Kinetics of Fertilization in the Sea Urchin: A Comparison of Methods

In order to study the rate of fertilization it is necessary to employ an agent which will both achieve a rapid functional separation of sperm from eggs and also allow the use of some method of assessing the numbers of unfertilized, fertilized and polyspermic eggs in a sample. Since present methods do not allow of a distinction at the time of fertilization between fertilizing sperm and others at the egg surface, all spermicides used must permit the eggs to remain viable long enough for the successful sperm to undergo some recognizable developmental change. In previous work the extent of fertilization has been assessed at first cleavage, and the only spermicides which have been used extensively are dilute sea water (Rothschild & Swann, 1952) and sea water containing traces of the detergent lauryl sulphate (Hagstrom & Hagstrom, 1954; Allen & Griffin, 1958). These two methods do not seem to have been compared on the same batch of eggs, which is surprising in view of reported differences both in the detailed shape of the fertilization rate curves and in their interpretation (Runnstrom, 1961). In this paper we report a comparison of these and other spermicides using a method of assessing fertilization based on counting male pronuclei. This method was mentioned in a preliminary report (Baker & Presley, 1966) and has been found to provide a more accurate measure of the degree of polyspermy than does the cleavage method.

Mature specimens of Echinus esculentus were obtained at Plymouth in late March and early April and kept in sea-water tanks for a few days until use. Mature specimens of Psammechinus miliaris were obtained from the Wash in late July and early August, transported to Cambridge and kept until use in an indoor sea-water aquarium with continuous circulation and aeration. Before use, specimens were rinsed for 30 sec. with 30 % artificial sea water (A.S.W.) to destroy any free spermatozoa and then washed with normal sea water. Release of eggs and sperm was induced by hemisecting the specimens and standing the gonad-containing half-shell over a collecting vessel. Eggs were shed into filtered natural sea water (E. escidentus) or artificial sea water (P. miliaris), and kept at the desired experimental temperature before use. Sperm was shed into a beaker surrounded by ice, containing a small quantity of a solution which was half molar with respect to both dextrose and glycine. Stock sperm so obtained was kept at o° C.

After collection eggs and sperm were tested for fertilization, and batches giving less than 95 % fertilization or containing more than 5 % eggs in the germinal vesicle stage were rejected. Before each experiment eggs were washed in A.S.W., and allowed to settle before re-washing ; all eggs were washed twice, and at the second washing were divided among test tubes so that each sample used throughout the experiment was representative of the whole population of eggs. Where pre-treatment with any agent was employed, the supernatant was removed and the treatment fluid was added for an appropriate period prior to the experiment. All solutions to which the eggs were exposed were kept with the eggs in a water bath at the experimental temperature, usually 16° C.

The density of the stock sperm solution was determined with a haemocytometer slide, after dilution with distilled water. Immediately before using sperm for a series of fertilizations, the stock suspension was agitated thoroughly and then diluted with A.S.W. to give the required sperm density. This diluted sperm was then kept at the experimental temperature.

The composition of A.S.W. was NaCl, 460 mm; KC1, 10 mm; MgCl₂, 55 mm; CaCl₂, 11 mm; NaHC0₃, 2·5 mm; pH 8·0.

The spermicidal fluids generally used (and the abbreviations used throughout this paper) were :

30-A.S.W. : 30% artificial sea water as above, pH 8·0.

Lauryl sulphate: 0·002% in A.S.W. pH 8·0.

KC1: A.S.W. in which the NaCl was replaced isosmotically with KC1, pH 8·0.

EDTA: 100 mm-Na-EDTA; 300 mm-NaCl; 10 mm-KCl adjusted to pH 8·0 with NaHC0₃.

UN0₃: A.S.W. containing 0·3mm uranyl nitrate (UO_a(NO₃)₂, 6H₂O), pH 6·3. This solution was prepared from a stock solution of 10 mm uranyl nitrate in distilled water by adding this to an appropriate quantity of A.S.W. Such a solution did not precipitate, though at higher molarities and at more alkaline pH a precipitate formed.

Nicotine (British Drug Houses) was freshly prepared before each experiment as stock solution of 5 drops (200 mg.) per 100 ml. A.S.W., pH 8·o (approximately 12·5 mm), and appropriately diluted before use.

In preliminary experiments spermicidal agents were tested for their effect on sperm motility, and subsequently for their ability to allow further development of fertilized eggs. None of the agents abolished sperm motility completely, less than 1 % of sperm remaining feebly motile for several minutes. The experimental procedure was therefore designed to effect a 1/100 dilution of sperm at the time of inactivation with spermicide. Uranyl nitrate was included because of Okazaki’s (1956) report that uranyl ions reversibly block fertilization without causing any immediate reduction in teperm motility. Preliminary results confirmed this report and uranyl nitrate was therefore compared with the other agents.

