- Email: [email protected]
Effects of divalent cations on the time course of post-tetanic decay of miniature endplate potential frequency in frogs
0304~522/9l $3.00f 0.00 Pergamon Press plc c 1991IBRO
~e~~~~cie~ee Vol. 40, NO. 3, pp. 879483, 1991 Printed in Great Britain
EFFECTS OF DIVALE~ CATIONS ON THE TIME COURSE OF POST-TETANIC DECAY OF MINIATURE ENDPLATE POTENTIAL FREQUENCY IN FROGS K. NARITA and I-I. KITA* Department of Physiology, Kawasaki Medical School, Kurashiki, Okayama 701-01, Japan Ahstrnct-Relatively high external Mg’+ specifically induces a shoulder in the post-tetanic decay of miniature endplate potential frequency at frog neuromuscular junctions. This effect is antagonized by Ca’+, but not by SrZ+ and Mn 2+. This shoulder formation is not caused by Mn*+ or dinitrophenol in the medium. It is suggested that Mgz+ enters nerve terminals and displaces Ca”+ from internal stores; Mg+ subsequently interferes with Ca2+ removal until the MgZ+ itself has been removed. The dinitrophenol result suggests that ATP-dependent Ca*+ extrusion is not permanently slowed. An equally likely and simpler explanation is that Mg*+ enters nerve terminals and the Mg2+ itself causes an increase in miniature endplate potential frequency. External Ca’+ competes with Mg 2+ for channel entry and prevents this effect. Tbe time course of the decay of miniature endplate potential frequency reflects the processes involved in Mg’+ extrusion and/or uptake.
The frequency of miniature endplate potentials EXPERIMENTAL PROCEDURES (MEPPs) increases during tetanic stimulation of the Experiments were performed on sciatic nerve-sartorius motor nerve in Ca2+-containing saline solutions,5 muscle preparations dissected from the frog, Rnnu nigro~acu~ur~. Following dissection, the preparations were probably due to an increased Ca*+ con~ntration soaked for several hours in 0 CaZ+-Mg2+ EGTA solution within the motor nerve terminals owing to its influx (see below) at YC, after which they were mounted and during tetanic stimulation.’ In solutions lacking stretched to about 120% of their slack length in a chamber Ca2+ but containing Sr’+, Mg*‘, Mn2+, Co2+ or NiZC with a capacity of 5 ml. The nerve was placed on a pair of as the sole divalent cation, tetanic stimulation prosilver wire electrodes and covered with a mixture of Vaseline duces similar increases in the MEPP frequency.6~8~‘1~14 and paraffin oil. The basic saline solution used for the dissection conThis has been explained by postulating that the tained (mM): NaCl, 100, KCI, 2; CaCl,, 2.5; MgCl,, 3; divalent cations enter the terminal through Ca*+ Tris-maleate buffer (DH 7.4. 8. The 0 Ca’+-M$+ EGTA channels during stimulation and then either release solution contained ImM I&Cl, and 1 mM sycoletherdiaminetetra-acetic acid (EGTA), and the other constituents Ca2+ from intraterminal stores or mimic the action of Ca2f 11.12.20 were unchanged except that CaCl, was excluded. Solutions After the cessation of tetanic stimulation in solutions containing Ca2+, Sr2+, Mn2+ or Mg2+, the elevated frequency gradually falls toward the initial level, along a roughly exponential time course, presumably due to sequestration of Ca2+ by the Ca2+ stores and extrusion of these cations from the terminals.” If, as has been suggested, the MEPP frequency is dependent on the concentration of Ca2+ within the terminal, the time course of the decay of the MEPP frequency should reflect the decline of the elevated concentration of the divalent cation in the terminal. This time course is influenced by divalent cations in the bathing solution.‘0~‘8
containing either M$+, Sti+ or Mn*+ as the sole divalent cation did not include EGTA. When the concentrations of MgZf, Srz2+or Mn2+ were increased in the solutions containing Mg*+, Sr*+ or Mn’+ in place of Ca*+, the osmolarities of all solutions were maintained equal by adding sucrose. Neostigmine bromide ( 10m6g/ml) was added to all solutions for easier detection of MEPPs. The solution temperature was kept at 19-22°C. When solutions were changed, 50 ml of new solution was allowed to flow into the chamber at a rate of lOml/min. MEPPs were recorded with an jntra~llular glass microelectrode and counted with a special device” to follow the time course of the decline of the frequency aAer supramaximal tetanic stimulation (50 Hz, 2-3 min). RESULTS Eflect
*To whom correspondence should be addressed. Abbreuiulions: [Caz+],, intracellular free calcium ion concentration; DNP, 2+dinitrophenol; EGTA, glycoletherdiaminetetra-acetic acid; MEPP, miniature endplate potential; [Mg*+],, intracellular free magnesium ion concentration.
of Mg2+ on post-tetanic decay
When the M$+ concentration was 1 mM, the time course for the decline of the elevated MEPP frequency produced by tetanic stimulation was roughly exponential. Increasing the Mg2+ concentration slowed the decline (Fig. 1). In 6 mM or 11 mM, the 879
X80
K.
