BIOCHEMICAL AND FUNCTIONAL CHARACTERIZATION
OF ZONADHESIN: A SPERM PROTEIN POTENTIALLY
MEDIATING SPECIES-SPECIFIC SPERM-EGG
ADHESION DURING FERTILIZATION
by
MING BI, B.S., M.S.
A DISSERTATION
IN
MEDICAL BIOCHEMISTRY
Submitted to the Graduate Faculty of Texas Tech University Health Sciences Center Ln Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY
Advisory Committee
Daniel M. Hardy (Chairperson) Beverly S. ChUton Charles H. Faust S. Sridhara Simon C. Williams
Accepted
AssociateTDean of the Graduate School of Biomedical Sciences Texas Tech University Health Sciences Center
May, 2002 © 2002, Ming Bi ACKNOWLEDGMENTS
This major accomplishment in my life was made possible by numerous supports from many people who helped me in many different ways. First of all, I would like to thank my mentor Dr. Daniel Hardy. I have learned a lot from him in last five years. He taught me not only academic knowledge and lab techniques in my research field, but also the way to think wisely and to maintain a correct scientific attitude. In addition, his confidence based on his broad knowledge always made me feel comfortable and encouraged during my down time. Beside his great help in study and research, he also helped me to overcome the culture shock and showed great patience in dealing with my poor English.
My thanks are also due to all other members in my committee, Dr. Chilton, Dr.
Faust, Dr. Sridhara, and Dr. Williams. They gave me lots of valuable suggestions for my research project, and also helped me to keep on track of my study and research plan. I also want to express my special thanks to Dr. Faust. As a graduate advisor for Medical
Biochemistry program, he always explained clearly to me about the requirement and procedure on course selection and graduation plan. Furthermore, I would never forget his warm greet in the airport on my first day in Lubbock.
My acknowledgements also go to my colleagues in the lab, especially John
Hickox. John has helped me a lot in daily lab work, and he has also done significant work in zonadhesin research. I want to thank Tony Cheung. It was Tony who showed me around and taught me lab techniques when I first joined the lab. I also want to thank my other colleagues, Nagesh Uppuluri, Michael Wassler, Steve Tardif, and Joshua Games. They always lent me a hand any time when I needed. Furthermore, 1 would like to thank all other faculty members, staff, and graduate students in my department, who have helped me here and there.
I would like to acknowledge the great work from our collaborators. First, I want to thank Dr. Gary Olson and Virginia Winfrey, of Vanderbilt University, for their excellent work on ultrastmctural localization of zonadhesin on spermatozoa using immunoelectron microscopy. I also want to thank Dr. Scott Whisnant, Dr. Heidi Brady,
Dr. Andy Herring, Nancy Haden, and Kelly Breazeale of the Department of Animal
Science, Texas Tech University for kindly providing animal semen, and personnel in
Texas Tech University Farm for continuous supply of pig semen.
Finally, I would like to thank my family. It is their continuous support that made me get this far. I would first thank my parents. They have sacrificed so much to support me to complete a series of education goals, financially and spiritually. I will always feel regret that I could not make it to see my mother before she passed away. It was her final advice and inspiration that kept me to stick to my academic plan and to complete my goal today. I also owe a great gratitude to my wife, Xin. Without her continuous unselfish support, it is impossible for me to finish my goal. I really appreciate her hard work in taking care of our son and having a flill-time job at the same time. I would thank my lovely son, Kevin. Making him proud of me is the strongest motivation for me to do my best to complete my goal.
To Xin and Kevin.
ni TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
ABSTRACT x
LIST OF FIGURES xii
LIST OF ABBREVIATIONS xv
CHAPTER
I. BACKGROUD AND INTRODUCTION 1
1.1 An Inttoduction to Mammalian Fertilization 1
1.2 Species Specificity of Sperm-Egg Adhesion 4
1.3 Role of Zona Pellucida Glycoproteins in Sperm-Egg Adhesion 10
1.4 Controversial Data on Zona Pellucida-Binding Candidates in Spermatozoa 15
1.5 Preliminary Studies on Zonadhesin 20
1.6 Multimerization and Functions of VWD Domains in
vWF and Mucins 25
1.7 Purpose of Study 29
II. EXPRESSION OF ZONADHESIN FRAGMENTS IN CULTURED MAMMALIAN CELLS 35
2.1 Introduction 35
2.2 Materials and Methods 37
2.2.1 Constmction of pCMV-5/DO-Dl Plasmid 37
iv 2.2.2 Transient Expression of rDO-Dl Zonadhesin in COS-1 Cells 38
2.2.3 Purification of rDO-Dl Zonadhesin from COS-1 Cell Extract 39
2.2.4 Purification of rDO-Dl Zonadhesin from Culture Media 40
2.2.5 Constmction of Recombinant Protein Containing Zonadhesin Mucin Domain 40
2.2.6 Constmction of Recombinant Protein Containing von
Willebrand Factor Signal Peptide 41
2.2.7 Expression and Purification of the DO-Dl Fusion Protein 42
2.2.8 Preparafion of DO-Dl Antisera 42
2.2.9 Electrophoresis and Western Blotting 43
2.3 Results and Discussion 43 2.3.1 rDO-Dl Was Constmcted and Expressed in COS-1 Cultured Mammalian Cells 43
2.3.2 rDO-Dl Was Purified from Cell Extract and Culture Media by Ni^"^ Affinity Chromatography 45
2.3.3 Constmcts with Mucin Domain and Constmcts with von Willebrand Factor Signal Peptide Were Generated 46
III. HETEROGENEOUS PROCESSING AND ZONA PELLUCIDA- BINDING ACTIVITY OF PIG ZONADHESIN 55
3.1 Introduction 55
3.2 Materials and Methods 57
3.2.1 Isolafion of Sperm Membrane Fraction 57
3.2.2 Isolation of Zona Pellucida 58 3.2.3 Zona Pellucida-Binding Assays 59
3.2.4 Expression and Purification of DO-Dl Fusion Protein 60
3.2.5 Preparation of DO-Dl Antisera 61
3.2.6 Expression and Purification of Dl and D3 Fusion Proteins 61
3.2.7 Preparation of Domain-Specific Antisera 62
3.2.8 Preparation of GST, GST-Dl, and GST-D3 Affinity Columns 62
3.2.9 Affinity Purification of Dl and D3 Antibodies 63
3.2.10 Preparation of D3 Immunoaffmity Column 63
3.2.11 Immunoaffmity Purification of Zonadhesin from Spermatozoa 64
3.2.12 Preparation of Antisera to Zonadhesin Holoprotein 65
3.2.13 Site-Directed Mutagenesis 65
3.2.14 Electrophoresis and Western Blotting 66
3.2.15 In FzYro Multimerization 67
3.2.16 Indirect Immunofluorescence 68
3.3 Results and Discussion 68
3.3.1 Zonadhesin Binding Sites Are Evenly Distributed on the Entire Zona Pellucida 68
3.3.2 Antibodies Were Developed to Characterize Zonadhesin Iso forms 69
3.3.3 Multiple Zonadhesin Forms Are Present in Freshly Ejaculated Spermatozoa 72
VI 3.3.4 The pi 05/45 Monomeric Form of Zonadhesin Binds Preferentially to the Zona Pellucida 74
3.3.5 Cysteines Other than Those in the CG(L/V)CG Motif Can Mediate Spontaneous Multimerization of Purified Zonadhesin Fragments/« Vitro 75
3.3.6. Zonadhesin Localizes to the Apical Heads of Pig
Spermatozoa 77
3.3.7 Summary and Discussion 77 rv. TEMPORAL AND FUNCTIONAL CHARACTERIZATION
OF ZONADHESIN PROCESSING AND LOCALIZATION 89
4.1 Introduction 89
4.2 Materials and Methods 93
4.2.1 Development of Anti-N Antisemm 93
4.2.2 Deglycosylation 94
4.2.3 Collection of Spermatozoa from the Epididymis 95
4.2.4 Detergent Extraction of Spermatozoa and Testis Tissue 95
4.2.5 Zona Pellucida Binding Assay 96
4.2.6 Immunoprecipitation 97
4.2.7 Electrophoresis and Westem Blotting 98
4.2.8 Bead Adhesion Assay 98
4.2.9 Indirect Immunofluorescence 99
4.2.10 Immunoelectron Microscopy 99
4.3 Results and Discussion 101 Vll 4.3.1 p300 Is a Processed Zonadhesin Gene Product That Includes a MAM Domain 101
4.3.2 p300 Includes a Mucin-Like Domain That is Glycosylated with 0-Linked Oligosaccharides 103
4.3.3 Both p45 and pl05 Are Glycosylated with N-linked Oligosaccharides 105
4.3.4 Zonadhesin Is Present in at Least Two Physicochemically Distinct Compartments of Pig Spermatozoa 106
4.3.5 p300 Undergoes Significant Changes during Sperm
Maturation in the Epididymis, but p45 and p 105 Do Not 107
4.3.6 Oligomeric Changes of Zonadhesin Occur in the Epididymis 110
4.3.7 The Zona Pellucida-Binding Activity of Zonadhesin Increases in the Epididymis 112 4.3.8 Zonadhesin Localizes to the Outer Acrosomal Membrane and the Acrosomal Matrix by Immunoelectron Microscopy 113
4.3.9. Binding of Zonadhesin Is Sufficient to Support Adhesion of Cell-Sized Stmctures to the Zona Pellucida 115
4.3.10 Zonadhesin Is Processed by Hydrolysis of an Asp-Pro Bond 116
V. MARKED VARIATION IN THE BIOCHEMICAL PROPERTIES OF ZONADHESIN IN SPERMATOZOA FROM A VARIETY OF MAMMALS 135
5.1 Introduction 135
5.2 Materials and Methods 139
5.2.1 Collection of Spermatozoa 139
5.2.2 Differential Detergent Extraction of Spermatozoa 139
Vlll 5.2.3 Indirect Immunofluorescence 140
5.2.4 Immunoelectron Microscopy 140
5.3 Results and Discussion 141
5.3.1 Zonadhesin Proteins Vary Significantly among MammaHan Species 141
5.3.2 Zonadhesin Localizes to the Apical Head of Spermatozoa of All Tested Species 144
VI. SUMMARY AND DISCUSSION 152
6.1 Overview 152
6.2 How does Zonadhesin "See" the Zona Pellucida? 153
6.3 What Factor(s) Determine Zonadhesin's Species Specificity? 155
6.4 Future Studies 157
REFERENCES 160
IX ABSTRACT
Fertilization is an important and unique process in organisms that generate offspring by sexual reproduction. Several fines of evidence support the idea that glycoproteins in the egg zona pellucida (ZP) mediate species-specific sperm-egg adhesion. In contrast, studies on the complementary molecules in spermatozoa remain controversial. Some potential candidate adhesion molecules have been identified, but none of them is supported by unequivocal evidence. A mosaic protein called zonadhesin is one of them. Zonadhesin was first found in pig sperm membrane extracts based on its ability to bind to the native ZP in a species-specific manner. The zonadhesin cDNA sequence reveals that the protein comprises several extracellular domain types found in other molecules known to mediate cell-cell interactions. Orthologs of pig zonadhesin in three other species possess domain stmctures similar to pig zonadhesin's, but with variations. Our overall hypothesis is that zonadhesin mediates species-specific adhesion between spermatozoa and the egg ZP. In this dissertation, I report biochemical and functional characterization of pig zonadhesin. I observed that zonadhesin forms disulfide- bonded oligomers and multimers, but the pl05/p45 monomer displays a preference for binding to the ZP. Zonadhesin undergoes heterogeneous processing and is targeted to more than one physicochemical compartment in developing sperm cells. Post- translational processing generates at least three zonadhesin components that are glycosylated with different types of oligosaccharides. Zonadhesin's oligomeric status changes during maturation of spermatozoa in the epididymis. Immunoelectron microscopy demonstrated that zonadhesin localizes to the outer acrosomal membrane and the acrosomal matrix. Moreover, I identified zonadhesin protein in eight additional species by Westem blotting, immunofluorescence, or both. Zonadhesin from each of these species localizes to the apical sperm head overlying the acrosome, but the MrS of these proteins' polypeptide chains vary significantly. In addition, a bead adhesion assay was developed for future tests of zonadhesin's ability to bind to homologous or heterologous ZP. Finally, preliminary work on expression of zonadhesin domain(s) initiated fiiture loss- and gain-of-fiinction transgenic studies. Potential practical benefits derived from this work may include development of new contraceptives that ftinction by blocking gamete interactions and diagnostic methods for human fertility and farm animal fecundity.
XI LIST OF FIGURES
1.1 Morphology of the sperm apical head in interactions with the
mammalian egg extracellular matrix 31
1.2 Hypothetical molecular mechanisms for species specificity 32
1.3 Schematic comparison of zonadhesin in three species 33
1.4 Schematic comparison of VWD domain-containing proteins 34
2.1 A schemafic outline for constmction of rDO-Dl plasmid 49
2.2 Detection of expressed rDO-Dl protein in transfected COS-1
cell extracts and culture media 50
2.3 Purification of expressed rDO-Dl protein from cell extracts 51
2.4 Purification of rDO-Dl protein from culture media 52
2.5 Schemafic outlines of constmcts containing a part of mucin domain
and constmcts containing a von Willebrand factor signal peptide 53
2.6 Detecfion ofpv6H-Dl protein in cell extracts and culture media 54
3.1 Direct binding of zonadhesin to intact zona pellucida 79
3.2 Domain stmcture, polypeptide composition, and predicted
characteristics of the pig zonadhesin precursor 80
3.3 Production and characterization of zonadhesin antisera and antibodies 81
3.4 Identification of multiple zonadhesin forms in membrane fractions
of freshly ejaculated spermatozoa 83
3.5 Differential binding of zonadhesin formXlsl to the zona pellucida 84
3.6 Differential binding of zonadhesin multimers to the zona pellucida 85 3.7 Multimerization of Dl and D3 ftision proteins in vitro 86
3.8 Light microscopic localization of zonadhesin in pig epididymal spermatozoa with the DO-Dl antisera 88
4.1 Identification of zonadhesin polypeptides containing the N (MAM) domain 119
4.2 Immunoprecipitation of biotinylated proteins from sperm membrane
with zonadhesin antibodies and antisera 120
4.3 Enzymatic deglycosylation of p300 121
4.4 Enzymatic deglycosylation of p45 andpl05 122
4.5 Differential detergent extraction of denuded spermatozoa 124
4.6 Properties of zonadhesin in spermatozoa at different maturation states 125 4.7 Two-dimensional SDS-PAGE of differential detergent extracts from caput epididymal and ejaculated spermatozoa 126
4.8 Zona pellucida-binding activity of zonadhesin in spermatozoa at different maturation states 127
4.9 Immunolocalization of zonadhesin on pig spermatozoa with anti-holoprotein antisemm 128
4.10 Ultrastmctural localization of zonadhesin in pig ejaculated spermatozoa 129
4.11 Ultrastmctural localization of zonadhesin in pig spermatids 132
4.12 Bead adhesion assay for testing zonadhesin's ability to function as an adhesion molecule 133
4.13 Determination of zonadhesin proteolytic processing sites 134
Xlll 5.1 Detection of zonadhesin polypeptides in spermatozoa of eight mammals 146
5.2 Detection of zonadhesin holoprotein in spermatozoa from three ungulates and one lagomorph 147
5.3 Two-dimensional electrophoresis of zonadhesin in spermatozoa
from six eutherian mammals 148
5.4 Localization of zonadhesin by indirect immunofluorescence 149
5.5 Ultrastmctural localization of hamster and bovine zonadhesin 150
xiv LIST OF ABBREVIATIONS
cDNA - complementary deoxyribonucleic acid
DFP - diisopropyl fluorophosphate
Die - differential interference contrast
DMEM - Dulbecco's modification of Eagle's medium
DNA - deoxyribonucleic acid
EDTA - ethylenediaminetetraacetic acid
EEM - egg extracellular matrix
ER - endoplasmic reticulum
FBS - fetal bovine semm
GalTase - galactosyl transferase
GCA - guanylyl cyclase-A
GSH - reduced glutathione
GSSG - oxidized glutathione
GST - glutathione-S-transferase lEM - immunoelectron microscopy
IF - immunofluorescence
IPTG - isopropylthiogalactoside mRNA - messenger ribonucleic acid
XV PCR - polymerase chain reaction
PSM - porcine submaxillary mucin
SDS - sodium dodecylsulfate
SDS-PAGE - SDS polyacrylamide gel electrophoresis
TCA - trichloroacetic acid
VERL - vitelline envelope receptor for lysin
VWD - von Willebrand factor D domain vWF - von Willebrand factor
ZP - zona pellucida
XVI CHAPTER I
BACKGROUND AND INTRODUCTION
1.1 An Introduction to Mammalian Fertilization
Fertilization is a cmcial and unique event in the organisms that generate offspring by sexual reproduction. In fertilization, two haploid cells—one from the male parent and the other from the female parent—unite to form a diploid zygote that subsequently develops into an individual whose genetic makeup is different from either parent
(Yanagimachi, 1994). Although it has been studied for more than a century, fertilization remains one of the least understood fundamental biological processes. Before 1950, most fertilization research focused on invertebrates and non-mammalian vertebrates because collection of gametes in large quantity from those organisms is more practicable than from mammals (Yanagimachi, 1994). However, knowledge acquired from non- mammalian fertilization, as well as improvement in research techniques, has made it possible to obtain more and more information on mammalian fertilization.
Mammalian fertilization has been extensively studied in the last several decades.
Data from numerous studies indicate that fertilization in mammals is a complicated process consisting of multiple steps that involve a variety of molecules on both gametes.
Although some events of mammalian fertilization are similar to those in invertebrates and other cold-blooded species, the major difference lies in that mammaHan fertilization occurs in the female reproductive tract instead of extemally as occurs with invertebrates, amphibians and fishes. For those organisms that release both eggs and spermatozoa into an open aquatic environment, gametes themselves must employ a special mechanism to avoid generating hybrids, which are typically sterile. Thus, there is an obvious adaptive advantage to species-specific fertilization in these organisms. In mammals, however, physiognomy and mate discrimination prior to mating impose the major barrier to heterologous fertilization. Thus, it would seem unnecessary for species specificity of mammalian fertilization to be built into gametes themselves. Nevertheless, many studies of mammalian fertilization in vitro have demonstrated that this process is also relatively species-specific (Yanagimachi, 1994).
Spermatozoa from most mammalian species do not have the ability to fertilize the egg right after completion of spermatogenesis in testis. Instead, they must first travel through the epididymis, a long, convoluted tubule that connects the testis and the vas deferens. At the time spermatozoa leave the testis and enter the epididymis, they are motionless and incapable of progressive motility. After they reach the caudal epididymis, where the cells are stored before ejaculation, spermatozoa are capable of moving actively and are ready for ejaculation (Yanagimachi, 1994). However, sperm maturation does not stop here, as even in ejaculated semen spermatozoa remain unable to fertilize the egg.
After mammalian spermatozoa are ejaculated into the female, they must traverse the reproductive tract. At this time, they undergo a cmcial process called capacitation, wherein they acquire the capacity to fertilize the egg. It has been long known in the field of fertilizafion research that freshly ejaculated spermatozoa cannot fertilize the egg in vitro. Only after they have been incubated in the female reproductive tract or in female reproductive tract fluid in vitro for a certain period do spermatozoa gain the capacity to fertilize the egg in vitro. Chemically-defined media have been developed for sperm capacitation in some mammalian species. Interestingly, each species requires its own optimal solution for efficient capacitation. Although the molecular basis of capacitation remains unclear, most researchers believe that changes in the sperm membrane, such as removal or alterafion of a stabiHzer or protective coat to sensitize the membrane, account for acquisition of fertilization ability for spermatozoa (Yanagimachi, 1994).
Although millions of spermatozoa are ejaculated into the female reproductive tract, only a small fraction of spermatozoa, perhaps 100 cells, are finally able to make their way to the vicinity of the ovulated egg in the ampulla of the oviduct (Check et al,
1990). Before they directly interact with the egg, spermatozoa must first penetrate the cumulus mass formed by cumulus cells and their secreted matrix. They then adhere to the zona pellucida (ZP), the porous extracellular matrix surrounding the egg. Interaction with the ZP causes spermatozoa to undergo the acrosome reaction, an exocytotic process that results in fusion between the sperm plasma membrane and the outer acrosomal membrane
(Fig. 1.1a c6 b). The acrosome is a large secretory granule overlaid by the sperm plasma membrane on the sperm cell's apical head. It contains various hydrolytic enzymes. In sea urchin, only those spermatozoa that have undergone the acrosome reaction are able to adhere to the extracellular matrix outside the eggs. However, in mammals, even though some conflicting data were reported on the timing of adhesion events between sperm and egg, the notion most extensively accepted is that only acrosome-intact spermatozoa are capable of adhering to the ZP in most if not all mammalian species. However, this view may be incorrect, and some investigators are now quesfioning it (Gerton, 2002) (see also
Chapter III). After the acrosome reaction, hydrolytic enzymes stored in the acrosome are released and are thought to digest a pathway for sperm penetration through the ZP (Fig. lAc&d). When the acrosome-reacted sperm cell reaches the surface of the egg plasma membrane, its equatorial segment fiases with the egg's plasma membrane to deliver the sperm nucleus into the egg. At the same time, the egg undergoes a cortical reaction that results in modification of the ZP to prevent additional spermatozoa from fertilizing the egg. The nuclei from the male and female gametes are merged and eventually complete the whole fertihzation process (Yanagimachi, 1994).
1.2 Species Specificity of Sperm-Egg Adhesion
Gamete adhesion during fertilization largely resembles cell-cell interaction among somatic cells. However, as mentioned in the previous section, gamete adhesion is distinguished by its species specificity (Yanagimachi, 1994). Most interactions of somatic cells are largely similar in all animal species (for example, signaling between neurons). In contrast, gamete interactions might differ markedly even between closely related species.
Regardless of whether one examines non-mammals or mammals, the interaction between gametes in species ranging from marine invertebrates to man displays some degree of species specificity. Spermatozoa efficiently adhere to and fertilize eggs from the same species, but they either fail to fertilize or at least inefficiently fertilize eggs from other species, sometimes even if the egg is from a closely related species. Although mammalian fertilization is the focus in this discussion, knowledge obtained from non- mammalian species has helped to estabUsh a general understanding of species-specific gamete recognition and thereby provided an important precedent for studies of mammaHan fertihzation. After all, species specificity of gamete recognition is much more evident in many non-mammals than in mammals. For mammals, because the interaction between sperm and egg occurs internally, there is almost no nattiral way for heterologous mammalian gametes to meet with each other. Therefore, it was unknown whether mammalian fertilization occurred in a species-specific manner until in vitro fertilization became a reality. However, many non-mammalian species (e.g., sea urchin, fish, and xenopus) spawn their gametes in an open aquatic environment (sea, river, or lake) often overlapped by gametes from other species, hence spermatozoa are able to get access to heterologous eggs under natural conditions. However, it is rare to see a hybrid created by heterologous fertilization among those organisms. Therefore, a species- specific recognition mechanism has been built in to the gametes during evolution. Due to the convenience in collecting gametes and the straightforward requirement for species specificity, most important concepts and terms on species-specific gamete recognition come first from research on broadcast spawning species, especially sea urchins and abalones. Therefore, it is necessary to introduce some critical information on species specificity of gamete recognition achieved from non-mammals.
Despite endeavors from many investigators, studies on fertilization of various species have not yet yielded a complete view of the biochemistry subserving gamete interactions in any one species. However, they have established a number of important concepts that collectively provide a framework for ongoing research. Research from a broad range of species has revealed that the interaction between sperm and the acellular investment immediately outside the plasma membrane of the egg plays a critical role in species-specific recognition and adhesion between sperm and egg. This stmcture is called vitelline layer in sea urchins, vitelline envelope in anuran amphibians, chorion in fishes, and zona pellucida (ZP) in most mammals (Bi et al, 2002). hiterestingly, it has been found that the components from all of these acellular stmctures share a conserved protein domain called the ZP domain (Prasad et al, 2000). This ZP domain is present in the major glycoprotein components of egg investments from all vertebrate species examined so far. In addition to the homologous ZP domain, the acellular investments surrounding the egg from various species share another common characteristic—imposing the last barrier for spermatozoa before they reach egg surface. In addition, adhesion of spermatozoa to this extracellular stmcture exhibits the highest degree of species specificity among all fertilization events. Thus, a general term—egg extracellular matrix
(EEM) (Bi et al, 2002) has been suggested to define the extracellular investment outside the egg to which spermatozoa adhere in a species-specific manner.
Adhesion of spermatozoa to the EEM is conceptually analogous to adhesion of somatic cells to the extracellular matrix. Both procedures are mediated by interaction of cell surface proteins with complementary components of an acellular stmcture, except that gamete adhesion shows relative species specificity (Yanagimachi, 1981, 1994;
Vacquier, 1998). The complementary molecules in the vitelline layer and spermatozoa that are involved in species-specific adhesion of spermatozoa to the egg have been extensively studied in sea urchin. For a broadcast-spawning organism like the sea urchin, the divergence of only those few proteins mediating sperm-egg interaction could be enough to establish barriers to fertilization and hence avoid generating hybrids. An acrosomal protein named bindin has been identified as the primary adhesive protein, and it interacts in a species-specific maruier with its cognate ligand in the egg's vitelline layer
(Belief et al, 1977). Indeed, the sea urchin acrosome is composed almost entirely of bindin (Glabe and Vacquier, 1977). Analysis of bindin cDNA sequences from various sea urchin species revealed that this molecule varies significantly among species. All bindins contain a conserved central domain, while the sequence flanking this domain displays strong species specificity that is achieved by variation in the number and location of short repeating amino acid sequence elements in this region. Interestingly, bindin appears to be a unique protein for sea urchin, because its ortholog has never been found in sperm from other species (Vacquier, 1998). Thus, not all of organisms utilize the same molecule to accomphsh the species-specific interaction between sperm and egg.
Abalone is another popular organism in studying species specificity of fertilization. An abalone sperm protein called lysin was discovered to be responsible for mediating the sperm-egg adhesion by binding to a large fibrous glycoprotein in the egg envelopes called VERL (vitelline envelope receptor for lysin) (Lewis et al, 1982). Like bindin in sea urchins, abalone lysin is also present in the acrosome instead of on the sperm surface, and it is released from the acrosome during acrosome reaction. Lysin binds species-specifically to VERL and dissolves a hole in the vitelline envelope in a nonenzymatic manner (Lewis et al, 1982). The sequences of seven species of Califomia abalone have been analyzed. Although these species diverged fairly recently on the evolutionary tree, they cannot fertilize across-species. The molecular evolution rate of abalone sperm lysin is much faster than other proteins in abalone and displays positive
Darwinian selection (Vacquier, 1998). The most likely hypothesis to explain the strong positive selection is that these sperm proteins are adapting to maintain proper interaction with their changing cognate egg surface receptor, VERL. VERL contains dozens of tandem repeats of a lysin receptor sequence motif (Swanson and Vacquier, 1997); such tandem repeat sequences have a tendency to undergo concerted evolution, a process by which unequal crossing-over and gene conversion propagate and homogenize the repeat sequences within a species. Through this process, repeat sequences become more similar to each other within a species than between any two species. This would force its complementary molecule lysin on the sperm to change to remain compatible with VERL.
Thus, the combination of concerted molecular evolution on one gamete and positive selection on the opposite gamete promotes the gametes to change in a much faster rate than other molecules (Swanson and Vacquier, 1998; Vacquier, 1998). This might explain why some closely related species cannot form hybrids even though the other proteins in the two species show high sequence identity.
In contrast to those non-mammalian species discussed above, our knowledge on species specificity of mammalian fertilization is much more limited. Two intrinsic difficulties have imposed such limitations. First, the whole fertilization process is much more complicated in mammals than in sea urchin and abalone; second, internal fertilization in mammals prevents a direct study on in vivo fertilization. However, research on non-mammalian species has helped us to understand some basic concepts and to propose putafive molecule-level mechanisms behind the species specificity of fertilization. In addition, the in vitro study of mammalian fertilization has also provided information that can be used to infer the mechanism of in vivo fertilization. A variety of evidence supports the function of ZP glycoproteins as the ligands to which spermatozoa adhere. Although the study of ZP-binding molecules on spermatozoa remains controversial, there is httle doubt that the species specificity of gamete recognition must come from species diversity in the molecular components that mediate sperm-EEM interactions. Such molecular diversity could arise in at least four ways (Bi et al, 2002). (1) Each species could have its own unique set of complementary adhesion molecules,
that is, completely different sets of gene products mediate adhesion in different species
(Fig. 1.2a). This approach is probably most straightforward to fulfill the species-specific
recognition between gametes. However, data obtained so far suggest that it is unlikely
each species uses completely different gene products to effect gamete interactions. The
conservation of ZP domains in the EEM of almost all species demonstrated that at least
on the egg side the binding proteins likely evolved from a common ancestral gene. The
fianction of bindin and lysin also indicates that one single protein is sufficient to impose
the fertilization barrier between closely related species. However, we cannot exclude the
possibility that two species distanced far enough in evolution could develop two
completely different genes to achieve species-specific fertilization. (2) Functionally
distinct adhesion molecules arose by evolutionary divergence of ancestral adhesion
molecule genes (Fig. 1.2b). hi this model, the stmctures of adhesion molecules on the
sperm cell surface evolve in concert with changes in the EEM. Such changes can occur because the primary selective pressure is for a given species' complementary sperm and egg adhesion molecules to stay compatible with each other, and not necessarily to stay the same as the ancestral molecules. Hence, the nature of the adhesion molecule pair would be relatively free to change so long as within-species compatibility was maintained. (3) The same gene products are present in gametes of multiple species, but the species-unique combinations of these molecules confer species specificity (Fig. 1.2c).
The key point of this model is that species-specific gamete recognition is not conferred by any single protein; instead, combinations of multiple molecules build distinct recognition complexes among different species, thereby imposing a barrier to prevent cross-species fertilization. (4) All of the above are tme; gene products common to all mammals but also highly divergent between species may act along with species-unique gene products to mediate adhesion (Fig. 1.2d).
Based on data obtained so far, it is too early to draw any conclusion on which possibility is more favored. To distinguish among these possible mechanisms of species specificity, interspecies comparisons of sperm-EEM adhesion molecules are required.
Therefore, we must first identify the molecules on both egg and sperm that mediate the species-specific sperm-egg adhesion. In the next two sections, progress on identification of such molecules in mammalian species will be discussed.
1.3 Role of Zona Pellucida Glycoproteins in Sperm-Egg Adhesion
Despite the intrinsic difficulty of mammalian fertilization studies, extensive research by a variety of groups in the last few decades has provided solid knowledge about molecules that may mediate species-specific recognition between sperm and egg.
Like non-mammalian organisms, among a number of steps in fertilization the adhesion of spermatozoa to the EEM (it is the ZP in the case of mammals) exhibits the greatest degree of species specificity (Peterson et al, 1980; Yanagimachi, 1981). The ZP serves as a major barrier for interspecific fertilization. Without the ZP, the direct exposure of egg plasma membrane to spermatozoa permits heterologous fertilization between some species (Check et al, 1990). Indeed, this fact is the basis of the zona-free hamster egg penetration test, which can be used to assess the function of spermatozoa from human and other species. The adhesion of spermatozoa to the egg is thought to be mediated by
10 complementary factors located in the egg ZP and on the sperm surface. Indeed, the early demonstration that heterologous spermatozoa do not adhere to the ZP of pig eggs was among the first evidence that sperm-ZP interaction was not merely a consequence of non-specific adhesiveness. In addition, the ZP plays multiple critical roles during fertilization. In addition to its function as a barrier for heterologous fertilization, the ZP serves as a trigger for the acrosome reaction, and also blocks polyspermy after
spermatozoa penetration. Furthermore, even after fertilization, the one-cell zygote and
early embryo remain surrounded by the ZP dming oviductal transport until the blastocyst
stag
11 inhibit spemi adhesion to the egg in vitro by saturating the binding sites on the acrosome- intact spemiatozoa (Bleil and Wassamian, 1980a). The O-linked oligosaccharides on
ZP3, instead of the polypeptide backbone, appear to play a critical role in mediating spemi-ZP adhesion, since even after extensix'e proteolysis of ZP3 the small fragments retain the ability to inhibit //; vitro fertilization, and removal of O-linked but not N-linked oligosaccharides eliminates ZP3's sperm binding activity (Flomian and Wassannan,
1985). ZP3 is also an acrosome reaction inducer (Bleil and Wassamian, 1983).
