This web only provides the extract of this article. If you want to read the figures and tables, please reference the PDF full text on Blackwell Synergy. Thank you.

- Review -

Segment boundaries of the adult rat epididymis limit interstitial signaling by potential paracrine factors and segments lose differential gene expression after efferent duct ligation

Terry T. Turner^1,2, Daniel S. Johnston³, Scott A. Jelinsky⁴, Jose L. Tomsig¹, Joshua N. Finger³

¹Department of Urology and ²Cell Biology, University of Virginia Health Science System, Charlottesville, VA 22908, USA
³Contraception, Women's Health and Musculoskeletal Biology, Wyeth Research, Collegeville, PA 19426, USA
⁴Biological Technologies, Molecular Profiling and Biomarker Discovery, Wyeth Research, Cambridge, MA 02140, USA

Abstract

The epididymis is divided into caput, corpus and cauda regions, organized into intraregional segments separated by connective tissue septa (CTS). In the adult rat and mouse these segments are highly differentiated. Regulation of these segments is by endocrine, lumicrine and paracrine factors, the relative importance of which remains under investigation. Here, the ability of the CTS to limit signaling in the interstitial compartment is reviewed as is the effect of 15 days of unilateral efferent duct ligation (EDL) on ipsilateral segmental transcriptional profiles. Inter-segmental microperifusions of epidermal growth factor (EGF), vascular endothelial growth factor (VEGFA) and fibroblast growth factor 2 (FGF2) increased phosphorylation of mitogen activated protein kinase (MAPK) in segments 1 and 2 of the rat epididymis and the effects of all factors were limited by the CTS separating the segments. Microarray analysis of segmental gene expression determined the effect of 15 days of unilateral EDL on the transcriptome-wide gene expression of rat segments 1_4. Over 11 000 genes were expressed in each of the four segments and over 2 000 transcripts in segment 1 responded to deprivation of testicular lumicrine factors. Segments 1 and 2 of control tissues were the most transcriptionally different and EDL had its greatest effects there. In the absence of lumicrine factors, all four segments regressed to a transcriptionally undifferentiated state, consistent with the less differentiated histology. Deprivation of lumicrine factors could stimulate an individual gene's expression in some segments yet suppress it in others. Such results reveal a higher complexity of the regulation of rat epididymal segments than that is generally appreciated. (Asian J Androl 2007 July; 9: 565_573)

Keywords: efferent duct ligation; dedifferentiation; proximal epididymis

Correspondence to: Dr Terry T. Turner, Department of Urology, University of Virginia School of Medicine, P. O. Box 800422, Charlottesville, VA 22908, USA.
Tel: +1-434-924-0429 Fax: +1-434-924-8311
E-mail: ttt@virginia.edu

DOI: 10.1111/j.1745-7262.2007.00302.x

1 Introduction

The epididymis has traditionally been divided the organ into the regions of caput, corpus and cauda with the "initial segment" of the caput sometimes separated for further definition. Earlier reports [1, 2] made it clear that each epididymal region of the common rodent models is made up of intra-regional segments or lobules of coiled tubule, and more recent investigations in the mouse [3] and rat [4] have demonstrated that the epididymides of those species are divided into 10 and 19 intraregional segments, respectively (Figure 1). Segmentation has been shown to exist in dogs [5], and the marmoset, a small primate species, shows histological evidence of epididymal segments as well (Terry T. Turner and D. Bomgardner, unpublished observations).

The connective tissue septa (CTS), which separate the segments, can establish borders for epididymal gene expression, protein presence or epithelial response to lumicrine factors [3, 6, 7]. The mechanism behind the sudden turn-on-turn-off of particular gene expressions at the CTS has remained unclear, but it has been hypothesized that CTS divide the epididymis into separate interstitial compartments, which allows inter-segmental paracrine signaling that can be unique to a specific segment [7].

