Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Magn Reson. Author manuscript; available in PMC 2012 March 1.
Published in final edited form as:
PMCID: PMC3045750

Off-Resonance Saturation MRI of Superparamagnetic Nanoprobes: Theoretical Models and Experimental Validations


Off-resonance saturation (ORS) is a new magnetic resonance imaging (MRI) method that has shown greatly improved contrast sensitivity for the detection of cancer-specific biomarkers by superparamagnetic nanoprobes in vivo. However, quantitative understanding of the ORS contrast mechanism and its dependence on the structural parameters of superparamagnetic nanoprobes are still lacking. Here we propose a quantitative model of ORS contrast and its experimental validation by superparamagnetic polymeric micelles (SPPM) with precisely controlled structural properties. Size selected, monodisperse Fe3O4 nanoparticles (6.1 ± 0.2 nm) were used to form a series of SPPM nanoprobes with specifically controlled corona thickness using 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxypoly(ethylene glycol) (DSPE-PEG) with different PEG molecular weights. Transmission electron microscopy and dynamic light scattering showed that SPPM were uniform in size. The average hydrodynamic diameters of SPPM with PEG lengths of 0.55, 1, 2, and 5 kD were 16.6 ± 2.8, 18.4 ± 2.9, 24.1 ± 3.4, and 28.9 ± 3.4 nm, respectively. MRI experiments at 7T determined that r2 values of SPPM with 0.55, 1, 2, and 5 kD PEG as corona were 201 ± 3, 136 ± 8, 107 ± 5, and 108 ± 8 Fe mM-1s-1, respectively. ORS intensity from Z-spectra of SPPM showed a significant correlation with the inverse of T2 relaxation rates (1/T2, s-1) of the SPPM nanoprobes regardless of the PEG corona thickness. These data provide the fundamental understanding of the structure-property relationships between the SPPM nanostructures and ORS sensitivity, which offers useful mechanistic insights for the future improvement of SPPM nanoprobes in cancer molecular imaging applications.

Keywords: magnetic resonance imaging, transverse relaxivity, off-resonance saturation, superparamagnetic polymeric micelles, PEG corona length, nanoprobes

1. Introduction

Recent development of novel MR probes and new imaging methods with improved imaging sensitivity and specificity have rapidly advanced the use of MRI in molecular and cellular imaging applications [1-4]. Compared to the low molecular weight, paramagnetic metal chelates such as Gd-DTPA, superparamagnetic nanoparticles (e.g., Fe3O4 [5-7], MnFe2O4 [8], FeCo [9] nanocrystals) have demonstrated substantially higher molar relaxivities and improved sensitivity for in vivo MR imaging. Surface functionalization of superparamagnetic iron oxide (SPIO) nanoparticles has provided multiple examples of targeted contrast agents that specifically recognize tumor markers such as the receptors of transferrin [10], folate [11], and Her-2/neu [12, 13]. Once bound to a targeted marker or after internalization into a cell, SPIO probes can create substantial disturbances in the local magnetic field leading to a rapid dephasing of protons and loss of MR signal intensity. T2- or T2*-weighted (T2/T2*-w) MR imaging methods are most commonly used for SPIO imaging. T2*-w images obtained by gradient echo (GRE) imaging sequences often have low signal-to-noise ratios. T2*-w MR imaging is also highly sensitive to the magnetic susceptibility produced by the presence of SPIO-based agents producing signal voids larger than the actual sizes of accumulated SPIO agents. Finally, image voids from SPIO agents may be difficult to differentiate from other tissue voids (e.g., air sacs in the lung or abdominal cavity) or void-like signals caused by severe magnetic anisotropy that can occur at tissue interfaces.

Due to the above limitations, several recent studies reported new imaging techniques to generate “positive” MRI contrast from SPIO-based agents [14-21], which can potentially improve imaging accuracy of these agents. Seppenwoolde et al. reported a technique called White Marker to track paramagnetic markers stationed in a catheter inserted into a pig's abdominal aorta [14]. The positive contrast was achieved by dephasing of background signal while the signal near the markers was not affected due to the dipole field generated by the markers. Cunnigham et al. reported the use of a similar method to obtain positive contrast from cells labeled with superparamagnetic nanoparticles [15]. In another work, SPIO-labeled stem cells in rat hearts were visualized using a sweep imaging with Fourier transformation (SWIFT) technique, in which nuclei were excited by a swept radiofrequency excitation with simultaneous signal acquisition [16]. Positive contrasts of SPIO-labeled stem cells were generated in the imaginary component, which can be registered with anatomical images of similar imaging planes. Recently, Zurkiya et al. and our research group have reported an off-resonance saturation (ORS) method to produce positive contrast of SPIO nanoparticles [19-21]. This technique involved the application of an RF pulse at an off-resonance frequency and the positive contrast was obtained by taking the ratio or subtraction of the images with and without the saturation.

