α-actinin-4, originally identified as an actin-binding protein associated with cell motility, invasion, and metastasis of cancer cells, appears to be overexpressed in various human epithelial carcinomas, including colorectal, breast, esophageal, ovarian, and non-small cell lung carcinomas. The authors evaluated whether α-actinin-4 might be appropriate as a molecular target for cancer gene therapy. In 64 primary oral squamous cell carcinomas (OSCCs) and 10 normal oral mucosal specimens, and in seven human OSCC cell lines, α-actinin-4 expression was evaluated immunologically and correlations with clinicopathologic factors were examined. Overexpression of α-actinin-4 was detected in 38 of 64 oral squamous cell carcinomas (70%); significantly more frequently than in normal oral mucosa. The expression of α-actinin-4 was significantly associated with invasion potential defined by the Matrigel invasion assay. Cancer cell lines with higher α-actinin-4 expression had greater invasive potential. An RNAi-mediated decrease in α-actinin-4 expression reduced the invasion potential. These results indicated that the overexpression of α-actinin-4 was associated with an aggressive phenotype of OSCC. The study indicated that α-actinin-4 could be a potential molecular target for gene therapy by RNAi targeting for OSCC.
Oral squamous cell carcinoma (OSCC) is the most common malignant tumor of the head and neck region and accounts for more than 90% of cancers of the oral cavity . The primary therapeutic modality for OSCC is surgery. Recent advances in surgical techniques and anticancer agents have improved tumor regression and survival rate, but wide surgical resection of OSCC causes oral dysfunction so new treatment strategies are required.
Neck lymph node metastasis is strongly related to a poor prognosis in SCC of the head and neck . Alterations in the expression of adhesion-related molecules are associated with poor prognosis in OSCC patients .
α-actinin-4, an actin-binding and crosslinking protein, is thought to play important roles in forming stress fibers, promoting cell adhesion and regulating cell shape and motility . Of the 4 isoforms of α-actinin, the expression of α-actinin-2 and α-actinin-3 is limited to skeletal and cardiac muscle sarcomeres . In contrast, α-actinin-1 and α-actinin-4, which are non-muscular α-actinins, are widely expressed. The latter 2 isoforms share a high degree of similarity (87% amino acid homology) but have different subcellular localizations . α-actinin-1 is localized within the end of actin stress fibers and adherens junctions, and associates them with the cell membranes . α-actinin-4 is present in the cytoplasm and nucleus and colocalizes with actin stress fibers . The expression of α-actinin-4 is highly concentrated at the leading edge of motile cells and in the cytoplasm of sharp cell extensions .
Recent reports have suggested that the increased expression of α-actinin-4 in the cytoplasm of various malignant neoplasms is correlated with poor prognosis, enhanced cell motility, advanced tumor stage, and lymph node metastasis . The relationship between α-actinin-4 expression and invasiveness or metastatic ability in OSCC and the mechanisms underlying any such relationship remain unknown.
In this study, the authors immunohistochemically examined α-actinin-4 expression in OSCC. They determined the clinicopathological significance of α-actinin-4 expression in relation to various parameters such as patient characteristics and histopathological findings. siRNA analysis strongly suggested that α-actinin-4 would be a potent molecular target for cancer gene therapy in OSCC.
Materials and methods
Paraffin-embedded sections were obtained from biopsy specimens of 64 patients with OSCC who underwent radical surgery. Tumor stage was classified according to the TNM classification of the International Union Against Cancer, histological differentiation was defined according to the WHO classification, and invasion pattern was determined according to Bryne’s classification . As controls, samples of normal oral epithelium were obtained after informed consent from 10 patients undergoing routine surgical removal of their third molars.
The following human OSCC cell lines were obtained from the Human Science Research Resource Bank (Osaka, Japan): SAS, SCC25, OSC20, HSC-2, HSC-3, HSC-4, Ca9-22, and the human keratinocyte cell line, HEKa, as a control. All cells were cultured under conditions recommended by their depositors.
