Comparative analysis of ionic currents in the somatic membrane of embryonic and newborn rat sensory neurons

Neur~~c~ence Vol. 58, No. 2, pp. 341-346, 1994 Printed in Great Britain

03~~522/94 S6.Oil+0.00 PergamonPressLtd c 1993 IBRO

COMPARATIVE ANALYSIS OF IONIC CURRENTS IN THE SOMATIC MEMBRANE OF EMBRYONIC AND NEWBORN RAT SENSORY NEURONS S.

A. FEDULOVA,P. G. KOSTYUKand N. S. VESELOVSKY*

Bogomoietz Institute of Physiology, 4 Bogomoletz Street, 252601 GSP, Ukrainian Academy of Sciences, Kiev 24, Ukraine Abstract-Inward currents in the somatic membrane of dissociated rat dorsal root ganglion neurons have been studied in two groups of animals (17 days of embryonic development and the first day after birth) by suction pipette (whole-cell configuration) and voltage-clamp techniques. Altogether 157 neurons were examined. Four components in the inward currents have been identified: fast tetrodotoxin-sensitive (lx,) and slow tetrodotoxin insensitive (I&J sodium, low-(&) and high-threshold (I&) calcium currents. The percentage of neurons demonstrating four types of inward currents [La, Ix, I&, I& increased from 21% in embryo to 61% in newborn. The percentage of neurons with If,,, I&. and I& increased from 4% in embryonic to 14% in the first day after birth. The percentage of cells with If,,, I& and I$ (without tetrodotoxin-insensitive IL) decreased from 56 to 11% in embryo and newborn rats, respectively. A statistically significant linear correlation was found between the densities of IL8 and IL currents for both ages. A correlation also occurred between the densities of Zhiland I&. A reciprocal relation between the densities of both types of calcium currents and the size of cell soma was found in the neurons with all four types of inward currents from newborn animals. A comparison of these data with previous study of inward currents during ~stnat~ development indicates that the most dramatic changes in their distributions and mean densities takes place some time after the birth of the animals.

One of the main features characte~zing the individual properties of a nerve cell is the pattern of electrical activity generated by the activation of different types of ionic channels in its membrane. The neuronal membrane has various combinations of voltage-operated channels and during neuronal development this combination may change substantial1y.3&9,16 Extensive investigations of such characteristics of action potential such as its duration and ionic nature1~“~r8have been made on embryonic neurons and excitable neuronal precursor cells of amphibians. The shortcoming of these investigations was the examination of developmental changes only within a few days. The patch clamp technique permitted the study of membrane currents in different types of cells such as embryonic neurons from amphibians,2 and neuronal precursor cells from chick and rat.“’ These investi-

*To whom correspondence should be addressed at: Depart-

ment of General Physiology of Nervous System, A. A. Bogomoletz Institute of Physiology, 4 Bogomoletz str., 252601 GSP, Ukrainian Academy of Sciences, Kiev 24, Ukraine. A~brev~tio~: DMEM, Dulbecco’s modified Eagle’s medium; DRG, dorsal root ganglion; EGTA, ethyleneglycolbis(aminoethylether)tetra-acetate; HEPES, N-2hydroxyethylpiperazine-N’-2-ethanesulfonic acid; TEA, tetraethylammonium; TTX, tetrodotoxin.

gations have d~onstrated a shift in the ionic nature of action potential from mainly calcium to predominantly sodium during development. In recent investigations two types of calcium conductance have been examined separately.5~6,s,14~19 It was shown that lowthreshold calcium current appears less frequently in adult cells than in young ce11s6~* As was shown in our preliminary investigations, substantial changes still occur in excitability of postnatal dorsal root ganglion (DRG) neurons. A significant decrease in the number of neurons demonstrating four types of inward currents (tetrodotoxin (TTX)-sensitive and TTX-insensitive sodium, lowand high-~reshold calcium currents) was found, and at the same time the number of neurons showing only two types of inward current (TTX-sensitive sodium and high-threshold calcium) increased. The mean density of high threshold calcium channels also increased during postnatal development. The TTXinsensitive sodium currents occurred more rarely in adult neurons than in newborn ones, although the mean density of this current increased during ontogenetic development6 The aim of the present study was to find out if specific changes of the excitatory mechanisms occur immediately after birth, using the same ex~~mental approach. Such data may give us important knowledge about the functional changes of sensory neurons during this most critical step of the ontogenetic development.

