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).

Imaging Infection

Session chair: Patrick Van Dijck (Leuven, Belgium); Patricia Wenk (Magdeburg)
Shortcut: PS 04
Date: Wednesday, 26 August, 2020, 10:00 a.m. - 11:30 a.m.
Session type: Parallel Session


Abstract/Video opens by clicking at the talk title.

10:00 a.m. PS 04-01

Introductory Lecture

Greetje Vande Velde1

1 KU Leuven, Leuven, Netherlands

10:18 a.m. PS 04-02

A novel MRI-based approach to study virulence differences of clinical Cryptococcus neoformans strains in a murine model

Liesbeth Vanherp1, Alexandre Alanio2, Amy Hillen1, Jennifer Poelmans1, Katrien Lagrou3, Greetje Vande Velde1, Uwe Himmelreich1

1 KU Leuven, Biomedical MRI, Department of Imaging and Pathology, Leuven, Belgium
2 Pasteur Institute, Molecular Mycology, Department of Mycology, Paris, France
3 KU Leuven, Clinical Bacteriology and Mycology, Department of Microbiology, Immunology and Transplantation, Leuven, Belgium


The virulence of the infective strains is an important contributor to the outcome of brain infections caused by Cryptococcus spp. (1). Typically, virulence is studied by infecting large groups of animals and recording survival or analyzing fungal burden of the infected organs ex vivo, but this approach does not provide insights in how the disease establishes and progresses. The aim of our study was to assess the virulence of different clinical strains and the associated variability in the development and presentation of brain infection in mice by in vivo MRI.


13 clinical C. neoformans strains or reference strain C. neoformans H99 were injected intravenously (50 000 cells) in Balb/C mice (n=3 per Cryptococcus strain) to induce cryptococcal meningo-encephalitis. 3D anatomical brain MRI scans (T2-weighted RARE, 94 µm resolution) were acquired every 2-4 days using a 9.4T preclinical MRI (Bruker BioSpin), starting from day 3 after infection and continuing until animals reached humane endpoints. The resulting images (n = 253) of the infected mouse brains were quantified using an in-house developed pipeline (MeVisLab, ITK-SNAP and ImageJ) to obtain multi-parametric disease- and morphology-related read-outs such as total brain volume, ventricle volume, volume of different disease-related features, number of brain lesions and average lesion volume.


Survival of the animals ranged from 7 to 39 days, with symptoms only being present in the last stages of the disease. In contrast, MRI already indicated differences in disease onset and presentation on day 3 (Fig.1). In general, cryptococcal infection of the brain induced enlargement of the ventricles and increased total brain volume, consistent with edema of the brain tissue and hydrocephalus. MRI showed the presence of parenchymal lesions, fluid accumulation around major vasculature and involvement of the meninges. Even when survival rates were similar, the relative contributions of these different parameters were highly variable between strains, but not between animals infected with the same strain. The phenotypical presentation of disease in mice was compared to in vitro virulence factors of the Cryptococcus strains (growth rate, …) and patient data (1) to potentially unravel factors that contribute to certain aspects of disease.


MRI provided longitudinal insights in disease development and progression in a preclinical model and the multiple MRI read-outs allowed for a detailed comparison with both in vitro virulence data and clinical data. This novel non-invasive approach maximizes the amount of information obtained from single animals and can provide a unique basis for understanding the impact of strain variability on disease severity and presentation.

