The micropyle: a sperm guidance system in teleost fertilization

A wide range of behaviourial, morphological and physiological strategies have evolved to ensure successful sperm-egg contact and fertilization in different species. In mammals where copulation and internal fertilization bring gametes into close proximity, receptor-mediated gamete recognition and adhesion is known to occur (see recent review by Wassarman, 1990). In addition to the large numbers of gametes usually produced, sperm also undergo acrosome reactions in order to release sufficient enzyme needed to breakdown the zona pellucida thereby making it possible for the fertilizing sperm to penetrate the egg at virtually any point on the zona pellucida.

In many non-mammalian aquatic and marine organisms, the primitive and wasteful condition of external fertilization prevails. It is most likely that gamete recognition at the molecular level is also a common feature in these organisms. In the sea urchin, for example, sperm have been shown to possess a protein molecule, bindin, which interacts with a glycoprotein receptor in the egg vitelline envelope in a speciesspecific manner (Vacquier and Moy, 1977).

The teleost sperm, unlike the mammalian sperm, lacks an acrosome. Sperm penetration is, therefore, entirely mechanical and dependent on the sperm squeezing through one funnel-shaped micropylar opening on the tough protective multilayered egg coat, the chorion. In order to increase the chances of sperm-egg contact, considerable reproductive effort and investment are made in the production of vast numbers of gametes, which are shed synchronously in close proximity, often following an elaborate species-specific courtship behaviour in response to physiological and/or environmental cues.

In his sperm redundancy theory, Cohen (1975, 1977) argues that only a very small proportion of the large numbers of sperm generally introduced into the female tract are suitable for fertilization. This is understandable as a proportion of every sperm sample analyzed is usually morphologically abnormal and has been shown to be selectively eliminated by the human cervical mucus (Barros et al. 1984). Assuming that Cohen’s hypothesis is also true in the teleost where fertilization is external, it seems highly unlikely (even with the vast numbers of sperm produced) that contact between.the few ‘suitable’ sperm and eggs should be entirely random in view of the added problems of sperm penetration occurring only at a single point on the egg chorion. Furthermore, the relatively short time within which fertilization must occur after gametes are in contact with water presents further constraint.

In this study, we present the first evidence of a sperm guidance system in a teleost. We propose that the development of the micropyle increases the chances of successful fertilization once some form of speciesspecific chemoattraction brings the sperm into close proximity with the egg.

Rosy Barb, Barbus conchonius, adults were obtained from local aquarists and sexes maintained separately under tropical aquarium conditions on a 12 h light: 12 h dark cycle. Prior to collecting gametes, 2 males and 3 females were put into different compartments of a breeding trap approximately 16 h before the next light period.

The transparent barrier separating the sexes was removed about 5 min after the onset of the light period and adults exhibiting prolonged mating behaviour were individually anaesthetized in ethyl rn-aminobenzoate, MS222 (Sigma, England) at final concentrations of 80–100 ppm in tank water until ventilation movement of the operculum stopped and the fish was completely immobilised in 3–7 min.

Fish were wiped dry of water and anaesthetics, held in damp tissue and stripped of gametes by the application of gentle pressure to the lower abdomen. The stripping process usually took less than 30 s and stripped fish almost always showed complete recovery on return to aerated tank water. Sperm were stripped into capillary tubes. The undiluted sperm retained normal motility for up to 2 h at room temperature but were always used within 10 min. The behaviour of sperm during in vitro fertilization was studied in fifteen batches of eggs (10–20 eggs per batch) which were stripped directly into specially designed microscope observation chambers where they were brought in contact with diluted suspension of sperm (20–25000 in 500 μ l of tank water).

Materials for scanning electron microscopy (SEM) were prepared according to standard procedures. Specimens were fixed for 24 h in freshly prepared 0.1M sodium cacodylate containing 2% glutaraldehyde at 4°C, rinsed twice in 0.1M cacodylate buffer at ph 7.2 and then postfixed for 2 h in 1% osmium tetroxide in cacodylate buffer. Dehydration was gradual in graded alcohol over the range of 30–100%. Samples were critical point dried at 31.5°C and then coated with gold-palladium in an SEM-PREP sputter coater and scanned using a Hitachi (F800) scanning electron microscope. Still 35 mm photographs were taken with an Olympus camera (C-35AD-2) coupled to an Olympus photomicroscope (BHS/ PM-10AD) using Kodak Technical Pan black and white film.

