| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Departments of Obstetrics and Gynecology and of Cell Biology and Anatomy, Cornell University Medical College, New York, New York 10021
INTRODUCTION
Spermatozoa and oocytes of many eutherian (placental) mammals, including the human, are now manipulated successfully in a variety of ways for reasons having to do particularly with human infertility and domestic animal production, as well as basic research. Even so, several evolutionarily novel features of their function and form largely defy explanation still [1]. Most obvious in the male is the absolute dependence of eutherian spermatozoa on the environment of the upper epididymis for final maturation, and on an androgen- and usually a scrotal temperature-regulated system in the distal epididymis for their storage. Eutherian spermatozoa also have to undergo capacitation in the female tract before they can fertilize and, again for reasons that are unclear, the eutherian sperm head has adopted a novel design quite different from that in all other vertebrates. The equally unusual character of the minute egg is reflected in the ubiquitous cell mass of the cumulus oophorus around it and in a relatively prominent zona pellucida. By virtue of its thickness and resilience, this coat greatly outstrips the thin vestment around the eggs of other vertebrates and of invertebrates that have provided models for much of fertilization research. Finally, eutherian spermatozoa fuse with the oolemma by way of a segment of plasma membrane over the central (equatorial) region of the head, rather than by the apex of the inner acrosomal membrane (IAM), as in most other groups.
What these novel gamete features mean for the mechanisms of fertilization in Eutheria is as yet uncertain. Over the last 50 years, a general picture has been gained in a few eutherian species of the way that spermatozoa approach, penetrate, and activate the egg. Even so, there is no consensus yet about the molecular events through which spermatozoa bind to the zona pellucida, fuse with the oolemma, and then induce cortical granule exocytosis and resumption of meiosis [2]. Until recently, the one central step about which there has been some agreement is the sperm's passage through the zona pellucida. Variant modes of egg coat penetration include that of tailless crustacean spermatozoa that generate forward movement only by acrosome eversion on contacting the egg coat [3]; and some insect and teleost fish spermatozoa lack an acrosome and penetrate via a micropyle. Otherwise, this step is generally considered to involve lysis of a hole in the egg coat by acrosomal enzymes, in conjunction with some forward thrust from the sperm tail. Other functions proposed for such enzymes in mammals include development of the acrosome reaction [4, 5], secondary sperm head binding to the zona pellucida [6, 7], and even sperm fusion with the oolemma [8].
How well substantiated is the zona lysin theory in eutherian mammals? In considering this question in light of the information available some 10 years ago, Hunter [9] stressed the need then for a balanced view that would at least invoke a role for both enzymatic and mechanical factors. More recently, Yanagimachi [10] has detailed much of the evidence for and against the concept of zona lysis. However, sperm lysis of a path through the eutherian zona is still treated as a fact in most general texts, and in many analyses of sperm function (e.g., [2, 11]). Even the occurrence of fertilization in "knockout" mice with a null mutation for the putative zona protease, acrosin, merely evoked the idea of enzyme redundancy and the goal of identifying alternative enzymes as candidates [12]. However, there are other inconsistencies that bring the premise of zona lysis into question. Such doubts come from largely un-noted idiosyncratic features of the behavior of fertilizing spermatozoa in the Eutheria, from a broader comparative perspective with regard to sperm head design and the particular character of the egg coat, and from the results of certain experiments. In considering this issue here, a synopsis is given of the evidence on which the lytic concept is based ("the evidence for"), followed by the comparative and experimental observations that collectively undermine that idea ("the evidence against"). For the most part, the citations refer to original observations and to some representative reviews that cover the now extensive literature on mammalian fertilization.
