Null Mutation of the Transcription Factor Inhibitor of DNA Binding 3 (Id3) Affects Spermatozoal Motility Parameters and Epididymal Gene Expression in Mice

Abstract

ID3 is a transcription factor that acts as a dominant-negative regulator of other transcription factors by sequestering them, thus rendering them unable to bind DNA. We have shown previously that ID3 is expressed in a unique, region-specific manner along the epididymis, a highly specialized tissue of the male reproductive tract that functions in the transport and maturation of spermatozoa. The goal of these studies was to test the hypothesis that ID3 plays a role in the epididymis in the region-specific regulation of gene expression that is responsible for establishing the microenvironment required to carry out sperm-related functions. The consequences of ID3 deficiency on epididymal histology and gene expression profiles, as well as spermatozoal motility parameters, were determined. Although ID3 deficiency (Id3^−/− mice) had no noticeable impact on epididymal histology, the targeted mutation adversely affected sperm motility parameters. Moreover, principal component analysis of microarray data indicated that the gene expression signatures for tissues obtained from Id3^−/− mice and their genotypic controls were distinct from each other in each epididymal region. The predominant effect of the Id3 null mutation was in the cauda region where the expression of many transcription factors, including Hoxb8 and Bclaf1, was markedly affected. ID3 may play an important role in the molecular circuitry involved in the establishment and maintenance of the region-specific differences in gene expression that are characteristic of the epididymis.

Transcription factor expression in the cauda epididymidis is significantly altered in animals bearing a null mutation of the Id3 gene thereby affecting gene expression and sperm motility.

INTRODUCTION

The epididymis is a component of the male excurrent reproductive tract; it is a highly specialized tissue that functions in the transport, maturation, protection, and storage of spermatozoa [1, 2]. The morphological and functional segmentation of the epididymis into four distinct regions is observed both histologically and at the molecular level by complex, region-specific gene expression [3, 4]. These unique profiles are presumably responsible for establishing the luminal microenvironments required for sperm-related functions [5, 6]. In spite of considerable effort, the precise molecular mechanisms that govern these events, as well as how the establishment of specific, regionalized expression of certain genes occurs, remain poorly defined.

The inhibitor of differentiation/DNA binding (ID) proteins consist of a family of four helix-loop-helix (HLH) transcription factors (ID1, ID2, ID3, and ID4) that lack a basic region, and therefore cannot exert their effects as transcriptional regulators through direct interaction with DNA [7–12]. Rather, IDs act as dominant-negative regulators of other transcription factors, including those of the ubiquitously expressed helix-loop-helix E box class, by forming quenching, non-DNA-binding, heterodimeric complexes [7, 13, 14]. This is thought to occur as the result of a higher binding affinity of ID proteins that prevents E proteins from interacting with each other. IDs have been shown to be implicated in multiple developmental, physiological, and pathophysiological processes [15–18]. These transcriptional regulators play a role in positively regulating proliferation and inhibiting differentiation in a wide variety of cell types and organ systems [19].

Although ID1, ID2, and ID4 are primarily expressed in smooth muscle cells surrounding epididymal tubules or in clear cells along the epididymis, ID3 is found primarily in the nuclei of principal cells [20]. Id3 was first identified as a gene induced in the early transcriptional response to growth factors and other signaling agents in mouse 3T3 cells [10]. Since then, it has been reported that ID3 is involved in vascular smooth muscle cell proliferation, skeletal muscle cell differentiation, osteogenesis, and neurogenesis, as well as tumor-induced angiogenesis [21–24]. Importantly, Id3 expression is not exclusive to proliferating cells; it can be found in terminally differentiated cells, such as Sertoli cells, where Id3 is induced by serum but not follicle-stimulating hormone [25]. Previously, we have shown that ID3 has a unique, highly region-specific expression profile along the adult epididymis, both at the mRNA and protein levels [20]. Principal cells are the primary cell type found along the epididymis, comprising up to 80% of the epithelial cell population [26]; it is this cell type that is primarily responsible for protein secretion into the lumen. Taken together, this suggests that ID3 may act as an important regulator of many downstream genes associated with the region-specific differences that are the hallmark of the epididymis.

If indeed ID3 is an important regulator of other differentially expressed genes along the epididymis, then one would predict that there would be important consequences on epididymal gene expression profiles if ID3 were removed from the system. Mice having a targeted deletion of the Id3 gene (Id3^−/−) have no obvious phenotypic malformations, although it was found that ID3 plays an essential regulatory role in thymocyte development by regulating the activity of E proteins during the final stages of thymocyte maturation [27]. We have obtained Id3^+/− animals and mated these to produce Id3-deficient animals.

