Novel Gd nanoparticles enhance vascular contrast for high-resolution magnetic resonance imaging

Abstract

Background: Gadolinium (Gd), with its 7 unpaired electrons in 4f orbitals that provide a very large magnetic moment, is proven to be among the best agents for contrast enhanced MRI. Unfortunately, the most potent MR contrast agent based on Gd requires relatively high doses of Gd. The Gd-chelated to diethylene-triamine-penta-acetic acid (DTPA), or other derivatives (at 0.1 mmole/kg recommended dose), distribute broadly into tissues and clear through the kidney. These contrast agents carry the risk of Nephrogenic Systemic Fibrosis (NSF), particularly in kidney impaired subjects. Thus, Gd contrast agents that produce higher resolution images using a much lower Gd dose could address both imaging sensitivity and Gd safety.

Methodology/principal findings: To determine whether a biocompatible lipid nanoparticle with surface bound Gd can improve MRI contrast sensitivity, we constructed Gd-lipid nanoparticles (Gd-LNP) containing lipid bound DTPA and Gd. The Gd-LNP were intravenously administered to rats and MR images collected. We found that Gd in Gd-LNP produced a greater than 33-fold higher longitudinal (T(1)) relaxivity, r(1), constant than the current FDA approved Gd-chelated contrast agents. Intravenous administration of these Gd-LNP at only 3% of the recommended clinical Gd dose produced MRI signal-to-noise ratios of greater than 300 in all vasculatures. Unlike current Gd contrast agents, these Gd-LNP stably retained Gd in normal vasculature, and are eliminated predominately through the biliary, instead of the renal system. Gd-LNP did not appear to accumulate in the liver or kidney, and was eliminated completely within 24 hrs.

Conclusions/significance: The novel Gd-nanoparticles provide high quality contrast enhanced vascular MRI at 97% reduced dose of Gd and do not rely on renal clearance. This new agent is likely to be suitable for patients exhibiting varying degrees of renal impairment. The simple and adaptive nanoparticle design could accommodate ligand or receptor coating for drug delivery optimization and in vivo drug-target definition in system biology profiling, increasing the margin of safety in treatment of cancers and other diseases.

Figure 1. Effects of serum on Gd-concentration dependent change in longitudinal (T₁) relaxation time and comparison of lipid-nanoparticle morphology between Gd-DTPA liposome and Gd-lipid nanoparticles.

Panel A: Gd-lipid nanoparticles containing mPEG₂₀₀₀-PE (○,●) or Gd-DTPA liposomes without (mPEG) (◊,♦), were exposed to serum (●,♦) and relaxation time was measured with a 3T MRI instrument as described in Table 1. The data were fitted using linear regression. The electron micrographs represent morphology of Gd-lipid nanoparticles (panel B) and Gd-DTPA liposomes (panel C). Samples were negatively stained with 1% phosphotunstate. Please note the small electron exclusion bodies surrounding the Gd-lipid nanoparticles that were not detectable with Gd-DTPA liposomes. The bars in panels B and C represent 100 nm.

Figure 2. Comparison of whole body MR image obtained using a 3T MR instrument.

The rats were intravenously given indicated doses of Gd, in Gd-DTPA-BMA (Omniscan, panel A), Gd-DTPA-DPC (Vasovist, panel B) or Gd-lipid nanoparticles (panels C-E). The mmole/kg dose of Gd were 0.05 for Omniscan (panel A), 0.03 for Vasovist (panel B), 0.00125, 0.01, 0.02 for Gd-lipid-nanoparticles (panels C, D & E, respectively). The DCE-MR images were collected at 15 min post Gd administration. The circles and arrows indicate Gd-dependent MR contrast in the bladder and the heart blood vessels. Also, the catheter line used for IV administration of Gd contrast is apparent as a high contrast line.

Figure 3. Time course of Gd-enhanced MR images for Gd-lipid nanoparticles.

The rats were administered intravenously with 0.01 mmole/kg of Gd-lipid nanoparticles. The MR images at indicated time were collected with a 3T MR scanner. The animal was also scanned before dosing (panel A), 5 min (panel B), 15 min (panel C) or 24 hr (panelD) after Gd nanoparticle dosing. By 15 min (C), appearance of high positive contrast due to Gd elimination into the bile which connects to the intestine (arrows) and the gut is apparent. Also by 24 hrs, all the Gd-lipid nanoparticles were eliminated (panel D). Please note that the image for 24 hr did not line up due to different animal position. The contrast at 24 hr is found mainly in the GI tract, some of which are contributed by fatty materials as well. Also, due to leakage at the site of Gd administration at the left femoral vein, Gd localization is apparent as a high contrast area. Please note the vaso-restriction ○ clearly apparent in panels B & C.

Figure 4. Time-course blood Gd kinectics in rats.

Rats were intravenously given 0.5 mmole/kg of ¹²⁵Gd either in Gd-lipid nanoparticles (Gd-LNP) or soluble Gd-DTPA and Gd concentrations in blood were analyzed and presented as µmole/ml. Data presented were mean ±a S.D of n = 4 per treatment group. * Indicates p<0.05 by student T test.

Figure 5. Comparison of Gd in LNP (hatched) vs soluble Gd-DTPA (filled) formulation excreted in feces and urine at 24 and 48 hrs.

Rats were intravenously given 0.5 mmole/kg of ¹²⁵Gd either in Gd-lipid nanoparticles (Gd-LNP) or soluble Gd-DTPA. The urine and feces collected from rats housed in metabolic cages were analyzed at 24 and 48 hr. Data were expressed as mean ±a S.D. in µmole/g [Gd] for each treatment group. * Indicates p<0.05 by student T test.

Figure 6. Schematic presentation of soluble and lipid nanoparticle associated Gd-DTPA molecular rotation and impact on relaxivity as well as their retention in healthy blood vasculature.

The medium circular Gd and small circular DTPA chelate in solution or H₂O exhibit high molecular rotation (∼10¹⁰ sec⁻¹) that influences water proton relaxation, which typically produce 4–6 mM⁻¹*sec⁻¹ relaxivity. When Gd-DTPA is bound to lipid nanoparticle surface with surface bound water molecules attracted by PEG and Gd's molecular rotation is reduced to ∼10^7–8 sec⁻¹, relaxivity is greatly increased to about ∼130 mM⁻¹*sec⁻¹ (not drawn to scale). Gd in Gd-LNP also reduces the cellular uptake and thus lowers the cellular and tissue accumulation in non-pathogenic tissues and cells which may relate to clinical presentation of fibrosis.