Expression, immunolocalization and sperm-association of a protein derived from 24p3 gene in mouse epididymis

Expression, Immunolocalization and

Sperm-Association of A Protein Derived

From 24p3 Gene in Mouse Epididymis

2_{Institute of Biochemical Science, College of Science, National Taiwan University, Taipei, Taiwan}

ABSTRACT The cDNA sequence for 24p3 protein in ICR mouse epididymal tissue was deter-mined by PCR using primers designed according to the cDNA sequence derived from 24p3 protein in mouse uterine tissue. In the present study, 24p3 protein was immunolocalized in the epithelial cells and lumen of mouse epididymis. Both immunoblot analysis for protein and northern blot analysis for mRNA level showed a declining gradient of24p3 expression from the caput to caudal region of the epididymis. The 24p3 protein was undetectable in the testis. These findings suggest that the 24p3 protein is a caput-initiated secretory protein in the mouse epididymis. A postnatal study revealed that24p3 gene expression occurred in mice at the age of 14 days, before the completion of epididymal differentiation. This expression remained at a constant level until epididymal differentiation was completed. We also found that the secreted 24p3 protein interacted predominantly with the acrosome of caudal spermatozoa. Our findings suggest that the epididymal 24p3 protein is a caput-initiated and sperm-associated gene product and may be important in the reproductive system. Mol. Reprod. Dev. 57:26±36, 2000.ß 2000 Wiley-Liss, Inc.

Key Words: caput-initiated; epididymis; lipocalin; secretory protein; spermatozoa

INTRODUCTION

Mammalian spermatozoa acquire their motility and the ability to fertilize an oocyte during passage through the epididymis. These maturational events are believed to be dependent on the microenvironment created by the absorptive and secretory functions of the epididymis (Orgebin-Crist and Fournier-Delpech, 1982; Amann et al., 1993; Hinton and Palladino, 1995). The principal components of this environment are speci®c proteins that are synthesized and secreted in certain regions (caput, corpus or cauda) of the epididymis (Garberi et al., 1979; Brooks, 1981; Devine and Carroll, 1985). These proteins may be important in spermatozoal changes that occur in the different regions of the epididymis during postnatal develop-ment and may also be involved in regulating the functional integrity of spermatozoa. Although the

regional-speci®c expression of epididymal proteins has been established for some time, the identity and function of these proteins in spermatozoa have not been elucidated. A number of secretory proteins from epididymal epithelium and the hormonal regulation of their synthesis have been previously studied (Rankin et al., 1992; Lefrancois et al., 1993; Bendahmane and Abou-Haila, 1994), but the biological functions of these proteins remain unclear. While the ontogeny of epididymal protein expression in rats has been studied (Brooks, 1987; Charest et al., 1989; Ueda et al., 1990) and rabbits (Toney and Danzo, 1989), few comparable studies of mice epididymal proteins have been reported.

We previously characterized a 25 kDa mouse uterine glycoprotein named mouse 24p3 protein, which is an estrogen-regulated lipocalin secreted from the uterine epithelium. Its cDNA sequence is identical to that of 24p3-cDNA, which has been cloned from primary cultures of SV40-infected kidneys in mice (Hraba-Renevey et al., 1989; Chu et al., 1997, 1998). Previously, we puri®ed 24p3 protein from mouse uterine luminal ¯uid and demonstrated by northern blot analysis that the 24p3 gene is normally expressed in the lung, spleen, uterus, vagina and epididymis of mice (Chu et al., 1996). In the reproductive organ of the male mouse, 24p3 gene expression is unique to the epididymis. The protein derived from the 24p3 gene has also been found in lipopolysaccharide-stimulated mouse PU5.1.8 macrophage cells (Meheus et al., 1993) and bFGF-stimulated 3T3-cells (Davis et al., 1991). Liu and Nilson-Hamilton (1995) showed that 24p3 protein is an acute phase protein in liver. The function of 24p3 protein in the reproductive tract has not been reported. In the present study, we found that the epididymal caput is responsible for 24p3 protein secretion and 24p3 gene expression. We also established that the 24p3

ß 2000 WILEY-LISS, INC.

Grant sponsor: National Science Council, Taiwan; Grant number: NSC87-2311-B001-114.

*Correspondence to: Sin-Tak Chu, Institute of Biological Chemistry, Academia Sinica, P.O. Box 23-106, Taipei, 10617, Taiwan.

E-mail: [email protected]

protein associates with epididymal spermatozoa. The 24p3 protein's synthesis and secretion in epididymis and association of 24p3 protein on spermatozoa may concern spermatozoal maturation.

MATERIALS AND METHODS Materials

Diethylstilbestrol dipropinate (DES) and aprotinin were obtained from Sigma (St. Louis, MO). [a-32_P]dATP

and125_{I-labeled anti-rabbit IgG prepared from donkeys}

were purchased from Amersham International (Bucks, U.K.). Anti-rabbit alkaline phosphatase and IgG-¯uorescein conjugate prepared from goats were obtained from Sigma (St. Louis, MO). All of the reagents and enzymes used in cDNA preparation, PCR, and the T7 DNA polymerase sequencing system were purchased from Promega (Madison, WI). A Geneclean kit was purchased from BIO101, Inc. (La Jolla, CA). The random primed DNA probes (Prime-A-Gene kit) were obtained from Promega (Madison, WI). All chemicals were of reagent grade.

