| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
research-article |
Biological Technologies,3 Molecular Profiling and Biomarker Discovery, Wyeth Research, Cambridge, Massachusetts 02140
Contraception,6 Women's Health & Musculoskeletal Biology, Wyeth Research, Collegeville, Pennsylvania 19426
Departments of Urology4 and Cell Biology,5 University of Virginia Health Science System, Charlottesville, Virginia 22908
ABSTRACT
Regional differences along the epididymis are essential for the establishment of the luminal environment required for sperm maturation. In the current study, 19 morphologically distinct segments of the rat epididymis were identified by microdissection. Total RNA was isolated from each segment and subjected to microarray analysis. Segmental analysis of epididymal gene expression identified more than 16 000 expressed qualifiers, whereas profiling of RNA from whole rat epididymis identified approximately 12 000 expressed qualifiers. Screening a panel of normal rat tissues identified both epididymal-selective and epididymal-specific transcripts. In addition, more than 3500 qualifiers were shown to be present and differentially upregulated or downregulated by more than fourfold between any two segments. The present study complements our previous segment-dependent analysis of gene expression in the mouse epididymis and allows for comparative analyses between datasets. A total of 492 genes was shown to be present on both the MOE430 (mouse) and RAE230_2 (rat) microarrays, expressed in the epididymis of both species, and differentially expressed by more than fourfold in between segments in each species. Moreover, in-depth quantitative RT-PCR analysis of 36 members of the beta defensin gene family showed highly conserved patterns of expression along the lengths of the mouse and rat epididymides. These analyses elucidate global gene expression patterns along the length of the rat epididymis and provide a novel evaluation of conserved and nonconserved gene expression patterns in the epididymides of the two species. Furthermore, these data provide a powerful resource for the research community for future studies of biological factors that mediate sperm maturation and storage.
epididymis, gene regulation
The mammalian epididymis is the site of post-testicular sperm maturation and storage and has been the subject of recent, comprehensive reviews [1, 2]. The conventional division of the epididymis, proximally to distally, into caput, corpus, and cauda regions has often been used in studies of epididymal biology, including studies of ion transport [3, 4], protein secretion/absorption [5, 6], gene expression [7, 8], and various aspects of sperm maturation [9, 10]. These regions are often further divided into proximal and distal halves for further definition (e.g., proximal or distal caput), although for nearly half a century the epididymides of a number of species have also been divided into zones based on histology [11–13]. The caput, corpus, and cauda regions of the epididymis can also be divided into intraregional segments, i.e., coilings of the single epididymal tubule into lobules bounded by connective tissue septae [14–16]. The potential functional importance of intraregional segments has been suggested by studies showing that individual gene or protein expression is often sharply bounded by the segment borders [7, 17, 18].
Recently, such findings have been broadly validated by microarray analyses of all ten intraregional segments of the mouse epididymis [19]. That study demonstrated widespread, segmental regulation across the organ, with over 2000 genes being shown to be differentially regulated by more than fourfold in at least two different segments. Cluster analysis revealed large sets of genes that are highly expressed only in specific segments, and principal component analysis of all gene expression revealed that certain segments (e.g., segments 1, 3, and 7) segregated from all the other segments [19].
The mouse is a species that is commonly used in epididymal studies, whereas the rat is also a commonly used model that offers advantages for certain types of investigations. For example, luminal fluids can be obtained by in vivo micropuncture of the rat epididymis with no contamination by fluids from other compartments [1, 20, 21], whereas this is an extremely difficult procedure in the mouse. Luminal fluid collection allows for assessment of secreted proteins and this, coupled with segmental gene and protein expression data, offers the possibility of detecting the secretion of known proteins at a much finer level than was previously possible. Moreover, it also offers the possibility of detecting novel secreted proteins and their underlying, potentially segment-specific gene transcription. In the present study, we examined the rat epididymis for intraregional segmentation, dissected the segments for RNA extraction, and used microarray technology to analyze segmental gene expression in the rat epididymis. Analysis of individual segments allows for: 1) increased sensitivity, thereby providing an improved ability to identify epididymal-specific transcripts that may identify novel contraception targets; 2) construction of a map detailing the location of expression of thousands of genes in the epididymis, which will facilitate the cloning of genes of interest and the identification and purification of native epididymal proteins; and 3) the identification of common biological processes that are conserved between the mouse and rat epididymides. Furthermore, the segmental gene expression data for the rat epididymis have been made available on the web (http://mrg.genetics.washington.edu) as a companion database to the already posted mouse epididymal transcriptome data.