Each rate-determination experiment included a sample of eggs placed unfertilized in 15 ml. of spermicide, followed after 15 sec. by the addition of a quantity of sperm twice as great as that present in samples from the corresponding rate experiment. Such a sample was taken as a ‘zero’ time point for each experiment, the procedure stemming from the practical difficulty of mixing eggs, sperm and spermicide thoroughly and simultaneously. 30-A.S.W. was the only spermicide which repeatedly failed to block fertilization in 100% of the eggs. With sperm densities of 10⁷ and 10⁸ sperm/ml., the percentage of eggs fertilized in the presence of 30-A.S.W. averaged 22 ± 8 and 40 ± 6 % respectively.

A further measure of the effect of the spermicides was obtained by measuring the oxygen consumption of sperm in the presence of the agents. The oxygen consumption was reduced to less than 10 % of its control value by those agents which immobilized sperm.

1·5 ml. of eggs was placed in a 5 ml. beaker in a water bath at the desired experimental temperature, and agitated by a small glass-covered stirrer for about 3 sec. before and after addition of sperm.

At the desired time intervals after addition of sperm, samples of the sperm-egg mixture were decanted into 30 ml. of spermicide, pre-cooled to the temperature of further development, which had been kept gently agitated in a 250 ml. beaker. In all experiments zero time was taken as the time of adding sperm to the eggs, and duration of fertilization as the time up to decanting into spermicide. After addition of the sample to the spermicide the mixture was rapidly transferred to a boiling tube, except in the case of 30-A.S.W. where the sea water was first rendered isosmotic after 30 sec. by the addition of 10 ml. of hypertonic (3 x isotonic) sea water. The tubes were placed in a water bath for 10 min. at 15° C. (E. esculentus), or for 10 min. at 10° C. (P. miliaris), and the eggs were left to develop. It should be noted that with most of the agents some sperm recover on return to normal sea water, and to avoid errors it is necessary for further development to take place in the inactivating fluid.

At the end of the development period a sample of eggs was withdrawn from the bottom of the boiling tube with a pipette of mouth bore 1 mm. diameter and placed on a 3 in. x 1 in. microscope slide. A in. square microscope slide, supported at each comer by a globule of Cow Gum (P. B. Cow (Li-Lo), Slough, Bucks, England), was lowered over the sample and pressed down until the eggs were lightly gripped. The slide was then placed in 25 % acetic acid, 75 % absolute alcohol for 24 hr. To avoid fixation artifacts it was necessary to use fresh fixative for each experiment.

After fixation the slides were placed in 0·5 % orcein in 45 % aqueous acetic acid for 2-3 hr., and then the staining solution was replaced with 10% aqueous acetic acid and the edges of the coverslip were sealed with molten paraffin wax. Preparations could be kept for several weeks without deterioration in a sealed chamber over 10% acetic acid.

Plate 1 shows eggs stained in this manner. Pronuclei were counted with a x 40 objective using light-ground phase-contrast optics which enabled visible expansion of the pronucleus, penetration below the surface of the egg, and the presence of an associated sperm aster to be used as distinguishing criteria. Routinely, 25 eggs in a sample were counted sequentially on the slide, starting at a randomly selected point; in some experiments 100 eggs from each sample were counted.

In preliminary experiments the cleavage method of assessing fertilization was compared with counts of male pronuclei. After withdrawing a sample of eggs for microscopy the spermicide was decanted off the remainder, which was placed in a beaker in an excess of filtered natural sea water in which further development took place at 17 ° C. An assessment was made of fertilization by the morphology of the first cleavage division. Text-fig. 1 shows the results of a comparison between cleavage counts and pronuclear counts for total fertilization and polyspermic fertilization. A linear correlation is apparent for total fertilization, though the scatter is wider than ideally expected. Where more than 20% of the eggs are polyspermic, pronuclear counts consistently give a higher score than does cleavage. With nicotine pre-treatment a high degree of polyspermy can be attained in all eggs (Clark, 1936), but at the time when control eggs are undergoing first cleavage, such polyspermic eggs are indistinguishable from unfertilized eggs except for the presence of a fertilization membrane (Pl. 1 D, E). Subsequently these eggs become irregular and divide into multiple cytoplasmic fragments when control eggs are at the 8 to 16-cell stage. A few of these eggs develop into swimming gastrulae of abnormal appearance. It was found to be impossible to score the degree of polyspermy in such eggs with accuracy by the cleavage method and the impression was gained that this difficulty was present in any egg with more than four male pronuclei.

