Ex Vivo and In Vivo Evaluation of Dodecaborate-Based Clusters Encapsulated in Ferumoxytol Nanoparticles

Abstract

Host-guest interactions represent a growing research area with recent work demonstrating the ability to chemically manipulate both host molecules as well as guest molecules to vary the type and strength of bonding. Much less is known about the interactions of the guest molecules and hybrid materials containing similar chemical features to typical macrocyclic hosts. This work uses in vitro and in vivo kinetic analyses to investigate the interaction of closo-dodecahydrododecaborate derivatives with ferumoxytol, an iron oxide nanoparticle with a carboxylated dextran coating. We find that several boron cluster derivatives can become encapsulated into ferumoxytol, and the lack of pH dependence in these interactions suggests that ion pairing, hydrophobic/hydrophilic interaction, and hydrogen bonding are not the driving force for encapsulation in this system. Biodistribution experiments in BALB/c mice show that this system is nontoxic at the reported dosage and demonstrate that encapsulation of dodecaborate-based clusters in ferumoxytol can alter the biodistribution of the guest molecules.

Figure 1.

A) Previous work describing the encapsulation of boron-rich compounds with beta-cyclodextrin (β-CD). B) Previous work describing the encapsulation of carbon-based small molecules in ferumoxytol (FMX). C) This work, the encapsulation of boron-rich compounds with ferumoxytol, a dextran-coated, iron oxide nanoparticle. The blue icosahedron represents various dodecaborate-based compounds and the red cone represents various carbon-based small molecules.

Figure 2.

Overview of dynamic dialysis experiment used to determine kinetic parameters. (A) Dialysis cup used for experiments. The MWCO filter allows for the passage of free boron clusters but not ferumoxytol nanoparticles. (B) Decrease of boron content over time (measured by ICP-OES) as the boron clusters leach out of the ferumoxytol nanoparticles. (C) Retention of iron content over time (measured by ICP-OES) as the nanoparticles are retained in the dialysis cup. (D) ¹¹B NMR of free Na₂B₁₂H₁₂ (blue trace) as well as Na₂B₁₂H₁₂ encapsulated in ferumoxytol (black trace). Both spectra are referenced internally to B(OH)₃ ca. 20 ppm. Error bars show one standard deviation from measurement replicates.

Figure 3.

Dynamic dialysis results for the release of 1 from ferumoxytol in phosphate-buffered saline (PBS). Data were fitted to exponential decay trendlines to model first order release kinetics. Data were recorded as single trials with error bars showing one standard deviation from measurement replicates.

Figure 4.

Dodecaborate derivatives screened. Kinetic data are shown in Table 1

Figure 5.

Biodistribution of ferumoxytol-encapsulated 1 (1@FMX, blue circles) compared to 1 (orange squares) alone and 2 (green triangles) in selected organs. Boron content was measured via ICP-OES and is reported as mass of boron per gram of wet tissue. All values were measured in triplicate and error bars show one standard deviation. Please see the Supporting Information for more experimental details as well as the complete biodistribution data. The inset on each graph shows the first 5 time points.