Ferroportin depletes iron needed for cell cycle progression in head and neck squamous cell carcinoma

Abstract

Introduction: Ferroportin (FPN), the only identified eukaryotic iron efflux channel, plays an important role in iron homeostasis and is downregulated in many cancers. To determine if iron related pathways are important for Head and Neck Squamous Cell Carcinoma (HNSCC) progression and proliferation, we utilize a model of FPN over-expression to simulate iron depletion and probe associated molecular pathways.

Methods: The state of iron related proteins and ferroptosis sensitivity was assessed in a panel of metastatic HNSCC cell lines. Stable, inducible expression of FPN was confirmed in the metastatic HNSCC lines HN12 and JHU-022 as well as the non-transformed normal oral keratinocyte (NOK) cell line and the effect of FPN mediated iron depletion was assessed in these cell lines.

Results: HNSCC cells are sensitive to iron chelation and ferroptosis, but the non-transformed NOK cell line is not. We found that FPN expression inhibits HNSCC cell proliferation and colony formation but NOK cells are unaffected. Inhibition of cell proliferation is rescued by the addition of hepcidin. Decreases in proliferation are due to the disruption of iron homeostasis via loss of labile iron caused by elevated FPN levels. This in turn protects HNSCC cells from ferroptotic cell death. Expression of FPN induces DNA damage, activates p21, and reduces levels of cyclin proteins thereby inhibiting cell cycle progression of HNSCC cells, arresting cells in the S-phase. Induction of FPN severely inhibits Edu incorporation and increased β-galactosidase activity, indicating cells have entered senescence. Finally, in an oral orthotopic mouse xenograft model, FPN induction yields a significant decrease in tumor growth.

Conclusions: Our results indicate that iron plays a role in HNSCC cell proliferation and growth and is important for cell cycle progression. Iron based interventional strategies such as ferroptosis or iron chelation may have potential therapeutic benefits in advanced HNSCC.

Keywords: Oral Epithelial Cells; cell proliferation; ferroportin; head and neck squamous cell carcinoma; iron metabolism; iron/cell proliferation; senescence.

Figure 1

HNSCC cell lines exhibit an increased dependence on iron and are susceptible to iron chelation and ferroptosis. (A) Cells were grown with increasing amounts of deferoxamine (DFO) for 72 hours and cell viability was assessed via CellTiter-Glo assay. (B) Cells were treated with the indicated amount of Dp44mT for 72 hours and cell viability was measured via CellTiter-Glo. (C–G) The indicated cell line was seeded and treated with a vehicle control (DMSO), Ferrostatin-1 (Fer-1, 10μM), the indicated ferroptosis inducer (IKE or ML162), or a combination of both for 72 hours. Cell viability was measured via CellTiter-Glo assay. Values were normalized to a non-treated control with DMSO. (H) Western Blot of TFR1, FPN, FTH1, with GAPDH as a loading control. Quantification of TFR1, FPN, and FTH1 was performed using GAPDH as a loading control and normalized to the NOKs (*P<0.05; **P<0.01; ***P<0.001).

Figure 2

Stable Expression of FPN in the HN12, JHU-022, and NOK cell lines. (A) Western blot of FPN in the HN12-FPN/Luc, JHU-022-FPN/Luc, and NOK-FPN/Luc after treatment with 0.5 μg/mL of doxycycline for 48 hours. For rescue experiments, hepcidin was added to cell cultures with doxycycline at 20μM. (B) HN12-FPN/Luc and JHU-022-FPN/Luc were stimulated with 0.5 μg/mL dox for 72 hours and blotted for iron related proteins transferrin receptor 1 (TfR1), ferritin heavy chain 1 (FTH1), or N-myc downstream regulated protein (NDRG1). Quantification of TFR1, FTH1, and NDRG1 was performed using GAPDH as a loading control and normalized to the Luc control line. (C) (D) Measurement of the labile iron pool in cells using 1 μM FerroOrange. Changes in the levels of labile iron are represented as ratios of induced (+dox) samples normalized to un-induced (- dox) samples. (*P<0.05). All data is representative of 3 biological replicates.

