The present research tried to find the differences among EBV-positive samples, HPV-positive samples, HPV/EBV coinfected samples, and controls concerning the expression levels of EBV and HPV genes, the physical status of HPV integration, and the expression levels of some inflammatory, tumor suppressor, anti-apoptotic and anoikis related mediators including IL-17, IL-11, IL-6, IL-1, IL-8, TNF-α, NF-κB, VEGF, TGF-β, ROS, and RNS, Rb, p53, Survivin, Bcl-2, CD44, TWIST, E-cad, N-cad, PTPN13, and SLUG.
The viral infections globally accounts for about 12% of cancers, predominantly (> 85%) in developing countries [11, 12]. The HPV and EBV are reportedly associated with 38% of all cancers linked with viruses [39]. Evidence has proved an associstion between the HPV infection and the cervical cancer prognosis and carcinogenesis. There is also reportedly a strong association between the HPV infection and other cancers as an etiologic factor [40]; as well, studies reported the presence of EBV in the cervical cancer [41] and the prostate cancer [26]. In addition, the EBV/HPV co-infection has also been detected in the breast cancer [42, 43]. In our study, the HPV and EBV were isolated from in 31.3% and 49.3% of the PCa group, as well as in 15% and 40% of the control group, respectively. The EBV/HPV co-infection was present in 14.9% of PCa patients, 7.5% of control samples. Moreover, the high-risk strains (HPV 16 and 18) were responsible for 50% and 30% of HPV/EBV co-infected PCa samples, respectively. Also, the most common EBV genotype in the study was genotype 2. The EBV infection has been reported in the prostate tissues samples. The presence of the EBV was reported in 8.8% of malignant and benign prostate tissue samples (31 of 352) in Sweden [24], and in 8% of malignant, benign and normal prostate tissues (16 of 200) in the United States [25]. A study reported the EBV infection in about 37% of prostate cancer patients (n = 19) [23]. HPV-18 and EBV (EBNA1) gene sequences were detected in high proportions and almost equal to normal, benign and prostate cancer samples [26]. The EBV/HPV co-infection was significantly more prevalent in prostate cancer (55%) when comparing with benign samples (15%) and normal prostate samples (30%) [26]. Moreover, the EBV and HPV act simultaneously to enhance the cultured cervical cell proliferation, indicating that the same may happens to prostate epithelial cells [26, 44], in line with the present experiments about the high-risk HPVs and about the EBV in PCa.
The HPV genome integration is the key genetic phenomenon in the cervical carcinogenesis. Some studies revealed that the EBV infection accelerates the HPV genome integration into cervical cell genome in the cervix, which increases their genomic instability [18, 45] and develops lymphoepithelioma-like cervical carcinoma in some cases [41]. In line with this hypothesis, our results showed the maximum HPV genome integration rate was detected in the HPV/EBV co-infected PCa group (8/10, 80%). Also, the purely integrated HPV (47.4%) was significantly higher than the purely episomal HPV (5.2%) in the tumor tissues positive for the HPV. The purely integrated HPV was significantly higher in HPV/EBV co-infected PCa samples compared to the mono HPV-infected PCa samples (P: 0.0009).
Since no studies have been performed on the effect of the HPV/EBV coinfection on PCa development, we compared inevitably our findings with the results obtained from other cancers. The EBV acts probably as a co-factor for the HPV to induce the pathology of uterine cervix, verified by Szkaradkiewicz et al. [46] who reported a possible sexually EBV transmission route. Also, It has been found that sexually transmissible infections have been associated with increased risks of PCa [47]. However, there is no convincing evidence to support the hypothesis that EBV or HPV infection plays a direct role in the pathogenesis of prostate cancer. Additionally, a recent meta-analysis by de Lima et al. shows that EBV infection posed a two-fold increased risk of precancerous cervical lesions and four-fold increased risk of cervical cancer in HPV positive women [45]. However, in the current study, there were no significant associations between the presence of EBV, HPV, and HPV/EBV co-infection with PCa ( P = 0.35, P = 0.06, and P = 0.34, respectively).
There are limited empirical techniques to analyze the molecular pathways regarding the EBV/HPV coinfection. However, EBV LMP1 mixed with HPV16 E6 proteins in the transformed mouse embryonic fibroblasts (MEFs) caused a decrease in the residues of DNA Damage Response (DDR), including p27, pRb and p53, as well as an elevation in the levels of checkpoint kinase 1 (CHK1), Akt and MAPK signaling and NF-κB signaling [48]. Similarly, as mentioned in result 3.5 section, mean expression level of p53 and Rb in co-infecction group were lower than mono HPV- and mono EBV-infected groups but was not significantly different between co-infected group with mono HPV-infected group, while, there was a significant between coinfected group with mono EBV-infected group. Besides, there was no significant difference between EBV-positive samples with EBV-negative specimens. Additionally, a significant reverse correlation was found between the HPV E7 and E6 mRNA expression levels with those of p53 and Rb, respectively (Table 7). Therefore, it can be concluded which lower concentration of Rb and p53 in coinfected PCa group probably more due to the simultaneous presence of HPV and EBV infections but maybe HPV infection effect was more than EBV infection.
