| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Male Reproductive Tract; |
Departments of Paediatrics,3 Physiology and Pharmacology,4 and Biology,5 The University of Western Ontario, Children's Health Research Institute, London, Ontario, Canada N6C 2V5
ABSTRACT
The seminal vesicle is a male accessory sex organ that develops from segments of the Wolffian duct adjacent to the urogenital sinus. It produces most of the seminal plasma in both humans and rodents. To date, very few transcription factors have been linked to the development and differentiation of seminal vesicles. In this study, we have examined the role of basic helix-loop-helix (BHLH) B8 transcription factor expressed at high levels in the adult seminal vesicle and during seminal gland differentiation. Immunofluorescent studies indicate that BHLHB8 is expressed within the epithelial layer of the seminal layer of the seminal vesicle following branching morphogenesis but prior to full maturation of cell morphology and function. Analysis of mice that do not express BHLHB8 (Bhlhb8–/–) indicates no deficiency in the initial development of the seminal vesicle. However, morphological and ultrastructural analysis indicates disruption of the epithelial cellular architecture. The seminal vesicle epithelial layer of 2-mo-old Bhlhb8–/– mice shows extensive cellular degeneration based on the appearance of reduced microvilli, altered granule size, and dilated endoplasmic reticulum and Golgi apparatus. The seminal vesicle epithelial cells also degenerate prematurely, as evidenced by disruption of nuclear architecture and significant accumulations of autophagic bodies. These results identify BHLHB8 as a regulator in establishing and stabilizing the secreting epithelial cells of the seminal vesicle.
autophagy, developmental biology, gene regulation, knock-out, male reproductive tract, Mist1, mouse, seminal vesicles, transcription
The seminal vesicles are paired accessory reproductive glands that exist in both male humans and rodents. They produce most of the seminal fluid that affects sperm capacitation and motility, and defects may have an effect on sperm fertility [1–4]. Seminal vesicles consist of an epithelial cell layer with surrounding mesenchyme and smooth muscle layer. Seminal vesicles first appear as buds from the Wolffian ducts next to the urogenital sinus between Embryonic Day (E) 15.5 and 18.5. Through branching morphogenesis, the rudiment expands significantly until approximately 2 wk after birth, at which time most of the branching is complete [5]. At this time, the epithelial cells begin differentiation and assume a highly secretory function. Through hypertrophy, the gland continues to grow into adult life.
One of the key events in seminal vesicle development is the signaling that occurs from the tissue mesenchyme that drives development, branching, and differentiation of the epithelial layer of the gland [6, 7]. This process is dependent on signaling through androgen receptors located in the mesenchyme [8]. Fibroblast growth factor (FGF) signaling from the surrounding mesenchyme is essential for proper expansion and branching of the gland [9–12], and the seminal vesicle branching mutant, svs, is caused by a deletion in FGF receptor 2 (Fgfr2) [13]. In addition, the targeted ablation of Hoxa10, Hoxa13, or Hoxd13 leads to incomplete invagination and branching of seminal vesicles [14–16]. The decrease in seminal vesicle branching is not accompanied by alterations in the epithelial cell layer, suggesting that, while the initial determination of the epithelium is dependent on mesenchyme-derived signals, there are factors that independently regulate seminal vesicle epithelial differentiation [14]. To date, no factor has been identified as being specifically important for seminal vesicle secretory epithelial differentiation.
BHLHB8, formerly referred to as MIST1 or DIMMED [17], belongs to the basic helix-loop-helix (bHLH) family of transcription factor proteins, and is expressed at high levels in all serous-secreting exocrine cells characterized to date, including pancreatic acinar cells, salivary and lacrimal gland acinar cells, lactating mammary acini, and chief cells of the stomach [18, 19]. Immunohistochemical analysis also reveals high levels of expression in adult seminal vesicles. Within the pancreas and salivary glands, targeted ablation of Bhlhb8 (Bhlhb8–/–) leads to disorganization of the acinar cells and alterations in gene expression [20–22]. In addition, Bhlhb8–/– pancreatic acini have a defect in calcium handling and regulated exocytosis [20, 23]. Bhlhb8–/– mice are particularly sensitive to pancreatic injury, and exhibit a decreased ability to activate the endoplasmic reticulum (ER) stress response [24]. While the analysis of the transcriptional activity of BHLHB8 so far has revealed only one putative transcriptional target, Gjb1 encoding connexin32, the complex phenotype observed in Bhlhb8–/– serous exocrine glands, suggests a multigenic disruption [20, 22]. To date, the seminal vesicles of Bhlhb8–/– mice have not been characterized. Given past findings that BHLHB8 functions to promote complete cell maturation in a number of serous exocrine cell types [20, 21, 25], we hypothesized that BHLHB8 is necessary for the complete differentiation of cell morphology and function of the secretory epithelium of seminal vesicles.
