Inhibition of hypoxia inducible factor 1 and topoisomerase with acriflavine sensitizes perihilar cholangiocarcinomas to photodynamic therapy

Abstract

Background: Photodynamic therapy (PDT) induces tumor cell death by oxidative stress and hypoxia but also survival signaling through activation of hypoxia-inducible factor 1 (HIF-1). Since perihilar cholangiocarcinomas are relatively recalcitrant to PDT, the aims were to (1) determine the expression levels of HIF-1-associated proteins in human perihilar cholangiocarcinomas, (2) investigate the role of HIF-1 in PDT-treated human perihilar cholangiocarcinoma cells, and (3) determine whether HIF-1 inhibition reduces survival signaling and enhances PDT efficacy.

Results: Increased expression of VEGF, CD105, CD31/Ki-67, and GLUT-1 was confirmed in human perihilar cholangiocarcinomas. PDT with liposome-delivered zinc phthalocyanine caused HIF-1α stabilization in SK-ChA-1 cells and increased transcription of HIF-1α downstream genes. Acriflavine was taken up by SK-ChA-1 cells and translocated to the nucleus under hypoxic conditions. Importantly, pretreatment of SK-ChA-1 cells with acriflavine enhanced PDT efficacy via inhibition of HIF-1 and topoisomerases I and II.

Methods: The expression of VEGF, CD105, CD31/Ki-67, and GLUT-1 was determined by immunohistochemistry in human perihilar cholangiocarcinomas. In addition, the response of human perihilar cholangiocarcinoma (SK-ChA-1) cells to PDT with liposome-delivered zinc phthalocyanine was investigated under both normoxic and hypoxic conditions. Acriflavine, a HIF-1α/HIF-1β dimerization inhibitor and a potential dual topoisomerase I/II inhibitor, was evaluated for its adjuvant effect on PDT efficacy.

Conclusions: HIF-1, which is activated in human hilar cholangiocarcinomas, contributes to tumor cell survival following PDT in vitro. Combining PDT with acriflavine pretreatment improves PDT efficacy in cultured cells and therefore warrants further preclinical validation for therapy-recalcitrant perihilar cholangiocarcinomas.

Keywords: cancer therapy; drug delivery system; extrahepatic cholangiocarcinoma; hypoxia; tumor targeting.

Figure 1. Hypoxia-related protein expression in an extrahepatic perihilar cholangiocarcinoma (Klatskin tumor) resection specimen

Serial histological sections were used for protein profiling in the same region. A. Hematoxylin and eosin staining of a cholangiocarcinoma section containing tumor mass (intense purple staining, circular structure, bottom left, marked with an asterisk in panels A-E), tumor stroma, and native tissue (e.g., pre-existent arterial structures). B. VEGF staining (brown), showing intense staining of the tumor mass and vascular endothelium (insert) as well as pre-existent biliary structures (insert, arrow). The insert corresponds to the demarcated region in the low-magnification image. C. CD105 staining (brown), showing no staining in the tumor mass and positive staining of the vascular endothelial cells in the tumor stroma (insert, arrow, indicates neovessel formation). D. Angiogenesis was further confirmed with CD31 (blue) and Ki-67 (red) double staining, showing that the blood vessels in the tumor stroma contain proliferating endothelial cells (insert, arrows). E. GLUT-1 staining (brown) was largely absent in the tumor mass and stroma, indicating that these regions were not affected by hypoxia. In some regions of the tumor, however, positive staining was observed (insert, arrow). F. Strong GLUT-1 staining was found in another region of the histological specimen. Magnification: 4× (A-E, scale bar = 100 μm) and 10× (F, scale bar = 40 μm).