Our previous study has demonstrated the effectiveness of ORS method for in vivo detection of tumor-targeted superparamagnetic polymeric micelles (SPPM) in tumor-bearing mice [21]. The ultrasensitive SPPM design provided a significant increase in sensitivity of detection by ORS MRI to the picomolar range of the nanoparticles. The ORS technique also improved the imaging accuracy compared to conventional T2*-weighted imaging method. This technique also allowed for visualization of image contrast without a pre-contrast acquisition. The ability to provide increased imaging sensitivity and accuracy while producing positive contrast has made ORS a promising technique for MR molecular imaging of SPPM. In this report, we describe the development of a theoretical model to quantitatively understand the ORS contrast mechanism, while experimentally validating the model using a set of structurally defined SPPM nanoprobes.

2. ORS Theory

2.1 Fast Diffusion and Quantitative ORS Model

When a SPPM particle is magnetized by an external magnetic field, the relaxation times of neighboring water protons are greatly shortened. First, we assume that all outer-sphere (OS) water molecules diffuse rapidly on the NMR time scale and that the bulk water signal can be characterized by a single set of longitudinal (T1) and transverse (T2) relaxation times. After applying a soft continuous wave RF pulse at a power level of ω1 (ω1 = 2πB1, B1 in Hz) and a frequency offset Ω for a duration of t, the magnetization of the attenuated water signal can be described by the Bloch equation:




and Mzo is the equilibrium Z-magnetization component in the static external magnetic field. The solution of Bloch equation can be expressed as


where Mo is the magnetization vector at initial state prior to saturation, and M is the magnetization at steady-state,


Note that the second term of Eq. 2 represents the magnetization difference between the initial and steady-state, modulated by a time-dependent function e-At (e-At converges to 0 at t→∞).

Equation 4 is central to the understanding of the steady-state Z-magnetization (Mz and thereafter referred to as Mz) and its relationship with MRI parameters. The dependence of Z-magnetization on RF saturation power (ω1), saturation frequency offset (Ω), and the relaxation times T1 and T2 can be predicted and quantified using this equation. Z-magnetization (Mz/Mzo) [22] of a solution with given relaxation times, T1 and T2, is highly influenced by offset saturation frequency (Ω). At a saturation frequency far away from the bulk water peak, |Ω| → ∞, Mz/Mzo is approximately 1 as the Ω2 T22 term becomes the dominant factor in the equation. In other words, the ORS intensity is negligible because the initial and steady state Z-magnetizations are equivalent. Conversely, a decrease in Mz/Mzo or more pronounced ORS contrast is expected, when the saturation frequency offset is approaching the on-resonance frequency of water, or |Ω| → 0 ppm. The Ω2 T22 term becomes less dominating and the relaxation times play a bigger role in dictating the ORS intensity resulting in smaller Mz/Mzo.

2.2 Approximation of ORS Models

In our previous work, we have shown ORS of SPPM solutions and its dependence on saturation frequency and SPPM concentration [21]. For in vivo ORS imaging, we used pre-determined imaging parameters such as saturation power ω1 and frequency offset Ω. With these known parameters, it is possible to predict and quantify the ORS intensity (Mz/Mzo). To test its feasibility, Equation 4 is further derived to propose a simplified model to quantify the ORS intensity of SPPM.

Equation 4 can be rearranged to



x=ω12T1T21+Ω2T22,which is further rearranged to:x=(ω1Ω)2T1T21Ω2+T22

Under ORS conditions with a relatively large Ω (e.g. Ω ≥ 900 Hz or 3 ppm), where 1/Ω2 [double less-than sign] T22, Eq. 5 becomes:


Equation 6 shows a linear relationship between the ORS intensity and longitudinal relaxation time (T1) and inverse of transverse relaxation time (1/T2) of SPPM solutions at a given saturation power and frequency. In tissue samples, Eq. 6 can be further derived to show the correlation between the ORS intensity and SPPM concentration.




