EMIM 2018 ControlCenter

Online Program Overview Session: PS-19

To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).

New Probes | Synthesis and Enabling Technologies

Session chair: Peter Brust - Leipzig, Germany; Daniel Jirak - Prague, Czech Repubic
Shortcut: PS-19
Date: Thursday, 22 March, 2018, 4:00 PM
Room: Lecture Room 03 | level -1
Session type: Parallel Session


Click on an contribution preview the abstract content.

4:00 PM PS-19-1

Introductory Talk by Eric Ahrens - La Jolla, USA

This talk provides an overview of state-of-the-art research and refers to the following presentations selected from abstract submissions.

4:20 PM PS-19-2

General in situ activation of [11C]CH3I to form 11C–C bonds in Negishi coupling reaction – proof of concept (#158)

L. Rejc1, N. Alonso2, V. Gómez-Vallejo1, J. Alcazar2, F. P. Cossio3, J. I. Andres2, J. Llop1

1 CIC biomaGUNE, Radichemistry and Nuclear Imaging, San Sebastian, Gipuzkoa, Spain
2 Janssen Research & Development, Lead Discovery Chemistry, Toledo, Spain
3 University of Basque Country, Department of Organic Chemistry I, San Sebastian, Gipuzkoa, Spain


The most common method for labelling with carbon-11 (11C), the use of [11C]CH3I, is limited to the production of [11C]methoxides, [11C]methylamines, and [11C]methylthioesters. The development of metal mediated reactions is expanding 11C-labelling options,1 however, the main limitation remains the need for specific precursors, which are challenging to prepare, require careful handling, and are often toxic. To avoid these limitations, we designed a general and efficient one-pot Negishi reaction via in situ formation of [11C]CH3ZnI to produce a variety of [11C]methylaryls and [11C]thymidine.


A series of aryl halides, obtained from Sigma Aldrich, was tested to investigate the generality of one-pot two-step Negishi reaction. [11C]CH3I was produced by iodination of cyclotron-produced [11C]CH4 using a recirculation system. [11C]CH3I was distilled into a zinc-filled cartridge,2 preloaded with a solution of an aryl halide and a mediator tetrakistriphenylphosphine palladium (0). The success of the reaction was evaluated by radio-HPLC analysis of the reaction mixture immediately after reaction. In case of [11C]thymidine, commercially available 5-iodo-2-deoxyuridine was used as the substrate and the crude reaction mixture was submitted to purification by semi-preparative HPLC, to produce pure [11C]thymidine, appropriate for use in animal studies.


Close-to-quantitative (>90%) trapping of [11C]CH3I in the zinc cartridge was achieved. [11C]CH3I underwent fast and quantitative reaction with zinc to form [11C]CH3ZnI. The latter was indirectly identified by the detection of [11C]CH4 in the radio-HPLC chromatogram and reaction kinetics studies. Subsequent Negishi coupling reactions in the presence of transition metal mediator were successful with a variety of aryl halides and 5-iodo-2-deoxyuridine (Figure 1, Table 1). The electron donating and electron withdrawing effects of aryl substituents narrate the reaction rate according to their reactivity;3 hence higher temperatures or the use of different metal complexes might be required in specific cases to achieve close-to-quantitative conversion of [11C]CH3I into the product


[11C]CH3I can efficiently be activated with zinc in situ followed by a fast Negishi coupling reaction with a wide variety of substrates. As a proof of concept, unprecedented synthesis of [11C]thymidine was chosen. Successful radiolabelling and isolation of this proliferation biomarker candidate proves the potential of this easy, general, and fast method to become an alternative for the production of new 11C-labelled imaging probes and further expand the possibilities for evolution of PET imaging.