AcknowledgmentWe are thankful for financial support to the Infect-ERA project CryptoVIEW. LV is an SB PhD fellow at Research Foundation Flanders (FWO).
[1] Alanio, A, Desnos-Ollivier, M, Dromer, F. ‘Dynamics of Cryptococcus neoformans-macrophage interactions reveal that fungal background influences outcome during cryptococcal meningoencephalitis in humans’, MBio. 2011;2(4):e00158–11.
Fig. 1: Longitudinal MRI data for mice infected with C. neoformans strains H99, AD4-76a and AD2-04a.
(A) T2-weighted MR images for 3 representative animals, (B) corresponding segmentations of the 3D scans, (C) a 3D model of the brain at the last time point, (D) cross-sectional quantitative data obtained at day 7, showing mean + SD of 3 animals. Total brain and ventricle volume are expressed as percentage increase relative to day 3 (left axis), while the volume of lesions, fluid accumulation and involvement of the meninges are reported in mm³ (right axis).
Keywords: Infection, MRI, Brain, Pathogenesis, Mouse model
10:30 a.m. PS 04-03

Developing a radiolabelled tracer to visualize Plasmodium falciparum infection in vivo

Janie Duvenhage1, 3, Thomas Ebenhan2, 3, Lyn-Marie Birkholtz1, Seike Garny1, Jan Rijn Zeevaart4, 5, 3

1 University of Pretoria, Department of Biochemistry, Genetics and Microbiology, Pretoria, South Africa
2 University of Pretoria, Department of Nuclear Medicine, Pretoria, South Africa
3 Nuclear Medicine Research Infrastructure, Preclinical Imaging Facility, Pretoria, South Africa
4 North West University, Preclinical Drug Development Platform, Potchefstroom, South Africa
5 South African Nuclear Energy Corporation, Radiochemistry, Pelindaba, South Africa


Malaria, caused by Plasmodium (P) falciparum, can lead to death in untreated patients. A major limitation of understanding this infection is the inability to visualise the intricate host-parasite mechanisms that governs malaria pathogenicity in situ. Non-invasive molecular imaging techniques could aid investigating malaria-related pathologies in disease targeted organs; hence we set out developing Zirconium-89 (89Zr) radiolabeled malaria-specific tracers, i.e. a Plasmodium-specific antibody (IIIB6) and single-chain variable fragments (scFv) to assess their biodistribution by PET/CT imaging.


IIIB6 was tested for its ability to recognize all intra-erythrocytic parasite stages after the stage-specific production and evaluation of each of its developmental stage with flow cytometry (FC) and confocal microscopy (CM). 89Zr-radiolabeling was achieved by conjugating IIIB6 to Bz-DFO-NCS. The stability and organ uptake of [89Zr]IIIB6 was studied in healthy mice by imaged-guided biodistribution over 24 h using μPET/CT. Its usefulness was validated ex vivo against [89Zr]h-R3, a humanized monoclonal antibody against the EGFR, and [89Zr]oxalate (control). Phage display-deriving scFvs were selected against enriched trophozoite/schizont-infected erythrocytes. Single soluble scFv clones were identified; sequenced and binding properties determined via surface plasmon resonance.


FC/CM-analyses showed IIIB6 recognising relevant P. falciparum-infected erythrocyte stages (Figure 1). μPET/CT image analysis indicated a pharmacological half-life of 9.6 ± 2.5 h for [89Zr]IIIB6. High hepatic tracer concentrations was determined at 2–24 h followed by spleen >kidneys >heart >stomach and >lung (Figure 2). Since [89Zr]IIIB6 activity in lung tissue decreased desirably over time, pulmonary sequestration of parasites could be a suggested imaging scenario with [89Zr]IIIB6-PET. [89Zr]IIIB6 and [89Zr]h-R3 had similar hepatic accumulation indicating that the liver was not a target organ for IIIB6 but rather served as part of the hepatobiliary excretion pathway (expected for antibodies). The ex vivo biodistribution studies shows no evidence of 89Zr-dissociation from IIIB6 as indicated by bone uptake compared to [89Zr]oxalate. Suitable scFv clones were identified warranting further evaluation.


A proof-of-concept radiolabelling of a P. falciparum-specific pan-reactive antibody bioconjugate for in vivo molecular imaging was achieved. [89Zr]IIIB6-PET/CT imaging demonstrates that further characterisation of this antibody is needed before possibly continuing in malaria-infected animals. The developed and characterized P. falciparum-specific scFvs could potentially succeed approaching malaria infections in real time with PET/CT.

