- Email: [email protected]
Purification and properties of major α-d -mannosidase in the luminal fluid of porcine epididymis
Biochimica et Biophysica Acta 1432 (1999) 382^392 www.elsevier.com/locate/bba
Puri¢cation and properties of major K-D-mannosidase in the luminal £uid of porcine epididymis Yin-Zhe Jin a , Francoise Dacheux d , Jean-Louis Dacheux d , Shiro Bannai a , Yoshiki Sugita e , Naomichi Okamura a;b;c; * b
a Institute of Basic Medical Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan College of Medical Technology and Nursing, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan c Tsukuba Advanced Research Alliance, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan d Laboratoire de Physiologie de la Reproduction, INRA, 37380 Monnaie, France e Ibaraki Prefectural University of Health Sciences, Ami-machi, Inashiki-gun, Ibaraki 300-0331, Japan
Received 15 March 1999; received in revised form 3 May 1999; accepted 3 May 1999
Abstract A lysosomal type K-D-mannosidase was successfully purified by DEAE-Sephacel, Red-Amicon and Superdex 200 column chromatographies from porcine cauda epididymal fluid. The purified enzyme consisted of 63 and 51 kDa subunits at equimolar amounts. It cleaved K1-2 linked mannosyl residues and less but significantly cleaved K1-3 and K1-6 linked mannosyl residues in the high-mannose oligosaccharides. The optimal pH to hydrolyze oligosaccharide was in the acidic pH range (pH 3.5V4.0). Total K-D-mannosidase activities in the porcine epididymal fluid increased from proximal to distal caput epididymis, which maintained to cauda epididymis. At least two kinds of K-D-mannosidase (lysosomal type enzyme and 135 kDa K-D-mannosidase (MAN2B2)) were contained in the porcine epididymal fluid. The activity of the lysosomal type enzyme is much higher than MAN2B2 at the physiological pH. These results suggest that the lysosomal type K-Dmannosidase is the predominantly active enzyme in the luminal fluid of porcine epididymis and that it participates in the glycoprotein modification on the sperm surface during epididymal transit. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: K-D-mannosidase; Epididymis; Sperm maturation
1. Introduction Mammalian sperm released from testis are unable to exert progressive motility and fertilizing ability. They must acquire these potential abilities during the passage through the epididymis, and this process
* Corresponding author. College of Medical Technology and Nursing, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan. Fax: +81-298-53-3426; E-mail: [email protected]
is termed sperm maturation. The sperm maturation involves many biochemical reactions including modi¢cation of the sperm surface glycoproteins [1^4], which is mainly attained by glycosyltransferase and glycosidase present in epididymal £uid and/or sperm plasma membrane [5^10]. In this connection, various types of K-D-mannosidase have been found in rat epididymis. For instance, lysosomal [11,12] and cytosolic [13] K-D-mannosidases were found in epididymal homogenate, and Golgi K-D-mannosidaseII-like enzyme was found in epididymal Golgi membrane [14]. Recently, a novel type K-D-mannosidase was
0167-4838 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 9 9 ) 0 0 1 1 7 - X
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
puri¢ed from both sperm plasma membrane and epididymal £uid [15,16]. Those K-D-mannosidases except lysosomal K-D-mannosidase show their optimal activities at neutral pH. For this reason, the role of the lysosomal acidic K-D-mannosidase activity in epididymal £uid has scarcely been postulated in the sperm maturation. In the preceding studies, we puri¢ed a novel 135 kDa K-D-mannosidase (MAN2B2) from the luminal £uid of the porcine cauda epididymis and cloned its cDNA [17^19]. Simultaneously, we found that epididymal £uid contains another type of K-Dmannosidase whose activity was higher than MAN2B2 at both acidic and neutral pH. In the present study, we puri¢ed the enzyme and identi¢ed it as a lysosomal type K-D-mannosidase, which suggests that it contributes to the sperm maturation as a major K-D-mannosidase in the porcine epididymal £uid.
383
K-D-mannosidase activity of each fraction (4 ml) was measured by hydrolysis of PNP-K-D-mannoside. The fractions containing K-D-mannosidase activity were pooled and applied on a Red-Amicon column (2.5U15 cm, Amicon Corp., USA). After standing the column for 30 min, proteins were separated by a stepwise elution: 0, 0.6, 1 and 1.5 M KCl in 20 mM Tris-HCl (pH 7.4) at a £ow rate of 1 ml/min. The enzyme protein eluted with 0.6 M KCl was concentrated by ultra¢ltration through an Amicon PM-10 ¢lter and then was separated by gel-¢ltration through a FPLC-Superdex 200 column (1.6U60 cm, Pharmacia, France). Elution was performed with 20 mM Tris-HCl (pH 7.4) containing 0.25 M NaCl at a £ow rate of 0.4 ml/min. Protein was monitored at 280 nm. The protein content of each fraction was analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were stained with Coomassie Brilliant Blue R-250. 2.3. Amino acid sequencing and homology search
2. Materials and methods 2.1. Preparation of the epididymal £uid Two di¡erent methods were used for the collection of the £uid from the epididymal lumen. For protein puri¢cation, the cauda epididymal £uid was collected by retrograde microperfusion of ductus epididymides of the cauda epididymis with PBS (140 mM NaCl, 2.7 mM KCl, 1.5 mM KH2 PO4 , 8.1 mM Na2 HPO4 ). For determination of the glycosidase activities in the £uid of the various epididymal regions, the epididymal £uids were collected by microperfusion of the ductus epididymides with air. The collected epididymal £uids were then centrifuged twice at 2500 rpm at 4³C for 15 min. The supernatant was centrifuged again at 16 000 rpm at 4³C for 20 min. The resulting supernatant was stored at 0³C. 2.2. Puri¢cation of K-D-mannosidase The cauda epididymal £uid was dialyzed against 20 mM Tris-HCl (pH 8.0), and then applied on a DEAE-Sephacel anion-exchange column (1.5U20 cm, Pharmacia, France). Proteins were eluted with a linear gradient (500 ml) of 0^0.75 M NaCl in 20 mM Tris-HCl (pH 7.4) at a £ow rate of 0.4 ml/min.
Two subunits of the puri¢ed enzyme protein were separated by SDS-PAGE and transferred onto a polyvinylidene di£uoride (PVDF) membrane. The N-terminal amino acid sequencing of the subunits was carried out by the automated Edman degradation, using an Applied Biosystems 477 A protein sequencer (Applied Biosystems, Roissy, France). Search for the proteins homologous to the puri¢ed protein was performed with the database of the National Center for Biotechnology Information. 2.4. Molecular mass determination Molecular mass of the subunits of the puri¢ed protein was determined by SDS-PAGE under both reduced and non-reduced conditions. The analysis of the molecular mass of the native enzyme was carried out by FPLC gel-¢ltration. The marker proteins such as BSA (67 kDa), aldolase (158 kDa), catalase (232 kDa) and ferritin (440 kDa) were applied to FPLC-Superdex 200 column (1.6U60 cm, Pharmacia, France) with the puri¢ed enzyme. The elution condition was the same as noted above for the enzyme puri¢cation. The molecular mass was obtained from a standard curve constructed with the marker proteins.
BBAPRO 35941 5-7-99
384
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 1. Structure of the high-mannose oligosaccharides used for the determination of K-D-mannosidase activities as substrates. Man, GlcNAc and PA indicate mannose, N-acetylglucosamine and pyridylamine residues, respectively.