Because 1/T2-endogenous [double less-than sign] r2[SPPM] because T2-endogenous is longer than 200 ms and SPPM probes have high r2 values, Eq. 7 then becomes:


A linear relationship between ORS intensity and SPPM concentration is depicted in Eq. 8. This equations shows that ORS intensity of strongly dependent on T2 relaxivity and concentration of SPPM probes in addition to its dependence on saturation power and offset frequency. It is important to note, however, that similar substitution of (1/T1-endogenous+r1[SPPM]) for 1/T1 will not simplify the equation because both terms will likely be in the same order of magnitude. This is because SPPM formulations are not effective T1 relaxation agents with low r1 values.

3. Experimental Section

3.1 Preparation of SPPM

Fe3O4 nanoparticles (or SPIO) were synthesized from iron(III) acetylacetonate in benzyl ether based on a published procedure [23]. After synthesis, the particles were size-selected by repeated precipitation with ethanol from hexane to yield monodisperse Fe3O4 nanoparticles (6.1 ± 0.2 nm) as confirmed by transmission electron microscopy (TEM). 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-methoxypoly(ethylene glycol) (DSPE-PEG) with PEG molecular weights of 0.55, 1, 2, 5 kD (Avanti Polar Lipids, Inc. Alabaster, Alabama) were used to produce SPPM following a published procedure with slight modification [24]. Briefly, a mixture of SPIO (100 μg) and DSPE-PEG (2.5 mg) in THF was dried with a flow of argon generating a thin film. The film was rehydrated with pH 7.4 HEPES (10 mM) buffer that contains 150 mM NaCl (1 mL) at 70 °C. The mixture was vortexed for 5 min and incubated at 70 °C for 3 h with intermittent vortexing. The resulting mixture was then diluted with deionized water (3 mL), filtered through a nylon filter membrane (size cutoff 0.2 μm), and then concentrated using an Amicon Centrifugal Filter. SPPMs were collected and the final volume was adjusted to 1 mL.

3.2 Characterization of SPPM

All formulations of SPPM were characterized by TEM and dynamic light scattering (DLS). Carbon-coated TEM grids were subject to a glow-discharge treatment using a plasma etcher. A drop of a SPPM solution was placed on the hydrophilic grid for 2 min. After the SPPM solution was removed, a solution of 2% phosphotungstic acid (PTA, pH = 7.4) was placed on the grid for 30 s. The grids containing SPPM samples were then analyzed on a JOEL 1200 EX TEM microscope at an accelerating voltage of 120 kV. For size distribution analysis, SPPM solutions were centrifuged at 5000 rpm for 1 min before subjecting to size measurements using a Viscotek 802 DLS instrument (Houston, TX) equipped with a He-Ne laser (λ = 825 nm). Scattered light was collected at 90° and data were analyzed from 50 measurements at 20 °C. Iron concentration of SPPM solutions were measured by atomic absorption spectroscopy (AAS). First, a known volume of SPPM solutions were digested in aqua regia at 70 °C for 3 h. The solution was then diluted and the actual iron contents were determined using a Varian SpectrAA 50 (Varian Inc., Palo Alto, CA) spectrometer. The calibration curve was obtained using an iron standard solution (Sigma Aldrich, St. Louis, MO).

3.3 Magnetic Resonance Imaging

All MR experiments were performed on a 7T small animal imager with a 40 mm i.d. Millipede coil (Varian, Inc, Palo Alto, CA). The SPPM sample was placed in a 1 mL plastic syringe with both ends sealed with paraffin wrap. The syringes contained SPPM solutions with different concentrations (50, 100, 250, 500, 1000 μM Fe) and were inserted into a 10 mL plastic syringe, which was filled with deionized water and sealed with paraffin. T1 measurements were carried out using an inversion recovery pulse sequence with TR = 15 s, TE = 2.13 ms, and flip angle = 6°. Inversion time (TI) was arrayed from 5.1 ms to 10 s. The null intensity point of the series of TI-varying MRI images for each solution was visually verified and was used to cross-check the T1 values obtained from data fitting the curves of MRI intensity versus TI. T2 times of SPPM solutions were measured using a fast spin echo pulse sequence with TR = 2 s and TE = 7, 15, 20, 30, 45, 60, 90, 150, 300, 500 ms. ORS experiments were carried out using a spin echo (SE) pulse sequence (TR = 2s, TE = 8.5 ms), with a pre-saturation B1 power of 1.75 μT for 0.5 s. The off-resonance saturation frequencies (Ω) were 15, ± 3, ± 1.5, ± 1.2, ± 0.9, ± 0.6, ± 0.3, or 0 kHz from bulk water. The image obtained with the saturation frequency of 15 kHz (or 50 ppm) was used as a reference image. All experiments were conducted in triplicate using 3 different SPPM sample preparation for each formulation.