AcknowledgmentNuclear Technologies in Medicine and the Biosciences Initiative (NTeMBI) a national technology platform developed and managed by the South African Nuclear Energy Corporation (Necsa) and the Department of Science and Technology.
[1] Kingston, HW, Ghose, A, Plewes, K, Ishioka, H, Leopold, SJ, Maude, RJ, Paul, S, Intharabut, B, Silamut, K, Woodrow, C, Day, NP (2017), Disease severity and effective parasite multiplication rate in falciparum malaria, Open forum infectious diseases Vol. 4, No. 4 p.oxf169
[2] Holland, JP, Divilov, V, Bander, NH, Smith-Jones, PM, Larson, SM, Lewis, JS 2010, 89Zr-DFO-J591 for immunoPET of prostate-specific membrane antigen expression in vivo. Journal of Nuclear Medicine, 51(8):1293-300
[3] Duvenhage,J, Ebenhan, T, Garny, S, Gonzalez, IH, Montana, RL, Price, R, Birkholtz LM, Zeevaart, JR 2019, Molecular imaging of a zirconium-89 labeled antibody targeting Plasmodium falciparum-infected human erythrocytes. Molecular Imaging and Biology, pp.1-9
Figure 1.
Flow cytometry analysis followed by confocal microscopic evaluation of IIIB6 to show binding of the antibody to Plasmodium falciparum-infected erythrocytes. The top histogram depicts early trophozoite stages containing a single nuclei (1N). 1N- >2N represents mature trophozoites stages and >2N represents schizonts multiple nuclear devisions have taken place. For confocal microscopy, IIIB6 was labeled with Alexa-Fluor-647 and the parasite DNA stained with Hoechst. The blue and red channels were merged to show binding of IIIB6 to the parasite-infected erythrocyte
Figure 2
Representation of maximum intensity projections (MIP)  and time-activity curves (TAC) derived from volume of interest analysis of the PET/CT images obtained from athymic nude mice injected with [89Zr]IIIB6. Following tracer injection, microPET/CT images were acquisited for 20 min at 2 h, 4 h, 6 h, 24 h (only 24 h PET/CT scan showed) and the TACs derived from volume of interest analysis of these PET/CT scans. Color scale: kilobecquerel per millilitre (blue = minimum, red = maximum.
Keywords: malaria, Zirconium[89], antibody, plasmodium
10:42 a.m. PS 04-04

Radiolabelled Anti-Staphylococcal antibodies for diagnosis of bacterial infection by immunoPET

Maria Isabel Gonzalez1, 2, Maria Isabel Gonzalez2, Martha Kestler1, Lorena Cusso1, 2, 3, Daniel Calle1, 2, Patricia Muñoz5, 1, Emilio Bouza1, 5, Manuel Desco Menéndez1, 2, 3, Beatriz Salinas1, 3, 4

1 Inst. de Investig. Sanitaria Gregorio Marañón, Experimental Medicine and Surgery Unit, Madrid, Spain
2 Centro Nacional de Investigaciones Cardiovasculares Carlos III, Advanced Imaging Unit, Madrid, Spain
3 Universidad Carlos III de Madrid, Bioengineering and Aerospace Engineering Dpt, Madrid, Spain
4 Centro de Investigación Biomédica en Red de Salud Mental (CIBERSAM), Madrid, Spain
5 Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES), Madrid, Spain


Staphylococcus aureus (S. aureus) is a clinically aggressive Gram-positive bacterium that is one of the most frequent causes of human infection [1]. The numerous complications arising therefrom require a rapid and accurate diagnosis to apply an effective treatment. The lack of specific and sensitive non-invasive diagnostic tools makes it difficult the detection of small infectious foci. As a possible solution, we present a specific anti α-toxin antibody radiotracer for the selective detection of S. aureus infection by immunoPET/CT imaging based on the role of a-toxin as main virulent factor.