All protocols were approved by the Animal Care Committee at The University of Western Ontario, and mice were handled according to the regulations stipulated by the Canadian Council on Animal Care. Mice deficient for Bhlhb8 (Bhlhb8–/–) were generated by targeted homologous recombination, as previously described [21]. The targeting vector contained the LacZ coding sequence for β-galactosidase inserted 8 bp upstream of the translational start site for Bhlhb8, and analysis to date has indicated that β-galactosidase expression appropriately mimics Bhlhb8 expression in mice that carry one targeted allele and one wild-type (WT) allele (Bhlhb8+/LacZ [18]). Bhlhb8–/– mice were backcrossed for at least seven generations into a C57BL/6 background and are, therefore, 99.5% genetically similar to the C57BL/6. All mice were fed with a standard diet and water ad libitum.
Immunofluorescence and β-Galactosidase Histochemistry
Seminal vesicles were dissected from WT, Bhlhb8+/LacZ, and Bhlhb8–/– mice at various postnatal (Postnatal Days [PNs] 1, 10, 15, and 28) and adult (2, 6, and 12 mo) time points, and embedded in Cryomatrix (Fisher), as previously described [26]. Tissues were cut to 6-µm thickness using a Shandon cryostat, mounted onto charged slides (Fisher Scientific), and stored at –20°C until use. β-Galactosidase histochemistry was performed as previously described [18] on PN 1–28 and adult tissue sections from Bhlhb8+/LacZ seminal vesicles. Histochemistry was repeated on at least two replicates for each time point. For immunofluorescence (IF) analysis, sections were fixed with 10% formalin for 10 min, washed several times with PBS, permeabilized with 0.1% TritonX-100, and blocked with 5% BSA + 0.1% TritonX-100 for 30 min. Sections were incubated with a rabbit antibody raised against BHLHB8 (1:500, [21]) diluted in blocking solution for 1 h at room temperature. As a negative control, primary antibodies were omitted. Sections were then washed three times with blocking solution and incubated in a goat anti-rabbit fluorescein isothiocyanate (Sigma, Oakville, ON) secondary antibody diluted 1:100 in blocking solution for 1 h at room temperature. Following incubation, sections were again washed three times in PBS; a third wash, containing 4',6'-diamidino-2-phenylindole (Sigma), was undertaken to stain the nuclei. Slides were mounted with Vectashield (Vector Laboratories, Burlington, ON) and IF was visualized with a Leica upright fluorescent microscope. All images were captured using the Openlab imaging system (Quorum Technologies, Guelph, ON).
Bhlhb8–/– and WT mice were killed at either 2 or 12 mo of age and perfused with 0.5% glutaraldehyde and 4% formaldehyde. Fixed seminal vesicles were embedded in Polybed/araldite embedding medium. In some cases, seminal vesicles were removed without perfusion and immediately placed in 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.2. Following overnight fixation, tissue was postfixed in 1% osmium tetroxide in 0.1 M phosphate buffer. Semi-thin sections were stained with toluidine blue before further processing for electron microscopy. Silver-gold ultra-thin sections were stained with 2% uranyl acetate in ethanol and Reynolds lead citrate. Sections were examined with a Philips Model 410 electron microscope at 60 kV. Secretory granule number was determined by counting the granules in at least 20 seminal vesicle epithelial cells from three different WT and Bhlhb8–/– mice.
