EMIM 2019
To search for a specific ID please enter the hash sign followed by the ID number (e.g. #123).

Imaging in Infection | Immunology

Session chair: Greetje Vande Velde (Leuven, Belgium); Johannes Schwenck (Germany)
 
Shortcut: PW22
Date: Friday, 22 March, 2019, 11:45 a.m.
Room: ALSH | level 0,BOISDALE | level 0,CARRON | level +1,DOCHART | level +1
Session type: Poster

Contents

Click on an contribution to preview the abstract content.

501

Real-time arthroscopic imaging of bacterial biofilms in a human post-mortem model using fluorescently labelled vancomycin (#390)

Jorrit W.A. Schoenmakers1, 2, Marjolein Heuker1, Marina López-Álvarez1, Wouter B. Nagengast3, Gooitzen M. van Dam4, Jan Maarten van Dijl1, Paul C. Jutte2, Marleen van Oosten1

1 University of Groningen, University Medical Center Groningen, Department of Medical Microbiology and Infection Prevention, Groningen, Netherlands
2 University of Groningen, University Medical Center Groningen, Department of Orthopaedics, Groningen, Netherlands
3 University of Groningen, University Medical Center Groningen, Department of Gastroenterology and Hepatology, Groningen, Netherlands
4 University of Groningen, University Medical Center Groningen, Department of Critical Care, Department of Surgery and Department of Nuclear Medicine and Molecular Imaging , Groningen, Netherlands

Introduction

A prosthetic joint infection (PJI) is a serious complication following a total joint replacement. Early and accurate diagnosis is vital to increase the chance of successful treatment and optimise antibiotic use1. Current diagnostic techniques to detect PJIs are, however, time consuming, often non-specific and unable to detect bacterial biofilm2. This proof-of-principle study explores the use of fluorescently labelled vancomycin (vancomycin-IRDye800CW; vanco-800CW)3 as a bacteria-targeting imaging modality to detect Gram-positive bacterial biofilms.

Methods

In this study, we have explored real-time optical imaging and visualised infected knee prostheses using a near-infrared (NIR) fluorescent fiber during arthroscopy in a human post-mortem model. Femoral components of two knee prostheses were coated on the lateral sides with Staphylococcus epidermidis biofilm. The biofilm-coated prostheses were surgically implanted on the distal femurs of a human cadaver, followed by standard watertight suturing. Vanco-800CW was injected inside each knee cavity and after incubation, flushed with NaCl 0.9%. Imaging was performed with both, white and NIR light before and after administration of the tracer. The post-mortem experiments were conducted according to institutional guidelines.

Results/Discussion

The ex vivo knee prosthesis, after incubation with vanco-800CW, shows a strong fluorescent signal from the lateral part containing biofilm, whereas no fluorescent signal was detected of the medial site. Images of the post-mortem experiment, prior to administration of the tracer, showed no fluorescence compared to the background. After injection of vanco-800CW, the biofilm-coated part of the prosthesis emitted a strong fluorescent signal, while the uncoated part showed no fluorescence compared to the background.

Conclusions

The promising results of the post-mortem experiment indicate that bacteria-targeted arthroscopic imaging of bacterial biofilms is possible and can potentially be achieved in clinical setting in the very near future. Aspects of this approach, like high resolution, the possibility to image in real-time and bacteria-specific tracers, make this imaging modality superior to current diagnostic modalities3.  

References

  1. Steward PS. Antimicrobial Tolerance in Biofilms. Microbiol Spectr 2015; 3(3).
  2. Tande AJ, Patel R. Prosthetic joint infection. Clin Microbiol Rev 2014; 27(2):302-345.
  3. van Oosten M, Schafer T, Gazendam JA, Ohlsen K, Tsompanidou E, de Goffau MC, et al. Real time in vivo imaging of invasive- and biomaterial-associated bacterial infections using fluorescently labelled vacomycin. Nat Commun 2013; 4:2584.
Keywords: arthoscopic, biofilm, fluorescent, bacteria-targeted
502

Illuminating Bacterial Quorum Sensing Using Optical Molecular Tools (#39)

Wiktor Szymanski1, 2, Mickel Hansen2, Willem Velema2, Jacques Hille2, Arnold Driessen3, Ben Feringa2

1 University Medical Center Groningen, Radiology, Groningen, Netherlands
2 University of Groningen, Stratingh Institute, Groningen, Netherlands
3 University of Groningen, GBB, Groningen, Netherlands

Introduction

Bacterial communication is of key importance in the regulation of bacterial symbiotic processes and pathogenesis of infections.[1] It is mainly based on quorum sensing (QS); a process in which bacteria produce and excrete QS autoinducers responsible for controlling cellular organization, virulence and biofilm formation, amongst others.[2] External control over the activity of QS autoinducers with light, coupled with optical imaging methods, would enable theranostic regulation of biofilm formation, which is becoming a major consideration in the treatment of bacterial infections.[3,4]

Methods

A small library of light-switchable QS inducers and inhibitors was prepared by introducing an azobenzene molecular photoswitch into the structure of known QS inducer molecules. The photochemical properties of the products were fully characterized using UV-Vis and NMR spectroscopy methods. Biological evaluation focused on the Las network of Pseudomonas aeruginosa, which can be quantified by the induction of LasQS as measured by a functional readout of bioluminescence in a QS reporter strain (E.coli JM109 pSB1075). In this strain, the compounds potentially bind to the transcriptional activator LasR to form a stable dimer, which can bind to the responsive promoter region of the LasQS system proceeding the luxCDABE-lasR promoter fusion reporter genes.

Results/Discussion

At the thermally stable states, several compounds from the library proved to be agonists of quorum sensing with 15-18% induction, compared to the native autoinducer. However, irradiation (λ = 365 nm for 5 min) and subsequent evaluation as quorum sensing agonists revealed a dramatic increase in activity. Especially the compound presented in Figure A stood out in this respect. Subsequent investigation to the dose response of its irradiated and non-irradiated forms (Figure B) revealed a stunning >700-times difference in activity between the irradiated and non-irradiated forms. This represents an unprecedented selectivity in photopharmacology.