anti S. aureus α-toxin antibody (α-ToxAb; 150KDa, 1 mg/ml) was conjugated for 30 min at 37ºC with DFO-NCS (pH 8.8, 30min at 37ºC)  in a 5 fold molar excess of the chelator. DFO- α-ToxAb conjugate was radiolabeled with 89Zr4+ for 1h at 25ºC at pH 5.5 and purified with centrifugal filters units. Radiochemical purity of 89Zr-αToxAb was stablished by radio HPLC (PBS, 0.2 ml/min). Pharmacokinetics of the tracer was evaluated by in vivo blood half-life and ex vivo biodistribution studies (1h, 48h and 72h p.i.). In vivo PET/CT imaging (200-250mCi, 400 mL, PBS) was performed in a local infection-inflammation model of SD rats (1h, 24, 48 and 72h). In this model, left hind leg received intramuscular injection of active S. aureus (0.4mL, 10*8CFU/mL) and right hind leg of an inactive strain of Saureus


Pure 89Zr-αToxAb antibody was synthetized (Fig1.A) yielding a radiochemical purity > 95%, (Fig. 1B) and specific activity of 10.4 mCi/mg. Pharmacokinetic evaluation of the radiolabeled α-toxin antibody confirmed a blood half-life of 2.33 d, typical of Ab with similar size (Fig. 1C). Ex vivo biodistribution studies (Fig. 2B) confirmed lung uptake and high circulation values (blood) at 1 h (2,3 ± 1,0  and 2,6 ± 04 %ID/g respectively) and hepatobiliary metabolism, with main accumulation in liver, spleen and kidneys at longer timepoints. In vivo PET/CT comparison with commercial 18F-FDG confirmed the capacity of our tracer to discern between infection and inflammation, even at short time points (Fig 2.1). These results were confirmed by ex vivo autoradiography (Fig. 2D). In vivo PET-CT evaluation of the 89Zr-DFO-α-toxin at later time points (24, 48 and 72h, Fig1D) showed main uptake in the infected region with no significant uptake in the inactive strain region or any other organ (Fig. 2)


We present the design and characterization of a novel inmunePET tracer specific of S. aureus based on the radiolabeling of anti a-toxin antibody. Results achieved in vivo confirmed the specificity of the radio-antibody toward Staphylococcal infections, allowing distinguishing between inflammatory and infectious processes and probing its potential as platform for diagnosis improving the current approaches employed in the clinic such as 18F-FDG.

AcknowledgmentThe Small Animal Imaging core, especially Yolanda Sierra, Alexandra de Francisco and Maria Felipe. This work was partially supported by Comunidad de Madrid (Y2018/NMT-4949 (NanoLiver – CM and S2017/BMD-3867 RENIM-CM, co-financed by European Structural and Investment Funds),  Ministry of Science, Innovation and Universities (ISCIII-FIS grants PI16/02037, co-financed by ERDF, FEDER, Funds from the European Commission, “A way of making Europe”) and Ayudas Fundación BBVA a Equipos de Investigación Científica 2018 (“Diagnosis and treatment follow-up of severe Staphylococcal Infections with Anti-Staphylococcal antibodies and Immune-PET “)
[1] Baddour, L.M. et al. (2007) Mayo Clin. Proc., 82(10), p. 1163-1164
Figure 1.

A) Radiosynthesis scheme of 89Zr-αToxAb. B) Representative HPLC radioactive chromatograms, C) Blood half-life, D) In vivo PET/CT imaging at 24,48 and 72h in MIP.

Figure 2.

A) PET-CT imaging in a local infection model of radiolabeled antibody, free 89Zr4+ and commercial 18-FDG. B) Ex vivo biodistribution estudies 1h post injection. C) Ex vivo autoradiography.

