This article is mentioned in:

Introduction

Head and neck squamous cell carcinoma (HNSCC) which includes cancers of the oral cavity, oropharynx, larynx and hypopharynx is the sixth most common cancer worldwide and has an incidence of approximately 600,000 cases per year (1). The current management and treatment of HNSCC involves multi-modality approaches of surgery, chemotherapy and radiotherapy (2). Despite recent advances in early detection, diagnosis and treatment, the 5-year survival for patients with HNSCC has remained at 50% for the past 30 years (3). Molecular-targeted therapy, based on molecular findings of the last 50 years, is one of the most promising gateways to the development of new strategies in oncology (4). Cetuximab, a monoclonal antibody to the epidermal growth factor receptor (EGFR), is the only molecular-targeted therapy to be routinely used in clinical practice for the treatment of recurrent and metastatic HNSCC (5). Based on a limited number of phase II and III trials that have investigated the efficacy of cetuximab in addition to cisplatin in patients who were refractory to platinum-based therapy, the combination appears to confer further benefit over anti-EGFR agents alone (6,7).

The first generation of immunotoxins developed 35 years ago, which heralded targeted therapy, employed chemical conjugations of antibodies and either intact toxins or toxins with attenuated cell-binding properties. We previously reported that recombinant fusion protein IL4 (38–37)-PE38KDEL (also termed IL4-PE), consisting of circularly permuted interleukin (IL)-4 and a mutated form of Pseudomonas exotoxin (PE), induced significant regression of established biliary tract tumors and significantly improved the survival of animals with disseminated tumors (8,9). In addition, IL4-PE was reported to be highly and specifically cytotoxic to glioma cell lines in vitro, and caused partial or complete regression of established human glioblastoma multiforme tumors in nude mice (10). IL-4 receptor α (IL-4Rα)-targeted protein-based immunotoxin was tested in the clinic for the treatment of human solid tumors (11,12). However, its clinical application faced many challenges, including non-specific toxicities and immunogenicity (13).

To overcome these issues, we previously developed a ‘hybrid peptide’, composed of target-binding and cytotoxic sequences containing cationic-rich D- and L-amino acids to form amphipathic partial α-helices that disrupt the cancer cell membrane selectively, and are stable when combined with a cancer-targeting moiety (14). It is known that peptide drugs are relatively easily synthesized using either recombinant or solid-phase chemical synthesis techniques and the production costs are generally affordable when compared to antibody-based therapeutics (14). IL-4Rα has been previously reported to have high expression on the surface of a variety of human solid tumors such as renal cell carcinoma, malignant melanoma and glioblastoma (15). Although the biological function of IL-4Rα expression on solid tumors remains unclear, this receptor may be an effective candidate for a novel molecular-targeted therapy.

In the present study, we examined the expression levels of IL-4Rα in both patient samples and HNSCC cell lines, and then explored the antitumor activity of the IL-4Rα-lytic hybrid peptide against HNSCC.

Materials and methods

Patient samples

HNSCC specimens were obtained from 5 patients (4 males, 1 female; mean age, 56.6 years), who underwent radical surgery at the Department of Oral and Maxillofacial Surgery, Tsukuba University Hospital, Japan from 2010 to 2011. Primary tumor sites were the tongue and the gingiva. For immunoblot analysis, we obtained HNSCC tissue from the cancerous lesion (cancer) and normal tissue from the normal area (normal) in other specimens from 1 patient. For immunohistochemistry, tissue was fixed in 10% formalin and paraffin embedded. The study protocol was approved in accordance with the ethics guidelines of the Tsukuba University (H23-61). All patients provided written informed consent for use of specimens.

Peptides

The following IL-4Rα-lytic hybrid peptide, the lytic peptide and IL-4 binding peptide were purchased from Invitrogen: the IL-4Rα-lytic hybrid peptide, KQLIRFLK RLDRNGGGKLLLKLLKKLLKLLKKK (underlined letters are D-amino acids); the lytic peptide, KLLLKLLKKLLKLL KKK; and the IL-4 binding peptide, KQLIRFLKRLDRN. All peptides were synthesized by the use of solid-phase chemistry, purified to homogeneity by reverse-phase high-pressure liquid chromatography and assessed by mass spectrometry. All peptides were dissolved in water and buffered to pH 7.4.

