This protocol outlines the steps required to perform ex vivo validation of in vivo near-infrared fluorescence xenograft imaging experiments in mice using fluorophore labelled nanobodies and conventional antibodies.
This protocol outlines the steps required to perform ex vivo validation of in vivo near-infrared fluorescence (NIRF) xenograft imaging experiments in mice using fluorophore labelled nanobodies and conventional antibodies.
First we describe how to generate subcutaneous tumors in mice, using antigen-negative cell lines as negative controls and antigen-positive cells as positive controls in the same mice for intraindividual comparison. We outline how to administer intravenously near-infrared fluorophore labelled (AlexaFluor680) antigen-specific nanobodies and conventional antibodies. In vivo imaging was performed with a small-animal NIRF-Imaging system. After the in vivo imaging experiments the mice were sacrificed. We then describe how to prepare the tumors for parallel ex vivo analyses by flow cytometry and fluorescence microscopy to validate in vivo imaging results.
The use of the near-infrared fluorophore labelled nanobodies allows for non-invasive same day imaging in vivo. Our protocols describe the ex vivo quantification of the specific labeling efficiency of tumor cells by flow cytometry and analysis of the distribution of the antibody constructs within the tumors by fluorescence microscopy. Using near-infrared fluorophore labelled probes allows for non-invasive, economical in vivo imaging with the unique ability to exploit the same probe without further secondary labelling for ex vivo validation experiments using flow cytometry and fluorescence microscopy.
In the present report, we describe the implementation of near-infrared fluorophore labelled probes for validation of in vivo xenograft imaging experiments by using ex vivo flow cytometry and fluorescence microscopy of the dissected xenograft tumors. We compare a single domain nanobody (s+16a, 17 kDa) 1 and a monoclonal antibody (Nika102, 150 kDa) 2,3 directed to the same target antigen for specific in vivo near-infrared fluorescence imaging in a lymphoma xenograft model. The target antigen ADP-ribosyltransferase ARTC2.2 is expressed as a GPI-anchored cell surface ecto-enzyme by lymphoma cells 4-9.
Nanobodies derived from camelid heavy-chain-only antibodies are the smallest available antigen-binding fragments 10,11. With only ~15 kDa, these small antibody fragments are soluble, very stable and are renally cleared from circulation 8,10. These properties make them particularly suited for specific and efficient targeting of tumor antigens in vivo 12-20. Common antigen targets of available nanobodies are the epidermal growth factor receptor (EGFR1 or HER-1), human epidermal growth factor type 2 (HER-2 or CD340), carcinoembryonic antigen (CEA) and vascular cell adhesion molecule-1 (VCAM-1) 21. Nanobody conjugates are promising tools for cancer immunotherapy and treatment of inflammatory diseases 22.
Recent studies have shown that nanobodies allow higher tumor-to-background (T/B)-ratios than conventional antibodies in in vivo molecular imaging applications 8,17,19. This is explained mainly by the relatively poor and slow tissue penetration of conventional antibodies, slow clearance from circulation, and long retention in non-targeted tissues 23. Moreover, excess of conventional antibodies leads to non-specific accumulation in target antigen-negative tumors caused by the enhanced permeability and retention (EPR) effect 24,25. This means that higher doses of conventional antibodies may increase not only specific signals but also nonspecific signals, thus reducing the maximum achievable tumor-to-background ratio. In contrast, increasing the dose of nanobodies increases the signals of antigen-positive tumors but not of normal tissue or antigen-negative tumors (unpublished data).
Beyond the comparison of nanobodies and conventional antibodies, we outline an intraindividual assessment of antigen-positive and -negative xenografts in the same mice for direct comparison of specific and nonspecific signals due to the EPR effect. The near-infrared fluorophore conjugated probes allowed us to exploit a single probe in vivo and ex vivo using near-infrared fluorescence imaging, flow cytometry, and fluorescence microscopy. Applying our protocols allows for nonradioactive, highly sensitive, and inexpensive optimization of in vivo molecular imaging experiments such as evaluation of new antibody constructs for specific tumor targeting.
The aim of this tutorial study is to highlight the use of NIRF-imaging for evaluation of new antibody constructs in preclinical molecular imaging.
In this protocol, all experiments were performed with a small-animal NIRF-Imaging system, a fluorescence activated cell sorter (FACS) flow cytometer, and a confocal microscope.
NOTE: Experiments were performed in accordance with international guidelines on the ethical use of animals and were approved by the local animal welfare commission of the University Medical Center, Hamburg.
1. Preparation of Tumor Cells, Mice, and Antibody Constructs
2. In Vivo Imaging
3. Harvesting and Preparation of Tumors
4. FACS Analysis
5. Microscopic Analysis
6. In Vivo Imaging Analysis
Fluorescently labelled probes allow for the combination of different NIRF-imaging techniques (Figure 1A). We aimed to perform in vivo NIRF-imaging, flow cytometry, and fluorescence microscopy sequentially in order to compare fluorescently labelled nanobodies and monoclonal antibodies for specific in vivo imaging (Figure 1B).
