Click on an contribution preview the abstract content.
Click on an contribution preview the abstract content.
Introductory Talk by Amnon Bar-Shir - Rehovot, Israel
This talk provides an overview of state-of-the-art research and refers to the following presentations selected from abstract submissions.
Unambiguous detection of atherosclerosis by pretargeted molecular imaging (#78)
I. Fernández-Barahona1, J. Pellico1, 2, J. Ruiz-Cabello2, 3, 4, F. Herranz1, 2
1 Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Madrid, Spain
Pretargeted imaging is based on the use of bioorthogonal tracers that selectively accumulate upon reaction with a pre-modified biomolecule in vivo. To date, this promising approach has been especially used for cancer diagnosis however, to our knowledge, it has not been applied to atherosclerosis. Here, we synthesized a bioorthogonal nano-radiotracer for in vivo pretargeted molecular imaging in a mouse model of atherosclerosis. We based this approach in our new platform for molecular imaging, 68Ga core-doped Fe2O3 nanoparticles, which provide PET and T1-MRI signals.
In the first place, TCO moiety was covalently attached to E-06, a naturally occurring mouse monoclonal IgM antibody targeting oxidized LDL, oxidized HDL and proteins covalently modified by oxidized phospholipids.
Secondly, we synthesized a new 68Ga-based nano-radiotracer (NRT). Citrate-coated extremely small iron oxide nanoparticles core-doped with 68Ga. NRT synthesis was carried out very rapidly in a microwave oven, taking only 15 minutes to obtain the pure, ready-to-use sample. We next incorporated the tetrazine (TZ) moiety to the NRT surface through amide formation between benzylamino-tetrazine and the carboxylic acid groups in citrate.
After full NRT characterization, in vivo imaging experiments were carried out in ApoE-/- and C57BL/6 mice. 24 hours after Ab injection, 68Ga-NRT-TZ was intravenously administered; some ApoE-/- mice were inyected with 68Ga-NRT as a bioorthogonal reaction control. PET/CT imaging of ApoE-/- mice 1 hour post 68Ga-NRT-TZ injection revealed specific localization in several consecutive planes of the aortic arch. In contrast, post 68Ga-NRT-TZ injection images from control mice showed no noticeable signal in the aortic arch nor in the whole aorta. PET/CT signal was also absent from ApoE-/- mice injected with unmodified 68Ga-NRT.
The convenient relaxometric values of 68Ga-NRT-TZ motivated us to check the ability of MRI to detect NRT accumulation in the aorta. Ex vivo T1-weighted MRI of aortas from ApoE-/- mice injected with 68Ga-NRT or 68Ga-NRT-TZ were acquired. Clear hyperintense areas were evident in lesions, showing higher intensity areas for the ApoE-/- mouse injected with 68Ga-NRT-TZ.
The production of a new kind of bioorthogonal nano-radiotracer through the synergistic combination of nanotechnology and radiochemistry proves the feasibility of in vivo pretargeted imaging. Our results demonstrate the ability of this approach to unambiguously detect atherosclerosis.
1. Pellico, J. et al. One-Step Fast Synthesis of Nanoparticles for MRI: Coating Chemistry as the Key Variable Determining Positive or Negative Contrast. Langmuir 33, 10239–10247 (2017). doi:10.1021/acs.langmuir.7b01759
2. Pellico, J. et al. In vivo imaging of lung inflammation with neutrophil-specific 68Ga nano-radiotracer. Sci. Rep. 7, 13242 (2017). doi:10.1038/s41598-017-12829-y
3. Pellico, J. et al. Fast synthesis and bioconjugation of 68 Ga core-doped extremely small iron oxide nanoparticles for PET/MR imaging. Contrast Media Mol. Imaging (2016). doi:10.1002/cmmi.1681
4. Pellico, J. et al. Hybrid pretargeted molecular imaging of atherosclerosis with bioorthogonal nano-radiotracers. Bioconjugate Chemistry. (under review)
This study was supported by a grant from the Spanish Ministry for Economy and Competitiveness (MEyC) (grant number: SAF2016-79593-P) and from Carlos III Health Research Institute (grant number: DTS16/00059). We thank Simon Bartlett for editorial assistance and manuscript preparation. The CNIC is supported by the Spanish Ministry of Economy, Industry and Competitiveness (MEIC) and the Pro CNIC Foundation, and is a Severo Ochoa Centre of Excellence (MEIC award SEV-2015-0505).
