Hum. Reprod. Advance Access published May 20, 2015 Human Reproduction, Vol.0, No.0 pp. 1 –9, 2015 doi:10.1093/humrep/dev112
ORIGINAL ARTICLE Reproductive biology
Epididymitis: ascending infection restricted by segmental boundaries A. Stammler 1, T. Hau 1, S. Bhushan2, A. Meinhardt 2, D. Jonigk 3, T. Lippmann 3, A. Pilatz4, I. Schneider-Hu¨ther 1, and R. Middendorff 1,*
*Correspondence address. E-mail: [email protected]
Submitted on March 19, 2015; resubmitted on April 17, 2015; accepted on April 24, 2015
study question: Is the regionalization of epididymitis related to epididymal segmentation? summary answer: We show for the first time that luminal ascent of bacteria is strictly gated by epididymal segment boundaries, involving ductal constriction adjacent to the infected area.
what is known already: The epididymal duct is a continuous, unbranched tube, coiled into segments that are divided by connective tissue septa. Sonographic analysis indicates that swelling associated with epididymitis is predominant in the cauda region. Epididymal segmentation has never been investigated in the context of pathological alterations. study design, size, and duration: We analyzed segment-specific changes in the epididymal duct in a mouse model and in men. In the mouse epididymitis model (3 days post-infection, injection of bacteria into the lumen of the vas deferens), two Escherichia coli strains were tested: a uropathogenic strain CFT073 (UPEC, n ¼ 7) and a fecal non-pathogenic strain NPEC470 (NPEC, n ¼ 5). Two control groups: phosphate-buffered saline, sham-treated animals (n ¼ 4) and untreated mice (n ¼ 8). In addition, segmentation was verified by ex vivo injection of dye into the interstitial spaces of untreated mouse epididymides. Histological findings were compared with specimens from epididymitis patients (n ¼ 10, age range 14–78, median 60 years) who underwent surgical intervention; control: samples from patients without epididymitis (n ¼ 16, age range 38–87, median 73 years). participants/materials, setting, and methods: We investigated the ascending infections by detailed histological analysis in correlation with local infection status in a mouse epididymitis model. As a proof of concept, rare patient material from two archives was analyzed: epididymides from patients who underwent surgical intervention for persisting epididymitis, and for control, histologically normal epididymides from men who underwent orchiectomy for therapy of prostatic carcinoma. main results and the role of chance: Luminal ascent of E. coli in mice was strictly gated by epididymal segment boundaries. In the mouse model, both strains of E. coli were detected exclusively in the distal cauda segment associated with damage of the epithelium and muscle layer. Ductal constriction occurred in the non-infected upstream segments of infected area, putatively blocking further luminal ascent of bacteria in UPEC-infected animals. Corresponding histological and morphological changes were found in epididymitis patients. The caput region was found to be unaffected in patients and the mouse model. limitations, reasons for caution: Patient samples represented advanced cases of epididymitis that made surgical intervention necessary.
wider implications of the findings: Our data demonstrate the impact of epididymal segmentation, presumably a protective response mechanism against infectious invasion and bacterial ascent, during epididymitis and affirm the importance of rapid intervention.
study funding/competing interest(s): This work was supported by grants from the State of Hessen (LOEWE-MIBIE) and the DFG (KFO 181). The authors have no conflicts of interest to declare. trial registration number: No clinical trial involved. Key words: epididymis / epididymitis / epididymal segmentation / ascending infection / claudin-1
& The Author 2015. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
1 Institute of Anatomy and Cell Biology, Signaling Group, Justus-Liebig-University Giessen, 35392 Giessen, Germany 2Institute of Anatomy and Cell Biology, Reproductive Biology Group, Justus-Liebig-University Giessen, 35392 Giessen, Germany 3Institute of Pathology, Hannover Medical School, 30625 Hannover, Germany 4Department of Urology, Pediatric Urology and Andrology, Justus-Liebig-University Giessen, 35392 Giessen, Germany
2
Introduction
Materials and Methods Patients and ethics statement Specimen from patients (n ¼ 10, age range 14 – 78, median 60 years) who underwent surgical intervention of persisting epididymitis were sourced from the Department of Pathology, Hannover Medical School, Germany, and their provision was approved by the local Ethics Commission (No. 1378-2012). Surgery was performed as the ultima ratio procedure in the case of persisting disease after conservative therapy. Samples of human epididymides, previously classified as histologically normal, were obtained from a local tissue bank and used as control (n ¼ 16, age range 38 – 87, median 73 years). Most patients were undergoing orchiectomy for suppression of androgen production during treatment of prostatic carcinoma, which was state-of-the-art therapy at the time of sample collection. Collection and
use of human tissue was approved by the Ethics Commission of the A¨rztekammer Hamburg, Germany.
Animal model and ethics statement For an established infectious bacterial epididymitis model, 10-week-old inbred male C57BL/6N mice were purchased from Charles River Laboratories (Sulzfeld, Germany) and infected with the uropathogenic E. coli strain UPEC FT 073 (Welch et al., 2002) and or human gut commensal E. coli strain NPEC470, which is referred to as a non-pathogenic strain in the urogential tract (Bhushan et al., 2008). Animal husbandry, animal experiments and handling of animals were conducted according to the governmental principles and endorsed by the Animal Ethics Committee of the Regierungspra¨sidium Giessen, Germany (permit no. GI 20/23-16/2009). Using the same method as Lang et al. (2013), 5 – 10 ml of suspension containing 40 000 – 80 000 bacteria in phosphate-buffered saline (PBS) were injected under sight control (30 G needle) through the muscular wall into the lumen of the proximal part of the vas deferens, in close proximity to the cauda epididymis. Three days after inoculation, mice were killed by isofluran overdosage. Epididymides were fixed in Bouin’s solution and embedded in paraffin. This stage of disease progression was chosen in order to mimic the stage of epididymitis in patients at first presentation [typically after 2 days of scrotal symptoms (Pilatz et al., 2013)]. Untreated C57/BL6 mice were killed by cervical dislocation according to government principles regarding the Care and Use of animals, with permission (G8151/591-00.33) of the local regulatory authorities. UPEC- (n ¼ 7) and NPEC-inoculated (n ¼ 5) individuals were analyzed and compared with two control groups, PBS sham-treated (n ¼ 4) and untreated mice (n ¼ 8).
Tracer injection Untreated C57/BL6 mice were killed by cervical dislocation and the epididymides were removed. About 20 – 50 ml saturated methylene blue tracer solution (dissolved in PBS) was injected through the capsule into the interstitial space at one single site per epididymis (n ¼ 9 animals). Injections were performed under sight control with a 27 G needle. On the basis of our detailed description of segmentation in mouse epididymis (Thong et al., 2014), the injection site was located in the cauda segments 10, 9 and 8.
Histological analysis For immunohistochemistry and Azan staining, 5 mm thick de-paraffinized mouse and human epididymal sections were used. For Azan staining, sections were stained with 0.1% (w/v) azocarmine G for 10 min and counter-stained with a 0.5% (w/v) aniline-blue and 2% (w/v) orange G solution for 1 h. For immunostaining, sections were incubated overnight at 48C with the primary antibodies anti-E. coli (Dako, Glostrup, Denmark, 1:2000), anticlaudin-1 (Invitrogen, Camarillo, CA, USA, dilution 1:150) and anti-smooth muscle actin (SMA) (Serotec, Oxford, UK, 1:100) diluted in PBS with 0.2% (w/v) bovine serum albumin and 0.1% (w/v) sodium azide. Immunoreactivity was visualized by using the EnVision kit (Dako) followed by the nickel– glucose oxidase approach (Davidoff et al., 1995). Each staining was performed at least three times. For claudin-1 and E. coli detection, sections were heat-treated by boiling in citrate-buffer (pH 6.0) by microwave (9 min at 700 W followed by 15 min at 450 W) before application of primary antibody. For negative controls, primary antibodies were omitted. Images were captured with an Axioskop 2 plus (Carl Zeiss, Go¨ttingen, Germany) supplied with AxioCam (Carl Zeiss) and AxioVision, version 4.8 (Carl Zeiss). Photos were processed with Photoshop CS4 (Adobe, San Jose, CA, USA).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
The epididymis connects the testis and vas deferens and is essential for sperm maturation and storage. Although the epididymal duct is a continuous tubule of 1 m length in the mouse and 6 m in humans, it is coiled into discrete portions that are separated by connective tissue septa, manifesting the segmented structure of the epididymis (Joseph et al., 2009; Hinton et al., 2011). In the human epididymis, 10 distinct zones of epithelial morphology have also been described as being bounded by connective tissue septa (Holstein, 1969). In the mouse, 10 epididymal segments can be distinguished on the basis of different entities: (i) characteristic epithelial morphology (Reid and Cleland, 1957; Abou-Haı¨la and Fain-Maurel, 1984; Thong et al., 2014), (ii) regions of distinct gene expression (Kirchhoff, 1999; Penttinen et al., 2003; Johnston et al., 2005; Jelinsky et al., 2007; reviewed by Turner et al., 2007); and (iii) at least in the caput region of rats, separated interstitial spaces with limited diffusion (Turner et al., 2003). However, the functional consequences of segmentation have barely been investigated. Especially for pathological situations, such as epididymitis, no knowledge about the influence of segmentation is available. Epididymitis is often the consequence of an ascending bacterial infection from the urinary tract via the vas deferens through the ductal lumen of the epididymis (Pilatz et al., 2011). Escherichia coli is the predominant causative pathogen in epididymal infections in males (88% of antibiotic-naive patients, Pilatz et al., 2014a). Although antibiotic treatment was shown to be an effective therapy to remove pathogens, 40% of patients remain oligo- or azoospermic after unilateral epididymitis (Rusz et al., 2012). About 20% of patients with azoospermia have a history of urogenital tract infection or inflammation (Jungwirth et al., 2012). Chronic epididymitis can result in reduced sperm count and motility (Haidl et al., 2008). But even epididymitis patients whose sperm quality has been classified as normal by conventional methods show an altered sperm protein composition (Pilatz et al., 2014b). Sonographic analysis during epididymitis indicates swelling predominantly in the cauda region (Lotti et al., 2011; Pilatz et al., 2013; Lotti and Maggi, 2013, 2015). Studies of E. coli-based epididymitis models in mice (Nashan et al., 1993; Lang et al., 2013, 2014) and other animals have focused on various aspects of the immune response, whilst structural changes have been only marginally regarded, even though the reason for obstruction, as well as for altered sperm protein composition, has not been clarified. Our study investigates the functional consequence of epididymal segmentation and segment-specific histological alterations in epididymitis.