NAKITA
and H. KIT,%
&&ct
0.1 0
, 5
,
, 15
IO Tim*
of Sr’ ’ on
the shouldu
The antagonistic action produced by C‘a,” W;IS not found with Sr’+. The shoulder produced by IOmM Mg2+ remained almost unchanged when the concentration of Sr’+ was raised to 0.2 and 0.4 mM (Fig. 3). Since Sr2+ is known to be much less effective than Ca*+ in evoking transmitter release,‘,‘” one might argue that Sr2+ might have an antagonistic effect if higher concentrations were used. In our exper20
(mln)
Fig. 1. The effect of Mg*+ on the time course of the decline of the MEPP frequency after cessation of tetanic stimulation for 3 min at 50 Hz. The preparation was soaked for 7.5 h in a refrigerated 0 CaZ+-Mg*+ EGTA solution containing 30 mM sucrose. The preparation was then transferred to a new solution of the same content and soaked in it for 30 min until the solution reached room temperature. After the equilibration, MEPPs were recorded for IOmin. Then the nerve was stimulated tetanically and the recording was continued until 14 min elapsed after the end of the tetanus. Thereafter the concentration of Mg’+ was increased from 1 to 6mM and recording of the tetanus-recovery sequence was repeated. After 6 mM, a further 5 mM of MgCl, was added and the post-tetanic decay in the MEPP frequency was followed until 21 min after stimulation. Then the preparation was returned to the initial solution and MEPPs were recorded until 19min after the tetanus. During the course of the experiment, the osmolarity of the solution was kept constant by changing the concentration of added sucrose. This figure is a representative example of five experiments.
curves displayed a shoulder, indicating that the decline in the MEPP frequency was initially slow and then accelerated (Fig. 1). The slowing of the initial decay was characteristic of relatively high Mg2+ concentration. In other experiments, it was observed that at 3SmM Mg*+, the time course of decay was prolonged, though the shape of the curve remained the same, while at 5 mM
iments, however, Sr2+ at concentrations higher than 0.4 mM evoked endplate potentials, which precluded measurements of MEPP frequencies. In one successful experiment in which 0.8 mM Sr’+ was used. no antagonistic effect was observed. l@ect of Mn”
on the shoulder and post-tetanic
decay
The shoulder produced by relatively high Mg2+ was similarly unaffected by the addition of 0.2-l .OmM Mn2+ (Fig. 4). The post-tetanic decay in Mn2+ solutions followed a rough exponential curve with upward parallel shifts as the Mn2+- concentration was increased (Fig. 5). Mn2’ in the concentrations used did not play any role in the shoulder formation. It was shown previously that Mn2+, even at concentrations of 5, 7.5 and 10 mM, did not produce a shoulder.” Effect of dinitrophenol
on post -tetanic decay
The post-tetanic decay followed a rough exponential curve when the medium contained 0.5 mM Mn2+ as the only divalent cation (Fig. 6). The addition of 3 ,uM 2,4-dinitrophenol (DNP), which is known to decouple oxidative phosphorylation, had little
Mg2+, a shoulder was definitely present. Antagonistic formation
action of Mg2+ and Ca2+ on shoulder
Increasing concentrations of Ca2+, from 0.05 mM, made the shoulder on the post-tetanic decay curve of the MEPPs produced by 8mM Mg2+ less evident, and it finally disappeared at 0.2mM Ca’+(Fig. 2). Thus Ca2+ antagonized Mg2+ in the formation of the shoulder. To demonstrate the shoulder a little more quantitatively, the ratios of the post-tetanic MEPP frequency at 3 min to the frequency immediately after the end of tetanic stimulation were measured (Table I). These ratios approached one as the Mg2+ concentration was increased, showing the formation of the shoulder. In solutions containing 8 or 10mM MgZC and varying concentrations of Ca2+, the ratios fell to smaller values as the Ca2+ concentration was raised, showing the disappearance of the shoulder.
Time
IminI
Fig. 2. Antagonistic action of Ca*+ on shoulder formation. The preparation was soaked for 4 h in a refrigerated 0 Ca’+-M&+ EGTA solution containing 25 mM sucrose. It was then transferred and bathed for 30 min in 0 Ca2+-MgZ+ EGTA solution with 8 mM MgCl, and 1.5 mM sucrose at room temperature. After recording MEPPs for 10 min, the nerve was stimulated for 2min at 50Hz and the MEPP frequency was followed until 12min after the stimulation. Afterwards, the concentration of Ca’+ was increased to 0.05, 0.1 and 0.2 mM with corresponding decreases in the sucrose concentration to maintain the same osmolarity and recording of the tetanus-recovery sequence was repeated in each solution. This figure is a representative example of seven experiments.
881
Cations and post-tetanic decay Table 1. Ratio of post-tetanic miniature endplate potential frequency at 3 min to frequency immediately (at 0 min) after end of tetanic stimulation Experiment
Mg2+ concentration (mM)
Ratio (3 min/O min)
In Mg2+ solutions
1
1 6 11 1 (recovery)
2
I 3.5 6 11
3
1 3.5 6 8.5 11
Experiment
Ca2+ concentration (mW
In solutions containing Mgr+ and Ca*+ I(10 mM Mg*+) 0 0.05 0.1 0.2 0.3 2 (10 mM Mg2+) 0 0.05 0.1 0.2 3 (8 mM M$+) 0 0.05 0.1 0.2
0.28 0.26 0.74 0.20 0.29 0.36 0.58 0.79 0.23 0.30 0.45 0.49 0.64 Ratio (3 min/O min) 0.68 0.59 0.27 0.27 0.14 0.73 0.65 0.33 0.17 0.81 0.32 0.18 0.13
The experiments in this table are not those shown in Figs 1 and 2. or no effect on the time course of the decay. An
,,....l....,....,....
0
5
10 Tim*
15
$0
I
(min)
Fig. 3. Effect of Srr+ on the shoulder. The preparation was soaked for 3.5 h in a refrigerated 0 Ca*+-Mg2+ EGTA solution containing 24 mM sucrose.. It was then transferred and immersed for 30 mm in the same solution with 10 mM MgCI, and 1.5 mM sucrose at room temperature. After 10 mm recording of MEPPs, the nerve was stimulated for 2 min at 50 Hz and recording was carried out until 16 min after the tetanus. Thereafter the concentration of Sr2+ was raised to 0.2 and 0.4mM with corresponding decreases in the sucrose concentration to maintain the same osmolarity and recording of the tetanus-recovery sequence was repeated at each concentration of Sti+. This figure shows a representative example of nine experiments.