Figure 3

Expression of Ferroportin inhibits growth and proliferation of HNSCC. HN12 (A), JHU-022 (C) and NOK (E) FPN expressing or Luc control expressing cells were seeded in 96 well plates at 1000 cells per well with 0.25 μg/mL dox and grown for 4 days. Growth was assessed using CellTiter-Blue reagent. HN12-FPN (B), JHU-022-FPN (D), and NOK-FPN (F) cells were seeded at 25,000 cells per well in a 6-well plate +/- 0.25 μg/mL dox. After 4 days cells were counted via hemocytometer. HN12-FPN (G) and JHU-FPN (H) were seeded in 6 well plates and treated with doxycycline (0.25 μg/mL), hepcidin (10 μM), or left untreated. Cells were grown for 4 days, after which cells were trypsinized and counted via hemocytometer. (*P<0.05; **P<0.01; ***P<0.001). All data is representative of 3 biological replicates.

Figure 4

Expression of Ferroportin inhibits colony formation of HNSCC cell lines. Clonogenic assay quantification and a representative plate of HN12 (A), JHU-022 (B), and NOK (C) FPN expressing or Luc control expressing cells. Cells were seeded at 250 cells per well with 0.25 μg/mL dox and colonies were counted after 7-8 days. (***P<0.001) Data and images are representative of 3 biological replicates.

Figure 5

Ferroportin expression causes cell cycle arrest in HNSCC. FPN or Luciferase control expressing HN12 (A), JHU-022 (B) and NOK (C) were grown with 0.5 μg/mL dox for 3 (HN12) or 4 days (NOK, JHU-022) and stained with propidium iodide to measure DNA content. The DNA content was assessed by flow cytometry and the cell cycle distribution was analyzed using FlowJo. All data is representative of 3 biological replicates.

Figure 6

Ferroportin expression in HNSCC cell lines downregulates key cell cycle related genes (A), (B) The mRNA levels of p21, cyclin A, cyclin B, cyclin D and cyclin E were assessed by qRT-PCR in HN12-FPN/Luc and JHU-022-FPN/Luc expressing cells grown in 0.5 μg/mL dox for 3 days. Levels of target genes are normalized to the levels of both GAPDH and beta-actin. Data is representative of 4 technical replicates performed in 3 biological replicates. (C) Western blot of p21 levels in HN12-FPN and HN12-Luc expressing cells. (D) Western blot of cyclinD1 levels in HN12-FPN/Luc and JHU-022/Luc expressing cells grown in 0.5 μg/mL dox for 3 days (E) HN12-FPN/Luc and JHU-022-FPN/Luc cells were stimulated with 0.5 μg/mL dox for 3 days and stained for p-γH2Ax (green) or DAPI (blue). Graphs represent the ratio of cells + for staining over the total number of quantified cells. Images are representative of 3 biological replicates. (*P<0.05; **P<0.01; ***P<0.001).

Figure 7

FPN expression induces senescence in HN12. (A), (B) EDU incorporation of HN12-FPN/Luc (A) and JHU-022-FPN/Luc (B) grown with 0.5 μg/mL dox for 3 days. (C, D). Beta-galactosidase activity of HN12-FPN/Luc (C) and JHU-022-FPN/Luc (D) after 7 days of exposure to 0.5 μg/mL dox. One μUnit is equal to 1 pmol of product generated per min. (E) HN12-FPN/Luc and JHU-022 FPN/Luc expressing cells were grown with 0.5 μg/mL dox for 37 days and stained for senescence using CellEvent Green Senescence Kit (F), (G) Clonogenic assay of HN12-FPN/Luc and JHU-022-FPN/Luc expressing cells grown for 72 hours with 0.5 μg/mL after recovery in dox free media for 48 hours. (*P<0.05; **P<0.01).

Figure 8

FPN expression inhibits growth of HN12 cells in an orthotopic xenograft model. HN12-FPN cells were implanted into the left cheek of mice and allowed to form tumors for 5 days. FPN expression was induced by adding 2mg/mL Doxycycline to the drinking water of mice. Control mice received normal drinking water. After 21 days, tumors were harvested and tumor weight (A) and tumor volume (B) were measured. For each group n=7 mice. (*P<0.05).