Characterization of Two Distinct Populations of Epididymosomes ...

BIOLOGY OF REPRODUCTION 83, 473–480 (2010) Published online before print 10 June 2010. DOI 10.1095/biolreprod.109.082438

Characterization of Two Distinct Populations of Epididymosomes Collected in the Intraluminal Compartment of the Bovine Cauda Epididymis1 Gilles Frenette, Julie Girouard, Olivier D’Amours, Nancy Allard, Laurence Tessier, and Robert Sullivan2 Centre de Recherche en Biologie de la Reproduction and De´partement d’Obste´trique-Gyne´cologie, Faculte´ de Me´decine, Universite´ Laval, Quebec City, Quebec, Canada INTRODUCTION

During their transit along the epididymis, mammalian spermatozoa acquire new proteins that are necessary for their acquisition of forward motility and fertility. By using the bovine model, we previously showed that small membranous vesicles named epididymosomes are secreted in the epididymal intraluminal compartment. Epididymosomes from caput and cauda are different, and interact sequentially with the transiting spermatozoa. In fact, selected proteins of epididymosomes are transferred to different subcompartments of the maturing spermatozoa. In this study, we investigate the possibility that different subpopulations of epididymosomes are present in the caudal portion of the epididymis. Through the use of discontinuous sucrose gradient ultracentrifugation, we isolated two distinct populations that differ in their protein and lipid compositions. Although they have similar diameters, the ultrastructural appearance of these two populations was very different. The low-density (Ld) vesicles are enriched in cholesterol, sphingomyelin, and ganglioside M1, suggesting the existence of detergent-resistant membrane domains or rafts. The high-density (Hd) vesicles show a high protein concentration, including ACTB and VAMP8. When each subpopulation of biotinylated cauda epididymosomes was coincubated with caput spermatozoa, a subset of biotinylated proteins was transferred to the sperm; the Ld and Hd vesicles transferring the same pattern of proteins. In vitro competition assays of protein transferred from Ld or Hd epididymosomes to sperm confirm the similarity in the selected transferred proteins. Electrospray tandem mass spectrometry (ES-MS/MS) analysis of proteins associated with the two populations of vesicles confirm the epididymal origin of some of them, the possible involvement of others in transmembrane signaling systems, and the identification of proteins for which functions in sperm physiology remain to be determined. Mass spectrometry analysis also revealed that ELSPBP1 and GBB2 were transferred from epididymosomes to spermatozoa. Results are discussed with regard to the functions of these two cauda epididymosome populations in sperm physiology.

In mammals, spermatozoa have to transit along the epididymis in order to acquire their full fertilizing potential. This process involves complex interactions between the intraluminal milieu and spermatozoa that occur in a sequential manner along the excurrent duct. Some epididymis-secreted proteins are added to spermatozoa, and have been shown to be involved in the acquisition of fertilizing ability and forward motility [1, 2]. These newly acquired proteins are classically referred to as coating proteins [2]. However, some sperm proteins acquired during the epididymal transit behave as integral membrane proteins, are glycosylphosphatidylinositol (GPI) anchored to the sperm plasma membrane [3–5], or can be integrated in intracellular subcompartment [6–8]. The mechanisms of acquisition of these proteins by the maturing spermatozoa remain puzzling. Within the epididymal intraluminal compartment, spermatozoa interact with small membranous vesicles called epididymosomes [9–11] that are thought to be secreted by principal cells using the unclassical apocrine pathway [12, 13]. A complex mixture of proteins is associated with these vesicles that differs from the soluble proteins found in the epididymal fluid used to prepare the epididymosomes, at least in bovine [14] and human [15]. The proteins in epididymosomes are segregated in detergent-soluble and detergent-resistant membrane domains that are involved in the compartmentalization of transferred proteins to different sperm subcellular and membranous domains [16, 17]. The protein transfer from epididymosomes to maturing spermatozoa shows certain specificity, is saturable, and is temperature, pH, and zinc dependent [18]. These transferred proteins have been shown to be involved in zona pellucida binding [3, 19], protection against reactive oxygen species [20], and involved in modulation of motility [21] (reviewed in [10, 22, 23]). We recently reported that epididymosomes collected in the caput and cauda bovine epididymis have different protein compositions, and that they interact differently with spermatozoa [14]. In the present study, we report that epididymosomes collected in the cauda epididymal fluid can be separated into two distinct populations differing in their biochemical composition, suggesting that they may play different roles.

epididymis, epididymosome, gamete biology, male reproductive tract, sperm maturation 1 Supported by Natural Sciences and Engineering Research Council of Canada (NSERC) grant to R.S. J.G. and O.D. are supported by a Ph.D. scholarship from the Canadian Institutes of Health Research and NSERC-Fonds de recherche sur la nature et les technologies, respectively. 2 Correspondence: Centre de Recherche du Centre Hospitalier Universitaire de Que´bec (CHUL), 2705 boulevard Laurier, room T1-49, Quebec City, Quebec G1V 4G2, Canada. FAX: 418 654 2765; e-mail: [email protected]

MATERIALS AND METHODS Biological Material Bovine tissues were obtained from a slaughterhouse. Immediately after slaughter, the testis were put on ice and brought to the laboratory within 2 h. Epididymides were dissected on arrival. Only epididymides with motile spermatozoa in the tubule of the distal cauda portion were used in this study. Fluids were collected as previously described [14].

Received: 11 November 2009. First decision: 15 December 2009. Accepted: 19 May 2010. Ó 2010 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

Epididymosomes Preparation Epididymosomes were prepared from cauda epididymal fluid as previously described [18]. Briefly, cauda epididymal fluid was collected by applying

473

Downloaded from www.biolreprod.org.