and when Mg *+ is the only divalent cation in the bathing solution, M$+ enters the terminal through the open channel in place of Ca2+.8~‘2~20 Although the dose-dependent tetanic rise in MEPP frequency in Mg2+ solutions indicates Mg2+ entry into the frog nerve terminal,8.‘4 the terminal is too small to determine directly whether its Ca*+ channels are permeable to M$+. Since the Ca*+ channels in frog skeletal muscle fibres show clearly measurable Mg2+ currents as well as Ba’+, Sr*+, Ca*+ and Mn*+
increase in the concentration of DNP to 5 PM, which was the highest concentration we could examine, shifted the decay curve upward and slowed the decay. The time course remained exponential and there was no shoulder (Fig. 6). DISCUSSION
The present work shows that the decline in the post-tetanic potentiation of the MEPP frequency was prolonged by relatively high M$+, which caused the formation of a shoulder on the decay curve (a convex curve). Among the divalent cations tested only Mg*+ was effective for the formation of a shoulder. It was also demonstrated that Mg*+ was antagonized by Ca*+ in this phenomenon, but not by Sr*+ and Mn*+. Furthermore, Mn*+ and a metabolic inhibitor, DNP, did not produce a shoulder, though the latter slowed the decay. It has been reported that the decay of post-tetanic potentiation slows when the Mg*+ concentration is elevated,14 but detailed analyses of the shape of the decay curve in the Mg2+ solutions have not been done. We believe that depolarization of the nerve terminal membrane caused by tetanic stimulation opens up the voltage-gated Ca*+ channels of the membrane,
c
p”
9 0.1-l
0
1
5
.
,
.
10 Tim*
I
15
.
,
I
20
(inin)
Fig. 4. Effect of MnZ+ on the shoulder. The preparation was soaked for 3.5 h in a refrigerated 0 Ca2+-Mg*+ EGTA solution containing 31 mM sucrose. It was then transferred and bathed for 1 h in the same solution with 8 mM MgCl, and 7.5 mM sucrose at room temperature. After 10min recording of MEPPs, the nerve was stimulated for 2 mm at 50 Hz and the recording was continued until 16 min after the stimulation. Thereafter the concentration of MnZ+ was increased from 0 to 0.2, 0.5 and 1 mM with corresponding concentrations of sucrose to maintain the osmolarity of the solutions and recording of the tetanus-recovery sequence was repeated in each solution. This figure is a representative example of six experiments in which 5, 8 or 10mM Mg2+ was used to produce the shoulder.
882
K. NARI,I’A and H.
Fig. 5. Effect of Mn2+ on post-tetanic decay in 0 Ca*+-MgZ+ EGTA solutions. The preparation was soaked for 4 h in a refrigerated 0 Ca*+-Mg’+ EGTA solution containing 15 mM sucrose and for 3 h in a fresh solution of the same composition at 20°C. After recording MEPPs for lOmin, the nerve was stimulated for 2 mm at 50 Hz and recording was done until 11 min after the tetanus. Thereafter the concentration of Mn*+ was increased to 0.2, 1 and 2.5 mM with corresponding decreases in the concentration of sucrose and recording of the tetanus-recovery sequence was repeated in these solutions. This figure shows a representative result from five experiments.
currents,‘~‘5 the existence of similar Ca*+ channels in the nerve terminal membrane is quite probable. The increased intraterminal concentration of free Mg”, [Mg2+li, could somehow increase the intraterminal concentration of free Ca”, [Ca’+],, probably by releasing Ca 2+ from intraterminal stores, the most probable candidates for which are the smooth endoplasmic reticulum and mitochondria.‘* After cessation of the tetanus, the increased [Ca2+ji gradually decreases to the resting level by both extrusion from the terminal and sequestration in the intraterminal Ca stores.3.‘0 The Ca2+ extrusion probably takes place mainly through the ATP-driven Ca*+ pump and Na2+-Ca2+ exchanger. The time course for the .
0.5mll
M”
. .
0.5mm 0.5mM
Y” Y”
+ 3* * 5.M
ONP
ow
5 t ;
O.Oli
0
.
.
I
I
10
5 Time
.
.
I
15
I
20
I
(main)
Fig. 6. Effect of DNP on post-tetanic decay in 0.5 mM Mn*+ solution. The preparation was soaked for 3.5 h in refiigerated 0 Ca*+-Mg*+ EGTA solution. It was then transferred and immersed for 30min in the same solution containing 0.5 mM Mn2+ at 20°C. After recording control MEPPs for 10 min, the nerve was tetanized for 2min at 50 Hz and MEPPs were recorded until 11 min after the tetanus. Thereafter 3 and 5 PM DNP were added and recording of the tetanus-recovery sequence was repeated in both solutions. This figure is a representative example of three experiments.