Seminal Fluid Collection and Protein Identification
The seminal fluid was removed from 12-mo-old WT and Bhlhb8–/– seminal vesicles and added to an equal volume of PBS. Samples were vortexed for 10 sec and then centrifuged at 14 000 rpm for 5 min to produce soluble, viscous, and insoluble fractions. The soluble fraction was removed and stored at –80°C. Equal volumes (10 µl) of seminal fluid samples were resolved by SDS-PAGE on 12% polyacrylamide gels. Following electrophoresis, gels were stained with Coomassie blue for 1 h and then destained overnight with several washes of 50% methanol/10% acetic acid for evaluation of protein banding. For protein identification, bands were excised, reduced with Dithiothreitol (DTT), and alkylated with iodoacetamide. In-gel tryptic digestion was then carried out as previously described [27]. Mass spectrometry (MS) analysis was performed on a Micromass Q-TOF Global mass spectrometer equipped with a Z-spray source and operating in positive ion mode. Peptides were separated using nanospray liquid chromatography (precolumn: 5 x 0.5 mm C18 [LC Packings]; analytical column: 75 µm x 150 mm [Waters]) on a Waters CapLC system. Data were acquired using MassLynx 4.0 and calibrated using an MS/MS spectrum of Glu-fibrinopeptide B. Data were processed using the Peptide auto function of MassLynx 4.0, and pkl files were used to search Mascot, GPM, and PEAKS.
RNA Isolation, RT-PCR, and Northern Blot
Seminal vesicles were isolated from 2- or 12-mo-old WT and Bhlhb8–/– mice and RNA isolated using Trizol (Invitrogen, Burlington, ON) per manufacturer's directions. To assess Bhlhb8 expression, 1 µg of total RNA was reverse transcribed using random primers and Superscript Reverse Transcriptase II (Invitrogen). The cDNA was amplified through 35 rounds of PCR with denaturing at 94°C for 30 sec, priming at 60°C for 40 sec, and extension at 72°C for 60 sec.
Primers specific for Bhlhb8 were: forward primer: 5' CGGGATCCGGCTGTCAGCGACTTGAT 3'; reverse primer: 5' ATTCGAGGCCGTACGCGC 3'. Primers specific for Actb (β-actin) were: forward primer: 5' ATGGAGAAGATCTGGCAC 3'; reverse primer: 5' CCATCATGAAGTGTGACG 3'.
For Northern blot analysis, 40 µg of total RNA was resolved on a 1% agarose gel and blotted to polyvinylidene fluoride membrane (GE Healthcare, Baie D'urie, PQ). The expression of gap junction membrane channel protein beta 1 (Gjb1), FXYD domain-containing ion transport regulator 3 (Fxyd3), LacZ, and 18S was determined by Northern blotting as described previously [24]. The probe for Gjb1 was provided by Dr. G.M. Kidder (University of Western Ontario), while Fxyd3 was obtained from MRC gene service (clone no. 6485767).
To contrast fertility between WT and Bhlhb8–/– male mice, 2-mo-old male mice were mated with three 2- to 4-mo-old C57 BL6 WT females for 1-wk intervals over a 6-mo period. By the end of the study, male mice were 8 mo of age. Each morning, female mice were examined for a copulatory plug. After 7 d, females were separated from males and the males mated to other female mice. The number of females with copulatory plugs was determined as a percentage of all females used in the study. All female mice were monitored for several weeks, and pregnancy rate was determined as the number of females that gave birth compared with the number of females with copulatory plugs. The number of mice per litter was determined on PN1.
