Mevalonate pathway activity as a determinant of radiation sensitivity in head and neck cancer

Abstract

Radioresistance is a major hurdle in the treatment of head and neck squamous cell carcinoma (HNSCC). Here, we report that concomitant treatment of HNSCCs with radiotherapy and mevalonate pathway inhibitors (statins) may overcome resistance. Proteomic profiling and comparison of radioresistant to radiosensitive HNSCCs revealed differential regulation of the mevalonate biosynthetic pathway. Consistent with this finding, inhibition of the mevalonate pathway by pitavastatin sensitized radioresistant SQ20B cells to ionizing radiation and reduced their clonogenic potential. Overall, this study reinforces the view that the mevalonate pathway is a promising therapeutic target in radioresistant HNSCCs.

Keywords: HNSCC; DNA damage response; mevalonate pathway; radiation sensitivity; senescence; statin.

Figure 1

Proteomic analysis reveals a distinct radioresistant HNSCC cell proteome. (A) Protocol schematic for mass spectrometry assay. Whole‐cell protein lysates were prepared from SCC61 (radiosensitive), JSQ3 (radioresistant), and SQ20B (radioresistant) HNSCC cell lines. Proteins were separated by gel electrophoresis, digested with trypsin, and analyzed via LC‐MS/MS. (B) Venn diagram showing proteins uniquely identified in one or two cell lines, or proteins shared between all cell lines. 201 proteins were shared by the two radioresistant cell lines (SQ20B and JSQ3) and 69 proteins were unique to radiosensitive cells (SCC61), while 4102 proteins were shared by all three cell lines. (C) Venn diagram showing distribution of up‐ and downregulated proteins shared between the two radioresistant cell lines, SQ20B and JSQ3, compared to the radiosensitive cell line, SCC61. In general, most proteins found in both radioresistant cell lines were similarly regulated, with 585 shared proteins upregulated and 803 shared proteins downregulated. Only 74 proteins displayed variable regulation. Only circled groups were considered for further analysis. (D) Representation of quantitative mass spectrophotometric intensity ratios of SQ20B/SCC61 (x‐axis) and JSQ3/SCC61 (y‐axis) shown on log₂ scale. A 1.5‐fold (0.58 log₂) change was considered statistically significant. Upregulated proteins, red. Downregulated proteins, green. Variably regulated proteins, gray; nonsignificant proteins, black. The data show a linear trend, with proteins up‐ or down‐regulated in the SQ20B vs. SCC61 cell line tending to be similarly regulated in JSQ3 vs. SCC61 samples.

Figure 2

Gene ontology analysis of proteomic data of mevalonate pathway activity and substrate validation. (A) Proteins were separated into functional categories by gene ontology (GO) terms related to mevalonate pathway activity, including mevalonate biosynthetic pathway, cholesterol biosynthesis, protein farnesylation, and protein geranylgeranylation. Protein data are shown as the log₂ fold‐change ratio for SQ20B/SCC61 (labeled SQ20B) and JSQ3/SCC61 (labeled JSQ3). (B) Amount of HMGCR protein is increased in radioresistant cells. Cell samples were collected during active growth, lysed, and a western blot was performed. (C) Mean cholesterol levels in SCC61, JSQ3, and SQ20B 48 h after cell passage (during active growth). 0.7 × 10⁶ cells were assayed per sample. Statistical significance was determined using the parametric unpaired t‐test. Error bars: SD, n = 3 biological replicates. Significance is indicated by asterisks (*P < 0.05). Cholesterol levels were significantly increased in the radioresistant cell lines.