Stammler et al.
3
Epididymal segmentation in ascending epididymitis
Morphometric and statistical analysis Inner ductal diameters, based on the inner border of the muscle layer, were measured using the AxioVision software. At least, three cross-sections of the duct were measured in each segment from each individual. The mean values
and SEM were calculated with MS Excel (Microsoft Office Version 2010, Microsoft, Redmond, WA, USA). P-values were calculated by the nonparametric Mann – Whitney test with GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 1 Morphological overview of the mouse epididymis. Interstitial injection of methylene blue tracer in untreated mouse epididymides revealed that septa in the cauda region act as diffusion barriers between Segment 10 and the vas deferens (A), Segments 10 and 9 (B) and Segments 9 and 8 (C). Sections of mouse epididymis inoculated with UPEC (D and E), NPEC (F and G), PBS sham-treated (H) and untreated mouse (I) processed with Azan staining (D, F, H, I) and E. coli immunolocalization (E and G). (D and E) Longitudinal sections (medial section plan) of a UPEC-infected epididymis. The distal cauda showed abscess-like swelling of the duct restricted to Segments 10 (seg 10) and ductal constriction restricted to Segment 9 and 8. Escherichia coli was detected in the lumen of the epididymal duct in Segment 10. (F and G) Longitudinal sections (frontal section plan) of an NPEC-infected epididymis. Escherichia coli were detected in the lumen of the epididymal duct in Segment 10. (H) Transverse sections of a PBS sham-treated epididymis. I: longitudinal sections (medial section plain) of an untreated epididymis. Scale bar: 1 mm (A– I) and 10 mm in detailed images (D).
4
Stammler et al.
Results To investigate whether the cauda segments in mouse are separate chambers, we injected methylene blue tracer into the interstitial space of cauda segments. We found that septa act as diffusion barriers between all three cauda segments (Fig. 1A– C). Aligning the location of segmental boundaries to E. coli-infected mouse epididymides, we found that the ascent of E. coli was restricted by segmentation (Fig. 1E and G). We compared the morphology of the epididymis of UPEC- and NPEC-infected mice (Fig. 1D–G) with controls, PBS sham-treated and untreated mice (Fig. 1H and I) in paraffin sections and found profound pathological alterations in the segments of cauda (Segments 8–10) of E. coli-inoculated animals. In UPEC- and NPEC-infected epididymides all ductal crosssections of the entire Segment 10 contained a dense population of E. coli in the lumen (Fig. 1E and G). No E. coli was detectable in the adjacent Segment 9 (Fig. 1E and G), indicating that luminal ascent of E. coli was restricted by segment boundaries. Escherichia coli was never detected in the sham and untreated controls (Supplementary data, Table S1). In the infected segments, no significant change of the ductal diameter was found (Fig. 2). On the contrary, in Segments 9 and 8 of UPEC-infected mice, which were devoid of bacteria, the ductal diameter was significantly reduced compared with that of controls (Fig. 2). In NEPC, the ductal diameter was not significantly different from sham-treated control. No significant changes were found in Segment 7 (Fig. 2) or in further upstream segments. On the basis of our data of infected and uninfected regions (Supplementary data, Table S1), we analyzed segment-bysegment histology of the epithelium (Fig. 3) and muscle layer (Fig. 4). In Segment 10 of E. coli-infected animals, the epithelium was absent from the ductal wall in most parts of the cross-sections, as shown by Azan staining (Fig. 3Ia and b). This was never observed in the respective
controls (Fig. 3 Ic and d) or in segments without bacteria (Fig. 3Ie –t). Immunolocalization of the tight junction protein claudin-1 (Fig. 3II) suggested severe epithelial damage, as indicated by detachment of the epithelium from the ductal wall to the lumen (Supplementary data, Fig. S1), whilst a population of atypically shaped claudin-1-positive epithelial cells seemed to remain at the ductal wall in some parts (Fig. 3IIa and b). In the non-infected segments 9 and 8 of UPEC- and NPEC-treated animals (Fig. 3IIc and d), the epithelium was found to be taller than in sham and untreated controls (Fig. 3Ie and f), displaying intense claudin-1 expression and maintaining a pseudostratified character. No changes were found in Segment 7 or upstream (Fig. 3IIm–t, Supplementary data, Table S1). In order to determine potential reasons for the alterations in tubular diameter, changes in the morphology of the muscle layer were assessed by using SMA immunolocalization as a marker of smooth muscle cells (Fig. 4). In Segment 10 of UPEC- and NPEC-infected mice (Fig. 4A and B), the smooth muscle layer was thicker than that in controls (Fig. 4C and D) and contained interspersed SMA-negative areas. No obvious changes were observed in the upstream segments (constricted or non-constricted) and in either control (Fig. 5E – T, Supplementary data, Table S1). In one UPEC-infected animal, the full spectrum of histological changes was additionally visible in Segment 9, combined with positive immuno-detection of E. coli in this segment but no bacteria were present in Segment 8 or upstream (Supplementary data, Table S1). Next we analyzed tissue from patients who had undergone surgical intervention owing to persisting epididymitis and found that histological changes were concentrated in the cauda region (Table I, Fig. 5). Azan staining of control samples (Fig. 5A –C) and in epididymitis (Fig. 5D–F) revealed the presence of septa in the caput, corpus and cauda epididymis. Even though samples represented an advanced stage of
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 2 Diameter of the epididymal duct. Diameter was measured in cross-sections of the epididymal duct at the inner border of the muscle layer of UPEC- (n ¼ 5 animals with three ductal cross-sections per segment), NPEC- (n ¼ 3, see above) and PBS-inoculated (n ¼ 3, see above) and untreated animals (n ¼ 8, see above). No significant differences were observed in Segment 10. In Segment 9 and 8 of UPEC-inoculated animals, the diameter of ductal cross-sections was significantly reduced compared with sham or untreated animals. (The animal with an infected Segment 9 was excluded from diameter statistics.) In NPEC-inoculated individuals, in Segment 9, the ductal diameter was reduced significantly compared with untreated controls. In Segment 7, no significant differences were observed among the four groups. SEM is indicated, P-values ≤ 0.005 calculated by the Mann –Whitney test were considered significant (*).
5
Epididymal segmentation in ascending epididymitis
NPEC- and PBS sham-treated control as well as from untreated animals stained with Azan (I) or claudin-1 immunolocalization (II). Distinct changes of morphology are strictly correlated with infection status and segmentation. In Segment 10 of UPEC-infected (a) and NPEC-animals (b), the epithelium appears to be absent from the ductal wall. Only atypically shaped epithelial cells remain at the ductal wall. In PBS-treated (c), the epithelium seems unaffected compared with untreated controls (d). (e) In Segment 9 of UPEC-inoculated animals, the ductal pseudostratified epithelium appears substantially taller and lacks stereocilia. Claudin-1 expression seems enhanced. (f) In Segment 9 of NPEC-inoculated animals, the epithelium appears narrower while lacking stereocilia. In PBS-treated (g), the epithelium seems unaffected compared with untreated controls (h). (i and j) In Segment 8 of UPEC-infected (a) and NPEC-animals (b), the epithelium appears taller and in UPEC-inoculated mice, claudin-1 expression seemed intensified. No stereocilia are visible. In PBS-treated (k), the epithelium seems unaffected compared with untreated control (l). (m –t) In the distal and proximal region of Segment 7, no differences were observed between the four experimental groups. Scale bar 10 mm in images (t) applies to all panels.
epididymitis, no pathological changes of the epithelium or muscle layer were found in the caput region (Table I). Fibrotic changes were observed in specimens from the cauda and corpus, typically combined with epithelial damage and an interspersed muscle layer (Table I). In the cauda region of epididymitis patients, areas with narrow ductal cross-sections, surrounded by dense connective tissue septa, were observed in proximity to pathologically degenerated, wide ductal cross-sections (Fig. 5C and 6A). Although all patients had received antibiotic treatment before the surgery, E. coli immunoreactivity was found in the cauda specimen of one patient (Patient IX, Table I). A detailed analysis was performed on all patients, and this is displayed in Fig. 6 for Patient IX. Narrow ductal cross-sections had an intact pattern of claudin-1 (Fig. 6B, inlay a), while in degenerated parts of the duct, claudin-1 was only detected in epithelial remnants (inlay b) and in detached epithelial cells (inlay c). Cross-sections of degenerated parts of the duct showed large E. coli-positive areas in the lumen associated with decayed ductal walls (Fig. 6C). Narrow parts of the duct showed a compact muscle layer (Fig. 6D, inlay a) while
around the degenerated parts of the duct, a completely different pattern of SMA was observed, including interspersion of SMA-negative areas peri-ductally (inlay b) and SMA-positive areas throughout the interstitial space.
