15th European Molecular Imaging Meeting
supported by:

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

New Methods & Methodology in Neuroimaging II

Session chair: Abraham Martin (Bilbao, Spain); Albrecht Stroh (Mainz, Germany)
 
Shortcut: PW20
Date: Friday, 28 August, 2020, 12:00 p.m. - 1:30 p.m.
Session type: Poster

Contents

Abstract/Video opens by clicking at the talk title.

255

Polarization-Resolved Second Harmonic Generation (P-SHG) for deep 3D characterization of collagen-based scaffolds

Dionysios Xydias1, 2, Sotiris Psilodimitrakopoulos1, Giorgos Ziakas2, 3, Tzeranis Dimitrios4, Emmanuel Stratakis1

1 Foundation for Research and Technology Hellas, Institute of Electronic Structure and Laser, Heraklion, Greece
2 University of Crete, Department of Materials Science and Technology, Heraklion, Greece
3 Foundation for Research and Technology Hellas, Institute of Molecular Biology and Biotechnology, Heraklion, Germany
4 University of Cyprus, Department of Mechanical and Manufacturing Engineering, Nicosia, Cyprus

Introduction

Collagen based scaffolds are widely used as substrates for the culture of various types of cells, both as in-vitro models for the study of cells and as clinical patches to promote cell regeneration and thus faster healing of wounded tissue. For this reason, it is important to investigate the interaction between the collagen scaffold and the cells, in order to characterize its suitability for various applications. Here, we present a novel, non-linear method for deep 3D characterization of collagen-fiber orientation, using Polarization-resolved Second Harmonic Generation (P-SHG).

Methods

Human umbilical vein endothelial cells (hUVECs) and mesenchymal stem cells (hMSCs) were cultured on monolayers, detached and seeded in collagen type-I based scaffolds. The cells were then stained with Hoechst and Alexa488 after 3, 5, 7 and 10 days in-vitro (DIV). The growth of the cell culture was recorded via 2-photon excitation fluorescence (2PEF). The scaffold collagen was simultaneously characterized by P-SHG microscopy. The imaging system is a fully motorized custom-built non-linear microscope equipped with a 1030 nm, 76 MHz, 70 fs pulsed laser and a raster-scanning system based on galvanometric mirrors sample excitation[1]. The emitted signal was separated with high pass filters and collected in three different photon multiplier tubes (PMTs) each with a dedicated band pass filter.

Results/Discussion

We compared cell migration and tissue development on all DIV, between samples that were fixed and stained and a living culture on which we carried repetitive measurements. We validated that our imaging system didn’t inflict any observable alterations on our samples after several hours of irradiation. In both cases, we performed P-SHG imaging up to 200 μm deep and noticed that the cell migration proceeded up to 50μm deep in the scaffold and the cells were well-attached on the collage fibers.
We demonstrate that P-SHG signal analysis can be used for the determination of the collagen fiber orientation, both on the surface and as deep as 200 μm inside the scaffold. As a result, we observe that as the DIV increase, cells interact with the scaffold, reorganizing its structure. Scaffold pores decrease in size and the nominal 90o relevant orientation between collagen fibers is gradually lost in cell migrated areas, while it is retained deep inside the scaffold, far from cellular tissue.

Conclusions

P-SHG combined with 2PEF microscopy is used as a minimally invasive, deep imaging technique for the observation of the interaction between collagen-based scaffolds and the developed cellular tissue and the characterization of its effect on the scaffold. Cells have successfully created a live tissue migrating 50 μm deep in the scaffolds. Comparing samples on different DIV we conclude that cells reorganized the collagen fibers of the scaffold.

Acknowledgment

Dionysios Xydias acknowledges financial support from the HELLAS-CH and the Niarchos-ARCHERS projects.
Sotiris Psilodimitrakopoulos acknowledges financial support from IQONIC-EU, H2020, under Grant Agreement N 820677.

References
[1] S.Psilodimitrakopoulos et al Light: Science & Applicatios. 2018, Vol. 7, 18005.
[2] S.Psilodimitrakopoulos et al  Scientific Reports. 2019, Vol. 9, 14285.
[3] G.M. Maragkakis et al Opto-Electronic Advances. 2019, Vol. 2, 11.
[4] S.Psilodimitrakopoulos et al Biomedical Optics Express. 2014, Vol. 5, 4362.
[5] A.Kourgiantaki et al Under Review
Block representation of the non-linear microscopy imaging setup

A λ/2 retarder plate is used to control the polarization of the source laser pulse. The beam is raster-scanned on the sample with a system of galvanometric mirrors and focused with an objective lens. The SHG signal is collected in the forward direction with a condenser lens, while the emitted fluorescence signals from Hoechst and Alexa488 dyes are separated by a 509 HP filter and collected in the backwards direction. The band pass filters used are 514/3 (green), 485/64 (blue) and 536/40 (red) for the SHG, Hoechst and Alexa488 signals, respectively.

Imaging results
100 μm thick 3D volume of scaffold with cells (SHG-green, Hoechst-blue and Alexa488-red) (a). Only SHG signal from same region as in a (b). Analysis of the P-SHG signal yields two pieces of information: Orientation of collagen fibrils (c) and anisotropy parameter (d).
Keywords: collagen based scaffold, polarization sensitive SHG, fibril orientation, collagen organization, scaffold characterization
256

Choice of anesthesia and data analysis method can critically improve sensitivity of 18F-FDG PET during epileptogenesis

Ina Jahreis1, Pablo Bascunana1, Tobias L. Ross1, Frank M. Bengel1, Marion Bankstahl2, Jens P. Bankstahl1

1 Hannover Medical School, Department of Nuclear Medicine, Hannover, Germany
2 Hannover Medical School, Department of Laboratory Animal Science, Hannover, Germany

Introduction

Preclinical positron emission tomography (PET) imaging is a powerful tool for longitudinal in vivo measurements. For preclinical PET, anesthesia is usually unavoidable but also influences the distribution of radiotracers like 2-[18F]-fluoro-deoxy-D-glucose (18F-FDG) used for imaging of glucose metabolism. For reliable data interpretation, structured analysis of anesthesia impact on basal glucose metabolism and disease-associated changes is needed.