Cell lines and culture conditions

The HNSCC cell lines (HSC-2, HSC-3, HSC-4, Ca9-22 and OSC-19) were purchased from the Japanese Collection of Research Bioresources (Osaka, Japan). The human normal keratinocyte cell line (HaCaT) was purchased from the American Type Culture Collection (Manassas, VA, USA). These cell lines were maintained in Dulbecco’s modified Eagle’s medium or RPMI-1640 containing 10% heat-inactivated fetal calf serum (Nichirei Biosciences Inc., Tokyo, Japan) and 1% penicillin-streptomycin in a humidified atmosphere with 5% CO2 at 37°C.

Immunoblot analysis

Immunoblot analysis was carried out as previously described (9). Briefly, whole-cell extracts were obtained using buffer containing 1% (v/v) Triton X-100, 0.1% (w/v) SDS, and 0.5% (w/v) sodium deoxycholate, separated by SDS-PAGE, and transferred onto a PVDF membrane. IL-4Rα antibody was used at dilution 1:200 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and actin as the internal control (Sigma-Aldrich, St. Louis, MO, USA). Proteins were visualized on Hyperfirm using an enhanced chemiluminescence/western blotting system (GE Healthcare, Piscataway, NJ, USA).

Quantitative real-time PCR

Total RNA of cells was isolated using High Pure RNA Tissue kit (Roche, Basel, Switzerland). For the reverse transcriptase reaction, 400 ng of the RNA sample was used. The reaction was carried out in a final volume of 10 μl of reaction mixture with Takara Kit II (Takara, Shiga, Japan). Aliquots (2 μl) of the cDNA samples were amplified in a final volume of 20 μl of PCR mixture containing SYBR Premix Ex Taq II (Takara). Quantitative real-time PCR was carried out using PRISM 7000 (Applied Biosystems, Carlsbad, CA, USA). The following primers were used: IL-4Rα forward, 5′-CTGACCTGGAGCAACCCGTATC-3′ and IL-4Rα reverse, 5′-GCAGACGGACAACACGATACAG-3′; GAPDH forward, 5′-GTCTTCACCACCATGGAGAAGGCT-3′ and GAPDH reverse, 5′-CATGCCAGTGAGCTTCCCGTTCA-3′.

Cell viability assay

Cell viability assay was performed as previously described (14). Briefly, cells were seeded into 96-well plates at 3×103 cells/well in 90 μl of medium and incubated at 37°C for 24 h. Each peptide (IL-4R-lytic hybrid peptide, lytic peptide or IL-4 binding peptide) diluted in 10 μl culture medium was added to the cells. After a 72-h incubation, the cell viability assay using WST-8 solution (Nacalai Tesque, Kyoto, Japan) was performed.

Immunohistochemistry

For immunostaining of IL-4Rα, 2-μm sections from patient samples were stained using the Vectastain kit according to the manufacturer’s instructions with the anti-IL-4Rα antibody (R&D Systems, Minneapolis, MN, USA).

Binding assay

The IL-4 binding peptide labeled with fluorescein isothiocyanate (FITC) was incubated with HSC-2 and HaCaT cells. Quantification of the binding activity of this peptide to HSC-2 and HaCaT cells treated with various concentrations for 30 min was carried out. Cells were washed twice with phosphate-buffered saline, and the peptides were detected using a flow cytometer (FACSCalibur, BD Biosciences, San Jose, CA, USA).

Antitumor activity of the IL-4Rα-lytic hybrid peptide in a human tumor xenograft mouse model in vivo

Animal experiments were carried out in accordance with the guidelines of Tsukuba University. HSC-2 cells (5×106) resuspended in 150 μl of phosphate-buffered saline were inoculated subcutaneously into the flank region of 4- to 6-week-old athymic female nude mice weighing 17–20 g. When the tumors reached 20–60 mm3 in volume, animals were assigned randomly to four groups. Saline (control), the IL-4Rα-lytic hybrid peptide (5 or 10 mg/kg), or the lytic peptide alone (5 mg/kg) was injected intratumorally (50 μl/injection) three times a week for a total of 9 times. Tumors were measured with a caliper, and the tumor volume (in mm3) was calculated using the following formula: Length × width2 × 0.5. All values are expressed as means ± SD.

Results

Expression of IL-4Rα in HNSCC tissue specimens

We first analyzed the expression levels of IL-4Rα in HNSCC tissue specimens. Patient characteristics including age, gender, primary site of tumor, TNM classification and differentiation are shown in Table I. Immunoblot analysis and immunohistochemistry were performed to investigate the expression levels of IL-4Rα in tissue specimens from HNSCC patients. Immunoblot analysis showed that IL-4Rα was expressed in all HNSCC specimens (tongue and gingival carcinomas) but not in the normal tissue specimens from the same patients (Fig. 1A). Similarly, immunohistochemical analysis using anti-IL-4Rα antibody showed IL-4Rα immunopositivity in the HNSCC cancerous epithelium but not in the normal epithelium in all patients. Fig. 1B-E shows the staining pattern for patient no. 2.