The objective of these studies was to determine the effect of ID3 deficiency on epididymal histology and gene expression profiles, as well as on spermatozoal motility parameters.

MATERIALS AND METHODS

Animals

Founder mice used for this study were kindly provided by Dr. Cornelis Murre, University of California, San Diego, CA. The Id3 gene was inactivated in embryonic stem (ES) cells by gene targeting; the HLH dimerization domain was replaced by a neomycin resistance marker gene in a transcriptional orientation opposite to the Id3 gene. Heterozygous mutant ES cells were isolated and used to generate Id3 chimeric mice, and these were backcrossed to C57BL/6J mice to generate Id3 heterozygous mice [27]. Breeding heterozygous males and females produced mice of all three genotypes. Mice were housed at the McIntyre Animal Resources Centre, McGill University, and were maintained under controlled temperature (22°C) and lighting (12L:12D), with food and water provided ad libitum. The primers for the PCR used for genotyping have been reported by Rivera et al. [27]. Two PCRs were done on each sample to identify wild-type (Id3^+/+), heterozygous (Id3^+/−), and Id3-deficient (Id3^−/−) mice. All animal studies were conducted in accordance with the principles and procedures outlined in the Guide to the Care and Use of Experimental Animals prepared by the Canadian Council on Animal Care (McGill Animal Care Committee protocol no. 206).

Sperm Counts

Hemocytometric counts of spermatozoa were done as previously described [28]. A single cauda epididymidis from each animal was weighed and homogenized (Polytron, setting 6; Brinkmann Instruments Inc., Westbury, NY) for 2 × 15 sec periods, with a 10-sec interval, in 3 ml of 0.9% NaCl, 0.1% thimerosal, and 0.5% Triton X-100. All counts (n = 4) were made in duplicate and averaged.

Spermatozoal Motility

Spermatozoa from the cauda epididymides were used immediately for computer-assisted sperm analysis (CASA), as previously described [29]. Briefly, the left cauda epididymidis of each animal was clamped off proximally, then cut out and rinsed in prewarmed Hank medium M199 (Invitrogen Canada Inc., Burlington, ON) supplemented with 0.5% bovine serum albumin, pH 7.4, at 37°C. It was then minced in fresh medium, the remaining cauda epididymal tissue removed, and the sperm left to disperse for several minutes. Subsequently, an aliquot of the sperm suspension was placed into a prewarmed 2X-CEL 80-μm-deep glass sperm analysis chamber and analyzed using a Hamilton-Thorne IVOS automated semen analyzer (Hamilton-Thorne Biosciences, Beverly, MA). Each of four samples (n = 4) was assayed in triplicate. Tracks were digitally recorded at 60 Hz under 4× dark-field illumination. Analysis was completed using IVOS settings determined by The Jackson Laboratory (courtesy of Hamilton-Thorne). The parameters analyzed include percent motility, percent progressive motility, average path velocity (VAP), straight line velocity (VSL), curvilinear velocity (VCL), and amplitude of lateral head displacement (ALH), as well as a derived parameter, straightness (STR = VAP/VCL × 100%).

Tissue Preparation for Light Microscope Immunohistochemistry

The right testis and epididymis were removed and immediately immersed in Bouin fixative. After 15 min, the epididymides were cut longitudinally such that all regions (initial segment, caput, corpus, and cauda) could be viewed in a single section. Thereafter, the epididymides were dehydrated in alcohol, embedded in paraffin, sectioned at 5 μm, and mounted on glass slides.

Hematoxylin-Eosin Staining

Sections were dewaxed in xylene, hydrated in a series of descending ethanol solutions, and stained by a routine hematoxylin-eosin staining technique as described previously [30].