Discussion We present data that indicate that bacterial infection invades the epididymis in a strictly segmental manner. Comparing a 3-day post-infection mouse model with specimen from patients who required surgical intervention in cases of persisting epididymitis, we found three robust concordances: (i) the caput epididymis was unaffected in both settings, (ii) morphological alterations were found to be restricted by septa in both settings and (iii) lesions of the epithelium and in the muscle layer were concentrated in the cauda epididymis. The different degrees of pathological alteration of caput and cauda correspond to clinical sonographic data (Lotti et al., 2011; Lotti and Maggi, 2013, 2015; Pilatz et al., 2013)
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 3 Lesions and alterations in the epididymal epithelium. Micrographs of the epididymal duct originating from mice 3 days after inoculation of UPEC-,
6
Stammler et al.
and epididymal swelling in a rat model (Lucchetta et al., 1983). SMApositive areas in the interstitial space of human specimens could be fibroblasts transformed to myofibroblasts in the course of fibrotic changes, which was not observed in the 3-day post-infection mouse model. The importance of epididymal segmentation in ascending epididymitis has not been described before this study. In E. coli-positive specimens of men and mice, the bacteria were detected in the lumen combined with profound structural damage of the duct, including detachment of the epithelium and lesions of the muscle layer, visually similar to structures previously characterized as ‘abscesses’ in men (Hori and Tsutsumi,
1995) and in a rat model (Ludwig et al., 2002). The epididymal epithelium mediates water balance (Thong et al., 2014), pH-regulation of luminal contents (Breton and Brown, 2013) and protein secretion (Dacheux, et al., 2006). The detachment of the epithelium might be the reason for altered sperm quality described in acute (Pilatz et al., 2014b) and chronic epididymitis (Haidl et al., 2008). In our systematically analyzed mouse model, in segments upstream of the luminally infected area, the ductal diameter was significantly decreased. In specimens of the cauda region from epididymitis patients, we observed areas with a constricted duct in proximity to a degenerated
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 4 Muscle layer in the epididymis visualized by SMA immunolocalization. Micrographs of the epididymal duct originating from mice 3 days after inoculation of UPEC-, NPEC- and PBS sham-treated controls as well as from untreated animals with immunolocalization of SMA. Distinct changes of morphology are strictly correlated with infection status and segmentation. In Segment 10 of UPEC- (A) and NPEC-infected animals (B), the muscle layer is substantially thickened and interspersed with SMA-negative cells/extracellular material. In PBS-sham treated (C), the epithelium seems unaffected compared with untreated controls (D). In Segment 9 of UPEC- (E) and NPEC-infected animals (F), the muscle layer is thickened and interspersed with SMA-negative cells/extracellular material. In PBS sham-treated (G), the epithelium seems unaffected compared with untreated controls (H). (I – T) In Segment 8 and the distal and proximal region of Segment 7, no differences were observed among the four experimental groups. Scale bar 10 mm in image (t) applies to all panels.
7
+ + 2
Interspersed Interspersed
Interspersed Interspersed None
Detached in infected duct Detached in infected duct
Neg. Neg. Neg. 2 2 4 Cauda Corpus Caput 25 X
11 12 13
2 1 Cauda Cauda 66 IX
9 10
None None None
+ +
Interspersed Detached in one segment Neg.
+
+ Interspersed None 2
2 Cauda
Unclear
14 VIII
8
65 VII
7
Pos. Pos.
2
Neg.
2
None 76 VI
6
Corpus
3
Neg.
Lesions in one segment
+
None 3 54 V
5
Corpus + cauda
Neg.
Lesions in one segment
2 None
None
Detached in one segment
Neg. 4
3
Caput
Corpus + cauda
73 IV
4
46 III
3
None
+ None
Neg.
2 None Lesions in one segment
Lesions in one segment Neg. 3 Corpus + cauda
Neg. 4 Caput + corpus
2 16 II
1 78 I
..........................................................................................................................................................................................................................................................
Escherichia coli immuno-detection Number of segments in the specimen Putative region of specimen Specimen no. Age (year) Patient no.
duct, separated by septa. We assume that epididymal duct constriction represents a protective response mechanism against bacterial ascent that has not been suggested before. In the mouse model, epithelial and muscular lesions in the infected area occurred regardless of E. coli strain, once introduced into the vas deferens. These findings support the hypothesis of Buckles et al. (2015) that uropathogenic virulence factors are mainly connected with the ability of bacteria to ascend. UPEC is described as uropathogenic (Welch et al., 2002) while NPEC is commensal in the human gut (Bhushan et al., 2008). In NPEC-infected animals, ductal constriction was not found to be significantly different from sham-treated controls, suggesting that UPEC effects are regulated by uropathogenic virulence factors. Lang et al. (2013) reported reduced expression of occludin in UPEC-inoculated animals. In contrast to occludin, we found that claudin-1 expression was not reduced as long as the epithelium was present. However, claudin-1, but not occludin, expression was found to be reduced after exposure to inflammatory cytokines of the transforming growth factor b-family in an in vitro epididymis model (Stammler et al., 2013), suggesting that tight junction proteins are differentially regulated by specific pathophysiological effects. Since the alterations were found to be homogeneous within each segment, the effect could be regulated by local factors via the interstitial
Table I Phenotypes of specimens from epididymitis patients.
Figure 5 Human epididymis and epididymitis, azan staining, typical samples from caput (A and D), corpus (B and E) and cauda (C and F) region of patients with normal epididymal histology (A– C) and epididymitis patients (D: Patient IV, E: Patient II, F: Patient VIII). All regions of the human epididymis comprise septa (marked by arrow heads in A – E and a dotted line in F) that restrict areas of distinct epithelial morphology. The corpus of epididymitis patient II (E) shows slight fibrotic alteration compared with controls. The cauda region of Patient VIII (F) shows regionalized alterations of the epididymal duct bordered by septa. Scale bar 2.5 mm applies to all panels.
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Epithelial damage in at least one segment
Muscle layer damage in at least one segment
Fibrotic features
Epididymal segmentation in ascending epididymitis
8
Stammler et al.
space. Tracer injection into the interstitial space demonstrated that the segments were separate compartments with limited diffusion across segment boundaries, as previously suggested in rat (Turner et al., 2003, 2007). Signaling seems to be limited by segment boundaries. The path of communication between segments remains unresolved. Bacterial sensors in the epithelium of the urethra have been described to cause constriction via cholinergic pathways (Deckmann et al., 2014) as previously described in other organs (Krasteva and Kummer, 2012). It is unclear if this mechanism also applies in the epididymis. In this study, we show for the first time that segmental boundaries in the epididymis restrict the ascent of bacteria, putatively by ductal constriction of the non-infected adjacent segments. Lesions in the epithelium and in the muscle layers acquired through epididymitis are strictly correlated with segmentation and infectious status. We provide the histological equivalent of previous sonographic studies and demonstrate histological changes that might cause altered sperm protein composition.
Acknowledgements
Supplementary data
Conflict of interest
Supplementary data are available at http://humrep.oxfordjournals.org.
None declared.
We are grateful to Andre Kaschtanow and Sabine Tasch for their excellent technical assistance and Dr Tali Lang for fruitful discussions.
Authors’ roles Study design (A.S., R.M.), animal model (A.M., S.B.), execution (I.S.-H., A.S., T.H., S.B.), collection of patient’s samples (T.L., D.J., R.M.), data analysis (A.S., T.H., R.M.), manuscript drafting (A.S., R.M.) and critical discussion (A.S., A.P., A.M., R.M.).
Funding This work was supported by grants from the State of Hessen (LOEWEMIBIE) and the DFG (KFO 181).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 6 Cauda region of epididymitis Patient IX, Specimen no. 9. Septum is marked by a dotted line. (A) Azan staining. Groups of structurally intact narrow cross-sections (*) were found in proximity to large, degenerated cross-sections (**). (B) Claudin-1-immunolocalization in narrow ducts revealed an intact epithelium (inlay a). Remnants of claudin-1-positive epithelium were found at the ductal wall of a degenerated duct (inlay b). A detached epithelium was observed in the lumen (inlay c). (C ) Escherichia coli was detected by immunolocalization in the lumen of some narrow ducts (inlay a) and in degenerated ducts (inlay b) as well as in proximity to these. (D) SMA-immunolocalization, narrow ducts show a compact muscle layer (inlay a) while the cross-sections of degenerated ducts exhibited a disrupted muscle layer with interspersion of SMA-negative areas (inlay b). Scale bar 1 mm, inlays 10 mm.