Animals

Outbred male mice were purchased from Charles River Laboratories (Wilmington, MA) and were main-tained and bred in the animal center at the College of Medicine, National Taiwan University. Animals were treated in accordance with the institutional guidelines for the care and use of experimental animals. The animals were killed by cervical dislocation and the epididymis was removed from each animal for further study.

RNA Isolation, cDNA Preparation and Northern Analysis

Total cellular RNA was isolated and double-stranded cDNAs were prepared on the polyadenylated fraction of epididymal RNA by a standard procedure (Sambrook et al., 1989) using AMV reverse transcriptase (Pro-mega, Madison, WI). Total cellular RNA was isolated and electrophoresed in 1% (w/v) agarose gel containing 3-[N-morpholino]propanesulfonic acid (MOPS) buffer (1 mM EDTA, 20 mM MOPS, 5 mM Na-acetate, pH 7.0) and 2.2 M formaldehyde. Transfer to nylon ®lter (hybond-N‡_{; Amersham) for blotting and northern}

analysis was performed as previously described (Tho-mas, 1980). Hybridization to the speci®c probe was performed overnight at 42_{C in a hybridization buffer.}

The32_{P-labeled random-priming kit used a template of}

a cDNA segment of the mouse 24p3 gene (579 bp) (Hraba-Renevey et al., 1989) or a cDNA segment of the mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene (1.25 kbp) inserted into the pGEM3 vector.

Polymerase Chain Reaction (PCR)

Based on the previously established structure of mouse 24p3 cDNA (Hraba-Renevey et al., 1989), we synthesized one oligonucleotide,

CTGGGCCTTGCCC-TGCTTGGGGTC, which represents nucleotide 44±67 of 24p3 cDNA and another oligonucleotide, GTTGT-CAATGCATTGGTCGGTGGG, which is complemen-tary to nucleotides 599±622 of 24p3 cDNA. These two oligonucleotides were employed as the primer pair for PCR, which ampli®ed the single-stranded cDNAs of epididymis with Taq polymerase for 30 cycles: 94_{C for}

30 sec; 58_{C for 30 sec; 72}_{C for 1 min. The reaction}

mixture was subjected to electrophoresis on 2.0% agarose gel. The ampli®ed DNA (579 bp), which was extracted from the gel with a Geneclean kit (BIO101, Inc.), was sequenced according to the sequencing system using both oligonucleotide of the primer pair as the primer for PCR. Each base was determined at least three times. The random-primed DNA probe was prepared using the Prime-A-Gene kit by the method of Feinberg and Vogelstein (1983).

Electrophoresis and Western-Blot Analysis The luminal ¯uid was collected by microperfusion of epididymal tubules. Epididymes were removed from the mice and separated into three parts as caput, corpus, and cauda. Then 200 ml phosphate-buffered saline (PBS), containing 10 mM EDTA and 30 mg/ml aprotinin, was added into each part of the intact-tissue and centrifuged at 300 g for 5 min to tightly compress the tissues. The supernatant was collected as luminal ¯uid and the pellet as epididymal ¯uid-free tissue. The epididymal ¯uid-free tissue was homogenized in PBS buffer as saline-extracted protein. Epididymal tissues along with or separated from luminal ¯uid in their ducts were homogenized in PBS/10 mM EDTA in the presence of aprotinin (30 mg/ml) and centrifuged at 100,000 g for 20 min in a Beckman airfuge (Beckman, Palo Alto, CA). The concentration of proteins in the clari®ed supernatant and collected luminal ¯uid were determined by the modi®ed Lowry method (Lowry et al., 1951) and resolved by SDS-PAGE [15% (w/v) acryla-mide] on a gel slab. Proteins were transferred from the gel to a poly(vinylidene di¯uoride) membrane (Bowen et al., 1980) in PBS at 4_{C for 18 hr by diffusion. The}

transferred proteins were detected with the 1 mg/ml of 24p3 protein-induced antibody, followed by125_I-labeled

donkey anti-rabbit IgG diluted to 1:5000 and ¯uoro-graphy. Using the 24p3 protein-induced antibody diluted to 1 mg/ml as the primary antibody and goat anti-rabbit IgG conjugated with horseradish peroxi-dase diluted (Sigma, St. Levis, MO) to 1:2000 as the secondary antibody, the sperm-extraction proteins were assayed on the immunoblot. The reactive bands were visualized by enhanced chemiluminescence (ECL) (RPN2132, Amersham Pharmacia Biotech UK Limited) with exposure to X-ray ®lm.

Antibody Preparation

The 24p3 protein was puri®ed from female uterine luminal ¯uid as previously described (Chu et al., 1996). The 24p3 protein in normal saline (500 mg/ml) was mixed with an equal volume of Freund's complete adjuvant. Each New Zealand white rabbit received a

subcutaneous injection of 1 ml of the mixture. After 4 weeks and 8 weeks, the rabbit was treated in the same way with 250 mg of antigen in a mixture of normal saline and incomplete adjuvant (1:1 by volume). Two weeks after the last boost, blood was collected from the ear vein and the serum fraction was applied to a protein A-sepharose column (Phamarica, Uppsala, Sweden) for IgG isolation. The partially puri®ed antibody was used for immunodetection of 24p3 protein throughout this study. The preimmuserum for control assay was collected from the ear vein by a capillary tube before immunization.

