5 Hox Genes and Meis 1, a Hox-DNA Binding Cofactor, Are Expressed ...

BIOLOGY OF REPRODUCTION 68, 644–650 (2003) Published online before print 23 October 2002. DOI 10.1095/biolreprod.102.009324

59 Hox Genes and Meis 1, a Hox-DNA Binding Cofactor, Are Expressed in the Adult Mouse Epididymis1 Daniela Bomgardner,2,3 Barry T. Hinton,4 and Terry T. Turner3,4 Departments of Urology3 and Cell Biology,4 University of Virginia Health Science System, Charlottesville, Virginia 22908 ABSTRACT

alogous genes). The position of a hox gene within a cluster correlates with the expression of that gene along an embryo’s A/P axis [2] and with the time that the gene is activated [3]. Mice with targeted disruptions of specific hox genes develop homeotic transformations due to the misspecification of body segments along the A/P axis of the skeleton and a variety of organ systems [4]. All hox genes share a DNA sequence, the homeobox, which encodes a 60 amino acid DNA-binding motif, referred to as the homeodomain. Proteins that contain such a domain are transcription factors because they control the transcription rate of their target genes [5]. Although the homeodomain is highly conserved each hox protein binds to specific DNA sequences. This fact gave rise to the idea that DNA binding cofactors must exist to aid in hox-DNA binding specificity. Indeed, another homeodomain protein, Meis 1, that favors binding with hox proteins from paralogous group 9–13 has been identified [6]. The combination of hox protein plus its cofactor or cofactors is believed to provide target specificity, stabilize the binding to the DNA, and control whether a hox protein will act as a transcriptional activator or repressor [7]. As the embryo is segmented in development, the adult epididymis is segmented in structure [8] and in its pattern of gene and protein expression [9]. Organ structure and function are closely related; thus the segmented nature of the epididymis may be important for sperm maturation and storage. It is currently believed that the complex interplay between regional (caput, corpus, and cauda) differences in androgen levels [10], activity of steroid metabolizing enzymes [11], and presentation of lumicrine signaling molecules [12] is responsible for regulating gene expression and protein synthesis, but the mechanisms of intraregional, segment-specific regulation have not been addressed. A strong line of evidence indicates an importance of hox genes in the male reproductive tract. Dolle´ et al. [13] reported the expression of 59 hox genes hoxd-10 and hoxd13 in the male reproductive tract of mammalian embryos. Since then, several 59 hox genes have been detected in the embryonic epididymis and vas deferens [14]. Further, hoxa10 and hoxa-11 knockout mice are sterile and the epididymis shows a homeotic transformation [15, 16]. In hoxa10 knockout mice the cauda epididymidis appeared to have transformed into the corpus, and in the hoxa-11 knockout mice the vas deferens displayed a partial transformation into a cauda. Although hox genes are not commonly studied after the natal period, their potential effectiveness beyond embryogenesis might play a pivotal role in the maintenance of segmental function in a variety of adult organs. For example, hoxb-8 expression has been detected in the adult mouse brain [17] where segmentation into hemispheres and lobes is important. Hox gene expression has also been de-

Hox genes determine the formation of segmented structures during development. The epididymis shows a segmented organization in its structure and function beyond embryogenesis. This study examined the adult mouse epididymis and vas deferens for expression of 59 hox genes and a hox-DNA binding cofactor. Reverse transcriptase-polymerase chain reaction (RTPCR) showed the expression of hoxa-9, hoxa-10, hoxa-11, hoxd9, and hoxd-10 in all regions including the vas deferens. Semiquantitative RT-PCR revealed highest mRNA levels for hoxa-11 in the distal part of the epididymis and vas deferens, and this was confirmed by Northern blot analysis. To determine protein presence an antibody raised against a peptide N-terminal to the homeodomain of hoxa-11 was produced in rabbits. The antibody recognized a band of approximately 37–39 kDa in Western blot analysis. Immunohistochemistry indicated the presence of hoxa11 in the nuclei of the epithelial cells with some staining in the cytoplasm. Staining was also detected in nuclei of interstitial cells throughout the entire organ and the vas deferens. A DNA binding cofactor for hoxa-11, Meis 1, was investigated for its presence in the epididymis. Semiquantitative RT-PCR identified both transcripts for Meis 1 (Meis 1a and Meis 1b) in all regions. Protein presence was confirmed by Western blot analysis, and this detected one band of approximately 53–55 kDa. Immunohistochemistry localized Meis 1 in the nuclei of interstitial cells throughout the entire organ and the vas deferens. Our study provides preliminary data from which we suggest the involvement of homeodomain transcription factors in the maintenance of segmental function of the adult epididymis and vas deferens. epididymis, male reproductive tract, vas deferens