Immunohistochemical staining and evaluation
Serial 4-μm thick specimens were taken from the tissue blocks. Sections were deparaffinized in xylene, soaked in target retrieval solution buffer (Dako, Glostrup, Denmark) and placed in an autoclave at 121 °C for 5 min for antigen retrieval. Endogenous peroxidase was blocked by incubation with 0.3% H 2 O 2 in methanol for 30 min. Immunohistochemical staining was performed using the Envision system (Envision+, Dako, Carpinteria, CA). The primary antibody used was directed against α-actinin-4 (4D10, Abnova, Taipei, Taiwan). The sections were incubated with the monoclonal antibody overnight at 4 °C. Reaction products were visualized by immersing the sections in diaminobenzidine (DAB) solution, and the samples were counterstained with Meyer’s hematoxylin and mounted. Negative controls were performed by replacing the primary antibody with phosphate-buffered saline. α-actinin-4 expression was defined as the presence of specific staining in the cytoplasm and nuclei of tumor cells. The immunoreactivity of α-actinin-4 was scored by staining intensity and immunoreactive cell percentage as follows : staining index 0, tissue with no staining; 1, tissue with faint staining or moderate staining in ≤25% of tumor cells; 2, tissue with moderate staining or strong staining in ≤25% of tumor cells; 3, tissue with strong staining in >25% of tumor cells. Overexpression of α-actinin-4 was defined as staining index ≥2.
RNA isolation and semiquantitative reverse transcription-polymerase chain reaction (RT-PCR)
Total RNA was isolated with TRIzol Reagent (Invitrogen, Carlsbad, CA) and first-strand cDNA was synthesized from 1 μg total RNA using Oligo d (T) primer (Invitrogen) and ReverTra Ace (Toyobo, Osaka, Japan). For PCR analysis, cDNA was amplified by Taq DNA polymerase (Takara, Otsu, Japan). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous expression standard. Each PCR program involved a 3-min initial denaturation step at 94 °C, followed by 28 cycles (for α-actinin-4), or 18 cycles (for GAPDH) at 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 1 min, on a PCR Thermal Cycler MP (Takara). Primer sequences were as follows: AGTATGAACGCAGCATCGTG for α-actinin-4(F); GGTGAGGATCTGGTTCTCCA for α-actinin-4(R); ATGTCGTGGAGTCTACTGGC for GAPDH(F); and TGACCTTGCCCACAGCCTTG for GAPDH(R). The amplified products were separated by electrophoresis on ethidium bromide-stained 2% agarose gels. Band intensity was quantified by Image J software.
The BioCoat Matrigel invasion chamber (Becton Dickinson, Bedford, MA) was used. This contains an internal chamber with an 8-μm porous membrane bottom that was coated with Matrigel. Six-well cell culture inserts and a six-well multiwell companion plate were used for the experiment. The membranes were rehydrated with warm serum-free medium for 2 h. The internal chamber was filled with 1.25 × 10 5 cells in medium containing 10% fetal bovine serum (FBS) as a chemoattractant. Cells were incubated for 72 h at 37 °C in a 5% CO 2 atmosphere. After incubation, non-invading cells were removed from the top of the wells with a cotton swab, and cells that transferred to the inverse surface of the membrane were subjected to Diff-Quick staining. Cells were counted under a microscope at 100× magnification. For the control cell count, cells that passed through a control chamber without Matrigel were counted. All experiments were performed in triplicate, and cell numbers at least in 4 fields/well were counted. The ratio of the cell count that passed through the Matrigel chamber to the control cell count was defined as the invasion index, expressed as a percentage.