Figure 1 IL-4Rα expression in HNSCC tissue specimens. (A) IL-4Rα protein levels in HNSCC tissues. Tissue sections were prepared from 5 patients who underwent radical resection of oral cancer. Cancerous (cancer) and normal (normal) tissue regions were obtained as described in Materials and methods. (B-E) Immunohistochemical analysis of HNSCC samples of patient no. 2. (B) H&E staining. (C-E) Anti-IL-4Rα antibody reactivity in normal area (D) and cancer area (E). Scale bars indicate 300 μm (B and C) and 50 μm (D and E).

Table I Clinicopathological features of the HNSCC patients.

Table I

Clinicopathological features of the HNSCC patients.

No.	Age (years)	Gender	Primary site	TNM classification	Differentiation of SCC
1	57	Male	Tongue	T1N0M0	Well
2	40	Male	Tongue	T2N0M0	Moderate
3	47	Female	Tongue	T2N0M0	Well
4	78	Male	Gingiva	T3N2bM0	Well
5	61	Male	Gingiva	T4aN2bM0	Moderate

[i] SCC, squamous cell carcinoma.

Expression of IL-4Rα in cultured HNSCC cell lines

We next investigated the expression levels of IL-4Rα in HNSCC cell lines by immunoblot and real-time PCR analyses. Immunoblot and real-time PCR analyses demonstrated that all HNSCC cell lines expressed IL-4Rα but HaCaT did not (Fig. 2A). We also examined mRNA expression levels of IL-4Rα by real-time PCR analysis (Fig. 2B). Relative expression levels of IL-4Rα in HSC-2 and HSC-3 cells were ~4-fold higher than that of HaCaT cells. The lowest expression level of IL-4Rα in HNSCC cell lines was found in OSC-19, however, this level was still 2-fold higher than that of HaCaT (Fig. 2B).

Figure 2 IL-4Rα expression in cultured HNSCC cell lines. (A) Protein levels of IL-4Rα in HNSCC cell lines and HaCaT normal keratinocytes. Specimens were prepared from five HNSCC cell lines: HSC-2, HSC-3, HSC-4, OSC-19 (tongue carcinoma), Ca9-22 (gingival carcinoma) and HaCaT cells. Actin was used as a loading control. (B) Expression levels of IL-4Rα mRNA in HNSCC and HaCaT cell lines. Real-time PCR was performed as described in Materials and methods using IL-4Rα-specific primers, and GAPDH served as an internal control.

Cytotoxic activity of the IL-4Rα-lytic hybrid peptide in HNSCC cell lines

To assess the in vitro cytotoxic activity of the IL-4Rα-lytic hybrid peptide in HNSCC and HaCaT cells, the WST assay was performed using HNSCC cell lines treated with the IL-4Rα-lytic hybrid peptide, lytic peptide or IL-4 binding peptide. HSC-2, HSC-3, HSC-4 and Ca9-22 cells were sensitive to the IL-4Rα-lytic hybrid peptide; the concentration that killed 50% of all cells (IC50) was <5 μM. The OSC-19 cell line was also sensitive to the IL-4Rα-lytic hybrid peptide with an IC50 of <10 μM. In contrast, optimal cell killing was not induced in HaCaT cells by either the lytic peptide or IL-4 binding peptide or IL-4Rα-lytic hybrid peptide (Fig. 3). The cytotoxic activity of the hybrid peptide was strongly enhanced when compared with that of the lytic peptide. The cytotoxic activity of the IL-4Rα-lytic hybrid peptide increased 4.0- to 13.2-fold when compared with that of the lytic peptide in the cancer cell lines (Table II). We examined the binding activity of the IL-4 binding peptide labeled with FITC to both HSC-2 and HaCaT cells by flow cytometry, and then found that exposure of HSC-2 to this peptide resulted in the increased binding activity of this peptide in a concentration-dependent manner (Fig. 4). These results suggest that the IL-4Rα-lytic hybrid peptide selectively kills cancer cells expressing IL-4Rα.

