Human herpesvirus-6 and epstein-barr virus infections at different histopathological grades of oral squamous cell carcinomas.

BACKGROUND: The aim of this study was to determine the prevalence and viral load of Epstein-Barr virus (EBV) and Human herpesvirus-6 (HHV-6) in different histopathologic grades of oral squamous cell carcinoma (OSCC).
METHODS: Forty-five formalin-fixed paraffin-embedded tissue section of OSCC patients were analyzed by quantitative real-time polymerase chain reaction for detection of EBV and HHV-6.
RESULTS: The mean age of the patients was 58.6 years, 69% of whom were female, and 31% were male. Overall, the positive rate for EBV and HHV-6 were 16.7% and 27.1%, respectively; and the mean viral load EBV was 27.9 × 10(3) and 38.5 × 10(3) for HHV-6. No correlation was demonstrated between the viral load of EBV DNA (P = 0.35) and HHV-6 (P = 0.38) at the different OSCC histopathologic grades.
CONCLUSIONS: These findings neither lend support to the hypothesis that EBV and HHV-6 are directly involved in OSCC nor rule out the possibility that these viruses play an indirect role in carcinogenesis in this area.

INTRODUCTION

Squamous cell carcinoma (SCC) is the most common malignancy of the oral cavity and includes 88% of mouth and lips malignancies.[1] Different factors have been attributed to cause oral SCC, among them internal factors such as systemic disease, malnutrition, general resistance, and iron deficiency anemia and external factors include tobacco and alcohol use, syphilis, poor oral hygiene, and sunlight (at vermilion) could be mentioned.[2]

Epstein–Barr virus (EBV) whose cell lines were determined by Tony Epstein and Yvonne Barr from Burkitt's lymphoma in 1965[3] belongs to the genus lymphocryptovirus of γ-herpesvirus family.[4] Its early exposure in infancy is usually asymptomatic, whereas it causes infectious mononucleosis in adolescence. In general, EBV infects more than 90% of the world adult population.[4] In B lymphocytes, EBV causes lytic infection, while, in epithelial cells, it results in latent infection.[5] EBV gene expression alters the biological properties of the infected cells in latent and lytic infections and may result in cancer in humans.[6] This virus is involved in several human malignancies including Burkitt's lymphoma, Hodgkin's lymphoma, and nasopharyngeal carcinoma.[7]

Human herpesvirus-6 (HHV-6) was firstly extracted from peripheral blood mononuclear cells of the patients with AIDS and lymphoproliferative disorders by Salahuddin in 1986.[8] According to genetic characteristics, the virus is classified as a β-herpesvirus, although based on biological properties it can be placed in the γ-herpesvirus subfamily.[9] Primary infection occurs during the first 2 years of life and is usually associated with undifferentiated febrile illness, although some children show the classical manifestation of roseola infantum (exanthema subitum).[10] HHV-6 encodes the U24 protein which has been implicated in the pathology of multiple sclerosis.[11] This virus is associated with various oral lesions and is reported to be present in different cells; its reactivation in epithelial tumors of the oral cavity has also been shown.[12] Since recognition of the role of viral factors seems important in the development of oral SCC (OSCC) and the role of viruses in oral cancer may vary among different countries, the present study evaluated the frequency of EBV and HHV-6 viruses in OSCC samples in Zahedan, Sistan and Baluchestan Province located in the south-east of Iran.

METHODS

General materials

All the diagnosed OSCC samples of Department of Pathology, and Zahedan School of Dentistry from 2004 until 2013 were evaluated. The histopathologic grading samples were performed by two oral and maxillofacial pathologists through microscopy in a three-grade scale of I to III. High mature tumors resembling their tissue of origin, that is squamous epithelium, were graded as well differentiated SCC or grade I. In contrast, tumors with high nuclear and cellular polymorphism and very low or no production of keratin were classified as poorly differentiated SCC or grade III and tumors with a microscopic appearance between these two grades were classified as moderately differentiated SCC.[13] Demographic information was also obtained from the patients files.