RNA interference (RNAi)
All siRNAs were purchased from Takara Bio Inc. (Otsu, Japan). Cells were transfected with double-strand RNA using TransIT-siQUEST ® transfection reagent (Mirus, Madison, USA) according to the manufacturer’s protocol. The OSC20 tongue cancer cell line was used for this experiment. Briefly, 1.0 × 10 5 OSC20 cells were plated in each well of six-well plates and allowed to grow for 24 h, till they reached 50% confluence. Cells were then transfected with siRNA at a concentration of 200 nM using the transfection reagent and serum-free medium. Following 24 h of incubation, serum-rich medium was added. The α-actinin-4 siRNA sequences were 5’-GGACAUGUUCAUCGUCCAUTT-3’ and 5’-TTCCUGUACAAGUAGCAGGUA-3’. The scrambled control siRNA sequences were 5’-CGUAUGCGCGUACUCUAAUTT-3’ and 5’-TTGCAUACGCGCAUGAGAUUA-3’. All sequences were submitted to the National Institutes of Health Blast program to ensure gene specificity.
Western blot analysis
Cells were harvested by trypsinization, washed, and precipitated by centrifugation. The Mammalian Cell Extraction Kit (BioVision Research Products, Mountain View, CA) was used for the extraction of proteins. All subsequent manipulations were performed on ice. The cells were incubated in extraction buffer mix. The lysed cells were centrifuged at 15,000 rpm for 3 min and the supernatant was collected as the cytoplasmic fraction. The protein concentration of each sample was measured with micro-BCA protein assay reagent (Pierce Chemical Co. Rockford, USA.). Samples were denatured in sodium dodecylsulfate (SDS) sample buffer and loaded onto 12.5% polyacrylamide gels. After electrophoresis, the proteins were transferred onto a polyvinylidine difluoride membrane and immunoblotted with anti-α-actinin-4 (4D10, Abnova, Taipei, Taiwan) or anti-β-actin (9F2, Abcam, Cambridge, UK; 0.02 μg/ml). Signals were detected using a horseradish peroxidase-conjugated secondary antibody (ECL antimouse IgG, Amersham Biosciences, Piscataway, NJ; 0.01 μg/ml), and visualized using an ECL Kit (Amersham Pharmacia Biotech, Buckinghamshire, UK).
Statistical analysis was performed using StatMate ® (ATMS Co., Tokyo, Japan). The associations between α-actinin-4 expression and clinicopathologic features were assessed by Fisher’s exact test. To determine significant prognostic factors related to survival, multivariate analysis was performed using the Cox proportional hazards regression model. Continuous data are given as mean ± standard deviation. Data sets were examined by one-way analysis of variance (ANOVA) followed by Scheffe’s post-hoc test. The correlation between α-actinin-4 mRNA expression and invasion index was determined using Pearson’s correlation coefficient. P values less than 0.05 were considered significant.
Correlation between α-actinin-4 overexpression and clinicopathologic features
Immunohistochemistry with an anti-α-actinin-4-specific monoclonal antibody was performed on a series of 64 patients with OSCC. Representative immunohistochemical staining is shown in Fig. 1 . Overexpression of α-actinin-4 was undetectable in the normal epithelium ( Fig. 1 A). In the SCC cells, strong α-actinin-4 staining was apparent at the invasive front and the diffuse invasive area. α-actinin-4 overexpression was detected significantly more frequently in OSCC (38 of 64; 70%) than in normal oral epithelium (0 of 10; 0%; P < 0.001, Table 1 ). α-actinin-4 overexpression was significantly more frequent in higher grade cancers (grades 3/4 vs. grades 1/2; P = 0.004, Table 1 ). Cox regression analysis was performed with the parameter of α-actinin-4 overexpression. There was no significant independent predictor of survival (Hazard ratio:1.50, 95%CI:0.769–2.94, P value:0.233).
|Number of samples||α-actinin-4 over expression||p value|
|Squamous cell carcinoma||64||26||38|
|T classification||T1 + T2||41||18||23||0.476|
|T3 + T4||23||8||15|
|N1 + N2||14||7||7|
|Pattern of invasion||Grades 1/2||39||22||17||0.004|