|Home | About | Journals | Submit | Contact Us | Français|
The role of estrogen and testosterone in the regulation of gene expression in the proximal reproductive tract is not completely understood. To address this question, mice were treated with testosterone or estradiol, and RNA from the efferent ducts and caput epididymides was processed and hybridized to Affymetrix M430 2.0 microarrays. Analysis of array output identified probe sets in each tissue with altered levels in hormone-treated versus control animals. Hormone treatment efficacy was confirmed by determination of serum hormone levels before and after treatment and by observed changes in transcript levels of previously reported hormone-responsive genes. Tissue-specific hormone sensitivity was observed with 2867 and 3197 probe sets changing significantly in the efferent ducts after estrogen and testosterone treatment, respectively. In the caput epididymidis, 117 and 268 probe sets changed after estrogen and testosterone treatment, respectively, demonstrating a greater response to hormone in the efferent ducts than in the caput epididymidis. Transcripts sharing similar profiles in the intact and hormone-treated animals compared with castrated controls were also identified. Ontology analysis of probe sets revealed that a significant number of hormone-regulated transcripts encode proteins associated with lipid metabolism, transcription, and steroid metabolism in both tissues. Real-time RT-PCR was used to confirm array data and to investigate other potential hormone-responsive regulators of proximal reproductive tract function. The results of this work reveal previously unknown responses to estrogen in the caput epididymidis and to testosterone in the efferent ducts, as well as tissue-specific hormone sensitivity in the proximal reproductive tract.
The proximal reproductive tract performs several functions that are required for fertility in the male, including the concentration of spermatozoa, maturation of spermatozoa into motile and fertile sperm, and storage of mature sperm . Two duct systems, the efferent ducts and the epididymis, carry out these diverse functions. In the mouse, the efferent ducts comprise three regions that connect the testis to the epididymis (the proximal, conus, and common regions), which are composed of two types of cells (ciliated and nonciliated) . The epididymis is composed of a single convoluted tubule and is divided into several segments, each serving different functions and having distinct gene expression profiles [3, 4]. Within the epididymis, the initial segment and the caput epididymidis are most proximal to the testis and are required for maturation of spermatozoa to fully motile sperm and for resorption of testicular fluid to concentrate the spermatozoa . The two distal segments, the corpus and the cauda, serve primarily to facilitate maturation of spermatozoa and to store mature sperm until use. The epididymis creates an environment that is permissive for maturation of spermatozoa by regulating the concentration of various peptides, enzymes, and solutes along the length of the epididymal lumen , and this process depends on the concentration of the spermatozoa entering the epididymis from the rete testis via the efferent ducts. Functionally, the efferent ducts serve to connect the testis and epididymis and to concentrate spermatozoa by absorbing up to 96% of the fluid produced by the testis . The factors regulating fluid resorption in the efferent ducts and epididymis and those involved in sperm maturation have not been fully elucidated; therefore, analysis of gene expression in these tissues should generate useful targets to explore.
Global transcriptome analysis of the normal proximal reproductive tract has been limited to the epididymis and its various segments, while analysis of treated tissue has focused on the effects of efferent duct ligation, gonadectomy, orchidectomy, vitamin E, oxidative stress, specific gene knockouts, and androgens on the adult epididymal transcriptome either as a whole or in a segment-specific manner [3, 7–15], as well as the effects of neonatal exposure to diethylstilbestrol and estradiol on the neonatal and adult epididymis . There is a lack of information regarding the transcriptome of the efferent ducts, as well as the effects of steroid hormones on the adult reproductive tract. Although it has been clear for decades that androgens have a key role in the development of the reproductive tract, they are required for adult tract function as well [17, 18]. Most important, several studies [1, 19, 20] have shown that estrogens have a role in regulating the efferent ducts and caput epididymidis gene expression and function. The evidence shows that a key modulator of the efferent ducts function is estrogen, while the epididymis appears to be more responsive to testosterone; however, the role of each hormone in the individual tissues of the proximal reproductive tract remains unclear.