Several compounds from the library showed good inhibitory activity against a native autoinducer, albeit with a small difference in activity between the irradiated and non-irradiated states. Notably another compound reported before by our group[3] proved to be the most potent photoswitchable inhibitor in our library with 57% inhibition of QS activity.

Conclusions

We present here a library of photoswitchable QS agonists and antagonists. Up to 71% QS induction was obtained, whereas the best inhibitor showed almost 60% inhibition. Our lead compound, showed privileged properties for further development in optical imaging-coupled photopharmacology, due to an unprecedented difference in activity between the inactive and light-activated state, which could be interconverted without apparent fatigue.

References

[1] Greenberg, E. P. Nature 2003, 424, 134–134.

[2] Welsh, M. A.; Blackwell, H. E. FEMS Microbiol. Rev. 2016, 40, 774–794.

[3] Van der Berg, J. P.; Velema, W. A.; Szymanski, W.; Driessen, A. J. M.; Feringa, B. L. Chem. Sci. 2015, 6, 3593–3598

[4] Hansen, M. J.; Hille, J. I. C.; Szymanski, W.; Driessen, A M.; Feringa, B.L. 2018 submitted

Acknowledgement

This work was financially supported by 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 program 024.001.035), and the European Research Council (Advanced Investigator Grant no. 694345 to B.L.F.).

Photoswitchable Quorum Sensing Inducer
A. The structure of the optimized compound together with its photoswitching process; B. (left) Dose-response curve for the irradiated and non-irradiated compound and their influence on quorum sensing induction; (right) reversible switching of potency by irradiation with light of different wavelengths.
Keywords: Quorum sensing, bacterial infections, photopharmacology, optical imaging
503

Ex vivo imaging of osteomyelitis and implant infections using fluorescently labelled vancomycin (#370)

Marjolein Heuker1, Marina López-Álvarez1, Gooitzen M. van Dam2, 3, Jan Maarten van Dijl1, Frank F.A. IJpma2, Marleen van Oosten1

1 University of Groningen, University Medical Center Groningen, Department of Medical Microbiology and Infection Prevention, Groningen, Netherlands
2 University of Groningen, University Medical Center Groningen, Department of Surgery, Groningen, Netherlands
3 University of Groningen, University Medical Center Groningen, Department of Critical Care and Department of Nuclear Medicine and Molecular Imaging, Groningen, Netherlands

Introduction

Osteomyelitis and implant-associated infections are severe complications after bone fracture treatment1. Clinical suspicion might result in removal of infected bone or biomaterials. However, there is no adequate tool to directly distinguish infection from sterile inflammation. Accurate visualization and diagnosis of an infection during revision surgery will improve clinical decision-making and the chance of a successful treatment. Here, we evaluated the specificity and sensitivity of vancomycin-IDRye800CW (vanco-800CW)2 in presumably infected patient materials obtained during revision surgery.

Methods

In this ex vivo study, we aim to establish a bacteria-targeted fluorescent imaging approach to discriminate between infected materials (such as tissue, bone marrow and infected plates and screws) and non-infected materials. Tissue and implants extracted during revision surgery were incubated with vanco-800CW for 15 minutes at 37oC and subsequently washed with phosphate-buffered saline. Fluorescent macroscopic imaging was performed with the IVIS Lumina II and an intraoperative camera system (SurgVision Explorer Air). Imaging results were confirmed using standard culture techniques, including diagnostic culture by sonication. Institutional review board permission for this study was obtained.

Results/Discussion

Our results with explant material from 8 patients demonstrate that sites infected with Gram-positive bacteria show a fluorescent vanco-800CW signal, whereas no fluorescent signal is detectable on non-infected materials. Gram-negative bacterial infections did not result in a fluorescent signal as expected.

Conclusions

Vanco-800CW is a specific and effective tracer for ex vivo detection of Gram-positive bacteria on implants and osteomyelitis. Exact sensitivity and specificity have still to be determined using a larger sample size. Fluorescent targeted imaging with vanco-800CW is a promising technique that can provide fast, accurate and real-time information, potentially allowing faster diagnosis and treatment of fracture-related infections.

References

  1. Engelsman, A. F. et al. The risk of biomaterial-associated infection after revision surgery due to an experimental primary implant infection. Biofouling (2010).
  2. van Oosten, M. et al. Real-time in vivo imaging of invasive- and biomaterial-associated bacterial infections using fluorescently labelled vancomycin. Nat. Commun. (2013).
Keywords: fluorescent, bacteria-targeted, osteomyelitis, biomaterials, traumatology
504

68Ga-siderophores for Pseudomonas aeruginosa infection imaging (#103)

Milos Petrik1, Eva Umlaufova1, Vladislav Raclavsky2, Andrea Palyzova3, Vladimir Havlicek3, 4, Zbynek Novy1, Marian Hajduch1, Clemens Decristoforo5

1 Palacky University, Institute of Molecular and Translational Medicine, Olomouc, Czech Republic
2 Palacky University, Department of Microbiology, Olomouc, Czech Republic
3 Czech Academy of Sciences, Institute of Microbiology , Prague, Czech Republic
4 Palacky University, Regional Centre of Advanced Technologies and Materials, Olomouc, Czech Republic
5 Innsbruck Medical University, Clinical Department of Nuclear Medicine, Innsbruck, Austria

Introduction

Pseudomonas aeruginosa (P.a.) infections have become a serious problem, especially in immunocompromised patients (1). The development of novel therapeutic and diagnostic strategies is intensively pursued. One of the promising diagnostic strategies for the detection of P.a. could be utilization of its siderophores. Siderophores are small iron-chelating molecules produced by almost all microorganisms (2). Replacing iron in siderophores by radiometal, such as gallium-68, opens approaches for targeted imaging of P.a. by means of PET. Here we report on 68Ga-siderophores for P.a. infection imaging.

