The MRT is comprised of the testis, epididymis, vas deferens, prostate, seminal vesicle, bulbourethral glands and the penile urethra (Figure 1
). Two main functions of the MRT are (1) the production, maturation and protection of spermatozoa from immune aggression and infectious agents and (2) the production of hormones. To achieve these goals, the MRT has evolved a series of structural and functional features which are far beyond the scope of the present review. For those seeking for a deeper understanding of these features, we recommend reading comprehensive reviews found in the literature [48
The testis may serve as a reservoir for viruses after their hematogenous dissemination [3
] and even after they have been cleared from the blood. This persistence has been clearly demonstrated for HIV in humans [52
]. Persistence in the testis requires that viruses cross the endothelial layer, reach the interstitial space, infect or ideally replicate in cells present in this location or in the seminiferous tubule wall, and evade or modulate immune responses. Viruses can alter the blood testicular barrier (BTB) making it leakier or infect Sertoli cells to reach the adluminal face of the seminiferous tubules either free or inside the motile spermatozoa. Thus, viruses must find a portal of entry into the testis, invade cells and modulate immune responses to favor their replication or persistence in the MRT.
The detection of ZIKV RNA in semen was a clear demonstration that this virus could reach the MRT, but the exact mechanisms of invasion, replication and persistence were unknown at that time [27
]. Understanding how ZIKV passes from the blood into the MRT and its interactions with different cells types might help in the development of vaccines and antiviral drugs that would prevent sexual transmission and the deleterious effects in pregnancy [53
]. Moreover, ZIKV poses a risk to semen banks and to human fertilization strategies as it can be shed persistently for months in semen of men who are infected by the virus [39
]. An aggravating situation in this context is that sperm preparation procedures that reduce the risk of HIV and HCV transmission are not effective against ZIKV [55
Several mechanisms can lead to altered or impaired sperm production, subfertility or infertility because of viral infections in the MRT. Reduced generation of germ cells, germ cell apoptosis, sperm cell abnormalities, low sperm cell generation, altered sperm cell capacitation, inflammation resulting in immune cell infiltration, activation of autoimmune responses, accessory glands infection and duct obstruction have been reported [56
]. Unfortunately, most of the studies on ZIKV biology in the MRT have been carried out in animal models, so care must be taken not to assume that the same interactions and outcomes translate to the human MRT. We will attempt to focus on studies carried on human patients and on human cell cultures in vitro to have a clearer picture of the potential effects of this virus on male reproductive health.
3.2. Zika Virus in the Epididymis
The human epididymal epithelial lining (Figure 2
) expresses the AXL co-receptor for ZIKV [92
] and this could result in the infection of these cells by free ZIKV particles arriving from the seminiferous tubules, as has been already detected in mouse models [99
]. Viral replication in the epididymal epithelium would amplify the infection in the MRT and could be a plausible explanation for the high viral loads detected in semen. In one study, the viral load in semen was 100,000 times higher (8.6 × 1010
copies/mL) than that in whole blood [28
As described above in the testis, the virus could also reach the epididymis from the blood and infected interstitial cells such as macrophages and DCs and gain access to the spermatozoa in the epididymal lumen. Here, the virus could adsorb to the membranes of spermatozoa or infect these cells in the acute phase of the disease when spermatozoa traffic for approximately 12 days in the lumen of the epididymis [55
The DC population in the normal epididymal intertubular space is primarily composed of immature plasmacytoid and myeloid DCs [100
]. These cells change to mature phenotypes upon persistent inflammation of the epididymis and could affect the integrity and permeability of the fragile BEB (Blood–Epididymis Barrier) promoting a Th17 inflammatory response during the course of ZIKV infection, as has been demonstrated for other infectious agents [101
]. As a result of inflammation, DCs could meet self-antigens present on spermatozoa, capture and present them to auto-reactive T cells increasing the chances for fertility disorders. This hypothesis has yet to be tested in long term follow up studies of ZIKV-infected males. The development of new methods with higher sensitivity to detect ZIKV RNA in the epididymis of mouse will probably contribute to further studies of the interactions of ZIKV in this organ in humans [102
3.3. Zika Virus in the Prostate Gland and in the Seminal Vesicle
Inflammation and infection of the testis and epididymis seem to be more relevant for male reproduction than inflammation or infection of accessory glands [83
]. However, in non-human primates ZIKV persists in the prostate and seminal vesicle after viremia has been resolved [103
]. In mouse models, ZIKV does not infect the prostate and seminal vesicle [104
], possibly because AXL or other flavivirus receptors are not present on the surface of the cells in these two glands.