Histology and Immunohistochemistry Tissues were ®xed in freshly prepared Bouin's solution [0.2% picric acid/2% (v/v) formaldehyde in PBS] overnight, dehydrated in ethanol, in®ltrated, and embedded in paraf®n. Each tissue section (7 mm) was mounted on a slide that had been precoated with Vetabond reagent (Burlingame, CA). Sections were dried at 45_{C, paraf®n was removed in xylene and the}

sections were rehydrated through a gradient from alcohol to distilled water. The rehydrated sections were placed in a saturated lead thiocyanate solution and heated as described by von Wasielewski and co-workers (1994). Slides were cooled at room temperature for 15 min, then rinsed in distilled water and PBS each for 15 min. The slides were immersed in a blocking solution (5% nonfat skimmed milk in PBS) in a moisture chamber at 25_{C for 1 hr and washed four times with}

PBST for 15 min each. The epididymal sections on slides were incubated with the 24p3-induced antibody diluted to 1 mg/ml in the blocking solution. Slides were gently agitated in four changes of PBST for 15 min each, then the antibody against 24p3 protein was immunodetected with alkaline phosphatase con-jugated anti-rabbit IgG diluted 1:1000 in the blocking solution. The chemical staining was con-ducted in the presence of 0.033% nitro blue tetra-zolium (NBT) and 0.0165% 5-bromo-4-chloro-3-indolyl phosphate (BCIP) in 100 mM Tris-HCl containing 100 mM NaCl and 5 mM MgCl2 at pH 9.0 for 15 min

at room temperature. After chemical staining, sections were washed with three changes of PBST for 15 min each.

Epididymal spermatozoa were smeared and air-dried on a glass slide and then immersed in methanol for 20 sec to ®x the spermatozoa on the slides. The slides were rinsed twice with PBST and then incubated with blocking solution for 1 hr before histochemical staining. To examine the 24p3 protein associated with sperma-tozoa, the spermatozoa were incubated with 24p3 protein (4.0 mM) for 60 min and then incubated with 24p3 protein-induced antibody diluted to 1 mg/ml in the blocking buffer for 60 min. After individual incubation, the slides were washed with three changes of PBST for 15 min each. The slides were then incubated with ¯uorescein-conjugated goat anti-rabbit IgG diluted to 1:400 in blocking solution for 60 min. After incubation, the slides were washed with three changes of PBST for

15 min each. Control slides were allowed to react with both the primary and secondary antibodies without 24p3 protein.

The specimens and spermatozoa on the slides were covered with 50% (v/v) glycerol in PBS and photo-graphed with a microscope equipped with epi¯uores-cence (AH3-RFCA; Olympus, Tokyo, Japan).

The 24p3 Protein Extraction From Spermatozoa

The 24p3 protein was extracted from spermatozoa as described by Rankin and co-workers (1992). Brie¯y, spermatozoa from caput epididymis (106 _{cells) were}

suspended in 0.5 ml Tyrode's without BSA containing 1  Protease Inhibitor Cocktail (PI Cocktail, Cat. No. 1697498, Roche Molecular Biochemicals, Germany) at room temperature and then centrifuged at 80 g for 4 min. The spermatozoa suspension was then divided into two equal parts. One part was further washed twice with HM containing PI Cocktail. After each washing, supernatants (low-salt wash) were collected as control samples (C1S and C2S) and the ®nal pellet was resuspended with an equal volume of double-strength Lammli electrophoresis buffer containing 0.125 M Tris-HCl (pH 6.8), 2% SDS, 20% glycerol, and 10% b-mercaptoethanol (SDS extract for control; C3S). The other part of the spermatozoa suspension was resuspended in 200 ml of HM containing 0.5 M NaCl (high-salt wash) for 30 min after low-salt buffer treatment. After high-salt washing, the suspen-sion was collected by centrifugation for 4 min at 500 g (HS). The spermatozoa extracted with the high-salt buffer were incubated in 200 ml of HM containing 0.1% Triton X-100 for 30 min on ice and then centri-fuged at 500 g for 10 min, and the supernatant was collected as extracted proteins (TS). The Triton-extracted pellet was ®nally Triton-extracted with an equal volume of double-strength Laemmli electrophoresis buffer (SDS extract; SS). All of the SDS extracts were spun for 4 min in a microcentrifuge to collect the extracted proteins for SDS-PAGE. The extracted suspensions were desalted into distilled water and concentrated to volumes of 30 ml by a Microcon-10 device (Amicon Corp., Danver, MA). Ten microliter of each protein extract was added with an equal volume of double-strength Laemmli buffer and sub-jected to SDS-PAGE and western blotting for analysis.

