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).

Disease Models - Translational Approaches in Neuroimaging II

Session chair: Dominik von Elverfeldt (Freiburg, Germany); Jürgen Goldschmidt (Magdeburg, Germany)
Shortcut: PW19
Date: Friday, 28 August, 2020, 12:00 p.m. - 1:30 p.m.
Session type: Poster


Abstract/Video opens by clicking at the talk title.


Evaluation of the therapeutic effects of Ambroxol in ischemic stroke rats

Kristin J. Patzwaldt1, Linda Schramm1, Yi Wang2, Sven Poli2, Bernd J. Pichler1, Salvador Castaneda Vega1, 3

1 Eberhard Karls University of Tübingen, Department of Preclinical Imaging and Radiopharmacy/ Werner Siemens Imaging Center, Tübingen, Germany
2 Eberhard Karls University of Tübingen, Department of Neurology & Stroke/ Hertie Institute for Clinical Brain Research, Tübingen, Germany
3 Eberhard Karls University of Tübingen, Department of Nuclear Medicine and Clinical Molecular Imaging, Tübingen, Germany


Ischemic Stroke (IS) is a sudden and devastating disease where blood flow is locally restricted or completely interrupted in a brain region causing neuronal cell death within minutes. Acute energy failure results in neuronal depolarization due to loss of function of sodium-potassium pumps and uncontrolled sodium and calcium influx into the cells terminating in brain swelling [1]. It was shown that Ambroxol had inhibiting effects on sodium and calcium channels [2]. Here we hypothesized that Ambroxol would have a therapeutic effect when applied shortly after IS induction in rats.


Middle cerebral artery occlusion stroke surgery was performed on rats (n=58) which were divided into three groups after a 60 min occlusion: Control-NaCl (NaCl) (n=23), Ambroxol (n=17) and Control-PEG400+NaCl (PEG) (n=18). Ambroxol treatment was applied intraperitoneally twice a day. T2 weighted image (T2WI) and Diffusion weighted image scans were performed up to 1 month after occlusion and apparent diffusion coefficient (ADC) maps were calculated. All images were co-registered and analyzed using an in-house developed machine learning algorithm that segments stroke regions. It was trained on the NaCl and tested on the Ambroxol and PEG groups. Kruskal-Wallis non-parametric voxel-wise p-value maps were used to show significant differences between therapy and control groups with an α of 0.05.


The mean ADC, T2WI and stroke probability maps of all animals per group at 24 h are shown in Figure (Fig) 1. ADC in the ipsilateral cortex of Ambroxol animals shows a stark contrast to the two control groups. This is measured by the mean stroke probability maps denoting less stroke-predicted regions in the cortex. The voxel-wise p-value maps showed statistically significantly different voxels (p<0.05) in striatal and cortical brain regions of animals receiving therapy, in comparison to both control groups (Fig 2). A voxel-wise p-value map calculated between both control groups showed a minimal significance between the voxels. This was consistent with the reduced ADC regions in the cortex of both control groups. The stroke volume evaluation showed a median stroke volume in the Ambroxol group of approximately half of that of control animals and a consistent variance among the groups (median: 0.07cm³±0.10cm³ compared to NaCl (median: 0.14cm³±0.08cm³) and PEG (median: 0.17cm³±0.09cm³).


Our preliminary results on the ongoing preclinical therapeutic trial confirmed our hypothesis that Ambroxol has a measurable therapeutic effect in ischemic stroke at 24 h post stroke occurrence. Further data analysis is still ongoing in order to determine if these effects translate into metabolical and functional improvements.

1 Woodruff T. et al., 2011, “Pathophysiology, treatment, and animal and cellular models of human ischemic stroke”, Molecular Neurodegeneration6(11)

2 Malerba M, Ragnoli B., 2008 „Ambroxol in the 21st century: pharmacological and clinical update“, Expert Opin. Drug Metab. Toxicol.4(8)
Keywords: MACO, Therapy, Ambroxol

Stem cell induced functional network stabilization after cortical stroke: A longitudinal resting state fMRI study in mice

Anuka Minassian1, Claudia Green1, Michael Diedenhofen1, Stefanie Vogel1, Simon Hess2, Maren Stoeber1, Marina Dobrivojević Radmilović1, 3, Dirk Wiedermann1, Peter Kloppenburg2, Mathias Hoehn1

1 Max-Planck-Institute for Metabolism Research, In-vivo-NMR Laboratory, Cologne, Germany
2 University of Cologne, Biocenter - Institute for Zoology, Cologne, Germany
3 University of Zagreb, School of Medicine, Zagreb, Croatia


Most stroke studies dealing with functional deficits and assessing stem cell therapy have used the filament model of the middle cerebral artery occlusion. This model produces massive hemispheric damage and can be seen as a model for severe clinical strokes. Recently, the anatomic and behavioral changes have been reported for the distal occlusion of the middle cerebral artery, generating a well circumscribed cortical lesion, which can thus be proposed as mild to moderate cortical stroke. This provides a new model for the assessment of stem cell treatment.