Specific quantification of 57Fe‑nanoparticles by combined MRI and mass spectrometric imaging (#146)
A. Beuker1, M. Masthoff1, R. Buchholz2, L. Wachsmuth1, F. Albers1, W. Heindel1, U. Karst2, M. Wildgruber1, C. Faber1
1 Translational Research Imaging Center, Institute for Clinical Radiology, University Hospital of Muenster, Muenster, North Rhine-Westphalia, Germany
Iron oxide nanoparticles (ION) are common contrast agents for (pre-)clinical MRI. However, MRI signal is always influenced by iron from endogenous sources. Hence, an exact correlation of administered ION with the MRI signal by T2 quantification is not possible in vivo. Here, we combine non-radioactive 57Fe-based ION MRI with laser-ablation-mass-spectrometry (LA-ICP-MS) for specific differentiation between endogenous iron (56Fe) and applied ION. We assess distribution of administered ION, apply 57Fe-ION for cell tracking and aim to correlate local ION concentration with T2 values.
Healthy C57BL/6 mice were i.v. injected with novel 57Fe-ION (NanoPET, Berlin; dose: 2,5 ml/kg; 16,7mg Fe/ml). After 2h, 1d, 3d, 7d and 30d p.i. MRI with T2/T2*-mapping of liver, spleen, kidneys and brain was performed on a 9,4T small-animal MRI. Mice were sacrificed and organs of interest were extracted for LA-ICP-MS to quantify the amount of 57Fe and the 56Fe/57Fe isotope ratio. Distribution of 57Fe-ION was validated by histology via mac-3 and Prussian blue staining, too.
57Fe-ION were also evaluated for cell tracking in a mouse model of local inflammation. C57BL/6 mice were injected s.c. with 100µl of a polyacrylamide-gel (pellet) to induce sterile inflammation. 57Fe-ION were injected i.v. after 4h, followed by MRI (1d, 3d, 7d p.i.) and LA-ICP-MS, now including the subcutaneous pellet.
Specific detection and quantification of 57Fe-ION by ex vivo LA-ICP-MS in liver, spleen and kidneys (see Fig.1 a-c) enabled for a particle distribution study. Comparing the detected 57Fe-signal with histological staining indicated that most ION were first internalised by tissue macrophages in liver and spleen. Over time, 57Fe-signal increased in the red pulp of the spleen but decreased in macrophages (see Fig.1 c-d).
Distribution of 57Fe-ION by LA-ICP-MS was correlated to organ specific T2 times by in vivo MRI (see Fig. 2). Observed signal decrease in liver, spleen, kidney and brain just p.i. was mainly due to intravascular accumulation of ION. While T2 of the brain was recovering quickly and no long-term changes could be detected, T2 of the liver and spleen remained low indicating storage of applied ION. T2 of kidneys slowly recovered, due to clearance of ION.
57Fe-ION were also detectable in the subcutaneous pellets, indicating migration of macrophages with internalised ION.
Using novel 57Fe-ION enables for specific detection and quantification of applied ION. Hereby, studying distribution and long-term fate of ION becomes feasible. From increasing 57Fe-signal in splenic red pulp, we assume 57Fe-iron is deposited in endogenous iron stores over time.
Additionally 57Fe-ION can specifically validate and quantify ION based cell tracking, shown by detecting internalised 57Fe-ION in inflammatory tissue.
