| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Division of Reproductive Biology, Department of Biochemistry and Molecular Biology, Johns Hopkins University, School of Hygiene and Public Health, Baltimore, Maryland 21205
ABSTRACT
Sperm-egg interactions occur at multiple levels on the egg surface, first with the egg's extracellular matrix and then with the egg's plasma membrane. The BioPore minisymposium on "The Egg Surface" at the 1999 annual meeting of the Society for the Study of Reproduction highlighted a series of events underlying successful interactions of the sperm with the egg: 1) composition, synthesis, and assembly of the mouse egg's extracellular matrix, the zona pellucida, during oogenesis; 2) oocyte maturation and development of the sperm-binding domain of mouse eggs; and 3) characterization of functional domains in different sperm ligands (fertilin-
and fertilin-ß in the mouse and lysin in the abalone) that recognize cognate binding sites on the egg surface. Data that were presented are reviewed here and discussed with respect to conserved and divergent features of gamete functions.
fertilization
INTRODUCTION
The BioPore minisymposium on "The Egg Surface" at the 1999 annual meeting of the Society for the Study of Reproduction reflected the complexities of getting sperm and egg together, but it also went above and beyond its title. Presentations in this session covered multiple molecules, multiple species, and multiple events of fertilization, and they highlighted several issues concerning the mechanisms of fertilization and evolution as well as species specificity in gamete functions and interactions.
In virtually all species, eggs are covered by multiple layers of extracellular vestments; thus, sperm-egg interactions begin with these extracellular layers and then continue at the level of the egg plasma membrane. The mammalian egg is surrounded by two extracellular layers: the zona pellucida (ZP, or zona), which is synthesized during oogenesis (primarily by the oocyte); and the cumulus oophorus, including an extracellular matrix (ECM) of hyaluronic acid secreted by the cumulus cells during oocyte maturation [1]. In an analogous fashion, the eggs of lower vertebrates (e.g., amphibians) and of invertebrates are enclosed in a vitelline envelope (VE) as well as an additional, outer layer known as the jelly coat or jelly layer (Fig. 1) [2]. Thus, the creation of a fertilization-competent egg surface involves the synthesis and assembly of the egg's ECM as well as the synthesis and expression of egg membrane receptors for sperm ligands.
|
The 1999 BioPore minisymposium discussed three key features of gamete function: 1) the composition, synthesis, and assembly of the mouse egg's ECM during oogenesis; 2) oocyte maturation and development of the sperm-binding domain of mouse eggs; and 3) characterization of functional domains in different sperm ligands for binding sites on the egg surface. Individually, each talk discussed gamete function leading up to and during fertilization. The breadth of the subject matter demonstrated the abilities of multiple methods of modern biological research (e.g., genetic, molecular, cellular, biochemical, structural) to shed light on the molecular basis of sperm-egg interactions, as well as how these interactions can occur in a species-specific manner. Selected topics related to this issue of species specificity and how gamete functions have diverged or been conserved through evolution are discussed at the end of this review.
OOGENESIS AND OOCYTE MATURATION: THE EGG'S PREPARATION FOR FERTILIZATION
The egg's preparation for fertilization begins long before the gametes meet. Jurrien Dean (NIH) described the expression of gene products during oogenesis and folliculogenesis that are critical for mammalian oocyte function, one of which is the set of genes required to form the ZP. In mice, the three zona proteins (i.e., ZP1, ZP2, ZP3) are encoded by single-copy genes, and each gene has a conserved E-box (CANNTG) in its proximal promoter. In vitro studies have identified a basic helix-loop-helix (bHLH) transcription factor, FIG (Factor in the Germline, alpha), that binds as a heterodimer with E12 (a ubiquitous bHLH factor) to the E-box in the ZP promoters. The protein-DNA interaction of FIG/E12 heterodimers with the ZP promoters is necessary, but not sufficient, for ZP gene transcription [3]. Expression of FIG can be detected as early as embryonic day 13 in urogenital ridges and persists in growing oocytes of adults. Female mice with a targeted disruption of the Fig gene do not form primordial follicles perinatally and rapidly lose oocytes during the first few days after birth. Adult females are sterile, with shrunken ovaries completely devoid of germ cells. Males, however, do not appear to be affected.