Methods

Here, we used naïve rats (n=7-11 per anesthesia) and rats during epilepsy development (n=4-12 per timepoint and anesthesia) known to exhibit cerebral glucometabolic changes to compare four anesthesia protocols: 18F-FDG uptake phase in conscious rats followed by a static scan, dynamic scans under continuous isoflurane, medetomidine-midazolam-fentanyl (MMF), or propofol anesthesia. Furthermore, we applied different analysis approaches: atlas-based regional analysis, statistical parametric mapping, and kinetic analysis.

Results/Discussion

In naive rats, significantly decreased cortical and thalamic 18F-FDG uptake was found under isoflurane (up to -23%, p=0.001) and propofol anesthesia (-22%, p<0.001) while MMF anesthesia led to a globally decreased brain uptake compared to conscious rats (up to -42% in piriform cortex, p<0.001). During epileptogenesis, MMF anesthesia was best distinctive for visualization of prominently increased glucometabolism in epilepsy-related brain areas (figure 1, up to +19% in thalamus, p=0.027). Kinetic modeling further increased sensitivity to detect changes at this time point, particularly for continuous isoflurane anesthesia (e.g. in amygdala: SUV: +13%, p=0.279 vs. Ki: +50%, p<0.001), but also for MMF anesthesia (Ki: +56% in thalamus, p<0.001). During chronic epilepsy, hypometabolism affecting almost the whole brain was detectable with all protocols.

Conclusions

This study exemplifies that careful selection of anesthesia protocols for functional imaging can prevent the generation of false negative data. Importantly, evaluation of anesthesia protocols should not only include healthy controls but also animals of the disease model of interest.

AcknowledgmentThis study was funded by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n°602102 (EPITARGET). I. Jahreis was supported by a scholarship from the Konrad-Adenauer-Stiftung e.V.
Figure 1. 18F-FDG uptake during epileptogenesis.

Coronal, sagittal and horizontal SPM t-maps resulting from voxel-wise comparisons of 18F-FDG uptake under baseline (BL) condition compared with uptake 7 d post status epilepticus (SE): (a) awake, (b) isoflurane, (c) MMF (medetomidine-midazolam-fentanyl) and (d) propofol anesthesia. Only clusters with significantly different voxels are shown (Student’s t-test, P<0.05, minimum cluster size of 100 voxels). Cold scale represents significantly decreased P-values, hot scale significantly increased P-values for each voxel.

Keywords: epilepsy, glucose metabolism, positron emission tomography
257

Hybrid PET-MRI evaluation of myelin biomarkers in a NHP model for MS: preliminary results

Marine Breuilly1, Marion Toussenot4, Stéphane Ployart4, Océane Wateau4, Thierry Billard2, Frédéric Bonnefoi2, Thibaut Iecker2, Justine Debatisse3, 5, Nicolas Costes2, Inés Merida2, Thomas Troalen5, Joachim Confais4, Fabien Chauveau1

1 Univ Lyon; Lyon Neuroscience Research Center; CNRS UMR5292; INSERM U1028, Univ Claude Bernard Lyon 1, LYON, France
2 CERMEP, LYON, France
3 Univ Lyon; CarMeN Laboratory; INSERM; INRA; INSA Lyon; Université Claude Bernard Lyon 1, LYON, France
4 CYNBIOSE, LYON, France
5 Siemens Healthcare SAS, SAINT DENIS, France

Introduction

Conventional MRI measurements in Multiple Sclerosis patients remain poorly correlated with disability and lack long-term prognostic value. A subsequent challenge is to identify the most reliable measure of myelin content in the white matter. Advanced MRI approaches (magnetization transfer, relaxometry…) are investigated (1). PET with repurposed amyloid tracers proved promising in translational studies (2), and allowed to quantify myelin changes in patients (3). Here we used an EAE model in NHP to concurrently evaluate candidate MRI markers and PET radiotracers, namely [11C]PiB and [18F]BF227.

Methods

Experimental Autoimmune Encephalitis (EAE) was induced in one adult female primate (Macaca fascicularis) through intradermal immunizations with recombinant human myelin oligodendrocyte glycoprotein (rhMOG) and Freund’s Incomplete Adjuvant (FIA) performed every 4 weeks. The animal underwent 3 simultaneous PET-MRI exams, under a Siemens Biograph mMR hybrid camera. PET consisted in a dynamic 60-min acquisition after [11C]PiB injection, immediately followed by 120-min acquisition after [18F]BF227 injection, while MRI included anatomical T1w, T2w and FLAIR. For analysis, T1w images were spatially normalized to a reference T1 template with corresponding atlas (4). T2w, FLAIR and PET images were registered to T1w.

Results/Discussion

The animal developed motor symptoms (transient unilateral left paresis) twenty days after the second immunization, and was treated with dexamethasone for one week to prevent aggravation. Imaging was performed at D4, D11 and D39 after the appearance of symptoms. Paresis, though attenuated in frequency and intensity, persisted until the final exam and sacrifice of the animal. Hyperintense WM lesion was detected on the right hemisphere (Fig. 1), which volume, measured on FLAIR images, decreased over time: 1170mm3; 500mm3; 220mm3 at D4, D11 and D39 respectively. Asymetrical WM uptake was noticed on late PET images at D11 (Fig. 2) of both [11C]PiB and [18F]BF227. Quantification of radiotracer will be performed as standardized uptake values (SUV) relatively to cerebellum after adequate definition of regions of interest (lesion core, border, and normal appearing white matter, in progress).

Conclusions

This preliminary report on one animal suggests that the EAE NHP model is relevant for testing PET-MRI biomarkers of myelin changes over the course of the disease. The study design will allow a back-to-back comparison of the “candidate” [18F]BF227 (previously evaluated in a rat model, 5), against the “reference” radiotracer [11C]PiB.