Immunohistochemistry

Sections were deparaffinized with xylene and hydrated through a series of graded ethanol solutions. During hydration, endogenous peroxidase activity was abrogated by treating the sections with 70% ethanol containing 1% (v/v) H₂O₂. Once hydrated, the sections were bathed in PBS containing glycine to block free aldehyde groups. Immunohistochemical staining of the sections was done using a Vectastain Elite ABC Kit (Vector Laboratories, Burlingame, CA). An affinity-purified rabbit anti-ID3 (1:1400; Santa Cruz Biotechnology, Santa Cruz, CA) polyclonal antibody was used. Nonspecific binding sites were blocked using 5% bovine serum albumin and goat serum blocking solution for 2 h. Reactions were visualized through the use of biotinylated goat anti-rabbit immunoglobulin G (IgG) secondary antibodies and diaminobenzidine tetrahydrochloride. Sections were counterstained with methylene blue, dehydrated in ethanol and xylene, and mounted with cover slips using Permount. Negative controls were treated the same way except that the primary antibody was preincubated with the corresponding blocking peptide (ID3; Santa Cruz Biotechnology).

RNA Extraction

Epididymides of Id3^+/+ and Id3^−/− mice were collected, divided into initial segment/caput and cauda regions, and immediately frozen in liquid nitrogen; the corpus regions were not used in further studies. These regions were selected for the analysis because the initial segment/caput regions express ID3 at relatively low levels in wild-type animals, whereas the cauda region shows markedly higher levels of expression [20]. Tissues were stored at −80°C until used for RNA extraction. Epididymal regions were crushed under liquid nitrogen, and total RNA was extracted and DNase treated using the RNeasy kit (Qiagen, Mississauga, ON), according to the manufacturer's instructions. The RNA quality was assessed by Genome Quebec using a Bioanalyzer (Agilent Technologies, Santa Clara, CA) and was quantified using a Nanodrop ND-1000 Spectrophotometer (Nanodrop Technologies, Wilmington, DE).

Gene Expression Analysis

Gene expression was assessed in the initial segment/caput and cauda epididymal regions of Id3^+/+ and Id3^−/− mice using Mouse Genome 230 2.0 microarrays (Affymetrix; n = 3–4 per region per genotype). Background subtraction and normalization of probe set intensities were done using the Robust Multi-array Analysis method [31], and only the genes with an expression value of 5 or higher were considered as expressed. The data were further normalized per gene to the median (GeneSpring v7.3; Agilent Technologies). The next filter applied was statistical significance between the two genotypes of each epididymal region (Id3^+/+ and Id3^−/−); this was tested by Wilcoxon-Mann-Whitney test with a P value cutoff set at 0.05. Transcripts from that list were then filtered for those that showed a minimum of a 2-fold change in expression. The NetAffx Analysis Center (Affymetrix) was used to correlate array results with annotation information, molecular function, and biological processes.

Quantitative Real-Time RT-PCR

Oligonucleotide primers were purchased from Qiagen for previously described sequences for Id3 (catalogue no. QT00248185), homeobox B8 (Hoxb8; catalogue no. QT00137830), Bcl2-associated transcription factor 1 (Bclaf1; catalogue no. QT00165396), and bone morphogenic protein receptor type 2 (Bmpr2; catalogue no. QT01057504). Quantitative PCR was done using the One-Step QuantiTect SYBR Green RT-PCR Kit (Qiagen) according to the manufacturer's instructions using a real-time fluorometric thermocycler (LightCycler; Roche Diagnostics, Laval, QC). Four to seven separate samples from each group (n = 4–7 per region per genotype) were assayed in duplicate, and all genes were normalized against Rn18s. The PCR thermal cycling conditions were as follows: reverse transcription at 50°C for 20 min, activation at 95°C for 15 min, followed by 50 cycles of denaturation at 94°C for 15 sec, annealing at 55°C for 30 sec, and extension at 72°C for 30 sec. Nonspecific amplification was monitored by melting curve analysis of each reaction.

Protein Extraction and Western Blotting

Epididymides were collected, divided into initial segment/caput and cauda regions, and immediately frozen in liquid nitrogen. Samples were placed in RIPA buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, and 50 mM Tris, pH 7.5) containing 10 μl/ml protease inhibitor cocktail and 20 μl/ml phosphatase inhibitor mix (Active Motif Inc., Carlsbad, CA). Samples were homogenized using an ultrasonicator (Sonics and Materials Inc., Newtown, CT) and centrifuged at 10 000 × g for 10 min at 4°C. The supernatants were used for immunoblotting. Proteins from each sample (25 μg; n = 4–5 per region per genotype) were resolved by 8% SDS-PAGE and then transferred onto equilibrated polyvinylidene difluoride membranes (Amersham Biosciences, Oakville, ON) by electroblotting. Membranes were blocked with 5% skim milk for 1 h at room temperature, and then probed at 4°C overnight with primary antibodies against ID3 (1:200; sc-490; Santa Cruz Biotechnology), HOXB8 (1:200; sc-82897; Santa Cruz Biotechnology), BCLAF1 (1:250; ab77989; Abcam, Cambridge, MA), and BMPR2 (1:500; ab10862; Abcam). Subsequent to TBS-T washes, membranes were probed with horseradish peroxidase-coupled donkey anti-rabbit or anti-goat IgG (1:2000). Membranes were developed by the enhanced chemiluminescence method, as per the manufacturer's instructions (Amersham Biosciences). The bands were quantified by densitometric analysis using a Chemi-Imager v5.5 imaging system (Alpha Innotech, San Leandro, CA). Each sample was normalized against β-actin and run in duplicate.