Protocols for the use of animals in these experiments were approved by the University of Virginia and Wyeth Research Animal Care and Use Committee and were in accordance with the National Institutes of Health standards established by the Guidelines for the Care and Use of Experimental Animals. Adult, male Sprague-Dawley rats were obtained from the University of Virginia Vivarium sources and maintained on a 12L:12D cycle with food and water ad libitum. The rats were anesthetized by i.p. injection of urethane (1 g/kg body weight). Unilateral castration was performed following ligation of the spermatic and vasal arteries. The epididymis was microdissected from the efferent ducts and vas deferens, and for convenience of dissection, was bisected at the narrowest point of the corpus region. Both epididymal halves were immediately placed in separate Petri dishes that contained ice-cold saline, and were defatted with sharp dissection. This and all subsequent manipulations were performed with the use of a dissecting microscope, while maintaining the dissection dish and medium on ice at all times. One operator proceeded to microdissect individual segments from the caput and proximal corpus regions, while another investigator microdissected individual segments from the distal corpus and cauda regions, as described below. The epididymal tunica albugenia were removed using sharp microdissection. Removal of the tunica and blunt microdissection along the plane of the segment-dividing connective tissue septae resulted in the separation of all detectable epididymal segments. The exact location of the segments and the track of the connective tissue septae borders varied somewhat between animals, although the general appearance of the segments and their septae was relatively consistent. As each segment was isolated, it was placed immediately in RNALater (>10x tissue volume; Ambion, Austin, TX) in a 1.5-ml Eppendorf tube on ice. Segments 1–11 and 13–19 were isolated as whole segments. Segment 12 was divided into approximately equal halves by the original bisection of the organ. These segments were designated 12 proximal (12P) and 12 distal (12D), yielding a total of 20 tissues obtained from each microdissected epididymis. Within 30 min of epididymal extirpation, all tissues were dissected and placed in separate 1.5-ml microcentrifuge tubes that contained more than 10x volume of RNALater. Following dissection of the first epididymis from each animal, the contralateral testis and epididymis were extirpated as described above, and the animal was then killed. The second epididymis was defatted and dissected as above, and the isolated tissues were combined with the tissues isolated from the first epididymis. The pooled tissues from both epididymides comprised one sample for subsequent RNA extraction. When all of the segments comprising one sample were collected, the samples were immediately stored at –80°C. This procedure was repeated until 5–8 samples of each segment were collected for RNA extraction. Mouse segments were isolated from adult C57B/6 mice, as described previously [19]. In total, five samples, representing 2–5 isolated segments, were collected for each of the ten mouse segments.
Immediately prior to processing, rat epididymal segments were thawed at room temperature and the RNALater was removed. Samples were washed twice with 1 ml of ice-cold TRIzol (Gibco BRL, Gaithersburg, MD). The tissue was homogenized with a PowerGen 700 automatic homogenizer (Fisher Scientific, Hampton, NH) in 600 µl of TRIzol. Total RNA was purified by chloroform extraction and further purified with an RNAeasy column (Qiagen, Valencia, CA). For all samples, the concentration of RNA was determined by absorbance at 260 nm (NanoDrop, Wilmington, DE) and quality was determined using an Agilent Bioanalyzer (Agilent, Palo Alto, CA). Mouse epididymal segments were microdissected as previously described [19] and placed in RNAlater. Tissue samples were homogenized in RLT buffer (Qiagen) that contained 1% beta-mercaptoethanol. Lysates were then layered on CsCl2 (5.7 M) and centrifuged overnight at 35 000 rpm. RNA pellets were resuspended in water and further purified with the RNAeasy isolation kit (Qiagen). DNase was added to all samples on the column and eluted with water.