ABSTRACT

474

FRENETTE ET AL.

retrograde air pressure into the vas deferens cannulated with a syringe. The fluid was diluted in 0.15 M NaCl and centrifuged at 1000 3 g to remove spermatozoa and twice at 3600 3 g for 20 min to remove any cellular debris. The 3600 3 g supernatant was subjected to ultracentrifugation at 120 000 3 g in a swinging bucket rotor (MLS-50; Beckman) at 48C for 2 h. The pellet containing epididymosomes was resuspended in NaCl and centrifuged again at 120 000 3 g for 2 h. Aliquots of epididymosomes were kept at 808C until use. The freezing-thawing cycle does not affect the protein transfer properties of epididymosome preparations (Supplemental Data available online at www. biolreprod.org). The freshly prepared pellet of cauda epididymal epididymosomes was resuspended in PBS and mixed with 1 volume of 85% (w/v) sucrose in PBS; 0.85 ml of this suspension was overlaid with 3 ml of 35% (w/v) and 1 ml of 5% (w/v) sucrose solutions in PBS. The discontinuous gradient was ultracentrifuged at 120 000 3 g for 16 h at 48C. Ten fractions of 480uL collected from the top of the gradient were analyzed. Two epididymosome populations distributed themselves along the discontinuous sucrose gradient. The low-density (Ld) population was recovered at the 5%/35% interface of the sucrose gradient and the high-density (Hd) vesicles were in the 42.5% sucrose cushion at the bottom of the tube. The two fractions containing Ld and Hd epididymosomes were diluted with 0.15M NaCl and centrifuged at 120 000 3 g for 2 h. The resulting pellets were washed once with NaCl solution, the resulting pellets were resuspended in small volume of NaCl 0.15M, and stored at 808C until use.

Epididymosomes Biotinylation and Competition Assays

Protein Precipitation, Electrophoresis, and Immunoblotting Proteins from epididymosome suspensions were precipitated in MeOH/ CHCl3 [24] and analyzed on SDS-PAGE according to Laemmli [25]. Electrophoreses were run on the basis of 10 lg of proteins per lane, as determined by Bio-Rad protein assay kit. The gels were stained with the Coomassie brilliant blue R-250 (Bio-Rad) or transferred to a nitrocellulose membrane (Bio-Rad) with a semidry graphite blotter system (Pharmacia). Nitrocellulose membranes were blocked overnight at 48C in PBS containing 5% (w/v) skim milk or 3% (w/v) bovine serum albumin. The following antibodies were used to probe the nitrocellulose membrane: anti-P25B [26]; anti-AKR1B1 (gift from Dr. M.A. Fortier); anti-MIF (gift from Dr. M. Nishibori); anti-NPC2 (gift from Dr. J.L. Gatti); anti-PGAM2 (gift from Dr. Y. Matuo and K. Uchida, Oriental Yeast Co.); anti-AK1 (Santa Cruz Biotechnology); anti-phosphatidylethanolamine binding protein (PEBP) 1 (Santa Cruz Biotechnology); anti-VAMP8 (Sigma-Aldrich); and anti-actin B (Sigma-Aldrich). The immune complexes were revealed with the corresponding secondary antibody coupled with HRP (Jackson Laboratory) and enhanced chemiluminescence (ECL) as a peroxidase substrate (Amersham).

Two-Dimensional Gel Electrophoresis and Electrospray Tandem Mass Spectrometry Identification of Proteins Following coincubations with biotinylated Ld and Hd epididymosomes, proteins of caput spermatozoa were extracted with 0.3% Triton X-100 in water and submitted to two-dimensional (2D) gel electrophoresis. Briefly, proteins were precipitated with 9 volumes of acetone at 208C. The pellets were dissolved in solution D (7 M urea, 2 M thiourea, 2% CHAPS, 2% Pharmalyte [v/v], and 40 mM DTT) and separated on a 13-cm linear Immobilin Dry Strip

Ganglioside M1 Detection Cholera toxin subunit b (CTB) is known to bind ganglioside M1 (GM1), a marker of detergent-resistant membrane domains, or rafts, present in epididymosomes [17]. Ld and Hd epididymosomes equivalent to 15 lg of protein were slot blotted in duplicates on polyvinylidene fluoride membranes. Membranes were blocked overnight at 48C in PBS containing 3% dry skim milk, washed in PBS, and incubated with CTB-HRP conjugate. The ECL was used as substrate, and the densitometry quantification of GM1 was performed with an image analyzer system.

Lipid Extraction and High-Performance Thin-Layer Chromatography Ld and Hd epididymosomes were extracted by the Folch method [27]. Briefly, equivalents of 50–150 lg protein of Ld and Hd epididymosomes suspended in 500 ll NaCl solution were mixed successively with 1 ml methanol and 2 ml chloroform. The organic and aqueous phases were separated by centrifugation at 500 3 g for 10 min. The organic phase was collected, and the upper aqueous phase was re-extracted once more with chloroform:methanol (2:1 [v/v]) and centrifuged at 500 3 g for 10 min at 48C. The two collected organic phases were pooled and evaporated under a nitrogen stream. The dried lipid pellets were dissolved in a small volume of chloroform:methanol (2:1 [v/ v]). The different classes of extracted phospholipids were separated by highperformance thin-layer chromatography (HPTLC; Merck) [9, 28] with a mixture of methyl acetate:propan-2-ol:chloroform:methanol:0.25% (w/v) aqueous KCl (25:25:25:10:9 [v/v]). A mixture of phospholipids and cholesterol was migrated in parallel to allow identification. The phospholipid spots were visualized by treating the HPTLC plates with a solution containing 10% (w/v) CuSO4 and 8% (v/v) H3PO4, and by heating at 1808C. Densitometric analyses of the phospholipid classes were carried out with an image analyzer system (Alpha Innotech Corporation, San Leandro, CA). Three samples of Ld and Hd epididymosomes were independently analyzed, and each analysis was performed in duplicate. Comparisons of lipid composition of epididymosome populations were based on a constant concentration of protein.