Both testosterone and estrogen mediate their main effects via their nuclear receptors. Localization investigations in the efferent ducts and the caput epididymidis indicate that at least some cell types of both tissues produce the androgen receptor and both isoforms of the estrogen receptor . Steroid radiolabeling findings indicate that both hormones bind elements in the efferent ducts and the caput epididymidis . In vivo analysis of androgen-treated efferent ducts resulted in an increase in fluid resorption, while estrogen treatment resulted in a decrease in fluid resorption . In addition, solute and water resorption (as well as secretion) has been shown to be regulated in part by testosterone in certain regions of the epididymis . Taken together, this evidence indicates that the proximal reproductive tract is responsive to both testosterone and estrogen and that the two sex hormones may regulate fluid resorption and solute transport. The development of the estrogen receptor knockout mouse (Esr1tm1Ksk, hereafter referred to as Esr1KO) confirmed the importance of estrogen in the proximal reproductive tract [25, 26]. These mice show a significant alteration of tract morphology and function [27, 28]. Efferent duct abnormalities included disruptions of nonciliated and ciliated cell morphology and tissue ultrastructure (lumen diameter). Although less severe than the efferent duct defects, the Esr1KO caput epididymidis displays abnormal cell morphology, particularly in the narrow cells. In addition to evidence from mouse models, work examining the effect of estrogen on single genes that are thought to be involved in fluid movement in the efferent ducts has demonstrated estrogen to be a potential modulator of fluid movement [29–32].
The objective of these studies was to expand on previous work by defining the effect of steroid hormone treatment on the caput epididymidis and the efferent duct transcriptomes by utilizing microarray analysis of castrated, sham-castrated, and hormone-treated adult males. To our knowledge, this work represents the first global examination of efferent duct gene expression and the effect of estrogen on the transcriptomes of adult proximal reproductive tract tissues.
Animal care and surgery protocols were approved by the Washington State University Animal Care and Use Committee in accord with the National Institutes of Health's standards established by the Guidelines for the Care and Use of Experimental Animals. Adult male BL/6–129 mice were maintained in a temperature- and humidity-controlled room with food and water ad libitum. Mice were castrated or sham castrated under ketamine/sedazine anesthetic (0.022 mg of ketamine and 0.01 mg of sedazine per animal), and blood from the orbital sinus was collected just before surgery for determination of baseline serum estrogen and testosterone levels. Sham castrations consisted of removal, manipulation, and replacement of the testis, epididymis, and epididymal fat pad. Fourteen days after surgery and before hormone treatment, blood was collected again for serum estrogen and testosterone analysis. Hormone treatment consisted of a single s.c. injection to the side of the spinal ridge of 100 μl of sesame oil (vehicle) containing 15 μg of estradiol (E2) or testosterone propionate (TP) given 14 days after castration. Control animals (sham castrated and castrated) were injected with vehicle alone. Twelve hours after treatment, blood was collected to determine serum estrogen and testosterone levels, and caput epididymidis and efferent duct tissue was harvested after removal of the epididymal fat pad. Of the caput epididymides, only the proximal and distal segments were collected (regions II and III), hereafter referred to as the caput epididymidis. Individual tissue samples were homogenized in 500 μl of TRIzol reagent (Invitrogen, Carlsbad, CA) and stored at −80°C until RNA extraction was performed.
Levels of testosterone and E2 were analyzed via radioimmunoassay by the Assay Core of the Center for Reproductive Biology at Washington State University. For testosterone analysis, the DSL-4100 (Diagnostic Systems Laboratories, Webster, TX) was used according to the manufacturer's recommendations. For E2 analysis, duplicate 100-μl samples and standards were extracted with 3 ml of methyl-tert-butyl ether for 1 min on a multiple tube vortexer. The solvent and aqueous phases were allowed to separate for 5 min, and the aqueous phase was frozen over liquid nitrogen. The solvent fraction was decanted into 12 × 75-mm borosilicate glass assay tubes and dried overnight under a fume hood. Any remaining solvent was removed at 57°C under air. Once dried, the samples and standards were reconstituted in 100 μl of FTA hemagglutination buffer (Becton-Dickinson and Co., Sparks, MD) made with ultra-PURE water (Invitrogen). Reconstituted samples were immediately assayed using a double-antibody radioimmunoassay (DSL-4400; Diagnostic Systems Laboratories). The lowest standard used was 5 pg/ml. Interassay coefficients of variation were 7.5% and 9.0% for TP and E2, respectively, and intraassay coefficients of variation were 4.6% and 5.3%, respectively.
