Biochimica et Biophysica Acta, 1i 19 (1992) 127-132 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/${15.00

! 27

BBAPRO 34105

Isolation and biochemical characterization of two isoforms of a boar sperm zona pellucida-binding protein Libia Sanz 1, Juan J. Calvete z.3, Wolfram Schiller 3 Karlheinz Mann 3 and Edda T/Spfer-Petersen ~ t Department of Dermatology, Androiogy Unit, Unil:ersityof Munich, Munich (Germany), 2 lnstituto de Quimica-F[sica C.S.I.C., Madrid (Spain) and J Max-Planck-histitut fiir Biochemie, Martinsried (Germany) (Received 22 April 1991 ) (Revised manuscript received 18 September 1991)

Key words: Boar sperm protein; Zona pellucida-bindingprotein; AWN-I; AWN-2

Protein-carbohydrate complementarity has been recognized as a general mechanism of gamete recognition and adhesion in the process of fertilization throughout the whole animal kingdom. It appears that carbohydrate-binding molecules on the anterior sperm head surface mediate the binding of the male gamete to certain glycoconjugates present on the egg's extracellular coat. Subtle differences in protein and carbohydrate conformation may confer to this interaction a species-specific character. The mechanism responsible for gamete recognition is, however, poorly understood. A step in its elucidation is the characterization of the complementary molecules on the egg and sperm surfaces. With this aim we report here the isolation and partial structural characterization of two isoforms of a zona pellucida-binding protein (which we call AWN-I and AWN-2) from boar spermatozoa, including partial sequence determination, assignment of disulphide bonds and identification of an N.terminal blocking group. AWN-1 and AWN-2 were isolated from acid extracts of washed ejaculated sperm and were present in seminal vesicle secretions, but absent in samples of epididymal fluid, suggesting a seminal vesicle origin for these sperm proteins. No analogous protein sequence could be found in the MIPS data bank, indicating that the AWN proteins may belong to a novel mammalian protein family involved in fertilization.

Introduction

Fertilization occurs through a multistep, speciesspecific process. The initial physiological events in the sequence of the highly complex interactions leading to the fusion of the male and female gamete are adhesion, recognition and binding of the spermatozoa to the egg surface (reviewed in Ref. 1). On the investing egg, binding is mediated by sperm receptors located in the zona peUucida, a thick glycoprotein network which surrounds all mammalian eggs [1]. The zona pellucida from mammalian species is composed of only 2-4 different glycoproteins (reviewed in Ref. 2). In mouse, the most studied model species in fertilization research, the zona pellucida glycoprotein 3 (ZP3) sc,wes

Correspondence: E. T6pfer-Petersen, Department of Dermatology, Andrology Unit, University of Munich, Frauenlobstr. 9/11, DMiinchen 2, Germany.

as the primary receptor for acrosome-intact sperm [3] through O-linked oligosaccharide chains containing an essential galactose residue at their non-reducing end [4]. The idea that complementary molecules mediate gamete interaction, first proposed by Lillie at the beginning of this century [5], is nowadays fully accepted: accumulating evidence suggests that carbohydratebinding proteins on the sperm plasma membrane surrounding the sperm head recognize complementary glycoconjugates on the oocyte's extracellular coat [6-8]. Bindin, a 24 kDa major protein located in the sea urchin sperm acrosome granule, which binds to a complementary egg vitelline membrane glycoprotein receptor, is the best characterized sperm receptor-binding protein [9,10]. Bindin specifically recognizes and binds (K a = 10 -8 M), sulfated fucose-containing polysaccharides found on proteoglycans of the egg surface [l 1]. Sulfate esters on the vitelline coat and arginine residues of bindin appear to be responsible for this interaction [12,13]. Bindin, although not an integral membrane