Spermatozoa Preparation

Tissues were minced gently in a modi®ed Tyrode's solution containing 125 mM NaCl, 2.7 mM KCl, 0.5 mM MgCl2, 0.36 mM NaH2PO4, 4.5 mM glucose, 0.09 mM

pyruvate, 25.0 mM lactate, penicillin (100 IU/ml), strep-tomycin (100 mg/ml), and 5  10ÿ 4_{% phenol red at pH}

7.4 (Fraser, 1985). The spermatozoa were extruded out of minced tissues by tearing with a pair of tweezers. The cell suspension was gently ®ltered through two layers of nylon gauze (400 mesh) and collected by centrifugation at 150 g for 10 min. The cells were

washed twice with the medium and resuspended in PBS (107_{cells/ml) before use.}

RESULTS

Sequence Analysis of Mouse Epididymal 24p3 cDNA

We ampli®ed cDNA prepared from the epididymis of mature male mice by PCR using the primer pair (Fig. 1, single underline) designed for ampli®cation of nucleo-tides 44±622 of 24p3 cDNA (see Materials and Methods section). A major DNA product ( 579 bp) was found in these nucleotide sequences identical to nucleotides 44± 622 of the uterine 24p3 cDNA sequence (Fig. 1). These data reveal that the deduced amino acid sequences of epididymal cDNA are similar to those of mouse uterine 24p3 protein, and further imply that 24p3 mRNA exists in the epididymis of mature male mice. The data strongly support that 24p3 gene expression occurs in the reproductive tract not only in female mice but also in male mice. The deduced amino acid sequence from the nucleotide sequence showed three conserved motifs from members of lipocalin superfamily (Fig. 1, double underline), which have been suggested to be involved in a ligand-binding site. As a result, it is possible for epididymal 24p3 protein to bind a hydrophobic

mole-cule in the reproductive tract. Although the function of mouse uterine 24p3 protein is yet to be determined precisely, it is found that a lipocalin in uterine ¯uid also exists in epididymis.

Regional Difference of 24p3 Gene Expression in the Epididymis

Using immunoblot analysis, 24p3 protein content intensity was found to be greatest in the caput, less intense in the corpus, and undetectable in the caudal region (Fig. 2A, lane 2±4). Similarly, 24p3 protein was present at higher levels in the luminal ¯uid of the caput and corpus compared to the caudal ¯uid (Fig. 2A, lane 5±7). The speci®c distribution of the transcripts for 24p3 protein in adult mouse epididymis was also determined by northern analysis. Comparisons of mRNA expression level of 24p3 gene were made among the caput, corpus, and caudal regions of mature mouse epididymis. Figure 2B (upper panel) shows the expres-sion of 24p3 mRNA in each part of the epididymis as assayed by northern blot analysis. The 24p3 gene was highly expressed in the caput (Fig. 2B, lane 1), declined sharply in the corpus (Fig. 2B, lane 2) and was almost undetectable in the cauda (Fig. 2B, lane 3). Expression of GAPDH mRNA (internal control) was constant in these regions (Fig. 2B lower panel). The data indicate

Fig. 1. The nucleotide sequence of mouse epididymal cDNA. The nucleotide sequence (44±622) was compiled from six independent mouse epididymal cDNA. Underline indicates the primer pairs used for PCR in the epididymal 24p3 gene study. The deduced amino acids

are shown by capital letters below the nucleotide sequence. Double underline indicates the three short motifs which are highly conserved between members of the lipocalin protein family (Flower et al., 1991).

that 24p3 gene expression in the caput region was more active than in the corpus epididymis and the caudal epididymis.

Figure 3 shows the results from testis and epididymis using light microscopy to describe the immunolocaliza-tion of 24p3 protein. No reactive staining was seen in the basal compartment of the seminiferous epithelium

or in the cytoplasmic processes between germ cells. None of the germ cells, including spermatogonia, early spermatocytes, spermatids, and spermatozoa showed immunoreactive staining for 24p3 protein (Fig. 3E). There was a complete absence of reactive staining over the entire testis and epididymis as preimmuserum was used (Fig. 3A±D). Signi®cant immunostaining was found in the epithelial cells of the caput (Fig. 3F, arrowhead E). Reactive stains were also present in luminal spermatozoa of the proximal caput (Fig. 3F, arrowhead S). In the corpus region, principal epithelial cells and luminal spermatozoa were weakly stained (Fig. 3G), and were undetectable in the caudal region (Fig. 3H). The results of northern blot analysis coin-cided with observations from western analysis and immunohistochemical study, all of which strongly suggest that the expression of 24p3 gene occurs mainly in the caput. These results clearly demonstrate that expression of the male mouse 24p3 gene and the secretion of 24p3 protein initiate in the caput region of the epididymis.

Developmental Pro®le of 24p3 Protein in Epididymis

It is well-known that the secretory proteins of the epididymis create a microenvironment for spermato-zoal maturation during the developmental period (Amann et al., 1993; Hinton and Palladino, 1995). In mice, the pubertal periodÐde®ned as the completion of spermatogenesis and the ®rst mating (Jean-Faucher et al., 1978)Ðoccurs between 35 and 40 days of age. We compared the levels of 24p3 protein and the implica-tions of these levels in the epididymis of male mice at different ages determine whether 24p3 gene expression accompanies the development of this reproductive organ. As shown in Fig. 4A, 24p3 protein synthesis was detectable in the epididymis at the age of 2 weeks (lane 2) and thereafter remained at a constant level throughout the maturation process from 3 to 7 weeks of age (lane 3±7). The levels of 24p3 mRNA in epididymis at different ages were also examined. As shown in Fig. 4B, 24p3 mRNA was detectable when mice were 2 weeks old (lane 1). The amount of mRNA in immature mouse was not signi®cantly different from that in mature mice (lane 2±6). The expression of the 24p3 gene in developing epididymis showed a very stable pattern during pre- and post-pubertal development. Immunohistochemical study of 24p3 protein was performed on tissue section from the caput region of the epididymis of 2-, 4-, 6-, and 12-week-old mice. Immunoreactive staining particles existed clearly at 2-weeks of age (Fig. 5E) and remained steadily visible until 12-weeks of age (Fig. 5F±H). No reactive staining was present in control sections (Fig. 5A±D). These results suggest that 24p3 protein may be synthesized by the translation of 24p3 mRNA at a constant level in the epithelium of the epididymis and that the sub-sequent secretion of 24p3 protein from the epithelial cells results in its accumulation in the lumen. These ®ndings imply that 24p3 gene expression is maintained