Sperm count (1.2bn per ml of undiluted sample) was done on digitised video images of sperm samples on a haemocytometer using a counting application on a VIDS IV computer image analysis package (Analytical Measuring Systems, Cambridge England). Time-lapse video (TLV) recordings of sperm motility patterns and early fertilization events were made using a high-resolution charge coupled device (CCD) colour camera coupled to an Olympus photomicroscope (BHS/PM-10AD) and a VHS recorder (JVC model BR-9000 UEK). Sperm behaviour was analyzed visually and by computer-aided image analysis of traces of actual recordings. A sperm was considered ‘guided’ if it travelled along a micropylar groove into the immediate vicinity of the micropylar pit which is the only point from which egg penetration could be achieved. On the other hand, sperm that transversed ridges and grooves, or travelled almost entirely on ridges were considered ‘unguided’ even if they eventually arrived at the immediate vicinity of the micropylar pit. Sample trajectories, based on actual motility patterns of ‘guided’ and ‘unguided’ sperm are shown in Fig. 1. A mathematical model, based on observed fertilization events, was developed demonstrating the role of the micropyle in achieving successful fertilization by increasing chances of sperm-egg contact.

The Rosy Barb egg has a maximum chorion diameter of approximately 100 microns and a large surface area of about 31,428 square microns. The micropylar region consists of a non-sticky sperm catchment area (SCA) of approximately 20 microns in diameter and surface area of 314 square microns. At the centre of the SCA is a funnel-shaped vestibule with a maximum diameter of 4.5 microns at the micropylar entrance and less than a micron at the bottom of the micropylar pit. The SCA is a system of 7–10 micropylar grooves and ridges which drain directly into the vestibule (Fig. 2). The Rosy Barb sperm head has a diameter of 1.0–1.3 microns, giving a maximum cross-sectional area of approximately one square micron, maximum head and mid-section lengths of 2 microns and an overall length of 13–17 microns, respectively. The anterior end is rounded and lacks an acrosome. Both the structure and size of the Rosy Barb sperm are comparable to the published data (head diameter of 1.8μm, flagellum length of 30ftm) for the closely related Zebra fish, Brachydanio rerio (Wolenski and Hart, 1987).

Analysis of sperm movement in the vicinity of the micropylar region of a total of over 60 unfertilized eggs showed that the fertilizing sperm, usually the first sperm to arrive, always travelled preferentially along the grooves. Using our criteria of guidance described above (see Materials and methods), 86% of sperm entering the vicinity of the micropylar region within the first 30 s of sperm-egg contact (mean=100) were guided along the micropylar groove into the immediate vicinity of the micropylar pit, giving a guidance efficiency (GE) of 0.86 (100% guidance as shown by the fertilizing sperm=GE of 1.0).

However, the behaviour of sperm coming into the micropylar region of water-hardened or already fertilized eggs showed a temporal reduction in guidance efficiency, with only 50% of sperm showing directional guidance behaviour 80s postfertilization (Fig. 3). A small proportion (20%) of such postfertilization guided sperms showed ‘searching’ behaviour once in the immediate vicinity of the micropylar pit: sperm repeatedly returned to the region of the micropylar pit after moving away from it (Fig. 4). Pattern of sperm agglutination in the micropylar region showed highly significant differential distribution between micropylar ridges and grooves: almost all the sperm agglutinated along the grooves (Fig. 5). Interestingly, sperm were not observed to agglutinate on any other area of the chorion other than the micropylar region.

The egg showed considerable increase in volume as a direct result of elevation of the fertilization membrane following sperm penetration or water activation. Such physical changes were also found to alter the physical structure of the micropylar region: the ridges and grooves became less distinct as a result of the egg becoming more rounded.