2.5. Enzyme assay K-D-Mannosidase activities of the chromatography fractions were determined by measuring released pnitrophenol in a standard incubation mixture (100 Wl) containing 25 Wl enzyme solution, 0.625 mM
PNP-K-D-mannoside and 0.1 M citrate and 0.2 M Na2 HPO4 (pH 4.6). After incubation for 20 min at 37³C, the reaction was stopped by the addition of 400 Wl of 1 M Tris-HCl (pH 10). Released p-nitrophenol from the PNP-glycoside was determined from the absorbance at 400 nm. Mannosidase activity toward high-mannose oligosaccharides was determined by hydrolysis of pyridylamino-Man9 GlcNAc2 (Man9 ), pyridylamino-Man8 GlcNAc2 (Man8 ), pyridylamino-Man5 GlcNAc2 (Man5 ) and pyridylamino-Man3 GlcNAc2 (Man3 ), whose structures are shown in Fig. 1. The oligosaccharides were incubated with the puri¢ed enzyme in 0.1 M citrate, 0.2 M Na2 HPO4 bu¡er (pH 4.6) or in 50 mM sodium phosphate bu¡er (pH 7.0) at 37³C. The resulting digest was applied to an APS-2 Hypersil column (250U4.6 mm, Keystone Scienti¢c, Inc., USA) according to the method of Liao et al. [20]. Elution was performed by a linear gradient from 70% acetonitrile, 30% 100 mM sodium phosphate (pH 4.0) to 49% acetonitrile, 51% 78 mM sodium phosphate (pH 4.0) at a £ow rate of 1 ml/min. Retention times of oligosaccharide standards such as Man3 , Man4 , Man5 , Man6 , Man7 , Man8 , and Man9 , were 29.6, 33.5, 37.5, 41.0, 44.6, 48.8, and 52.9 min, respectively. Fluorescence of the digest was detected by £uorescence spectrophotometer
Fig. 2. Regional distribution of K-D-mannosidase activities in the luminal £uid of porcine epididymis. Enzyme activities toward Man8 were measured both at pH 4.6 (a) and 7.0 (b). 7.5 Wl of epididymal £uid obtained from proximal caput (region 2, E2), distal caput (region 4, E4), mid corpus (region 6, E6) or cauda (region 8, E8) were incubated with 20 pmoles of substrate in 50 Wl of incubation bu¡er at 37³C for 60 min (a) or 6 h (b). Protein concentrations of the epididymal £uid of each region are as follows : region 2, 14.9 mg/ml; region 4, 13.6 mg/ml; region 6, 13.1 mg/ml; region 8, 12.5 mg/ml. EF, epididymal £uid. Data are means þ S.E. from three separate experiments. Signi¢cance, as assessed by t-test analysis, is given as *P 6 0.005, **P 6 0.05, and ***P 6 0.1 compared with values for E2.
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
(F-2000, HITACHI, Japan) with excitation at 320 nm and emission at 400 nm. K-D-mannosidase activity was calculated by the reduction of the peak area of pyridylamino substrate at appropriate time of incubation. 3. Results 3.1. Distribution of K-D-mannosidase activities in the luminal £uids of various epididymal regions Modi¢cation of the carbohydrate moiety of the glycoproteins on the sperm surface is known to be one of the prominent features involved in the epididymal sperm maturation. It is also known that several glycosidases and glycosyltransferases are present in the epididymal £uids. It was found that L-N-ace-
385
tylglucosaminidase was the most active glycosidase in the porcine epididymal £uid, followed by K-D-mannosidase, L-D-galactosidase, and K-L-fucosidase in decreasing order (data not shown). Other glycosidases such as L-D-glucosidase, L-D-mannosidase, LD-fucosidase, K-D-glucosidase and K-D-galactosidase exhibited only trace activities at the physiological pH. In this study, we aimed at characterizing K-D-mannosidase in the porcine epididymal £uid. First, the distribution of the K-D-mannosidase activities in the epididymal £uid was analyzed. As shown in Fig. 2, the activities hydrolyzing oligosaccharide (Man8 ) signi¢cantly increased from proximal caput epididymis to distal caput epididymis and the high activity was maintained to cauda epididymis. The K-D-mannosidase activities in the epididymal £uid were higher at pH 4.0 than at pH 7.0 examined so far.
Fig. 3. Separation of K-D-mannosidase activity by DEAE-Sephacel anion-exchange column chromatography. The porcine cauda epididymal £uid was fractionated by DEAE-Sephacel chromatography as described in Section 2. (a) The elution pro¢le of the enzyme activity (solid line, E : measured at pH 4.6, O: measured at pH 6.5) and epididymal protein (broken line). Fractions I and II indicate lysosomal type and 135 kDa K-D-mannosidases, respectively. Black bar shows the fractions collected for the further puri¢cation. (b) The proteins in each fraction analyzed by SDS-PAGE.
BBAPRO 35941 5-7-99
386
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 5. Protein contents in each puri¢cation step analyzed by SDS-PAGE. Lane St, molecular weight standard; lane 1, luminal £uid of cauda epididymis; lane 2, DEAE-Sephacel fraction; lane 3, Red Amicon fraction; lane 4, Superdex 200 fraction. Inset shows density of bands corresponding to the 63 and 51 kDa subunits on lane 4. The gel was scanned into Macintosh computer with £at-bed scanner (EPSON GT-8000) using Adobe Photoshop 3.0 software at 200 dots per inch. The density of protein band was measured by pixel counting using NIH Image 1.56 software.
3.2. Puri¢cation of K-D-mannosidase
Fig. 4. Puri¢cation of the lysosomal type K-D-mannosidase by Red Amicon (a) and Superdex 200 (b) chromatographies. Solid line shows the enzyme activity and broken line shows the protein elution. The fractions indicated by black bars were collected as K-D-mannosidase fraction. F, ferritin; C, catalase; A, aldolase; B, BSA.
K-D-Mannosidase was puri¢ed from the porcine cauda epididymal £uid by successive column chromatographies as described in Section 2. Fig. 3 shows the elution pro¢le of epididymal proteins and enzyme activity in the fractions from the DEAE-Sephacel column. An K-D-mannosidase which was active both at pH 7.0 and 4.6 was eluted at about 0.15% NaCl in fraction 12^24 (Fraction I). On the other hand, the enzyme which was active only at pH 7.0 and had been identi¢ed as MAN2B2 [17^19] was eluted at about 0.25^0.3% NaCl in fraction 40^56 (Fraction II). Other fractions did not show K-D-man-
Table 1 Puri¢cation of lysosomal type K-D-mannosidase from porcine cauda epididymal £uid Puri¢cation steps
Total protein (mg)
Total act. (nmole/min)
Enzyme recovery (%)
Speci¢c act. (nmole/min/mg protein)
Puri¢cation ratio
CEFa DEAEb RAc SDd
98.1 3.8 0.53 0.013
3930 1980 0610 630
100 50 41 16
40 520 3030 48500
1 13 76 1210
a
Cauda epididymal £uid was prepared from three boars. DEAE-Sephacel column chromatography. c Red-Amicon column chromatography. d Superdex 200 FPLC. b
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 6. Analysis of K-D-mannosidases puri¢ed in the presence or absence of various proteinase inhibitors by SDS-PAGE. K-Dmannosidases puri¢ed in the presence (lane 1) or the absence (lane 2) of proteinase inhibitors were analyzed by SDS-PAGE under either reduced (a) or non-reduced (b) conditions. Lane St, molecular weight standard.
nosidase activity toward either PNP-K-D-mannoside (Fig. 3a) nor oligosaccharide (Man8 ) substrate (data not shown). The activity of K-D-mannosidase in Fraction I was more than 3 times as high as that of MAN2B2 even at pH 7.0 (Fig. 3a). Further puri¢cation of the K-D-mannosidase in Fraction I was performed by Red-Amicon and FPLC-Superdex 200 chromatographies. In the two
Fig. 7. N-terminal amino acid sequences of the 63 (a) and 51 (b) kDa subunits of puri¢ed porcine K-D-mannosidase. N-terminal amino acid sequences of the subunits were compared with the corresponding regions of the human (U60266), bovine (L31373) and feline (AF010191) lysosomal K-D-mannosidases.
387
Fig. 8. Substrate speci¢city of the puri¢ed K-D-mannosidase. The puri¢ed enzyme (1 Wg protein) was incubated with various PNP-glycosides (0.625 mM) in the reaction mixture (pH 4.6) at 37³C for 15 min. Released p-nitrophenol from substrates was determined from the absorbance at 400 nm. Glycosidase activities for each substrate were expressed as % of activity for PNP-K-D-mannoside.
steps of puri¢cation, the enzyme protein was successfully separated from other epididymal proteins (Fig. 4a and b). Finally, 0.013% of protein and 16% of the activity were recovered from the original cauda epididymal £uid (Table 1). The molecular mass of the native enzyme estimated by FPLC-Superdex 200 gel-¢ltration was 275 kDa (Fig. 4b). Fig. 5 shows the SDS-PAGE patterns of the enzyme fractions during the puri¢cation. The puri¢ed enzyme protein which was eluted as a single peak from FPLC-Superdex 200 column showed two bands at 63 and 51 kDa on the SDS-PAGE (Fig. 5, lane 4) under the reduced conditions. The ratio of density of the two bands on SDS-PAGE stained with Coomassie Brilliant Blue R-250 was 1.2:1, suggesting that those are present in equimolar amounts. In order to exclude possibilities that the ¢nal fraction contains two di¡erent proteins or non-speci¢cally degraded fragments, all the puri¢cation steps were performed in the presence of various proteinase inhibitors: 1 mM PMSF, EDTA, benzamidine and 1 Wg/ml of leupeptin, pepstatin, phosphoramidon, chymostatin, aprotinin and elastatin. As shown in Fig. 6, the puri¢ed enzyme also contained two bands on SDS-PAGE under the reduced conditions. Furthermore, both enzymes puri¢ed in the absence and the presence of proteinase inhibitors formed a single band at 260 kDa on SDS-PAGE under the non-reduced conditions. As shown in Fig. 7, N-terminal amino acid se-
BBAPRO 35941 5-7-99
388
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
3.3. Properties of the puri¢ed K-D-mannosidase Substrate speci¢city of the puri¢ed enzyme was shown in Fig. 8. The puri¢ed enzyme releases p-nitrophenol speci¢cally from PNP-K-D-mannoside. The optimal pH to hydrolyze both PNP-K-D-mannoside and high-mannose oligosaccharide (Man8 ) was in the acidic pH range, 3.5^4.0 (Fig. 9). As shown in Fig. 10, the intermediate digests of
Fig. 9. E¡ect of pH on the K-D-mannosidase activity. The puri¢ed K-D-mannosidase equivalent to 1 Wg protein was incubated with 0.625 mM PNP-K-D-mannoside (a) or 0.4 WM Man8 (b) in 50 Wl of 50 mM sodium phosphate bu¡er at various pHs at 37³C for 20 (PNP) or 45 (Man8 ) min. Enzyme activities were determined as in Figs. 8 and 10.