After the onset of ZP gene expression, the resultant proteins must be assembled into a functional ECM. Genetic studies from the Dean lab indicate that normal zona matrix assembly requires all three ZP glycoproteins. Although homozygous Zp1-null mice can form a zona matrix with just ZP2 and ZP3, the zona matrix is structurally abnormal. Approximately 10% of growing follicles have ectopic localization of granulosa cells in the perivitelline space. Eggs lacking ZP1 are ovulated after hormonal stimulation and can be fertilized, but Zp1-null females have decreased fecundity, with litters only half the size of normal. This embryonic loss results from precocious hatching of early embryos from structurally abnormal ZP [4]. In contrast to the somewhat subtle phenotype of females lacking ZP1, female mice lacking ZP3 are completely sterile [5, 6]. In fact, Zp3-null mice cannot form a zona matrix, few eggs are recovered after hormonal stimulation, and no two-cell embryos are detected after mating. An earlier in vitro transient assay using antisense oligonucleotides to degrade either ZP2 or ZP3 transcripts indicated that loss of either ZP2 or ZP3 protein prevented incorporation of the other into the ZP matrix [7]. Thus, the phenotype of Zp2-null mice likely will be similar to that observed in Zp3-null mice (i.e., an absent zona matrix and infertility). Interestingly, however, zona matrix assembly does not appear to require that the three ZP glycoproteins come from the same species; this observation is based on studies with transgenic mice that produce chimeric human-mouse zona matrices. ZP2 and ZP3 are very similar in the human and the mouse, with 61% and 67% amino acid identity, respectively (although differences do occur in the posttranslational modifications). The experiments were designed to test whether expression of human ZP3 could compensate for the lack of mouse ZP3 in the Zp3 -\- mice. In the animals expressing human ZP3 and mouse ZP1 and ZP2 [8], the zona matrix is assembled during oogenesis, and the eggs are ovulated. Mouse sperm can still bind to the chimeric ZP of these eggs, leading to successful fertilization, pregnancy, and the birth of live pups. In contrast, human sperm do not bind to the mouse eggs with a ZP containing human ZP3.
A final preparation for fertilization that follows oogenesis and folliculogenesis is release of the oocyte to the oviduct and meiotic maturation. In the mouse, meiotic maturation is accompanied by development of cell polarity, resulting in the establishment of a subdomain of the egg plasma membrane (known as the microvillar region) to which sperm bind. Results from the lab of Janice Evans (Johns Hopkins University) show that this morphological polarity is mirrored by molecular polarity, with several specific molecules being segregated to either the microvillar or the amicrovillar region and the remainder to the plasma membrane and cortex overlying the meiotic spindle. Several molecular markers of the egg polarity become localized to their specific subdomain (microvillar or amicrovillar) during oocyte maturation [9]. Three of the markers for the microvillar region are molecules involved in sperm-egg membrane interactions (i.e., binding sites for recombinant fertilin-ß, recombinant fertilin-, and a ß1-integrin epitope implicated as a binding site for fertilin-ß) [10, 11]. Establishment of cell polarity in the mouse egg correlates with the cortical localization of chromatin, and this appears to depend, at least partially, on intracellular calcium transients that occur during meiotic maturation. Drug treatments that perturb the localization of chromatin (cytochalasin or colchicine) also perturb the localizations of these molecular markers of egg polarity, which agrees with results that show these treatment perturb morphological polarity [9]. Treatment of oocytes during maturation in vitro with the cell-permeable calcium chelator BAPTA-AM also perturbed development of normal cell polarity in the egg.