3.4 Data Analysis

The MR images were analyzed using ImageJ (NIH). The relaxivity values (i.e. r1 and r2) were calculated using a least-squares curve fitting of the reciprocal of relaxation times (i.e. 1/T1 and 1/T2) versus iron concentrations ([Fe]). For ORS experiments, the Mz/Mzo division images were generated by pixel-by-pixel division of the saturation image by the reference image. The Z-spectra were generated by plotting Mz/Mzo versus saturation frequencies from the bulk water peak. Comparison of apparent Mz/Mzo values obtained from MRI experiments with theoretical calculations based on Eq. 4 was achieved using MATLAB (MathWorks, Natick, MA).

4. Results and Discussion

A set of single SPIO-loaded micelles encapsulated by PEGylated phospholipids was used in this study. A monodisperse distribution of SPIO nanoparticles (Fig. 2a) was employed to ensure the uniformity of the resulting micelle samples. PEG corona thickness was controlled by the use of DSPE-PEG lipids with varying PEG molecular weights from 0.55 to 5 kD. Each SPPM contained only one SPIO nanoparticle due to the short hydrophobic distearoyl chains of the phospolipid. TEM analyses revealed that all of the SPPM samples had a core-shell structure (Fig. 2), where staining with 2% PTA solution showed bright circles of the SPPM's hydrophobic cores encasing black SPIO nanoparticles. SPPMs are highly uniform with comparable core sizes (~ 13 nm) across all formulations (Fig. 2d-f). This is consistent with expected results due to each formulation containing the same disteraoyl lipid segment for SPIO encapsulation. DLS analysis showed the hydrodynamic diameters of SPPM became larger with an increase in PEG lengths. More specifically, the average diameters were 16.6 ± 2.8, 18.4 ± 2.9, 24.1 ± 3.4, and 28.9 ± 3.4 nm for the SPPM with PEG corona lengths of 0.55, 1, 2, 5 kD, respectively. The data confirmed that higher molecular weight of PEG will result in larger hydrated layer and thus larger hydrodynamic diameter of SPPM. The difference in hydrodynamic diameter between each SPPM formulation as a result of differing PEG molecular weights agrees well with an estimation of micelles' corona length based on PEG's molecular weight and radius of gyration [25]. This series of SPPM, with well-controlled structures, provided an excellent experimental system for the validation of the proposed ORS theory.

Figure 2
(a) TEM of as-synthesized Fe3O4 (6.1 ± 0.2 nm), (b) size distribution of SPPM by DLS, (c-f) TEM of SPPM with PEG molecular weights of 550 D, 1 kD, 2 kD, or 5 kD after staining with 2% PTA solution, respectively. The insets show the amplified images ...

MR relaxivities of all SPPM formulations were first investigated. T1 and T2 relaxation times of SPPM solutions were measured with Fe concentrations ranging from 50 μM to 1 mM. T1 and T2 relaxivity values (r1 and r2) are summarized in Fig. 3. T2 relaxivity decreased with an increase of the PEG molecular weight despite all the SPPM formulations containing the same SPIO core. Average r2 values (n=3) were 201 ± 3, 136 ± 8, 107 ± 5, and 108 ± 8 s-1 mM-1 Fe for SPPMs with 0.55, 1, 2, 5 kD PEG as corona, respectively. These results are consistent with those from LaConte et al., in which a similar trend was observed in the study of an effect of PEG length on T2 relaxation rate (1/T2) of monocrystalline iron oxide (MION) nanoparticles [26]. In contrast to transverse relaxation, the PEG length did not significantly affect the longitudinal relaxation of SPPM where r1 values are statistically equivalent (0.7-0.8 s-1 mM-1 Fe) from different formulations (Fig. 3).