If a large deviation from a Poisson distribution is expected, the latter method is probably a better way of estimating m from the sample.

Distribution functions may be calculated either taking the unfertilized eggs as the initial condition, in which case a deviation from the expected Poisson distribution will represent a departure from the initial fertilization rate for re-fertilization; or the unfertilized eggs in the sample may be discounted, and the monospermic eggs taken as the initial condition for the calculation of a distribution, deviation from which may represent a change in the rate of re-fertilization with successively more sperm entering the egg. It should be noted that in this latter case the theoretical distribution will only be approximate, since in practice the monospermic eggs to be re-fertilized do not appear simultaneously, and hence not all eggs have been susceptible to polyspermy for the same length of time.

The degree of fertilization can be expressed either as the percentage of eggs fertilized or as the mean number of male pronuclei (sperm) per egg, for any sample. The former gives a measure of the rate of fertilization, while the latter provides extra information about the degree of polyspermy. Results obtained with different spermicides on the same batch of eggs of E. esculentus are shown in Text-fig. 2 A, B. It can be seen that the curves fall into three types: (1) 30-A.S.W. which gives a high initial rate of fertilization with a large incidence of polyspermy in the period from 15 to 40 sec., (2) uranyl nitrate which shows no apparent fertilization until after 15 sec., but at long time intervals shows the same degree of fertilization as the other agents; (3) KC1, EDTA, and lauryl sulphate, all of which are essentially similar in giving an apparent increase in rate during the first 15 sec., with the rate subsequently declining. Essentially similar results were obtained with P. miliaris, a marked lag in fertilization being apparent with UNO₃, although in this species the curve obtained with 30-A.S.W. more closely approached those of the other spermicides in the KC1 group (Text-fig. 3).

It was evident from the preliminary experiments on efficacy of kill (see Methods section) that at high sperm densities 30-A.S.W. was not able to produce complete inactivation of sperm at the instant of application. It may therefore be suspected that this is one factor contributing to the apparently higher rate of fertilization at short time intervals shown by 30-A.S.W. in comparison with the other agents. A second effect of 30-A.S.W. is the promotion of polyspermy. This phenomenon is evident in Text-fig. 2B. With E. esculentus comparison of the sperm distribution in egg samples fertilized for the same time intervals at the same sperm density showed that with 30-A.S.W., especially at short time intervals, there was always present a greater number of polyspermic eggs than with the other agents (Table 1). At high sperm densities it was often found that samples where the sperm were inactivated in the time intervals from 5 ‒ 40 sec. showed a greater degree of polyspermy than did samples from the same experiment but at later times. With eggs pre-treated with nicotine, which promotes a high degree of polyspermy (Hertwig & Hertwig, 1887; Clark, 1936; Rothschild & Swann, 1950; Baker & Presley, 1966), this polyspermy-promoting effect of 30-A.S.W. at high sperm density is very marked (Text-fig. 4).

Such an effect was not found where the other agents were employed to inactivate the sperm. In the case of P. miliaris, the polyspermy-promoting action of 30-A.S.W. was much less marked, a significant excess of polyspermy at short time intervals over long time intervals never being obtained, although the initial rate of fertilization as determined with 30-A.S.W. was always apparently higher than with the other agents.

In both P. miliaris and E. esculentus inactivation of sperm with uranyl nitratecontaining sea water produces rate curves which differ from those obtained with the other agents in that there is a marked lag before fertilization becomes apparent. A similar difference between uranyl nitrate and the other spermicides is seen in the case of eggs treated with nicotine prior to fertilization (Fig. 3B). After nicotine treatment a marked degree of polyspermy is attained with all agents at time intervals greater than 40 sec., but at intervals up to 15 sec. UNO₃-inactivation gives unfertilized eggs, while the other agents yield considerable polyspermy.