2 Segmentation in the rat epididymis and the partition of signaling by the mitogen activated protein kinase (MAPK) pathway

Epididymal CTS restrict diffusion of molecules from the interstitium of one segment to the next [7]. This barrier property potentially provides a physiological basis for segmented epithelial function because CTS can limit the influence of interstitial cell-epithelial signaling molecules (e.g. growth factors) to the segment in which they were secreted.

Growth factors are polypeptides that promote a variety of cell activities, usually in a paracrine or endocrine fashion [8]. Growth factors mediate their effects primarily through stimulation of the mitogen activated protein kinase (MAPK) pathway [9]. MAKP is a highly conserved kinase that upon stimulation by phosphorylation accumulates in the cell nucleus and promotes the phosphorylation of numerous proteins and downstream kinases, including transcription factors [10]. Several investigators have explored possible roles of a variety of growth factors and growth factor receptors in the epididymis [11], but more emphasis has been placed on expression and localization rather than on sorting out any specifics of regulation. Because previous investigators have demonstrated the presence of vascular endothelial growth factor (VEGFA), basic fibroblast growth factor 2 (FGF2), and epidermal growth factor (EGF) or their receptors in the interstitium or tubules of the epididymis, we recently verified the expression of these specific genes in the rat epididymis using real-time reverse transcription polymerase chain reaction (RT-PCR) (data not shown) and used these molecules to challenge the hypothesis that CTS restrict interstitial signaling to the segment in which the molecule first appears. The response signal used was the phosphorylation of MAPK after growth factor infusion into a specific segment in vivo.

The experiments were done using segments 1 and 2 of the rat epididymis [12] (Figure 2A). Preliminary, in vitro experiments determined that MAPK phosphorylation occurred in both segments in response to all three growth factors and that the timing of the response to growth factors was both immediate (within 2 min) and prolonged (lasting out to 90 min; data not shown). Following those experiments, in vivo microperifusion experiments were performed wherein rats were killed and prepared for in vivo micropuncture as previously described [13].

Microperifusions (i.e. infusions into the interstitial space surrounding the tubules of a specific segment) were carried out using sharpened glass micropipettes filled with approximately 20 µL growth factor solution (10^-6 mol/L EGF, FGF2, or VEGFA in phosphate-buffered saline [PBS] + 0.1% lissamine green as a tracking dye). The loaded pipette was attached to a micromanipulator and to an infusion pump via a PE50 cannula. Under a dissecting microscope the tunica albuginea of either segment 1 or segment 2 of the rat epididymis was punctured with the micropipette. The pipette tip was left in the segment's interstitial space and 15 µL of the selected growth factor solution was infused over an approximate 5-min time period. The pipette was left in place until the end of the experiment and the epididymis was covered with mineral oil to prevent dehydration and to help maintain proper tissue temperature. Control experiments were carried out using vehicle alone as the perifusion medium. To test the effect of CTS disruption on the effect of perifused growth factors, 0.1% collagenase type I was added to the VEGFA solution and used for in vivo perifusion into segments 1 and 2 in different epididymides for 90 min, as described above. Perifusions with medium lacking VEGFA but containing 0.1% collagenase were included as negative controls.

After each in vivo perifusion, segments 1 and 2 were collected separately and analyzed for MAPK phosphorylation by Western blot using an antibody to active MAPK (kindly provided by Dr Thomas Sturgill, Department of Pharmacology, University of Virginia, VA, USA) that recognizes both the MAPK1 and MAPK3 isoforms [14]. For the detection of total MAPK (i.e. phosphorylated and nonphosphorylated forms), a rabbit anti-MAPK antibody was used (Santa Cruz, CA, USA). Tissue samples (50 μg protein/lane) were electrophoresed in 12% PAGE gels, transferred to nitrocellulose membranes, and stained with Ponceau S to verify the transfer. Membranes were blocked with 3% non-fat dry milk in PBS and probed with the primary antibodies according to standard protocols. Membranes were washed and incubated with peroxidase-labeled, goat anti-rabbit antibody and detection was carried out using a commercial chemiluminescent substrate.