quence of the 63 kDa subunit was found to be 60%, 80% and 80% identical with the corresponding residues deduced from the cDNAs coding human (U68567, U05572, U60266) [20^22], bovine (L31373) [23] and feline (AF010191) [24] lysosomal K-D-mannosidases, respectively. Furthermore, N-terminal sequence of the 51 kDa subunit was 55%, 80% and 85% identical with the corresponding residues of human, bovine and feline lysosomal K-D-mannosidases, respectively.
Fig. 10. Separation and identi¢cation of oligosaccharides digested by K-D-mannosidase in the porcine epididymal £uid. 0.4 WM Man9 (a), Man8 (b) or Man5 (c) were incubated with the puri¢ed K-D-mannosidase (5 Wg protein) at 37³C under various conditions: Man9 , at pH 4.6 for 5 min; Man8 , at pH 7.0 for 2 h; Man5 , at pH 4.6 for 20 min, respectively. Resulting digest was separated by APS-2 Hypersil column as described in Section 2. Representative elution pro¢les are shown. K-D-mannosidase activity was determined by the reduction of the peak area of substrate.
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
389
4. Discussion
Fig. 11. Digestion of Man9 and Man5 by the puri¢ed K-D-mannosidase. 0.4 WM Man9 (a) or Man5 (O) was incubated with the puri¢ed enzyme (5 Wg protein) either at pH 4.6 (a) or pH 7.0 (b) for various intervals of time at 37³C. Resulting digest was separated by APS-2 Hypersil column. Enzyme activity was determined as in Fig. 10.
Man9 , Man8 and Man5 produced by the puri¢ed enzyme were separated by an APS-2 Hypersil column and the K-D-mannosidase activity was determined by the reduction of the peak area of the substrate as described in Section 2. At pH 4.6, Man9 was completely hydrolyzed to Man8 VMan1 within 20 min, as shown in Fig. 11a. However, it took two hours to hydrolyze Man5 completely. Similar results were obtained at pH 7.0 (Fig. 11b), although the rate of the hydrolysis was much slower than at acidic pH.
Several studies have demonstrated that changes in the carbohydrate moiety occur on the sperm plasma membrane during epididymal transit [25^27]. Although the carbohydrate composition of the sperm surface is not fully elucidated yet, high contents of mannosyl residues are observed in N linked oligosaccharide core and terminus of the various glycoproteins. Therefore, those residues are possibly the target of K-D-mannosidase in the modi¢cation of the sperm surface. We have recently cloned a novel 135 kDa K-Dmannosidase (MAN2B2) which has optimal pH at 6.5 and is not active at acidic pH [17^19]. In addition to this enzyme, we found another K-D-mannosidase in the epididymal £uid, whose activity is much higher than MAN2B2 both at acidic and neutral pHs. In the present study, we could purify the K-D-mannosidase in only three chromatographic steps, starting with the luminal £uid of porcine epididymis. The puri¢ed enzyme consists of 63 kDa and 51 kDa subunits at equimolar amounts. As shown in Fig. 6, even when the enzyme was puri¢ed in the presence of various proteinase inhibitors, composition of subunits was just the same as the enzyme puri¢ed in the absence of inhibitors. Furthermore, under the non-reduced condition, the enzymes puri¢ed both in the absence and the presence of the proteinase inhibitors were moved as a single band at 260 kDa on SDS-PAGE. It is unlikely that non-speci¢c degradation of the enzyme occurred to produce two fragments during the puri¢cation procedures. The puri¢ed enzyme was also shown to have an apparent molecular mass of 275 kDa by molecular sieve chromatography. This value corresponds to the apparent molecular weights for the lysosomal K-Dmannosidase of rat liver (335 000 by molecular sieve chromatography and 200 000 by sucrose density centrifugation) [28] or epididymis (220 000 by molecular sieve chromatography) [12]. These results indicate that the puri¢ed enzyme is composed of tetramer: (63 kDa subunit)2 c(51 kDa subunit)2 . As shown in Fig. 7, N-terminal sequences of both 63 and 51 kDa subunits had very high homology with the corresponding residues of the human, bovine and feline lysosomal K-D-mannosidases [20^24]. All these data together suggest that the K-D-manno-
BBAPRO 35941 5-7-99
390
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
sidase puri¢ed from the porcine epididymal £uid is a homologue of the lysosomal enzyme. It is also indicated that the precursor of the puri¢ed enzyme is cleaved at two sites, releasing a signal peptide and producing 63 and 51 kDa subunits. The N-terminus of the 63 kDa subunit is in good agreement with the cleavage site of the signal peptide of several lysosomal K-D-mannosidases [20^24]. On the other hand, the N-terminus of the 51 kDa subunit is located ¢ve residues before the position of the potential cleavage site as predicted from human and bovine lysosomal K-D-mannosidases [20^23]. Emiliani et al. [29] have puri¢ed human lysosomal K-D-mannosidase, but the ¢nal fraction contained 67, 59, 55 and 30 kDa fragments on SDS-PAGE, showing a high microheterogeneity of the enzyme. Among those fragments, the partial amino acid sequence of the 30 kDa fragment was found to correspond to the amino acid sequence predicted by human lysosomal K-D-mannosidase cDNA (U05572) [21]. In addition, Nilssen et al. [22] have suggested that the human enzyme is synthesized as a single-chain precursor which is processed into three glycopeptides of 70, 42 and 15 kDa and that the 70 kDa peptide is further partially proteolysed into three more peptides. The bovine and feline mature enzymes have been also suggested to contain ¢ve and three peptides, respectively [23,24]. The reason for the discrepancy in the subunit composition between the enzyme puri¢ed from the porcine epididymal £uid and other lysosomal K-D-mannosidases is not clear. It is interesting to know whether the di¡erent subunit composition causes the secretion of the enzyme into the epididymal lumen instead being targeted to lysosome. It is possible that during the puri¢cation processes some lysosomal proteinases digested lysosomal K-D-mannosidases. On the other hand, it has been known that epididymal £uid contains proteinase inhibitors, which possibly protect against the degradation of the native form of the K-D-mannosidase. It thus seems that epididymal £uid is a very good source for obtaining the lysosomal K-D-mannosidase homologue without producing microheterogeneity in the enzyme preparation. The puri¢ed K-D-mannosidase cleaved both PNPK-D-mannoside and high-mannose oligosaccharides most actively at pH 4.0. At pH 3.0, the enzyme cleaved little oligosaccharide while signi¢cant