To examine the spatial expression of BHLHB8 within seminal vesicles, IF analysis with a BHLHB8-specific antibody was performed on seminal vesicle sections from 2-mo-old mice (Fig. 1, A and B). BHLHB8 expression was limited to the epithelial lining of the tissue with no expression in the surrounding mesenchyme. In addition, it appears that BHLHB8 was expressed in all epithelial cells at this time point. To determine when BHLHB8 is first expressed during seminal vesicle development, IF for BHLHB8 was performed at four different stages of differentiation (Fig. 1, C–J). At PN1, seminal vesicles had begun to develop and were observed as two buds protruding from the Wolffian duct adjacent to the urogenital sinus. At PN10, the seminal vesicle was undergoing branching morphogenesis, while, by PN15, branching had ceased but the epithelial lining was still not morphologically or functionally mature [5]. By PN28, the epithelial cells of the seminal vesicle were functional, and reached adult maturity by PN40 [28]. IF analysis for BHLHB8 revealed limited nuclear expression within only a few cells of the epithelial lining of the seminal vesicles beginning at PN10 (Fig. 1E). No expression was observed at PN1 (Fig. 1C), indicating that BHLHB8 expression does not precede initial determination of the seminal vesicle. What little staining that was observed at PN1 was not nuclear and was also observed in negative control slides, suggesting that it represented background staining. By PN15 (Fig. 1G), the majority of epithelial cells were BHLHB8 positive, and no expression of BHLHB8 was observed in the surrounding mesenchyme. This pattern of expression was similar at PN28 (Fig. 1I), except that, at this time point, all epithelial cells were BHLHB8 positive. No expression of BHLHB8 was observed outside of the epithelial layer.
|
To assess the expression of the Bhlhb8 gene in an alternative but complimentary way, we characterized the expression of β-galactosidase in mice heterozygous for the Bhlhb8-targeted allele. This allele has the LacZ gene inserted in place of the Bhlhb8 coding region 8 bp upstream of the start codon [21]. Therefore, expression of LacZ should mirror Bhlhb8 gene activity, which we have previously confirmed [18]. In corroboration with the IF data, no LacZ expression was observed at PN1 (Fig. 1K), indicating that the Bhlhb8 gene was not active at this time point. By PN10, only a very small percentage of cells in the epithelial layer were X-gal positive (Fig. 1L). By PN15, LacZ expression was more extensive, observed throughout the epithelial lining, which was also observed in adult organ (Fig. 1, M and N). As observed for the BHLHB8 protein, expression of the Bhlhb8-targeted allele was not expressed outside of the epithelial layer. Based on this expression analysis, the expression of BHLHB8 delineated seminal vesicle secretory epithelial differentiation, predating the complete maturation and functional ability of these cells.
To assess the importance of BHLHB8 function in seminal vesicle differentiation, we examined mice that do not express Bhlhb8 (Bhlhb8–/–). These mice are viable and fertile, and seminal vesicles are readily observed, indicating that BHLHB8 is not necessary for initial determination or branching morphogenesis to occur. Examination of Bhlhb8–/– seminal vesicles by IF (Fig. 2, A–D) and RT-PCR (Fig. 2E) revealed a complete absence of Bhlhb8 expression, although the Bhlhb8 locus is still active based on LacZ expression (Fig. 2F). Our previous work had documented a link between BHLHB8 function and the expression of Gjb1 (gene encoding Connexin32 [Cx32]) in pancreatic and salivary gland acinar cells [22]. Therefore, we compared the accumulation of Gjb1 RNA in seminal vesicles of 2-mo-old WT and Bhlhb8–/– mice by Northern blot analysis (Fig. 2F). Consistent with a role for BHLHB8 in Gjb1 expression, the absence of BHLHB8 resulted in a concomitant decrease in Gjb1 expression. We have also documented a decrease in pancreatic Fxyd3 expression in Bhlhb8–/– tissue [24]. In contrast to what occurs in the pancreas, there was an increase in Fxyd3 mRNA accumulation in Bhlhb8–/– seminal vesicles (Fig. 2F). These results confirm that the loss of BHLHB8 protein leads to an alteration in the gene expression pattern of seminal vesicle epithelial cells.
|
We next compared the tissue and cell morphology of seminal vesicles in 2-mo-old WT and Bhlhb8–/– mice (Fig. 3). Low-magnification microscopy of toluidine blue-stained, semi-thin sections revealed no overt differences in structure between the two groups (Fig. 3, A–C), but the cells within the Bhlhb8–/– epithelium were not uniform in appearance. Both WT and Bhlhb8–/– epithelium presented a simple columnar epithelium with a regular row of nuclei. However, many Bhlhb8–/– cells had altered staining intensity, either of the nuclei (Fig. 3B) or cytoplasm (Fig. 3C). At higher magnification, the cellular appearance of Bhlhb8–/– epithelial cells supports the concept that these cells have disrupted cellular architecture (Fig. 3, D and E). There is reduced ruffling of the luminal border and the appearance of atypical granules that extend throughout the cell.
