of Yeshiva University, New York, N.Y.
[JOURNAL OF CELLULAR PHYSIOLOGY Vol. 67, No. 1, February 1966]
Although a number of electrophysiological studies have been carried out on eggs from several species, most workers have been primarily interested in membrane changes during egg activation. The experiments described in this report were designed to compare the electropotential difference between the external environment and the cytoplasm of amphibian eggs at ovulation, activation and first cleavage. A preliminary report of this work has been presented (Morrill and Watson, '62).
Electrodes of the Ag-AgCl type were used. The indifferent electrode was connected via a KCI agar bridge to the solution bathing the egg and grounded through a calibrator. The liquid junction of the exploring electrode was a pyrex capillary tube drawn to a 0.5 to 6 p tip (5 to 30 megohms resistance) filled with 2.5 M KCI. The exploring electrode was connected to the grid of a cathode follower (CK 5886 electrometer tube) with a grid current less than 1 x 10-11 amps. All recordings were made with a direct-coupled amplifier and an Offner pen recorder. The standard sensitivity employed was 10 mV or more per centimeter of pen deflection so that potential changes as small as I mV could be measured. All experiments were carried out at room temperature (20 to 24o).
The full strength Ringer's solution contained per liter: 111 mM NaCl, 1.9 mM KCl, 1.1 mM CaCl2, 2.4 mM NaHCO, and 0.08mM NaH2PO4.
It should be noted that the egg differs from other animal cells, in that it has an outer or vitelline membrane closely applied to the plasma membrane. Continuous with the plasma membrane is a dense granular cytoplasmic layer which together with the plasma membrane make up the egg cortex. The measurements reported here represent the transcortical potential.
As shown in column two, the cortical membrane potential of the ovarian oocyte was measured relative to a salt solution isotonic with adult frog serum. After ovulation, however, the R. pipiens egg is no longer viable in isotonic Ringer's solution. Fertilization and development, which normally occur in pond water, may be carried out in the laboratory in 0.1 strength Ringer's solution. Studies in this laboratory have shown that if eggs are fertilized in 0.1 strength Ringer's solution and transferred to 0.1 strength increments of Ringer's solution, cleavage proceeds normally up to 0.5 strength Ringer's solution but is blocked or irregular in 0.6 strength or stronger Ringer's solution. In the present study the transmembrane potential of the oocyte in isotonic Ringer's solution was compared with that in ovulated unfertilized and fertilized eggs in 0.1 strength Ringer's solution.
It was possible to keep individual cells, both ovarian and ovulated, impaled with a micropipette for at least six hours with no change in the steady transmembrane potential. Usually the potentials were stable within a few seconds after entry. Isolated ovarian oocytes and ovulated unfertilized eggs could be stored'at room temperature with aeration for at least eight hours with no apparent effect on the steady-state potentials. Normally, however, the membrane potential measurements were carried out immediately after isolation of the eggs or oocytes.
When microelectrodes with 3 to 6 microns tip diameters were used, the steady potential was maintained during penetration of the cell until the electrode tip reached the opposite plasma membrane. However, in penetrations with electrodes having tips less than 1 micron, we observed the appearance of a new decreased steady potential difference as the electrode tip penetrated 0.2 to 0.7 mm below the surface of the animal pole. This is the example shown in the lower record of figure 2. The new steady potential was found to be -37 ±2 mV (SEM, 10 determinations) for oocytes having a transmembrane potential of -63 mV. Furthermore, withdrawal of the exploring electrode and re-entry into the same oocyte yielded a repetition of the initial -63 mV transmembrane potential followed by a shift to the -37 mV potential at approximately the same depth as found on the first penetration.
Histological sections indicated that the -37 mV potential appeared at approximately the depth of the large nucleus, or germinal vesicle, which is about 0.5 mm in diameter and displaced toward the animal pole in R. pipiens oocytes. The position of the tip of the exploring electrode was determined as follows: The electrode was inserted into the -37 mV zone, the solution bathing the oocyte removed, and the oocyte with electrode in place frozen at dry ice temperature. Frozen sections in the plane of the electrode demonstrated the tip of the electrode to be within the nucleus.
Our failure to detect a nuclear potential with electrodes having 3 to 6 micron tips is consistent with the observation (cf. Kanno and Loewenstein, '63) that the nucleus is often pushed for some distance through the cell interior by electrodes with so-called blunt tips, but that electrodes with finer (i.e., less than 1 micron) tips will penetrate the nucleus readily.