Finally, our future aim is to correlate organ specific T2 times with the 57Fe-ION-amount and thus establish a specific validation of MR-iron quantification.
Fig. 1: Distribution maps for 57Fe in Laser-ablation-mass-spectrometry (LA-ICP-MS)
LA-ICP-MS of 10µm cryoslices was performed of explanted organs after in vivo intravenous application of 57Fe-ION. Exemplary semiquantitative distribution maps for 57Fe are shown for a) liver and b) kidney. Quantitative distribution maps for 57Fe in spleen are shown for c) 3d and d) 30d post injection. Laser spot size was 25µm.
Fig. 2: Time course of T2-relaxation times in different tissues
T2-relaxation times for liver, spleen, kidneys and brain at different time points after 57Fe-ION i.v. injection (2h, 1d, 3d, 7d, 30d). Naïve animals did not receive any injection. Data is presented as mean ± standard deviation (at least n=3 for each time point).
Keywords: ION, SPIO, MRI, T2-Mapping, cell-tracking, iron-quantification
Fishing for Aldehydes: Expanding the Probe Tackle Box with Conditional CEST-MRI Lures (#261)
M. Suchy1, T. Dang1, Y. Truong1, C. Lazurko1, W. Oakden3, W. Lam3, G. Facey1, G. Stanisz3, 4, A. Shuhendler1, 2
1 University of Ottawa, Chemistry & Biomolecular Sciences, Ottawa, Ontario, Canada
Aldehydes are regularly produced in cells through tightly regulated processes necessary for life. Low homeostatic concentrations are required for immune responses, genetic regulation, and signal transduction mechanisms . Cell stress can throw aldehyde levels into dysregulation, resulting in the initiation and progression of a variety of disease and injury states . While some aldehydes have been investigated for assessing injury  or cancer therapy response , for example, the use of reactive carbonyls as imaging biomarkers is infrequent and mostly limited to ex vivo detection.
Building upon previous work establishing rapid, catalyst-free trapping of aldehydes using N-amino anthranilic acids , we have developed activatable molecular MRI contrast agents providing imaging signal through chemical exchange saturation transfer magnetic resonance imaging (CEST-MRI) . These hydrazine-containing agents, collectively termed Hydrazo-CEST, are CEST-inactive until they trap reactive carbonyls to form hydrazones (Fig. 1a). Upon hydrazone formation, proton exchange from the ring-proximal nitrogen falls into the CEST regime and produces high CEST-MRI contrast (%MTRasym>20%).
Using a variety of control probes, we have identified that both the hydrazone proton and the ortho-carboxylic acid are necessary for CEST-MRI. The CEST-MRI signal was sensitive to the electronics of both hydrazine and carbonyl components: the more electron-withdrawing the combined substituents, the lower the contrast (Fig. 1b), which was correlated with alterations in proton exchange rates, deviating from that required for contrast production. Aldehyde trapping proceeded rapidly under physiological conditions (kobs~0.1 min-1) and to near completion (>90%) within 15 min (Fig. 1c). The ortho-carboxylic acid substantially increased the degree of reaction completion and the stability of the hydrazone product relative to control compounds lacking the ortho-carboxylic acid or containing an ortho-methyl ester. This effect derived from a key intramolecular hydrogen bond. Finally, we have verified that our N-amino anthranilic acid probes were biocompatible even at high doses (Fig. 1d).
The development of Hydrazo-CEST, comprised of a set of CEST-MRI probes conditionally activated when bound to bioactive carbonyls, and the rational investigation of the chemical determinants of CEST signal production, places the mapping of aldehydes in vivo by MRI within reach. These probes could provide novel diagnostic and prognostic strategies for diseases and injuries that are initiated and evolved through the biogenesis of aldehydes, including heart disease, neurodegeneration, and traumatic brain injury.