Tissues were collected and fixed either at 4°C for 4 h in 4% paraformaldehyde/0.25% glutaraldehyde or overnight at 4°C in Bouin fixative. After fixation, tissues were washed and embedded in paraffin. Five-micrometer sections were cut, stained with hematoxylin-eosin, and imaged using a Nikon (Natick, MA) Microphot.
Total RNA was extracted using the standard TRIzol protocol, quantified in an ND-1000 Spectrometer (NanoDrop, Wilmington, DE), and quality assessed using a Bioanalyzer 2100 (Agilent, Palo Alto, CA). Only samples with an RNA integrity number (determined using Agilent 2100 Expert Software) of greater than 7.0 were used for further analysis. This measure of quality takes into account both the ratio of ribosomal bands and the presence of degraded products and as such is an accurate measure of quality.
Samples from two different animals of approximately 10 μg of caput epididymidis RNA were labeled using a GeneChip One-Cycle Target Labeling Kit (Affymetrix, Santa Clara, CA) to produce labeled cRNA, which was then hybridized to the Affymetrix GeneChip Mouse Genome M430 2.0 microarray. Samples from two different animals of 50 ng of efferent duct RNA were labeled with the Ovation Biotin RNA Amplification and Labeling System (NuGEN Technologies, San Carlos, CA) and hybridized to the M430 2.0 microarray. Hybridization quality was assessed using GeneChip Operating Software version 1.4 (Affymetrix). Production of cDNA and cRNA hybridization to arrays and evaluation of hybridization quality were completed by the Laboratory for Biotechnology and Bioanalysis I at Washington State University.
Array output was normalized via the robust multiarray method, and data analysis was conducted using GeneSpring version 7.3 (Agilent). Genes were identified as hormone regulated if they 1) had a raw score of greater than 50 in at least one sample, 2) were determined to be significantly different versus castrated controls by ANOVA (P = 0.05) with a Benjamini and Hochberg false discovery rate multiple test correction, and 3) showed a 2-fold or greater increase or reduction versus castrated controls.
For real-time PCR, forward and reverse oligonucleotide primers were designed for specificity to target genes using Primer Express 2.0 (Applied Biosystems, Foster City, CA). Primers for real-time RT-PCR are given in Table 1. Two-step real-time RT-PCR was used to determine hormone regulation of microarray-identified transcripts using previously described techniques . Briefly, 300 ng of RNA was reverse transcribed using Superscript III RT (Invitrogen) per the manufacturer's instructions. For each treatment/tissue combination, RNA samples in triplicate were used to synthesize cDNA to be used as a template for real-time PCR. Analysis was carried out in a 96-well plate on a 7500 Fast Real-Time PCR System (Applied Biosystems). Reactions for each sample were carried out in triplicate, with each reaction containing equal amounts of cDNA, 1× Fast SYBR GREEN Mastermix (Applied Biosystems), and 600 nM of forward and reverse primers for each analyzed gene. Reaction conditions were as follows: 20-sec incubation at 95°C, followed by 40 cycles of 3-sec incubation at 95°C and 30-sec incubation at 60°C. The threshold cycle (CT) was calculated using the 7500 software version 2.0.1 (Applied Biosystems), and output was analyzed using the CT method described previously . Data are presented as average fold-change by treatment, and error is reported as the SEM. Statistical analysis was completed using JMP 7.0.1 (SAS Institute Inc., Cary, NC).
Probe IDs identified in the microarray analysis as hormone regulated were uploaded to DAVID Bioinformatics Resources (http://david.abcc.ncifcrf.gov/), and their biological process annotation was defined using the functional annotation tools. Only biological process gene ontology terms defined as significantly overrepresented by DAVID were included for further analysis.
Microarrays were used to investigate global gene expression in two regions of the proximal reproductive tract, the efferent ducts and the caput epididymidis, after treatment with one of two hormones, estrogen in the form of E2 and testosterone in the form of TP. The resulting microarray output was used to identify transcripts with altered levels under hormone treatment versus castrated controls within each tissue. Thus, hormone-regulated genes are defined as those whose probe sets had a raw signal exceeding 50 in at least one sample, in which the average signal represented at least a 2-fold difference between hormone-treated and castrated controls, and where the raw signals were found to be significantly different between treated and castrated controls by ANOVA (P = 0.05). Hormone-regulated probe sets were then examined to determine if the annotated genes were linked with specific biological processes and if specific processes were overrepresented.
