Expression of TIMP-2 in HPV-16 infected oral squamous cell carcinoma in patients in Iraq

Aim: To determine the expression of tissue inhibitors of metalloproteinases (TIMP-2) in oral squamous cell carcinoma (OSCC) and the difference in its expression level between positive and negative HPV-16 (human papilloma virus- 16) OSCC patients. Methods: This study was conducted on 33 biopsies obtained from patients with OSCC and 10 normal oral mucosa as controls. In situ hybridization (ISH) was used to investigate the presence of HPV-16, while immunohistochemistry (IHC) was used to estimate the expression level of TIMP-2. Results: The TIMP-2 was expressed in 27 (81.8%) of OSCC sections with no significant difference between its expression level in HPV-16 positive and HPV-16 negative OSCC cases (p=0.058). TIMP-2 was found to be highly expressed in OSCC sections, and the presence of HPV was not related to its overexpression. Conclusions: The percentage of samples that appeared to accommodate detectable HPV-16 was high, but no significant difference was observed in relation to TIMP-2 expression level. Future studies with a larger number of patients are highly recommended to address the possible association between TIMp-2 and OSCC positive HPV-16.


Introduction
The major neoplasm worldwide is oral cancer. It is accompanied with high death rate, which is associated with a routine late discovery in its development. The recurrence is a serious prognostic influence in patients with oral squamous cell carcinoma (OSCC) [1][2][3] . According to data from the American Oral Cancer Foundation published in 2014, over 640,000 new cases are found every year 4 , whereas there were 128,000 deaths due to oral cancer worldwide in 2008 according to the International Agency for Research on Cancer (IARC) 5,6 .
In Iraq, cancer made up 10% of total deaths of all ages. Oral cancer constituted approximately 4.5% of all cancer cases and OSCC represents about 91.5% of all oral cancer and 37% of head and neck cancer according to Iraqi cancer registry 7 . The prevalence of oral cancer during [2003][2004][2005][2006] in Iraqi governorates in relation to sex, age and site has been estimated, and the study found that the highest prevalence (55.81%) was observed in tongue, while the lowest (3.87%) was found in the floor of the mouth (Al-Kawaz, 2010) 8 . Moreover, Baghdad governorate was found to have the highest prevalence of OSCC (41.08%) 8 . Mohammed et al. (2015) 9 , detected human herpesvirus-1 antigen in OSCC and apparently healthy control in Iraq. The detection of HHV-1 antigen, chemicals and radiation, found to play important role in the development of oral cancers in Iraq.
Epidemiological and molecular studies have determined that the persistence of infection with high-risk human papillomaviruses (HR-HPVs) types is the most common risk factor for the development of head and neck malignancies [10][11][12] . HPVs are small non-enveloped, double-stranded, circular DNA viruses that encode two viral oncoproteins, E6 and E7. The best-characteristic properties of E6 and E7 proteins from HR-HPVs are their ability to mediate the degradation of p53 and pRb, respectively. These viral proteins also interact with other several cellular factors and lead to revoke normal cell cycle checkpoint stand cell death mechanisms 13 . A possible malignant disorder is the proliferative verrucous leukoplakia (PVL), which is often undergone malignant transformation to OSCC 14,15 . Akrish et al., (2014) 16 reported and compared the histologic, immunohistochemical and clinical features as well as the survival rates of carcinoma arising in patients with PVL (p-scca) and conventional squamous cell carcinoma (c-scca). The authors found that P-scca revealed considerably better prognostic factors and limited survival rates and longer duration of disease in compare to c-scca. They also suggested that p-scca may show a distinct entity, which may have clinical implications on treatment 16 .
Development of most solid tumors is characterized by an increase in secretion and activation of matrix metalloproteinases (MMPs) produced by either the tumor cells or tumor-associated fibroblasts. These proteins play a master role in diverse physiological processes and diseases such as the homeostatic tissue remodeling and cancer 17 . Proteolytic activity of MMPs can be regulated at different levels including gene expression, protein compartmentalization zymogen to active enzyme conversion and the presence of specific inhibitors 18 .
Tissue or extracellular MMPs are regulated by endogenous inhibitors named tissue inhibitor metalloproteinase family 1 to 4 (TIMP-1-4). Among them, TIMP-2 is unique because it may function both as an MMP activator and inhibitor 17 . Although TIMPs are known to act as inhibitors of MMPs, they exhibit other biological functions. Indeed, TIMPs play complex divergent roles in metastasis, functioning both as anti-invasive agents and as prognostic indicators. Another study suggested that the overexpression of TIMPs is correlated with poor outcome in colorectal and breast carcinomas 18 .
In OSCC, MMP-9 and TIMP-2 expressions have shown a predictive value for tumor metastases and cause-specific survival. As in other tissue carcinomas, high expression of TIMP-2 is the most independent factor for worse prognosis in early-stage oral SCC 10,12,17 . However, expression of TIMP-2 (MMP) in HPV-16 infected OSCC patients is still incomplete and needs further investigations 14 , whereas, the reliable markers for progression of high-risk HPVinfected epithelium to malignancy are not yet available 19 . Chandolia et al., (2016) 20 also approved that OSCC shows higher MMP-9 expression as compared to oral epithelial dysplasia followed by epithelium from normal oral mucosa. However, no correlation was found between the histological grades of OSCC.
A previous immunohistochemical study reported the presence of HPV type 16 infected patients with OSCC in Iraq 21 .
Overexpression of p53 tumor suppressor gene showed significant prevalence of high oncogenic HPV genotypes in patients with OSCC in Iraq. They found herald marks for the spread of HPV among Iraqi general population, which played an important role in oral carcinogenesis 21 . The expressions of p53 and its related gene MDM4 in oral, laryngeal and cutaneous squamous cell carcinoma have also been investigated using tissue microarray in Iraq 22 . It was found that p53 and MDM4 were frequently overexpressed in SCC cases with significant correlation between these markers, which was considered as an indicator for prognostic factor in regard to tumor grading 22 . Several molecular markers have been suggested as reliable prognostic cancer biomarkers for the classification of OSCC 19,[21][22][23][24][25][26] .
Considering all the previous facts regarding OSSC in Iraq, this study intended to investigate the expression of TIMP-2, and the difference in its expression level between positive and negative HPV-16 OSCC cases.

