Herpes simplex virus inoculation in murine rete testis results in irreversible testicular damage

Abstract

This study aimed to establish the influence of herpes simplex virus (HSV) on testis morphology and germ cell development using a model of ascending urogenital HSV infection in mice. Adult C57BL/6J mice were inoculated with 100 plaque-forming units of HSV1 in rete testis. Viral proteins and HSV DNA were detected from 3 days postinoculation (DPI), while capsids and virions could be visualized at 6 DPI. Infectious activity of HSV was revealed by rapid culture method in testes from 3 to 14 DPI, and virus DNA by PCR – from 3 to 100 DPI. Germ and Sertoli cells were infected during the early stages of the infection, whereas interstitial cells only occasionally contained the virus at 21 and 45 DPI. Microscopic analysis revealed severe degeneration of the germinal epithelium in the infected testes. By 21 DPI, testes became atrophic and most Sertoli cells were destroyed. No testicular regeneration and no spermatozoa in the epididymis were observed at 45 and 100 DPI. From 3 DPI, inflammatory cells accumulated in the interstitium between damaged tubules; a significant increase in the number of CD4⁺, CD8⁺ T lymphocytes and F4/80⁺ cells was observed in the infected testes. This study shows that in the case of HSV retrograde ascent into seminiferous tubules, the acute viral infection results in irreversible atrophy of the germinal epithelium, orchitis and infertility. These results may be used to further study viral orchitis and the influence of HSV on spermatogenesis and male fertility.

According to the European Association of Urology (EAU), infertility affects 15% of reproductive-aged couples with roughly half of these cases being attributed to a male factor (EAU 2010). Accumulating evidence indicates that viral infections contribute to male infertility (Garolla et al. 2013). For example, mumps virus and human immunodeficiency virus (HIV) in patients with AIDS cause inflammatory reactions in the testis and subsequent impairment of spermatogenesis (Dejucq & Jégou 2001). Another widespread virus, herpes simplex virus (HSV-1, 2), has been found in the semen and testes of men, but data on its prevalence can vary across studies (Wald et al. 1999; Kapranos et al. 2003; Bezold et al. 2007). The presence of HSV-1, 2 in semen has been associated with a decrease in sperm concentration and actively motile spermatozoa, as well as an increased number of degenerating and immature germ cells, and abnormal spermatozoa (Kapranos et al. 2003; Bezold et al. 2007; Bocharova et al. 2007; Klimova et al. 2010). Moreover, a significant association has been found between detection of HSV in semen and infertility (Bocharova et al. 2007; Klimova et al. 2010). HSV and another member of the Herpesviridae, human cytomegalovirus (CMV), were found to replicate and cause damage to somatic and germ cells in testis organotypic culture (Naumenko et al. 2011a,b2011b). Despite the significance of these observations, it remains unclear whether HSV can cause testicular infection in vivo. A natural route of HSV entry into the testis is also yet to be established. It has been shown that HSV does not infect testes via the hematogenous route (Burgos et al. 2005). Additionally, in the case of interstitial injection, murine CMV has been detected in endothelial and Leydig cells, but not inside the seminiferous tubules. It is possible that Sertoli cells, by the production of anti-viral factors or by their junction system, prevent herpes virus occurrence in the seminiferous tubules (Tebourbi et al. 2001). The lack of a suitable animal model has hampered the investigations in this field.

Here, we report the development of a mouse model of testicular HSV infection. In this model, the virus is injected into seminiferous tubules of adult mice via rete testis, mimicking the retrograde path of the virus from male urogenital tract into the testis. The aim was to characterize testicular HSV infection and clarify the influence of HSV on testis morphology and spermatogenesis.

Materials and methods

Animals

Adult male C57Bl/6J mice (weighing 21–25 g) of 8 weeks of age were kept at 22 °C with 12-h light/12-h dark schedule and fed with standard food pellets and water ad libitum. A total of 75 mice were used.