  1. a) S. Kealey, J. Passchier, M. Huiban, Chem. Commun. 2013, 49, 11326-11328; b) K. Dahl, C. Halldin, M. Schou, Clinical and Translational Imaging 2017, 5, 275-289.
  2. N. Alonso, L. Z. Miller, J. M. Muñoz, J. Alcazar, D. T. McQuade, Adv. Synth. Catal. 2014, 356, 3737-3741.
  3. Z.-B. Dong, G. Manolikakes, L. Shi, P. Knochel, H. Mayr, Chem. Eur. J. 2010, 16, 248–253
Figure 1
Reaction for [11C]methylaryl synthesis using Negishi cross-coupling reaction, via in situ formation of 11CH3ZnI.
Table 1

Conversion values obtained for different aryl halides and triflates and 5-iodo-3-deoxyuridine. In all cases, T= 70°C, t=5 min; conversion calculated as the ratio between the area of the peak corresponding to 11C-labelled compound and the sum of the areas of all the peaks in the radiochromatogram.

Keywords: carbon-11, Negishi reaction, [11C]thyimidine
4:30 PM PS-19-3

A new generic platform for PET imaging based on 68Ga-labeled Pept-insTM: from infectious disease to cancer imaging (#166)

M. Siemons1, 2, L. Khodaparast2, L. Khodaparast2, J. Lecina1, F. Claes2, R. Gallardo2, M. Ramakers2, F. Rousseau2, J. Schymkowitz2, G. Bormans1

1 Laboratory for Radiopharmaceutical Research, KU Leuven, Department of Pharmaceutical and Pharmacological Sciences, Leuven, Belgium
2 Switch Laboratory, VIB‐KU Leuven Center for Brain & Disease Research, Leuven, Belgium


The Pept-in technology is based on a highly specific beta-sheet aggregation of a target protein, induced by short amyloidogenic peptides called Pept-ins (7-20 AA). Pept-In sequences are derived from aggregation-prone regions present in the target protein. They provide a generic platform for diagnostics similar to antibodies and derivatives. However, the rapid production by solid-phase peptide synthesis results in lower cost for production and a higher production throughput with lower batch-to-batch variation compared to antibodies and derivatives.


Gallardo et al. (2016, Science) designed vascin which inhibits VEGFR2 function by inducing aggregation of this receptor with high specificity. Furthermore, Khodaparast et al. (unpublished data) showed that Colpeptin1 induced proteostatic collapse in E.coli leading to bacterial cell death. Vascin and Colpeptin1 were modified for 68Ga-labeling by synthesis of [68Ga]Ga-NODAGA-PEG4-vascin and [68Ga]Ga-NODAGA-PEG2-Colpeptin1. Since proline is a known beta-sheet structure breaker, proline-mutated variants were synthesized as controls to obtain [68Ga]Ga-NODAGA-PEG4-vascin(Pro) and [68Ga]Ga-NODAGA-PEG2-Colpeptin1(Pro). Dynamic µPET imaging and biodistribution studies were performed in a mouse melanoma tumor model for vascin and an E. coli muscle infection model for radiolabeled Colpeptin1.


Radiolabeled vascin and Colpeptin1 provided good target-to-background contrast in PET images with high specificity, since for vascin tumor-to-muscle SUV ratios (T/M SUVR) were significantly higher than for its proline-mutated variant (Fig. 1a) and the same observation was made for Colpeptin1 muscle-to-muscle SUV ratios (muscle SUVR) compared to its proline-mutated variant (Fig. 1b). The possibility of the aspecific EPR effect being responsible for in vivo accumulation through self-aggregates, was countered by injection of radiolabeled Colpeptin1 into the melanoma tumor model (Fig. 1a), confirming that the presence of the target protein is required for retention of the radiolabeled Pept-in. Furthermore, radiolabeled Colpeptin1 did not show accumulation at sites of LPS-induced sterile inflammation nor in inactive E. coli infected muscle. Biodistribution studies proved favorable pharmacokinetic properties for both radiolabeled vascin and Colpeptin1.


The excellent in vivo specificity of Pept-ins observed in two totally different disease models, i.e. cancer and infectious disease, highlights the immense potential of a platform based on this technology for diagnostic and therapeutic applications.