Samples collection

This cross-sectional research protocol was approved by ethical committee of Research Deputy of the Zahedan University of Medical Sciences. Three sections were prepared from the formalin-fixed paraffin-embedded (FFPE) tissue blocks and were put in 1.5 ml microtubes for DNA extraction which was performed using the specific kit of RecoverAll™ Total Nucleic Acid Isolation Kit for FFPE (Ambion, Carlsbad, California, USA). In brief, paraffin-embedded tissues are deparaffinized using xylene and wash the pellet twice with ethanol. Next, add protease digestion with an incubation time tailored for recovery of DNA. Nucleic acids are purified using a rapid glass-fiber filter methodology that includes an on-filter nuclease treatment and are eluted into water.

The DNAs were then evaluated using specific primers for HHV-6 and EBV of the AmpliSens® EBV/cytomegalovirus (CMV)/HHV-6-screen-FRT polymerase chain reaction (PCR) kit (Moscow, Russia). The PCR condition was showed in Table 1. The fluorescent quantitative real-time PCR amplification curve of HHV-6 showed in Figure 1.

Table 1

The PCR conditions amplification of EBV and HHV-6

Figure 1

The fluorescent quantitative real-time polymerase chain reaction amplification curve of Human herpesvirus-6

Statistical analysis

Data analysis was run by SPSS software version 17 (Chicago, IL, USA). Data are presented as mean ± SD Chi-square, ANOVA, Kruskal–Wallis, Mann–Whitney tests and Spearman correlation coefficient were used for the comparison of statistical differences between the groups, and P < 0.05 was considered statistically significant.

RESULTS

The mean age of the patients whose samples were evaluated was 58.6 ± 13.9 years. Of 45 studied samples, 31 patients (69%) were female with a mean age of 56.4 ± 12.8 years and 14 patients (31%) were male with a mean age of 63 ± 15.3 years. The gender of three patients with OSCC was unknown.

Of all samples, 21 (43.8%) cases were from grade I, 19 (39.6%) grade II, and 8 (16.7%) grade III and no significant correlation was found between the mean age of the patients and the histopathological grades (ANOVA test, P = 0.85).

Epstein–Barr virus positive was found in 8 (16.7%) samples and HHV-6 positive in 13 (27.1%) samples. The mean viral loads of EBV and HHV-6 in different histopathologic grades of positive samples are shown in Table 2. No correlation was demonstrated between the viral load of EBV (P = 0.35) and HHV-6 (P = 0.38) at the different OSCC histopathologic grades.

Table 2

Viral loads comparison of EBV and HHV-6 in different histopathological grades in positive samples

According to Spearman correlation coefficient, a significant correlation was found between the viral load of EBV and HHV-6 in all cases (r = 0.37, P = 0.009), while no significant correlation was found between the viral load of EBV and HHV-6 in positive cases (r = 0.001, P = 0.999).

There was also no relationship between the viral load of EBV and HHV-6 in positive cases and patient's age, (Spearman correlation coefficient, r = 0.156, P = 0.71 and r = 0.096, P = 0.78 respectively). The viral loads of EBV and HHV-6 in positive samples are shown separately for two genders in Table 3. According to the Chi-square test, no significant statistical association was observed between the prevalence of HHV-6 and gender (P = 0.48). Due to the low number of EBV positive cases, it could not be analyzed.

Table 3

Viral loads comparison of EBV and HHV-6 in two genders in positive samples

Generally, 7.7% of OSCC samples were observed in lips, 7.7% in tongue, and 32.7% in the buccal mucosa, 1.9% in the floor of the mouth, 17.3% in the maxillary gingiva, and 32.7% in the mandibular gingiva. All OSCC samples in tongue were negative for EBV and HHV-6. Considering the lack of data in different locations, the viral loads only described in different location and data analysis was not performed [Table 4].