INTRODUCTION

All vertebrates possess a common mechanism of development that involves hox genes. These genes are key regulators during development because they determine the patterning and segment identity along the anterior/posterior (A/P) axis of the embryo [1]. In mammals, 39 hox genes are distributed along four clusters designated Hox A, B, C, and D. Based on sequence similarity of the hox genes within the separate clusters, hox genes have been aligned from 39 to 59 into 13 paralogous groups (e.g., hoxa-9, hoxb-9, hoxc-9, and hoxd-9 are parThis work was supported by grants F32HD40706-01, DK 45179, and U54 HD28934 from the National Institutes of Health. Correspondence: Daniela Bomgardner, Department of Urology, Box 800422 (300 Lane Road), University of Virginia Health Science System, Charlottesville, VA 22908. FAX: 434 924 8311; e-mail: [email protected]

1

2

Received: 12 July 2002. First decision: 7 August 2002. Accepted: 3 September 2002. Q 2003 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

644

645

Hox GENES IN ADULT MOUSE EPIDIDYMIS

TABLE 1. PCR primer sequences, their annealing temperatures, product sizes and GenBank accession numbers. (T 5 temperature; acc. 5 accession). Primer hoxa-9 hoxa-10 hoxa-11 hoxd-9 hoxd-10 Meis 1 a/b

Sequence of primers (59 → 39) AAGAAAGGACTGTCCGTCTCCC CTAGCTTCCACAATCACAATGGG GTGAGTTCTGGGGCAGAGGC AGTTCCAAAGGCGAAAATGC CATACTCCTACTCCTCCAACCTGC CCCACCGTGCTATAGAAATTGG CACTACGGGATTAAGCCTGAAACC TTTGGGTCAAGTTGCTGCTGC CCTATGGAATGCAAACCTGTGG ATATCCAGGGACAGGAACCTCG CAGTTGGCACAAGATACAGGACTTACC CAAGAATGGAACTCCATAGTCCTTGC

tected in other adult tissues, such as the mammary gland [18], the uterus during implantation [19], and the prostate [20]. That hox genes determine segmental patterning during development, that the epididymis is maintained as a segmented structure in adulthood, and that hox gene expression has been shown in adult tissues, have led us to the hypothesis that hox genes play a role in the maintenance of segmental function of the adult epididymis. Our study was designed to 1) determine whether a number of 59 hox genes are expressed in the adult epididymis, 2) examine the regional expression pattern of a selected hox gene and its protein, and 3) investigate the presence and regionalization of the DNA-binding cofactor, Meis 1, at the mRNA and protein level. MATERIALS AND METHODS

Animals and Tissue Preparation Adult male (56–60 days) C57BL/6 mice were obtained from the University of Virginia vivarium and maintained on a 12L:12D light cycle with water and food ad libitum. Mice were killed and the epididymides and the vasa deferentia were removed. For RNA and protein isolation tissues were divided into initial segment, caput, corpus, and cauda of the epididymis and the vas deferens. These regions were pooled from three animals and the tissues were immediately snap-frozen in liquid nitrogen and stored at 2808C. For immunohistochemical procedures, whole epididymides and vas deferentia were fixed in Bouin fixative or 4% (w/v) paraformaldehyde (PFA) for 6–8 h. All experiments complied with the regulations set forth by the Animal Welfare Act (Public Law 91-579), the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996) published by the Department of Health and Human Services, and the policies and procedures of the University of Virginia Institutional Animal Care and Use Committee.

