Fluorescent Deferoxamine Complexes of Cu(II) and Zr(IV): Insights in the Development of Dual Imaging Probes

Abstract

Deferoxamine (DFO) is widely regarded as benchmark chelator for ⁸⁹Zr(IV), a commonly used PET (positron emission tomography) tracer. We have introduced a novel fluorescent coumarin derivative of DFO (DFOKC), characterized by chelating unit and fluorophore covalently linked via a lysine molecule. This design introduces a free primary amine group, which, in perspective, can be functionalized with biological vectors, potentially improving tumor tissue selectivity. Its acid-base and metal coordination properties toward Cu(II) and Zr(IV) ions were thoroughly characterized using UV-Vis and fluorescence emission spectroscopy. DFOKC strongly coordinates both metal ions, forming somewhat more stable complexes than DFO, while retaining fluorescence emission, thus enabling dual-mode optical and PET imaging. Biodistribution assays conducted on NIH-3T3 fibroblasts, and MDA-MB-231 mammary adenocarcinoma cell lines demonstrated that the presence of primary amine groups favors Zr-DFOKC complex cell internalization via pinocytosis compared to the parent molecule DFOC, in which the fluorophore is linked to the amine group of DFO. Furthermore, crystal violet and MTT assays revealed no cytotoxic effects or mitochondrial impairment, even at concentrations higher than those typically used for radio-diagnostic applications. These results strongly support the potential of DFOKC as a versatile and promising tool for dual imaging, offering significant advantages in molecular imaging.

Keywords: Cu(II) complexes; Zr(IV) complexes; deferoxamine; fluorescent PET probes; optical/PET dual mode imaging.

Figure 1

Drawing of DFOC and DFOKC fluorescent ligands.

Scheme 1

Synthetic pathway of DFOKC.

Figure 2

Absorbance values at 220 and 434 nm of DFOKC superimposed to the distribution diagram of the species present in solution (DFOKC is indicated with L, [L] = 10⁻⁵ M).

Figure 3

UV‐Vis spectra of DFOKC at different pH values ([DFOKC] = 10⁻⁵ M, 0.1 M aqueous NMe₄Cl).

Figure 4

Fluorescence emission spectra of DFOKC at different pH values. Inset: variation of fluorescence emission intensity at 475 nm as a function of pH ([DFOKC] = 10⁻⁵ M, 0.1 M aqueous NMe4Cl, λ _exc = 400 nm).

Figure 5

Distribution diagram of the species present in solution of DFOKC in the presence of 1 equivalent of Cu(II) (DFOKC is indicated with L, [L] = [Cu(II)] = 10⁻⁵ M).

Figure 6

Fluorescence emission spectra of DFOKC in the presence of increasing amounts of Cu(II). Inset: variation of fluorescence emission intensity at 475 nm as a function of Cu(II) equivalents ([DFOKC] = [Cu(II)] = 10⁻⁵ M, 0.1 M aqueous NMe₄Cl, λ _exc = 400 nm).

Figure 7

UV‐Vis spectra of DFOKC at different pH values in the presence of 1 equivalent of Zr(IV) ([DFOKC] = [Zr(IV)] = 10⁻⁵ M, 0.1 M aqueous NMe4Cl).

Figure 8

Absorbance values at 265 and 433 nm of DFOKC in the presence of 1 equivalent of Zr(IV) superimposed to the distribution diagram of the species present in solution (DFOKC is indicated with L, making explicit the charged complexes present in solution. [L] = [Zr(IV)] = 10⁻⁵ M).

Figure 9

a) Observed (red/green cross) versus calculated (blue/violet squares) absorbances at selected wavelengths as generated by the HypSpec fitting of spectroscopic data for the determination of the Zr(IV) binding constants (reported in Table 3). b) Sample of a whole observed (red line) versus calculated (blue squares) spectrum (pH = 7.09) as resulting from the HypSpec treatment of the ligand's spectra to obtain the set of Zr(IV) binding constants reported in Table 3.

Figure 10

Fluorescence emission spectra of DFOKC in the presence of increasing amounts of Zr(IV). Inset: variation of fluorescence emission intensity at 475 nm as a function of Zr(IV) equivalents ([DFOKC] = [Zr(IV)] = 10⁻⁵ M, 0.1 M aqueous NMe₄Cl, λ _exc = 400 nm).

Figure 11

Cell viability assay by crystal violet vital dye uptake on NIH‐3T33 fibroblasts and MDA‐MB 231 human breast adenocarcinoma cells exposed to the noted compounds for 24 hours at increasing concentrations. No significant decrease in cell viability was detected at any concentrations assayed. C, control. Bars are the mean ± SEM of 3 replicate experiments. (one‐way ANOVA and Newman Keuls post‐test).

Figure 12

MTT mitochondrial metabolism assay on NIH‐3T3 fibroblasts and MDA‐MB 231 human breast adenocarcinoma cells exposed to the noted compounds for 24 hours at increasing concentrations. No significant changes in mitochondrial respiratory metabolism were detected. C, control. Bars are the mean ± SEM of 3 replicate experiments. (one‐way ANOVA and Newman Keuls post‐test).

Figure 13

Fluorescent (λ 540 nm emission) images of NIH‐3T3 fibroblasts and MDA‐MB 231 breast adenocarcinoma cells exposed to the noted compounds (0.1 µM) for 10 minutes. A dotted cyan fluorescence can be seen within the cells. A faint, diffuse cyan fluorescence can also be seen in the extracellular medium. Confocal microscopy, magnification ×630. The bar graphs show the number of fluorescent dots per cell area: NIH‐3T3 fibroblasts show no statistically significant differences among the tested compounds, whereas MDA‐MB 231 cancer cells show a significant increase in the cells treated with both Cu‐ and Zr‐DFOKC as compared with the parental compounds Cu‐ and Zr‐DFOC (*p < 0.05, **p < 0.01; one‐way ANOVA and Newman Keuls post‐test).

Figure 14

Representative ultrastructural images of MDA‐MB 231 human breast adenocarcinoma cells treated with the noted compounds (0.1 µM) for 10 minutes. The cells show a normal organelle complement with no signs of damage. Plasma membrane pits (arrows) and microvesicles in the peripheral cytoplasm (asterisks) are suggestive of pinocytosis. Some vesicles appear to fuse with cortical endosomes (e). The images of pinocytosis were more frequently observed in the cells exposed to CuDFOKC. Magnification is indicated under the bars.

Figure 15

Survival curves calculated from clonogenic assay of MDA‐MB 231 human breast adenocarcinoma cells treated or not with the noted compounds (0.1 µM) and exposed to X‐irradiation at increasing doses. (Two‐way ANOVA: p = 0.62, not significant).