Methods

Radiolabelling of selected siderophores (three different types of pyoverdines and deferoxamine) with Ga-68 was performed using acetate buffer. Stability in human serum, towards competitive chelator and iron challenge as well as protein binding and partition coefficient values were determined. In vitro uptake of 68Ga-siderophores was tested in various microbial cultures including different P.a. strains. In vivo behaviour of 68Ga-siderophores was studied in normal Balb/c mice by ex vivo biodistribution study and PET/CT imaging. Mouse and rat P.a. infection models were developed and used for the evaluation of 68Ga-siderophores under study by means of PET/CT.

Results/Discussion

All studied siderophores were labelled with 68Ga with high (>95%) radiochemical purity. The resulting complexes showed hydrophilic properties (log P = -3), low protein binding (<5%) and high stability in human serum (>95%) up to 120 min incubation time. In vitro uptake of 68Ga-siderophores was highly dependent on type of microbial culture. 68Ga-pyoverdines displayed different levels of uptake in tested P.a. strains, while 68Ga-deferoxamine was taken up by all P.a. strains under study. In normal mice, all studied 68Ga labelled siderophores showed similar results manifested by rapid renal excretion and low blood values even at a short time period (90 min) after the application. PET/CT imaging of selected 68Ga-siderophores in P.a. infected animals showed focal accumulation of 68Ga-pyoverdine I and 68Ga-deferoxamine in infected tissues.

Conclusions

Studied siderophores can be labelled with Ga-68 with high affinity and radiochemical purity. 68Ga-siderophores showed very similar and satisfactory in vitro characteristics. In vitro uptake of 68Ga-siderophores in tested microbial cultures was diverse even within the microbial species. 68Ga-siderophores revealed excellent pharmacokinetics and can be used for imaging of P.a. infections.

References

1. de Bentzmann S, Plésiat P. The Pseudomonas aeruginosa opportunistic pathogen and human infections. Environ. Microbiol. 2011;13:1655-1665.

2. Hider RC, Kong X. Chemistry and biology of siderophores. Nat. Prod. Rep. 2010;27:637-657.

 

Acknowledgement

We gratefully acknowledge the financial support of Technology Agency of the Czech Republic (Project No. TE01020028) and Ministry of Education Youth and Sports of the Czech Republic (Project No. LO1304).

Keywords: Siderophores, Pseudomonas aeruginosa, Imaging
505

Ex vivo Laser-based Endomicroscopy of Infections in Explant Human Lungs (#216)

Usma Koser1, Marjolein Heuker2, Anne-Marie G. de Smet3, Michiel E. Erasmus4, Erik A. Verschuuren5, Wouter B. Nagengast6, Kev Dhaliwal7, Gooitzen M. van Dam8, Jan Maarten M. van Dijl2, Marleen van Oosten9

1 University Medical Center Groningen, University of Groningen, Intensive Care Unit and Faculty of Medical Sciences, Groningen, Netherlands
2 University Medical Center Groningen, University of Groningen, Department of Medical Microbiology and Faculty of Medical Sciences, Groningen, Netherlands
3 University Medical Center Groningen, Intensive Care Unit , Groningen, Netherlands
4 University Medical Center Groningen, Department of Cardiology and Thorax Surgery, Groningen, Netherlands
5 University Medical Center Groningen, Department of Pulmonolgy and Tuberculosis, Groningen, Netherlands
6 University Medical Center Groningen, Department of Gastroenterology and Hepatology, Groningen, Netherlands
7 Univeristy of Edinburgh, ESPRC IRC Hub Queens Medical Reserach Institute , Edinburgh, United Kingdom
8 University Medical Center Groningen, University of Groningen, Department of Surgery Division of Surgical Oncology, Groningen, Netherlands
9 University Medical Center Groningen, Deaprtment of Medical Microbiology, Groningen, Netherlands

Introduction

Pneumonia remains a leading cause of death1,2. Newly designed rapid diagnostic tools enable greater accuracy in clinical decision making and optimise antimicrobial management thus leading to better clinical outcomes. Visualisation of the distal respiratory tract evolved following the development of probe-based confocal laser endomicroscopy and molecular imaging3. Additional fluorescence techniques and microbe specific agents can provide unique insights into processes in the alveolar space and can rapidly identify structural change and microbial presence during infection4.

Methods

From April 2018 we received explanted lungs or lobes from patients undergoing lung transplant or resection for malignancy at the University Medical Center Groningen (UMCG). The protocol was ethically approved and initially trialled with a porcine, slaughterhouse lung model.

We establish mechanical ventilation, with varying pressures, through direct intubation of the available airway and perform bronchoscopy examination of the specimens. Cellvizio and the alveoflex, fiber-based bundle (Mauna Kea Technologies, Paris) were used with optical molecular imaging agents e.g. BAC 15 BAC 26, capable of identifying specific bacterial and fungal targets. Thereafter, image algorithm analysis was performed on the recorded datasets, using previously described techniques in MATLAB (The Mathworks Inc.)7.

Results/Discussion

Preliminary laboratory and animal lung models were used to optimise our protocol and image bacteria in the distal lung. Subsequent to this, we performed 17 procedures on human lung explants at the UMCG. Thus far, we have obtained specimens from 7 lung cancer patients, 1 rejected donor-lung following ex vivo lung perfusion (EVLP) and 8 double explanted lungs (aetiologies: chronic obstructive airways disease – COPD, cystic fibrosis – CF, pulmonary hypertension and interstitial lung diseases). With the use of antimicrobial specific fluorescence imaging agents, Gram-positive and Gram-negative bacteria in the alveolar space were visualised. This correlates with established microbiology culturing and identification methods following recognised sampling techniques (lavage and endobronchial brushing).

Conclusions

We have established a working model to ventilate end-stage, diseased, human lung explants and image the distal respiratory tract using optical molecular imaging techniques. New insights regarding the presence of bacteria in the alveolar space of explants in varying diseases are provided. Further clinical human studies are required to validate and assess the translatability of these novel techniques for real time infection imaging in patients.