128 protein, associates firmly with the sperm acrosomal membrane allowing it to serve as an anchorage for the sperm at the egg surface [14]. in contrast to bindin, relatively little is known about sperm surface receptors for egg in other species. An understanding of the mechanisms underlaying speciesspecific receptor-ligand interactions requires, however, a detailed analysis of the structure-function relationship between the complementary molecules. Putative high- and low-molecular-weight sperm zona pellucidabinding proteins have been recognized in various species (reviewed in Ref. 6-8). Studies from different laboratories have recently shown that the pig's high molecular mass (53-55 kDa) zona-binding protein was identical with acrosin [15,16] and that this biological activity was associated with an N-terminal peptide of its heavy chain [17]. Recently we have described a method for the identification and isolation of zona pellucida-binding proteins from ejaculated boar sperm [18]. In the present work we report the isolation, partial sequence analysis and location of post-translational modifications of two isoforms of an 18 kDa boar sperm zona pellucida-binding protein. The interaction of the sperm protein with its homologous zona pellucida was inhibited by fucoidan, showing that the sperm protein possess carbohydrate-binding activity. Comparison of the peptide sequences obtained from the boar proteins with the known sequences of mammalian lectins did not exhibit any similarity. Thus, they may belong to a novel mammalian carbohydrate-binding protein family involved in fertilization.

iected from 40 tzm filters and solubilized in 0.2 M NH4HCO 3 pH 9.0 at 72°C for 20 min. Electroblotting onto nitrocellulose sheets was performed following Ref. 22. After blocking, the blots were incubated 2 h at 37°C with biotinylated zona pellucida prepared as described [15], washed, incubated 1 h at 37°C with streptavidin peroxidase (1 : 1000, v/v), washed with 20 mM Tris-HCl, 0.5 M NaCI (pH 7.4) and finally developed with this buffer containing 20% (v/v) methanol, 1 mg/ml 4-chloro-l-naphtol (BioRad) and 15/~1 H 2 0 2. For inhibition experiments the blots were first incubated in the presence of 1 m g / m l fucoidan. Amino acid and amino sugar analyses were done with a Biotronik LC 5001 analyzer after hydrolysis at l l0°C in 6 M HCI for 24 h and 4 M HCI for 4 h, respectively. Reduction of disulphide bridges was done for 2 min at 100°C adding to the protein solution (5 m g / m l in 100 mM Tris-HCl, 1 M guanidine/HCl, pH 7.8) 2-mercaptoethanol up to a final concentration of 1% (v/v). Subsequent alkylation of cysteine residues was performed with a 2 molar excess of 4-vinylpyridine over reducing agent, for 30 rain at room temperature. Peptides were sequenced either with a Beckman 890C sequencer or with an Aplied Biosystems 473A following the manufacturers instructions. Mass spectra were recorded with a MAT 900 mass spectrometer (Finnigan MAT, Bremen) equipped with S.I.M.S. ionization.

Materials and Methods

Results and Discussion

Ejaculated boar spermatozoa were collected, washed and extracted as previously described [19]. Seminal vesicle and epididymal fluids were generous gifts from Professor D. Cechova (Prague, Czechoslovakia) and Professor A.E. Friess (Bern, Switzerland), resp~:etively. Isolation of zona pellucida-binding proteins was done following the procedure used in Ref. 18. The last purification step was performed by reverse-phase HPLC on a Lichrospher RP-100 (Merck, Darmstadt, Germany) C18 column (25 × 0.4 cm, 5 t~m particle size) eluting at 1 ml/min with a linear stepwise gradient of 0.1% TFA in (A) water and (B) acetonitrile (isocratically (20% B) for 5 min, followed by 20-38% B in 15 min, 38-46% B in 32 min and 46-70% B in 24 min). SDS-gel electrophoresis was done according to Ref. 20. Zona pellucidae were isolated according to ReL 21. Briefly, zona-encased oocytes were recovered from frozen-thawed ovaries by passage through nylon screens of decreasing pore size (2000-80 tzm). Following homogenization of zona-encased oocytes with a teflon/glass homogenizer, zona-pellucidae were col-