Incubation w ith purified ZP3 can induce spermatozoa to undergo an acrosome reaction that is distinguished from spontaneous acrosome reactions by its sensitivity to pertussis toxin (Ward et al, 1992). There is also evidence suggesting that the modification of ZP3
(Wassarman, 1987; Miller et al, 1993) and/or ZP2 (Bleil et al, 1981) after the cortical reaction is responsible for the prevention of polyspermy. In addition to ZP3's adhesion function, mouse ZP2 was shown to function as the secondary binding factor to support the sustained adhesion of acrosome-reacted spermatozoa to the ZP (Bleil et al, 1988).
Studies on the ZP from a wide variety of mammals demonstrated that all of those
ZPs are composed of a limited number of glycoproteins that are closely related to mouse
ZPl, ZP2, and ZP3 (Yanagimachi, 1994). Even the extracellular matrices surrounding the eggs of non-mammals such as fish, birds, and amphibians are also composed of glycoproteins homologous to mammalian ZP glycoproteins. However, the glycoprotein serving as the sperm receptor in other species is not always the ortholog of mouse ZP3
(Sacco el al, 1989), suggesting that species variation of spemi ligand on the ZP might contribute to species specificity of sperm-ZP adhesion.
12 Because generating null alleles by gene targeting provides strong direct loss-of- function evidence, investigators have employed this technique to disrupt the gene of each mouse ZP glycoprotein, and thereby assess their functions in vivo. As expected, female mice with ZP3 null alleles {mZP3'') were infertile (Liu et al, 1996; Rankin et al, 1996).
Furthemiore, both growing oocytes and unfertilized eggs in these knockout mice were found to lack a ZP Therefore, ZP3 is not only essential for fertilization; it is also an indispensable stmctural component for assembly of the ZP matrix structure. In contrast, the ZPl null female mouse {luZPl''') is fertile (Omu cl al, 1995), and a ZP is observed surrounding the egg. Morphologically, the major difference between mZPl''and wild type mice is that the ZP in inZPF' mice is thinner and more porous, and an accentuated perivitelline space is observed during folliculogenesis. However, even though 80% of the
ZPl null females were fertile, these fertile animals consistently gave birth to litters that were about half the size of those from normal females. The decreased fecundity may largely result from early embryonic loss due to the lack of protection provided by a normal ZP during passage down the oviduct.
Recently, mice with a disrupted ZP2 gene have been generated (Rankin et al,
2001). In mZP2~'' mice, few eggs are detected in the oviduct after stimulation with gonadotropins, and no two-cell embryos are recovered after mating ZP2-null females with nonnal male mice. The stmctural defect is more severe than that observed in ZPl- null mice, but not quite as severe as that observed in ZP3-null mice. Although zona-free oocytes matured and fertilized in vitro can progress to the blastocyst stage, the developmental potential of blastocysts derived from /?!Z/'2"'''appears compromised, and live births have not been observed after transfer to foster mothers (Rankin et al, 2001).
13 Hence, it appears that the importance of ZP2 in keeping ZP function falls between ZP3 and ZPl. These results from gene targeting mutagenesis indicated that both ZP2 and ZP3 are required for successful fertilization and for assembly of a functional ZP matrix, whereas ZPl is not absolutely required for spemi adhesion or ZP assembly but is important for successful protection of early stage embryos.
In addition to offering loss-of-function evidence, establishment of ZP glycoprotein knockout mice lines also pa\cd a road for studying species specificity of sperm-egg interaction. The research group that generated ZPl, ZP2 and ZP3 knockout mice carried out a rescue experiment on the genetic background of the ZP3 null mouse
(Rankin el al, 1998). They inserted the human ZP3 gene into the mZP3'' mouse genome to generate a heterospecific ZP3-rescued mouse. The transgenic mouse was found to be fertile. Human ZP3, instead of mouse ZP3, was expressed, and an apparently normal ZP was observed. In addition, the human ZP3 was glycosylated like the ZP3 in normal human ZP, and unlike that of nomial mouse ZP3, indicating that glycosylation of ZP3 is solely based on the amino acid sequence of the polypeptide instead of on species-specific post-translational processing. Interestingly, in vitro fertilization experiments demonstrated that mouse spermatozoa could specifically adhere to the ZP of rescued transgenic mice, but human spemiatozoa failed to do so. These results indicated that an orthologous ZP glycoprotein can rescue function in fertilization, but it cannot change the species specificity of sperm-ZP adhesion. Two possible reasons may account for these results. First, specific carbohydrate modification in the Golgi apparatus may detennine the ZP's species-specific sperm adhesion activity. However, given the fact that ZP3 was glycosylated as human ZP3 instead of mouse ZP3, glycosylation is unlikely to be solely
14 responsible for the preservation of species specificity of mouse gamete recognition.
Second, ZP3 by itself may not dictate species specificity of sperm adhesion to the ZP; the matrix structure assembled by all three glycoproteins instead confers the species specificity of this interaction.
1.4 Controversial Data on Zona Pellucida-Binding Candidates in Spermatozoa
Compared to the relatively unequivocal role of ZP glycoproteins in spemi-egg
adhesion, the nature of the complementary factor(s) on the spemi surface is much more
controversial. Dozens of candidates have been postulated to serve as the mediator for
adhesion of spermatozoa to the ZP, but only a few of these proteins have been
extensively studied (Snell and White, 1996). These few proteins, including P-
galactosyltransferase (GalTase) (Lopez, 1991; Miller and Shur, 1994), sp56 (Bookbinder
et al, 1995), acrosin, and zonadhesin (Hardy and Garbers, 1994, 1995; Gao and Garbers,
1998; Hickox et al, 2001) display certain properties that make them more favored than
others. However, even for these favored candidates, no single molecule is supported by
convincing evidence that it is the sole mediator of spemi-ZP adhesion.
Among these candidates, GalTase has the longest and most extensively
researched history (Shur, 1989). It has been studied mostly in mouse. GalTase binds to
the carbohydrate moiety of ZP3, specifically to oligosaccharides that temimate with N-
acetylglucosamine, in an enzyme-substrate binding manner (Miller ct al, 1992). The
evidence suggesting that GalTase mediates sperm-ZP adhesion has come primarily from
three observations: (1) purified GalTase and antibody against GalTase inhibit adhesion of
15 spermatozoa to the ZP in a dose-dependent mamier (Shur and Neely, 1988); (2) GalTase can bind directly to radiolabeled, purified ZP3 (Miller et al, 1992); and (3) antibody against GalTase induces acrosome reactions (Gong et al, 1995). However, spemi from transgenic mice that overexpress GalTase on the spemi surface bind more radiolabeled
ZP3 but display a reduced avidity for the whole ZP compared to wild type cells
('I'ouakini et al, 1994), indicating that interaction with a single ZP glycoprotein is not parallel to the interaction with the whole ZP. Moreover, recent data from GalTase knockout mice revealed that the GalTase'' mice are fertile (Lu and Shur, 1997). In vitro fertilization also showed that more GalTase''' spermatozoa adhere to the ZP than wild type spermatozoa. Therefore, GalTase is not the sole sperm protein required for either mouse fertilization or adhesion of spemiatozoa to the ZP Although GalTase has been identified on the sperm surface in other species, there is no evidence to suggest that
GalTase binds to ZP3 in a species-specific maimer. In fact, since the binding site on the
ZP for GalTase is its enzymatic substrate, which displays little if any species variation, it is unlikely that GalTase is able to confer the species specificity of sperm-ZP adhesion.
Several lines of evidence also support the putative role of another candidate, sp56, which was identified because it was specifically radiolabeled by a photoactivatable crosslinker covalentiy bound to ZP3 (Bleil and Wassarman, 1990). Like GalTase, the major evidence that sp56 is a ZP adhesion molecule comes from the observation that pre incubation of spermatozoa with purified sp56 or antibodies to sp56 interferes with sperm-ZP adhesion (Cheng et al, 1994). sp56 was previously reported as a peripheral membrane protein (Bookbinder et al, 1995), but recent data from immunoelectron microscopy suggested that sp56 is largely, if not exclusively, located in the acrosomal
16 mattix in both mouse and guinea pig spennatozoa (Foster et al, 1997; Kim et al., 2001).
Another problem preventing sp56 from being considered as the universal ZP adhesion factor is that orthologs of sp56 were not be detected in some species, such as human
(Bookbinder et al, 1995; Snell and White, 1996). These data indicated that sp56 is unlikely to serve as a signal transducer (because it is not a membrane protein) or a general ZP binding factor in all mammals (apparent lack of an ortholog in certain species). Like GalTase, no evidence supports that sp56 binds to the ZP in a species- specific manner.
Another protein worth mentioning is proacrosin. Although it is not considered as a likely ZP adhesion molecule for acrosome-intact spermatozoa due to its localization, proacrosin does exhibit some functions during sperm-ZP adhesion and subsequent ZP penetration (Bleil et al, 1988). Because it is mainly localized in the acrosomal matrix, which is masked from direct ZP interaction by the plasma membrane of acrosome-intact spermatozoa, some investigators considered it a secondary ZP adhesion protein during fertilization (Yanagimachi, 1994). Nevertheless, proacrosin cannot be excluded to play an important role in spemi-ZP adhesion because it does bind specifically to the ZP.
However, proacrosin binding is highly promiscuous; pig proacrosin appears to bind equally well to pig, bovine, and mouse ZP, as well as to Xenopus oocytes envelope, eliminating its potential function in conferring species specificity to sperm-ZP adhesion
(Hardy and Garbers, 1994). Acrosin is a serine protease that has long been postulated to digest the ZP and thereby facilitate penetration by motile spermatozoa (Urch et al,
1985a, b). Some investigators also observed that acrosin sustained binding power during sperm penefration through the ZP (Yanagimachi, 1994). Two independent research
17 groups have generated null alleles of the mouse proacrosin gene (Baba et al, 1994).
Unexpectedly, Acr" mice are fertile. //; vitro fertilization demonstrated that Acr'' spemiatozoa can penetrate the ZP, but this process is significantly delayed compared to wild type spemiatozoa. Therefore, e\en though much //; vitro evidence suggested that proacrosinVacrosin plays a role in ZP adhesion and penetration, it is not absolutely required for //; vivo fertilization, indicating that fertilization is indeed a complicated process in which multiple, redundant i'actors are likely involved. It seems that in some cases mice with targeted deletion of molecules that are thought to function in fertilization are fertile. Although such results may indicate that those molecules actually have no function in fertilization //; vivo, they probably also reflect the complexity of fertilization and the redundancy of the sperm molecules that are involved in fertilization.
All of these ZP binding candidates on spermatozoa were initially identified based upon their interaction with solubilized glycoprotein or denatured ZP instead of the native
ZP matrix. As discussed above, the species variation in the ZP glycoproteins that serve as the sperm ligands indicates that a single ZP glycoprotein may not function exclusively in this capacity for all mammalian species. Recent data on the pig ZP showed that neither pig ZPB (ortholog of mouse ZPl) nor ZPC (ortholog of mouse ZP3) is able to bind to pig spermatozoa by itself, but these glycoproteins can form heteroduplexes in vitro that bind avidly to pig sperm membranes (Yurewicz et al, 1998). Therefore, the sperm adhesion activity is more likely conferred by the whole ZP stmcture instead of a single glycoprotein. As far as the species specificity is concerned, results from the human ZP3- rescued inZP3''' mice indicated that ZP3 by itself could not confer the species-specific adhesion between spermatozoa and the ZP. As long as an intact ZP is assembled outside the egg, it restores the original species specificity even though it contains heterologous
ZP3 (Rankin et al, 1998). Therefore, it is likely that the whole ZP matrix structure constructed by all the ZP glycoproteins, instead of ZP3 only, confers species specificity of spemi-ZP adhesion. Thus, in the research for egg binding proteins on the sperm membrane, molecules that bind directly to the native ZP may be more likely to ha\ e a natural function in spemi-egg adhesion than molecules that bind only to a single ZP glycoprotein.
ft IS also possible that a single sperm factor is not sufficient to mediate ZP adhesion. In fact, the events that follow spemi-ZP adhesion are relatively simple on the
ZP side, as the ZP largely plays a passive role and does not undergo significant change during or right after sperm-ZP interaction. After all, the ZP is an extracellular matrix primarily composed of only three glycoproteins in most species, but the composition of the spemi membrane is far more complicated. Upon interaction with the ZP, acrosome- intact spermatozoa undergo acrosome reactions, a process that requires a cascade of signal transduction events. Hydrolytic enzymes are subsequently released, and the whole sperm cell experiences both changes in morphology and motility in preparation for ZP penetration (Yanagimachi, 1994). Therefore, it is not a surprise that the study of ZP binding factor(s) of spermatozoa has been more difficult, diverse and controversial.
Although we cannot exclude the possibility that one single protein will eventually emerge as the exclusive and universal egg binding protein, it seems more likely that multiple sperm proteins function during adhesion to the ZP For example, one factor may serve to mediate initial interaction with the ZP, a second factor may mediate signal transduction in
19 the acrosome reaction, and yet a third factor may function as the species-specific recognizer for adhesion between homologous gametes.
1.5 Preliminary Studies on Zonadhesin
Zonadhesin is a multiple-domain spenn protein with properties that might account for species specificity of spemi-ZP interaction. This protein was first identified in pig spemi membrane extracts in 1994 and named zonadhesin due to its ability to bind to the
ZP ("zona" + "adhesion") (Hardy and Garbers, 1994). The rationale behind the discovery of zonadhesin was to identify the protein(s) that bind to the native, particulate ZP in a species-specific manner. Particulate ZP was chosen as the affinity matrix because the native stmcture, not individual ZP glycoproteins probably plays a critical role in species- specific sperm-egg recognition and adhesion. In contrast, previous approaches that lead to identification of other ZP-binding candidates almost exclusively used solubilized ZP glycoproteins as ligands (Lopez et al, 1991; Miller and Shur, 1994; Bookbinder et al,
1995).
Although several proteins were observed initially to bind to particulate ZP, a protein migrating in SDS-PAGE at M, 150,000 under nonreducing conditions was the only polypeptide that remained bound with high apparent affinity to the native ZP after several rounds of detergent washing. Analysis by SDS-PAGE under disulfide-reduced conditions demonstrated that this protein is composed of two disulfide-bonded subunits, one migrating at Mr 105,000 (pl05) and the other migrating at Mr 45,000 (p45) (Hardy and Garbers, 1994). These two subunits were later found to be encoded by a single
20 mRNA transcript (Hardy and Garbers, 1995). In addition to its high avidity, binding of zonadhesin to the ZP was observed to be species-specific. Both p45 and pi05 from pig spemiatozoa bound specifically to the pig ZP, but failed to bind to the ZP isolated from other tested species, such as human, mouse, bovine or Xenopus (Hardy and Garbers,
1994). Zonadhesin was the first and only ZP-binding candidate on mammalian spemiatozoa showing the potential to confer the species specificity of spemi-egg interaction.
A 7785-bp pig zonadhesin cDNA has been cloned from testis cDNA libraries and fully sequenced. The cDNA contains a single major open reading frame of 7488 bp. The sequence predicts a 29-amino acid putative signal peptide, a 2418-amino acid extracellular region, a single transmembrane segment, and a 36-amino acid cytoplasmic tail. All tryptic fragments from both pi05 and p45 can be found in the deduced amino acid sequence. Dotplot analysis and sequence comparisons revealed that the extracellular region of zonadhesin contains multiple domains (Fig. 1.3). In the extracellular region, one and a half domains homologous to MAM domains are located right after a signal peptide.
The MAM domains are followed by a region of highly repetitive sequence that comprises dozens of consensus sequence repeats usually present in mucins. Following this mucin- like domain, five tandem homologous domains show high sequence identity to von
Willebrand factor (vWF) D domains (VWD) (Hardy and Garbers, 1995).
MAM domains are named after three membrane proteins containing this domain: meprins (M) (Jiang et al, 1992), A5 antigen (A) (Beckniann and Bork, 1993), and receptor protein-tyrosine phosphatase \x (RPTPfi, M) (Zondag et al, 1995). A typical
MAM domain contains about 160 amino acids including four conserved cysteine residues
21 and some conserved hydrophobic and aromatic amino acids (Gao and Garbers, 1998).
Meprins are zinc-metalloproteases in which the MAM domain is likely to promote fomiation of dimers and oligomers among meprins subunits (Takagi et al, 1991; Jiang et al, 1992), In A5 antigen, which is a developmentally regulated cell suriace molecule, the
MAM domain is involved in mediating cell-cell interactions (Takagi ei al, 1991). MAM domains in RPTP|a are essential for hemophilic cell-cell interaction and for specificity of those interactions (Takagi et al, 1991; Zondag et al, 1995). hi general, proteins containing MAM domains are involved in cell-cell interactions.
In addition to von Willebrand factor, VWD domains have been found in a number of other proteins, such as human intestinal mucin MUC2 (Gendler and Spicer, 1995),
Xenopus laevis integumentary mucin FIM-B.l (Joba and Hoffmann, 1997), and mouse irmer ear matrix protein a-tectorin (Legan et al, 1997). Mucins belong to a large glycoprotein family that modulates cell-cell interactions in somatic tissues. Most VWD- domain proteins are also involved in cell-cell or cell-extracellular stmcture interaction.
Both p45 and pi05 contain VWD domains, and initial data showed that p45 and pi05 form high Mr multimers as well as the Mr 150,000 heterodimer (Hardy and Garbers,
1994). Study of vWF and mucins demonstrated that VWD domains play an important role in multimerization of these proteins (Sadler, 1998). More details on VWD domain and multimerization in vWF and mucins will be presented in the next section. From this discussion, it is apparent that all of zonadhesin's extracellular domains have the potential to be involved in cell-cell interactions, indicating a potential role for zonadhesin in mediating such interactions.
22 Cleavage of the putative signal peptide at serine 29 would produce a 2447-amino
acid mature, nascent polypeptide chain with a calculated molecular mass of 267 kDa
(Hardy and Garbers, 1995). This predicted molecular mass is much larger than the sum
(Mr 150,000) of its known subunits, p45 and pl05, indicating that post-translational processing is involved in producing mature zonadhesin subunits. Either some parts of the zonadhesin precursor are removed and degraded by proteolytic processing, or other zonadhesin coniponeiit(s) comprising the missing domains are actually present in spermatozoa but lack the ability to bind to the ZP Northem Blots of poly (A)+ RNAs from several pig tissues disclosed that testis is the only tissue in which zonadhesin niRNA can be detected, and in situ hybridization further localized zonadhesin expression more precisely to haploid spermatids (Hardy and Garbers, 1995). Spermatid-specific expression of zonadhesin supports the potential role of zonadhesin in reproduction- related process.
Orthologs of pig zonadhesin have been cloned and sequenced in mouse (Gao and
Garbers, 1998), rabbit (Lea et al, 2001) and human (Wilson et al, 2001; Cheung et al, in preparation). Zonadhesin in those three species contains the similar domain stmcture as pig zonadhesin, but mouse zonadhesin exhibits variations in D domains by containing
20 additional tmncated D domains between D3 and D4 (Fig. 1.3). In contrast to the one and half MAM domains in pig, three full MAM domains are present in mouse zonadhesin
(Gao and Garbers, 1998). Like pig zonadhesin, both mouse and human zonadhesin niRNA are expressed specifically in testis, supporting its potential role in mammalian fertilization. The similarity and variation of domain structures among species suggest that
23 point mutations within the domains as well as shuffled domain structure may confer the species specificity of its ZP binding activity.
In both human and mouse, zonadhesin variant fragments that may result from alleraative splicing or gene rearrangement were observed during sequencing of their cDNAs (Gao and Garbers, 1998; Wilson et al, 2001; Cheung et ul, in preparation). In mouse, PCR products lacking partial D3 domain or containing extra partial D3 domains w ere detected, indicating possible niRKA alternative splicing. Existence of rare PCR products with partial reversed-order domains may suggest occasional gene rearrangement e\'ents (Gao and Garbers, 1998). Gene arrangement and alternative splicing appear more evident in human zonadhesin. Six mRNA variants of zonadhesin exist in human, each having different splicing ex'cnts in the D4 domain (Cheung et al, in preparation). These rearrangements lead to a premature termination of the protein in four of the variants such that the EGF-like domain and the transmembrane segment are not found. The absence of a transmembrane segment in some variants suggests that zonadhesin may also exist as a secreted protein, as well as a membrane protein. The presence of all types of deletions m a "hot region" of the D4 domain suggested that the deletions likely do not reflect random individual variation or cloning artifacts. Instead, it seems likely to reflect a specific gene rearrangement or mRNA altemative splicing in zonadhesin expression. Characterization of the human zonadhesin gene now indicates that the six variants represent altematively spliced forms of the mRNA (Wilson et al, 2001; Cheung et al, in preparation).
Although it is too eariy to make any conclusion, the gene rearrangement or altemative mRNA splicing of human zonadhesin may indicate that zonadhesin is located in more
24 than one compartment and perform multiple functions corresponding to its localization in spemiatozoa.
1.6 Multimerization and Functions of VWD Domains in vWF and Mucins
One of the most distinct characteristics of zonadhesin is that it contains both mucin-like and vWF D domains (VWD). In addition, it shares similarity in domain structure with vWF and a few types of mucins (Fig. 1.4). Like zonadhesin, both vWF and mucins are very large molecules. Three factors contribute to the large sizes of vWF and mucins. First, the nascent polypeptides themselves comprise a very large number of amino acid residues; second, they (especially mucins) usually undergo heavy glycosylation in the endoplasmic reticulum (ER) and the Golgi apparatus; third, they fomi multimers by intemiolecular disulfide bonds, and such multimerization is usually required for their normal functions (Sadler, 1998; Perez-Vilar and Hill, 1999).
Accordingly, information on vWF and mucins is relevant to characterization and functional studies of zonadhesin.
vWF was first found by Erik von Willebrand in research of a common inherited bleeding disorder now called von Willebrand disease, ft is a large, multimeric glycoprotein found in blood plasma, platelet a-granules, and subendothelial connective tissue. vWF has two functions in hemostasis—mediating the adhesion of platelets to subendothelial connective tissue and binding blood clot
25 residues. A 22-residue signal peptide is removed first, and subsequent proteolytic processing generates a mature vWF protein of 2050 residues and a propeptide of 741 residues. Prepro-vWF includes 14 domains and four of them are fiill D domains (Sadler,
1998). Most of the vWF domains are rich in cysteine, and all of those cysteine residues
appear to be oxidized in disulfide bonds in the secreted protein (Marti et al, 1987). Upon franslocation into the endoplasmic reticulum, prepro-vWF is glycosylated by both N-
linked and O-linked oligosaccharides, and then forms tail-to-tail disulfide-bonded dimers through its C-terminal CK domain (Marti et al, 1987; Voorberg et al, 1991). Following transport into the Golgi apparatus, provWF (vWF precursor with only the signal peptide removed) dimers are subjected to further glycosylation and sulfation, and then form additional head-to-head disulfide bonds near the N-terminal domains to generate multimers with masses greater than 20 million Daltons. Finally, the propeptide is removed and mature vWF multimers are secreted or stored in Weibel-Palade bodies of endothelial cells (Sadler, 1998).
Although the vWF propeptide is not present in the mature protein, it is required for vWF multimerization (Verweij et al, 1987; Wise et al, 1988). In addition, the acidic environment of the Golgi apparatus is also cmcial for muhimer formation (Spom et al,
1986). The vWF propeptide contains two D domains (Dl and D2); both of them include a conserved sequence CGLCG. The vicinal cysteines in this sequence are critical for multimer formation; even insertion of a single additional amino acid residue between two cysteines dismpts the formation of vWF muhimers but has no impact on the formation of tail-to-tail dimers (Mayadas and Wagner, 1992). Because a similar sequence motif is present in the active sites of enzymes that are involved in catalyzing formation of
26 disulfide bonds during protein folding, the ftinction of the vWF propeptide probably mirrors the activity of disulfide isomerase, which is to facilitate proper intermolecular disulfide bond formation by catalyzing protein disulfide interchange (Mayadas and
Wagner, 1992). Mature vWF has two destinations in the cell's protein trafficking pathway: 95% is secreted constitutively, whereas the rest is stored in cytoplasmic granules called Weibel-Palade bodies (Spom et al, 1986). Secreted vWF constitutes mostly small multimers and a high proportion of unprocessed pro VWF subunits, but the vWF in Weibel-Palade bodies is primarily very large multimers (Spom et al, 1986).
Even though small multimers and pro VWF are secreted into blood circulation, only large multimers have a long half-life in the circulation and retain the ability to bind to platelets and Factor VIII. It is apparent that multimerization is essential for vWF's function in blood clotting, and the large multimers are probably released from Weibel-Palade bodies in response to bleeding (Sadler, 1998). Indeed, one type of von Willebrand disease (type
2A) is caused by lack of vWF multimers in circulation (Lyons et al, 1992).
In addition to vWF, another family of glycoproteins that contain VWD domains is mucins. Mucins are major glycoprotein components of the mucous, which primarily functions to protect epithelial cells that are exposed to extemal substances and to help materials travel smoothly through a tract (Perez-Vilar and Hill, 1999). Although a variety of different-sized mucins exist in heterogeneous form, one of the common properties is that they are all extensively glycosylated by O-linked oligosaccharides on threonine and serine residues in tandem repeated sequences that vary in number, length and amino acid sequence (Gendler and Spicer, 1995). These heavily 0-glycosylated repeat sequences are also present in other proteins where they are called mucin-like domains, hiterestingly,
27 some secreted mucins contain three N-terminal cysteines-rich VWD domains, and some
have a fourth VWD domain at the C-terminal, separated by a large mucin domain, hi
contrast to mucin domains, these VWD domains are usually N-glycosylated (Perez-Vilar
and Hill, 1999).
Like vV^, the mature mucins also form extensive intermolecular disulfide bonds.
Study of mucin assembly is more difficult than vWF due to the enormous size of the monomers. Both nascent zonadhesin and prepro-vWF are very large molecules compared to most other proteins, but some well studied mucins have twice or more times as many amino acid residues as zonadhesin and vWF. For instance, MUC5B in human has 5662 amino acid residues (Desseyn et al, 1997), and its counterpart in pig, PSM (porcine submaxillary mucin), contains 13288 amino acid residues (Eckhardt et al, 1997).
Although it is almost impossible to study expression and muftimerizafion of these huge proteins directly, investigators have obtained insights into multimer formation by expressing recombinant mucin domains in mammalian cells, followed by characterization of the expressed proteins with SDS-PAGE and chromatography under reducing and nonreducing conditions. Similar to vWF, mucins form disulfide-bonded dimers through their C-terminal CK-domains (Perez-Vilar et al, 1996; Perez-Vilar and Hill, 1998b), and the dimers then form disulfide-bonded multimers through their N-terminal VWD domains (Perez-Vilar et al, 1998). Amid these two events, N-glycosylation and extensive
0-glycosylafion take place. Studies on PSM revealed that three cysteines conserved in all
CK domains are required for dimerization (Perez-Vilar and Hill, 1998b). In addition, the relatively acidic environment in the Golgi apparatus and the conserved sequence of the
CGLCG motif are also critical for successful multimerization (Perez-Vilar and Hill,
28 1998a). One major difference between multimerization of mucins and vWF is that the propeptide containing the Dl and D2 domains is cleaved from mature vWF multimers but retained in mucins.
In addition to vWF and mucins, a newly found protein in the mouse inner ear named a-tectorin also contains four full and one partial VWD domain (Legan et al,
1997). Compared to the VWD domains in mucins and vWF, a-tectorin and zonadhesin share much higher amino acid sequence identity (26%)), and those VWD domains are even arranged in the same order (i.e., all in tandem). Interestingly, a-tectorin also contains one ZP domain (Legan et al, 1997). Moreover, the ZP domain is also present in another related irmer ear protein, P-tectorin (Legan et al, 1997). The significance of containing sperm protein domains and an egg protein domain in one protein is unclear at this time, but this fact suggests a possible link between VWD domains and ZP domains, thereby further supporting a potential function of zonadhesin D domains in ZP-adhesion.
1.7 Purpose of Study
Even though fertilization has been under intensive study for almost 100 years, the molecule-level mechanisms behind gamete recognition during fertilization remain to be clarified. Especially for ZP-binding candidates on mammalian spermatozoa, the literature is full of controversial and even conflicting data, which may result from the intrinsic differences among various species or from different research approaches.
Some unique characteristics of zonadhesin, especially its ability to bind to the native ZP in a species-specific manner, make it a top candidate for mediating species-
29 specific sperm-egg recognition and adhesion. To understand the fianction of zonadhesin in fertilization, it is necessary to characterize both biochemical and functional aspects of the protein. Key questions to be answered include how zonadhesin is expressed and post- translationally processed, what factor(s) influence zonadhesin's maturation and ftinction, at which step zonadhesin functions as an adhesion molecule in mediating sperm-egg recognition, and so on.
In this dissertation I report on extensive studies of pig zonadhesin, including biochemical characterization, ZP binding activity, and ultrastmctural localization on spermatozoa. All of these studies were aimed at addressing a single overall hypothesis: zonadhesin mediates egg-sperm adhesion in a species-specific manner. The purpose of my research was to help us to understand zonadhesin processing and maturation, and provide more information on its potential role in fertilization. Potential practical benefits derived from this work may include development of efficient contraceptives and diagnostic methods for human fertility and farm animal fecundity.
30 Fig. 1.1 Morphology of the sperm apical head in interactions with the mammalian egg extracellular matrix. Panel a: Interaction of the plasma membrane overlying the head of a sperm cell that has an intact acrosome. Panel b: Interaction of the outer aspect of the acrosomal matrix that has emanated through discontinuities in the plasma and outer acrosomal membranes in the very early stages of acrosomal exocytosis. Panel c: Interaction of acrosomal matrix remnants later in the progression of acrosomal exocytosis. Panel d: Interaction of the inner acrosomal membrane after completion of acrosomal exocytosis.
31 Species A Species B Species C SP ^^ SP ^ SP
EEM EEM EEM
SP ^^ SP ^ SP b H^ WW fifi EEM EEM EEM
SP ^ SP ^ SP
EEM EEM EEM
SP ^ SP ^ SP d ^1_^1_^1 EEM EEM EEM
Fig. 1.2 Hypothetical molecular mechanisms for species specificity. Interactions between sperm and EEM of three species (A, B, C) are depicted. Row a: Each species has a unique adhesion molecule pair. Row b: Each species has the same adhesion molecule pair that has developed species-specific function through divergence from a common ancestral protein. Row c: Each species uses a unique combination of proteins shared by the three species. Row d: Combinafion of 6 and c.
32 MAM DO Pig II 1 M D1 D2 D3 D4 / 500 aa SP
MAM DO Human 1 2 3 M Dl D2 D3 D4 / SP Mouse
MAM DO DO-like Domains 1 2 3 M Dl D2 D3 D4 / SP Plasma Membrane
Fig. 1.3 Schematic comparison of zonadhesin in three species. MAM represents the MAM domain; M represents the mucin-like domain; Dl, D2, D3, and D4 represent the full VWD domains; DO represents the tmncated, partial VWD domain; SP represents the putative signal peptide; and the black bar represents the putative transmembrane segment and cytoplasmic tail.
33 MAM DO Pig zonadhesin 1 H M 1 1 01 02 03 04 '
[ )C Mouse a-tectorin II ^^ 02 03 04' ZP 1
Human preprovWF 01 02 10 3 04'
0' Human IV1UC2 f | 01 02 llfirP^II ^ III Mucin_23aa repeat 04
Fig. 1.4 Schematic comparison of VWD domain-containing proteins. MAM represents the MAM domain; M represents the mucin or mucin-like domain; DO, Dl, D2, D3, D4, and D4' represent the various full and partial VWD domains; and ZP represents the zona pellucida domain conserved in EEM glycoproteins from a broad range of species. Black boxes represent the domains unique to the specific proteins that have no sequence similarity to zonadhesin.