Figure 3
T1 (top) and T2 (bottom) relaxivities (r1 and r2) of SPPM with different PEG corona lengths. The difference in r1 values across all formulations is not statistically significant. The single and double asterisks (* and **) indicate statistical significance ...

Next, we measured the dependence of ORS intensity on the off-resonance saturation frequency (Ω) at different SPPM concentrations. The Ω values were varied from 0 to 10 ppm on both positive and negative sides of the bulk water resonance frequency. Figure 4 shows a representative ORS image analysis of a SPPM sample with 5 kD PEG. SPPM-induced image darkening was visible (Fig. 4a) when the saturation frequency was closer to the water peak (|Ω| → 0 ppm). The fire-color-coded images are ORS images obtained by taking a division of spin-echo images with and without saturation (Mz/Mz0). Figure 4b plotted the ORS intensity as a function of Ω at four different SPPM concentrations. It is important to note that experimental data (red squares in Fig. 4b) and theoretical calculations (blue curves) provide a close fit for all four different SPPM concentrations. The theoretical data were calculated using Eq. 4 and other pre-determined parameters (e.g. T1 and T2 of each SPPM solution and pre-saturation (B1) power: ω1 = 1.75 μT or 75 Hz). These data validate the quantitative ORS models as derived from Bloch equations based on the steady-state assumption under the current pre-saturation conditions (ω1 = 1.75 μT, t = 500 ms). These results complement the ORS model by Hu and Zurkiya [19], where the authors elegantly demonstrated that water diffusion instead of water exchange-mediated magnetization transfer as the primary causation of ORS contrast. In their model, the retention time of water molecules in magnetic isosurfaces approximated by the dipole field is a key factor in generating the ORS contrast. Therefore, the diffusion rate of water molecules is an important driving force of the ORS contrast generating mechanism. It should be noted that a gradient echo sequence with short pre-saturation pulse (6 ms Gaussian pulse) was used in the reported study, which may not permit the steady-state assumption as allowed in our study (see supplemental Figure S1 for the reaching of steady state at different B1 power and saturation durations). Future work is necessary to examine the specific ORS acquisition conditions and boundary conditions that would allow for the quantitative correlation with MR relaxation parameters of the superparamagnetic nanoprobes.

Figure 4
ORS analysis of SPPM at different saturation frequency offsets. (a) Proton density images of SPPM at different concentrations obtained with RF saturation frequencies (top) and their corresponding division images (Mz/Mz0, bottom). (b) Simulated and experimental ...

Under our current model, the complete quantitative ORS expression (Eq. 4) can be greatly simplified at a relatively large Ω (e.g. Ω ≥ 900 Hz or 3 ppm). Under this condition, the ORS intensity (Mz/Mz0) becomes linearly correlated with T1/T2 (Eq. 6). First, we plotted the ORS intensity as a function of iron concentration for different SPPM nanoprobes. The left panel of Figure 5 shows that SPPM with the shortest PEG corona (i.e. 0.55 kD) demonstrated the highest concentration-dependent ORS intensity over SPPM with longer PEG corona. The ORS sensitivity (slope) increased with decrease of PEG lengths, i.e., PEG 0.55 kD > PEG 1 kD > PEG 2 kD ≈ PEG 5 kD. Similar to the T2 relaxivity study, SPPM samples with a PEG corona of 2 or 5 kD showed an almost identical ORS dependence on [Fe]. This trend is well supported by Eq. 8 showing the relationship between ORS intensity and transverse relaxivity values of SPPM nanoprobes. Interestingly, when the ORS intensity was plotted as a function of T2 relaxation rates (right panel, Fig. 5), all the data points from different SPPM samples collapsed into a single linear correlation. Such linear relationship can be explained by the low longitudinal relaxivity of all SPPM formulations, which will result in comparable T1 times due to the small r1 values (0.7-0.9 Fe mM-1s-1 with a standard deviation of approximately 0.1 Fe mM-1s-1). This result also suggests that the difference in T1 relaxivites between SPPM formulations will be small relative to endogenous T1 times in samples or tissues resulting in equivalent T1 times for all samples. Since saturation power (ω1) and frequency (Ω) are pre-determined parameters, the T1(ω1/Ω)2 term becomes a constant. The resulting ORS intensity is then linearly correlated with T2 relaxation rate as predicted by Eq. 6. These data suggest that for SPPM nanoprobes, T2 relaxation rate is a predominant factor that directly affects the ORS sensitivity of detection.