Statistical Analysis

Unless otherwise indicated, all statistical analyses were done using Student t-test. Nonparametric tests were applied when equal variance testing failed or when sample sizes were too small to formally test for normality. Data are presented as means plus or minus standard errors of the mean. For all analyses, the level of significance was set at P ≤ 0.05.

RESULTS

Animal and Tissue Weights

Null mutation of the Id3 gene elicited no apparent changes in animal weight or behavior (data not shown). Additionally, no significant differences in reproductive tissue weights were noted between Id3-deficient mice and their genotypic controls; however, the spleen weight was increased in Id3^−/− animals (Fig. 1).

FIG. 1.

Reproductive organ and accessory tissue weights in wild-type (dark gray bars) and ID3-deficient (light gray bars) mice. Data are shown as mean ± SEM. *P ≤ 0.05. Sem. Ves., seminal vesicles.

Spermatozoal Motility

There was no significant change in sperm count between the cauda epididymides of Id3^+/+ and Id3^−/− animals (data not shown). The effect of ID3 deficiency on multiple sperm motility parameters was assessed by CASA using a Hamilton-Thorne IVOS semen analyzer (Fig. 2). Although ID3-deficient males displayed the same proportion of motile sperm as wild-type animals, the proportion of these that were progressively motile showed a trend towards reduction in the null animals, although this was not statistically significant (P = 0.100; Fig. 2a). Interestingly, all three velocity parameters examined—VAP, VSL, and VCL—were statistically significantly reduced in sperm from ID3-deficient males compared with wild-type animals (Fig. 2b). Although no change was observed in the sperm motility straightness parameter, the amplitude of lateral head displacement was significantly reduced in sperm from Id3-deficient animals compared with their genotypic controls (Fig. 2c).

FIG. 2.

Spermatozoal motility parameters in wild-type (Id3^+/+; dark gray bars) and Id3 null mutant (Id3^−/−; light gray bars) mice. The parameters analyzed include (a) progressive motility (%); (b) average path velocity (VAP; the total distance along the smoothed average path for each sperm divided by the time elapsed, in μm/sec), straight line velocity (VSL; the straight line distance from the beginning to the end of each sperm's track divided by the time elapsed, expressed in μm/sec), and curvilinear velocity (VCL; the track speed, or total distance covered by each sperm divided by the time elapsed, in μm/sec); (c) a derived parameter, straightness (STR = VAP/VCL × 100%), as well as amplitude of lateral head displacement (ALH, in μm). Each of four samples (n = 4) was assayed in triplicate. Data are shown as mean ± SEM. *P ≤ 0.05.

Epididymal Histology and ID3 Expression Profile

Histological analysis of the epididymal epithelium did not reveal any apparent differences between Id3^+/+ and Id3^−/− animals (Fig. 3). ID3 was expressed in a unique region-specific manner in the epididymis; both mRNA and protein expressions were relatively low in the initial segment/caput region and highest in the cauda epididymidis (Fig. 4, a and b). Array data indicate only a 19% decrease in Id3 expression in the initial segment/caput region of Id3^−/− animals as compared with wild-type controls, whereas a 58% decrease is obtained by real-time PCR. This is not surprising considering the low level of Id3 expression in this region at baseline. Although the arrays showed a 71% decrease in Id3 expression in the cauda region of Id3-deficient animals, analysis by real-time PCR indicates a 99% decrease in expression (Fig. 4a). Importantly, these results are highly consistent with ID3 protein expression determined by Western blot (Fig. 4b). Interestingly, ID3 deficiency did not result in a compensatory increase in Id1, Id2, or Id4 mRNA expression, as indicated by analysis of expression arrays (Supplemental Fig. S1, all Supplemental Data are available online at www.biolreprod.org). Importantly, whereas ID3 is expressed in principal cells in the cauda epididymidis of wild-type animals, Id3 null mutant mice do not express significant levels of ID3 in any epididymal cell types or regions (Fig. 4, b–d).