Ethical approval

The study was approved by the local Ethics Committee of the Ivanovsky Institute of Virology of Ministry of Health of Russian Federation.

Virus

HSV-1 (strain F) was provided by the Russian Federation State Collection of Viruses at the Ivanovsky Institute of Virology (Ministry of Health, Russian Federation). HSV was propagated and titrated by plaque assay using Vero cells. Vero cells were grown in Eagle's minimal essential medium (MEM) with Earle's BSS (PanEco, Moscow, Russia) supplemented with 10% foetal bovine serum (FBS) (Biolot, St. Petersburg, Russia) and gentamicin (PanEco).

Inoculation of testes with HSV

Mice were anaesthetized with chloral hydrate (Sigma-Aldrich, St Louis, MO, USA) (400 mg/kg, i.p.). The inoculation was based on the introduction of fluid via the rete testis, as previously described (Ogawa et al. 1997). In brief, a small incision was made in the lower abdominal and muscle wall. Testes and epididymides were withdrawn from the body cavity. Under a dissecting microscope, the efferent ducts were identified and isolated. One hundred plaque-forming units (PFU) of HSV in 15 μl MEM containing 0.02% trypan blue (Biolot) were microinjected into the efferent ducts of each testis and flowed through the rete testis to fill the seminiferous tubules. Approximately 50–70% of the tubules were filled with viral solution as determined by trypan blue. Control mice were injected with the same volume of MEM. Mice were killed at 2, 3, 6, 10, 14, 21, 45 and 100 days postinfection (DPI). Testes, epididymides, kidneys, livers and brains were examined.

Analysis of virus

Rapid culture method (RCM)

RCM was used for the detection of HSV infectious activity in the testes. The testes from infected (3, 6, 10, 14, 21, 45 and 100 DPI) and mock-infected mice were dissociated in MEM. Testis cell suspensions were added to confluent monolayers of Vero cells and incubated for 1 h at 37 °C in an atmosphere of 95% air/5% CO₂. The cells were washed in serum-free culture medium, incubated for 48 h in Eagle's MEM (plus Earle's BSS with 2% FBS), washed twice in phosphate-buffered saline (PBS) and fixed in cold acetone. HSV was identified by immunofluorescent staining with monoclonal antibody (Mab) against HSV-1 gB protein (Klimova et al. 1999).

In situ hybridization

Testes at 2, 3, 6 and 45 DPI were fixed in 4% paraformaldehyde solution (PFA) for 24 h and embedded in paraffin. Histological sections (4 μm) were made from each testis. Sections were deparaffinized and rehydrated with xylene and graded ethanol, then treated with 3% hydrogen peroxide solution for 15 min, proteinase K (Dako, Glostrup, Denmark) for 1.5 min, glycine/PBS (2 mg/ml) for 5 min, 4% PFA for 5 min, glycine/PBS (2 mg/ml) for 5 min, dehydrated in graded ethanol and air-dried. Hybridization buffer with HSV-biotinylated DNA probe (1 μg/ml, Enzo Life Sciences, Farmingdale, NY, USA) was then added to the sections, which were covered with coverslips, heated for 5 min at 95 °C and hybridized for 18–20 h at 37 °C. For colorimetric detection after hybridization, UltraSensitive Enhanced Hrp-DAB in situ detection system (Enzo) was used according to the manufacturer's protocol. The sections were analysed using light microscopy (Model BZ-9000, Keyence, Osaka, Japan).

