Human papillomavirus proteins are found in peripheral blood and semen Cd20+ and Cd56+ cells during Hpv-16 semen infection

Once considered a sexually transmitted virus principally affecting females, human papillomavirus (HPV) currently represents an important factor responsible for both couple infertility and cancer development, as demonstrated by recent experimental and epidemiological data [1]. Cervical cancer is the most common cause of mortality related to human papillomavirus [1] but HPV infection also raises important concerns for men. In fact, HPV has been associated with most male anal cancers (approximately 90%) and a subset of penile (50%) and oral cancers (10–72%) [2]. Additionally, the presence of HPV has been widely reported in semen [3], where it has been associated with impairment of sperm motility [4], supporting the hypothesis that HPV infection is an emerging risk factor for male infertility. Previous studies from our group have documented that HPV can be detected in different cell populations within human semen. By the use of fluorescence in situ hybridization (FISH) analysis, HPV-DNA has been detected on both sperm and non-spermatozoal cells [5‐7]. These latter include Sertoli cells, squamous epithelial cells and ‘round cells’, such as spermatozoal precursors, erythrocytes and leukocytes [8]. While the presence of Sertoli, immature sperm and epithelial cells is generally considered the result of natural cell turnover and epithelial exfoliation, the presence of increased levels of sperm leukocytes has been traditionally associated with concomitant inflammation or infection of the genito-urinary tract [9]. However, the detection of HPV-DNA in non-spermatozoal cells has not been further investigated. Recent findings from our lab showed that male patients with HPV 16 in semen had a round shaped subpopulation of cells positive for FISH-HPV, not morphologically ascribable to spermatozoa or epithelial cells. A cytological analysis with May Grunwald-Giemsa staining suggested these cells to be of myeloid origin and subsequent cytochemical analysis confirmed the population to be CD45-positive leukocytes. Despite no clear tropism of HPV to white blood cells has been demonstrated, this observation is in agreement with previous reports showing papillomavirus-DNA associated with circulating leukocytes in both human and animal models [10‐12]. The possible access of papillomavirus to peripheral blood leukocytes has been very recently strenghtened by experimental data from animal models, showing that BPV-DNA and transcripts can be found in blood cells a few days after intradermal inoculation of horses with BPV1 [13]. Moreover in humans, several studies conducted in female patients with cervical cancers, have shown that HPV DNA can be found in peripheral blood, sera, plasma and arterial cord blood [14, 15]. In these studies, authors concluded that HPV detected in blood cells of cancer patients was not, or not-only, subtended by metastatic cells released at a later stage of disease but was also associated to blood mononuclear cells. Moreover, circulating HPV DNA was detected also in healthy blood donors [11]. However, the vast majority of these studies did not show HPV inside blood cells. Most or all of the studies merely focused on the expression of HPV DNA in blood, or on the outside surface of cells [11]. This latter aspect, prompted us to consider HPV infection as a triggering event for virus diffusion into blood stream, even in the absence of evident HPV related diseases. The aim of the present study was to better characterize HPV infected round cells in semen and to evaluate the possible presence of HPV infected cells in blood.

The seminal parameters of the HPV16-infected patients and control subjects assessed are reported in Table 1. We observed lower sperm motility and an increased frequency of positive SperMar test in HPV16-positive samples. No significant differences in other seminal parameters were observed between the two groups. FISH analysis for HPV, performed on patient semen samples, showed a 26.2 ± 5.9 mean percentage of HPV-DNA bearing sperm. Among non-spermatozoal cells, we observed a mean of 19.3 ± 8.4% HPV bearing cells. The morphological features of these cells were clearly visible by microscopy and were variable, ranging from large-irregular shaped cells with disrupted nuclear margins of likely epithelial origin, to round-regular shaped nucleated cells with limited cytoplasm, possibly of the myeloid lineage (Figure 1a I and II). No infected cells were observed in semen samples from controls.