Low-molecular-weight sperm carbohydrate-binding proteins, which interact with zona pellucida component(s), have been identified in a number of mammalian species (see Refs. 6-8 for review). However, no structural data of these proteins have been yet reported. We have developed both a specific assay for identification of zona pellucida-binding proteins [15] and a purification scheme which yields several hundreds of micrograms of pure zona pellucida-binding proteins per 100 mg of total acid-extracted sperm proteins [18]. Fig. 1 shows the final reverse-phase HPLC purification step. When the individual proteins were tested for their ability to bind biotinylated pig zona pellucida [15] we found that all but peak 2 had this biological activity. The structural characterization of the proteins in peaks 1, 2 and 3, which we call 'the AQN family', since all three members of this family contain the N-terminal amino acid sequence Ala-GlnAsn [18], will be described elsewhere [23]. Here we report the isolation and structural characterization of the other two major boar sperm proteins isolated in peaks 4 and 5 (Fig. 1).

129 I

~

e "'~

0.3

q,

0.2

acid extracts of washed ejaculated sperm and seminal vesicle fluid, but were absent from epididymal extracts. This suggests that the 18 kDa boar proteins are actually secretory components of the seminal vesicle which are coated to the sperm surface at the time of ejaculation.

30 20

10

20

30

~0

50

60

min

Fig. 1. Reverse-phase HPLC isolation of low-molecular-weight boar sperm proteins. Acid-extracted boar spermatozoa proteins were fractionated by size-exclusion chromatography [18] and the low-molecular-weight zona pellucida-binding protein fractions were collected and further purified on a Lichrospher C-18 column as described under Materials and Methods.

Isolation and biological origin of the two zona pellucidabinding proteins SDS-polyacrylamide gel electrophoresis of the isolated proteins in peaks 4 and 5 showed that both migrate with the same apparent molecular size of 18 kDa (Fig. 2). The electrophoretic mobility of neither of them changed after disulphide bridge reduction, indicating that each protein consisted of a single polypeptide chain. Both proteins bound solubilized biotinylated zona pellucida in blotting analysis and this interaction was inhibited by both fucoidan (Fig. 2) and non-labeled zona pellucida. Moreover, the individual proteins inhibited the attachment of capacitated boar spermatozoa to their homologous oocyte in an in vitro assay and were located by immunofluorescence microscopy on the anterior sperm head (Sanz et al., in preparation). Similar results have been recently reported by Hanqing et al. [24]. Altogether, this suggests that the 18 kDa boar sperm proteins may mediate the binding of sperm to component(s) of the zona pellucida through a carbohydrate-binding mechanism. It has been well established that the composition of the plasma membrane of mammalian spermatozoa changes during maturation in the epididymis, after mixing with seminal vesicle secretion components at ejaculation, and during capacitation in the female genital tract (reviewed in Ref. 25). Some of these differences are due to uptake, modification, or loss, of whole (glyco)protein molecules. To establish the biological origin of the boar proteins, the possible presence of the 18 kDa proteins in seminal vesicle and epididymal fluids was investigated by HPLC separation of the components of both fluids, followed by electrophoresis, eleetroblotting, zona pellucida-binding analysis and amino acid and N-terminal sequence analyses of the isolated components (data not shown). The 18 kDa zona pellucida-binding proteins were found in both

Biochemical characterization of the two zona pellucidabinding glycoproteins Amino acid and amino sugar analyses (Table I) provided the first evidence that both proteins are actually very similar, or identical molecules. Their amino acid compositions are nearly identical. Both proteins are proline-rich and do not contain any methionine residue. They contain some glucosamine as amino sugar, indicating that only N-linked oligosaccharide chain(s) are to be expected. Neither glycoprotein contained free sulphydryl groups, but four ethylpyridyl-cysteine residues were recovered after reduction of disulphide bonds and reaction with vinylpyridine, indicating the presence of two disulphide bonds.

94 67 43 30 21

I

14

?

:? !

j

a

b

~ .....

,/ .....

.):,'

c

:,d:J~ '

de

Fig. 2. 15-25% polyacrylamide-SDS gel electrophoresis of the purified proteins in Fig. i. Lanes b and c correspond to the proteins in peaks 4 and 5, respectively. Lanes d and e show the binding of biotinylated pig zona pellucid~, to the protein in lane b in the absence and presence of fucoidan, respectively. The same result was obtained with the protein in lane c. Lane a, molecular weight markers, from top to bottom: phosphorylase b, bovine serum albumin, ovoalbumin, carbonic anhydrase, soybean trypsin inhibitor and lysozyme. The numbers on the left represent molecular mass in kDa.

