The TP53 gene is known as the “sentinel gene” due to its ability to identify cell damage and to coordinate complex mechanisms to mediate cell repair, protecting genome stability and, consequently, cell homeostasis. Therefore, it is not surprising that TP53 tumor suppressor is the most frequently mutated gene in human tumors. Most of the alterations described are missense mutations, whereby the protein loses its primary function, or it acquires oncogenetic functions (Datta et al., 2017; Wang and Sun, 2017; Sammons et al., 2020). In addition, the p53 mutant protein may also facilitate the adaptation of tumor cells to the disadvantageous environment that arises as the tumor grows (Mantovani et al., 2019).

The association between human papillomavirus (HPV) and some cancers, including cervical, head and neck, vulvar, anorectal, and penile squamous cell carcinomas (SCC), is well characterized by the canonical mechanism involving the HPV oncogenes E6 and E7 and p53 and RB1 proteins (de Martel et al., 2017). During the transformation process, HPV inserts its genetic material into host human integration sites (IS), which have been identified in regions harboring cancer-related genes, as well as in regions presenting copy number alterations (CNAs) (Busso-Lopes et al., 2015; Macedo et al., 2020; Pinatti et al., 2021). Indeed, integration of DNA-copy number alterations and other omics data have shown that DNA methylation, mRNA, and miRNA expressions alterations affect coding-genes and miRNAs located within or near the HPV common integration sites (Barzon et al., 2014; Groves and Coleman, 2018; Rosa et al., 2019; Pinatti et al., 2021).

In the last decade, it has been also demonstrated that the wild-type p53 protein plays primarily its role as a transcription factor by regulating a large network of protein-coding genes and non-coding RNAs, including miRNAs, both inducing or repressing their targets (Hermeking, 2012; Fischer, 2017). In addition, the p53 protein regulates miRNA processing, primarily through its central DNA-binding domain, a target site of most cancer-specific mutations. Interestingly, miRNAs can also regulate p53 expression by matching in the seed region into the 3′UTR of TP53 mRNA, directly inducing the repression of TP53 or its regulators (Hermeking, 2012; Hermeking et al., 2014). Although several tumor-specific alterations in the p53-miRNA network have been described in different cancers (Hermeking, 2012; Datta et al., 2019), there is no data on miRNAs targeting of TP53 gene in HPV-associated penile cancer.

PeCa is a rare carcinoma in developed countries, but it presents higher incidence rates in South America, Asia, and Africa, where limited economic and social conditions play a large impact leading to delay in diagnosis, and treatment initiation. In Brazil, specifically in the Northeast region that is particularly affected by low socio-economic conditions and educational levels and high frequency of HPV infection, presents a high incidence of PeCa, with patients presenting additional comorbidities, which contributes to a high incidence of mortality rates (Macedo et al., 2020; Silva et al., 2021) However, even in countries that are not impacted by major economic limitations, the incidence and mortality rates of PeCa has increased, mainly among younger patients (Hansen et al., 2018). Hence, the increased occurrence of PeCa, irrespectively of the countries’ socioeconomic conditions, has suggested that HPV infection is possibly the main triggering mechanism for tumor development, in addition to poor hygiene of the genital region, phimosis, uncircumcision, and chronic inflammation (Christodoulidou et al., 2015; Kidd et al., 2017; Adashek et al., 2019).

PeCa treatment options are limited. No effective target therapy is available, mainly due to the scarcity of knowledge on the molecular pathways involved in the development and progression of these tumors. Limited data is available on the role and mechanisms of miRNA deregulation in PeCa, including those that disrupt miRNAs targets that regulate critical genes associated with the action of HPV, such as the TP53 and RB1 genes. Considering the high frequency of HPV infection in patients with advanced PeCa in Maranhão State, in Northeast of Brazil, in the present study miRNAs expression analysis was performed in high-risk HPV-associated PeCa with TP53 and RB1 mRNA downregulated expression, as previously reported by our group in >80 and 60% of the patients, respectively (Macedo et al., 2020).

This study opens the opportunity to better understand the role of TP53 and RB1 transcriptional regulators in HPV-associated penile carcinomas and brings much needed knowledge on the molecular tumorigenesis of this still-neglected tumor.

It is well known that the integration of the human papillomavirus (HPV) can occur at or near cancer-related genes (Durst et al., 1987). However, it is not completely understood the mechanisms by which the HPV virus controls its integration into the host cell genome and the molecular consequences that ultimately lead to the development and progression of the HPV infected tumors. Studies have used high-performance technologies to identify virus integration sites in the host genome to better understand the molecular alterations that occur in the host cell, leading to the loss of its genomic stability (Akagi et al., 2014; Bodelon et al., 2016; Liu et al., 2016; Gao et al., 2017; Rosa et al., 2019). The most well-known example is the canonical HPV/TP53/RB1 signaling pathway initiated by the viral E2 disruption. This leads to the loss of the negative feedback control of the viral oncoproteins E6 and E7, mediating ubiquitination and degradation of the p53 and pRb proteins, respectively (Squarzanti et al., 2018). Other authors proposed that HPV integration also directly causes activation of oncogenes or inactivation of tumor suppressors, as reported in HPV-related squamous cell carcinomas (Parfenov et al., 2014; Hu et al., 2015).