341

EXPERIMENTAL

PROCEDURES

A 50 r

Experiments were performed on rats. Unidentified neurons isolated from lumbar dorsal root ganglia of 17-day embryos and animals of the first day after birth were used. This report is based on recordings from 78 embryonic and 79 neonatal neurons. Embryos were removed from timed 17 days (+ 12 hj pregnant anesthetized rats. Ganglia from the embryos were treated with 0.05% protease (Sigma, Type XIV) in Dulbecco’s modified Eagle’s medium (DMEM, Sigma) for 8 min at 33°C. For newborn rats the treatment took 16 min at the same temperature and enzyme concentration. Ganglia were thoroughly washed and dissociated in DMEM, and isolated neurons were plated on coated glasses. The primary culture was held in a CO, incubator for 2--3h at 35°C; within this time adhesion of neurons to glass was completed. Cells were used for experiments during the first 5-6 h in vitro.

A_*

1

0

300

-i

600

900

60 r

.

B

0

:

40

‘.

20

. *

l l

.

*

.

Whole-cell patch-clamp experiments were carried out in normal extracellular solution containing (mM): NaCl 140, KC1 5, CaCf, 2, MgCI, 2, HEPES 10, pH was adjusted to 7.3 with NaOH. For recording slow sodium and calcium currents we used additional pipettes for superfusion of the cell with normal extracellular solution containing lo-’ M TTX and sodium-free extracellular solution containing (mM): choline chloride 130, tetraethylammonium (TEA)chloride 10, MgCl, 2, CaCl, 2, HEPES 10, pH was adjusted to 7.3 with KOH. Voltage-clamp experiments were carried out with the solution in the pipette (mM): CsCl 60, CsOH 60, EGTA 10, HEPES 20, TEA-Cl 2, MgCI, 2, pH was adjusted to 7.3 using DL-asparaginic acid. Electroph_wiologicnl

measurements

All the experiments were done using a single-electrode voltage-clamp amplifier, the working frequencies were in the range 25-50 kHz. The resistances of patch-clamp pipettes were about f-2 MSZ. Before each registration of ionic currents the single-electrode voltage-clamp amplifier was adjusted to obtain the square-form of membrane potential shift for excluding a “hump”. The measurements of the capacitance of somatic membrane were done in two ways: (i) by calculating the integral of the area under the capacitance charge current;6 and (ii) by calculating the capacity from the equation of maximum frequency resolution.

1

0

1200 900 600 300 Fig. 1. Correlation between somatic membrane capacity and area of the DRG neurons from embryo (A) and neonatal (B) rats. Abscissa: area of cellular soma (pm*); ordinate: membrane capacity (pF).

newborn animals was 663.3 & 166.7pm2. The increase in the mean surface of the neuronal soma as well as the mean capacity of cells were statistically significant. As we mentioned previously,6 certain combinations of inward currents in the somatic membrane allowed the division of the heterogeneous population of all the investigated neurons into five homogeneous subpopulations, and the same approach was used in the present study. Neurons, which had in their somatic membrane all four types of inward currents: TTX-sensitive fast and TTX-insensitive slow sodium, 20

In the present investigation we took into account four types of inward currents: TTX-sensitive fast (IL,) and TTX-insensitive slow (fk,) sodium currents, low-(IL*) and high-(Z&) threshold calcium currents. The recorded inward currents from embryonic and neonatal neurons did not show any additional electrophysiological peculiarities compared to those previously described for adult neurons.6,7.12 The size of all neurons examined was estimated under the microscope. The capacity of the neuronal membrane (considered as a value proportional to the surface of the somatic membrane) was also measured and used to define the current densities. The correlation between the membrane capacity and the somatic area measured under the microscope is shown in Fig. 1. The distribution of neurons according to the surface of their soma for both groups investigated is shown in Fig. 2. The mean surface of embryonic neurons was 469.6 & 161.7 ,um2, and for those from

A

c

RESULTS

. 10

.