Sample collection
The study was conducted on 33 retrospective cases of OSSC taken from the Maxillofacial Center in Surgical Specialty Hospital in Baghdad between 2007 and 2009. The age of patients ranged between 30 and 95 years. None of the cases had received radiotherapy or chemotherapy. As negative controls, ten normal oral tissues, were obtained from a buccal mucosa of individuals underwent plastic surgery at the same hospital. This study was approved by the Ethical Committee / College of the Medicine/ University of Baghdad (No.2012, 11-12 Ar). The expression of HPV-16 and TIMP-2 were investigated using In situ hybridization (ISH) and immunohistochemistry (IHC) respectively.

Immunohistochemistry
Fixed paraffin tumor and control tissues were sectioned into 5 µm thickness and stained according to the TIMP-2 kit manufacture's protocol (Chemicon International, USA). The slides were placed in 10mM citric acid buffer at pH 6.0 and underwent antigen retrieval for 10 min at 680 W in a microwave oven. Mouse monoclonal antibody against TIMP-2 in 5Ug/ml (Chemicon International, USA) was then added and kept overnight at room temperature. After application of peroxidase labeled secondary antibody and the DAB, the sections were finally counterstained and mounted. About 10 representative areas containing 100 SCC cells were analyzed. Immunostaining was scored using combined quantitative (percentage of tumor cells with immunoreactivity) and qualitative criteria (intensity of staining: none, weak, or strong). These parameters were then combined as: 0= no staining; 1= weak (<10% of tumor cells); 2= weak (≥10% of tumor cells), and 3=strong (≥10% of tumor cells) 27 . Paraffin-embedded sections of colon cancer were immunostained as positive control.

In Situ Hybridization (ISH)
ISH procedure was done using DNA probe Hybridization/ detection System in Situ Kit (Maxim Biotech, USA), according to manufacturer's protocol. The biotinylated cDNA probe for HPV-16 (Maxim biotech, USA) was diluted to 7% (0.7 µL of the probe diluted in 9.3 µL of hybridization solution). Positive control was made with housekeeping gene probe, while the negative control made with hybridization solution without probe. Determination of positive reaction was made with nuclear and/or cytoplasmic blue staining of cells 28 .
Nuclear hybridization signals for HPV-16 were observed in 24 (72.7%) of OSCC paraffin-embedded tissues, while 6 (18.2%) cases appeared negative (Table 1). No detectable signals were seen in the sections from the control group. The immunoreactivity of TIMP-2 observed in 24 (72.7%) and 3 (9.1%) of positive and negative HPV-16 OSCC cases, respectively (Table 1). Immunoreactivity to TIMP-2 was localized mainly on the cell membrane and the cytoplasm of tumor cells (Figure 1. A, B). It was also detected on some stromal cells surrounding the tumor cells. No detectable signals were observed in the sections from the control group ( Figure 2). The total expression of TIMP-2 was detected in 27 (81.8%) of OSCC cases (Table 1) and its expression scores are illustrated in Table 2. There was no significant difference in TIMP-2 expression between HPV-16 infected and non-infected cases (p=0.058). In addition, the expression score of TIMP-2 revealed no significant difference between cases with and without lymph node metastases, p = 0.176 (Table 3), whereas, its expression showed no significant difference considering the tumor grade and stage, p = 0.838 and p = 0.169, respectively (Table 4).