34 CHAPTER II
EXPRESSION OF ZONADHESIN FRAGMENTS IN
CULTURED MAMMALIAN CELLS
2.1 Introduction
Existing data suggest that zonadhesin plays a role in mediating species-specific
adhesion between spermatozoa and the egg ZP during fertilization. However, these data
derive primarily from in vitro protein binding experiments and characterization of the
zonadhesin protein; additional cell adhesion ftmctional studies are needed to evaluate
zonadhesin's role in fertilization. Loss-of-fiinction evidence would come from
experiments wherein spermatozoa failed to interact normally with the egg ZP if
zonadhesin was eliminated from the cells; this type of evidence would demonstrate the
"indispensability" of zonadhesin in sperm-egg interaction. In contrast, gain-of-function
evidence would come from experiments wherein ectopic expression of zonadhesin
enabled cultured mammalian cells that normally do not express the protein to act like
spermatozoa and adhere to the egg ZP; this type of evidence would demonstrate the
"sufficiency" of zonadhesin in sperm-egg interaction. Loss-of-function studies typically
include characterization of specific gene dismpted mice, which are currently being
generated by my colleagues. Although interference of in vitro fertilization with purified proteins or anfibodies is also commonly used to obtain loss-of-function evidence, the
initial inhibition experiment using antisera to zonadhesin failed to interfere with in vitro
sperm-egg adhesion. As I will discuss later, these inhibition experiments probably are
35 not suitable for study of zonadhesin due to its localization in spermatozoa. For a gain-of- function experiment, because zonadhesin is expressed only in testis (Hardy and Garbers,
1995), we can express and target zonadhesin to the plasma membrane in cultured mammalian cells, which normally do not express zonadhesin. In principle, this could transform the cultured cells into sperm-like cells with respect to zonadhesin. If these zonadhesin-expressing cells specifically bound to the ZP, we would have strong evidence that zonadhesin mediates sperm-ZP adhesion during fertilization. In addition, the sperm like cultured cells could be used to study species specificity by testing the adhesion between zonadhesin-expressing cultured cells and heterologous ZP. Heterologous zonadhesins could also be fiised to the sperm membrane; the species specificity of such mosaic cells would provide direct information on how species specificity is conferred at molecular level.
Compared to most other proteins, the zonadhesin precursor is predicted to be a very large nascent polypeptide that undergoes post-translational processing to generate the two known polypeptides, pi05 and p45, as the active proteins. As I have discussed in the previous chapter, for huge molecules like some secreted mucins, it is a practical approach to study their expression, targeting and post-translational modification by expressing fragments of large molecules in cultured mammalian cells (Perez-Vilar and
Hill, 1999). Because zonadhesin is a multiple-domain protein, it is reasonable first to express individual domains or domain combinations to determine which domain or domains are involved in ZP binding activity. The native, particulate pig ZP can be used as the immobilized material. If certain expressed domains or domain combinations retain the ability to bind species-specifically to the ZP, we can constmct constant expression
36 plasmid to express the pepfide in large amount. By comparing the activities of these expressed proteins, it should be possible to identify individual domains or combinations of domains that mediate zonadhesin's functions in sperm-egg interaction.
Here I describe expression of zonadhesin fragments in cultured COS-1 cells. I successfiiUy expressed a fragment of zonadhesin that includes DO and Dl domains in cultured COS-1 cells. The expressed zonadhesin peptide (rDO-Dl) was purified by Ni^"^ affinity column from cell extracts because of a hexahistidine tag located at its C-terminus.
However, even though a signal peptide was inserted at the N-terminus of the rDO-Dl plasmid, only a small amount of rDO-Dl peptide was secreted into the media. 1 also developed a number of additional constmcts for secretion of expressed proteins, but none of them worked as expected. This failure of secretion may reflect the complexity of zonadhesin's processing in the cell's pathway, suggesting that it is a prerequisite to understand the protein's biochemical characteristics before any attempt for gain-of- function experiment.
2.2 Materials and Methods
2.2.1 Constmction of pCMV-5/DO-Dl Plasmid
Constmction of the DO-Dl domain plasmid began with digestion of pig zonadhesin cDNA clone M2 with Sail and ^coRI and subsequent isolafion of the excised
1.6 kb DNA fragment spanning the DO and Dl domains (DO-Dl fragment). A 100 bp
PCR product containing a translation initiation codon and the signal peptide of receptor guanylyl cyclase-A (GCA, to direct expressed protein to secrete into culture media) was
37 generated using sense and antisense primers containing Kpnl and Xhol restriction sites
(sense primer: 5'-TGCCCGGTACCTGCACTCGCTGA-3'; antisense primer: 5'-
TTGGTCAGCTCGAGCACCACAGCCA-3'), respectively, and a rat GCA cDNA clone as template. After Kpnl/Xhol digestion, the Xhol overhang was ligated to the compatible
5' Sail overhang of the DO-Dl fragment. A synthetic adapter with an ^coRI overhang,
six histidine codons (6-his, to enable its purification by Ni^"^ affinity chromatography), a
stop codon, and a BamUl overhang was ligated to the 3'-end of the BamHl digested DO-
Dl fragment. This recombinant DO-Dl insert was then ligated into Xp/zI/5amHI-digested pCMV-5 (a mammahan expression vector that works especially well in SV40- transformed simian COS cell lines) to generate the rDO-Dl (recombinant DO-Dl zonadhesin) pCMV-5 constmct.
2.2.2 Transient Expression of rDO-Dl Zonadhesin in COS- 1 Cells
COS-1 cells were cultured in a 75 cm^ vented tissue culture flask (Coming, Inc.) in 15 ml of DMEM media (Gibco, Inc.) containing 10%o fetal bovine semm (FBS, Gibco,
Inc.) and Ix antibiofic-antimycotic solution (100 units/ml penicillin, 0.25 |ag/ml amphotericin B and 100 )ag/ml streptomycin, Mediatech, Inc.) in a cell culture incubator
(37°C, 5% CO2). A transfection cocktail was made by adding 50 |al 10 mg/ml DEAE- dextran (filter-sterilized) drop by drop into 1 ml Ca^^Mg^^ free PBS buffer (10 mM
NaP04, 150 mM NaCl, pH 7.4) containing 5 |ig pCMV-5/DO-Dl (or 5 |ag pCMV-5 vector without insert as a control). After COS-1 cells grew to reach 70-90% confluence, the culture media were removed and the cells were washed twice with PBS. The
38 transfection cocktail was dripped evenly onto the flask wall where the cells were growing. The flask was then incubated for 30 min in cell culture incubator, tifting every
10 min. The cells were shocked by adding 10 ml pre-warmed DMEM media (without
FBS and antibiotic solution) containing 10% DMSO and incubating for 2.5 min at 22°C.
Finally, 15 ml of regular DMEM media containing 10% FBS and Ix antibiotic- antimycotic solution were added, and the cells were incubated for 24 h. After replacement with fresh media, the cells were incubated for an additional 48 h. The media were collected directly and centrifuged briefly to remove any cellular debris. After washing twice with PBS, the transfected COS-1 cells were harvested with a scraper and resuspended in 2 ml of PBS. For analysis by Westem blotting, cells were dissolved directly in Ix SDS-PAGE sample buffer under disulfide-reduced conditions, while the culture media were 1:1 diluted in 2x SDS-PAGE sample buffer under disulfide-reduced conditions.
2.2.3 Purification of rDO-Dl Zonadhesin from COS-1 Cell Extract
Cellular proteins were solubilized in 20 mM NaP04, pH 7.4 containing 0.5 M
NaCl and 1% hydrogenated Triton X-100, and then separated by Ni^"^ affinity chromatography (HiTrap chelating column charged with Ni2S04, Phamacia, Inc.). The hexahistidine tag on the expressed rDO-Dl enabled the recombinant protein to bind specifically to the Ni^"^. The 1 ml column was equilibrated in starting buffer (20 mM
NaP04, pH 7.4, 0.5 M NaCl, 20 mM imidazole, 0.2% hydrogenated Triton X-100) at a flow rate of 1 ml/min. After sample loading at 1 ml/min, the column was washed with
39 starting buffer until A280 retumed to the base line. Bound proteins were eluted with a 10 ml linear 20-500 mM imidazole gradient at a flow rate of 1 ml/min and the collected fractions were analyzed with Westem blotting.
2.2.4 Purification of rDO-Dl Zonadhesin from Culture Media
Purification of the rDO-Dl protein from culture media of transfected cells largely followed the same procedure as purification from cell extracts. Briefly, 15 ml of rDO-Dl plasmid-transfected or non-recombinant pCMV-5-transfected culture media were filtered with a 0.45 |j,m syringe filter, centrifiiged 10 min at 10,000 x g to remove cells, and then loaded onto a starting buffer equilibrated Hi-Trap Ni""" chelating column at a flow rate of
1 ml/min. The flowing-through media were recycled on the column four times to maximize binding of the hexahistidine tag-containing proteins to the column. After extensive washing with starting buffer (until A280 retumed to baseline), the bound protein was eluted with a 10 ml linear 20-500 mM imidazole gradient at a flow rate of 1 ml/min and then the collected fractions were analyzed with Westem blotting.
2.2.5 Constmction of Recombinant Protein Containing Zonadhesin Mucin Domain
Two constmcts with GCA signal peptides and a part of zonadhesin mucin domain were generated based on the pig zonadhesin cDNA clone Ul. A 2.0 kb cDNA fragment containing parts of the M domain, the DO domain, and the Dl domain (M-Dl) was generated by digesting clone Ul with Hindlll and EcoRl, and a 3.2 kb cDNA fragment
40 containing an additional D2 domain (M-D2) was generated by digesting clone Ul with
Hindlll and Ncol. A 226 bp PCR product encoding part of the 5'-untranslated region and the signal peptide of GCA was generated using sense and antisense primers containing
Kpnl and Hindlll restriction sites, respectively. After KpnllHindlll digestion, the GCA signal peptide was ligated to the 5' Hindlll overhang of both M-Dl and M-D2 fragments.
A synthetic adapter with an ^coRI overhang, six histidine codons, a stop codon, and a
BamEl overhang was ligated to the 3'-end of the M-Dl fragment to generate the rM-Dl insert, and another synthetic adapter with an Ncol overhang, six histidine codons, a stop codon, and a BamVil overhang was ligated to the 3'-end of the M-D2 fragment to generate the rM-D2 insert. Finally the two inserts were ligated into A!p«I/5awHI-digested pCMV-5 to produce the rM-Dl and rM-D2 plasmids.
2.2.6 Constmction of Recombinant Protein Containing von Willebrand Factor Signal Peptide
A 160 bp fragment encoding the vWF signal peptide (including the initiation codon) was generated by PCR from a human vWF cDNA clone (kind gift of Professor
Denise Wagner) and inserted into the EcoRUKpnl sites of pCMV-5, and a synthesized
DNA fragment with six histidine codons and a stop codon was inserted into the ClallSaK sites of the plasmid 3' to the vWF signal peptide. Notl and CM sites between these two regions were used to insert various PCR-generated zonadhesin domains and domain combinations, because neither of these sites is present in the pig zonadhesin cDNA. This newly constmcted plasmid specialized for subcloning and expressing cassettes of zonadhesin domains was named pv6H. A pig zonadhesin cDNA fragment containing the
41 Dl domain was inserted to obtain the pv6H-Dl constmct used for expression in COS-1 cells.
2.2.7 Expression and Purificafion of the DO-Dl Fusion Protein
The 1.7 kb ^coRl fragment of pig zonadhesin cDNA clone M2 (in pBluescript) was subcloned into the EcoRI site of pET-23d. The 5' sticky end of this fragment came from the EcoRl site in the adapter used to constmct the cDNA library (Hardy and
Garbers, 1995), and its 3' sticky end from the ^coRI site at nucleotides 4063-4068 of the zonadhesin composite cDNA (Genbank Accession # U40024). This constmct specified an Mr 64,000 fusion protein comprising 20 amino acids of N-terminal vector-encoded protein, 19 amino acids of C-terminal vector-encoded protein (including a hexahistidine tag), and zonadhesin amino acids Pro^^^-Ser'^^'* The fusion protein was expressed in E. coli strain BL21/DE3 by induction with 0.5 mM isopropylthiogalactoside (IPTG) for 2 h at 37°C, and isolated from inclusion bodies by preparative SDS-PAGE and electroelution.
2.2.8 Preparation of DO-Dl Antisera
Asp-Pro bonds of the purified DO-Dl fusion protein were hydrolyzed for 36 h with 70%) formic acid at 37°C. The final hydrolysates contained a mixture of proteins with MrS corresponding to partial hydrolysis products predicted from the deduced amino acid sequence, including an Mr 33,000 core polypeptide. Hydrolysates were lyophilized to remove formic acid prior to injection. Two female NZW rabbits were immunized
42 (i.m.) with 0.2-0.5 mg protein each in 1 ml Freund's complete adjuvant (=day 0). Booster injections (i.m.) on day 45 consisted of 0.2-0.5 mg protein each in 1 ml Freund's incomplete adjuvant. Antisera were recovered from blood obtained by terminal exsanguinations on day 58.
2.2.9 Electrophoresis and Westem Blotting
SDS-PAGE and Westem blotting were done as described previously (Laemmli,
1970; Morrissey, 1981; Hardy et al, 1987; Towbin et al, 1979), using 8% gels. For
Westem blotting, proteins in gels were transferred to nitrocellulose by electroblotting.
For primary antibody probing, DO-Dl antisera were diluted 1/50,000 in TBST. Bound antibody was detected with HRP-conjugated secondary antibody (Biosource fritemational, Camarillo, CA) diluted 1/50,000 in TBST, and development by chemiluminescence (SuperSignal, Pierce Chemical Co.).
2.3 Results and Discussion
2.3.1 rDO-Dl Was Constmcted and Expressed in COS-1 Cultured Mammalian Cells
I initially planned to express various combinations of zonadhesin D-domains in cultured mammalian cells. The selected expression plasmid was pCMV-5, which is a highly efficient expression vector for cultured mammalian cells, especially in SV-40 transformed simian COS cell lines. The basic strategy was to digest the pig zonadhesin cDNA at appropriate restriction sites near the boundaries of D-domains to obtain different D-domain combinations. When appropriate sites were not present, PCR was
43 used to generate the desired zonadhesin cDNA inserts. Next, an insert encoding a signal peptide and a synthetic staffer encoding a hexahistidine tag and a stop codon were ligated to the 5'- and 3'-ends, respectively, of the zonadhesin fragments. The signal peptide was expected to direct secretion of the expressed protein into culture media, and from which it could be purified by Ni^"^ affinity chromatography. We expected to develop constmcts overlapping with each other by a single D-domain, and the ZP-binding activity of those constmcts would disclose the function of these single or combinations of D-domains.
Eventually, we would express the D0-D4 recombinant protein that is essentially equivalent to zonadhesin purified from spermatozoa but without undergoing proteolysis that produces p45 and pi05.
The first constmct with an insert containing DO and Dl domain was generated based on the strategy above (Fig. 2.1). This constmct included a PCR-generated signal peptide of receptor guanylyl cyclase-A at the 5'-end of DO-Dl fragment and a hexahistidine tag with a stop codon at the 3'-end of DO-Dl fragment. This pCMV/DO-Dl plasmid was transfected into cultured COS-1 cell by DEAE-dextran, and expressed rDO-
Dl protein in either cell extract or cultured media was detected on Westem blots using the DO-Dl antisera (Fig. 2.2). The DO-Dl antisera used here were developed against bacterially expressed protein containing DO-Dl domain, and they detect both p45 and pi 05. The reactivity of these antisera will be discussed in more detail in Chapter III. A broad Mr 60,000 immunoreactive band was observed in the extract of cells transfected with rDO-Dl (recombinant DO-Dl zonadhesin) plasmid, but was not present in control cells either untransfected or transfected with non-recombinant pCMV-5. The size of this protein was consistent with the molecular mass deduced from the DNA sequence of the
44 insert. Thus, the desired rDO-Dl protein was successfully expressed in COS-1 cells.
However, non-specific signal from semm proteins interfered with direct detection of secreted rDO-Dl in culture media.
2.3.2 rDO-Dl Was Purified from Cell Extract and Culture Media by Ni^"^ Affinity Chromatography
When designing the plasmid constmcts to express zonadhesin domain(s), we anticipated the need to purify the expressed proteins in a simple and efficient way. A hexahistidine tag was added right before the stop codon in the same reading frame with the zonadhesin DO-Dl fragment. Since the hexahistidine tag is able to bind specifically to
Ni *, I was able to purify the expressed rDO-Dl by one-step Ni^"^ affinity chromatography. Moreover, purifying expressed rDO-Dl protein from culture media might also overcome the interference from semm proteins that non-specifically bound to the DO-Dl antisera in attempts to detect expressed rDO-Dl in culture media. In cell extracts, the rDO-Dl protein (Mr 60,000) eluted primarily in column fraction 11 (Fig.
2.3), where the imidazole concentration is -400 mM. Thus the hexahistidine tag was expressed correctly in the rDO-Dl plasmid and enabled purificafion by Ni^"^ affinity chromatography. In culture media, a Mr 60,000 immunoreactive band was detected in column fractions 13, 14 and 15, where the imidazole concentration is 350 mM to 400 mM (Fig. 2.4b), the same point as in cell extract purification. This band was not observed in fractions of media from cells transfected with non-recombinant pCMV-5 (Fig. 2.4a).
Thus, rDO-Dl protein was secreted into culture media. However, the rDO-Dl protein was present in culture media at a very low concentration, suggesting that the signal peptide
45 used was not efficient to direct the secretion of rDO-Dl into culture media. Moreover, we could not exclude the possibility that the rDO-Dl protein came from the dead broken
COS-1 cells.
Because our ultimate goal was to target the zonadhesin recombinant protein with a transmembrane segment to the membrane for gain-of-fiinction experiments, it was essential to be able to drive secretion of the expressed proteins into culture media in the case of recombinant proteins that lacked a transmembrane segment. In addition, to determine the ftinction of single zonadhesin domains or domain combinations, the purified zonadhesin polypeptides needed to be in their native conformations to retain their ZP-binding activity. However, purification of expressed proteins that are trapped in the cultured cells usually requires harsh conditions that were likely to denature the expressed proteins. Therefore, it was essential that the expressed proteins be secreted into culture media in a large amount to facilitate their purification for fiiture functional study.
Although the GCA signal peptide worked well in many other expression systems, my result indicated that it is probably not suitable for directing secretion of zonadhesin polypeptides. Therefore, we designed another set of plasmid constmcts for directing secretion of expressed zonadhesin polypeptides.
2.3.3 Constmcts with Mucin Domain and Constmcts with von Willebrand Factor Signal Peptide Were Generated
A number of factors influence the targeting and secretion of expressed proteins in culttired mammalian cells, such as glycosylation, sulfation and oligomerization. At the time when I conducted the expression of zonadhesin domains, we did not know how
46 zonadhesin was processed to generate two known subunits (p45 and pi05) or whether glycosylation of the protein occurs in the ER or Golgi apparatus. The failure of secretion of expressed zonadhesin DO-Dl domains could result from the incorrect glycosylation of the recombinant proteins in the cultured cells. The sequence of tryptic fragments of p45 and pi05 gave us some clue about what domains those subunits probably contained; p45 was likely to include the DO and Dl domains, while pi05 probably spanned the D2 through D4 domains. Although both the MAM domains and the mucin-like domain are removed in post-translational processing of zonadhesin precursor, those removed domains, especially the usually O-glycosylated mucin-like domain, may be involved in the correct targeting of zonadhesin. Therefore, we added part of the mucin-like domain to two new constmcts (Fig. 2.5a). In addition to the part of the mucin-like domain, one constmct contained the DO and Dl domains (rM-Dl) and the other contained the DO, Dl, and D2 domains (rM-D2). Unfortunately, the presence of the mucin domain not only failed to direct secretion of expressed zonadhesin polypeptides into media, they made the constmcts un-expressible (data not shown).
Another possible cause of failure in secretion of expressed zonadhesin peptides is that the GCA signal peptide is not suitable for targeting of zonadhesin fragments in cultured COS-1 cells. I therefore replaced the GCA signal peptide with the signal peptide of vWF, because we expected this protein's D-domains to be stmcturally similar to those of zonadhesin, and the vWF signal peptide was shown to drive successfiil secretion of recombinant vWF expressed in cultured COS-1 cells. I generated a universal expression cassette plasmid with the human vWF signal peptide (N-terminal side) and a hexahistidine tag followed by a stop codon (C-terminal side) flanking Notl and Clal sites
47 for expression of various zonadhesin domains (Fig. 2.5b). I was then able simply to drop the different zonadhesin fragments into this cassette plasmid with the correct reading frame to express a variety of combinations of zonadhesin domains. The first zonadhesin domain I expressed was the Dl domain generated by PCR. The plasmid with Dl insert
(pv6H-Dl) was expressed successfully in COS-1 cells (Fig. 2.6). The antibody used to probe the Westem blot was an affinity-purified anti-Dl antibody that specifically recognizes the zonadhesin Dl domain and detects only p45. Its development and reactivity will be discussed in detail in Chapter III. However, we did not detect any expressed Dl peptide in the culture media, suggesting that the vWF signal peptide also did not work properly in directing secretion of zonadhesin recombinant proteins. 1 also tried expressing those zonadhesin domain constmcts in other cell lines, such as CHO cells and Hela cells, but the expected recombinant zonadhesin proteins also failed to express correctly in those cell lines.
Collectively, these cultured cell expression results demonstrated that our zonadhesin recombinant protein could be expressed in cultured mammalian cells, but some as yet unknown mechanism in the cell's protein trafficking system prevented the recombinant proteins from being secreted into culture media. Since we understand very little about zonadhesin's biochemical properties, it is almost impossible for us to find a way to solve the problem. Therefore, it is necessary to characterize the zonadhesin biochemically and functionally first. These studies also identified Hmitations in the utility of our DO-Dl antisera, so 1 developed several new reagents that are described in detail in
Chapter III and Chapter IV.
48 Xho Sal CiTCGAC GA'G'CTIG EcoRl
Kpnl BamH I
Fig. 2.1 A schemafic outline for constmction of rDO-Dl plasmid. The insert of the constmct comprises three parts: a GCA signal peptide generated by PCR, a zonadhesin cDNA fragment spanning the DO and Dl domains, and a synthetic hexahistidine tag followed by a stop codon. This insert was ligated into the multiple cloning site {KpnUBamHl) of the pCMV-5 plasmid.
49 Untransfected Transfected with Transfected with COS-1 Cell pCMV-5 Only pCMV-5/ D0-D1
1#
rD0-D1 r Mr 60,000 L
CMC M C M C = Cell Extract M = Culture Medium
Fig. 2.2 Detection of expressed rDO-Dl protein in transfected COS-1 cell extracts and culture media. Shown are Westem blots probed with the DO-Dl antisera. pCMV-5/DO-Dl or non-recombinant pCMV-5 transfected, as well as untransfected cells and media were harvested and loaded onto S% SDS-PAGE. An Mr 60,000 immunoreactive band is present in the extract of cells transfected with rDO-Dl plasmid, but not present in the extract of untransfected cells or of cells transfected with non-recombinant pCMV-5 vector. Thus the desired rDO- Dl protein was successfully expressed in COS-1 cells. However, non-specific signal from semm proteins interfered with direct detection of secreted rDO-Dl in the cell culture media.
50 Io UJ Fraction Number .OJ O 1 5 6 7 8 9 10 11 12 18
rD0-D1 Mr 60,000
Fig. 2.3 Purification of expressed rDO-Dl protein from cell extracts. Shown is a Westem blot probed with the DO-D1 antisera. Cellular proteins expressed by rDO-D 1 transfected COS-1 cells were separated by Ni^"^ affinity column. After extensive washing, bound proteins were eluted with a 10 ml linear 20-500 mM imidazole gradient. The collected column fractions were loaded onto S% SDS-PAGE and detected by the DO-Dl antisera. Note that the rDO-Dl protein eluted primarily in column fraction 11, where the imidazole concentration is -400 mM.
51 ^ Fraction Number
r® a ^67 8 9 10 11 12 13 14 15
rD0-D1^
^ Fraction Number O 6 7 8 9 10 11 12 13 14 15
rDO-Dl ^ ^rDO-Dl
Fig. 2.4 Purification of rDO-Dl protein from culture media. Shown are Westem blots probed with the DO-D 1 antisera. The culture media from non-recombinant pCMV-5 vector transfected cells {panel a) or pCMV5/D0-Dl transfected cells {panel b) were loaded onto a Ni ^ chelating chromatography column to purify secreted rDO-Dl from culture media. The bound proteins were eluted with a 10 ml linear 20-500 mM imidazole gradient. The collected fractions were analyzed by Westem blots probed with the DO-Dl antisera. Note that a Mr 60,000 band was detected in column fractions 13, 14 and 15, where the imidazole concentrafion is 350 mM to 400 mM, in the culture media from pCMV-5/DO-Dl transfected cells {panel b), but not in the corresponding fractions when media from cells transfected with non-recombinant pCMV-5 were fractionated under idenfical condifions {panel a). Thus, rDO-Dl protein appeared to be secreted into culture media, but only in very small amounts.
52 a rM-D1 plasmid
Kpn I Hind I EcoR I BamH I 226 bp I 2.0 Kb GCA M DO D1 6 His+Ter
rM-D2 plasmid
Kpn I Hind Ncol BamH 226 bp I 3.2 Kb GCA M DO D1 D2 6 His+Ter
pv6H cassette plasmid
EcoRl Notl Cla I BamH I I,— SPofvWF -; Domains of Zonadhesin i-6-His+Ter
pv6H-D1 EcoRl Notl Cla I BamH I I I SPofvWF D1 Domain -6-His+Ter
Fig. 2.5 Schematic outlines of constmcts containing a part of mucin domain and constmcts containing a von Willebrand factor signal peptide. Panel a shows two constmcts containing an extended GCA signal peptide, a part of the mucin domain and a hexahistidine tag. The upper constmct contains the DO and Dl domains (rM-Dl), while the lower constmct contains the DO, Dl, and D2 domains (rM-D2). Panel b shows a cassette plasmid containing the signal peptide of vWF and a hexahistidine tag between which any zonadhesin domain or domain combination can be dropped in {upper). The first constmct developed from this cassette plasmid contained the zonadhesin Dl domain (pv6H-Dl, lower).
53 M. 60,000-
Mr 40,000
Cell Extract Culture Media
Fig. 2.6 Detection of pv6H-Dl protein in cell exttacts and culture media. Shown is a Westem blot probed with an anti-Dl antibody (detail on development of this antibody is presented in Chapter III). The cell extract or culture media from cells transfected with pCMV-5 vector, pv6H-Dl or pCMV-5/DO-Dl were separated by %% SDS-PAGE. Note that Mr 60,000 and Mr 40,000 bands were detected in the rDO-Dl-transfected and pv6H-Dl-transfected cell extracts, respectively. Therefore, the rDO-Dl and pv6H-Dl proteins were successfiiUy expressed in COS-1 cells, but failed to be secreted into culture media.
54 CHAPTER III
HETEROGENEOUS PROCESSING AND ZONA
PELLUCflDA-BINDING ACTIVITY OF
PIG ZONADHESIN
3.1 Introduction
Adhesion of mammalian spermatozoa to the zona pellucida (ZP) is a complex process mediated by binding of sperm proteins to complementary ligands in the ZP
(Yanagimachi, 1994; Bi et al, 2002). The complexity of this process derives partly from cellular changes that occur during gamete interactions. Spermatozoa undergo physiological changes in the female reproductive tract that are required for fertilization and are collecfively called capacitation (Yanagimachi, 1994; Jaiswal and Eisenbach,
2002). Although the molecular basis of capacitation is only partly understood, in some if not all species avidity of sperm-ZP adhesion increases as capacitation progresses. After capacitation is completed, membranes involved in initial adhesion events are lost from the sperm surface during the acrosome reaction, but adhesion is sustained by interaction of newly exposed stmctures with the ZP (Yanagimachi, 1994; Bi et al, 2002). Unique adhesion molecule pairs likely function at different times during fertilization, and the activities of these molecules may change as fertilization progresses (Bi et al, 2002). It is therefore important to assess the biochemical and functional properties of sperm adhesion molecules at each stage in the fertilization process.
Several sperm proteins that may mediate adhesion to the ZP have been identified and characterized (Bi et al, 2002). Among these molecules zonadhesin is unique in its
55 ability to bind directly and in a species-specific manner to native, particulate ZP (Hardy and Garbers, 1994, 1995). Zonadhesin from pig (Hardy and Garbers, 1995), mouse (Gao and Garbers, 1998), rabbit (Lea et al, 2001), and human (Wilson et al, 2001) spermatozoa is a mosaic protein with a predicted Type I integral membrane topology. In each of these species, the protein's large extracellular region comprises primarily three domain types (MAM, mucin, and von Willebrand D [VWD]) that are present in other adhesion molecules (Wagner, 1990; Beckmann and Bork, 1993; Varki, 1994). Although the domain stmctures of zonadhesin from these four mammals have been predicted from cDNA sequences, relatively little is known about the proteins' biochemical and functional properties.
The active form of pig zonadhesin in membrane fractions of capacitated, epididymal spermatozoa is a two-chain molecule with disulfide-bonded Mr 105,000 and
Mr 45,000 polypeptides, both of which are derived from a predicted 2467 amino acid nascent precursor (Hardy and Garbers, 1994, 1995). High Mr forms of zonadhesin have also been observed, suggesting the possible formation of covalent oligomers (Hardy and
Garbers, 1994). This possibility was further implied by the presence in the pig zonadhesin Dl, D2, and D3 domains of a conserved CG(L/V)CG sequence motif (Hardy and Garbers, 1995) that is important for the oligomerization and proper function of von
Willebrand factor (Mayadas and Wagner, 1992) and for the oligomerization of porcine submaxillary mucin (Perez-Vilar and Hill, 1998a, b, 1999). These observations suggested that the protein at a minimum undergoes limited proteolysis and possibly also oligomerization as occurs in the functional maturation of vWF and other D-domain proteins (Wagner, 1990). However, it is unclear when during sperm maturation such
56 post-translational processing occurs, or whether it is important for zonadhesin's ZP- binding activity.
Here we report that heterogeneous post-translational processing gives rise to multiple isoforms of pig zonadhesin in freshly ejaculated spermatozoa. Among these, only forms comprising the pi05 and p45 polypeptides possess ZP-binding activity, and the monomeric pi05/45 form binds more avidly than do higher order covalent oligomers.
Furthermore, we find that zonadhesin binds uniformly to homologous ZP and localizes to the apical head of pig spermatozoa. These properties ftirther support a function for zonadhesin in sperm adhesion to the egg's extracellular matrix.
3.2 Materials and Methods
3.2.1 Isolation of Sperm Membrane Fraction
Freshly ejaculated boar semen was immediately mixed with an equal volume of an extender designed especially for pig semen (AndroHep). The rest of the procedure was usually performed within two hours after ejaculation. The extended semen was passed through four layers of cheesecloth to remove debris. Sperm motility was checked by light microscopy, and it normally was above 80%. The boar semen was subsequently diluted with an equal volume of PBS (phosphate-buffered saline, containing 10 mM NaP04, 150 mM NaCl, pH 7.4) and then subjected to centriftigation at 300 x g for 15 min at room temperature. The pelleted spermatozoa were resuspended in HNE/DFP (20 mM
NaHEPES, 1 mM NaEDTA, 130 mM NaCI, 0.5 mM DFP, pH 7.5, 0°C). Subsequently a steel Parr bomb was used to dismpt sperm membranes by cavitation after being kept at
57 650 psi N2 pressure for 30 min on ice (Haden et al, 2000). Sperm membranes from the dismpted cells were separated from the rest of the sperm components by a two-step differential centtifugation. The first centriftigation was performed at 1,500 x g for 10 min at 2°C. The pellet was discarded, and the supematant suspension was subjected to the second centriftigation at 300,000 x g (Beckman Ti-70.1) for 45 min at 2°C. The cmde membrane pellet was resuspended in 10 ml of ice cold HE/DFP (20 mM HEPES, ImM
NaEDTA, 0.5 mM DFP, pH 7.5) by Dounce homogenization (20 strokes) on ice, and washed once by centrifugation at 300,000 x g (Beckman Ti-70.1) for 45 min at 2°C
(Hardy and Garbers, 1994, 1995; Haden et al, 2000). The membrane pellet was resuspended in HE/DFP buffer and stored at -80°C. In some experiments, solutions also contained 1 mM iodoacetamide to inhibit thiol proteases and to prevent thiol oxidation.