- .

5 . 15 :. 0 150 300

20 15 $0 5

. . li,. . l__Ji1 B

r

A

i t

l

I

r-7

.

.

0

600 750 900

.

l

150 300 450 800 750 900 1050

Fig, 2. Distribution histograms for neurons of the embryo (A) and neonatal (B) rats piotted against their somatic area. Abscissa: somatic area (pm2); ordinate: number of neurons with corresponding somatic area. Filled circles indicate expected frequency after Kolmogorov-Smimov test.

Ionic currents in the embryonic 20

and newborn

rat sensory neurons

343

Table 1. Quantitative distribution of dorsal root ganglion neurons from embryonic and neonatal rats in different subpopulations

15

Subpopulations III II

I

10 Embryo

16 20.5%*** 48 60.8%

5 Newborn 1 0

150

300 450

600

n

B

10 5

0

150 300 450

600

v

15 19.2% 8 10.1%

3 3.8%

750 900

20 c

I

44 56.4%*** 9 11.4%

3 3.9%* 11 13.9%

IV

750 900

Fig. 3. Distribution of the neurons from the first (A) and third (B) subpopulations according to the area of their soma (filled area) plotted over the distribution histograms from Fig. 2A. Abscissa and ordinate arc the same as in Fig. 2.

low- and high-threshold calcium currents, formed the first subpopulation. Neurons without detectable lowthreshold calcium currents formed the second subpopulation. Neurons without TTX-insensitive slow sodium current formed the third subpopulation. Finally neurons, which had in the somatic membrane only TTX-sensitive fast sodium and high-threshold calcium currents, formed the fourth subpopulation. In only three cells from newborn rats a combination of TTX-insensitive slow sodium and high-threshold

calcium lation,

currents

was

observed

(the

fifth

subpopu-

see Ref. 6).

The distribution of neurons from animals of different ages according to occurrence of the above mentioned subpopulations is shown in Table 1. Results of statistical analysis of these data using G-test for equality of two neighbouring groups are also shown in Table 1. Here and elsewhere we used a single asterisk to indicate the 95% confidence limit; two asterisks, 99%; and three asterisks, 99.9% confidence limits. The most prominent alteration occurred in the number of neurons of the first and third subpopulations: a statistically significant increase of cells exhibiting TTX-insensitive ZNaand an increase of cells exhibiting all types of inward currents. Since we investigated unidentified neurons to make this finding more reliable we checked the possibility of a loss of neurons with a certain size during development, as this could influence the number of neurons in these subpopulations. Figure 3 shows the distributions of neurons from the first and third subpopulations according to the surface of their soma plotted over the general distribution of somatic surfaces taken C

D

Fig. 4. Distribution of densities of inward currents in the DRG neurons from embryonic (A, B, C, D) and neonatal (E, F, G, H) rats. Abscissa: density of inward currents (pA/pF), ordinate: number of neurons with corresponding current density. (A, E) TTX-sensitive, and (B, F) for TTX-insensitive sodium currents; (C, G) low- and (D, H) for high-threshold calcium currents.

344

S. A.

FEDLLOVA et al

Table 2. Mean densities (p, A/F) of ionic currents and mean somatic surface (S, pm*) in four main subpopulations of dorsal root ganglion neurons I pfNa Embryo Neonatal p ‘Na Embryo Neonatal p’Ca Embryo Neonatal phCa Embryo Neonatal s Embryo Neonatal

II

391 308

389 272

119** 195

230 211

III 345 279

I

. . .