A total of 53 nude mice were used. 32 mice underwent distal occlusion of the middle cerebral artery (dMCAO) for cortical stroke of moderate severity (1), with 18 mice for stroke only and another 14 mice additionally received stem cells. Human neural stem cells were implanted adjacent to the ischemic lesion, one week after stroke induction. A third group (n=12) served for sham occlusion. Stable graft vitality was monitored with bioluminescence imaging. In all animals, the functional networks were studied with resting state functional magnetic resonance imaging (rs-fMRI) for 12 weeks after stroke or sham intervention. Differentiation of the neural stem cells was analyzed by patch clamp electrophysiology.


Following stroke induction, we found a pronounced and continuously rising hypersynchronicity of the sensorimotor networks including both hemispheres after dMCAO (Fig.1), in contrast to the filament MCAocclusion model which produces severe strokes, where massive reduction of the functional connectivity, reflecting hyposynchronicity, had been reported earlier (2). The sham occluded animals presented no alterations of functional network strength relative to baseline. In the stem cell treated ischemic animals functional connectivity did not show hypersynchronicity but was globally only slightly reduced below baseline at 2 weeks post stroke, returning thereafter completely to baseline. Bioluminescence signal intensity of grafted neural stem cells remained stable throughout the whole 12 weeks observation period clearly indicating a persistent vitality of the graft, both in healthy and in ischemic brains, and patch clamp recording showed neuronal differentiation of the graft at 14 weeks.


This novel hypersynchronicity appears as a hallmark of mild cortical strokes, in contrast to strokes with striatal involvement with hyposynchronicity (1), pointing to the potential of differential diagnosis between mild and severe strokes by rs-fMRI. Stem cell treatment led to persistent normalization of the functional sensorimotor networks, demonstrating the high potential of stem cell treatment for functional outcome improvement after stroke.

AcknowledgmentStipend to AM by the University of Cologne Emerging Groups Initiative (CONNECT group) implemented into the Institutional Strategy of the University of Cologne and the German Excellence Initiative.
[1] Minasian et al. (2019) Scientific Reports 9, 6823

[2] Green et al. (2018) Journal of Neuroscience 38, 1648-61
Functional connectivity changes
Fig.1 Schematic changes of whole brain functional network strength during 12 weeks after stroke induction, referenced to the pre-stroke baseline situation.
Keywords: moderate severity stroke model, resting state fMRI, functional connectivity, network stabilization, stroke-induced hyperconnectivity

Magnetic Resonance Imaging evaluation of the role of hyperglycemia in experimental subarachnoid haemorrhage

Ana Joya1, 2, Sandra Plaza3, Daniel Padro3, Jordi Llop2, Pedro Ramos-Cabrer3, 4, Abraham Martín1, 4

1 Achucarro Basque Center for Neuroscience, Leioa, Spain
2 CIC biomaGUNE, Radiochemistry and Nuclear Imaging, San Sebastian, Spain
3 CIC biomaGUNE, Magnetic Resonance Imaging, San Sebastian, Spain
4 Ikerbasque Basque Foundation for Science, Bilbao, Spain


Spontaneous subarachnoid haemorrhage (SAH) is a devastating cerebrovascular disorder due to the rupture of an intracranial aneurysm with high mortality and morbidity1. SAH can lead to devastating outcomes for patients, including cognitive decline and delayed cerebral ischemia2. The occurrence of hyperglycemia has been linked with the worsen outcome after SAH3. Despite this, the mechanisms involved have been scarcely evaluated so far. Hence, the aim of this study is to evaluate the effect of hyperglycemia in a rat model of SAH using Magnetic Resonance Imaging (MRI) and neurological evaluation.


In vivo MRI imaging studies using different sequences: T2W, T2*(SWI), T2map, T1W high resolution (pre and post contrast) and Dynamic Contrast Enhanced (DCE) were carried out to evaluate the effect of hyperglycemia on bleeding, brain damage, blood brain barrier disruption (BBBd) and neurofunctional outcome at days 1 and 3 after SAH onset. Two hyperglycemic ranges were induced by intraperitoneal administration of 2,5 ml or 1,5 ml of dextrose at 30 minutes before SAH. Blood samples were withdrawn before (basal) and immediately after SAH (30 minutes after the injection) to measure glucose levels. Finally, the neurological outcome after SAH was measured using a neurological test based in a 0 (highest handicap) to 18 (normal) point-scale at days 1 and 3 after SAH.