Epididymal segmentation in ascending epididymitis
References
Escherichia coli modulates innate immunity to suppress Th1-mediated inflammatory responses during infectious epididymitis. Infect Immun 2014;82:1104 – 1111. Lotti F, Maggi M. Interleukin 8 and the male genital tract. J Reprod Immunol 2013;100:54 – 65. Lotti F, Maggi M. Ultrasound of the male genital tract in relation to male reproductive health. Hum Reprod Update 2015;21:56 – 83. Lotti F, Corona G, Mancini M, Filimberti E, Degli Innocenti S, Colpi GM, Baldi E, Noci I, Forti G, Adorini L et al. Ultrasonographic and clinical correlates of seminal plasma interleukin-8 levels in patients attending an andrology clinic for infertility. Int J Androl 2011;34:600 – 613. Lucchetta R, Clavert A, Meyer JM, Bollack C. Acute experimental E. coli epididymitis in the rat and its consequences on spermatogenesis. Urol Res 1983;11:117 – 120. Ludwig M, Johannes S, Bergmann M, Failing K, Schiefer HG, Weidner W. Experimental Escherichia coli epididymitis in rats: a model to assess the outcome of antibiotic treatment. BJU Int 2002;90:933– 938. Nashan D, Jantos CA, Ahlers D, Bergmann M, Schiefer HG, Sorg C, Nieschlag E. Immuno-competent cells in the murine epididymis following infection with Escherichia coli. Int J Androl 1993;16:47 – 52. Penttinen J, Pujianto DA, Sipila¨ P, Huhtaniemi IT, Poutanen M. Discovery in silico and characterization in vitro of novel genes exclusively expressed in the mouse epididymis. Mol Endocrinol 2003;17:2138 – 2151. Pilatz A, Bu¨rger M, Wagenlehner F, Hossain H, Weidner W. Acute epididymitis: etiology and inflammatory parameters in a prospective study. J Urol 2011;185:e569. Pilatz A, Wagenlehner F, Bschleipfer T, Schuppe H-C, Diemer T, Linn T, Weidner W, Altinkilic B. Acute epididymitis in ultrasound: results of a prospective study with baseline and follow-up investigations in 134 patients. Eur J Radiol 2013;82:e762 – e768. Pilatz A, Hossain H, Kaiser R, Mankertz A, Schu¨ttler CG, Domann E, Schuppe H-C, Chakraborty T, Weidner W, Wagenlehner F. Acute epididymitis revisited: impact of molecular diagnostics on etiology and contemporary guideline recommendations. Eur Urol 2014a, Epub ahead of print. Pilatz A, Lochnit G, Karnati S, Paradowska-Dogan A, Lang T, Schultheiss D, Schuppe H-C, Hossain H, Baumgart-Vogt E, Weidner W et al. Acute epididymitis induces alterations in sperm protein composition. Fertil Steril 2014b;101:1609 –1617.e1 – 5. Reid B, Cleland K. The structure and function of the epididymis: I. The histology of the rat epididymis. Austr J Zool 1957;5:223 – 256. Rusz A, Pilatz A, Wagenlehner F, Linn T, Diemer T, Schuppe H-C, Lohmeyer J, Hossain H, Weidner W. Influence of urogenital infections and inflammation on semen quality and male fertility. World J Urol 2012;30:23–30. Stammler A, Mu¨ller D, Tabuchi Y, Konrad L, Middendorff R. TGFbs modulate permeability of the blood – epididymis barrier in an in vitro model. PLoS ONE 2013;8:e80611. Thong A, Mu¨ller D, Feuerstacke C, Mietens A, Stammler A, Middendorff R. Neutral endopeptidase (CD10) is abundantly expressed in the epididymis and localized to a distinct population of epithelial cells—its relevance for CNP degradation. Mol Cell Endocrinol 2014;382:234 – 243. Turner TT, Bomgardner D, Jacobs JP, Nguyen QAT. Association of segmentation of the epididymal interstitium with segmented tubule function in rats and mice. Reproduction 2003;125:871 – 878. Turner TT, Johnston DS, Jelinsky SA, Tomsig JL, Finger JN. Segment boundaries of the adult rat epididymis limit interstitial signaling by potential paracrine factors and segments lose differential gene expression after efferent duct ligation. Asian J Androl 2007;9:565 – 573. Welch RA, Burland V, Plunkett G, Redford P, Roesch P, Rasko D, Buckles EL, Liou SR, Boutin A, Hackett J et al. Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc Natl Acad Sci USA 2002;99:17020 – 17024.
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Abou-Haı¨la A, Fain-Maurel MA. Regional differences of the proximal part of mouse epididymis: morphological and histochemical characterization. Anat Rec 1984;209:197– 208. Bhushan S, Tchatalbachev S, Klug J, Fijak M, Pineau C, Chakraborty T, Meinhardt A. Uropathogenic Escherichia coli block MyD88-dependent and activate MyD88-independent signaling pathways in rat testicular cells. J Immunol 2008;180:5537 – 5547. Bhushan S, Hossain H, Lu Y, Geisler A, Tchatalbachev S, Mikulski Z, Schuler G, Klug J, Pilatz A, Wagenlehner FME et al. Uropathogenic E. coli induce different immune response in testicular and peritoneal macrophages: implications for testicular immune privilege. PLoS ONE 2011;6:e28452. Breton S, Brown D. Regulation of luminal acidification by the V-ATPase. Physiology (Bethesda) 2013;28:318 – 329. Buckles EL, Luterbach CL, Wang X, Lockatell CV, Johnson DE, Mobley HLT, Donnenberg MS. Signature-tagged mutagenesis and co-infection studies demonstrate the importance of P fimbriae in a murine model of urinary tract infection. Pathog Dis 2015, 73. Dacheux J-L, Belghazi M, Lanson Y, Dacheux F. Human epididymal secretome and proteome. Mol Cell Endocrinol 2006;250:36 – 42. Davidoff MS, Middendorff R, Mayer B, Holstein AF. Nitric oxide synthase (NOS-I) in Leydig cells of the human testis. Arch Histol Cytol 1995;58:17–30. Deckmann K, Filipski K, Krasteva-Christ G, Fronius M, Althaus M, Rafiq A, Papadakis T, Renno L, Jurastow I, Wessels L et al. Bitter triggers acetylcholine release from polymodal urethral chemosensory cells and bladder reflexes. Proc Natl Acad Sci USA 2014;22:8287– 8292. Haidl G, Allam JP, Schuppe H-C. Chronic epididymitis: impact on semen parameters and therapeutic options. Andrologia 2008;40:92– 96. Hinton BT, Galdamez MM, Sutherland A, Bomgardner D, Xu B, Abdel-Fattah R, Yang L. How do you get six meters of epididymis inside a human scrotum? J Androl 2011;32:558– 564. Holstein AF. Morphologische Studien am Nebenhoden des Menschen. In: Bargmann W, Doerr W (eds). Zwanglose Abhandlungen aus dem Gebiet der normalen und pathologischen Anatomie. Stuttgart, Germany: Thieme Medical Publisher, 1969. Hori S, Tsutsumi Y. Histological differentiation between chlamydial and bacterial epididymitis: nondestructive and proliferative versus destructive and abscess forming—immunohistochemical and clinicopathological findings. Hum Pathol 1995;26:402–407. Jelinsky SA, Turner TT, Bang HJ, Finger JN, Solarz MK, Wilson E, Brown EL, Kopf GS, Johnston DS. The rat epididymal transcriptome: comparison of segmental gene expression in the rat and mouse epididymides. Biol Reprod 2007;76:561 – 570. Johnston DS, Jelinsky SA, Bang HJ, DiCandeloro P, Wilson E, Kopf GS, Turner TT. The mouse epididymal transcriptome: transcriptional profiling of segmental gene expression in the epididymis. Biol Reprod 2005;73:404–413. Joseph A, Yao H, Hinton BT. Development and morphogenesis of the Wolffian/epididymal duct, more twists and turns. Dev Biol 2009;325:6–14. Jungwirth A, Giwercman A, Tournaye H, Diemer T, Kopa Z, Dohle G, Krausz C. European Association of Urology guidelines on Male Infertility: the 2012 update. Eur Urol 2012;62:324– 332. Kirchhoff C. Gene expression in the epididymis. Int Rev Cytol 1999; 188:133– 202. Krasteva G, Kummer W. ‘Tasting’ the airway lining fluid. Histochem Cell Biol 2012;138:365 – 383. Lang T, Dechant M, Sanchez V, Wistuba J, Boiani M, Pilatz A, Stammler A, Middendorff R, Schuler G, Bhushan S et al. Structural and functional integrity of spermatozoa is compromised as a consequence of acute uropathogenic E. coli-associated epididymitis. Biol Reprod 2013;89:59. Lang T, Hudemann C, Tchatalbachev S, Stammler A, Michel V, Aslani F, Bhushan S, Chakraborty T, Renz H, Meinhardt A. Uropathogenic
9
ORIGINAL ARTICLE Reproductive biology
Epididymitis: ascending infection restricted by segmental boundaries A. Stammler 1, T. Hau 1, S. Bhushan2, A. Meinhardt 2, D. Jonigk 3, T. Lippmann 3, A. Pilatz4, I. Schneider-Hu¨ther 1, and R. Middendorff 1,*
*Correspondence address. E-mail: [email protected]
Submitted on March 19, 2015; resubmitted on April 17, 2015; accepted on April 24, 2015
study question: Is the regionalization of epididymitis related to epididymal segmentation? summary answer: We show for the first time that luminal ascent of bacteria is strictly gated by epididymal segment boundaries, involving ductal constriction adjacent to the infected area.