Fig. 2. Immunoblotting and northern blot analysis of mouse epididymal ¯uid and epididymal homogenate with 24p3 protein-induced antibody and 24p3 cDNA probe. (A) Protein extracts (50 mg), prepared from various populations, were separated by SDS/PAGE and processed for the immunoblotting procedure as described in Materials and Methods. Lane 1, crude uterine ¯uid from mouse uterus; lane 2, the saline extract of the epididymal caput; lane 3, the saline extract of the epididymal corpus; lane 4, the saline extract of the epididymal cauda; lane 5, the ¯uid of epididymal caput; lane 6, the ¯uid of epididymal corpus; lane 7, the ¯uid of epididymal cauda. The right-hand arrow indicates the position of 24p3 protein. (B) Northern blot analysis of epididymal RNA from adult mice. Total RNAs (50 mg), prepared from different regions of epididymal homogenate, were run on 1% agarose/formaldehyde gel and transferred to a nylon mem-brane. The membrane was probed with32_{P-labeled random-primed} DNA from a cDNA segment of either mouse 24p3 protein (upper) or GAPDH (below). The level of GAPDH mRNA served as an internal control. Lane 1, RNA from epididymal caput; lane 2, RNA from epididymal corpus; lane 3, RNA from epididymal cauda.

at a constant level in mice throughout the develop-mental period.

The 24p3 Protein Interacts With Spermatozoa The results shown in Fig. 3 imply that the 24p3 protein may associate with spermatozoa. To examine this possibility, spermatozoa from the caput epididymis were sequentially extracted as described in Materials and Methods. Figure 6 shows that 24p3 protein was present in the supernatants of the low-salt and high-salt washes (lane C1S, C2S, and HS), and that a substantial amount of 24p3 protein was present in the

Triton X-100 extracts of caput spermatozoa (lane TS) and SDS-extract for Triton-extracted spermatozoa (lane SS). These results indicate that the sperm-associated protein was ®rmly bound, requiring solubi-lization of the spermatozoal membrane by Triton X-100 (lane TS) or SDS (lane C3S) to be extracted. To determine the interaction of 24p3 protein with sperma-tozoa, spermatozoa from the caudal segment of mouse epididymis were extruded as described in Materials and Methods. The control slide had no reactive staining with the FITC-conjugated secondary antibody (Fig. 7A). When the caudal spermatozoa were preincubated

Fig. 3. Immunohistochemical localization of 24p3 protein in mouse testis and epididymal sections. Tissues were ®xed in Bouin's solution and thick sections were stained with 24p3 protein-induced antiserum at a concentration of 1 mg/ml as described in Materials and Methods. (A and E) testis, (B and F) caput of epididymis, (C and G) corpus of

epididymis, (D and H) cauda of epididymis. A±D represent the control sections reacted with preimmuserum. E±H indicate the immunor-eactive staining of 24p3 protein in tissues. Arrowhead E and arrowhead S indicate the epithelium of epididymis and luminal spermatozoa, respectively. Magni®cation  100. Bar ˆ 100 mm.

with 24p3 protein before the performance of indirect histochemical staining, a crescent ¯uorescence zone on the anterior region of the spermatozoal acrosome was visible, implying that this was the 24p3 protein-association site on the acrosomal region (Fig. 7B). Figures 7C and D show the caudal spermatozoa morphology under phase contrast microscopy. These results revealed the association site of the 24p3 protein on the spermatozoal acrosome.

DISCUSSION

Spermatozoa maturational changes occur in the epididymis. The particular importance of epididymal secreted-proteins in promoting these changes has been well-established in several types of animals (Orgelin-Crist and Jahad, 1979; Gonzalez et al., 1984) but not in