Normoglycemic (n=19) and hyperglycemic rats (n=21 for 2,5 ml and n=14 for 1,5 ml) were used to calculate the survival rate. Hyperglycemic animals showed lower survival rates at different days after SAH in comparison with normoglycemic animals. In addition, these results evidenced direct correlation between blood glucose levels and mortality. Besides, hyperglycemic rats showed a significant increase of the haemorrhagic volume after SAH in comparison with normoglycemic animals using MRI (Figure 1). Within the hyperglycemic group, higher blood glucose levels after SAH promoted an earlier increase of stroke lesions, oedema and BBBd. Finally, hyperglycemia induced a major impairment of neurofunctional evolution after SAH. Therefore, these results suggest that hyperglycemia worsens SAH evolution through the increase of bleeding, brain damage and neurofunctional outcome at the acute and sub-acute phases following SAH.


These results suggest that hyperglycemic condition during SAH worsen the pathological evolution of preclinical SAH increasing the mortality rate, volume of haemorrhage, injury around the lesion and disruption of the BBB. Likewise, these findings are in agreement with the effect of high blood glucose levels after clinical SAH on the increase of complications and the higher risk of death or functional disability.


The authors would like to thank the Foundation-La Marató de TV3 for financial support.

[1] Macdonald, et al., Spontaneous subarachnoid haemorrhage. Lancet, 2017. 389(10069): p. 655-666.
[2] Lucke-Wold BP et al. Aneurysmal Subarachnoid Hemorrhage and Neuroinflammation: A Comprehensive Review. Int J Mol Sci. 2016;17(4):497.
[3] Kruyt, N.D., et al., Hyperglycemia and clinical outcome in aneurysmal subarachnoid hemorrhage: a metaanalysis. Stroke, 2009. 40(6): p. e424-30.
Figure 1

Hyperglycemia triggers an increase in the haemorrhagic volume after SAH in comparison with normoglycemic animals using T2W and SWI-MRI

Keywords: Subarachnoid haemorraghe, SAH, MRI, hyperglycemia, blood brain barrier

Long-term follow-up of a rat model of transient ischemic stroke: protocol design and characterization of neurofunctional and imaging endpoints in the context of stem cell therapy

Chloe Dumot1, Lucille Capin2, Violaine Hubert1, Elodie Ong1, Radu Bolbos3, 5, Fabien Chauveau4, 5, Emmanuelle Canet-Soulas1, Celine Auxenfans2, Marlene Wiart1, 5

1 CarMeN Laboratory, Institut National de la Santé et de la Recherche Médicale U1060, INRA U1397, Université Lyon 1, INSA Lyon, Lyon, France
2 Tissue and cell bank, HCL, Lyon, France
3 Cermep, Lyon, France
4 BIORAN Team, Lyon Neurosciences Research Center, CNRS UMR5292, Inserm U1028, Université Claude Bernard Lyon 1, Lyon, France
5 CNRS, Lyon, France


Stem cell therapy is a promising ischemic stroke treatment; however, there is little consensus regarding the most appropriate protocol and relevant endpoints for pre-clinical evaluation [1]. The global aim of the study was to design a protocol for the long-term assessment of stroke outcome in a rat model of transient middle cerebral artery occlusion (tMCAO) using staircase task [2] coupled with diffusion tensor imaging (DTI) [3]. The specific aim of the study was to assess neurofunctional and imaging endpoints in sham vs tMCAO rats and to calculate sample size for a future therapeutic study.


Fig 1A presents the protocol. In brief, 25 male Sprague Dawley rats were trained for the staircase task and then submitted to surgery. To assess the safety of intracerebral administration, stem cells were injected at week 1 (W1) in half of tMCAO rats. The staircase task was performed 4 times a week during 5 weeks. The number of pellets retrieved with forepaws were averaged to obtain one score per week. Side bias was calculated as: contra/(ipsi + contra). MRI was obtained at D3 and D35 post-surgery. Lesion size was measured on T2-weighted imaging. The conventional DTI metrics (AD, RD, MD and FA) were measured in the ipsilateral and contralateral striatum. Student’s t tests and Pearson correlation were used for statistical analysis. Data are presented as mean ± standard deviation.


Fig 1B presents the CONSORT-like chart: 10 rats were included in the analysis, 6 tMCAO and 4 shams. Fig 2A-B shows results for sham vs tMCAO rats. The ipsilateral forepaw of tMCAO rats (corresponding to brain spared side) regained sham level at W2, while the contralateral one (brain ischemic side) showed prolonged deficits (side bias at W5: tMCAO 19±17% vs sham 51±5%, p=0.003). The lesion shrank between D3 and D35 (tMCAO: -39±37%). DTI metrics AD, RD and MD significantly increased at D35 in the ipsilateral striatum of tMCAO rats (Fig 2B for MD; D3 vs D35, ipsilateral side: p=0.007; D35, ipsi vs contralateral side: p=0.001). There was an inverse correlation between MD and side bias at W5 (R2=0.87). Fig 2C-D shows results for treated vs untreated rats. There was no difference between groups in staircase task (Fig 2C; side bias at W5: treated 22±18% vs untreated 16±18%, p=0.34), lesion shrinkage (treated -52±17% vs untreated -26±51%, p=0.22) and DTI metrics (Fig 2D for MD; D35: p=0.28).