what is known already: The epididymal duct is a continuous, unbranched tube, coiled into segments that are divided by connective tissue septa. Sonographic analysis indicates that swelling associated with epididymitis is predominant in the cauda region. Epididymal segmentation has never been investigated in the context of pathological alterations. study design, size, and duration: We analyzed segment-specific changes in the epididymal duct in a mouse model and in men. In the mouse epididymitis model (3 days post-infection, injection of bacteria into the lumen of the vas deferens), two Escherichia coli strains were tested: a uropathogenic strain CFT073 (UPEC, n ¼ 7) and a fecal non-pathogenic strain NPEC470 (NPEC, n ¼ 5). Two control groups: phosphate-buffered saline, sham-treated animals (n ¼ 4) and untreated mice (n ¼ 8). In addition, segmentation was verified by ex vivo injection of dye into the interstitial spaces of untreated mouse epididymides. Histological findings were compared with specimens from epididymitis patients (n ¼ 10, age range 14–78, median 60 years) who underwent surgical intervention; control: samples from patients without epididymitis (n ¼ 16, age range 38–87, median 73 years). participants/materials, setting, and methods: We investigated the ascending infections by detailed histological analysis in correlation with local infection status in a mouse epididymitis model. As a proof of concept, rare patient material from two archives was analyzed: epididymides from patients who underwent surgical intervention for persisting epididymitis, and for control, histologically normal epididymides from men who underwent orchiectomy for therapy of prostatic carcinoma. main results and the role of chance: Luminal ascent of E. coli in mice was strictly gated by epididymal segment boundaries. In the mouse model, both strains of E. coli were detected exclusively in the distal cauda segment associated with damage of the epithelium and muscle layer. Ductal constriction occurred in the non-infected upstream segments of infected area, putatively blocking further luminal ascent of bacteria in UPEC-infected animals. Corresponding histological and morphological changes were found in epididymitis patients. The caput region was found to be unaffected in patients and the mouse model. limitations, reasons for caution: Patient samples represented advanced cases of epididymitis that made surgical intervention necessary.
wider implications of the findings: Our data demonstrate the impact of epididymal segmentation, presumably a protective response mechanism against infectious invasion and bacterial ascent, during epididymitis and affirm the importance of rapid intervention.
study funding/competing interest(s): This work was supported by grants from the State of Hessen (LOEWE-MIBIE) and the DFG (KFO 181). The authors have no conflicts of interest to declare. trial registration number: No clinical trial involved. Key words: epididymis / epididymitis / epididymal segmentation / ascending infection / claudin-1
& The Author 2015. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
1 Institute of Anatomy and Cell Biology, Signaling Group, Justus-Liebig-University Giessen, 35392 Giessen, Germany 2Institute of Anatomy and Cell Biology, Reproductive Biology Group, Justus-Liebig-University Giessen, 35392 Giessen, Germany 3Institute of Pathology, Hannover Medical School, 30625 Hannover, Germany 4Department of Urology, Pediatric Urology and Andrology, Justus-Liebig-University Giessen, 35392 Giessen, Germany
2
Introduction
Materials and Methods Patients and ethics statement Specimen from patients (n ¼ 10, age range 14 – 78, median 60 years) who underwent surgical intervention of persisting epididymitis were sourced from the Department of Pathology, Hannover Medical School, Germany, and their provision was approved by the local Ethics Commission (No. 1378-2012). Surgery was performed as the ultima ratio procedure in the case of persisting disease after conservative therapy. Samples of human epididymides, previously classified as histologically normal, were obtained from a local tissue bank and used as control (n ¼ 16, age range 38 – 87, median 73 years). Most patients were undergoing orchiectomy for suppression of androgen production during treatment of prostatic carcinoma, which was state-of-the-art therapy at the time of sample collection. Collection and
use of human tissue was approved by the Ethics Commission of the A¨rztekammer Hamburg, Germany.
Animal model and ethics statement For an established infectious bacterial epididymitis model, 10-week-old inbred male C57BL/6N mice were purchased from Charles River Laboratories (Sulzfeld, Germany) and infected with the uropathogenic E. coli strain UPEC FT 073 (Welch et al., 2002) and or human gut commensal E. coli strain NPEC470, which is referred to as a non-pathogenic strain in the urogential tract (Bhushan et al., 2008). Animal husbandry, animal experiments and handling of animals were conducted according to the governmental principles and endorsed by the Animal Ethics Committee of the Regierungspra¨sidium Giessen, Germany (permit no. GI 20/23-16/2009). Using the same method as Lang et al. (2013), 5 – 10 ml of suspension containing 40 000 – 80 000 bacteria in phosphate-buffered saline (PBS) were injected under sight control (30 G needle) through the muscular wall into the lumen of the proximal part of the vas deferens, in close proximity to the cauda epididymis. Three days after inoculation, mice were killed by isofluran overdosage. Epididymides were fixed in Bouin’s solution and embedded in paraffin. This stage of disease progression was chosen in order to mimic the stage of epididymitis in patients at first presentation [typically after 2 days of scrotal symptoms (Pilatz et al., 2013)]. Untreated C57/BL6 mice were killed by cervical dislocation according to government principles regarding the Care and Use of animals, with permission (G8151/591-00.33) of the local regulatory authorities. UPEC- (n ¼ 7) and NPEC-inoculated (n ¼ 5) individuals were analyzed and compared with two control groups, PBS sham-treated (n ¼ 4) and untreated mice (n ¼ 8).
Tracer injection Untreated C57/BL6 mice were killed by cervical dislocation and the epididymides were removed. About 20 – 50 ml saturated methylene blue tracer solution (dissolved in PBS) was injected through the capsule into the interstitial space at one single site per epididymis (n ¼ 9 animals). Injections were performed under sight control with a 27 G needle. On the basis of our detailed description of segmentation in mouse epididymis (Thong et al., 2014), the injection site was located in the cauda segments 10, 9 and 8.
Histological analysis For immunohistochemistry and Azan staining, 5 mm thick de-paraffinized mouse and human epididymal sections were used. For Azan staining, sections were stained with 0.1% (w/v) azocarmine G for 10 min and counter-stained with a 0.5% (w/v) aniline-blue and 2% (w/v) orange G solution for 1 h. For immunostaining, sections were incubated overnight at 48C with the primary antibodies anti-E. coli (Dako, Glostrup, Denmark, 1:2000), anticlaudin-1 (Invitrogen, Camarillo, CA, USA, dilution 1:150) and anti-smooth muscle actin (SMA) (Serotec, Oxford, UK, 1:100) diluted in PBS with 0.2% (w/v) bovine serum albumin and 0.1% (w/v) sodium azide. Immunoreactivity was visualized by using the EnVision kit (Dako) followed by the nickel– glucose oxidase approach (Davidoff et al., 1995). Each staining was performed at least three times. For claudin-1 and E. coli detection, sections were heat-treated by boiling in citrate-buffer (pH 6.0) by microwave (9 min at 700 W followed by 15 min at 450 W) before application of primary antibody. For negative controls, primary antibodies were omitted. Images were captured with an Axioskop 2 plus (Carl Zeiss, Go¨ttingen, Germany) supplied with AxioCam (Carl Zeiss) and AxioVision, version 4.8 (Carl Zeiss). Photos were processed with Photoshop CS4 (Adobe, San Jose, CA, USA).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
The epididymis connects the testis and vas deferens and is essential for sperm maturation and storage. Although the epididymal duct is a continuous tubule of 1 m length in the mouse and 6 m in humans, it is coiled into discrete portions that are separated by connective tissue septa, manifesting the segmented structure of the epididymis (Joseph et al., 2009; Hinton et al., 2011). In the human epididymis, 10 distinct zones of epithelial morphology have also been described as being bounded by connective tissue septa (Holstein, 1969). In the mouse, 10 epididymal segments can be distinguished on the basis of different entities: (i) characteristic epithelial morphology (Reid and Cleland, 1957; Abou-Haı¨la and Fain-Maurel, 1984; Thong et al., 2014), (ii) regions of distinct gene expression (Kirchhoff, 1999; Penttinen et al., 2003; Johnston et al., 2005; Jelinsky et al., 2007; reviewed by Turner et al., 2007); and (iii) at least in the caput region of rats, separated interstitial spaces with limited diffusion (Turner et al., 2003). However, the functional consequences of segmentation have barely been investigated. Especially for pathological situations, such as epididymitis, no knowledge about the influence of segmentation is available. Epididymitis is often the consequence of an ascending bacterial infection from the urinary tract via the vas deferens through the ductal lumen of the epididymis (Pilatz et al., 2011). Escherichia coli is the predominant causative pathogen in epididymal infections in males (88% of antibiotic-naive patients, Pilatz et al., 2014a). Although antibiotic treatment was shown to be an effective therapy to remove pathogens, 40% of patients remain oligo- or azoospermic after unilateral epididymitis (Rusz et al., 2012). About 20% of patients with azoospermia have a history of urogenital tract infection or inflammation (Jungwirth et al., 2012). Chronic epididymitis can result in reduced sperm count and motility (Haidl et al., 2008). But even epididymitis patients whose sperm quality has been classified as normal by conventional methods show an altered sperm protein composition (Pilatz et al., 2014b). Sonographic analysis during epididymitis indicates swelling predominantly in the cauda region (Lotti et al., 2011; Pilatz et al., 2013; Lotti and Maggi, 2013, 2015). Studies of E. coli-based epididymitis models in mice (Nashan et al., 1993; Lang et al., 2013, 2014) and other animals have focused on various aspects of the immune response, whilst structural changes have been only marginally regarded, even though the reason for obstruction, as well as for altered sperm protein composition, has not been clarified. Our study investigates the functional consequence of epididymal segmentation and segment-specific histological alterations in epididymitis.
