The Application of Heptamethine Cyanine Dye DZ-1 and Indocyanine Green for Imaging and Targeting in Xenograft Models of Hepatocellular Carcinoma

Abstract

Near infrared fluorescence (NIRF) imaging has strong potential for widespread use in noninvasive tumor imaging. Indocyanine green (ICG) is the only Food and Drug Administration (FDA) -approved NIRF dye for clinical diagnosis; however, it is unstable and poorly targets tumors. DZ-1 is a novel heptamethine cyanine NIRF dye, suitable for imaging and tumor targeting. Here, we compared the fluorescence intensity and metabolism of DZ-1 and ICG. Additionally, we assayed their specificities and abilities to target tumor cells, using cultured hepatocellular carcinoma (HCC) cell lines, a nude mouse subcutaneous xenograft model of liver cancer, and a rabbit orthotopic transplantation model. We found that DZ-1 accumulates in tumor tissue and specifically recognizes HCC in subcutaneous and orthotopic models. The NIRF intensity of DZ-1 was one order of magnitude stronger than that of ICG, and DZ-1 showed excellent intraoperative tumor targeting in the rabbit model. Importantly, ICG accumulated at tumor sites, as well as in the liver and kidney. Furthermore, DZ-1 analog-gemcitabine conjugate (NIRG) exhibited similar tumor-specific targeting and imaging properties, including inhibition of tumor growth, in HCC patient-derived xenograft (PDX) mice. DZ-1 and NIRG demonstrated superior tumor-targeting specificity, compared to ICG. We show that DZ-1 is an effective molecular probe for specific imaging, targeting, and therapy in HCC.

Keywords: hepatocellular carcinoma; heptamethine cyanine dyes; intraoperative navigation; near-infrared fluorescence; optical imaging; xenograft model.

Figure 1

Uptake of DZ-1 and indocyanine green (ICG) in hepatocellular carcinoma (HCC). (A) Confocal microscope analyses of DZ-1 dye uptake by HCC. Original magnification: 400×; scale bars: 50 μm; (B) Ratio of near infrared fluorescence (NIRF, red) dye uptake intensity in HCC at successive time points. Data are presented as mean ± standard deviation (SD) (n = 5); (C,D) Co-localization of DZ-1 or ICG (red) and LysoTracker (green, top panel) or MitoTracker (yellow, bottom panel) in Hep3B cells, as determined by confocal microscopy. Cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, blue). Original magnification: 1200×; Scale bars, 20 μm.

Figure 1

Uptake of DZ-1 and indocyanine green (ICG) in hepatocellular carcinoma (HCC). (A) Confocal microscope analyses of DZ-1 dye uptake by HCC. Original magnification: 400×; scale bars: 50 μm; (B) Ratio of near infrared fluorescence (NIRF, red) dye uptake intensity in HCC at successive time points. Data are presented as mean ± standard deviation (SD) (n = 5); (C,D) Co-localization of DZ-1 or ICG (red) and LysoTracker (green, top panel) or MitoTracker (yellow, bottom panel) in Hep3B cells, as determined by confocal microscopy. Cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, blue). Original magnification: 1200×; Scale bars, 20 μm.

Figure 2

The distribution of DZ-1 and ICG in the organs of subcutaneous tumor model. (A) Dual ex vivo BLI/NIRF imaging of mice with Hep3B-Luc subcutaneous xenografts treated with varying doses of ICG or DZ-1; (B) Quantification of NIRF intensity within the tumor area (per cm²) of subcutaneous tumor xenografts; (C) NIRF/BLI signal intensity correlation in mice (n = 5) with subcutaneous tumor xenografts (right); (D,E) Distribution intensity per cm² of DZ-1 and ICG in the organs of subcutaneous tumor models at successive times.

Figure 3

Metabolism of DZ-1 and ICG in the organs of subcutaneous tumor model. (A,B) NIRF intensities per unit area (cm²) in subcutaneous tumor xenograft mice subjected to DZ-1 or ICG were measured at 0.5, 2, 3, 6, 10, 24, and 48 h; (C,D). Ratio of tumor NIRF intensity to background NIRF intensity at various time points after injection of DZ-1 or ICG in subcutaneous tumor xenograft mice.

Figure 4

Comparison of ICG and DZ-1 imaging in orthotopic liver tumors. (A) Representative NIRF optical images of tumor sites by injection of DZ-1 (0.5 μmol/kg) in vivo. Scale bar: 20 μm. (B) Representative NIRF optical images of tumor sites by injection of ICG (10 μmol/kg) in vivo. However, there were more differences between the bioluminescence signal range and fluorescence in the tumor site in vitro. Scale bar: 20 μm. (C) A large area of carcinogenesis of liver identified by ICG (10 μmol/kg). The ex vivo signal range of bioluminescence in the organ is consistent with that of fluorescence.

Figure 5

The application of DZ-1 and ICG in surgical exploration. (A) DZ-1 imaging using NIRF optical fiber intraoperative guidance in rabbits with VX2 liver cancer (left). Ex vivo NIRF imaging of rabbit liver cancer using small animal optical imaging system (right). Hematoxylin-eosin (H&E) staining of the fluorescent tissue sections; (B) NIRF intensity/tumor area (per cm²) of DZ-1 uptake in liver tumor of rabbit; (C) ICG imaging using NIRF optical fiber intraoperative guidance in rabbits with VX2 liver cancer (left). Ex vivo NIRF imaging of rabbit liver cancer using small animal optical imaging system (right). H&E staining of the fluorescent tissue sections. 10× and 20×, Scale bars, 20 μm; (D) NIRF intensity/tumor area (per cm²) of ICG uptake in liver tumor of the rabbits. Blue circle indicate the tumor location.

Figure 6

Imaging and targeting of DZ-1 drug conjugate in a liver cancer patient-derived xenograft (PDX) model. (A) H&E, NIRF, and immunohistochemistry (IHC) analyses of liver cancer tissues derived from both PDX mouse models and original patient samples. Original magnification: 400×; scale bars represent 20 μm; (B) DZ-1 dye uptake by Hep3B cells with a prior exposure to either HIF1α stabilizers (DMOG), or OATP inhibitor (BSP). Scale bar, 50 μm; (C) NIRF optical imaging of PDX models established by implanting 3 different human liver cancer specimens to subcutaneous of nude mice. Strong fluorescent signal was detected at subcutaneous tumor site; (D–F) Inhibition of NIRG on the tumor growth from liver cancer PDX mouse, including C64003, C34566, and B66873. d represents the treatment time.