We characterized neurofunctional and imaging endpoints for monitoring long-term outcome in tMCAO vs sham rats and checked the innocuity of intracerebral cell administration. The protocol may be simplified with endpoints evaluation at 3 time points only (before tMCAO, W1 and W5). 23 rats per group are needed to detect an intermediate bias value of 35±20% at W5 (0.8 power and 0.05 alpha error). Optimal cell therapy regimen remains to be determined.

AcknowledgmentThe authors thank Jean-Baptiste Langlois of the Animage platform (CERMEP, Lyon) for his technical assistance. We would like to thank Claire Rome from Grenoble Institut des Neurosciences (Grenoble, France) and Gaël Malleret from Centre de Recherche en Neurosciences de Lyon (Lyon, France) for fruitful exchanges about the therapeutic and neurobehavioral protocol. This research was funded by the French national research agency (ANR) project Breakthru (ANR18-CE19-0003) and was performed within the framework of the RHU MARVELOUS (ANR16-RHUS-0009) of University Claude Bernard Lyon (UCBL), within the program “Investissements d’Avenir”.
[1] 'Stem Cell Therapies as an Emerging Paradigm in Stroke (STEPS): bridging basic and clinical science for cellular and neurogenic factor therapy in treating stroke’, Stroke, 2009. 40(2): 510-515.
[2] Trueman, R.C., et al., Systematic and detailed analysis of behavioural tests in the rat middle cerebral artery occlusion model of stroke: Tests for long-term assessment. J Cereb Blood Flow Metab, 2017. 37(4):1349-1361.
[3] Rudrapatna, S.U., et al., Can diffusion kurtosis imaging improve the sensitivity and specificity of detecting microstructural alterations in brain tissue chronically after experimental stroke? Comparisons with diffusion tensor imaging and histology. Neuroimage 2014. 15(97):363-73.
Figure 1- Overview of experimental protocol (A) and CONSORT-like chart (B).

The staircase task was performed as recommended by Trueman JCBFM 2017. One pellet was placed in each stair (maximum of pellet retrieval: 7 by side, arrows). Rats were placed in the box for ten minutes and filmed to assess the number of pellets retrieved with the forepaw on each side. 500,000 human adipose-derived stem cells were injected in a 10-µL volume. DTI metrics (axial diffusivity or AD, radial diffusivity or RD, mean diffusivity of MD, fractional anisotropy or FA) were measured with DSI studio ((http://dsi-studio.labsolver.org/) in the striatum. All analyses were performed blindly.

Figure 2- Neurofunctional and imaging endpoints.

(A) Staircase task in sham vs tMCAO rats. Graph shows the number of pellets retrieved with the forepaw by side (ipsilateral forepaw: brain spared side & contralateral: brain ischemic side) for each group. Surgery was performed so as to impair the dominant side; (B) Mean diffusivity (MD) from DTI data in the striatum of sham vs tMCAO rats. P<0.05 *D3 vs D35 ipsilateral side of tMCAO, §D35 contralateral vs ipsilateral side of tMCAO. (C) Staircase task in treated vs untreated group; (D) MD in ipsilateral striatum of tMCAO rats: treated vs non-treated group. All lesions have striatum involvement.

Keywords: stroke, dti, neurobehavior, chronic stage, cell therapy

Donepezil restores glucose metabolism in a mouse model of Alzheimer’s disease: an 18F-FDG positron emission tomography (PET) imaging study.

Gaëlle Hugon1, Sebastien Goutal2, Maud Goislard1, Nicolas Tournier1, Alexandra Winkeler1

1 UMR 1023, Service Hospitalier Frédéric Joliot, CEA, Inserm, Université Paris Sud, CNRS, Université Paris-Saclay, Orsay, France
2 MIRCen, CEA/IBFJ/DRF/LMN; UMR CEA CNRS 9199-Université Paris Saclay, Fontenay-aux-Roses, France


Donepezil (DPZ) is a reversible acetylcholinesterase inhibitor used to limit cognitive decline in patients with Alzheimer’s disease (AD). The efficacy of DPZ to improve cognitive function has been demonstrated in preclinical and clinical studies (1). However, cholinergic side effects often occur, which may affect the quality of life. Nowadays, the use of this drug is controversial because only modest beneficial effects are associated with treatment. Here we investigate treatment response of DPZ on glucose metabolism using 18F-FDG PET, recognized as a clinical biomarker of AD progression (2).


A mouse model of AD with a unilateral intracerebroventricular (icv) injection of peptide Aβ25-35 (3) was used to investigate low dose (0.25 mg/kg) effect of DPZ on the cerebral glucose consumption. Three groups of Swiss mice were treated once daily by oral gavage. Groups were defined as follows: i) wt-vehicle; ii) Aβ-vehicle; iii) Aβ-DPZ low dose. 18F-FDG PET imaging was performed 7 days after initiation of the treatment. 18F-FDG was administered intraperitoneally in awake animals, and static PET acquisitions were performed 50 min post radiotracer injection. Standard Uptake Values were calculated and normalised to the cerebellum. Statistical parametric mapping was performed for voxel-wise comparison of brain PET images.