|
To gain a better understanding of potential cellular defects in the secretory epithelium cells of the seminal vesicle, cell ultrastructure was examined by transmission electron microscopy (TEM; Fig. 4). TEM applied to WT samples revealed large secretory granules containing an outer clear area and inner, electron-dense core. In Bhlhb8–/– epithelial cells, these granules were still apparent, but appeared to be much smaller. Quantitative analysis of normal secretory granules within the epithelial cells indicated an increase in the number of granules within Bhlhb8–/– epithelium (28.3 ± 1.4 granules/cell section) compared with WT epithelium (20.3 ± 1.0 granules/cell section) (Fig. 4F). In addition, a second type of granule was observed within Bhlhb8–/– epithelial cells (Fig. 4, B and E). These granules were electron dense and may represent degenerative vesicles typical of damaged cells. At the same time, Bhlhb8–/– epithelial cells displayed distended ER and limited microvilli suggestive of decreased exocytosis ability. Therefore, morphological comparisons indicated that the absence of BHLHB8 resulted in altering the cellular phenotype of seminal vesicle epithelial secretory cells.
|
Previous analysis of pancreatic tissue in Bhlhb8–/– mice identified a progressive phenotype in this organ as the animal ages [21]. To determine if similar changes occur in Bhlhb8–/– seminal vesicles, tissue from 12-mo-old WT and Bhlhb8–/– mice was compared morphologically (Fig. 5). Toluidine blue staining revealed extensive disruption of the seminal vesicle epithelium. Numerous large granules were observed within the majority of epithelial cells (Fig. 5B). In addition, a number of cells within the Bhlhb8–/– seminal vesicle epithelium displayed a densely stained nucleus and degenerative cellular appearance (Fig. 5C). TUNEL analysis indicated that there was no increase in apoptotic cells within the Bhlhb8–/– epithelium, so we would conclude that these were necrotic cells (data not shown).
|
Ultrastructural analysis of the seminal vesicle epithelium in WT and Bhlhb8–/– 12-mo-old animals confirmed the extensive alterations in cell morphology in Bhlhb8–/– tissue (Fig. 6). Similar to observations in younger animals, the secretory epithelial cells in Bhlhb8–/– mice contained fewer and smaller secretory granules. In addition, there were numerous granules of a secondary nature that extended throughout the cells. These abnormal granules exhibited a swirled appearance typical of autophagic vesicles and, in some cases, were as large as the nuclei (
7 µm diameter). In many cases, epithelial cell nuclei had lost their spherical structure and were also more electron dense, although they still displayed obvious areas of euchromatin and heterochromatin.
|
The morphological changes associated with the absence of BHLHB8 expression indicated that deficits in seminal vesicle function should occur. Toluidine blue staining also suggested that the seminal fluid composition is different between WT and Bhlhb8–/– mice. The luminal contents in the WT seminal vesicle were separated from the epithelial lining, leaving a clear zone between the cells and the fluid (Fig. 7A). The seminal fluid in Bhlhb8–/– seminal vesicles did not show this same clear zone, suggesting a difference in the seminal fluid composition and decreased viscosity (Fig. 7B). To determine whether any differences exist between the seminal plasma of WT and Bhlhb8–/– mice, seminal plasma was isolated from 10-mo-old mice and equal volumes resolved by SDS-PAGE. Following Coomassie blue staining, WT and Bhlhb8–/– seminal fluid exhibited different banding patterns. Numerous additional bands were observed in the Bhlhb8–/– samples, including strong bands at 53, 40, and 37 kDa. In addition, a 25-kDa band in the WT samples was significantly reduced in Bhlhb8–/– samples. To confirm the identity of these proteins, MS was performed. From this analysis, 1,4-
-D-glucan glucanohydrolase (53 kDa; salivary and hepatic precursor for -amylase), seminal vesicle secretory protein 2 (i.e., semenoclotin; 40 kDa), and carcinoembryonic antigen (CEA)-related cell adhesion molecule 10 (CEACAM10; 37 kDa) were all identified as having increased accumulation in Bhlhb8–/– seminal fluid. Conversely, WT seminal fluid contained higher accumulations of seminal vesicle secretory (SVS) proteins IV and V (25 kDa). Therefore, the seminal fluid composition of the Bhlhb8–/– mice was altered, indicating the importance of BHLHB8 in the attainment of complete seminal vesicle (SV) epithelial cell maturation and function.