Electron Microscopy The two populations of epididymosomes were fixed for 16 h in PBS containing 2.5% glutaraldehyde. The fixed particles (Ld and Hd) were then


Freshly prepared Ld and Hd cauda epididymosomes were resuspended in 200 ll PBS and mixed with 1 mg of Sulfo-NHS-LC-biotin (Pierce). After 105 min at room temperature, the vesicle suspensions were diluted with 20 volumes of 0.15 M NaCl and ultracentrifuged at 120 000 3 g for 2 h at 48C. The pellets were washed by ultracentrifugation, resuspended in NaCl, and aliquots were stored at 808C. The in vitro assay of protein transfer from epididymosomes to spermatozoa was performed according to Frenette et al. [14]. Freshly collected caput spermatozoa were washed several times with NaCl and incubated at a concentration of 30 3 106/150 ll 0.15 M NaCl, 10 mM MES-PIPES (pH 6.5), together with suspensions of biotinylated epididymosomes, for 3 h at 378C [18]. In competition assays, a 40–60 time excess of unbiotinylated Ld or Hd epididymosomes was added to the protein transfer assay. The final concentrations of epididymosomes in the coincubation media were similar to the epididymal intraluminal concentration of the vesicles (740 6 340 lg/ml). Biotinylated proteins transferred from epididymosomes to spermatozoa were detected by probing Western blots of sperm protein 1% Triton X100 extracts (equivalent of 7 3 106 spermatozoa per lane) with neutravidin-horseradish peroxidase conjugate (NA-HRP).

(pH 3–10) with the IPGphor II Isoelectric Focusing System (GE Healthcare, Buckinghamshire, U.K.). For the second dimension (SDS-PAGE), the strips were equilibrated in buffer E (6 M urea, 75 mM Tris [pH 8.8], 25% glycerol [w/ v], 2% SDS) containing 65 mM DTT, followed by incubation in buffer E containing 135 mM iodoacetamide, respectively. The second dimension was run on 12% polyacrylamide gels, and transferred proteins were revealed as described above. Total protein extracts of biotinylated caput and cauda epididymosomes were run in parallel. These electrophoretic patterns were revealed to visualize biotinylated proteins. Considering that the 2D electrophoretic patterns of biotinylated proteins transferred by the Ld and Hd epididymosomes are identical, only proteins transferred from Hd epididymosomes to caput spermatozoa were further studied. Coomassie blue-stained proteins of the epididymosomes electrophoretic patterns corresponding to the biotinylated proteins transferred to sperm were excised from the gels and digested with trypsin for electrospray tandem mass spectrometry (ES-MS/MS) peptide identification as previously described [15]. Mass spectrometry was performed by the Proteomics platform of the Eastern Quebec Genomic Center (Quebec City, QC, Canada). Peptides were separated by online reversed-phase nanoscale capillary liquid chromatography (LC) and analyzed by (ES-MS/MS). Mascot (Matrix Science, London, U.K.) version 2.2.0 was used to search the bovine Uniref 100_14_7_bos_taurus_9913 database, assuming peptides were generated by trypsin digestion. MS/MS-based peptide and protein identifications were validated with Scaffold (version scaffold_2_04_06; Proteome Software Inc., Portland, OR). Protein identification was considered significant if it contained at least three unique identified peptides with P . 95%. Protein probabilities were assigned by Protein Porphet algorithm. The MS/MS peptides databank was revisited to identify biotinylated peptides of epididymosomes potentially transferred to spermatozoa. The biotinylated derivative (NHS-LC) was specified as the variable modification of lysine ( þ339.5 Da). Scaffold (version Scaffold-2_02) was used to validate MS/MS-based peptide and protein identification. Peptide identifications were accepted if they could be established at greater than 95.0% probability, as specified by the Peptide Prophet algorithm. The biotinylated spectra were manually validated.

TWO POPULATIONS OF EPIDIDYMOSOMES

475

FIG. 1. Electrophoretic and Western blot analyses of cauda epididymal epididymosomes separated along a discontinuous sucrose gradient. A) Coomassie blue-stained electrophoretic patterns of proteins of cauda epididymosomes distributed along a discontinuous sucrose gradient. B) Western blot analyses of P25B, PEBP1, AKR1B1, AK1, and MIF in each gradient fraction. Fractions were collected from the upper gradient and are numbered from 1 to 10. Molecular mass standards are indicated on the left of the figure.

washed with 0.1 M cacodylate buffer (Na(CH3)2AsO23H2O, pH 7.3) and rinsed with water. The specimens were placed on Ni/formvar grids and stained with phosphotungstic acid, washed with water, and dried. Analyses were performed on a JEM-1230 transmission electron microscope at 80 kV (JEOL, Montreal, QC, Canada). Images were captured at 120 0003 magnification. This electron microscopy method allows visualization of smears of intact epididymosomes, thus allowing estimation of the diameter of these membranous vesicles.

RESULTS Preliminary results of epididymosome ultracentrifugation on different sucrose gradients indicated that two populations of membranous vesicles can be separated according to their density (data not shown). Based on these observations, ultracentrifugation was performed on a three-step sucrose gradient. In these conditions, vesicles distributed themselves at the 5%/35% sucrose interface and in the denser fraction of the gradient. Coomassie blue-stained gels of each gradient fraction indicated that protein-containing vesicles were present in fractions 3 and 4, and in the dense fractions 8–10. Intensity of protein staining was much greater in fractions 8–10 when compared with the fractions collected at the 5%/35% interface. Protein electrophoretic patterns of fractions 3–4 and 8–10 suggests that these two portions of the sucrose gradient