Five micrograms of total RNA were used to generate biotin-labeled cRNA using an oligonucleotide T7 primer in a reverse transcription reaction followed by in vitro transcription with biotin-labeled UTP and CTP. Aliquots of 10 µg of cRNA were fragmented and hybridized to RAE230 2.0 arrays (Affymetrix, Santa Clara, CA) representing more than 31 000 transcripts. Hybridized arrays were stained according to the manufacturers protocols on a Fluidics Station 450 and scanned on an Affymetrix scanner 3000. All array images were visually inspected for defects and quality. Arrays with excessive background, low signal intensity, or major defects within the array were eliminated from further analysis. This resulted in the number of replicates shown in Table 1. Signal values were determined using Gene Chip Operating System 1.0 (GCOS; Affymetrix). For each array, all probe sets were normalized to a mean signal intensity value of 100. The default GCOS statistical values were used for all analyses. Signal values and absolute detection calls were imported into the Genesis 2.0 (GeneLogic, Gaithersburg, MD) or Decision Site 8.1 (Spotfire, Somerville, MA) software.
|
A qualifier was considered to be detectable if the mean expression in any segment was greater than 50 signal units and the percentage of samples with a Present (P) call, as determined by the GCOS default settings, was greater than or equal to 67%. Normalized signal values were transformed to log-base10 and ANOVA was performed. A qualifier was considered to be segmentally regulated if the difference between two segments met the following criteria: 1) the qualifier had to be detected in at least 67% of the samples of at least one of the segments; 2) the fold-change between at least two segments was at least 4.0; and 3) the ANOVA test gave P 
0.01. These conditions were met by 3946 qualifiers, which were used for further analysis. For some computations, the expression values for each qualifier were normalized to a mean of zero and a standard deviation of 1 (z-score normalization). This allowed a direct comparison of patterns within the data without respect to absolute expression levels.
Identification of Defensin Genes
The sequences of all the rat and mouse beta defensin genes were downloaded from GenBank [22]. The mouse defensin gene family consists of 53 members, including three psuedogenes, and the rat defensin gene family consists of 44 members, including one psuedogene. BLAST searches and sequence alignment identified 36 unambiguous orthologs among the mouse and rat defensins.
Rat to Mouse Qualifier Mapping
Orthologous genes were identified as follows: 1) probe sets were matched to gene symbols using the cross-species array comparison (version 12/18/2005, available at www.affymetrix.com). Mouse gene symbols were mapped to rat gene symbols using Homologene (build 46.1) [23], and 6198 orthologous genes were identified that had representation on both the mouse and rat chips. Multiple qualifiers for each species may map to a single gene. Since qualifiers that represent the same gene often show similar expression patterns, and in order to simply the analysis, the qualifier with the highest expression value within a gene cluster for each species was selected for further analysis.
Qualifiers expressed in any segment were screened for tissue selectivity by comparison with a Wyeth Research database of the RAE230 genechip profiling data from 21 normal rat tissues. The data were acquired from a source generated by Wyeth Research (Cambridge, MA) using standardized procedures and internal controls to minimize variation. Each tissue comprised between 3 and 12 replicas. The collection included tissue samples from the adrenal gland, bladder, brain, colon, embryo, eye, heart, kidney, liver, lung, ovary, oviduct, pituitary gland, prostate, salivary gland, small intestine, spleen, stomach, testis, thymus, and uterus. All of the transcriptional profiling data were normalized to a mean signal intensity value of 100 in GCOS. Qualifiers were considered epididymis-selective if the following conditions were met: 1) the qualifier had to have a 67% P call in at least one epididymal segment; 2) the expression of the qualifier in the epididymis was greater than 50 signal units; and 3) the mean expression of the qualifier in any segment was threefold higher than the mean expression in any of 21 other tissues. Qualifiers were considered to be epididymis-specific if the following conditions were met: 1) the qualifier was epididymis-selective; 2) the qualifier was never detected (0% P call) in any of the 21 other tissues; and 3) the qualifier was expressed at a level of less than 50 signal units in all 21 tissues.
The genes were annotated and characterized using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) [24]. Using the functional annotation tool, each gene was annotated based on a number of sources, including but not limited to GO ontology, Kegg pathways, and Biocarta pathway. The Fisher exact test was used to determine if a particular category was overrepresented in the dataset. Related genes are usually annotated with the same functional categories and therefore, functional categories can be grouped based on the co-occurrence of sets of genes, allowing a reduction in the number of unique categories and aiding in the identification of biologically relevant processes. The list of genes that were differentially expressed in the epididymis of both species was annotated and the functional categories were grouped. To aid in interpretation, representative categories from selected groups were chosen. The complete list of categories is available in Supplemental Table S5 (all supplemental tables are available online at www.biolreprod.org).