KITA
decay in [Ca”], near the release sites would be reflected in the time course of the decline of the post-tetanic MEPP frequency.“’ We believe that the shoulder on the decay curve of the post-tetanic MEPP frequency is formed in the following way. After tctanic stimulation, the increased [Mg2+], would lower the activity of the ATP-driven Ca2+ pump, because DNP prolonged the exponential decay of the post-tetanic MEPP frequency at 5 PM (Fig. 6). Suppression of the pump would result in the slowing of Ca’+ extrusion. ATP-driven Ca*+ transport systems, which include the surface membrane Ca*’ pump and the sarco(endo)plasmic reticulum Ca” pump, require for their functioning the presence of free Mg*+ on the &-side, the side from which Ca2+ moves.*,‘* However, the effect of Mg *+ is biphasic and a high Mg’+ concentration is inhibitory,4s’3 probably due to competition with Cal+ at the transport site.22 This finding supports the above-mentioned view regarding the role of relatively high Mg2’ in shoulder formation, Once the [Mg”], falls to a critical level, probably by extrusion and/or sequestration, extrusion of Cal’ would be accelerated by the normal functioning of the Ca’+ pump and the time course for the post-tetanic MEPP frequency would show a rapid fall following the shoulder. Since it has been reported that the Na+ concentration affects post-tetanic potentiation through the Na+--Ca’+ exchanger and the effect of Mg’+ on its activity is inhibitory,‘7,‘8.2’ the exchanger probably contributes to shoulder formation. However, our experiment, in which the effect of halving the Na’ concentration on the post-tetanic decay was examined, could not elucidate this point. The difference between Mg2+ and DNP in shoulder formation is considered to be as follows. DNP suppresses the ATP-driven pump at an almost constant rate with time, which produces a prolonged exponential decay curve. In contrast, the depression due to elevated [Mg2+], decreases with time after tetanic stimulation, so its depressive action on the pump is stronger initially, and then gradually becomes weaker. When the [Mg’+], falls below a certain critical level, the inhibitory action of Mg2+ is lost and the pump recovers its normal activity. This causes a shift in the post-tetanic MEPP frequency from an upper, slower exponential curve to a lower, faster one during an earlier phase of decay, resulting in the formation of a shoulder. A simpler, alternative explanation is that Mg2+, which enters nerve terminals during tetanic stimulation, acts directly on the release machinery to cause quanta1 release of transmitter. After the tetanus, the elevated [Mg2+li decays as the accumulated Mg’ + is extruded from the terminals at a rate that is initially slow, and then accelerates. The process of the clearing of Mg ‘+ is reflected in the time course of post-tetanic MEPP frequency showing a shoulder. The antagonin the shoulder ism between Mg2+ and Ca”
Cations and post-tetanic decay formation is considered to be caused by competition of the two ions for channel entry. At present, however, we have no evidence in our preparation to determine which explanation is more likely, although the secondary effect of Mg*+ on transmitter release is stressed.8s’4 Mg*+, which is abundant in the cells, acts as a regulator of the action of Ca*+ in cellular functions.’ Mg*+ maintains transmitter release, i.e., synaptic activity as shown by the slowing of the decline in the post-tetanic MEPP frequency to form a shoulder on its decay curve. Recently it has been reported
883
that cytoplasmic Mg*+ at millimolar concentrations plays an important role in Ca*+-triggered exocytosis in permeabilized rat pheochromocytoma cells.23 Prolongation of the post-tetanic synaptic activity is equivalent to so-called short-term potentiation.24 Mg2+ is considered to contribute to short-term potentiation. Acknowledgements-We
are grateful to Kumazawa for preparation of the figures Miwako Ichiyasu for typing the manuscript. supported in part by Project Research Kawasaki Medical School.
Miss Satoko and to Miss The work was Grants from
REFERENCES 1. Almers W. and Palade P. T. (1981) Slow calcium and potassium currents across frog muscle membrane: measurements with a Vaseline-gap technique. J. Physiol., Land. 312, 159-176. 2. Asturias F. J. and Blasie J. K. (1989) Effect of Mg2+ concentration of Ca2+ uptake kinetics and structure of the sarcoplasmic reticulum membrane. Biophys. J. 55, 739-753. 3. Blaustein M. P. (1988) Calcium and synaptic function. In Handbook of Experimental Pharmacology, Vol. 83. Calcium in Drug Actions (ed. Baker P. F.), pp. 275-304. Springer, Berlin. 4. Caride A. J., Rega A. F. and Garrahan P. J. (1986) The reaction of Mg2+ with the Ca2+-ATPase from human red cell membranes and its modification by Ca2+. Biochim. biophys. Actn 863, 165-177. 5. de1 Castillo J. and Katz B. (1954) Statistical factors involved in neuromuscular facilitation and depression. J. Physiol., Lond. 124, 574-585. 6. Dodge F. A. Jr, Miledi R. and Rahamimoff R. (1969) Strontium and quanta1 release of transmitter at the neuromuscular junction. J. Physiol., Lond. 200, 267-283. 7. Francis L. P., Lennard R. and Singh J. (1989) Effect of magnesium on acetylcholine-evoked secretory responses and calcium mobilization in isolated rat pancreas. J. Physiol., Lond. 415, 94P. 8. Hurlbut W. P., Longenecker H. E. Jr. and Mauro A. (1971) Effects of calcium and magnesium on the frequency of miniature end-plate potentials during prolonged tetanization. J. Physiol., Lond. 219, 17-38. 9. Katz B. (1969) The Release of Neural Transmitter Substances. Charles C. Thomas, Springfield, IL.
10. Kita H., Narita K. and Van der Kloot W. (1980) Effects of temperature on the decline in miniature end-plate potential frequency following a tetanus. Brain Res. 190, 435-445. 11. Kita H., Narita K. and Van der Kloot W. (1981) Tetanic stimulation increases the frequency of miniature end-plate potentials at the frog neuromuscular junction in Mn2+-, Co2+- and Ni’+-saline solutions. Brain Res. 205, 11l-121. 12. Kita H., Narita K. and Van der Kloot W. (1982). The relation between tonicity and impulse-evoked transmitter release in the frog. J. Physiol., Land. 325, 213-222. 13. Klinger R., Wetzker R., Fleischer I. and Frunder H. (1980) Effect of calmodulin, Ca2+ and Mg2+ on the (Ca2+ + Mg2+)ATPase of erythrocyte membranes. Cell Calcium 1, 229-240. 14. Lev-Tov A. and Rahamimoff R. (1980) A study of tetanic and post-tetanic notentiation of miniature end-date potentials at the frog neuromuscular junction. J. Physiol., Lond. 309, 247-273. _ 15. McCleskev E. W. and Almers W. (1985) The Ca channel in skeletal muscle is a laree Proc. natn. Acad. Sci. U.S.A. _ Y oore. 1 I
82, 7149-7153.