"THE EVIDENCE FOR"
Reproduction was initially something of a laggard compared to other aspects of mammalian physiology and, at the time of increasing focus on mammalian fertilization more than 40 years ago (e.g., [13, 14]), knowledge of its mechanisms was minimal. However, a natural expectation existed that the general principles of fertilization recognized in other orders, including the formation of a hole in the egg coat by the lytic action of acrosomal enzymes [15], would apply to mammals as well. Austin and Bishop commented in 1957, "A mammalian zona lysin has yet to be demonstrated, but a small hole left in the zona after passage of the spermatozoon testifies to its probable existence" [14]; and in 1958 they noted that "the perforatorium may carry a lysin capable of altering the zona substance in such a way as to permit the spermatozoon to pass through into the perivitelline space" [16]. Thus, the belief that spermatozoa utilize acrosomal enzymes to penetrate the much thicker zona pellucida of eutherian mammals was, in a sense, preordained. Since then, a number of hydrolytic and other enzymes have been detected within the eutherian acrosome [17], and support for the concept of zona lysis has come from a) the effects of acrosomal extracts or individual enzymes on the integrity of the zona, b) the incontrovertible fact that an acrosome reaction is required for sperm penetration through the zona, and c) fertilization suppression in the presence of protease inhibitors.
a) Effects of Enzymes on the Zona
The door to the anticipated zona lytic function of the eutherian acrosome really opened with the observation that lipoglycoprotein extracts of rabbit, ram, and bull sperm acrosomes could disperse much of the cumulus/corona and slowly distort, or in a few cases eventually dissolve, the rabbit zona pellucida [18]. Subsequently, the most potent acrosomal enzyme that sometimes dissolved the rabbit zona in about 90 min [19], a serine protease (E.C. 3.4.21.10), was named "acrosin" [20]. There have been many demonstrations since that soluble extracts of the acrosome can distort and sometimes finally solubilize the zona pellucida [21–23]. However, as illustrated later here, release of the content of the acrosome at the surface of the eutherian zona does not appear to change that surface. Moreover, homologous acrosin does not solubilize the sheep zona [24], and although the zona of the mouse is relatively sensitive to mouse acrosin [25], it is penetrated by acrosin-deficient spermatozoa [12, 26]. Nonetheless, acrosin is seen generally as the key zona lysin, though it has also been proposed that other enzymes (e.g., arylsulphatase A and ß-N-acetylhexosaminidase) might act, singly or in concert with acrosin [11, 27–29].
In interpreting the effects of acrosomal extracts or individual enzymes, a first caveat is raised by the extreme susceptibility of the relatively prominent marsupial zona to protease action. While certainly thinner (usually ~2–4 µm) than the eutherian zona (~8–16 µm), the marsupial egg coat dissolves completely in only 2–8 sec in a 0.1% solution of trypsin or chymotrypsin [30, 31], as compared to 
2 min for that of various Eutheria [21]. Thus, although the eventual response of the eutherian zona to acrosomal extracts and specific proteases first seemed consistent with its penetration being a lytic event, it is in fact relatively resistant in that respect.
b) Need for the Acrosome Reaction
Spermatozoa with intact acrosomes adhering to the zona pellucida have been seen to deeply indent it sometimes, for example in the human [32] and rabbit (unpublished results). Nevertheless, with the exception of rare heterologous gamete combinations [33], an acrosome reaction is a prerequisite for sperm penetration through the zona. It should be recognized, however, that the acrosome reaction not only releases enzymes but obviously changes the apical profile of the sperm head from a blunt to a tapering edge that would favor penetration dependent on physical thrust. Moreover, as has been noted [4, 5], the enzymes themselves may play a part in completion of the reaction, and subsequently in a secondary phase of sperm binding to the zona surface [6, 7], that is nonenzymatic [34].
c) Effects of Inhibitors
One of the main pieces of evidence for lysis in penetration of the eutherian zona is the fact that fertilization fails to occur in the presence of enzyme inhibitors [20]. Fertilization suppression in the presence of various protease inhibitors was first observed in rabbits, and subsequently in rodents, monkeys, and, recently, humans [35, 36]. However, the use of such inhibitors introduces two important elements that make their action difficult to interpret in terms of zona penetration. These agents generally prevent the acrosome reaction [4, 5, 35, 37] or dispersal of the acrosome matrix [38–40], and they may compromise the prior step of sperm binding to the zona, given the possible enzyme dependence of this key event [6, 7].
"THE EVIDENCE AGAINST"
Several lines of evidence now implicitly question the likelihood that penetration of the zona pellucida matrix in eutherian mammals depends upon acrosin and/or other acrosomal hydrolases. Such doubts come from a) comparative observations of the disposition and behavior of fertilizing spermatozoa, b) perspectives on penetration as a function of sperm head design and character of the zona, and c) experimental outcomes.
a) The Behavior of Fertilizing Spermatozoa
It has been overlooked that the mode of egg coat penetration differs radically in eutherian mammals from that of the subtherian groups in which a lytic role for acrosomal enzymes seems well founded (e.g., sea urchin [41, 42]; ascidian [43–46]; mollusc [47–49]; anuran amphibian [50–52]; chicken [53–55]).