This work was supported by an NSERC Discovery Grant RGPIN 2015-05796 (A.J.S.), the Canada Research Chairs Program 950-230754 (A.J.S.), the Canadian Foundation for Innovation (A.J.S.), and the Canadian Institutes of Health Research PJT376892 (A.J.S.).
Keywords: Chemical Exchange Saturation Transfer, Contrast Agent, Aldehydes, Activatable Imaging Probe
Using single fluorinated agent for multiplexed imaging with 19F-CEST MRI (#337)
R. Shusterman-Krush1, L. Avram1, B. C. Gibb2, A. Bar-Shir1
1 Weizmann Institute of science, Rehovot, Israel
The complexity of biology attracts scientists from a wide range of fields. Although optical imaging sensors are widely used to study such complexity in a multicolor imaging fashion, their light signal source calls for alternatives. In MRI, diaCEST and paraCEST based probes have been used for multiplexed imaging by exploiting the Dw of a labile proton. Here, we show that combining the CEST approach with 19F-MRI allows obtaining “multicolor” imaging of multiple targets using a 19F-agent. Based on the distinct Dws obtained in 19F-MR, a novel platform for multiplexed MR imaging is presented.
All NMR and MRI experiments were performed on 9.4T scanners (Bruker, Germany). 19F-CEST NMR data were acquired as previously described. 1H-MRI was acquired using a FLASH sequence. For 19F-MRI, RARE sequence was used TR/TE=6000/3.64 ms; 10 mm slice thickness; FOV=3.2×3.2 cm2; matrix size=64×64. For the 19F-CEST MRI the frequency of the pre-saturation pulse (B1=3.6 μT/ 3000 ms) was swept from Δω=+5ppm to Δω=-5ppm in 100Hz steps.
In the present study, two supramolecular systems composed of molecular hosts (either cucurbit[n]uril, CB; or octa acid, OA, Fig. 1a) and 19F-guest (fluoroxene, Fig. 1a) were used. Initially, the 19F-CEST characteristics of CB:fluoroxene and OA:fluoroxene systems in PBS were studied (Fig. 1b). A large CEST effect was obtained (45-50% signal change) for both studied samples. While for the CB:fluoroxene system the effect was at Dw of +1.6 ppm (Fig. 1b, purple), the effect for OA:fluoroxene system was at Dw of -2.0 ppm (Fig. 1b, green). This observation encouraged us to use this platform for multiplexed 19F-MRI. A phantom of 4 different host:guest pairs (1:100) was prepared as shown in Fig. 2. 1H-MRI and 19F-MRI did not reveal any difference between the samples. However, by performing the 19F-CEST MRI experiment, a multicolor imaging characteristic was obtained. A pseudo-multicolor imaging map demonstrates the capability of using the proposed platform for multiplexed MR imaging.
We have demonstrated the feasibility of applying the 19F-CEST methodology on a variety of host:guest systems to amplify 19F-MR signals of low concentration targets (up to 500 nM of imaged target). By targeting the molecular hosts (CB or OA) to multiple low-concentration targets the proposed platform has the potentiality to monitor such targets, simultaneously, using single 19F-probe.
 G. Liu, M. Moake, Y.-e. Har-el, C. M. Long, K. W. Y. Chan, A. Cardona, M. Jamil, P. Walczak, A. A. Gilad, G. Sgouros, P. C. M. van Zijl, J. W. M. Bulte, M. T. McMahon, Magnetic Resonance in Medicine 2012.
 G. Ferrauto, D. D. Castelli, E. Terreno, S. Aime, Magnetic Resonance in Medicine 2013.
 L. Avram, M. A. Iron, A. Bar-Shir, Chemical Science 2016.
chemical structure and 19F-CEST NMR analysis
Figure 1: (a) Chemical structure of cucurbituril (CB), octa acid (OA), and Fluoroxene, followed by a schematic description of the multicolored host-guest probes system. (b) 19F-CEST-NMR characterization of the CB:fluoroxene (1:100 ratio) and OA:fluoroxene (1:100 ratio) systems.