SPERM MOLECULES INVOLVED IN INTERACTIONS WITH THE EGG SURFACE
The molecular basis of sperm-egg interactions was addressed from two different vantage points in studies involving two different experimental systems. Structure-function analysis of two mammalian sperm ligands, fertilin- and fertilin-ß, has identified key functional domains of these sperm proteins. Fertilin- and fertilin-ß both participate in gamete membrane interactions, and both are members of the molecular family known as ADAMs (for a disintegrin and a metalloprotease domain) with a modular domain structure including a metalloprotease domain, a disintegrin domain, and a cysteine-rich domain [12, 13]. Results from the Evans lab suggest that fertilin- and fertilin-ß appear to use different functional domains to mediate the interaction of sperm with the egg plasma membrane. A tripeptide sequence, ECD (Glu-Cys-Asp), in the disintegrin domain appears to be the "business end" of mouse fertilin-ß. When comparing the amino acid sequences of fertilin-ß and snake disintegrins, the ECD sequence in fertilin-ß aligns closely, albeit not exactly, with the RGD (Arg-Gly-Asp) tripeptide of snake disintegrins that mediates the interactions of these molecules with specific integrin receptors [14]. The terminal aspartic acid residue of this ECD sequence was especially critical for the function of this subdomain; this observation was based on studies of point-mutated versions of this region [15]. In contrast to fertilin-ß, mouse fertilin- appears to be more complex, using both its disintegrin domain and its cysteine-rich domain to interact with the egg surface [16]. Further analysis of the fertilin- disintegrin domain identified a short amino acid sequence, DLEECDCG, which is different (i.e., 50 amino acids farther toward the amino terminus) from the region that aligns with the ECD-containing region of the fertilin-ß disintegrin domain. Taken together, these data suggest that ADAM proteins can have multiple functional domains (i.e., cysteine-rich domain and different portions of the disintegrin domain) that mediate cell-cell interactions.
A different perspective on gamete recognition proteins came from studies in the lab of Victor D. Vacquier (Scripps Institution for Oceanography, University of California at San Diego) involving an abalone sperm ligand, lysin. Lysin is released from the abalone sperm acrosome and mediates species-specific dissolution of the egg VE by interacting with an envelope component known as the VE Receptor for Lysin (VERL). Dissolution of the VE occurs via a nonenzymatic process, apparently through dimers of lysin interacting with VERL and competing for bonds that hold the VERL fibers together, thus causing the fibers to unravel and allowing the sperm access to the egg plasma membrane. Subsequently, an 18-kDa protein on the acrosomal process appears to mediate fusion of the sperm with the egg membrane; this idea is based on the observation that the 18-kDa protein can induce liposome fusion [17].
Interaction between the sperm lysin and the VE of the abalone egg occurs in a species-specific manner. Abalone and other marine invertebrates are broadcast spawners, and fertilization occurs externally. Thus, because multiple species overlap in breeding seasons and zonation, a mechanism for species specificity in gamete interactions must exist. Seven different species of abalone living off the coast of California have been studied as a model to elucidate how molecules that mediate species-specific fertilization could have evolved. The VERL is a very large glycoprotein (~1000 kDa) composed of at least 25 tandem repeats of 153 amino acids. Analysis of these repeats in multiple abalone species revealed that VERL evolves as other proteins with repetitive sequence elements, by "concerted evolution," involving unequal crossing over and gene conversion between repeats. The result of concerted evolution is that the repeats within a species homogenize and the repeats between species diverge rapidly. The working model of how VERL and lysin change between species is as follows: A point mutation in a VERL repeat occurs, and concerted evolution randomly replaces the original VERL repeats with this point-mutated version of the repeat within a species. (Indeed, sequence comparisons indicate that each VERL repeat is more similar to other repeats within the same species than it is to repeats in other species [18].) As concerted evolution propagates the new VERL repeat, selective pressure forces lysin to adapt to these changes in VERL. The final result is that lysin and VERL within a species remain compatible with each other but diverge from their homologues in other species. In fact, the sperm proteins lysin and the 18-kDa protein (in the abalone) as well as bindin (in the sea urchin) appear to be some of the most rapidly evolving genes [19–21].
SPECIES SPECIFICITY OF GAMETE FUNCTIONS AND INTERACTIONS
In general, the interactions of gametes occur in a species-specific manner. Differences in reproductive behaviors certainly serve as a barrier to cross-fertilization, although differences in interaction-mediating molecules also ensure that a sperm fertilizes an egg from a female of the same species, particularly for broadcast-spawning aquatic invertebrates. Some of these molecular differences were highlighted in the three presentations in this BioPore minisymposium.