AcknowledgmentImaging was performed at CERMEP imaging platform, with the help of the staff: Jean-Batiste Langlois and Marco Valdebenito (animal care and monitoring), Francois Liger (radiochemistry), Dr Sophie Lancelot and Dr Didier Lebars (radiopharmacists), Christian Tourvielle (cyclotron operator), Dr Franck Lamberton (MRI sequences) and Dr Jérôme Redouté (image processing). This work was supported by the French National Agency of Research (grant number ANR-18-CE17-0013-01).
References
[1] O'Muircheartaigh J et al., ‘Quantitative neuroimaging measures of myelin in the healthy brain and in multiple sclerosis.’, Hum Brain Mapp. 2019 May;40(7):2104-2116
[2] Stankoff B et al., ‘Imaging central nervous system myelin by positron emission tomography in multiple sclerosis using [methyl-¹¹C]-2-(4'-methylaminophenyl)- 6-hydroxybenzothiazole.’, Ann Neurol. 2011 Apr;69(4):673-80
[3] Bodini B, Stankoff B, ‘Imaging Central Nervous System Demyelination and Remyelination by Positron-Emission Tomography.’, Brain Plast. 2016 Dec 21;2(1):93-98
[4] Ballanger B et al., ‘A multi-atlas based method for automated anatomical Macaca fascicularis brain MRI segmentation and PET kinetic extraction.’, Neuroimage. 2013 Aug 15;77:26-43
[5] Zhang M et al., ‘Quantitative longitudinal imaging of demyelination and remyelination in the lysolecithin-induced rat model of multiple sclerosis using [18F]-BF227 PET and MRI.’ J Nucl Med 2017;58(suppl 1):345, 64th Annual Meeting of the Society of Nuclear Medicine and Molecular Imaging (SNMMI), Denver, Colorado, USA.
WM lesion in a EAE NHP at D4, D11 and D39 after first MS relapse

This figure exhibits the transverse view of FLAIR MRI centered on the largest lesion localised in the right hemisphere. The signal tends to decrease over time, in concordance with a progressive remission state (neurological deficit decrease over time). Left to right column: D4, D11 and D39.

[11C]PiB - FLAIR versus [18F]-BF227 - FLAIR at D11
This figure exhibits the transverse view of FLAIR MRI centered on the largest lesion localised in the right hemisphere, together with the late PET image with [11C]PiB (left), and late PET image with [18F]-BF227 (right) for the PET-MRI exam at D11. The asymetry of radiotracer fixation is clearly visible between left and right. Unit of colorbar scale is given in SUV.
Keywords: Myelin, PET-MRI, Multiple sclerosis, MRI, EAE NHP
258

Towards a potential PET tracer targeting ALPHA-synuclein in the brain

Sabrina Buss1, Laura Kuebler1, Andreas Maurer1, Daniel Bleher1, Felix Schmidt3, Andrei Leonov4, Sergey Ryazanov4, Bernd J. Pichler1, Deniz Kirik2, Christian Griesinger4, Armin Giese3, Kristina Herfert1

1 University of Tuebingen, Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Tuebingen, Baden-Württemberg, Germany
2 Lund University, Brain Repair and Imaging in Neural Systems (BRAINS), Lund, Sweden
3 MODAG GmbH, Munich, Germany
4 Max Planck Institute for Biophysical Chemistry, Goettingen, Germany

Introduction

An imaging tracer to visualize α-synuclein (αSYN) in the brain would be a game changing tool for an early detection and therapy control in Parkinson´s disease (PD) research. In previous studies, we identified the compound MODAG-001, showing desired αSYN binding characteristics in vitro and favorable in vivo kinetics. Here, we present a modified structure (MODAG-005) with improved properties that has a higher affinity and selectivity in fibril binding experiments. In addition, first encouraging experiments in human brain tissue homogenates and slices are presented.

Methods

Saturation binding assays were performed using [3H]MODAG-005 and recombinant human fibrils of αSYN, Aß1-42 and tau46. 11C-labeling was performed and the in vivo kinetics and metabolites in plasma and brain were assessed. Brain homogenate binding assays were established in mouse brain tissue inoculated with decreasing concentrations of fibrils to determine the sensitivity limitations of the assay. Binding experiments were performed in human brain homogenates of PD cases and autoradiography experiments in brain slices from dementia with Lewy bodies (DLB).

Results/Discussion

MODAG-005 showed very high affinity to pure αSYN fibrils (Kd<0.2nM), moderate affinity to tau46 (Kd>15nM) and low affinity to Aß1-42 fibrils (Kd>100nM). In vivo, [11C]MODAG-005 showed a good BBB penetration (SUV>1.5) in the mouse followed by a fast clearance (half-life clearance 10 min). One metabolite was identified 5 and 15 min after tracer injection in the mouse brain, however, only at low levels (16% after 15 min). A sensitivity analysis in mouse brain homogenates revealed that low homogenate levels are needed to reduce the amount of non-specific binding for the trade-off of reduced levels of αSYN. First human homogenate binding experiments showed a higher binding in PD tissue in comparison to a healthy control. Autoradiography experiments in DLB cases showed a blockable binding pattern similar to those of paraffin embedded tissue blots.

Conclusions

In fibril binding experiments MODAG-005 showed high specificity towards αSYN and high selectivity over Aß1-42 and tau46 as well as favorable in vivo kinetics in the mouse brain. Binding experiments in human brain tissue are ongoing, but preliminary experiments are encouraging and point to a potential binding to αSYN in human brain tissue.