Genes that were differentially expressed in both the mouse and rat were clustered in hierarchical clusters using GeneCluster [25] and visualized using Treeview [25]. The raw expression data were log2-transformed. Each gene was z scored normalized and clustered by hierarchical clustering using an average linkage algorithm with a correlation similarity metric.
Primers for Real-Time RT-PCR Analyses
Primers were chosen from the published sequences for Clu (clusterin), Cst8 (cystatin 8, also known as cysteine-related epididymal-specific protein or Cres), Gpx3 (glutathione peroxidase 3), and Lcn5 (lipocalin 5 or epididymal retinoic acid binding protein, ERABP). Primers were designed for all the mouse and rat defensin family members. All primers were designed using Primer Express (Applied Biosystems, Foster City, CA). The primers chosen for each transcript and the accession number for each sequence are available in Supplemental Table S1.
RNA Expression Analysis by Real-Time RT-PCR
Briefly, targeted mRNA from rat or mouse epididymal segments was analyzed by RT-PCR using 2.5 ng total RNA in a final volume of 25 µl that contained 300 nM of the target-specific PCR primers (Invitrogen, Carlsbad, CA), 100 nM of a fluorescently-labeled oligonucleotide probe (Eurogentec, San Diego, CA), and 1x Quantitect Probe RT-PCR Mix (Qiagen). Reverse transcription was performed for 30 min at 48°C, followed by 40 thermal cycles of 30 sec at 94°C and 1 min at 60°C using a 7900HT Fast Real-Time PCR System (Applied Biosystems). Target mRNA was normalized to 18S ribosomal RNA as determined using TaqMan Ribosomal RNA Control Reagents (Applied Biosystems).
Segmentation of the Rat Epididymis
The rat epididymis consists of 19 distinct, consistently identified segments. Slight variations were observed between epididymides with respect to the size and shape of specific segments. This is similar to what has been reported previously for the mouse epididymis [19], a schematic of which is shown in Figure 1A for comparison with the present findings in the rat (Fig. 1B) along with an example of the histological appearance of the proximal rat segments (Fig. 1C). While the gross anatomical structure of the epididymides (caput, corpus, cauda) was conserved between the two species, the absolute number and the arrangement of individual segments differed.
|
Microarray Analysis of the Rat Epididymal Transcriptome
Microarray analysis of the individual segments of the entire rat epididymis identified 16 360 distinct qualifiers (Table 1). This represented a comprehensive view of the rat epididymis and resulted in the detection of approximately 53% of the qualifiers on the array. Segmental analysis of the rat epididymis led to the identification of 3532 expressed qualifiers that were not detectable in the whole epididymal RNA samples, an increase in sensitivity of 28%. On average, an additional 1237 expressed qualifiers were identified in each segment (from 941 qualifiers in segment 8 to 1618 in segment 19; Table 1). The microarray data illustrate that there was no significant change in the number of qualifiers expressed within any of the segments (Table 1).
The analyses described in this study identified 9864, 3000, 329, and 171 qualifiers that were differentially expressed 2-, 5-, 50-, and 100-fold or more, respectively, between any two segments. This is consistent with the large number of genes that are differentially expressed between mouse epididymal segments [19] and in the different regions of the rat [26] and human [27] epididymis. Owing to the large number of differentially expressed qualifiers (9864 showing twofold expression), we chose to characterize further and analyze only those qualifiers that were differentially expressed by at least fourfold (P < 0.01 by ANOVA) between any two segments. This allowed the analysis of a reasonable number of qualifiers while focusing on those genes that are highly and significantly differentially expressed between individual segments. These criteria resulted in 3946 qualifiers, representing 24 % of the genes expressed in the epididymis. A heat map of the expression patterns across the epididymis in which qualifiers are ordered by the segment with maximal expression shows that the majority of these genes show a clear, punctuated, expression pattern in which relatively high expression is restricted to a segment or to a subset of adjacent segments (Fig. 2A).
|
Hierarchical clustering grouped individual segments with similar expression profiles. The clustering indicates that the first two segments are distinct from each other and from succeeding segments, which suggests a relatively sharp transition in segment function as luminal spermatozoa enter the epididymis. The clustering dendrogram (Fig. 2B) also shows that corpus segments 12 and 13 organize more closely with the cauda segments than with the caput segments. Based on this analysis, it appears that the 19 segments can be combined into nine groups with similar expression patterns. The groupings consist of segment 1, segment 2, segments 3–6, segments 7–9, segments 9–11, segments 12–13, segments 14–15, segments 16–17, and segments 18–19. As with segment 9, occasional samples fell outside their group but this variation is expected given the variation in size and shape of individual segments from different animals.