Specific in vivo accumulation of 68Ga-labeled Pept-ins at their target site

(a) PET image in a mouse melanoma tumor model. PET T/M SUVR bar plots of radiolabeled vascin (n= 8), vascin(Pro) (n= 5) and Colpeptin1 (n= 4). (b) PET image in a foreleg muscle infection model. PET muscle SUVR of radiolabeled Colpeptin1 (n= 5) and Colpeptin1(Pro) (n= 6). Data are expressed as mean ± SD. Statistical significance was calculated by using one-way ANOVA or unpaired student’s t-test.

4:40 PM PS-19-4

Small CaF2 nanocrystals as nano-sized tracers for in vivo 19F-MRI (#179)

I. Ashur1, H. Allouche-Arnon1, A. Bar-Shir1

1 The Weizmann Institute of Science, Department of Organic Chemistry, Rehovot, Israel


Fluorine-19 MRI agents are widely used as tracers for cellular imaging. In recent years, perfluorocarbon (PFC) nanoemulsions have been used successfully as 19F-tracers in various applications1-4 for preclinical studies and clinical practices. However, PFCs lack the merits of inorganic nanocrystals (small size, dense material, high crystallinity, surface modifiability, etc.). Here we show that small (8 nm) water-soluble CaFnanoparticles (NPs) allow to obtain high-resolution 19F-NMR signals. This makes them suitable as 19F-MRI tracers and allows a "hot-spot" display. 


Synthesis and characterization: PEGylated CaF2 nanofluorides (CaF2-PEG, Fig. 1) were synthesized using a solvothermal method. The PEGylated NPs were fluorescently labeled by FITC and SCY3-based fluorophores and fully characterized.

19F-NMR and 19F-MRI: High-resolution 19F-NMR spectra collected with a 9.4 T NMR spectrometer. T1 and T2 relaxation times of the NPs 19F were determined. In vitro and in vivo MRI performed on vertical 9.4 T MRI scanner. 1H-MRI: A FLASH sequence with TR/TE=360/4 ms. 19F-MRI: 3D ultrashort TE protocol  (3D-UTE) with a flip angle of 10o, TR/TE=150/0.02 ms, FOV=3.2×3.2×3.2 cm3, matrix=32×32×32, and NA=8. Mice were immunized by footpad injection of immunogenic emulsion. MRI data were collected 10 days post immunization and after NPs injection to the footpad.


Figure 1 shows that PEGylated CaF2 NPs (Fig. 1a) are crystalline and small (Fig. 1b) with overtime colloidal stability (Fig. 1c). Figure 1d shows that small CaF2 NPs could be detected with high-resolution 19F-NMR. The ability to monitor the NPs with 19F-MRI was demonstrated on a phantom (Fig. 1e). A clear 19F-MR signal of NPs was acquired by using UTE sequence, overcoming their relative short T2 and allowing a “hot-spot” display. The performance of our newly proposed NPs as imaging tracers for in vivo 19F-MRI was tested in mice model of inflammation. The immunized mice were subjected to 1H- and 19F-MRI that acquired pre- and post- injection of fluorescently labeled PEGylated CaF2 NPs that preserved their 19F-NMR properties (Fig 2a). By using 19F-UTE-MRI, a clear 19F-signal was observed at the popliteal lymph node ROI (LN, Fig 2b-c, 2 repsentative mice). FACS analysis of excised cells from the LN revealed that the majority of the NPs were accumulated in macrophages and dendritic cells.


We demonstrated, for the first time, that small fluoride-based nanocrystals (specifically, CaF2 NPs) freely tumbling in solution can be studied while in solution with high-resolution 19F-NMR and be used as nano-tracers for 19F-MRI. The proposed nanocrystals elucidate a novel type of 19F-tracers that combine the advantages of using nanocrystals (small, high 19F-equivalency, maximal 19F-density, and surface modifiability) with the merits of 19F-MRI tracers.


