Comparison of Two Ultrasmall Superparamagnetic Iron Oxides on Cytotoxicity and MR Imaging of Tumors

Abstract

Purpose: This study was performed to compare the cytotoxicity and magnetic resonance (MR) contrast in diverse cultured cells and xenograft tumors models of two ultra-small superparamagnetic iron oxides (USPIOs), thermally cross-linked superparamagnetic iron oxide nanoparticles (TCL-SPION) and monocrystalline iron oxide nanoparticles (MION-47).

Materials and methods: Transmission electron microscopy (TEM) images and R(2) relaxivity values of the TCL-SPION and MION-47 were obtained and the cell viability and cell growth velocity of treated and untreated human fibroblasts and human umbilical vein endothelial cells (HUVEC) were evaluated. The effect of TCL-SPION and MION-47 on the secretion of interlukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α), the production of nitric oxides and the mitochondrial membrane potentials in murine macrophage cells (RAW264.7) was compared. Human hepatocellular carcinoma cells (HepG2, 5x10(5)) were subcutaneously injected into nude mice (BALB/c) and in vivo MR imaging of tumors before and after injection with TCL-SPION or MION-47 (12.5 mg Fe/kg) was performed on a 1.5 Tesla MRI scanner.

Results: On TEM images, the average core diameter of TCL-SPION was 9 nm whereas that of MION-47 was 5 nm. TCL- SPION (345.0 ± 6.2 mM(-1)sec(-1)) had higher relaxivity (R(2)) than MION-47 (130.7 ± 1.1 mM(-1)sec(-1)). Significant changes in cell viability and growth were not found in human fibroblasts and HUVEC exposed to TCL-SPION and MION-47. However, IL-6 and TNF-α secretions increased dose-dependently and significantly in the macrophages treated with MION-47 or TCL-SPION. TCL-SPION had a lower stimulatory effect on IL-6 secretions than did MION-47 (P <0.05) and nitric oxides were produced in the macrophages by MION-47 but not TCL-SPION. A change in the mitochondrial membrane potential of the macrophages was observed 24 hours after the exposure, and MION-47 induced more collapses of the mitochondrial membrane potential than did TCL-SPION. In the in vivo MR imaging, 33.0 ± 1.3% and 7.5 ± 0.4% signal intensity decrease on T(2)*-weighted images was observed in the tumors injected with TCL-SPION and MION-47, respectively.

Conclusion: Due to the modified surface properties and larger core size of its iron oxide nanoparticles, TCL-SPION achieves lower cytotoxicity and better tumor MR contrast than MION-47. Our study suggests that TCL-SPION may be used as a new platform for tumor imaging and therapy monitoring.

Keywords: MION-47; Magnetic resonance imaging; TCL-SPION; Tumor targeting.; Ultra-small superparamagnetic iron oxides.

Figure 1

Physicochemical parameters and comparison of magnetic properties. (A) A schematic representation of TCL-SPION and MION-47. (B) TEM image of TCL-SPION and MION-47. (C) A T₂-weighted MR image of TCL-SPION and MION-47 at various iron oxide concentrations. At identical iron oxide concentrations, TCL-SPION had a more intense MR contrast effect than MION-47. (D) Plots of the R₂ (1/T₂) value. The relaxivity (R₂) of TCL-SPION (345.0 ± 6.2 mM^-1sec^-1, blue) was higher than that of MION-47 (130.7 ± 1.1 mM^-1sec^-1, red).

Figure 2

In vitro evaluation of cytotoxicity in human normal cells. (A, B) The viabilities of human fibroblasts and human umbilical vein endothelial cells (HUVEC) were assessed by MTT assay. A reduction in the viability of human fibroblasts and HUVEC was not observed after 24 hours of exposure to increasing concentrations of TCL-SPION and MION-47 (up to 1 mg Fe/mL). (C, D) The growth velocity was assessed by a trypan blue assay. The human fibroblasts and HUVEC were exposed to 100 µg Fe/mL of TCL-SPION or MION-47 separately for 3 days, and there was no effect on the cell growth velocity. The data are presented as the means ± standard errors from at least three independent experiments.

Figure 3

In vitro evaluation of cytokines release and nitric oxide production in macrophage. The levels of (A) IL-6 and (B) TNF-α released from mouse macrophage cells (RAW264.7). RAW264.7 cells were treated with TCL-SPION or MION-47 at concentrations of 25 µg Fe/mL and 50 µg Fe/mL for 24 hours. The IL-6 and TNF-α releases in the RAW264.7 cells were dose-dependently and significantly increased by adding TCL-SPION and MION-47 (*, P < 0.001, vs. control). MION-47 showed a greater stimulatory effect on IL-6 release than did TCL-SPION, while the TNF-α stimulatory effect was similar. (C) Nitric oxide production in macrophages. Macrophages were treated TCL-SPION or MION-47 at concentrations of 25 µg Fe/mL, 50 µg Fe/mL and 100 µg Fe/mL for 24 hours. The nitrite levels were assayed by the Griess reaction. MION-47 induced a dose-dependent and significant increase in nitric oxide production (*, P < 0.001, vs. control), but TCL-SPION did not have any stimulatory effect on nitric oxide production. All data are presented as the means ± standard errors from at least three independent experiments.

Figure 4

Mitochondrial membrane potential of macrophage. (A) Flow cytometry analysis of JC-1 in macrophages. Macrophages were treated with TCL-SPION or MION-47 at a concentration of 50 µg Fe/mL for 6 hours and 24 hours. The TCL-SPION and MION-47 treatments gradually reduced the population of healthy cells (the JC-1 aggregates). (*, P < 0.05, vs. control) (B) Fluorescence image of JC-1 aggregates (red) and JC-1 monomers (green) in macrophages. The MION-47 treated cells emitted more green fluorescence than did the TCL-SPION treated cells. All data are presented as the means ± standard errors from at least three independent experiments.

Figure 5

In vivo MR imaging. (A) T₂*-weighted images of tumor-bearing mice injected with 12.5 mg Fe/kg of TCL-SPION or MION-47. After injection with TCL-SPION, a noticeable darkening appeared in the tumor area, indicating a large accumulation of TCL-SPION within the tumor. In the mice injected with MION-47, there was no considerable difference in the pre-injection and post-injection tumor signal intensities. (B) The changes in signal intensity measured on T₂*-weighted tumor images. The tumor signal intensity of the mice injected with TCL-SPION was decreased by 33.0 ± 1.3%. MION-47 did not cause a significant change in signal intensity after injection (7.5 ± 0.4%). (C) Prussian blue staining of tumor microsections. More TCL-SPION was accumulated in the tumors than MION-47. All data are presented as the means ± standard errors from at least three independent experiments.