Both the MAPK1 and MAPK3 isoforms were detected in both segments 1 and 2 of the rat epididymis [15] (Figure 3). The average molecular masses of these two isoforms were approximately 44.5 kDa and 43.8 kDa for MAPK1 and MAPK3, respectively (n = 5, each). No other bands were revealed by the anti-active MAPK antibody or by the anti-total MAPK antibody in either nonstimulated or growth factor-stimulated epididymides.

Mitogen activated kinase activation was observed only in the specific segment perifused with either of the three growth factors, EGF, FGF2 or VEGFA, not in the adjacent, non-perifused segment (Figure 3). MAPK activation persisted through the entire 90-min experimental period, but whether the segment perifused was segment 1 or 2, the adjacent segment did not typically show detectable levels of MAPK phosphorylation (Figure 3). Total MAPK, as detected by western blot, was present with equivalent signal in both segments irrespective of perfusate (not shown). Preliminary experiments at 30 and 60 min after growth factor perifusion gave the same result (i.e. at no time after growth factor presentation was MAPK phosphorylation detected in the non-perifused segment). As with the in vitro experiments, no MAPK phosphorylation was detected in segments perifused with medium alone (not shown). Of note, the response of both segment 1 and segment 2 to the perifusion of any one of the three growth factors appeared similar (Figure 3).

To test the hypothesis that CTS were responsible for retaining the response to growth factors within the perifused segments, segments were perifused with VEGFA in the presence of collagenase. Ninety minutes after perifusion of the VEGFA/collagenase solution into either segment 1 or 2, MAPK activation was observed not only in the perifused segment but in the adjacent segment as well [15] (Figure 4). Perifusions containing collagenase alone did not activate MAPK in either segment.

These studies were conducted using growth factors for which previous evidence suggested they or their receptors were present in the epididymis. We used epididymal segments 1 and 2 because the CTS between these two segments divide clearly different epithelia [1] with different responses to stimuli [16]. The nomenclature "segments 1 and 2" is the result of a simple numbering of all 19 segments of the rat epididymis similar to the numbering of the 10 segments of the murine epididymis [3]. Rat segments 1 and 2 correspond to zones 1A and 1B in the terminology of Reid and Cleland [1]. MAPK phosphorylation was used as a test of growth factor action because MAPK activation is the hallmark of growth factor signaling [9].

Interestingly, and consistent with our hypothesis that the CTS provide a functional barrier isolating paracrine effects within individual segments, MAPK activation induced by EGF, FGF2 and VEGFA occurred only in the perifused segments in vivo, whether the perifusion was in segment 1 or segment 2 (Figure 3).

Connective tissue septa are formed by eosinophilic strands with fibers, suggesting collagen as a major component. Accordingly, perifusion experiments with VEGFA plus collagenase were conducted to permit the degradation of collagen and the reduction of the intersegmental barrier's resistance to molecular movement. Under these conditions, VEGFA activated MAPK in both the perifused segment and the adjacent segment, again irrespective of which segment was perifused (Figure 4).

Exposure of CTS to collagenase allowed MAPK activation activity in the segment adjacent to the perifused segment. This suggests a reduction in the intersegmental barrier and a passage of VEGFA from the perifused segment into the adjacent segment. Such movement never occurred with any growth factor in the absence of collagenase, thus reinforcing the idea that under normal conditions the physical integrity of the CTS prevents substances originally appearing in one segment from freely diffusing into the next. The growth factors used in the present study (EGF [6 kDa], FGF2 [26 kDa] and VEGFA [45 kDa]) are all at or above the molecular mass of molecules previously shown to be largely retained by intraregional segments [7]; therefore, the CTS are capable of restricting the movement of a variety of paracrine molecules that may be secreted within a particular segment.

