Silencing of high-mobility group box 2 (HMGB2) modulates cisplatin and 5-fluorouracil sensitivity in head and neck squamous cell carcinoma

Abstract

Dysregulation of protein expression is associated with most diseases including cancer. MS-based proteomic analysis is widely employed as a tool to study protein dysregulation in cancers. Proteins that are differentially expressed in head and neck squamous cell carcinoma (HNSCC) cell lines compared to the normal oral cell line could serve as biomarkers for patient stratification. To understand the proteomic complexity in HNSCC, we carried out iTRAQ-based MS analysis on a panel of HNSCC cell lines in addition to a normal oral keratinocyte cell line. LC-MS/MS analysis of total proteome of the HNSCC cell lines led to the identification of 3263 proteins, of which 185 proteins were overexpressed and 190 proteins were downregulated more than twofold in at least two of the three HNSCC cell lines studied. Among the overexpressed proteins, 23 proteins were related to DNA replication and repair. These included high-mobility group box 2 (HMGB2) protein, which was overexpressed in all three HNSCC lines studied. Overexpression of HMGB2 has been reported in various cancers, yet its role in HNSCC remains unclear. Immunohistochemical labeling of HMGB2 in a panel of HNSCC tumors using tissue microarrays revealed overexpression in 77% (54 of 70) of tumors. The HMGB proteins are known to bind to DNA structure resulting from cisplatin-DNA adducts and affect the chemosensitivity of cells. We observed that siRNA-mediated silencing of HMGB2 increased the sensitivity of the HNSCC cell lines to cisplatin and 5-FU. We hypothesize that targeting HMGB2 could enhance the efficacy of existing chemotherapeutic regimens for treatment of HNSCC. All MS data have been deposited in the ProteomeXchange with identifier PXD000737 (http://proteomecentral.proteomexchange.org/dataset/PXD000737).

Keywords: Antimetabolite; Biomarker; Biomedicine; Drug resistance; In vitro labeling.

Figure 1. Workflow employed for iTRAQ-based quantitative proteomics of HNSCC cell lines

200 μg of protein from HNSCC cell lines was digested and iTRAQ labeled using 8plex iTRAQ reagents. SCX fractionation followed by LC-MS/MS analysis was carried out using LTQ-Orbitrap Velos mass spectrometer interfaced with Proxeon Easy nLC II system. The data was searched against Human RefSeq 50 using Proteome Discoverer software suite. Candidate proteins were validated using immunohistochemistry.

Figure 2. Representative MS/MS spectra of peptides from proteins overexpressed in HNSCC cell lines

(A) Ezrin (B) Insulin-like Growth Factor 2 mRNA-binding Protein 3 (C) SET nuclear oncogene (D) High Mobility Group Box 2.

Figure 3. Immunohistochemical validation of HMGB2 overexpression in HNSCC tissues

Representative sections from HNSCC tissue microarrays stained with anti-HMGB2 antibody (A) Expression of high mobility group box 2 in HNSCC primary tissue (B) Expression of high mobility group box 2 in representative non-tumor oral squamous mucosa.

Figure 4. Cell viability of HNSCC lines in response to cisplatin and 5-FU

(A) HNSCC cell lines were transfected with scrambled siRNA or HMGB2 siRNA (10nM and 20nM) and Western blot analysis was performed using anti-HMGB2 antibody. β-actin was used as a loading control. JHU-O11, JHU-O22, JHU-O28 JHU-O29 and FaDu were treated with indicated concentrations of cisplatin (B) and 5-FU (C) and MTT assays were performed.

Figure 5. Silencing of HMGB2 enhances sensitivity to cisplatin and 5-FU in HNSCC cells

HNSCC cells (A) JHU-O22, (B) JHU-O28, (C) JHU-O29 and (D) FaDu were transfected with HMGB2-siRNA or scrambled siRNA and treated with cisplatin and 5-FU for 72 h and assayed for cellular survival using MTT assay. * denotes p-value <0.05 and ** denotes p – value <0.005