Immunofluorescent analysis

Testes at 3, 6, 10, 14, 21 and 45 DPI were embedded in Tissue-Tec OTC compound (EMS, Hatfield, PA, USA), snap-frozen in liquid nitrogen and stored at −80 °C. Frozen sections (4 μm) were made from each testis with an interval of approximately 200 μm between sections. For confocal microscopy, 25-μm sections were made. Frozen tissue sections were fixed with 4% PFA and permeabilized with 1% Triton X-100. Tissue sections were incubated with 3% bovine serum albumin (BSA) for 30 min at 37 °C, followed by incubation with primary antibodies for 1 h at 37 °C. Rinsed samples were incubated with secondary antibodies for 30 min at 37 °C in the dark, counterstained with DAPI (Sigma-Aldrich) or propidium iodide (PI) (Sigma-Aldrich) and mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA). Mouse Mab against HSV1 gB protein, rabbit polyclonal anti-HSV1 antibody (Abcam, Cambridge, UK), rabbit polyclonal anti-Wilms tumour 1 protein (WT1) antibody (LSBio, Seattle, WA, USA) were used as primary antibodies. Anti-mouse FITC-conjugated (Dako), anti-rabbit FITC-and Alexa Fluor 594-conjugated (Jackson ImmunoResearch, West Grove, PA, USA) antibodies were used as secondary antibodies. For gB staining, before incubation with the primary antibody, sections were incubated in Unconjugated AffiniPure Fab Fragment Goat Anti-Mouse IgG (H+L) (Jackson ImmunoResearch) for 1 h at 37 °C. Sections were analysed by fluorescence microscopy (Model DMRXA2, Leica Microsystems, Wetzlar, Germany). The total number of tubules and the number of gB⁺ and Wt1⁺ tubules were recorded for three sections from each testis. Tissue sections (25 μm) were examined under a confocal microscope (Model TCS SP5 STED, Leica).

Electron and immunoelectron microscopy

For electron microscopy (EM), testes and epididymides of infected and mock-infected animals were fixed in 2.5% glutaraldehyde in 0.1 M buffered sodium cacodylate (pH 7.4) for 24 h at 4 °C, postfixed in 1% OsO₄ in 0.1 M sodium cacodylate buffer at room temperature for 40 min, dehydrated in gradual ethanol concentrations and, subsequently, submitted to progressive impregnations in Epon resin (Sigma-Aldrich). Polymerization was carried out at 60 °C for 48 h. Ultrathin sections were cut in an ultratome III (LKB, Bromma, Sweden).

For immunoelectron microscopy (IEM), fixation was performed with 0.1% glutaraldehyde and 4% PFA for 40 min. Dehydrated samples were embedded in LR White resin (Sigma-Aldrich). Sections of selected areas were cut and mounted on nickel grids. Ultrathin sections were incubated for 1 h in 5% BSA in PBS to block non-specific binding. Rabbit polyclonal anti-HSV1 antibody (Abcam) served as primary antibody during overnight incubation at 4 °C. As secondary antibody, 10-nm gold particle-conjugated donkey anti-rabbit polyclonal antibody (Abcam) was used.

Both Epon and LR White sections were stained with uranyl acetate and lead citrate (Sigma-Aldrich) and examined in a JEM-100 S electron microscope (JEOL, Tokyo, Japan) at 80 kV.

PCR

DNA was extracted from the testes at 3, 6, 10, 14, 21, 45 and 100 DPI and from kidneys, livers and brains at 45 and 100 DPI, using the protocol of the DNA-sorb-B Amplisense DNA extraction kit (InterLabService, Moscow, Russia) and real-time PCR performed using the Amplisense HSV I, II-FL kit (InterLabService) and Rotor-Gene Q (Qiagen, Hilden, Germany), according to the manufacturer's protocol. Calculation of Ct and preparation of standard curve were performed by Rotor-Gene Operating Software, version 1.8 (Corbett Research, Doncaster, Australia).

Histology

Testes at 3, 6, 14, 21, 45 and 100 DPI were fixed with modified Davidson's fluid (mDF) for 24 h and embedded in paraffin. Histological sections (7 μm) were made from each of the testis with an interval of 200 μm between sections. All sections were stained with haematoxylin and eosin and examined under light microscopy (Model BZ-9000, Keyence). The total number of seminiferous tubules and the number of normal tubules were recorded for three histological sections from each testis.