3.2.2 Isolation of Zona Pellucida
Porcine ZP were isolated from sliced ovaries by stepwise sieving through screens
(Dunbar et al, 1980), and then further purified by ultracentriftigation through Percoll
(Amersham Pharmacia) gradients (Hardy and Garbers, 1994). After sieving sliced ovaries through four screens with mesh sizes of 1 mm, 210 [im, 140 f^m and 70 p,m, the oocytes were recovered from the 70 [im screen with homogenization buffer (HB; 25 mM triethonolamine, 150 mM NaCl, 1 mM MgCl2, 1 mM CaCb, pH 8.5) containing 0.25 mg/ml hyaluronidase, 0.25 mg/ml DNase I, 1% Triton X-100 and 0.5 mM DFP. The oocytes were then Dounce homogenized 10 strokes with the tight pestle on ice. Sodium deoxycholate was added to a final concentration of 1%, and the suspension was incubated
58 on ice for 30 min. The oocyte homogenate was mixed with isotonic Percoll to produce a
50%, (v/v) Percoll suspension and ultracentrifuged for 40 min at 90,000 x g (35,000 rpm,
Beckman 70.1 Ti rotor) at 2°C. A single, sharp ZP band formed at approximately 50%
from the top of the Percoll gradient. The ZP band was recovered with a glass pipette, washed at least 1000-fold with 1% (w/v) Triton X-100 in buffer HNE (20 mM HEPES,
130 mM NaCl, ImM EDTA, pH 7.5) by centriftigation (2,500 x g, 10 min, 2°C), resuspended in 1% CHAPS/HNE at 1-4 mg of protein/ml, and stored frozen at -20°C.
The ZP preparation was assessed with phase contrast microscopy to verify its composition. The ZP preparation contained mostly ZP fragments with very little adherent granular material.
3.2.3 Zona Pellucida-Binding Assays
Detergent-solubilized proteins from sperm membrane fractions were mixed with isolated ZP, and zonadhesin that bound directly to the particulate, native ZP was detected either by Westem blotting (Hardy and Garbers, 1994, 1995) or by epifluorescence. For localization of binding sites, sperm proteins were biotinylated (Hardy and Garbers, 1994) prior to solubilization and incubation with ZP. The ZP with bound sperm proteins were washed extensively with 20 mM NaHEPES, 0.5 M NaCl, 1 mM EDTA, 1%, (v/v) Triton
X-100, 0.5%o (w/v) Na deoxycholate, 0.1% SDS, pH 7.5 (mRIPA) (Hardy and Garbers,
1994), and then bound, biotinylated proteins were detected by incubating for 15 min at
22°C with Texas Red-labeled streptavidin (Molecular Probes) diluted 10,000-fold in 10 mM Tris-HCl, 150 mM NaCl, 0.1% (v/v) Tween 20, pH 7.5 (TBST). After washing 3x5
59 min with TBST, ZP were dropped on coverslips, air dried, mounted with Fluoromount G
(Electron Microscopy Sciences), and viewed by epifluorescence. ZP-bound forms of zonadhesin were characterized by Westem blotting. Biotinylated zonadhesin polypeptides that remained bound to ZP after washing with mRIPA were detected by probing blots with HRP-streptavidin (Hardy and Garbers, 1994). Altematively, zonadhesin polypeptides that remained bound after washing with 1% CHAPS/HNE were
detected by probing blots with specific antisera as described below.
3.2.4 Expression and Purification of DO-Dl Fusion Protein
The 1.7 kb EcoRl fragment of pig zonadhesin cDNA clone M2 (in pBluescript)
was subcloned into the EcoRl site of pET-23d. The 5' sticky end of this fragment came
from the ^coRI site in the adapter used to constmct the cDNA library (Hardy and
Garbers, 1995), and its 3' sticky end from the EcoRI site at nucleotides 4063-4068 of the
zonadhesin composite cDNA (Genbank Accession # U40024). This construct specified
an Mr 64,000 fusion protein comprising 20 amino acids of N-terminal vector-encoded
protein, 19 amino acids of C-terminal vector-encoded protein (including a hexahistidine
tag), and zonadhesin amino acids Pro'^^^-Ser'^^^ The ftision protein was expressed in E.
coli strain BL21/DE3 by induction with 0.5 mM isopropylthiogalactoside (IPTG) for 2 h
at 37°C, and isolated from inclusion bodies by preparative SDS-PAGE and
electroelution.
60 3.2.5 Preparation of DO-Dl Antisera
Asp-Pro bonds of the purified DO-Dl fusion protein were hydrolyzed for 36 h with 70%o formic acid at 37°C. The final hydrolysates contained a mixture of proteins with MrS corresponding to partial hydrolysis products predicted from the deduced amino acid sequence, including an Mr 33,000 core polypeptide. Hydrolysates were lyophilized to remove formic acid prior to injection. Two female NZW rabbits were immunized
(i.m.) with 0.2-0.5 mg protein each in 1 ml Freund's complete adjuvant (=day 0). Booster injections (i.m.) on day 45 consisted of 0.2-0.5 mg protein each in 1 ml Freund's incomplete adjuvant. Antisera were recovered from blood obtained by terminal exsanguinations on day 58.
3.2.6 Expression and Purification of Dl and D3 Fusion Proteins
Glutathione-S-transferase (GST) fusion proteins comprising in part amino acids
Ser^^^-Met^^^ of the Dl domain or amino acids Ile'^^^-Pro'^^^ of the D3 domain were expressed in E. coli strain BL21. PCR products (Dl sense primer: 5'-AGTGGATCCA
GCACCTTCTCTGG-3';D1 antisense primer: 5'-ATAGAATTCTGCTAGGCCGTGT
TG-3'; D3 sense primer: 5'-CATCGGATCCCAGGTCAAGTTTGACGG-3'; D3 antisense primer: 5'-GGGGAATTCTAGGCCGCCTG-3'; underiined bases denote mismatches introduced to create restriction sites and stop codons) encoding the Dl and
D3 domain segments were directionally cloned into the BamUl and ^coRI sites of pGEX-2T. The correct reading frame was confirmed by DNA sequencing using primers flanking the inserted DNA fragments. Fusion protein expression was induced with 0.1
61 mM IPTG at 37°C for 2 h. After washing the bacteria with 10 mM NaP04, 150 mM
NaCl, pH 7.4 (PBS), soluble fusion proteins were extracted by sonicating cell pellets in
PBS containing 0.5 mM DFP, 1.0 mM EDTA, 10 |aM E-64 and 0.2 % Triton X-100. Cell lysates were applied to a glutathione (GSH)-Sepharose column (15 ml bed volume) equilibrated at 22°C in PBS. Non-binding proteins were washed through with PBS, and fusion proteins were eluted with 5 mM GSH in 50 mM Tris-HCl, pH 8.0. Eluted fusion proteins were present at concentrations of 5-8 mg/ml in the pooled, peak fractions, and with prior disulfide bond reduction migrated as single bands in SDS-PAGE (10%o gels).
Total yields of purified fusion protein were 40-45 mg/500 ml culture.
3.2.7 Preparation of Domain-Specific Antisera
Four female NZW rabbits were immunized (i.m.) with 1 mg purified fusion protein/animal (two with GST-Dl, two with GST-D3) emulsified in 0.5 ml Freund's complete adjuvant (=day 0). Booster injections consisted of 1 mg purified fusion protein/animal emulsified in 0.5 ml Freund's incomplete adjuvant (i.m.) on day 42, and 1 mg soluble protein/animal in PBS (s.c.) on days 49 and 70. Antisera were recovered from blood obtained by terminal exsanguinations on day 81.
3.2.8 Preparation of GST. GST-Dl, and GST-D3 Affinity Columns
Purified GST (100 mg), GST-Dl (70 mg), and GST-D3 (70 mg) were dialyzed at
4°C for >16h in 0.1 M NaHCOs, 0.5 M NaCl, pH 8.3 to remove GSH and exchange into conjugation buffer. Dialyzed proteins were each coupled at 10 mg/ml swelled gel to
62 CNBr-activated Sepharose 4B (Amersham Pharmacia). After washing by suction on a glass filter to remove uncoupled proteins, remaining activated groups were blocked with
1 M ethanolamine, and the conjugated resins were washed with three cycles of ahemating pH (0.1 M sodium acetate, 0.5 M NaCl, pH 4.0 and 0.1 M NaHCOs, 0.5 M NaCl, pH
8.3). The affinity matrices were poured into 1 cm diameter glass columns, equilibrated in
PBS containing 0.02%, NaNs, and then stored at 4°C.
3.2.9 Affinity Purification of Dl and D3 Antibodies
Antibodies to GST were removed by passing 20 ml of antisera through a 10 ml bed volume GST-Sepharose column equilibrated at 22°C in PBS. Antibodies to zonadhesin Dl or D3 domains were then affinity purified from their anti-GST-depleted sera by chromatography on GST-Dl or GST-D3 columns respectively (7 ml bed volume each, equilibrated in PBS at 22°C). Elation of bound antibodies with 0.2 M NaCitrate,
0.15 M NaCl, pH 3.0 was monitored continuously by A28o- Peak fractions were pooled and immediately adjusted to pH 7 by addition of 1 M Tris (unbuffered). Antibodies to
GST that were removed in the initial depletion steps were similarly eluted from GST-
Sepharose and recovered for use as affinity-purified control antibodies. All purified antibodies were stored at -70°C.
3.2.10 Preparafion of D3 Immunoaffmity Column
Twenty mg of affinity-purified antibody to the D3 domain (11.4 mg from rabbit
#RI28 and 8.6 mg from #R129) were desalted into 0.1 M NaHCOs, 0.5 M NaCl, pH 8.3
63 (=coupling buffer) in two mns on four tandem 5 ml HiTrap desalting columns
(Amersham Pharmacia). Desalted protein (15 mg in 8.8 ml) was coupled to 0.43 g (dry weight) of freshly swollen CNBr-activated Sepharose 4B. Protein assay of uncoupled protein confirmed that more than 95% of the antibody (>14.5 mg) was coupled to the affinity matrix (1.5 ml packed volume), which after blocking and washing as for the fiasion protein affinity matrices was equilibrated in PBS containing 0.02%, NaNs and stored at 4°C.
3.2.11 Immunoaffmity Purification of Zonadhesin from Spermatozoa
Sperm membrane fractions (100 mg protein) were Dounce homogenized in 10 ml of 1%, SDS/HNE and incubated for 30 min at 22°C. The homogenate was diluted to 100 ml with HNE containing 0.5 mM DFP, 0.56%, (w/v) sodium deoxycholate, and 1.1%
(v/v) hydrogenated Triton X-100 to produce the composition of mRIPA, a detergent solution in which zonadhesin retains its ZP binding activity (Hardy and Garbers, 1994).
After ultracentrifugation for 1 h at 100,000 x g, 2°C, up to 50 ml of the supematant solution (=mRIPA extract) was applied to a 1.5 ml anti-D3 column, and non-bound proteins were washed through with mRIPA until A280 (monitored continuously) retumed to baseline. The column was ftirther washed with 10 ml HNE containing 1%, (v/v) hydrogenated Triton X-100, and then protein was eluted with 10 ml of 0.2 M NaCitrate,
0.15 M NaCl, l%o (v/v) hydrogenated Triton X-100, pH 3.0. Eluted protein fractions containing purified zonadhesin were pooled, adjusted to pH 7 with 1 M Tris (unbuffered), and stored at -70°C.
64 3.2.12 Preparation of Antisera to Zonadhesin Holoprntf^in
Two female NZW rabbits were immunized (i.d.) with 100 ^g purified zonadhesin
holoprotein/animal emulsified in 1 ml Freund's complete adjuvant (=day 0). After
boosfing with 100 p.g purified protein/animal emulsified in 1.2 ml Freund's incomplete
adjuvant (i.m.) on day 45, antisera were recovered from blood obtained by terminal
exsanguinations on day 60.
3.2.13 Site-Directed Mutagenesis
Mutations in the pGEX-2T constmct encoding GST-D3 were introduced with T4 polymerase-based GeneEditor (Promega Corp.), and those in the constmct encoding
GST-Dl were introduced with PCR-based QuickChange (Stratagene hic). The primer for generating the double C^S mutant of the C'™''GVC'^'^G motif in GST-D3 (=C1709S,
C1712S) was 5'-ACGGAAGGACCTCCGGCGTGAGCGGGAACTTCA-3' (altered bases in mutagenesis primer sequences are underlined). Primers for generating mutants of the C^^GLC^^^G motif in GST-Dl were: 5'-CTCTGGCAAACTTCTGGTCTCTGTGG
CG-3' (C933S sense); 5'-CGCCACAGAGACCAGAGAGTTTGCCAGAG-3' (C933S antisense); 5'-CTCTGTGGTCTCAGTGGCGACTATGACGG-3' (C936S sense); 5'-
CCGTCATAGTCGCCACTGAGACCACAGAG-3' (C936S antisense); 5'-CTCTGGCA
AACTCTCTGGTCTCAGTGGCGACTATG-3' (C933S, C936S sense); 5'-CATAGTCG
CCACTGAGACCAGAGAGTTTGCCAGAG-3' (C933S, C936S antisense). The procedures to generate mutants mostly followed the instmctions of the QuickChange Kit.
The nicked vector DNA incorporating the desired mutation was transformed into
65 elecfrocompetent DH-lOB cells by electroporation. Mutated constmcts were verified by double stranded DNA sequencing using insert-flanking pGEX primers (sense: 5'-
AATCGGATCTGGTTCCG-3'; antisense: 5'-CGTCAGTCAGTCACGAT-3'). The selected pGEX-2T plasmid containing the mutated Dl antigenic region was transformed into BL-21. Mutant fusion proteins were then expressed and purified by GSH affinity chromatography as for the wild type proteins.
3.2.14 Electrophoresis and Westem Blotting
SDS-PAGE and Westem blotting were done as described previously (Laemmh,
1970; Towbin et al, 1979; Morrissey, 1981; Hardy et al, 1987). For two-dimensional
SDS-PAGE, samples were first loaded onto gradient (4-12%) tube gels (diameter = 1mm) and mn at a constant current of 0.5 mA per tube. Upon completion of first dimensional electrophoresis, the tube gel was carefully recovered from the glass tube with a special syringe (Hoefer Inc.), and then was put on the top of 4-12%, gradient slab SDS-PAGE gels. Melted 1%, agarose containing l%o SDS and 100 mM DTT was then added on top.
The slab gel (1.5 mm thick) was mn at a constant current of 20 mA. The reducing agent
DTT moved into the gel and reduced proteins in the sample (Hardy and Garbers, 1994).
The purpose of using gradient gels (4-12%, linear) instead of straight percentage gels was to resolve both high Mr zonadhesin isoforms and their constituent polypeptides. For
Westem blot, proteins were ttansferred from SDS-PAGE gels to nitrocellulose by electroblotting. Subsequent antibody treatments were performed in TBST (10 mM Tris-
HCl, pH 7.5, 150 mM NaCl, 0.1% Tween 20), including both washing and antibody dilution. Dilutions used to detect zonadhesin on Westem blots of pig sperm membranes
66 were: DO-Dl antisera-1/50,000; Dl antibody-1/5,000; D3 antibody-1/50,000, holoprotein antisera-1/50,000, each in TBST. Bound antibody was detected with HRP-conjugated secondary antibody (Biosource International) diluted 1/50,000 in TBST, and development by chemiluminescence (SuperSignal, Pierce Chemical Co.).
3.2.15//; F;Yro Multimerization
Wild type and mutant ftision proteins were purified by GSH affinity chromatography in the presence of ImM DTT to stabilize the expressed proteins in their monomeric forms. The purified ftision proteins were then desalted on Sephadex G50 spin columns equilibrated in 50 mM Tris-HCl, pH 7.4 containing ImM DTT. For time course studies, oxidized glutathione (GSSG) was added to desafted fusion protein (1 )ig in Tris-
DTT) to a final concentration of 25 mM, which upon reaction with DTT produced a 24 mM GSSG/2 mM GSH redox buffer with an effective potential in solution (Eh) at pH 7.4 of-246 mV. Reactions were incubated at 37°C, and disulfide bond formafion was terminated at various times by adding iodoacetamide to 60 mM. To determine dependence of multimer formation on Eh, 5-25 mM GSSG was added to reactions to produce redox buffers ranging from -269 to -246 mV Eh. After incubating for 2 hours at
37''C, reactions were terminated with iodoacetamide as for time course studies. Multimers in the terminated reactions were separated by SDS-PAGE and detected by staining with
Coomassie Blue.
67 3.2.16 Indirect Immunofluorescence
All steps for immunolocalization experiments were done at 22°C. Spermatozoa were recovered from pig epididymides and washed with HNE buffer as described previously (Hardy and Garbers, 1994). For immunofluorescence, cells were smeared on coverslips, air dried, then fixed in 100% methanol for 30 min. After blocking 30 min with
10%, (v/v) heat-inactivated goat semm (HIGS) in PBS, coverslips were floated 1 h on DO-
Dl antisera diluted 1/400 in PBS/HIGS. After washing coverslips with PBS, bound antibody was detected by incubating 1 h with Texas Red-conjugated antibody to rabbit immunoglobulin (Biosource Intemational) diluted 1/400 in PBS/HIGS. After a final wash with PBS, coverslips were mounted with Fluoromount G and viewed by epifluorescence and phase contrast microscopy.
3.3 Results and Discussion
3.3.1 Zonadhesin Binding Sites Are Evenly Distributed on the Entire Zona Pellucida
Zonadhesin from membrane fractions of capacitated, epididymal spermatozoa bound directly and with high affinity to intact ZP (Fig. 3.1). The bound zonadhesin comprised pi05 and p45 polypeptides (Fig. 3.1a) as previously observed (Hardy and
Garbers, 1994). Although earlier work established the species specificity of this interaction, the distribution of zonadhesin binding sites in the pig ZP has not been characterized. We therefore visualized ZP-bound zonadhesin in situ by affinity fluorescence (Fig. 7>.lb 8LC). Zonadhesin protein was detected on the entire ZP, indicating that its binding sites were not regionalized in the ZP stmcture. In addition, the
68 relative evenness of the labeling suggested the binding sites are intrinsic to the ZP and not associated with adherent materials from the cumulus cell matrix or other potential contaminants of the ZP preparation.
3.3.2 Antibodies Were Developed to Characterize Zonad hesin Isoforms
The locations of pi 05 and p45 tryptic peptides in the sequence of the pig zonadhesin precursor indicated that p45 comprises in part the Dl domain, and that pi05 comprises in part the D2 and D3 domains (Fig. 3.2) (Hardy and Garbers, 1995). To detect zonadhesin isoforms in spermatozoa and to characterize their polypeptide compositions, we prepared various domain-directed antisera and affinity-purified antibodies. The precursor's deduced sequence specified numerous potentially antigenic regions, including segments located in approximately the same positions within the Dl and D3 domains that exhibited high predicted hydrophilicity, surface probability, and flexibility (Fig. 3.2).
Accordingly, we raised antisera to a Gene 10 fusion protein spanning the DO and Dl domains (P^^-'-S'^^'*), as well as to two GST fusion proteins comprising in part the short, antigenic segments identified in the Dl and D3 domains (Ser^^^-Met^^^ and fle'^^'^-Pro'^^^ respectively; Figs. 3.2 and 3.3). We also raised antisera to whole zonadhesin isolated from membrane fractions of pig spermatozoa (Fig. 3.3).
The DO-Dl fusion protein was recovered from inclusion bodies and purified by preparative SDS-PAGE (Fig. 3.3a). Rabbit anfisera to this recombinant protein recognized the vector-encoded amino acids at the N- and C-termini of the fusion protein but did not cross-react with zonadhesin (not shown). We therefore hydrolyzed the
69 purified fusion protein at Asp-Pro bonds to remove the short, vector-encoded peptides.
The Asp-Pro hydrolysis products corresponded to those predicted by the precursor sequence, thereby confirming that the hydrolysate contained the desired zonadhesin fragments (Fig. 3.3a). This hydrolysate was used to prepare antisera to the DO-Dl domains.
In contrast to the DO-Dl fusion protein, the GST-Dl and GST-D3 proteins were expressed in soluble form and purified by chromatography on GSH-Sepharose (Fig.
3.3a). Antisera to these expressed, recombinant proteins were fractionated by chromatography on Dl and D3 affinity columns to produce reagents for specific detection of these domains. The DO-Dl antisera and the Dl antibody each recognized both the DO-Dl and GST-Dl ftision proteins but not the GST-D3 protein (Fig. 3.3a).
Similarly, the D3 antibody recognized the GST-D3 protein but not the other two ftision proteins. The absence of cross-reactivity with other D-domains indicated that these reagents were suitable for domain-specific detection of zonadhesin polypeptides.
We used the D3 antibody to prepare an immunoaffmity column for purification of zonadhesin from membrane fractions of freshly ejaculated spermatozoa (Fig. 3.3b).
Without prior reduction of disulfide bonds, the purified zonadhesin migrated in SDS-
PAGE primarily as an Mr 150,000 protein (pi50) and a less prominent Mr >300,000 protein. When protein disulfides were reduced, the same zonadhesin preparation migrated primarily as p45, pi05, and an Mr 300,000 polypeptide (p300). This purified zonadhesin preparation was used to raise antisera to the fiilly processed and disulfide-bonded holoprotein, and to confirm the reactivity and specificity of antisera and affinity-purified antibodies (Fig. 3.3b).
70 Figure 3.4 shows the reactivity of the four immunoreagents with zonadhesin isoforms on Westem blots of membrane fractions from pig ejaculated spermatozoa.
Antisera to the hydrolyzed DO-Dl protein recognized primarily pl50 on blots of non- reduced proteins (Fig. 3.4a). In contrast to this relatively weak interaction with a single, disulfide-bonded fomi of zonadhesin, the DO-Dl antisera detected several polypeptides of disulfide-reduced zonadhesin, including the pi 05 and p45 components previously described, as well as an Mr 300,000 protein (p300) and at least two other polypeptides of intermediate size (designated p60-90). The reaction of the DO-Dl antisera with p45 was consistent with the presence of p45 tryptic peptides in the Dl domain (Fig. 3.2). The reaction of this antisemm also with pi 05 indicated that the N-terminus of pi 05 is likely upstream of S'^^"* (the C-terminus of the expressed DO-Dl fragment). The absence of the p60-90 polypeptides from the holoprotein purified by D3 immunoaffinity chromatography (compare to Fig. 3.3) indicates that these DO-Dl-reactive polypeptides neither contain nor are covalentiy associated with a D3 polypeptide.
Like the DO-Dl antisera, the Dl- and D3-specific antibodies also detected pl50 zonadhesin in non-reduced sperm proteins (Fig. 3.4a). However, in contrast to the complex pattem of disulfide-reduced proteins the DO-Dl antisera recognized, the Dl antibody recognized only p45. Similarly, the D3 antibody bound primarily to pi05 and more weakly to an Mr 60,000 polypeptide. Affinity-purified control antibodies to GST did not bind significantly to sperm proteins (not shown). The antisemm to the zonadhesin holoprotein detected pi50 in non-reduced sperm proteins, but primarily recognized proteins with Mr »300,000. Overall, the DO-Dl antisera reacted much more weakly with non-reduced forms of zonadhesin than it did with the reduced, constituent polypeptides.
71 In contrast, the Dl and D3 antibodies bound similariy to both non-reduced and reduced zonadhesin, and the antisemm to the zonadhesin holoprotein reacted strongly with non- reduced zonadhesin (Figs. 3.3b and 3.4a), but did not recognize the protein's separated, disulfide-reduced polypeptides. The differential binding of our antibodies to reduced and non-reduced zonadhesin likely reflects stmctural properties determined in part by the many intramolecular disulfide bonds present in von Willebrand D-domains (Marti et al,
1987). The cross-reaction of the antibodies to the GST-Dl and GST-D3 proteins with non-reduced zonadhesin extracted from spermatozoa, both on blots and in the isolation of the holoprotein by immunoaffinity chromatography, ftirther suggested that the tertiary stmctures of these soluble fusion proteins are similar to that of the native protein.
3.3.3 Multiple Zonadhesin Forms Are Present in Freshly Ejaculated Spermatozoa
Because of the complicated pattem detected by the DO-Dl antisera under disulfide-reduced conditions and the tendency of forming disulfide bonds among zonadhesin subunits, two-dimensional SDS-PAGE was employed to resolve zonadhesin oligomers. The first dimension was under nonreducing conditions, while the second dimension was under disulfide-reduced conditions. This approach is a powerful method to resolve the complicated associations of proteins that possess disulfide bonds. Proteins lacking any type of disulfide bond (intermolecular or intramolecular) reside on the diagonal of the gel, as long as gels used in both dimensions have the same percentage of polyacrylamide. Those proteins that form intermolecular disulfide bonds appear under the diagonal, while those proteins that form intramolecular disulfide bonds migrate above the
72 diagonal because the intramolecular disulfide bonds make them more folded and therefore more mobile than their fully unfolded conformations.
Two-dimensional SDS-PAGE revealed that the p60-90 zonadhesin polypeptides migrated with the same mobility in each dimension (i.e., on the gel diagonal), and are therefore not covalentiy bound to other polypeptides. In contrast, pi05 and p45 were components of Mr 150,000, 300,000, and >900,000 complexes stabilized by intermolecular disulfide bonds (Fig. 3.4b). The presence of both pl05 and p45 in the Mr
150,000 non-reduced protein (pi50) indicated that this zonadhesin form comprised primarily one each of the two polypeptides, consistent with both polypeptides' being derived by proteolysis from a single precursor molecule. Similar compositions of the Mr
300,000 and >900,000 non-reduced proteins, in particular the presence of mostly pl05 and p45 in a ratio similar to that of pi 50, suggested these large complexes are covalent dimer and hexamer respectively of pi 50 (Fig. 3.4b). The relative amounts of the Mr
150,000, 300,000, and >900,000 zonadhesins did not change substantially when 1 mM iodoacetamide was included in membrane isolation buffers to inhibit potentially artifactual thiol-disulfide exchange (data not shown). Thus, the Mr 300,000 and 900,000 zonadhesins likely exist as tme multimers in the sperm cell. Other VWD domain proteins form oligomers that are functionally important. For example, the absence of the largest vWF multimers causes a mild clotting deficiency distinct from other forms of von
Willebrand disease (Ruggeri and Zimmerman, 1980), presumably because the higher inherent avidity of muftivalent interactions is necessary for proper clot formation.
Although vWF multimerizes extensively (Spom et al, 1986; Mayadas and Wagner,
1989), porcine submaxillary mucin preferentially forms trimers in transfected cells
73 (Perez-Vilar et al, 1998; Perez-Vilar and Hill, 1998a). The apparent absence in spermatozoa of zonadhesin multimers other than dimer and hexamer indicates its oligomerization, like that of porcine submaxillary mucin, is more limited and specific than the mulfimerization of vWF.
3.3.4 The pi 05/45 Monomeric Form of Zonadhesin Binds Preferentially to the Zona Pellucida
To test the ZP-binding activity of the zonadhesin isoforms present in ejaculated cells, we solubilized membrane fractions of spermatozoa in 1% CHAPS/HNE, incubated the extract with particulate ZP, and then removed non-binding proteins by sequential washing with CHAPS/HNE and mRIPA. Under these conditions, zonadhesin was the only protein from membrane fractions of capacitated, epididymal spermatozoa that remained bound to the ZP (Fig. 3.1) (Hardy and Garbers, 1994). In contrast to previous
ZP-binding studies that used biotinylated sperm proteins, however, in this experiment
ZP-bound zonadhesin was detected on Westem blots with the DO-Dl antisera. Only zonadhesin forms comprising the pl05 and p45 polypeptides bound to the ZP (Fig. 3.5a), and these pi05/45 zonadhesin molecules were separable by anion exchange chromatography from other forms that lacked ZP binding activity (Fig. 3.5b). In two- dimensional SDS-PAGE (Fig. 3.6), the predominant ZP-bound form of pi 05/45 zonadhesin migrated with Mr 150,000 in the first dimension (disulfides not reduced). The
Mr 300,000 (dimer) and >900,000 (hexamer) proteins also bound to the ZP, but a significant proportion of these zonadhesin forms remained in the non-binding fraction
(Figs. 3.6a & 3.6^?). Thus the pl05/45 monomeric form of zonadhesin bound
74 preferentially to the ZP in vitro, although pi05/45 multimers did also possess ZP-binding activity.
3.3.5 Cysteines Other than Those in the CG("L/V)CG Motif Can Mediate Spontaneous Multimerization of Purified Zonadhesin Fragments In Vitro
At high concentrations in storage, the purified Dl and D3 ftision proteins formed viscous gels that liquefied upon addition of 10 mM DTT. This observation suggested that the fusion proteins, which each contained the CG(L/V)CG sequence motif (Fig. 3.7a), had spontaneously formed intermolecular disulfide bonds even though a mild reductant was present (the 5 mM GSH used to elute them from GSH-Sepharose). SDS-PAGE without prior reduction of disulfides revealed that covalent multimers were indeed present in stored preparations of both ftision proteins, but not in identically stored GST
(Fig. 3.1b). Including DTT in isolation buffers preserved the proteins in their monomeric forms (Fig. 3.7c). Addition of oxidized DTT at concentrations up to 100 mM, which raised the effective potentials in solution (Eh) at pH 7.4 as high as -292 mV, did not induce formation of covalent multimers in vitro (not shown). However, addition of 25 mM oxidized glutathione (GSSG) to produce a -246 mV redox buffer induced rapid formation of disulfide-bonded multimers of both proteins (Dl, Fig. 3.7c; D3 not shown).
Most of the Dl ftision protein was converted to multimers within 30 min, and multimer formation continued until very little monomeric protein remained (Fig. 3.7c). To determine if the vicinal cysteines in the CG(L/V)CG sequence motif were important for multimer formation, we compared the multimerization kinetics of C^S mutants to those
75 of the wild type proteins. Single mutants of the Dl protein (C''"GLCG^SGLCG and
CGLC G-^CGLSG) formed multimers at the same apparent rates as the wild type proteins (not shown). The double mutant of the Dl protein (C^"GLC'^^''G^SGLSG) also formed multimers at -246 mV with the same kinetics as the wild type protein (Fig. 3.7c).
Furthermore, the multimerization of the wild type and double mutant Dl fusion proteins were indistinguishable at various Eh ranging from -269 mV to -246 mV (Fig. 3.1d).
Neither the wild type nor the mutant protein multimerized significantly at -269 mV, but multimer formation was clearly evident when Eh was raised to -259 mV. Double mutation of the C'™''GVC''''^G motif in the D3 fusion protein to SGVSG also did not perturb multimer formation (not shown). Collectively, these results demonstrate that the expressed Dl and D3 fragments of pig zonadhesin spontaneously form intermolecular disulfide bonds, and that the reaction is dependent on Eh but not on the vicinal cysteines in the CG(L/V)CG sequence motif known to be important for vWF multimer formation.