mouse. Previously, we demonstrated that 24p3 protein, an estrogen-regulated lipocalin, is expressed and secreted by epithelial cells in the uterus (Chu et al., 1996; Huang et al., 1999). In the present study, we found an mRNA sequence that is translatable into 24p3 protein expressed in mouse epididymis. This ®nding is consistent with a previous report of 24p3 gene expres-sion in male mice (Chu et al., 1996). The 24p3 protein identi®ed in the present study was found in epididymal tissue homogenate and luminal ¯uid, suggesting that the synthesis and secretion of the protein was most abundant in the caput region. Northern blot and immunohistochemical analysis demonstrated that 24p3 protein's synthesis and tissue localization pat-terns are different from other epididymal-secreted proteins (Rankin et al., 1992; Cornwall and Hann, 1995). Since spermatozoal maturation occurs early in the proximal epididymis and completes in the cauda (Amann et al., 1993; Hinton and Palladino, 1995), we hypothesized that the caput-initiated secretory 24p3 protein was a good candidate for investigation of spermatozoal maturation in the reproductive tract. Jimenez et al. (1990) and Ghyselinck and Dufaure (1990) reported the immunolocalization of a 24 kDa secretory protein in mouse epididymis that binds to the spermatozoa and shares sequence homology with glutathione peroxidase. Vernet and co-workers (1997) also presented a detailed study of the distribution of the peroxidase protein (GPX5) in mouse epididymis. Ran-kin and co-workers (1992) reported a 25 kDa secretory protein (MEP9) in the proximal and mid-caput of mouse epididymis whose antibody cross-reacted with a 25 kDa testicular antigen (MTP). Araki and co-workers (1992) characterized MTP as a member of the phospholipid-binding protein family, and suggested that MTP may have a role in lipid metabolism during spermatozoal maturation. However, according to gene sequencing and immunoreactive localization data, 24p3 protein and these other identi®ed proteins are distinct, hence this epididymal caput-synthesized and secreted protein may serve a biological function quite different from other proteins within the epididymis. Our Triton-extracted study indicated an association of 24p3 protein with spermatozoa, and immunocytochemical studies demonstrated that 24p3 protein is associated with the anterior acrosomal membrane of epididymal spermato-zoa. We speculate that 24p3 protein existing in uterine luminal ¯uid (Chu et al., 1996) may associate with spermatozoa in the female genital tract; i.e., that spermatozoal processing occurs during epididymal transit and/or that physiological changes occur in sperm within the female reproductive tract. Further studies are necessary to determine whether the 24p3 protein affects spermatozoal activity.

Sexual hormones appear to be essential for main-taining the synthesis and secretion of reproductive proteins (Ghyselinck et al., 1989; Lefrancois et al., 1993), whereas some secretory proteins are hormone-independent (Defelein et al., 1996). In the present study, we found that 24p3 protein is expressed in

Fig. 4. Ontogeny of 24p3 protein synthesized by mouse epididymis. (A) Proteins (50 mg) of epididymal homogenate were resolved by 15% SDS-PAGE and then immunodetected by western blot procedures with antibody to mouse 24p3 protein and followed by125_I-labeled anti-rabbit IgG detection. Lane 1, positive control of crude uterine ¯uid from female mouse; lanes 2±7, the epididymal protein extract from 2-, 3-, 4-, 5-, 6- and 7-week-old male mice. (B) Northern blot analysis of mouse total RNA. Total RNA (50 mg) prepared from epididymal homogenate was run on a 1%-agarose-formaldehyde gel, transferred to a nylon membrane and probed with32_{P-labeled random-primed} DNA from a cDNA segment of either mouse 24p3 (upper) or GAPDH (below). The level of GAPDH mRNA served as an internal control. Lane 1 to lane 6 indicate the level of 24p3 mRNA from 2±7 week-old mice.

mouse epididymis from the prepubertal period to adulthood. Our immunohistochemical study showed that 24p3 protein appears in 14-day-old mice prior to differentiation of the epididymis, which is completed at around day 20 (Abou-Haila and Fain-Maurel, 1985) while the androgen content is still low (Jean-Faucher et al., 1985). Indeed, in the mouse epididymis, the content of androgen is age-dependent. Androgen increases progressively during development, and reaches its highest level at 40 days of age and then remains relatively constant until 90 days of age (Jean-Faucher et al., 1985). In the present study, the 24p3

protein and mRNA levels in mouse epididymis were constant from the age of 2±12 weeks. There was no signi®cant correlation between androgen content and the level of 24p3 gene expression in the epididymis during development. Based on these results, 24p3 protein might be considered a constitutive gene product in the luminal environment for spermatozoa from rete testis in male mouse. One way to investigate the in¯uence of sexual hormones on the production of 24p3 protein in vivo would be to examine the effect of injection of short-term sexual hormones into castrated male mice.

Fig. 5. Epididymal localization of 24p3 protein during postnatal development. Epididymal tissues of mice of different ages were ®xed in Bouin's solution and stained with antibody to 24p3 protein at a dilution of 1 mg/ml as described in Materials and Methods. A±D represents the control sections of 2-, 4-, 6-, and 12-week-old mouse

epididymis. E±H shows reactive staining of 2-, 4-, 6-, and 12-week-old mouse epididymis. The arrowhead E indicates the location of 24p3 protein in the epithelium of epididymis. Magni®cation  100. Bar ˆ 100 mm.

Fig. 6. Sequential extraction of caput spermatozoa. Proteins were extracted as described in Materials and Methods, then resolved by SDS-PAGE, transferred to nitrocellulose membrane and probed with 24p3 protein-induced antibody. Lane LF, caput luminal ¯uid; lane C1S, the ®rst extract with low-salt buffer; lane C2S, the second extract

with low-salt buffer; C3S, SDS extracted protein of spermatozoa with twice low-salt wash; HS, 24p3 protein in high-salt extract; TS, the Triton X-100 extractable proteins; SS, SDS extract for Triton-extracted spermatozoa. The arrow indicates the 24p3 protein.