16. Meiri U. and Rahamimoff R. (1971) Activation of transmitter release by strontium and calcium ions at the neuromuscular junction. J. Physiol., Land. 215, 709726. 17. Meiri H., Zelingher J. and Rahamimoff R. (1986) A possible involvement of the NaCa exchanger in regulation of transmitter release at the frog neuromuscular junction. In Calcium, Neuronal Function and Transmitter Release (eds
Rahamimoff R. and Katz B.), pp. 2399256. Martinus Nijhoff, Boston, MA. 18. Misler S., Falke L. and Martin S. (1987) Cation dependence of posttetanic potentiation of neuromuscular transmission. Am. J. Physiol. 252, C55C62. 19. Narita K. (1984) A device for measuring miniature end-plate potential frequency. Kawasaki Igakkai Shi 10, 62-69 [in
Japanese]. 20. Narita K., Kawasaki F. and Kita H. (1990) Mn and Mg influxes through Ca channels of motor nerve terminals are prevented by verapamil in frogs. Bruin Res. 510, 289-295. 21. Sanchez-Armass S. and Blaustein M. P. (1987) Role of sodium-calcium exchange in regulation of intracellular calcium in nerve terminals. Am. J. Physiol. 252, C595C603. 22. Schatzmann H. J. (1989) The calcium pump of the surface membrane and of the sarcoplasmic reticulum. A. Rev. Physiol. 51, 473-485. 23. Wegenhorst U., Gratzl M., Fohr K. J. and Ahnert-Hilger
G. (1989) Millimolar concentrations of free magnesium enhance exocytosis from permeabilized rat pheochromocytoma (PC 12) cells. Neurosci. Left. 106, 300-304. 24. Zucker R. S. (1989) Short-term synaptic plasticity. A. Rev. Neurosci. 12, 13-31. (Accepted 9 August 1990)
~e~~~~cie~ee Vol. 40, NO. 3, pp. 879483, 1991 Printed in Great Britain
EFFECTS OF DIVALE~ CATIONS ON THE TIME COURSE OF POST-TETANIC DECAY OF MINIATURE ENDPLATE POTENTIAL FREQUENCY IN FROGS K. NARITA and I-I. KITA* Department of Physiology, Kawasaki Medical School, Kurashiki, Okayama 701-01, Japan Ahstrnct-Relatively high external Mg’+ specifically induces a shoulder in the post-tetanic decay of miniature endplate potential frequency at frog neuromuscular junctions. This effect is antagonized by Ca’+, but not by SrZ+ and Mn 2+. This shoulder formation is not caused by Mn*+ or dinitrophenol in the medium. It is suggested that Mgz+ enters nerve terminals and displaces Ca”+ from internal stores; Mg+ subsequently interferes with Ca2+ removal until the MgZ+ itself has been removed. The dinitrophenol result suggests that ATP-dependent Ca*+ extrusion is not permanently slowed. An equally likely and simpler explanation is that Mg*+ enters nerve terminals and the Mg2+ itself causes an increase in miniature endplate potential frequency. External Ca’+ competes with Mg 2+ for channel entry and prevents this effect. Tbe time course of the decay of miniature endplate potential frequency reflects the processes involved in Mg’+ extrusion and/or uptake.
The frequency of miniature endplate potentials EXPERIMENTAL PROCEDURES (MEPPs) increases during tetanic stimulation of the Experiments were performed on sciatic nerve-sartorius motor nerve in Ca2+-containing saline solutions,5 muscle preparations dissected from the frog, Rnnu nigro~acu~ur~. Following dissection, the preparations were probably due to an increased Ca*+ con~ntration soaked for several hours in 0 CaZ+-Mg2+ EGTA solution within the motor nerve terminals owing to its influx (see below) at YC, after which they were mounted and during tetanic stimulation.’ In solutions lacking stretched to about 120% of their slack length in a chamber Ca2+ but containing Sr’+, Mg*‘, Mn2+, Co2+ or NiZC with a capacity of 5 ml. The nerve was placed on a pair of as the sole divalent cation, tetanic stimulation prosilver wire electrodes and covered with a mixture of Vaseline duces similar increases in the MEPP frequency.6~8~‘1~14 and paraffin oil. The basic saline solution used for the dissection conThis has been explained by postulating that the tained (mM): NaCl, 100, KCI, 2; CaCl,, 2.5; MgCl,, 3; divalent cations enter the terminal through Ca*+ Tris-maleate buffer (DH 7.4. 8. The 0 Ca’+-M$+ EGTA channels during stimulation and then either release solution contained ImM I&Cl, and 1 mM sycoletherdiaminetetra-acetic acid (EGTA), and the other constituents Ca2+ from intraterminal stores or mimic the action of Ca2f 11.12.20 were unchanged except that CaCl, was excluded. Solutions After the cessation of tetanic stimulation in solutions containing Ca2+, Sr2+, Mn2+ or Mg2+, the elevated frequency gradually falls toward the initial level, along a roughly exponential time course, presumably due to sequestration of Ca2+ by the Ca2+ stores and extrusion of these cations from the terminals.” If, as has been suggested, the MEPP frequency is dependent on the concentration of Ca2+ within the terminal, the time course of the decay of the MEPP frequency should reflect the decline of the elevated concentration of the divalent cation in the terminal. This time course is influenced by divalent cations in the bathing solution.‘0~‘8
containing either M$+, Sti+ or Mn*+ as the sole divalent cation did not include EGTA. When the concentrations of MgZf, Srz2+or Mn2+ were increased in the solutions containing Mg*+, Sr*+ or Mn’+ in place of Ca*+, the osmolarities of all solutions were maintained equal by adding sucrose. Neostigmine bromide ( 10m6g/ml) was added to all solutions for easier detection of MEPPs. The solution temperature was kept at 19-22°C. When solutions were changed, 50 ml of new solution was allowed to flow into the chamber at a rate of lOml/min. MEPPs were recorded with an jntra~llular glass microelectrode and counted with a special device” to follow the time course of the decline of the frequency aAer supramaximal tetanic stimulation (50 Hz, 2-3 min). RESULTS Eflect
*To whom correspondence should be addressed. Abbreuiulions: [Caz+],, intracellular free calcium ion concentration; DNP, 2+dinitrophenol; EGTA, glycoletherdiaminetetra-acetic acid; MEPP, miniature endplate potential; [Mg*+],, intracellular free magnesium ion concentration.
of Mg2+ on post-tetanic decay
When the M$+ concentration was 1 mM, the time course for the decline of the elevated MEPP frequency produced by tetanic stimulation was roughly exponential. Increasing the Mg2+ concentration slowed the decline (Fig. 1). In 6 mM or 11 mM, the 879
X80
K.