Figure 5
ORS intensity as a function of Fe concentration (left) and T2 relaxation rates (right) for SPPMs with different PEG corona thickness. Imaging parameters: TR = 2 s, TE = 8.5 ms, saturation power (ω1) = 1.75 μT, saturation duration = 500 ...

Our previous report has demonstrated the effectiveness of the ORS method to image αvβ3 integrins on tumor vasculature in a tumor-bearing animal by cRGD-encoded SPPM nanoprobes [21] Compared to conventional T2/T2 *-weighted imaging, the ORS method does not require pre-contrast scanning and, moreover, the contrast sensitivity can be greatly enhanced by the direct comparison between the “ON” and “OFF” images (i.e. with and without the pre-saturation) for SPIO-specific contrast. Despite these advantages, ORS imaging, like many other MRI techniques, can still benefit from a highly sensitive design of contrast probes for improved detection. Results from this study established the quantitative correlation of ORS intensity with the imaging acquisition conditions and microscopic structures and MR relaxation properties of the nanoprobes. This knowledge has several valuable implications for the optimized design of MR nanoprobes. First, ORS intensity is directly correlated to T1 and T2 of the nanoprobe and, in a simplified model (Eq. 6), is proportional to the ratio of T1/T2. This suggests that an ideal ORS contrast probe should be a “pure” T2 agent with minimal capability for T1 relaxation. Simulations of ORS intensity (Mz/Mzo) using Eq. 4 clearly show that the ORS effect is larger (Mz/Mzo is smaller) when T1 is longer (Fig. S1). However, longer T1 values also require a longer pre-saturation period to reach steady-state. The current SPPM design, where the as-synthesized SPIO is coated with hydrophobic oleic acids and further encapsulated in the hydrophobic core of the micelles, can effectively minimize direct contact with water molecules and result in significantly reduced T1 relaxivity as demonstrated by the experimental data (r1 = 0.7-0.9 Fe mM-1s-1, Fig. 3a) [27]. In comparison, the hydrophilic SPIO nanoparticles (e.g. by base precipitation method in dextran matrix) had higher T1 relaxivity (e.g. r1 = 23.9 Fe mM-1s-1 for Feridex) [28]. Consequently, results from Feridex and Gd-doped Feridex solutions show reduced ORS sensitivity (see Fig. S2) as a result of shorter T2 time and longer T1 time of Feridex compared to those of SPPM. On the other hand, strategies to increase the T2 relaxivity of the contrast agents will also lead to enhanced ORS detection. Development of highly magnetized superparamagnetic nanoparticles (e.g. FeCo [9], MnFeO4 [8], Zn-SPIO [29]) or SPIO-clustered micelles [30] will considerably increase the sensitivity of ORS detection. Second, although results from this study show that SPPM with shorter PEG length (e.g. 0.55 kD) have higher ORS sensitivity, it may not have sufficient stability in blood. For example, a 2 kD PEG length was used in the liposomal delivery of doxorubicin, which has demonstrated prolonged circulation and blood half-life of the nanoparticles [31]. Consequently, an optimal PEG length needs to be investigated to provide the best compromise between ORS sensitivity and desirable pharmacokinetics. Third, use of higher B1 power (ω1) and smaller frequency offset (Ω) can also further increase ORS sensitivity (Eqs. 4, 6, and 8). However, increase of B1 power should be carefully examined to stay below the limit of maximum specific absorption rate (i.e. 4 W/kg body weight) for RF exposure [32, 33], in particular for future clinical applications. Finally, higher magnetic fields are also beneficial to ORS imaging as shorter T2 times and longer T1 times of MRI contrast agents at higher fields [34] contribute to higher ORS contrast enhancement as suggested in Eqs. 6 and 8.

The combination of the ORS method and the ultrasensitive SPPM design is a great asset for MR molecular imaging applications, in which improved sensitivity and detection accuracy are highly desirable. Much work still needs to be performed in order for clinical translation of this technique. One shortcoming for current exogenous imaging agents such as SPIO-based probes and PARACEST complexes [35] is their limited targeting specificity to the disease locations. However, a number of strategies using cell-specific ligands have been reported to improve the accumulations of these agents at targeted sites such as tumorous tissues [8, 21] and atherosclerotic plaques [36, 37]. Our SPPM design offers greatly improved MR sensitivity as evidenced by their high relaxivity values [30]. ORS imaging allows for additional benefits in accurate detection of SPPM. This combination may offer considerable advantages over other class of MR agents with limited detection sensitivity and contribute to the existing arsenals of imaging tools for molecular diagnosis of cancer and other diseases.