In these subtherian groups, acrosomal enzymes produce a focal dissociation of the thin egg coat. As shown in Figures 1–3, local disassembly of the egg coat fibers by acrosomal lysin(s) creates a hole of somewhat greater diameter than that of the sperm head, through which this then passes. In molluscs, egg coat lysin acts in a nonenzymatic (stoichiometric) manner [56] that probably depends on the disruption of hydrogen bonds [48], whereas the content of the chicken acrosome acts enzymatically to induce a focal "depolymerization" of the vitelline envelope [53, 54].
|
In Eutheria, by contrast, the interaction of spermatozoa with the egg coat appears quite different, with some variation among the many genera. Local release of soluble acrosomal enzymes has no evident effect on the zona surface, a penetration path through unchanged zona matrix then being established by a sperm head devoid of acrosome content. In some shrews, at least, the acrosome is shed already within the cumulus, and binding to the zona is established apparently by barbs of the perforatorium [57]—structures that, as a matter of some interest, occur also in other insectivores (unpublished results), elephant shrews [58], and even megabats [59]. Moreover, it is intriguing that a zona protein, ZPC, is synthesized by cells of the mature bovine follicle [60], since the cumulus oophorus can induce the acrosome reaction in at least some "higher" eutherian mammals as well. Therefore, sometimes acrosomal hydrolases may be released or exposed before contact with the zona.
In many cases, however, the sperm head reacts after it binds to the zona surface, on which it then tends to lie more or less horizontally before penetration (Fig. 1). Not uncommonly, one surface of the head suggests a slight convexity and the other an essential flatness or even a slight concavity; and in the pig, cow, sheep, horse, and rabbit at least, it is the latter surface that appears to face the zona at the onset of successful penetration ([61, 62]; P.J. Dziuk, personal communication). Whether this orientation to the zona depends in any way on an asymmetry of the sperm head surface properties [63] is not known. Although the initial width and angle of the subsequent penetration path can vary [64], a fairly typical example for many species is that in Figure 4, its downward curvature being due probably to the asymmetrical profile of the head [61, 62].
|
In a few eutherians studied, zona binding appears to be enhanced after the acrosome reaction by enzymes that are linked to the insoluble matrix/outer acrosomal membrane (OAM) complex [65–67]. The latter relationship acts to maintain elevated enzyme concentrations locally at the binding site [68–70], and could provide the sperm-related base from which proacrosin or other enzymes can cross-link with the zona. Significantly, however, the reacted OAM/matrix complex also lifts away from the now bare IAM (Fig. 5). In attempting to reconcile this with the concept of zona lysis, I first suggested that a residual lytic activity might function on the IAM [64]. This, however, has not been borne out. Using a gelatin substrate, Gaddum-Rosse and Blandau [71] observed that the abundant proteolytic activity of the intact guinea pig acrosome was quite absent in > 95% of moving spermatozoa after reaction and loss of the acrosome (Fig. 6), with only a faint residue persisting in the remainder. They commented then that "the presumed role of acrosin (proteases) in zona penetration cannot be fully appreciated until its presence on the heads of capacitated acrosome-reacted spermatozoa can be unequivocally established." But this has proven difficult too. Acrosin or proteinase precursor is readily detected within the acrosome and in the matrix/OAM complex of reacted spermatozoa, and sometimes on the equatorial segment of reacted sperm heads, probably through secondary adsorption [72–78]. Some PH-20 molecules apparently remain anchored in association with the naked IAM [79]. However, notwithstanding a claim for the human [80], there is no convincing evidence for protease activity at any concentration over the leading edge or either surface of the IAM after loss of the reacted shroud and matrix of the acrosome.
