In vivo and ex vivo imaging

Various exosomes used in dual-targeting investigation were adjusted to the same concentration (300 μg/mL, 3.69×1010 particles/mL), and then labeled with fluorescent dye DiR or DiD (final dye concentration 0.05 mM). After 2 hours of incubation at 37℃, unbounded dye was removed using Exosome Spin Columns (Invitrogen) by following the corresponding instruction book, and the comparative fluorescence signal intensity of obtained eluates were reconfirmed by Kodak In-vivo Imaging System FX Pro. Different from nanovesicles, aMT-mv numbers were calculated with flow cytometry (Beckman) and unbound dye were removed from vesicle by centrifugation (10000 g). The same particle number and injection amount (3.69×109, 30 μg) of above vesicles were subcutaneously injected on the back of tumor-bearing mice. For in vivo imaging, mice were imaged using the Kodak In-vivo Imaging System FX Pro at different time points to evaluate the DiR (or DiD) fluorescence signal distribution at LNs and tumors. For ex vivo detection, mice were euthanized and dissected at 48 h, the LNs and tumors were also imaged by Kodak In-vivo Imaging System FX Pro. In our preliminary experiment, the aMT-exos accumulating to inguinal LNs (near aMT-exos injection site) and axillary LNs (near tumor site) accounted for more than 95% of that draining to LNs. Therefore, we isolated inguinal LNs and axillary LNs for intensive study in our manuscript. These tissues were further made into frozen sections for analysis to further evaluate the exosomes infiltration.