NAKITA
and H. KIT,%
&&ct
0.1 0
, 5
,
, 15
IO Tim*
of Sr’ ’ on
the shouldu
The antagonistic action produced by C‘a,” W;IS not found with Sr’+. The shoulder produced by IOmM Mg2+ remained almost unchanged when the concentration of Sr’+ was raised to 0.2 and 0.4 mM (Fig. 3). Since Sr2+ is known to be much less effective than Ca*+ in evoking transmitter release,‘,‘” one might argue that Sr2+ might have an antagonistic effect if higher concentrations were used. In our exper20
(mln)
Fig. 1. The effect of Mg*+ on the time course of the decline of the MEPP frequency after cessation of tetanic stimulation for 3 min at 50 Hz. The preparation was soaked for 7.5 h in a refrigerated 0 CaZ+-Mg*+ EGTA solution containing 30 mM sucrose. The preparation was then transferred to a new solution of the same content and soaked in it for 30 min until the solution reached room temperature. After the equilibration, MEPPs were recorded for IOmin. Then the nerve was stimulated tetanically and the recording was continued until 14 min elapsed after the end of the tetanus. Thereafter the concentration of Mg’+ was increased from 1 to 6mM and recording of the tetanus-recovery sequence was repeated. After 6 mM, a further 5 mM of MgCl, was added and the post-tetanic decay in the MEPP frequency was followed until 21 min after stimulation. Then the preparation was returned to the initial solution and MEPPs were recorded until 19min after the tetanus. During the course of the experiment, the osmolarity of the solution was kept constant by changing the concentration of added sucrose. This figure is a representative example of five experiments.
curves displayed a shoulder, indicating that the decline in the MEPP frequency was initially slow and then accelerated (Fig. 1). The slowing of the initial decay was characteristic of relatively high Mg2+ concentration. In other experiments, it was observed that at 3SmM Mg*+, the time course of decay was prolonged, though the shape of the curve remained the same, while at 5 mM
iments, however, Sr2+ at concentrations higher than 0.4 mM evoked endplate potentials, which precluded measurements of MEPP frequencies. In one successful experiment in which 0.8 mM Sr’+ was used. no antagonistic effect was observed. l@ect of Mn”
on the shoulder and post-tetanic
decay
The shoulder produced by relatively high Mg2+ was similarly unaffected by the addition of 0.2-l .OmM Mn2+ (Fig. 4). The post-tetanic decay in Mn2+ solutions followed a rough exponential curve with upward parallel shifts as the Mn2+- concentration was increased (Fig. 5). Mn2’ in the concentrations used did not play any role in the shoulder formation. It was shown previously that Mn2+, even at concentrations of 5, 7.5 and 10 mM, did not produce a shoulder.” Effect of dinitrophenol
on post -tetanic decay
The post-tetanic decay followed a rough exponential curve when the medium contained 0.5 mM Mn2+ as the only divalent cation (Fig. 6). The addition of 3 ,uM 2,4-dinitrophenol (DNP), which is known to decouple oxidative phosphorylation, had little
Mg2+, a shoulder was definitely present. Antagonistic formation
action of Mg2+ and Ca2+ on shoulder
Increasing concentrations of Ca2+, from 0.05 mM, made the shoulder on the post-tetanic decay curve of the MEPPs produced by 8mM Mg2+ less evident, and it finally disappeared at 0.2mM Ca’+(Fig. 2). Thus Ca2+ antagonized Mg2+ in the formation of the shoulder. To demonstrate the shoulder a little more quantitatively, the ratios of the post-tetanic MEPP frequency at 3 min to the frequency immediately after the end of tetanic stimulation were measured (Table I). These ratios approached one as the Mg2+ concentration was increased, showing the formation of the shoulder. In solutions containing 8 or 10mM MgZC and varying concentrations of Ca2+, the ratios fell to smaller values as the Ca2+ concentration was raised, showing the disappearance of the shoulder.
Time
IminI
Fig. 2. Antagonistic action of Ca*+ on shoulder formation. The preparation was soaked for 4 h in a refrigerated 0 Ca’+-M&+ EGTA solution containing 25 mM sucrose. It was then transferred and bathed for 30 min in 0 Ca2+-MgZ+ EGTA solution with 8 mM MgCl, and 1.5 mM sucrose at room temperature. After recording MEPPs for 10 min, the nerve was stimulated for 2min at 50Hz and the MEPP frequency was followed until 12min after the stimulation. Afterwards, the concentration of Ca’+ was increased to 0.05, 0.1 and 0.2 mM with corresponding decreases in the sucrose concentration to maintain the same osmolarity and recording of the tetanus-recovery sequence was repeated in each solution. This figure is a representative example of seven experiments.
881
Cations and post-tetanic decay Table 1. Ratio of post-tetanic miniature endplate potential frequency at 3 min to frequency immediately (at 0 min) after end of tetanic stimulation Experiment
Mg2+ concentration (mM)
Ratio (3 min/O min)
In Mg2+ solutions
1
1 6 11 1 (recovery)
2
I 3.5 6 11
3
1 3.5 6 8.5 11
Experiment
Ca2+ concentration (mW
In solutions containing Mgr+ and Ca*+ I(10 mM Mg*+) 0 0.05 0.1 0.2 0.3 2 (10 mM Mg2+) 0 0.05 0.1 0.2 3 (8 mM M$+) 0 0.05 0.1 0.2
0.28 0.26 0.74 0.20 0.29 0.36 0.58 0.79 0.23 0.30 0.45 0.49 0.64 Ratio (3 min/O min) 0.68 0.59 0.27 0.27 0.14 0.73 0.65 0.33 0.17 0.81 0.32 0.18 0.13
The experiments in this table are not those shown in Figs 1 and 2. or no effect on the time course of the decay. An
,,....l....,....,....