To confirm the efficacy of hormone treatment, pretreatment and posttreatment serum hormone levels were determined for each animal. Pretreatment and posttreatment serum hormone levels for each group are given in Supplemental Table S1 (available at www.biolreprod.org). No difference was found between the pretreatment and posttreatment hormone levels in the castrated controls. However, in the E2-treated animals, serum E2 levels increased from an average of 35.2 ± 3.6 pg/ml before treatment to 3396.3 ± 1078.7 pg/ml after treatment. Serum testosterone concentrations increased from less than 0.05 ng/ml before TP treatment to an average of 7.7 ± 2.9 ng/ml after TP treatment. Sham-castrated animals had serum testosterone levels ranging from 2.0 to 9.9 ng/ml and E2 levels ranging from less than 6.25 to 30.5 pg/ml.
Histological sections of hematoxylin-eosin-stained efferent ducts and caput epididymides were analyzed to determine if significant changes in tissue morphology or cell population occurred as a result of hormone treatment. The proximal regions of the efferent ducts were examined in each of the hormone treatments and were compared with those of castrated animals. No significant differences in cellular morphology or composition were observed between hormone-treated and castrated animals. In the caput epididymides (regions II and III), no significant differences in cell composition and only minor decreases in brush border width and principal cell height were observed when comparing hormone-treated and castrated animals. Images of castrated, hormone-treated, and sham-castrated efferent ducts and caput epididymides are shown in Supplemental Figures S1 and S2.
The hormone-regulated gene lists identified in this study were queried for genes known to be sex hormone responsive. In all tissues and both hormone treatments, probe sets associated with genes previously demonstrated to be direct targets of either estrogen or testosterone were identified by our analysis. For example, collagen, type 1, alpha 2 (Col1a2) was shown to be up-regulated in all E2-treated tissues. Col1a2 expression is induced in immature uterus in response to E2 treatment . Additional probe sets associated with genes previously reported as estrogen regulated and identified in one or both E2-treated tissues include aquaporin 5 (Aqp5) , RAS-like estrogen-regulated growth inhibitor (Rerg) , and cystic fibrosis transmembrane conductance regulator homolog (Cftr) . Also identified as TP regulated in this study were probe sets associated with the following genes that have been previously shown to be regulated by testosterone : distintegrin and metallopeptidase domain 7 (Adam7), glutamate oxaloacetate transaminase 2 mitochondrial (Got2), and prostaglandin-endoperoxide synthase 1 (Ptgs1).
The numbers of transcripts differentially expressed in hormone-treated tissue relative to controls for each of the tissue and hormone combinations are given in Table 2. In the efferent ducts, 2328 transcripts had altered levels with E2 treatment, and 2588 transcripts had altered levels with TP treatment. In the caput epididymides, 97 transcripts had altered levels with E2 treatment, and 206 transcripts had altered levels with TP treatment. Transcripts with levels most dramatically altered by E2 and TP treatment in each tissue are given in Tables 3 and and4,4, respectively.
Transcripts with significantly altered levels in sham-castrated (intact) animals compared with castrated animals were identified and compared with transcripts having significantly altered levels after hormone treatment. Transcripts with similar expression patterns in both intact animals and hormone-treated animals compared with castrated controls were then identified. For example, aquaporin 1 (Aqp1) transcript levels are 3.64-fold higher in E2-treated efferent ducts compared with castrated controls and are 3.19-fold higher in intact animals compared with castrated controls. The numbers of transcripts found to be regulated in a similar manner in both intact and hormone-treated animals for each hormone and tissue combination are given in Table 5. In the caput epididymides, 29 transcripts had similarly altered levels in intact and E2-treated animals, and 121 transcripts had similarly altered levels in intact and testosterone-treated animals. In the efferent ducts, 95 transcripts had similarly altered levels in intact and E2-treated animals, and 141 transcripts had similarly altered levels in intact and testosterone-treated animals. Examples of genes encoding these transcripts are given in Tables 6 and and77.