The nature of the difference between the UNO₃ rate-curves and the others was investigated in more detail in a comparison between KC1 and UNO₃ on normal eggs. The separation between the curves is temperature-sensitive (Text-fig. 5), temperature having a somewhat greater effect on the UNO₃ curve (Q_10-20 = 5·2 ) than on the KC1 curve (Q_10-20 = 3·6)-The KC1 in routine use differs from UNO₃ in pH (KC1, pH = 8·0 ; UNO₃, pH = 6·3) as well as in ionic composition. When K-rich A.S.W pH 6·3, was compared as an inactivating agent with the same solution containing UNO₃, the separation between the rate curves was still present, and the KC1 curve was very similar to that obtained with the K-rich A.S.W. in routine use. The difference between the curves was abolished when uranyl nitrate A.S.W. to which had been added 5 MM EDTA without change of pH was employed as the inactivating agent. Thus it may be concluded that the uranyl ions are responsible for the differences between the curves. Eggs which have been fertilized for 15 sec. and then inactivated with K-rich A.S.W. containing uranyl nitrate and allowed to develop in that agent show no evidence of fertilization; however, if after short periods of inactivation in that agent the eggs are transferred to uranyl-free K-rich A.S.W., a number of eggs develop male pronuclei (Baker & Presley, 1969). This suggests that K-rich A.S.W. alone is unable to maintain the block on pronuclear development produced by UN0₃ and indicates that the two agents act at different stages in the process, uranyl ions acting later than potassium.

With KC1, EDTA, and lauryl sulphate as the inactivating agents a rate curve of sigmoid shape is obtained, indicating a lag in fertilization rate in the initial 5 sec. This lag is followed by a rapid increase in fertilization rate and then a progressive decline at time intervals when most of the eggs are fertilized. In the case of E. esculentus, the curve as determined with EDTA often showed a slightly lower degree of fertilization than did KC1 and lauryl sulphate at the same time intervals ; this was not the case with P. miliaris, and the difference may not be significant as the microscopic appearance of the E. esculentus eggs showed surface damage with EDTA, and errors of scoring may therefore be present. It was concluded that KC1 and lauryl sulphate were the most satisfactory agents for determinations of fertilization rate, as they acted rapidly but did not produce a spurious polyspermy at short time intervals. However, the sigmoid shape of the curves makes impossible accurate calculation of the initial rate of fertilization for comparison with re-fertilization, after the manner of Rothschild & Swann (1952).

One possible explanation of the initial lag retardation of the sperm penetration by the jelly coat. The jelly coat of samples of eggs was removed by gentle shaking in A.S.W., pH 5 · 3 (Hagstrom, 1956), until the eggs were free of jelly as judged by microscopic examination of the packing of a sample in a watch-glass and the absence of trapping of sperm in the jelly coat of the eggs in this sample on test fertilization. A comparison of fertilization rates of such jelly-free eggs with those of normal eggs from the same female, using KC1 and lauryl sulphate as inactivating agents, showed that in the absence of jelly fertilization was more rapid but the sigmoid nature of the curve was still evident (Text-fig. 6 A). Similar results were obtained with eggs pre-treated with nicotine (Text-fig. 6B).

Evidence that re-fertilization of eggs takes place at a lower rate than that of initial fertilization was found from analysis of the distribution of sperm among samples of 100 eggs. Text-fig. 7 compares this distribution with that predicted for a Poisson distribution (see Methods section). It can be seen that the number of polyspermic eggs in samples having a large proportion of fertilized eggs is considerably less than that predicted by chance from the proportion of unfertilized eggs in the sample. In the case of re-fertilization of monospermic eggs the results suggest that there can be no marked decline in the rate of third and subsequent fertilizations over the second.

To exclude the possibility that during the period of fertilization there was a marked decline in the fertilizing power of the sperm, rate determinations were made on normal unfertilized eggs using either supernatant sperm which had been used to fertilize another sample of eggs for 180 sec. previously or sperm which had been exposed for 180 sec. to egg-free A.S.W. in which unfertilized eggs had been shaken. Fertilization rates so obtained were as fast as those obtained with fresh sperm at comparable densities.

It is of interest to compare the time-relationships of the fertilization rate curves with the initiation and completion of fertilization-membrane production, because the elevation of the fertilization membrane is generally thought to provide a complete block to further fertilization.

In the course of our experiments with nicotine it became clear that nicotine can promote polyspermy if applied either before or just after fertilization (Baker & Presley, 1966). The final loss of sensitivity to nicotine seems to correlate well with the time of fertilization-membrane elevation and provides a simple method of estimating the time of appearance of fertilization membranes in a sample of eggs. The duration of nicotine sensitivity in a population of eggs in relation to the cortical response to fertilization was investigated by fertilizing samples of eggs under identical conditions and adding nicotine at various time intervals after fertilization to give a final concentration of 12· 5 MM, mixing being obtained by hand agitation. In each case fertilization proceded for a total of 180 sec., except that where nicotine was added at 180 sec. inactivation was at 240 sec. ; in the subsequent scoring (after KC1 inactivation) eggs showing five or more pronuclei were counted as having been in a nicotine-sensitive state at the time of addition of nicotine, such eggs not being present in samples where nicotine-free A.S.W. was added with agitation. The time course of the cortical response was studied in normal eggs at the same sperm density and temperature by the formaldehyde fixation method of Allen & Griffin (1958). Text-fig. 8 shows the time relationship of the fertilization curves as determined with KC1, UNO₃, the cortical responses, and the duration of the nicotine-sensitive stage.