FIG. 3.

Light micrographs showing sections of the adult mouse epididymis stained with hematoxylin-eosin. Initial segment sections (a and b), caput epididymidis sections (c and d), cauda epididymidis sections (e and f). a, c, and e), Id3^+/+ animals. b, d, and f) Id3^−/− animals. Original magnification ×40.

FIG. 4.

a) Relative expression of Id3 mRNA in the initial segment/caput and cauda epididymidis of wild-type (dark gray bars) and Id3-deficient (light gray bars) mice from microarray data (Array; n = 3–4 per region per genotype) and quantitative real-time RT-PCR (RT; n = 4–7 per region per genotype). b) Relative expression of ID3 protein by Western blot analysis in the initial segment/caput and cauda epididymidis of wild-type and ID3-deficient mice (n = 4–5 per region per genotype). Light micrographs showing sections of the adult mouse epididymis from Id3^+/+ (c) and Id3^−/− (d) animals stained with an antibody for ID3. Data are shown as mean ± SEM. *P ≤ 0.05. WT, wild type. Original magnification ×40 (c, d).

Impact of Id3 Null Mutation on Epididymal Gene Expression

We assessed the impact of ID3 deficiency on epididymal gene expression using whole-mouse genome Affymetrix 230 2.0 microarrays. Principal component analysis (PCA) was used to reduce the dimensionality of the gene expression data sets (n = 3–4 per group) to allow general relationships between experimental groups to be more easily discerned (Fig. 5). Consistent with previous reports indicating the region-specific expression of many genes along the epididymis [32–34], PCA indicated evident differences, representing approximately 70% of the variance, in the transcriptional profiles of wild-type initial segment/caput and cauda epididymal regions. The PCA also indicated that the epididymal gene expression signatures for the two genotypes examined (Id3-deficient and wild-type controls) were distinct from each other, occupying very different spaces within the PCA plot. Importantly, although the microarray chips representing the wild-type epididymal regions were clustered together within the PCA plot, those representing the gene expression signatures of the null epididymides were more scattered, indicating less concerted changes in gene expression. These less concerted changes were noted particularly in the cauda region of the ID3-deficient animals.

FIG. 5.

Principal component analysis was used to reduce the dimensionality of the gene expression data sets (n = 3–4 per group) to allow general relationships between experimental groups (epididymal region and genotype) to be more easily discerned. IS/Cap, initial segment/caput.

Scatter plots allow the visualization of the correlation of all transcripts between two microarray chips, whether from the same or different epididymal regions and genotypes. The tight distribution of data points around the identity line demonstrated that transcript expression levels were comparable across experimental replicates (Fig. 6, a and c). Plotting the expression of transcripts from wild-type initial segment/caput epididymidis against ID3-deficient initial segment/caput epididymidis transcripts showed a similar tight distribution (Fig. 6b). A more scattered profile indicated changes in transcript levels due to ID3 deficiency in the cauda epididymidis (Fig. 6d).

FIG. 6.

Scatter plots allowing the visualization of the correlation of all probe sets between two microarray chips. Two microarray chips from the initial segment/caput (a) and cauda (c) epididymal regions of wild-type animals. Correlation of probe sets from wild-type initial segment/caput epididymidis plotted against ID3-deficient caput epididymidis transcripts (b). Scattered profile indicating changes in transcript levels due to ID3 deficiency in the cauda epididymidis (d).

Of the 45 101 probe sets on the microarray chip, 23 422 (51.9%) were considered expressed in the wild-type initial segment/caput epididymidis. Of these, 23 058 were also expressed in the ID3-deficient tissue, with an additional 705 probe sets expressed uniquely in the null group (Fig. 7a). A total of 24 445 probe sets (54.2%) were expressed in the wild-type cauda epididymidis, and 869 were solely expressed in the Id3 null cauda epididymidis (Fig. 7a). Similar percentages of expressed probe sets on microarray chips were found to be expressed in the mouse epididymis by Johnston et al. [35]; however, different microarray chips, segmentation schemes of the epididymis, as well as filtering criteria were used, making direct comparisons not doable. Although only 41 probe sets (33 up-regulated and 8 down-regulated) were significantly affected in the initial segment/caput epididymidis (Supplemental Table S1), 76 probe sets (55 up-regulated and 21 down-regulated) were significantly affected in the null cauda epididymidis as compared with genotypic controls (Supplemental Table S2), showing changes greater than 2-fold increases and 50% decreases (Fig. 7b). Interestingly, we observed that of the 63 probe sets that were uniquely affected in the cauda region showing greater than 2-fold increases or 50% decreases, 19% (12 of 63 probe sets) were annotated as regulators of transcription, compared with only 7% (2 of 28 probe sets) in the initial segment/caput epididymidis (Fig. 7c). Additionally, 3 of the 13 probe sets that were altered in both epididymal regions were transcription factors.