References

1 Mizgerd, Joseph P. "Acute lower respiratory tract infection." New England Journal of Medicine 358.7 (2008): 716-727.

2 World Health Organization (WHO). The global burden of disease: 2004 update. Geneva: WHO; 2008 [accessed 2016 Jun 29]. Available from: http://www.who.int/healthinfo/global_burden_disease/GBD_report_2004update_full.pdf

3 Thiberville, Luc, et al. "In vivo imaging of the bronchial wall microstructure using fibered confocal fluorescence microscopy." American journal of respiratory and critical care medicine 175.1 (2007): 22-31.

4 Mills, Bethany, Mark Bradley, and Kevin Dhaliwal. "Optical imaging of bacterial infections." Clinical and translational imaging 4.3 (2016): 163-174.

5 Akram, A. R., et al. "A Labelled-Ubiquicidin Antimicrobial Peptide for Immediate In Situ Optical Detection of Live Bacteria in Human Alveolar Lung Tissue."

6 Akram, Ahsan R., et al. "In situ identification of Gram-negative bacteria in human lungs using a topical fluorescent peptide targeting lipid A." Science Translational Medicine 10.464 (2018): eaal0033.

7 Seth, Sohan, et al. "Estimating Bacterial Load in FCFM Imaging." Annual Conference on Medical Image Understanding and Analysis. Springer, Cham, 2017.

Acknowledgement

Proteus, Edinburgh, UK, have provided intellectual insight, discussion and the molecular agents for this work. We would like to acknowledge the assistance of the Cardiology and Thorax Surgery and Pathology departments of the UMCG in providing specimens according to an ethically approved protocol. Furthermore, the abundance of services provided by the clinical skills department of the UMCG.

Keywords: Pneumonia, bacteria, Optical Molecular Imaging.
506

Fungal cell density in cryptococcal brain lesions affects read-outs from quantitative, multiparametric MRI (#227)

Liesbeth Vanherp1, 2, Jennifer Poelmans1, 2, Kristof Govaerts1, 2, Katrien Lagrou3, Greetje Vande Velde1, 2, Uwe Himmelreich1, 2

1 KU Leuven, Biomedical MRI, Leuven, Belgium
2 KU Leuven, MoSAIC, Leuven, Belgium
3 KU Leuven, Laboratory of Clinical Bacteriology and Mycology, Leuven, Belgium

Introduction

Cryptococcal brain lesions are often difficult to distinguish from other pathologies (e.g. tumors and abscesses) on anatomical MRI. By using quantitative, multiparametric MRI, we studied the typical MR features of C. neoformans H99 (CN) and C. gattii R265 (CG) brain lesions to identify potential markers for differential diagnosis. To explain differences in the MR properties between both strains, we assessed in vivo fungal cell density and size using fibered confocal fluorescence microscopy (FCFM) and ex vivo microscopy.

Methods

Cryptococcal brain lesions were induced by stereotactic injection of 104 GFP-expressing CN H99 or CG R265 cells in the striatum of female Balb/c mice. Using a 9.4T preclinical MRI scanner (94/20 Biospec, Bruker Biospin), we acquired T2-weighted anatomical brain scans (2D, axial and coronal), diffusion-weighted images (b-values 31 and 1561 s/mm³), T1 maps (RARE-VTR with 5 TRs) and T2 maps (MSME with 16 TEs). For FCFM (Cellvizio, Mauna Kea), the fiberoptic probe was mounted on a stereotactic frame and inserted into the lesion. After sacrificing the animals, colony-forming unit (CFU) counting on brain homogenates was performed and the size of the fungal cells and their surrounding capsule was measured after india ink staining using a microscope with digital camera (Leica DM LS2 with ICC50W)1.

Results/Discussion

Lesions were characterized by higher apparent diffusion coefficients (ADC), T1 and T2 relaxation times than in the contralateral (normal) hemisphere. The ADC and T2 relaxation times were significantly higher in CG lesions than in CN lesions, without differences in the T1 relaxation time. FCFM showed that the cryptococcal cell density was higher in lesions caused by CN H99 than by CG R265. The number of cells (CFU) per mm³ of lesion material (MRI) was significantly higher in CN lesions, confirming more densely packed cells in CN lesions. India ink staining of the homogenized brain tissue showed that CG R265 cells are in general larger than CN H99 cells, with a thicker capsule around the cell and a larger cell body size, leading to a lower cell density compared to CN. The larger fluid-filled intercellular spaces and higher amount of the highly hydrated capsular material in CG lesions can explain the higher ADC and T2 relaxation time.

Conclusions

Increased ADCs can potentially be used to differentiate cryptococcomas from e.g. bacterial abscesses that show restricted diffusion2. T2 relaxation times and ADCs differed between CG R265 and CN H99 lesions, which was related to the capsule size and associated fungal density in the respective lesions. This shows that pathogen characteristics can impact MR read-outs and that in vivo MRI can potentially provide indirect measures for cell density.

References

  1. Dragotakes, Q. & Casadevall, A. Automated Measurement of Cryptococcal Species Polysaccharide Capsule and Cell Body. J. Vis. Exp. e56957–e56957 (2018). doi:10.3791/56957
  2. Nadal Desbarats, L. et al. Differential MRI diagnosis between brain abscesses and necrotic or cystic brain tumors using the apparent diffusion coefficient and normalized diffusion-weighted images. Magn. Reson. Imaging 21, 645–50 (2003).

Acknowledgement

We are thankful for financial support from the European Commission for the Infect-ERA project CryptoVIEW. LV is an SB PhD fellow at Research Foundation Flanders (FWO).