contained homogeneous populations of epididymosomes (Fig. 1A). Western blots were performed on each sucrose gradient fraction to detect five proteins known to be associated with cauda epididymal epididymosomes: P25B, PEBP1, AKR1B1, AK1, and MIF. These proteins were immunodetected in all sucrose fractions in which proteins were detected by Coomassie blue staining of SDS-PAGE. The relative intensity between the fractions was variable from one protein to the other. Whereas the quantity of protein associated with Ld fractions was lower than that associated with Hd vesicles, P25B was immunodetected at a similar intensity in fractions 3– 4 and 8–10. AK1 signal was more intense in Ld fractions. PEBP1, AKR1B1, and MIF were more intense in 8–10, with levels of detection proportional to the quantity of protein detected in Coomassie blue-stained electrophoretic patterns of each gradient fraction (Fig. 1B). Based on the similarity of the electrophoretic patterns of the gradient fractions, and on the Western blots results, fractions 3–4 and 8–10 were pooled and considered as the Ld and Hd populations of cauda epididymosomes, respectively. According to the sucrose density cushion, the relative densities of the Ld and Hd epididymosomes were 1.02–1.15 and .1.15, respectively. When the two populations of epididymosomes were submitted to SDS-PAGE analysis on the basis of similar protein concentrations, electrophoretic patterns of Ld and Hd epididymosomes were very different. In Hd vesicles, two intense bands at .100 and 45 kDa were detected, whereas, in the Ld population, the major protein band was at 32 kDa (Fig. 2A). We have also been able to separate two populations of vesicles with the caput epididymal fluid. In contrast to the


FIG. 2. A) Coomassie blue-stained electrophoretic patterns of Ld and Hd cauda epididymosomes. Molecular mass standards are indicated on the left of the figure. B) Western blot analyses of different preparations of Ld (Ld1–Ld3) and Hd (Hd1–Hd4) epididymosomes. Western blots were probed with antibodies directed against AKR1B1, PEBP1, PGAM2, NPC2, P25B, AK1, ACTB, VAMP8, and MIF. ND, not determined.

476

FRENETTE ET AL.

cauda epididymosomes, the electrophoretic patterns of the caput vesicles differing in density were similar, but different from the two cauda epididymal epididymosomes populations (data not shown). The cauda epididymosome proteome contains more than 400 different proteins (data not shown). We have investigated the presence of some of these proteins in the Ld and Hd


FIG. 3. A) Western blot analyses of biotinylated proteins of Ld and Hd epididymosomes collected in the cauda epididymal fluid at two different concentrations. B) Western blot detection in duplicate of biotinylated proteins from Ld and Hd epididymosomes from the cauda epididymidis transferred to unbiotinylated caput spermatozoa. C and D) Western blot of 2D electrophoresis of biotinylated proteins from Ld (C) and Hd (D) cauda epididymosomes transferred to caput spermatozoa. E) Coincubation of caput spermatozoa with Ld or Hd cauda epididymosomes generates similar results as illustrated by the merged electrophoretic patterns illustrated in (C) and (D). Numbers in (E) correspond to protein spots identified by ES-MS/MS (Table 1). Molecular mass standards are indicated on the left of the figure.

populations of epididymosomes collected in the cauda epididymidis. Western blots were performed on different preparations of Ld (Ld1–Ld3) and Hd (Hd1–Hd4) epididymosomes on the basis of equivalent concentrations of protein. Some proteins, such as PEBP1 and AKR1B1, were associated at similar concentrations with the two populations of epididymosomes. Some other proteins, PGAM2, NPC2, P25B, and AK1, were immunodetectable at higher concentrations in Ld epididymosomes, whereas actin beta (ACTB), VAMP8, and MIF were associated in higher concentrations with the dense epididymosomes collected in the cauda segment (Fig. 2B). As for Coomassie blue-stained gels, biotinylated surfaceexposed proteins of the two populations of cauda epididymosomes showed different electrophoretic patterns (Fig. 3A). We have shown previously that epididymosomes transfer selected proteins to epididymal spermatozoa [18]. When coincubated in vitro with caput spermatozoa, Ld and Hd cauda epididymosomes transferred a similar electrophoretic pattern of biotinylated proteins, even though their total protein composition differed (Fig. 3, B–E). When transferred proteins were visualized by probing incubated spermatozoa with peroxidase-coupled avidin, 77%–85% of spermatozoa were labeled regardless of the subpopulation (Ld and Hd) of cauda epididymosomes with which they were coincubated. In both cases, the pattern of labeling was similar to that already reported for proteins transferred from unfractionated cauda epididymosomes to caput sperm [18]. In a competition assay, when unbiotinylated epididymosomes were added in excess during in vitro coincubation of biotinylated epididymosomes with caput spermatozoa, the amount of biotinylated proteins transferred to spermatozoa decreased. Interestingly, both Ld and Hd unbiotinylated epididymosomes competed in a similar way when caput spermatozoa were coincubated with one or the other population of biotinylated epididymosomes (Fig. 4). When submitted to 2D gel electrophoresis, five major spots of biotinylated proteins transferred from epididymosomes to spermatozoa were visualized (Fig. 3, C–E). The Coomassie blue-stained protein spots of epididymosomes corresponding to these biotinylated proteins transferred to caput spermatozoa were analyzed by mass spectrometry. Table 1 lists the ES-MS/ MS identification of these proteins of cauda epididymosomes. A total of 14 different proteins were identified. For each protein, 3–25 different peptides were identified covering 8%– 85% of the amino acid sequence of the bovine proteins. For nine of the identified proteins, peptides were found in one excised Coomassie blue-stained protein spot. For the other five proteins, their corresponding peptides were present in more than one protein spot of the 2D electrophoretic pattern of epididymosomes. The proteins of the epididymosomes identified by ES-MS/MS were not necessarily transferred to spermatozoa in the coincubation in vitro experiments. The ES-MS/MS data bank was revisited to identify biotinylated proteins of epididymosomes corresponding to the one transferred to spermatozoa. Out of the 14 proteins, only epididymal sperm binding protein 1 (ELSPBP1) and guanine nucleotide-binding protein G(1)/G(S)/G(T) subunit beta-2 (GBB2) appeared to be biotinylated (Table 1). HPTLC was performed to analyze the lipid composition of Ld and Hd epididymosomes collected in the cauda epididymidis (Fig. 5A). The cholesterol concentration per milligram of proteins was 2-fold higher in Ld epididymosomes when compared with Hd epididymosomes. Similarly, phosphatidylethanolamine, phosphatidylcholine, and sphingomyeline were enriched in Ld epididymosomes by Ld:Hd ratios of 1.7, 2.9, and 2.0, respectively. Phosphatidylserine was found at similar