|
To determine if the alterations in seminal fluid affect the copulation rate, formation of copulatory plug, and reproductive fecundity, WT and Bhlhb8–/– male mice were mated to WT female mice over a period of several months. In that time (2–6 mo of age), there was no significant difference in the number of copulatory plugs identified, the number of litters generated (Fig. 7D), or the number of offspring per litter (data not shown). This implies that the changes in Bhlhb8–/– seminal fluid did not affect male reproductive function and fertility.
The seminal vesicle is an accessory organ of the male reproductive tract derived from the Wolffian ducts and dependent on androgen signaling for differentiation and maturation. During development, signals from the surrounding mesenchyme, including FGF signaling, are critical for determination and branching of the seminal vesicle [5, 12]. However, little is known about the epithelial cell-specific factors that are necessary for differentiation. In this study, we have identified a mediator of epithelial cell maturation. This protein, BHLHB8, is a transcription factor necessary for the complete maturation of several exocrine cell types, including pancreatic acinar cells [21], salivary gland acinar cells, lacrimal cells [20], and chief cells of the stomach [20, 25]. Consistent with studies of BHLHB8 in other organs, the absence of BHLHB8 in the seminal vesicle epithelium leads to altered cellular organization, functional deficits in secretion, and, over time, degeneration of the seminal vesicle.
BHLHB8, which has been named MIST1 (rodent) or DIMMED (Drosophila) in the past, belongs to a large family of transcription factors that share a common bHLH motif that is responsible for DNA binding [17, 29, 30]. This protein belongs to the B class of bHLH proteins in that BHLHB8 exhibits a restricted pattern of expression. Typically, class B bHLH proteins heterodimerize with class A bHLH proteins, which have a ubiquitous pattern of expression, such as E12. This dimerization leads to binding of a consensus "E-box" DNA sequence, CANNTG, and target gene activation. However, recent evidence suggests that the preferential binding by BHLHB8 is as a homodimer, and overexpression of a dominant negative form of the protein leads to similar phenotypic deficits that are observed in the Bhlhb8–/– mouse [31, 32].
The initial function attributed to BHLHB8 was as a negative regulator of skeletal muscle differentiation [33]; however, expression analysis revealed no expression of BHLHB8 during myogenesis in vivo, and the Bhlhb8–/– mouse exhibits no myogenic defects [21]. While it is possible that there may be functional redundancy between a closely related bHLH protein and BHLHB8, to date, no paralogue has been identified. The most similar bHLH protein to BHLHB8 is neurogenin 3, which exhibits less than 70% identity at the protein level within the bHLH domain alone. Therefore, it is likely that the primary role of BHLHB8 is in serous exocrine cell differentiation.
Consistent with a role for BHLHB8 in seminal vesicle epithelial cell differentiation, BHLHB8 expression predates the terminal differentiation and functional maturation of the SV epithelium. Therefore, the role of BHLHB8 is downstream of FGF signaling and occurs after determination of the seminal vesicle epithelium. This is similar to the expression pattern and putative role for Bhlhb8 in other developing organs—the gene is activated after determination but before differentiation [19, 21]. As observed in other serous-secreting cell types, the expression of BHLHB8 is maintained at high levels throughout adulthood, possibly suggesting a role for BHLHB8 in stabilizing the mature phenotype. In addition, BHLHB8 is expressed throughout the entire epithelial layer, so it is unlikely to correlate to the different columnar cells previously described by Badia et al. [34].