Flow cytometric analysis

For quantitative analysis of leucocytes in the testis interstitium, testes at 10 and 21 DPI were decapsulated and incubated with type IV collagenase (1 mg/1.5 ml, Sigma-Aldrich) at 37°C for 15 min. Collagenase was inactivated and seminiferous tubules were allowed to settle. Supernatant containing testicular interstitial cells was washed with PBS, and red blood cells were depleted by osmotic lysis with ammonium chloride (160 mM NH₄Cl, 170 mM Tris–HCl, pH 7.2). Cells were washed, centrifuged, stained with CD8a-PE, CD4-PE and F4/80-PE antibodies (Biolegend, San Diego, CA, USA) and analysed by flow cytometry (CellLabQuanta SC, Beckman Coulter, Brea, CA, USA). The absolute number of positive cells per testis was calculated from percentages obtained by flow cytometric analysis and the total number of interstitial cells.

Statistical analysis

Results were expressed as mean ± SEM. Comparisons between groups were assessed by the nonparametric Mann–Whitney test, and P < 0.05 was considered significant.

Results

Infectious activity of HSV

Single cells and typical plaques stained with Mab to gB were observed in Vero cells cultured with all homogenates of testes at 3, 6 and 10 DPI. Viral plaques were rarely detected (1 of 3 males) in testes at 14 DPI and were absent at 21, 45 and 100 DPI. Neither stained cells nor plaques were observed in uninfected control samples.

Intratesticular localization of HSV

By in situ hybridization, viral DNA was first detectable in spermatocytes at 2 and 3 DPI, when slight degenerative changes developed in the germinal epithelium (Figure 1a,b). At these time points, viral signals were also detected in a few round spermatids (Figure 1b), spermatogonia and Sertoli cells (Figure 1c). Stronger signals of HSV DNA appeared in Sertoli cells, spermatogonia, some spermatocytes (Figure 1d) and a few elongated spermatids (Figure 1e) at 6 DPI. At the late period of HSV infection (45 DPI), viral DNA was found in Sertoli cells located in degenerating seminiferous tubules (Figure 1f) and in a small number of interstitial cells.

Figure 1.

In situ hybridization of HSV DNA on testis sections at 3 DPI (a–c), 6 DPI (d, e) and 45 DPI (f). Signals of viral DNA are clearly seen in spermatogonia and Sertoli cells located near the basement membrane of seminiferous tubules, spermatocytes (arrowheads), round spermatids (arrows) and elongated spermatids (asterisks). Sections are counterstained with haematoxylin. Scale bars = 20 μm.

HSV glycoprotein gB⁺ seminiferous tubules were first observed in mice at 3 DPI, with the number of positive tubules increasing up to fourfold by 6 DPI (Figure 2). A few gB⁺ tubules were detected at 10 and 14 DPI, but none could be detected at 21 and 45 DPI. Immunostaining with polyclonal antibody against all major HSV glycoproteins and one core protein (anti-HSV1 antibody) revealed weak signals in a few interstitial cells (Figure 3a) and some seminiferous tubules (Figure 3b) at 21 and 45 DPI.

Figure 2.

Dynamics of testicular HSV infection. Vertical axis indicates the percentage of gB⁺ seminiferous tubules in the testes of infected mice; horizontal axis indicates days postinfection (DPI). The number of gB⁺ tubules was counted in three cryosections of the testis, and the percentage of gB⁺ tubules relative to the total number of tubules examined was calculated. Data are mean ± SEM; n = 3 testes per time point. The largest number of gB⁺ tubules was detected at 6 DPI (*P < 0.05 in relation to other time points).

Figure 3.