|
|
It has been suggested that at least the first stage of sperm head intrusion into the zona may be facilitated by acrosomal enzymes [10, 81]. However, even that appears very doubtful. As noted, binding of an intact sperm head does not seem to produce any local change in the zona surface, during or after the acrosome reaction. In a variety of eutherian species, transmission electron microscopy (TEM) views typically show a zona surface visibly unchanged beneath the reacted acrosome, the putative point of enzyme release (Fig. 5). Moreover, sperm-associated clefts in the zona surface considered to result from enzyme action [81] are in no way a consistent feature of the initial phase of sperm entry, as seen in the scanning electron microscope (SEM) (Fig. 7) or the TEM.
|
In summary, there are essential differences between eutherians and most other groups in the way that spermatozoa interact with the egg coat (Fig. 1). Rather than creating a preformed hole in a very thin coat, the release of acrosomal lysin(s) from the reacting eutherian sperm head has no apparent local effect on the zona. As the sperm head penetrates, the IAM appears devoid not only of visible matrix but also of acrosin. The penetration path extends on occasion beyond sperm heads caught within the zona [64, 82], as if its matrix were being split (Fig. 8), and this was first interpreted in the light microscope as a sperm head "filament" [61]. Neither this slit nor the complete penetration path shown in Figure 4 conveys any suggestion of the lytic dissociation of matrix associated with sperm entry in subtherian groups.
|
b) Gamete Design
To paraphrase Williams [83], the emergence of a new design in biology can be attributed to a long period of selection for a particular role, and, allied to the laws of physical science it can explain, or at least imply, that role. Gametes, and therian gametes in particular, exemplify this. The move to internal fertilization in vertebrates has generally been accompanied by a more complex organization of the sperm tail, by elongation of the sperm nucleus [84–86], and by a transition to the more arginine-rich protamines [87]. However, Williams's principle applies further in the case of therian mammals. Whereas the gametes of monotreme mammals have remained essentially reptilian [88, 89], the advent of marsupials and eutherians witnessed not only a major reduction in egg size but—central to this discussion—a divergent or two-tiered hypertrophy of the egg coats paralleled by novel but radically different sperm head designs. In marsupials, the novel elements in the sperm's architecture seem to optimize application of the acrosomal content to a more protease-sensitive zona, whereas the unusual design of the eutherian sperm head seems implicitly related to physical thrust.
The marsupial situation Compared to the ~0.5- to 1.0-µm-thick egg envelope in subtherian groups, the zona is distinctly thicker at ~2–4 µm in most marsupial genera. An angulated relationship with the head allows the tail of the marsupial spermatozoon to apply directed pressure to the zona surface through the whole face of the acrosome. Nonetheless, that unusual endpoint—a "sandwich" application of the whole acrosome content between one face of the fertilizing sperm head and the zona surface—depends on slightly different strategies in representative American and Australian marsupials (Fig. 9). Both have an eccentrically placed acrosome and a central nuclear insertion of the tail. However, American didelphid spermatozoa adopt a fixed head-tail angle apparently offset by a final pairing in the epididymis [90, 91], their separation occurring prior to fertilization [30, 92]. The Australian pattern, as exemplified by Sminthopsis, involves a mobilization of the tail-head junction, transforming a previously streamlined profile to a potential "T" shape in spermatozoa at the fertilization site [31]. The fact that these angled configurations apparently allow marsupial spermatozoa to apply the whole content of the reacting acrosome against a highly protease-sensitive zona pellucida suggests the involvement of acrosomal lysins in penetrating it. Consistent with this, penetrating spermatozoa of the American marsupials, Didelphis and Monodelphis, leave a wide hole in the zona that mirrors the flat profile of the head [30, 92]. However, since spermatozoa revert to a streamlined form during penetration of the somewhat thicker (~4.0 µm) zona in Sminthopsis [31, 93] (Fig. 9), that step in Australian marsupials may require more of a mechanical element as well.
|
The eutherian situation In most eutherians studied, the thick (~8–16 µm) zona pellucida behaves elastically [94], is cross-linked to a degree by intermolecular -S-S- bonds [95–97], has an intrinsic rigidity or structural "memory" (an empty zona often remains spherical), and, as noted earlier, is far more resistant to proteolysis than that of marsupials. There has been some discussion as to whether the directed thrust that capacitated spermatozoa can mount is sufficient in itself to overcome the inertia of the eutherian zona matrix [94, 98, 99]. However, penetration involves not merely directed thrust but, importantly, a levered or oscillating action of the head [1, 65, 99], specifically permitted by its particular design.
