Fertilization potential and electrical properties of the Xenopus laevis egg

DEVELOPMENTAL

BIOLOGY

107, 395-406 (1985)

Fertilization Potential and Electrical Properties of the Xenopus laevis Egg DENNIS Zoology Received

Department, April

J. WEBB’

AND RICHARD

University

of Cal@rnia,

20, 1984; accepted

in revised

NUCCITELLI Davis, fwm

Califor?zia

September

95616 7, 1984

The membrane potential of Xenopus eggs was monitored continuously from prior to fertilization until early cleavage. A rapid decay of the initial potential of -33.1 ? 8.1 (SD) mV (N = 14) upon impalement to a value of -19.3 k 4.2 (SD) mV (N = 68) suggested that insertion of the first electrode caused depolarization. Outward and inward rectification were observed when the resting potential was made more positive than about 5 mV or more negative than about -30 mV. Eggs were not activated by this level of current injection. Fertilization and activation evoked a membrane depolarization which was influenced by the external Cl- concentration, the nature of the halide species, and 4,4-diisothiocyanostilbene-2,2-disulfonic acid. Smaller transient depolarizations were associated with the initial stages of the fertilization potential but not with activation. Only when the fertilization potential was significantly diminished, as in high external Cl- or in the presence of Br- or I- solutions did polyspermy ensue. The input resistance of the unfertilized egg was 13.2 + 9.8 MB (N = 26) and decreased about 200-fold at the peak of the fertilization potential to 0.0’77 + 0.020 MD (N = 9). Ninety minutes after the onset of the fertilization potential and about 6 min after the start of furrow formation the membrane began a series of cleavage cycle-associated hyperpolarizations. These were unaffected by either the external Cll concentration or other halide species. Reduction in amplitude of the fertilization potential had no apparent effect upon the normal elevation of the fertilization envelope or upon cleavage and later development. The fast electrical block to polyspermy appears to have a lower threshold in Xenopus compared with other species and is also effective at negative membrane potentials. o 19% Academic

Pruss, Inc.

INTRODUCTION

Fertilization in many species is associated with a change in the membrane potential referred to as the fertilization potential (Hagiwara and Jaffe, 1979). The fertilization potential has been implicated in the prevention of polyspermy in certain marine species including the sea urchin (Jaffe, 1976), echiuroid worm (Gould-Somero et al, 1979), and starfish (Miyazaki and Hirai, 1979), as well as in certain freshwater species such as anuran amphibians (Cross and Elinson, 1980; Grey et a,!, 1982; Charbonneau et al, 1983). In all these species the fertilization potential is a depolarization of the membrane to a less negative and usually positive potential due to an increased conductance to specific ions. In the marine species it results predominantly from an influx of the cations Na+ and Ca2+ while in the freshwater species (anurans), it is apparently achieved by an efflux of Cl-. The eggs of amphibians so far studied possess a relatively less negative membrane potential (Maeno, 1959; De Laat and Bluemink, 1974; Cross and Elinson, 1980; Iwao et al, 1981; Kado et al, 1981; Webb and Nuccitelli, 1981b; Grey et aL, 1982; Charbonneau et &, 1983). This may reflect the true potential, but it is

more likely an underestimate due to the depolarizing effects of such factors as impalement-induced damage and the failure to take into account the electrode tip potential in the low-ionic-strength solutions often used. In addition, even though a high membrane resistance is often measured in these eggs, this will still be less than the true value due to the injection of current pulses too large in amplitude or too short in duration. Most of the recent electrical measurements in developing anurans have been obtained from eggs of Rana. Eggs of Xenopus laevis have somewhat different electrical properties. The popularity of Xenopue as a model system for the study of developmental questions makes it important that these properties are carefully determined. We report here on the changes in membrane potential and resistance during fertilization and activation of eggs of the frog, Xenopus and the nature of the ionic conductances involved. Some of the difficulties encountered when making microelectrode measurements on such eggs are also discussed. A preliminary account of some of this work has appeared in abstract form (Webb and Nuccitelli, 1981a, 1982a). MATERIALS

Animals ’ Present address: Laboratoire de Cytologie Experimentale, Universite de Rennes I, Campus de Beaulieu, 35042 Rennes Cedex, France.

AND METHODS

and Gametes

Xenopus females were primed with 100 IU of human chorionic gonadotropin (hCG) (Sigma Chemical Co., 395

0012-1606/85 $3.00 Copyright All rights

0 1985 by Academic Press, Inc. of reproduction in any form reserved.

396

DEVELOPMENTAL BIOLOGY

St. Louis, MO.), injected subcutaneously 24 to 72 hr before a booster dose of 800 IU. The frogs were kept at 20°C or below overnight and eggs could be expressed between 9 and 20 hr after the booster injection. The onset time and duration of egg production could be increased by maintaining the animals below room temperature. Eggs were squeezed out into a dry petri dish and then six of them were carefully arranged on the outer surface of a clean, dry, plastic petri dish lid which served as the experimental chamber. Jelly-coated eggs stick firmly to the plastic surface and were flooded a few minutes later with standard Fl solution. Mature males were anesthetized on ice before being decapitated and pithed. The excised testes were blotted dry and kept in high-ionic-strength solution (3x Fl) at 4°C. To inseminate eggs, a small piece of testes (about 2 mm3) was cut off, blotted, rinsed in a small volume of the solution being tested, blotted again, and then placed in 3 drops of the test solution. It was then macerated with scalpels and sucked repeatedly into a Pasteur pipet before being added to the eggs in the chamber. As a rule the impaled eggs were not perfused continuously in order to avoid mechanical disturbances but the bathing solution (about 3 ml) was frequently exchanged. Solutions