,

700

65 66 240*** 662

601

A

299 238

4.5 4

535* 661

70

IV

.. . -

4.3 3.6 56 51

90 60

45* 35 478* 696

-

B 30 51

.

420** 641

. from

Fig.

2. It shows

that

the

mode

of the

distri-

butions of the first and third subpopulations did not differ from the mode of distribution of somatic surfaces for all neurons investigated and this took into account a loss of small (or large) neurons. The overall distributions of current densities in neurons from embryonic and newborn rats are shown in Fig. 4. As can be seen, some of them are not monomodal, making their statistical comparison difficult. Therefore it was better to compare the mean densities of currents in embryonic and neonatal neurons in the subpopulations mentioned above. Table 2 presents the mean densities of each inward current in these subpopulations and the statistically significant differences between them. Statistically significant changes of the current densities were only observed in two cases: density Z& in the third subpopulation decreased and density of ZR, increased during the period of birth.

0

2

4

6

6

10

Fig. 5. Examples of direct (A) and inverse (B) correlation. (A) Direct correlation between densities of T’I’X-sensitive sodium (abscissa, pA/pF) and low-threshold calcium (ordinate, pA/pF) currents for embryonic neurons from the third subpopulation (n = 44). (B) Inverse correlation between low-threshold calcium current density (abscissa, pA/pF) and membrane capacity (ordinate, pF) for neonatal neurons from the first subpopulation (n = 46).

We also carried out a correlation analysis for current densities and sizes of cells in all subpopulations. Since such analysis gives a reason only for a numerous subpopulation, we calculated the Braves-Pirson correlation coefficient r only for the third subpopulation of embryonic cells, and the first subpopulation of neurons from neonatal animals.

% I

6or _

111

Table 3. The Braves-Pirson correlation coefficients (r) for the first and second subpopulations in embryo and neonatal dorsal root ganglion neurons Variables x

Y

Subpopulation I p ‘Na p”Na p ‘Na phCa p”Na phCa p’Ca p’Ca phCa phCa

S Subpopulation p’Na S All neurons p ‘Na p’Na p’Na p’Ca p’Ca S

S C S C

C

Embryo Corr(r) -0.04 -0.06

Neonatal Corr(r)

0.15 0.51* 0.33 -0.18

0.41** -0.30* -0.43** -0.35*

-0.12 0.74**

-0.41** 0.56***

III phCa C

0.40** 0.56***

pSNa phCa p hCa S C C

0.1 0.41*** 0.53* 0.24 -0.06 0.63***

-5

0.47*** 0.36*

1

7

45

90

7.

_~ 15

-5

1

7

45

90

-5

1

7

45

90

60 50 40

10

30 20

5

10

0.35* 0.72*’ 0.47*** 0.32** 0.38** -0.32* -0.43*** 0.58***

0

-5

1

7

45

90

0

Fig. 6. Changes in the proportion of rat DRG neurons, revealing different combinations of inward current channels during ontogenesis. The subpopulations revealed I k, + &+&+I& (I), r~~+r~,+I& (II), &,+&+I;, (III), and IL, + 1k (IV). Abscissa: days before and after birth. Asterisks indicate IO (*), 5 (**) and 1 (***) % significant levels.

Ionic currents in the embryonic and newborn rat sensory neurons 80 60 40 20 0

-5

1

7 45 90

80 60

Ei :

Fig. 7. Changes in the proportion of rat DRG neurons, revealing low-thr~hoid calcium (A) and OX-insensitive sodium (B) currents in all neurons investigated during ontogenesis. Abscissa: days before and after birth.