Stammler et al.
3
Epididymal segmentation in ascending epididymitis
Morphometric and statistical analysis Inner ductal diameters, based on the inner border of the muscle layer, were measured using the AxioVision software. At least, three cross-sections of the duct were measured in each segment from each individual. The mean values
and SEM were calculated with MS Excel (Microsoft Office Version 2010, Microsoft, Redmond, WA, USA). P-values were calculated by the nonparametric Mann – Whitney test with GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 1 Morphological overview of the mouse epididymis. Interstitial injection of methylene blue tracer in untreated mouse epididymides revealed that septa in the cauda region act as diffusion barriers between Segment 10 and the vas deferens (A), Segments 10 and 9 (B) and Segments 9 and 8 (C). Sections of mouse epididymis inoculated with UPEC (D and E), NPEC (F and G), PBS sham-treated (H) and untreated mouse (I) processed with Azan staining (D, F, H, I) and E. coli immunolocalization (E and G). (D and E) Longitudinal sections (medial section plan) of a UPEC-infected epididymis. The distal cauda showed abscess-like swelling of the duct restricted to Segments 10 (seg 10) and ductal constriction restricted to Segment 9 and 8. Escherichia coli was detected in the lumen of the epididymal duct in Segment 10. (F and G) Longitudinal sections (frontal section plan) of an NPEC-infected epididymis. Escherichia coli were detected in the lumen of the epididymal duct in Segment 10. (H) Transverse sections of a PBS sham-treated epididymis. I: longitudinal sections (medial section plain) of an untreated epididymis. Scale bar: 1 mm (A– I) and 10 mm in detailed images (D).
4
Stammler et al.
Results To investigate whether the cauda segments in mouse are separate chambers, we injected methylene blue tracer into the interstitial space of cauda segments. We found that septa act as diffusion barriers between all three cauda segments (Fig. 1A– C). Aligning the location of segmental boundaries to E. coli-infected mouse epididymides, we found that the ascent of E. coli was restricted by segmentation (Fig. 1E and G). We compared the morphology of the epididymis of UPEC- and NPEC-infected mice (Fig. 1D–G) with controls, PBS sham-treated and untreated mice (Fig. 1H and I) in paraffin sections and found profound pathological alterations in the segments of cauda (Segments 8–10) of E. coli-inoculated animals. In UPEC- and NPEC-infected epididymides all ductal crosssections of the entire Segment 10 contained a dense population of E. coli in the lumen (Fig. 1E and G). No E. coli was detectable in the adjacent Segment 9 (Fig. 1E and G), indicating that luminal ascent of E. coli was restricted by segment boundaries. Escherichia coli was never detected in the sham and untreated controls (Supplementary data, Table S1). In the infected segments, no significant change of the ductal diameter was found (Fig. 2). On the contrary, in Segments 9 and 8 of UPEC-infected mice, which were devoid of bacteria, the ductal diameter was significantly reduced compared with that of controls (Fig. 2). In NEPC, the ductal diameter was not significantly different from sham-treated control. No significant changes were found in Segment 7 (Fig. 2) or in further upstream segments. On the basis of our data of infected and uninfected regions (Supplementary data, Table S1), we analyzed segment-bysegment histology of the epithelium (Fig. 3) and muscle layer (Fig. 4). In Segment 10 of E. coli-infected animals, the epithelium was absent from the ductal wall in most parts of the cross-sections, as shown by Azan staining (Fig. 3Ia and b). This was never observed in the respective
controls (Fig. 3 Ic and d) or in segments without bacteria (Fig. 3Ie –t). Immunolocalization of the tight junction protein claudin-1 (Fig. 3II) suggested severe epithelial damage, as indicated by detachment of the epithelium from the ductal wall to the lumen (Supplementary data, Fig. S1), whilst a population of atypically shaped claudin-1-positive epithelial cells seemed to remain at the ductal wall in some parts (Fig. 3IIa and b). In the non-infected segments 9 and 8 of UPEC- and NPEC-treated animals (Fig. 3IIc and d), the epithelium was found to be taller than in sham and untreated controls (Fig. 3Ie and f), displaying intense claudin-1 expression and maintaining a pseudostratified character. No changes were found in Segment 7 or upstream (Fig. 3IIm–t, Supplementary data, Table S1). In order to determine potential reasons for the alterations in tubular diameter, changes in the morphology of the muscle layer were assessed by using SMA immunolocalization as a marker of smooth muscle cells (Fig. 4). In Segment 10 of UPEC- and NPEC-infected mice (Fig. 4A and B), the smooth muscle layer was thicker than that in controls (Fig. 4C and D) and contained interspersed SMA-negative areas. No obvious changes were observed in the upstream segments (constricted or non-constricted) and in either control (Fig. 5E – T, Supplementary data, Table S1). In one UPEC-infected animal, the full spectrum of histological changes was additionally visible in Segment 9, combined with positive immuno-detection of E. coli in this segment but no bacteria were present in Segment 8 or upstream (Supplementary data, Table S1). Next we analyzed tissue from patients who had undergone surgical intervention owing to persisting epididymitis and found that histological changes were concentrated in the cauda region (Table I, Fig. 5). Azan staining of control samples (Fig. 5A –C) and in epididymitis (Fig. 5D–F) revealed the presence of septa in the caput, corpus and cauda epididymis. Even though samples represented an advanced stage of
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 2 Diameter of the epididymal duct. Diameter was measured in cross-sections of the epididymal duct at the inner border of the muscle layer of UPEC- (n ¼ 5 animals with three ductal cross-sections per segment), NPEC- (n ¼ 3, see above) and PBS-inoculated (n ¼ 3, see above) and untreated animals (n ¼ 8, see above). No significant differences were observed in Segment 10. In Segment 9 and 8 of UPEC-inoculated animals, the diameter of ductal cross-sections was significantly reduced compared with sham or untreated animals. (The animal with an infected Segment 9 was excluded from diameter statistics.) In NPEC-inoculated individuals, in Segment 9, the ductal diameter was reduced significantly compared with untreated controls. In Segment 7, no significant differences were observed among the four groups. SEM is indicated, P-values ≤ 0.005 calculated by the Mann –Whitney test were considered significant (*).
5
Epididymal segmentation in ascending epididymitis
NPEC- and PBS sham-treated control as well as from untreated animals stained with Azan (I) or claudin-1 immunolocalization (II). Distinct changes of morphology are strictly correlated with infection status and segmentation. In Segment 10 of UPEC-infected (a) and NPEC-animals (b), the epithelium appears to be absent from the ductal wall. Only atypically shaped epithelial cells remain at the ductal wall. In PBS-treated (c), the epithelium seems unaffected compared with untreated controls (d). (e) In Segment 9 of UPEC-inoculated animals, the ductal pseudostratified epithelium appears substantially taller and lacks stereocilia. Claudin-1 expression seems enhanced. (f) In Segment 9 of NPEC-inoculated animals, the epithelium appears narrower while lacking stereocilia. In PBS-treated (g), the epithelium seems unaffected compared with untreated controls (h). (i and j) In Segment 8 of UPEC-infected (a) and NPEC-animals (b), the epithelium appears taller and in UPEC-inoculated mice, claudin-1 expression seemed intensified. No stereocilia are visible. In PBS-treated (k), the epithelium seems unaffected compared with untreated control (l). (m –t) In the distal and proximal region of Segment 7, no differences were observed between the four experimental groups. Scale bar 10 mm in images (t) applies to all panels.
epididymitis, no pathological changes of the epithelium or muscle layer were found in the caput region (Table I). Fibrotic changes were observed in specimens from the cauda and corpus, typically combined with epithelial damage and an interspersed muscle layer (Table I). In the cauda region of epididymitis patients, areas with narrow ductal cross-sections, surrounded by dense connective tissue septa, were observed in proximity to pathologically degenerated, wide ductal cross-sections (Fig. 5C and 6A). Although all patients had received antibiotic treatment before the surgery, E. coli immunoreactivity was found in the cauda specimen of one patient (Patient IX, Table I). A detailed analysis was performed on all patients, and this is displayed in Fig. 6 for Patient IX. Narrow ductal cross-sections had an intact pattern of claudin-1 (Fig. 6B, inlay a), while in degenerated parts of the duct, claudin-1 was only detected in epithelial remnants (inlay b) and in detached epithelial cells (inlay c). Cross-sections of degenerated parts of the duct showed large E. coli-positive areas in the lumen associated with decayed ductal walls (Fig. 6C). Narrow parts of the duct showed a compact muscle layer (Fig. 6D, inlay a) while
around the degenerated parts of the duct, a completely different pattern of SMA was observed, including interspersion of SMA-negative areas peri-ductally (inlay b) and SMA-positive areas throughout the interstitial space.