amounts of PNP-K-D-mannoside were degraded by it. On the other hand, the enzyme cleaved very little PNP substrate but was e¤cient in cleaving oligosaccharide substrate at the neutral pH. The exact reason for these di¡erences has not been known, but there have been several reports of di¡erences in pH dependency when PNP substrates have been compared with natural one [12]. pH of the epididymal £uid has been reported to be around 5.8V7.2, which is lower than serum pH [30^ 32]. We found that pH of the porcine epididymal £uid gradually declined from caput to cauda; 6.94 þ 0.16 in caput, 6.84 þ 0.19 in corpus and 6.73 þ 0.17 in cauda. These results suggest that the puri¢ed enzyme can be physiologically active toward natural substrate in the epididymal £uid. We determined the substrate speci¢city of the puri¢ed enzyme using the high-mannose oligosaccharides Man9 and Man5 as substrates. The puri¢ed enzyme hydrolyzed both substrates both at acidic and neutral pHs. Such rather broad substrate speci¢city is a characteristic for the lysosomal K-D-mannosidase [20,33]. But the rate of K1-3 and K1-6 cleavage in Man5 was much slower than the rate of K1-2 cleavage in Man9 . Tulsiani et al. [12,14^16] have extensively studied the properties of the rat epididymal K-D-mannosidases. They puri¢ed an acidic K-D-mannosidase from the acid extract of rat epididymis and identi¢ed it as a lysosomal K-D-mannosidase [12]. Rat epididymal K-D-mannosidase contains two heavy chains (60 kDa) and two light chains (31 kDa), which has similar subunit composition to the enzyme puri¢ed from the porcine epididymal £uid. But the substrate speci¢cities of rat and porcine enzymes were slightly di¡erent. It has been shown that rat lysosomal K-D-mannosidase hydrolyzes high-mannose oligosaccharides (Man9ÿ5 ) and the branched trimannosyl oligosaccharide (Man(K1-6)[Man(K1-3)]Man(L1-4)GlcNAc) very slowly [12]. On the other hand, the lysosomal type K-D-mannosidase secreted into the porcine epididymal £uid quickly hydrolyzed K1-2 mannose residues in the high-mannose oligosaccharides. In contrast, the mannosidase hydrolyzed the branched oligosaccharide (Man5 and Man3 ) at a slower rate, ¢nally yielding pyridylamino-Man(L1-4)GlcNAc(L1-4)GlcNAc(Man1 ). It is suggested that the secretory type of lysosomal K-D-mannosidase is
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
processed di¡erently from the lysosome-targeted enzyme and has a di¡erent substrate speci¢city. As shown in Fig. 3, the activity of lysosomal type K-D-mannosidase (Fraction I) toward PNP substrate is higher than that of MAN2B2 (Fraction II) both at acidic and neutral pHs. Furthermore, the activity of the puri¢ed lysosomal type enzyme toward oligosaccharide substrate at neutral pH (Fig. 11) was also con¢rmed to be higher than that of MAN2B2 determined previously [18]. Though MAN2B2 cleaves K1-2 bond but neither K1-3 nor K1-6 bond [18], the puri¢ed lysosomal type enzyme can hydrolyze all. We could not detect any other K-D-mannosidase activities in the epididymal £uid, though several types of the K-D-mannosidases have been found in the mammalian epididymis [13^16]. The present results suggest that lysosomal type K-D-mannosidase is responsible for the major part of K-D-mannosidase activity which can hydrolyze K1-2, K1-3 and K1-6 bonds in the luminal £uid of porcine epididymis. K-D-mannosidase activity, as well as those of other glycosidases, increases in distal caput and proximal corpus region as shown in Fig. 2. There is growing evidence indicating that distal caput is a unique part where sperm ¢rst acquire their motile and fertilizing abilities though not perfect [34]. Recently, we have found that procathepsin L, which is a lysosomal proteinase, is expressed in distal caput and is secreted into the extracellular £uid instead being targeted to lysosome, and this process may be regulated by platelet-derived growth factor [35]. These results indicate that, although the overall function of epididymis is principally regulated by androgens, there are other additional mechanisms regulating the locally restricted function of epididymis. The expression and secretion mechanisms of the lysosomal type K-Dmannosidase in the porcine epididymis need to be elucidated. This is the ¢rst report that the secretory type of the lysosomal K-D-mannosidase was puri¢ed from the luminal £uid of porcine epididymis. The present results suggest that this K-D-mannosidase contributes to major part of K-D-mannosidase activity in the luminal £uid of epididymis, which may be closely related to the glycoprotein modi¢cation on the surface of maturating sperm.
391
Acknowledgements This work was supported in part by grants from the Ministry of Education (05044121 and 10680682) and by a grant from University of Tsukuba Project Research to N.O., and partly by Tsukuba Advanced Research Alliance (TARA Sakabe-Shoun project).
References [1] M.C. Orgebin-Crist, in: D.M. deKretser, G. Spera (Eds.), Morphological Basis of Human Reproductive Function, Acta Medica, Plenum Press, New York, 1987, pp. 155^174. [2] J.C. Hall, G.J. Killian, Biol. Reprod. 36 (1987) 709^718. [3] R. Yanagimachi, in: E. Knobil, J. Neill (Eds.), The Physiology of Reproduction, Vol. 1, Raven Press, New York, 1988, pp. 135^186. [4] R. Jones, in: S.R. Milligan (Eds.), Oxford Reviews of Reproduction Biology, Vol. 2, Clarendon, Oxford, 1989, pp. 285^337. [5] A. Bernal, J. Torres, A. Reyes, A. Rosado, Biol. Reprod. 23 (1980) 290^293. [6] D.W. Hamilton, Biol. Reprod. 23 (1980) 377^385. [7] B. Tadolini, T.J. Wilson, P.R. Reddy, H.G. Williams-Ashman, Adv. Enzyme Regul. 15 (1976) 319^336. [8] M.D. Skudlarek, M.C. Orgebin-Crist, Biol. Reprod. 35 (1986) 167^178. [9] M.D. Skudlarek, D.R. Tulsiani, M.C. Orgebin-Crist, Biochem. J. 286 (1992) 907^914. [10] D.R. Tulsiani, M.D. Skudlarek, M.K. Holland, M.C. Orgebin- Crist, Biol. Reprod. 48 (1993) 417^428. [11] S.M. Snaith, G.A. Levvy, Biochem. J. 114 (1969) 25^33. [12] D.R. Tulsiani, V.D. Coleman, O. Touster, Arch. Biochem. Biophys. 267 (1988) 60^68. [13] P. Dutta, G.C. Majumder, Biochem. J. 218 (1984) 489^494. [14] M.D. Skudlarek, M.C. Orgebin-Crist, D.R. Tulsiani, Biochem. J. 277 (1991) 213^221. [15] D.R. Tulsiani, M.D. Skudlarek, M.C. Orgebin-Crist, J. Cell Biol. 109 (1989) 1257^1267. [16] D.R. Tulsiani, M.D. Skudlarek, S.K. Nagdas, M.C. Orgebin-Crist, Biochem. J. 290 (1993) 427^436. [17] N. Okamura, F. Dacheux, A. Venien, S. Onoe, J.C. Huet, J.L. Dacheux, Biol. Reprod. 6 (1992) 1040^1052. [18] N. Okamura, M. Tamba, H.J. Liao, S. Onoe, Y. Sugita, F. Dacheux, J.L. Dacheux, Mol. Reprod. Dev. 42 (1995) 141^ 148. [19] K. Ohata, N. Okamura, M. Kojima, H. Yasue, Mamm. Genome 8 (1997) 158^159. [20] Y.F. Liao, A. Lal, K.W. Moremen, J. Biol. Chem. 271 (1996) 28348^28358. [21] V.L. Nebes, M.C. Schmidt, Biochem. Biophys. Res. Commun. 200 (1994) 239^245.
BBAPRO 35941 5-7-99
392
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
[22] O. Nilssen, T. Berg, H.M.F. Riise, U. Ramachandran, G. Evjen, G.M. Hansen, D. Malm, L. Tranebjarg, O.K. Tollersrud, Human Mol. Genet. 6 (1997) 717^726. [23] O.K. Tollersrud, T. Berg, P. Healy, G. Evjen, U. Ramachandran, O. Nilssen, Eur. J. Biochem. 246 (1997) 410^419. [24] T. Berg, O.K. Tollersrud, S.U. Walkley, D. Siegel, O. Nilssen, Biochem. J. 328 (1997) 863^870. [25] R.H. Hammerstedt, A.D. Keith, S. Hay, N. Deluca, R.P. Amann, Arch. Biochem. Biophys. 196 (1979) 7^12. [26] J.L. Courtens, S. Fournier-Delpech, J. Ultrastruct. Res. 68 (1979) 136^148. [27] G.L. Nicolson, N. Usui, R. Yanagimachi, H. Yanagimachi, J.R. Smith, J. Cell Biol. 74 (1977) 950^962. [28] D.J. Opheim, O. Touster, J. Biol. Chem. 253 (1978) 1017^ 1023.
[29] C. Emiliani, S. Martino, J.L. Stirling, B. Maras, A. Orlacchio, Biochem. J. 305 (1995) 363^366. [30] D.W. Carr, M.C. Usselman, T.S. Acott, Biol. Reprod. 33 (1985) 588^595. [31] C.R. Ca£isch, T.D. DuBose Jr., J. Toxicol. Environ. Health 32 (1991) 49^57. [32] N. Okamura, Y. Tajima, A. Soejima, H. Masuda, Y. Sugita, J. Biol. Chem. 260 (1985) 9699^9705. [33] S. al Daher, R. de Gasperi, P. Daniel, N. Hall, C.D. Warren, B. Winchester, Biochem. J. 277 (1991) 743^751. [34] J.L. Dacheux, M. Paquignon, Reprod. Nutr. Dev. 20 (1980) 1085^1099. [35] N. Okamura, M. Tamba, Y. Uchiyama, Y. Sugita, F. Dacheux, P. Syntin, J.L. Dacheux, Biochim. Biophys. Acta 1245 (1995) 221^226.