477


FIG. 4. Densitometric determination of biotinylated proteins of Ld (A) and Hd (B) cauda epididymosomes transferred to caput spermatozoa in absence (Ctr) or presence of excess amount of unbiotinylated Ld or Hd epididymosomes. Quantities of transferred proteins are measured as arbitrary optical density units, and are expressed as a percentage of control consisting in coincubation in the absence of unbiotinylated epididymosomes (Ctr). Vertical bars express SD of four independent experiments. Different letters indicate statistical differences at P , 0.05.

DISCUSSION Through the use of ultracentrifugation on a discontinuous density gradient, we have been able to isolate two distinct populations of epididymosomes from the bovine cauda epididymal intraluminal compartment. The vesicles of the two populations are heterogeneous in size, but are in the same range of diameters (50–250 nm), as revealed by electron microscopy, and have buoyant densities of 1.02–1.15 and .1.15. Fornes et al. have previously described two populations of vesicles from the rat cauda epididymal fluid with similar buoyant densities [29]. In contrast to our observations, the vesicles described in that report differed in size, having diameters of 0.2–1.4 and 3.0–3.5 lm. Obviously, these vesicles are too big to be secreted by an apocrine pathway and to interact with spermatozoa the way epididymosomes do.

TABLE 1. ES-MS/MS identification of biotinylated proteins of cauda epididymosomes. Spot no. 1 1 1 1 2 2 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 5 5 a b

Protein identification

Accession no.

GBB2: Guanine nucleotide-binding protein G(I)/G(S)/G(T) subunit beta-2 GBB1: Guanine nucleotide-binding protein G(I)/G(S)/G(T) subunit beta-1 RGN: Regucalcin ELSPBP1: PREDICTED: similar to epididymal sperm binding protein E12 ELSPBP1: PREDICTED: similar to epididymal sperm binding protein E12 ANXA4: Annexin A4 ELSPBP1: PREDICTED: similar to epididymal sperm binding protein E12 CAPZB: F-actin-capping protein subunit beta NP: Purine nucleoside phosphorylase CLIC5: Chloride intracellular channel protein 5 AP2A2: AP-2 complex subunit alpha-2 PDXP: Pyridoxal phosphate phosphatase ANXA4: Annexin A4 GBB2: Guanine nucleotide-binding protein G(I)/G(S)/G(T) subunit beta-2 ELSPBP1: PREDICTED: similar to epididymal sperm binding protein E12 GBB1: Guanine nucleotide-binding protein G(I)/G(S)/G(T) subunit beta-1 AKR1B1: Aldose reductase LDHB: L-lactate dehydrogenase B ANXA1: Annexin A1 ACTB: Actin, cytoplasmic 1 RGN: Regucalcin ELSPBP1: PREDICTED: similar to epididymal sperm binding protein E12 ANXA4: Annexin A4

P11017 P62871 Q9TTJ5 UPI0000EBDDA3 UPI0000EBDDA3 P13214 UPI0000EBDDA3 P79136 P55859 UPI000179D34C Q0VCK5 Q3ZBF9 P13214 P11017 UPI0000EBDDA3 P62871 P16116 Q5E9B1 P46193 P60712 Q9TTJ5 UPI0000EBDDA3 P13214

Only the peptides with 95% minimum identification probability have been considered. Detection of biotinylated peptides; the number of peptides is in parentheses.

MW 37 37 33 26 26 36 26 34 32 49 104 32 36 37 26 37 36 37 39 42 33 26 36

kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa kDa

No. of peptidesa

Cover (%)

25 14 6 4 10 15 10 8 6 4 4 3 3 22 11 8 5 3 3 4 4 10 18

76 74 25 17 50 52 52 39 37 15 8 15 12 85 50 57 20 11 11 13 18 52 61

Biotinylated peptidesb No No No No Yes No Yes No No No No No No Yes Yes No No No No No No Yes No

(5) (4)

(1) (5)

(3)


concentrations in both populations of cauda epididymosomes (Fig. 5B). Slot blots probed with CTB revealed that the concentration of GM1 was 2.2-fold higher in Ld than in Hd cauda epididymosomes when compared on the basis of a constant protein concentration (Fig. 5, B and C). Ld and Hd epididymosomes from the cauda epididymidis were compared at the ultrastructural level. In both cases, the two populations were characterized by heterogeneous vesicles with a diameter varying between 50 and 250 nm. The epididymosome preparation procedures yielded pure populations of vesicles with no other organelles or cell debris, and were devoid of extravesicular, amorphous, electron-dense material. The Hd population appeared as spherical particles with a relatively high electron-dense appearance. The Ld vesicles showed similar heterogeneity in their diameter, and have a collapsed appearance with a relatively light electrondense contrast (Fig. 6).

478

FRENETTE ET AL.