5. Conclusions

In summary, we have established the theoretical framework and quantitative expressions to describe the ORS contrast mechanism of SPPM nanoprobes. The proposed ORS theory was well supported by the experimental data from a series of precisely controlled SPPMs with varied PEG lengths. Results showed that the thickness of PEG corona is a critical parameter that greatly affected MR relaxivity and ORS sensitivity of SPPM. The proposed ORS theory was able to quantitatively predict the contrast enhancement of a given SPPM solution with known relaxation times T1 and T2. Results from this study provide the mechanistic insights on the ORS contrast sensitivity of SPPM nanoparticles, which will assist the future development of SPPM as ultrasensitive MRI nanoprobes for in vivo imaging applications.

Figure 1
(a) Illustration of mechanism of SPPM-induced ORS contrast, and (b) a schematic representation of SPPM with different PEG corona lengths but the same Fe3O4 core. In ORS contrast, an off-resonance RF saturation pulse only affects the SPPM-containing water ...

Supplementary Material



We thank the financial support from the following agencies: National Institutes of Health (R21 EB005394 and R01 CA129011 to JG, NIH CA15531 to ADS) and Robert A. Welch Foundation (AT-584 to ADS). C. Khemtong was supported by a Department of Defense Breast Cancer Research Program Multidisciplinary Postdoctoral Award (W81XWH-06-1-0751).