Induction of AD by icv injection of Aβ25-35 peptide induced a significant decrease of 18F-FDG uptake in the Aβ-vehicle group at injection side. Despite expected circulation of this peptide through the brain ventricular system, this lateralisation effect may be due to low diffusion of Aβ25-35 peptide. This phenomenon was observed in rats after bilateral icv injection of Aβ25-35 peptide, demonstrating a diffusion gradient from the ventricles to the different brain regions (4). In another study, icv Aβ25-35 injection induced memory impairment, and DPZ treatment (0.1mg/kg) restored some memory functions (5). Yet, DPZ effect on glucose consumption in vivo was not investigated in this model. Interestingly, we found that Aβ-DPZ low dose mice did not show significant differences compared to the wt-group at injection side, indicating restoration of cerebral glucose consumption after DPZ treatment. (Figure 1)


In this study, 18F-FDG PET underlines at the same time the decrease of glucose consumption after icv Aβ25-35 injection and its restoration after DPZ low dose treatment to basal level as observed in wt mice. Thus, 18F-FDG PET is a pertinent strategy to assess the efficacy of DPZ treatment and may be of use for other treatment strategies in this icv mouse model of AD.

[1] Shin CY, Kim H-S et al, 2018. 'The Effects of Donepezil, an Acetylcholinesterase Inhibitor, on Impaired Learning and Memory in Rodents'. Biomol Ther. 1;26(3):274–81.
[2] Shimada A, Hashimoto H et al, 2011. 'Evaluation of therapeutic response to donepezil by positron emission tomography'. Osaka City Med J. 57(1):11–9
[3] Maurice T, Lockhart BP et al, 1996. 'Amnesia induced in mice by centrally administered beta-amyloid peptides involves cholinergic dysfunction'. Brain Res.15;706(2):181–93.
[4] Zussy C, Brureau A et al, 2011. 'Time-course and regional analyses of the physiopathological changes induced after cerebral injection of an amyloid β fragment in rats'. Am J Pathol. 179(1):315–34.
[5] Tsunekawa H, Noda Y, et al, 2008. 'Synergistic effects of selegiline and donepezil on cognitive impairment induced by amyloid beta (25-35)'. Behav Brain Res. 19;190(2):224–32
Figure 1
Keywords: 18F-FDG PET, Alzheimer's mouse model, Donepezil

Metabolic changes in focal brain ischemia in rats treated with human induced pluripotent stem cell-derived neural precursors confirm the beneficial effect of transplanted cells

Natalia Ziółkovská1, 2, Nataliya Romanyuk3, Pavla Jendelová3, Daniel Jirák2, 1

1 Charles University, First Faculty of Medicine, Prague, Czech Republic
2 Institute for Clinical and Experimental Medicine, Department of Computed Tomography, Magnetic Resonance and Clinical Experimental Spectroscopy, Prague, Czech Republic
3 Czech Academy of Sciences, Institute of Experimental Medicine, Department of Neuroregeneration, Prague, Germany


Transplantation of induced pluripotent stem cell-derived neural precursor (iPSC-NP) represents an intriguing therapeutic approach to neural regeneration1. In this study, we transplanted iPSC-NPs in ischemic stroke rat model and monitored its effect on neuroregeneration using magnetic resonance imaging (MRI; lesion volume changes) and spectroscopy (MRS; metabolite concentration changes: Inositol (Ins), Taurine (Tau), Glycerophosphocholine+Phosphocholine (GPC+PCh), N-acetyl-aspartate+N-acetyl-aspartyl-glutamate (NAA+NAAG), Creatine+Phosphocreatine (Cr+PCr) and Glutamate+Glutamine (Glu+Gln))2.


Focal cerebral ischemia was induced by temporary right middle cerebral artery occlusion (MCAO). Cell-treated animals were transplanted with human iPSC-NP 7 days after MCAO. Both grafted and control animals were divided into groups based on lesion size assessed from T2-weighted MR images (RARE; TR/TE=3/36 ms; ST=9 min 36 s) acquired at 4.7 T: small control lesion-SCL, small treated lesion-STL, big control lesion-BCL, big treated lesion-BTL. The effect of iPSC-NPs on metabolic changes was determined by 1H-MRS (single-voxel PRESS sequence; TR/TE=2.5/13.2 ms; ST=21 min 30 s; water suppression) of both hemispheres 1 month and 4 months after MCAO. Spectra were evaluated using the LC Model.