The protein and lipid compositions vary greatly between the two populations of epididymosomes; this may explain why they behave differently when centrifuged in a density gradient. When comparing the two populations of epididymosomes, the Ld vesicles are enriched in cholesterol and in phospholipids, with the exception of phosphatidylserine. Furthermore, their high concentrations of cholesterol, sphingomyelin, and GM1 are the signature of a high content of detergent-resistant membrane domains, or raft domains. This contributes to the Ld of these vesicles when compared with the Hd population of cauda epididymosomes. In contrast, the higher protein content of the dense population of vesicles present in the cauda epididymidis may contribute to their higher density, and


FIG. 5. Lipid composition of Ld and Hd epididymosomes prepared from the cauda epididymidis. A) HPTLC of lipids extracted from Ld and Hd epididymosomes from the cauda epididymidis. Considering that Hd epididymosomes are enriched in proteins, HPTLC of Ld and Hd epididymosomes were performed on the basis of 6 lg and 12 lg protein, respectively. Lipids were revealed with 10% (w/v) CuSO4 and 8% (v/v) H3PO4. Phospholipids standards (Std) are to the left of the figure. Chol, cholesterol; PE, phosphatidylethanolamine; PS, phosphatidylserine; PC, phosphatidylcholine; SM, sphingomyelin. B) Densitometric quantification of lipid composition of Ld (hatched bars) and Hd (dotted bars) epididymosomes. Results are expressed as arbitrary units per milligram protein. Vertical bars express SD of three independent determinations. *Statistical differences at P , 0.05. C) Slot blot determination of GM1 content of two preparations of Ld and Hd epididymosomes prepared from the cauda epididymidis.

explain why they have a more electron-dense appearance at the ultrastructural level. Although they are heterogeneous in size, the two populations of epididymosomes are devoid of other organelles, cellular debris, or any other organic material. The analysis of their protein content can thus provide some information with regard to their potential roles in sperm maturation. The differences in their protein electrophoresis patterns clearly support the hypothesis that at least two distinct populations of epididymosomes are present in the intraluminal compartment of the bovine cauda epididymidis. The same amounts of AKR1B1 and PEBP1 are associated with both populations of epididymosomes. AKR1B1 is an aldose reductase, known to be secreted by the epididymal epithelium that reduces glucose in sorbitol [30]. Sorbitol being a linear sugar, it cannot cross the sperm plasma membrane. We previously hypothesized that this enzyme deprives spermatozoa of carbohydrate energy sources, maintaining epididymal sperm in a quiescent state during its transit along the excurrent duct [31]. PEBP1 has been described as a decapacitating factor [32]. Like AKR1B1, it thus maintains spermatozoa in a quiescent state before ejaculation. Both types of cauda epididymosomes thus seem to have common roles in the sperm reservoir function of the distal part of the epididymis. While present in both epididymosome preparations, NPC2 (also known as human epididymis protein 1), P25B, and AK1 are more abundant in Ld epididymosomes. NPC2 transcript is highly expressed in the epididymis [33], is found in high soluble concentration in seminal plasma [16], and has been hypothesized to be responsible for sperm membrane cholesterol efflux occurring during epididymal maturation [34] and after ejaculation [16]. This is based on the existence of a cholesterol pocket in the three-dimensional structure of NPC2 [35], and on consequences of a mutation in NPC2 gene resulting in an autosomal Niemann-Pick disease characterized by a late endosomal/lysosomal storage disorder due to a defect in intracellular cholesterol transport [36, 37]. The Ld cauda epididymosomes being enriched in cholesterol, it is not surprising that NPC2 is preferentially associated with this subpopulation of epididymal vesicles. However, whether or not the NPC2 associated with epididymosomes modulates sperm plasma membrane cholesterol content during epididymal maturation remains to be determined. P25B is another epididymis-secreted protein associated in higher concentration with the Ld epididymosomes. P25B is the bovine ortholog [26] of proteins described in human (DCXR, also known as P34H) [38], hamster (P26H) [39, 40], mouse [41], and monkey [42]. These proteins have been proposed to be involved in sperm zona pellucida-binding events. In human, spermatozoa of 40% of men presenting with idiopathic infertility are devoid of P34H, and are unable to bind zona pellucida in vitro [43]. In bovine, low levels of P25B is a predictor of low fertility rates [26, 44]. Spermatozoa acquire P25B during their transit along the epididymis. We recently showed that P25B is GPI anchored to raft domains of both epididymosomes and epididymal spermatozoa and, by an in vitro assay, that P25B is transferred from raft domains of cauda epididymosomes to raft domains of caput spermatozoa [17]. Higher concentrations of cholesterol, sphingomyelin, and GM1 strongly suggest that Ld cauda epididymosomes are enriched in detergent-resistant membrane domains or rafts. It is thus logical that P25B, as well as AK1 [16], another protein associated with raft in epididymal spermatozoa, are found in higher amounts in Ld epididymosomes. Considering that raft domains are involved in the compartmentalization of proteins in both epididymosomes and spermatozoa [17, 23], Ld epididymosomes may coordinates the


479 FIG. 6. Transmission electron micrograph of Ld (A) and Hd (B) epididymosomes from the cauda epididymidis. Micrographs are shown at two different magnifications; bars ¼ 100 nm.

epididymosomes (compare Fig. 3, C–E, of this study with Fig. 2 of Frenette et al. [18]). This would suggest that these vesicles do not fuse with spermatozoa, but transfer only a selected subset of proteins. The bovine genome sequencing being almost completed (http://bovinegenome.org/), we have been able to identify proteins associated with Hd and Ld epididymosomes, but their transfer to the maturing spermatozoa has yet to be confirmed. Only the transfer to spermatozoa of ELSPBP1 and GBB2 has been confirmed by mass spectrometry analysis of biotinylated proteins. The list of these epididymosomal proteins transferred to spermatozoa is only a partial one, due to the limitations of the procedures used in this study. In fact, the listed proteins are restricted to those rare surface proteins with a free-surfaceexposed amine group available for biotinylation. ELSPBP1 possesses four fibronectin-binding domain 2 molecules, and it has been shown that it binds to spermatozoa via the choline group of membrane phospholipids. However, the function of this epididymis-specific protein in sperm maturation remains to be determined. It has been recently proposed that ELSPBP1 prevents premature activation of swelling-activated channels, a mechanisms involved in sperm cell volume control during the epididymal transit [45]. The identification of GBB2 as a biotinylated protein transfer to spermatozoa suggests that epididymosomes may be involved in some maturing process of sperm membrane signaling pathways. In conclusions, at least two different populations of epididymosomes are present in the intraluminal compartment of the cauda epididymidis. Although the biochemical composition of these two populations of vesicles suggests that they interact differently with spermatozoa, the proteins transferred from these two types of epididymosomes to spermatozoa in vitro have similarities. Further work is needed to improve our understanding of the functions of epididymosomes in sperm maturation.