Figure 3

Inhibition of mevalonate pathway by pitavastatin inhibits cellular proliferation and cholesterol levels. (A) Colony formation assay of SCC61, SQ20B, and JSQ3 treated with statins. SCC61, SQ20B, and JSQ3 cells were treated with 2.5, 5, or 10 μm pitavastatin, simvastatin, lovastatin, atorvastatin (lipophilic), or pravastatin or rosuvastatin (hydrophilic) for 1 h prior to IR. After 4 days of culture, crystal violet staining was performed and plates were imaged. Lipophilic statins, notably pitavastatin, and reduced colony formation in cell lines. (B) ATP luminescence viability assay was performed on SCC61, JSQ3, and SQ20B cells to measure viability 48 h after exposing cells to 0, 4, 6, 8, or 10 μm PIT. Three biological replicates were measured; mean ATP luminescence and standard deviation are shown. Viability was reduced by PIT in all cell lines, but markedly so in the radioresistant cell lines, particularly SQ20B. Statistical significance was determined using the parametric unpaired multiple t‐test. Significant conditions are indicated by asterisks (*P < 0.05).

Figure 4

Elevation of LDL receptor levels and LDL uptake characterize radioresistant HNSCC cells. (A) Immunostaining of LDL receptor, LDLR, on living HNSCC cell lines analyzed by flow cytometry. (B) Uptake of fluorescently labeled LDL‐DyLight 488 by HNSCC cells analyzed by flow cytometry. (C) Microscopy images of biological replicates of cell populations treated as shown in (B); scale bars, 20 μm. Elevated levels of LDL receptors and LDL uptake are evident in the radioresistant cell lines JSQ3 and SQ20B. (D) Statistical analysis of LDLR staining intensity. (E) Statistical analysis of LDL‐Dy488 uptake intensity. Mean fluorescence intensities are shown; error bars = SEM. Significance is indicated by asterisks (*P < 0.05). Statistical significance of E and D was determined using Mann–Whitney U‐test.

Figure 5

Inhibition of mevalonate pathway by pitavastatin reduces viability, increases DNA damage, and induces senescence upon ionizing radiation in radioresistant HNSCC. (A) ATP‐based cellular proliferation assay of cell lines treated with IR (10 Gy) to demonstrate cell line radiosensitivity. Radioresistant cell lines JSQ3 and SQ20B exhibited lower ATP levels at 48 hr than SCC61 (radiosensitive) cells, which continued proliferating normally (*P < 0.05 vs radiosensitive cells at same time point) Statistical significance was determined using the parametric unpaired multiple t‐test. (B) ATP‐based proliferation assay for cell lines treated with IR (10 Gy) ± PIT (0–10 μm). Radioresistant cell lines demonstrated significantly reduced ATP levels with PIT treatment (*P < 0.05 vs radiosensitive cells at same treatment condition). Statistical significance was determined using the parametric unpaired multiple t‐test. (C) Representative images of ‘Comet’ DNA double‐strand break (DSB) assay data with mean percent of DNA in comet tail ± SEM shown. (D) Plots representing mean percent of DNA in comet tail ± SEM for 100 cells per treatment condition. Significance is indicated by asterisks (*P < 0.05). Statistical significance was determined using Mann–Whitney U‐test. Results indicate that PIT enhances persistent DNA DSBs induced by IR in radioresistant SQ20B cells. (E) Flow cytometric senescence assay data of cells treated with 10 μm of PIT alone (PIT), IR only (10 Gy), or in combination with IR (PIT + IR). Vehicle‐treated cells (VEH) were stained as controls. Cells were analyzed by flow cytometry 6 days after treatment. The combination of PIT + IR dramatically elevated the percent of senescent cells above IR alone in the SQ20B radioresistant cells.

Figure 6

Statin use improves outcomes of radiotherapy in HNSCC patients. Analysis was conducted using data from 518 medical records of stage III‐IVB HNSCC patients, a fraction of whom were incidentally being treated with statins concomitantly with their radiotherapy. Time to local control or distant control was determined from last date of radiotherapy. Survival curves were plotted using the Kaplan–Meier method, and significance was assessed using the Log Rank test. Data indicate that incidental use of statins was statistically associated with better local, but not distant, control in HNSCC patients.