Discussion
According to a recent review article 6 , there are shortage in the studies regarding the correlation between OSCC, detection of HPV and expression of strictly related tumor markers. However, published articles have shown that high oncogenic HPV genotypes have an essential etiologic role in the development of OSCC worldwide 17,[28][29][30][31][32][33][34][35][36] .
The correlation between HPV and OSCC was first suggested by Syrjänen et al., (1983) 14 , who detected koilocytotic atypias in malignant oral lesions by optical microscopy and the presence of viral DNA by means of ISH 14 .
The current study revealed positive nuclear hybridization signals for HPV-16 in 24 (72.7%) of OSCC cases. This result is in accordance with those of previous studies, which reported a similar detection rate among HNSCC tumors with 90% of the HPV types identified as HPV-16 34,35 . A previous study also confirmed that oral infection with HPV increased the risk of tumorigenesis of oropharyngeal cancer independent of tobacco use 35 . The result of the present study is also in agreement with a previous study in Iraq, which reported the significant prevalence of high oncogenic HPV genotypes in OSCC patients 21 .
The high-risk HPV adopts mechanism to generate the malignant progression of previously benign lesions. Earlier investigations have proved that high-risk HPVs and the two HPV-related oncoproteins E6 and E7 can immortalize and transform oral keratinocytes in vitro 34,35 . Pérez-Sayáns García et al. (2012) 32 stated that "the expression of TIMPs in OSCC is higher in tumors than in normal tissue, which correlates with an increase of metastatic risk and regional lymph node affectation". However, the role of TIMP-2 in cervical carcinoma has been approved to be down-regulated by acute expression of HPV oncoproteins that may favor deregulated MMP activity in the context of HPV infection 15,30 . In this study, immunoreactivity to TIMP-2 was observed in 27 (81.8%) of OSCC cases. This result is in accordance with previous studies, and reflect a pathological role of TIMP-2 in OSCC 17,18,[36][37][38][39] . The finding of increased TIMP-2 expression in oral cancer might be explained by the growth-promoting activity of TIMPs on a variety of cell types, or the induction of TIMPs by secreted MMPs from tumor-host interaction in the extracellular milieu 39 .
In addition, many immunohistochemical studies addressed the positive role for TIMP-2 in tumor progression and metastasis 3,16,18,[21][22][23]26 . It is believed that TIMP-2 suppresses the tumor invasion and metastases by inhibiting MMP-2. High activation of MMP-2 in OSCC has been reported to be associated with high expression of TIMP-2 in tumor cells 26,39,40 TIMP-2 over-expression has been linked with local tumor invasion, nodal status and clinical stage, as well as with disease-free survival in oral cancer, especially in tongue SCC 41 .
In this study, 72.7% of OSCC appeared to harbor detectable HPV-16, whereas, no significant difference was observed in relation between positive HPV-16 and TIMP-2 expression level. However, the p value was borderline (p = 0.058). The results of this study also revealed no significant association between level of TIMP-2 expression and tumor stage, tumor grade or lymph node metastases. However, many other researchers found significant relationships, and such studies include the following correlations between: A) the expression of TIMP-2 and tumor size or lymphatic metastasis 9,41 . B) the expression of TIMP-1 and lymphatic metastasis 9,41,42 ; C) the expression of MMP-2 and histologic grade or lymphatic metastasis 4,14,39,[42][43][44][45] .
In conclusion, this study investigated the expression of TIMP-2 in HPV-16 infected tumor tissues of OSCC in Iraq, and found that the percentage of samples that appeared to accommodate detectable HPV-16 reached up to 72.7%. However, no significant difference was observed in relation to TIMP-2 expression level. Future studies with a larger number of patients are highly recommended to address the possible association between TIMp-2 and OSCC positive HPV-16.