Unless stated otherwise fertilization and activation were carried out in standard Fl solution, modified from that of Hollinger and Corton (1980) and containing in mM: NaCl 31.25, KC1 1.75, NaHzP04 0.5, NaOH 1.9, CaClz 1.0, MgC& 0.06, tricine 10, pH adjusted to 7.8. This solution differs slightly from that used in an earlier study (Webb and Nuccitelli, 1981a,b) in that CaClz was increased from 0.25 to 1.0 mM in an attempt to improve the sealing of the egg membrane around the inserted glass electrode. Chloride-free solution was prepared by replacing chloride salts with their sulfate equivalents. This was then mixed with the standard Fl to give a series of solutions with different concentrations of chloride. Choline chloride or NaCl was added to Fl solution to increase Cl- to 60 mM. The impermeability of the membrane to sulfate was tested by comparing the results obtained in the 1 mM Cl- solution using either sulfate or methanesulfonate as the substitute. F-, Br-, and I- solutions were made up by replacing the NaCl of the standard Fl with the appropriate sodium halide. DIDS (4,4-diisothiocyanostilbene-2,2-disulfonic acid, Calbiochem-Behring, San Diego, Calif. and Sigma Chemical Co.) was made up immediately before each experiment to 1 mM in cold Cl-free solution and kept

VOLUME107,1985

in the dark and at 4°C as much as possible. A23187 (Calbiochem-Behring) was prepared when required from a stock solution in ethanol (1 mg/ml) to give 1 to 2 X 10e6 M. The final maximum ethanol level in the test solution of 0.1% or less had no effect when applied alone. Electrical

Measurements

Glass micropipets were filled with a solution containing 3 M KCl, 10 mM EDTA, and 10 mM potassium citrate. In general a fresh microelectrode was used for each egg, with a low tip potential (about -5 mV) and resistance (about 10 MO) when measured in Fl solution. A Ag/AgCl pellet (W-P Instruments, New Haven, Conn.) or wire connected the microelectrodes to a twochannel Biodyne AM4 amplifier (Santa Monica, Calif.), or to an AD311J high-input impedance operational amplifier (Analog Devices, Norwood, Mass.). Electrodes were mounted on Narashige (MP-2) or Leitz micromanipulators and potentials were monitored on an oscilloscope (Tektronix Model 5111, Beaverton, Oreg.) and recorded on a four-channel chart recorder with a 0.25~set full-scale response time (Multicorder MC 6601, Watanabe Instruments, Irvine, Calif.). In order to monitor membrane resistance, current was injected through a second electrode using an anapulse stimulator (W-P Instruments). The series resistance of the Fl solution/agar bridge and Ag/AgCl wire or pellet used to ground the chamber solution was less than 4 kQ. Impalement of the eggs was observed with a Wild stereomicroscope while elevation of the fertilization envelope could be followed with an inverted Zeiss microscope. Eggs were subjected to a number of rinses with the test solution over a period of at least 10 min prior to impalement. Three or four eggs were impaled during a typical experiment. Electrodes were generally left in place until the early blastula stage and the embryos were often retained until the swimming tadpole stage. Values are stated as the mean + SD (number of measurements). Significant differences were assessed by Student’s t test. RESULTS

The Membrane

Potential

of the Mature

Oocyte

In order to illustrate some of the problems involved in measuring the membrane potential accurately, mature Xen,opu.s oocytes were impaled with three different electrodes at the same time (Fig. 1). The first egg activated as the third electrode was inserted (arrow 3). Following withdrawal of the electrodes from this egg, two more eggs were impaled by all three electrodes

WEBB

AND

NUCCITELLI

Fertilization

397

in %nops

58 intervals . . . . . . .. . . .. . . . . .

2omh

sperm

Potential

..

sperm

FIG. 1. The potentials recorded by three microelectrodes as they approach and impale the same egg prior to fertilization. During the time period shown here three different eggs were impaled with the same three electrodes consecutively. Arrows: (1) Penetration of jelly layers and vitelline envelope of the first egg by all three electrodes. (2) First egg impaled by two electrodes (bottom two traces). (3) First egg “prick activates” upon insertion of third (top trace) electrode. (4) Penetration of jelly layers and vitelline envelope of the second egg. (5) Second egg impaled by all three electrodes. (6) Fertilization of second egg about 3 min after insemination. (7) Electrodes withdrawn from second egg. (8) Penetration of jelly layers and vitelline envelope of third egg. (9) Third egg impaled by all three electrodes. (10) Fertilization of third egg. Note faster time scale. Recording at this faster chart speed revealed a series of transient depolarizations or steps at the onset of the fertilization potential. The gap in the traces after arrow 7 indicates that a fresh batch of eggs were placed in the chamber. The time scale was reduced again at the last 5-set interval.

and subsequently fertilized. Quite large potential shifts associated with penetration of the egg investment layers preceded act&l impalement and confirmed previous reports (Palmer and Slack, 1970; Ito, 1972; Cross and Elinson, 1980). These may result from a decrease in the electrode tip potential following slight breakage or unclogging of the tip, or due to the presence of a different ionic strength solution in the perivitelline space (Maeno, 1959). These shifts, essentially of the zero potential, will markedly affect the measured membrane potential if zero potential is taken as that with the tip in the solution surrounding the eggs. The membrane potentials of the Xenm egg recorded by the three electrodes were in close agreement only if zero potential was taken as that potential at which the tip was considered to be up against the plasma membrane. This zero potential was similar to that with the tip in the solution surrounding the eggs only if electrodes with small tip potentials and resistances were used. Large shifts in the zero potential upon electrode withdrawal were eliminated by penetration of the investment layers of another egg (Fig. 1, arrows 4 and 8). Drifts in the membrane potential recorded by one particular electrode could also be attributed to changes in the tip potential as seen, for example, in

the small hyperpolarization of the third egg just after impalement (arrow 9, top trace). In spite of this, the subsequent fertilization potential, including the small transient depolarizations (arrow lo), had a similar amplitude and shape to that measured by the other two electrodes. This also provides evidence for the isopotentiality of the Xenow egg membrane since the electrodes were widely separated in the animal hemisphere. Impalement of an egg with a single electrode, whether using negative capacitance overtuning or tapping of the electrode holder, elicited a membrane potential often accompanied by an overshoot which decayed to a relatively more stable but less negative level (Fig. 2B). The mean value of the overshoot potential in the unfertilized egg was -33.1 f 8.1 mV (14) while the stable membrane potential was -23.8 f 7.7 mV (41). Insertion of a second electrode into the same egg rarely elicited an overshoot, the membrane potential very quickly reaching a level close to that recorded by the first electrode. The mean value of the membrane potential with two electrodes inserted was -19.3 f 4.2 mV (68) in eggs subsequently fertilized. Upon entry of the second electrode the potential recorded by the first usually decreased slightly (Figs. 2A, B) but would

398

DEVELOPMENTAL

B -a

20min

fy+-t

BIOLOGY

OmV.-

-/--

-OmV

5MS2 IOmVL 0.5s ‘-/

---~-

”

,.

