Introduction

Development of bimodal ¹H/³¹P magnetic resonance responsive probe reflecting anatomy and biochemical activity of environment at the same time is extremely challenging². The advantage of these modalities is that they can provide complementary information about physiological conditions in living tissue.

The aim of this project is to develop and test a conceptually new class of biodegradable responsive phosphorus–containing contrast agent for ¹H/³¹P MR. Presented probe is based on phytate (myo-inositol-1,2,3,4,5,6-hexakisphosphate) crosslinked with Ca²⁺ ions (CaIP₆) and doped with different concentrations of Fe³⁺ ions. High content of phosphorus in phytate enables its detection by ³¹P MRS/MRI. Linked paramagnetic Fe³⁺ ions broaden the phosphorus signal, making it invisible on ³¹P MR and act as a probe for ¹H MR only. When the link is broken and Fe³⁺ is cleaved out, the ³¹P signal appears due to its narrowing in the spectrum. This detachment can act as a response to external physiological changes (lower pH in cancer tissue¹, bacteria producing iron chelating siderophores³ etc.). Here we present a pilot in vitro study focusing on the effect of linked iron on ³¹P MR signal and its chelation using iron-binding bacterial siderophore – deferoxamine (DFOA).

Methods

Presented nanoparticles (100–300 nm) are based on CaIP₆ (phosphorus c= 1mM) doped with different concentrations of paramagnetic Fe³⁺ ions (iron c= 0, 0.68, 2.04, 2.73, 5.43, 13.6 mM,) linked by Ca²⁺ ions replacement. In vitro simulation of Fe³⁺ release was performed with DFOA (2.73 mM Fe³⁺+DFOA; [DFOA]:[CaIP₆] ratio 1:1)⁴, a compound possessing high affinity to Fe³⁺ ions. We chose 2.73 mM Fe³⁺ probe for testing, because of its ³¹P MR signal widening and adequate iron concentration to avoid over–chelation. Sufficiently high molecular weight of contrast agents (above renal threshold) assures sufficiently long blood circulation.

Firstly probes were measured by 7 T ³¹P NMR for signal confirmation and then tested at 4.7 T scanner using self-made ¹H/³¹P RF coil, which is wide tuned for both ¹H and ³¹P nuclei, so that anatomical reference acquired by ¹H MR could be obtained. ¹H MR (FLASH, repetition time/echo time TR/TE=100/6 ms) was firstly applied for probes positioning. Probes MR properties were than assessed by ³¹P MRI/MRS. Single pulse sequence was used for obtaining ³¹P T1 relaxation times (TR=160–3000 ms, 3000 acquisitions) and for spectra comparison between probes containing different Fe³⁺ concentrations (TR=500 ms, scan time ST=16 h 40 m). For ³¹P MRI chemical shift imaging (CSI) sequence was optimized (TR=500 ms, ST=1 h, resolution 2.5x2.5x5.8 mm³). For visualization of Fe³⁺ influence on MR signal ¹H MRI of phantoms was measured (T₂-weighted imaging, TR/TE=2000/24 ms). Signal-to-noise ratio (SNR) was calculated from both ¹H and ³¹P MR images and from ³¹P MR spectra. Pilot testing of highly concentrated 2.73 mM Fe³⁺ contrast cytotoxicity of probe was assessed using Alamar Blue assay (4T1 mammary carcinoma cell line; 48 h incubation).

Results

¹H MRI of phantoms containing high amount of iron appear darker on T₂-weighted images, with the lowest SNR=1.2 calculated from the probe doped with the highest Fe³⁺ concentration and the highest SNR=86.2 calculated for probe containing no iron. Effect of iron chelation was observed in ¹H MRI (2.73 mM Fe³⁺ probe SNR=25.4; 2.73 mM Fe³⁺+DFOA probe SNR=34.7) (Fig. 1). We confirmed linked iron ions impact on ³¹P MRS signal widening with increasing Fe³⁺ concentrations and no signal obtained from concentrations higher than 2.73 mM Fe³⁺. Influence of DFOA was visible in ³¹P MRS data of 2.73 mM Fe³⁺+DFOA, showing signal increase (SNR=72.2) after ferric ions recomplexation by siderophore, comparing to 2.73 mM Fe³⁺ without DFOA (SNR=13) and to probe with no iron doping (SNR=45.9) (Fig.2). Signal restorations was also observed on ³¹P NMR, where a significant increase in signal intensity was observed in probe with DFOA. In ³¹P MRI signal was observed for low Fe³⁺ concentrations, with the highest SNR=57.1 for 0.68 mM Fe³⁺ (Fig. 1). ³¹P T₁ relaxation times are decreasing with iron doping increase (0.68 mM Fe³⁺: T₁=487 ms; 2.04 mM Fe³⁺: T₁=305 ms). Cytotoxicity testing confirms that our probe is not significantly influencing cells viability.

Discussion

Our results confirms the possible use of this new contrast agent as a ¹H/³¹P MR responsive probe offering two imaging modalities. The processed data confirmed the effect of iron ions and iron chelating compounds (DFOA) on ³¹P MR signal, which changes can be used as a response for sensing early tissue pathology and biochemical processes in living organism. Small inconsistences in the SNR results can be explained due to the low sensitivity at magnetic field strength 4.7 T, however should not affect the detectability of ³¹P MR signal even in the lower magnetic fields used in clinical practice.

Conclusion

We present bimodal ¹H/³¹P MR responsive probe as the proof of principle. This new class of biodegradable responsive phosphorus-containing probe confirms sufficient sensitivity for exogenously triggered ³¹P MRI/MRS at 4.7 T. Our results indicates no cytotoxicity of testes probes; however further detailed testing on different cell lines should be performed.

Acknowledgements

The study was supported by the Charles University, GA UK No 358119; Institute for Clinical and Experimental Medicine IKEM, IN00023001; Charles University, First Faculty of Medicine; Ministry of Education of the Czech Republic through the SGS project no. 21332/115 of the Technical University of Liberec.