BBAPRO 35941 5-7-99
Puri¢cation and properties of major K-D-mannosidase in the luminal £uid of porcine epididymis Yin-Zhe Jin a , Francoise Dacheux d , Jean-Louis Dacheux d , Shiro Bannai a , Yoshiki Sugita e , Naomichi Okamura a;b;c; * b
a Institute of Basic Medical Sciences, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan College of Medical Technology and Nursing, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan c Tsukuba Advanced Research Alliance, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan d Laboratoire de Physiologie de la Reproduction, INRA, 37380 Monnaie, France e Ibaraki Prefectural University of Health Sciences, Ami-machi, Inashiki-gun, Ibaraki 300-0331, Japan
Received 15 March 1999; received in revised form 3 May 1999; accepted 3 May 1999
Abstract A lysosomal type K-D-mannosidase was successfully purified by DEAE-Sephacel, Red-Amicon and Superdex 200 column chromatographies from porcine cauda epididymal fluid. The purified enzyme consisted of 63 and 51 kDa subunits at equimolar amounts. It cleaved K1-2 linked mannosyl residues and less but significantly cleaved K1-3 and K1-6 linked mannosyl residues in the high-mannose oligosaccharides. The optimal pH to hydrolyze oligosaccharide was in the acidic pH range (pH 3.5V4.0). Total K-D-mannosidase activities in the porcine epididymal fluid increased from proximal to distal caput epididymis, which maintained to cauda epididymis. At least two kinds of K-D-mannosidase (lysosomal type enzyme and 135 kDa K-D-mannosidase (MAN2B2)) were contained in the porcine epididymal fluid. The activity of the lysosomal type enzyme is much higher than MAN2B2 at the physiological pH. These results suggest that the lysosomal type K-Dmannosidase is the predominantly active enzyme in the luminal fluid of porcine epididymis and that it participates in the glycoprotein modification on the sperm surface during epididymal transit. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: K-D-mannosidase; Epididymis; Sperm maturation
1. Introduction Mammalian sperm released from testis are unable to exert progressive motility and fertilizing ability. They must acquire these potential abilities during the passage through the epididymis, and this process
* Corresponding author. College of Medical Technology and Nursing, University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan. Fax: +81-298-53-3426; E-mail: [email protected]
is termed sperm maturation. The sperm maturation involves many biochemical reactions including modi¢cation of the sperm surface glycoproteins [1^4], which is mainly attained by glycosyltransferase and glycosidase present in epididymal £uid and/or sperm plasma membrane [5^10]. In this connection, various types of K-D-mannosidase have been found in rat epididymis. For instance, lysosomal [11,12] and cytosolic [13] K-D-mannosidases were found in epididymal homogenate, and Golgi K-D-mannosidaseII-like enzyme was found in epididymal Golgi membrane [14]. Recently, a novel type K-D-mannosidase was
0167-4838 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 9 9 ) 0 0 1 1 7 - X
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
puri¢ed from both sperm plasma membrane and epididymal £uid [15,16]. Those K-D-mannosidases except lysosomal K-D-mannosidase show their optimal activities at neutral pH. For this reason, the role of the lysosomal acidic K-D-mannosidase activity in epididymal £uid has scarcely been postulated in the sperm maturation. In the preceding studies, we puri¢ed a novel 135 kDa K-D-mannosidase (MAN2B2) from the luminal £uid of the porcine cauda epididymis and cloned its cDNA [17^19]. Simultaneously, we found that epididymal £uid contains another type of K-Dmannosidase whose activity was higher than MAN2B2 at both acidic and neutral pH. In the present study, we puri¢ed the enzyme and identi¢ed it as a lysosomal type K-D-mannosidase, which suggests that it contributes to the sperm maturation as a major K-D-mannosidase in the porcine epididymal £uid.
383
K-D-mannosidase activity of each fraction (4 ml) was measured by hydrolysis of PNP-K-D-mannoside. The fractions containing K-D-mannosidase activity were pooled and applied on a Red-Amicon column (2.5U15 cm, Amicon Corp., USA). After standing the column for 30 min, proteins were separated by a stepwise elution: 0, 0.6, 1 and 1.5 M KCl in 20 mM Tris-HCl (pH 7.4) at a £ow rate of 1 ml/min. The enzyme protein eluted with 0.6 M KCl was concentrated by ultra¢ltration through an Amicon PM-10 ¢lter and then was separated by gel-¢ltration through a FPLC-Superdex 200 column (1.6U60 cm, Pharmacia, France). Elution was performed with 20 mM Tris-HCl (pH 7.4) containing 0.25 M NaCl at a £ow rate of 0.4 ml/min. Protein was monitored at 280 nm. The protein content of each fraction was analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were stained with Coomassie Brilliant Blue R-250. 2.3. Amino acid sequencing and homology search
2. Materials and methods 2.1. Preparation of the epididymal £uid Two di¡erent methods were used for the collection of the £uid from the epididymal lumen. For protein puri¢cation, the cauda epididymal £uid was collected by retrograde microperfusion of ductus epididymides of the cauda epididymis with PBS (140 mM NaCl, 2.7 mM KCl, 1.5 mM KH2 PO4 , 8.1 mM Na2 HPO4 ). For determination of the glycosidase activities in the £uid of the various epididymal regions, the epididymal £uids were collected by microperfusion of the ductus epididymides with air. The collected epididymal £uids were then centrifuged twice at 2500 rpm at 4³C for 15 min. The supernatant was centrifuged again at 16 000 rpm at 4³C for 20 min. The resulting supernatant was stored at 0³C. 2.2. Puri¢cation of K-D-mannosidase The cauda epididymal £uid was dialyzed against 20 mM Tris-HCl (pH 8.0), and then applied on a DEAE-Sephacel anion-exchange column (1.5U20 cm, Pharmacia, France). Proteins were eluted with a linear gradient (500 ml) of 0^0.75 M NaCl in 20 mM Tris-HCl (pH 7.4) at a £ow rate of 0.4 ml/min.
Two subunits of the puri¢ed enzyme protein were separated by SDS-PAGE and transferred onto a polyvinylidene di£uoride (PVDF) membrane. The N-terminal amino acid sequencing of the subunits was carried out by the automated Edman degradation, using an Applied Biosystems 477 A protein sequencer (Applied Biosystems, Roissy, France). Search for the proteins homologous to the puri¢ed protein was performed with the database of the National Center for Biotechnology Information. 2.4. Molecular mass determination Molecular mass of the subunits of the puri¢ed protein was determined by SDS-PAGE under both reduced and non-reduced conditions. The analysis of the molecular mass of the native enzyme was carried out by FPLC gel-¢ltration. The marker proteins such as BSA (67 kDa), aldolase (158 kDa), catalase (232 kDa) and ferritin (440 kDa) were applied to FPLC-Superdex 200 column (1.6U60 cm, Pharmacia, France) with the puri¢ed enzyme. The elution condition was the same as noted above for the enzyme puri¢cation. The molecular mass was obtained from a standard curve constructed with the marker proteins.
BBAPRO 35941 5-7-99
384
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 1. Structure of the high-mannose oligosaccharides used for the determination of K-D-mannosidase activities as substrates. Man, GlcNAc and PA indicate mannose, N-acetylglucosamine and pyridylamine residues, respectively.
2.5. Enzyme assay K-D-Mannosidase activities of the chromatography fractions were determined by measuring released pnitrophenol in a standard incubation mixture (100 Wl) containing 25 Wl enzyme solution, 0.625 mM
PNP-K-D-mannoside and 0.1 M citrate and 0.2 M Na2 HPO4 (pH 4.6). After incubation for 20 min at 37³C, the reaction was stopped by the addition of 400 Wl of 1 M Tris-HCl (pH 10). Released p-nitrophenol from the PNP-glycoside was determined from the absorbance at 400 nm. Mannosidase activity toward high-mannose oligosaccharides was determined by hydrolysis of pyridylamino-Man9 GlcNAc2 (Man9 ), pyridylamino-Man8 GlcNAc2 (Man8 ), pyridylamino-Man5 GlcNAc2 (Man5 ) and pyridylamino-Man3 GlcNAc2 (Man3 ), whose structures are shown in Fig. 1. The oligosaccharides were incubated with the puri¢ed enzyme in 0.1 M citrate, 0.2 M Na2 HPO4 bu¡er (pH 4.6) or in 50 mM sodium phosphate bu¡er (pH 7.0) at 37³C. The resulting digest was applied to an APS-2 Hypersil column (250U4.6 mm, Keystone Scienti¢c, Inc., USA) according to the method of Liao et al. [20]. Elution was performed by a linear gradient from 70% acetonitrile, 30% 100 mM sodium phosphate (pH 4.0) to 49% acetonitrile, 51% 78 mM sodium phosphate (pH 4.0) at a £ow rate of 1 ml/min. Retention times of oligosaccharide standards such as Man3 , Man4 , Man5 , Man6 , Man7 , Man8 , and Man9 , were 29.6, 33.5, 37.5, 41.0, 44.6, 48.8, and 52.9 min, respectively. Fluorescence of the digest was detected by £uorescence spectrophotometer
Fig. 2. Regional distribution of K-D-mannosidase activities in the luminal £uid of porcine epididymis. Enzyme activities toward Man8 were measured both at pH 4.6 (a) and 7.0 (b). 7.5 Wl of epididymal £uid obtained from proximal caput (region 2, E2), distal caput (region 4, E4), mid corpus (region 6, E6) or cauda (region 8, E8) were incubated with 20 pmoles of substrate in 50 Wl of incubation bu¡er at 37³C for 60 min (a) or 6 h (b). Protein concentrations of the epididymal £uid of each region are as follows : region 2, 14.9 mg/ml; region 4, 13.6 mg/ml; region 6, 13.1 mg/ml; region 8, 12.5 mg/ml. EF, epididymal £uid. Data are means þ S.E. from three separate experiments. Signi¢cance, as assessed by t-test analysis, is given as *P 6 0.005, **P 6 0.05, and ***P 6 0.1 compared with values for E2.