Discussion We present data that indicate that bacterial infection invades the epididymis in a strictly segmental manner. Comparing a 3-day post-infection mouse model with specimen from patients who required surgical intervention in cases of persisting epididymitis, we found three robust concordances: (i) the caput epididymis was unaffected in both settings, (ii) morphological alterations were found to be restricted by septa in both settings and (iii) lesions of the epithelium and in the muscle layer were concentrated in the cauda epididymis. The different degrees of pathological alteration of caput and cauda correspond to clinical sonographic data (Lotti et al., 2011; Lotti and Maggi, 2013, 2015; Pilatz et al., 2013)
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 3 Lesions and alterations in the epididymal epithelium. Micrographs of the epididymal duct originating from mice 3 days after inoculation of UPEC-,
6
Stammler et al.
and epididymal swelling in a rat model (Lucchetta et al., 1983). SMApositive areas in the interstitial space of human specimens could be fibroblasts transformed to myofibroblasts in the course of fibrotic changes, which was not observed in the 3-day post-infection mouse model. The importance of epididymal segmentation in ascending epididymitis has not been described before this study. In E. coli-positive specimens of men and mice, the bacteria were detected in the lumen combined with profound structural damage of the duct, including detachment of the epithelium and lesions of the muscle layer, visually similar to structures previously characterized as ‘abscesses’ in men (Hori and Tsutsumi,
1995) and in a rat model (Ludwig et al., 2002). The epididymal epithelium mediates water balance (Thong et al., 2014), pH-regulation of luminal contents (Breton and Brown, 2013) and protein secretion (Dacheux, et al., 2006). The detachment of the epithelium might be the reason for altered sperm quality described in acute (Pilatz et al., 2014b) and chronic epididymitis (Haidl et al., 2008). In our systematically analyzed mouse model, in segments upstream of the luminally infected area, the ductal diameter was significantly decreased. In specimens of the cauda region from epididymitis patients, we observed areas with a constricted duct in proximity to a degenerated
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 4 Muscle layer in the epididymis visualized by SMA immunolocalization. Micrographs of the epididymal duct originating from mice 3 days after inoculation of UPEC-, NPEC- and PBS sham-treated controls as well as from untreated animals with immunolocalization of SMA. Distinct changes of morphology are strictly correlated with infection status and segmentation. In Segment 10 of UPEC- (A) and NPEC-infected animals (B), the muscle layer is substantially thickened and interspersed with SMA-negative cells/extracellular material. In PBS-sham treated (C), the epithelium seems unaffected compared with untreated controls (D). In Segment 9 of UPEC- (E) and NPEC-infected animals (F), the muscle layer is thickened and interspersed with SMA-negative cells/extracellular material. In PBS sham-treated (G), the epithelium seems unaffected compared with untreated controls (H). (I – T) In Segment 8 and the distal and proximal region of Segment 7, no differences were observed among the four experimental groups. Scale bar 10 mm in image (t) applies to all panels.
7
+ + 2
Interspersed Interspersed
Interspersed Interspersed None
Detached in infected duct Detached in infected duct
Neg. Neg. Neg. 2 2 4 Cauda Corpus Caput 25 X
11 12 13
2 1 Cauda Cauda 66 IX
9 10
None None None
+ +
Interspersed Detached in one segment Neg.
+
+ Interspersed None 2
2 Cauda
Unclear
14 VIII
8
65 VII
7
Pos. Pos.
2
Neg.
2
None 76 VI
6
Corpus
3
Neg.
Lesions in one segment
+
None 3 54 V
5
Corpus + cauda
Neg.
Lesions in one segment
2 None
None
Detached in one segment
Neg. 4
3
Caput
Corpus + cauda
73 IV
4
46 III
3
None
+ None
Neg.
2 None Lesions in one segment
Lesions in one segment Neg. 3 Corpus + cauda
Neg. 4 Caput + corpus
2 16 II
1 78 I
..........................................................................................................................................................................................................................................................
Escherichia coli immuno-detection Number of segments in the specimen Putative region of specimen Specimen no. Age (year) Patient no.
duct, separated by septa. We assume that epididymal duct constriction represents a protective response mechanism against bacterial ascent that has not been suggested before. In the mouse model, epithelial and muscular lesions in the infected area occurred regardless of E. coli strain, once introduced into the vas deferens. These findings support the hypothesis of Buckles et al. (2015) that uropathogenic virulence factors are mainly connected with the ability of bacteria to ascend. UPEC is described as uropathogenic (Welch et al., 2002) while NPEC is commensal in the human gut (Bhushan et al., 2008). In NPEC-infected animals, ductal constriction was not found to be significantly different from sham-treated controls, suggesting that UPEC effects are regulated by uropathogenic virulence factors. Lang et al. (2013) reported reduced expression of occludin in UPEC-inoculated animals. In contrast to occludin, we found that claudin-1 expression was not reduced as long as the epithelium was present. However, claudin-1, but not occludin, expression was found to be reduced after exposure to inflammatory cytokines of the transforming growth factor b-family in an in vitro epididymis model (Stammler et al., 2013), suggesting that tight junction proteins are differentially regulated by specific pathophysiological effects. Since the alterations were found to be homogeneous within each segment, the effect could be regulated by local factors via the interstitial
Table I Phenotypes of specimens from epididymitis patients.
Figure 5 Human epididymis and epididymitis, azan staining, typical samples from caput (A and D), corpus (B and E) and cauda (C and F) region of patients with normal epididymal histology (A– C) and epididymitis patients (D: Patient IV, E: Patient II, F: Patient VIII). All regions of the human epididymis comprise septa (marked by arrow heads in A – E and a dotted line in F) that restrict areas of distinct epithelial morphology. The corpus of epididymitis patient II (E) shows slight fibrotic alteration compared with controls. The cauda region of Patient VIII (F) shows regionalized alterations of the epididymal duct bordered by septa. Scale bar 2.5 mm applies to all panels.
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Epithelial damage in at least one segment
Muscle layer damage in at least one segment
Fibrotic features
Epididymal segmentation in ascending epididymitis
8
Stammler et al.
space. Tracer injection into the interstitial space demonstrated that the segments were separate compartments with limited diffusion across segment boundaries, as previously suggested in rat (Turner et al., 2003, 2007). Signaling seems to be limited by segment boundaries. The path of communication between segments remains unresolved. Bacterial sensors in the epithelium of the urethra have been described to cause constriction via cholinergic pathways (Deckmann et al., 2014) as previously described in other organs (Krasteva and Kummer, 2012). It is unclear if this mechanism also applies in the epididymis. In this study, we show for the first time that segmental boundaries in the epididymis restrict the ascent of bacteria, putatively by ductal constriction of the non-infected adjacent segments. Lesions in the epithelium and in the muscle layers acquired through epididymitis are strictly correlated with segmentation and infectious status. We provide the histological equivalent of previous sonographic studies and demonstrate histological changes that might cause altered sperm protein composition.
Acknowledgements
Supplementary data
Conflict of interest
Supplementary data are available at http://humrep.oxfordjournals.org.
None declared.
We are grateful to Andre Kaschtanow and Sabine Tasch for their excellent technical assistance and Dr Tali Lang for fruitful discussions.
Authors’ roles Study design (A.S., R.M.), animal model (A.M., S.B.), execution (I.S.-H., A.S., T.H., S.B.), collection of patient’s samples (T.L., D.J., R.M.), data analysis (A.S., T.H., R.M.), manuscript drafting (A.S., R.M.) and critical discussion (A.S., A.P., A.M., R.M.).
Funding This work was supported by grants from the State of Hessen (LOEWEMIBIE) and the DFG (KFO 181).
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Figure 6 Cauda region of epididymitis Patient IX, Specimen no. 9. Septum is marked by a dotted line. (A) Azan staining. Groups of structurally intact narrow cross-sections (*) were found in proximity to large, degenerated cross-sections (**). (B) Claudin-1-immunolocalization in narrow ducts revealed an intact epithelium (inlay a). Remnants of claudin-1-positive epithelium were found at the ductal wall of a degenerated duct (inlay b). A detached epithelium was observed in the lumen (inlay c). (C ) Escherichia coli was detected by immunolocalization in the lumen of some narrow ducts (inlay a) and in degenerated ducts (inlay b) as well as in proximity to these. (D) SMA-immunolocalization, narrow ducts show a compact muscle layer (inlay a) while the cross-sections of degenerated ducts exhibited a disrupted muscle layer with interspersion of SMA-negative areas (inlay b). Scale bar 1 mm, inlays 10 mm.
