Suppression of migration, invasion, and metastasis of cisplatin-resistant head and neck squamous cell carcinoma through IKKβ inhibition

Abstract

We explored the role of the transcription factor, NF-κB, and its upstream kinase IKKβ in regulation of migration, invasion, and metastasis of cisplatin-resistant head and neck squamous cell carcinoma (HNSCC). We showed that cisplatin-resistant HNSCC cells have a stronger ability to migrate and invade, as well as display higher IKKβ/NF-κB activity compared to their parental partners. Importantly, we found that knockdown of IKKβ, but not NF-κB, dramatically impaired cell migration and invasion in these cells. Consistent with this, the IKKβ inhibitor, CmpdA, also inhibited cell migration and invasion. Previous studies have already shown that N-Cadherin, an epithelial-mesenchymal transition (EMT) marker, and IL-6, a pro-inflammatory cytokine, play important roles in regulation of HNSCC migration, invasion, and metastasis. We found that cisplatin-resistant HNSCC expressed higher levels of N-Cadherin and IL-6, which were significantly inhibited by CmpdA. More importantly, we showed that CmpdA treatment dramatically abated cisplatin-resistant HNSCC cell metastasis to lungs in a mouse model. Our data demonstrated the crucial role of IKKβ in control of migration, invasion, and metastasis, and implicated that targeting IKKβ may be a potential therapy for cisplatin-resistant metastatic HNSCC.

Keywords: Cisplatin resistance; HNSCC; Head and neck squamous cell carcinoma; IKKβ; IKKβ inhibitor; Invasion; Metastasis; Migration; NF-κB.

Figure 1.. Cisplatin-resistant HNSCC cells have elevated IKKβ/NF-κB signaling and stronger abilities to migrate and invade.

A: Cal27 and Cal27CP cells were lysed and the phosphorylation status and total protein levels of IKKα/β and p65, as well as β-actin, were analyzed by Western blot. B and C: Cal27 and Cal27CP cell migration (B) and invasion (C) were monitored by xCELLigence system. SFM: serum-free medium.

Figure 2.. IKKβ knockdown impairs cisplatin-resistant HNSCC cell migration and invasion.

A: Cal27CP cells were transfected with non-target siRNA, siRNA IKKβ, or siRNA p65 for 48 hours and the expression of IKKβ, p65 and β-Actin was tested by Western blot. B and C: Cal27CP cells were transfected with non-target siRNA, siRNA IKKβ, or siRNA p65 for 48 hours and cell migration (B) and invasion (C) were monitored by xCELLigence system. SFM: serum-free medium. D: Cal27CP cells transfected with non-target siRNA, siRNA IKKβ, or siRNA p65 for 24 hours were seeded in 12 well plates (800/well) and colony formation was observed after 10 days. E: IKKβ regulates cell migration and invasion independent of NF-κB, whereas it may regulate cell proliferation through NF-κB migration in Cal27CP cells.

Figure 3.. IKKβ inhibitor, CmpdA, inhibits cisplatin-resistant HNSCC cell migration and invasion.

A and B: Cal27CP cells were treated with IKKβ inhibitor, CmpdA (2 μM) and cell migration (A) and invasion (B) were monitored. C: UMSCC74B cells were treated with increasing doses of CmpdA and cell invasion was monitored.

Figure 4.. CmpdA inhibits N-Cadherin expression in cisplatin-resistant HNSCC cells.

A: Cal27 and Cal27CP cells were lysed and the expression of E-Cadherin, N-Cadherin, and β-actin was detected by Western blot. B: Cal27CP and UMSCC74B cells were treated with increasing doses of CmpdA for 24 hours and the levels of phosphor-p65, p65, E-Cadherin, N-Cadherin, and β-actin were detected by Western blot.

Figure 5.. CmpdA inhibits IL-6 expression in cisplatin-resistant HNSCC cells.

A: Cal27 and Cal27CP cells were lysed and the levels of IL-6, phospho-STAT3, Stat3, and GAPDH were detected by Western blot. B: Cal27CP and UMSCC74B cells were treated with increasing concentrations of CmpdA for 24 hours and the levels of IL-6, phospho-STAT3, Stat3, and GAPDH were detected by Western blot.

Figure 6.. CmpdA abates Cal27CP cell metastasis to lungs in mice.

A: 1×10⁶ YFP/luc-Cal27CP cells pre-treated for 72 hours with vehicle control or 2.5 μM CmpdA were injected intravenously into NRG mice. On day of injection, mice were imaged and grouped into the control group (Group 1; n=5) and the treatment group (Group 2; n=6) so that mean intensity of luminescence signaling in lungs was similar. Mice were then treated intraperitoneally (IP) three times per week with 10% DMSO (Group 1) or 10 mg/kg CmpdA (Group 2), for three weeks. Luminescence signaling (photon intensity) in lungs was monitored. B: Intensity of luminescence signaling in lungs at days 1 (start) and 22 (end) were compared (*P˂0.05). C: 11 mice labeled from 1 to 11 (control group: 1–5; and treatment group: 6–11) were euthanized at day 22, lungs were washed and collected, and fluorescence images (red) were taken. D: Fluorescence images on lungs were measured (**P˂0.01). E: Percent body weight changes in mice treated with vehicle control or CmpdA were recorded from days 1 to 22. F: Lung lesions in mice treated with vehicle control or CmpdA (mice 1–11 in Figure D) were counted and compared. G: Hematoxylin and eosin (H&E) staining of paraffin sections of lungs from control and CmpdA treated mice are shown at the same magnification. Yellow arrows indicate metastatic lung tumors.