Mice were injected with 50 μg of nanobody and monoclonal antibody to evaluate the specificity of the fluorescently labelled constructs for in vivo imaging. The results showed specific labeling of antigen-positive tumors with both nanobody and monoclonal antibody at 6 hr after injection (Figure 2). ROI analyses of the antigen-positive tumors showed a much higher T/B ratio of ~12 for the nanobody compared to ~6 for the monoclonal antibody. Moreover, the nanobody showed no nonspecific signal in the antigen-negative tumors, whereas the monoclonal antibody showed nonspecific confounding signals in the antigen-negative tumors.
Besides the nonspecific signal of the negative tumors, the monoclonal antibody also induced nonspecific background signals in the entire animal. This is likely due to excessive free circulating antibodies, which are too large to be renally excreted. Contrariwise, animals injected with nanobodies showed nonspecific signals only in the kidneys due to the renal elimination of the small nanobodies.
Flow cytometry analyses of tumor cell suspensions showed specific labeling of antigen-positive tumor cells with both AF680-conjugates 6 hr after injection. The stronger fluorescence signal of the nanobody labelled cells compared to monoclonal antibody labelled cells reflects the in vivo NIRF-imaging results. Importantly, the flow cytometric analyses reveal that there is no nonspecific labelling of antigen-negative cells with either of the two constructs (Figure 3).
Fluorescence microscopy of tumor cryosections showed a strong and almost homogenous labeling of antigen-positive cells with the nanobody 6 hr after injection. Contrariwise, the monoclonal antibody showed a much weaker and rather inhomogeneous staining (Figure 4A). Antigen-negative tumors show no staining 6 hr after injection of the nanobody, whereas antigen-negative tumors injected with the conventional antibody show nonspecific scattered staining in the interstitial space (Figure 4B).
Figure 1: Flourescence imaging and antibody constructs. (A) Imaging setup for evaluation of AF680-conjugates: In vivo NIRF-imaging followed by flow cytometry and fluorescence microscopy. (B) Schematic of AlexaFluor680 labelled nanobody s+16a (red) and monoclonal antibody Nika102 (blue). Orange stars indicate the AlexaFluor680 fluorochromes.
Figure 2: In vivo NIRF-imaging. Images of the fluorescence signal of antigen-positive (+) and antigen-negative (-) tumors in mice that have been injected with nanobody s+16a (A) and monoclonal antibody Nika102 (B). In vivo imaging was performed before (0 h) and 6 h after injection. Signal intensities are displayed as radiant efficiency (p/sec/cm2/sr) / (µW/cm2).
Figure 3: Ex vivo FACS analyses of cell bound antibody constructs from tumor cell suspensions. (A) Gating strategy for FACS analyses of tumor cells. (B) Histograms display the amount of the intravenously injected AF680-conjugated nanobody s+16a and antibody Nika102 specifically bound to the tumor cells in vivo. Antigen-negative tumors are displayed as unfilled histograms and antigen-positive tumors are displayed as filled histograms.
Figure 4: Ex vivo fluorescence microscopy. (A) Vue d'ensemble fluorescence microscopy of entire antigen-positive tumor cryosections 6 h after injection of s+16a680 or Nika102680. Signal intensities of the in vivo intravenously injected AF680-conjugates without any secondary-labelling agents are displayed in red. Ex vivo counterstained nuclei are displayed in blue and vessels in green. Dotted lines indicate outer margins of the entire tumors. (B) Close-up fluorescence microscopy of antigen-positive and antigen-negative tumors.
We used near-infrared fluorophore labelled nanobodies and conventional monoclonal antibodies directed against the same target on lymphoma cells for a multimodal comparison of in vivo and ex vivo analyses. We showed that nanobodies are well suited as diagnostic tools for rapid and specific in vivo detection of lymphomas.
In vivo, s+16a680 allowed a fast and more specific detection of ARTC2-positive xenografts. Apart from the different kinetics for best tumor visualization in vivo, the major drawback of Nika102680 was the high nonspecific signal from ARTC2-negative tumors and nonspecific background signals.
Ex vivo flow cytometric analyses of dispersed cells from dissected tumors showed no nonspecific binding to ARTC2-negative lymphoma cells of injected AF680-conjugates. Ex vivo fluorescence revealed strong and almost homogenous staining of cells in ART2C-positive tumor sections in case of nanobody s+16a, confirming that the nanobody was able to reach even remote areas within the tumor after 6 h. In contrast, the monoclonal antibody showed weaker and inhomogeneous staining of cells in ARTC2-positive tumors after 6 hr. Better imaging results with the conventional antibody can be achieved after 24 hr or 48 hr (data not shown). In order to perform a thorough comparison of two differently sized constructs, imaging at different time points (serial-imaging) has to be performed to identify the optimal imaging time point for each construct.