0
5
10 Tim*
15
$0
I
(min)
Fig. 3. Effect of Srr+ on the shoulder. The preparation was soaked for 3.5 h in a refrigerated 0 Ca*+-Mg2+ EGTA solution containing 24 mM sucrose.. It was then transferred and immersed for 30 mm in the same solution with 10 mM MgCI, and 1.5 mM sucrose at room temperature. After 10 mm recording of MEPPs, the nerve was stimulated for 2 min at 50 Hz and recording was carried out until 16 min after the tetanus. Thereafter the concentration of Sr2+ was raised to 0.2 and 0.4mM with corresponding decreases in the sucrose concentration to maintain the same osmolarity and recording of the tetanus-recovery sequence was repeated at each concentration of Sti+. This figure shows a representative example of nine experiments.
and when Mg *+ is the only divalent cation in the bathing solution, M$+ enters the terminal through the open channel in place of Ca2+.8~‘2~20 Although the dose-dependent tetanic rise in MEPP frequency in Mg2+ solutions indicates Mg2+ entry into the frog nerve terminal,8.‘4 the terminal is too small to determine directly whether its Ca*+ channels are permeable to M$+. Since the Ca*+ channels in frog skeletal muscle fibres show clearly measurable Mg2+ currents as well as Ba’+, Sr*+, Ca*+ and Mn*+
increase in the concentration of DNP to 5 PM, which was the highest concentration we could examine, shifted the decay curve upward and slowed the decay. The time course remained exponential and there was no shoulder (Fig. 6). DISCUSSION
The present work shows that the decline in the post-tetanic potentiation of the MEPP frequency was prolonged by relatively high M$+, which caused the formation of a shoulder on the decay curve (a convex curve). Among the divalent cations tested only Mg*+ was effective for the formation of a shoulder. It was also demonstrated that Mg*+ was antagonized by Ca*+ in this phenomenon, but not by Sr*+ and Mn*+. Furthermore, Mn*+ and a metabolic inhibitor, DNP, did not produce a shoulder, though the latter slowed the decay. It has been reported that the decay of post-tetanic potentiation slows when the Mg*+ concentration is elevated,14 but detailed analyses of the shape of the decay curve in the Mg2+ solutions have not been done. We believe that depolarization of the nerve terminal membrane caused by tetanic stimulation opens up the voltage-gated Ca*+ channels of the membrane,
c
p”
9 0.1-l
0
1
5
.
,
.
10 Tim*
I
15
.
,
I
20
(inin)
Fig. 4. Effect of MnZ+ on the shoulder. The preparation was soaked for 3.5 h in a refrigerated 0 Ca2+-Mg*+ EGTA solution containing 31 mM sucrose. It was then transferred and bathed for 1 h in the same solution with 8 mM MgCl, and 7.5 mM sucrose at room temperature. After 10min recording of MEPPs, the nerve was stimulated for 2 mm at 50 Hz and the recording was continued until 16 min after the stimulation. Thereafter the concentration of MnZ+ was increased from 0 to 0.2, 0.5 and 1 mM with corresponding concentrations of sucrose to maintain the osmolarity of the solutions and recording of the tetanus-recovery sequence was repeated in each solution. This figure is a representative example of six experiments in which 5, 8 or 10mM Mg2+ was used to produce the shoulder.
882
K. NARI,I’A and H.
Fig. 5. Effect of Mn2+ on post-tetanic decay in 0 Ca*+-MgZ+ EGTA solutions. The preparation was soaked for 4 h in a refrigerated 0 Ca*+-Mg’+ EGTA solution containing 15 mM sucrose and for 3 h in a fresh solution of the same composition at 20°C. After recording MEPPs for lOmin, the nerve was stimulated for 2 mm at 50 Hz and recording was done until 11 min after the tetanus. Thereafter the concentration of Mn*+ was increased to 0.2, 1 and 2.5 mM with corresponding decreases in the concentration of sucrose and recording of the tetanus-recovery sequence was repeated in these solutions. This figure shows a representative result from five experiments.
currents,‘~‘5 the existence of similar Ca*+ channels in the nerve terminal membrane is quite probable. The increased intraterminal concentration of free Mg”, [Mg2+li, could somehow increase the intraterminal concentration of free Ca”, [Ca’+],, probably by releasing Ca 2+ from intraterminal stores, the most probable candidates for which are the smooth endoplasmic reticulum and mitochondria.‘* After cessation of the tetanus, the increased [Ca2+ji gradually decreases to the resting level by both extrusion from the terminal and sequestration in the intraterminal Ca stores.3.‘0 The Ca2+ extrusion probably takes place mainly through the ATP-driven Ca*+ pump and Na2+-Ca2+ exchanger. The time course for the .
0.5mll
M”
. .
0.5mm 0.5mM
Y” Y”
+ 3* * 5.M
ONP
ow
5 t ;
O.Oli
0
.
.
I
I
10
5 Time
.
.
I
15
I
20
I
(main)
Fig. 6. Effect of DNP on post-tetanic decay in 0.5 mM Mn*+ solution. The preparation was soaked for 3.5 h in refiigerated 0 Ca*+-Mg*+ EGTA solution. It was then transferred and immersed for 30min in the same solution containing 0.5 mM Mn2+ at 20°C. After recording control MEPPs for 10 min, the nerve was tetanized for 2min at 50 Hz and MEPPs were recorded until 11 min after the tetanus. Thereafter 3 and 5 PM DNP were added and recording of the tetanus-recovery sequence was repeated in both solutions. This figure is a representative example of three experiments.