The above results can be used as proof of principle only. In nature, secreted growth factors commonly become quickly bound to interstitial binding molecules (e.g. heparan sulfate proteoglycan and FGF-binding protein 1), which might restrict diffusion, depending on local conditions [17, 18]. In the present study, molar excess concentrations of growth factors (μmol/L instead of nmol/L) were used in the perifusion medium to overcome this potential binding, thus leaving unbound growth factor available for diffusion through the CTS. The resulting data supports the possibility of intersegmental, paracrine signaling and begs the question of which signaling molecules are important in which segments of the epididymis and what specific processes they regulate. That research is underway. It is important to recall, however, that paracrine signaling within individual segments is only one aspect of cell regulation in the epididymis. Endocrine regulation of the tissue is well known, and will not be addressed here. However, non-androgenic lumicrine factors from the testis are known to be important for both epididymal development [19, 20] and maintenance [21_23]. Efferent duct ligation (EDL) obstructs the flow of testicular products into the epididymal tubule and causes the loss of testicular lumicrine factors (e.g. androgen binding protein, FGF2 and sperm membrane proteins [24]) from all points distal to the ligation. EDL also induces several alterations in the epididymal tubule [24, 25]. At present, reports of the effects of EDL on the epididymis focus on relatively few genes and give reference only to the standard epididymal regions or to the initial segment. Microarray analysis of gene expression allows quantitative detection of thousands of gene transcripts, and when performed on individual segments rather than entire regions of the epididymis increases the sensitivity of transcript detection. Because the deprivation of testicular lumicrine factors alters epithelial structure dramatically, as well as secreted proteins [22], gene expression [23] and epithelial cell apoptosis [16_26], especially in the proximal segments of the epididymis [21, 27], and because microarray analysis of individual segments allows a broadly based, yet highly specific, evaluation of epididymal gene expression, we have used this approach to examine the effect of lumicrine signaling on the segmented function of the rat epididymis.

3 Rat epididymal segments lose differential gene expression in the absence of lumicrine factors from the testis

Animals were anesthetized as above and unilateral efferent duct ligation and contralateral sham operation were performed as described previously [16, 28]. Fifteen days later, the testes and epididymides were exposed and the epididymides were removed and epididymides were immediately placed in ice-cold saline in a Petri dish under a dissecting microscope. The epididymal tunica albuginea was removed from the caput epididymidis using sharp microdissection and the first four epididymal segments (Figure 2) were isolated. As each segment was isolated it was immediately placed in RNALater (>10 × tissue volume; Ambion, Austin, TX, USA) on ice. All four segments of each epididymis were in RNALater within 15 min of epididymal extirpation, and this procedure was repeated until each epididymal segment was represented by five separate samples for RNA extraction.

Five additional animals were subjected to unilateral EDL and 15 days later both the contralateral control epididymides and the ipsilateral, EDL epididymides were extirpated, paraffin embedded, and stained with hematoxolyn-eosin for subsequent measurement of tubule diameter and epithelial height.

RNA was extracted in ice cold TRIzol by conventional procedures and purified with RNAeasy columns (Qiagen, Valencia, CA, USA). RNA quantity was determined by absorbance at 260 nm and quality was determined using an Agilent Bioanalyzer (Palo Alto, CA, USA).

3.1 Microarray processing

Five µg total RNA were used to generate biotin-labeled cRNA using an oligo T7 primer in a reverse transcription reaction followed by in vitro transcription reaction with biotin-labeled UTP and CTP. Ten µg cRNA were fragmented and hybridized to RAE230 2.0 arrays (Affymetrix, Santa Clara, CA, USA). Hybridized arrays were stained according to manufactures protocols on Fluidics Station 450 and scanned on Affymetrix Scanner 3000. Arrays with excessive background, low signal intensity, or major defects within the array were eliminated from further analysis. The final number of arrays used for analysis was three for all control segments and five for all EDL segments.

Signal values were determined by 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 Expressionist Analysis 3.0 (Genedata; Basel, Switzerland) for analysis.