130 TABLE I

Amino acid and amino sugar composition of AWN-I and AWN-2 (peaks 4 and 5 front Fig. 1, respecticely) * Determined as the ethylpyridyl derivative of cysteine. N.D., not detected. Residue ASX THR SER GLX PRO GLY ALA CYS VAL MET ILE LEU TYR PHE LYS HIS ARG GALN GLCN

Mol/100 moi amino acids AWN- 1

AWN-2

6.7 5,0 6,5 6.1 12.1 10,7 5,4 2.9 5.1 N.D. 8.6 6.6 3.8 5.1 6.1 2.2 6.4 N.D. 1.8

6.8 4.6 5.8 6.4 11.6 9.8 5.2 3.0 5.2 N.D. 8.1 6.4 3.4 5.4 5.9 1.8 5.9 N.D. 1.9

preted as indicating that the presence of the N-terminal blocking group did not allow chymotrypsin to cleave the zW-N3 bond. This would totally explain the observed differences in the chymotryptic maps of both glycoproteins. Since the only difference between peak 4 and peak 5 proteins (Fig. 1) seems to be the presence of an N-terminal blocking group, they can be considered as isoforms, which we will call AWN-1 and AWN2, respectively.

Mass spectrometric characterization of the N-terminal blocking group of A WN-2 To characterize the nature of the N-terminal blocking group of AWN-2,-peptide C'-13 was analyzed by mass spectrometry and a M + H += 2639.3 was obtained. This value corresponded to the peptide 1AWNRRSRSCGGVI2 disulphide bridged to 22NSDGPQKDCVW 32 (M + H + calculated = 2596.2) plus 43 a.m.u. This result indicated that the blocking group may be an acetyl group and provided further confirmation of the existence of a disulphide bridge between cysteines 9 and 30. Partial structural characterization PRAWN To gain further information about the structure of the AWN isoforms, AWN-1 was cleaved with trypsin

N-terminal sequence analysis yielded the first 25 residues of the protein in peak 4: AWNRRSRSCGGVLRDPPGKIFNS DG... The protein in peak 5 had a blocked N-terminus.

°'3]11

0.2

The two glycoproteins are isoforms When the proteins were cleaved with chymotrypsin and the peptide maps compared, it was evident that they were almost identical (Fig. 3). Two peptides (C-3 and C-7] of peak 4 were absent in the corresponding chromatogram of the peak 5 digest, while a new component (C'-13) was found in the latter. Amino terminal sequence analysis of the eo-eluting chymotryptic peptides from each glycoprotein showed them to have identical sequences (Table II). This clearly indicated that the major structure of both proteins was identical. To define the structural difference(s) between them, analysis of the unique peaks in each chymotryptic digest was performed. N-terminal sequence analysis of C-3 and C-7 showed AW and NRRSRSXGGVL + NSDGPQKDXVW (in a 1:1 molar ratio), respectively (Table II). They align with positions 1-2, 3-13 and 22-32, respectively, in the N-terminal sequence of the intact protein. The two sequences in peptide C-7 suggested the existence of a disulphide bridge within the N-terminal part of the protein. The amino terminal sequence of C'-13 was NSDGPQKDXV. The fact that peptide C'-13 contained a single amino-terminal sequence was inter-

3

~

Ill Ii ~

1

. . ....

8 I0

7

70 50

. ' " ""

11"12

30 "7" i I

10

30

50

70

90 fmin]

rO 0

~c 0.3,

C'-

tj

70 ~c

11

p

50

0.2, 13

30 tO

10

30

50

70

90 Imin)

Fig. 3. Comparison by reverse-phase HPLC of the chymotryptic peptide maps obtained after digestion of the protein in peak 4 (upper panel) and the protein in peak 5 (lower panel) of Fig. l. The arrows indicate the position of those peptide peaks unique to each protein. Amino terminal sequences of peptides (2-1 to C12 (and C'-13) are in Table iI.

