Targeting the SphK1/S1P/PFKFB3 axis suppresses hepatocellular carcinoma progression by disrupting glycolytic energy supply that drives tumor angiogenesis

Abstract

Background: Hepatocellular carcinoma (HCC) remains a leading life-threatening health challenge worldwide, with pressing needs for novel therapeutic strategies. Sphingosine kinase 1 (SphK1), a well-established pro-cancer enzyme, is aberrantly overexpressed in a multitude of malignancies, including HCC. Our previous research has shown that genetic ablation of Sphk1 mitigates HCC progression in mice. Therefore, the development of PF-543, a highly selective SphK1 inhibitor, opens a new avenue for HCC treatment. However, the anti-cancer efficacy of PF-543 has not yet been investigated in primary cancer models in vivo, thereby limiting its further translation.

Methods: Building upon the identification of the active form of SphK1 as a viable therapeutic target in human HCC specimens, we assessed the capacity of PF-543 in suppressing tumor progression using a diethylnitrosamine-induced mouse model of primary HCC. We further delineated its underlying mechanisms in both HCC and endothelial cells. Key findings were validated in Sphk1 knockout mice and lentiviral-mediated SphK1 knockdown cells.

Results: SphK1 activity was found to be elevated in human HCC tissues. Administration of PF-543 effectively abrogated hepatic SphK1 activity and significantly suppressed HCC progression in diethylnitrosamine-treated mice. The primary mechanism of action was through the inhibition of tumor neovascularization, as PF-543 disrupted endothelial cell angiogenesis even in a pro-angiogenic milieu. Mechanistically, PF-543 induced proteasomal degradation of the critical glycolytic enzyme 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3, thus restricting the energy supply essential for tumor angiogenesis. These effects of PF-543 could be reversed upon S1P supplementation in an S1P receptor-dependent manner.

Conclusions: This study provides the first in vivo evidence supporting the potential of PF-543 as an effective anti-HCC agent. It also uncovers previously undescribed links between the pro-cancer, pro-angiogenic and pro-glycolytic roles of the SphK1/S1P/S1P receptor axis. Importantly, unlike conventional anti-HCC drugs that target individual pro-angiogenic drivers, PF-543 impairs the PFKFB3-dictated glycolytic energy engine that fuels tumor angiogenesis, representing a novel and potentially safer therapeutic strategy for HCC.

Keywords: Angiogenesis; Glycolysis; Hepatocellular carcinoma; PF-543; PFKFB3; Sphingosine kinase.

Fig. 1

PF-543 inhibits SphK1 activity in DEN-treated mice. A Total and phospho-SphK1 levels were examined using immunohistochemical staining in tumorous (T) and para-tumorous (para-T) tissues of human HCC specimens; scale bar = 50 μm; n = 10. B Schematic illustration of treatment schedules for in vivo studies, created with BioRender.com. C Body weight over 12 weeks of vehicle (veh) or PF-543 (PF) treatment. D Liver mass was weighed. E phospho(p)-SphK1 SphK1 and SphK2 protein levels in liver tissues were determined using Western blotting. F Hepatic levels of ceramide (Cer), sphingosine (Sph), and S1P were analyzed using lipidomics. C-F n = 9. Data are expressed as mean ± SD. *p < 0.05; ***p < 0.001

Fig. 2

Inhibition of SphK1 suppresses HCC progression in DEN-treated mice. DEN-injected mice were treated with vehicle (veh) or PF-543 (PF) for 12 weeks. A The number and maximal diameter of visible liver tumors were quantified from macroscopic images. B The number and maximal diameter of intrahepatic liver tumors were quantified from the scanning of the entire H&E-stained liver tissue sections. C Cell proliferation in non-tumorous (NT) and tumorous (T) liver tissues was stained and quantified using Ki67 immunohistochemistry; scale bar = 50 μm. Data are expressed as mean ± SD. n = 9. *p < 0.05; **p < 0.01; ***p < 0.001