Typical MR images with visible lesion volume are shown in the figure 1. The greatest lesion decrease was observed in SCL group, probably due to a spontaneous recovery of less extensively damaged brain tissue not subjected to invasive transplantation. The initial lesion size increase followed by a decrease in BTL group was possibly caused by necrotic processes slowing down in time and grafted cells having time to incorporate to the lesioned tissue. Most of measured metabolite levels were higher on the contralateral side and differ in time (Fig. 2). Spectroscopic results from lesion side correlates with volumetric measurements with the highest correlation observed in both time points for NAA+NAAG (1 month: r=-0.83, p<0.001; 4 months: r=-0.87, p<0.001).


Our data indicates that iPSC-NP transplantation may be an interesting tool for post-ischemic tissue repair especially in big lesions. Metabolite concentration changes measurements may help to elucidate the biochemical changes that occur during early stages of ischemic stroke and benefit early diagnosis. Further research into the ability of iPSC-NPs for the long-term restoration of lesioned tissue is necessary.

AcknowledgmentThe study was supported by Institute for Clinical and Experimental Medicine IKEM, IN00023001; Charles University, First Faculty of Medicine.
[1] Baker et al., 2017, Sci Rep 7(1), 10075.
[2] Saunders, D.E., 2000, Br Med Bull 56(2), 334-345.
Fig. 1

T2-weighted MR images of lesions at three time points in all groups.

Fig. 2
Statistically significant differences in metabolite concentrations. (A) Metabolite concentrations from both hemispheres compared 4 months after MCAO in BTL group. (B) Metabolite concentrations from contralateral hemispheres compared in time in BTL group; *p<0.05, **p<0.01
Keywords: iPSC-NPs, stroke, magnetic resonance imaging, magnetic resonance spectroscopy, metabolic changes

Beyond the trisomy: a magnetic resonance exploration of the Dp1Tyb Down syndrome mouse model brain

Maria Elisa Serrano Navacerrada1, Eugene Kim1, Diana Cash1, Bernard Siow2, Nisha Singh3, Andre Strydom4, Darryl Hayward5, Dorota Gibbins5, Elizabeth M. Fisher6, Victor Tybulewicz5

1 King's College of London, Neuroimaging, London, United Kingdom
2 The Francis Crick Institute, In vivo imaging, London, United Kingdom
3 University of Oxford, Pharmacology, London, United Kingdom
4 King's College of London, Forensic & Neurodevelopmental Sciences, London, United Kingdom
5 The Francis Crick Institute, Immune Cell Biology Laboratory & Down Syndrome Laboratory, London, United Kingdom
6 University College London, Institute of Neurology, London, United Kingdom


Down syndrome (DS) is one of the most common birth defects, affecting 1/1000 live births worldwide. The main cause of DS in humans is clear: the trisomy of chromosome 21 (Hsa21). However, the complexity of modelling this disease both genetically and phenotypically has led to the development of multiple DS models [1]. Amongst them, Dp1Tyb mice present the main human-like genetic alterations and some DS-like phenotypes, such as cardiac defects, and learning and memory deficits [2]. Nevertheless, some neurobiological aspects of this model are still unknown, requiring further research.


Dp1Tyb mice (DS model) were bred at the Francis Crick Institute (London, UK) [3]. These animals have three copies of the entire 23 Mb region of Mmu16 that is orthologous to a large region of Hsa21. For this exploratory study, we used five Dp1Tyb mice and eight wildtype (WT) littermates of both sexes. These animals were scanned at three months old with a 9.4T Bruker BioSpec horizontal bore scanner. Two MR sequences were employed: 1) single voxel 1H spectroscopy (PRESS, to study changes in the metabolites in the hippocampus); 2) T1-weighted 3D structural imaging (MP2RAGE, to study morphometric changes in the brain). The data were pre-processed and analysed with different software and toolboxes, including FID-A, LCModel, ANTs, QUIT and FSL.


The Student’s t-test highlighted significant differences between Dp1Tyb and WT mice in the concentration of three hippocampal metabolites (see Figure 1). Glutamine (Gln) was significantly increased in the Dp1Tyb mice, compared with WTs (t(11) = -3.305, p < .01). In contrast, the concentrations of N-Acetylaspartate (NAA) and Taurine (Tau) were significantly lower in the Dp1Tyb mice, compared to WT (t(11) = 2.983, p < .05 ; and t(11) = 2.852, p < .05 respectively). Furthermore, differences in the visually observable brain shape between Dp1Tyb and WT mice were quantified at a voxel-wise level using deformation-based morphometry. This group comparison highlighted a reduction in the volume of some DS brain regions, such as the substantia nigra, the subiculum, and the hippocampus (see Figure 2).


The Dp1Tyb mice appear to capture the main anatomic [3] and biochemical characteristics [4] observed in DS. Interestingly, the alterations in NAA and Gln have also been reported in other disorders such as Alzheimer’s disease [5]. Therefore, the characterisation of this model could help us investigate the genotype-phenotype relationship observed in DS, offering insight about the common mechanisms behind different diseases.


VT was funded by the Francis Crick Institute which receives its core funding from the Medical Research Council, Cancer Research UK and the Wellcome Trust, and, with EF, by a grant from the Wellcome Trust.