  1. Ahrens, E.T., Flores, R., Xu, H. & Morel, P.A. In vivo imaging platform for tracking immunotherapeutic cells. Nat Biotechnol 23, 983-987 (2005).
  2. Flogel, U. et al. In vivo monitoring of inflammation after cardiac and cerebral ischemia by fluorine magnetic resonance imaging. Circulation 118, 140-148 (2008).
  3. Janjic, J.M., Srinivas, M., Kadayakkara, D.K. & Ahrens, E.T. Self-delivering nanoemulsions for dual fluorine-19 MRI and fluorescence detection. J Am Chem Soc 130, 2832-2841 (2008).
  4. Ruiz-Cabello, J. et al. In vivo "hot spot" MR imaging of neural stem cells using fluorinated nanoparticles. Magn Reson Med 60, 1506-1511 (2008).
  5. Ahrens, E.T., Helfer, B.M., O'Hanlon, C.F. & Schirda, C. Clinical cell therapy imaging using a perfluorocarbon tracer and fluorine-19 MRI. Magn Reson Med 72, 1696-1701 (2014).
Water soluble PEGylated CaF2 NPs.

a, Schematics of PEGylated CaF2 NPs. b, TEM images of the NPs. c, DLS of the NPs after purification (red) and 40 days later (blue). d, High-resolution 19F-NMR of the NPs. e, MRI of phantom containing solutions with or without the NPs in two concentrations (1H-MRI, top panel). Middle panel: 19F-MRI acquired by 3D-UTE sequence (TR/TE=150/0.02 ms). Bottom panel: “hot spot” display of the 19F-data.

In vivo imaging of fluorescently-labeled PEGylated CaF2 NPs in a model of inflammation.
a, Fluorescently labeled PEGylated NPs, their fluorescent properties, and high-resolution 19F-NMR. b, Anatomical 1H-MRI of studied mice (left); matched 19F-MRI (middle) and pseudo-color maps of 19F-MRI data on the 1H-MRI (right). c, FACS analysis for specific cell populations (NPs in red, and PBS in black). 
4:50 PM PS-19-5

A Design of Experiments (DoE) Approach Towards the Optimization of Copper-Mediated Radiofluorination Reactions of Arylstannanes. (#227)

G. D. Bowden1, A. Maurer1, B. J. Pichler1

1 Eberhard Karls University of Tübingen, Werner Siemens Imaging Center, Tübingen, Baden-Württemberg, Germany


Copper-mediated radiofluorination reactions (CMRFs) are multicomponent reactions that provide access to previously hard to synthesize tracers.1 They involve many factors that affect radiochemical conversion (RCC), which makes reaction optimization difficult and time-consuming. Unlike the traditional “One Variable at a Time” method, DoE is a statistical approach to reaction optimization that models and maps reaction space across multiple factors simultaneously.2 This study aimed to determine the usefulness of DoE to radiochemical reaction development and optimization using a model CMRF. (Fig 1)


A fractional factorial (Res V+) design was constructed using MODDE Go to model the effects of temperature, solvent volume, catalyst loading, ligand loading, and atmosphere on RCC. 24 runs were performed in 4 blocks of 6, which were included as blocking factors to account for variances in 18F processing. [18F]Fluoride was trapped on QMA resin and eluted with 50 µg K2CO3, 10 mg KOTf in 550 µl H2O.1 Aliquots (250-300 MBq) were transferred to 6 vials and dried azeotropically with acetonitrile under a stream of argon. Pre-made reaction mixtures with BiPhSnBu3 (4,5 nmol) were then added and left to stir under the required atmospheric and temperature conditions for 10 min. Reactions were quenched with 1 ml of H2O. RCCs were assessed using radioTLC and product identity was confirmed using HPLC.


The obtained RCCs were fitted to a model in MODDE Go using multiple linear regression (MLR). The summary of fit statistics (R2 = 0,88 (observed % variance), Q2 = 0,65 (predicted % variance)) indicated a valid model. The factor screening study indicated that temperature and solvent volume did not have a significant effect on RCC, while catalyst and ligand loading were significant factors. Interestingly, argon atmospheres had a slightly positive yet non-significant influence on RCC, a result contrary to the literature-published protocols which perform these reactions under air. There were no significant differences between experimental blocks. These results were used to design a more detailed Response Surface Optimization (RSO) study for a new tracer currently in development. RSO modelling of the results of this study (17 runs) showed a quadratic function for catalyst and ligand loading factors as well as an interaction between ligand loading and substrate concentration. (Fig 2)


DoE is a powerful approach to radiochemical reaction optimization that has the potential to reveal new insights into the behaviour of new 18F chemistry. It was used to show which experimental factors were most important to a model CMRF. It was also used to optimize and predict the experimental factors that will aid in the development of an improved synthesis for a novel tracer currently under development. Further work is currently being done to apply this approach to a number of novel tracers that have up until now suffered from low yields and poor synthesis performance.