In the absence of BHLHB8, the seminal vesicle epithelium is altered morphologically, similar to previous observations in the pancreas and salivary glands, and suggests that alterations in granule morphogenesis are a primary defect due to the absence of BHLHB8 [20]. There is also distension of the ER and Golgi apparatus and pockets of less electron-dense cells throughout the epithelial lining in 2-mo-old mice. These cells may represent clear cells that have been previously characterized in boar seminal vesicles, and are believed to represent new epithelial cells that have not fully differentiated [34]. Interestingly, at later time points, we observed a larger number of densely stained cells, which are believed to represent older degenerating cells. Therefore, it would appear that Bhlhb8–/– seminal vesicle epithelial cells are not fully mature, and our results indicate that the seminal vesicles in Bhlhb8–/– animals degenerate at a higher rate. Interestingly, the phenotype of the seminal vesicle epithelium in Bhlhb8–/– mice is remarkably similar to mice that have undergone chronic exposure to alcohol. Mice fed an ethanol-laced diet for 6 mo develop altered cell morphology with decreases in cell and granule size, reduced microvilli, and the appearance of autophagic vesicles [35, 36]. This phenotype is presumably due to malnutrition in the alcohol-fed mice.
Recent evidence also suggests that the appearance of autophagic vesicles is the result of accumulating misfolded proteins [37, 38]. The increased load of improperly folded proteins triggers the unfolded protein response (UPR), and the distended appearance of the ER in Bhlhb8–/– seminal vesicle epithelial cells also suggests UPR-inducing conditions [39]. The UPR serves to reduce the level of proteins within the cell via two mechanisms [40]. First, there is an increase in proteins that promote ER-associated protein degradation. Second, there is a decrease in general protein translation through the phosphorylation of eIF2. We have documented a decreased ability for Bhlhb8–/– pancreatic acini to stimulate the ER stress response and UPR following initiation of pancreatic disease [24], and recent studies indicate that the Bhlhb8 gene is a direct target of the transcription factor, X-box binding protein 1 [41], a key factor in the ER stress response. While the Bhlhb8–/– phenotype indicates that the ER stress pathway is compromised to some extent, future experiments will need to specifically examine Bhlhb8–/– seminal vesicles for the expression of chaperones and ER degradation-enhancing -mannosidase-like protein, which are mediators of the UPR [40].
Whether the phenotypes we have observed are directly due to the loss of BHLHB8 transcriptional control, or to secondary effects, remains to be determined. Given the fact that similar molecular processes are involved in the exocytosis pathway of all serous-secreting exocrine cells, it seems likely that BHLHB8 regulates the expression of proteins specifically involved in this pathway, and we and others have identified altered expression of several genes involved in regulated exocytosis [20, 23]. However, restoration of these proteins in the context of Bhlhb8–/– acinar cells does not rescue the phenotype [20]. An additional molecular target of BHLHB8 is the Gjb1 gene, which encodes Cx32 [22]. Cx32 is the only connexin identified to date to be expressed in the secretory epithelial lining of the seminal vesicle [42, 43]. The expression of Cx32, which is involved in direct intercellular communication via gap junctions, is greatly diminished, supporting previous work that suggests that BHLHB8 is a positive regulator of Gjb1 gene expression in serous exocrine tissue. Recent reports indicate that Cx32 can bind to discs large homolog (DLGH) 1, a member of the membrane-associated guanylate kinase scaffolding proteins that are involved in epithelial polarization and cell-cell adhesion [44]. While a scenario including Cx32 and DLGH1 could provide an attractive mechanism by which the loss of BHLHB8 leads to disrupted cell organization and cell communication, it is unlikely that an alteration in Cx32 expression alone would account for the phenotype observed in Bhlhb8–/– mice. To date, there have been no studies describing seminal vesicle defects in Gjb1–/– mice, and a targeted ablation of Dlgh1 results in complete absence of the seminal vesicle, but no disruption in cell-cell complexes [44]. Gjb1–/– mice exhibit significant phenotypic abnormalities in organs (i.e., liver and brain) in which BHLHB8 is not normally expressed [45, 46], as well as relatively moderate defects in pancreatic acini compared with the Bhlhb8–/– phenotype [47]. Most likely, the phenotype of the Bhlhb8–/– mouse is the result of multiple molecular changes. Microarray analysis comparing the gene expression patterns of seminal vesicles from WT and Bhlhb8–/– mice will be necessary to fully understand the molecular changes that occur following loss of BHLHB8 transcriptional activity.