One of the important defense strategies of innate immunity against pathogens is inflammation that helps to heal damaged tissue [49]. However, inflammation during pathogenesis can play a dual role. Prolonged chronic inflammation promotes tissue damage and autoimmune disorders, in addition to can lead to the accumulation of damage to DNA and give rise to cancers [50, 51]. Chronic or recurrent inflammation can be developed following the persistent viral infection. The pivotal pathogenic mechanism in the development of cancer caused by EBV and HPV is to induce the cytokine impact and the chronic inflammation [14] but the function of HPV/EBV co-infection in PCa is unknown. In addition, the NF-κB has a key role in coordinating the expression of genes related to the chronic inflammatory diseases [52]. A wide variety of factors such as viral infection, oxidative stresses, cytokines (TNF-α, IL-1β), and tumor promoters can activate NF-κB. Also, NF-κB modulates expression of the various genes including cytokines (e.g. IL-1, 6, 8, TNF) and cell cycle regulatory molecules [49, 52]. A crosstalk between inflammation and cancer mediates by NF-kB thoroughly accepted [49]. It has been observed that an elevated EBER1 level in cervical cells may play a role in the transition from the inflammation to the oncogenesis of cervical cancer caused by HPV through the modulation of innate immunity [53]. The results of the current study showed that all inflammatory factors had significantly higher expression level in HPV-positive specimens when comparing with the HPV-negative ones (Table 5, result 3.4 section), while the concentration of VEGF, IL-11, IL-6, IL-1 and NF-kB was significantly elevated in the EBV-positive specimens in comparison with the EBV-negative ones (Table 6). There was also a significant association among the expression levels of LMP-1, E7 and E6 with NF-kB as well as between IL-6 and LMP-1 (Table 7). These results indicate that a possible association between inflammation with HPV and EBV in PCa. The level of NF-κB, TNF-α, IL-6, IL-17, VEGF, ROS and RNS in HPV/EBV co-infected PCa samples was higher than non-coinfected PCa samples. Besides, the mean expression level of TNF-α, IL-8, RNS, ROS, TGF-β, VEGF and IL-17 in co-infection positive PCa samples was significantly higher than mono EBV positive samples. Also, the inflammatory factors had greater mean expression level in the HPV/EBV co-infection group when comparing with mono HPV and mono EBV-infected groups but no significant difference was reported between coinfection and mono HPV groups. Therefore, by comparing the mean expression level of inflammatory factors between co-infection and mono-infection (Table 6, result 3.4 section) it can be concluded that HPV/EBV coinfection can increase the mean expression of inflammatory factors compared to mono HPV or mono EBV infection.
In a study by Grace et al., a significantly positive correlation was observed between the HPV-induced SCC and the expression levels of Bcl2/p53 proteins [54], confirming that the HPV could develop the cervical lesions by deregulating the induction of apoptosis [54]. In fact, it can be said that the high-risk HPV-E6 oncoprotein is able to enhance the Bcl-2 protein expression in the cervical cancer through the elimination of inhibitory activity of p53 over the Bcl-2 [55]. Survivin and Bcl-2 (as anti-apoptotic proteins), which widely expressed in most tumor tissues, are critical factors to regulate the progression of cell cycle and to impede the apoptosis [56]. In a study by Guo et al., the p53-induced survivin upregulation is promoted by LMP1 through an elevation activity of survivin promoter and activity of p53-survivin DNA binding, which means that the complexity of p53 regulation in Survivin occurs via the viral oncoprotein of LMP1 in NPC. Their model of p53-induced G1/S cell cycle progression could upregulate the LMP1-mediated expression of survivin in the pathogenesis of nasopharyngeal carcinoma [57]. Moreover, an LMP1-induced upregulation of Bcl-2 has been reported in B cells [58], while LMP1 has no effect on Bcl-2 levels in the leukemic T cell line Jurkat [59]. In a study of Muzio et al., the evaluation of survivin expression, oral premalignant lesions and oral carcinoma in the presence of HPV infection showed significantly higher expression level of survivin in the samples positive for HPV when comparing with those negative for HPV. The evidence suggested that the expression level of survivin may be influenced directly or indirectly by the HPV [60]. In this study, it was observed that the Survivin and Bcl-2 expression level in the HPV-infected, EBV-infected and HPV/EBV-coinfected samples were significantly higher than in HPV-negative, EBV-negative and non-HPV/EBV-coinfected samples, respectively. Additionally, in the current study, there was a direct association between the E6 and E7 expression level and the Survivin and Bcl-2 expression level as well as between the expression levels of LMP-1 and Survivin. The EBV/HPV co-infection may initiate a neoplastic transformation of carcinogenesis [22]. Moreover, an in vivo interaction occurs between HPV and EBV as well as an interaction is there between EBV and HPV oncoproteins [44]. According to result of this study, the Survivin and Bcl-2 expression level significantly was higher in HPV/EBV coinfected PCa group than in non coinfected PCa group as well as there was not significantly different between mean expression level of Survivin in HPV/EBV co-infected PC with mono HPV- and mono EBV-infected PCa samples. In general, these results suggest that the co-presence of HPV and EBV infection may lead to the resistance of cancer prostate cells to apoptosis, although probably the effect of the HPV infection is greater than that of EBV.