1.        Makaravage, K. J., Brooks, A. F., Mossine, A. V., Sanford, M. S. & Scott, P. J. H. H. Copper-Mediated Radiofluorination of Arylstannanes with [18F]KF. Org. Lett. 18, 5440–5443 (2016).

2.        Murray, P. M. et al. The application of design of experiments (DoE) reaction optimisation and solvent selection in the development of new synthetic chemistry. Org. Biomol. Chem. 14, 2373–2384 (2016).

Figure 1:
The general reaction scheme explored by DoE. The factors below the arrow were investigated in the initial factor screening study: Catalyst loading, 2) Ligand loading, 3) Solvent volume, 4) Temperature and 5) Atmosphere. The range of the investigated factors are displayed in brackets.
Figure 2:
The reaction response surface generated by the Response Surface Optimization study revealed a quadratic relationship between catalyst and ligand loading. From this model, optimal reaction conditions can be estimated.
Keywords: Radiochemistry, [18F]Fluoride, Design of Experiments, Tracer Development
5:00 PM PS-19-6

Bioengineered bacterial encapsulins for multiscale in vivo molecular imaging (#295)

F. Sigmund1, 2, C. Massner1, 2, 4, P. Erdmann6, A. Stelzl1, 2, H. Rolbieski1, 2, H. Fuchs3, M. Hrabe de Angelis3, M. Desai7, S. Bricault7, A. Jasanoff7, V. Ntziachristos1, 5, J. Plitzko6, G. G. Westmeyer1, 2, 4

1 Helmholtz Zentrum München, Institute of Biological and Medical Imaging, Neuherberg, Bavaria, Germany
2 Helmholtz Zentrum München, Institute of Developmental Genetics, Neuherberg, Bavaria, Germany
3 Helmholtz Zentrum München, Institute of Experimental Genetics, Neuherberg, Bavaria, Germany
4 Technical University of Munich, Department of Nuclear Medicine, Munich, Bavaria, Germany
5 Technical University of Munich, Chair for Biological Imaging, Munich, Bavaria, Germany
6 Max Planck Institute of Biochemistry, Department of Molecular Structural Biology, Martinsried, Bavaria, Germany
7 Massachusetts Institute of Technology, Department of Biological Engineering, Camebridge, Massachusetts, United States of America


Compartmentalization, the spatial separation of processes into closed subspaces, is an important biological principle that enables the maintenance of specific local conditions which facilitate reactions and interactions in confined environments [1]. Whereas eukaryotic organisms usually employ membrane-enclosed organelles to achieve compartmentalization, bacteria have evolved proteinaceous nanocompartments that can spatially confine chemical reactions and sequester metals. Herein, we express bacterial encapsulins in mammalian systems for multimodal molecular imaging across different scales.


We expressed the encapsulin shell protein of Myxococcus xanthus [2], its native cargo proteins and a set of engineered cargo molecules in HEK293T cells and murine brains. We characterized their expression, self-assembly and cargo loading of the shell using co-Immunoprecipitation (Co-IP), Blue-Native-PAGE (BN-PAGE), and gel-based staining methods as well as fluorescence and bioluminescence imaging. Furthermore, we imaged HEK293T cells expressing melanin-producing encapsulins using multispectral optoacoustic tomography (MSOT). We conducted relaxometry measurements of cells expressing iron-loaded encapsulins and in vivo MR imaging in rat brains of xenografted cells at 7 Tesla. Finally, we imaged encapsulins in HEK293 cells using cryo-electron tomography (cryo-ET).