7OMa FIG. 2. Membrane potential of the egg upon impalement. (A) Two examples of spontaneous withdrawal of the electrode and repenetration of the same egg gave rise to an essentially unchanged membrane potential. (B) Two examples, from the oscilloscope, of an egg impaled with one electrode (bottom trace) followed by a second electrode (top trace). (C) Slow recovery of the membrane potential and resistance following insertion of a second electrode (arrow) into an already impaled egg. Dots indicate peaks of membrane potential excursions after injection of 0.5-nA hyperpolarizing current for 10 set each minute.

VOLUME

107, 1985

In Fig. 3 the voltage-current relationship is shown for small (Fig. 3A) and large (Figs. 3B, C) current pulses. The diameter, stage of maturation, and membrane potential were very similar and therefore, the mean steady-state voltage changes obtained from a number of eggs have been plotted against current in Fig. 3E. The mean potential of the unfertilized eggs was -20 mV and taken as zero current with inward hyperpolarizing current to the left. The S-shaped curve resembles that obtained in mature starfish eggs (Miyazaki, 1979). The curve is linear when small current pulses are injected but changes at potentials more positive than about +5 mV and more negative than about -30 mV. This suggests the presence of both outward and inward rectification as previously found in the starfish egg (Miyazaki, 1979). Calculation of high membrane resistances by current injection is not reliable with current pulses greater than 1 nA (Kado et aZ., 1981). This is because the nonlinear nature of the voltage-current relationship (Fig. 3E) will have a drastic effect on the large voltage excursions elicited (Fig. 3B). Furthermore, the long time constant of the Xenopus egg membrane often requires a pulse duration of 10 set or more. On a number of occasions when large E

601 V mV

recover almost completely although this could take some time (Fig. 2C). These observations suggest that the first impalement results in a significant leakage conductance. The second and subsequent electrodes have much less effect on the now depolarized membrane. The true membrane potential is probably closer to the overshoot peak recorded upon initial impalement. In some earlier measurements of the membrane potential (date 10/N-3182) single and double impalements of the unfertilized egg gave values of -23.8 f 7.7 mV (41) and -20.7 + 7.8 mV (17), respectively. These values were obtained when the potential had reasonably stabilized. The Membrane Resistance and Capacitance of the Mature Oocyte Measurement of the resistance of the Xenm egg membrane is fraught with similar sources of error as measurement of the potential. Inadequate sealing of the membrane around the electrode will give rise to a leakage conductance that will serve to diminish the membrane resistance directly and perhaps indirectly by activating voltage or ion sensitive channels.

btcl FIG. 3. Voltage-current relationship of egg showing membrane potential responses to injection of: (A) small current pulses; (B) large current pulses; (C) large current pulses with hyperpolarizing pulses eventually leading to an abrupt depolarization; (D) large current pulses injected at the peak fertilization potential; (E) voltagecurrent relationship. Mean of steady-state voltage excursions + SD. Up to 19 eggs used for a particular pulse. Pulse duration: (A, C) 10 set; (B, D) 5 sec. Vertical bar: Current (A) 2.3 nA; (B-D) 23 nA. Voltage (A) 20 mV; (B, C) 50 mV; (D) 5 mV. Measurements were made during July and August 1982.

WEBB

AND

NUCCITELLI

Fertilization

hyperpolarizing current pulses were injected the membrane voltage response would diminish abruptly. An example is shown in Fig. 3C. At the end of the current pulse the membrane was depolarized by a few millivolts but recovered, albeit slowly. Such occurrences never gave rise to egg activation and may be attributable to an increased Cl- conductance activated by Cl- derived from the KCl-filled electrode during current injection (Chenoy-Marchais, 1982; Coombes et al, 1955). This could also account for inward rectification at potentials more negative than -30 mV. The slope of the linear portion of Fig. 3E for small current amplitudes gives a value for the input resistance of about 10 MQ. The mean calculated from injection of hyperpolarizing current pulses of less than 1 nA was 13.2 + 9.8 MS2 (26) (Table 1). This is somewhat larger than the value of 2.47 MQ for Xenopus eggs matured in vitro (Kado et al, 1981) and about twice that of Rana pipiens (Cross and Elinson, 1980). However, this is certainly less than the true value for the unimpaled egg due to a leakage component around the electrodes. Assuming the Xenow oocyte to be a sphere with a diameter of 1.3 mm and a surface area of 0.053 cm2, the specific resistance of the unfertilized egg membrane is 0.70 MO -cm2 when the input resistance is 13.2 MB. This specific resistance must be taken as a minimum value not only for the reasons already mentioned concerning impalement-induced changes, but because surface irregularities such as microvilli will increase the effective surface area (De Laat and Bluemink, 1974). The membrane capacitance was estimated from the membrane time constant of seven eggs. The mean

ELECTRICAL

PROPERTIES

Date

-19.3 -17.7

6/82-lo/82

potential

6182-10182 measurements

to

FERTILIZATION

made

in standard

ACTIVATION

resistance

(MQ

After

+ 4.1 f 3.9

(68) (29)

0.077 f 0.02 (9)

(min

Activation

repolarization

-19.6 -16.9

+ 2

6.1 (68) 4.8 (29)

+ SD (N)) 20.7 + 18.0 (9)

_+ SD (N)) to

Duration of depolarization

4.8 + 4.2 (57) Fl solution.