Immunofluorescent localization of HSV proteins in the testes of infected mice. Viral proteins (red) are seen in the interstitium (a) and in the seminiferous tubule (b) at 21 DPI (a) and 45 DPI (b). Nuclei are counterstained with DAPI (blue). (c–f) Representative confocal microscopy images of the mouse testis at 6 DPI. Viral proteins (green) are seen in the cytoplasm of individual Sertoli cells (c, d); in the group of Sertoli cells and germ cells near the basement membrane of the seminiferous tubule (e); in all layers of the germinal epithelium (f). Nuclei are counterstained with PI (red). The primary antibody used was rabbit polyclonal anti-HSV1 antibody. Scale bars = 100 μm for (a, b, f); 20 μm for (c); 30 μm for (d); 40 μm for (e).

Using confocal microscopy, we investigated the localization of viral proteins in the germinal epithelium at 6 DPI. In some cases, only individual Sertoli cells were immunostained (Figure 3c, d) but typically groups of Sertoli cells and germ cells near the basement membrane of seminiferous tubules were labelled (Figure 3e). In some cases, HSV proteins were found in all layers of the germinal epithelium (Figure 3f). Positive signals for viral DNA and proteins were never found in the testes of mock-infected animals.

Ultrastructural analysis performed at 6 DPI revealed viral capsids in Sertoli cells (Figure 4a,b), spermatocytes (Figure 4a,c), elongated spermatids (Figure 4d,e), spermatogonia and peritubular cells. Enveloped viral particles were found in intercellular spaces of seminiferous tubules. IEM confirmed HSV presence in infected testes: polyclonal anti-HSV1 antibody immunostained either capsids in nucleus (Figure 4f,h) or dense bodies containing viral proteins in the cytoplasm (Figure 4f,g) of infected cells. Testes from mock-infected animals were found to be free of viral particles.

Figure 4.

EM and IEM detection of HSV in infected testis. (a–e) EM of HSV-infected testis. (a) Basal part of the seminiferous tubule from the infected mouse at 6 DPI. (b, c) Magnified fragments of (a), demonstrating HSV capsids (arrows) in Sertoli cell nucleus (b) and in spermatocyte nucleus (c). (d) Luminal compartment of the seminiferous tubule from the infected animal. (e) Magnified fragment of (d) – the cytoplasmic droplet of the elongating spermatid, containing viral capsids (arrows). (f–h) IEM of the HSV-infected testis using rabbit polyclonal anti-HSV1 antibody. (f) Immunolabelled viral capsids (arrows) in Sertoli cell nucleus and dense bodies (arrowheads) in cytoplasm. (g, h) Magnified fragments of (f). 1 – basement membrane, 2 – interstitium, 3 – Sertoli cell, 4 – spermatocyte, 5 – elongating spermatid. Scale bars = 5 μm for (a, d), 1 μm for (b, c, e); 2 μm for (f); 250 nm for (g, h).

Viral DNA detection by PCR

Real-time PCR was used to assess the presence of HSV DNA in testes. HSV DNA was detected in the testes of all infected animals at all time points. It should be noted that viral DNA was detected in the testes of infected mice at the later stages of infection (21, 45 and 100 DPI), when infectious virus was absent and signals of HSV proteins were rare and weak. To investigate virus spread throughout the body, kidneys, livers and brains of infected animals were also analysed by real-time PCR at 45 and 100 DPI. HSV DNA was found in one of three analysed kidneys, two of three livers and one of three brains at 45 DPI, but was absent in all samples at 100 DPI.

Effect of HSV on testicular morphology and spermatogenesis

All infected mice remained alive and displayed no visible signs of HSV infection such as bristling of fur, slight weakness and loss of movement. However, from 6 DPI, the testes of infected mice showed a variable degree of damage; testes were hyperaemic and smaller than in the mock-infected mice. Accordingly, the weight of testes was found to be significantly decreased compared to the control animals (Table 1). By 21 DPI, testes were covered with a thick layer of fibrous material and atrophy was grossly evident. The infected testes remained atrophic up to 100 DPI. From 3 to 14 DPI, damaged spermatozoa and degenerating immature germ cells were found in the epididymal fluid of infected mice. At 21, 45 and 100 DPI, spermatozoa were absent in the epididymides.