Keywords: radiotracer, immunoPET, bacterial infection, Staphylococcus aureus, antibody
10:54 a.m. PS 04-05

Light-switchable antibiotics: towards theranostics of bacterial infections

Willem A. Velema2, Michael R. Wegener2, Mickel J. Hansen2, Piermichele Kobauri2, Arnold J. M. Driessen3, Ben L. Feringa2, Wiktor Szymanski1, 2

1 University Medical Centre Groningen, Medical Imaging Center, Groningen, Netherlands
2 University of Groningen, Stratingh Institute for Chemistry, Groningen, Netherlands
3 University of Groningen, Groningen Biomolecular Sciences and Biotechnology Institute, Groningen, Netherlands


Lack of spatiotemporal control over the activity of drugs is a growing concern. In the case of antibiotics, the buildup of active substances in the environment leads to the emergence of bacterial resistance, a major threat for humanity.[1]
A potential solution lies in the design of drugs whose activity can be remotely controlled with high precision. Photopharmacology[2,3] is a rapidly growing field that uses light to reversibly switch on and off the drug activity. In combination with optical imaging, which uses light-emitting probes, photopharmacology lends itself perfectly to theranostics.


Here, our work on developing light-controlled antibiotics will be presented[4-5], with special focus on the design, photochemistry (including red-light activation for better tissue penetration) and antimicrobial activity.
In the photopharmacological approach, bioactive molecules are modified with molecular photoswitches, i.e. moieties that respond to radiation of a certain wavelength of light by changing their shape and properties. Those changes can then be translated to different potency of drugs to which the photoswitch has been introduced. The switching process can be reversed thermally or with light of a different color. The presentation will discuss three generations of light-controlled antibiotics.


1st generation[4], in which a photoswitch was incorporated into a quinolone antibiotic, resulting in a molecule that had low antimicrobial potency in the resting state, and could be activated with UV light to the highly potent isomer. This isomer had a half-life of ~2 hours, which is useful in a treatment where a drug is activated prior to administration and it switches off in time to prevent the build-up of active substance in the environment.
2nd generation [5], in which a photoswitchable trimethoprim analogue was created that could be reversibly switched on with red light. Since such light penetrates deeper into the tissues, this design offers prospects for local activation to avoid antibiotic toxicity to e.g. gut microbiome.
3rd generation [in progress], in which molecular docking and dynamics simulations are being used to rationalize the difference in potency between the photoisomers and develop more potent molecules with larger difference in potency that can be achieved with light.


This research enables light control over antibiotic potency, with the aim to avoid side effects and the emergence of bacterial resistance in the environment. With the first successful designs,[4-5] a proof of principle has been established for molecules that enable control inside and outside of the patient’s body. In the future, we will focus on theranostics application where light-emitting optical tracers can be used to activate the antibiotic.


We gratefully acknowledge generous support from NanoNed, The Netherlands Organization for Scientific Research (NWO-CW, Top grant to B.L.F. and NWO VIDI grant no. 723.014.001 for W.S.), the Royal Netherlands Academy of Arts and Sciences (KNAW), the Ministry of Education, Culture and Science (Gravitation programme 024.001.035), and the European Research Council (Advanced Investigator Grant no. 694345 to B.L.F.).

[1] OECD, 2018, 'Stemming the Superbug Tide: Just A Few Dollars More', OECD Health Policy Studies, OECD Publishing, Paris
[2] Velema, WA, Szymanski, W, Feringa, BL, 2014, 'Photopharmacology: Beyond the Proof of Principle', J. Am. Chem. Soc., 136, 2178
[3] Hoorens, MWH, Szymanski, W, 2018, 'Reversible, spatial and temporal control over protein activity using light', Trends Biochem. Sci., 43, 567
[4] Velema, WA, van der Berg, JP, Hansen, MJ, Szymanski, W, Driessen, AJM. Feringa, BL 2013 'Optical Control of Antibacterial Activity', Nature Chem., 5, 924
[5] Wegener, M, Hansen, MJ,  Driessen, AJM, Szymanski, W, Feringa, BL 2017, 'Photocontrol of Antibacterial Activity: Shifting from UV to Red Light Activation', J. Am. Chem. Soc., 139, 17979
Figure 1.