Conservation or Divergence of Composition, Synthesis, and Assembly of the Egg ECM
The egg's extracellular coverings not only provide a protective covering for the egg and embryo but can also mediate species-specific interactions with sperm. The molecular composition of invertebrate egg ECMs appears to differ significantly from that of vertebrate egg ECMs. Thus far, ZP1-, ZP2-, or ZP3-like proteins have not been detected in invertebrate VEs. Instead, invertebrate VEs are composed of completely different proteins, such as VERL in the abalone [22] and the 350-kDa receptor for sperm, VE-B, and other proteins in the sea urchin [23–25]. In contrast, homologues of ZP1, ZP2, and ZP3 (also known as ZPA, ZPB, and ZPC [26]) have been identified in multiple mammalian orders, including rodents, primates, cetartiodactylates (bovine, porcine), and carnivores (canine, feline). In addition, ZP protein homologues have been identified in several nonmammalian species, including the frog [27–29], chicken [30, 31], and even fish [32–36], which diverged from mammals 650 million years ago. However, in spite of the mammalian ZP, the amphibian and avian VE, and the fish chorion being composed of proteins similar to ZP1, ZP2, and ZP3, important differences still exist between these egg ECMs. It should be noted that ZP proteins are apparently synthesized only by oocytes in the mouse and frog [27, 28, 37], whereas transcripts and protein of ZP3 (in bovine, porcine, and nonhuman primates [38–40]) and ZP1 (in rabbit [41]) have been detected in granulosa cells as well as in oocytes. In chickens, granulosa cells seem to be the sole site of synthesis of ZP3 [30, 31], and in some, but not all, fish species, ZP3-like products are made in the liver [32, 35]. Furthermore, the fish chorion differs from other egg ECMs, because sperm do not actually bind to it but instead gain access to the egg plasma membrane through the micropyle, a thin tunnel in the chorion. In Xenopus, the egg ECM is modified during transit through the oviduct, which results in exposure of the egg ECM component gp69/64 (a homologue of ZP2 [29]), and sperm only bind to the ECM of oviposited eggs [2]. Thus, in this species, posttranslational modifications of egg ECM subsequent to the synthesis and assembly of the component proteins are required for the egg ECM to support sperm binding [42].
Mechanisms of egg coat assembly are at least partially conserved among species. Chimeric ZP assembled in vivo from human and mouse zona proteins (mouse ZP1, mouse ZP2, and human ZP3) form sufficient matrices to provide adequate protection to the egg and embryo and can support mouse sperm binding [8]. Surprisingly, human sperm do not bind to these chimeric zonae, despite in vitro data that has implicated ZP3 as being the primary sperm receptor in mice. Fully grown Xenopus sp. oocytes injected with mRNA encoding either mouse ZP1, ZP2, or ZP3 incorporate individual mouse proteins into their VEs [43]. Although mouse sperm do not bind to these chimeric frog-mouse egg coats, this may reflect the limited amounts of mouse protein in the VE or the need to have more than one zona protein to support mouse sperm binding. Taken together, these data indicate that homologous proteins from different species can assemble to make a chimeric egg coat, even with components from two species that diverged 350 million years ago. However, full function of the egg coat, including sperm binding, may require proper three-dimensional assembly and may be influenced by the sites of synthesis of the subcomponents and their specific posttranslational modifications.
Studies of ZPs and their assembly into egg coats also provide insights into other ECMs. Each zone protein has a 260-amino-acid "ZP domain" that is also present in components of other specialized vertebrate ECMs, notably - and ß-tectorins that form the tectorin membrane of the inner ear [44]. The apparent absences of zona proteins in invertebrate VEs and of ZP domains in the C. elegans genome [45, 46] suggest that this ZP protein motif emerged after the split of coelomates from pseudocoelomates and, possibly, is vertebrate specific. With the impending completion of genome sequences in additional metazoans (Drosophila, human), it will be interesting to see if ZP domain modules are present in other organisms and, if so, what roles these proteins play in the formation of ECMs.