Table 1.

Effect of herpes simplex virus on testis weight

Control	Days postinfection
	3	6	14	21	45	100
87.5 ± 2.5	85.6 ± 3.7	68.7 ± 5.8*	35.5 ± 1.6*	22.9 ± 3.0*	32.0 ± 0.7*	35.2 ± 3.4*

Data are expressed as mean (mg) ± SEM; n = 4–6 testes per time point.

^*P < 0.05 in relation to control.

Microscopic analysis revealed the first testicular lesions as early as 3 DPI. Degenerating germ cells from various stages of spermatogenesis were found in the lumina of some seminiferous tubules (Figure 5a). Disorganization of the germinal epithelium (Figure 5b), epithelial sloughing and arrest of spermatogenesis (Figure 5a) occurred in a few tubules. By 6 to 10 DPI, the degree of testicular damage increased; testes were characterized by a reduction in the thickness of the germinal epithelium (Figure 5c) and severe injury of germ and Sertoli cells (Figure 5d). At 14 and 21 DPI, many tubules turned atrophic, with some containing only Sertoli cells or Sertoli cells with a few degenerating germ cells (Figure 5e). In addition, cases of tubules occupied by cellular debris could be identified (Figure 5f). Partial disappearance of Sertoli cells was observed in many seminiferous tubules (Figure 5g), and in extreme cases, Sertoli cells were completely destroyed. The basement membrane of tubules lacking Sertoli cells appeared to disintegrate, and interstitial cells began to migrate towards the lumen of empty tubules (Figure 5g). At later time points (45 and 100 DPI), the tubules of infected mice were a mixture of tubules containing only Sertoli cells, Sertoli cells with a few germ cells, some tubules with fibrosis and a few normal tubules (Figure 5h). Quantitative analysis of the testicular histopathology confirmed qualitative observations and revealed a dramatic decrease in the number of normal seminiferous tubules between 3 and 6 DPI and between 6 and 14 DPI (Figure 6). No testicular damage was observed in the mock-infected mice (Figure 5i) at all time points.

Figure 5.

Testicular histopathology. (a) Degenerating germ cells in the lumina of seminiferous tubules (asterisks) and the arrest of spermatogenesis (arrow) at 3 DPI. (b) Disorganization of the germinal epithelium at 3 DPI. (c) Reduction in the thickness of the germinal epithelium (below dashed line), seminiferous tubules occupied by degenerating cells (asterisks) at 6 DPI. (d) Damaged Sertoli cells (arrowheads), multinucleated germ cells (arrow) and pyknotic germ cells (P) in abnormal tubules at 6 DPI. (e) Seminiferous tubules containing either only Sertoli cells or Sertoli cells and a few germ cells at 14 DPI. (f) Cellular debris (D) in tubules at 14 DPI. (g) Decrease in the number of Sertoli cells in seminiferous tubules at 21 DPI, arrows indicate margins of disintegrated tubules. (h) Testicular atrophy at 100 DPI, the normal tubule is indicated (asterisk). (i) Normal testicular morphology in mock-infected mice. Sections are stained with haematoxylin and eosin. Scale bars = 100 μm for (a, c, e, f, h, i); 40 μm for (b, g); 20 μm for (d).

Figure 6.

Quantitative analysis of testicular morphology during HSV infection. Vertical axis indicates the percentage of normal seminiferous tubules in the testes of infected mice; horizontal axis indicates days postinfection (DPI). The number of normal tubules (without degenerative changes) was counted in three histological sections of the testis, and the percentage of normal tubules relative to the total number of tubules examined was calculated. Data are mean ± SEM; n = 3 testes per time point. *P < 0.05 in relation to the previous time point.