The potential difference between the nucleus and cytoplasm was found to average +25 ±2 mV for 8 to 12 oocytes from each of four frogs. In contrast, the average potential difference across the plasma membrane in oocytes from the same four frogs varied from -55 to -68 mV. Thus the potential difference between nucleoplasm and cytoplasm seems more consistent than that between the cytoplasm and the external environment.
Compared to the ovarian oocyte, the vitelline and plasma membranes of the ovulated egg were relatively difficult to penetrate with the microelectrode. Studies with other amphibian eggs (cf. Kanno and Loewenstein, '63) have shown that the plasma membrane may form a sleeve around the electrode and that the electrode may push the membrane through the entire cell without entering the cytoplasm. Similar problems have been encountered with ovulated R. pipiens eggs. In general, rapidly tapering electrodes were found to be most successful for recording from ovulated unfertilized and fertilized eggs.
When R. pipiens eggs are simply pricked with a glass needle they subsequently rotate and within a few hours show puckering of the surface or abortive and irregular cleavage furrows. Activation of the egg could be detected electrically by the appearance of a slow monophasic change in the membrane potential similar to that reported for Bufo eggs (Maeno, '59). As shown in the upper record of figure 3, the activation potential appeared after a 0.65 mm penetration of the egg. When activation occurred, the potential usually increased within ten seconds and attained a maximum value of 50 to 80 mV within 1 to 2 minutes. The potential then declined slowly, usually taking 5 to 10 minutes to reach a steady state. As seen in the lower record of figure 3, no such potential change was recorded during the penetration of a fertilized uncleaved egg.
In order to determine the conditions under which penetration could be effected without activation, a series of electrodes of varying tapers were inserted into ovulated unfertilized eggs. Frequency of activation was found to be correlated with the diameter of the electrode at the egg surface, rather than with the absolute depth of penetration. As reported for Oryzias eggs (Hori,'58), a minimum shaft diameter of 15 to 20 [i at the egg surface was required for activation. Since the tip (0.5 to 3 micron) of the electrode for electrical recording was much smaller than the 15 to 20 micron required for activation, the initial penetration did not activate the egg and indicates that the positive potential seen upon initial penetration of the egg represents the resting membrane potential of the ovulated unfertilized egg.
As with the oocyte, the surface of the egg dimpled as the exploring electrode was slowly lowered. Usually the tip of the microelectrode penetrated the vitelline and plasma membranes spontaneously or entry could be facilitated by tapping the experimental table. As can be seen in figure 3, the cytoplasm of the ovulated unfertilized egg was positive relative to 0.1 strength Ringer's solution. The potential difference across the plasma and vitelline membranes was found to vary from +20 to +50 mV in preparations from different animals but were uniform in eggs obtained from an individual animal. The average standard error of the mean was ±2 mV for membrane potentials of 12 ovulated unfertilized eggs from each of five animals. In comparison, the cytoplasm of the ovarian oocyte remained about 50 mV negative, when transferred to 0.1 strength Ringer's solution, although the oocyte swelled and became very fragile to electrode penetration.
By using the most gradually tapering electrodes, it was possible to effect a deep penetration of the cytoplasm without causing activation. As the electrode tip penetrated 0.3 to 0.6 mm below the animal pole surface, a potential drop followed by a new decreased steady potential was always seen (unless obscured by the appearance of the activation potential) and the recording resembled that seen upon penetration of a membrane. The new decreased steady potential was stable for at least one hour in the ovulated unfertilized egg. Withdrawal of the exploring electrode and re-entry into the same egg again yielded a higher transmembrane potential drop as the electrode tip penetrated 0.3 to 0.6 mm below the egg surface. A similar potential drop was observed during penetration of the vegetal pole. No cytological structure within the egg has been described which corresponds to an ion diffusion barrier 0.3 to 0.6 mm below the surface of the ovulated unfertilized egg. This does not represent a nuclear potential, since the nucleus undergoes two maturation divisions during ovulation and in the ovulated unfertilized egg the metaphase spindle lies close to the animal pole and is greatly reduced in size relative to the oocyte nucleus (cf. Porter, '39). Furthermore, the potential drop 0.3 to 0.6 mm below the egg surface seems to be independent of the activation potential since, as shown in the bottom record of figure 3, a similar potential change appears after a 0.3 mm penetration of the newly fertilized egg.