Our D1 and D3 fragments contained two and three additional cysteines, respectively, downstream of the two in the CG(L/V)CG motif, one or more of which must form cystine in the multimerization of these proteins. Studies on porcine submaxillary mucin have identified half-cystines that are not in the CG(L/V)CG motif but are nonetheless important for the dimerization of fragments (Perez-Vilar and Hill, 1998b) or multimerization of full D-domains (Perez-Vilar and Hill, 1998a). Nevertheless, when they are expressed in eukaryotic cells, C^X mutants of the CG(L/V)CG motif in vWF
(Mayadas and Wagner, 1992) or in submaxillary mucin (Perez-Vilar and Hill, 1998a) generally exhibit defects in multimer formation. Our results show that cysteines other
76 than those in the CG(L/V)CG can mediate spontaneous muhimerization of purified zonadhesin fragments //; vitro. Further studies, including testing whether mutants of downstream cysteines (C"^ and C^^^) in GST-Dl form muhimers, will be necessary to determine if our results reflect differences in methods or tme differences in multimer formation between zonadhesin and other D-domain proteins.
3.3.6. Zonadhesin Localizes to the Apical Heads of Pig Spermatozoa
The DO-Dl antisera detected pig zonadhesin on the apical heads of methanol- fixed, epididymal spermatozoa (Fig. 3.8). All cells were strongly labeled when they were prepared by methanol fixation, which both denatures proteins and dismpts membranes.
Thus pig zonadhesin is present in the anterior head of pig spermatozoa, which is a part of the cell that interacts with the ZP. However, because methanol fixation dismpts cell membranes, we cannot discem from this experiment whether the pig protein is present on the sperm cell surface. These results are consistent with those reported by Gao and
Garbers (1998), who used antisera to the unique, partial D domain repeats of mouse zonadhesin to detect the protein on the apical heads of paraformaldehyde-fixed spermatozoa.
3.3.7 Summary and Discussion
These studies are relevant both to the potential function of zonadhesin in mammalian fertilization and to the functions of VWD domains in diverse proteins. To our knowledge no other sperm protein exhibits the physicochemical heterogeneity of
77 zonadhesin. In ovine spermatozoa, hyaluronidase is present as a remarkably heterogeneous mixture of disulfide-bonded multimers (Harrison, 1988a, b; Harrison and
Gaunt, 1988). Nevertheless, even this protein does not undergo the combined proteolytic removal of protein domains, generation of specific constituent polypeptides, and multimerization that occur in the processing of zonadhesin. In addition, hyaluronidase does not exhibit such marked variation in all species (Hunnicutt et al, 1996), whereas we find that zonadhesin is heterogeneous in spermatozoa from all mammals examined
(Chapter V). Our detection of differences in ZP binding activity among the various zonadhesin forms suggests that the heterogeneous processing of this protein is functionally important, just as proteolytic activation and heterogeneous multimerization are important for the proper function of vWF (Wagner, 1990). Unlike vWF muhimers, however, zonadhesin multimers appear to bind less avidly than the monomer. Thus multimerization of zonadhesin could represent a mechanism for storing the protein in a latent form that can be activated when it is required for interaction with the ZP, or it could reflect an additional function of the protein as a scaffold or other stmctural element of the sperm head. Further studies will be required to determine if zonadhesin's ZP binding activity is dynamically regulated during fertilization, for example, as a component of sperm capacitation.
These results described in this chapter were recently published (Hickox et al.,
2001).
78 a
p105- ^Kr
p45- m
df- /b
Fig. 3.1 Direct binding of zonadhesin to intact zona pellucida. Panel a: biotinylated polypeptides in a membrane fraction of capacitated, epididymal spermatozoa that bound to the pig ZP. CHAPS-solubilized sperm proteins were incubated with particulate ZP. The pig ZP with bound proteins were washed sequentially with CHAPS/HNE and mRIPA, and bound, biotinylated proteins were detected on Westem blots (10%, SDS-PAGE, disulfides reduced) by probing with HRP- streptavidin as described previously (Hardy and Garbers, 1994, 1995). Note that the pi 05 and p45 major polypepfides of zonadhesin remained bound under these conditions. Peptides comprising a minor portion of the bound zonadhesin (Hardy and Garbers, 1994) migrated at the dye front (df). Panel b: epifluorescence image of zonadhesin bound to the pig ZP. Bound, biotinylated zonadhesin on intact ZP from the experiment shown in panel a was detected in situ by probing with Texas Red-streptavidin. The labeled ZP was then viewed by fluorescence microscopy. Note that the zonadhesin-derived fluorescence is uniform, and associated with all regions of the ZP fragments. Panel c: DIC image of the ZP shown in panel b.
79 Amino acid number 1 T I T T T 400 800 1200 1600 2000 2400
p45 peptides pi 05 peptides Mill I I N I Wl |DO D1 D2 D3 TM Expressed fragments ^ D0-D1 D3 Dl
3.4-
Antigenic Index iiMii^iiiiiwii^ Jameson-Wolf
Surface Probability Emini
4fflBIHilBMBIIIIICZM»OHI3-«iIltffflBIBtHiH^^ Flexible Regions Karplus-Schuiz ,|j/||tly<^ Hydrophilicity Kyte-Doolittle
Fig. 3.2 Domain stmcture, polypeptide composifion, and predicted characteristics of the pig zonadhesin precursor. The N-terminal domain (designated N) is composed of one partial and one full MAM domain, whereas this region of mouse and human zonadhesins comprises three tandem MAM domains. Vertical bars mark the locations in the precursor of tryptic peptides that were previously isolated from p45 and pi05 and sequenced (Hardy and Garbers, 1995). Note that p45 must include much of the Dl domain, and that pi 05 includes most or all of the D2 and D3 domains. Horizontal bars denote segments in the tandem VWD domains that were expressed as ftision proteins for production of antisera. Two potentially antigenic segments in the Dl and D3 domains (short horizontal bars) exhibited high predicted hydrophilicity, flexibility, and surface probability.
80 a -A^' •S cS Sliver G' stain Western blots
Mr 60,000 m
Coomassie stains anti- anti- anti- antl- D0D1 Dl D3 holo
Silver stains Western blots NR R NR R NR R NR R NR R
wm -p300 Ii -
pi 50- •M fr-* -p105 '• - ••• m
fit -p45 - •>
. —^ antl-DODI anti-D1 anti-D3 antl-holo
Fig. 3.3 Production and characterization of zonadhesin antisera and antibodies. Panel a: preparation and specificity of antisera and antibodies to recombinant fusion proteins. Shown are protein stains and Westem blots of SDS-PAGE (10%o gels, protein disulfides reduced) as indicated. Purified, recombinant zonadhesin ftasion proteins (lanes labeled G10-D0D1-H6, GST-Dl, and GST-D3) each migrated primarily as single, Coomassie blue stained bands in overloaded gels. To generate the immunogen for production of anfisera to the DO-Dl domains, the purified Gene 10 fusion protein spanning p^^^.s'224 ^f pjg zonadhesin (lane labeled G10-D0D1-H6), was partially hydrolyzed at Asp-Pro bonds (lane labeled DP-hydrolyzed). The three predominant bands visible in the hydrolyzed preparation (asterisks) corresponded to P'^''-D"^' (51,500 Da, upper band), a mixture of P™-D"''^ and P^O^.DH?! (42,500 and 42,300 Da, respectively, middle band), and pSO^-D""^ (33,200 Da, lower band). GST ftision proteins (lanes
81 Fig. 3.3 Continued
labeled GST-Dl and GST-D3) were used to prepare affinity-purified antibodies to segments spanning Ser^^^-Met^"^^ of the pig zonadhesin Dl domain and Ile'^^"- Pro of the D3 domain. Specificity of the antisera and antibodies was determined by Westem blotting mixtures of the ftision proteins. Arrowheads mark the locations of the three fusion proteins in a mixture of 50 ng each partially pure protein (lane labeled "Silver stain"). Note that the anti-DO-Dl antisera and the anti-Dl antibody readily recognized the G10-D0D1-H6 and GST-Dl proteins but not GST-D3 in mixtures of 10 ng each protein (lanes overlined with "Westem blots"). Similarly, the anti-D3 antibody recognized GST-D3 but not the other fusion protein. Note also that antisera to the purified, disulfide-bonded zonadhesin holoprotein did not recognize the disulfide-reduced fusion proteins imder these conditions. All developed blots for panel a were exposed to film for 40 s. Panel b: composition and immunoreactivity of processed zonadhesin holoprotein purified from membrane fractions of pig ejaculated spermatozoa. Shown are protein stains and Westem blots of SDS- PAGE [4-12%, linear gradient gels, protein disulfides either reduced (lanes labeled "R") or not reduced (lanes labeled "NR") as indicated]. Note the different mobilities of the purified protein's constituent polypeptides when separated without and with prior reduction of disulfide bonds (lanes overlined with "Silver stains"). Note also that the different antisera and antibodies each recognized a unique subset of the disulfide reduced and non-reduced polypeptides (lanes overlined with "Westem blots", containing 220 ng zonadhesin/lane). The developed anti-DO-Dl, anti-Dl, and anti-D3 blots of panel b were exposed to film for 40 s, whereas the anti-holoprotein blot was exposed for 5 s.
82 a NR NR R NR R NR * I
pi 50 p150- p150- •pi 05
-p45
Probe: antl-D0-D1 anti-Dl anti-D3 anti-holoprotein
Disulfides not reduced (4%-8%)
I I IgM IVIyosin IgG 900 kDa 200 kDa 150 kDa
Fig. 3.4 Idenfification of multiple zonadhesin forms in membrane fractions of freshly ejaculated spermatozoa. Proteins were separated by SDS-PAGE, Westem blotted, and zonadhesin forms were detected with antisera or affinity-purified antibodies. Panel a: Reaction of DO-Dl antisera, Dl antibody, D3 antibody, and zonadhesin holoprotein antisera with zonadhesin separated on 4-12% linear gradient gels. Note the differences in polypeptides recognized depending on whether protein disulfides were reduced (lanes labeled "R") or not reduced (lanes labeled "NR"), and the Mr^300,000 disulfide non-reduced protein (asterisks) detected by the D3 antibody and the holoprotein antisera. Panel b: Reaction of DO-Dl antisera with zonadhesin polypeptides separated by two-dimensional SDS-PAGE. Note the presence of p60-90 on the diagonal, and the migration in the first dimension of pl05 and p45 as Mr 150,000, 300,000, and 900,000 complexes.
83 <4^r <^ ^
p105
1r
p45
Fig. 3.5 Differential binding of zonadhesin forms to the zona pellucida. Panel a: Binding of zonadhesin from membrane fractions of uncapacitated cells to the ZP. Proteins were separated by SDS-PAGE (10%, gel), Westem blotted, and zonadhesin was detected with the DO-Dl antisera. Note that of the several zonadhesin polypeptides present in the detergent-solubilized membrane preparation (lane labeled "Extracf), only pi 05 and p45 were components of the protein that bound to the ZP (lane labeled "ZP+bound"). The apparent immunoreactivity present between pi05 and p45 in the "ZP+bound" lane was non-specific signal that was also evident in a mock binding reaction that contained ZP but no sperm proteins (lane labeled "ZP only"). Panel b: Separation of zonadhesin forms by ion exchange chromatography. Zonadhesin comprising the pi05 and p45 polypeptides eluted at 0.2 M NaCl (lane labeled "DEAE pool I"), whereas other forms that lacked ZP-binding activity eluted at >0.5 M NaCl (lane labeled "DEAE pool II").
84 Disulfides not reduced (4%-8%)
(A -*» o (A
Q. C o a Mr : >900,000 300,000 150,000 Disulfides not reduced (4%-8%) (A c_ -t, w a. c o a Mr : >900,000 300,000 150,000 Fig. 3.6 Differential binding of zonadhesin muhimers to the zona pellucida. Binding experiments were performed as for Fig. 3.5a. Proteins in the non-binding and ZP-bound fractions were separated by two-dimensional electrophoresis, and zonadhesin forms were detected on Westem blots with the DO-Dl antisera. Panel a: Non-binding fraction. Panel b: Binding fraction. Note that only zonadhesin forms comprising the pi05 and p45 polypeptides exhibited ZP-binding activity. Note also that the proportion of pl05/45 monomer (Mr 150,000 when disulfides were not reduced) that bound to the ZP was higher than those of the other pi05/45 forms (dimer and hexamer). 85 a (S_SJ> ^ SGSSTFSGKLCGLCGDYDGD• IRCQVKFDGRGFLEVEIPKA ~~, c: SSNDNQKPDGSPAKDEKELG• YYGRTCGVCGNFNDEEEDEL• MMPSDALALDDVMYVDSWRD •SSWQTSEDADQQCEENQVSP• ~i ^ KEIDPNCQEDDRKTEAESQE -^ C PSCNTA-COO- ^ QPSANCRPADLERAQEQCQAA-GOO Dl Fusion Protein DS Fusion Protein #ol_o^ ^ multimers •»• _36kDa- «f8i»- -26kDa- Disulfides Reduced Not reduced Wild type (CGLCG) mulfimers GST-Dl- {33kDa) Double mutant (SGLSG) 25mMGSSG - - + + + t + Time 0 18h 15' 30' 60' 90' 120' d Wild type Double mutant ,i»»lil»«. til [GSSG] 0 5 10 15 25 0 5 10 15 25 (mM) Fig. 3.7 Multimerization of Dl and D3 fiision proteins in vitro. Panel a: Amino acid sequence of Dl and D3 fusion proteins. The conserved CG(L/V)CG sequence motif is underlined. Panel b: Spontaneous multimerization of purified GST-DI and GST-D3 in storage. Proteins were separated by SDS-PAGE (10% gel, silver stained) with (left gel) or without (right gel) prior reduction of disulfide bonds. 86 Fig. 3.7 Continued Note the high Mr multimers present in the stored Dl and D3 preparations, but not in identically stored preparations of GST. Panel c: Time course of Dl wild type and double mutant (C^^GLC^^^G^SGLSG) multimerization at -246 mV Eh. The fusion proteins were purified in the presence of 1 mM DTT to inhibit premature multimer formation. GSSG (25 mM) was then added to the purified proteins (upper gel=wild type, lower gel=double mutant) to initiate multimer formation. Reactions were terminated with iodoacetamide at intervals ranging from 15 min to 18 h as shown, and multimers were detected by SDS-PAGE (10%, gels). Note that GSSG induced significant multimer formation within 30 min, whereas in the absence of GSSG neither protein formed multimers even after prolonged incubation (compare GSSG -, Time 0 to GSSG -, Time 18 h). Panel d: Dependence of multimer formafion on Eh. Purified GST-Dl wild type and double mutant proteins were incubated with various concentrations of GSSG as indicated to promote multimer formation. Eh ranged from -269 mV (at 5 mM added GSSG) to -246 mV (at 25 mM added GSSG). Reactions were terminated as for panel c, and multimers were detected by SDS-PAGE on a 4-12% linear gradient gel. Note the presence of multimers in all reactions containing >10 mM added GSSG (Eh >259 mV). 87 Fig. 3.8 Light microscopic localization of zonadhesin in pig epididymal spermatozoa with the DO-Dl antisera. Shown is a phase contrast/epifluorescence (Texas Red) double exposure image of localization by immunofluorescence on methanol- fixed cells. Note the strong labeling on the apical heads of all cells. 88 CHAPTER rv TEMPORAL AND FUNCTIONAL CHARACTERIZATION OF ZONADHESIN PROCESSING AND LOCALIZATION 4.1 Introduction The previous chapter reported several aspects of zonadhesin's biochemical properties and activity. Availability of zonadhesin antibodies facilitated ftirther study of zonadhesin's processing and ZP-binding activity. The specific reactivity of anti-Dl and anti-D3 antibodies confirmed that developing antibody against antigenic regions predicted by the Jameson-Wolf algorithm is a practical approach for obtaining zonadhesin domain-specific antibodies. This strategy became more valuable after we leamed that zonadhesin undergoes post-translational processing to generate more than two final, mature forms. Two-dimensional SDS-PAGE probed with the DO-Dl antisera revealed that a protein migrating around Mr 300,000 participated in formation of high Mr multimers with zonadhesin's two subunits, p45 and pl05 (Fig. 3.4b). Data obtained from tryptic fragment sequencing and reactivity of anti-Dl and anti-D3 antibodies suggested that p45 and pi 05 include only VWD domains but no MAM or mucin-hke domains (Fig. 3.2). Thus, a question arose about the fate of those domains during post-translational processing. This question was especially important because both MAM domains and mucin-like domains in other proteins are involved in cell-cell or cell-extracellular matrix interaction. The fate of those two domains in zonadhesin might therefore be related to 89 zonadhesin's ZP-binding function. Accordingly, it became necessary to determine whether p300 contains zonadhesin MAM or mucin-like domains. Mucin-like domains are normally glycosylated by O-linked oligosaccharides. If p300 contained a mucin-like domain, one of its properties would be O-linked glycosylation. In addition, the cDNA sequences of both pig and mouse zonadhesin reveal that some N-linked glycosylation sites are present in their VWD domains. Therefore, it is very likely that zonadhesin undergoes glycosylation. As all of the major components of the ZP are glycoproteins, and their carbohydrates appear to be functionally important, it is interesting to determine whether their potential binding partner is also glycosylated. Furthermore, it would help to put forward a binding mechanism between zonadhesin and the ZP. Therefore, specific deglycosylation experiments should be performed to determine whether zonadhesin polypeptides are glycoproteins, and if they are, what types of oligosaccharide (O-linked or N-linked) are attached to those polypeptides. In addition, a complete deglycosylation would make it possible to determine the precise molecular mass of each zonadhesin subunit. Spermatozoa are not fertilization ready when they first pass from testis to epididymis. The epididymis can be divided into three regions in the order of their distance from testis: caput, corpus, and cauda. Spermatozoa in the caput epididymis are almost incapable of motion. Mature spermatozoa are stored in the caudal epididymis before they are ejaculated. In most mammalian species, only those spermatozoa in the caudal epididymis are able to move actively and travel through the female reproductive ttact to reach the egg (Yanagimachi, 1994). Transit through the epididymis is a critical step for sperm maturation. Furthermore, the redox environment in the epididymis 90 promotes formation of disulfide bonds in many molecules, such as some proteins in the nucleus (Yanagimachi, 1994). Because disulfide bond formation is related to zonadhesin's ZP binding activity, determining the oligomeric status of zonadhesin components in different regions of the epididymis will be helpfiil to understanding the process of zonadhesin's functional maturation. Moreover, it is also interesting to leam whether zonadhesin's ZP binding activity changes correspondingly during sperm transit through the epididymis. Zonadhesin's ultrastmctural locahzation in spermatozoa is cmcial for its function in fertilization. Although zonadhesin was originally found in sperm membrane preparations, the biochemical methods (cavitation) used to isolate the sperm plasma membrane cannot completely avoid contamination with acrosomal membranes. Indirect immunofluorescence has long been a typical approach to determine the localization of proteins on the membrane. However, due to the complicated membrane stmcture in spermatozoa, especially the closely apposed plasma and outer acrosomal membranes, immunofluorescence cannot provide high enough resolution to judge whether the sperm plasma membrane is broken. The most direct approach would be labeling live spermatozoa instead of labeling fixed spermatozoa, but this approach also mns the risk of artifacts from cell dismption. Compared to regular immunofluorescence, immunoelectron microscopy is able to provide the resolution required for ultrastmctural localization of zonadhesin on sperm. In this chapter, I report the results achieved in these aspects of zonadhesin's stmcture and function. We compared the post-translational modifications and binding activity of pig zonadhesin from testicular, epididymal, and ejaculated spermatozoa. The 91 p45 and pi 05 polypeptides of zonadhesin active forms, comprising primarily the Dl and D2-D3 domains respectively, were modified with N-linked oligosaccharide. A p300 polypeptide comprising in part the predicted mucin domain was modified with O-linked oligosaccharide. Detergent extraction experiments identified both a membrane-associated zonadhesin pool and an extraction-resistant pool that was solubilized only by denaturants. The proportion of both multimeric zonadhesin and active zonadhesin in the extraction- resistant fraction increased as spermatozoa matured in the epididymis. However, the sizes of zonadhesin polypeptides were the same at all sperm maturation states. Thus, the protein's major post-translational processing events occur largely in the testis, but assembly of the processed protein into an extraction-resistant stmcture is not complete until spermatozoa reach the caput epididymis. N-terminal sequencing of p45 and pi05 identified Asp-Pro bonds in tandemly repeated VWD domains of the zonadhesin precursor as potential cleavage sites in the protein's functional maturation. Zonadhesin in ejaculated cells localized by immunoelectron microscopy to the outer acrosomal membrane and outer aspect of the acrosomal matrix. In spermatids zonadhesin exhibited a unique, "omega-like" localization pattem over the developing acrosome of round spermatids. Thus, during sperm maturation the zonadhesin precursor is extensively modified by glycosylation and unusual proteolytic events. The modified protein is assembled into particulate stmctures in the distal parts of the acrosome where in the inifial stages of acrosomal exocytosis it could mediate species-specific adhesion to the zona pellucida. 92 4.2 Materials and Methods 4.2.1 Development of Anti-N Antiserum Preparation and use of some antibodies (anti-Dl and anti D3) and antisera (anti- DO-Dl and anti-holoprotein) for detecting zonadhesin were described in Chapter 111. A new domain-specific antibody (anti-N antibody) against an antigenic region in the pig zonadhesin MAM domains was developed using similar procedures. First, an antigenic region was selected using the Jameson-Wolf algorithm (Lasergene software, DNASTAR, Inc.). A PCR product (sense primer: 5'-TTGGATCCTGTGAAGAGAGCTTTC-3'; antisense primer: 5'-GGGCCGGAATTCCTATCTGAAAGACTGG-3') encoding this predicted antigenic region (C''*^-R^'^) was then directionally cloned into the BamHl and EcoRI sites of pGEX-2T. After the plasmid was transformed into E.coli stain BL21 by electroporation, expression of GST fiision protein was induced with 0.1 mM IPTG. After washing the bacteria with 10 mM NaP04, 150 mM NaCl, pH 7.4 (PBS), soluble ftision proteins were extracted by sonicating cell pellets in PBS containing 0.5 mM DFP, 1.0 mM EDTA, 10 ^M E-64 and 0.2% (v/v) Triton X-100. Cell lysates were applied to a glutathione (GSH)-Sepharose column (15 ml bed volume) equilibrated at 22°C in PBS. Non-binding proteins were washed through with PBS, and ftision proteins were eluted with 5 mM GSH in 50 mM Tris-HCl, pH 8.0. Eluted fusion protein migrated as single band in SDS-PAGE (10%, gels) under disulfide-reduced conditions. To develop antisera against this fusion protein, two female NZW rabbits were immunized (i.m.) with 1 mg purified fiision protein/animal emulsified in 0.5 ml Freund's complete adjuvant (=day 0). Booster injections consisted of 1 mg purified fiision protein/animal emulsified in 0.5 ml 93 Freund's incomplete adjuvant (i.m.) on day 42, and 1 mg soluble protein/animal in PBS (s.c.) on day 56. Antisera were recovered from blood obtained by terminal exsanguinations on day 70. 4.2.2 Deglycosylation A deglycosylation kit from Prozyme, Inc was used to deglycosylate zonadhesin. The procedure basically followed that specified by deglycosylation kit manual. For the analysis of glycosylafion type by Westem blotting, 5 |al of sperm membrane extract (-35 fag of protein) was diluted in 25 [il of HE (50 mM HEPES, 1 mM EDTA). 10 p,l 5X reaction buffer (0.25 M sodium phosphate, pH 7.0) and 2.5 fil denaturation solution (2% SDS, 1 M P-mercaptoethanoI) were then added and mixed gentiy. The solution was heated to 100°C for 5 min and cooled to room temperature, and reactions (50 [il final) were started by adding 2.5 |il 2% Triton X-100 solution and different combinations of deglycosylation enzymes, including PNGase F, sialidase A, Endo-0-Glycosidase, P(l-4) galactosidase, and Glucosaminidase (1 |al each). The deglycosylation reaction was performed at 37°C for 16 h. Deglycosylated zonadhesin polypeptides were detected by Westem blotting with domain-specific antibodies (anti-Dl, anti-D3, and anti-N). For time course analysis, 19.8 ^g of affinity-purified zonadhesin (450 |al) was concentrated by TCA precipitation. The pellet was re-dissolved with 2% (w/v) SDS, and then incubated with enzymes as described above. At intervals (30min, 1 h, 2 h, 4 h), 10 fxl aliquots were removed from the reaction solution, and the reaction was terminated by adding an equal volume of 2x SDS-PAGE sample buffer and immediate freezing. After SDS-PAGE, 94 separated proteins were stained first with Coomassie blue, and then double stained with silver. 4.2.3 Collection of Spermatozoa from the Epididymis Four testes were removed from two 6-month old boars and kept on ice for 30 minutes for transport to the lab. Epididymides were removed from the testes with surgical scissors. Caudal spermatozoa were flushed (retrograde) from caudal epididymis and released into HNE (20 mM HEPES, 130 mM NaCl, 1 mM EDTA, pH 7.5). Caput and corpus epididymides were separated from each other and then minced with surgical scissors. The minced tissues were kept in HNE and shaken gently for 20 minutes. Tissue debris was removed by passing spermatozoa through two layers of cheesecloth. The motility of spermatozoa was checked by light microscopy. Testes were cut into pieces and frozen immediately at -80°C. 4.2.4 Detergent Extraction of Spermatozoa and Testis Tissue Spermatozoa ejaculated or collected from different regions of the epididymis were washed once with HNE. After pelleting by a brief spin (300 x g, 10 min), the cells were first extracted with 10 volumes of 1% (v/v) Triton X-100 solution (1% Triton X-100, 150 mM NaCl, 1 mM EDTA, 10 |iM E-64, 0.5 mM DFP, 10 mM Tris-HCl, pH 7.5) for 10 min at 22°C, then centrifuged at 3,000 x g for 15 min at 22°C. The supematant solution was collected as Triton X-100 extract, and the pellet was re-extracted with 10 volume of 2% (w/v) SDS solution (2% SDS, 1 mM EDTA, 10 |aM E-64, 0.5 mM DFP, 10 mM Tris- 95 HCl, pH 6.8) following the same procedure. After centrifugation, the supematant solution of the re-extraction was collected as SDS extract. These detergent extracts of spermatozoa from different maturation states were then analyzed by Westem blotting and probed with various zonadhesin antibodies or antisera. To extract proteins from testis tissue, 100 mg of frozen testis tissue were ground into powder. The ground, frozen tissue was then sequentially extracted with 1 ml of detergent solution (Triton X-100 or SDS) to extract proteins. For detergent extraction of denuded spermatozoa, the sperm pellet from cavitation was extracted first with 10 volumes of l%o (v/v) Triton X-100 solution and then re-extracted with 10 volumes of 2%, (w/v) SDS solution. 4.2.5 Zona Pellucida Binding Assay The direct binding assay using particulate ZP was originally developed for the discovery of zonadhesin in pig spermatozoa (Hardy and Garbers, 1994, 1995). This assay has proved powerful in functional studies of pig zonadhesin. Briefly, 20 p.g isolated pig ZP (8 |al) in 1% Triton X-100/HNE was centriftiged at 15,000 x g for 1 min to remove excess buffer. 200 |J,1 of Triton X-100 extract of spermatozoa (200 \xg of protein) was then mixed with the ZP pellets, and the resultant suspension was rocked for 1 hour at 22°C. ZP fragments with bound sperm protein were washed quickly at 22°C with 3 x 200 li.1 of wash buffer (1% Triton X-100/HNE or Ix mRIPA) on a manifold equipped with a 96-well 0.45 |am hydrophilic membrane filtration plate. The washed ZP/sperm protein complexes were then solubilized in 40 ^1 Ix SDS-PAGE sample buffer (no reducing 96 agent), and separated by one-dimensional or two-dimensional SDS-PAGE, and characterized by Westem blotting with specific antibodies. 4.2.6 Immunoprecipitation Proteins in sperm membrane preparations were biotinylated on amino groups with sulfosuccinimidyl-6-(biotinamido) hexanoate (Pierce Chemical Co.), then solubilized at 10 mg/ml in l%o (w/v) SDS. The solubilized, biotinylated proteins were then diluted to Img/ml with HNE containing 1%, (v/v) Triton X-100, 0.5%, (w/v) sodium deoxycholate, 10 \xM E-64, and 100 \xM aprotinin (Boehringer Mannheim). To preclear, 1 [il of irrelevant antisemm (against an insect protein) was added to 400 |al of diluted extract, and the mixture was incubated for 1 h at 0°C. Protein A-Sepharose (10 |il of a 50% slurry; Amersham/Pharmacia) was then added, the suspension was rocked for 1 h at 4°C, and the beads were collected by centriftigation (10,000 x g, 15 sec, 4°C). The precleared supematant solution was then used for immunoprecipitation with anti-Dl antibody (5 p-l), anti-D3 anti-body (1 jil), anti-N antisemm (1 ^1), or holoprotein antisemm (1 |il). Protein A-Sepharose was then added and the suspension rocked as for preclearing. The beads from preclearing and from immunoprecipitation were each washed by centrifugation with 3x 1 ml of HNE containing 1%, (v/v) Triton X-100, and bead-bound immune complexes were eluted with 50 ^1 of 1 x SDS-PAGE sample buffer. Biotinylated proteins were detected on Westem blots by Streptavidin-peroxidase and chemiluminescence development. 97 4.2.7 Electrophoresis and Westem Blotting SDS-PAGE and Westem blotting were done as described in Chapter III (Hardy and Garbers, 1994). Gradient gels (4-12% linear) were used for each dimension to resolve both high Mr zonadhesin isoforms and their constituent polypeptides. Dilutions used to detect zonadhesin on Westem blots of pig sperm membranes were: Dl antibody- 1/5,000; D3 antibody-1/50,000, anti-N antisemm-1/50,000, holoprotein antisera- 1/50,000, each in TBST. In some experiments, a combination of Dl antibody and D3 antibody was used to detect p45 and pi05 simultaneously on one blot. To obtain similar strengths of signal, the anti-Dl antibody was diluted 1/5000, whereas the anti-D3 antibody was diluted 1/100,000. Bound antibody was detected with HRP-conjugated secondary antibody (Biosource Intemational) diluted 1/50,000 in TBST, and development by chemiluminescence. 4.2.8 Bead Adhesion Assay Sperm membrane proteins (~ 2 mg) were solubilized in 800 |il l%o CHAPS/HNE (1% CHAPS, 20 mM HEPES, 130 mM NaCl, ImM EDTA, 10 |aM E-64, and 100 |iM aprotinin, pH 7.5), and then evenly divided into two microcentrifuge tubes (400 ]il each). Particulate ZP (-10 fig in 4 ^1) was added to each tube. After constant rocking for 2 h at 22°C, the ZP was separated from unbound sperm membrane protein by centrifuging at 15,000 x g for 15 min at 4°C. The ZP pellet was washed 3 times with 3 x l%o CHAPS/HNE and then resuspended in 400 ^il of 1% CHAPS/HNE. One aliquot was rocked constantly with added 2 p,l anti-holoprotein antisera; the other was rocked with 2 98 |il unrelated antisera (against an insect protein) to work as a negative control. After the ZP pellet was washed with 3x1% CHAPS/HNE, 1 ^1 solution containing suspended fluorescent beads (1 [am in diameter, PolySciences, Inc) conjugated with secondary antibody (goat anti-rabbit) was added. The mixture was rocked for 1 h, and then the ZP was separated from the more buoyant fluorescent beads by a brief spin (~ 10 sec). The ZP with adherent beads was viewed by epifluorescence and phase contrast microscopy. 4.2.9 Indirect Immunofluorescence All steps for immunolocalization experiments were done at 22°C. Freshly ejaculated pig spermatozoa were washed once with HNE (Hardy and Garbers, 1994), smeared on coverslips, air dried, then fixed in 100%, methanol for 30 min. After blocking 30 min with 10% (v/v) heat-inactivated goat semm (HIGS) in PBS, coverslips were floated 1 h on zonadhesin holoprotein antisemm diluted 1/400 in PBS/HIGS. After washing coverslips with PBS, bound antibody was detected by incubating 1 h with Texas Red-conjugated antibody to rabbit immunoglobulin (Biosource Intemational) diluted 1/400 in PBS/HIGS. After a final wash with PBS, coverslips were mounted with Fluoromount G and viewed by epifluorescence and phase contrast microscopy. 4.2.10 Immunoelectron Microscopy Both post-embedding and pre-embedding immunolabeling approaches were employed to determine ultrastmctural localization of zonadhesin by electron microscopy. 