Ⅰ. INTRODUCTION

Concepts on the correlation between chronic inflammation and cancer are not new and have been widely accepted. For example, inflammatory bowel diseases are known to be related to the development of colon cancer, and cholangitis is linked to the pathogenesis of cholangiocarcinoma^1-3). Chronic periodontitis is the most common chronic inflammatory disease of oral cavity, and among oral cancers, more than 90% are oral squamous cell carcinoma⁴⁾. Thus, it is highly probable that these two types of diseases mutually affect each other. In spite that numerous studies provide and support emerging views on the role of chronic inflammation in development of cancers, correlation between chronic periodontitis and oral cancers has not been actively studied. It has been continuously revealed that infectious agents are main causative factors leading to inflammation-associated cancer^1,2). Hepatitis B and C viruses are well known predisposing factors for hepatocellular carcinoma, and infection with a specific subtype of human papilloma virus may initiate cervical cancer^5,6). Though numerous examples of infectious agents are found in association with cancers, most microbial agents which are involved in carcinogenesis are not bacterial but viral ⁸⁾. Not many bacterial agents were recognized as a carcinogen, and the role of bacterial agentinduced inflammation in the process of tumor development has not been defined. Only H. pylori, a causative factor in chronic gastritis, was established as a carcinogen in gastric cancer⁷⁾. In this study, we used oral squamous cell carcinoma cells and gram-negative anaerobic bacteria, Porphyromonas gingivalis, a major pathogen causing chronic periodontitis, for the study on the effect of chronic periodontitis on oral cancer progression.

Experimental studies have reported that P. gingivalis can invade a variety type of host eukaryotic cells, including gingival epithelial cells and endothelial cells^8-10). Thus, P. gingivalis, which are present in or near the area of oral mucosa, can invade and may provoke genetic changes in adjacent oral cancer cells. Considering the involvement of H. pylori in gastric cancer progression, investigations on changes in gene expression of oral cancer cells to invading P. gingivalis may be a meaningful strategy to define a role of P. gingivalis in oral cancer progression as well as a link between chronic periodontitis and oral cancer.

Epithelial-mesenchymal transition (EMT), a phenotypic change which epithelial cells acquire mesenchymal characteristics, is a commonly observed in the cancer progression. Through EMT cancer cells gain not only higher migratory and invasive properties, but also resistance to chemotherapeutic reagents^11-15). Information on P. gingivalis-induced changes in the expression level of genes which lead to EMT would contribute to the development of therapeutic modalities against intractable oral cancer. Accordingly, we observed changes of transcriptional profiles in oral cancer cells using microarrays to monitor EMT-related genes influenced by P. gingivalis infection.

Ⅱ. MATERIALS AND METHODS

Bacteria and culture conditions

The P. gingivalis strain 381 was cultured anaerobically in trypticase soy broth, supplemented with yeast extract (1 mg/ml), hemin (5 μg/ml), and menadione (1 μg/ml) at 37°C overnight.

Oral cancer cell culture

Ca9-22 oral squamous cell carcinoma (OSCC) cells from Japanese Collection of Research (BioSource Cell Bank, Japan) were cultured as a monolayer in minimum essential medium (MEM; Hyclone, Logan, UT) containing 10% fetal bovine serum (FBS; Atlas Biologicals, Fort Collins, CO) and 1% penicillin-streptomycin (GIBCO-BRL, Rockville, MD, USA) at 37°C in a humidified 5% CO₂-95% air incubator.

Bacterial infection

Before infection, P. gingivalis were washed twice and resuspended in phosphate-buffered saline (PBS). The number of bacteria in suspension was determined in a spectrophotometer. Oral cancer cells were then incubated with P. gingivalis at a multiplicity of infection (MOI) of 100 for 2 hours in a CO₂ incubator and then washed with PBS three times to remove non-invaded adherent and suspended bacteria. Those P. gingivalis–infected oral cancer cells were incubated under a normal cell culture condition before harvest.

RNA isolation and analysis of gene expression by the EMT PCR array

To analyze the expression profiles of EMT-related genes, total RNAs from non-infected control and P. gingivalisinfected oral cancer cells were extracted using Trizol (Invitrogen Life Technologies, Carlsbad, CA, USA) and purified by the RNeasy MiniElute Cleanup kit (Qiagen, Valencia, CA, USA). For quality control, RNA purity was evaluated by monitoring OD 260/280 ratio before amplification using Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, USA).

cDNAs were synthesized using the RT² first strand kit and were then added to RT² qPCR Master Mix, which contains SYBR Green and reference dye. Real-time PCR was performed using the ABI 7500 thermal cycler. A RT² Profiler PCR array panel was used to investigate the changes 84 EMT-related key genes. Fold changes in mRNA expression were defined with the ΔΔCt method. Differences between non-infected and P. gingivalis-infected oral cancer cells were analyzed using Array Data Analysis Web Portal from the following address: http://www.sabiosciences.com/pcrarraydataanalysis.