Fig. 7. Indirect immuno¯uorescence demonstration of the 24p3-protein binding zone on epididymal spermatozoa. Fresh epididymal spermatozoa were smeared and ®xed by methanol on glass slides. (A) The slide was incubated with 1 mg/ml 24p3 protein-induced antibody and followed with ¯uorescein-conjugated goat anti-rabbit IgG (1:400 dilution) as control. (B) The fresh cells were incubated in blocking buffer in the presence of 40 mM 24p3 protein for 1 hr and allowed to

react with 24p3 protein-induced antibody in the blocking buffer (1 mg/ ml) for 1 hr, followed by ¯uorescein-conjugated goat anti-rabbit IgG. A ¯uorescence microscope was used to observe the slides. (C) and (D) are as observed under phase contrast microscope under the same conditions as (A) and (B), respectively. The arrows indicate sperma-tozoal acrosome. Magni®cation  1000. Bar ˆ 10 mm.

The presence of high levels of epididymal retinoic acid±binding protein in the epithelial cell and lumen of the epididymis was thought to indicate that this protein was a retinoid transporter and hence a factor necessary for spermatozoal maturation (Sundaram et al., 1998; Lareyre et al., 1998). A sperm coating lizard epididymal secretory protein (LESP), also a lipocalin of epididymal secretory protein, was suggested to be a hydrophobic molecule transporter during spermatozoal maturation (Laurent et al., 1993). The de®nitive roles of retinoic acid and hydrophobic molecules in spermatozoal maturation are not yet clear, but they are thought to be important. We previously described 24p3 protein as a lipocalin in uterus, and that the protein showed an ability to bind small hydrophobic ligands such as fatty acids, cholesteryl ester, and retinoids (Chu et al., 1998). It seems reasonable to speculate that the association of 24p3 protein on spermatozoal acrosome may serve as a hydrophobic molecule transfer protein in the epididymal lumen and be involved in spermatozoal maturation.

Although the exact function of mouse 24p3 protein remains unknown, the observations that this protein is synthesized early in epididymal caput, associates with spermatozoa, and is secreted from uterus suggest that 24p3 protein plays an important role both in sperma-tozoal processing and fertilization. Such a possibility warrants further investigation.

REFERENCES

Abou-Haila A, Fain-Maurel MA. 1985. Postnatal differentiation of the enzymatic activity of the mouse epididymis. Int J Androl 8: 441±458.

Amann RP, Hammerstedt RH, Veeramachaneni DNR. 1993. The epididymis and sperm maturation: a perspective. Reprod Fertil Dev 5:361±381.

Araki Y, Vierula ME, Rankin TL, Tulsiani DRP, Orbebin-Crist MC. 1992. Isolation and characterization of a 25-kilodalton protein from mouse testis: sequence homology with a phospholipid-binding protein. Biol Reprod 47:832±843.

Bendahmane M, Abou-Haila A. 1994. Synthesis, characterization and hormonal regulation of epididymal proteins during postnatal development of mouse. Differentiation 55:119±125.

Bowen B, Steinberg J, Laemmli VK, Weintraub H. 1980. The detection of DNA-binding proteins by protein blotting. Nucleic Acid Res 8: 1±20.

Brooks DE. 1981. Secretion of proteins and glycoproteins by the rat epididymis: regional differences, androgen-dependence and effects of protease inhibitors, procaine and tunicamycin. Biol Reprod 25:1099±1114.

Brooks DE. 1987. Developmental expression and adrogenic regulation of the mRNA for major secretory proteins of the rat epididymis. Mol Cell Endocrinol 53:59±66.

Charest NJ, Petrusz P, Ordronneau P, Joseph DR, Wilson EM, French FS. 1989. Developmental expression of an androgen regulated epididymal protein. Endocrinol 125:942±947.

Chu ST, Huang HL, Chen JM, Chen YH. 1996. Demonstration of a glycoprotein derived from the 24p3 gene in mouse uterine luminal ¯uid. Biochem J 316:545±550.

Chu ST, Lin HJ, Chen YH. 1997. Complex formation between a formyl peptide and 24p3 protein with a blocked N-terminus of pyrogluta-mate. J Peptide Res 49:582±585.

Chu ST, Lin HJ, Huang HL, Chen YH. 1998. The hydrophobic pocket of 24p3 protein from mouse uterine luminal ¯uid: fatty acid and retinol binding activity and predicted structural similarity to lipocalins. J Peptide Res 52:390±397.

Cornwall GA, Hann SR. 1995. Specialized gene expression in the epididymis. J Androl 16:379±383.

Davis TR, Tabatabai L, Bruns K, Hamilton RT, Nilsen-Hamilton M. 1991. Basic ®broblast growth factor induces 3T3 ®broblasts to synthesize and secrete a cyclophilin-like protein and b2 -microglo-bulin. Biochim Biophys Acta 1095:145±152.

Defelein M, Grapey D, Schaeffer T, Chin-chance C, Bushman W. 1996. PAX-2: A developmental gene constitutively expressed in the mouse epididymis and ducts defferens. J Urol 156:1204±1207.

Devine KM, Carroll J. 1985. Peptide synthesis by the murine epididymis. Biochem Soc Trans 13:506±507.

Feinberg AP, Vogelstein B. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high speci®c activity. Anal Biochem 132:6±13.

Flower DR, North ACT, Attwood TK. 1991. Mouse oncogene protein 24p3 is a member of the lipocalin protein family. Biochim Biophys Res Commun 180:69±74.

Fraser LR. 1985. Albumin is required to support the acrosome reaction but not capacitation in mouse spermatozoa in vitro. J Reprod Fertil 74:185±196.