Reverse Transcriptase-Polymerase Chain Reaction Total RNA from each region of the epididymis and vas deferens was isolated using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions. All RNA samples were treated with DNase I (Invitrogen) to eliminate DNA contamination. Four micrograms of total RNA were reverse transcribed using Superscript II reverse transcriptase (RT) and random hexamers (Invitrogen) to obtain cDNA as described by the manufacturer. One microliter of RT product was used to perform the polymerase chain reaction (PCR) with primers specific to hoxa-9, hoxa-10, hoxa-11, hoxd-9, hoxd-10, hoxd-11, and Meis 1a/Meis 1b (Table 1). To amplify hoxa-11 and Meis 1a and Meis 1b using relative-quantitative RT-PCR, primers and competimers for 18S rRNA (489 base pairs [bp] or 324 bp; Ambion, Austin, TX) were added. Pilot experiments determined the ratio of competimers to primers for 18S rRNA to adjust the PCR products to the same linear range. Specific primer sets for all examined hox genes and Meis 1a/b were designed using Gene Runner (Hastings Software Inc., Hastings, NY) and were purchased from Invitrogen Life Technologies. The number of cycles used for amplification,

Number of cycles

Annealing temperature (8C)

Product size (bp)

30

60.3

488

M28449

29

68.9

353

NMp008263

34

59.1

262

NMp010450

30

61

344

pNM013555

30

56.6

320

NMp013554

30

57

658/528

GenBank accession number

U33630

the annealing temperature, product size, and GenBank accession number for each primer set are listed in Table 1. All reactions had an elongation temperature of 688C and were carried out with Platinum Taq DNA polymerase High Fidelity (Invitrogen) in a thermal cycler (Biometra, Go¨ttingen, Germany). Negative controls consisted of the omission of the Superscript II RT and the replacement of cDNA with water. Each RT-PCR experiment was repeated four times. All RT-PCR products were separated on 1% agarose gels. The identity of each amplified RT-PCR product was verified by sequence analysis after cloning into pCR BluntII-TOPO vector (hoxa-9, hoxa-10, hoxa-11, hoxd-9, and hoxd-10) and pCR 2.1-TOPO (Meis 1a and Meis 1b). Both vectors were purchased from Invitrogen.

Northern Blot Analysis For the analysis of hoxa-11, total RNA was extracted from the different regions of the epididymis and vas deferens as described above. For each region, 20 mg of total RNA was electrophoresed on an agarose/formaldehyde gel at constant voltage. The RNA was then transferred to a ZetaProbe membrane (Bio-Rad, Hercules, CA) by capillary action and the membrane was baked for 2 h at 808C. To prepare the probe, restriction enzyme digestion using EcoRI released hoxa-11 (262 bp) cDNA from its cloning vector. Labeling of the cDNA probes with [a32P]dATP (ICN Biomedicals, Aurora, CA) was carried out using the Prime-It II Random Labeling Kit (Stratagene, La Jolla, CA) according to the manufacturer’s instruction. The membrane was hybridized overnight at 67.88C in Modified Church and Gilbert buffer (0.5 M phosphate buffer pH 7.2, 7% SDS [w/ v], and 10 mM EDTA). Following hybridization, the membrane was washed in 2% saline-sodium citrate (SSC; 13 SSC 5 0.15 M sodium chloride and 0.015 M sodium citrate) and 0.1% SDS and exposed using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). To determine equivalency of load and transfer for hoxa-11, membranes were doublelabeled with glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The experiments were performed three times.

Western Blot Analysis For protein isolation, tissues were homogenized in liquid nitrogen. Upon evaporation of the liquid nitrogen, radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris, 150 mM NaCl, 10 mM EDTA, 1.0% NP-40, 0.5% sodium deoxycholate, and 0.1% SDS) including protease inhibitors (2 mg/ml aprotinin, 100 mM leupeptin, 10 mM E-64, and 1 mM PMSF) was added to the samples. Samples were allowed to sit on ice for 10 min before centrifugation (12 000 3 g for 10 min at 48C). The supernatant was transferred to a fresh tube. Protein samples (20 mg for hoxa-11, 40 mg for Meis 1) were diluted in Laemmli sample buffer [21], boiled for 5 min, and electrophoresed on 10% polyacrylamide gels. Separated proteins were electrotransferred to nitrocellulose membranes (Bio-Rad). To ensure equal loading of protein samples, membranes were stained with Ponceau S (Sigma, St. Louis, MO). Membranes were then blocked in Tris-buffered saline (TBS) with 5% milk and 0.1% Tween-20 for 1 h and incubated overnight at 48C with the primary antibody against hoxa-11 (1:8000) or Meis 1 (1: 1000) in TBS with 5% milk and 0.1% Tween-20. The hoxa-11 antibody was generated in rabbits against a region N-terminal to the homeodomain (acetyl-ASSDYPGDKNAEKGPQ-amide; QCB, Hopkinton, MA). This peptide sequence was used to search the National Center for Biotechnology Information database; no other proteins contained this specific sequence. A guinea pig antibody against Meis 1 was kindly provided by Dr. Corey Largman (Departments of Dermatology and Medicine, VA Medical