While our data confirmed a defect in the molecular and cellular makeup of the seminal vesicle epithelium, there was evidence of functional defects as well. The seminal fluid from Bhlhb8–/– mice have increased accumulation of 1,4--D-glucan glucanohydrolase, semenoclotin, and CEACAM10, and decreased accumulation of SVS proteins IV and V compared with WT mice. Semenoclotin is involved in formation of the copulatory plug [48], and recent reports suggest that it maintains sperm in an uncapacitated state [4], while CEACAM10 enhances sperm motility in vitro [49]. Therefore, the increased accumulation of these two proteins may counteract each other. Both CEACAM10 and semenoclotin production are dependent on androgen activity [49, 50], so the changes that we observed in Bhlhb8–/– seminal vesicles were likely not due to loss of androgen signaling. Also, while it may be attractive to suggest that we have identified a transcriptional regulator for these proteins in BHLHB8, it is more likely that alteration in their accumulation in the seminal fluid is a secondary defect, resulting from abnormal epithelial cell secretion and maturation.
Surprisingly, all of the deficits observed in the seminal vesicle epithelium of the Bhlhb8–/– mouse do not result in alterations in fertility. Seminal vesicle fluid is responsible for coagulation of the ejaculate, can alter sperm motility and capacitation, and acts as an immunosuppressant in the female reproductive tract (reviewed in [51]). Increased viscosity of the seminal fluid decreases sperm motility [52], and altered semenogelin degradation leads to premature capacitation [53], both resulting in decreased fertility. However, these studies were carried out in human samples, which may have a decreased tolerance to changes in the seminal fluid composition. Alternatively, it is possible that the mice involved in the fertility studies were too young (2 and 8 mo of age) to observe an effect on fertility. Possibly, older Bhlhb8–/– mice would have decreased reproductive capacity, especially because the phenotype appears to become more severe with age. Alternatively, the fertility of Bhlhb8–/– mice may be impaired under stressful conditions. Since Bhlhb8–/– seminal vesicles share similar phenotypic abnormalities to mice fed an alcohol-laced diet, likely they should exhibit the same effects on fertility. Cicero et al. [54] suggest that decreased reproductive ability is linked to exposure to alcohol; however, these findings were for rats treated with an acute exposure to ethanol. Studies that reflect a change in seminal vesicle epithelium are for chronically fed rodents and, in these cases, no affect on male reproductive behavior was identified [55]. Finally, it is possible that the extent of the phenotype is not severe enough to prevent seminal vesicle function. Specific removal of the entire seminal vesicle results in reduced reproductive ability, but not complete infertility [56]. Therefore, mild alterations to seminal vesicle function may have no effect on fertility. In fact, a 30% loss in sperm motility and 90% decrease in spermatozoa counts do not affect fertility rates in rats treated with chemotherapeutic agents [57, 58].
In conclusion, this study identifies a novel transcription factor involved in seminal vesicle development specifically linked to epithelial cell differentiation. Further work is underway to elucidate the molecular pathways that BHLHB8 regulates at the transcriptional level.
ACKNOWLEDGMENTS
The authors are grateful to Drs. Gabriel DiMattia, Thomas Drysdale, Gerald Kidder, and Andrew Watson for their thoughtful input into the study. In addition, they thank Dr. Gilles Lajoie for his assistance with mass spectroscopy.
FOOTNOTES
1Supported by operating grant MOP 53083 from the Canadian Institutes of Health Research (CIHR) and the Natural Sciences and Engineering Research Council of Canada. C.P. is supported by a salary award from the CIHR. 
Correspondence: 2Christopher Pin, Dept. of Paediatrics, University of Western Ontario, Children's Health Research Institute, 5th Floor, Victoria Research Laboratories, London, ON, Canada N6C 2V5. FAX: 519 685 8186; e-mail: cpin{at}uwo.ca
Received: 12 July 2007.
First decision: 15 August 2007.
Accepted: 21 September 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