Transcripts associated with genes that regulate fluid or solute transport were identified among those transcripts with similarly altered levels in intact and hormone-treated animals. Within the caput epididymides, E2 treatment resulted in altered levels of transcripts derived from two genes involved in solute transport (Cftr and solute carrier 26, member 3 [Slc26a3]), and TP treatment resulted in altered levels of transcripts derived from four solute carriers and from potassium channel, potassium voltage-gated channel, shaker-related subfamily, member 1 (Kcna1). Estradiol treatment of the efferent ducts resulted in altered transcript levels of four solute or fluid transporters, including Aqp1 and solute carrier family 16 (monocarboxylic acid transporter), member 4 (Slc16a4). Testosterone propionate treatment in the same tissue altered transcript levels of eight solute carriers and Aqp1. Intact animals compared with castrated controls were found to have altered transcript levels of 13 solute carriers, two potassium channels, Cftr, aquaporin 9 (Aqp9), and a chloride channel in the caput epididymides, as well as altered transcript levels of 17 solute carriers and Aqp1 in the efferent ducts.
Real-time RT-PCR results of transcripts with similarly altered levels in intact and hormone-treated animals are shown in Figures 1–3. In the efferent ducts (Figs. 1 and and2),2), the E2-responsive gene ELOVL family member 6, elongation of long-chain fatty acids (yeast) (Elovl6) was shown to have a 45-fold increase with E2 treatment and a 47-fold increase in intact animals compared with castrated controls. Elovl6 also had a 128-fold increase with TP treatment. Also in the efferent ducts, Slc26a3 demonstrated a 6-fold increase after TP treatment and a 5-fold increase in intact animals compared with castrated controls. Estradiol-treated animals showed a 6-fold decrease in Slc26a3 compared with castrated controls. Finally in the efferent ducts, solute carrier family 9 (sodium/hydrogen exchanger), member 3 (Slc9a3) had a 9-fold increase and a 13-fold increase in TP-treated and intact animals, respectively, compared with castrated controls and a 3-fold decrease in E2-treated animals compared with castrated controls. In the caput epididymides (Fig. 3), E2-repressed Slc26a3 had 2-fold and 3-fold increases in castrated and TP-treated animals, respectively, compared with intact and E2-treated animals, while E2-induced Cftr had 2-fold increases in intact and E2-treated animals. The E2- and TP-responsive lipase endothelial (Lipg) had a 1.5-fold increase after E2 treatment, a 5-fold increase after TP treatment, and a 10-fold increase in intact animals compared with castrated controls. Also in the caput epididymides, the TP-responsive myoinositol 1-phosphate synthase A1 (Isyna1) showed a 4-fold increase after TP treatment and a 5-fold increase in intact animals compared with castrated controls.
Ontology analysis was performed of probe sets identified as significantly altered with hormone treatment and of probe sets identified as both significantly altered with hormone treatment and expressed similarly in intact animals by DAVID . Several biological processes regulated by the sex hormones in the proximal reproductive tract were identified.
To identify specific genes mediating physiologically relevant processes, lists of probe sets associated with specific biological processes were queried. Processes identified as significantly regulated by one or both hormones in one or both tissues included transcription, steroid metabolism, and lipid metabolism.
A number of probe sets associated with regulation of transcription were identified as E2 and TP regulated in the efferent ducts and as TP regulated in the caput epididymidis. In both tissues, probe sets associated with transcription had predominantly lower signals with hormone treatment. In the caput epididymides, TP treatment decreased the signal of 17 probe sets associated with transcription, including several associated with nuclear receptors. In the efferent ducts, a large number of probe sets associated with transcription factors were identified as altered in hormone treatment, including sets associated with many members of the nuclear receptor family, several estrogen receptor or estrogen receptor-related receptors, and several Sry box (SOX)-containing genes. In all, 67 probe sets associated with transcription factors had altered signal with E2 treatment and 73 with TP treatment. Many of these probe sets had altered signal under both hormone treatment regimens.
Ontology analysis demonstrated that probe sets associated with steroid metabolism were most commonly altered by TP in the caput epididymidis and by E2 in the efferent ducts. In the caput epididymides, TP increased the signal of probe sets associated with genes such as sulfotransferase family 1E, member 1 (Sult1e1) and cytochrome P450, family 51 (Cyp51) and decreased the signal of probe sets associated with genes such as hydroxysteroid (17-beta) dehydrogenase 11 (Hsd17b11) and sterol O-acyltransferase 1 (Soat1). In the efferent ducts, E2 increased the signal of probe sets associated with genes involved in steroid metabolism such as cytochrome b5 reductase 1 (Cyb5r1), hydroxysteroid (17-beta) dehydrogenase 11 (Hsd17b11), and sterol-C5-desaturase (fungal ERG3, delta-5-desaturase) homolog (Saccharomyces cerevisae) (Sc5d).