FIG. 7.

Number of probe sets expressed in the wild-type and ID3-deficient mouse initial segment/caput and cauda epididymidis (a). Gene expression changes in the initial segment/caput and cauda epididymidis as a result of ID3 deficiency. Numbers of transcripts showing changes greater than 2-fold increases and 50% decreases are included (b). These transcripts are further represented in (c) changes in transcription factor expression in each epididymal region.

Effect of Id3 Targeted Mutation on Probe Sets Grouped by Biological Function

The predominant effect of the Id3 null mutation on gene expression in the epididymis was in the cauda region, where the expression of many transcription factors was markedly affected. A particularly noteworthy pattern of expression was observed when looking at the microarray mRNA expression profiles of the majority of these transcripts; they showed changes greater than a 2-fold increase or 50% decrease in the cauda epididymidis and smaller fold changes in the initial segment/caput regions (Fig. 8a). Conversely, the principal grouping of altered probe sets based on biological function in the initial segment/caput region was composed of the gene ontology (GO) annotation corresponding to the response to stress (12.2%; 5 of 41 probe sets); some of these probe sets were also altered in the cauda region (Fig. 8b).

FIG. 8.

Probe sets with GO annotations corresponding specifically to the regulation of transcription (a) or the response to stress (b) showing changes greater than 2-fold increases or 50% decreases in either one or both of the epididymal regions examined of wild-type (dark gray bars) and ID3-deficient (light gray bars) mice. *Changes greater than 2-fold increases or 50% decreases.

Hoxb8, Bclaf1, and Bmpr2 mRNA and Protein Expression

Quantitative real-time PCR was done in order to assess the levels of expression of Hoxb8, Bclaf1, and Bmpr2 mRNAs, two transcription factors, and one regulator of Id3, respectively, in the initial segment/caput and cauda epididymal regions of wild-type and ID3-deficient mice. The patterns of expression corresponded with those generated by the microarrays, with the effect of genotype being most pronounced in the cauda epididymidis. ID3 deficiency caused a 1.5-fold increase in Hoxb8 mRNA expression in the initial segment/caput region, whereas expression was increased by more than 3.6-fold in the cauda (Fig. 9a). Whereas Bclaf1 expression increased by 1.7-fold in the more proximal region, we observed a 2.7-fold increase in the cauda (Fig. 9c). A slightly higher fold-change due to ID3 deficiency was detected in the cauda epididymidis for Bmpr2 expression (Fig. 9e). We also analyzed the protein levels of HOXB8, BCLAF1, and BMPR2 in wild-type and Id3 null initial segment/caput and cauda regions. HOXB8 protein levels were not affected by the Id3 null mutation in the initial segment/caput region, but showed a significant, 2.1-fold increase in the cauda region (Fig. 9b). BCLAF1 protein expression was significantly increased in null tissues of both initial segment/caput and cauda regions (Fig. 9d). BMPR2 protein expression was also significantly increased in null tissues of both initial segment/caput and cauda regions; however, a marked change of more than 1.7-fold due to genotype was observed in the cauda region (Fig. 9f).

FIG. 9.

Hoxb8, Bclaf1, and Bmpr2 relative expression compared using microarrays (Array; a, c, and e; n = 3–4 per group) and quantitative real-time RT-PCR (RT; a, c, and e; n = 4–7 per group) in the initial segment/caput and cauda epididymidis of wild-type (dark gray bars) and ID3-deficient (light gray bars) mice. Effects of ID3 deficiency on HOXB8, BCLAF1, and BMPR2 protein expression (b, d, and f; n = 4–5 per group). Data are shown as mean ± SEM. *P ≤ 0.05.