AND

f SD (N)) +3.0 +0.55

Insemination fertilization

29.3 + 17.8 (64) were

(mV

f 4.2 (68) 2 6.6 (30)

Timing

a All

Fertilization and activation are accompanied by a depolarization of the membrane and a massive drop in the resistance (Figs. 4 and 5; Table 1). The initial phase of the fertilization potential in Xenopua is comprised of a number of small transient depolarizations (spikes) before the potential eventually reaches the peak (Fig. 1, arrow 10 and Fig. 4, top trace). This may take up to 20 sec. The spikes varied in number and size from egg to egg but were always less than the peak fertilization potential. Although detectable in nearly every case of fertilization, such eggs were always found to be monospermic by the criteria of a single observable sperm entry site or normal cleavage divisions. The peak fertilization potential from 68 eggs was +3.0 + 4.1 mV (Table 1). This potential is within the linear portion of the voltage-current relationship even if the potential excursions following current injection are taken into account (Fig. 3E). Therefore, the membrane resistance prior to and during fertilization can be compared even though the potentials are different. A dramatic 200-fold decrease in resistance to 0.077 + 0.02 MO (9) was measured during fertilization (Fig. 5A). This is almost 10 times the decrease observed with R. p@iens (Cross and Elinson, 1980). Upon repo-

13.2 f 9.8 (26)

Impalement fertilization

and Resistance

Peak depolarization

Membrane Fertilization

The Membrane Potential Changes at Fertilization

Unfertilized

6/82-lo/82 10/81-3182

399

in Xenopus

value of 0.045 & 0.007 (SD) PF gave a specific capacitance of 0.85 pF/cm2 using a surface area of 0.053 cm2. This is somewhat smaller than the corrected value of 1.44 pF/cm2 found for Xenopus eggs matured in vitro (Kado et aL, 1981).

TABLE 1 EGG DURING

OF Xenqpus

Membrane Fertilization Activation

Potential

includes

15.2 f 3.2 (55) ionophore

and spontaneous.

Duration positive 5.3 -t 3.5 (52)

400

DEVELOPMENTAL

BIOLOGY

VOLUME

lo?,1985

-3oL t A23107 FIG. 4. Fertilization after addition. Bar:

potential (top trace) and activation 20 min until arrows, then 20 set (top

potential (bottom trace) at fast chart trace) or 2 min (bottom trace).

larization the membrane potential returned to about the same value as in the unfertilized egg (Table 1). The resistance was now 20.7 + 18.0 MQ (9) and not significantly different to that prior to fertilization, although much larger than the 0.85 MQ reported by De Laat and Bluemink (1974) in precleavage Xenopus eggs. Swelling of the egg jelly during hydration increases the distance that the sperm have to travel. This may partly account for the difficulty of fertilizing amphibian

speed.

A23187

was rinsed

off about

10 min

eggs that have been exposed for some time to lowionic-stength solutions (Cross and Elinson, 1980). In the present study eggs were usually left in the solution being tested for at least 10 min prior to impalement. Fertilization was possible in Fl solution for at least 30 min after impalement (Table 1). The mean interval between insemination and fertilization was 4.8 + 4.2 min (57). The duration of the fertilization potential was 15.2 f 3.2 min (55). It remained more positive than zero

lOOKa

cleavage 3rd

2nd

4th

5th

30nA 1

B 10r

20M8

50Kfi

6M!J

0.5nA

FIG. 5. Membrane potential and resistance during (A) fertilization and cleavage (insemination at arrow), (B) spontaneous activation (arrow). Current pulse rate and amplitude were increased during the fertilization and activation potentials as indicated. The kQ resistances were the lowest values calculated for these two eggs at the peak depolarizations.

WEBB

AND

NUCCITELLI

Fertilization

for 5.3 f 3.2 min (52) but in about 30% of the 52 eggs, never became positive. Repolarization usually consisted of an initial slow phase followed by a shorter fast phase (Fig. 5A). Activation of eggs with A23187 evoked similar changes in membrane potential and resistance, however, there were no spikes (Fig. 4). Instead there was usually a very gradual depolarization for some seconds before a faster phase toward the peak value. The concentration of A23187 was much lower than that applied to R. @@ens (Cross, 1981) but similar to that used on R. temporaria (Charbonneau et ak, 1983). For Xenopus ionophore concentrations as low as lop7 M were effective, although the onset of activation was delayed. Spontaneous activation was sometimes encountered (Fig. 5B), usually after the membrane had depolarized to a relatively low level, and occurred in a single step. The remarkable similarity in the membrane potential and resistance changes up until first cleavage is illustrated for a fertilized and spontaneously activated Xenopus egg (Fig. 5). The membrane potential of the fertilized egg was monitored for at least 3 hr. The difference of 2.5 mV between the peak depolarization for fertilized and activated eggs (Table 1) was significant (P < 0.01). This may reflect a greater influence of a delayed hyperpolarizing conductance on activated eggs due to the slower rise time of the activation potential. We have insufficient data for determining whether the resistance at the peak potential was significantly different between fertilized and activated eggs. Certainly, in both, the peak potential is less positive than the equilibrium potential for Cl-, Eel (see later). The spikes (Fig. 4) provided a useful early indicator for distinguishing between fertilization and activation, and have not been reported in other amphibian species. Each cleavage division is accompanied by a hyperpolarization and decrease in resistance (Fig. 5A) as previously reported for both Rana and Xenopus (Woodward, 1968; De Laat et aL, 1974). In Xenoms this hyperpolarization begins about 6 min after the furrow begins to form. Activated eggs failed to cleave normally. Instead of the regular transient hyperpolarizations, the activated egg would display irregular changes of the membrane potential (Fig. 5B). The Efect of the External Chloride on the Membrane Potential

Concentration

The membrane potential of Xenopus eggs before fertilization and after repolarization was essentially unaffected by variations in [Cl$, (Fig, 6). On the other hand, the peak fertilization potential was dramatically affected and indicates that fertilization is normally

Potential

401

in Xenopus

60.I

I I

30. i: 5

4

E UJ

1

0I

-301

I

. 1 external

10 chloride conc.(mM)

, 100

OmV

1

5

10

20

J 35

60mM Cl-

FIG. 6. Upper: Membrane potential before (0), during peak (a), and after (0) fertilization versus the external Cl- concentration (plotted on a logarithmic scale). Bars indicate SD. Between 7 and 30 eggs were fertilized at each Cl- concentration. Linear regression of the points 10 to 60 m&f yields a slope of -46.3 with a correlation coefficient of 1.0. Measurements were made from April 1981 to March 1982. Lower: Examples of fertilization potentials at each Cl- concentration.