Of the 16 360 qualifiers detected in the epididymis, we determined that 423 were expressed threefold higher in the epididymis compared to 21 other normal tissues (epididymal selective qualifiers), and 110 were not detectable in any of the other 21 normal tissue epididymis-specific qualifiers (Table 1 and data not shown). Included in these 110 qualifiers are the known epididymal specific genes Cst11 [28], Adam7 [29], and Lcn5 [30]. We also identified a number of epididymis-specific genes the roles of which have yet to be established, including Adora1 for adenosine A1 receptor [31], Gdf15 for growth differentiation factor 15 [32], and Kcng3 for potassium voltage-gated channel subfamily G, member 3 [33]. The complete list of 110 qualifiers, their segment distributions, and expression levels are available in Supplemental Table S2.
Since many of the genes showed segment-dependent expression (Fig. 2A), we selected a number of these genes for qRT-PCR analysis, to confirm the results obtained from the microarray. Cst8 [34], Lcn5 [35], Clu [7], and Gpx3 [36] are examples of the consistency between our microarray and qRT-PCR data (Fig. 3), as well as between the microarray data and the expression values reported in the literature.
|
Comparison of Rat and Mouse Epididymal Gene Expression Profiles
We compared the expression patterns in the rat to our previous data generated from the mouse [19]. An accurate comparison of the genes expressed in the rat and mouse epididymis required the identification of orthologous sequences present on both the MOE430 mouse array used in the previous work [19] as well as the RAE230 array used in this study. At least two complicating factors make cross array analyses challenging. First, multiple qualifiers from each array may represent distinct transcripts from a single gene and second, the different arrays may not necessarily represent the same transcripts or region of a gene. Therefore for further comparison analysis, we moved from a qualifier-based approach to a gene-based approach. For each qualifier on the mouse and rat arrays, a representative unigene accession number was assigned. Mouse unigene clusters were then mapped to rat unigene clusters through the use of HomoloGene [23]. Six thousand one hundred ninety-eight orthologous genes were identified as being represented on both arrays. In many cases multiple qualifiers from a single species may map to the same gene, and in these cases a single representative qualifier with the highest expression signal for each orthologous gene was selected to represent the gene in further analyses. Of the 6198 orthologous genes, 3481 genes were determined to be expressed in the epididymides of both species (as defined in the Materials and Methods) and 492 (13%) were differentially expressed in both species (threefold change and ANOVA P < 0.05); these genes were used for further analysis.
Gene Regulation in the Mouse and Rat Epididymides
Many genes in the mouse and rat show segment-selective expression. K-means clustering was used to determine genes with conserved expression patterns between species. Using K-means clustering, we grouped the 492 genes that were differentially expressed in both species into twenty patterns. The groups were arbitrarily segregated into two classes; one group showed similar patterns of expression based on relative segment position and the other group showed dissimilar patterns of expression (Fig. 4). Thirteen groups representing 379 genes were identified as having similar expression patterns across the entire epididymis while eight groups representing 112 genes had dissimilar expression patterns. Figure 4 shows the heat map of the expression in each segment and the average expression in each segment from each of the individual clusters. Genes with similar patterns of expression are shown (Fig. 4, A and B). Only twelve clusters are shown in Figure 4B, since one cluster had only one member. As one proceeds along the length of the epididymis, the majority of these genes show similar changes in expression in the mouse and rat. For example, cluster 1 represents 37 genes whose expression is limited to the proximal segments in both species, and cluster 12 represents 100 genes whose expression is limited to the terminal segments in both species. Several clusters (clusters 3, 4, 7, and 9) represent slightly different but similar expression patterns, while their relative expression patterns are similar. Figure 4, C and D represents the seven patterns of expression for the 112 genes that have different relative expression patterns in the mouse and rat epididymis.