The influence of HSV infection on Sertoli cells was assessed by immunofluorescent detection of the Sertoli marker protein Wt1. At 14 DPI, many seminiferous tubules contained no Wt1⁺ cells, whereas Wt1⁺ cells could be detected in all tubules of mock-infected testes (Figure 7). Quantitative analysis of Wt1 expression revealed that only 48.8 ± 12.5% of seminiferous tubules from infected mice contained Wt1⁺ cells (P < 0.05).

Figure 7.

Representative images of immunofluorescent staining with the antibody against Sertoli cell marker Wt1 in sections of control mouse testes (a, b) and infected testes at 14 DPI (c, d). Nuclei are stained with DAPI. It is seen that only about half of the seminiferous tubules contain Sertoli cells at 14 DPI. Scale bars = 200 μm.

Development of orchitis

At as early as 3 DPI, degenerating seminiferous tubules were surrounded by a small number of leucocytes (Figure 5b). At 6 DPI, light and electron microscopy revealed how the interstitium of infected testes had been infiltrated by a large number of leucocytes (Figure 8a,b), with macrophages penetrating the basement membrane visible in some cases. Ultrastructural analysis of spermatozoa obtained from epididymides of infected mice at 6 DPI revealed up to 20% of injured gametes inside macrophages (Figure 8c). The number of inflammatory cells reached a maximum at 10 DPI. Inflammation decreased at 21 DPI, and at 45 and 100 DPI, almost no infiltrating cells could be detected.

Figure 8.

Leukocytosis in the pathogenesis of HSV-induced orchitis. (a) Leucocyte infiltration of HSV-infected testis at 6 DPI, light microscopy. (b) Leucocyte infiltration of testicular interstitium at 6 DPI, EM. (c) Several sperm heads inside the macrophage from the epididymis of the infected mouse at 6 DPI, EM. 1 – basement membrane, 2 – interstitium, 3 – macrophage, 4 – spermatozoon. Scale bars = 100 μm for (a), 5 μm for (b, c).

The number of leucocytes in infected testes was determined by flow cytometry. At 10 DPI, a significant increase in the number of T lymphocytes and F4/80⁺ (murine macrophage-specific antigen) cells was observed in infected testes, compared to mock-infected mice (Figure 9a,b). At 21 DPI, the number of CD8⁺ T lymphocytes returned to control levels (Figure 9a,c). Although the numbers of CD4⁺ T lymphocytes and F4/80⁺ cells remained high at 21 DPI, the levels were significantly reduced compared with 14 DPI (Figure 9). The CD4/CD8 ratio changed during the course of HSV infection, from 1.65 in mock-infected mice to 1.29 and 3.76 in infected mice at 10 and 21 DPI respectively.

Figure 9.

Flow cytometric analysis of testicular interstitial cells at 10 and 21 DPI. (a) Representative flow cytometry histograms of CD4, CD8 and F4/80 expression in the testis of mock-infected (control) mice and HSV-infected mice at 10 and 21 DPI. The percentages of stained cells are indicated. (b) Absolute number of T lymphocytes and F4/80⁺ cells in the testes. (c) Absolute number of CD4⁺ and CD8⁺ T lymphocyte subsets in the testes.*P < 0.05 in relation to control, ^#P < 0.05 in relation to 10 DPI.

Discussion

Here, we present the development of a novel animal model of HSV infection by inoculation of seminiferous tubules via rete testis. This technique mimics retrograde ascent of pathogens along the urogenital tract via rete testis into seminiferous tubules, considered a natural route of testicular infection (Bhushan et al. 2009). The technique of rete testis injection was first described by Brinster and colleagues (Brinster & Avarbock 1994; Brinster & Zimmermann 1994) and is now widely used for transplantation of spermatogonial stem cells into seminiferous tubules. To our knowledge, this is the first reported use of this technique for the investigation of virus influence on testis morphology and spermatogenesis.