The specifics of this design can vary in minor ways. For example, many ultrastructural studies have revealed that the human sperm head really lacks a perforatorium, that it has a somewhat wider sagittal profile, and that its nucleus has a relatively low complement of protamine-related cysteine [100], on which the -S-S- based rigidity of the sperm head appears to depend [101]. One can perhaps infer from these features that the human sperm head is marginally the least well adapted among the Eutheria for thrust. However, in virtually all, including man and other hominoids, the sperm head has evolved far from an ancestral type that is likely to have been nonstabilized and round in cross section, as in reptiles and many birds. Penetration seems very likely to depend, on the one hand, on the hyperactivated form of motility seen in capacitated spermatozoa [10, 102]. However, a physical mode of penetration is implied also in at least three novel features of the eutherian sperm head (Fig. 10). As noted above, the uniquely flat profile of the head (nucleus) permits the lateral oscillating movement seen in penetrating the zona (Fig. 11) (hamster [65, 99]; human [1]). The force exerted per unit area would seem to be optimized by the sharp frontal profile, and the scything action of the oscillating head must compound the stress that its leading border applies to the zona matrix, very much like the movement of a sawing knife. This flattened configuration of the nucleus is fundamentally no different in the murid and cricetid rodents that have a falciform sperm head, or in the hydramine rodents in which the anterior nucleus and head cytoskeleton forms two or three prongs [103]. Second, the perinuclear theca and the nuclear protamine display an unusual complement of thiols that become cross-linked during epididymal maturation [104, 105]. Such -S-S- cross-linking not only stabilizes but also rigidifies the sperm head [101]. Third, the IAM—the interface with the zona matrix during penetration—displays an unusual crystalline lattice organization and a relative toughness or resilience [106]. The latter features appear able to counter reactive forces and shear stress that may develop over the leading surface of the spermatozoon, and they add further weight to the concept of eutherian zona penetration as a physical event.
|
|
An unexpected deviation manifested by some species in the genus Bandicota (bandicoot rats) also supports the scenario proposed above. In B. indica and B. savilei (otherwise conventional murid rodents), the sperm head is round in cross section [107]. This design would seem to preclude lateral oscillation of the sperm head, limiting spermatozoa to the lytic "drilling" of the egg coat typical of subtherian groups; and in accord with that, the sperm head displays an acrosome of inordinate bulk that has no equatorial segment. At ~4 µm (W.G. Breed, personal communication), the zona is much the thinnest observed among Eutheria, and when oocytes (n = 12) recovered from eCG/hCG-stimulated follicles of B. indica were placed in 0.1% trypsin, the zona dissolved in ~15 sec (J.M. Bedford, unpublished results)—a sensitivity close to that of the marsupial zona. Thus, an unusually trivial and protease-sensitive zona in this Bandicota species is paralleled by a sperm head that displays a unduly large acrosome but lacks typically eutherian features considered here to reflect a physical penetration mode.
c) Experimental Outcomes
Doubts about sperm lysis of the eutherian zona that come from the behavior and design of the gametes are reinforced by three studies cited briefly below.
First, a significantly greater resistance of the rabbit zona to serine proteases, brought about by fertilization itself, or by the lectin wheat germ agglutinin, in no way retarded sperm penetration through it [108]. Similarly, penetration of the mouse zona in vitro was prevented by a protease inhibitor only when this was added to the medium before, not after, sperm binding to the zona surface [109]. The suggestion from these experiments—that serine protease action may be unimportant for zona penetration—is strengthened by the observation that acrosin-deficient spermatozoa are able to penetrate the zona pellucida and fertilize the mouse egg [12, 26], as are those lacking another acrosomal enzyme, galactosyltransferase [110]. Such outcomes do not in themselves rule out the possibility that zona penetration could be effected by the joint activity of other acrosomal hydrolases on sulfated components and on glycosidic linkages within the zona substance. However, these enzymes appear to have little or no visible dissociative effect on the eutherian zona.
THE ZONA PELLUCIDA
The associations outlined above carry a strong implication of cause and effect between the character of the zona pellucida, the unusual way in which the gametes interact, and the novel design of the therian sperm head, whether marsupial or eutherian. But why has the zona pellucida developed as such a thick elastic coat in eutherian mammals, and what is the structural basis of its unusual shell-like resilience?