|
Examples of specific genes with similar expression patterns in the rat and mouse are shown in Figure 5. Cystatin TE (Fig. 5A) shows expression in the initial segments of both species, Ca4 (Fig. 5B) shows expression in the caput and corpus (but not in the cauda) of both species, and Crabp1 (Fig. 5C) shows expression in the cauda of both species.
|
Although significant insights can be gained by evaluating the regulation of individual genes, the power of array-based studies lies in the ability to assess the expression of all genes, allowing for the elucidation of expression patterns of entire gene families. In order to identify categories of overrepresented genes, the list of 380 coregulated genes were annotated using DAVID 2.1 [24] with the Functional Annotation Tool, and the most significant categories were identified. Table 2 lists some of the overrepresented categories, and the complete analysis is available in Supplemental Table S5. The top differentially expressed categories include those with known regulation in the epididymis, including transporter activity, glutathione transferase activity, and protease inhibitor activity. A number of other categories are also evident, including lipid metabolism, extracellular matrix (ECM), and carboxylic acid metabolism. Specific examples of gene families with similar expression patterns between the two species are protease inhibitor and ECM genes (Fig. 6). In both the mouse and the rat epididymis, the ECM genes are expressed specifically at relatively high levels in the terminal few segments (Fig. 6, C and D). Genes that encode protease inhibitors have two conserved patterns of regulation, either high expression in the proximal segments or high expression in the terminal segments (Fig. 6, A and B)
|
|
Expression of the Defensin Gene Family Across Epididymes
The defensin gene family is highly regulated throughout the mouse [19] and rat epididymis [22]. Given the dual role of beta defensins in immunity [37] and sperm maturation [38–40] and their segment-dependent expression in the mouse, we further explored the expression of this family of genes in both rat and mouse epididymal segments. Recent reports have identified 44 beta defensin genes in the rat and 52 genes in the mouse [22], and only three of the rat beta defensin genes are represented on the array. Therefore, we took a systematic TaqMan-based approach to examine the expression of each of the beta defensin genes in each of the rat and mouse segments. Sequence comparisons and alignment identified 36 orthologous beta defensins genes, 29 of which were unambiguously detected in both species. Figure 7 shows the relative expression of the 29 defensins in each individual rat and mouse segment. All 29 genes had similar expression patterns across the rat and mouse epididymes (Fig. 7), with peak expression in similar relative segment positions.
|
In the present study, we have determined the segment-dependent regulation of over 30 000 rat transcripts using DNA microarray technology validated with multiple, individual gene analyses using Real-Time RT-PCR. In the process, we determined that the epididymis of the Sprague-Dawley rat is organized into nineteen different segments, three or four in what is often referred to as the initial segment, five comprising the remainder of the caput, four comprising the corpus, and six comprising the cauda (Fig. 1).
Previous studies have examined regional transcriptome expression in the rat epididymis [26], whereas our segmental approach allows for a more systematic analysis with a higher sensitivity than has previously been possible. Our analysis identified over 16 000 transcripts expressed in the epididymis, with close to 4000 genes being highly regulated along the length of the epididymis. This is comparable to the segmental regulation seen in the mouse [19], and the regional analysis of the human epididymis [27] implies the same will be true there. How such discrete segmental gene expression occurs remains unclear, but recent results have demonstrated that connective tissue septae (CTS) partition the epididymal interstitium into successive compartments [16]. This raises the possibility that the septae serve to limit the effects of paracrine signals to the segment in which a specific paracrine secretion occurs, which may explain how epithelial gene expression can be initiated or terminated so sharply at a CTS. Indeed, growth factors microinfused into individual segments in vivo exert their effects in the infused segment but not in adjacent segments [41].
Our previous work has characterized the epididymal transcriptome of the mouse, and the results of the present study have allowed us to perform a detailed comparison of the expression patterns in two divergent species (Fig. 4). Although the mouse and rat epididymides carry out common functions of sperm maturation and sperm storage, rats have almost twice as many segments as the mouse, suggesting that differences may exist in the number of segments that perform a specific task. This is likely, since with the exception of segment 1, high levels of expression of particular genes tended to be spread over multiple segments (Fig. 2), whereas in the mouse epididymis, peak gene expression is more commonly associated with a single segment [19].