646

BOMGARDNER ET AL. Center and University of California, San Francisco, CA). This antibody has been tested and reported previously [2]. Membranes were washed in TBS/0.1% Tween-20 and incubated with 1:5000 diluted peroxidase-conjugated goat anti-rabbit (for hoxa-11; Vector Laboratories, Burlingame, CA) or donkey anti-guinea pig (for Meis 1; Jackson ImmunoResearch Laboratories Inc., West Grove, PA) secondary antibodies for 1 h at room temperature. Peroxidase activity was detected using enhanced chemiluminescence (SuperSignal West Pico; Pierce Chemical Co., Rockford, IL). Control experiments consisted of 1) omission of the primary antibody and 2) preabsorption of the antibody with the specific hoxa-11 peptide or the whole Meis 1 protein. All experiments were repeated four times.

Immunohistochemistry Epididymides including vasa deferentia that had been fixed in PFA or Bouin were paraffin embedded for subsequent immunohistochemical localization of hoxa-11 and Meis 1, respectively. Sections (4 mm) were deparaffined, progressively rehydrated, washed in water followed by PBS, and treated with 3% hydrogen peroxide in methanol for 30 min to destroy endogenous peroxidase activity. For antigen retrieval, sections were boiled for 20 min in unmasking solution (Vector Laboratories) and then blocked for 1 h in PBS containing 10% normal goat (for hoxa-11) or horse (for Meis1) serum (Vector Laboratories), 0.5% Fish skin gelatin (FSG; Sigma) and avidin blocking solution (4 drops/ml; Vector Laboratories). Primary antibodies were diluted in PBS containing 0.5% FSG and 5% normal serum (goat or horse). The polyclonal rabbit antibody hoxa-11 (QCB, see above) was diluted 1:6000 and the polyclonal guinea pig antibody Meis 1 was diluted 1:200, and both were incubated overnight at 48C. Sections were then washed in PBS/0.5% FSG and incubated for 1 h in biotinylated goat anti-rabbit or goat anti-guinea pig secondary antibody (Vector Laboratories) diluted 1:200 in PBS, 0.5% FSG, and 5% normal serum. Sections were washed in PBS/0.5% FSG and the biotinylated secondary antibodies were visualized with avidin-biotin-peroxidase complex (Elite ABC Kit; Vector Laboratories) and diaminobenzidine (Sigma) as a chromogen. Sections were counterstained with hematoxylin, dehydrated, and mounted in Vectamount permanent mounting medium (Vector Laboratories). The specificity of the immunoreactions was verified by omission of the primary antibody or by preabsorption of the antibody with an excess of hoxa-11 peptide (QCB) or Meis 1 full-length protein. All experiments were repeated four times.

RESULTS

Expression of hox Genes in Adult Mouse Epididymis and Vas Deferens

FIG. 1. Hox gene expression in all regions of adult mouse epididymis and vas deferens. A) RT-PCR analysis of hoxa-9, hoxd-9, hoxa-10, and hoxd-10. B) Relative-quantitative RT-PCR analysis of hoxa-11. C) Northern blot analysis of hoxa-11. L, 100-bp ladder; the brightest band represents 600 bp. C1, No RT control; C2, water control.

Of the six hox genes examined by RT-PCR, five (hoxa9, hoxa-10, hoxa-11, hoxd-9, and hoxd-10 were expressed in all regions of the epididymis and the vas deferens (Fig. 1, A and B). Sequence analysis demonstrated the authenticity of all RT-PCR products. Hoxd-11 expression was not detected even after additional manipulation of the primers and PCR conditions (Table 1). With the exception of hoxd9, all the analyzed genes appeared to qualitatively exhibit an expression pattern that predominated more distally than proximally (Fig. 1, A and B). Because hoxa-11 knockout mice exhibit a homeotic transformation of the distal epididymis and vas deferens [4] we decided to focus further on that specific gene. Subsequently, we examined hoxa-11 gene expression in more detail with relative-quantitative RT-PCR and Northern analysis. Relative-quantitative RT-PCR (Fig. 1B) substantiated the preliminary findings that were indicated by standard RTPCR, which is not shown because of redundancy. Relatively low expression of hoxa-11 was detected in the initial segment followed by no to low expression in the caput and corpus, and relatively high expression in the cauda and vas deferens (Fig. 1B). In the representative analysis illustrated in Figure 1B, hoxa-11 expression in the cauda and vas deferens was approximately 9-fold and 13-fold higher than in the initial segment. Quadruplicate experiments consistently revealed a similar pattern. RT-PCR products were never