99 For the post-embedding immunolabeling approach, spermatozoa were fixed at 4"C with 4%o formaldehyde, 0.25%o glutaraldehyde in 0.1 M sodium phosphate buffer, pH 7.4. The spermatozoa were then rinsed in phosphate buffer, dehydrated through an ethanol series and embedded in LR White Resin. Thin sections were mounted on nickel grids and blocked in Tris-saline (150 mM NaCl, 25 mM Tris-HCl, pH 8.0, 0.1% Tween 20) containing 2.5% BSA, 5% goat semm, and 0.1% fish gelatin. The sections were then incubated in a 1/100 dilution of zonadhesin holoprotein antisemm or non-immune semm diluted in blocking solution for approximately one hour. After three rinses in blocking solution, the grids were incubated in 10 nm gold-conjugated, affinity-purified, goat anti- rabbit IgG for one hour. The grids were then rinsed in Tris-saline, followed by PBS, and then fixed with 1% glutaraldehyde in 0.1 M sodium phosphate buffer. The sections were finally rinsed with water and stained with uranyl acetate and lead citrate. For the preembedding immunolabeling approach, pig spermatozoa extended in AndroHep were processed the day after collection. The spermatozoa were centrifiiged for 5 minutes in a clinical centrifuge and the pellet resuspended in PBS. Two different treatments were then utilized. For intact spermatozoa, an aliquot of sperm suspension was diluted into 2 volumes of 2%, glutaraldehyde in 0.1 M sodium phosphate, pH 7.4 and fixed on ice for 20 min. For permeabilized spermatozoa, the spermatozoa were diluted into 5 volumes of 0.25%, Triton X-100 in PBS and extracted on ice for 20 min. The permeabilized spermatozoa were centriftiged, rinsed in PBS, and then fixed in 2% glutaraldehyde in O.I M NaP04, pH 7.4 for 20 min on ice. After the pretreatments, both intact and permeabilized sperm samples were processed idenfically as following procedures: (a) rinse in PBS containing 50 mM glycine to block free aldehydes; (b) block 100 in the same blocking solution as in the post-embedding protocol; (c) incubate in primary antibody diluted by blocking solution as above; (d) rinse in blocking solution (3x); (e) incubate in secondary antibody; (f) rinse in Tris-saline and then PBS; (g) fix sperm pellet in 1% glutaraldehyde in 0.1 M sodium phosphate; (h) post-fix in 1%, OSO4 in phosphate buffer; and (i) dehydrate, embed in epon resin, stain thin sections with uranyl acetate and lead citrate. The immunolabeled sperm samples were then viewed with electron microscope. Between each incubation step the spermatozoa were pelleted by a brief gentle centriftigation of 30 seconds at 3,000 x g in a microcentrifiige. 4.3 Results and Discussion 4.3.1 p300 Is a Processed Zonadhesin Gene Product That Includes a MAM Domain As described in Chapter III, the DO-Dl antisera were developed by immunizing rabbits with bacterially expressed protein spanning the DO and Dl domains of pig zonadhesin. These antisera detected not only both p45 and pi05, but also a Mr 300,000 band (p300) under disulfide-reduced conditions on a low percentage SDS-PAGE gel (Fig. 3.3b). Two-dimensional SDS-PAGE Westem blot (l"-D: disulfides not reduced; 2"'^-D: disulfides reduced) demonstrated that this Mr 300,000 band was covalentiy bound to p45 and pl05 in the form of high Mr disulfide-bonded multimers (> Mr 900,000) (Fig. 3.4b). In addition, a protein with the same Mr was also observed on silver stained SDS-PAGE (disulfides reduced) loaded with purified zonadhesin (Fig. 3.3b), indicating that this protein is related to zonadhesin. The sequence of p45 and pi05 tryptic fragments (Hardy and Garbers, 1995) and the reactivity of the anti-Dl and anti-D3 antibodies (Fig. 3.2) 101 indicated that p45 and pi 05 are composed largely of D domains. However, the fate of the MAM and mucin-like domains is unknown. One possibility is that these domains are removed and digested when the zonadhesin precursor undergoes post-translational processing, while an altemative possibility is that they remain present in an unknown zonadhesin gene product that fails to bind to the ZP. To clarify this issue, an antigenic region (predicted by the Jameson-Wolf algorithm) located in the MAM domains was bacterially expressed as a GST ftisionprotei n that was subsequently purified by GSH affinity chromatography, following the same procedure for generation of Dl and D3 antigenic fiision proteins. The anti-N antisemm was developed by immunizing rabbits with the purified MAM domain ftision protein. The reactivity of this antisemm was tested by probing Westem blots of pig sperm membrane extract (Fig. 4.1). Under disulfide- reduced conditions, the anti-N antisemm detected a band around Mr 300,000 and several bands with Mr between 40,000 and 65,000. Under non-reducing conditions, a single band around 350,000 was detected (Fig. 4.1a). A 2-D (l"-D: disulfides not reduced; 2"'^-D: disulfides reduced) Westem blot probed with anti-N showed that this Mr 300,000 band is also involved in formation of high Mr disulfide-bonded multimer (Fig. 4.1b). This is consistent with what we observed previously detected with the DO-Dl antisera (compare to Fig. 3.4b), indicating that the Mr 300,000 band detected by anti-N antisemm is the p300 protein. To test whether this Mr 300,000 band (p300) detected by the anti-N antisemm is tmly associated with p45 and pi05, we immunoprecipitated the biotinylated sperm membrane proteins with various zonadhesin antibodies, including anti-Dl, anti-D3, anti- N antisemm, and holoprotein antisemm (Fig. 4.2a). The immunoprecipitation showed 102 that p45, pl05, and antisemm against pig zonadhesin holoprotein are able to precipitate p300 as well as p45 and pl05, but the anti-N antisemm mainly precipitated p300. The amount of p45 and pi05 precipitated by the anti-N antisemm was much less than that of p300. To determine the oligomeric status of p300, the immunoprecipitate obtained with anti-N antisemm was subjected to analysis with 2-D SDS-PAGE (Fig. 4.2b, l"-D: disulfides not reduced; 2"'^-D: disulfides reduced). We observed that the anti-N antisemm did precipitate small amounts of p45 and pi05, but a significant amount of p300 appears close to the diagonal of the 2-D gel, indicating that in addition to participating in formation of the multimer with p45 and pl05, p300 also forms disulfide bond with small sized protein. This is consistent with what we observed in Fig. 4.1. In summary, these immunochemical data point to a conclusion that p300 is a processed zonadhesin gene product that includes zonadhesin MAM domains, and probably also the DO domain (because p300 is detected by the DO-Dl antisera). 4.3.2 p300 Includes a Mucin-Like Domain That Is Glycosylated with 0-Linked Oligosaccharides Given the proximity of the MAM and mucin-like domains in zonadhesin, the processed zonadhesin component containing MAM domain is likely to include also mucin-like domain, especially if p300 tmly includes the DO domain. In addition, the size of p300 is much larger than the deduced molecular mass of the MAM domain region, and it is even larger than the deduced molecular mass from the entire unprocessed zonadhesin polypeptide. Nevertheless, p300 is unlikely to be the unprocessed zonadhesin nascent polypeptide itself; otherwise, both anti-Dl and anti-D3 antibody should be able to detect 103 it. Since mucin-like domains are normally heavily glycosylated by O-linked oligosaccharides, 1 tested whether the oversized p300 results from extensive glycosylation on its mucin-like domain. To test the glycosylation hypothesis, deglycosylation enzymes that specifically remove O-linked oligosaccharide were employed to determine whether p300 contains O- linked oligosaccharides. Sperm membrane proteins were treated with Endo-0- glycosidase, which removes oligosaccharide core stmctures bound to serine and threonine, and three additional enzymes [sialidase A, P(l,4)-Galactosidase, and Glucosaminidase], which remove some types of modifications on the O-linked oligosaccharide core stmcture. The anti-N antisemm detected a smear of bands less than Mr 200,000, instead of p300, on Westem blots loaded with sperm membrane extracts that had been specifically deglycosylated with O-linked deglycosylation enzymes (Fig. 4.3a), indicating that p300 does contain O-linked oligosaccharides. The smeared bands were not a consequence of proteolysis because the control (under the same treatment but without addition of deglycosylation enzymes) maintained an intact p300.1 also treated sperm membrane extract with PNGase F, a highly efficient deglycosylation enzyme that specifically removes almost all types of N-linked oligosaccharides, to test whether p300 contains N-linked oligosaccharides. PNGase F treatment did not change p300 migration at all (Fig. 4.3a), indicating that N-linked oligosaccharides are not present in p300, or at least not present in a significant amount. Therefore, p300 is mainly glycosylated by O- hnked oligosaccharides. To ensure that deglycosylation was complete in 16 hours incubation, I conducted a time course test with purified zonadhesin. The same pattem of p300 migration change was evident on SDS-PAGE double-stained with Coomassie blue 104 and silver (Fig. 4.3b). The deglycosylation appeared to be complete in 1 hour, long before the 16 hours of digestion used for Fig. 4.3a. Interestingly, neither pi05 nor p45 in the punfied zonadhesin changed migration after O-linked deglycosylation, indicating these two zonadhesin components do not contain O-linked oligosaccharides. The smear of bands, instead of a single band that usually represents a bare peptide backbone, probably resuhed from the inefficiency of deglycosylation, because O-linked glycosylation is very complicated and no known enzyme can completely remove all such oligosaccharides. Thus, the actual size of the p300 bare peptide may be much smaller than Mr 160,000, the lowest band observed after enzymatic O-linked deglycosylation. 4.3.3 Both p45 and pl05 Are Glycosylated with N-linked Oligosaccharides Since zonadhesin deduced sequence reveals that some N-linked glycosylation sites are present in D domains, 1 also conducted deglycosylation experiment on the D domain-containing polypeptides p45 and pi05. PNGase F-treated sperm membrane extract was loaded onto SDS-PAGE under disulfide-reduced conditions and Westem blots were probed with the anti-Dl antibody (Fig. 4.4a, left) or the anti-D3 antibody (Fig. 4.4a, right). Migration of both p45 and pi 05 changed significantly, suggesting that both of them are specifically glycosylated by N-linked oligosaccharides. A deglycosylation time course using purified zonadhesin confirmed this conclusion (Fig. 4.4b). Treatment with PNGase F did not change migration of p300 at all, verifying that p300 is probably devoid of N-hnked oligosaccharides. Incubation with deglycosylation enzymes for intervals ranging from 30 minutes to 4 hours showed that the N-deglycosylation was 105 completed within 1 hour, hi contrast to deglycosylation of O-linked oligosaccharides, it is hkely that the size of the deglycosylated bands represent the actual sizes of the bare peptides of p45 and pi 05, because PNGase F is able to remove all types of N-linked oligosaccharides. Therefore, all known pig zonadhesin gene products are glycosylated by certain types of carbohydrate. Since all of the major proteins in the ZP are also glycoproteins, the carbohydrate may play a role in mediating zonadhesin-ZP interaction. However, the carbohydrate on the ZP glycoprotein appears to fail to confer species specificity to recognition between sperm and egg. As mentioned in Chapter 1, even though human ZP3 gene-rescued mZP3'^' mice expressed human ZP3 glycoprotein that is glycosylated like regular human ZP3, its egg adheres to mouse spermatozoa, instead of human spermatozoa. Therefore, the species-specific adhesion between sperm and egg appears not to be conferred exclusively by carbohydrates. 4.3.4 Zonadhesin Is Present in at Least Two Physicoche mically Distinct Compartments of Pig Spermatozoa Soon after zonadhesin was first identified in sperm membrane preparations, we detected zonadhesin in the sperm pellet from low speed centriftigation after cavitation (denuded spermatozoa). Extracting denuded spermatozoa directly with SDS-PAGE sample buffer yielded a similar pattem of zonadhesin subunits as that observed in sperm membranes. To determine the nature of zonadhesin's association with denuded spermatozoa, differential detergent extraction was employed. Denuded spermatozoa were first extracted with mild detergent 1%, (v/v) Triton X-100 solution (in HNE), and the sperm pellet was re-extracted with 2% (w/v) SDS solution (in HNE) (Fig. 4.5a). Those 106 proteins that are extractable with Triton X-100 normally either associate with lipid bilayers or are in soluble form in the cell's cytoplasm; the proteins that are resistant to Triton X-100 extraction and are exttacted only with a strong detergent such as SDS are usually entrapped tightly in the cytoskeleton or some other matrix. The results from differential detergent extraction showed that the zonadhesin profile in Triton X-100 extracts of denuded spermatozoa display a similar pattem as sperm membrane (Fig. 4.5a), indicating that a considerable amount of zonadhesin cannot be recovered in sperm membrane preparations obtained by cavitation. Because zonadhesin has two closely apposed membrane systems—the plasma membrane and the outer acrosomal membrane, we speculated that the Triton-extractable zonadhesin may also be present in the acrosomal membrane or in the cytosol in a soluble form, as well as in sperm plasma membrane. Presence of zonadhesin in SDS extracts further suggested that the protein localizes not only to the membrane or in soluble form, but also to a cell matrix in spermatozoa. Localization data obtained from immunofluorescence (Fig. 3.8) demonstrated that zonadhesin was localized on the apical head of spermatozoa overlying the acrosome, suggesting that the acrosomal matrix is a likely location for the zonadhesin extracted with SDS. 4.3.5 p300 Undergoes Significant Changes during Sperm Maturation in the Epididymis, but p45 and pi05 Do Not Spermatozoa experience significant changes when they travel through the epididymis. The most dramatic change is that the cells develop from an immotile state to a state capable of progressive movement. The acrosome also experiences morphological 107 changes in some species, and the molecular configurations of the acrosomal matrix changes correspondingly. In addition, some proteins, such as nuclear protamines , form extensively cross-linked disulfide bonds (Yanagimachi, 1994). To determine the timing of zonadhesin post-ttanslational processing events such as proteolytic digestion, glycosylation, and oligomer formation, we compared the properties of zonadhesin in spermatozoa from different regions of the epididymis (caput, corpus, and cauda), ejaculated semen, and testis tissue. Results reported in the previous section indicated that zonadhesin in sperm membrane preparations does not represent all forms of zonadhesin. Zonadhesins targeted to different destinations in the sperm cell may have different functions. Therefore, when comparing zonadhesin forms in spermatozoa from different epididymal regions, it was necessary to distinguish the molecules targeted to different destinations in the cells. For each sperm sample, I made differential detergent extracts to separate the zonadhesin associated with membranes and the zonadhesin embedded in the cell matrix. Spermatozoa were extracted with l%o Triton X-100 solution (representing the membrane-associated sperm proteins), and the pellet was re-extracted with 2%o SDS solution (representing matrix-associated sperm proteins). These differential detergent extracts from spermatozoa at different maturation states were then analyzed on Westem blots probed with three domain-specific antibodies against pig zonadhesin (anti-Dl, anti- D3, or anti-N). The resuUs are shown in Fig. 4.6. Equal quantities of proteins were loaded in each lane, ft is apparent that none of the zonadhesin polypeptides, p45 (Fig. 4.6a), pi05 (Fig. 4.6b), or p300 (Fig. 4.6c), exhibited significant Mr change in either Triton X- 100 extract or SDS extract from spermatozoa that have left the testis (epididymal or ejaculated spermatozoa). Thus, most, if not all, proteolytic processing and glycosylation 108 of zonadhesin's three known components have been completed within the testis. Moreover, the fact that zonadhesins in the Triton X-100 and SDS extracts displayed identical MrS suggests that zonadhesins targeted to the different compartments undergo similar proteolytic processing and glycosylation. However, there was a significant difference between zonadhesin in tesfis tissue and zonadhesin in spermatozoa. First, in the Triton X-100 extract, both p45 and pi05 in the testis were slightly larger than in spermatozoa, and the anti-Dl antibody also detected a few additional higher Mr bands (Fig. 4.6a & b), indicating the probable removal of certain oligosaccharides from p45 and pi05 before spermatozoa exit the testis. Second, p45 and pi05 were detected only in the Triton X-100 extract, and not in the SDS extract (Fig. 4.6a & b) of testis, indicating that zonadhesin does not become tightly embedded in the matrix compartment until spermatozoa reach the epididymis. Since the detergent extracts of testis contain proteins from immature spermatids at various developmental states, this observation may reflect the fact that cell matrix material has not completely formed in the majority of spermatids in testis. Changes in p300 were even more dramatic than those in p45 and pi05. The protein was not present in the Triton X-100 extract until spermatozoa were ejaculated, indicating that p300 was originally targeted only to the cell matrix, unlike p45 and pl05 that appeared targeted to both membrane and cell matrix. After spermatozoa transited the epididymis, a large amount of p300 seemed to be released from the cell matrix to membrane or to cytosol. In addition to the size changes in zonadhesin polypeptides, the strength of signal also displayed differences among spermatozoa at different maturation states, ft was apparent that more p45 and pi05 immunoreactivity was present in caudal and ejaculated 109 spermatozoa than in caput and corpus spermatozoa. This may reflect the presence of more zonadhesin embedded into cell matrix after spermatozoa mature in the epididymis. However, this experiment was not fully calibrated for a direct comparison of the amounts of zonadhesin polypeptides. For each lane, the same mass of total protein in the extracts was loaded, so the stronger signal may just indicate that the relative concentration of the polypeptides in that extract is higher than in other spermatozoa, instead of the direct reflection of the absolute amount of those zonadhesin polypeptides in the detergent extract. 4.3.6 Oligomeric Changes of Zonadhesin Occur in the Epididymis Because oligomerization influences zonadhesin's ZP binding activity (Yanagimachi, 1994), and the epididymis is known to promote disulfide bond formation for some sperm proteins (Fig. 3.6), I also checked the oligomeric status of zonadhesin in spermatozoa at different stages of maturafion by two-dimensional SDS-PAGE and Westem blotting. Sperm transit through the epididymis is a continuous process, so to simplify the 2-D gel analysis 1 picked spermatozoa only from two points, their starting point in the caput epididymis and their ending point in freshly ejaculated semen. Moreover, I used combinations of anti-Dl and anti-D3 antibodies to detect zonadhesin on Westem blot so both p45 and pi05 could be viewed on one blot. This made it possible to compare the oligomeric status of p45 and pi05 side by side. The Westem blots probed with the combination of anti-Dl and anti-D3 antibodies yielded resuUs comparable to those obtained with the anti-DO-Dl antisera in differential detergent extraction 110 experiments (Fig. 4.5b). Because the titers of the anti-Dl and anti-D3 antibodies differ, the signal strength likely does not reflect the protein ratio between p45 and pi05. Dilutions of the anti-Dl and anti-D3 antibodies were adjusted to give similar p45 and pl05 signal strengths. To get signals comparable to the SDS extract (Fig. 4.7a & b, right), three times more Triton X-100 extract protein (Fig. 4.7a & b, left) were loaded, indicating that the majority of zonadhesin is in the matrix compartment. In the Triton X- 100 extract of ejaculated spermatozoa, zonadhesin formed more low-Mr pi 05/45 monomer (Mr 150,000, formed by one molecule p45 and one molecule of pi 05) than high-Mr muUimer (Mr > 900,000) (Fig. 4.7b, left), but in the corresponding SDS extract, high-Mr muhimers were the major form of zonadhesin (Fig. 4.7b, right). In contrast, the ratio of low Mr oligomers to high Mr multimers was reversed for caput spermatozoa (Fig. 4.7a). Therefore, the epididymis promotes assembly of zonadhesin in the matrix compartment into high Mr multimers, whereas zonadhesin in the membrane compartment shifts in the opposite direction to a higher proportion of low-Mr oligomers. As described in Chapter III, the form of zonadhesin that most actively binds to the ZP is the Mr 150,000 pi05/45 molecule. An increase of active, pi05/45 zonadhesin in the membrane compartment may represent preparation for subsequent ZP-binding events following ejaculation. Formation of more zonadhesin multimers in the matrix compartment may help to stabilize the acrosomal matrix during acrosomal exocytosis, thereby sustaining adhesion as the sperm cell penetrates the ZP. Ul 4.3.7 The Zona Pellucida-Binding Activity of Zonadhesin Increases in the Epididymis The only known function of zonadhesin so far relates to its ZP-binding activity, so it is cmcial to determine whether this activity undergoes any changes in the process of sperm maturation. Membrane-associated zonadhesin is most likely to mediate initial interaction with the ZP, so it makes sense to check the ZP-binding activity of zonadhesin in Triton X-100 extracts. Furthermore, zonadhesin in SDS extracts is not suitable for ZP binding assays because the protein has been denatured, and the high concentration (2%) of SDS would prevent non-covalent interaction between zonadhesin and the native, particulate ZP. Under disulfide-reduced condifions (Fig. 4.8b), the amount of ZP-bound zonadhesin from Triton X-100 extracts was increased in ejaculated spermatozoa compared to epididymal or testicular spermatozoa, whereas there was no apparent increase in the total amount of zonadhesin in the extracts. This increased ZP-binding activity for zonadhesin in ejaculated spermatozoa was more evident under nonreducing conditions (Fig. 4.8a). This is consistent with the result that the amount of ftmctional pi 05/45 zonadhesin was increased in the Triton X-100 extract of ejaculated spermatozoa (Fig. 4.7b). Collectively, these studies on zonadhesin maturation demonstrated that although individual polypeptide components of zonadhesin do not change significantly when spermatozoa transit the epididymis, zonadhesin does undergo considerable changes in its oligomeric status and ZP-binding activity. It is likely that the ZP-binding activity of zonadhesin is regulated by its oligomeric status instead of changes in any single polypeptide component of the holoprotein. 112 4.3.8 Zonadhesin Localizes to the Outer Acrosomal Mem brane and the Acrosomal Matrix by hnmunoelectron Microscopy Availability of purified zonadhesin enabled us to develop an antisemm against the zonadhesin holoprotein. As discussed previously, the antisemm to zonadhesin holoprotein only detected zonadhesin that had not undergone disulfide bond reduction (Hickox et al, 2001) (see also Chapter III). This result suggested that the holoprotein antisemm likely recognized zonadhesin in its native conformation, and would therefore be suitable for ulttastmctural localization of zonadhesin on spermatozoa. Immunofluorescence using methanol-fixed, freshly ejaculated pig spermatozoa probed with holoprotein antisemm (Fig. 4.9) yielded the same resuft as Fig. 3.8 in which spermatozoa were probed with the DO-Dl antisera. In both experiments zonadhesin localized to the apical sperm head overlying the acrosome. Neither indirect immunofluorescence viewed by light microscopy nor biochemical approach alone is capable of providing a resolution high enough to determine whether a protein localizes to the sperm surface. One reason is that the sperm plasma membrane can easily be damaged before or during fixation. Only immunoelectron microscopy (lEM) is capable of providing unequivocal evidence on ultrastmctural localization, because we are able to observe directly whether the sperm plasma membrane is damaged. There are two approaches to immunolabel spermatozoa for lEM, pre-embedding or post-embedding. Pre-embedding immunolabeled spermatozoa can provide convincing evidence on whether the studied molecule is present on the sperm surface, because this approach immunolabels the spermatozoa before they are embedded and sectioned. An intact sperm membrane will prevent antibodies and gold particles from 113 interacting with targets in the interior of spermatozoa. A control of spermatozoa permeabilized with Triton X-100 allows labeling of intracellular targets because the sperm membrane is destroyed and reagents can reach the molecules inside the cell. Thus, if gold particles cannot be detected in non-permeabihzed spermatozoa but detected in permeabilized controls, it can be concluded that the molecule is not present on the sperm surface. lEM on spermatozoa pre-embedding-labeled with antisera to the zonadhesin holoprotein showed that no gold particles were detected on the surface of spermatozoa that had intact plasma membranes (Fig. 4.10a); only those spermatozoa with damaged membranes, either occurring inadvertently (Fig. 4.10a) or pre-permeabilizing with Triton X-100 (Fig. 4.10ZJ), were labeled by gold particles on the acrosome. Therefore, zonadhesin does not localize to the surface of sperm plasma membrane. lEM using post-embedding-labeled spermatozoa provided additional information on zonadhesin's subcellular localization within the sperm cell. In this approach, spermatozoa are embedded and sectioned before they are labeled with primary antibody, so the gold particles mark the position where the studied molecule is located inside the cell. Using this method, zonadhesin was almost exclusively detected on the outer part of the acrosome (Fig. 4.10c), indicating that the protein largely localizes to the outer acrosomal membrane and the adjacent, outer aspect of the acrosomal matrix. We also employed lEM to localize zonadhesin in testis spermatids (Fig. 4.11). In round spermatids (Fig. 4.1 la), zonadhesin largely localized to the acrosome, but not in the developing granule. Rather it is exhibited a unique, "omega-hke" localization pattem over the developing acrosomes of these cells. Some gold particles were also present in the Golgi apparatus outside the acrosome, consistent with the expected targeting pathway 114 for sorting and transporting zonadhesin to the acrosome. hi elongated spermatids (Fig. 4.116), zonadhesin was almost exclusively detected on the outer part of the acrosome that may represent the outer acrosomal membrane. This result was consistent with my biochemical resufts that zonadhesin could not be detected in SDS extracts of testis. Therefore, targeting of zonadhesin in cell sorting pathway is a complicated process. Certain forms of zonadhesin were targeted to the membrane, while the destination for other forms of zonadhesin is probably the acrosomal matrix. During this process, both glycosylation and oligomerization may play important roles. As discussed in Chapter 1, vWF also has two destinations in cell sorting system: secreted into circulation or stored in cytoplasmic granules, Weibel-Palade bodies. Although spermatozoa do not have Weibel-Palade bodies, the acrosome is also a type of secretory granule. Therefore, it is possible that targeting of zonadhesin in cell sorting pathway follows a similar mechanism as vWF. 4.3.9. Binding of Zonadhesin Is Sufficient to Support Ad hesion of Cell-Sized Stmctures to the Zona Pellucida I next used the pig zonadhesin holoprotein antisemm to develop a bead adhesion assay to test whether the binding of zonadhesin to the ZP is capable of supporting adhesion of cell-sized stmcture. The experiment was designed to test whether the interaction between zonadhesin and the ZP is able to mediate the adhesion of l-iam fluorescent beads (conjugated with goat-anti-rabbit antibody) to the ZP. Sperm membranes were solubilized in detergent and then incubated with the ZP. After extensively washing, the holoprotein antisemm and secondary antibody-conjugated 115 fluorescent beads were added. After another round of extensive washing, the particulate ZP was viewed by fluorescence microscopy. The control was incubated with irrelevant antisera (against an insect protein) in replacement of the holoprotein antisemm. The result showed that when the holoprotein antisemm was applied, fluorescent beads were distributed around the entire ZP fragments (Fig. 4.12c cfe d), whereas no beads bound to the ZP incubated with control antisera (Fig. 4.12a cS: b). Therefore, zonadhesin-ZP binding is sufficient to support the adhesion of cell-sized stmctures to the ZP, and zonadhesin binding sites on the ZP are accessible to such a large stmcture. This bead assay also provides an efficient visual approach for fiiture cross-species tests of zonadhesin's ZP-binding activity. 4.3.10 Zonadhesin Is Processed by Hydrolysis of an Asp- Pro Bond To determine the N-terminal processing sites of p45 and pi05, we performed automated Edman degradation on purified p45 and pi 05. The sequence of the first ten amino acid residues of p45 was PHYLTFDGRR, which is located near the beginning of the Dl domain of zonadhesin cDNA. Thus, the proteolytic processing site is between Asp*°^ and Pro^°^. The sequence obtained for the first nine amino acids of pi 05 was PHYLTFDGA, which is similariy located near the beginning of the D2 domain, indicating that this processing site is also an Asp-Pro bond between Asp''^' and Pro"^^. The polypeptides generated by processing at these two sites are completely compatible with the results obtained from tryptic fragment sequencing (Hardy and Garbers, 1995) and reactivity of domain-specific antibodies (see Chapter III). Interestingly, the 116 sequences of eighteen amino acids around these hydrolytic digestion sites are highly conserved (almost identical) in p45 and pi05 (Fig. 4.13a). Upon searching through the deduced sequence of zonadhesin, we found an additional such conserved sequence in the D4 domain. The cDNA sequence between the processing site in the D2 domain where pi05 starts and the potential processing site with conserved sequence in the D4 domain encodes a polypeptide with a molecular mass of 90 kDa, which is similar to the size of deglycosylated pi05. Therefore, we speculate that the zonadhesin precursor undergoes three specific cuts at Asp-Pro bonds in the conserved sequences, thereby generating four proteins (from N-terminal to C-terminal), p300, p45, pl05, and a fourth zonadhesin component that remains to be identified at this time (Fig. 4.136). Developing a domain- specific antibody against the D4 domain would help to identify the fourth component that is likely to include a transmembrane segment and a cytoplasmic tail. However, Asp-Pro bond is an unusual digestion site for protease, as most hydrolytic digestion sites are located following a basic amino acid residue. We have searched existing proteases database but failed to find any known protease with this specificity. Initially we were concemed that TCA precipitation of purified zonadhesin prior to N-terminal sequencing would artificially digest the zonadhesin protein at Asp- Pro bonds, as such bonds are the most labile of all peptide bonds to acid. However, the bacterially expressed DO-Dl fiision protein did not undergo any such hydrolysis with the same freatment of TCA precipitation, even though it contains the Dl domain conserved sequence with the Asp-Pro bond. This observation suggests that TCA precipitation camiot result in the specific hydrolysis of this Asp-Pro bond. Therefore, the Asp-Pro bond in the conserved sequence likely is the tme processing site to generate zonadhesin 117 subunits. However, we don't know whether this unusual cut resuUs from an unknown proteolytic enzyme or from the acidic environment in the acrosome (although the latter is without precedent). The information reported in this Chapter is sufficient to constmct a processing model for zonadhesin (Fig. 4.136). Zonadhesin is predicted to be expressed as a 267 kDa nascent peptide that is proteolytically processed by an unknown enzyme to generate four peptides: p300, p45, pl05, and a fourth polypepfide containing the partial D4 domain, ttansmembrane segment, and cytoplasm tail. At the same time, zonadhesin polypeptides also undergo glycosylation with O-linked oligosaccharides on p300 and N-linked ohgosaccharides on p45 and pi05. All these events apparently occur inside the testis. Zonadhesin also forms oligomers and multimers through disulfide bonds. This process probably starts in the testis, continues when spermatozoa travel through the epididymis, and is completed in ejaculated spermatozoa. 