DISCUSSION

Although the role of ID3 has been explored in many biological contexts, including skeletal muscle repair, immune cell fate determination, and atheroprotection, its role in the epididymis has not been examined previously. In these studies, we assessed the impact of ID3 deficiency on spermatozoal motility, epididymal histology, and epididymal gene expression. Despite normal reproductive organ weights, the spleen weight was increased in Id3^−/− animals; this is not surprising, since ID3 is a known negative regulator of B-cell differentiation and the white pulp of the spleen contains lymphoid nodules rich in these lymphocytes. Interestingly, although epididymal sperm counts remained unaltered in the presence of the Id3 null mutation, we observed a consistent and statistically significant decline in the ALH as well as the three velocity parameters, VAP, VSL, and VCL. This indicates that ID3 and/or its binding partners may be implicated in the acquisition of full spermatozoal motility potential.

A PCA of whole-genome microarrays from both Id3^+/+ and Id3^−/− initial segment/caput and cauda epididymides suggests that ID3 deficiency results in important consequences on epididymal gene expression. This is particularly true in the cauda, where ID3 expression within the epididymis is the highest. Therefore, it is not surprising that roughly twice the number of probe sets were significantly affected by ID3 deficiency in the cauda region compared with the initial segment/caput regions, where ID3 expression is relatively low.

Of particular interest, we observed greater variation/differential magnitudes of gene expression changes in cauda epididymides from Id3-deficient animals (in both our microarray and qRT-PCR experiments). This may be an indication of general gene dysregulation resulting from the Id3 null mutation in this region. Perhaps the regulatory action of ID3 and subsequent activation of specific gene expression programs depend on the cellular context and the presence or absence of collaborating transcription factors. This pattern of differential magnitudes of response to the targeted disruption of the Id3 gene in the cauda is not conserved at the protein level, indicating that perhaps these are more tightly regulated.

Furthermore, in the cauda epididymidis many of the probe sets that were altered by ID3 deficiency are implicated in physiological development (proliferation, differentiation, and apoptosis), whereas the predominantly affected group in the initial segment/caput regions, based on biological function, is the response to stress transcripts. Interestingly, microarray results of primary aortic vascular smooth muscle cells showed that the expression of many genes involved in immune trafficking and apoptosis, as well as cellular movement, proliferation, and adhesion, were modulated significantly in Id3^−/− mice as compared with wild-type controls [36]. Although few studies have examined the impact of Id3 deletion on global gene expression in a particular tissue as we have done here, many have ascertained the transcriptional profiles of specific target genes or described a range of phenotypic outcomes in the absence of Id3. Humoral immune responses and B-cell proliferation were shown to be impaired in ID3-deficient mice; ID3 has been shown to be a key player, along with E2A and PAX5, in the B-cell activation program [37, 38].

We selected three probe sets that were significantly altered due to ID3 deficiency and validated the results generated from our microarray analysis using qRT-PCR; we confirmed the impact of ID3 deficiency on the expression of Hoxb8, Bclaf1, and Bmpr2, selected on the basis of their relationship, whether direct or indirect, with ID3.

Precise spatial and temporal expression of the homeobox (HOX) family of transcription factors is of paramount importance in axis positioning and tissue determination during embryonic development. This family of regulatory genes is involved in the patterning of the anterior-posterior axial skeleton in the developing embryo [39]; deviations from a wild-type axial skeletal pattern, or homeotic transformations, result when the expression or function of these genes is altered [40–42]. Although in the past it was more common to see studies exploring Hox gene expression in developing embryonic tissues, more recent studies have implicated Hox gene expression in the maintenance of segmental phenotypes and functions in several adult tissues. The anterior-posterior axis of the male reproductive tract comprises the epididymis, vas deferens, seminal vesicles, coagulating gland, prostate gland, and bulbourethral glands; a striking anterior-to-posterior Hoxb13 expression gradient is observed in the rat [43]. Unique Hox code expression has also been shown to explain regional tissue specification; therefore, the fact that Hox genes have been detected in the epididymis is not surprising, considering the segmented nature of this organ [44–46]. Importantly, Hoxa10 and Hoxa11 knockout mice are infertile, and the epididymis shows homeotic transformations; the cauda transforms into the corpus and the vas deferens transforms into a cauda, respectively [47]. A number of additional Hox genes, including Hoxa9, Hoxa10, and Hoxa11, as well as their paralogues in the D cluster, are expressed along the adult epididymis, with generally increasing mRNA profiles from proximal to distal regions of the duct [48]. Our transcript of interest, Hoxb8, has not, to our knowledge, been studied previously in the epididymis. It is, however, detected in the mouse brain [49], where it may be implicated in segmental homeostasis of this organ in the adult. In line with this notion, HOXB8 may be involved in the maintenance of epididymal regionalization in the adult mouse. Id3 deletion may affect the expression of other unknown factors involved in this maintenance, leading to a compensatory increase in Hoxb8 expression. Alternatively, ID3 deficiency may lead to a decreased sequestration of activating signals regulating the expression Hoxb8.