Figure 2: (a) 1H-MRI and 19F-MRI of a phantom containing CB:fluoroxene (blue), CB:fluoroxene (magenta), β-cyclodextrin:fluoroxene (orange) and OA:fluoroxene (green) host-guest systems. (b) 19F-CEST map overlaid on 1H-MRI exhibiting significant effect for the CB:fluoroxene (Dw=+1.6 ppm) and OA:fluoroxene (Dw=-2.0 ppm) systems. (c) The MTRasym plots obtained from the map shown in b.
Keywords: molecular MRI, multicolored imaging, 19F-CEST MRI, host:guest
Extracellular glutamate and intracellular calcium recording with fiber optic and simultaneous fMRI (#386)
Y. Jiang1, X. Chen1, 2, P. Pais1, 3, X. Yu1
1 Max Planck Institute for Biological Cybernetics, Tübingen, Germany
Here, we expressed genetically encoded fluorescent reporter iGluSnFR1 for extracellular glutamate (Glu) sensing and genetically encoded calcium indicator GCaMP6f for calcium sensing in both neurons and astrocytes, and applied two channel fiber optic recording system in combination with blood oxygenation level-dependent signal (BOLD) fMRI. This platform offers us a more direct interpretation of neuronal transient with fMRI, thus, would expand our understanding of the signal propagation through the neuron-glia-vessel network couple to BOLD fMRI signals.
All images were acquired with a 14.1 T/26cm horizontal bore magnet (Magnex), interfaced to an AVANCE III console (Bruker) and equipped with a 12 cm gradient set, capable of providing 100 G/cm with a rise time of 150 us (Resonance Research). A transreceiver surface coil was used to acquire fMRI images. fMRI scans with block design were performed using 3D Echo planar imaging sequence: TR, 1.5 s, TE,11.5 ms, 1.92X1.92X1.92 cm3, FOV, 48X48X48 matrix, 400X400X400 um3 spatial resolution. The reporter iGluSnFR and GCaMP6f were expressed by AAV5 virus in the two hemisphere forepaw somatosensory cortex (FP-S1) with Syn or GFAP promoter. Fiber optic (200 mm) was inserted into the area which expressed the cortex for fluorescent signal recording.
Neuronal calcium and Glu signals with simultaneous fMRI from the FP-S1 of two hemispheres were acquired, respectively. Evoked neuronal calcium and Glu spikes were shown to follow each electrical pulses (Fig 1A), while the Glu spikes have earlier onset time and faster time to peak response in comparison with neuronal calcium. Also, amplitude of the evoked Glu spike increased proportionally to the amplitude of BOLD signals as the function of the stimulation intensity (Fig. 1B). The simultaneous fMRI BOLD maps and the time course of BOLD signal were shown (Fig. 1C).
Similar to previous study2, the astrocytic calcium signal is an integrated unitary spike, which has slower onset than the Glu spikes(Fig. 2A). Interestingly, we also observed the baseline drop of the Glu signal during the stimulation, which shows earlier onset with extended longer tail than the astrocytic signal. Also noteworthy is that the BOLD signals detected from both hemispheres are similar to each other(Fig.2B).
Concurrent glutamate and calcium recording was established with the BOLD fMRI brain mapping in anesthetized rats. This platform would lead to a better understanding of neurovascular coupling through the neuro-glial-vascular network in the animal brain. Future study will further clarify the neurovascular coupling events in the neuro-glial-vascular network and specify the source for the Glu baseline drop of during stimulation.
1. Marvin, J. S., et al. An optimized fluorescent probe for visualizing glutamate neurotransmission. Nat methods 10, 162-170, doi:10.1038/nmeth.2333 (2013).