Probe sets associated with lipid metabolism genes were regulated by both hormones in the efferent ducts and predominantly by TP in the caput epididymidis. In the efferent ducts, 72 and 30 probe sets associated with lipid metabolism had altered levels with TP and E2 treatments, respectively. Included among these are Lipg, apolipoprotein E (Apoe), Elovl6, and arachidonate 12-lipoxygenase (Alox12). In the caput epididymidis, probe sets associated with lipid metabolism having altered signal with TP treatment included Isyna1, 3-hydroxy-3-methyglutaryl-coenzyme A reductase (Hmgcr), and Lipg.
The male proximal reproductive tract is essential for male fertility, as it concentrates spermatozoa and provides an environment permissive for spermatozoa maturation. It has been demonstrated using various models that both estrogen and testosterone are capable of regulating the function of the proximal reproductive tract [6, 17, 23, 25, 27]; however, an analysis of the transcriptome-wide effect of both hormones on the two primary tissues of the proximal reproductive has not been available. To our knowledge, this work represents the first global transcriptome analysis examining the effects of both testosterone and estrogen on the efferent ducts and caput epididymidis. As such, the results described herein lead both to an improved understanding of the role of the two hormones in regulating proximal tract function and to identification of new target genes that may be important in maintaining optimal physiological function in these tissues.
Hormone treatment with both testosterone and E2 was confirmed to be effective. After introduction of each hormone, a measurable increase of that hormone was observed, and resulting levels were similar to or greater than those in intact animals. In addition, transcripts known to be regulated by a specific hormone were identified in each tissue under the expected hormone treatment regimen. Taken together, these results demonstrate that hormone treatment was effective and reflective of a normal hormonal state.
The total number of transcripts with altered levels under a given hormone treatment regimen relative to the other hormone within a tissue revealed a notable pattern. In the efferent ducts, both E2 and TP altered signals of similar numbers of probe sets; in the caput epididymides, TP treatment resulted in a much greater number of probe sets with altered signal than E2 treatment when hormone-treated tissue was compared directly with castrated tissue. These results confirm that the tissues of the male proximal reproductive tract respond differently to the two hormones, with the efferent ducts being equally sensitive to both hormones and the caput epididymidis being more responsive to testosterone. This observation is compatible with previous findings demonstrating that the transcriptome of the distal tissues of the caput epididymidis is more responsive to testosterone than the initial segment . The determination of transcripts with similar expression levels in both intact and hormone-treated animals provided insight into genes with potential physiological importance, as they are clearly regulated in the intact system. However, because the initial segment of the caput epididymidis regresses after castration [41, 42], the hormone response in this system may be altered compared with that in intact animals. Although it is likely that genes expressed similarly in intact and hormone-treated animals are truly regulated by hormones in the intact system, they may represent genes that are normally expressed in the initial segment and are only expressed in the more distal segments under abnormal conditions (such as a loss of the initial segment combined with hormone treatment). Thus, closer examination of specific targets identified in this work is required to determine the exact mechanism of hormone control in both castrated and intact animals.
In the case of genes identified as similarly regulated in both hormone-treated and intact animals, more transcripts had altered levels with TP treatment than with E2 treatment in both tissues. However, the fraction of transcripts with altered levels in TP-treated tissue compared with E2-treated tissue was greater in the caput epididymidis than in the efferent ducts. These results support the notion that estrogen has a more important role in the efferent ducts than in the caput epididymidis; however, both hormones are capable of influencing gene expression in both tissues.
Solute and fluid transporters are important players in modulating the major physiological functions of the proximal reproductive tract, namely, fluid uptake and establishment of the correct luminal environment. Thus, identification of hormone-regulated genes associated with these processes provides information about how testosterone and estrogen affect these important functions. Examination of solute and fluid transporters expressed in intact animals and in hormone-treated castrated animals revealed several notable results. First, although some of the solute carrier family members with altered transcript levels in intact animals relative to castrated controls have a similarly altered pattern in hormone-treated animals, many do not. Thus, other factors are involved in regulating the expression of these solute carriers. Second, many more solute transporters appear to have tightly regulated expression than the few that have been individually examined for their roles in the proximal reproductive tract such as Slc26a3 (a chloride anion exchanger, previously known as DRA) and Slc9a3 (a sodium/hydrogen exchanger, previously known as NHE-3). These unexamined solute transporters represent potentially important regulators of fluid and solute movement in these tissues and must be studied further to clarify their role. Overall, hormonal regulation of characterized and previously unreported fluid and solute transporters appears to be common; however, comparisons between intact and hormone-treated animals indicate that it is not the only mode of regulation for these potentially important modulators of tissue function.