Table 3 summari~

the results of such an analysis and Fig. 5 illustrates examples of direct correlation

between the densities of TTX-sensitive sodium and high-threshold calcium currents (Fig. 5A) and an inverse correlation between the densities of both types of Ca currents and the size of cell soma significant for neurons from newborn animals and absent for embryonic cells (Fig. 5B). DISCUSSION

The present series of experiments on neurons from embryos and animals at the first day after birth are the continuation of previously publish~ observations for the same neurons during postnatal development6 To analyse the experimental data, we divided the total heterogeneous population of neurons into more homogeneous subpopulations which probably contained neurons similar in their functional properties. A criterion for grouping was the existence of certain combinations of inward currents in their somatic membrane.6 Such a division was quite important for the analysis: neurons, which belonged to different subpopulations, showed different types of correlation between the densities of currents and different mean values of the latter. These features were constant for the subpopulations mentioned within the observation period, and alterations were found only in the redistribution in the number of neurons in these subpopulations. All embryonic neurons were ~h~acterized by direct correlation between different variables: this embryonic correlation disappeared in postnatal neurons, indicating that the ways of channel expression in embryonic cells is quali~tively different from that in neonatal or adult neurons. The establishment of an inverse correlation between the densities of both types of calcium currents and the somatic area in neonatal neurons obviously reflect the maturation of their functional individual-

345

ity. The high density of calcium currents in small neurons will increase the duration of action potentials. In fact, microelectrode recordings have shown that the action potentials in small neurons have longer duration correlating with smaller conduction velocity of their axons4.13 In respect of the expression of both types of I,-, in different cells, no substantial changes occur after birth. Both low- and high-voltage activated Ca channels occur in about the same number of cells. The present investigation has shown that 77% of embry onic neurons and 72% of neurons from newborn animals (I and III subpopulations) had low-threshold calcium currents. The situation changes dramatically if these data are compared with our earlier findings about channel expression during postnatal development.6 The percentages decreased to 47% in neurons from eight- to lo-day-old, to 34% in 45-SO-day-old and to 18% in 90-lOO-day-old animals (Fig. 6). Similar results were described for skeletal muscle by Gonoi and Hosegawa.* These authors as well as Beam and Knudson3 also recorded a decrease in mean density of low-threshold calcium current during postnatal development. However, Lovinger and White14 showed a postnatal increase of this current amplitude in rat DRG neurons. In our experiments we found that change in the mean density of lowthreshold calcium currents was statistically nonsignificant in all subpopulations (see Table 2 in this paper and Table 2 in Ref. 6) but the total number of neurons with this current decreased with time (Fig. 7A). It should be mentioned that only 24% of embryonic neurons had TV-insensitive sodium current (I and II subpopulation); expression of these channels also substantially increased during birth-75% of cells (Fig. 7B). As we have shown previously, the quantity of neurons with ~-insen~tive sodium current decreased during postnatal development from 35% in neurons from eight- to lo-day-old animals to 23% in 45-50 and to 14% in 90-lOO-day-old ones. The TTX-insensitive sodium current was studied earlier in chick cardiomyocites at different stages of embryonic development. l6 It has been shown that the excitability of cardiomyocites at early stages of embryonic development is determined by TTX-insensitive sodium conductive, and only at a later stage does the action potential of these cells become TTXsensitive.16McLean and Sperelakis” have shown that dissociated cardiomyocites from chick embryo kept in culture lose TTX-sensitive sodium conductan~, and TTX-insensitive sodium conductance appears. Although our findings indicate a similar type of developmental change of sodium conductance in DRG neurons, some qu~titative differences are obvious: embryonic DRG neurons showed TTXinsensitive sodium conductance less frequently than neurons from newborn animals and only during postnatal development did the number of such cells progressively decrease. This may represent the

346

S. A. FEDULOVA et al.

difference between the conditions encountered by a cell in vitro, and cells which normally develop in viva. CONCLUSION

The experimental data demonstrate that principal changes occur in the ionic mechanisms

no of

electrical excitability of rat sensory neurons during the period of birth. The most obvious is the establishment of an inverse correlation between the densities of both types of Ca currents and the size of the cell, probably indicating the end of functional maturation of large and small neurons in DRG.