647

where nuclei appeared almost black (Fig. 3C). Staining returned to moderate levels in the vas deferens (Fig. 3D). Sections incubated with the antibody that had been preabsorbed with the hoxa-11 peptide showed nuclear staining to have been eliminated and only a very light staining was exhibited exclusively in the cytoplasm (data not shown). The tissue was devoid of any staining when incubated with secondary antibody only (Fig. 3E). FIG. 2. Relative-quantitative RT-PCR analysis of Meis 1. One primer set recognizes both transcripts, Meis 1a and Meis 1b. L, 100-bp ladder; the brightest band represents 600 bp. C1, No RT control; C2, water control.

Presence of Meis 1 Protein in Adult Mouse Epididymis and Vas Deferens

observed in the no-RT and water control lanes of any gel (Fig. 1B). The QuantumRNA 18S Internal Standards (Ambion) showed a 489-bp product and confirmed equal loading. To confirm these results and to determine the presence of any additional transcripts, Northern blot analysis was performed using a 262-bp hoxa-11 cDNA probe. A single band of approximately 2.5 kilobases was detected exclusively in the cauda epididymis and vas deferens (Fig. 1C). This was the expected size of the transcript based on previous reports [4]. Blots were also probed with GAPDH, which was used as a load control (Fig. 1C). Expression was not detected in the initial segment, but in the example used in Figure 1C hoxa-11 expression in the vas deferens was 25-fold greater than in the cauda. This pattern was repeated in triplicate experiments.

The guinea pig Meis 1 antibody recognized one band of approximately 53–55 kDa (Fig. 4A) in all regions of the epididymis and the vas deferens (Fig. 4A), but with very faint to no detection in the cauda region. The size of this band is consistent with the reported Meis 1b form [2]. This band was abolished when the antibody was preincubated with the full-length protein and when the secondary antibody was used alone (Fig. 4A). Immunohistochemistry detected Meis 1 in the nuclei of the cells of the epididymal interstitium (Fig. 4, B–D). The epithelium was devoid of any staining. This staining pattern could be observed throughout the entire epididymis (Fig. 4, B and C) as well as in the vas deferens (Fig. 4D). The intensity of this staining did not appear to change along the organ. Staining was not detected on sections that had been incubated with the secondary antibody only (Fig. 4E). Similarly, when the antibody was preabsorbed with the fulllength Meis 1 protein, staining did not occur (data not shown).

Expression of the DNA Binding Cofactor Meis 1 in Adult Mouse Epididymis and Vas Deferens

Meis 1 gene expression was detected in all regions examined using relative-quantitative RT-PCR (Fig. 2). Two transcripts, Meis 1a (658 bp) and Meis 1b (528 bp) were identified in all regions of the epididymis and in the vas deferens. All control reactions were negative (Fig. 2). The QuantumRNA 18S Internal Standards (Ambion) showed a 324-bp product and confirmed equal loading. As with hoxa11, Meis 1 expression in the cauda and vas deferens exceeded that in more proximal segments. In the example illustrated in Figure 2, total Meis 1 expression in the cauda and vas deferens was 1.9-fold and 1.5-fold, respectively, of that in the initial segment. Similar results occurred in five separate experiments. Presence of hoxa-11 Protein in Adult Mouse Epididymis and Vas Deferens

The affinity-purified polyclonal rabbit anti-hoxa-11 antibody was used for Western blot analysis. One specific band of approximately 37–39 kDa was detected in all regions of the adult mouse epididymis including the vas deferens (Fig. 3A). Preabsorption with the immunizing peptide as well as the secondary antibody alone did not detect any bands (Fig. 3A). Immunolocalization of hoxa-11 detected its presence in the epithelium and interstitium of the entire adult mouse epididymis and vas deferens (Fig. 3, B–D). In epithelial cells, staining was observed most intensely in the nuclei, but staining in the cytoplasm was also apparent (Fig. 3, B– D). Nuclear localization was also evident in the cells of the epididymal interstitium (Fig. 3, B–D). Moderate levels of staining were observed in the initial segment and proximal caput (Fig. 3B) but became progressively stronger toward the cauda epididymidis in both interstitium and epithelium,