A gene transcript was considered detectable if its mean expression in any segment was greater than 50 signal units and the percentage of samples with a present call as determined by GCOS default settings was ³ 67% in the samples within a group. Such transcripts are hereafter referred to a "qualifiers". Normalized signal values were transformed to the log base 10 and pair-wise comparative analysis of the qualifiers in each segment was performed. A qualifier was considered to be segmentally regulated if two conditions were met: (i) the qualifier had to be detected in at least 67% of the samples of at least one of the segments analyzed; and (ii) a significant difference at the level of P £ 0.01 based on the Welch test had to exist between the control and EDL values. Qualifiers meeting these conditions were used for further analysis. For some computations, the expression values for each qualifier were normalized to a mean of 0 and a standard deviation of 1 (z-score normalization). This allowed direct comparison of patterns within the data without respect to absolute expression levels. Principal component analysis (PCA) was performed on the transformed data and visualized using Spotfire 7.2 (Somerville, MA, USA). The outcome of such an analysis is a set of variables visualized in a 2-D or 3-D space, a process useful for functional and biological interpretation of complex datasets [29].

3.2 Real-time reverse transcription polymerase chain reaction (qRT-PCR) analysis

cDNA samples for real-time PCR analyses were synthesized by random priming in a final volume of 20 µL using the Superscript III First Strand Synthesis System(Invitrogen; Carlsbad, CA, USA) for RT-PCR according to the manufacturer's instructions. The cDNA samples corresponding to each of the epididymal segment samples as well as whole epididymis were diluted to a final volume of 200 µL to produce a 10 × cDNA stock.

Real-time RT-PCR (qRT-PCR) analysis was carried out using FAM labeled flourogenic LUX primers from Invitrogen. Primers were chosen from published sequences for prostaglandin D₂ synthase (Ptgds), CD52 antigen (Cd52), glutathione peroxidase-3 (Gpx3), cystatin 8 (Cst8; also known as CRES for cysteine-related, epididymal specific protein), defensin β1 (Defb1), 5α-reductase I (Srd5a1), phosphatidylethanolamine binding protein (Pebp1), lipocalin 5 (Lcn5; also known as ERABP or epididymal retinoic acid binding protein) and superoxide disumtase-1 (Sod1). Flourophore-labled primers were synthesized by Invitrogen and non-labeled primers were synthesized by Wyeth Research (Collegeville, PA, USA).

Real-time RT-PCR reactions were conducted using Platinum Quantitative PCR SuperMix-UDG (Invitrogen), according to the manufacturer's instructions. PCR reactions were run on an automated flourometer in a 96 well format. PCR conditions for all reactions was as follows: 1 cycle of 48ºC for 30 min., 95ºC for 10 min. followed by 40 cycles of 95ºC for 15 s and 60ºC for 1 min. Relative expression was determined by using the C_T method [30] using Sequence Detector software, version 1.6.3. The cDNA samples of each segment and of whole epididymis were evaluated in triplicate with the primer pair for each gene. Results were normalized to 18S ribosomal RNA expression and expressed as a ratio of expression of each gene in each segment compared to the expression of that gene in the whole epididymis.

3.3 Results of the efferent duct ligation study

The rat caput epididymidis (Figure 2A) becomes considerably reduced in size 15 days after EDL and the surface appearance of the segments is less distinct (Figure 2). A total of 2 255 qualifiers were regulated in response to EDL in segment 1 (Table 1), or approximately 18% of the total number of qualifiers detected in that EDL segment. The number of qualifiers either upregulated or downregulated after EDL declined sequentially in each more distal segment, with segment 4 showing only 420 genes (approximately 4% of total) regulated in response to EDL (Table 1). The caliber of the changes varied from gene to gene and from segment to segment, with EDL again having its largest effects in segment 1 where 1 595 genes were induced to at least a 2-fold change, 150 to at least a 10-fold change and 11 to at least a 100-fold change. By segment 4, the numbers of gene expressions changing at those same levels were 420, 38, 15 and 3, respectively. The proportion of those altered gene expressions that was upregulated versus downregulated stayed roughly the same (e.g. approximately 50, 30 and 15% were upregulated at the 2-, 5-, and 10-fold levels, irrespective of segment) (not shown). All changes at the 100-fold level were downregulations.