Keywords: alpha-synuclein, PET tracer, Parkinson's disease
259

Accelerating Neuroimaging Research: [18F]FB-HTL as a Reporter Gene Tracer with Potential for in vivo Brain Imaging

Sophie Stotz1, 2, Bernd J. Pichler1, 2, Andreas Maurer1, 2

1 Eberhard Karls University Tuebingen, Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Tuebingen, Germany
2 Eberhard Karls University Tuebingen, Cluster of Excellence iFIT (EXC 2180) "Image Guided and Functionally Instructed Tumor Therapies", Tuebingen, Germany

Introduction

While the blood-brain-barrier (BBB) protects the brain from infiltration of undesired molecules, it complicates the development of diagnostic imaging agents targeting brain functions. To bypass the time-consuming developmental phase of such agents, reporter gene systems with respective small-molecule ligands specifically designed to cross the BBB can be used. This work establishes a reporter gene system using the well-characterized HaloTag [1] and an 18F-labeled ligand ([18F]FB-HTL) that, due to its small size (MW=346 Da) and lipophilic properties (logP=3.1), is able to diffuse into the brain.

Methods

As cell model, HEK293 cells were stably transfected with a transmembrane domain fused HaloTag (HEK-Halo) and its expression on the cell surface was verified by Western blot analysis and fluorescence microscopy using untransfected HEK293 cells as control. To prove the in vitro functionality of the reporter gene system, cells were incubated with [18F]FB-HTL, and SDS-PAGE autoradiography was performed with the cell lysate. Tracer uptake and blocking with non-radioactive FB-HTL was quantified in the cell model using gamma-counting. In a pilot study to assess in vivo functionality, NOD.CB17-Prkdcscid mice carrying either HEK-Halo or HEK293 subcutaneous xenografts were subjected to dynamic PET/MR measurement and subsequent ex vivo biodistribution analysis.

Results/Discussion

Western blot analysis of cell lysates verified expression of the HaloTag, and confocal microscopy with a commercial fluorescent ligand confirmed the main localization at the cell surface. SDS-PAGE autoradiography confirmed specific binding of [18F]FB-HTL to the tag. In vitro binding studies showed significant (p<0.0001) uptake of [18F]FB-HTL by the HEK-Halo cells compared to HEK293 control cells which was blockable by co-incubation with non-radioactive FB-HTL (p<0.0001). The acquired PET/MR data showed combined renal and hepatobiliary clearance of the tracer, and a slightly increased uptake in the HEK-Halo xenografts (SUVHEK-Halo=0.98±0.213 (n=3); SUVHEK293=0.766±0.191 (n=5)). Additionally, non-specific uptake into the brain (SUVbrain=0.789±0.282 (n=8)) was observed. The mean tumor-to-muscle ratio measured ex vivo was approximately two-fold higher in HEK-Halo xenografts (1.842±0.432) compared to HEK293 control xenografts (1.053±0.304; n.s.).

Conclusions

The results indicate that [18F]FB-HTL, together with the HaloTag, serves as a functional reporter gene system for in vivo imaging. Non-specific brain uptake and clearance, as verified by the PET images and the time-activity curves, shows the potential of [18F]FB-HTL for brain imaging. Further studies will assess specific brain uptake by induction of the expression of the protein in the brain with viral vectors and subsequent PET/MR imaging.

Acknowledgment

We thank Johannes Kinzler and Natalie Mucha for excellent technical assistance.

References
[1] Los, GV, Encell, LP, McDougall, MG, Hartzell, DD, Karassina, N, Zimprich, C, Wood, MG, Learish, R, Ohana, RF, Urh, M and Simpson, D, 2008, 'HaloTag: a novel protein labeling technology for cell imaging and protein analysis' ACS chemical biology, 3(6), pp.373-382.

Keywords: Reporter Gene, HaloTag, PET Imaging, Blood-brain Barrier
260

Keeping an eye on stroke

Helena Justić1, Anja Barić1, Marin Radmilović2, Siniša Škokić1, Jelena Kežić1, Marina Dobrivojević Radmilović1

1 University of Zagreb School of Medicine, Croatian Institute for Brain Research, Zagreb, Croatia
2 Sestre milosrdnice University Hospital Center, Department of Ophthalmology, Zagreb, Croatia

Introduction

The retina shares embryonic origin, anatomy and vasculature with the brain, and can provide insight into the cerebrovascular changes after stroke. Several studies have shown evidence of retinal dysfunction following middle cerebral artery occlusion (MCAO). However, the disruption of retinal blood supply via MCAO remains poorly investigated since there is no clear consensus on the origin of the ophthalmic artery in rodents. The aim of our study was to clarify this issue by in vivo assessment of vascular perfusion and the resulting cerebral and retinal ischemic injury caused by MCAO.

Methods

Male C57Bl6J mice underwent a 30-minutes MCAO followed by reperfusion. Seven days prior and 24 hours after surgery the animals were scored for neurological deficit and subjected to fluorescein angiography to visualize the changes in the retinal vasculature followed by magnetic resonance imaging using a 7T BioSpec 70/20 USR system. The scans included a high-resolution T2-weighted anatomical scan of the brain and the ipsilateral eye, angiography and a T2-map scan of the brain. The obtained images were analyzed with ImageJ to establish the size of the ischemic lesion, hemispheres size and retinal thickness. On the same day the animals were sacrificed, the brains and eyes isolated and processed for histological analysis.

Results/Discussion

Even though previous studies showed different methods of retinal vasculature occlusion, none of them provided anatomical evidence of the origin of the ophthalmic artery in mice or the status of vascular perfusion after occlusion. Middle cerebral artery occlusion significantly reduced the blood flow to the MCA and completely reduced the blood flow through the pterygopalatine artery, which gives rise to the ophthalmic artery in our C57bl6J mice. The anatomical scan used for the eye had sufficient spatial resolution but inadequate contrast to distinguish between retinal and choroidal layers through the entire globe. The radially measured total retinal/choroid layer thickness 24 hours after MCAO was significantly increased. The thickening of the retina and choroid correlates with the severity of brain edema.

Conclusions

The middle cerebral artery occlusion method proved to be a suitable model for induction of brain and retinal ischemia in mice, due to simultaneous occlusion of the MCA and the pterygopalatine artery. The severity of retinal ischemia assessed by the retinal/choroid layer thickness correlates with the severity of the brain ischemia. With this model, we could get an insight into the pathophysiological changes in the brain by observing the retina.