DISCUSSION

Hox genes determine the formation of segmented structures during development [1]. The epididymis continues to exhibit a segmented organization into adulthood [8, 9] and this feature of the organ contributes to the luminal microenvironment required for sperm maturation and storage [22]. Approaches that try to identify mechanisms resulting in this segment-specific gene and protein expression are limited, but 59 hox genes are candidates for being involved in this process. Their transcription and translation in the adult epididymis has not been investigated. We examined the expression of hoxa-9, hoxa-10, and hoxa-11 as well as their paralogues from the D cluster, hoxd-9, hoxd-10, and hoxd-11 in adult mouse epididymis and vas deferens. Expression for all but hoxd-11 was detected. It was interesting that RT-PCR showed a regionalized expression pattern for all chosen genes, except for hoxd-9. Generally, higher mRNA levels were detected in the distal part of the epididymis and in the vas deferens compared with the proximal region (Fig. 1, A and B). These data gave a first indication that 59 hox genes are expressed in adult epididymis and might play a role as in other adult tissues as well [18–20]. The use of various primer sets and different PCR conditions did not detect hoxd-11 expression. Whether this result was due to the experimental conditions or to the absence of hoxd-11 expression remains uncertain. Based on the above results plus the known epididymal transformation in hoxa-11 knockout mice [16] and homeobox swap experiments in which it was shown that other hox genes cannot substitute for hoxa-11 [23] we decided to focus attention on that gene and gene product. Relative-quantitative RT-PCR showed an expression pattern with low mRNA levels in the initial segment, low to none in the caput and corpus, and highest mRNA amounts in the cauda

648

BOMGARDNER ET AL.

FIG. 3. Hoxa-11 presence in adult mouse epididymis and vas deferens. A) Western blot analysis of tissue lysates isolated from all different regions. C1, Omission of the primary antibody; C2, preabsorption of the antibody with the hoxa-11 peptide. B–E) Immunohistochemical localization on tissue sections. B) Initial segment, C) cauda, and D) vas deferens. E) Incubated with secondary antibody only. Bars 5 100 mm.

and the vas deferens (Fig. 1B). Northern blot analysis confirmed the expression of hoxa-11 in the cauda and vas deferens (Fig. 1C). Expression was not detected in the initial segment, probably due to the lower sensitivity of this method. Previously, the existence of three transcripts for hoxa11 had been reported [16] using mRNA of a whole mouse embryo. However, using mRNA purified from developing limbs detects only one band (2.5 kb) [16]. Similarly, we detected one transcript of approximately 2.5 kb, which indicates the possibility that hoxa-11 expression is regulated differentially in the different tissues. To determine the presence of hoxa-11 protein, an antibody was raised against a region N-terminal to the hoxa11 homeodomain to ensure specificity of the immunogen. The hoxa-11 cDNA encodes a 313 amino acid protein with a predicted molecular mass of 34.5 kDa. The use of another hoxa-11 antibody had previously demonstrated the presence

of a doublet (;35–39 kDa). The presence of the doublet was speculated to be the result of posttranslational modification [24]. In contrast, our affinity-purified antibody recognized a single band of approximately 35–37 kDa in Western blots containing epididymal lysates isolated from all regions of the adult epididymis as well as the vas deferens (Fig. 3A). These data indicated that hoxa-11 is both transcribed and translated in the epididymis and lends credence to the possibility that the transcription factor has functional relevance. Hoxa-11 protein was detected in the caput and corpus epididymidis, where we did not observe mRNA expression. Several possibilities may explain this phenomenon. First, the antibody might bind to sequences or conformations of other proteins nonspecifically; however, we were able to block antibody binding by preabsorption of the antibody with the hoxa-11 peptide (see Materials and Methods) in


649

FIG. 4. Meis 1 presence in adult mouse epididymis and vas deferens. A) Western blot analysis of tissue lysates isolated from all different regions. C1, Omission of the primary antibody; C2, preabsorption of the antibody with the full-length Meis 1 protein. B–E) Immunohistochemical localization on tissue sections. B) Initial segment, C) cauda, and D) vas deferens. E) Incubated with secondary antibody only. Bars 5 100 mm.