118 a b NR R 4-12% disulfides not reduced -p300- 205- f • N) 121- W C_ :^ QJ 70- 0 —1 CD Q. p40- c 52- i o 65 0 Q. 34- 900 300 150 Fig. 4.1 Identification of zonadhesin polypeptides containing the N (MAM) domain. Proteins in membrane fractions of pig ejaculated spermatozoa were separated by one-dimensional {panel a, NR= disulfides not reduced, R = disulfides reduced) or two-dimensional {panel b) SDS-PAGE, and zonadhesin polypeptides were detected on Westem blots with antisera to the pig zonadhesin N (MAM) domain antigenic region. Note the single Mr 350,000 non-reduced polypeptide and the Mr 300,000 and 45-60,000 reduced polypeptides identified {panel a). Two- dimensional analysis revealed that the Mr 300,000 and 45-60,000 reduced polypeptides participate in formation of high Mr disulfide-bonded multimers {panel b). 119 a -p300 -p105 -p45 4%-12% disulfides not reduced p300- Q. C_ S, Q. CD p105- cr> s Q. C o (D CL p45- Fig. 4.2 Immunoprecipitation of biotinylated proteins from sperm membrane with zonadhesin antibodies and antisera. Shown are Westem blots detected with Streptavidin-peroxidase and chemiluminescence. Panel a: zonadhesin forms were immunoprecipitated with anti-Dl, anti-D3, anti-N, or anti-holoprotein as designated on the top of the gel (4-12%, SDS-PAGE). The control lane was loaded with the supematant from IP with irrelevant antisera. Panel b: the proteins immunoprecipitated with anti-N were loaded onto 2-D SDS-PAGE. Note that anti-Dl, anti-D3, and anti-holoprotein all co-precipitate p45, pl05, and p300, whereas anti-N does not pull down comparable amount of p45 and pi05 as the other three reagents, indicating that a large amount of p300 is not associated withp45 and pi05. 120 a Me ^Hf pH| Hi w^ Mji •• •?» *** ^f ^M "** MB Mi f^ —4 205- 200- |p300 116- -p105 121- 97- 66- 70- ••t iHi «•» #rt 4Mi ««* -p45 45- 52- 31- -.V:SP^ -l^^- P _ _ + _ .- + — 0+M __ + ++ + 0 _ _ _ + -h + — Time(h)0 4 1 2 4 16 M h + — Fig. 4.3 Enzymatic deglycosylation of p300. Panel a: proteins in sperm membranes were deglycosylated for 16 h (tie) at 37°C with a variety of combinations of deglycosylation enzymes (O = endo-0-glycosidase, P = PNGase F, M = mixture of sialidase A, p (l-4)-galactosidase, and glucosaminidase). The deglycosylated proteins were loaded onto SDS-PAGE (disulfides reduced), and zonadhesin polypeptides were detected with anti-N antisemm. One lane was loaded with sperm membrane proteins without any pre-treatment (to), and another control lane was loaded with membrane extract under the same treatment as other lanes except without addition of enzymes to assure that protein mobility changes are not caused by proteolysis during the long incubation at 37°C. Note that the mobility of p300 increased only when endo-0-glycosidase was present (helped by other enzymes that hydrolyze O-linked oligosaccharide), indicating that a majority of its oligosaccharides are O-linked. Panel b: a time course assay was performed with purified zonadhesin. Shown is a silver-stained SDS-PAGE (4- 12%, linear gradient, protein disulfides reduced) of purified zonadhesin deglycosylated with endo-0-glycosidase combined with a mixture of other O- linked deglycosylation enzymes (labeled as 0+M). Reactions ran for 1, 2, 4, or 16 h as indicated. Control lanes were loaded with zonadhesin under same treatment except without addition of enzymes. Note that deglycosylation of the purified protein was completed within 1 hour, with no fiirther changes in mobility apparent in 15 hours of additional incubation. 121 a *0 t^g M6 205- 121- 70- :ipio5 52- ::p45 p - - + -1- - - o - - - + + - M - - - + - - IPw ^^W P^W Wmn, w^n -pSOO 200- 116- 97- ]p105 66- 45- ]p45 PNGase F - - + + + Time(h) 0 4 12 4 Fig. 4.4 Enzymatic deglycosylation of p45 and pl05. Panel a: proteins in sperm membrane fractions were deglycosylated for 16 hours (tie) at 37°C with a variety of combinations of deglycosylation enzymes (O = endo-0-glycosidase, P = PNGase F, M = mixture of sialidase A, p (l-4)-galactosidase, and glucosaminidase). The deglycosylated sperm membrane proteins were loaded onto SDS-PAGE (4-12%o gel, protein disulfides reduced), and zonadhesin polypeptides were detected with the anti-Dl antibody {panel a, left) or the anti- D3 antibody {panel a, right). One lane was loaded with sperm membrane 122 Fig. 4.4 Confinued proteins without any pre-treatment (to), and another control lane was loaded with membrane extract incubated for 16 h except without addition of enzymes to assure that protein mobility changes are not caused by proteolysis during the long incubation at 37°C. Note that the motihties of p45 and pi05 increased only upon incubation with PNGase F, indicating that they are N-glycosylated. Panel b: a time course assay was performed with purified zonadhesin. Shown is silver- stained SDS-PAGE (4-12%o gel, protein disulfides reduced) of purified zonadhesin deglycosylated with PNGase F. Reactions ran for 1, 2, and 4 h as indicated. Two controls were included based on the same consideration as in panel a. Note that the deglycosylation of N-linked ohgosaccharides was completed within 2 h. 123 a <9 f^ <<^^ ^^ -p300 -p105 -p45 Fig. 4.5 Differential detergent extraction of denuded spermatozoa. Shown are Westem blots (4-12% SDS-PAGE, protein disulfides reduced) probed with the DO-Dl antisera {panel a) or anti-Dl/anti-D3 antibodies {panel b). Sperm membranes were isolated from the low speed supematant suspension following cavitation (lanes labeled membranes), and the sperm pellet (denuded spermatozoa) was extracted differentially with 1% Triton X-100 (lanes labeled Triton X-100), and then the pellet was re-extracted with 2%, SDS (label as SDS). Note that the Triton X-100 extract exhibited a heterogeneous zonadhesin pattem similar to sperm membranes, but the SDS extract contained primarily p45, pl05, and p300 {panel a), indicating that zonadhesin is present in more than one compartments in spermatozoa. Moreover, a combination of anti-Dl and anti-D3 antibodies is suitable for studying changes in p45 and pi05 {panel b). 124 •io i?>!? a lo^'M 7 ^S-^-^^o .^W 208- 208- 119- 119- -p105 94- 94- MtfMltM tM tm i» tnnti^ 51- ••••• -p45 51- 35- 35- Extract: Triton X-100 SDS Extract: Triton X-100 SDS «? ^'-M*^ -p300 Extract: Triton X-100 SDS Fig. 4.6 Properties of zonadhesin in spermatozoa at different maturation states. Spermatozoa collected from pig testis, caput epididymis, corpus epididymis, cauda epididymis, and ejaculated semen (as labeled on top of each lane) were extracted with 1% Triton X-100 (lanes underiined with Triton X-100) first and then the pellet was re-exttacted with 2%, SDS (lanes underlined with SDS). Shown are Westem blots (4-12% linear gradient, protein disulfides reduced) probed with anfi-Dl {panel a), anti-D3 {panel b), or anti-N {panel c). Note that neither p45 nor pi05 was detected in the testis SDS extract, nor did they change significantly in size or detergent exttactability as the spermatozoa passed through the epididymis. However, the amount of both p45 and pi 05 increased in SDS extracts of caudal and ejaculated spermatozoa, compared to caput and corpus spermatozoa. Note also that p300 was present in the Triton X-100 extract of only ejaculated spermatozoa {panel c). 125 a 4-12% disulfides not reduced 4-12% disulfides not reduced » ^—. ro 208- Q. 119- 94- a. 0) w —I 51- "* %• CD Q. 35- c o Triton X-100 extract SDS extract ^ 4-12% disulfide s not reduced^ 4-12% disulfides not reducec^ Triton X-100 extract SDS extract Fig. 4.7 Two-dimensional SDS-PAGE of differenfial detergent extracts from caput epididymal and ejaculated spermatozoa. Caput epididymal {panel a) or freshly ejaculated {panel b) spermatozoa were first extracted with 1% Triton X-100 {left, panels a & b), and then the pellets were re-extracted with 2%, SDS {right, panels a & b). Extracted proteins were separated by two-dimensional SDS-PAGE, and zonadhesin was detected on Westem blots with a mixture of anti-Dl and anti-D3 antibodies. Note that the multimerization states of zonadhesin in the Triton X- 100 and SDS extracts differed, and that the proportion of multimeric zonadhesin in the SDS extract increased as spermatozoa matured in the epididymis. 126 ^•^ .^ ^.^ ^» s- J" ^ •.& > i? ji? ;.& > i? ?!? a .^cf/0^4 MfV^^ ^^'o^ o^cfi^ ^^0^0#^^^#.^ o^ *# i 208- 208- 119- -p150 119_ -H!! •pi 05 94- 94- mM 51- -p45 35- . • ZP-bound total ZP-bound total disulfides not reduced disulfides reduced Fig. 4.8 Zona pellucida-binding activity of zonadhesin in spennatozoa at different maturation states. Shown are Westem blots probed with anti-Dl/anti-D3 antibodies. Proteins in pig testis, caput, corpus, or cauda epididymal spermatozoa, or ejaculated cells (as labeled above each lane) were extracted with l%o Triton X-100, and 200 \xg of sperm protein or 600 fag of testis protein were mixed with 20 p-g of pig particulate ZP. The ZP with bound sperm proteins was then washed twice with Triton X-100 and then three times with mRIPA. Bound proteins (ZP-bound) were separated by SDS-PAGE (4-12%, linear gradient gels), and zonadhesin polypeptides {panel a, disulfides not reduced; panel b, protein disulfides reduced) on Westem blots were detected with a mixture of anti-D 1 and anti-D3 antibodies. Note that the proportion of zonadhesin with ZP binding activity appeared to increase as spermatozoa matured. 127 a Fig. 4.9 Immunolocalization of zonadhesin on pig spermatozoa with anti-holoprotein antisemm. Freshly ejaculated pig sperm cells were fixed with 100%, methanol, and then labeled with the pig zonadhesin holoprotein antiserum. Bound antibodies were detected with a Texas Red-conjugated secondary antibody. The paired DIC (differential interference contrast, panel a) and fluorescenceimage s {panel b) (Planapo lOOx objective) showed that zonadhesin was detected only at the apical sperm head overlying the acrosome. 128 a '''',:k(/A Fig. 4.10 Ultrastmctural locahzation of zonadhesin in pig ejaculated spermatozoa. Panel a, pre-embedding-labeling of unpermeabilized cells. Panel b, pre-embedding- labeling of Triton X-100 permeabilized cells. Panel c, post-embedding-labeled cells. 129 - ^{- ^^^4Cf '-•»-''>' Fig. 4.10 Continued 130 Fig. 4.10 Continued 131 a Fig. 4.11 Ultrastmctural localization of zonadhesin in pig spermatids. Panel a, view of the developing acrosome overlying the nucleus of a round spermatid. Note the presence of zonadhesin at the developing outer acrosomal membrane and in adjacent, granular acrosomal material, but not in the dense core of acrosomal matrix. Panel b, longitudinal section through the apical head of an elongating spermatid. Note the continued apparent association of zonadhesin with the outer acrosomal membrane and outer aspects of the acrosomal contents. 132 Fig. 4.12 Bead adhesion assay for testing zonadhesin's ability to ftmction as an adhesion molecule. Sperm membrane proteins (400 |ig) were solubilized in 1% CHAPS, and then incubated with the particulate pig ZP (10 |j,g) under constant rocking for 2 hours at 22°C. The ZP with bound proteins were pelleted and washed 3 times with mRIPA. The botmd proteins were labeled with pig zonadhesin holoprotein antisemm and detected by addition of fluorescentbead s (1 |am) conjugated with secondary antibody {panel c, d). The negative control was incubated with an irrelevant antiserum in replacement of the anti-holoprotein antisemm {panel a, b). The ZP was either viewed by phase conttast {panel a. c) or epifluorescence microscopy {panel b, d). Note the heavy coating of adherent beads in panel d. 133 a Dl Domain "^YGSAT CSVYG DPHYL TFDGR RFNFM820 D2 Domain HSIQGSAT CTVSG DPHYL TFDGA LHHFT^^os D4 Domain ""^SSNL CSVFG DPHYR TFDGL SYRFQ^^ss Zonadhesin precursor anti-N anti-Dl anti-D3 \ / niMAM M DO D1 D2 D3 D4 Processed polypeptides proteolysis p300 p45 p105 ^806 3807 D1191 31192 31975 31976 MAM M DO D1 D2 D3 D4 • Mill 1 1 III 1 1 p45 tryptic peptides pi 05 tryptic peptides Fig. 4.13 Determination of zonadhesin proteolytic processing sites. Panel a: conserved sequences that include potential processing sites in the Dl, D2, and D4 domains. The N-terminal amino acid sequences of p45 and pi05 obtained by Edman degradation are underlined in the Dl and D2 domains, respectively. Conserved amino acid residues are shaded gray. Panel b: N-terminal sequences of p45 and pi05 demonstrated that p45 starts with PHYLTFDGRR, and pi05 starts with PHYLTFDGA. Comparison of these sequences with the deduced sequence of the zonadhesin precursor revealed that the D^^'^.p^'^'' and D"''-P"^^ peptide bonds are hydrolyzed to generate the N- and C-termini of p45, and the N-terminus of pl05. A similar sequence (D'''"PHYRTFDG) is present 85 kDa downstream in D4; we speculate that hydrolysis of the D'^^^-P' ''^ bond generates the C-terminus of pi 05. This model is compatible with our previous resuUs on the location of tryptic peptides in p45 and pi 05, and the reactivity of various antibodies with zonadhesin fragments. 134 CHAPTER V MARKED VARIATION IN THE BIOCHEMICAL PROPERTIES OF ZONADHESIN IN SPERMATOZOA FROM A VARIETY OF MAMMALS 5.1 Introduction Zonadhesin is a sperm protein that binds in a species-specific manner to the extracellular matrix (zona pellucida) of the oocyte (Hardy and Garbers, 1994, 1995). Zonadhesin cDNAs from four mammals (pig, mouse, human, and rabbit) encode proteins with similar but non-identical domain stmctures (Hardy and Garbers, 1995; Gao and Garbers, 1998; Lea et al, 2001; Wilson et al, 2001; Cheung et al, in preparation). All of them contain MAM domains, a mucin-like domain, VWD domains, a putative transmembrane segment, and a short cytoplasmic tail. Furthermore, the arrangement of these domains largely follows the same order. However, species-specific variation in zonadhesin domain stmctures does exist among these four species. For instance, mouse zonadhesin contains 20 tandem partial VWD domains between the D3 and D4 domains, and this expansion of partial domains is not present in pig, human, or rabbit (Hardy and Garbers, 1995; Gao and Garbers, 1998; Lea et al, 2001; Cheimg et al, in preparation); in humans, six splicing variants that differ in the fourth VWD domain were observed (Cheung et al, in preparation). Furthermore, extensive heterogeneity in the processing of the pig zonadhesin precursor gives rise to multiple mature forms of the protein with differing avidities for the zona pellucida (Hickox et al, 2001) (see also Chapter III). Because these four mammals are evolutionarily distant from one another, zonadhesin is 135 unlikely to be a species- or order-unique molecule; instead, it is probably a universal molecule present in spermatozoa of most if not all mammals. The multiple extracellular domains of zonadhesin render it a molecule with the potential to evolve species- specifically, as both domain rearrangement and single mutations inside the domains could contribute to the species variation that imposes a barrier to cross-species fertilization. Glycosylation of zonadhesin proteins is an additional feature that may play a role in species-specific gamete recognition. Although the deduced amino acid sequences and domain stmctures of zonadhesins in these four species are recognizably similar, the processed proteins of zonadhesin display great differences in their molecular masses. There are three known processed polypeptides of pig zonadhesin, p45, pi 05, and p300, which forms extensive intermolecular disulfide bonds to generate pl05/45 monomer (Mr 150,000), pl05/45 dimer (Mr 300,000) and high Mr multimers (Mr > 900,000) (Hardy and Garbers, 1994; Hickox et al, 2001) (see also Chapter III). However, mouse zonadhesin displayed very different oligomeric forms. Three bands, with the Mr of 100,000, 200,000 and 250,000, were detected under disulfides not reduced conditions with antisera to the partial VWD domain expansion (Gao and Garbers, 1998). Unfortunately, no result has been reported on mouse zonadhesin proteins under disulfide-reduced conditions, making it impossible to compare the processed zonadhesin proteins between mouse and pig. The post- translational processing of zonadhesin's nascent polypeptide is therefore an additional important factor that can account for the species-variation of zonadhesin. To test our overall hypothesis for zonadhesin studies—that zonadhesin mediates the sperm-ZP adhesion in a species-specific manner, it is necessary to determine whether 136 zonadhesin orthologs can be detected in more mammalian species and thereafter to perform an extensive comparison of zonadhesin among those species. Results of such studies would help us to put forward a working model for zonadhesin's function in mediating species-specific sperm-ZP adhesion. A number of approaches may be employed to perform extensive comparison of zonadhesin among species. First, we could determine the presence and size of the zonadhesin mRNA in other species by Northem blotting, but the obtained information would be very limited. It would be suitable for determining whether zonadhesin orthologs are present in other species, but would provide no information on DNA sequence or protein domain stmcture. A second approach would be to use a fragment of the pig zonadhesin cDNA to obtain zonadhesin cDNA clones from other species by screening their testis cDNA libraries. Ultimately we could determine their full cDNA sequences, as has been done for mouse, rabbit, and human. However, this approach is extremely laborious and time consuming. For such a molecule encoded by a large mRNA, it easily takes a year to clone and fully sequence a zonadhesin cDNA in one species. Thus, it is not practicable for a single student to make an extensive zonadhesin cDNA sequence comparison among a variety of mammalian species. In addition, such studies would yield only sequence data, and perhaps no new information about the proteins in these species. A third approach would be to analyze directly the processed zonadhesin proteins and their oligomeric forms using antisera against pig zonadhesin. Zonadhesin sequences vary significantly from species to species. Nevertheless, zonadhesin is a very large molecule; some regions are very likely to be similar between species. Therefore, it is reasonable to assume that antibodies against pig zonadhesin have a good chance to recognize 137 zonadhesin proteins in other species, hi addition, the antibodies against pig zonadhesin can also be used to localize zonadhesin in spermatozoa from other species. Therefore, the third approach is the most practicable one to perform extensive species-comparison of zonadhesin. A number of antibodies and antisera have been raised against pig zonadhesin, including domain-specific antibodies, such as anti-Dl and anti-D3 antibodies, and antisemm against native zonadhesin, such as the holoprotein antisemm (Hickox et al, 2001) (see also Chapter III). The domain-specific antibodies are suitable for identifying polypeptide of disulfide bond reduced zonadhesin, while the holoprotein antisemm is suitable for detecting zonadhesin oligomers in their native states so as to locahze zonadhesin on spermatozoa. Here I report characterization of zonadhesin protein in spermatozoa from eight eutherian mammals. Zonadhesin localized to the apical head of spermatozoa from bull, horse, rabbit, rat, hamster, mouse, mole, and man. Immunoelectron microcopy further localized zonadhesin to the acrosomes of bovine and hamster spermatozoa. The protein shared several general physicochemical properties (proteolytically processed forms present as multimers in particulate fractions) among the different species. However, the sizes of zonadhesin's processed polypeptides varied dramatically among species, as did the partitioning of the protein between compartments that were or were not extractable with mild detergent. Thus inter-species variation in zonadhesin is produced not only by point mutations and domain stmcture differences specified by its mRNAs, but also by species differences in processing and association of the protein with other sperm stmctures. 138 5.2 Materials and Methods 5.2.1 Collection of Spermatozoa Spermatozoa were collected from eight mammahan species for detection and analysis of zonadhesin in these species by Western blotting. For these species, including pig, horse, and bovine, freshly ejaculated semen was collected and immediately mixed with an equal volume of appropriate extender. Normally, spermatozoa in extender can live for at least 48 hours. The collected spermatozoa were subjected to differential detergent extraction within four hours. Human spermatozoa in liquefied semen were washed with PBS and subjected to detergent extracfion within two hours of donation. As for the other four species, including mouse, rat, hamster, and rabbit, spermatozoa were collected from caudal epididymides. The caudal epididymis in rabbit, rat, and hamster are large enough to use surgical scissors to make several slices, and spermatozoa were then released into PBS (10 mM NaP04, 150 mM NaCl, pH 7.4) by squeezing the epididymis in a plastic dish. Because mouse caudal epididymides are too small to handle like other species, a needle was used to collect spermatozoa from the snipped epididymides and washed in PBS. Spermatozoa collections from these four species were performed in the lab, so the spermatozoa were subjected to differential detergent extraction immediately. 5.2.2 Differential Detergent Extraction of Spermatozoa Differential detergent extraction of spermatozoa from those mammahan species largely followed the procedure for pig spermatozoa described in previous chapter. For spermatozoa collected from caudal epididymides and released directly in PBS buffer, 139 they were pelleted by gentle centrifugation at 300 x g. For ejaculated spermatozoa incubated in extenders, 10 volumes of PBS buffer were added before a brief centrifugation at 300 x g. Spermatozoa were washed once in HNE (20 mM HEPES, 130 mM NaCl, ImM EDTA, pH 7.5) at 22°C. 1% Triton X-100 solution (1% Triton X-100, 150 mM NaCl, 1 mM EDTA, 10 |aM E-64, 0.5 mM DFP, 10 mM Tris-HCl, pH 7.5) was then added to pelleted spermatozoa at a ratio of 10:1 (by volume) to ensure that the amount of detergent was sufficient to extract all of those sperm proteins soluble in Triton X-100. The pellet was re-extracted with 10 volumes of 2% SDS solution (2% SDS, 1 mM EDTA, 10 pM E-64, 0.5 mM DFP, 10 mM Tris-HCl, pH 6.8). The supematant extracts from centrifligations were collected as the Triton X-100 and SDS extracts, respectively. 5.2.3 Indirect Immunofluorescence The spermatozoa used for immunofluorescence were collected from either caudal epididymides (rat, mouse, rabbit, hamster, and mole) or freshly ejaculated semen (bovine, horse, and human). After being fixed and permeabilized with methanol, spermatozoa from these species were probed with pig zonadhesin holoprotein antisemm and Texas red conjugated secondary antibody, and viewed by epifluorescence microscopy. 5.2.4 Immunoelectron Microscopy lEM was used to determine the ultrastmctural localization of zonadhesin on bovine and mouse spermatozoa. The procedure largely followed the protocol described in the previous chapter (see section 4.2.10). 140 5.3 Results and Discussion 5.3.1 Zonadhesin Proteins Vary Significantly among Mammalian Species I used two pig zonadhesin antibodies to detect zonadhesin proteins in a variety of mammalian species. One of them was the domain-specific antibody anti-D3, which specifically detects pi05 in pig spermatozoa; the other reagent was the antisemm against the pig zonadhesin holoprotein, which was developed against affinity-purified zonadhesin holoprotein. This antisemm detects only native zonadhesin under nonreducing conditions (Hickox et al, 2001) (see also Chapter III). As I discussed in the previous chapter, pig zonadhesin forms are present in two cell compartments of spermatozoa, membrane-associated and acrosome matrix- associated. Since the zonadhesin targeted to different destinations may have different flinctions, I separated those two types of zonadhesin by differential detergent extraction to identify zonadhesin proteins in the various mammalian species. The freshly ejaculated spermatozoa (pig, horse, bovine and human) or caudal spermatozoa (rabbit, mouse, rat, hamster) were extracted with l%o Triton X-100 first, and the pellet was re-extracted with 2%o SDS solution. The anti pig D3 antibody detected a Mr 105,000 band in SDS extracts of spermatozoa from bovine, horse, rabbit, and hamster, but a Mr 60,000 in mouse, human, hamster, and rat (Fig. 5.1). In hamster, human and rat, one or more additional bands were detected in SDS extracts of spermatozoa. Several bands that varied significantiy from one species to another were also detected in Triton X-100 extracts of spermatozoa, but for some other species no distinct band was present in Triton X-100 extracts. Unlike pig sperm extract, in which the anti-D3 antibody detected the same-sized 141 band (pi05) in both Triton X-100 and SDS extracts, the band(s) detected in the Triton X- 100 extracts of other species exhibited different sizes from that detected in their subsequent SDS extracts. This resuU suggests that zonadhesin undergoes species-specific post-translational processing that leads to dramatic species variation of zonadhesin proteins. The Mr of human zonadhesin detected by the anti pig D3 antibody is consistent with that detected with an antibody against the human zonadhesin D3 domain (Cheung et al, in preparation), suggesting that the anti pig D3 antibody is able to recognize corresponding zonadhesin proteins in other mammals. Zonadhesins of those species (pig, horse, and bovine) belonging to the Order Ungulata shared a zonadhesin D3 polypeptide of the same Mr (Mr 105,000); while zonadhesin in rodent also shares a same Mr zonadhesin protein (Mr 60,000). Therefore, the proteolytic processing of zonadhesin appears to exhibit a certain degree of Ordinal specificity. However, within the same Order, processed zonadhesin proteins also varied from one another. For example, hamster zonadhesin contained two additional higher Mr D3 polypeptide bands that were not present in mouse or rat. Interestingly, comparison of zonadhesin cDNAs among pig, mouse, and human revealed that human zonadhesin exhibits a lower sequence identity with mouse zonadhesin than with pig zonadhesin, but the one of the detected human zonadhesin D3 polypeptides has the an Mr similar to a mouse zonadhesin D3 polypeptide. Thus, the species divergence of zonadhesin is sophisticated, not only consisting of variation in the nascent gene products, but also of variation in the products' post- translational proteolytic processing. Both processes could be involved in building a species-specific barrier among mammalian species. 142 The antiserum against the pig zonadhesin holoprotein was used to detect zonadhesin in other species under disulfide-not-reduced conditions (Fig. 5.2). In contrast to the anti-D3 antibody, the pig holoprotein antisemm only detected pig zonadhesin oligomers under nonreducing conditions, failing to react with disulfide reduced zonadhesin polypeptides. Thus, we expected this antisemm to detect zonadhesin orthologs in other species under nonreducing conditions, hi the SDS extract of spermatozoa from two ungulate species, horse and bovine, the pig holoprotein antisemm detected a Mr 300,000 band that was also present in SDS extracts of pig spermatozoa. These proteins are hkely to be the disulfide-bonded oligomers of zonadhesin proteins. However, the pig zonadhesin holoprotein anfisemm detected two bands, Mr 200,000 and Mr 100,000, in rabbit Triton X-100 extractable fractions but no bands from the Triton X- 100 resistant fractions. Unfortunately, this antisemm did not to detect any band on blots of rodent species or human. Since we observed that high Mr multimers hardly enter the top of the gel (4%,), the failure of detecting zonadhesin in these species is either because pig holoprotein antisemm only reacted with high Mr multimers in those species, or because zonadhesin does not form lower Mr oligomers in these species. Because formation of disulfide-bonded ohgomers and multimers may be important in regulation of zonadhesin's ftinction, we performed 2-D Westem blotting analysis to determine whether and how zonadhesins in those tested species form intermolecular disulfide bonds. In most species, the zonadhesin proteins detected by the anti-D3 antibody participated in formation of high Mr multimers that barely entered the gel (Fig. 5.3). However, only a low Mr oligomer was detected in horse spermatozoa, and no oligomer was detected in human spermatozoa. Therefore, even the closely related 143 species may form dramatically different disulfide-bonded oligomers or muhimers of zonadhesin. Collecfively, these species comparison data obtained by Westem blotting demonstrated that zonadhesin diverges among species at several levels, including post- ttanslational proteolytic processing, formation of disulfide-bonded oligomers, and existence in membrane and matrix compartments of spermatozoa, in addition to known variation in sequences specified by their cDNAs. 5.3.2 Zonadhesin Localizes to the Apical Head of Spermatozoa of All Tested Species The pig zonadhesin holoprotein antisemm has been used to localize zonadhesin on pig spermatozoa by indirect immunofluorescence and immunoelectron microscopy because it recognizes native zonadhesin (Chapter III). To determine the localization of zonadhesin in other mammahan species, we also used the pig holoprotein antisemm to detect zonadhesin in spermatozoa from a variety of mammalian species by indirect IF and IBM. In every species that we have checked, including three rodents (mouse, hamster and rat), two ungulates (horse and bovine), as well as rabbit, human, and mole, zonadhesin was detected on the apical head of spermatozoa overiying the acrosome by IF (Fig. 5.4). This result was consistent with the localization of pig, mouse, and rabbit zonadhesin reported previously (Gao and Garbers, 1998; Hickox et al, 2001; Lea et al, 2001). lEM was employed to determine the ultrastmctural localization of zonadhesin in two species, hamster and bovine. Spermatozoa from these two species were fixed and post-embedding immnunolabeled with the pig holoprotein antisemm, so the gold particles represent the 144 location of zonadhesin both on and inside the spermatozoa (Fig. 5.5). Like pig zonadhesin (Fig 4.10), zonadhesins of hamster (Fig. 5.5a) and bovine (Fig. 5.5b) both localized to the outer acrosomal membrane and in the acrosomal matrix. The conclusion from these results is that zonadhesin exists in most mammalian species. Although zonadhesins in some species may exhibit the similar domain stmctures, the processed proteins could vary dramatically. Furthermore, even if zonadhesin proteins of similar size were present in two species, the protein might still display species specificity through varied forms of disulfide-bonded oligomers and multimers. These multiple levels of regulation of a molecule may provide a hint about the puzzle of the relatively small number of genes in human genome. The human genome is estimated to comprise 26,000 to 38,000 genes (Venter et al, 2001), which is a much smaller number than many investigators expected. How could these less than 40,000 genes account for the sophisticated development and complexity of a human being? Interspecies comparisons of zonadhesin told us that post-translational proteolytic processing might play an important role in generating different molecules that are encoded by similar genes. Therefore, in addition to regulation on transcription level and translation level, post-translational level regulation might also play a critical role in generating many more molecules from the only 26,000 to 38,000 genes, thereby increasing the repertoire of gene products available for development and evolutionary success of complex organisms. 145 pig horse bovine mouse hamster rat human rabbit TS TS TS TSTSTS Fig.5.1 Detection of zonadhesin polypeptides in spermatozoa of eight mammals. Proteins in spermatozoa from pig, horse, bovine, mouse, hamster, rat, human and rabbit were extracted sequentially with 1% Triton X-100 (lanes labeled 7) and then 2% SDS (lanes labeled S). The extracted proteins were separated by SDS-PAGE (4- 12%) linear gradient gels, protein disulfides reduced), and zonadhesin polypeptides on Westem blots were detected with affinity-purified antibodies to the pig D3 domain. 146 horse bovine rabbit Fig. 5.2 Detection of zonadhesin holoprotein in spermatozoa from three ungulates and one lagomorph. Proteins in pig, horse, bovine, and rabbit spermatozoa were extracted sequentially with 1% Triton X-100 (lanes labeled T) and then 2% SDS (lanes labeled S). The extracted proteins were separated by SDS-PAGE (4-12%, linear gradient gel, protein disulfides not reduced), and zonadhesin on Westem blots was detected with antisera to the pig zonadhesin holoprotein. Note that zonadhesin from each ungulate species migrated as an Mr >300,000 band and was present in both Triton-extractable and -resistant fractions, whereas zonadhesin from rabbit only present in Triton-extractable fractions and migrated as Mr 200,000 and Mr 100,000 bands. 147 Bovine 4..,2o/„ disulfides not reduced Rabbit 4..|2o/„ disulfides not reduced ^^ j^ 12 % disulfide s 1 reduce d 1 ' Horse 4.12% disulfides not reduced Mouse 4.12% disulfides not reduced 'it .(>. 