131 TABLE It N-terminal amino acid sequotces of the tryptic (T-), chymotryptic (C-) attd VS-dericed (V-) peptides obtained from A WN-! and A WN-2 (Figs. 3 and 4) Pep- N-terminal sequence tide T1 T2 T3 T4 T$ T6 T7 C1 C2 C3 C4 C5 C6 C7 C8 C9 £10 Cll C12 C13' V1

R Q T I I A TE I F N S D G PQK YY A S P F L I P L P F P YADPEG Y Y Y A A S P F L I VKPHFH V V L A I VLR SRSXGG P Q K D I F N S D G T I K V K P HF RAS LRTSGQ AM R S S S N I GPPGS FHVVLA I XGGVL NRRSRS KDXVW NSDGPQ L P F P Y ADPEGP G P L P F YYADPE P L P F P DADPEG DGPPGS E I I SXGKEY P P G S VELLDG SXGKEY N S O G P Q KDXV EK Q T I I A T

Y F E D P E G P L P F P Y F E

X V W T I

K V K P H

F P Y Y

E

I I G K

I X G G

I S L

and with V8 proteinase, and the proteolytic fragments were separated by reverse-phase HPLC. Fig. 4 shows a typical separation of tryptic peptides from AWN-1. Cleavage with V8 (chromatogram not shown) was quite ineffective and only two peptides were recovered. Table II shows the N-terminal sequences obtained. Peptide T-7 overlaps with peptide T-6 defining, thus, the first 46 residues of the glycoprotein. Peptide C-10 contained aspartic acid at the position where tyrosine was found in T-4, T-5 and C-9. Therefore, the poss~ility that AWN-1 and AWN-2, are polymorphic proteins can not be ruled out at this stage.

Peptide C-I1 is a further chymotryptic degradation product of C-12. Both contained two sequences in a 1:1 molar ratio, suggesting the presence of a second disulphide bridge. This point was further confirmed by mass spectrometry. For peptide C-12 a protonated molecular ion (M + H +) of 2840.4 was found. This established that the disulphide bridge pattern in AWN1 (and AWN-2) is between nearest-neighbour cysteines.

Comparison cf the AWN-1 peptide sequences with animal carbohydrate-binding proteins The relative position of the AWN-! peptides within the primary structure of AWN-1 was established by sequence comparison with the amino acid sequences of AQN-1 and AQN-3, two members of the AQN-family of boar sperm zona pellucida-binding proteins [18], which are structurally and functionally related to A W N - I / 2 and whose primary structure have been completed recently [23] (Sanz et al., in preparation). The amino acid sequences obtained for the AWN-1 peptides did not exhibit any significant analogy with the "known primary structures of the carbohydratebinding domains of other animal carbohydrate-binding proteins. This includes the soluble (S-), Ca2+-depen dent (C-), thiol-dependent (S-Lac) and the selectin families (Refs. 26-31, and references cited therein). This may indicate that the AWN proteins (together with the AQN proteins) belong to a novel protein family involved in the interaction between gametes through a carbohydrate-binding mechanism. Our data may help in the complete elucidation of the structure of the members of this protein family. Characterization of the carbohydrate-binding site within the structure of AWN-1/2, determination of its specificity and identification of the zona pellucida counter-receptor carrying the carbohydrate structure(s) recognized by AWN-1/2 are other challenges for future investigations directed to a better understanding of the molecular basis of gamete interaction. Acknowledgements The authors wish to thank Professor D. Cechova and Professor A.E. Friess for gener9us gift of seminal vesicle and epididymal fluids, respectively. The provision of a Max Planck Gesellschaft fellowship to JJC is gratefully acknowledged. This work was supported by the Deutsche Forschungsgemeinschafl (TO 114/1-2) (ET-P) and by grant BMFT 0318824A (WS).

References 10

30

50

70

9o ¢minJ

Fig. 4. Separation by reverse-phase HPLC of the tryptic peptides (T-) of AWN-I. The amino terminal sequences obtained for peptides T-1 to T-7 are in Table If.