Western blot analysis using one-dimensional (Fig. 3A) and two-dimensional gel (data not shown) electrophoresis, and the antibody detected a single band at the correct molecular weight. This strongly indicated that the binding of our antibody was specific for hoxa-11. Second, a paracrine mechanism has been demonstrated for homeodomain proteins in vitro [25]. In a similar fashion, hoxa-11 might be synthesized and secreted in the initial segment, where hoxa-11 mRNA was found, and internalized in the corpus or cauda. These possibilities will need to be addressed in future experiments. Immunohistochemical localization of hoxa-11 with this same antibody resulted in prominent staining in nuclei of epididymal epithelial and interstitial cells. This is expected because hoxa-11 is a transcription factor. This staining pattern was found throughout the entire epididymis and in the vas deferens, which is in accordance with the detected pro-

tein presence in the Western blot analysis. We also observed cytoplasmic staining in epithelial cells. In preabsorption controls, the nuclear staining was completely abolished, indicating that the antibody recognized hoxa-11 protein there. A weak signal was still detected in the cytoplasm, indicating some nonspecific binding of the antibody. As described earlier, hox proteins require cooperation with other proteins to bind to their target DNA. Meis 1 in particular aids 59 hox proteins, such as hoxa-11, to gain this specificity [6]. Using relative-quantitative RT-PCR we detected the transcripts of both Meis 1a and Meis 1b (Fig. 2). This points to the possibility that Meis 1 is an active cofactor of hoxa-11 in the adult epididymis. To determine the protein presence we used a polyclonal guinea pig antibody that recognizes both proteins, Meis 1a and Meis 1b [26]. Western blot analysis showed a single band of approximately 53–55 kDa (Fig. 4A), which is consistent with the

650

BOMGARDNER ET AL.

reported Meis 1b form [27]. Meis 1b protein presence was noticed in all regions of the epididymis, and this is interesting in the light that hoxa-11 forms very strong DNA binding complexes specifically with Meis 1b [6]. Using the same antibody, Meis 1 was localized in the nuclei exclusively of interstitial cells. Because both hoxa11 and Meis 1b were both found in interstitial cells these homeodomain proteins possibly interact in this compartment, whereas hoxa-11 in the epithelium might be either inactive or require another DNA binding cofactor. This possibility remains to be determined. We speculate that hoxa-11 and Meis 1 interact and bind to target DNA and that this target DNA encodes for factors that influence the epididymal epithelium. The possibility of an interstitial-epithelial cross talk is virtually unexamined in the epididymis. Also, little attention has been paid to downstream candidates of hox proteins. Among the few identified, cell adhesion molecules appear to be one target group, in particular NCAM [28]. We have detected NCAM gene and protein expression in adult mouse epididymis, interestingly with higher mRNA levels in the distal region. Future studies will determine whether NCAM or other cell adhesion molecules are downstream targets in the adult epididymis. In conclusion, a hox transcription factor, hoxa-11, and its potential DNA binding cofactor, Meis 1, have been identified in adult mouse epididymis. Both were found at the mRNA and protein levels. The interaction of both proteins and their function in epididymis will be determined in our future studies. ACKNOWLEDGMENTS We express appreciation to Dr. Corey Largman for his generous gift of the Meis 1 antibody, protein, and serum and to Jason P. Jacobs for his excellent technical help. We thank Dr. Isa Hussaini for his help with Northern blot analysis.

REFERENCES 1. Favier B, Dolle´ P. Developmental functions of mammalian Hox genes. Mol Hum Reprod 1997; 3:115–131. 2. Graham A, Papalopulu N, Krumlauf R. The murine and Drosophila homeobox gene complexes have common features of organization and expression. Cell 1989; 57:367. 3. Duboule D. Temporal colinearity and the phylotypic progression: a basis for the stability of a vertebrate Bauplan and the evolution of morphologies through heterochrony. Dev Suppl 1994; 120:135–142. 4. Redline RW, Neish A, Holmes LB, Collins T. Homeobox genes and congenital malformation. Lab Invest 1992; 66:659–670. 5. Levine M, Hoey T. Homeoproteins are sequence specific transcription factors. Cell 1988; 55:537–540. 6. Shen WF, Montgomery JC, Rozenfeld S, Moscow JJ, Lawrence HJ, Buchberg JM, Largman C. Abd-like Hox proteins stabilize DNA binding by the Meis1 homeodomain proteins. Mol Cell Biol 1997; 17: 6448–6458. 7. Affolter M, Marty T, Vigano MA. Balancing import and export in development. Genes Dev 1999; 13:913–915.