As noted, the marsupial zona is significantly thinner than that of Eutheria, varying from 
2.0 µm in American to ~4.0 µm in most Australian species [111, 112], up to a maximum in the koala and wombat of ~8.0 µm (J. Chapman and W.G. Breed, personal communication)—the low of the eutherian range. In the present context, however, its 100-fold greater protease sensitivity as measured in two marsupial genera, noted earlier, seems perhaps more significant. The marsupial zona acquires a "mucoid" coat soon after fertilization, and then an outer shell membrane as it passes through the uterotubal junction [113, 114]. The zona then disappears as the unilaminar blastocyst begins to expand, with the distensible shell membrane remaining as the limiting coat for about two thirds of gestation until implantation [115].
In Eutheria, except for the lagomorphs (see below), the zona persists as the only egg vestment into at least the initial stage of blastocyst development—a phase that usually involves a significant expansion. This expansion increases the surface area of the blastocyst, and coincidentally it reduces the thickness of the elastic zona to one fifth or less of that around the egg, e.g., in cows [116, 117], sheep [118], fissipede carnivores [119], horses [120], and humans ([121]; Fig. 12). Although expansion is minimal in the mouse and rat, by Day 6 or 7 after ovulation the equally small egg of the shrew, Blarina brevicauda, develops as an expanded blastocyst of at least a 15 times greater diameter, invested still by a now much thinner zona (O.B. Mock and J.M. Bedford, unpublished results). On reflection, this general picture suggests that the thick resilient elastic character of the eutherian zona pellucida allows it to accommodate the stretching and thinning brought about by expansion of the blastocyst. It is very unlikely that the 0.5–1.0-µm more diffuse egg coat characteristic of most other groups could withstand such distension.
|
Why the eutherian zona always persists at least into the early uterine phase of blastocyst development is not entirely clear. It is evident that the zona acts in most species as a significant block to polyspermy after cortical granule exocytosis and, coincidentally, often develops a greater resistance to protease action and/or to deformation ("hardening"), perhaps through further cross-linking of its constituent molecules [122, 123]. The rabbit and other lagomorphs constitute a special case in that the zona acquires an external mucin coat beginning about 10 h after ovulation. In the absence of any zona block to polyspermy, this mucin layer could function to stop further penetration by ampullary spermatozoa at a time prior to first cleavage when the vitelline block has weakened, as eventual support for the highly stretched zona, and/or for some other aspect of implantation. In any case, though C.R. Adams observed that rabbit blastocysts from which the zona had been removed did not survive (cited in [124]), investigation of the function of the zona after fertilization has been limited primarily to small rodents. Zona-free rodent eggs can rarely be recovered more than 2 h or so after their transfer to the oviduct [125–127]. This was also the case for intact rabbit eggs transferred after pretreatment with 0.1% trypsin or chymotrypsin, unless first exposed to protease inhibitors [128]. On the other hand, morulae transferred without a zona sometimes eventually implanted normally in the mouse uterus [126, 129], and some eggs fertilized and/or cultured in vitro without a zona developed normally after transfer as blastocysts in mice and cattle [130, 131]. However, there is evidence that the zona may optimize normal early development by maintaining an appropriate cell-cell contact configuration beginning at the four-cell stage [132]. Moreover, culture of pig embryos suggests that the zona also can act to regulate the rate of cell division in the early embryo, and particularly cell allocation to the inner cell mass and trophectoderm [133, 134]. Thus, while apparently not essential for development per se, the zona probably acts to favor development until the formation of intercellular junctional complexes at compaction, as well as to protect or maintain a free independent existence for the conceptus in the tubal environment.
Whatever the zona's role during blastocyst development, hatching from it is soon followed by implantation in species such as the mouse, rat, and human. Not uncommonly, however, the hatched blastocyst continues to grow for a variable period where the preimplantation period is relatively prolonged; and in several animals, external acellular material elaborated in at least some cases by the trophoblast may become a limiting coat after the loss of the zona [119, 135,136].