We demonstrate that eukaryotically expressed encapsulins not only auto-assemble into 32 nm large nanospheres at high density and without toxic effects but that self-targeting and encapsulation of cargo molecules still efficiently occur in mammalian cells (Figure 1 a). We furthermore show localized enzymatic reactions inside the nanocompartment useful for optical and optoacoustic imaging (Figure 1 b), as well as confined iron biomineralization that labels cells for detection by MRI in vivo in mouse brains (Figure 1 c) and electron microscopy (EM) (Figure 1 d), demonstrating the potential of encapsulins as genetic markers across modalities. In addition, the iron-sequestration inside the nanoshells affords magnetic manipulation of cells genetically labelled with encapsulins (Figure 1 e).


In conclusion, we have introduced encapsulins as a versatile method to genetically engineer orthogonal compartments in mammalian cells that can self-convert into nanomaterials with attractive properties. These features offer applications in molecular imaging across multiple scales from subcellular resolution by EM, via fluorescence to non-invasive optoacoustic imaging and in vivo MRI. Genetically controlled encapsulation of multi-component processes in eukaryotic cells is a fundamental biotechnological capability with important implications for mammalian cell engineering and cellular therapy.


[1] DeLoache, W. C. & Dueber, J. E. Compartmentalizing metabolic pathways in organelles. Nat. Biotechnol. 31, 320–321 (2013).

[2] McHugh, C. A. et al. A virus capsid-like nanocompartment that stores iron and protects bacteria from oxidative stress. EMBO J. 33,​ 1896–1911 (2014).


We are grateful for support from the European Research Council under grant agreements ERC-StG: 311552 (F.S., A.S., H.R., G.G.W.).

Figure 1 a-b
(a) Encapsulins express and auto-assemble in mammalian cells as shown by BN-PAGE and electron microscopy. Cargo targeting was confirmed via Co-IP and subsequent silver-stained SDS-PAGE. Encapsulin over-expression did not cause reduced cell viability. (b) Targeting of engineered cargo enzymes as shown on BN-PAGE. Targeting of tyrosinase yields nanomelanosomes that can be dected via optoacoustics. 
Figure 1 c-e
(c) Encapsulins can effectively load iron inside mammalian cells as shown by Prussian-Blue staining on BN-PAGE and can be detected via MRI after xenografting into rat brains. (d) Encapsulins can serve as markers for cryo electron tomography (cryoET). (e) Cells expressing iron-loaded encapsulins can be magnetically sorted. 
Keywords: Reporter genes, Synthetic biology, Nanocompartments, MRI, Electron microscopy, Encapsulins
5:10 PM PS-19-7

Using Magnetic Relaxation for Color Magnetic Particle Imaging: First In Vivo Experimental Results (#410)

D. Hensley1, 2, Z. W. Tay1, X. Y. Zhou1, P. Chandrasekharan1, B. Zheng1, P. W. Goodwill2, S. M. Conolly1, 3

1 University of California, Berkeley, Bioengineering, Berkeley, California, United States of America
2 Magnetic Insight, Inc., Alameda, California, United States of America
3 University of California, Berkeley, Electrical Engineering and Computer Sciences, Berkeley, California, United States of America


Reporting physiologic contrast is of great interest to magnetic particle imaging (MPI) [1, 2], especially as the field explores molecular imaging applications. One avenue to such contrast is through the rich dynamic magnetic physics of the tracers used in MPI [3]. Signal is generated in MPI when magnetic tracers rotate in response to movement of a sensitive magnetic field region. The non-instantaneous response of the tracers is termed magnetic relaxation. A canonical way of reporting such information is via color images [4,5]. Here we show the first in vivo color MPI results.


0.229 mg of Chemicell tracer was administered to a 0.2 kg female Fisher rat via tail vein injection. This was followed by an injection of 875,000 (0.08 mg Fe) macroaggregated albumin and MPI tracer conjugates [6] using the Perimag MPI tracer (MAA-perimag). Two reference markers (2 uL each) were placed above the rat’s chest. The rat was then scanned at four excitation amplitudes of 10, 15, 17.5, and 20 mT using our in house MPI scanner [7] (4 x 3.75 x 10 cm FOV with respiratory gating). X-ray imaging was performed with a Kubtec Xpert 40. The Chemicell tracer clears to the liver while the MAA-perimag gets stuck in the lungs, allowing us to test our color algorithm’s ability to disambiguate the two organs. Fig. 1 describes the color MPI algorithm we applied to the data.