8. Setchell BP, Maddocks S, Brooks DE. Anatomy, vasculature, innervation, and fluids of the male reproductive tract. In: Knobil E, Neill JD (eds.), The Physiology of Reproduction, 2nd ed. New York: Raven Press; 1994: 1063–1175. 9. Kirchhoff C. Gene expression in the epididymis. Int Rev Cytol 1999; 188:133–202. 10. Brooks DE. Effect of androgens on protein synthesis and secretion in various regions of the rat epididymis, as analyzed by two-dimensional gel electrophoresis. Mol Cell Endocrinol 1983; 29:255–270. 11. Robaire B, Viger RS. Regulation of epididymal epithelial cell functions. Biol Reprod 1995; 52:226–236. 12. Hinton BT, Palladino MA. Epididymal epithelium: its contribution to the formation of a luminal fluid microenvironment. Microsc Res Tech 1995; 30:67–81. 13. Dolle´ P, Izpisua-Belmonte JC, Brown JM, Tickle C, Duboule D. Hox4 genes and the morphogenesis of mammalian genitalia. Genes Dev 1991; 5:1767–1776. 14. Rao M, Wilkinson MF. Homeobox genes and the male reproductive system. In: Robaire B, Hinton BT (eds.), The Epididymis. New York: Kluwer Academic/Plenum Publishers; 2002:269–283. 15. Benson GV, Lim H, Paria BC, Satokata I, Dey SK, Maas RL. Mechanisms of reduced fertility in Hoxa-10 mice: uterine homeosis and loss of maternal Hoxa-10 expression. Development 1996; 122:2687– 2696. 16. Hsieh-Li HM, Witte DP, Weinstein M, Branford W, Li H, Small K, Potter SS. Hoxa 11 structure, extensive antisense transcription, and function in male and female fertility. Development 1995; 121:1373– 1385. 17. Greer JM, Capecchi MR. Hoxb8 is required for normal grooming behavior in mice. Neuron 2002; 33:23–34. 18. Chen F, Capecchi MA. Paralogous mouse Hox genes, Hoxa9, Hoxb9, and Hoxd9, function together to control development of mammary gland in response to pregnancy. Proc Natl Acad Sci U S A 1999; 96: 541–546. 19. Daftary GS, Taylor HS. Implantation in the human: the role of HOX genes. Semin Reprod Med 2000; 18:311–320. 20. Sreenath T, Orosz A, Fujita K, Bieberich CJ. Androgen-independent expression of hoxb-13 in the mouse prostate. Prostate 1999; 41:203– 207. 21. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227:680–685. 22. Cornwall GA, Lareyre JJ, Matusik RJ, Hinton BT, Orgebin-Christ MC. Gene expression and epididymal function. In: Robaire B, Hinton BT (eds.), The Epididymis: From Molecules to Clinical Practice. New York: Kluwer Academic/Plenum Publishers; 2002: 269–283. 23. Zhao Y, Potter SS. Functional comparison of the hoxa 1, hoxa 10 and hoxa 11 homeoboxes. Dev Biol 2002; 244:21–36. 24. Gendron RL, Paradis H, Hsieh-Li HM, Lee DW, Potter SS, Markoff E. Abnormal uterine and glandular function associated with maternal reproductive defects in hoxa-11 null mice. Biol Reprod 1997; 56: 1079–1105. 25. Prochiantz A. Homeodomain-derived peptides. In and out of the cells. Ann N Y Acad Sci 1999; 886:172–179. 26. Lawrence HJ, Rozenfeld S, Cruz C, Matsukama K, Kwong A, Buchberg AM, Largman C. Frequent co-expression of the HOXA9 and MEIS1 homeobox genes in human myeloid leukemias. Leukemia 1999; 13:1993–1999. 27. Shen WF, Rozenfeld S, Kwong A, Ko¨mu¨ves LG, Lawrence HJ, Largman C. HOXA9 forms triple complexes with PBX2 and MEIS1 in myeloid cells. Mol Cell Biol 1999; 19:3051–3061. 28. Edelman GM, Jones FS. Developmental control of N-CAM expression by Hox and Pax gene products. Philos Trans R Soc Lond B Biol Sci 1995; 349:305–312.