Figure 1: Representative images for CN H99 and CG R265 lesions.
ADC maps, T2 maps, FCFM images and India ink microscopy images are shown for a C. neoformans H99 and C. gattii R265 lesion.
Keywords: Fungal infection, quantitative MRI, density
507

The use of PET-CT in imaging tuberculosis in non-human primates (#419)

Marieke Stammes1, Krista Haanstra1, Mohamed Khayum1, Richard Vervenne1, Michel Vierboom1, Frank Verreck1

1 Biomedical Primate Research Centre, Parasitology, section Tuberculosis, Rijswijk, Netherlands

Introduction

Tuberculosis (TB) remains the number one killer of infectious diseases. Worldwide, a person dies of TB about every 20 seconds and antibiotic resistance poses a growing global health challenge also in TB. The interaction between the pathogen and the host is complex, and many aspects underlying disease susceptibility and protective host defence remain elusive still. A better understanding of disease dynamics and immune mechanisms is key to the development of improved therapies. PET-CT 18F-FDG based imaging has opened new opportunities, especially in non-human primates (NHPs), to clarify this.

Methods

PET-CT was performed at week 0, 4, 6, 8 and 10 post-infection mycobacterium tuberculosis (Mtb). Images were obtained using a preclinical MultiScan LFER 150 PET-CT (Mediso Medical Imaging Systems) on anesthetized rhesus macaques (RM; n=24) as NHP model of TB. All CTs were acquired with a breath-hold with and without CT contrast (Omnipaque 300, 2 ml/kg). To detect the metabolically, inflammatory, activity of TB lesions, 45min after intravenous injection of 100 MBq 18F-FDG a 20min static PET scan was acquired. Image were analysed using Osirix MD viewer. ROIs were drawn on every lesion and data were exported as standardized uptake value (SUV) within these granulomas.

Results/Discussion

Due to progressive TB some RM reached a humane endpoint before wk10 which were euthanized immediately without performing an endpoint PET-CT. In total 80 scans were recorded for analyses (24, 23, 20, and 13 scans on week 4, 6, 8 and 10, respectively).  We determined on CT the lesion size and on PET the uptake of 18F-FDG in those granulomas, after which we performed a correlation analysis. We found a clear correlation between the lung pathology score (at endpoint) vs SUV of 0.72, 0.75, 0.91 and 0.88 at the four different timepoints, week 4,6, 8 and 10 post-infection, respectively, with p-value <0.0001 for all. A similar trend was observed for bacterial tissue load at endpoint vs the lung pathology score. The first is measured via the amount of colony forming units (CFUs)) from the lung.  The lung pathology score is semi quantative and based on the number and size of granulomas per lung lobe with a maximum score of 10 per lobe times 7 lobes.

Conclusions

Our results demonstrate the utility of PET-CT to monitor TB infection dynamics over time and implicitly the ability to monitor treatment response in a non-invasive manner over time. Using imaging in combination with conventional pathology assessment and microbiological measurements, we shall gain a deeper understanding of both the development of active TB and the underlying pathogen-host interaction in this clinically relevant NHP model of TB.

Figure 1
representative PET-CT images of TB in different stages of development
Figure 2
correlation plots of the SUV at the different scan timepoints vs the lung PA score at endpoint
Keywords: PET-CT, non-human primates, tuberculosis
508

Syntheses, validation and preclinical evaluation of [68Ga]Ga-NODAGA-/DOTA-CDP1 bioconjugates as positron emission tomography infection imaging agents. (#351)

Amanda H. Mdlophane1, 2, 3, Thomas Ebenhan1, 3, Balungile Madikizela5, Sanah Nkadimeng5, Thavi Govender6, Mike M. Sathekge1, 3, Jan R. Zeevaart4, 2, 3

1 University of Pretoria & Steve Biko Academic Hospital, Nuclear Medicine, Pretoria, South Africa
2 The South African Nuclear Energy Corporation (Necsa), Radiochemistry, Pretoria, South Africa
3 Preclinical Imaging Facility, Nuclear Medicine Research Infrastructure (NuMeRI), Brits, South Africa
4 Department of Science and Technology, Preclinical Drug Development Platform, North West University, Potchefstroom, South Africa
5 University of Pretoria, Phytomedicine Proramme, Department of Paraclinical Sciences, Faculty of Veterinary Science, Pretoria, South Africa
6 University of KwaZulu-Natal, Catalysis and Peptide Research Unit, School of Health Sciences and School of Chemistry and Physics, Durban, South Africa

Introduction

Antimicrobial peptides as infection imaging agents are of increased interest in Nuclear Medicine. Infection imaging provides a solution for often misdiagnosed, complicated infections and monitoring of therapy. The aim of this study is to validate the radiosynthesis of [68Ga]Ga-NODAGA-/DOTA-CDP1, to investigate bacterial binding and mammalian cell toxocity, and to establish the pharmacokinetic tracer profile in healthy mice using micro PET-CT imaging.

Methods

Sytheses of [68Ga]Ga-NODAGA-/DOTA-CDP including optimisation for pH, molarity, incubation time and temperature, and product purification was conducted (1). [68Ga]Ga-DOTA-CDP1 integrity and protein binding was investigated at ambient temperature and 37ºC, [68Ga]Ga-chloride and [68Ga]Ga-DOTATATE were internal refernces. DOTA-CDP1 cytotoxicity was investigated against HepG2 and Vero cells (2).Bacterial cell binding of [68Ga]Ga-DOTA-CDP1 with/without ecxess DOTA-CDP1 was determined in the following microbes: Escherichia coli (EC), Staphylococcus aureus (SA) and Mycobacterium smegmatis (MS). The pharmacokinetic profiling protocol for [68Ga]Ga-DOTA-CDP1 in healthy BALB/c mice employed  45 min dynamic PET/CT imaging followed by 1 h and 2 h static PET-CT and subsequent ex vivo biodistribution (3).