The choice of spermicidal agents is restricted by the need for the eggs to develop after treatment, while the sperm must be prevented from effecting further fertilization. It is not known at what stage in its interaction with the egg a sperm becomes insensitive to a particular spermicide, and it is quite possible that after transfer to a spermicide there is a reservoir of sperm close to the egg which can continue to effect fertilization for some time. Any error resulting from this will show a greater degree of fertilization at a given time point than is actually the case, and so a tendency for one agent to give a lower fertilization rate than another may signify its greater efficacy.

Errors in the method of scoring fertilization by cleavage count may arise from parthenogenetic activation of unfertilized eggs, asynchrony of cleavage in the population, or where polyspermic eggs divide normally or fail to cleave. In the case of the microscopic pronuclear count there may be a tendency to err on the low side in scoring, as sperm which have entered an egg but lagged in subsequent development may not be counted when the criteria of development are strictly applied. Thus when this method indicates a high degree of polyspermy, great significance may be attached to the finding.

All our results have indicated that the rate curves fall into three groups: that as determined with 30-A.S.W. ; those with KC1, lauryl sulphate and EDTA; and that with UNO₃. It is clear that the results obtained with 30-A.S.W. may include a systematic error due to the lack of immediate efficacy, and also the tendency to promote polyspermy at short time intervals where high sperm densities are employed. KC1, lauryl sulphate and EDTA all act at an earlier stage than UNO₃, and all agents act at an earlier stage than the visible cortical response and the disappearance of the nicotinesensitive stage. We have presented evidence elsewhere (Baker & Presley, 1969) that the difference between the UNO₃ curves and the others is due to UNO₃ blocking fertilization at a different stage from the other agents.

With KC1, lauryl sulphate and EDTA our finding of curves of a sigmoid shape both in the presence and absence of jelly are in agreement with those of Hagstrom (1956), but not with those of Rothschild & Swann (1952), whose results were obtained with dilute sea water as an inactivating agent. An initial ‘shock’ causing a lag in sperm motility on change of environment might contribute to the shape of the rate curve in the early stages, as might the necessity for the successful sperm to pass through an intermediate stage of penetration during a time interval of the order of seconds before ceasing to be susceptible to the spermicide. It is clear, however, that the removal of the bulk of jelly coat does not completely abolish this lag. The presence of a similar lag in the fertilization of nicotine-treated eggs suggests that the mechanism of sperm entry in this case is similar to that in normal fertilization.

Rothschild & Swann (1952) used rate curves of similar shape to those obtained by us with 30-A.S.W. to compare the initial rate of fertilization with that of re-fertilization in P. miliaris. They concluded that, within 3 sec. of fertilization, a rapid change was propagated round the egg surface, reducing the fertilization rate to ca. 1/20 of the initial rate. Our results with 30-A.S.W. suggest that a tendency to promote polyspermy, coupled with a lack of immediate efficacy, make for a considerable error in the analysis of rates determined with this agent in E. esculentus, though less so in P. miliaris. Despite this, in both species, our findings on the analysis of sperm distribution are in general agreement with those of Rothschild & Swann that re-fertilization takes place at a lower rate than initial fertilization. Despite the approximations in the calculation of probability, it is also very clear that in E. esculentus the entry of subsequent sperm does not markedly reduce the re-fertilization rate below that produced by the first fertilizing sperm, a finding which is in keeping with the concept of an ‘all or none’ fast partial block to polyspermy produced by the normal egg in response to the first fertilization, and which also suggests that no marked decline in the fertilizing power of the sperm suspension takes place during the time in which the eggs remain susceptible to sperm entry.

It should be noted that these experiments do not establish at what time point in the interaction between sperm and egg a ‘fast blocked’ condition of the egg may arise; if it is possibe for UNO₃ to block sperm entry as much as 30 sec. after an egg would have appeared fertilized as determined by KC1, the possibility cannot be excluded that the mode of exclusion of supernumerary spermatozoa by the egg involves action at some similarly late point in sperm entry.

During this work R.P. was in receipt of a grant for technical assistance from the Medical Research Council.