AcknowledgmentThe study is supported by the Croatian Science Foundation project BRADISCHEMIA (UIP-2017-05-8082) and co-financed by the European Union through the European Regional Development Fund, Operational Programme Competitiveness and Cohesion, grant agreement No. KK.01.1.1.01.0007, CoRE – Neuro. The work of doctoral student Anja Barić has been fully supported by the “Young researchers' career development project – training of doctoral students” of the Croatian Science Foundation funded by the European Union from the European Social Fund. Multimodal imaging was done at Laboratory for Regenerative Neuroscience - GlowLab, University of Zagreb School of Medicine.
References
[1] Kumar V. Understanding Retinal Changes after Stroke. Open J Ophthalmol. 2017;7(4):281–292.
[2] Luan H, Roberts R, Sniegowski M, Goebel DJ, Berkowitz BA. Retinal Thickness and Subnormal Retinal Oxygenation Response in Experimental Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2006;47(1):320-328.
[3] Allen RS, Sayeed I, Cale HA, et al. Severity of middle cerebral artery occlusion determines retinal deficits in rats. Exp Neurol. 2014;254:206–215.
Keywords: MRI, angiography, brain, retina, ischemia
261

High resolution patterns of beta-amyloid deposition in the whole mouse brain revealed by non-invasive dynamic optoacoustic tomography

Ruiqing Ni1, Xosé L. Dean-Ben1, 2, Gloria Shi1, Zhenyue Chen1, 2, Markus Rudin1, Roger Nitsch3, Daniel Razansky1, 2, Jan Klohs1

1 ETH Zurich & University of Zurich, Institute for Biomedical Engineering, Zurich, Switzerland
2 University of Zurich, Institute for Pharmacology and Toxicology, Zurich, Switzerland
3 University of Zurich, Institute for Regenerative Medicine, Zurich, Switzerland

Introduction

The abnormal deposition of fibrillar beta-amyloid (Aβ) deposits in the brain is a major histopathological hallmark of Alzheimer’s disease (AD). High-resolution and yet non-invasive imaging of amyloid deposition in 3D across the whole brain is still lacking. Here, we demonstrate a novel method to quantify brain Aβ fibrils by using optoacoustic tomography (OAT) in two transgenic mouse models of cerebral amyloidosis

Methods

Transgenic arcAβ and APP/PS1mice at 18-24 months of age and age-matched non-transgenic littermates (NTL) were used (1). For in vivo amyloid OAT (2), an oxazine derivative AOI987 (3) were injected i.v. through tail vein. Data was acquired over 120 minutes and analyzed after image reconstruction and linear unmixing for oxyhemoglobin, deoxyhemoglobin and AOI987. Volume-of-interest analysis was performed after registering the OAT dataset with magnetic resonance imaging data and anatomical brain atlas using PMOD. Immunohistochemical staining using anti-Aβ antibody 6E10, conformation antibody OC and AOI987 were performed for validation.

Results/Discussion

We showed that in vivo OAT detects with high specificity and sensitivity the fibrillar Aβ deposits in the cortical, hippocampal and thalamic brain regions of arcAβ and APP/PS1 mice compared to non-transgenic littermates. The AOI987 signal detected using in vivo and ex vivo OAT in arcAβ and APP/PS1 mouse brain correspond with immunohistochemical staining of Aβ deposits distribution in brain tissue sections. Confocal microscopy showed co-localization of AOI987 and 6E10 signals to parenchymal and vessel wall in brain tissue sections from arcAβ and APP/PS1 mice.

Conclusions

We demonstrated a new high-resolution in vivo 3D amyloid imaging platform mouse model of AD amyloidosis. This facilitates mechanistic studies and the monitoring of putative treatments targeting Aβ deposits.

Acknowledgment

The authors acknowledge technical support from Dr. Joachim Hohl, Simon and Dr. Gabriella Bodizs of the Scientific Center for Optical and Electron Microscopy (ScopeM) of ETH Zurich, Ms Priyanka Ravikumar at Institute for Regenerative Medicine, Univeristy of Zurich. JK received funding from the Swiss National Science Foundation (320030_179277), in the framework of ERA-NET NEURON (32NE30_173678/1), the Synapsis foundation and the Vontobel foundation. RN received funding from the University of Zurich Forschungskredit (Nr. FK-17-052), and Synapsis foundation career development award (2017 CDA-03).

References
[1] Ni R, et al. (2018), Cortical hypoperfusion and reduced cerebral metabolic rate of oxygen in the arcAβ mouse model of Alzheimer’s disease. Photoacoustics 10, 38-47
[2] Dean-Ben XL, et al. (2016), Functional optoacoustic neuro-tomography for scalable whole-brain monitoring of calcium indicators. Light Sci Appl 5, e16201
[3] Hintersteiner M, et al. (2005), In vivo detection of amyloid-beta deposits by near-infrared imaging using an oxazine-derivative probe. Nat Biotechnol 23, 577-583
Keywords: Alzheimer' disease, optoacoustic tomography, amyloid-beta, non-invasive imaging, mouse model
262

Preclinical evaluation of [11C]UCB-J for PET-imaging of synaptic vesicle glycoprotein 2A in mice

Maxime Schreurs1, Esther Kooijman1, Martien P. J. Mooijer1, Matthijs Verhage2, Albert D. Windhorst1, Guus A. M. S. van Dongen1, Danielle J. Vugts1, Wissam Beaino1

1 Amsterdam UMC, Vrije Universiteit Amsterdam, Radiology & Nuclear Medicine, Neuroscience, Amsterdam, Netherlands
2 Amsterdam UMC, Clinical genetics, Center for Neurogenomics and Cognitive Research, Amsterdam, Netherlands

Introduction

Synaptic loss is associated with neurodegenerative diseases such as Alzheimer’s disease and leads to a reduction in synaptic vesicle glycoprotein 2A (SV2A) expression. SV2A is expressed in synapses across the brain and is used as a potential synaptic density marker. SV2A levels can be assessed by PET imaging using the radiotracer [11C]UCB-J1. PET imaging of SV2A in mouse models can be of importance to better understand synaptic dysfunction in neurodegenerative diseases and help for novel therapies. Here, we evaluated the specificity of [11C]UCB-J for the assessment of SV2A expression in mice.