Figure 2. Analysis of HIF-1α activation after PDT

A. SK-ChA-1 cells were incubated with increasing concentration of ZnPC-ETLs, treated with PDT, and maintained at normoxic (red bars) or hypoxic (blue bars) culture conditions. Cell viability was determined 24 hours post-PDT (n = 6 per group). B. SK-ChA-1 cells were treated with PDT (10 μM ZnPC-ETLs, final lipid concentration) or received a control (CTRL) treatment, after which the cells were placed in a hypoxic chamber up to 240 minutes (min) post-PDT. HIF-1α protein levels were determined using Western blotting. As a positive control, cells were incubated with 500 μM CoCl₂ for 24 hours (top panel). Next, the HIF-1α protein bands and their corresponding β-actin protein bands were quantified using ImageJ software [74] and each HIF-1α value was divided by its corresponding β-actin value. All values were normalized to the positive control (CoCl₂) (bottom panel). C. SK-ChA-1 cells were either left untreated (grey bars) or treated with PDT (white bars), and subsequently placed at hypoxic conditions for 4 hours. Thereafter, downstream targets of HIF-1 were analyzed with qRT-PCR (n = 3 per group). Readers are referred to the experimental section for the significance of the statistical symbols.

Figure 3. Intracellular ACF localization

A–D. SK-ChA-1 cells were either left untreated or treated with PDT and subsequently incubated with ACF for 4 hours under normoxic (A, B) or hypoxic culture conditions (C, D). ACF localization was determined using confocal microscopy (ACF in green; Nile Red (membrane staining) in red).

Figure 4. Combination treatment of ACF with PDT

A. Evaluation of ACF stability using increasing concentrations of ZnPC-containing cell phantoms (ZnPC-CPs) with or without irradiation. ACF degradation was monitored using fluorescence spectroscopy (n = 4 per concentration). B. Cells were incubated with ACF for 24 hours, after which the uptake of ACF was determined using fluorescence spectroscopy. Data were normalized to protein content (n = 4 per concentration). C. ACF toxicity was determined after 24-hour incubation under either normoxic (red line) or hypoxic (blue line) conditions using the WST-1 method (n = 4 per group). Treatment efficacy of ACF and ACF + PDT was tested in SK-ChA-1 cells after 4 hours at D. normoxic and E. hypoxic culture conditions (n = 6 per group). (F, G) Relative caspase 3/7 activity was determined 4 hours after PDT at incubation at F. normoxic or G. hypoxic culture conditions (n = 6 per group). H. Lactate production by SK-ChA-1 cells treated with ACF and ACF + PDT was evaluated after 24 hours at normoxic (red bars) or hypoxic (blue bars) culture conditions (n = 6 per group). I–P. Analysis of DNA damage after control (CTRL), ACF, PDT, and ACF + PDT treatment. Cells were kept for 4 hours under normoxic (I-L) or hypoxic conditions (M-P) post-treatment. Cells were stained with DAPI (nuclei, blue) and phospho-H2AX (DNA double-strand breaks, red). The arrowhead in panel P indicates apoptosis. Readers are referred to the experimental section for the significance of the statistical symbols.

Figure 5. Gene expression analysis after control (CTRL), ACF, PDT, and ACF + PDT treatment

Gene expression levels were obtained by qRT-PCR from SK-ChA-1 cells as analyzed 0 hours, 2 hours, or 4 hours post-treatment under hypoxic conditions. The plotted heat map data represents the log₂-transformed fold change of each data point in relation to the 0-hour normoxic CTRL. Upregulated genes are depicted in red, downregulated genes in green. Numeric values are provided in Table S2.

Figure 6. Response to ACF after prolonged exposure

A–D. SK-ChA-1 cells were incubated with ACF for either (A, B) 24 hours or (C, D) 48 hours, after which the cell cycle profile was analyzed with flow cytrometry using propidium iodide staining (n = 3 per group). E. Flow cytometric analysis of SK-ChA-1 cells that were incubated with ACF for either 24 hours (in grey) or 48 hours (in white), after which the fraction of apoptotic (annexin V-positive) and F. necrotic (TO-PRO-3-positive) cells was determined (n = 3 per group). (G, H) SK-ChA-1 cells were exposed to ACF for G. 24 hours or H. 48 hours and intracellular DCF fluorescence was determined as a measure of ROS production. I–P. Analysis of DNA damage after control (CTRL) or ACF treatment. SK-ChA-1 cells received ACF or CTRL treatment for (I-L) 24 hours or (M-P) 48 hours, and were subsequently stained with DAPI (nuclei, blue) and phospho-H2AX (DNA double-strand breaks, red).