2. Wang, M., He, Y., &Yu, X. 2017. A novel role of intrinsic astrocytic calcium spikes to mediate brain states through central/dorsal thalamic nuclei. ISMRM 2017.
Tuebingen Universtity and Graduate Training Center of Neuroscience Tuebingen.
Fig 1. Characterizations Glu and neuronal calcium responses and BOLD signal in rat.
(A) Temporal features of sensory response of the evoked Glu and calcium signal by forepaw stimuli (3mA, 2Hz,4s). (B) The evoked Glu spike increased proportionally to the amplitude of BOLD signals as the function of the stimulation intensity. (C) The fMRI images and representative time course of iGluSnFR (left hemisphere) and GCaMP (right hemisphere) expressed in rat somatosensory cortex.
Fig 2. Concurrent glutamate and astrocytic calcium recording with fMRI.
(A) The glutamate and astrocyte calcium transient (upper panel) induced by multiply stimuli (2 Hz,4s) and the overall signal time course (lower panel). (B) The whole brain fMRI map and time course of iGluSnFR and astrocytic GCaMP expressed in rat somatosensory cortex by evoked forepaw electrical stimulations.
Keywords: glutamate fMRI, multi-modal fMRI, neuron-glia-vessel network
Sensing intracellular calcium ions using a manganese-based MRI contrast agent (#422)
A. Barandov1, B. B. Bartelle1, A. Jasanoff1
1 Massachusetts Institute of Technology, Department of Biological Engineering, Cambridge, Massachusetts, United States of America
Calcium ions are essential to signal transduction in virtually all cells. Although optical probes for intracellular calcium imaging have been in use for decades, the development of probes for noninvasive monitoring of intracellular calcium signaling in deep tissue and intact organisms remains an important challenge . To address this problem, we designed and synthesized a new class of contrast agents (ManICS1-AM) that can be internalized and trapped in cells, and that enables intracellular calcium levels to be monitored by magnetic resonance imaging (MRI).
MRI data were acquired on a 7 T Bruker Biospec system. HEK293 cells (Freestyle 293-F, Thermo Fischer Scientific) were cultured and prepared for relaxometry, cell uptake and calcium response assessments. For stimulation experiments, cells were incubated with 10 µM ManICS1-AM for 2 h to allow for effective labeling and AM ester cleavage. Pharmacological stimulation was conducted by adding 5-10 µM of stimulants thapsigargin, charbocol, calcimycin, or arachidonic acid (Sigma-Aldrich).
To enable readouts of intracellular Ca2+ fluctuations by MRI, we developed a cell permeable probe, ManICS1-AM, derived from our previously reported intracellular contrast agents  and calcium specific chelator . ManICS1-AM is readily internalized and retains in cytosol upon enzymatic cleavage of acetomethoxy (AM) esters (Fig. 1). The calcium sensor, ManICS1, exhibits relaxivity of 3.8-5.1 mM-1s-1 in vitro, a 34% change dependent on Ca2+ concentrations over its full titration range from 0 to 1 mM (Fig. 2) and an EC50 of 5 µM once loaded into cells. At concentrations of 100 µM the agent is able to transduce 1µM intracellular calcium signals into 2% changes in T1-weighted MRI, which is easily detectable by fMRI. Importantly, our sensor shows signal changes of 7% as ManICS1-loaded cells were stimulated chemically (Fig. 2E-F). The sensor enables to map contrast agent-mediated calcium responses using rapid MRI pulse sequences with optimal spatiotemporal resolutions.
Here we are reporting the first example of intracellular calcium sensitive contrast agent, which is retained in cells through well-understood mechanisms and shows significant T1-weighted MRI signal changes to intracellular calcium fluctuations induced by chemical stimulants. Relying on our cell data, our calcium sensor ManICS1-AM is a powerful tool for monitoring calcium signaling events in brain which is the subject of our on going research.
 Gingerberger C. et al. Neuron, 2012, 73, 862.