Fig. 3

SphK1 inhibition or ablation reduces vessel density in HCC tumors. A DEN-injected mice were treated with vehicle (veh) or PF-543 (PF) for 12 weeks. Blood vessels in non-tumorous (NT) and tumorous (T) liver tissues were stained and quantified by CD31 immunohistochemistry; scale bar = 50 μm; n = 9. B Blood vessels in DEN-injected wild-type (WT) and Sphk1 knockout (KO) mice were examined by CD31 immunohistochemistry; scale bar = 50 μm; n = 5. Data are expressed as mean ± SD. **p < 0.01; ***p < 0.001

Fig. 4

Inhibition of SphK1 impairs angiogenesis in HUVECs. Primary HUVECs were treated with PF-543 at the indicated concentrations for 16 h. A and B Tube formation was induced by 50 ng/ml VEGF-A (A) or conditioned medium collected from Huh7 HCC cell culture (B). Quantification of tube formation is presented as the number of junctions, segment length, and total branching length. kpx, 1000 pixels; scale bar = 200 μm (A) or 100 μm (B); n = 3. C Sprouting assays were performed in three-dimensional spheroids. Quantitation of the sprouting is presented as cumulative sprout length and number of sprouts per spheroid; scale bar = 100 μm; n = 10. D Cell migration was determined by transwell assay, and migrated cells were stained with crystal violet; scale bar = 100 μm; n = 3. E Cell viability was determined using MTS assay; n = 4. F Cell death was assessed using flow cytometry with propidium iodide (PI) staining; PI-, living cells (green); PI+, dead cells (red); n = 3. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001

Fig. 5

SphK1 promotes angiogenesis by regulating the glycolytic modulator PFKFB3. A Primary HUVECs were treated with 5 µM PF-543 for 16 h, prior to stimulation with 50 ng/ml VEGF-A for 15 min. B Following 1 h pre-treatment with 10 µM MG-132, primary HUVECs were treated with PF-543 at 5 µM for 16 h. A and B Western blotting analyses of the indicated proteins; n = 3. C and D Glycolytic rate, capacity and reserve were determined using Seahorse real-time glycolytic stress assay in PF-543-treated (C) or shRNA-mediated SphK1 knockdown (D) primary HUVECs. ECAR, extracellular acidification rate; n = 5. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001

Fig. 6

S1P is required for glycolysis and angiogenesis in an S1P₁-dependent manner. Primary HUVECs were treated with 5 µM PF-543 (PF) and 0.5 µM S1P (S) for 16 h, with 2 µM S1P₁ antagonist W146 added 1 h prior to these treatments. A Levels of ceramide (Cer), sphingosine (Sph) and S1P were analyzed using lipidomics; n = 4. B phospho(p)-SphK1, SphK1 and PFKFB3 protein levels were determined using Western blotting; n = 3. C Glycolytic rate, capacity and reserve were examined using Seahorse real-time glycolytic stress assay. ECAR, extracellular acidification rate; n = 6. D Tube formation was quantified as the number of junctions, segment length, and total branching length. kpx, 1000 pixels; scale bar = 200 μm; n = 4. E Cell migration was determined by transwell assay, and migrated cells were stained with crystal violet; scale bar = 100 μm; n = 3. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001

Fig. 7

Model depicting the anti-HCC actions of PF-543. Elevated pro-angiogenic factors, exemplified by VEGF-A, act through their receptors to drive sprouting angiogenesis in tumor endothelial cells (EC), promoting tumor neovascularization and HCC progression. PFKFB3 serves as a molecular switch in this process, dictating the glycolytic energy supply that is essential for sprouting angiogenesis. The selective SphK1 inhibitor PF-543 abrogates S1P production, subsequently turning off the PFKFB3-mediated glycolytic switch, which leads to the inhibition of sprouting angiogenesis and eventually suppression of HCC progression. The graphical abstract was created with BioRender.com. . The 3D conformer of PF-543 was adapted from PubChem, pubchem.ncbi.nlm.nih.gov, CID 66,577,038