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
(F-2000, HITACHI, Japan) with excitation at 320 nm and emission at 400 nm. K-D-mannosidase activity was calculated by the reduction of the peak area of pyridylamino substrate at appropriate time of incubation. 3. Results 3.1. Distribution of K-D-mannosidase activities in the luminal £uids of various epididymal regions Modi¢cation of the carbohydrate moiety of the glycoproteins on the sperm surface is known to be one of the prominent features involved in the epididymal sperm maturation. It is also known that several glycosidases and glycosyltransferases are present in the epididymal £uids. It was found that L-N-ace-
385
tylglucosaminidase was the most active glycosidase in the porcine epididymal £uid, followed by K-D-mannosidase, L-D-galactosidase, and K-L-fucosidase in decreasing order (data not shown). Other glycosidases such as L-D-glucosidase, L-D-mannosidase, LD-fucosidase, K-D-glucosidase and K-D-galactosidase exhibited only trace activities at the physiological pH. In this study, we aimed at characterizing K-D-mannosidase in the porcine epididymal £uid. First, the distribution of the K-D-mannosidase activities in the epididymal £uid was analyzed. As shown in Fig. 2, the activities hydrolyzing oligosaccharide (Man8 ) signi¢cantly increased from proximal caput epididymis to distal caput epididymis and the high activity was maintained to cauda epididymis. The K-D-mannosidase activities in the epididymal £uid were higher at pH 4.0 than at pH 7.0 examined so far.
Fig. 3. Separation of K-D-mannosidase activity by DEAE-Sephacel anion-exchange column chromatography. The porcine cauda epididymal £uid was fractionated by DEAE-Sephacel chromatography as described in Section 2. (a) The elution pro¢le of the enzyme activity (solid line, E : measured at pH 4.6, O: measured at pH 6.5) and epididymal protein (broken line). Fractions I and II indicate lysosomal type and 135 kDa K-D-mannosidases, respectively. Black bar shows the fractions collected for the further puri¢cation. (b) The proteins in each fraction analyzed by SDS-PAGE.
BBAPRO 35941 5-7-99
386
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 5. Protein contents in each puri¢cation step analyzed by SDS-PAGE. Lane St, molecular weight standard; lane 1, luminal £uid of cauda epididymis; lane 2, DEAE-Sephacel fraction; lane 3, Red Amicon fraction; lane 4, Superdex 200 fraction. Inset shows density of bands corresponding to the 63 and 51 kDa subunits on lane 4. The gel was scanned into Macintosh computer with £at-bed scanner (EPSON GT-8000) using Adobe Photoshop 3.0 software at 200 dots per inch. The density of protein band was measured by pixel counting using NIH Image 1.56 software.
3.2. Puri¢cation of K-D-mannosidase
Fig. 4. Puri¢cation of the lysosomal type K-D-mannosidase by Red Amicon (a) and Superdex 200 (b) chromatographies. Solid line shows the enzyme activity and broken line shows the protein elution. The fractions indicated by black bars were collected as K-D-mannosidase fraction. F, ferritin; C, catalase; A, aldolase; B, BSA.
K-D-Mannosidase was puri¢ed from the porcine cauda epididymal £uid by successive column chromatographies as described in Section 2. Fig. 3 shows the elution pro¢le of epididymal proteins and enzyme activity in the fractions from the DEAE-Sephacel column. An K-D-mannosidase which was active both at pH 7.0 and 4.6 was eluted at about 0.15% NaCl in fraction 12^24 (Fraction I). On the other hand, the enzyme which was active only at pH 7.0 and had been identi¢ed as MAN2B2 [17^19] was eluted at about 0.25^0.3% NaCl in fraction 40^56 (Fraction II). Other fractions did not show K-D-man-
Table 1 Puri¢cation of lysosomal type K-D-mannosidase from porcine cauda epididymal £uid Puri¢cation steps
Total protein (mg)
Total act. (nmole/min)
Enzyme recovery (%)
Speci¢c act. (nmole/min/mg protein)
Puri¢cation ratio
CEFa DEAEb RAc SDd
98.1 3.8 0.53 0.013
3930 1980 0610 630
100 50 41 16
40 520 3030 48500
1 13 76 1210
a
Cauda epididymal £uid was prepared from three boars. DEAE-Sephacel column chromatography. c Red-Amicon column chromatography. d Superdex 200 FPLC. b
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
Fig. 6. Analysis of K-D-mannosidases puri¢ed in the presence or absence of various proteinase inhibitors by SDS-PAGE. K-Dmannosidases puri¢ed in the presence (lane 1) or the absence (lane 2) of proteinase inhibitors were analyzed by SDS-PAGE under either reduced (a) or non-reduced (b) conditions. Lane St, molecular weight standard.
nosidase activity toward either PNP-K-D-mannoside (Fig. 3a) nor oligosaccharide (Man8 ) substrate (data not shown). The activity of K-D-mannosidase in Fraction I was more than 3 times as high as that of MAN2B2 even at pH 7.0 (Fig. 3a). Further puri¢cation of the K-D-mannosidase in Fraction I was performed by Red-Amicon and FPLC-Superdex 200 chromatographies. In the two
Fig. 7. N-terminal amino acid sequences of the 63 (a) and 51 (b) kDa subunits of puri¢ed porcine K-D-mannosidase. N-terminal amino acid sequences of the subunits were compared with the corresponding regions of the human (U60266), bovine (L31373) and feline (AF010191) lysosomal K-D-mannosidases.
387
Fig. 8. Substrate speci¢city of the puri¢ed K-D-mannosidase. The puri¢ed enzyme (1 Wg protein) was incubated with various PNP-glycosides (0.625 mM) in the reaction mixture (pH 4.6) at 37³C for 15 min. Released p-nitrophenol from substrates was determined from the absorbance at 400 nm. Glycosidase activities for each substrate were expressed as % of activity for PNP-K-D-mannoside.
steps of puri¢cation, the enzyme protein was successfully separated from other epididymal proteins (Fig. 4a and b). Finally, 0.013% of protein and 16% of the activity were recovered from the original cauda epididymal £uid (Table 1). The molecular mass of the native enzyme estimated by FPLC-Superdex 200 gel-¢ltration was 275 kDa (Fig. 4b). Fig. 5 shows the SDS-PAGE patterns of the enzyme fractions during the puri¢cation. The puri¢ed enzyme protein which was eluted as a single peak from FPLC-Superdex 200 column showed two bands at 63 and 51 kDa on the SDS-PAGE (Fig. 5, lane 4) under the reduced conditions. The ratio of density of the two bands on SDS-PAGE stained with Coomassie Brilliant Blue R-250 was 1.2:1, suggesting that those are present in equimolar amounts. In order to exclude possibilities that the ¢nal fraction contains two di¡erent proteins or non-speci¢cally degraded fragments, all the puri¢cation steps were performed in the presence of various proteinase inhibitors: 1 mM PMSF, EDTA, benzamidine and 1 Wg/ml of leupeptin, pepstatin, phosphoramidon, chymostatin, aprotinin and elastatin. As shown in Fig. 6, the puri¢ed enzyme also contained two bands on SDS-PAGE under the reduced conditions. Furthermore, both enzymes puri¢ed in the absence and the presence of proteinase inhibitors formed a single band at 260 kDa on SDS-PAGE under the non-reduced conditions. As shown in Fig. 7, N-terminal amino acid se-
BBAPRO 35941 5-7-99
388
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
3.3. Properties of the puri¢ed K-D-mannosidase Substrate speci¢city of the puri¢ed enzyme was shown in Fig. 8. The puri¢ed enzyme releases p-nitrophenol speci¢cally from PNP-K-D-mannoside. The optimal pH to hydrolyze both PNP-K-D-mannoside and high-mannose oligosaccharide (Man8 ) was in the acidic pH range, 3.5^4.0 (Fig. 9). As shown in Fig. 10, the intermediate digests of
Fig. 9. E¡ect of pH on the K-D-mannosidase activity. The puri¢ed K-D-mannosidase equivalent to 1 Wg protein was incubated with 0.625 mM PNP-K-D-mannoside (a) or 0.4 WM Man8 (b) in 50 Wl of 50 mM sodium phosphate bu¡er at various pHs at 37³C for 20 (PNP) or 45 (Man8 ) min. Enzyme activities were determined as in Figs. 8 and 10.
quence of the 63 kDa subunit was found to be 60%, 80% and 80% identical with the corresponding residues deduced from the cDNAs coding human (U68567, U05572, U60266) [20^22], bovine (L31373) [23] and feline (AF010191) [24] lysosomal K-D-mannosidases, respectively. Furthermore, N-terminal sequence of the 51 kDa subunit was 55%, 80% and 85% identical with the corresponding residues of human, bovine and feline lysosomal K-D-mannosidases, respectively.