Fig. 1 (f) shows imaging results and ROI quantification of our color algorithm applied to a known vial phantom, demonstrating that our algorithm is able to identify, disambiguate, and quantify co-localized tracers. Fig. 2 is the first demonstration of color MPI in vivo with Fig. 2 (a) showing MPI coronal slices, including a standard x-space MPI image, separate MAA-perimag (Tracer 1) and Chemicell (Tracer 2) images provides by the color algorithm, and combined colorized images overlaid with an X-ray reference. Fig. 2 (b—e) show different slices of the 3D colorized data set, further demonstrating the ability of the algorithm to parse the lungs and liver, each tagged with a different tracer. We see that the color MPI algorithm successfully parses the MAA-perimag tracer in the lungs from the Chemicell tracer in the liver, converting a standard MPI image with no discernible distinction between the lungs and liver into a colorized image with clear distinction between these organs.


We have reported the first in vivo color MPI results. These and other recent results in the field are beginning to paint a clear picture of how the rich relaxation physics of MPI tracers can enable new molecular imaging applications. More fully exploiting these physics will require improvements in areas such as tailored tracer design, how we encode relaxation information, and how we formulate reconstruction algorithms. Longer term, these approaches may enable applications such as monitoring of cell metabolism, targeted binding contrast, and imaging local micro-environmental conditions.


[1]       B. Gleich and J. Weizenecker. Tomographic imaging using the nonlinear response of magnetic particles. Nature, 435(7046):1214-1217, 2005. doi: 10.1038/nature03808.

[2]       P.W. Goodwill and S. M. Conolly. The X-space formulation of the magnetic particle imaging process: 1-D signal, resolution, bandwidth, SNR, SAR, and magnetostimulation. IEEE transactions on medical imaging, 29.11: 1851–1859, 2010.

[3]       L. R. Croft, et al. Low drive field amplitude for improved image resolution in magnetic particle imaging. Medical physics, 43.1: 424–435, 2016.

[4]       J. Rahmer, A. Halkola, B. Gleich, I. Schmale, and J. Borgert. First experimental evidence of the feasibility of multi-color magnetic particle imaging. Physics in medicine and biology, 60(5), 1775, 2015.

[5]       D. Hensley, P. Goodwill, L. Croft, and S. Conolly. Preliminary experimental x-space color MPI. Magnetic particle imaging (IWMPI), 2015 5th international workshop on, 2015.

[6]       X.Y. Zhou, et al. First in vivo magnetic particle imaging of lung perfusion in rats. Physics in Medicine and Biology, 62(9), 3510, 2017.

[7]       P. Goodwill, K. Lu, B. Zheng, and S. Conolly An x-space magnetic particle imaging scanner. Review of Scientific Instruments, 83(3), 033708, 2012.


We would like to acknowledge funding support from NIH 5R01EB019458-03, NIH 5R24MH106053-03, UC Discovery Grant 29623, W. M. Keck Foundation Grant 009323, and NSF GRFP.

Color Magnetic Particle Imaging Approach

(a) Multiple MPI scans are taken, each with a different excitation amplitude. (b,c) Tracers exhibit different image domain behavior at different excitation amplitudes per their relaxation properties. (d,e) Colorized images are produced by solving a pixel-wise inverse problem. (f) Initial vial phantom data and ROI quantification to test performance, especially unmixing of co-localized tracer.

First In Vivo Color MPI Results

(a) Standard and Color MPI coronal slice images. Separate MAA-perimag (Tracer 1) and Chemicell (Tracer 2) images obtained from the color algorithm are shown along with a combined image overlaid with X-ray reference. (b—e) Color images from different slices of the 3D dataset further demonstrating the separation of the lungs from the liver, which is not possible in the standard MPI images.