REFERENCES

1. Baccaglini P. I. (1978) Action potential of embryonic dorsal root ganglion neurons in Xenopus tadpoles. J. Physiol.. Lond. 283, 585-604. 2.

3. 4. 5. 6. I. 8. 9. 10.

Bader C. R., Bertrand D., Dupin E. and Kato A. C. (1983) Development of electrical membrane properties in cultured avian neural crest. Nature 305, 808-810. Beam K. G. and Knudson C. M. (1988) Effect of postnatal development on calcium currents and slow charge movement in mammalian skeletal muscle. J. gen. Physiol. 91, 799-815. Cameron A. A., Lean J. D. and Snow P. J. (1986) The electrophysiological and morphological characteristics of feline dorsal root ganglion cells. Brain. Res. 362, l-6. Carbone E. and Lux H. D. (1984) A low voltage-activated calcium conductance in embryonic chick sensory neurons. Biophys. J. 46, 413418. Fedulova S. A., Kostyuk P. G. and Veselovsky N. S. (1991) Ionic mechanisms of electrical excitability in rat sensory neurons during postnatal ontogenesis. Neuroscience 41, 303-309. Fedulova S. A., Kostyuk P. G. and Veselovsky N. S. (1985) Two types of calcium channels in the somatic membrane of newborn rat’s dorsal root ganglion neurons. J. Physiol., Land. 359, 431446. Gonoi T. and Hosegawa S. (1988) Postnatal disappearance of transient calcium channels in mouse skeletal muscle: effect of denervation and culture. J. Physiol., Lond. 401, 617-637. Gottmann K., Dietzel I. D., Lux H. D., Huck S. and Rohrer H. (1988) Development of inward currents in chick sensory and autonomic neuronal precursor cells in culture. J. Neurosci. 8, 3722-3732. Gottmann K., Dieter I. D., Lux H. D. and Ruedel C. (1989) Proton-induced Na current develops prior to voltage-dependent Na and Ca currents in neuronal precursor cells from chick dorsal root ganglion. Neurosci. Lett. 99, 90-94.

11. Grantyn R., Peruansky M., Rodrigues-T burg A. and Lux H. D. (1989) Expression of depolarizing voltage and transmitter-activated currents in neuronal precursor cells from the rat brain is preceded by proton-activated sodium current. Devl Brain Res. 49, 150-155. 12. Kostyuk P. G., Veselovsky N. S. and Tsyndrenko A. Y. (1981) Ionic currents in the somatic membrane of rat dorsal root ganglion neurons-I. Sodium currents. Neuroscience 6, 2423-2430. 13. Ho L. K., Kyungson C., MO C. J. and Coggeshall R. E. (1986) Correlation of cell body size, axon size and signal conduction velocity for individually labeled dorsal root ganglion cells in the cat. J. camp. Neural. 243, 335-346. 14. Lovinger D. M. and White G. (1989) Postnatal development of burst firing behavior and the low threshold transient calcium current examined using freshly isolated neurons from rat dorsal root ganglia. Neurosci. Lett. 102, 50-57. 15. McLean M. J. and Sperelakis N. (1974) Rapid loss of sensitivity to tetrodotoxin by chick ventricular myocardial cells after separation from the heart. Expl Cell Res. 86, 351-364. 16. Sperelakis N. and Shigenobu K. (1972) Changes in membrane properties of chick embryonic hearts during development. J. gen. Physiol. 60, 430453. 17. 18. 19.

Spitzer N. S. and Lamborghini J. E. (1974) The development of the action potential mechanism of amphibian neurons isolated in culture. Proc. natn. Acad. Sci. U.S.A. 73, 164-1645. Spitzer N. C. (1979) Ion channels in development. A. Rev. Neurosci. 2, 363-397. Yaari Y., Hamon B. and Lux H. D. (1987) Development of two types of calcium channels in cultured mammalian hippocampal neurons. Science 235, 680682. (Accepted 9 August 1993)