Evidence that these anatomical sites might be permissive to ZIKV comes from two studies of ZIKV shedding in vasectomized men. In these cases, the surgical procedure cuts and seals the ductus deferens that carries spermatozoa from the testis and epididymis towards these glands. Therefore, the source of the virus would have to be ahead of the surgical incision in the prostate, seminal vesicles or bulbourethral glands. One cannot rule out a possible role of urethral epithelium in ZIKV replication in the MRT.
In the first case report, infectious ZIKV particles were detected in the ejaculate of a vasectomized man, 69 days after symptom onset with probable transmission to the sexual partner [105
]. In the second case report, ZIKV RNA was detected in the ejaculate of a vasectomized man, up to 77 days after symptom onset and the virus was proven to be infectious in cell culture up to 21 days after symptom onset [60
]. No abnormal findings could be noted on digital palpation of the prostate of the vasectomized men and the level of the Prostate Specific Antigen (PSA) was normal suggesting that ZIKV infection, if present in the prostate of this patient, did not significantly alter the inflammatory status or the physiology of the gland. The microscopical analyses of the ejaculate revealed the absence of spermatocytes indicating that the surgical procedure had been successful. Round cells, most probably leukocytes, were detected. Some of these round cells were peroxidase positive (0.05 × 106
/mL) suggestive of neutrophil infiltration which has also been described in Rhesus
macaques infected by ZIKV [106
]. These round cells could be harboring ZIKV particles attached to their surface or carrying the virus inside their cytoplasm, but this was not determined in this study.
It is important to highlight that prostatitis—a disease commonly associated with bacterial infections—has been reported in male patients infected with ZIKV and DENV [107
] and that these viral agents must be remembered in the differential diagnosis of this disease in endemic areas and in travelers returning from countries where these viruses circulate. This symptom will prompt the patient to seek medical assistance creating an opportunity for sampling for ZIKV detection. Hematospermia, microhematospermia [111
] and perineal pain are also indicative of prostatitis and have been reported in males with ZIKV infection [27
]. However, most male infections course without this complaint [27
] suggesting that only a small number of male patients will probably have ZIKV in the prostate or that specific host–virus interactions might be involved in the control of the infection in the prostate. It would be advisable in such cases to needle biopsy the prostate to determine the inflammatory status of the gland and to test for the presence of Zika virus antigens.
A metastatic prostatic adenocarcinoma cell line (LNCaP) proved to be permissive to ZIKV infection [93
] with greater expression of the viral NS1 protein than seen in a testicular cell line (833KE). Viral load and the number of infected cells significantly increased from day 1 to day 5 after inoculation. No cytopathic effects were noticed after infection of these cells with a contemporary or historical ZIKV strain pointing to host cell survival and persistent infection potential of these cells. Altogether, these results suggest that the prostate might be a more efficient site for ZIKV amplification and persistence than initially thought. Additional studies are needed to determine the precise potential of this gland to serve as a viral reservoir in infected male patients.
3.4. Zika Virus in Semen
Semen is the vehicle that carries ZIKV particles from the MRT to the FRT. Viruses can travel as free particles in seminal plasma, adsorbed to cell membranes or inside the cytoplasm of leukocytes and spermatozoa. Many case reports of sexual transmission of ZIKV from male-to-female and from male-to-male have been published in a clear demonstration of the infectivity of ZIKV particles present in human semen [17
Shedding of ZIKV particles in semen is variable between individuals [98
]. In a prospective cohort in Puerto Rico, shedding was detected in 73% of symptomatic men and started as early as seven days post-infection [55
]. Shedding can occur concomitantly with virus presence in urine and blood or after viral clearance for these fluids [22
]. Viral shedding can also be intermittent or persistent [114
] for up to 188 days (six months) with viral load reaching 50,000 copies/mL in the first two weeks and ranging from 1000 to 10,000 copies/mL at the end of this period [54
]. The final shedding date indicates elimination of the virus by the immune system [115
] but might as well represent one more episode of intermittent shedding.
One interesting aspect of ZIKV shedding in semen is that asymptomatic patients can sometimes excrete longer (68 days) than symptomatic ones [113
]. For that reason, men diagnosed with ZIKV infection should avoid attempting conception for at least six months and keep systematic use of condoms in all sexual intercourses in this period. The high rate of asymptomatic infection following ZIKV infection, estimated in 80% of infected individuals [116
], leads to a gross underestimation of the risk for sexual transmission.