Garberi JC, Kohane AC, Cameo MS, Blaquier JA. 1979. Isolation and characterization of speci®c rat epididymal proteins. Mol Cell Endocrinol 13:73±82.

Ghyselinck NB, Dufaure JP. 1990. A mouse cDNA sequence for epididymal androgen-regulated proteins related to glutathione peroxidase. Nucleic Acid Res 18:7144±7149.

Ghyselinck NB, Jimenez C, Courty Y, Dufaure JP. 1989. Androgen-dependent messenger RNA(s) related to secretory proteins in the mouse epididymis. J Reprod Fert 85:631±639.

Gonzalez EF, Cuasnicu PS, Piazza A, Pineiro L, Blaquier JA. 1984. Addition of an androgen-free epididymal protein extract increases the ability of immature hamster spermatozoa to fertilize in vivo and in vitro. J Reprod Fertil 71:433±437.

Hinton BT, Palladino MA. 1995. Epididymal epithelium: its contribu-tion to the formacontribu-tion of a luminal ¯uid microenvironment. Microscopy Res Technique 30:67±81.

Hraba-Renevey S, Turler H, Kress M, Salomon C, Weil R. 1989. SV40-induced expression of mouse gene 24p3 involves a posttranscrip-tional mechanism. Oncogene 4:601±608.

Huang HL, Chu ST, Chen YH. 1999. Ovarian steroids regulate 24p3 expression in mouse uterus during the natural estrous cycle and the preimplantation period. J Endocrinol 162:11±19.

Jean-Faucher C, Berger M, de Turckheim M, Veyssiere G, Jean CL. 1978. Developmental patterns of plasma and testicular testosterone in mice from birth to adulthood. Acta Endocrinol 89:780±788. Jean-Faucher C, Berger M, de Turckheim M, Veyssiere G, Jean CL.

1985. Testosterone and dihydrotestosterone levels in the epididymis vas deferens and preputal gland of mice during sexual maturation. Int J Androl 8:44±57.

Jimenez C, Ghyselinck NB, Depeiges A, Dufaure JP. 1990. Immuno-chemical localization and association with spermatozoa of andro-gen-regulated proteins of Mr 24,000 secreted by the mouse epididymis. Biol Cell 68:171±174.

Lareyre JJ, Zheng WL, Zhao GQ, Kasper S, Newcomer ME, Matusik RJ, Ong DE, Orgebin-Crist MC. 1998. Molecular cloning and hormonal regulation of a murine epididymal retinoic acid-binding protein messenger ribonucleic acid. Endocrinol 139:2971±2981. Laurent M, Dufaure JP, Depeiges A. 1993. LESP, an

androgen-regulated Lizard epididymal secretory protein family identi®ed as a new member of the lipocalin superfamily. J Biol Chem 268: 10274±10281.

Lefrancois AM, Jimenez C, Dufaure JP. 1993. Developmental expression and androgen regulation of 24 kDa secretory proteins by the murine epididymis. Int J Androl 16:147±154.

Liu Q, Nilson-Hamilton M. 1995. Identi®cation of a new acute phase protein. J Biol Chem 270:22565±22570.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265±275.

Meheus LA, Fransen LM, Raymackers JG, Blockx HA, van Beeumen JJ, van Ban SM, van de Voorde A. 1993. Identi®cation by microsequencing of lipopolysaccharide-induced proteins secreted by mouse macrophage. J Immunol 151:1535±1547.

Orgebin-Crist MC, Fournier-Delpech S. 1982. Evidence for maturational changes during epididymal transit. J Androl 3: 429±433.

Orgebin-Crist MC, Jahad N. 1979. The maturation of rabbit epididymal spermatozoa in organ culture: stimulation by epididy-mal cytoplasmic extracts. Biol Reprod 21:511±515.

Rankin TL, Tsuruta KJ, Holland MK, Griswold MD, Orgebin-Crist M. 1992. Isolation, immunolocalization, and sperm-association of three proteins of 18, 25 and 29 kilodaltons by the mouse epididymis. Biol Reprod 46:747±766.

Sambrook J, Fritsch EF, Maniatis T. 1989. In: Fritsch EF, Maniatis T, editors. Molecular cloning: A laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, p7.19±7.50. Sundaram M, Sivaprasadarao A, van Aalten DMF, Findlay JBC.

1998. Expression, characterization and engineered speci®city of rat epididymal retinoic acid-binding protein. Biochem J 334: 155±160.

Thomas PS. 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Nat Acad Sci USA 77:5201±5205.

Toney TW, Danzo BJ. 1989. Estrogen and androgen regulation of protein synthesis by the immature rabbit epididymis. Endocrinol 25:231±242.

Ueda H, Hirano T, Fujimoto S. 1990. Changes in protein secretory patterns during the development of the rat epididymis. Zool Sci 25:681±690.

Vernet P, Faure J, Dufaure JP, Drevet JR. 1997. Tissue and developmental Distribution, dependence upon testicular factors and attachment to spermatozoa of GPX5, a murine epididymis-speci®c glutathione peroxidase. Mol Reprod Dev 47:87±98. von Wasielewski R, Werner M, Nolte M, Wilkens L, Georgii A. 1994.

Effects of antigen retrieval by microwave heating in formalin-®xed tissue sections on a broad panel of antibodies. Histochemistry 102:165±172.