Methods

PET scans were acquired on a MEDISO nanoscan PET/CT or PET/MRI (Budapest, Hungary) for 1h immediately after i.v. tail vein injection of [11C]UCB-J (9.5 ± 2.1 MBq, Am=69.5 ± 37.1 GBq/µmol) in mice. A baseline scan was performed in wild-type (WT) mice (C57BL/6, n=3) to determine total uptake of [11C]UCB-J in the brain. Blocking was performed in the same mice by injecting Levetiracetam i.v. (LEV, 30 mg/kg), a SV2A specific antagonist, 2-4 minutes prior to [11C]UCB-J administration. To further confirm the specificity of [11C]UCB-J, brain uptake was assessed in SV2A heterozygous knock-out mice (HKO, n=3), in which SV2A expression is about 50% lower. Dynamic and static reconstruction was performed using a TeraTomo reconstruction algorithm. Data are presented in standard uptake value (SUV).

Results/Discussion

Dynamic reconstruction and comparison of the area under the curve (AUC) of the time activity curves (TACs), demonstrated that blocking with LEV significantly reduced the uptake of [11C]UCB-J uptake in the brain by 54% (baseline AUC=281 vs block AUC=131). The HKO group demonstrated a 38% reduction in uptake compared to the WT mice (HKO AUC=175 vs baseline AUC=281) (Fig. 1A).
Static reconstruction (0-60 min post injection), demonstrated that the uptake of [11C]UCB-J after LEV blocking was lower in the brain (baseline 4.8 ± 0.2 vs block 2.5 ± 0.2 SUV0-60min, respectively; p<0.01), while in the SV2A HKO group the uptake of [11C]UCB-J was also lower compared to the WT mice (3.1 ± 0.6 vs 4.8 ± 0.2 SUV0-60min respectively; p<0.01). The reduction in brain uptake of [11C]UCB-J in HKO mice was less pronounced compared to blocking with LEV, this difference was however not significant (Fig. 1B). The representative PET images are shown in Fig 1C.

Conclusions

It is shown by both dynamic and static reconstruction that [11C]UCB-J binds specifically to SV2A. These results are crucial to further validate and use [11C]UCB-J as a tracer in mouse models.

References
[1] Finnema, SJ, Nabulsi, NB 2016,'Imaging synaptic density in the living human brain.' Science Translational Medicine, Vol. 8, Issue 348, pp. 348ra96:AAAS
Figure 1. PET imaging results of [11C]UCB-J in WT and SV2A HKO mice
(A) Whole brain TACs at baseline, blocking with LEV and in SV2A HKO expressed as SUV Mean ± SD. (B) Whole brain SUV of static (0-60 min) PET reconstruction, ** p<0.01. (C) Representative PET-MRI images of [11C]UCB-J brain uptake at baseline,  blocking study with LEV and in SV2A HKO mouse, expressed as SUV0-60 min. MRI image (left), PET image (middle) and MRI images overlaid with PET (right).
Keywords: synaptic density, [11C]UCB-J, SV2A, PET, mouse model
263

Inversion-Recovery T1w MRI Enables Accurate Localization of Cysts After Traumatic Spinal Cord Injury

Sinisa Skokic1, Pasquale Romanelli2, 4, Patrick Heimel3, Lara Bieler2, 4, Christina Kreutzer2, 4, Pia Zaunmair2, 4, Dominika Jakubecova2, 4, David Hercher3, Marina Radmilovic Dobrivojevic1, Sebastien Couillard-Despres2, 4, 5

1 University of Zagreb School of Medicine, Croatian Institute for Brain Research, Zagreb, Croatia
2 Paracelsus Medical University, Institute of Experimental Neuroregeneration, Salzburg, Austria
3 Ludwig Boltzmann Institute for Experimental and Clinical Traumatology, Vienna, Austria
4 Spinal Cord Injury and Tissue Regeneration Center Salzburg, Salzburg, Austria
5 Austrian Cluster for Tissue Regeneration, Vienna, Austria

Introduction

MRI imaging represents the best technique for monitoring the progression of secondary damage after traumatic spinal cord injury and evaluating the effects of therapy. Most common MRI techniques for tSCI assesment are diffusion imaging for tractography and myelination status, and T2/T2* imaging for morphological assessment. However, volumetric analysis of tSCI lesions based on T2/T2* images is difficult due to poor contrast between different tissues at the injury site. Here, we propose a T1-based MRI technique which delivers better contrast than T2/T2*w scans and enables automatic segmentation.

Methods

A total of 16 rats underwent tSCI procedure using the Infinite Horizon impactor. 8 weeks post-injury, the animals were perfusion-fixed with 4%-PFA and their spinal cords isolated. The samples were imaged ex vivo with 7T MRI using 3D T2w and T2*w sequences and the proposed T1w-IR sequence, all at 80 um isotropic resolution. Image post-processing was carried out in FIJI/ImageJ with a custom-built program for automatic segmentation and volumetric measurements of the cysts, which appear as hyperintense in the images. Measurements were validated by an expert evaluator. The samples were afterwards soaked in Accupaque for 48h and imaged with uCT at 6 um isotropic resolution. Finally, the tissue was histologically processed in order to validate the findings observed with imaging modalities.

Results/Discussion

In T2/T2*-weighted images, the whole injury site appears as one homogeneous hyperintense region, with little contrast between different tissue types present. The addition of a selective inversion recovery (IR) pulse, with adequate IR pulse timing, suppresses the signal of glial scar tissue which surrounds the cystic cavities. At the same time, signal originating from within the cyst is mostly unaffected, and grey and white matter signals only partly suppressed. The resulting images have sufficient contrast to lend themselves to automatic or semi-automatic segmentation of the cysts.