associated with a net Cl- efflux (Grey et aL, 1982). The slope of -46.3 mV for the linear portion of Fig. 6 (10 to 60 mM Cl-) is less than that expected for a purely Nernstian relationship but is similar to the slope of -48 mV reported in R. pipiens over the range 40 to 113 mM Cl- (Cross, 1981). However, in Rana the slope was less steep than in Xenopus over the lower range of Cl-. Eel is +15 mV in standard Fl solution (Cl- activity = 29.0 mM) if calculated using the intracellular Clactivity (53.5 mM) of the fertilized Xenopus egg (De Laat et aL, 1974); this activity being similar to that of

402

DEVELOPMENTAL BIOLOGY

oocytes matured in vitro (Kusano et al, 1982). This is markedly more positive than the peak potential measured at either fertilization or activation and suggests the involvement of another ion conductance change. The intracellular Na+ activity in the Xenopus egg is about 20 mM (Webb, unpublished result) which gives an ,?&a of +9.5 mV in Fl solution. However, when methanesulfonate was used as the Cl- substitute instead of sulfate (Na2S04), in a series of fertilizations in 1 mM Cl-, the peak potential of +45.2 + 3.6 mV (12) was very similar (Fig. 6). This was in spite of a twofold higher Na+ concentration in the sulfate solution. As in the case of Rana (Cross, 1981), a Na+ conductance increase appears to be ruled out during the fertilization potential of Xenopus. A more likely candidate is K+ since Ek will be about -90 mV in Fl solution, once again using a value for the intracellular K+ activity (51.6 mM) obtained in the fertilized egg (De Laat et al, 1974). In fact, an increased K+ conductance at fertilization has already been observed in Rana (Schlichter and Jaffe, 1984). Outward rectification would be expected to diminish the fertilization potential amplitude as the membrane potential passed +5 mV (Fig. 3E). This should occur around 20 mM [Cl-l, according to Fig. 6, yet the slope remains remarkably linear down to 10 mil4 Cl-. Repolarization was associated with a shoulder when the Cl- level was less than 20 mM (Fig. 6). This potential also increased with decreasing [Cl-]0 but at a lower rate than the peak potential. There was no noticeable effect of [Cl-l, upon the hyperpolarizations accompanying each cleavage. The Eflect of Other Halides Fertilization Potential

on the

The major effect of substituting Cl- with the other halides is seen during the Xenopus fertilization potential. The membrane potential before and after was essentially unaffected (Figs. 7 and 8). Fluoride does cause a slight depolarization but this could be due to its action as a metabolic inhibitor. With increasing radius of the halide ion the amplitude of the fertilization potential was reduced, and, in the case of I-, reversed direction (Fig. 8). Fertilization in Br- solution hardly evoked any potential change and only by monitoring the resistance was a conductance increase apparent. The increased conductance during fertilization of Xenms eggs is therefore not selective to just Cland is probably not restricted to the halide ions (Ito, 1972). The permeability of the egg membrane during fertilization increased for the halides in the sequence F < Cl < Br < I. A similar sequence for just Cl-, and Br- and I- was seen before with Xenopus (Grey et al.,

VOLUME 107, 1985

30 2:

F 20min

M ,.

&-I

-20 IJ

S Or

..

Cl

h 1Or $0 UF

. ._. ..1... ..,,...... u

.“.,

- . ....._...................~,,,,,~,, ........A

Br

FIG. 7. Membrane potential and resistance from fertilization to cleavage in solutions containing 35 mikf of the sodium salt of either fluoride, chloride, bromide, or iodide. Each solution also contained 1 mM calcium chloride and the other components of Fl solution except the sodium. Dots indicate peaks of membrane potential excursions following injection of a 5-nA hyperpolarizing current pulse for 2 set each minute.

1982) and Rana eggs (Cross and Elinson, 1980). Bromide and I- enter the egg with the least difficulty, thereby counteracting the depolarizing effect of any Cl- efflux. The hydrated ion will be the permeant form and as surface charge density decreases with increase in the ionic radius, so the radius of the hydrated ion will decrease (Wright and Diamond, 1977; Conway, 1981). Eggs fertilized in solutions containing Br- and Iwere almost always polyspermic, displaying multiple sperm entry sites and multiple furrows as first cleavage was attempted. Exceptions did occur and such monospermic eggs developed into normal embryos. Thus the different halides had no detrimental effect upon early development when fertilization was monospermic. The similar hyperpolarizations at cleavage in the various halide solutions (Fig. 7), support the view that they are mainly due to the increased K+ conductance appearing with the new membrane (De Laat and Bluemink, 1974; Kline et aL, 1983). Experiments in which the fertilization potential was virtually abolished either by raising external Cl- or replacing Cl- with Br- or I-, demonstrate that the potential change itself is not only apparently unimportant for the cortical reaction in Xenopus, since fertilization envelopes elevated and eggs rotated normally

WEBB

40-

AND

NUCCITELLI

Fertilization

F I

Cl 1 Br

bl P1 -40 ’ 1.0

1.5 2.0 ionic radius A

2.5

FIG. 8. Membrane potential before (0), during peak (O), and after (Cl) fertilization or activation in different halide media versus the ionic radius of the halide. Ionic radii values from Conway (1981). Bars indicate SD. Between 9 to 30 eggs were used for each halide. Linear regression slope: -67.9. Correlation coefficient: 0.99. Measurements were made from February to November 1981.

(Grey et al, 1982), it is also unnecessary for normal development. As a further test of the anionic nature of the conductance change at activation, eggs were treated from 20 to 60 min prior to activation with the Cl-/ anion exchange inhibitor, DIDS (Knauf et aL, 1977). Fertilization was not attempted in the DIDS solution to avoid any complicating action on the sperm. DIDS (1 mM) dramatically reduces the ionophore-induced activation potential in Cl--free Fl solution (Fig. 9). The peak potential was +19.6 + 8.0 mV (26), while in an identical solution without DIDS the activation potential amplitude was +41.0 f 0.02 mV (6). The membrane potential was not significantly affected by DIDS before (-18.3 ? 6.3 mV (26)) or after (-15.4 + 6.1 mV (22)) activation. In contrast to its action on Xenm eggs, DIDS had no affect on the activation potential amplitude in eggs of R. pip&ns (Schlichter and Jaffe, 1984).