Results/Discussion

[68Ga]Ga-DOTA-CDP1 showed >90% integrity after 2h incubation in PBS or serum whereas [68Ga]Ga-NODAGA-CDP1 was lacking stability (p<0.01). LC50 analysis indicated DOTA-CDP1 as non-toxic in both mammalian cell-lines with 86-89% [68Ga]Ga-DOTA-CDP1 unbound. Rapid 0h tracer binding in EC occurred  (40+/- 6.8%)  and MS (18+/-2.6%; p<0.01), at 1h uptake increased slowly  to 45+/-5.3 and 20+/-3.3%, respectively (p<0.02), EC uptake was higher than SA (29+/-8;p<0.003). PET imaging showed moderate blood pool clearance over 1h. Static [68Ga]Ga-DOTA-CDP1 PET showed significantly decreased heart SUV comparing images at 1h (2.53+/-0.5)  and 2h (1.5+/-0.3; p<0.01). Liver / spleen uptake increased with time. Ex vivo biodistribution showed lower liver and lung activity [68Ga]Ga-DOTA-CDP1 (15.7+/-7.6 and 11.3+/-4) compared to [68Ga]Ga-chloride (18.2+/-10.5 and 18.3+/-11.8) and significantly higher uptake in heart (7.5+/-1.7; p<0.003), blood pool (18.3+/-10.6; p<0.004) and brain (0.66+/-0.2; p<0.01).

Conclusions

[68Ga]Ga-DOTA-CDP1 exhibits good bench and serum stability. The concentration used in bacterial uptake is non-toxic due to low binding of [68Ga]Ga-DOTA-CDP1 to mammalian cells, thereby increasing bioavailability to host infection sites.Accumulation observed in the bacterial stratins at different sensitivities gives promise in distinguishing between infections. Biodistribution in healthy mice show delayed clearance from blood and blood rich organs

References

  1. Ebenhan, T., et al. (2014). Nucl Med Biol 41(5):390-400
  2. Mosmann, T. (1983).J Immunol Methods 65(1-2):55-63
  3. Mokaleng, B.B., et al. (2015). Biomed Res Int 2015:284354

 

Keywords: CDP1, infection imaging, PET/CT infection imaging, Ga68, infection probes
509

Mapping the dysregulation of acid-base homeostasis upon sepsis-induced shock by CEST-MRI (#63)

Lorena Consolino1, Alice Antonello1, Amerigo Pagoto1, Pietro Irrera2, 1, Laura Conti1, Elisabetta Bolli1, Dario Longo1

1 University of Torino, Department of Molecular Biotechnology and Health Sciences, Torino, Italy
2 University of Campania , Biostructures and Bioimages Institute, CNR, Napoli, Italy

Introduction

Sepsis is a systemic inflammatory response to a suspected infection that often leads to multi-organ dysfunction. Among the different clinical signs, metabolic acidosis is mainly a consequence of renal failure that impairs the homeostasis of the systemic acid-base regulation [1]. Recently, several MRI approaches were proposed to monitor sepsis-induced changes in renal physiology [2,3]. In this work, MRI-pH mapping was applied to investigate whether variations of pH balance in kidneys might be a predictive and specific biomarker for early diagnosis of sepsis-induced shock.

Methods

Male C57BL/6J mice (n=12) were i.p. administered with 10 mg/kg body weight (b.w).of LPS from E. coli to induce sepsis condition or with 1 ml of 4% thioglycollate (TG) broth as a control model of local inflammation. After 18 hours, MR images were acquired on kidney coronal sections on a 7T Bruker MRI scanner (FOV: 3 cm, in-plane resolution: 234 µm, slice thickness: 1.5 mm). MRI-CEST pH mapping [4] was performed using a fast spin-echo sequence with a RF saturation pulse of 3 μT x 5s and Z-spectra acquired before and after the injection of 1.5 g I/kg b.w. iopamidol into the tail vein. After MRI acquisition, mice were sacrificed, several organs collected and stained with H&E. Concentration of cyto/chemochines and creatinine were detected in mouse sera by cytometric beads and ELISA kits.

Results/Discussion

T2w MR-images of kidneys failed to report macroscopic lesions in both LPS and TG mice. Conversely, CEST-pH mapping allowed detecting renal dysfunctional changes. Extracellular pH values significantly raised upon LPS, whereas any pH changes were reported in TG and vehicle (mean pH: 6.48 and 6.83, p<0.001 for LPS;6.65 and 6.70 for TG; 6.55 and 6.64 for vehicle, before and after treatment.Fig.1A-B). Besides common sepsis-induced systemic signs (dishevelled hair, reduced mobility, diarrhoea), LPS-endotoxic shock stimulated a marked recruitment of proinflammatory mediators, as IL-6 and MCP-1, whereas a marginal increase was detected in TG mice (Fig.2A). In addition, only LPS mice reported a strong body weight loss (Δweight= -2.2 g), which variation significantly correlated with ΔpH changes (r=-0.77, P<0.05.Fig.2B). Increased concentration of serum creatinine (5.52 ng/µL) confirmed LPS-induced renal functional impairment in comparison to TG (0.46 ng/µL) and vehicle (1.2 ng/µL, Fig.2C)

Conclusions

These results showed that renal pH variations occur in LPS-treated mice and can be non-invasively visualized by the proposed CEST-MRI approach. In particular, alterations of renal pH balance are specific for systemic sepsis and are not detected in a local inflammation model induced by TG. Therefore, this study suggests that further evaluation of renal pH mapping might be considered as a potential tool for diagnosis of sepsis.

References

  1. Kellum JA., Critical care and resuscitation 2004, 6:197-203
  2. Tran M. et al., Clin Invest. 2011,121 :4003–4014
  3. Liu J. et al., NMR in Biomed 2018, 31:e3942
  4. Longo DL et al., NMR in Biomedicine. 2017;30:e3720

Acknowledgement

The authors acknowledge the framework of the EU COST Action CA16103 - PARENCHIMA “Magnetic Resonance Imaging Biomarkers for Chronic Kidney Disease”.