Statistical analysis

Data were obtained from at least 3 separate experiments, and the results were expressed as the mean values ± SD. Genes of a fold regulation greater than 2 or smaller than -2 were considered to display meaningful change. Statistical significance of differential gene expression between two groups was determined using Student t-tests. P values less than 0.05 were considered statistically significant.

Ⅲ. RESULTS

As discussed above, it can be suspected that chronic periodontitis may contribute to aggressiveness of oral cancer cells by modulating biologic behaviors of the cells through changes in EMT-related genes. The present study was done to define the role of chronic periodontitis on the oral cancer progression. To mimic chronic periodontitis, oral cancer cells were infected with P. gingivalis, a major pathogen in chronic periodontitis, at an MOI of 1, 10, and 100. The presence of P. gingivalis within oral cancer cells was first confirmed by observing P. gingivalis 16S ribosomal RNA from the cancer cells in a dose-dependent manner (Fig. 1A). The expression of 16S rRNA, an indicator of P. gingivalis, in oral cancer cells which was infected at an MOI of 100, were decreased in a time-dependent manner (Fig. 1B).

Gene expression differences between P. gingivalis-infected and non-infected OSCC cells were compared using RT² PCR Array which profiles levels of 84 EMT-related genes. Only genes with a fold change greater than twofold and statistical significance were listed. Genes of multiple signaling pathways which drive EMT were upregulated in P. gingivalis-infected oral cancer cells in compared to non-infected cells. For example, P. gingivalis-infected oral cancer cells were found to 5 fold overexpress NOTCH1, one of the Notch family of trans-membrane receptors, relative to non-infected cells. In addition, factors of Wnt/β-catenin signaling pathway as well as BMPs were significantly upregulated. TIMP1, an inhibitor of matrix metalloproteinase, were downregulated (-2.5 fold), suggesting that invasive ability of P. gingivalis-infected oral cancer cells might be promoted by that MMPs were unrestrained (Table 1 & 2).

The array data showed significant changes in many genes involved in tumor invasion and metastasis, supporting that both chronic inflammation and EMT are known to be related to the aggressiveness of cancer cells, such as invasive and metastatic abilities. The upregulated expression of several genes in P. gingivalis-infected OSCC cells was further validated by real-time PCR. The result on such as expression of MMP2 gene strongly suggests that P. gingivalis may increase invasiveness of oral cancer cells by degrading extracellular matrix components within tumor microenvironment. In addition, expressions of important transcription factors and its upstream and/or downstream effectors in the process of EMT, such as wnt and TCF4, were found to be significantly upregulated in P. gingivalis-infected oral cancer cells, implying that P. gingivalis could strongly induce EMT (Fig. 2).

Ⅳ. DISCUSSION

Inflammatory stimuli, which are induced by immune cells as well as the presence of microbial agent in tumor microenvironment, are known to modulate numerous factors that are involved in promoting EMT of cancer cells. The cells which encounter those stimuli exhibit higher durability and resistance to therapeutic agents by acquiring EMT characteristics^11-15). In the present study, we aimed to investigate whether inflammatory insults by P. gingivalis, which is a representative pathogen in chronic periodontitis, could evoke transcriptional changes in oral cancer cells using a microarray analysis. Before considering EMT-related changes, induction of inflammatory signal by P. gingivalis infection was first identified by observing increased expression of SERPINE1. Then we found significant differences in the EMT-related gene expression profiles of P. gingivalis-infected oral cancer cells, including the genes SNAI1 and SNAI3, which are major transcription factors involved in EMT. It is well that these upregulated transcriptional factors suppress the expression of E-cadherin and increase the expression of vimentin or N-cadherin, mesenchymal markers. Increased expression of CDH2(N-cadherin gene), which is a subsequent phenomenon of SNAI1 increase, was also observed, suggesting that P. gingivalis infection leads to formation of aggressive oral cancer cells that can be highly invasive and metastatic.

Wnt/β-catenin signaling is one of the most important pathways that are involved in EMT as well as the aggravation of cancer malignancy. Binding of wnt ligands to cell membrane activates the signaling pathway, and β-catenin is stabilized and translocated into nucleus. Wnt/β-catenin signaling pathway activate expression of numerous genes by forming a complex with nuclear TCF4¹⁶⁾. The increased expression of β-catenin/ TCF4 complex-responsive genes play a role in tumor progression and acquisition of aggressive behavior of cancer cells^17,18). Upregulated WNT ligands in P. gingivalis-infected OSCC cells in the present study suggest that periodontal pathogen may contribute to oral cancer progression by activating Wnt/β-catenin signaling. P. gingivalis infection additionally activate transcriptional activity of β-catenin in oral cancer cells by upregulating TCF4 gene expression, emphasizing the role of chronic periodontitis in oral cancer progression.

Though a direct correlation between P. gingivalis infection and aggressiveness of oral cancer cells cannot be concluded from this study and more reasonable evidences for the association are needed, this study showed alterations of EMT-related genes in P. gingivalis-infected oral cancer cells, suggesting a putative contribution of chronic periodontitis to the aggressiveness of OSCC cells. Further study based on the current findings would reveal the role of chronic periodontitis in oral cancer progression and provide a useful insight into the development of therapeutic modalities against inflammation-related oral cancer