12 % disulfi d "^i ^ OJ duce d • \ ' Hamster 4.12% disulfides not reduced ^ Human 4.12% disulfides not reduced *. « 12 % disulfide s I reduce d 1 \ ' • - Fig. 5.3 Two-dimensional electrophoresis of zonadhesin in spermatozoa from six eutherian mammals. Proteins in SDS extracts of bovine, rabbit, horse, mouse, and hamster spermatozoa were separated by two-dimensional SDS-PAGE (protein disulfides not reduced in the first dimension, reduced in the second dimension; 4- 12% linear gradient gels each dimension). Zonadhesin polypeptides on Westem blots were detected with affinity-purified antibody to the pig D3 domain. Note that zonadhesin polypepfides were in high Mr, disulfide-bonded complexes in each species except human. 148 Bovine Rat Horse Rabbit Fig. 5.4 Localization of zonadhesin by indirect immunofluorescence. Spermatozoa from human, mouse, mole, hamster, bovine, rat, horse, and rabbit were fixed with methanol, and zonadhesin was detected with antisemm to the pig zonadhesin holoprotein. Note the fluorescence signal conforming to the apical heads of spermatozoa from each of the eight species. 149 a i- Fig. 5.5 Ultrastructural localization of hamster and bovine zonadhesin. Shown are hamster {panel a) and bovine {panel b) spermatozoa post-embedding-labeled with antisera to the pig zonadhesin holoprotein. 150 Fig. 5.5 Continued 151 CHAPTER VI SUMMARY AND DISCUSSION 6.1 Overview The discovery of zonadhesin was an encouraging development in a research field where several groups of investigators are actively looking for the sperm molecule that mediates the adhesion of spermatozoa to the ZP Two features of zonadhesin distinguished it from other ZP-binding candidates. First, it binds to the native, particulate ZP, while most other investigators used denatured ZP or solubilized ZP glycoproteins as the binding ligand to identify ZP-binding molecules. Second, the binding of zonadhesin to the ZP occurs in a species-specific manner. No other ZP-binding candidate has been observed to bind in such a manner. The results reported in this thesis help to increase our understanding of zonadhesin biochemical properties and ftinction. I found that zonadhesin is an even more complicated molecule than previously appreciated in a number of respects. The zonadhesin cDNA encodes a 267-kDa precursor that is much larger than any other ZP-binding candidate discovered so far. On top of the inherent complexity of such a large molecule, 1 found that zonadhesin undergoes heterogeneous post-translational processing, including proteolytic processing and different types of glycosylation. In addition, zonadhesin polypeptides form extensive interchain disulfide bonds to generate a variety of oligomers and multimers, which are so large that it is difficult to characterize them with routine methods. Finally, zonadhesin is targeted to two physicochemically distinct compartments by the protein trafficking pathway; both membrane- and matrix-associated compartments were identified. On the one hand, the 152 complexity of zonadhesin increases the difficulty in characterizing its properties and funcfions; on the other hand, this complexity may reflect the intrinsic complexity of species-specific sperm-egg adhesion. These studies on zonadhesin also raise a number of interesting issues about the putative ftinction of zonadhesin. 6.2 How Does Zonadhesin "See" the Zona Pellucida? Ultrastmctural localization of zonadhesin by immunoelectron microscopy revealed that it is not present on the surface of the sperm plasma membrane; instead, it is localized in the outer acrosomal membrane and in the outer aspect of the acrosomal matrix. This would raise a question about how zonadhesin gets access to the ZP before the start of acrosomal exocytosis. Although some earlier reports stated that mammalian spermatozoa adhere to the ZP only after the acrosome reaction is completed, the commonly accepted notion is that only acrosome-intact spermatozoa can specifically adhere to the ZP. Because the sperm plasma membrane imposes a barrier for zonadhesin to "see" the ZP before the acrosome reaction, how could zonadhesin ftilfill the function of mediating the sperm-egg adhesion in a species-specific manner? Altematively, does the localization of zonadhesin in the acrosome eliminate its possible role in sperm-egg adhesion during fertilization? In my opinion, zonadhesin does have the opportunity to interact with the ZP before the acrosome reaction is completed. Many investigators appreciate that the acrosome reaction does not occur in an "all or none" manner. Instead, the exocytosis occurs first regionally and then spreads to the whole sperm head that overiies the 153 acrosome. Therefore, a possible way for zonadhesin to fiinction is that the original attachment of spermatozoa to the ZP is not in a species-specific manner. This "touch" is probably mediated by other protein(s) on the sperm surface and likely causes regional fusion between the sperm plasma membrane and the outer acrosomal membrane. The regional exocytosis at the point where the sperm cell contacts the ZP exposes the zonadhesin on the outer acrosomal membrane or the zonadhesin emanating from the outer aspect of the acrosomal matrix. The species-specific recognition subsequently takes place between the ZP and zonadhesin. If the zonadhesin and the ZP are from the same species and therefore are compatible, they would "bind" tightly with each other. This species-specific binding event would increase in avidity as the regional exocytosis spreads to a whole-sperm-wide exocytosis to complete acrosome reaction, thereby exposing all available zonadhesin. However, if spermatozoa and the egg come from different species, zonadhesin would fail to bind specifically to the ZP and spermatozoa will be released from the ZP, with their acrosome's completing exocytosis unproductively. Although there is no evidence so far to support this proposed working model for zonadhesin, it is a reasonable explanation for zonadhesin's potential ftinction. Indeed, it makes sense to hide the ZP-binding molecule from the female immune system until the time when they start performing their function. In addition, this scenario may explain why our anti-DO-Dl antisera failed to interfere with sperm-egg adhesion in eariy experiments (data not shown), because there is no chance for antibodies against zonadhesin to pre-occupy the ZP-binding site of zonadhesin that is inaccessible before inifiation of the acrosome reaction. Therefore, typical inhibition experiments with 154 zonadhesin antibodies are unlikely to succeed as they sometimes do for other ZP-binding candidates. 6.3 What Factor(s) Determine Zonadhesin's Species Specificity? The initial experiment that lead to discovery of zonadhesin disclosed a unique characteristic of zonadhesin in that it binds to the native ZP in a species-specific manner. The extensive studies on ZP glycoproteins have demonstrated that the species specificity of the ZP likely is not conferred by any single ZP glycoprotein. Although considerable evidence supports a funcfion of mouse ZP3 as the molecule that mediates the adhesion of spermatozoa to the ZP, replacement of mouse ZP3 with human ZP3 does not change the original species specificity of recognition between sperm and egg (Rankin et al, 1998). Instead, the whole stmcture of the ZP is more likely to confer the species specificity of its sperm adhesion acfivity. If zonadhesin is the complementary molecule for the ZP, is it able to confer the species specificity by itself? If so, by what manner? The complexity of zonadhesin we leamed during characterization of this protein makes us believe that variations generated in several levels can contribute to the species specificity of zonadhesin. (1) On the gene level. Comparison of sequenced zonadhesin cDNAs in four species revealed that the domain stmcture could significantly vary among mammalian species (e.g., between pig zonadhesin and mouse zonadhesin). In addition, since zonadhesin contains multiple extracellular domains, including a number of fiill or partial VWD domains, even without significant changes within each domain, the rearrangement or shuffling of those domains can produce varieties of possible domain 155 combinations. Moreover, some variants that are likely generated by gene rearrangement have been observed, suggesting that within the same species different sets of genes or nascent gene products may exist in different individuals. (2) On the transcription level. Dunng sequencing of mouse and human zonadhesin cDNA, it has been observed that some mRNA variants are present. Those variants were caused by either differential transcription or rearrangement on the gene level. Therefore, even though the zonadhesin genes are highly homologous, the transcribed mRNAs might be divergent. (3) On the post-translational processing level. We have observed that the processed zonadhesin polypeptides varied significantly between pig and human, even though the cDNA domain stmctures from the two species are similar. The antibody against the pig D3 antigenic region recognized pi05 in pig spermatozoa, but recognized a band around Mr 60,000 in human spermatozoa that was confirmed by a domain-specific antibody to the human D3 antigenic region, demonstrating that polypeptides containing the same domain may vary significantly in different species. Thus, the discrete post-translational processing events can increase dramatically the number of possible final protein products. Those divergent polypeptides generated from the same mRNA transcript may explain how a relatively small number of genes in the human genome can generate enough proteins required to produce an organism of the complexity of a human being. (4) On the supramolecular (oligomerization) level. Zonadhesin forms various oligomers and multimers through extensive intermolecular disulfide bonds. In almost every mammalian species we have tested, zonadhesin formed high Mr multimers. However, low Mr oligomers varied considerably among different species. In pig, the Mr 150,000 pi05/45 monomer comprising one molecule of pi 05 and one molecule of p45 showed the highest ZP- 156 binding preference in comparison to other forms of muhimers. Thus, oligomerization plays an important role in regulation of zonadhesin's ZP-binding activity. Collectively, the variations possibly produced in those levels have the potential to make zonadhesin a molecule with species-specific function. 6.4 Future Studies The results reported here have established a good understanding of the biochemical properties and function of zonadhesin. The antibodies developed against zonadhesin also provided the required reagents for ftirther study on zonadhesin. However, more evidence is needed to test rigorously our overall hypothesis—zonadhesin mediates the sperm-egg recognition in a species-specific manner. Future studies could be performed in these two directions as outlined below. Ultrastmctural localization of zonadhesin disclosed that the protein is localized in the acrosome, so it hides from any reagents from outside of sperm plasma membrane in the acrosome-intact spermatozoa before the onset of acrosomal exocytosis, making the regular antibody inhibition experiment impossible. The results generated by antibody inhibition experiments provide a type of loss-of-ftinction evidence, while the most convincing loss-of-function evidence normally comes from gene knockout experiments. If spermatozoa of the zonadhesin knockout mice lost the ability to adhere to the ZP and thereby failed to fertilize the egg, it will be strong loss-of-function evidence that zonadhesin mediates adhesion of spermatozoa to the egg. During the last decade, generation of mice with specifically dismpted genes has become a popular approach to 157 study the functions of proteins. Generating gene knockout mice not only provides an approach to acquire direct loss-of-fiinction evidence, but also paves a way to get gain-of- function evidence by re-introducing the same molecule back into the genome to rescue the lost fimcfion. It is especially useful in studying the species specificity of orthologs in different species. For instance, in the case of studies on ZP3, human ZP3 introduced into the ZP3 knockout mice did rescue the ZP3's function, but it failed to change the species specificity of sperm adhesion (Rankin et al, 1998). Therefore, one can conclude that ZP3 by itself is not sufficient to confer the species specificity of sperm-ZP adhesion. The same approach can be applied to zonadhesin for studying its species-specific ZP-binding activity, thereby demonstrating whether zonadhesin itself is enough to impose a species- specific barrier to block heterologous sperm-egg adhesion during mammalian fertilization. In addition to the molecules discussed in Chapter I (ZPl, ZP2, ZP3, acrosin, GalTase), several other molecules related to fertilization have been dismpted in mouse. Fertilin P (on spermatozoa) and CD9 (on the egg) are both reported to be involved in sperm-egg membrane ftision. Fertilin (3 knockout males (Mathur et al, 1986) and CD9 knockout females (Stanwell-Smith et al, 1983) are largely infertile, hiterestingly, the spermatozoa of fertilin p knockout mice failed to adhere to the ZP, although fertilin has not previously been shown to ftinction in sperm-ZP adhesion. Therefore, those gene knockout mice may also provide much information that cannot be obtained from traditional biochemical or physiological methods. 158 The initial ZP binding experiments demonstrated that pig zonadhesin only bound to the pig ZP but not to the ZP isolated from other species (Hardy and Garbers, 1994). However, due to lack of a specific reagent to detect zonadhesin in other species at that time, we could not get information on whether zonadhesin of other species also displays similar species-specific ZP-binding activity. The resuUs in Chapter V suggested that the antisemm to pig zonadhesin holoprotein is an appropriate reagent to detect zonadhesin in other species. Although we may use the original ZP-binding assay to study the species- specific ZP-binding activity of zonadhesin in other species, the bead adhesion assay described in Chapter IV provides an easier approach to achieve this goal. Because this assay mimics the adhesion of spermatozoa to the ZP, the result is more convincing and straightforward. One can isolate the native ZP from several species, such as mouse, bovine and horse, and then use bead adhesion assay to test whether zonadhesin from each of these species bind to the homologous ZP from the same species and the heterologous ZP from other species. Other studies on zonadhesin may include some typical approaches in fertilization studies, such as interfering with in vitro fertilization using purified zonadhesin or expressed zonadhesin domains, identifying the ZP glycoprotein that is the complementary molecule for zonadhesin, and so on. Although zonadhesin was found with an approach that is different from previous approaches in this field, it will be helpfiil to study this molecule using the traditional methods to provide comparable results with other ZP-binding candidates. Since we believe that adhesion of zonadhesin to the ZP is mediated by more than one factor, it will also be interesting to understand the relationship and interaction between zonadhesin and other potential candidates. 159 REFERENCES Baba T., Azuma, S., Kashiwabara, S., and Toyoda, Y. (1994). Sperm from mice carrying a targeted mutation of the acrosin gene can penetrate the oocyte zona pellucida and effect fertilization. J. Biol. Chem. 269, 31845-31849. Beckmann, G., and Bork, P. (1993). An adhesive domain detected in functionally diverse receptors. Trends Biochem Sci 18, 40-41. Bellet, N.F., Vacquier, J.P., and Vacquier, V.D. (1977). Characterization and comparison of "bindin" isolated from sperm of two species of sea urchins. Biochem. Biophys Res. Comm. 79, 159-165. Bi, M., Wassler, M.J., and Hardy, D.M. (2002). Sperm adhesion to the extracellular matrix of the egg. hi Fertilization, D.M. Hardy, ed. (San Diego: Academic Press) pp 153-180. ^^' Bleil, J.D., Beall, C.F., and Wassarman, P.M. (1981). Mammalian sperm-egg interaction: fertilization of mouse eggs triggers modification of the major zona pellucida glycoprotein, ZP2. Dev. Biol. 86, 189-197. Bleil, J.D., Greve, J.M., and Wassarman, P.M. (1988). Identification of a secondary sperm receptor in the mouse egg zona pellucida: role in maintenance of binding of acrosome-reacted sperm to eggs. Dev. Biol. 128, 376-385. Bleil, J.D. and Wassarman, P.M. (1980a). Mammalian sperm-egg interaction: identification of a glycoprotein in mouse egg zonae pellucidae possessing receptor activity for sperm. Cell 20, 873-882. Bleil, J.D. and Wassarman, P.M. (1980b). Stmcture and function of the zona pellucida: identification and characterization of the proteins of the mouse oocyte's zona pellucida. Dev. Biol. 76, 185-202. Bleil, J.D. and Wassarman, P.M. (1983). Sperm-egg interactions in the mouse: sequence of events and induction of the acrosome reaction by a zona pellucida glycoprotein. Dev. Biol. 95, 317-324. Bleil, J.D. and Wassarman, P.M. (1990). Identification of a ZP3-binding protein on acrosome-intact mouse sperm by photoaffinity crosslinking. Proc. Natl. Acad. Sci. USA 87, 5563-5567. Bookbinder, L.H., Cheng, A., and Bleil, J.D. (1995). Tissue- and species-specific expression of sp56, a mouse sperm fertilization protein. Science 269, 86-89. 160 Check, J.H., Nowroozi, K., Lee, M., Adelson, H., and Katsoff, D. (1990). Evaluation and treatment of a male factor component to unexplained infertility. Arch. Androl. 25, 199-211. Cheng, A., Le, T., Palacios, M., Bookbinder, L.H., Wassarman, P.M., Suzuki, F., and Bleil, J. D. (1994). Sperm-egg recognition in the mouse: characterization of sp56, a sperm protein having specific affinity for ZP3. J. Cell Biol. 125, 867-878. Cheung, T.L., Wassler, M.J., Comwall, G.A., and Hardy, D.M. Multiple intra-species variants of human zonadhesin. in preparation. Chia, S.E., Ong, C.N., Lee, S.T., and Tsakok, F.H. (1992). Blood concentrations of lead, cadmium, mercury, zinc, and copper and human semen parameters. Arch. Androl. 29, 177-183. Desseyn, J.L., Aubert, J.P., Van, S., I, Porchet, N., and Laine, A. (1997). Genomic organization of the 3' region of the human mucin gene MUC5B. J. Biol. Chem. 272, 16873-16883. Dunbar, B.S., Wardrip, N.J., and Hedrick ,J.L. (1980). Isolation, physicochemical properties, and macromolecular composition of zona pellucida from porcine oocytes. Biochemistry 2, 356-365. Eckhardt, A.E., Timpte, C.S., DeLuca, A.W., and Hill, R.L. (1997). The complete cDNA sequence and stmctural polymorphism of the polypeptide chain of porcine submaxillary mucin. J. Biol. Chem. 272, 33204-33210. Florman, H.M. and Wassarman, P.M. (1985). O-linked oligosaccharides of mouse egg ZP3 account for its sperm receptor activity. Cell 41, 313-324. Foster, J.A., Friday, B.B., Maulit, M.T., Blobel, C, Winfrey, V.P., Olson, G.E., Kim, K.S., and Gerton, G.L. (1997). AM67, a secretory component of the guinea pig sperm acrosomal matrix, is related to mouse sperm protein sp56 and the complement component 4-binding proteins. J. Biol. Chem. 272, 12714-12722. Gao, Z. and Garbers, D.L. (1998). Species diversity in the stmcture of zonadhesin, a sperm-specific membrane protein containing multiple cell adhesion molecule-hke domains. J. Biol. Chem. 273, 3415-3421. Gendler, S.J. and Spicer, A.P. (1995). Epithehal mucin genes. Annu. Rev. Physiol. 57, 607-634. Gerton, G.L. (2002). Function of the Sperm Acrosome. hi Fertihzation, D.M.Hardy, ed. (San Diego: Academic Press), pp. 265-302. 161 Glabe, C.G. and Vacquier, V.D. (1977). Species specific agglutination of eggs by bindin isolated from sea urchin sperm. Nature 267, 836-838. Goldsmith, J.R. (1997). Dibromochloropropane: epidemiological findings and current questions. Annals NY Acad. Sci. 837, 300-306. Gong, X., Dubois, D.H., Miller, D.J., and Shur, B.D. (1995). Activation of a G protein complex by aggregation of beta-l,4-galactosyltransferase on the surface of sperm. Science 269, 1718-1721. Greve, J.M. and Wassarman, P.M. (1985). Mouse egg extracellular coat is a matrix of intercoimected filaments possessing a stmctural repeat. J. Mol. Biol. 181, 253-264. Haden, N.P., Hickox, J.R., Scott, W.C, and Hardy, D.M. (2000). Systematic characterization of sperm-specific membrane proteins in swine. Biol. Reprod. 63, 1839-1847. Hardy, D.M. and Garbers, D.L. (1994). Species-specific binding of sperm proteins to the extracellular matrix (zona pellucida) of the egg. J. Biol. Chem. 269, 19000-19004. Hardy, D.M. and Garbers, D.L. (1995). A sperm membrane protein that binds in a species-specific manner to the egg extracellular matrix is homologous to von Willebrand factor. J. Biol. Chem. 270, 26025-26028. Hardy, D.M., Wild, G.C., and Tung, K.S. (1987). Purification and initial characterization of proacrosins from guinea pig testes and epididymal spermatozoa. Biol. Reprod. 37, 189-199. Harrison, R.A. (1988a). Hyaluronidase in ram semen. Quantitative determination, and isolation of muUiple forms. Biochem. J. 252, 865-874. Harrison, R.A. (1988b). Preliminary characterization of the multiple forms of ram sperm hyaluronidase. Biochem. J. 252, 875-882. Harrison, R.A. and Gaunt, S.J. (1988). Multiple forms of ram and bull sperm hyaluronidase revealed by using monoclonal antibodies. J. Reprod. Fertil. 82, 777- 785. Hickox, J.R., Bi, M., and Hardy, D.M. (2001). Heterogeneous processing and zona pellucida binding activity of pig zonadhesin. J. Biol. Chem. 276, 41502-41509. Hunnicutt, G.R., Mahan, K., Lathrop, W.F., Ramarao, C.S., Myles, D.G., and Primakoff,P. (1996). Stmctural relationship of sperm soluble hyaluronidase to the sperm membrane protein PH-20. Biol. Reprod. 54, 1343-1349. Jaiswal, B.S. and Eisenbach, M. (2002). Capacitation. hi Fertilization, D.M. Hardy, ed. (San Diego: Academic Press), pp. 57-118. 162 Jiang, W., Gorbea, CM., Flannery, A.V., Beynon, R.J., Grant, G.A., and Bond, J.S. (1992). The alpha subunit of meprin A. Molecular cloning and sequencing,' differential expression in inbred mouse strains, and evidence for divergent evolution of the alpha and beta subunits. J. Biol. Chem. 267, 9185-9193. Joba, W. and Hoffinann, W. (1997). Similarities of integumentary mucin B.l from Xenopus laevis and prepro-von Willebrand factor at their amino-terminal regions J Biol. Chem. 272, 1805-1810. Kim, K.S., Cha, M.C., and Gerton, G.L. (2001). Mouse sperm protein sp56 is a component of the acrosomal matrix. Biol. Reprod. 64, 36-43. Laemmli, U.K. (1970). Cleavage of stmctural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lea, LA., Sivashanmugam, P., and O'Rand, M.G. (2001). Zonadhesin: characterization, localization, and zona pellucida binding. Biol. Reprod. 65, 1691-1700. Legan, P.K., Rau, A., Keen, J.N., and Richardson, G.P. (1997). The mouse tectorins. Modular matrix proteins of the inner ear homologous to components of the sperm-egg adhesion system. J. Biol. Chem. 272, 8791-8801. Lewis, C.A., Talbot, C.F., and Vacquier, V.D. (1982). A protein from abalone sperm dissolves the egg vitelline layer by a nonenzymatic mechanism. Dev. Biol. 92, 227- 239. Liu, C, Litscher, E.S., Mortillo, S., Sakai, Y., Kinloch, R.A., Stewart, C.L., and Wassarman, P.M. (1996). Targeted dismption of the mZP3 gene resuhs in production of eggs lacking a zona pellucida and infertility in female mice. Proc. Natl. Acad. Sci. USA 9i, 5431-5436. Lopez, L.C., Youakim, A., Evans, S.C, and Shur, B.D. (1991). Evidence for a molecular distinction between Golgi and cell surface forms of beta 1,4-galactosyltransferase. J. Biol. Chem. 266, 15984-15991. Lu, Q. and Shur, B.D. (1997). Sperm from beta 1,4-galactosyltransferase-null mice are refractory to ZP3-induced acrosome reactions and penetrate the zona pellucida poorly. Development 124, 4121-4131. Lyons, S.E., Bmck, M.E., Bowie, E J., and Ginsburg, D. (1992). Impaired intracellular transport produced by a subset of type IIA von Willebrand disease mutations. J. Biol. Chem. 267, 4424-4430. Marti, T., Rosselet, S.J., Titani, K., and Walsh, K.A. (1987). Identification of disulfide- bridged substmctures within human von Willebrand factor. Biochemistry 26, 8099- 8109. 163 Mathur, S., Pathak, S., Kandel, A., Ziegler, J., Rust, P.F., and Wihamson, H.O. (1986). Leukocyte migration inhibitory factor (LIF) to sperm from autoimmune men in infertile couples. J. Immunol. 136, 4444-4450. Mayadas, T.N. and Wagner, D.D. (1989). hi vitro multimerization of von Willebrand factor is triggered by low pH. bnportance of the propolypeptide and free sulfliydryls J. Biol. Chem. 264, 13497-13503. Mayadas, T.N. and Wagner, D.D. (1992). Vicinal cysteines in the prosequence play a role in von Willebrand factor multimer assembly. Proc. Nati. Acad Sci USA 59 3531- 3535. Miller, D.J., Gong, X., Decker, G., and Shur, B.D. (1993). Egg cortical granule N- acetylglucosaminidase is required for the mouse zona block to polyspermy J Cell Biol. 123, 1431-1440. Miller, D.J., Macek, M.B., and Shur, B.D. (1992). Complementarity between sperm surface beta-l,4-galactosyltransferase and egg-coat ZP3 mediates sperm-egg binding. Nature 357, 589-593. Miller, D.J. and Shur, B.D. (1994). Molecular basis of fertihzation in the mouse. Seminars in Developmental Biology 5, 255-264. Morrissey, J.H. (1981). Silver stain for proteins in polyacrylamide gels: A modified procedure with enhanced uniform sensitivity. Anal. Biochem. 117, 307-310. Omu, A.E., Dashti, H., Mohamed, A.T., and Mattappalhl, A.B. (1995). Significance of trace elements in seminal plasma of infertile men. Nutrition 11, 502-505. Paufler, S.K. and Foote, R.H. (1969). Effect of triethylenemelamine (TEM) and cadium chloride on spermatogenesis in rabbits. J. Reprod. Fertil. 19, 309-319. Perez-Vilar, J., Eckhardt, A.E., DeLuca, A., and Hill, R.L. (1998). Porcine submaxillary mucin forms disulfide-linked multimers through its amino-terminal D-domains. J. Biol. Chem. 273, 14442-14449. Perez-Vilar, J., Eckhardt, A.E., and Hill, R.L. (1996). Porcine submaxillary mucin forms disulfide-bonded dimers between its carboxyl-terminal domains. J. Biol. Chem. 271, 9845-9850. Perez-Vilar, J. and Hill, R.L. (1998a). Identification of the half-cystine residues in porcine submaxillary mucin critical for multimerization through the D-domains. Roles of the CGLCG motif in the Dl- and D3-domains. J. Biol. Chem. 273, 34527- 34534. 164 Perez-Vilar, J. and Hill, R.L. (1998b). The carboxyl-terminal 90 residues of porcine submaxillary mucin are sufficient for forming disulfide-bonded dimers. J. Biol. Chem. 273, 6982-6988. Perez-Vilar, J. and Hill, R.L. (1999). The stmcture and assembly of secreted mucins J Biol. Chem. 274, 31751-31754. Peterson, R.N., Russell, L., Bundman, D., and Freund, M. (1980). Sperm-egg interaction: evidence for boar sperm plasma membrane receptors for porcine zona pellucida. Science 207, 73-74. Prasad, S.V., Skinner, S.M., Carino, C, Wang, N., Cartwright, J., and Dunbar, B.S. (2000). Stmcture and function of the proteins of the mammalian Zona pellucida. Cells Tissues Organs 166, 148-164. Rankin, T., Familari, M., Lee, E., Ginsberg, A., Dwyer, N., Blanchette-Mackie, J., Drago, J., Westphal, H., and Dean, J. (1996). Mice homozygous for an insertional mutation in the Zp3 gene lack a zona pellucida and are infertile. Development 722, 2903-2910. Rankin, T.L., O'Brien, M., Lee, E., Wigglesworth, K., Eppig, J., and Dean, J. (2001). Defective zonae pellucidae in Zp2-null mice dismpt folliculogenesis, fertility and development. Development 128, 1119-1126. Rankin, T.L., Tong, Z.B., Castle, P.E., Lee, E., Gore-Langton, R., Nelson, L.M., and Dean, J. (1998). Human ZP3 restores fertility in Zp3 null mice without affecting order-specific sperm binding. Development 725, 2415-2424. Ruggeri, Z.M. and Zimmerman, T.S. (1980). Variant von Willebrand's disease: characterization of two subtypes by analysis of multimeric composition of factor Vlll/von Willebrand factor in plasma and platelets. J. Clin. Invest. 65, 1318-1325. Sacco, A.G., Yurewicz, E.G., Subramanian, M.G., and Matzat, P.D. (1989). Porcine zona pellucida: association of sperm receptor activity with the alpha-glycoprotein component of the Mr = 55,000 family. Biol. Reprod. 41, 523-532. Sadler, I.E. (1998). Biochemistry and genetics of von Willebrand factor. Annu. Rev. Biochem. 67, 395-424. Shur, B.D. (1989). Galactosyltransferase as a recognition molecule during fertihzation and development, fri The Molecular Biology of Fertihzation, H.Schatten and G.Schatten, eds. (San Diego: Academic Press), 37-71. Shur, B.D. and Neely, CA. (1988). Plasma membrane association, purification, and partial characterization of mouse sperm beta 1,4-galactosyltransferase [published erratum appears in J Biol Chem 1989 Mar 5;264(7):4264]. J. Biol. Chem. 263, 17706- 17714. 165 Snell, W.J. and White, J.M. (1996). The molecules of mammalian fertilization. Cell 85, 629-637. Spom, L.A., Marder, V.J., and Wagner, D.D. (1986). Inducible secretion of large, biologically potent von Willebrand Factor multimers. Cell 46, 185-190. Stanwell-Smith, R., Thompson, S.G., Haines, A.P., Ward, R.J., Cashmore, G., Stedronska, J., and Hendry, W.F. (1983). A comparative study of zinc, copper, cadmium, and lead levels in fertile and infertile men. Fertil. Steril. 40, 670-677. Swanson, W.J. and Vacquier, V.D. (1997). The abalone egg vitelline envelope receptor for sperm lysin is a giant multivalent molecule. Proc. Nati. Acad. Sci. USA 94, 6724- 6729. Swanson, W.J. and Vacquier, V.D. (1998). Concerted evolution in an egg receptor for a rapidly evolving abalone sperm protein. Science 257, 710-712. Takagi, S., Hirata, T., Agata, K., Mochii, M., Eguchi, G., and Fujisawa, H. (1991). The A5 antigen, a candidate for the neuronal recognition molecule, has homologies to complement components and coagulation factors. Neuron 7, 295-307. Towbin, H., Staehelin, T., and Gordon, J. (1979). Electrophoretic ttansfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Nati. Acad. Sci. USA 76, 4350-4354. Urch, U.A., Wardrip, N.J., and Hedrick, J.L. (1985a). Limited and specific proteolysis of the zona pellucida by acrosin. J. Exp. Zool. 233, 479-483. Urch, U.A., Wardrip, N.J., and Hedrick, J.L. (1985b). Proteolysis of the zona pellucida by acrosin: the nature of the hydrolysis products. J. Exp. Zool. 236, 239-243. Vacquier, V.D. (1998). Evolution of gamete recognition proteins. Science 257, 1995- 1998. Varki, A. (1994). Selectin ligands. Proc. Nati. Acad. Sci. USA 97, 7390-7397. Venter, J.C, et al (2001). The sequence of the human genome. Science 297, 1304-1351. Verweij, C.L., Hart, M., and Pamiekoek, H. (1987). Expression of variant von Willebrand factor (vWF) cDNA in heterologous cells: requirement of the pro-polypeptide in vWF multimer formation. EMBO J. 6, 2885-2890. Voorberg, J., Fontijn, R., Calafat, J., Janssen, H., van Mourik, J.A., and Pannekoek, H. (1991). Assembly and routing of von Willebrand factor variants: the reqmrements for disulfide-linked dimerization reside within the carboxy-terminal 151 amino acids. J. CellBiol. 775, 195-205. 166 Wagner, D.D. (1990). Cell biology of von Willebrand factor. Ann. Rev. Cell Biol. 6, 217- 246. Ward, C.R., Storey, B.T., and Kopf, G.S. (1992). Activafion of a Gi protein in mouse sperm membranes by solubilized proteins of the zona pellucida, the egg's extracellular matrix. J. Biol. Chem. 267, 14061-14067. Wassarman, P.M. (1987). The biology and chemistry of fertihzation. Science 235, 553- 560. Wassarman, P.M. (1988). Zona pellucida glycoproteins. Ann. Rev. Biochem. 57 415- 442. Wilson, M.D., Riemer, C, Martindale, D.W., Schnupf, P., Boright, A.P., Cheung, T.L., Hardy, D.M., Schwartz, S., Scherer, S.W., Tsui, L.C, Miller, W., and Koop, B.F. (2001). Comparative analysis of the gene-dense ACHE/TFR2 region on human chromosome 7q22 with the orthologous region on mouse chromosome 5. Nucleic Acids Res 29, 1352-1365. Wise, R.J., Pittman, D.D., Handin, R.L, Kaufman, R.J., and Orkin, S.H. (1988). The propeptide of von Willebrand factor independently mediates the assembly of von Willebrand multimers. Cell 52, 229-236. Yanagimachi, R. (1981). Mechanisms of fertilization in mammals. In Fertilization and Embryonic Development In Vitro, L.Mastroianni and J.D.Biggers, eds. (New York: Plenum Press), pp. 81-182. Yanagimachi, R. (1994). Mammalian fertihzation. In The Physiology of Reproduction, E.Knobil and J.D.Neill, eds. (New York: Raven Press, Ltd.), pp. 189-317. Youakim, A., Hathaway, H.J., Miller, D.J., Gong, X., and Shur, B.D. (1994). Overexpressing sperm surface p 1,4-galactosyltransferase in transgenic mice affects multiple aspects of sperm-egg interactions. J. Cell Biol. 126, 1573-1583. Yurewicz, E.G., Sacco, A.G., Gupta, S.K., Xu, N., and Gage, D.A. (1998). Hetero- oligomerization-dependent binding of pig oocyte zona pellucida glycoproteins ZPB and ZPC to boar sperm membrane vesicles. J. Biol. Chem. 273, 7488-7494. Zondag, G.C, Koningstein, G.M., Jiang, Y.P., Sap, J., Moolenaar, W.H., and Gebbink, M.F. (1995). Homophilic interactions mediated by receptor tyrosine phosphatases mu and kappa. A critical role for the novel extracellular MAM domain. J. Biol. Chem. 270, 14247-14250. 167