Figure 1. MRI measurements of renal extracellular pH in LPS, TG and vehicle treated mice.
A) Bargraphs showed increased pH values upon LPS administration (P<0.001). No significant changes were reported on pH measurements in TG and vehicle mice. B) Representative T2w MR images (upper panel) and color-enconded pH parametric maps overimposed on T2w images of kidney (bottom panel) in vehicle, LPS an TG mice.
Figure 2. Ex-vivo validation of LPS-induced septic shock.
A) Sera concentration (pg/mL) of proinflammatory cytochines at 0, 1.5, 3, 6, and 24 hours after LPS (left) and TG (right) treatment. B) Significant inverse correlation between pH (ΔpH) and b.w. (Δgrams) variations in LPS mice is observed (r=-0.77, P<0.05). C) Sera quantification of creatinine (ng/uL) showed increased values in LPS mice in comparison to TG and vehicle mates.
Keywords: sepsis, pH imaging, CEST-MRI
510

Nuclear imaging of infection: in vitro assessment of 2-[18F]-fluorodeoxysorbitol tracer for the detection of bacterial infection (#261)

LIsanne M. Braams1, Jurgen W.A. Sijbesma2, Hendrikus H. Boersma3, Jan Maarten van Dijl1, Philip H. Elsinga2, Andor W.J.M. Glaudemans2, Riemer H.J.A. Slart2, Marleen van Oosten1

1 University of Groningen, University Medical Center Groningen, Department of Medical Microbiology and Infection Prevention, Groningen, Netherlands
2 University of Groningen, University Medical Center Groningen, Department of Nuclear Medicine and Molecular Imaging, Groningen, Netherlands
3 University of Groningen, University Medical Center Groningen, Department of Clinical Pharmacy and Pharmacology, Groningen, Netherlands

Introduction

Imaging of inflammation and infection is widely done with the positron emission tomography (PET) tracer fluorine-18-fluorodeoxyglucose (18F-FDG). However, 18F-FDG cannot reliably distinguish bacterial infection from a tumour or sterile inflammation1. Therefore, there is a need for bacterial targeted tracers2. Sorbitol is a sugar analogue only metabolized by bacteria of the Enterobacteriaceae family and not by mammalian cells, which makes it an ideal candidate for targeted bacterial PET imaging3.

Methods

In this study, a sorbitol based PET tracer 2-[18F]-fluorodeoxysorbitol (18F-FDS) is synthesised from 18F-FDG through a single reduction step according to established methods4. Purity analysis is done by thin liquid chromatography. 18F-FDS is tested in vitro on a wide variety of 12 clinically relevant isolates, 11 infected clinical blood samples and 2 infected clinical ascites samples. Samples were collected from the department of Medical Microbiology at the University Medical Centre Groningen, the Netherlands. Heatkilled cultures and sterile patient materials were used as controls. The uptake of 18F-FDS was assessed by measuring activity (Bq) with a gamma-counter and correlating this to the number of colony forming units (CFU).

Results/Discussion

Clinical isolates from the family Enterobacteriaceae accumulated 18F-FDS substantially in comparison to the heatkilled controls. Additionally, Pseudomonas aeruginosa and Corynebacterium jeikeium showed relevant uptake, which has not been described before. The uptake in Enterobacteriaceae was more than a two-fold increase compared to the tested Gram-positive bacteria. In both blood and ascites samples, a high18F-FDS uptake correlated with infection. The highest uptake of 18F-FDS was seen in Klebsiella pneumonia, both in clinical strains as well as in infected blood samples.

Conclusions

18F-FDS was successful in detecting infected patient materials which makes it a promising bacterial-targeted PET tracer for clinical translation.

References

  1. Love C, Palestro CJ. Radionuclide imaging of infection. J Nucl Med Technol. 2004;32(2):47-57; quiz 58-59.
  2. Glaudemans AWJM, Slart RHJA, van Dijl JM, van Oosten M, van Dam GM. Molecular imaging of infectious and inflammatory diseases: a terra incognita. J Nucl Med. 2015 May 1;56(5):659–61.
  3. Weinstein EA, Ordonez AA, Demarco VP, Murawski AM, Pokkali S, Macdonald EM, et al. Imaging Enterobacteriaceae infection in vivo with 18 F-fluorodeoxysorbitol positron emission tomography.
  4. Li Z-B, Wu Z, Cao Q, Dick DW, Tseng JR, Gambhir SS, et al. The Synthesis of 18 F-FDS and Its Potential Application in Molecular Imaging. B Acad Mol Imaging Mol Imaging Biol. 2007;10:92–8.
Keywords: PET, Enterobacteriacea, Sorbitol
511

In vivo imaging of infection and immunity in nonhuman primate models of infectious diseases (#295)

Thibaut Naninck1, Céline Mayet1, Sabine Tricot1, Sophie Luccantoni1, Nidhal Kahlaoui1, Roger Le Grand1, Catherine Chapon1

1 CEA/Jacob, IDMIT, Fontenay-aux-Roses, France

Introduction

Although non-human primates (NHP) provide a good model for infectious diseases, their exploration for viral transmission and dissemination by in vivo imaging has not been used extensively. This can be explained by the limited access to adequate structures for imaging these large animals with a high resolution while having specific regulations according to the level of confinement for the human pathogens. Our main objectives are to develop minimally invasive technologies for the longitudinal monitoring of infections, host response and treatments in NHP.

Methods

IDMIT (Infectious disease models for innovative therapies) is a national infrastructure (www.idmitcenter.fr ) founded by five major academic institutions in France (the CEA, the INSERM, ANRS, the Université Paris Sud and the Institut Pasteur) which is located at Fontenay-aux-Roses (France). The infrastructure develops NHP models for immune related disorders and human infectious diseases for preclinical evaluation of human vaccines, immunotherapies and anti-microbial treatments. IDMIT has implemented a unique set of facilities for in vivo imaging in BSL2 and BSL3 constrains (near infrared fluorescence, probe based confocal endomicroscopy, echography, two-photon microscopy and PET-CT).