First (A,B) and Second generation (C-E) of photoresponsive antibiotics. Stuctures (A, C), photopresponsiveness to visible light (D) and antibacterial activity is shown with patterning experiments (B) and microbial growth curves (E).

Keywords: theranostics, photopharmacology, antibiocs, optical control, molecular design
11:06 a.m. PS 04-06

Multimodal imaging in experimental cerebral malaria reveals early stage accumulation of infected erythrocytes and altered brain perfusion

Patricia Wenk1, Rituparna Bhattacharjee2, 4, Annika Michalek3, Dirk Schlüter6, 4, 7, Eike Budinger2, Anja M. Oelschlegel2, 5, Nishanth Gopala6, 4, Jürgen Goldschmidt2

1 Leibniz Institute for Neurobiology, Department of Functional Architecture of Memory, Magdeburg, Germany
2 Leibniz Institute for Neurobiology, Systems Physiology of Learning, Magdeburg, Germany
3 Leibniz Institute for Neurobiology, Special Laboratory for Non-invasive Imaging, Magdeburg, Germany
4 Otto-von-Guericke University, Institute of Medical Microbiology and Hospital Hygiene, Magdeburg, Germany
5 Otto-von-Guericke University, Institute of Anatomy, Medical Faculty, Magdeburg, Germany
6 Hannover Medical School, Institute of Medical Microbiology and Hospital Epidemiology, Hannover, Germany
7 Helmholtz Centre for Infection Research, Braunschweig, Germany


Cerebral malaria (CM), one of the most common neurological diseases, is a severe complication of human malaria caused by Plasmodium falciparum. It has been hypothesized that intravascular sequestration and congestion of infected-RBCs (iRBCs) play a key role in a cascade of events leading to a prothrombotic state, brain edema and death. Direct evidence for this is lacking in humans as well as in animal models. We here study, in P. berghei ANKA (PbA)-infected mice as a model of experimental CM (ECM), brain iRBC accumulation and its potential consequences for brain perfusion using SPECT and MRI.


Mice were infected i.p. with PbA-iRBCs. For whole body and brain SPECT-imaging of iRBC-distributions, iRBCs enriched from blood of infected mice, were labeled with 99mTechnetium. 99mTc-labeled iRBCs were injected i.v. into infected (n=8) and non-infected controls (n=8) at day 5 p.i. Brain perfusion was studied using 99mTcHMPAO-SPECT at day 5 and day 7 p.i. and with MR angiography (MRA). Flow-compensated FLASH-based TOF images were acquired in combination with velocity maps on day 5, 6 and 7 p.i. T2-weighted RARE images were used as anatomical reference and for edema detection.


SPECT-imaging of 99mTc-labeled iRBCs showed substantial and highly significant increases of brain iRBC-contents in infected as compared to control mice. Labeling was most intense in areas of high blood volume as the venous sinuses. SPECT-perfusion experiments showed areas of decreased cerebral blood flow in periventricular regions at day 5 p.i., increasing in extent at day 7 p.i. Velocity maps indicated reductions in flow velocity increasing in severity from day 6 to day 7 p.i. Concordant with severe clinical symptoms, T2-MRI revealed brain edema on day 7 p.i., a time point at which TOF-angiographies showed markedly reduced venous outflow. To the best of our knowledge, our data provide the first in vivo evidence for brain-wide early stage intracerebrovascular iRBC accumulation in experimental CM and in CM in general. Our SPECT and MR-findings of reduced cerebral blood flow suggest that flow deficits can precede MR-detectable edema.


In humans, brain edema is a hallmark of CM and fatal outcomes are related to edema severity. Controversies exist about the underlying pathophysiology and the validity of animal models. Our data argue for similar pathologies in mouse and man and are in line with early theories on CM pathology in humans proposing that intravascular iRBC sequestration and congestion might lead to reduced venous efflux that in turn could contribute to brain edema.

Keywords: Cerebral Malaria, SPECT/MRI, MR-Angiography, Brain Perfusion, Blood Flow