association of epididymal proteins with different functional structures of maturing spermatozoa. Obviously, Hd epididymosomes have biochemical characteristics different from those of Ld vesicles prepared from cauda epididymal fluid. Lipid composition suggests a much lower density of membrane raft domains, and a higher content in proteins. Western blots probed to detect proteins already known to be associated with epididymosomes reveal that very high concentrations of ACTB are associated with Hd epididymosomes. This form of actin is a constituent of the cytoskeleton, and may explain the more elevated electron density of Hd vesicles visualized by electron microscopy. ACTB being barely detectable in Ld epididymosomes, the two populations of epididymal vesicles are probably structurally very different. Due to their high concentration of the cytoskeleton-form of actin (ACTB), the Hd epididymosomes are probably more rigid and resistant than the Ld epididymosomes. This is probably the reason why they do not have the collapsed appearance characterizing the Ld vesicles when observed at the electron microscopy level. Based on their biochemical compositions, we can hypothesize that the two populations of epididymosomes of the cauda epididymis play different roles in epididymal physiology. By using in vitro coincubation assays, we previously showed that selected biotinylated proteins are transferred from epididymosomes to spermatozoa [18], and that caput and cauda epididymosomes transfer distinct subsets of proteins to caput spermatozoa in vitro [14, 23]. Surprisingly, the Ld and Hd subpopulations of cauda epididymosomes transfer the same pattern of proteins, even though they differ in their Coomassie blue-stained or biotinylated protein electrophoretic patterns. In fact, the 2D electrophoretic pattern of biotinylated proteins from Ld and Hd cauda epididymosomes transferred to caput spermatozoa are similar to that of proteins transferred when caput spermatozoa are incubated with unfractionated cauda

480

FRENETTE ET AL.

ACKNOWLEDGMENTS We acknowledge Mr. R. Janvier (Universite´ Laval) for transmission electron microscopy processing of epididymosomes. Dr. Sylvie Bourassa from the proteomic platform of our institution is also acknowledged for her help in MS/MS results analysis. We also wish to acknowledge Dr. M.A. Fortier (Universite´ Laval), Dr. M. Nishibori (Japan), Dr. J.L. Gatti (France), and Drs. Y. Matuo and K. Uchida (Oriental Yeast Co.) for generous gifts of antibodies against AKR1B1, MIF, NPC2, and PGAM2, respectively. Dr. P. Leclerc is acknowledged for critical reading of the manuscript.

REFERENCES

22.

23.

24.

25. 26.

27.

28. 29.

30. 31. 32.

33.

34.

35.

36.

37. 38.

39.

40. 41.

42.

43.

44. 45.


1. Cuasnicu P, Cohen D, Ellerman D, Busso D, DaRos V, Morgenfeld M. Changes in sperm proteins during epididymal maturation. In: Robaire B, Hinton BT (eds.), The Epididymis. From Molecules To Clinical Practice. New York: Kluwer Academic/Plenum Publishers; 2002:389–404. 2. Cooper TG. Interactions between epididymal secretions and spermatozoa. J Reprod Fertil Suppl 1998; 53:119–136. 3. Legare C, Berube B, Boue F, Lefievre L, Morales CR, El-Alfy M, Sullivan R. Hamster sperm antigen P26h is a phosphatidylinositol-anchored protein. Mol Reprod Dev 1999; 52:225–233. 4. Zhang H, Martin-Deleon PA. Mouse epididymal Spam1 (pH-20) is released in the luminal fluid with its lipid anchor. J Androl 2003; 24:51– 58. 5. Kirchhoff C, Hale G. Cell-to-cell transfer of glycosylphosphatidylinositolanchored membrane proteins during sperm maturation. Mol Hum Reprod 1996; 2:177–184. 6. Eickhoff R, Wilhelm B, Renneberg H, Wennemuth G, Bacher M, Linder D, Bucala R, Seitz J, Meinhardt A. Purification and characterization of macrophage migration inhibitory factor as a secretory protein from rat epididymis: evidences for alternative release and transfer to spermatozoa. Mol Med 2001; 7:27–35. 7. Eickhoff R, Baldauf C, Koyro HW, Wennemuth G, Suga Y, Seitz J, Henkel R, Meinhardt A. Influence of macrophage migration inhibitory factor (MIF) on the zinc content and redox state of protein-bound sulphydryl groups in rat sperm: indications for a new role of MIF in sperm maturation. Mol Hum Reprod 2004; 10:605–611. 8. Frenette G, Legare C, Saez F, Sullivan R. Macrophage migration inhibitory factor in the human epididymis and semen. Mol Hum Reprod 2005; 11:575–582. 9. Rejraji H, Sion B, Prensier G, Carreras M, Motta C, Frenoux JM, Vericel E, Grizard G, Vernet P, Drevet JR. Lipid remodeling of murine epididymosomes and spermatozoa during epididymal maturation. Biol Reprod 2006; 74:1104–1113. 10. Saez F, Frenette G, Sullivan R. Epididymosomes and prostasomes: their roles in posttesticular maturation of the sperm cells. J Androl 2003; 24: 149–154. 11. Yanagimachi R, Kamiguchi Y, Mikamo K, Suzuki F, Yanagimachi H. Maturation of spermatozoa in the epididymis of the Chinese hamster. Am J Anat 1985; 172:317–330. 12. Hermo L, Jacks D. Nature’s ingenuity: bypassing the classical secretory route via apocrine secretion. Mol Reprod Dev 2002; 63:394–410. 13. Aumuller G, Wilhelm B, Seitz J. Apocrine secretion—fact or artifact? Anat Anz 1999; 181:437–446. 14. Frenette G, Girouard J, Sullivan R. Comparison between epididymosomes collected in the intraluminal compartment of the bovine caput and cauda epididymidis. Biol Reprod 2006; 75:885–890. 15. Thimon V, Frenette G, Saez F, Thabet M, Sullivan R. Protein composition of human epididymosomes collected during surgical vasectomy reversal: a proteomic and genomic approach. Hum Reprod 2008; 23:1698–1707. 16. Girouard J, Frenette G, Sullivan R. Seminal plasma proteins regulate the association of lipids and proteins within detergent-resistant membrane domains of bovine spermatozoa. Biol Reprod 2008; 78:921–931. 17. Girouard J, Frenette G, Sullivan R. Compartmentalization of proteins in epididymosomes coordinates the association of epididymal proteins with the different functional structures of bovine spermatozoa. Biol Reprod 2009; 80:965–972. 18. Frenette G, Lessard C, Sullivan R. Selected proteins of ‘‘prostasome-like particles’’ from epididymal cauda fluid are transferred to epididymal caput spermatozoa in bull. Biol Reprod 2002; 67:308–313. 19. Frenette G, Sullivan R. Prostasome-like particles are involved in the transfer of P25b from the bovine epididymal fluid to the sperm surface. Mol Reprod Dev 2001; 59:115–121. 20. Rejraji H, Vernet P, Drevet JR. GPX5 is present in the mouse caput and