In our in vitro fertilization system where eggs were stripped directly onto fertilization chambers and fertilized in situ, preliminary observations consistently showed association between the point of sperm entrance and blastodisc formation: the blastodisc formed directly underneath the micropyle in all cases where eggs remained undisturbed following sperm penetration as typically represented in Fig. 6. However, if the eggs were disturbed then egg rotation within the chorion occurred and the relationship between the micropyle and blastodisc was lost.

There are various biochemical, biophysical, morphological and physiological factors that affect directionality in sperm motility and ensure sperm-egg contact in almost all studied animal models, especially those animals where fertilization is internal. These include fluid current and ciliary movement within, and contraction of, the reproductive tract (Hawk, 1983; Fujihara et al. 1983), diffusible components from the egg (Rossignol and Lennarz, 1983) and various other biophysical and biochemical factors in the immediate environment of the sperm (Katz et al. 1989). A significant proportion of these mechanisms are maternal contributions of evolutionary significance in that they ensure speciesspecific fertilization. In this study, we have presented the first evidence of sperm guidance in an externally fertilized system, and one where sperm entry into the egg is only achieved through a specialized and predetermined point, the micropyle.

Heterogeneity, with respect to sperm reception and penetration, of the egg surface has been demonstrated in a wide range of invertebrate and vertebrate animal models such as ascidians (Phallusis mammillata, Speksnuder et al. 1989), freshwater bivalves (Unio elongatulus, Focarelli et al. 1988), teleosts (Brachydanio, Wolenski and Hart, 1987) and anurans (Discoglossus pictus, Talevi and Campanella, 1988). Furthermore, our results, which are also consistent with results of the above studies on point of sperm entry, showed a correlation between point of sperm entry and formation of the blastodisc in undisturbed eggs. It would appear that a specialized point of sperm entry is a retained primitive feature which, in addition to admitting the fertilising sperm and preventing polyspermy, also contributes to some major morphogenetic decisions such as determination of the animal hemisphere (Speksnuder et al. 1989) and axis formation (Danilchik and Black, 1988) in animal development.

Based on theoretical calculations and assuming a sperm head cross-sectional area the same size as the micropylar pit, the probability (P_random) of any single sperm randomly migrating through the micropyle in the absence of sperm guidance is given by the equation:

With guidance, however, the probability of a sperm migrating through the micropyle once it was within the sperm catchment area would be:

Difference between the probability of ‘guided’ and ‘unguided’ sperm penetrating the micropylar pit P_(delta) is attributed to sperm guidance role of the micropylar region and given by the equation:

Given the observed sperm guidance efficiency of 1.0 (for the fertilizing sperm) to 0.86 within the first critical 30 s, and assuming that all sperm are equally capable of fertilizing an egg, our results suggest that the sperm catchment area enhances the chances of fertilization by as much as 99.7% once sperm were in the immediate vicinity of the micropylar region. In the absence of sperm guidance, it would require an enormous increase in sperm number in order to achieve the same level of fertilization.

We suggest that some form of chemical attractant emanates from the micropylar pit. The searching behaviour around the micropylar entrance may, therefore, result from traces of such a chemical attractant even after the fertilization membrane has elevated and the micropylar entrance blocked by non-fertilizing sperms (Fig. 7). The presence of any chemical attractant was, however, not the main focus of the study and therefore not investigated experimentally.

Our morphometric measurements of the micropylar pit and sperm head agree with the earlier view (Hart and Donovan, 1983) that polyspermy in the closely related Zebra fish was prevented by the physical plugging of the micropyle by the fertilizing sperm head and other non-fertilizing sperm arriving postfertilization; thus the micropyle becomes the primary physical barrier against polyspermy. Furthermore, it is plausible that the observed decrease in guidance efficiency and physical morphological changes in the egg postfertilization have some contributory roles to preventing polyspermy.

The molecular and ultrastructural basis of sperm guidance and motility along the micropylar grooves remain unclear and analysis of the mechanism a logical sequel to this study.

We would like to express our gratitude to Peter Thorogood for making available research facilities and laboratory space without which the work would not have been possible, and for his guidance and encouragement in the preparation of the manuscript. We also thank Norman Maclean, David Penman and Gary Carvalho for their helpful discussions and suggestions at various stages of the work. Heather Caldwell helped with photography. D.A. was supported by a Wellcome grant to Peter Thorogood.