Fig. 10. Separation and identi¢cation of oligosaccharides digested by K-D-mannosidase in the porcine epididymal £uid. 0.4 WM Man9 (a), Man8 (b) or Man5 (c) were incubated with the puri¢ed K-D-mannosidase (5 Wg protein) at 37³C under various conditions: Man9 , at pH 4.6 for 5 min; Man8 , at pH 7.0 for 2 h; Man5 , at pH 4.6 for 20 min, respectively. Resulting digest was separated by APS-2 Hypersil column as described in Section 2. Representative elution pro¢les are shown. K-D-mannosidase activity was determined by the reduction of the peak area of substrate.
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
389
4. Discussion
Fig. 11. Digestion of Man9 and Man5 by the puri¢ed K-D-mannosidase. 0.4 WM Man9 (a) or Man5 (O) was incubated with the puri¢ed enzyme (5 Wg protein) either at pH 4.6 (a) or pH 7.0 (b) for various intervals of time at 37³C. Resulting digest was separated by APS-2 Hypersil column. Enzyme activity was determined as in Fig. 10.
Man9 , Man8 and Man5 produced by the puri¢ed enzyme were separated by an APS-2 Hypersil column and the K-D-mannosidase activity was determined by the reduction of the peak area of the substrate as described in Section 2. At pH 4.6, Man9 was completely hydrolyzed to Man8 VMan1 within 20 min, as shown in Fig. 11a. However, it took two hours to hydrolyze Man5 completely. Similar results were obtained at pH 7.0 (Fig. 11b), although the rate of the hydrolysis was much slower than at acidic pH.
Several studies have demonstrated that changes in the carbohydrate moiety occur on the sperm plasma membrane during epididymal transit [25^27]. Although the carbohydrate composition of the sperm surface is not fully elucidated yet, high contents of mannosyl residues are observed in N linked oligosaccharide core and terminus of the various glycoproteins. Therefore, those residues are possibly the target of K-D-mannosidase in the modi¢cation of the sperm surface. We have recently cloned a novel 135 kDa K-Dmannosidase (MAN2B2) which has optimal pH at 6.5 and is not active at acidic pH [17^19]. In addition to this enzyme, we found another K-D-mannosidase in the epididymal £uid, whose activity is much higher than MAN2B2 both at acidic and neutral pHs. In the present study, we could purify the K-D-mannosidase in only three chromatographic steps, starting with the luminal £uid of porcine epididymis. The puri¢ed enzyme consists of 63 kDa and 51 kDa subunits at equimolar amounts. As shown in Fig. 6, even when the enzyme was puri¢ed in the presence of various proteinase inhibitors, composition of subunits was just the same as the enzyme puri¢ed in the absence of inhibitors. Furthermore, under the non-reduced condition, the enzymes puri¢ed both in the absence and the presence of the proteinase inhibitors were moved as a single band at 260 kDa on SDS-PAGE. It is unlikely that non-speci¢c degradation of the enzyme occurred to produce two fragments during the puri¢cation procedures. The puri¢ed enzyme was also shown to have an apparent molecular mass of 275 kDa by molecular sieve chromatography. This value corresponds to the apparent molecular weights for the lysosomal K-Dmannosidase of rat liver (335 000 by molecular sieve chromatography and 200 000 by sucrose density centrifugation) [28] or epididymis (220 000 by molecular sieve chromatography) [12]. These results indicate that the puri¢ed enzyme is composed of tetramer: (63 kDa subunit)2 c(51 kDa subunit)2 . As shown in Fig. 7, N-terminal sequences of both 63 and 51 kDa subunits had very high homology with the corresponding residues of the human, bovine and feline lysosomal K-D-mannosidases [20^24]. All these data together suggest that the K-D-manno-
BBAPRO 35941 5-7-99
390
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
sidase puri¢ed from the porcine epididymal £uid is a homologue of the lysosomal enzyme. It is also indicated that the precursor of the puri¢ed enzyme is cleaved at two sites, releasing a signal peptide and producing 63 and 51 kDa subunits. The N-terminus of the 63 kDa subunit is in good agreement with the cleavage site of the signal peptide of several lysosomal K-D-mannosidases [20^24]. On the other hand, the N-terminus of the 51 kDa subunit is located ¢ve residues before the position of the potential cleavage site as predicted from human and bovine lysosomal K-D-mannosidases [20^23]. Emiliani et al. [29] have puri¢ed human lysosomal K-D-mannosidase, but the ¢nal fraction contained 67, 59, 55 and 30 kDa fragments on SDS-PAGE, showing a high microheterogeneity of the enzyme. Among those fragments, the partial amino acid sequence of the 30 kDa fragment was found to correspond to the amino acid sequence predicted by human lysosomal K-D-mannosidase cDNA (U05572) [21]. In addition, Nilssen et al. [22] have suggested that the human enzyme is synthesized as a single-chain precursor which is processed into three glycopeptides of 70, 42 and 15 kDa and that the 70 kDa peptide is further partially proteolysed into three more peptides. The bovine and feline mature enzymes have been also suggested to contain ¢ve and three peptides, respectively [23,24]. The reason for the discrepancy in the subunit composition between the enzyme puri¢ed from the porcine epididymal £uid and other lysosomal K-D-mannosidases is not clear. It is interesting to know whether the di¡erent subunit composition causes the secretion of the enzyme into the epididymal lumen instead being targeted to lysosome. It is possible that during the puri¢cation processes some lysosomal proteinases digested lysosomal K-D-mannosidases. On the other hand, it has been known that epididymal £uid contains proteinase inhibitors, which possibly protect against the degradation of the native form of the K-D-mannosidase. It thus seems that epididymal £uid is a very good source for obtaining the lysosomal K-D-mannosidase homologue without producing microheterogeneity in the enzyme preparation. The puri¢ed K-D-mannosidase cleaved both PNPK-D-mannoside and high-mannose oligosaccharides most actively at pH 4.0. At pH 3.0, the enzyme cleaved little oligosaccharide while signi¢cant
amounts of PNP-K-D-mannoside were degraded by it. On the other hand, the enzyme cleaved very little PNP substrate but was e¤cient in cleaving oligosaccharide substrate at the neutral pH. The exact reason for these di¡erences has not been known, but there have been several reports of di¡erences in pH dependency when PNP substrates have been compared with natural one [12]. pH of the epididymal £uid has been reported to be around 5.8V7.2, which is lower than serum pH [30^ 32]. We found that pH of the porcine epididymal £uid gradually declined from caput to cauda; 6.94 þ 0.16 in caput, 6.84 þ 0.19 in corpus and 6.73 þ 0.17 in cauda. These results suggest that the puri¢ed enzyme can be physiologically active toward natural substrate in the epididymal £uid. We determined the substrate speci¢city of the puri¢ed enzyme using the high-mannose oligosaccharides Man9 and Man5 as substrates. The puri¢ed enzyme hydrolyzed both substrates both at acidic and neutral pHs. Such rather broad substrate speci¢city is a characteristic for the lysosomal K-D-mannosidase [20,33]. But the rate of K1-3 and K1-6 cleavage in Man5 was much slower