Another issue highlighted by this study is the problem of terminology. The term ‘initial segment' has arisen from the classical work of Reid and Cleland [11], who described six zones of the rat epididymis and referred to zone 1 as the ‘initial zone.' This work together with suggestions made by others [42] has led to the use of the term initial segment to refer to the first perceived segment of the epididymis, especially in the rat and mouse. Reid and Cleland [11] divided their initial zone into subzones 1a, 1b, and 1c, which correspond to segments 1, 2, and 3 of the present study (Fig. 1). Thus, what is commonly referred to as the initial segment is actually at least three different segments. This causes a problem when reference is made to the initial segment of the mouse and the initial segment of the rat, especially when making functional comparisons. The initial segment of the mouse truly is the initial (the first) segment of the epididymis, whereas the traditional initial segment of the rat consists of multiple segments. The true initial segment of the rat epididymis is segment 1 only, as is the case in the mouse (Fig. 1), since this comparability is true from both morphological and gene expression viewpoints. The tubule morphologies in segment 1 of rats and mice are similar in that they have wider lumens and greater epithelial height than those of their following segments. Furthermore, segment 1 of the mouse epididymis has a transcription profile that distinguishes it from subsequent segments [19] and hierarchical clustering of the transcriptional profiles of rat segments yields a similar result (Fig. 2A). Studies using segments 1, 2, and 3 (subzones 1a, 1b, and 1c as defined by Reid and Cleland [11]) and sometimes segment 4 as the initial segment of the rat epididymis will confound comparisons of initial segment function in these two species.
Over 400 differentially expressed genes are common to both species, with over 300 genes showing similar expression patterns. It is recognized that these values probably present an under-representation of the true overlap, since our studies are confounded by the technical issues raised when comparing two different species whose genome sequences are at various stages of completion at the time of gene array design. Thus, the oligonucleotide sequences used to quantify gene expression do not always allow direct cross-identification. The probes of the two arrays may monitor different regions of a gene or an alternatively spliced region or may comprise a different probe-sequence context, preventing identical signals from both arrays. Given these caveats, we have mapped over 6000 genes between the arrays, and we have shown that more than 400 of these genes are regulated in both species. Our pathway mapping and annotation of the genes regulated in both species will ultimately allow insights into the mechanism of action of the epididymis. Furthermore, by concentrating on genes with similar expression patterns, we have enriched for pathways that are conserved between species, which hopefully represent critical pathways for epididymal function.
We have attempted to map mouse segments to rat segments based on transcriptional profiles. However, given the difference in the number of segments (10 vs. 19), it was difficult to determine definitively the corresponding segments between the mouse and rat. Instead, our analysis concentrated on comparing the patterns of expression along the length of the epididymis (Figs. 2 and 3) using clustering algorithms as an unbiased approach to group genes with similar patterns of expression. The clusters in the rat and mouse (Fig. 4) illustrate repeatable patterns of regulation, some of which are shared (Fig. 4B) and some of which are not shared (Fig. 4D) between the rat and mouse. Interestingly, the majority of genes that are differentially expressed in the mouse and rat have similar patterns of expression, and these patterns are dependent upon the relative number of segments from the initial segment (Fig. 4C). These groupings suggest that these patterns of expression are highly conserved. It will be of interest to determine the effects of testicular factors on the epididymal transcriptome, especially in the more proximal segments, in which lumicrine regulation has been investigated [1, 43, 44].
Many well-studied epididymal genes show similar patterns of expression, including Cst6, Cst8, Cst11, Ros1, Gpr64, Lcn2, Srd5a1, Clu, and Gpx3, while Galgt1 shows high-level expression in mouse initial segments and high-level expression in rat cauda. Many of the genes that show differential patterns of expression show subtle shifts in peak expression or are expressed at low levels in one of the species. Therefore, further investigation of the differential expression patterns is needed. Furthermore, these genes do not fall into any obvious category that represents known epididymal biology. Several genes that show differential expression patterns are involved in fatty acid metabolism, including Fah, Fads1, Elovl6, Acox1, and Acox3, which show high-level expression in mouse initial segments and high-level expression in rat terminal segments. Given that many genes involved in fatty acid metabolism are expressed throughout the length of the epididymis, the role of these differences in expression patterns is unclear.