Among the qualifiers responding to EDL, there were many with relatively low expression in control segments, but which were moderately upregulated after EDL (Figure 5A). At the same time, there were many qualifiers with relatively high expression in control segments, but which were sharply reduced in expression after EDL (Figure 5B).

The net effect of EDL was to eliminate segmental differences in gene expression among the segments studied. Microarray results were corroborated by qRT-PCR of selected genes known to be present in the epididymis (Figure 6). Cst8, Defb1 and Srd5a1were all genes downregulated by EDL according to microarray analysis and similar results were detected by qRT-PCR. Other genes, Ptgds, Cd52, Gpx3 and Pebp1 among them, also showed highly similar results on the two different types of analysis.

Principal component analysis grouped samples with similar transcriptome expression patterns. When displayed in 2-D space, the analysis illustrates that rat segments 1 and 2 stand separately as unique gene-expression units of the epididymis (Figure 7). Control segments 3 and 4 span common space and are not different from each other, but EDL eliminates segmental differences and sets all 4 segments apart from their control counterparts (Figure 7).

4 Summary

It is well known that interruption of luminal contribution from the testis alters epididymal gene and protein expression [24, 25]; however, the breadth of the changes have been difficult to assess because only a score or so of specific genes or proteins have been studied in the epididymis after EDL or after orchidectomy with androgen supplementation. Microarray analysis of gene expression has previously been used in rats [31] and mice [32, 33] to evaluate more broadly the changes in epididymal gene expression after orchidectomy with androgen replacement, and these studies have documented widespread change in gene expression throughout the epididymis. Unfortunately, direct comparisons among these studies have been difficult because of the use of different species, different epididymal dissection patterns, and different micro-array platforms. Direct comparison of those studies with the present study is difficult, as well, but the aim was to use a segment-by-segment evaluation of gene expression changes in the proximal segments after an EDL of sufficient time to allow the initial wave of epithelial apoptosis [16, 27] to diminish and for the epithelium to come to a post-EDL steady state.

Unilateral EDL was used, which allowed preservation of circulating androgens while eliminating lumicrine factors from the ipsilateral epididymis. The results make it clear that thousands of gene expressions are altered by loss of lumicrine factors, and that those factors can be suppressive (> 1 200 qualifiers upregulated after loss of lumicrine factors) or stimulatory (> 1 000 qualifiers downregulated after loss of lumicrine factors; Table 1). It has been demonstrated previously that loss of testicular factors can upregulate or downregulate individual gene expressions in select regions of the rat [31] and murine [32, 33] epididymis, but the present study focused specifically on the first four segments of the rat caput epididymidis because those segments receive intraluminal factors most directly from the testis. It seemed likely these segments of the epididymis would be most sensitive to lumicrine factors [24].

The first three segments of the rat epididymis make up what has historically been referred to in the rat as the "initial segment", initially identified by Reid and Cleland [1] as zones 1a, 1b and 1c. The initial segment nomenclature presents difficulties described elsewhere [4], not the least of which is that under that single heading all three or even four segments of the rat "initial segment" have been assayed together as if they are a single unit. They are not. The segments are transcriptionally different (Figures 2 and 3), and the more proximal the segment, the more profoundly its gene transcription is affected by EDL (Table 1, Figure 5). Nearly 20% of the segment 1 transcriptome was altered after 15 days of EDL, and this proportion declined to approximately 4% by segment 4 (Table 1).

