In vivo tumor imaging using a near-infrared-labeled endostatin molecule

Purpose: Endostatin is a 20-kD C-terminal fragment of collagen XVIII and is a potent inhibitor of angiogenesis. Imaging technologies that use near-infrared (NIR) fluorescent probes are well suited to the laboratory setting. The goal of this study was to determine whether endostatin labeled with a NIR probe (Cy5.5) could be detected in an animal after intraperitoneal injection and whether it would selectively localize in a tumor. Methods: Endostatin was conjugated to Cy5.5 monofunctional dye and purified from free dye by gel filtration. LLC, a murine tumor, was implanted in C57BL/6 mice. The tumors were allowed to grow to 350 mm(2), at which point the mice were injected with 100 mug/100 muL endostatin-Cy5.5 and imaged at various points under sedation. Imaging was performed using a lightproof box affixed to a fluorescent microscope mounted with a filter in the NIR bandwidth (absorbance maximum 675 run and emission maximum 694 nm). Images were captured by a CCD and desktop computer and stored as 16-bit Tiff files. The mice were also serially imaged for uptake into the tumor and washout from the tumor. Results: After intraperitoneal injection, endostatin-Cy5.5 was quickly absorbed, producing a NIR fluorescent image of the tumors at 24 h that persisted through 7 days. However, the signal peaked at 42 h after injection. Control animals included mice containing green fluorescent protein (GFP) under the control of an actin promotor, which expresses GFP in every cell; tumor-free mice injected with endostatin-Cy5.5; mice with tumors that were not injected with endostatin-Cy5.5; and mice with tumors injected with dye alone. In the four sets of control animals, no NIR photon emissions were detected at 24 hours or 5 days. Only the GFP mouse was detected using the GFP filter. Unlike previous analogous studies with 4-N-(S-glutathionylacetyl)amino) phenylarsenoxide (GSAO)-Cy5.5 in which the tumor image faded with time, the endostatin-Cy5.5 NIR signal was emitted from the tumor up to 7 days after injection, the last time point examined. Conclusion: The results of this study demonstrated that endostatin covalently bound to Cy5.5 will migrate from a distant intraperitoneal injection site to a tumor. These data indicate that endostatin-Cy5.5 is appropriate for selectively imaging tumors in experimental animals. Furthermore, data suggest that the anti-angiogenic effect of endostatin occurs through a local mechanism of action, within the tumor or tumor vasculature, rather than through a systemic mechanism. (C) 2004 Elsevier Inc.