[1] Herault, Y, et al. 2017, ‘Rodent models in Down syndrome research: impact and future opportunities’, Dis Model Mech., 10 (10): 1165–86
[2] Lana-Elola, E, et al. 2016, ‘Genetic dissection of Down syndrome-associated congenital heart defects using a new mouse mapping panel’, Elife 5, e11614.
[3] Beacher, F, et al. 2010, ‘Brain anatomy and ageing in non-demented adults with Down's syndrome: an in vivo MRI study’, Psychological medicine, 40.4: 611-619.
[4] Lin, A-L, et al. 2016, ‘1H-MRS metabolites in adults with Down syndrome: effects of dementia’, NeuroImage: Clinical, 11: 728-735.
[5] Lamar, M, et al. 2011, ‘Down syndrome with and without dementia: an in vivo proton magnetic resonance spectroscopy study with implications for Alzheimer's disease’, Neuroimage, 57.1: 63-68.
Figure 1

Differences between 3-month-old Dp1Tyb (DS model) and wildtype mice (WT) in the hippocampal concentration of three metabolites, calculated from the in vivo MRS spectrum. The plots display the individual values for each animal, and the mean for each group is represented by an horizontal line. Statistically significant group differences have been calculated with the Student’s t-test, with *p < .05 **p < .01

Figure 2
A map of voxel-wise differences in volume between 3-month-old Dp1Tyb (Down syndrome model) and wildtype mice calculated from in vivo MR images, overlaid on the T1-weighted study-specific template. The color of the overlay indicates the percent volume difference (cool colors indicate reduced volume in Dp1Tyb mice compared to wildtypes) and the opacity of the overlay indicates the significance of the volume difference. Black contours demarcate regions where the FWE-corrected p < .05.
Keywords: Down syndrome, Dp1Tyb model, single voxel 1H spectroscopy, T1-weighted 3D MR

In vivo tracer biodistribution, pharmacokinetic and targeted PET/CT imaging in dogs using a Gallium-68 labelled NK1 receptor imaging agent for localization of chronic pain

Janine Suthiram1, 2, Thomas Ebenhan2, 3, Jan Rijn Zeevaart4, 1, 3, Mike M. Sathekge5, 3

1 The South African Nuclear Energy Corporation (Necsa SOC Ltd.), Radiochemistry, Pelindaba, South Africa
2 University of Pretoria, Department of Nuclear Medicine, Pretoria, South Africa
3 Nuclear Medicine Research Infrastructure/NPC, Preclinical Imaging Facility, Pretoria, South Africa
4 Preclinical Drug Development Platform, Preclinical Drug Development Platform, Potchefstroom, South Africa
5 University of Pretoria, Nuclear Medicine & Steve Biko Academic Hospital, Pretoria, South Africa


NK1 receptor availability has been implicated as playing a role in chronic pain disorders [1]. Pre-clinical evaluation of the effect of systemic blockade of NK1 receptors in rodents, though promising, did not have a similar analgesic effect in patients [2]. This study aims to evaluate biodistribution studies of Gallium-68 (68Ga) radiolabelled DSP, an 11-mer peptide with specificity for the NK1 receptor, in healthy dogs and in dogs with chronic osteoarthritic symptoms. For clinical translation, domesticated animals (e.g., cats and dogs) are considered to be better models than rodents.


DSP (50 μg), buffered to a pH of 3.5‑4, was incubated for 15 min at 95 ˚C with 1 ml of  68Ga-activity obtained by fractionated elution from a tin-dioxide-based 68Ga/68Ge generator (0.6M HCl) (312 MBq ± 119 MBq; n=15). The percentage labeling efficiency (% LE) was determined using ITLC-SG paper (0.1 M citrate, pH 5) to determine free 68Ga and 68Ga-product; for colloidal 68Ga detection:  2: 1 M ammonium acetate/methanol 1:1 (v/v).  Sterile filtered [68Ga]Ga-DSP was administered intravenously to healthy and diseased dogs. Tracer biodistribution was evaluated within 150 min using 3 static PET/CT image acquisitions. The SUV (g/mL) quantification was achieved by 3D-VOI (volume of interest) analysis. Tracer represented in arterial blood and urine samples was quantified.


Instant labelling of [68Ga]Ga-DSP (99.0±0.7%, n=4) yielded doses with a specific activity of 18±4 MBq/nmol. Colloidal-68Ga evaluation was negative. Highest tracer uptake occurred in the urinary bladder (5.6 MBq, SUV =44 g/ml) and kidneys, i.e. rapid renal excretion. Due to first pass effects, considerable tracer uptake was observed in liver (3.3 MBq, SUV =3.5 ) and intestines (1.8 MBq, SUV =0.9 ), decreasing gradually over 120 min. Minimal uptake was seen in other organs (SUV≤ 0.5) (Figure 1). 3/3 dogs demonstrated abnormal tracer uptake in the leg, pelvis and shoulder regions. Intense unilateral [68Ga]Ga-DSP accumulation was seen in the hip region of a dog with chronic osteoarthritic symptoms. Target-to-nontargeted ratios for diseased dogs increased from 2.15±0.52 at 60 min p.i. to 2.76±0.72 at 120 min p.i. Time-activity-curves (n=6) yielded a pharmacological t½ of 11-15 min and an excretion rate of 15-47 MBq/hour for [68Ga]Ga-DSP; 29-46% of activity recovered in urine over 150 min.