Disclosure Statement: We declare no conflict of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Jun YW, Lee JH, Cheon J. Chemical design of nanoparticle probes for high-performance magnetic resonance imaging. Angew Chem Inter Edit. 2008;47:5122–5135. [PubMed]
2. Khemtong C, Kessinger CW, Gao JM. Polymeric nanomedicine for cancer MR imaging and drug delivery. Chem Commun. 2009;24:3497–3510. [PMC free article] [PubMed]
3. Na HB, Song IC, Hyeon T. Inorganic Nanoparticles for MRI Contrast Agents. Adv Mater. 2009;21:2133–2148.
4. Weissleder R, Pittet MJ. Imaging in the era of molecular oncology. Nature. 2008;452:580–9. [PMC free article] [PubMed]
5. Bulte JW, Kraitchman DL. Iron oxide MR contrast agents for molecular and cellular imaging. NMR Biomed. 2004;17:484–99. [PubMed]
6. Thorek DL, et al. Superparamagnetic iron oxide nanoparticle probes for molecular imaging. Ann Biomed Eng. 2006;34:23–38. [PubMed]
7. Wang YXJ, Hussain SM, Krestin GP. Superparamagnetic iron oxide contrast agents: physicochemical characteristics and applications in MR imaging. Eur Radiol. 2001;11:2319–2331. [PubMed]
8. Lee JH, et al. Artificially engineered magnetic nanoparticles for ultra-sensitive molecular imaging. Nat Med. 2007;13:95–99. [PubMed]
9. Seo WS, et al. FeCo/graphitic-shell nanocrystals as advanced magnetic-resonance-imaging and near-infrared agents. Nat Mater. 2006;5:971–6. [PubMed]
10. Hogemann-Savellano D, et al. The transferrin receptor: a potential molecular imaging marker for human cancer. Neoplasia. 2003;5:495–506. [PMC free article] [PubMed]
11. Sonvico F, et al. Folate-conjugated iron oxide nanoparticles for solid tumor targeting as potential specific magnetic hyperthermia mediators: synthesis, physicochemical characterization, and in vitro experiments. Bioconjug Chem. 2005;16:1181–8. [PubMed]
12. Artemov D, et al. MR molecular imaging of the Her-2/neu receptor in breast cancer cells using targeted iron oxide nanoparticles. Magn Reson Med. 2003;49:403–8. [PubMed]
13. Funovics MA, et al. MR imaging of the her2/neu and 9.2.27 tumor antigens using immunospecific contrast agents. Magn Reson Imaging. 2004;22:843–50. [PubMed]
14. Seppenwoolde JH, Viergever MA, Bakker CJ. Passive tracking exploiting local signal conservation: the white marker phenomenon. Magn Reson Med. 2003:784–90. [PubMed]
15. Cunningham CH, et al. Positive contrast magnetic resonance imaging of cells labeled with magnetic nanoparticles. Magn Reson Med. 2005;53:999–1005. [PubMed]
16. Zhou R, et al. SWIFT detection of SPIO-labeled stem cells grafted in the myocardium. Magn Reson Med. 2010;63:1154–61. [PMC free article] [PubMed]
17. Gilad AA, et al. MR tracking of transplanted cells with “positive contrast” using manganese oxide nanoparticles. Magn Reson Med. 2008;60:1–7. [PMC free article] [PubMed]
18. Mani V, et al. Gradient echo acquisition for superparamagnetic particles with positive contrast (GRASP): sequence characterization in membrane and glass superparamagnetic iron oxide phantoms at 1.5T and 3T. Magn Reson Med. 2006;55:126–35. [PubMed]
19. Zurkiya O, Hu X. Off-resonance saturation as a means of generating contrast with superparamagnetic nanoparticles. Magn Reson Med. 2006;56:726–732. [PubMed]
20. Ren J, et al. SPIO MRI Contrast Can Be Enhanced by Off-resonance Saturation. The Fifth Annual Meeting of the Society for Molecular Imaging; 2006; Waikoloa, Hawaii, USA.
21. Khemtong C, et al. In vivo Off-Resonance Saturation Magnetic Resonance Imaging of alpha(v)beta(3)-Targeted Superparamagnetic Nanoparticies. Cancer Res. 2009;69:1651–1658. [PMC free article] [PubMed]
22. Grad J, Bryant RG. Nuclear Magnetic Cross-Relaxation Spectroscopy. J Magn Reson. 1990;90:1–8.
23. Sun SH, et al. Monodisperse MFe2O4 (M = Fe, Co, Mn) nanoparticles. J Am Chem Soc. 2004;126:273–279. [PubMed]
24. Dubertret B, et al. In vivo imaging of quantum dots encapsulated in phospholipid micelles. Science. 2002;298:1759–1762. [PubMed]
25. Sutton D, et al. Doxorubicin and beta-lapachone release and interaction with micellar core materials: experiment and modeling. Exp Biol Med. 2007;232:1090–9. [PubMed]
26. LaConte LEW, et al. Coating thickness of magnetic iron oxide nanoparticles affects R-2 relaxivity. J Magn Reson Imaging. 2007;26:1634–1641. [PubMed]
27. Duan HW, et al. Reexamining the effects of particle size and surface chemistry on the magnetic properties of iron oxide nanocrystals: New insights into spin disorder and proton relaxivity. J Phys Chem C. 2008;112:8127–8131.
28. Weissleder R, et al. Superparamagnetic iron oxide: pharmacokinetics and toxicity. AJR Am J Roentgenol. 1989;152:167–73. [PubMed]
29. Barcena C, et al. Zinc ferrite nanoparticles as MRI contrast agents. Chem Commun. 2008;19:2224–2226. [PubMed]
30. Ai H, et al. Magnetite-loaded polymeric micelles as ultrasensitive magnetic-resonance probes. Adv Mater. 2005;17:1949–1952.
31. Sutton D, et al. Functionalized micellar systems for cancer targeted drug delivery. Pharm Res. 2007;24:1029–46. [PubMed]
32. Bottomley PA, et al. Estimating radiofrequency power deposition in body NMR imaging. Magn Reson Med. 1985;2:336–49. [PubMed]
33. Hoult DI, Chen CN, Sank VJ. The field dependence of NMR imaging. II. Arguments concerning an optimal field strength. Magn Reson Med. 1986;3:730–46. [PubMed]
34. Uematsu H, et al. A direct comparison of signal behavior between 4.0 and 1.5 T: a phantom study. Eur J Radiol. 2003;45:154–159. [PubMed]
35. Viswanathan S, et al. Alternatives to gadolinium-based metal chelates for magnetic resonance imaging. Chem Reviews. 2010;110:2960–3018. [PMC free article] [PubMed]
36. Lipinski MJ, et al. MRI to detect atherosclerosis with gadolinium-containing immunomicelles targeting the macrophage scavenger receptor. Magn Reson Med. 2006;56:601–10. [PubMed]
37. Nahrendorf M, et al. Noninvasive vascular cell adhesion molecule-1 imaging identifies inflammatory activation of cells in atherosclerosis. Circulation. 2006;114:1504–11. [PubMed]