Results/Discussion

So far, we have developed non-invasive and longitudinal fluorescence imaging approaches using probe based confocal endomicroscopy to study the behavior of skin antigen presenting cells following intradermal immunization with different vaccine vectors in order to better understand the mechanisms leading to the induction of cellular and humoral immune responses after vaccination1-2. Furthermore, non-invasive in vivo imaging procedures were developed using probe based confocal endomicroscopy coupled with bronchoscopy to track bacterial localization and cellular interactions with host cells in the lower respiratory tract of challenged and naturally infected animals in a model of fluorescent bordetella pertussis in baboons3. Development of PET-CT approaches to track pathogens and inflammation related to infection is ongoing.

Conclusions

This type of approaches could now be used for extended characterization of diverse cell types in vivo and their interactions with other vaccine antigens and/or pathogens. Furthermore the implementation of the PET-CT system will allow us to study whole body dissemination of pathogens with a high sensitivity. 

References

1.           Todorova, B., Salabert, N., Tricot, S., Boisgard, R., Rathaux, M., Le Grand, R. et al. Fibered Confocal Fluorescence Microscopy for the Noninvasive Imaging of Langerhans Cells in Macaques. Contrast Media Mol Imaging. 2017, 3127908  (2017).

2.           Todorova, B., Adam, L., Culina, S., Boisgard, R., Martinon, F., Cosma, A. et al. Electroporation as a vaccine delivery system and a natural adjuvant to intradermal administration of plasmid DNA in macaques. Sci Rep. 7, 4122  (2017).

3.           Naninck, T., Coutte, L., Mayet, C., Contreras, V., Locht, C., Le Grand, R. et al. In vivo imaging of bacterial colonization of the lower respiratory tract in a baboon model of Bordetella pertussis infection and transmission. Sci Rep. 8, 12297  (2018).

Acknowledgement

We warmly thank the IDMIT infrastructure staff for excellent technical assistance, morte specifically the Animal Science and Welfare core facility. 

Keywords: infectious diseases, nonhuman primate, PET-CT, inflammation, optical imaging
512

Design of a Staphylococcus aureus-specific imaging marker for infective endocarditis (#525)

Laura Kreylos1, Martin Grewer1, Christian N. Schwarz1, Hélène Van de Vyver2, Christiane Geyer1, Carsten Höltke1, Silke Niemann2, Cornelius Faber1

1 University Hospital Muenster, Department of Clinical Radiology, Muenster, North Rhine-Westphalia, Germany
2 University Hospital Muenster, Institute of Medical Microbiology, Muenster, North Rhine-Westphalia, Germany

Introduction

Staphylococcus aureus-induced infective endocarditis (IE) is a life threatening disease with high mortality rates up to 30% [1]. Due to the currently prevalent long diagnostic latency (29 ± 35 days [2]) and unspecific diagnostic methods of IE, there is an urgent need for improved detection tools. Here, we assess the binding affinity of peptide fragments from specific bacteriophages and antimicrobial peptides to S. aureus and design a reference fluorescence imaging probe for use in diagnosis of infective endocarditis.

Methods

Binding efficiency of peptide candidates (­~20 aa) to the S. aureus´ surface was assessed using ELISA (n=8) and fluorescence measurements (n=6) with GFP-expressing bacteria.

Peptide candidate ubiquicidin (UBI29-41) was synthetized with a polyhistidin-tag and conjugated to a nickel-containing fluorophore (NTA-Atto 488, Sigma Aldrich).

Antimicrobial capacity, binding affinity, specificity and detection thresholds were assessed in incubation experiments with naïve S. aureus. A bacterial pellet (108 CFU) was incubated with different concentrations of the fluorescent probe and observed in fluorescence microscopy and fluorescence reflectance imaging (FRI). Binding specificity was assessed in competition experiments with unlabelled UBI 29-41.

Results/Discussion

Among the investigated peptide fragments the peptide UBI showed the best pathogen affinity. UBI29-41 was therefore used for design of a reference imaging probe. In growth experiments with different concentrations of S. aureus bacteria (104-108CFU) in presence of different amounts (0,125 – 2mmol/l) of UBI29-41, the peptide fragment did not exert antimicrobial effects.

In bacterial labeling assays the imaging probe showed pronounced fluorescence intensity as observed in fluorescence microscopy and FRI. In contrast, the bacteria did not emit a fluorescent signal after incubation with the uncoupled fluorophore. After incubation with 0.1 mmol/l of the probe, bacteria showed well detectable signal. In competition experiments a signal difference of 62% between the preincubated bacteria (100 fold excess unlabeled UBI29-41) and the pellet only incubated with the dye was observed, indicating a specific binding on the bacterial surface.

Conclusions

Our results show that our imaging marker binds specifically to S. aureus. The coordinative binding of the His-tagged UBI 29-41 to the Ni-containing fluorophore is stable under the tested conditions, and suggests that, Ni-coated iron oxide nanoparticles (IONP) can be coupled to UBI29-41 to detect bacterial infections via MRI imaging. An UBI-based MRI marker appears likely to reproduce previous results obtained with an UBI-based SPECT-tracer [3].

References

[1] Baddour LM, et al. (2015): Infective Endocarditis in Adults: Diagnosis, Antimicrobial Therapy, and Management of Complication. Circulation, 132 (15): 1435-1486

[2] Habib, Gilbert, et al. (2015): 2015 ESC Guidelines for the management of infective endocarditis. European heart journal, 36 (44): 3075–3128

[3] Sathekge, Mike, et al. (2018): Molecular imaging in musculoskeletal infections with 99mTc-UBI 29-41 SPECT/CT. Ann Nucl Med (2018) 32: 54.

Design of an imaging marker

Left: Scheme for labeling the bacteria by using peptide fragments coupled to IONP (Iron Oxide Nanoparticles) and Nickel containing fluorophore. Right: Labeled bacterial biofilm in fluorescence microscopy.

FRI competition experiment

108 CFU S. aureus bacteria pellet incubated with
A)  0,01 mmol/l UBI-Atto-Dye (2 h 37°C)

B) 1mmol/l UBI (1 h, 37°C) + 0,01 mmol/l UBI-Atto-Dye (1 h, 37°C)

C) PBS (Control)

Keywords: Infection Imaging, FRI, MRI, S. aureus, contrast agent