Potential

403

in Xenqus

ization potential of Xenopus is due to a transient increase in the anionic conductance of the egg membrane revealing a similar mechanism to that already demonstrated in other anuran amphibians (Maeno, 1959; Ito, 1972; Cross and Elinson, 1980; Charbonneau et al, 1983). It likewise contrasts with the cationic conductance increase observed in certain marine species at fertilization (Hagiwara and Jaffe, 1979). The halide selectivity of the anionic channel is in the order F < Cl < Br < I and suggests that the greatest permeability is to the ion, I-, having the smallest hydrated ionic radius or the lowest energy of hydration. Third, since the slope of the fertilization potential versus the logarithm of [Cl-],, is not Nernstian, then the possible involvement of another, hyperpolarizing, ionic conductance change can be inferred. Fourth, and perhaps most important, both membrane potential and resistance appear to be markedly affected by electrode insertion, particularly the first electrode. This may have brought about a membrane conductance change in addition to the leakage conductance. True Membrane

Potential

The Xenom egg membrane potential of -19 mV reported here is comparable to the value of Grey et al. (1982). It is almost double that measured during an intracellular pH study (Webb and Nuccitelli, 1981b) and now closer to En for the unfertilized egg in Fl solution. However, it may still not reflect the true value bearing in mind the potential decay upon initial impalement (Fig. 2; Chambers and de Armendi, 1979; Huelser and Schatten, 1982). The true potential is probably at least as negative as -33 mV, the potential measured immediately upon impalement. All amphibian species studied until now reveal a relatively low egg

DISCUSSION

Among the points emerging from this work are first, the high membrane resistance before and after fertilization indicating that these large eggs of Xen0pu.a possess one of the highest specific membrane resistances yet reported. Furthermore, during fertilization the resistance decreases about 200-fold. Second, the fertil-

A23187 FIG. 9. Action of DIDS in Cl--free Fl solution.

on the A23187-induced

activation

potential

404

DEVELOPMENTAL

BIOLOGY

membrane potential. Unfortunately, initial impalement is rarely shown while a wide variety of solutions have been employed by different workers. In addition, the electrode tip potential is not always mentioned and this becomes particularly important when eggs are impaled in low ionic strength solution. It is also a fallacy to assume that impalement can have had little damaging effect if insertion of a second electrode gives essentially the same membrane potential as the single electrode impalement. Most of the damage will already have been done by the first electrode (Hagiwara and Jaffe, 1979; Nuccitelli and Grey, 1984). Xenopa.s fertilization usually takes place soon after release of the gametes with some associated body fluid. The natural solution for fertilization therefore, may be between the low ionic strength freshwater and higher ionic strength body fluid. Eggs that have been in freshwater for a few minutes are certainly more difficult to fertilize. In fact, electrodes were inserted after the addition of sperm suspension to the unhydrated eggs of Rana (Cross and Elinson, 1980). However, this may not allow adequate time for recovery from impalement or for exchange of the test solution at the membrane. The solution used in the present study enabled optimum fertilization at least 30 min after hydration of the egg jelly and electrode insertion (Table 1). Fast Block to Polyspermg Throughout this study Xenopus fertilization in the standard Fl solution was monospermic, the only exceptions being cases in which the membrane potential had depolarized to a particularly low level, around -10 mV. Such eggs often activated spontaneously but fertilization could occur up to 30 min after insemination although usually not before some repolarization to around -16 mV. Even these cases were rarely polyspermic. In order to induce polyspermy the fertilization potential had to be drastically reduced in amplitude by increasing external Cl- or in the presence of Br- or I- (Grey et aL, 1982). The fast electrical block to polyspermy occurs at a positive membrane potential that varies for different marine species (Jaffe et ak, 1982). In Xenow this potential was apparently quite negative and did not have to become positive in order to prevent polyspermy. This has also been noticed in R. temporaria (Charbonneau et al., 1983), and also in the starfish egg which blocked at -5 mV (Miyazaki and Hirai, 1979). If the membrane potential of the Xenopus egg is close to the fast block potential then a depolarization of just a few millivolts, including steps, should be sufficient to prevent polyspermy. In a solution of low Cl- such as freshwater the amplitude of the fertilization potential may exceed +40 mV. This is a

VOLUME

107, 1985

long way from the fast block and resting membrane potential and indicates a more than adequate depolarization for preventing polyspermy. Moreover, if the membrane potential of the unimpaled egg is significantly more negative than the value given in Table 1, then the amplitude of the fertilization potential would be even greater. A stepwise depolarization of the kind illustrated in Fig. 1, from a level of say -70 mV would increase the chances of polyspermy. Activation

Potential

Activation of XenoF eggs using the calcium ionophore A23187 (Fig. 4) or by Ca2’ injection (Hollinger et aZ., 1979) indicates an early involvement of the divalent ion in the membrane conductance changes and the cortical reaction. This supports similar evidence obtained with Rana (Cross, 1981; Charbonneau et aL, 1983). Recent direct evidence for a wave of increased free Ca2+ following fertilization in Xenopus has been obtained with Ca2+-selective microelectrodes (Busa and Nuccitelli, 1984). There is also evidence for a Ca2+activated Cl- channel in immature oocytes. A Ca2+controlled Cl- current present in full-grown Xenopus oocytes, fell to nearly zero after maturation (Robinson, 1979). A transient Ca’+- activated conductance increase, probably Cl-, was elicited upon depolarization of voltage clamped Xenopus oocytes to potentials more positive than about -20 mV (Miledi, 1982; Barish, 1983). A transient rise in free Ca2+ has been reported at maturation and ionophore-induced activation in the Xenopus oocyte (Wasserman et aL, 1980). However, changes in internal Ca” are closely associated with internal pH changes (Meech and Thomas, 1977) which themselves may regulate ion conductance (Strickholm, 1981; Moody and Hagiwara, 1982) including Cl- channels (Hanke and Miller, 1983). The transient acidification that accompanies Xenopus fertilization (Webb and Nuccitelli, 1981b), parallels the transient rise in Ca’+, so that either ion could be involved in the initiation of the fertilization potential. At present there is no evidence to support a voltage-activated fertilization potential in amphibian eggs. Depolarizing and hyperpolarizing pulses failed to activate Xenopus eggs under current clamp conditions (Fig, 3). Activation of amphibian eggs using massive current levels either intracellularly (Charbonneau et aL, 1983) or extracellularly (Iwao et al, 1981) are most probably a result of general membrane breakdown. Although both ionophore and sperm elicited a membrane depolarization the two events are not identical (Fig. 4 and Table 1). The transient depolarizations were never observed during activation which with ionophore always began with a smooth gradual depo-