1 Wassarman, P.M. (1987) Science 235, 553-560. 2 Wassarman, P.M. (1988) Annu. Rev. Biochem. 57, 415-442. 3 BleiL J.D. and Wassarman, P.M. (1980) Cell 20, 873-882.

132 4 Wassarman, P.M. (1989) in Carbohydrate Recognition in Cellular Function (Bock, G. and Harnett, S., eds.), Wiley & Sons, Chichester, U.K., pp. 135-155. 5 Lillie, F.R. (1913) Science 38, 524-528. 6 0 ' R a n d , M.G. (1988) Gamete Res. 19, 315-328. 7 Macek, M.B. and Shur, B.D. (1988) Gamete Res. 20, 93-109. 8 Miller, DJ. and Ax, R.L. (1990) Molecular Reproduction And Development 26, 184-198. 9 Vacquier, V.D. and Moy, G. (1977) Proc. Natl. Acad. Sci. USA 74, 2456-2460. 10 Minor, J.E., Gao, 13. and Davidson, E.H. (1989) in The Molecular Biology of Fertilization (Schatten, H. and Schatten, G., eds.), Academic Press, San Diego, pp. 73-88. 11 DeAngelis, P.L. and Glabe, C.G. (1987) J. Biol.Chem. 262, 13946-13952. 12 DeAngelis, P.L. and Glabe, C.G. (1988) Biochemistry 27, 81898194. 13 DeAngelis, P.L. and Glabe, C.G. (1990) Biochim. Biophys. Acta 1037, 100-105. 14 Kennedy, L., DeAngelis, P.L. and Glabe, C.G. (1989) Biochemistry 28, 9153-9158. 15 T6pfer-Petersen, E. and Henschen, A. (1988) Biol. Chem. Hoppe Zeyler 369, 69-75. 16 Jones, R., Brown, C.R and Lancaster, R.T. (1988) Development 102, 781-792.

17 T6pfer-Petersen, E., Steinberg, M., Ebner von Eschenbach, C. and Zucker, A. (1990) FEBS Len. 265, 51-54. 18 Jonakov~t, V., Sanz, L., Calvete, JJ., Henschen, A., C~chova, D. and T6pfer-Petersen, E. (1991) FEBS Lett. 280, 183-186. 19 C~chova, D., T6pfer-Petersen. E. and Henschen, A. (1988) FEBS Lett. 241,136-140. 20 Laemmli, U.K. (1970) Nature (London) 227, 680-685. 21 Dunbar, B.S., Bardrit, N.J. and Hedrick, J.L. (1980) Biochemistry 19, 356-365. 22 Towbin, H., Staehelin, T. and Gordon, J. (1979) Proc. Natl. Aead. Sci. USA 76, 4350-4354. 23 Sanz, L., Calvete, JJ., Mann, K., Sch~ifer, W., Schmid, E.R. and T6pfer-Petersen, E. (1991) FEBS Lett. 291, 33-36. 24 Hanqing, M., Tai-Ying, Y. and Zhao-Wen, S. (1991) Mol. Reprod. Dev. 28, 124-130. 25 Yanagimachi, R. (1988) in The Physiology of Reproduction (Knobil, E. and Neili, J., eds.), Raven Press, New York, pp. 135-185. 26 Clerch, L.B., Whitney, P., Hass, M., Brew, K., Miller, 1"., Werner, R. and Massaro, D. (1988) Biochemistry 27, 692-699. 27 Drickamer, K. (1988) J. Biol. Chem. 263, 9557-9560. 28 Hoyle, G.W. and Hill, R.L. (1991) J. Biol. Chem. 266, 1850-1857. 29 Springer, T.A. (1990) Nature 246, 425-433. 30 McEver, R.P. (1990) Blood Cells 16, 73-84. 31 Gitt, M.A. and Barondes, S.H. (1991) Biochemistry 30, 82-89.

Isolation and biochemical characterization of two isoforms of a boar sperm zona pellucida-binding protein.

Protein-carbohydrate complementarity has been recognized as a general mechanism of gamete recognition and adhesion in the process of fertilization thr...
559KB Sizes 0 Downloads 0 Views