 Barandov A. et al. J. Am. Chem. Soc. 2016, 138, 5483.
 Tsien R. et al. J. Cell Biol. 1982, 94, 325.
The authors thank funding came from the MIT Simons Center and NIH grants R21-MH102470 and U01-NS090451 to A.J.
Figure2. ManICS1 reports calcium-dependent MRI signal changes in cells.
(A) T1-weighted signal changes as a function of [Ca2+] in buffer. (B) Relaxivity values corresponding to the titration series in A. (C) Washout time course of sensors for preincubated HEK293 cells. (D) Distribution of manganese in cells incubated with sensors. (E) Ca2+ responses in cells labeled with ManICS1-AM (left) or Fura-2-AM (right). (F) Ca2+ response of cells incubated with ManICS1-AM.
Keywords: MRI contrast agents, Intracellular calcium imaging
Synthesis of fluorinated curcumin-based molecules for detecting amyloid plaques by 19F-MRI (#426)
R. Stefania1, L. Tei2, F. Garello1, U. Fasoglio1, M. Tripepi1, G. Forloni3, A. Deluigi3, C. Balducci3, E. Terreno1
1 University of Torino, Department of Molecular Biotechnology for Health Sciences, Torino, Italy, Italy
The detection of Aβ plaques is one strategy for Alzheimer’s Disease (AD) diagnosis. Amyloid imaging can be successfully pursued by 19F-MRI using fluorinated curcumin-based compounds.1 The developed probes contained a limited number of 19F atoms/molecule, and, furthermore, they were directly conjugated to the aromatic portion of the molecule, thus potentially reducing the detection sensitivity due to the broadening of the 19F signal. This work aims at synthesizing a series of novel F-containing curcumins with a high number of 19F nuclei, suitably spaced from the aromatic part of the molecule.
The preparation of F-containing curcumins (Fig. 1) started from the synthesis of mono and bi-carboxylic acid derivatives of curcumin followed by an amide coupling with two different perfluorinated amines, one of them incorporated a carboxylic functional group to improve solubility. All the compounds obtained were then purified by HPLC, dissolved at 10 mg/mL in DMSO or in a mixture of normal saline and Cremophor®, and characterized by 1H and 19F NMR. Following T1 and T2 measurement at 7 T, 19F-MR images of the synthesized compounds were acquired to prove their potential application in vivo. Finally, to verify the ability to bind Aβ plaques, different slices of APP-PS1 mouse brain tissues were incubated with the four compounds and then imaged by confocal microscopy.
With the exceptions of bi-F18 (characterized in DMSO), the compounds synthesized were soluble in normal saline added with Cremophor. T1 and T2 values obtained for the water soluble compounds lied in the proper range for ensuring a good MRI detection (Table 1), and a phantom containing the three compounds at 10 mg/mL was imaged in 10 min with a good signal-to-noise ratio, proportional to the number of fluorine atoms per molecule. The deposition of the compounds on brain slices explanted from APP-PS1 mice (transgenic AD model) showed a reduced affinity of the derivatives with respect to the parent curcumin. However, a qualitative assessment of the fluorescent signal allowed to rank the affinity of the compounds towards the plaques in the order: mono-F9 > mono-carboxy-F9 > bi-F18 > bi-carboxy-F18. Most likely, the higher affinity showed by the mono-derivatives is due to the preservation of one phenolic group in the aromatic portion of the structure involved in the binding to the plaque.
The compounds synthesized in this work displayed a good potential for in vivo applications either in terms of water solubility or 19F-MRI signal detection. As far as the affinity towards plaques is concerned, the preliminary results highlighted the importance to maintain at least one phenolic group of curcumin. On this basis, compound mono-F8 appears to be the more promising to be tested in vivo on preclinical AD models for amyloid imaging by 19F-MRI.
Keywords: Curcumin, 19F-MRI, Amyloid imaging, Alzheimer