Epididymal segmentation in ascending epididymitis
References
Escherichia coli modulates innate immunity to suppress Th1-mediated inflammatory responses during infectious epididymitis. Infect Immun 2014;82:1104 – 1111. Lotti F, Maggi M. Interleukin 8 and the male genital tract. J Reprod Immunol 2013;100:54 – 65. Lotti F, Maggi M. Ultrasound of the male genital tract in relation to male reproductive health. Hum Reprod Update 2015;21:56 – 83. Lotti F, Corona G, Mancini M, Filimberti E, Degli Innocenti S, Colpi GM, Baldi E, Noci I, Forti G, Adorini L et al. Ultrasonographic and clinical correlates of seminal plasma interleukin-8 levels in patients attending an andrology clinic for infertility. Int J Androl 2011;34:600 – 613. Lucchetta R, Clavert A, Meyer JM, Bollack C. Acute experimental E. coli epididymitis in the rat and its consequences on spermatogenesis. Urol Res 1983;11:117 – 120. Ludwig M, Johannes S, Bergmann M, Failing K, Schiefer HG, Weidner W. Experimental Escherichia coli epididymitis in rats: a model to assess the outcome of antibiotic treatment. BJU Int 2002;90:933– 938. Nashan D, Jantos CA, Ahlers D, Bergmann M, Schiefer HG, Sorg C, Nieschlag E. Immuno-competent cells in the murine epididymis following infection with Escherichia coli. Int J Androl 1993;16:47 – 52. Penttinen J, Pujianto DA, Sipila¨ P, Huhtaniemi IT, Poutanen M. Discovery in silico and characterization in vitro of novel genes exclusively expressed in the mouse epididymis. Mol Endocrinol 2003;17:2138 – 2151. Pilatz A, Bu¨rger M, Wagenlehner F, Hossain H, Weidner W. Acute epididymitis: etiology and inflammatory parameters in a prospective study. J Urol 2011;185:e569. Pilatz A, Wagenlehner F, Bschleipfer T, Schuppe H-C, Diemer T, Linn T, Weidner W, Altinkilic B. Acute epididymitis in ultrasound: results of a prospective study with baseline and follow-up investigations in 134 patients. Eur J Radiol 2013;82:e762 – e768. Pilatz A, Hossain H, Kaiser R, Mankertz A, Schu¨ttler CG, Domann E, Schuppe H-C, Chakraborty T, Weidner W, Wagenlehner F. Acute epididymitis revisited: impact of molecular diagnostics on etiology and contemporary guideline recommendations. Eur Urol 2014a, Epub ahead of print. Pilatz A, Lochnit G, Karnati S, Paradowska-Dogan A, Lang T, Schultheiss D, Schuppe H-C, Hossain H, Baumgart-Vogt E, Weidner W et al. Acute epididymitis induces alterations in sperm protein composition. Fertil Steril 2014b;101:1609 –1617.e1 – 5. Reid B, Cleland K. The structure and function of the epididymis: I. The histology of the rat epididymis. Austr J Zool 1957;5:223 – 256. Rusz A, Pilatz A, Wagenlehner F, Linn T, Diemer T, Schuppe H-C, Lohmeyer J, Hossain H, Weidner W. Influence of urogenital infections and inflammation on semen quality and male fertility. World J Urol 2012;30:23–30. Stammler A, Mu¨ller D, Tabuchi Y, Konrad L, Middendorff R. TGFbs modulate permeability of the blood – epididymis barrier in an in vitro model. PLoS ONE 2013;8:e80611. Thong A, Mu¨ller D, Feuerstacke C, Mietens A, Stammler A, Middendorff R. Neutral endopeptidase (CD10) is abundantly expressed in the epididymis and localized to a distinct population of epithelial cells—its relevance for CNP degradation. Mol Cell Endocrinol 2014;382:234 – 243. Turner TT, Bomgardner D, Jacobs JP, Nguyen QAT. Association of segmentation of the epididymal interstitium with segmented tubule function in rats and mice. Reproduction 2003;125:871 – 878. Turner TT, Johnston DS, Jelinsky SA, Tomsig JL, Finger JN. Segment boundaries of the adult rat epididymis limit interstitial signaling by potential paracrine factors and segments lose differential gene expression after efferent duct ligation. Asian J Androl 2007;9:565 – 573. Welch RA, Burland V, Plunkett G, Redford P, Roesch P, Rasko D, Buckles EL, Liou SR, Boutin A, Hackett J et al. Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc Natl Acad Sci USA 2002;99:17020 – 17024.
Downloaded from http://humrep.oxfordjournals.org/ at North Dakota State University on May 26, 2015
Abou-Haı¨la A, Fain-Maurel MA. Regional differences of the proximal part of mouse epididymis: morphological and histochemical characterization. Anat Rec 1984;209:197– 208. Bhushan S, Tchatalbachev S, Klug J, Fijak M, Pineau C, Chakraborty T, Meinhardt A. Uropathogenic Escherichia coli block MyD88-dependent and activate MyD88-independent signaling pathways in rat testicular cells. J Immunol 2008;180:5537 – 5547. Bhushan S, Hossain H, Lu Y, Geisler A, Tchatalbachev S, Mikulski Z, Schuler G, Klug J, Pilatz A, Wagenlehner FME et al. Uropathogenic E. coli induce different immune response in testicular and peritoneal macrophages: implications for testicular immune privilege. PLoS ONE 2011;6:e28452. Breton S, Brown D. Regulation of luminal acidification by the V-ATPase. Physiology (Bethesda) 2013;28:318 – 329. Buckles EL, Luterbach CL, Wang X, Lockatell CV, Johnson DE, Mobley HLT, Donnenberg MS. Signature-tagged mutagenesis and co-infection studies demonstrate the importance of P fimbriae in a murine model of urinary tract infection. Pathog Dis 2015, 73. Dacheux J-L, Belghazi M, Lanson Y, Dacheux F. Human epididymal secretome and proteome. Mol Cell Endocrinol 2006;250:36 – 42. Davidoff MS, Middendorff R, Mayer B, Holstein AF. Nitric oxide synthase (NOS-I) in Leydig cells of the human testis. Arch Histol Cytol 1995;58:17–30. Deckmann K, Filipski K, Krasteva-Christ G, Fronius M, Althaus M, Rafiq A, Papadakis T, Renno L, Jurastow I, Wessels L et al. Bitter triggers acetylcholine release from polymodal urethral chemosensory cells and bladder reflexes. Proc Natl Acad Sci USA 2014;22:8287– 8292. Haidl G, Allam JP, Schuppe H-C. Chronic epididymitis: impact on semen parameters and therapeutic options. Andrologia 2008;40:92– 96. Hinton BT, Galdamez MM, Sutherland A, Bomgardner D, Xu B, Abdel-Fattah R, Yang L. How do you get six meters of epididymis inside a human scrotum? J Androl 2011;32:558– 564. Holstein AF. Morphologische Studien am Nebenhoden des Menschen. In: Bargmann W, Doerr W (eds). Zwanglose Abhandlungen aus dem Gebiet der normalen und pathologischen Anatomie. Stuttgart, Germany: Thieme Medical Publisher, 1969. Hori S, Tsutsumi Y. Histological differentiation between chlamydial and bacterial epididymitis: nondestructive and proliferative versus destructive and abscess forming—immunohistochemical and clinicopathological findings. Hum Pathol 1995;26:402–407. Jelinsky SA, Turner TT, Bang HJ, Finger JN, Solarz MK, Wilson E, Brown EL, Kopf GS, Johnston DS. The rat epididymal transcriptome: comparison of segmental gene expression in the rat and mouse epididymides. Biol Reprod 2007;76:561 – 570. Johnston DS, Jelinsky SA, Bang HJ, DiCandeloro P, Wilson E, Kopf GS, Turner TT. The mouse epididymal transcriptome: transcriptional profiling of segmental gene expression in the epididymis. Biol Reprod 2005;73:404–413. Joseph A, Yao H, Hinton BT. Development and morphogenesis of the Wolffian/epididymal duct, more twists and turns. Dev Biol 2009;325:6–14. Jungwirth A, Giwercman A, Tournaye H, Diemer T, Kopa Z, Dohle G, Krausz C. European Association of Urology guidelines on Male Infertility: the 2012 update. Eur Urol 2012;62:324– 332. Kirchhoff C. Gene expression in the epididymis. Int Rev Cytol 1999; 188:133– 202. Krasteva G, Kummer W. ‘Tasting’ the airway lining fluid. Histochem Cell Biol 2012;138:365 – 383. Lang T, Dechant M, Sanchez V, Wistuba J, Boiani M, Pilatz A, Stammler A, Middendorff R, Schuler G, Bhushan S et al. Structural and functional integrity of spermatozoa is compromised as a consequence of acute uropathogenic E. coli-associated epididymitis. Biol Reprod 2013;89:59. Lang T, Hudemann C, Tchatalbachev S, Stammler A, Michel V, Aslani F, Bhushan S, Chakraborty T, Renz H, Meinhardt A. Uropathogenic
9