What determines the relative protease insensitivity and elastic shell-like character of the eutherian zona is unknown. Some constituent glycoproteins of the egg coat may have been conserved throughout vertebrate evolution, since similar gene sequences are found in amphibia and even in teleost fish [137, 138], but the physical character of the egg coat can vary greatly. For example, the fibrous ultrastructure of the mature anuran coat [139] can be distinguished easily from that of the chicken and as easily from the uniform texture of the zona in eutherian mammals. The components and to some extent the organization of the latter have been characterized, particularly for the mouse and pig [137, 140]. The fact that the zona of the fertilized eutherian (hamster) egg is immediately dissociated by SDS implies that its resilient character does not depend primarily on covalent bonds [123]. However, since dithiothreitol alone eventually dissociated the zona of the rabbit and mouse [95–97], but had no effect on that of a marsupial [31], disulfide bonds may be important among the cross-links on which the special physical nature of the eutherian zona presumably depends.
COMMENT
In view of the evidence now at hand, it can no longer be assumed that sperm penetration of the substance of the zona pellucida in eutherian mammals depends on the lytic action of acrosomal enzymes. The present arguments do not prove that lysis has no role in this, but the contrary evidence is too striking to justify continuing supposition that it does. Summarized simply, it appears that the unusual thickness and elastic resilience of the eutherian zona, which allow it to accommodate by stretching and thinning during blastocyst expansion (Fig. 12), at the same time may preclude a penetration mode based on acrosomal lysins. The collective evidence suggests that changes in the character of the eutherian zona that allow it to remain during blastocyst expansion have modified the strategy for its penetration from a mechanism based partly on lysis to one based on cutting thrust. Thus, ultimately, certain functional needs of the preimplantation embryo may have been the selective force for development of the particular character of the zona pellucida—and in turn for the novel architecture of the eutherian sperm head and of the novel means whereby it penetrates the zona. These arguments do not necessarily imply that acrosomal enzymes no longer have any role in Eutheria. Experiments involving protease inhibitors, or targeted deletions of acrosin or galactosyltransferase, suggest that cooperative enzyme activity plays some part in development of the acrosome reaction; and enzyme(s) complexed to acrosome matrix probably promote binding to the zona surface.
In reevaluating this issue, it appears that too little attention has been paid to unusual features of eutherian eggs and especially spermatozoa, and of the interaction between them. The lysin theory was predicted for eutherian mammals when there was little appreciation of the degree to which the design and precise behavior of their gametes differ from those of other groups, ranging from invertebrates to birds (Fig. 13). Not only does the eutherian zona surface remain intact where the sperm head binds and reacts; that binding finally appears to involve an acrosomal matrix component that has not been recorded in the types of spermatozoa that obviously lyse a hole in the egg coat. Complexed to interrupted OAM, this matrix may stabilize the reacted shroud and/or act as a scaffold from which an enzyme such as proacrosin/acrosin could cross-link with the zona surface. However, during the natural acrosome reaction, this visible complex lifts away from the IAM, which interacts with the zona matrix as a bare though stable membrane devoid of any significant concentration of acrosin at least. Furthermore, not only do acrosomal lysins seem to operate as soluble agents in other groups, but an IAM undergoing rapid lateral oscillations may need to be devoid of protease/zona receptor molecules in penetrating a matrix to which they could cross-link. Therefore, the notion that IAM-bound lysin(s) facilitates zona penetration in Eutheria may be flawed even as an idea.
|
There is no indication, either, that the path left by penetrating eutherian spermatozoa results from lytic dissociation of the zona matrix. Indeed, the immediate dissociation response of the marsupial zona underscores the fact that the eutherian zona is relatively insensitive to serine proteases. Although eventual solubilization by acrosome extracts or acrosin has been cited as indirect evidence for lysis as the means of its penetration, the zona shows no response to the acrosomal content released where the sperm head binds and reacts. Even the slower fertilization by acrosin- or galactosyltransferase-deficient mouse spermatozoa [12, 26, 110] apparently reflects a delay in development of the acrosome reaction and/or in functional binding, not an inability to penetrate the zona or to undergo fusion with the oolemma and activate the egg. In the case of humans, TEM studies often show sperm heads within narrow channels comparable to those generally seen in animal eggs [141, 142]. However, in human eggs often necessarily fixed 48–60 h after insemination, spaces suggestive of local lysis sometimes are present, though with some evidence of degeneration [32]. Therefore, since the sperm head of the human and other hominoids may seem somewhat less well designed perhaps for a scenario involving physical thrust, a reexamination of human material fixed soon after insemination might be justified. Otherwise, the evidence points to physical penetration through the resilient zona matrix by a sperm head devoid of lytic activity at the leading edge of the IAM; and on occasion it has been possible to detect distortion of the zona matrix fibrils by the advancing sperm head [106]. The summary diagram in Fig. 13 illustrates the drastic change in the general form of the eutherian sperm that favors the hypothesis of oscillating physical thrust. Although they may be aberrant examples in the eutherian scheme, situations such as the precocious loss of the acrosome in the musk shrew and the anomalous gamete features displayed by some bandicoot rats provide further evidence for this view.