Figure 3 Cytotoxic activity of the IL-4Rα-lytic hybrid peptide in HNSCC and HaCaT cell lines in vitro. The five HNSCC cell lines (HSC-2, HSC-3, HSC-4, Ca9-22 and OSC-19) and HaCaT cells (3×103 cells) were cultured with the indicated concentrations of peptides for 72 h, and cytotoxic activity was assessed using WST-8 reagent (●, IL-4Rα-lytic hybrid peptide; ○, lytic peptide; ▲, IL-4 binding peptide). Data represent means ± SD (bars) of triplicate determinations.

Figure 4 Binding assay of HSC-2 and HaCaT cells treated with the FITC-labeled IL-4 binding peptide. Quantification of the binding activity of this peptide to HSC-2 and HaCaT cells treated with various concentrations for 30 min. IL-4 binding peptide at 0 μM represents untreated cells used as a control. Fold-change in fluorescence intensity is the extent of binding of the FITC-labeled IL-4 binding peptide to each cell line, relative to the fluorescence intensity values. Data are presented as means ± SD (error bars) from triplicate determinations.

Table II Cytotoxic activity of each peptide in HNSCC and HaCaT cell lines.

Table II

Cytotoxic activity of each peptide in HNSCC and HaCaT cell lines.

	IC50 (μM)		IC50 ratio
Cell lines	IL-4Rα-lytic hybrid peptide	Lytic peptide	Lytic/IL-4Rα-lytic hybrid peptide
HNSCC cells
HSC-2	1.9±0.1	8.2±0.2	4.4
HSC-3	2.3±0.6	17.4±1.2	7.4
HSC-4	2.8±0.8	33.2±6.1	11.9
Ca9-22	2.2±0.7	29.6±5.8	13.2
OSC-19	8.2±5.4	33.3±13.1	4.0
Normal cells
HaCaT	37.6±15.3	43.4±13.6	1.2

[i] IC₅₀ values (peptide concentration inducing 50% inhibition of control cell growth) are represented as means ± SD from triplicate determinations.

In vivo antitumor activity of the IL-4Rα-lytic hybrid peptide in a human HNSCC xenograft mouse model

Following the observation that the IL-4Rα-lytic hybrid peptide exhibits a marked cytotoxic effect on HNSCC cells in vitro (Fig. 3), the antitumor activity of the hybrid peptide was assessed in a xenograft model of human HNSCC. HSC-2 cells were inoculated subcutaneously into athymic nude mice, and the animals were subsequently treated with the IL-4Rα-lytic hybrid peptide by intratumoral injection. As shown in Fig. 5, tumors grew aggressively in the control mice injected with saline alone, reaching a volume of >1000 mm3 by day 24. In contrast, mice treated with the IL-4Rα-lytic hybrid peptide showed significant tumor regression at both dosages: mean tumor volumes were 491 mm3 (5 mg/kg) and 283 mm3 (10 mg/kg) on day 24. Moreover, tumors in mice injected with the lytic peptide grew rapidly similar to tumor growth in the control mice with saline alone, reaching a volume of 1104 mm3 (Fig. 5). No other abnormalities, such as loss of appetite and body weight, were observed in mice injected with the IL-4Rα-lytic hybrid peptide (data not shown). Histological analysis also showed no side effects in tissues from the major organs, including the liver and kidney, which were obtained from mice treated with intratumoral administration of the IL-4Rα-lytic hybrid peptide (data not shown). These results demonstrated that the IL-4Rα-lytic hybrid peptide exhibited effective antitumor activity in a mouse xenograft model of HNSCC.

Figure 5 Antitumor activity of the IL-4Rα-lytic hybrid peptide in a xenograft model. (A) HSC-2 cells were inoculated subcutaneously into nude Balb/c mice. Animals received an intratumoral peptide injection or saline alone [□, control; lytic peptide (○, 5 mg/kg); IL-4Rα-lytic hybrid peptide (●, 5 mg/kg or ▲, 10 mg/kg)] three times per week from day 3 as indicated by the arrows. Data are expressed as mean ± SD (bars), n=6 animals per group. *P<0.01. (B-E) Images show tumor growth on day 24 after tumor inoculation: (B) saline alone; (C) lytic peptide; (D) IL-4Rα-lytic hybrid peptide (5 mg/kg); (E) IL-4Rα-lytic hybrid peptide (10 mg/kg).

Discussion

The main treatment options for patients with HNSCC currently involve surgery, radiotherapy and chemotherapy, alone or in combination. Despite significant advances in HNSCC treatment, survival rates and prognosis have improved only moderately over the years (16). Systemic chemotherapy remains the only effective treatment option, but it is associated with significant rates of toxicity in HNSCC patients, who usually have a high prevalence of co-morbidities and problematic lifestyle habits (17). Ideally, future therapies should act over the short term, to minimize damage to healthy cells and target tumor compartments that have the highest sensitivity.