WEBB

AND

NUCCITELLI

Fertilization

larization similar to that seen in Rana (Cross, 1981; Schlichter and Elinson, 1981). A23187 application and Ca2+ injection are also able to activate oocytes of Xenw (Belanger and Schuetz, 1975; Hollinger et aL, 1979) and other anurans (Elinson, 1977) at a much earlier stage than they can be fertilized. Unfortunately, membrane potential was not monitored in these studies to assess whether a Cl- conductance increase was evoked. Nevertheless, Cl- and K+ channels do appear to be present in the immature Xenopus oocyte. Prolonged treatment with acetylcholine elicited a transient depolarization of long duration with a remarkably similar form to the fertilization potential described here. The peak potential was also less than &I. Moreover, a Kf conductance increase could also be evoked by catecholamines. The long latency of these two responses suggested to the authors the involvement of a second substance released intracellularly (Kusano et al, 1982). The sensitivity of the oocyte to both substances was lost following maturation at which time sperm are able to elicit a Cl- and possibly a K+ conductance increase. These studies suggest that although the specific membrane receptors may change during maturation, a common intermediate mechanism involved in channel activation may exist intracellularly. This idea appears even more attractive in light of the recent proposition of a single “class of multistate Clchannel coupled to either GABA or glycine receptors” following a patch clamp study of mouse neurones (Hamil et al, 1983). The authors thank Dr. L. A. Jaffe, Dr. N. L. Cross, Dr. R. D. Grey, Dr. D. Kline, Dr. R. Dohmen, and Dr. L. Schlichter for commenting on earlier drafts of the manuscript. This work was supported by NSF Grant PCM 8118174 to R.N. REFERENCES BARISH, M. E. (1983). A transient calcium-dependent chloride current in the immature Xenopus oocyte. J. PhysioL (London) 342,309-325. BELLANGER, A. M., and SCHUETZ, A. W. (1975). Precocious induction of activation responses in amphibian oocytes by divalent ionophore A23187. Dev. BioL 45, 378-381. BUSA, W. B., and NUCCITELLI, R. (1984). An elevated free cytostolic Ca2+ wave follows fertilization in eggs of the frog Xen0pu.s laevvis. J. Cell BioL In press. CHAMBERS, E. L., and DE ARMENDI, J. (1979). Membrane potential, action potential and activation potential of eggs of the sea urchin, Lytechinus variegatus. Exp. Cell Res. 122, 203-219. CHARBONNEAU, M., MOREAU, M., PICHERAL, B., VILAIN, J. P., and GUERRIER, P. (1983). Fertilization of amphibian eggs: A comparison of electrical responses between anurans and urodeles. Dev. BioL 98,304-318. CHENOY-MARCHAIS, D. (1982). A Cll conductance activated by hyperpolarization in Aplysia neurones. Nature (London) 229, 359-361. CONWAY, B. E. (1981). Studies in physical and theoretical chemistry. “Ionic Hydration in Chemistry and Biophysics,” pp. 774. Elsevier, New York.

Potential

in Xenopus

405

J. S., ECCLES, J. C., and FATT, P. (1955). The specific ionic conductances and the ionic movements across the motoneuronal membrane that produce the inhibitory post-synaptic potential. J. Physiol (London) 130,326-373. CROSS, N. L. (1981). Initiation of the activation potential by an increase in intracellular calcium in eggs of the frog, Rana p&ens. COOMBES,

Dew. BioL 85, 380-384.

N. L., and ELINSON, R. P. (1980) A fast block to polyspermy in frogs mediated by changes in the membrane potential. Dev. BioL 75, 187-198. DE LAAT, S. W., and BLUEMINK, J. G. (1974). New membrane formation during cytokinesis in normal and cytochalasin-B-treated eggs of Xenopus laevis. J. Cell BioL 60, 529-540. DE LAAT, S. W., BUWALDA, R. J. A., and HABETS, A. M. M. C. (1974). Intracellular ionic distribution, cell membrane permeability and membrane potential of the Xenopua egg during first cleavage. Exp. Cell Res. 89, 1-14. ELINSON, R. P. (1977). Fertilization of immature frog eggs: Cleavage and development following subsequent activation. J. Embryol. Exp. MorphoL 37, 187-201. GOULD-SOMERO, M., JAFFE, L. A., and HOLLAND, L. Z. (1979). Electrically mediated fast polyspermy block in eggs of the marine worm, Urechis caupa J. Cell BioL 82, 426-440. GREY, R. D., BASTIANI, M. J., WEBB, D. J., and SCHERTEL, E. R. (1982). An electrical block is required to prevent polyspermy in eggs fertilized by natural mating of Xenopus 1aevi.s. Dew. BioL 89, 475-484. HAGIWARA, S., and JAFFE, L. A. (1979). Electrical properties of egg cell membranes. Annu. Rev. Biophys. Bioeng: 8, 385-416. HAMIL, 0. P., BORMANN, J., and SAKMANN, B. (1983). Activation of multiple-conductance state chloride channels in spinal neurones by glycine and GABA. Nature (London) 305, 805-808. HANKE, W., and MILLER, C. (1983). Single chloride channels from Torpedo electroplax. Activation by protons. J. Gen PhysioL 82, 2545. HOLLINGER, T. G., and CORTON, G. L. (1980). Artificial fertilization of gametes from the South African clawed frog, Xenopus laevti. Gamete Res. 3, 45-57. HOLLINGER, T. G., DUMONT, J. N., and WALLACE, R. A. (1979). Calcium-induced dehiscence of cortical granules in Xenopus laevti oocytes. J. Exp. Zool. 210, 107-116. HUELSER, D., and SCHATTEN, G. (1982). Bioelectric responses at fertilization: Separation of the events associated with insemination from those due to the cortical reaction in sea urchin, Lytechinus CROSS,

variegatus.