The results of this study have identified, for the first time, the presence of HPV16 in CD45+ leucocytes in the semen of infected patients. Furthermore, HPV was also detected in a very restricted subpopulation of circulating PBMC bearing a similar immunophenotype. Although these findings are in contrast with the known tropism of HPV for epithelial cells [20], other authors have reported HPV spreading both from cancers and infected sites [10, 20]. Pao et al. [21], first reported the presence of different types of HPV [6, 11, 16, 19] in peripheral blood mononuclear cells from female patients affected by HPV infections of the genito-urinary tract [21]. In addition, in a recent study performed on HIV patients, 11 of 57 blood samples were positive for HPV-16 DNA [14]. The authors stated that the HPV genome detected in positive samples existed as an episomal form, albeit at a low DNA copy number. Moreover, in another study assessing a cohort of healthy Australian blood donors, the presence of HPV-DNA in peripheral leukocytes was reported in approximately 8% of subjects [11]. Presence of circulating HPV can be explained by the immune response. Local and systemic viral infections are controlled by innate and adaptive immune mechanisms, where both humoral responses, which prevent disease recurrence [22], and cell-mediated immune response are triggered. In particular for cell-mediated immunity, NK cells together with CD1d-restricted NK-T cells (iNKT cells) are the main cell populations of the innate immune system responsible for the clearance of HPV infection [23]. NK cells constitutively express IFNγ, granzyme and perforin. They are ‘primed’ to initiate an anti-viral immune response by detecting decreased MHC-I expression on infected cells, in the case of HPV, due to repression by the E5 and E7 viral-oncoproteins at specific sites of the MHC-I processing and presentation pathway [22‐25]. In parallel, the activation of antigen presenting cells (APC) and subsequent antigen presentation in the draining lymph nodes is necessary for the initiation of a primary adaptive response to an infecting agent. In the case of HPV, the local APCs responsible are the Langherhans cells of the epidermis, dendritic cells present in the dermis, and occasionally CD20+ B-lymphocytes [25‐28]. Data from animal models of papillomavirus infection suggest that APCs take up virus-like-particles through the FcγRIII receptor [29], degrade antigen within proteasomes and incorporate it into MHC II on the cell surface. Thereafter, APCs trigger the T cell-mediated adaptive response of CD4+ and CD8+ populations [29, 30], in part by the release of a typical pattern of inflammatory cytokines such as interleukin-1β, interleukin-12, tumour necrosis factor-α, and interleukin-6 [31]. In this study, the presence of human papillomavirus markers in a subset of semen leukocytes from a cohort of HPV-genitourinary infected patients has been documented. Both HPV-16 E6 and L1 proteins co-localized with the viral DNA. Notably, cells displaying molecular hallmarks of HPV were frequently found to be CD20+ B-lymphocytes and CD56+ natural killer-like (NK-like) cells, two leukocyte populations recruited during viral infection. In addition, in peripheral blood of 25% of patients, the presence of HPV-bearing cells was found by FISH and further confirmed by single-step PCR. In agreement with the results observed in semen, DNA and protein markers of HPV in peripheral blood were mainly associated with CD20+ and CD56+ cells. It is possible that signals for viral DNA and L1-protein could be likely due to the endocytic uptake of virions, given that they are mostly observed as punctate patterns in the cytoplasm. However, if this was the case, virion endocytosis does not explain the detection of E6 protein and it is highly unlikely that the observed E6 signal is an artefact since it is only seen in HPV DNA positive cells. These evidences raise the hypothesis that NK-like cells and B lymphocytes represent possible targets of HPV, in according to previous studies [32, 33]. In fact on a side, Renoux et al. [32] showed by in vitro studies that cytotoxic activity and cytokine production by NK cells appeared to be linked to rapid HPV-VLP entry into these cells by macropinocytosis. Since CD16 blockade inhibited this process, the authors concluded that CD16 is necessary for HPV–VLP internalization [32], even if a role for CD16 as a primary receptor for HPV infection has not yet been demonstrated. On the other side, the competency of HPV to be uptaken by B lymphocytes cannot be excluded. In fact, this possibility is supported by the fact that heparan-sulfate proteoglycans, theorised to potentially be one of the primary receptors involved in HPV entry, are expressed by B lymphocytes and are a requirement for normal cell maturation, differentiation and function [33]. However, recent studies suggested that HPV entry into target cells is actually the result of a concerted interaction between virus-capside proteins and more than one receptor on cell surface like α6-integrin or annexin A2-heterotetramer [34, 35]. Thus further investigation will be necessary to clarify this aspect.

It has been postulated that HPV in polymorphonuclear cells may be involved in vertical transmission of the virus, following the detection of HPV-DNA in reproductive and placental cells, as well as in virgins, infants and children [36, 37]. Indeed, since HPV-DNA has been detected in amniotic fluid [37], placenta and the umbilical cord [36], the hematogenous route of chorionic and placental tissue infection and subsequent spread to amniotic cells and possible ingestion by the fetus has been hypothesized [36, 37]. Since sporadic fetal brain malformations (eg, dysplasias) could be detected as early as 24 weeks gestation [38], it has been suggested the transplacental spread of HPV16 as a plausible mechanism for entry into the brain during early fetal cortical development, even in the absence of overt clinical infection. Interestingly, HPV-16 protein expression has very recently been reported in patients affected by focal cortical dysplasia type IIB (FCDIIB) [39]. In particular, HPV16-E6 protein was robustly expressed in cerebral cortex specimens of all FCDIIB patients examined, most notably in the cytoplasm of balloon cells (BC). E6 protein signals were not detectable in regions without BC or in control tissue specimens. Moreover, E6 expression in BC displayed a strong association with components of the mammalian target of rapamycin complex 1-signalling pathway, whose abnormal activation is linked to FCDIIB etiology [40]. Taken together, these observations pose an open question regarding the meaning of HPV markers found in distinct blood leukocyte populations as the result of cell-mediated immunity or the involvement in virus spreading from the original site of infection.

In addition to the low number of patients enrolled, the main drawback of this study is represented by the extremely low percentage of HPV positive leukocytes recovered in seminal fluid and peripheral blood. A possible reason for these findings is that semen immune cells can become unproductively infected with HPV16 when trafficking through the seminal compartment in HPV16-infected men, and these cells may traffic back into circulation and be detected at low frequency in peripheral blood. This condition limited our ability to isolate HPV bearing leukocytes and further evaluate their infectiveness. Thus a mechanistic explanation about how virus enters circulating mononuclear cells remains an open question. Moreover, the assessment of HPV DNA on peripheral blood performed by INNO LiPA was not able to confirm the results obtained by FISH and Single-step PCR analysis (data not shown). A possible explanation of this discrepancy may reside in the different method sensitiveness.

In conclusion, this is the first study reporting the presence of HPV-antigens together with HPV-DNA in circulating PBMC from male patients with HPV-16 infection of the genitourinary-tract. This evidence raises major concerns about the risk of developing cancers to distal organs.