Like other previous studies, the results reported here emphasize that in vivo molecular imaging with labelled nanobodies allows rapid and specific same-day tumor imaging with high tumor-to-background ratios 12-15,17-19. Contrariwise, conventional antibodies result in low tumor-to-background ratios and nonspecific signals from antigen-negative tumors early after injection due to their slow clearance from the body. In order to obtain optimal imaging results with conventional antibodies, imaging time points 24 hr or even 48 hr after injection are commonly needed. These findings are in accord with previous studies that have suggested that conventional antibodies with proven therapeutic benefit have limited utility in molecular imaging 17,19,26. Therefore conventional antibodies might be rather suited for therapeutic purposes due to their long plasma half-life while nanobodies are rather suited for imaging purposes due to their rapid clearance from the circulation. These differences are due to the fact that any excess of the smaller nanobodies (15-17 kDa) is rapidly cleared via renal elimination while excess of larger conventional antibodies (150 kDa) is retained in the circulation. So the major advantage of nanobodies for molecular imaging is the low background signal at early imaging time points regardless of the injected dose. This allows same day imaging and could be translatable to the clinical setting. Contrariwise, conventional antibodies have to be exactly titrated to minimize nonspecific background signals, while maintaining enough specific signal from the targeted tissue (unpublished data).
One of the limitations of the in vivo NIRF-imaging technique is the low penetration depth which generally allows only imaging of subcutaneous but not of orthotopic tumor models. However, this limitation might be overcome in an experimental setting by the recently developed tomographic photo-acoustic techniques that allow whole-body imaging of living mice 27. Another limitation of the NIRF-imaging technique is the assessment of the tissue dose as compared to radionuclide-mediated imaging. However, the nanobodies may be radiolabelled for positron emission tomography (PET) imaging of xenograft models and exact quantitative assessment of tracer biodistribution. Indeed, our NIRF-imaging results are in accordance with a recent study that compared nanobodies and conventional antibodies for PET imaging. The authors also came to the conclusion that nanobodies allow same-day imaging with high tumor-to-background ratios 15.
However, only the labelling of antibody constructs with the near-infrared fluorescent dye AF680 allowed us the comprehensive in vitro, in vivo and ex vivo near-infrared fluorescence imaging comparison using flow cytometry, fluorescence microscopy, and NIRF-imaging. For this reason, and because it is nonradioactive, highly sensitive, inexpensive, and uses comparatively easy-to-produce targeted probes, we advocate the use of the NIRF-imaging technique for evaluation of new antibody constructs in preclinical molecular imaging.
The authors have nothing to disclose.
This work was supported by the graduate school ‘Inflammation and regeneration’ of the Collaborative Research Centre 841 of the Deutsche Forschungsgemeinschaft (Alexander Lenz, Valentin Kunick, William Fumey), by the Collaborative Research Centre 877 of the Deutsche Forschungsgemeinschaft (Friedrich Koch-Nolte), by the Werner Otto Foundation (Peter Bannas), by the Wilhelm Sander Foundation (Peter Bannas, Friedrich Koch-Nolte), and by the Deutsche Forschungsgemeinschaft (Martin Trepel, Friedrich Haag and Friedrich Koch-Nolte). We thank the University Cancer Center Hamburg (UCCH) In Vivo Optical Imaging Core Facility and staff at UKE for consultation and their high quality service. The Core Facility was supported in part by grants from Deutsche Krebshilfe (German Cancer Aid).
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
AF680 protein labelling kit | Invitrogen | A20172 | |
Anti-CD16/CD32-antibody | BioXCell | BE0008 | |
Anti-CD31-antibody | Santa Cruz | sc-1506 | labeled with secondary antibody with AF488 |
Anti-CD45-antibody V450 | BD Biosciences | 560501 | |
AxioVision LE software | Zeiss | www.zeiss.com | |
Basement membrane matrix | BD Biosciences | 354234 | Alternative product can be used |
Cell strainer 70µm | Corning | 431751 | Alternative product can be used |
Confocal microscope | Leica | www.leica-microsystems.com | Leica SP5 with the following lasers: He-Neon for AF680, Argon Laser for AF488, and a 405-Diode for DAPI |
DAPI | Molcular Probes | D1306 | |
DC27.10 cells | laboratory specific | n/a | Other cells with different surface targets can be used |
DPBS | Sigma Aldrich | D8662 | |
FACS Canto II | BD Biosciences | www.bdbiosciences.com | |
Flourescence mircoscope | Zeiss | www.zeiss.com | Zeiss Axiovert 200 with Filter Set #32 for AF680: 000000-1031-354 |
ImageJ software | NIH | http://imagej.nih.gov/ij/ | |
IVIS 200 | Perkin Elmer | www.perkinelmer.com | Alternative in vivo imaging system can be used |
Leica LAS software | Leica | www.leica-microsystems.com | Software specific to microscope used |
Living Image software | Perkin Elmer | www.perkinelmer.com | Software specific to imaging system used |
Needles 30 G | BD Biosciences | 305128 | Alternative product can be used |
Nika102-antibody AF680 | laboratory specific | Other antibodies against different surface targets can be used | |
Paraformaldehyde | Sigma Aldrich | P6148 | Potential hazards: carcinogenic, can irritate the eyes and skin, contact may cause drying of the skin and/or allergic dermatitis |
s+16a-nanobody AF680 | laboratory specific | Other antibodies against different surface targets can be used | |
Syringes 1ml | Braun | 916 1406 V | Alternative product can be used |