Keywords: magnetic particle imaging, molecular imaging, color MPI, image contrast, image reconstruction
5:20 PM PS-19-8

Chemical design of innovative LnIII-chelates for T1 and CEST MRI applications (#481)

L. Tei2, L. Leone2, G. Ferrauto1, 4, D. Delli Castelli1, 4, Z. Baranyai3, M. Botta2

1 Università di Torino, Department of Molecular Biotechnology for Health Sciences, Torino, Italy
2 Università del Piemonte Orientale, Dipartimento di Scienze e Innovazione Tecnologica, Alessandria, Italy
3 University of Debrecen, Department of Inorganic and Analytical Chemistry, Debrecen, Hungary
4 University of Torino, Molecular Imaging Center, Torino, Italy


Ln(HPDO3A) complexes (Fig 1A) have been useful MRI diagnostic tools for several years: the Gd-complex is the MRI agent ProHance, whereas Yb(HPDO3A) has been used as paraCEST agent for cell labelling and for pH and temperature mapping.[1] The presence of different exchanging isomers in solution (square, SAP, and twisted square antiprismatic, TSAP) explains the relaxometric and CEST behaviour of the Gd and Yb complexes, respectively.[2] The two coordination isomers of Yb(HPDO3A) allow a ratiometric method for pH mapping but reduce the signal achievable thus hampering its use for cell labelling.


HPDO3A was modified with the aim to increase the population of the fast (water) exchanging coordination isomer (TSAP). Thus, we designed a new polyaminocarboxylic macrocyclic ligand where the three acetic pendant arms of HPDO3A were substituted by three methyl acetic arms (HPDO3MA, Fig 1A). A complete 1H and 17O NMR relaxometric study on the GdIII complex, a detailed CEST characterization and a dissociation kinetic study on the YbIII complex were carried out. Finally, cell labelling experiments with the YbIII complex were performed and compared to Yb(HPDO3A) to show the better performances of the novel CEST agent.


The synthesis started by the preparation of the tris methyl acetic substituted macrocycle followed by the insertion of the hydroxypropyl pendant arm by reaction with 2-(R)-propylene oxide. 1H HR-NMR analyses revealed the presence of different isomeric species in solution for the EuIII complex whereas the YbIII complex showed only the TSAP isomer. Thus, Yb(HPDO3MA) shows only one CEST peak at 126 ppm with ST of 17.2%, corresponding to the sum of the ST effects of Yb(HPDO3A) SAP and TSAP isomers (Fig 1B). This allowed the labelling of TS/A cells with a CEST effect of 15%, almost double than that obtained for Yb(HPDO3A). Moreover, kinetic studies on Yb(HPDO3MA) showed that its kinetic inertness is about twice as high that of Yb(HPDO3A) (t1/2 = 2.2×107 h vs 1.2×107 h). Finally, the relaxivity of Gd(HPDO3MA) at 20 MHZ and 298 K is 5.1 mM-1 s-1 and analysis of NMRD profiles and 17O NMR data showed the occurrence of a faster water exchange rate than for Gd(HPDO3A) (kex = 3.3×107 s-1).


We have designed and characterized a new macrocyclic ligand able to form LnIII complexes with improved performance than LnHPDO3A. In case of the Gd-complex, a faster water exchange rate was determined, whereas for Yb(HPDO3MA) the presence of only one isomer in solution allowed to double the CEST effect in cell labelling experiments. Finally, the kinetic inertness of the latter complex, higher than Yb(HPDO3A), is promising for further application in molecular imaging protocols.


1. Magn. Reson. Med. 2013, 69, 1703; Angew. Chem., Int. Ed. 2011, 50, 1798; Magn. Reson. Med. 2014, 71, 326.

2. Inorg. Chem. 2013, 52, 7130

Figure 1

A: chemical structures of Ln(HPDO3A) and Ln(HPDO3MA); B and C: Z- and ST%-spectra of Yb(HPDO3A) and Yb(HPDO3MA)

Keywords: Lanthanides, paraCEST, kinetic inertness, relaxometry, cell labelling