ZIKV RNA can be detected in the seminal plasma (the liquid supernatant after centrifugation of semen) and in the cellular components of semen. Interestingly, in all negative semen samples no RNA could be detected in the cellular pellets. Nonetheless, viral RNA was detected in two percent of seminal plasma negative samples. Furthermore, in 16% of the seminal plasma positive samples, no RNA could be detected in the cells [55
]. These results confirm that ZIKV can travel free in seminal plasma or attached/inside the cells present in semen. Similar results were reported elsewhere [117
]. Studies must be conducted to determine if free virus particles can cause infection in the FRT or if the virus must be transmitted inside spermatozoa during fertilization.
Spermatozoa were purified by gradient centrifugation and ZIKV RNA was found in 25% of the samples, all of which came from shedders with high viral loads (>5 log copies per mL) in seminal plasma suggesting that higher viral loads lead to higher infection rates in spermatozoa [55
]. Moreover, 64% of the motile spermatozoa samples tested positive for ZIKV RNA and the virus from one of the samples could be cultured in Vero E6 cells. Other researchers detected viral RNA only in the cellular fraction of semen and were unable to culture the virus [54
]. These results clearly demonstrate that spermatozoa from infected males maintain their ability to swim and can carry infectious virus particles to the FRT during fertilization which could lead to ZIKV congenital syndrome in a percentage of the infected fetuses [118
]. Unfortunately, one cannot be sure of the precise localization of ZIKV particles in this study, either adsorbed to or inside spermatozoa.
The presence of ZIKV antigens inside the head of spermatozoa was unambiguously demonstrated using immunohistochemical techniques associated with confocal and emission depletion microscopy [98
]. At least 3.52% of the spermatozoa carried the virus and the authors argued that the virus might have originated from Sertoli cells which are known to have the AXL co-receptor. However, in mice AXL is not essential for ZIKV infection, clinical signs, viral load, tropism or histological changes in major organs [119
]. Thus, the virus could have infected germ cells, spermatocytes or spermatozoa in the testis or during its traffic and storage in the epididymis as hypothesized earlier in this article.
Much more concern arises when one considers the presence of infective ZIKV particles in spermatozoa [55
] because they can directly deliver the virus into the cytoplasm of the oocyte during fertilization [120
] which can lead to miscarriage or congenital ZIKV syndrome. Intravaginal exposure to ZIKV in mouse models can lead to fetal brain infection [121
]. There is a possibility that during outbreaks, such as the one in French Polynesia and the more recent one in the Americas, this mode of transmission was responsible for some of the microcephaly cases reported. In Brazil, where the use of condom is not universal, almost 3000 cases of congenital ZIKV syndrome have been reported and confirmed by the Ministry of Health [122
]. With increased knowledge and information, men could plan to avoid unprotected sexual intercourses, semen donation or planned pregnancy for at least six months after disease onset.
Some negative effects of Zika virus presence in the MRT have been reported [55
]. Shedders had significantly lower semen volume, lower total sperm count and lower total motile sperm at least in one occasion and a significant increase in spermatozoa anomalies in two occasions. One can only infer that the infection of the MRT interferes, at least temporarily, with the level of spermatogenesis, sperm morphology, sperm fitness for fertilization and affects the production of seminal plasma by the annexed glands. Further studies are urgently needed to elucidate the long-term consequences of this infection in human males.
In terms of ZIKV epidemiology, shedding of infective particles in semen is what matters the most not only because of its possible adverse effects on the fetus, but also because it allows the transmission of the virus in areas where the mosquito vectors do not thrive. Inoculation of virus particles from semen samples in cell cultures in vitro and checking for increasing viral load has been used to determine viral viability. Using this approach, it was demonstrated that infective particles were shed up to 69 days after symptom onset [105
] and sexual transmission was probable in one of these cases [105
]. Sexual transmission has been reported to occur as late as 41 days after the onset of illness in an index case [23
The detection of infectious virus particles has not always been successful and might be due to technical issues. Negative cultures might reflect the absence of infective particles, technical inadequacy, insufficient technical skills, viral neutralization, viral persistence inside semen cells, viral destruction by processing, among other possibilities. A good approach is to free all possible ZIKV particles present in the cell fraction before inoculation onto the cell layer in vials.
It is known that the semen can influence the vaginal environment, the female immune responses, the establishment of a successful pregnancy and viral transmission [123
]. Since infected cells in the MRT produce several cytokines and chemokines it would be interesting to measure their concentrations in semen samples of ZIKV-infected males to determine their potential effects on the FRT and pregnancy outcomes.