Segments of the epididymis that are highly differentiated in control animals become transcriptionally undifferentiated after EDL (Figures 5 and 7). Fifteen days after EDL, none of the segments studied were transcriptionally different from each other in an overall statistical sense, but they were all different from their respective control tissues (Figure 7). In other words, EDL induces a loss of differentiation of the proximal segments of the rat epididymis. The implications of this are that even in the patent male tract, testicular secretions in addition to testosterone are required for epididymal function in a way that is far more broad than previously appreciated. A more complete determination of important lumicrine molecules is called for, as well as a elucidation of their role both in direct signaling to the tubule epithelium and in indirect signaling to cells in the epididymal interstitium. Those latter cells might play an important role in tubule regulation through the paracrine signaling mechanisms discussed previously.

Therefore, at least two different signaling processes are important in regulating the rat epididymis in addition to endocrine regulation: (i) a paracrine signaling process wherein growth factors or other signaling molecules traffic between interstitial and epithelial cells within a given segment, signaling limited to individual segments by the CTS forming the segment borders; and (ii) a lumicrine signaling process wherein intra-luminal molecules can pass from segment to segment, yet are still required for the direct or indirect regulation of thousands of gene expressions in a segment-specific manner. These processes speak of the complexity of the epididymal tubule and call for further investigations into the specific molecules involved in both the lumicrine and the paracrine signals that regulate the epididymal epithelium.

Acknowledgment

This work was supported by National Institutes of Health grant DK45179 (T. T. Turner) and Wyeth Research.

References

1 Reid BL, Cleland KW. The structure and function of the epididymis: histology of the rat epididymis. Aust J Zool 1957; 5: 223_46.

2 Abou-Haila A, Fain-Maurel MA. Regional differences of the proximal part of mouse epididymis: morphological and histochemical characterization. Anat Rec 1984; 209: 197_208.

3 Johnston DS, Jelinsky SA, Bang HJ, DiCondeloro P, Wilson E, Kopf GS, et al. The mouse epididymal transcriptome: transcriptional profiling of segmental gene expression in the epididymis. Biol Reprod 2005; 73: 404_13.

4 Jelinsky SA, Turner TT, Bang HY, Finger JN, Solarz MK, Wilson E, et al. The rat epididymal transcriptome: comparison of segmental gene expression in the rat and mouse epididymides. Biol Reprod 76: 561_70.

5 Kirchhoff C. The dog as a model to study human epididyimal function at the molecular level. Mol Human Reprod 2002; 8: 695_701.

6 Kirchhoff C. Gene expression in the epididymis. Int Rev Cytol 1999; 188: 133_202.

7 Turner TT, Bomgardner D, Jacobs JP, Nguyen QA. Association of segmentation of the epididymal interstitium with segmented tubule function in rats and mice. Reproduction 2003; 125: 871_8.

8 Wilkinson MG, Millar JBA. Control of the eukaryotic cell cycle by MAP kinase signaling pathways. FASEB J 2000; 14: 2147_57.

9 Roux PP, Blenis J. ERK and p38 MAPK-activated protein kinases: a family of protein kinases with diverse biological functions. Microbiol Mol Biol Rev 2004; 68: 320_44.

10 Schaeffer HJ, Weber M. Mitogen-activated protein kinases: specific messages from ubiquitous messengers. Mol Cell Biol 1999; 19: 2435_44.

11 Tomsig JL, Turner TT. Growth factors and the epididymis. J Androl 2006; 27: 348_57.

12 Turner TT, Johnston DS, Finger JN, Jelinsky SA. Differential gene expression among the proximal segments of the rat epididymis is lost after efferent duct ligation. Biol Reprod 2007; 77: in press.

13 Turner TT, Miller DW, Avery EA. Protein synthesis and secretion by the rat caput epididymidis in vivo: influence of the luminal microenvironment. Biol Reprod 1995; 52: 1012_9.

14 Rossomando AJ, Sanghera JS, Marsden LA, Weber MJ, Pelech SL, Sturgill TW. Biochemical characterization of a family of serine/threonine protein kinases regulated by tyrosine and serine/threonine phosphorylations. J Biol Chem 1991; 266: 20270_5.