Table 4

The viral load of EBV and HHV-6 distribution by location in positive samples

DISCUSSION

Oral cancer is the sixth most common cancer in the body including SCC as the most common one. SCC has different etiological causes and viruses which have been proposed as an etiological factors.[2]. The present study aimed to evaluate the prevalence and viral load of EBV and HHV-6 infection in different histopathologic grades of OSCC.

In this study, the prevalence of EBV in OSCC was 16.7% which was higher than a study performed in the northeast of Iran.[14] It may be due to the difference of studied population, since they had only examined OSCC in patients <40 years of age.[14] The prevalence of EBV infection on OSCC has been reported from 0% to 100% in different areas of the world.[151617181920] The reason for this difference may arise from geographic diversity of different samples. For instance, Tsuhako has reported different frequencies of EBV-positive OSCC in different regions of Japan.[21] To determine EBV in these studies, various techniques such as; PCR, in situ hybridization, southern blot analysis, and immunohistochemistry have been used. The difference in the prevalence has been attributed by some researchers to different sensitivities of the assessment techniques[1520] and the statistical methods used.[1622]

Epstein–Barr virus has also been proposed to be associated with malignancies such as Burkitt's lymphoma, nasopharyngeal carcinoma, gastric cancer, and salivary glands cancers.[23] EBV encodes several proteins with transformation potential including EBV nuclear antigens 2 and 3 (EBNA2 and EBNA3) and latent membrane protein 1 and 2 (LMP1 and LMP2).[24] In vitro studies have shown that LMP1 can inhibit p53-mediated apoptosis in epithelial cells.[23] The high prevalence of EBV DNA in tumoral tissues can be due to infiltration of EBV-infected B lymphocytes and macrophages or the higher sensitivity of tumor keratinocytes to EBV infection altered by the immunological environment in the presence of the tumor.[25] Decreasing local, and general immunity can also increase the prevalence; accordingly, studies have reported high prevalence of EBV in clinically normal oral mucosa of immunocompromised patients and hence support this theory.[15]

Failure to detect EBV DNA in some studies has been described by the “hit and run” theory, according to which viral DNA is required only for malignancy induction, and it disappeared during uncontrolled cell cycles of the host[22] In some studies, the viral DNA has been isolated from patients with OSCC via PCR; however, the expression of viral proteins could not be determined.[2125] Various descriptions have been proposed in this regard, for example, the identified viral genomes in carcinoma cells may remain un-translated, or they may originate from oropharynx and be present in the saliva and hence contaminate tumor samples without being a carcinogen.[22] It has been also proposed that the viral DNA may reside in infiltrating lymphocytes.[26] Thus, the role of EBV as a risk factor still remains controversial.

The mean viral load of EBV and HHV-6 in positive samples of OSCC was 27 ± 31 × 103 and 38 ± 42 × 103, respectively. According to the correlation of histopathologic grades of OSCC with EBV prevalence, various studies have obtained different results. For instance, no significant association has been reported between histopathologic grades of OSCC and EBV by Sand et al.[15] Despite stating the lack of correlation between histopathologic grades and EBV prevalence by Maeda and Shamaa et al., they observed higher EBV prevalence in lower histopathologic grades of OSCC.[2327] Given the increased nuclear atypia in OSCC, Gonzalez-Moles et al. have shown a higher prevalence of EBV.[17] In contrast to these studies, Kobayashi et al. have reported a higher prevalence of EBV in higher histopathologic grades of OSCC, which was statistically significant.[16] In the present study, no significant correlation was observed between the mean viral load of EBV and histopathologic grades of OSCC in positive samples (P = 0.77).

Given the infectious nature of EBV and its tendency to spread, isolation of the virus from multiple lesions has raised the hypothesis of “field cancerization” in tumors of the head and neck.[28] This hypothesis was first proposed in 1953 and implies that the surrounding epithelium of tumor samples taken from the upper gastro-respiratory path has histopathologically changed.[29] However, further studies are needed to confirm this hypothesis.