Gamete

Res. 5, 363-377.

S. (1972). Effects of media of different ionic composition on the activation potential of anuran egg cells. Dev. Growth D&x 14, 217-227. IWAO, Y., ITO, S., and KATAGIRI, C. (1981). Electrical properties of toad oocytes during maturation and activation. Dev. Growth wer. 23, 89-100. JAFFE L. A. (1976). Fast block to polyspermy in sea urchin eggs is electrically mediated. Nature (London) 261, 68-70. JAFFE, L. A., GOULD-SOMERO, M., and HOLLAND, L. (1979). Ionic mechanism of the fertilization potential of the marine worm, Urechis caupo (Echiura). J. Gen. PhysioL 73, 469-492. JAFFE, L. A., GOULD-SOMERO, M., and HOLLAND, L. (1982). Studies of the mechanism of the electrical polyspermy block using voltage clamp during cross species fertilization. J. Cell BioL 92, 616-621. JAFFE, L. A., and ROBINSON,K. R. (1978). Membrane potential of the unfertilized sea urchin egg. Dev. BioL 62, 215-228. KADO, R. T., MARCHER, K., and OZON, R. (1981). Electrical membrane properties of the Xenopus Levis oocyte during progesterone-induced meiotic maturation. Dev. BioL 84, 471-476. ITO,

406

DEVELOPMENTAL BIOLOGY

KLINE, D., ROBINSON, K. R., and NUCCITELLI, R. (1983). Ion currents and membrane domains in the cleaving Xenopus egg. J. CeU BioL 97, 1753-1761. KNAUF, P. A., FUHRMAN, G. F., ROTHSTEIN, S., and ROTHSTEIN, A. (1977). The relationship between anion exchange and net anion flow across the human red blood cell membrane. J. Gen PhysioL 69, 363-386.

KUSANO, K., MILEDI, R., and STINNAKRE, J. (1982). Cholinergic and catecholaminergic receptors in the Xe-nqpus oocyte membrane. J. Physiol 328, 143-170. MAENO, T. (1959). Electrical characteristics and activation potential of BzGfo eggs. J. Gen Physiol 43, 139-157. MEECH, R. W., and THOMAS, R. C. (1977). The effect of calcium injection on the intracellular sodium and pH of snail neurones. J. Physiol (Lmdm) 265, 867-879. MILEDI, R. (1982). A calcium-dependent transient outward current in Xerwpus laeuis oocytes. Proc R. Soc Lmd Ser. B 215,491-497. MIYAZAKI, S. (1979). Fast polyspermy block and activation potential: Electrophysiological basis for their changes during oocyte maturation of a starfish. Dev. BioL 70,341-354. MIYAZAKI, S., and HIRAI, S. (1979). Fast polyspermy block and activation potential: Correlated changes during oocyte maturation of a starfish. Dew. Bid 70.32’7-340. MOODY, W. J., and HAGIWARA, S. (1982). Block of inward rectification by intracellular H+ in immature oocytes of the starfish Media&w aequalis. J. Gen Physiol 79, 115-130. NUCCITELLI, R., and GREY, R. D. (1984). Controversy over the fast, partial, temporary, block to polyspermy in sea urchins: A reevaluation. Dev. BioL 103, 1-17. PALMER, J. F., and SLACK, C. (1970). Some bio-electric parameters of early Xenqpus embryos. J. EmbryoL Exp. MwphoL 24,535-553.

VOLUME 107, 1985

ROBINSON,K. R. (1979). Electrical currents through full-grown and maturing Xenqpus oocytes. Proc. NatL Acad Sci USA 76,837~841. SCHLICHTER,L. C., and ELINSON, R. P. (1981). Electrical responses of immature and mature Rana pipiens oocytes to sperm and other activating stimuli. Dev. BioL 83,33-41. SCHLICHTER, L. C., and JAFFE, L. A. (1984). Fertilization-induced chloride and potassium conductances in eggs of the frog, Ranu pipkns. Biophya J. 45,23a. STRICKHOLM, A. (1981). Ionic permeability of K, Na and Cl in potassium-depolarized nerve. Biuphys. J. 35,677-697. WASSERMAN, W. J., PINTO, L. H., O’CONNOR, C. M., and SMITH, L. D. (1980). Progesterone induces a rapid increase in Ca2+ of Xenopus laevis oocytes. Proc. NatL Acad Sci. USA 77, 1534-1536. WEBB, D. J., and NUCCITELLI, R. (1981a). The fertilization potential of Xenopus la&s. J. Cell BioL 91, 182a. WEBB, D. J., and NUCCITELLI, R. (1981b). Direct measurement of intracellular pH changes in Xenopzcs eggs at fertilization and cleavage. J. Cell BioL 91, 562-567. WEBB, D. J., and NUCCITELLI, R. (1982a). The fertilization potential of Xenopus laevis and Ranu pipiem J. Cell BioL 95, 163a. WEBB, D. J., and NUCCITELLI, R. (1982b). Intracellular pH changes accompanying the activation of development in frog eggs: Comparison of pH microelectrodes and “P-NMR measurements. In “Intracellular pH: Its Measurement, Regulation and Utilization in Cellular Functions” (R. Nuccitelli and D. W. Deamer, eds.), pp. 293-324. Alan R. Liss, New York. WOODWARD,D. J. (1968). Electrical signs of new membrane production during cleavage of Rana pipiena eggs. J. Gen PhysioL 52,509-531. WRIGHT, E. M., and DIAMOND, J. M. (1977). Anion selectivity in biological systems. PhysioL Rev. 57, 109-156.