Production and secretion of inflammatory cytokines are among the main events that take place upon viral infection. These molecules coordinate host cell-mediated immune response by recruiting cellular elements from the immune system and by regulating gene expression on target cells [34, 35]. The pleiotropic cytokine TNF is a key regulator of inflammation of the epithelia with a well-documented capacity to induce growth arrest in normal or HPV16-immortalized keratinocytes, mainly in the G0/G1 phase of the cell cycle . Conversely, we have previously reported that HPV18-immortalized and both HPV16 or HPV18-transformed cell lines are resistant to TNF-induced growth arrest [22, 23].
In order to address the yet unknown molecular bases of this difference we applied cDNA microarray technology to compare the global gene expression profiles of TNF-sensitive normal and HPV16-immortalized keratinocytes with that of TNF-resistant HPV18-immortalized ones. Some limitations of this study are the use of a reduced number of samples and the existence of differences in cell culture conditions which are inherent to our experimental setting, i.e. the differences in cell density at the different time-points described above. However, using this approach we identified a group of genes that clearly distinguish both cells groups (Figure 2 and Table 1). This indicates that TNF-sensitive cell lines share a group of genes which are regulated in a way that clearly differentiate them from the TNF-resistant one.
On the other hand, when we analyzed changes in global gene expression as a function of TNF treatment we observed that HPV16 and HPV18 samples could not be distinguished from each other while normal keratinocytes could be readily discriminated (Figure 3). This observation suggests that the presence of either HPV16 or 18 has an impact on TNF-regulated gene expression. In line with these observations, several studies have shown that HPV positive cells exhibit impaired TNF pathways [37, 38]. Moreover, it has been reported that the effects of TNF on HPV-harboring cells depends on variables as cell type studied, the virus type present and culture conditions (i.e., growth factors availability). This cytokine is capable of inducing the proliferation of HPV16-immortalized human cervical epithelial cells cultures in the absence of growth factors through an autocrine, EGF receptor-dependent, pathway . Besides, TNF can upregulate E6/E7 RNA expression and cyclin-dependent kinase activity in these cells . Conversely, it has been reported that TNF exerts a potent cytostatic effect on HPV16-immortalized keratinocytes while HPV18-immortalized as well as cervical carcinoma-derived HPV-positive cell lines remain unaffected [22, 23]. Furthermore, it has been observed that increased tumorigenicity of human keratinocytes transformed by HPV16 is associated with resistance to TNF cytostatic effect . Finally, it was demonstrated that TNF downregulates HPV18 transcription in non-malignant HeLa-fibroblasts hybrids, while viral expression in tumorigenic hybrids segregants as well as in parental HeLa cells remained undisturbed . On the other hand, it has been consistently observed that TNF negatively regulates normal keratinocytes proliferation in monolayer [22, 23, 36] as well as in organotypic cell cultures [41, 42]. Altogether, these data support the notion that acquisition of resistance to TNF by HPV-infected cells may represent an important step towards malignancy.
Despite the existence of similarities between the two high-risk HPV types used to generate the cell lines studied, the fact that HPV16 and HPV18 are different viruses that exhibit clear differences in their biological activities must be highlighted. For instance, epidemiological studies have shown that HPV18 is more associated to cervical adenocarcinomas while HPV16 is more prevalent in squamous cell carcinomas [43–45]. Furthermore, compared to other HPV types HPV18 has been associated with increased transforming potential in cell culture systems and with poorer cancer prognosis at the clinical level [26, 27, 46, 47]. On the other hand, HPV16 exhibits a greater potential to establish persistent infections that can progress to high-grade lesions [48, 49]. Although we cannot explain the molecular bases of the differences in gene expression between these cell lines, we believe that this may reflect the divergences that exist between these HPV types.
We next searched for genes that best distinguish between TNF-sensitive and TNF-resistant cells by pair-wise comparison both before and after cytokine treatment for 3 or 60 hours. By this means we identified 48 and 52 different genes, respectively, that set apart TNF-sensitive from TNF resistant cells (Figure 4, Table 3, additional files 4 and 5). The functional characterization of these genes shows that they are involved in critical cellular processes such as regulation of proliferation, differentiation and cell adhesion. Altogether, the differential expression of these genes may contribute to the differential response to the cytostatic effect of TNF observed in these cells.
Two genes, namely KLK7 and SOD2, were selected for further analysis based on their reported function and expression profile (Figure 5A). KLK7 expression pattern was validated by Northern blot and showed that it is equally expressed in TNF-treated or untreated HPV18-immortalized cells but is not detected in PHK or HPV16-immortalized cells (Figure 5B). Kallikreins are a sub-group of serine proteases with different physiological functions. In humans, kallikreins are encoded by 15 structurally similar, steroid hormone-regulated genes that co-localize to chromosome 19q13.4, representing the largest cluster of contiguous protease genes in the entire genome [50–52]. These proteins mediate the proteolytic degradation of cohesive intracellular structures associated to epithelial differentiation. Recent data also suggest that kallikreins may be causally involved in carcinogenesis, particularly in tumor metastasis and invasion, and, thus, may represent attractive drug targets to consider for therapeutic intervention . Consistent with our findings, it has been observed that KLK7 expression is up-regulated in cervical tumors as well as in cells lines derived from them. On the other hand, normal keratinocytes express low levels of this protein [53, 54]. Furthermore, KLK7 expression has been found up-regulated in breast  and ovary tumors  and is being considered a new tumor progression marker.
The superoxide dismutase 2 (SOD2) expression pattern was also validated by Northern blot (Figure 5B). This gene is up-regulated in TNF-sensitive but not in TNF-resistant cells. The superoxide dismutase 2 (SOD2) belongs to a family of enzymes involved in the conversion of superoxide radicals in molecular oxygen. Reactive oxygen metabolites have multifactorial effects on the regulation of cell growth and malignant invasion. Furthermore, numerous in vivo studies have shown that the superoxide dismutases can be highly expressed in aggressive human solid tumors [57–59].
Previous reports have shown that activation of the transcription factor NF-κB is essential for the induction of SOD2 by TNF and IL-1β [60, 61]. Here we show that TNF-sensitive cells exhibit higher levels of activated NF-κB than TNF-resistant ones after cytokine treatment (Figure 6 and additional file 6). Several studies have shown that NF-κB is a negative regulator of keratinocytes proliferation in the epidermis, and that it plays an important role in cell differentiation and tissue homeostasis [62–64]. In stratified epithelia NF-κB is found in the cytoplasm of proliferating cells from the basal layer while it is detected in the nuclei of non-proliferating cells from the upper layers. Furthermore, it has been observed that NF-κB superexpression is associated with epidermal hypoplasia while its down-regulation promotes hyperplasia . Overall, these data suggest that alterations in TNF-mediated NF-κB activation pathways can play a role in the development and progression of HPV-associated epithelial and mucosal lesions.