The mean age of patients in the present study was 58 ± 13 years and consistent with the other studies;[152327] no relationship was observed between age and the prevalence of EBV and the viral load. Although not statistically significant, the prevalence of EBV was higher in men than women. Maeda et al. did not find a significant correlation between the prevalence of EBV in OSCC and gender,[23] whereas Shamaa et al. reported a significantly higher prevalence in men.[27] Other cancers related to EBV such as Hodgkin's lymphoma, Burkitt's lymphoma, and nasopharyngeal carcinoma are also 2-3 times more common in men. This difference in EBV-related gastric cancers has been attributed to different exposures over a lifetime, occupational risk factors, and biological factors or hormones in both males and females. Based on information, testosterone suppresses the immune system, and thus men are more prone to infection and its outcomes.[30] Environmental factors such as the use of smokeless tobacco products can increase the risk of oral cancer through increased EBV replication and associated inflammatory processes in the oral cavity.[31] Perhaps higher prevalence of smoking among men in this region is due to the increased prevalence of this virus in OSCC of men. However, Sand and D’Costa et al. did not observe a significant association between chewing tobacco, snuff use, alcohol consumption, and smoking and the prevalence of EBV in OSCC samples.[1518] In the present study, no significant correlation was found between OSCC location and EBV prevalence; this finding is consistent with the studies of Maeda and Sand et al.[1523] However, Gonzalez-Moles et al. has reported a significantly higher prevalence of EBV in the lateral surface of the tongue.[17]

HHV-6 was identified in some malignancies such as oral cancer, cervical carcinoma, and Hodgkin's and non-Hodgkin's lymphomas.[32] HHV-6 produces the oncoprotein of open reading frame-1 which binds to wild-type p53 and disrupts cell cycle regulation. Neoplastic transformation of nontumoral human epidermal keratinocytes by HHV-6 DNA has been also shown.[32] In the present study, the prevalence of HHV-6 in the examined samples was 27%. Using PCR, Yadav et al. reported the frequency of HHV-6 in OSCC in 1994 and 1997 as 67% and 57%, respectively,[3334] which are much higher than our result. Significant increase of HHV-6 antibody levels in OSCC patients compared to controls indicates a possible role of HHV-6 in the pathogenesis of OSCC.[35] In the present study, no significant correlation was found between the mean viral load of HHV-6 and histopathologic grades of OSCC (P = 0.72). Similarly, in a study by Yadav et al., no correlation was observed between the presence of HHV-6 antigen and differentiation of OSCC through immunohistochemistry. Yadav et al. had studied the presence of HHV-6 in OSCC in three methods; the highest HHV-6 was found with immunohistochemistry and in situ hybridization and the lowest with PCR; the reason of which was attributed to the fragility and susceptibility of that part of DNA amplified by PCR due to formalin-fixation treatment.[34] There was no significant correlation between the viral load of HHV-6 in OSCC and age and gender in this study. In addition, no study examining the association of age, gender with the prevalence of HHV-6 in OSCC was found to be compared with ours.

HHV-6 is capable of activating other HHVs such as EBV, CMV, and papillomaviruses. Razavi et al. have reported a significant association between the presence of HHV-6 and EBV genomes in salivary gland neoplasms.[36] In addition, a significant correlation was suggested by Zhou et al. between HHV-6 and EBV infections and histopathologic progress of angioblastic T-cell lymphoma.[37] In the present study as well, a correlation existed between the mean viral load of EBV and HHV-6 in OSCC.

CONCLUSIONS

In total, a significant correlation was not found between the mean viral load of EBV and HHV-6 and histopathologic grades of OSCC. These findings, therefore, do not lend support to the hypothesis that EBV and HHV-6 are directly involved in OSCC. The data do not rule out the possibility that these viruses play an indirect role in disease pathogenesis that they use a ‘hit and run’ mechanism or are involved in a small minority of cases.