Exactly where marsupials stand in this regard is uncertain. The marsupial zona has undergone a more modest hypertrophy, it disappears with the onset of blastocyst expansion, and it is fortified well before that stage by other coats. While the precise mechanisms of zona penetration may differ between American and Australian forms (Fig. 9), the design of their spermatozoa seems to favor application of the whole acrosome content against a highly protease-sensitive zona, implying some involvement of lysis in its penetration. It is interesting, therefore, that the complement of hyaluronidase, acrosin, arylsulfatase, and hexosaminidase in the marsupial (Didelphis) acrosome exceeds that in the rabbit acrosome [143]. In fact, the later evolution of eutherian mammals has often witnessed a significant reduction in the size of the acrosome. There is much variation in this regard among both the Chiroptera and Rodentia. However, in insectivores—shrews [144–147], moles, hedgehog ([148]; O.B. Mock and J.M. Bedford, unpublished results)—and elephant shrews [58], the rostrum of the acrosome is exaggerated in comparison with that of higher mammals such as carnivores, ungulates, and lagomorphs. Within the primates, the evolutionary transition from the prosimian to the simian and finally to the hominoid line also has been accompanied by a notable reduction in this regard [149].
Finally, acceptance that the eutherian zona is generally penetrated by physical means could explain the evolution of the acrosome's equatorial segment. The particular constitution of the exposed IAM [106] may provide a necessary counter to the reaction and the shear forces that oscillating thrust against the zona may create, but not the labile quality required for the traditionally fusogenic role of this membrane. The evolution of an alternative more-posterior fusion site, displaced away from the sperm head's leading edge, conversely provides one more, if indirect, hint of reactive forces generated in the prior step of zona penetration. Support for this is found in the rare case of B. indica, whose cylindrical sperm head, large acrosome, and protease-sensitive zona all imply a lytic mode of zona penetration—but where there is no equatorial segment.
CONCLUSIONS
In eutherian mammals, acrosomal enzymes could help some spermatozoa to penetrate the cumulus oophorus, and they appear to facilitate development of the acrosome reaction and then functional binding of the spermatozoon to the zona pellucida. Acrosomal enzymes have been considered to function primarily, however, in lysis of a path in the eutherian zona pellucida, as in other groups ranging from invertebrates to birds. This last supposition is now brought into question for eutherian mammals as a result of comparative focus on the details of gamete interaction, through a wider comparative perspective on the several novel features of eutherian gamete design, and by certain experimental results. The weight of this evidence indicates that eutherian spermatozoa have evolved a wholly new strategy for zona penetration based on cutting thrust. This novel strategy appears to have evolved in response to the resilient elasticity and thickness of the eutherian zona—unusual features that allow it to accommodate the stretching and thinning undergone during preimplantation expansion of the blastocyst but that appear to preclude a conventional lytic mode of its penetration.
|
|
ACKNOWLEDGMENTS
I am very grateful to Drs. K.J. Betteridge, W.G. Breed, and G.E. Olson for helpful comments, and to Pauline Thomas for the illustrations.
FOOTNOTES
1 Correspondence: J.M. Bedford, Box 30, Cornell University Medical College, 1300 York Avenue, New York, NY 10021. FAX: 212 746 8589; mbedford{at}mail.med.cornell.edu 
Accepted: August 17, 1998.
Received: April 23, 1998.
REFERENCES
-Aminobenzamidine, an acrosin inhibitor, inhibits mouse sperm penetration of the zona pellucida but not the acrosome reaction. J Reprod Fertil 1982; 65:185–194.[Abstract]