15 Tomsig JL, Usanovic S, Turner TT. Growth factor-stimulated mitogen-activated kinase (MAPK) phosphorylation in the rat epididymis is limited by segmental boundaries. Biol Reprod 2006; 75: 598_604.

16 Turner TT, Riley TA. p53 independent, region-specific epithelial apoptosis is induced in the rat epididymis by deprivation of luminal factors. Mol Reprod Dev 1999; 53: 188_97.

17 Ornitz DM. FGFs, heparan sulfate and FGFRs: complex interactions essential for development. Bioessays 2000; 22: 108_12.

18 Tassi E, Al-Attr A., Aigner A, Swift MR, McDonnell K, Karavanov A, et al. Enhancement of fibroblast growth factor activity by an FGF-binding protein. J Biol Chem 2001; 276: 40247_53.

19 Alexander NJ. Prenatal development of the ductus epididymis in the rhesus monkey: the effects of fetal castration. Am J Anat 1972; 135: 119_34.

20 Abe K, Takano H, Ito T. Response of the epididymal duct in the corpus epididymidis to efferent or epididymal duct ligation in the mouse. J Reprod Fert 1984; 64: 69_72.

21 Fawcett DW, Hoffer AT. Failure of exogenous androgens to prevent regression of the initial segment of the rat epididymis after efferent duct ligation or orchiectomy. Biol Reprod 1979; 29: 162_81.

22 Holland MK, Vreeburg JT, Orgebin-Crist MC. Testicular regulation of epididymal protein secretion. J Androl 1992; 13: 266_73.

23 Palladino, MR, Hinton BT. Expression of multiple g-glutamyl transpeptidase messenger ribonucleic acid transcripts in the adult epididymis is differentially regulated by androgens and testicular factors. Endocrinology 1994; 135: 1225_33.

24 Robaire R, Hinton BT, Orgebin-Crist MC. The Epididymis. In: Neill JD, editor. Physiology of Reproduction, 3rd edition. New York: Elsevier; 2006: 1071_148.

25 Cornwall GA, Lareyre JJ, Matusik RJ, Hinton BT, Orgebin-Crist BT. Gene expression and epididymal function. In: Robaire B, Hinton BT, editors. The Epididymis: From Molecules to Clinical Practice. New York: Kluwer Academic/Plenum Publishers; 2002: 169_79.

26 Fan X, Robaire R. Orchidectomy induces a wave of apoptotic cell death in the epididymis. Endocrinology 1998; 139: 2128_36.

27 Nicander L, Osman DL, Ploen L, Bugge HP, Kvisgaard KN. Early effects of efferent ductule ligation on the proximal segment of the rat epididymis. Int J Androl 1983; 6: 91_102.

28 Turner TT, Bomgardner D, Jacobs JP. Sonic hedgehog pathway genes are expressed and transcribed in the adult mouse epididymis. J Androl 2004; 25: 514_22.

29 Alter O, Brown PO, Botstein D. Singular value decomposition for genome-wide expression data processing and modeling. Proc Natl Acad Sci U S A 2000; 97: 10101_6.

30 Pontius J, Wagner L, Schuler G. Unigene: a unified view of the transcriptome. In: The NCBI Handbook. National Center for Biotechnology Information, Bethesda, MD. 2003.

31 Ezer N, Robaire B. Gene expression is differentially regulated in the epididymis after orchidectomy. Endocrinology 2003; 144: 975_88.

32 Chauvin TR, Griswold MD. Androgen-regulated genes in the murine epididymis. Biol Reprod 2004; 71: 563_9.

33 Sipilä P, Pujianto DA, Shariatmadari R, Nikkita J, Lehtoranta M, Huhtaniemi IT, et al. Differential endocrine regulation of genes enriched in initial segment and distal caput of the mouse epididymis as revealed by genome-wide expression profiling. Biol Reprod 2006; 78: 240_51.