In our previous study, we showed downregulated mRNA expression of the TP53 and RB1 genes in 86 and 65% of high-risk HPV-associated PeCa, respectively. In the present subset of cases, we evaluated miRNAs’ expression, and observed that 73 and 69% were downregulated for both genes, respectively, suggesting the existence of other regulatory mechanisms in addition to the canonical HPV/TP53/RB1 pathway (Macedo et al., 2020). Although not all the cases presented with expression alterations in these genes, these results were recently corroborated by Furuya et al., who also described TP53 reduced expression levels in penile tumors.

Compared to other cancers (Santos et al., 2018; Datta et al., 2019; Hussen et al., 2021; Liu et al., 2021), few studies have described epigenetic events in penile tumors, whether evaluating miRNAs (Zhang et al., 2015; Hartz et al., 2016; Kuasne et al., 2017; Peta et al., 2017; Ayoubian et al., 2021; Furuya et al., 2021) or by evaluating methylation patterns (Feber et al., 2015; Kuasne et al., 2015; Marchi et al., 2017). Changes by both mechanisms could justify the downregulation of TP53 and RB1, however only RB1 has been reported to be hypermethylated (Marchi et al., 2017). Additionally, as most pathogenic variants of these genes have been described in coding regions, the mRNA downregulation of TP53 and RB1 does not appear to be due to mutations (Feber et al., 2016; Wang et al., 2019; Chahoud et al., 2021). Furthermore, few studies have evaluated patients’ cohorts with a high incidence of HPV infection, remaining poorly known the impact of HPV infection in disrupting mRNA/miRNA networks in penile tumors (Zhang et al., 2015; Hartz et al., 2016; Kuasne et al., 2017; Ayoubian et al., 2021; Furuya et al., 2021). In the present study, our main goal was to determine whether altered miRNAs expression could be associated with the down-regulation of the TP53 and RB1 expression in the etiopathogenesis of HPV-associated PeCa. This goal is of critical relevance to these particular virus associated with infected tumors, considering that the patient cohort investigated, from the State of Maranhão in Northeastern Brazil, is characterized by advanced PeCa and a high rate of HPV infection (>90%), as we reported previously (Coelho et al., 2018; Macedo et al., 2020). In addition, this study can provide useful information to target HPV-specific molecular pathways in human cancers.

In the present study, all patients were tested for HPV by nested-PCR followed by DNA sequencing. Using two highly sensitive methods we successfully detected HPV infection in 100% of men with PeCa, all of them with high-risk subtypes. Despite the high HPV prevalence in all human populations, occurring as hundreds of types, subtypes, and variants, many of them are not associated with cancer. On the other hand, it is well established the correlation between high-risk HPV and severe dysplasia, in situ and invasive cancer, usually as monoclonal lesions due to clonal selection from less advanced precursors (Pontén and Guo, 1998). This may explain why we were able to detect HPV in 100% of the primary tumor since all of them presented high-risk genotypes.

It is well established that p53 regulates the expression of both protein-coding genes and non-coding RNAs (Hermeking, 2012; Fischer, 2017). TP53-regulated miRNAs can mediate tumor suppression in response to cellular stress; similarly, the expression and activity of p53 can also be under the control of miRNAs (Hermeking, 2012). More than 20 miRNAs have been described to directly regulate p53 via canonical bindings (seed) in 3′UTR (reviewed by Liu et al., 2017). Down regulation of TP53 through seed sequences induce phenotypes that are consistent with loss of p53 function, such as reduced apoptosis, cellular senescence, increased invasion, and growth of tumor cells (Hermeking, 2012; Deng and Sui, 2013; Hermeking et al., 2014). Despite the increasing number of miRNAs that form the TP53 mRNA/miRNAs interaction network, there is no information on TP53-repressor miRNAs in HPV-associated PeCa. Our data revealed a total of 507 differentially expressed miRNA (miRDE) between the tumor and non-tumor tissue of HPV-infected PeCa patients, of which 494 were downregulated and 13 were upregulated. Among the 13 miRDE upregulated, five (let-7a-5p, miR-130a-3p, miR-15b-5p, miR-21-5p and miR-25-3p) were found overexpressed in 100% of the tumors analyzed. Moreover, miR-130a-3p, miR-15b-5p and miR-31-5p were predicted as novel regulator for TP53 gene; while miR-142-3p, miR-200c-3p, miR-205-5p, miR-223-3p, miR-22-3p, miR-25-3p and miR-31-5p for RB1.