Bcl2-associated factor 1 (BCLAF1) is a transcription factor originally shown to interact with antiapoptotic members of the BCL2 family. Sustained overexpression of BCLAF1 in HeLa cells induced apoptosis and repressed transcription [50]. Initially described as a proapoptotic factor and transcriptional repressor, BCLAF1 has also been implicated in various processes not typically linked to BCL2 functions [51–54]. Although BCLAF1 is dispensable for cell viability, it is necessary for postnatal viability in mice; analysis of Bclaf1^−/− mice has revealed that this gene plays an essential role in both maintaining immune homeostasis of T- and B-cell lineages and proper lung development [51]. In the Bclaf1 knockouts, lung development arrests before the differentiation of bronchoalveolar epithelial cells; however, the defect is the result of a general effect on cell growth or apoptosis. Interestingly, recent evidence suggests that BCLAF1 may serve a regulatory role in pre-mRNA splicing or mRNA processing; Bclaf1^−/− fibroblasts have altered levels of spliced transcripts derived from adenovirus E1A pre-mRNA compared with wild-type cells [52–54]. Interestingly, GO annotations corresponding specifically to immune function and transcription activation were affected by ID3 deficiency in the initial segment/caput and cauda regions of the epididymis, respectively; therefore, this may account for the up-regulation of BCLAF1 in our model.

A strong line of evidence indicates that ID3 is a bone morphogenetic protein (BMP) target gene [55, 56]; Id3 contains BMP receptor-specific phosphorylated mothers against decapentaplegic homolog (SMAD) response elements. Signaling of BMPs occurs via heteromeric complexes of type I and type II serine/threonine kinase receptors, and both receptor types are known to be essential for signal transduction [57–62]. Previous studies have established a role for BMPs in the maintenance of epididymal integrity [63]. Targeted mutation of the Bmp7 and Bmp8a gene (Bmp7^+/−; Bmp8a^−/−) results in significant degeneration of the distal caput and cauda epididymal epithelium, as well as decreased sperm counts and motility [64]. Consequently, decreased litter numbers and sizes result from crossing mutant males with wild-type females. This compromised fertility is also seen in Bmp4^+/− mutants on a largely C57BL/6 background [65]. Interestingly, the epididymal defect observed in these animals, consisting of a thinner, vacuolated epithelium with a saw-toothed appearance, was seen only in the corpus epididymidis. Recent studies have shown that skeletal muscle regeneration after injury was both delayed and diminished in Id1^+/−Id3^−/− mice [66]. In wild-type animals, expression of both Id1 and Id3 was up-regulated after injury, as were upstream components of the BMP signaling pathway, including Bmpr2 and Smad1/5/8. Id3 mRNA transcription and protein translation were induced following BMP2 and BMP4 treatment of MEMM cells [67]. BMP2 acts via ID2 or ID3 and SMAD3 to synergistically regulate follicle-stimulating hormone β subunit (Fshb) transcription with activins [68]. In line with these studies, our results indicate that ID3 deficiency in the initial segment/caput and cauda epididymides resulted in an increase in BMPR2 expression both at the mRNA and protein levels. Collectively, these data offer a model in which BMP regulates and acts through ID3.

Thus, targeted disruption of the Id3 gene impacted spermatozoal motility as well as epididymal gene expression profiles without affecting epididymal histology. Gene expression changes in the mouse initial segment/caput and cauda epididymidis regions as a result of ID3 deficiency indicate that ID3 may be an important regulator of epididymal gene expression. Further, the Id3 null mutation resulted in dysregulation of the expression of Hoxb8, Bclaf1, and Bmpr2, genes with established roles in transcriptional regulation.

Taken together, these data support the hypothesis that ID3 may sit near the top of a hierarchical transcriptional cascade and exert its effects by controlling the expression of transcription factors that, in turn, coordinate discreet gene expression programs within the epididymis. ID3 may contribute to the molecular basis for the region-specific differences in gene expression in the epididymis, a hallmark of the tissue.