Tangential Flow Ultrafiltration Allows Purification and Concentration of Lauric Acid-/Albumin-Coated Particles for Improved Magnetic Treatment

Abstract

Superparamagnetic iron oxide nanoparticles (SPIONs) are frequently used for drug targeting, hyperthermia and other biomedical purposes. Recently, we have reported the synthesis of lauric acid-/albumin-coated iron oxide nanoparticles SEON(LA-BSA), which were synthesized using excess albumin. For optimization of magnetic treatment applications, SPION suspensions need to be purified of excess surfactant and concentrated. Conventional methods for the purification and concentration of such ferrofluids often involve high shear stress and low purification rates for macromolecules, like albumin. In this work, removal of albumin by low shear stress tangential ultrafiltration and its influence on SEON(LA-BSA) particles was studied. Hydrodynamic size, surface properties and, consequently, colloidal stability of the nanoparticles remained unchanged by filtration or concentration up to four-fold (v/v). Thereby, the saturation magnetization of the suspension can be increased from 446.5 A/m up to 1667.9 A/m. In vitro analysis revealed that cellular uptake of SEON(LA-BSA) changed only marginally. The specific absorption rate (SAR) was not greatly affected by concentration. In contrast, the maximum temperature Tmax in magnetic hyperthermia is greatly enhanced from 44.4 °C up to 64.9 °C by the concentration of the particles up to 16.9 mg/mL total iron. Taken together, tangential ultrafiltration is feasible for purifying and concentrating complex hybrid coated SPION suspensions without negatively influencing specific particle characteristics. This enhances their potential for magnetic treatment.

Keywords: hyperthermia; nanoparticle concentration; nanoparticle purification; specific absorption rate (SAR); superparamagnetic iron oxide nanoparticles (SPIONs); tangential ultrafiltration.

Figure 1

Effect of ultrafiltration on protein content and hydrodynamic aggregate size of SEON^LA-BSA. The relative dry mass (A) of SEON^LA-BSA (V = 5 mL) was calculated after ultrafiltration with different volumes of ultrapure water as the washing agent. The mass displayed is relative to the mass of the precursor SEON^LA in order to discriminate the mass of albumin from the mass of the particle cores; (B) Photon cross-correlation spectroscopy (PCCS) measurements displayed as volume mean diameter of the samples from (A). All measurements were performed in triplicate.

Figure 2

Linear plots of desorption isotherms of albumin from SEON^LA-BSA. Freundlich (A) and Langmuir (B) isotherms showed a good determination coefficient R². Apparently, the Freundlich model was able to better describe the desorption of albumin from the particle matrix with an R² of 0.9907 and a slope of 1/n = −1.06.

Figure 3

pH-dependent zeta potential and hydrodynamic size measurements. The development of (A) the zeta potential and (B) the hydrodynamic diameter (z-average) depending on the pH as measured by DLS. SEON^LA-BSA before filtration is depicted in red; SEON^LA-BSA after filtration with a double excess volume of ultrapure water is depicted in blue. All measurements were performed in triplicate 30 s after sonication.

Figure 4

Efficiency of the concentration and the effect on the hydrodynamic aggregate size of SEON^LA-BSA. The iron content of SEON^LA-BSA (A) depending on the ratio of the starting volume/end volume (concentration factor); (B) PCCS measurements of the hydrodynamic diameters of the samples from (A). All measurements for (A,B) were performed in triplicate.

Figure 5

Increase of the saturation magnetization of SEON^LA-BSA of the samples. The saturation magnetization increases in a linear way (R² = 0.9937) with the respective iron content of the samples, which was measured in triplicate using the aforementioned UV-VIS method.

Figure 6

Blood stability assays. Exemplary light microscopy images of (from left to right) concentrated SEON^LA-BSA (c(Fe) = 16.9 ± 1.43 mg/mL), PBS and SEON^LA (c(Fe) = 17.2 ± 0.69 mg/mL) diluted 1:2 in whole blood stabilized with EDTA. The arrows mark exemplary particle aggregates.

Figure 7

Influence of tangential ultrafiltration on cellular uptake of SEON^LA-BSA. Total iron oxide concentration of filtrated (blue bars) and un-filtrated (red bars) SEON^LA-BSA in Jurkat cells after 48 h of incubation. The uptake was significantly reduced by filtration for all three concentrations tested (* p < 0.05; ** p < 0.005). All measurements were performed in triplicate; the background was subtracted.

Figure 8

SAR and T_max of SEON^LA-BSA after filtration and/or concentration. Specific absorption rates (blue bars) and maximum heating temperatures (yellow dots) of un-filtrated SEON^LA-BSA and filtrated and concentrated SEON^LA-BSA. All measurements were performed in triplicate.