Azido-galactose outperforms azido-mannose for metabolic labeling and targeting of hepatocellular carcinoma

Abstract

Metabolic glycoengineering of unnatural monosaccharides provides a facile method to label cancer cells with chemical tags for glycan imaging and cancer targeting. Multiple types of monosaccharides have been utilized for metabolic cell labeling. However, the comparison of different types of monosaccharides in labeling efficiency and selectivity has not been reported. In this study, we compared N-azidoacetylgalactosamine (GalAz) and N-azidoacetylmannosamine (ManAz) for metabolic labeling of HepG2 hepatocellular carcinoma in vitro and in vivo. GalAz showed higher labeling efficiency at low concentrations, and outperformed ManAz in metabolic labeling of HepG2 tumors in vivo. GalAz mediated labeling of HepG2 tumors with azido groups significantly improved the tumor accumulation of dibenzocyclooctyne (DBCO)-Cy5 and DBCO-doxorubicin conjugate via efficient Click chemistry. This study, for the first time, uncovered the distinct labeling efficiency and selectivity of different unnatural monosaccharides in liver cancers.

Figure 1.

(a) Schematic illustration of GalAz- or ManAz-mediated metabolic labeling of cancer cells with azido groups and subsequent detection by DBCO-Cy5 via Click chemistry. (b) CLSM images of HepG2 cells after treated with GalAz (200 μM), ManAz (200 μM), and PBS, respectively for 72 h and stained with DBCO-Cy5 (20 μM, red) for 40 min. Cell nuclei were stained with DAPI (blue). Scale bar: 10 μm. (c) Flow cytometry profiles of HepG2 cells following the same treatment in (b). (d) In vitro labeling kinetics of GalAz in HepG2 cells. HepG2 cells were incubated with different concentrations of GalAz (10 μM, 50 μM, 200 μM, 1 mM, and 5 mM) for different time (1, 6, 12, 24, 48, 72, 96, and 120 h), and stained with DBCO-Cy5 (20 μM) for 40 min. Average Cy5 fluorescence intensity was measured on a GE-analyzer. (e) Comparison of in vitro labeling kinetics of GalAz and ManAz at a concentration of 50 μM and 1 mM, respectively.

Figure 2.

(a) Structure of ¹⁴C-Gal and ¹⁴C-Man, as representatives of GalAz and ManAz, respectively. (b) ¹⁴C radioactivity of proteins extracted from HepG2 cells after treatment with ¹⁴C-Gal, ¹⁴C-Man, and PBS, respectively. (c) Biodistribution of ¹⁴C-Gal and ¹⁴C-Man in athymic nude mice bearing subcutaneous HepG2 tumors at 6 h and 5 days p.i., respectively. Tissues were harvested from mice, homogenized, lysed, and measured for the radioactivity on a Tricarb liquid scintillation counter. All numerical data were presented as mean ± SD (n=4) and analyzed by Student’s t-test (two-tailed, 0.01 < *P ≤ 0.05, 0.001 < **P ≤ 0.01, and ***P ≤ 0.001).

Figure 3.

(a) Time frame of in vivo metabolic labeling and targeting study. Athymic nude mice bearing subcutaneous HepG2 tumors were i.v. injected with GalAz (200 mg/kg), ManAz (200 mg/kg), and PBS, respectively once daily for three days. DBCO-Cy5 (5 mg/kg) was i.v. injected on day 4. Tissues were harvested at 48 h p.i. of DBCO-Cy5 for Cy5 quantification. (b) Biodistribution of Cy5 in tissues at 48 h p.i. of DBCO-Cy5. Data were presented as mean ± SD (n = 4–5) and analyzed by Student’s t-test (two-tailed, 0.01 < *P ≤ 0.05, 0.001 < **P ≤ 0.01, and ***P ≤ 0.001). (c) Representative CLSM images of tumor tissue sections from different groups. Cell nuclei were stained with DAPI (blue). Scale bar: 10 μm.

Figure 4.

(a) Structure of pH-responsive DBCO-doxorubicin conjugate, DBCO-hz-Dox. (b) Time frame of acute efficacy study. Athymic nude mice bearing HepG2 tumors were i.v. injected with Gal-N₃ (200 mg/kg) or Man-N₃ (200 mg/kg) or PBS once daily for three days. DBCO-hz-Dox (8.0 mg/kg, Dox equivalent) was i.v. injected on day 4. Tissues were harvested from mice at 48 h p.i. of DBCO-hz-Dox. (c) Representative TUNEL staining sections of HepG2 tumors from mice treated with GalAz + DBCO-hz-Dox, ManAz + DBCO-hz-Dox, DBCO-hz-Dox, GalAz, ManAz, and PBS, respectively. Scale bar: 50 μm. (d) Quantification of TUNEL stains via ImageJ. The apoptosis index was determined as the ratio of apoptotic cell number (TUNEL, red) to the total cell number (DAPI, blue). 20 tissue sections were counted per tumor; n=5. Statistical significance analysis was performed by Student’s t-test (two tailed, 0.01 < *P ≤ 0.05, 0.001 < **P ≤ 0.01, and ***P ≤ 0.001).