Our results show that HSV causes acute testicular infection, which reaches its maximum at 6 DPI and decreases significantly as early as 10 DPI. Spermatocytes are the first cells to be infected before HSV spreads to the basement membrane of the seminiferous tubules to infect spermatogonia and Sertoli cells. Although a previous study was unable to identify infected spermatozoa after incubation of semen with HSV in vitro (Pallier et al. 2002), we here provide evidence that HSV can infect haploid germ cells; HSV DNA and viral capsids are detected in a few round and elongated spermatids.

Impermeability of the blood–testis barrier for herpes viruses has been proposed (Tebourbi et al. 2001; Burgos et al. 2005). In this study, HSV was found only in seminiferous tubules and was unable to pass through the Sertoli cell barrier into interstitial tissue at early time points. However, at 21 and 45 DPI, when Sertoli cells had been damaged in the course of HSV infection, viral DNA and proteins were occasionally detected in the intertubular space.

Infectious virus was not recovered from testes after 21 DPI. HSV glycoprotein gB, which is involved in viral cell entry (Spear & Longnecker 2003), was also never observed in infected testes since this time point. However, HSV DNA was detected at the later stage of HSV infection by real-time PCR and hybridization in situ. Moreover, staining with polyclonal antibody against all major HSV glycoproteins and one core protein revealed weak signals in infected testes at 21 and 45 DPI. Taken together, these data suggest that abortive infection, in which a proportion of HSV genes are expressed but no infectious virions are produced (Kinchington et al. 2012), occurred at the later time points.

Despite the use of a very low HSV dose for infection (approximately 10⁻⁵ PFU per testis cell), severe degeneration of the germinal epithelium was achieved. In addition, strong signals of viral DNA and proteins were detected in the damaged tubules, whereas no or weak signals were detected in the intact tubules of infected animals, suggesting that viral replication is a direct factor of testicular damage.

Testicular inflammation during the course of HSV infection could also account for seminiferous tubule degeneration. In the current work, interstitium was infiltrated by a large number of CD4⁺, CD8⁺ T lymphocytes and F4/80⁺ (murine macrophage-specific antigen) cells at the early stage of infection. It is commonly accepted that regulation of spermatogenesis involves cytokines normally associated with inflammatory processes, such as interleukins IL-1a, IL-6, interferons, tumour necrosis factor and activin (O'Bryan & Hedger 2008). These regulators are expressed in the testis under normal, non-inflammatory conditions, and in the course of inflammation, up-regulation of cytokine expression may disrupt germ cell development and play a key role in fertility disorders. Other viruses have been described as causative factors of testicular inflammation; for example orchitis after encephalomyocarditis virus infection has been reported in mice (Yamanouchi-Ueno et al. 2004) and in Sendai virus-infected rats (Melaine et al. 2003).

A major effect of testicular HSV infection is severe Sertoli cell damage. HSV DNA, proteins and viral capsids were detected in Sertoli cells throughout the course of infection, and at 14 DPI, only 48.8% of seminiferous tubules contained Sertoli cells. Sertoli cells support and nourish germ cells, regulate spermatogenesis and form the blood–testis barrier. Because each Sertoli cell can support only a fixed number of germ cells, the number of Sertoli cells is directly related to sperm counts. Therefore, alterations in Sertoli cell function and the number of Sertoli cells may lead to irreversible testicular damage. Indeed, infected testes remained atrophic at 45 and 100 DPI, with no signs of testicular regeneration and no spermatozoa in the epididymides.

In conclusion, our results demonstrate that HSV inoculation of seminiferous tubules via rete testis is a promising model for the study of viral orchitis and the influence of HSV on spermatogenesis and male fertility.

Conflict of interest

The authors declare no conflict of interest.

Funding source

This work was supported by the grant from Russian Foundation for Basic Research (RFBR), Research Project No. 12-04-32258mol_a.