviation of the determined molecular weights given by Debye fitting.

Figure 11

Autocorrelation functions of dynamic light scattering (DLS) measurements of BMP-2 in (A) MES buffer and (B) PBS buffer. (C) DLS size distribution of BMP-2 in MES buffer pH 5 by volume and (E) by intensity and (D) DLS size distribution of BMP-2 in PBS pH 7.2 by volume and (F) by intensity.

Figure 12

Single frames from microscopic light scattering videos (with same magnification) of samples containing 13.2 µg/mL BMP-2 during NTA in (A) MES buffer pH 5 and (B) PBS buffer pH 7.2. Red circles indicate individual intensity tracks. Blue dotted lines show large aggregates. (C) Mean NTA number size distribution (n = 5) of BMP-2 aggregates in MES buffer pH 5.

Figure 13

Schematic presentation of the apparent aggregation equilibrium of BMP-2 at pH 5 and pH 7.4.