21.

cauda epididymidis lumen at three different locations. Mol Reprod Dev 2002; 63:96–103. Frenette G, Lessard C, Madore E, Fortier MA, Sullivan R. Aldose reductase and macrophage migration inhibitory factor are associated with epididymosomes and spermatozoa in the bovine epididymis. Biol Reprod 2003; 69:1586–1592. Sullivan R, Saez F, Girouard J, Frenette G. Role of exosomes in sperm maturation during the transit along the male reproductive tract. Blood Cells Mol Dis 2005; 35:1–10. Sullivan R, Frenette G, Girouard J. Epididymosomes are involved in the acquisition of new sperm proteins during epididymal transit. Asian J Androl 2007; 9:483–491. Wessel D, Flugge UI. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal Biochem 1984; 138:141–143. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227:680–685. Parent S, Lefie´vre L, Brindle Y, Sullivan R. Bull subfertility is associated with low levels of a sperm membrane antigen. Mol Reprod Dev 1999; 52: 57–65. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957; 226:497–509. Vitiello F, Zanetta JP. Thin-layer chromatography of phospholipids. J Chromatogr 1978; 166:637–640. Fornes MW, Sosa MA, Bertini F, Burgos MH. Vesicles in rat epididymal fluid. Existence of two populations differing in ultrastructure and enzymatic composition. Andrologia 1995; 27:233–237. Oates PJ. Polyol pathway and diabetic peripheral neuropathy. Int Rev Neurobiol 2002; 50:325–392. Frenette G, Lessard C, Sullivan R. Polyol pathway along the bovine epididymis. Mol Reprod Dev 2004; 69:448–456. Nixon B, MacIntyre DA, Mitchell LA, Gibbs GM, O’Bryan M, Aitken RJ. The identification of mouse sperm-surface-associated proteins and characterization of their ability to act as decapacitation factors. Biol Reprod 2006; 74:275–287. Kirchhoff C. Specific gene expression in the human and non-human primate epididymis. In: Robaire B, Hinton BT (eds.), The Epididymis. From Molecules To Clinical Practice. New York: Kluwer Academic/ Plenum Publishers; 2002; 201–218. Legare C, Thabet M, Gatti JL, Sullivan R. HE1/NPC2 status in human reproductive tract and ejaculated spermatozoa: consequence of vasectomy. Mol Hum Reprod 2006; 12:461–468. Friedland N, Liou HL, Lobel P, Stock AM. Structure of a cholesterolbinding protein deficient in Niemann-Pick type C2 disease. Proc Natl Acad Sci U S A 2003; 100:2512–2517. Naureckiene S, Sleat DE, Lackland H, Fensom A, Vanier MT, Wattiaux R, Jadot M, Lobel P. Identification of HE1 as the second gene of Niemann-Pick C disease. Science 2000; 290:2298–2301. Vanier MT, Millat G. Structure and function of the NPC2 protein. Biochim Biophys Acta 2004; 1685:14–21. Legare C, Gaudreault C, St.-Jacques S, Sullivan R. P34H sperm protein is preferentially expressed by the human corpus epididymidis. Endocrinology 1999; 140:3318–3327. Gaudreault C, Legare C, Berube B, Sullivan R. Hamster sperm protein, p26h: a member of the short-chain dehydrogenase/reductase superfamily. Biol Reprod 1999; 61:264–273. Gaudreault C, El Alfy M, Legare C, Sullivan R. Expression of the hamster sperm protein P26h during spermatogenesis. Biol Reprod 2001; 65:79–86. Begin S, Berube B, Boue F, Sullivan R. Comparative immunoreactivity of mouse and hamster sperm proteins recognized by an anti-P26h hamster sperm protein. Mol Reprod Dev 1995; 41:249–256. Lamontagne N, Legare C, Gaudreault C, Sullivan R. Identification and characterization of P31m, a novel sperm protein in Cynomolgus monkey (Macaca fascicularis). Mol Reprod Dev 2001; 59:431–441. Boue F, Sullivan R. Cases of human infertility are associated with the absence of P34H an epididymal sperm antigen. Biol Reprod 1996; 54: 1018–1024. Sullivan R. Male fertility markers, myth or reality. Anim Reprod Sci 2004; 82–83:341–347. Sahin E, Petrunkina AM, Ekhlasi-Hundrieser M, Hettel C, Waberski D, Harrison RA, Topfer-Petersen E. Fibronectin type II-module proteins in the bovine genital tract and their putative role in cell volume control during sperm maturation. Reprod Fertil Dev 2009; 21:479–488.