Species Specificity of Sperm Interactions with Egg ECM or Plasma Membrane
Some of the molecular interactions between sperm ligands and egg ECM or plasma membrane molecules, in both vertebrates and invertebrates, occur in a species-specific manner; other interactions between gamete proteins are nonspecies-specific. For example, some ZP-binding proteins on sperm, such as zonadhesin, appear to function in a species-specific manner (i.e., zonadhesin isolated from pig sperm binds to the pig zona but not to the mouse zona or the Xenopus VE), whereas as other sperm proteins, such as acrosin, recognize the zonae of multiple species [47]. As discussed during this minisymposium, abalone lysins show a very impressive level of species-specific interaction between lysin on sperm and VERL on the egg's VE. In fact, portions of lysins (e.g., two tracks of basic amino acids) are quite invariable, whereas N-termini and C-termini are divergent. Studies using recombinant forms of chimeric lysins from two different abalone species have confirmed that the N- and C-termini, as well as one internal region (residues 103–108), mediate species-specific dissolution of VERL [48].
Species specificity of gamete membrane interactions is evident even in mammals. In in vitro studies, zona-free mammalian eggs tend to fuse only with sperm from the same or a closely related species; the exception to this is the egg of the golden hamster, which can fuse with the sperm from many mammalian orders when the zona is removed from the egg [49]. As discussed during this minisymposium, fertilin-ß is a sperm ligand that participates in sperm-egg membrane binding, with the ECD tripeptide in the disintegrin domain being very important for adhesion of this protein to its cognate receptor on the egg membrane [15]. However, in contrast to the highly divergent functional domains in abalone lysin, the ECD functional tripeptide of the fertilin-ß disintegrin domain is very well conserved in the homologues of fertilin-ß in all species (see Table 1 in [13]), making it seem to be unlikely that fertilin-ß alone mediates species-specific gamete membrane interactions. However, many different molecules are hypothesized to be involved in sperm-egg membrane binding and/or sperm-egg fusion [13], and several of these could mediate the species-specific occurrence of gamete membrane interactions. Nevertheless, identification of fertilin-ß's functional domain may provide insight into how some ADAM family members function as cell adhesion molecules, in both sperm-egg adhesions [16, 50, 51] and other cell-cell adhesions. Interestingly, the CD of the ECD is completely conserved in comparable regions of the sperm ADAM's fertilin-ß, fertilin- (ADAM1), cyritestin (ADAM3), and other ADAM family members, including an ADAM-like protein in S. pombe (accession Z98849) and two of the seven putative C. elegans ADAMs (ADM-1 [52] and Wormpep database identification C04A11.4 [53]). Whether this conservation is biologically significant, however, remains to be demonstrated.
CONCLUSIONS
Impressive levels of both conservation and divergence are evident in gamete functions. Some gamete features, such as vertebrate egg ECMs and sperm ADAM proteins, appear to be relatively consistent through multiple species. The conservation in these mammalian ADAM sperm ligands contrasts strikingly with the divergence of lysin between species in the same genus. This is a fascinating difference between fertilin-ß and lysin, considering that primates and rodents (two orders with identical ECD tripeptides in the functional domains in their fertilin-ß molecules) diverged 80 million years ago, yet the different abalone species within in the same genus (Haliotis) diverged only 10–20 million years ago. This also contrasts with the conservation of ZP proteins as components in egg ECMs among vertebrate classes that diverged 650 million years ago. These observations could be representative of the different pressures on species with different reproductive strategies and/or different pressures on different genes. The studies discussed in the BioPore minisymposium have interesting implications for our understanding about the basis of species specificity in gamete interactions, and they clearly illustrate the wide variety of insights that can be gleaned from studying gamete function in multiple experimental systems.
ACKNOWLEDGMENTS
I gratefully acknowledge the contributions and comments from my fellow speakers in the BioPore minisymposium, Vic Vacquier (session chair), and Jurrien Dean.
FOOTNOTES
First decision: 10 March 2000.
1 Correspondence: Janice P. Evans, Division of Reproductive Biology, Department of Biochemistry and Molecular Biology, Johns Hopkins University School of Public Health, 615 N. Wolfe St., Room 3606A, Baltimore, MD 21205. FAX: 410 614 2356; jpevans{at}jhsph.edu 
Accepted: March 20, 2000.
Received: February 15, 2000.
REFERENCES
, a germ cell specific transcription factor involved in the coordinate expression of the zona pellucida genes. Development 1997; 124:4939–4947.[Abstract] subunit to mouse eggs: evidence for function as a cell adhesion molecule in sperm-egg binding. Dev Biol 1997; 187:94–106.[CrossRef][Medline] and ß in fertilization. Biol Reprod 1998; 59:145–152.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