than the rate of K1-2 cleavage in Man9 . Tulsiani et al. [12,14^16] have extensively studied the properties of the rat epididymal K-D-mannosidases. They puri¢ed an acidic K-D-mannosidase from the acid extract of rat epididymis and identi¢ed it as a lysosomal K-D-mannosidase [12]. Rat epididymal K-D-mannosidase contains two heavy chains (60 kDa) and two light chains (31 kDa), which has similar subunit composition to the enzyme puri¢ed from the porcine epididymal £uid. But the substrate speci¢cities of rat and porcine enzymes were slightly di¡erent. It has been shown that rat lysosomal K-D-mannosidase hydrolyzes high-mannose oligosaccharides (Man9ÿ5 ) and the branched trimannosyl oligosaccharide (Man(K1-6)[Man(K1-3)]Man(L1-4)GlcNAc) very slowly [12]. On the other hand, the lysosomal type K-D-mannosidase secreted into the porcine epididymal £uid quickly hydrolyzed K1-2 mannose residues in the high-mannose oligosaccharides. In contrast, the mannosidase hydrolyzed the branched oligosaccharide (Man5 and Man3 ) at a slower rate, ¢nally yielding pyridylamino-Man(L1-4)GlcNAc(L1-4)GlcNAc(Man1 ). It is suggested that the secretory type of lysosomal K-D-mannosidase is
BBAPRO 35941 5-7-99
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
processed di¡erently from the lysosome-targeted enzyme and has a di¡erent substrate speci¢city. As shown in Fig. 3, the activity of lysosomal type K-D-mannosidase (Fraction I) toward PNP substrate is higher than that of MAN2B2 (Fraction II) both at acidic and neutral pHs. Furthermore, the activity of the puri¢ed lysosomal type enzyme toward oligosaccharide substrate at neutral pH (Fig. 11) was also con¢rmed to be higher than that of MAN2B2 determined previously [18]. Though MAN2B2 cleaves K1-2 bond but neither K1-3 nor K1-6 bond [18], the puri¢ed lysosomal type enzyme can hydrolyze all. We could not detect any other K-D-mannosidase activities in the epididymal £uid, though several types of the K-D-mannosidases have been found in the mammalian epididymis [13^16]. The present results suggest that lysosomal type K-D-mannosidase is responsible for the major part of K-D-mannosidase activity which can hydrolyze K1-2, K1-3 and K1-6 bonds in the luminal £uid of porcine epididymis. K-D-mannosidase activity, as well as those of other glycosidases, increases in distal caput and proximal corpus region as shown in Fig. 2. There is growing evidence indicating that distal caput is a unique part where sperm ¢rst acquire their motile and fertilizing abilities though not perfect [34]. Recently, we have found that procathepsin L, which is a lysosomal proteinase, is expressed in distal caput and is secreted into the extracellular £uid instead being targeted to lysosome, and this process may be regulated by platelet-derived growth factor [35]. These results indicate that, although the overall function of epididymis is principally regulated by androgens, there are other additional mechanisms regulating the locally restricted function of epididymis. The expression and secretion mechanisms of the lysosomal type K-Dmannosidase in the porcine epididymis need to be elucidated. This is the ¢rst report that the secretory type of the lysosomal K-D-mannosidase was puri¢ed from the luminal £uid of porcine epididymis. The present results suggest that this K-D-mannosidase contributes to major part of K-D-mannosidase activity in the luminal £uid of epididymis, which may be closely related to the glycoprotein modi¢cation on the surface of maturating sperm.
391
Acknowledgements This work was supported in part by grants from the Ministry of Education (05044121 and 10680682) and by a grant from University of Tsukuba Project Research to N.O., and partly by Tsukuba Advanced Research Alliance (TARA Sakabe-Shoun project).
References [1] M.C. Orgebin-Crist, in: D.M. deKretser, G. Spera (Eds.), Morphological Basis of Human Reproductive Function, Acta Medica, Plenum Press, New York, 1987, pp. 155^174. [2] J.C. Hall, G.J. Killian, Biol. Reprod. 36 (1987) 709^718. [3] R. Yanagimachi, in: E. Knobil, J. Neill (Eds.), The Physiology of Reproduction, Vol. 1, Raven Press, New York, 1988, pp. 135^186. [4] R. Jones, in: S.R. Milligan (Eds.), Oxford Reviews of Reproduction Biology, Vol. 2, Clarendon, Oxford, 1989, pp. 285^337. [5] A. Bernal, J. Torres, A. Reyes, A. Rosado, Biol. Reprod. 23 (1980) 290^293. [6] D.W. Hamilton, Biol. Reprod. 23 (1980) 377^385. [7] B. Tadolini, T.J. Wilson, P.R. Reddy, H.G. Williams-Ashman, Adv. Enzyme Regul. 15 (1976) 319^336. [8] M.D. Skudlarek, M.C. Orgebin-Crist, Biol. Reprod. 35 (1986) 167^178. [9] M.D. Skudlarek, D.R. Tulsiani, M.C. Orgebin-Crist, Biochem. J. 286 (1992) 907^914. [10] D.R. Tulsiani, M.D. Skudlarek, M.K. Holland, M.C. Orgebin- Crist, Biol. Reprod. 48 (1993) 417^428. [11] S.M. Snaith, G.A. Levvy, Biochem. J. 114 (1969) 25^33. [12] D.R. Tulsiani, V.D. Coleman, O. Touster, Arch. Biochem. Biophys. 267 (1988) 60^68. [13] P. Dutta, G.C. Majumder, Biochem. J. 218 (1984) 489^494. [14] M.D. Skudlarek, M.C. Orgebin-Crist, D.R. Tulsiani, Biochem. J. 277 (1991) 213^221. [15] D.R. Tulsiani, M.D. Skudlarek, M.C. Orgebin-Crist, J. Cell Biol. 109 (1989) 1257^1267. [16] D.R. Tulsiani, M.D. Skudlarek, S.K. Nagdas, M.C. Orgebin-Crist, Biochem. J. 290 (1993) 427^436. [17] N. Okamura, F. Dacheux, A. Venien, S. Onoe, J.C. Huet, J.L. Dacheux, Biol. Reprod. 6 (1992) 1040^1052. [18] N. Okamura, M. Tamba, H.J. Liao, S. Onoe, Y. Sugita, F. Dacheux, J.L. Dacheux, Mol. Reprod. Dev. 42 (1995) 141^ 148. [19] K. Ohata, N. Okamura, M. Kojima, H. Yasue, Mamm. Genome 8 (1997) 158^159. [20] Y.F. Liao, A. Lal, K.W. Moremen, J. Biol. Chem. 271 (1996) 28348^28358. [21] V.L. Nebes, M.C. Schmidt, Biochem. Biophys. Res. Commun. 200 (1994) 239^245.
BBAPRO 35941 5-7-99
392
Y.-Z. Jin et al. / Biochimica et Biophysica Acta 1432 (1999) 382^392
[22] O. Nilssen, T. Berg, H.M.F. Riise, U. Ramachandran, G. Evjen, G.M. Hansen, D. Malm, L. Tranebjarg, O.K. Tollersrud, Human Mol. Genet. 6 (1997) 717^726. [23] O.K. Tollersrud, T. Berg, P. Healy, G. Evjen, U. Ramachandran, O. Nilssen, Eur. J. Biochem. 246 (1997) 410^419. [24] T. Berg, O.K. Tollersrud, S.U. Walkley, D. Siegel, O. Nilssen, Biochem. J. 328 (1997) 863^870. [25] R.H. Hammerstedt, A.D. Keith, S. Hay, N. Deluca, R.P. Amann, Arch. Biochem. Biophys. 196 (1979) 7^12. [26] J.L. Courtens, S. Fournier-Delpech, J. Ultrastruct. Res. 68 (1979) 136^148. [27] G.L. Nicolson, N. Usui, R. Yanagimachi, H. Yanagimachi, J.R. Smith, J. Cell Biol. 74 (1977) 950^962. [28] D.J. Opheim, O. Touster, J. Biol. Chem. 253 (1978) 1017^ 1023.
[29] C. Emiliani, S. Martino, J.L. Stirling, B. Maras, A. Orlacchio, Biochem. J. 305 (1995) 363^366. [30] D.W. Carr, M.C. Usselman, T.S. Acott, Biol. Reprod. 33 (1985) 588^595. [31] C.R. Ca£isch, T.D. DuBose Jr., J. Toxicol. Environ. Health 32 (1991) 49^57. [32] N. Okamura, Y. Tajima, A. Soejima, H. Masuda, Y. Sugita, J. Biol. Chem. 260 (1985) 9699^9705. [33] S. al Daher, R. de Gasperi, P. Daniel, N. Hall, C.D. Warren, B. Winchester, Biochem. J. 277 (1991) 743^751. [34] J.L. Dacheux, M. Paquignon, Reprod. Nutr. Dev. 20 (1980) 1085^1099. [35] N. Okamura, M. Tamba, Y. Uchiyama, Y. Sugita, F. Dacheux, P. Syntin, J.L. Dacheux, Biochim. Biophys. Acta 1245 (1995) 221^226.
BBAPRO 35941 5-7-99