We have chosen the beta defensin gene family to study in greater detail. The beta defensin family members are small cationic peptides, many of which have broad-spectrum antibacterial activity and the expression of which is predominately localized to the male reproductive tract. Some beta defensins are associated with sperm and may be involved in sperm maturation [38, 39]. The defensin gene family comprises a large cluster on rat chromosomes 16, 15, 9, and 3 and has conserved syntenic sequences in the mouse. The sequences and chromosomal localization are highly conserved and in the present study, we show that the mRNA expression along the segments of the epididymis is also highly conserved. Interestingly, the pattern of expression is highly correlated with the chromosomal location (Supplemental Table S6).
Given the variability in the number of segments in different species, one of our goals is to examine the epididymides of additional species to understand the breadth of the segmentation phenomenon in epididymal structure. It will be interesting to repeat these studies in humans, to appreciate the full impact of segmental regulation of human epididymal gene expression.
In summary, the data obtained in the present study provide a comprehensive view of the potential mechanisms of epididymal function and regulation of the luminal environment necessary for sperm development in two model organisms. Furthermore, the amount of data generated creates a vast resource for further data mining to understand the mechanisms leading to sperm maturation and the microenvironment necessary for sperm storage. All of the data from this investigation will be made publicly available in a searchable format via the Mammalian Reproductive Genetics database (http://mrg.genetics.washington.edu) when this report goes to press. These findings may ultimately help to identify novel targets for male contraception and to improve our understanding of the various forms of male infertility.
ACKNOWLEDGMENTS
The authors thank Leeying Wu for confirmation of the beta defensin gene sequences.
FOOTNOTES
1Supported by NIH grant DK45179 to T.T.T. 
Correspondence: 2Scott A. Jelinsky, Wyeth Research, 87 Cambridge Park Drive, Cambridge, MA 02140. FAX: 617 665 7519; e-mail: sjelinsky{at}wyeth.com
Received: 15 September 2006.
First decision: 23 October 2006.
Accepted: 5 December 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. A. Cornwall New insights into epididymal biology and function Hum. Reprod. Update, March 1, 2009; 15(2): 213 - 227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.Q. Lin, J.Y. Li, H.Y. Wang, J. Liu, C.L. Zhang, W.T. Wang, J. Liu, N. Li, and S.H. Jin Cloning and identification of a novel sperm binding protein, HEL-75, with antibacterial activity and expressed in the human epididymis Hum. Reprod., September 1, 2008; 23(9): 2086 - 2094. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Thimon, E. Calvo, O. Koukoui, C. Legare, and R. Sullivan Effects of Vasectomy on Gene Expression Profiling along the Human Epididymis Biol Reprod, August 1, 2008; 79(2): 262 - 273. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Johnston, W. W. Wright, P. DiCandeloro, E. Wilson, G. S. Kopf, and S. A. Jelinsky Stage-specific gene expression is a fundamental characteristic of rat spermatogenic cells and Sertoli cells PNAS, June 17, 2008; 105(24): 8315 - 8320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. T. Page, J. K. Amory, and W. J. Bremner Advances in Male Contraception Endocr. Rev., June 1, 2008; 29(4): 465 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Turner De Graaf's Thread: The Human Epididymis J Androl, May 1, 2008; 29(3): 237 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Dean, J. M. Good, and M. W. Nachman Adaptive Evolution of Proteins Secreted during Sperm Maturation: An Analysis of the Mouse Epididymal Transcriptome Mol. Biol. Evol., February 1, 2008; 25(2): 383 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Allen and A. C. Spradling The Sf1-related nuclear hormone receptor Hr39 regulates Drosophila female reproductive tract development and function Development, January 15, 2008; 135(2): 311 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Thimon, O. Koukoui, E. Calvo, and R. Sullivan Region-specific gene expression profiling along the human epididymis Mol. Hum. Reprod., October 1, 2007; 13(10): 691 - 704. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. T. Turner, D. S. Johnston, J. N. Finger, and S. A. Jelinsky Differential Gene Expression among the Proximal Segments of the Rat Epididymis Is Lost after Efferent Duct Ligation Biol Reprod, July 1, 2007; 77(1): 165 - 171. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