Several studies have shown up-regulated expression of these miRNAs in several types of tumors. Overexpression of let-7a-5p has been observed in HCV-related cirrhosis (Petkevich et al., 2021) and liver cancer and ovarian cancer, where it presents a non-invasive diagnostic potential (Liu et al., 2021). Corroborating our data, some studies have also suggested that TP53 is a target of let-7a-5p (Balakrishnan et al., 2014; Pillai et al., 2014; Nunez Lopez et al., 2019; Zhou et al., 2019). MiR-130a-3p is recognized as a miRNA with tumor suppressor action (Kong et al., 2018; Song et al., 2021), that may act directly (Causin et al., 2021) or indirectly (Hu et al., 2021) in cancer progression. On the other hand, miR-15b-5p has generally been described to act on cell proliferation mechanisms, such as the ones involving the LATS2 (Liu et al., 2020), BCL-2 (Zhang et al., 2015), and PTPN4/STAT3 pathways (Liu et al., 2020).

Interestingly, down-regulation of TP53 by miR-25 resulted in a decrease in apoptosis in HCT116 colon cancer cells, A549 cells, NSCLC, and multiple myeloma cells (Kumar et al., 2011). In lung cancer, miR-25 was observed to promote cell proliferation and also inhibit apoptosis by down-regulating the expression of the MOAP1/TP53 axis genes (Wu et al., 2015). Moreover, recent evidence shows that miR-25-3p may also act with LncRNAs on a LINC00858/miR-25/SMAD7 axis modulating TP53-wild expression in colorectal carcinoma (Zhan et al., 2020). (Wang et al., 2021) also observed that exosomal miR-25-3p induced cell proliferation and resistance to temozolomide in glioblastoma through down-regulation of FBXW7, promoting c-Myc and cyclin E expression. MiR-21-5p, also observed up-regulated in this study, was shown by (Huang et al., 2021) to negatively regulate the tumor suppressor PDCD4 and cause resistance to by Osimertinib by interfering with MEK/ERK signaling.

These results are in concordance with the suppressive effect of these miRNAs in the HPV-related genes observed in PeCa in the present study. Our computational analysis revealed that TP53 and RB1 have 131 and 490 target sites for the 13 upregulated miRNAs, respectively. The highest number of miRNA binding sites were identified in coding regions, and not in UTR regions, as reported by (Hafner et al., 2010) Furthermore, 98.5 and 97.3% of the sites in TP53 and RB1, respectively, are in non-canonical seedless regions, presenting high levels of complementarity and conservation. Although most miRNA targets have sites that are perfectly complementary to the seed region, it has been shown that miRNAs can directly interact with seedless binding sequences, even improving their function (Shin et al., 2010). (Lal et al., 2009) for example, presented evidence of cell proliferation control by miR-24 in the E2F2/MYC axis through seedless binding in the 3′UTR region. (Park et al., 2017). also showed that destabilization of miRNAs targets is dramatically increased when binding occurs in non-canonical seedless regions. Therefore, we propose that the 13 miRNAs overexpressed in PeCa directly repress TP53 and RB1 by silencing their messenger RNA at different binding sites, especially in non-canonical seedless regions.

Although a unique miRNA may have a pivotal role in a particular pathway, most miRNAs act targeting multiple mRNAs, affecting the same or several gene pathways. Considering the 13 upregulated miRNAs, we predicted six main pathways by the enrichment analysis. Viral carcinogenesis (hsa05203), in which TP53 and RB1 act, was the main pathway affected (p < 1.00 × 10-325). MiR-205-5p, miR-16-5p, miR-22-3p, miR-93-5p, let-7a-5p, and miR-130a-3p were involved in most of the pathways affected. Glioma and cell cycle pathways, in addition to viral carcinogenesis, were previously shown to be regulated by other miRNAs identified in cytobands affected by CNVs in the same population from Maranhão State (Silva et al., 2021). Thus, these current findings reinforce the involvement of these pathways in HPV-associated penile tumorigenesis.

It is worth highlighting that 84.6% of miRDE were observed to be located in HPV integration sites (HPV-IS), including the five miRDE overexpressed in 100% of cases (sites at 3q25.33, 7q22.1, 9q22.32, 11q12.1, and 17q23.1). Several viruses mediate tumorigenesis by expressing viral oncogenes or activating host oncogenes through the integration of viral DNA into the human genome (Lee and Dutta, 2009; Tuna and Amos, 2017). We have recently shown that chromosomal regions with gene copy number alterations (CNA) are present in HPV-IS, such as 2p12-p11.2 and 14q32.33 (observed in 100% of PeCa patients), which can also affect the expression of miRNAs located in these regions (Macedo et al., 2020; Silva et al., 2021). These regions were also described in other HPV-associated tumors (Wentzensen et al., 2004; Kumar Gupta and Kumar, 2015; Holmes et al., 2016; Liu et al., 2016). This data shows the close connection of CNAs and miRNA deregulation located in HPV-IS. Altogether, our present data, support that miRNAs located in HPV-IS can directly repress genes related to HPV infection, such as TP53 and RB1, highlighting HPV insertion as one of the factors that trigger epigenetic mechanisms.

In this study, we suggest that the HPV-related genes, TP53 and RB1, are directly down-regulated by 13 miRNAs located in high-risk HPV integration sites, notably for the HPV16 subtype, present in 72% of the PeCa patients studied. Considering that the expression and activity of TP53 and RB1 can be under the control of miRNAs, our findings provide a new understanding of the role of high-risk HPV infection in penile tumorigenesis through an epigenetic mechanism.