Eggs with a potential difference of +30 mV (SEM, 10 determinations) across the plasma membrane were found to have a potential difference of +13 ±1 mV between the deeper cytoplasm and the external medium. For comparison, as shown in table 1, eggs with a plasma membrane potential difference of + 41 ±1 mV the potential difference between the deeper cytoplasm and the external medium was found to be + 26 ± mV. Thus the deeper cytoplasm was about 15 to 17 mV negative, relative to the cortical cytoplasm. A potential change has also been observed after insertion of the electrode tip deep into the yolk region of the mature unfertilized Oryzias egg (Hori, '58). Hori has suggested that this difference in potential corresponds to the difference in chemical composition reported by Aketa ('54) between the cortical region of the ovulated Oryzias egg and the deeper yolk region. Unfortunately, comparable histochemical studies have not been carried out in ovulated amphibian eggs.
The cytoplasm of the fertilized egg remained positive relative to 0.1 strength Ringer's solution until the beginning of cleavage. As division proceeded, the cell slowly depolarized and the cytoplasm became 50 to 60 mV negative at completion of the first cleavage. The potential difference between the cytoplasm of one or more cells from each of ten embryos at the twocell stage and an external solution of 0.1 strength Ringer's solution was found to average - 58 ±2 mV as given in table 1. The magnitude of the cytoplasmic potential remained relatively constant after repeated cleavage. The example shown in figure 4 is at the eight-cell stage and demonstrates a complete transverse penetration of the egg, passing through three of the eight cells.
The cortical membrane potential reported here for mature R. pipiens oocytes is in good agreement with the value given by Maeno ('59) for mature Bufo oocytes. It is interesting that much lower membrane potentials (-10 to -20 mV) have been reported by Kanno and Loewenstein ('63) for immature Xenopus oocytes. These latter workers found that, as a rule, membrane potential increased with increasing size of the immature oocyte. If their plot of membrane potential as a function of oocyte diameter is extrapolated to the diameter of the mature oocyte (ca. 2,000 micron) the calculated membrane potential of -55 mV is found to be consistent with the observed values for mature R. pipiens oocytes (this study) and mature Bufo oocytes (Maeno, '59).
In contrast to the positive nuclear membrane potential reported here, Kanno and Loewenstein ('63) found no measurable potential across the nuclear membrane in immature oocytes of newts and frogs. Since the cortical membrane potential apparently increases during oogenesis, the nuclear membrane potential may also increase from a low initial value. A positive nuclear membrane potential has been reportecf by Gelfan ('31) for the germinal vesicles of starfish eggs.
A comparison of the electrical properties of the cortical membrane of the mature oocyte and ovulated unfertilized egg of Bufo have been carried out by Maeno ('59), but the ovulated unfertilized egg was studied in isotonic salt solutions. It can be shown that the eggs of most if not all freshwater animals require hypotonic environments for normal fertilization and development (Holtfreter, '43; Needham, '63). Positive cortical membrane potentials have been previously observed in ovulated unfertilized Oryzias eggs relative to isotonic salt solutions (Maeno et al., '56; Hori, '58). From subsequent experiments with Oryzias eggs, Ito and Maeno ('60) reported that the egg membrane was not positive but slightly negative. The earlier reported positive potentials were explained by these authors as a failure to penetrate the egg membrane with the formation of a plug at the tip of the electrode. We have found that the membrane potential of the ovulated unfertilized frog egg decreased with increasing external NaCl concentration. The transcortical potentials were found to vary from -3 to +12 mV in eggs from different frogs when studied in isotonic Ringer's solution. As mentioned above, eggs of fresh water animals require hypotonic salt solutions for normal development. Since Oryzias is a fresh water fish, it would be necessary to repeat the experiments on these eggs in hypotonic salt solutions to determine the membrane potential under physiological conditions. Transient membrane potential changes have been reported at fertilization or activation for marine invertebrate eggs (Tyler et al., '56; Hiramoto, '59) and for eggs of fish (Hori, 58; Ito and Maeno, '60; Ito, '62) and toad (Maeno, '59). The findings may be summarized as follows: The negative resting potential of the marine invertebrate egg membrane undergoes a slow depolarization returning to the same or more negative resting potential after fertilization in sea water. Both fish and toad egg membranes, on the other hand, undergo a slow hyperpolarization from a negative resting potential upon activation in isotonic salt solution. Both the frog egg membrane (this study) and the toad egg membrane (Maeno, '59) undergo a slow hyperpolarization from a positive resting potential when activated in a hypotonic salt solution. Thus in its natural (or physiological) environment the amphibian egg cytoplasm may be nearly 100 mV positive relative to pond water for a brief period at fertilization.