The developed automatic segmentation procedure was tested for robustness and precision by measuring the volume on orthogonal sagittal cross-sections of the same sample. Preliminary tests indicate an acceptable level of variation in total measured cyst volume between measurements. False positive and false negative detections on average amount to only a few percent of the total measured volume.

Conclusions

Automatic segmentation is preferred over manual segmentation for assessing spinal cord injury in the chronic phase both time-wise and accuracy-wise, due to the presence of multiple very small, irregularly shaped cysts. T1w-IR sequence proposed here isolates the cysts from the surrounding tissue, thus enabling automatic segmentation and volumetric analysis, in contrast to standard T2/T2*-based sequences which lack the specificity and contrast.

Acknowledgment

This work was supported by Croatian-Austrian Scientific & Technological Cooperation grant 2018/2019 of the Ministry of Science and Education of the Republic of Croatia, Austrian Agency for International Cooperation in Education and Research, Project No. HR 02/2018 and Paracelsus Medical University Research Fund PMU-FFF R-19/01/117-ROM. MRI scans were performed at the Laboratory for Regenerative Neuroscience – GlowLab, University of Zagreb, Croatia. This publication was co-financed by the European Union through the European Regional Development Fund, Operational Programme Competitiveness and Cohesion, grant agreement No. KK.01.1.1.01.0007, CoRE – Neuro.

References
[1] Wu T-L, Byun NE, Wang F, Mishra A, Janve VA, Chen LM, Gore JC, 2016, ‘Longitudinal assessment of recovery after spinal cord injury with behavioral measures and diffusion, quantitative magnetization transfer and functional magnetic resonance imaging’, NMR in Biomedicine, e4216, 1-14, Wiley
[2] Wilkins N, Skinner NP, Motovylyak A, Schmit BD, Kurpad S, Budde MD, 2020, ‘Evolution of MRI as Predictors and Correlates of Functional Outcome After Spinal Cord Contusion Injury in the Rat’, Journal of Neurotrauma, ePub Ahead of Print, Mary Ann Limbert, Inc.
[3] Song W, Song G, Zhao C, Li X, Pei X, Zhao W, Gao Y, Rao J-S, Duan H, Yang Z, 2018, ‘Testing Pathological Variation of White Matter Tract in Adult Rats after Severe Spinal Cord Injury with MRI’, BioMed Research International, e2018, 1-13, Hindawi
Keywords: MRI, spinal cord injury, rat, ex vivo, inversion recovery
264

Mesenchymal Stem Cell-Derived Exosomes for Drug Delivery and PET Imaging

Arunkumar Pitchaimani1, Miguel Moreira1, Annalisa Palange1, Beatriz P. Samper2, Roberto Moratta1, Felipe P. Cardoso2, Paolo Decuzzi1

1 Istituto Italiano di Tecnologia, Nanotechnology for Precision Medicine, GE, Italy
2 University of Navarra, CIMA, Navarra, Spain

Introduction

Extracellular vesicles like the nano-sized EXOsomes (EXO) from mammalian cells play a key role in various transcellular communication processes including genetic exchange. Particularly Mesenchymal Stem cells (MSCs) exosomes have functions similar to that of its parent cell in repairing tissue damage, the immune system modulation, etc.1,2 Interestingly, it has the ability to cross the BBB upon systemic injection and accumulates in the brain parenchyma.3 Herein, we investigate the role of MSC EXO in delivering therapeutic doxorubicin and PET imaging agents (64Cu) non-invasively via the nose.

Methods

Human adipose derived MSCs were isolated following the protocols developed at CIMA, University of Navarra (Spain). Further, EXO were isolated from the MSCs via a differential ultracentrifugation. The protein profile was analyzed using western blots and LC-MS (Proteinase-K method). Rhodamine-EXO were used for studying differential cellular uptake in macrophages (RAW 264.7), neuronal cells (Neuro-2A) and brain cancer cells (U87-MG) via confocal and flow cytometry. Doxorubicin was loaded into EXO and tested its encapsulation efficiencies and release rates using HPLC . Finally, exosomes were also labeled with lipid-DOTA molecules reacted with 64Cu, using fusogenic liposomes. Administration of EXO to mice was realized non-invasively through the nasal cavity in Nu/Nu mice.

Results/Discussion

The exosomes isolated from the hADSCs were spherical in shape with an average diameter of 119 ±7 nm under cryo-EM. The dynamic light scattering analysis showed exosomes with a hydrodynamic diameter of ~ 130 ± 18 nm and the surface zeta potential around – 30 ± 2mV. The characteristic functional protein markers exposed of the exosome membrane, CD-9 and CD-81, were clearly identified through western blot and mass spectroscopy analyses. Further, confocal microscopy and flow cytometry analyses showed that significant levels of naïve exosome were taken up by RAW 264.7, Neuro-2A and U87-MG cells, as compared to the PK treated EXO. The radiolabeling efficacy of EXO was found to be ~70% and showed a high stability for over 24h. Upon intra-nasal administration, EXO were slowly released from the nasal cavity at the systemic level with peak accumulations in the stomach, intestine, liver and tumor. The cytotoxicity of DOX-EXO was comparable to that of the free drug.

Conclusions

This preliminary work demonstrates that human adipose stem cell-derived exosomes can be efficiently loaded with a chemotherapeutic molecule and a radioactive agent. The biodistribution of the exosomes can be monitored over time using Nuclear Imaging and the therapeutic activity of doxorubicin loaded exosomes is comparable with the free drug molecule.

