Fast and precise targeting of single tumor cells in vivo by multimodal correlative microscopy.

Mouse handling

Mounting the cranial window, intracardiac injection of tumor cells and anesthesia To perform intravital imaging of the cerebral cortex, a cranial window was grafted onto an 8–10-week-old immunodeficient female mouse ( Mus musculus ) (NU/NU nude mouse, Charles River Laboratories International, Sulzfeld, Germany), as previously described. After a 3-week healing period, the mouse received a 100 µl solution of cytoplasmic GFP-expressing JIMT1 cells (1×10 7 cells/ml in PBS) through intracardiac injection through the left ventricle. Intravital imaging was performed immediately following injection, and on the second and third day post-injection. For imaging, the mouse was anesthetized by using isoflurane. To visualize the blood vessels, 100 µl of TRITC–dextran (500 kDa, Sigma-Aldrich) was administered to the mouse through injection into the tail vein. The mouse was placed on a custom-made holder (Deutsches Krebsforschungszentrum, Heidelberg, Germany) and remained anesthetized by using an isoflurane gas anesthesia system. All animal experiments were conducted in agreement with the regulations for animal experimentation dictated by the state Baden-Württemberg (Germany), and were approved by the state authorities.

Subcutaneous injection of tumor cells

An immunodeficient adult 8- to 10-week-old mouse (Rj:NMRI-Foxn1nu/Fox1nu, Janvier labs, Saint-Berthevin, France) was injected with the D2A1 Lifeact-YPet cell line, as previously described ( Karreman et al., 2014 ).

Two-photon excitation IVM and NIRB

IVM and NIRB through the cranial window The mouse was mounted on an upright confocal microscope (Zeiss LSM 7, controlled by the software Zen 2012, Carl Zeiss Microscopy GmbH, Jena, Germany) and kept at a steady temperature of 32°C by means of a heating pad. The holder kept the mouse head immobile and allowed analysis of the same imaging area over days by keeping track of the x , y coordinates of the microscope stage. IVM was performed at an excitation wavelength of 850 nm (680–1080 nm Chameleon Ultra II laser, Coherent, Santa Clara, CA) using a ×20 Zeiss W Plan Apochromat N.A. 1 objective (Carl Zeiss Microscopy GmbH). The GFP and TRITC emission wavelengths (510 and 573 nm, respectively) were collected by two hybrid non-descanned detectors. As the perfusion fixation leads to a wash out of the TRITC–dextran, fluorescent lectin from Lycopersicon esculentum (Sigma-Aldrich Chemie GmbH, Munich, Germany) was injected into the mouse tail vein in order to record the 3D organization of the blood vessels after fixation. The mouse was perfusion-fixed through intracardiac injection with 2.5% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) and 2% formaldehyde (Electron Microscopy Sciences) in 0.1 M PHEM buffer (comprising 60 mM PIPES, 25 mM HEPES, 10 mM EGTA and 2 mM MgCl, pH adjusted to 6.9). Following fixation, the ROI was imaged again based on the stored stage x , y -coordinates. NIRB was performed with the same laser that had been used for IVM, tuned to a 750 nm wavelength. Above the ROIs, at the level of the brain surface, a 150×150 µm 2 area was scanned in a single focal plane until the NIRB square became clearly visible through emission of autofluorescence in the green channel. Around the ROI, three bigger 300×300 µm 2 NIRB squares were drawn in non-symmetric positions to facilitate orientation and retrieval of the ROI upon dissection ( Fig. 2 ). The brain was removed from the skull and post-fixed by immersion in the same fixative at 4°C overnight. The following day, the fixative was replaced with 0.1 M PHEM buffer, and the brain was stored at 4°C until further processing. IVM and NIRB analysis of subcutaneous tumors in the mouse ear IVM and NIRB were performed as previously described ( Karreman et al., 2014 ) on the anesthetized mouse. After the IVM data acquisition, the mouse was fixed through intracardiac perfusion of 2.5% glutaraldehyde and 2% formaldehyde in 0.1 M PHEM buffer. The ear was dissected and immersed in the same fixative at 4°C overnight. The next day, the ear biopsy was transferred to 0.1 M PHEM buffer and stored at 4°C until further processing.

Processing for electron microscopy

To improve the penetration of the chemicals during processing, small samples (