Real-time RT-PCR was used to confirm hormone-induced altered transcript levels of several genes with known or hypothesized functions in the proximal reproductive tract. In addition, the analyses queried previously uncharacterized genes whose function may be important for proximal reproductive tract physiology.
Three genes with altered transcript levels after hormone treatment were examined via real time RT-PCR in the efferent ducts: Slc26a3, Slc9a3, and Elovl6. Two of these, Slc26a3 and Slc9a3, have clearly defined roles in the efferent ducts; however, this work represents the first description of their positive regulation by a sex hormone in this tissue, to our knowledge. Slc26a3 encodes a chloride anion exchanger, previously known as DRA, that along with other ion transporters regulates ion movement across the efferent duct epithelium. Repression of Slc26a3 by estrogens in the efferent ducts has been demonstrated using Esr1KO and ICI-treated animals ; however, its positive regulation by testosterone in the same tissue has not been examined until this work. Real-time RT-PCR analysis showed diametric hormone regulation of Slc26a3 expression, indicating that both hormones may be important in regulating its expression in the intact animal. Like Slc26a3, Slc9a3 encodes an ion transporter that is hypothesized to be important in the regulation of fluid homeostasis in the efferent ducts, although reports of its regulation by estrogens have been confounding. Slc9a3 message is known to be decreased in the Esr1KO model ; however, no repression is observed with ICI treatment, indicating that Slc9a3 is not directly regulated by estrogen receptor . Results herein demonstrated no significant effect on Slc9a3 message by E2 treatment and a significant increase in expression after TP treatment. Clearly, the regulation of Slc9a3 expression by one or both hormones needs to be explored further. Regulation of key ion transporters by one or both hormones indicates that both estrogen and testosterone may be important in maintaining the correct luminal environment of the efferent ducts. Elovl6 expression, function, and hormone responsiveness have not been examined in the efferent ducts prior to this work, to our knowledge. Elovl6 codes for an elongase for very long fatty acids, and it has been shown that the Drosophila homolog Baldspot (also known as neighbor of abl, noa) is essential for postmeiotic sperm development and is expressed in the supportive somatic cells of the male Drosophila gonad . Identification of a homologous function for Elovl6 in the mammalian reproductive tract would indicate that regulation of fatty acid or lipid metabolism has an important role in the efferent ducts. Regulation of Elovl6 by E2 shows that estrogen may regulate multiple aspects of efferent duct function, including fatty acid metabolism.
In the caput epididymidis, four genes with altered transcript levels after hormone treatment were examined. Two of the four have been hypothesized to have at least some role in regulation of reproductive tract function, while two have no known function in this tissue. Slc26a3 is regulated by estrogen in the efferent ducts ; however, a direct decrease in the level of transcript by E2 treatment has not been shown in the caput epididymidis until this work. As in the efferent ducts, TP treatment results in an increase in Slc26a3 message. Regulation of Slc26a3 expression by estrogen in the caput epididymidis demonstrates the importance of estrogen in the regulation of the luminal environment of the caput epididymidis, as well as the efferent ducts, while diametric regulation of Slc26a3 by both sex hormones in the caput epididymidis shows that a balance of both may be required for optimal tissue function. The role of Cftr, a chloride channel, in caput epididymal physiology has been extensively studied [44, 45]; however, reports of its regulation by estrogen have been confounding. It has been shown that estrogen represses Cftr expression in the efferent ducts , while it also induces expression in the female reproductive tract . Our work shows that Cftr is up-regulated by E2 in the caput epididymidis in a manner similar to the regulation in intact animals, supporting the notion that the regulation of this transporter by estrogen is important for maintaining the luminal environment of the caput epididymidis. Two transcripts of unknown function in the reproductive tract were also examined via real-time RT-PCR in the caput epididymidis, Isyna1 and Lipg. Both likely have some role in reproductive processes. Isyna1 encodes the first enzyme in the pathway producing myoinositol, an osmolyte that has unusually high concentrations in the seminiferous tubule and epididymal fluids . While Sertoli cells in the testis are the likely source of myoinositol in seminiferous tubule fluid, additional myoinositol in the epididymal fluid is likely produced by the epithelial cells of the epididymis as demonstrated by tissue enzyme activity . Regulation by testosterone of Isyna1 expression and the known role of Isyna1 in the