Optimised [68Ga]Ga-DSP labeling was accomplished with high radiochemical purity. In vivo, the tracer exhibited minimal non-specific organ uptake and rapid renal elimination. We were able to successfully demonstrate the targeting of  “painful sites” in a dogs with chronic pain. These findings merit further preclinical and clinical testing concerning the use of [68Ga]Ga-DSP as a PET/CT tracer for the diagnosis of chronic pain disorders.


The authors wish to thank the Nuclear Medicine Department at Steve Biko Academic Hospital for access to the 68Ga/68Ge generator and Delene van Wyk for her role as the radiographer during the pre-clinical study.

[1] Peterson, M, et al., 2013, PET-Scan Shows Peripherally Increased Neurokinin 1 Receptor Availability in Chronic Tennis Elbow: Visualizing Neurogenic Inflammation?, 8, e75859.
[2] Traub, RJ, 1996, The spinal contribution of substance P to the generation and maintenance of inflammatory hyperalgesia in the rat, Pain, 67, 151-161.
Figure 1

Representative PET/CT maximum intensity projection (MIP) image, 60 min post administration of [68Ga]Ga-DSP.

Keywords: NK1 receptors, chronic pain, gallium-68, imaging guided biodistribution, pre-clinical

Machine learning protocols and network analysis of Ca2+ fluorescence imaging after ALS IgG action on cultured astrocytes and neurons

Pavle Andjus1, Andrej Korenić1, Milena Milošević1, Dunja Bijelić1, Bilal E. Kerman3, Abulkerim Çapar2, Srdjan Antic4, Wim V. Drongelen5

1 University of Belgrade Faculty of Biology, Center for laser microscopy, Belgrade, Serbia
2 Istanbul Technical University, Istanbul, Turkey
3 Istanbul Medipol University, Istanbul, Turkey
4 University of Connecticut Health, Dept. of Neuroscience, Farmington, United States of America
5 University of Chicago, Dept. of Pediatrics, Chicago, United States of America


The project “Automated Functional Screening of immunoglobulins IgGs for Diagnostics of Neurodegenerative Diseases” (AUTOIGG) aims towards the design of an innovative automated multifunctional device for diagnostics of neurodegenerative diseases. The main goal is to develop cellular models and procedures with patient IgGs for diagnostic procedures related to neurodegenerative diseases (particularly amyotrophic lateral sclerosis - ALS). We will report on the preliminary signal analysis protocols for Ca2+ imaging data obtained with fluorescent probes in rodent neural cells in culture.


Rat astrocytes in culture were loaded with Fluo-4 AM and murine hippocampal pyramidal neurons in culture with Oregon Green 488 BAPTA-1 AM. Recordings on astrocytes were made on Evolve 512 EMCCD Digital Camera System (Photometrics), while neurons were imaged at DIC15 (when demonstrating action potentials) in 5% CO2/95% O2 ACSF with low resolution, high sensitivity NeuroCCD camera (RedShirtImaging). The analysis was performed by developing machine learning protocols on astrocytes or applying cross-correlation analysis on neuronal signalling (van Drongelen, 2007).


Raw data from primary rat cortical astrocytes were processed with Unsupervised Machine Learning algorithms in order to train a classifier to reliably predict a class for any given test trace. Clustering with Decision Trees along with Block Bootstrapping for time series to model the traces, as well as k-Nearest Neighbours (Fig.1) classification gave us satisfactory preliminary results.
After ALS IgG application the cross-correlograms of the Ca2+ signaling in hippocampal pyramidal neurons in culture revealed a synchronization that was apparent in highly correlated activity (correlation coefficient > 0.9) across cell pairs at physiological delays. Control human IgGs (commercial or from a healthy subject) have not shown similar highly synchronous behaviour.


Thus, in addition to an innovative technical solution for the use of cell models in a lab-on-chip device we have started developing robust signal analysis protocols on cells in vitro as induced by immune humoral factors that may serve for better stratification of patients for personalized medicine.

AcknowledgmentSupported by H2020 MSCA-RISE project No 778405 - AUTOIGG.
[1] van Drongelen W. 2007. Signal Processing for Neuroscientists: Introduction to the Analysis of Physiological Systems. Amsterdam: Elsevier
Cortical astrocyte activity

Starting Ca2+ trace upper left, and increasing number of the closest training examples (clockwise).

Keywords: ALS, calcium fluoreescence imaging, machine learning, neuronal cell cultures, astrocytes