The concept of a ‘magic bullet’ proposed by Paul Ehlrich over 100 years ago has led to the search for agents that can selectively target cancer cells (18). Immunotoxins are proteins used to treat cancer and are composed of an antibody fragment linked to a toxin (19). Several disadvantages of these conventional immunotoxins for clinical use include immunogenicity, undesirable toxicity, manufacturing difficulties, short half-lives and neutralizing antibody production (20,21). However, peptides can be produced affordably by chemical synthesis, with a cost comparable to that of producing protein drugs. Moreover, since peptides are easy to produce, a wide variety of candidate peptides combining moieties for targeting and for toxicity can be tested in preclinical settings. We previously linked two functional peptide domains to produce a novel chimerical peptide termed a ‘hybrid peptide’, which was designed as a bifunctional peptide that binds to receptors or proteins overexpressing in cancer cells and consequently disrupts the cancer membrane (14,22,23). In the present study, we focused on IL-4Rα as recent evidence suggests that IL-4Rα is preferentially expressed on the surface of a variety of solid tumors including HNSCC (24).

The high degree of antitumor activity of the IL-4Rα-lytic hybrid peptide in HNSCC correlated with the expression of IL-4Rα in vitro. All HNSCC tumor specimens showed specific immunohistochemical staining for IL-4Rα, and western blot analysis revealed expression of IL-4Rα. However, IL-4Rα expression was not observed in normal tissue specimens from the same patients (Fig. 1). These data are consistent with previous reports that HNSCC cells express IL-4Rα on their cell surface and confirm that IL-4Rα is expressed in situ(25,26). These results also indicate that this receptor may be an attractive target for the treatment of HNSCC.

Previous results suggest that IL-4 receptor-targeted cytotoxin may provide an effective therapeutic option for HNSCC (24,26). In the present study, the in vitro cytotoxicity of the IL-4Rα-lytic hybrid peptide was examined in five HNSCC cell lines (Fig. 3). HSC-2 cells, which showed the highest level of IL-4Rα expression in western blot analysis, also showed the highest sensitivity to the IL-4Rα-lytic hybrid peptide (Figs. 2 and 3). Normal HaCaT cells with low IL-4Rα expression were not sensitive to the IL-4Rα-lytic hybrid peptide (Fig. 3). These results suggest that the cytotoxic effect of the IL-4Rα-lytic hybrid peptide correlates well with the level of IL-4Rα expression.

In the present study, although the growth rate of HSC-2 was rapid, it was found that intratumoral administration of the hybrid peptide at 10 mg/kg, dramatically inhibited the growth of HSC-2 tumors in vivo (Fig. 5). Histological analysis also showed no abnormal changes in the tissues of major organs obtained from the mice injected with the hybrid peptide (data not shown). For clinical use, local injection may be effective for tumors such as HNSCC, and combination with prior chemoradiotherapy should be developed. These observations indicate that abundant IL-4Rα expression in HNSCC tumors would facilitate efficient targeting by the IL-4Rα-lytic hybrid peptide.

In conclusion, IL-4Rα was overexpressed in both tumor specimens from patients with HNSCC and in HNSCC cell lines in vitro. The overexpressed IL-4Rα on HNSCC cells could be successfully targeted with the IL-4Rα-lytic hybrid peptide in vitro and in vivo. Future investigations using cancer progenitor cells isolated from primary malignant tissues at different stages during cancer progression and metastatic disease may help to identify new biomarkers for the development of more effective diagnostic and prognostic methods and targeted therapies (27). Additional studies should be performed to reveal the antitumor activity of the IL-4Rα-hybrid peptide in animal models, and perhaps a phase I clinical trial should be undertaken to study its antitumor activity.

Acknowledgements

We thank Megumi Kawamoto, Kumi Kodama, Nana Ohkubo, Aya Torisawa (Department of Pharmacoepidemiology, Kyoto University), and Airi Ueda (Department of Clinical Sciences, and Molecular Cellular Physiology, Tsukuba University) for technical assistance or advice with the tissue culturing and in vivo experiments. This study was supported by Grants-in-Aid for Scientific Research (B) (grant no. 24390449), Young Scientists (A) (grant no. 2368009), and Challenging Exploratory Research (grant no. 23659934) from the Japan Society for the Promotion of Science (JSPS).

References