Metastasis refers to the removal of tumor cells from their original location and then their dissemination to and proliferation in distant locations [61]. To this end, epithelial cancer cells must achieve a flexible migratory phenotype. In this regard, the epithelial-mesenchymal transition acts as a developmental pathway for tumor cells to reach this phenotype [62]. One of the physiological barriers for the tumor metastasis is anoikis, but most tumor cells become resistant to the anoikis [63].
According to the findings, the Slug and Twist (EMT- inducing transcription factors) are responsible for the anoikis resistance, in contrast, E-cadherin significantly inhibits anoikis resistance [63]. The CD44 as a cell membrane receptor with multifunctional potential is highly expressed in many cancers such as PCa and accounts for cell-cell adhesion and tumor metastasis and invasion [64]. The tumor aggression may be inhibited by the PTPN13 phosphatase. The invasion of different cancer cells is suppressed by the PTPN13 [65]. In a study by Castilla et al., the PTPN13 gene silencing elevated the expression level of invasion-related genes in the PCa cells [66]. The HPV-E6, by losing the PTPN13, triggers the anchorage-independent growth in the human epithelial cells [67, 68]. Accordingly, the present study aimed to analyze the expression levels of SLUG, Twist, N-cadherin, E-cadherin and PTPN13 proteins were in prostate tissue lysates. As mentioned in result 3.5, the mean expression level of N-cad, SLUG and TWIST were significantly higher in the cases positive for HPV compared to that in the groups negative for HPV, while the PTPN13 and E-cad level in the HPV-positive PCa group was significantly downregulated compared with HPV-negative PCa group (Table 5). Based on the findings, the E6 and E7 expression levels had a significant negative and positive correlation with E-cadherin and N-cadherin/TWIST/SLUG, respectively (Table 7). It should be noted that no significant difference in the concentrations of PTPN13, TWIST, SLUG, N-cad and E-cad was found between EBV-positive and EBV-negative groups (Table 4). However, the mean CD44 expression level was significantly greater in EBV-positive samples than in EBV-negative ones and there was a positive correlation between the LMP-1 and LPM-2 expression level and the CD44 expression level (Tables 4 and 7) but no significant difference was seen between the HPV-positive and HPV-negative groups. It has been reported that the LMP1 down-regulates the E-cadherin expression and up-regulates TWIST and other transcription factors linked to the cell motility [69, 70]. The EBV-protein, LMP-1, leads to anoikis resistance by inducing the expression of anti-apoptosis proteins survivin, CD44, the inhibitor of DNA binding 1 (Id1), Bim and ROS [63, 71]. Moreover, the co-expression of E6 and LMP1 triggers some processes, including tumor-formation ability, anchorage-independent growth, resistance to apoptosis and cell proliferation, in nude mice when comparing with the expression of E6 or EBNA1 alone [72]. The LMP1 and HR-HPV E6 co-expression is associated with more aggressive malignant tumors, including cervical SCC and breast adenocarcinoma [73, 74]. In current study, the level of TWIST, and N-cad in HPV/EBV-coinfected PCa group compared to non-coinfected PCa group were increased, while the E-cad had significantly decreased expression level. However, no significant difference was reported in the mean expression level of anoikis-related factors between the PCa group positive for co-infection and the PCa group positive for mono HPV. As well as, the mean level of CD44 expression was not significant differences between HPV/EBV coinfection PCa group vs mono EBV-positive PCa group. Thus, the increase in mean expression level of CD44 in the coinfection group is most likely due to the presence of EBV virus. In conclusion, the existence of the HPV and EBV is probably the cause of these changes and viral infection plays important role in contribute to anoikis resistance and PCa development.
One of the limitations of this study was we had to sample a peripheral area of surgically dissected benign prostatic hyperplasia because of lacking access to normal or healthy prostate samples.