AcknowledgmentThis work was partially supported by the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 754490.
References
[1] Bose, R. J.; Kim, B. J.; Arai, Y.; Han, I.-B.; Moon, J. J.; Paulmurugan, R.; Park, H.; Lee, S.-H. Bioengineered Stem Cell Membrane Functionalized Nanocarriers for Therapeutic Targeting of Severe Hindlimb Ischemia. Biomaterials 2018, 185, 360–370
[2] Gao, C.; Lin, Z.; Jurado-Sánchez, B.; Lin, X.; Wu, Z.; He, Q. Stem Cell Membrane-Coated Nanogels for Highly Efficient In Vivo Tumor Targeted Drug Delivery. Small 2016, 12 (30), 4056–4062.
[3] Betzer, O.; Perets, N.; Angel, A.; Motiei, M.; Sadan, T.; Yadid, G.; Offen, D.; Popovtzer, R. In Vivo Neuroimaging of Exosomes Using Gold Nanoparticles. ACS Nano 2017, 11 (11), 10883–10893.
Mesenchymal Stem Cell-Derived Exosomes for Drug Delivery and PET Imaging
Role of Mesenchymal stem cell exosome in delivering therapeutic doxorubicin and PET imaging agents (64Cu) non-invasively via nose to brain.
Keywords: Mesenchymal stem cells, Exosomes, PET, Doxorubicin
265

Localisation of Kisspeptin and Neurokinin B Receptors in rat brain and peripheral tissues

Robert B. Reeve1, Thomas Ebenhan2, Robert P. Millar3, Jillene Visser2, Janke Kleynhans2, Zulfiah Moosa5, Jan Rijn Zeevaart4

1 University of Pretoria, Physiology, Faculty Health Sciences, Pretoria, South Africa
2 Nuclear Medicine Research Infrastructure, Preclinical Imaging Facility, Pretoria, South Africa
3 University of Pretoria, Immunology, Faculty of Health Sciences, Pretoria, South Africa
4 Nuclear Corporation of South Africa, Radiochemistry, Pretoria, South Africa
5 University of Pretoria, Anatomy & Physiology, Faculty of Veterinary Science, Pretoria, South Africa

Introduction

Reproduction is controlled by Gonadotropin Releasing Hormone (GnRH) which stimulates the release of gonadotrophins from the anterior pituitary gland 1. Upstream regulation of GnRH release has been shown to be due to Kisspeptin (KP) and Neurokinin B (NKB).  KP/NKB are found predominantly in the arcuate nucleus of the hypothalamus and is secreted onto GnRH neurons 2. KP may have other effects in the brain involving appetite, metabolism and reproductive behaviours 3. This study uses radiolabeled KP / NKB analogues to study receptor localization and tracer biodistribution with noninvasive imaging.

Methods

Biological activity for DOTA-KP/-NKB was assessed in a dose response (range 10^(-6)-10^(-12) M) manner using an inositol phosphate (IP) assay using KiSS1R and NK3R transfected HEK 293-T cells. A generic radiolabeling method was adapted and tested with ITLC for 68Ga-product, free- and colloidal-68Ga, respectively 4. Labelling efficiency, purity and tracer stability (3 h) was quantified using HPLC.  Sprague-Dawley rats were then administered intravenously with [68Ga]Ga-DOTA-KP/-NKB (8-40 MBq). Positron Emission Tomography/Computed Tomography (microPET/CT) was commenced immediately after administration and continued for a 30-min initial scan followed by a 15-min scan after 105 min post tracer administration. Ex vivo organ biodistribution was measured by gamma counting and expressed as %ID/g.

Results/Discussion

IP results showed that both DOTA-neuropeptides showed comparable potencies as compared to their unconjugated counterparts, thus biological activity was retained. At 5 nM DOTA-KP showed already good 68Ga-radiolabeling whereas DOTA-NKB displayed suboptimal radiolabeling. HPLC revealed that DOTA-KP and -NKB labeled at 20 and 30 nM displayed 84% and 76% labelling efficiency, respectively. Unlike [68Ga]Ga-DOTA-KP, [68Ga]Ga-DOTA-NKB showed a decrease in stability after 3 hours. microPET/CT images and ex vivo results suggested there was little uptake of [68Ga]Ga-DOTA-KP by the brain (0.02 %ID/g) and the highest uptake was seen in the kidneys 2.96 %ID/g. The stomach, reproductive organs and the small intestine showed notable uptake of [68Ga]Ga-DOTA-KP (0.44, 0.35 and 0.38 %ID/g; respectively); a target blockade co-injecting 50-fold cold DOTA-KP suggested that there was no specific uptake. [68Ga]Ga-DOTA-NKB uptake was seen primarily within the kidneys along with negligible uptake in the brain.

Conclusions

Adequate radiolabelling solutions were established for [68Ga]Ga-DOTA-KP/-NKB. The injection did not result in any cerebral tracer uptake. Ex vivo  biodistribution of [68Ga]Ga-DOTA-KP was relevant to kidneys, reproductive organs, small intestines, stomach and spleen; however, [68Ga]Ga-DOTA-NKB uptake differed notably for serum, small intestines, thyroid, and femur. Both peptides show effective renal excretion, which is expected for these peptides.

AcknowledgmentThe authors acknowledge the National Research Foundation, the Nuclear Medicine Research Infrastructure for funding as well as my supervisors for their guidance.
References
[1] Pinilla L, Aguilar E, Dieguez C, Millar RP, Tena-Sempere M, 2012, ‘Kisspeptins and Reproduction: Physiological Roles and Regulatory Mechanisms’, Physiol Rev, 92, 1235–316
[2] Roa J, Aguilar E, Dieguez C, Pinilla L, Tena-Sempere M, 2008, 'New frontiers in kisspeptin/GPR54 physiology as fundamental gatekeepers of reproductive function', Frontiers Neuroendocrinology, 29, 48-69
[3] Newton CL, Anderson RC, Katz AA, Millar RP, 2016, 'Loss-of-function mutations in the human luteinizing hormone receptor predominantly cause intracellular retention', Endocrinology, 157, 4364-77
[4] Ebenhan T, Chadwick N, Sathekge MM, Govender P, Govender T, Kruger HG, 2014, Peptide synthesis , characterization and 68 Ga-radiolabeling of NOTA-conjugated ubiquicidin fragments for prospective infection imaging with PET / CT, 41, 390-400
Keywords: Neuropeptides, Localistion, biodistribution, kisspeptin, DOTA