regulation of osmolarity suggest that testosterone may alter the fluid microenvironment of the caput epididymidis via this important enzyme. However, TP treatment does not recapitulate Isyna1 expression in intact animals, and E2 treatment has no effect, demonstrating that normal Isyna1 expression in the caput epididymidis requires additional factors besides the sex hormones. Lipg encodes a lipase associated with cellular uptake of fatty acids and cholesterol. Lipg expression has been described in the principal cells of the epididymidis, and a potential role in regulating sperm cell motility has been proposed . Lipg regulation by both estrogen and testosterone indicates that both hormones can modulate the ability of the epididymal epithelium to produce cholesterol. Further examination of the regulation of both Isyna1 and Lipg by the sex hormones and the role that their products have in epididymal function should provide valuable information regarding osmolyte regulation and lipid metabolism in the caput epididymidis. Taken together, the real-time RT-PCR results in the caput epididymidis demonstrate several modes of hormone regulation in the intact animal: regulation by both hormones in an opposite manner, regulation by both hormones in a similar manner, and regulation by one hormone alone or in concert with other factors. These observations indicate that a balance of both hormones is required for optimal expression of the selected genes in the caput epididymidis and thus may also be required for optimal tissue function.
Ontology analysis was used to identify potential hormone-regulated functions in addition to solute and fluid transport in the proximal reproductive tract. This analysis identified overrepresented probe sets associated with transcription, steroid metabolism, and lipid metabolism. In both tissues, probe sets associated with transcriptional regulators commonly had lower signals after hormone treatment compared with castrated controls. In the caput epididymidis, which appears to be more sensitive to testosterone than E2, most of the probe sets associated with transcription were altered by TP treatment, whereas in the efferent ducts, which demonstrated equal sensitivity to both hormones, almost equivalent numbers of probes sets associated with transcription are altered by E2 and TP treatment. It is notable that most transcriptional regulators identified in this ontology analysis had decreased levels after hormone treatment. It is possible that many hormone-responsive transcriptional regulators in the proximal reproductive tract do not act predominantly to activate transcription but rather to repress constitutive expression.
Both hormones altered the signal of probe sets associated with lipid metabolic processes in the efferent ducts, and testosterone alone influenced similar probe sets in the caput epididymidis. Even in the efferent ducts, twice as many probe sets associated with lipid metabolism had altered signal with TP treatment than with E2 treatment, indicating that testosterone may have a more important role in regulating lipid metabolism in the proximal reproductive tract than estrogen. In the case of steroid metabolism, the classic regulator of efferent duct function, estrogen, was the predominant regulator in that tissue, while testosterone had the most effect in the caput epididymidis. Taken together, these results imply that both estrogen and testosterone have important roles in regulating transcription and lipid metabolism in the efferent ducts, while these roles are predominantly regulated by testosterone in the caput epididymidis.
Before this study, examination of hormone regulation of the proximal reproductive tract transcriptome was limited to the analysis of androgen-treated epididymis. However, in the efferent ducts, regulation by estrogen has previously appeared to have a greater role than that of testosterone, while the opposite was true in the caput epididymidis. This work aimed to increase our understanding of the effect that both sex hormones have on the proximal reproductive tract by identifying genes regulated by one or both sex hormones, by confirming hormone regulation of several biologically important genes, and by examining the ontological classes most commonly regulated by the two hormones of interest. The results of this study support the idea that both hormones have an important role in regulation of proximal reproductive tract function, with a greater role by testosterone than by estrogen in the caput epididymidis and a clear role for other factors in both tissues. This work serves as a starting point for further examination of the hormone responsiveness of the proximal reproductive tract transcriptome and provides valuable insight into potentially novel genes associated with proximal reproductive tract function.
The authors would like to thank Derek Pouchnik and the Laboratory for Bioanalysis and Biotechnology I (LBBI) for GeneChip processing and Lizhong Yang for assistance with microarray quality assessment.
1Supported by a grant from NICHD HD 10808.