Ashman et al. have recently reported ('64) that the membrane resistance of the daughter cells produced from first cleavage in Aste-Has forbesi eggs was consistently lower than the resistance of the membrane of the fertilized uncleaved egg. This finding is consistent with the change in membrane potential in R. pipiens eggs at first cleavage and suggests permeability changes during cleavage in both marine and fresh water eggs.
The changes in membrane potential at ovulation, activation and first cleavage may be a reflection of corresponding changes in the ion selectivity of the oocyte and egg membranes. Maeno has suggested ('59) that the activation potential results from a transient increase in chloride ion permeability. Ion analyses of oocytes and ovulated unfertilized eggs indicate that the membrane potential changes at ovulation are accompanied by a marked increase per liter of cell water in intracellular calcium and chloride and by a decrease in intracellular potassium (Morrill, '65). Naora et al. have reported ('62) that the sodium and potassium content of the oocyte nucleus is several times higher than the cytoplasm. At ovulation, the large germinal vesicle is broken down as the maturation divisions are carried out, which must result in an alteration of the chromosomal ionic environment. Examination of the ion permeability changes at the egg membrane as well as the changes in intracellular ion concentrations could provide further information on the sequence of events initiating the development of the vertebrate embryo. Such studies are now in progress.
The authors are indebted to Dr. Adele B. Kostellow and Professor V. E. Amassian for their technical advice and interest in this study. Special thanks are given to Miss Jean Rosenthal for her collaboration in a number of the experiments.
Ashman, R. F., Y. Kanno and W. R. Loewenstein 1964 The intracellular electrical coupling at a forming membrane junction in a dividing cell. Science, 145: 604-605.
Gelfan, S. 1931 Electrical potential difference across the nuclear membrane of the starfish egg. Proc. Soc. Exp. Biol. Med., 29: 58-59.
Hiramoto, Y. 1959 Electric properties of echinoderm eggs. Embryologia, 4: 219-235.
Holtfreter, J. 1943 Properties and functions of the surface coat in amphibian embryos. J. Exp. Zool., 93: 251-323.
Hori, R. 1958 On the membrane potential of the unfertilized egg of the medaka, Oryzias latipes and changes accompanying activation. Embryologia, 4: 79-91.
Ito, S. 1962 Resting potential and activation potential of the Oryzias egg. II. Change of membrane potential and resistance during fertilization. Embryologia, 7: 47-55.
Ito, S., and T. Maeno 1960 Resting potential and activation potential of the Oryzias egg. I. Response to electrical stimulation. Kumamoto J. Sci., Ser. B, Sec. 2 (5): 100-107.
Kanno, Y., and W. R. Loewenstein 1963 A study of the nucleus and cell membranes of oocytes with an intracellular electrode. Exp. Cell Res., 31: 149-166.
Loewenstein, W. R., and Y. Kanno 1963 The electrical conductance and potential across the membrane of some cell nuclei. J. Cell Biol., 16: 421-425.
Maeno, T. 1959 Electrical characteristics and activation potential of Bufo eggs. J. Gen. Physiol., 43: 139-157.
Maeno, T., H. Morita and M. Kuwabara 1956 Potential measurements on the eggs of Japanese killi-fish, Oryzias latipes. Mem. Fac. Sc., Kyushu Univ., Ser. E, 2: 87-94.
Morrill, G. A. 1965 Water and electrolyte changes in amphibian eggs at ovulation. Exp. Cell Res., 40: 664-667.
Morrill, G. A., and D. E. Watson 1962 Transmembrane potential changes during fertilization of R. pipiens eggs. Fed. Proc., 21: 149.
Naora, H., H. Naora, M. Izawa, V. G. AUfrey and A. E. Mirsky 1962 Some observations on differences in composition between the nucleus and cytoplasm of the frog oocyte. Proc. Nat. Acad. Sci., 48: 853-859.
Needham, J. 1963 In: Chemical Embryology, Hafner Pub. Company, New York, Vol. 11: 777-793.
Porter, K. R. 1939 Androgeneic developmentin the egg of Rana pipiens. Biol. Bull., 77: 233-257.
Tyler, A., A. Monroy, C. Y. Kao and H. Grundfest 1956 Membrane potential and resistance of the starfish egg before and after fertflization. Biol. Bull., 111: 153-177.
This work was supported by United States Public Health Service research grants GM 10757 and GM 04715 from the National Institute of General Medical Sciences. Dr. Morrill and Dr. Watson were supported in part by United States Public Health Service training grants 5TI-GM-395 and 2TI-NB-5304, respectively.
Dr. Watson's present address: Biomedical Department, University of California, Livermore, California.