KITA
decay in [Ca”], near the release sites would be reflected in the time course of the decline of the post-tetanic MEPP frequency.“’ We believe that the shoulder on the decay curve of the post-tetanic MEPP frequency is formed in the following way. After tctanic stimulation, the increased [Mg2+], would lower the activity of the ATP-driven Ca2+ pump, because DNP prolonged the exponential decay of the post-tetanic MEPP frequency at 5 PM (Fig. 6). Suppression of the pump would result in the slowing of Ca’+ extrusion. ATP-driven Ca*+ transport systems, which include the surface membrane Ca*’ pump and the sarco(endo)plasmic reticulum Ca” pump, require for their functioning the presence of free Mg*+ on the &-side, the side from which Ca2+ moves.*,‘* However, the effect of Mg *+ is biphasic and a high Mg’+ concentration is inhibitory,4s’3 probably due to competition with Cal+ at the transport site.22 This finding supports the above-mentioned view regarding the role of relatively high Mg2’ in shoulder formation, Once the [Mg”], falls to a critical level, probably by extrusion and/or sequestration, extrusion of Cal’ would be accelerated by the normal functioning of the Ca’+ pump and the time course for the post-tetanic MEPP frequency would show a rapid fall following the shoulder. Since it has been reported that the Na+ concentration affects post-tetanic potentiation through the Na+--Ca’+ exchanger and the effect of Mg’+ on its activity is inhibitory,‘7,‘8.2’ the exchanger probably contributes to shoulder formation. However, our experiment, in which the effect of halving the Na’ concentration on the post-tetanic decay was examined, could not elucidate this point. The difference between Mg2+ and DNP in shoulder formation is considered to be as follows. DNP suppresses the ATP-driven pump at an almost constant rate with time, which produces a prolonged exponential decay curve. In contrast, the depression due to elevated [Mg2+], decreases with time after tetanic stimulation, so its depressive action on the pump is stronger initially, and then gradually becomes weaker. When the [Mg’+], falls below a certain critical level, the inhibitory action of Mg2+ is lost and the pump recovers its normal activity. This causes a shift in the post-tetanic MEPP frequency from an upper, slower exponential curve to a lower, faster one during an earlier phase of decay, resulting in the formation of a shoulder. A simpler, alternative explanation is that Mg2+, which enters nerve terminals during tetanic stimulation, acts directly on the release machinery to cause quanta1 release of transmitter. After the tetanus, the elevated [Mg2+li decays as the accumulated Mg’ + is extruded from the terminals at a rate that is initially slow, and then accelerates. The process of the clearing of Mg ‘+ is reflected in the time course of post-tetanic MEPP frequency showing a shoulder. The antagonin the shoulder ism between Mg2+ and Ca”
Cations and post-tetanic decay formation is considered to be caused by competition of the two ions for channel entry. At present, however, we have no evidence in our preparation to determine which explanation is more likely, although the secondary effect of Mg*+ on transmitter release is stressed.8s’4 Mg*+, which is abundant in the cells, acts as a regulator of the action of Ca*+ in cellular functions.’ Mg*+ maintains transmitter release, i.e., synaptic activity as shown by the slowing of the decline in the post-tetanic MEPP frequency to form a shoulder on its decay curve. Recently it has been reported
883
that cytoplasmic Mg*+ at millimolar concentrations plays an important role in Ca*+-triggered exocytosis in permeabilized rat pheochromocytoma cells.23 Prolongation of the post-tetanic synaptic activity is equivalent to so-called short-term potentiation.24 Mg2+ is considered to contribute to short-term potentiation. Acknowledgements-We
are grateful to Kumazawa for preparation of the figures Miwako Ichiyasu for typing the manuscript. supported in part by Project Research Kawasaki Medical School.
Miss Satoko and to Miss The work was Grants from
REFERENCES 1. Almers W. and Palade P. T. (1981) Slow calcium and potassium currents across frog muscle membrane: measurements with a Vaseline-gap technique. J. Physiol., Land. 312, 159-176. 2. Asturias F. J. and Blasie J. K. (1989) Effect of Mg2+ concentration of Ca2+ uptake kinetics and structure of the sarcoplasmic reticulum membrane. Biophys. J. 55, 739-753. 3. Blaustein M. P. (1988) Calcium and synaptic function. In Handbook of Experimental Pharmacology, Vol. 83. Calcium in Drug Actions (ed. Baker P. F.), pp. 275-304. Springer, Berlin. 4. Caride A. J., Rega A. F. and Garrahan P. J. (1986) The reaction of Mg2+ with the Ca2+-ATPase from human red cell membranes and its modification by Ca2+. Biochim. biophys. Actn 863, 165-177. 5. de1 Castillo J. and Katz B. (1954) Statistical factors involved in neuromuscular facilitation and depression. J. Physiol., Lond. 124, 574-585. 6. Dodge F. A. Jr, Miledi R. and Rahamimoff R. (1969) Strontium and quanta1 release of transmitter at the neuromuscular junction. J. Physiol., Lond. 200, 267-283. 7. Francis L. P., Lennard R. and Singh J. (1989) Effect of magnesium on acetylcholine-evoked secretory responses and calcium mobilization in isolated rat pancreas. J. Physiol., Lond. 415, 94P. 8. Hurlbut W. P., Longenecker H. E. Jr. and Mauro A. (1971) Effects of calcium and magnesium on the frequency of miniature end-plate potentials during prolonged tetanization. J. Physiol., Lond. 219, 17-38. 9. Katz B. (1969) The Release of Neural Transmitter Substances. Charles C. Thomas, Springfield, IL.
10. Kita H., Narita K. and Van der Kloot W. (1980) Effects of temperature on the decline in miniature end-plate potential frequency following a tetanus. Brain Res. 190, 435-445. 11. Kita H., Narita K. and Van der Kloot W. (1981) Tetanic stimulation increases the frequency of miniature end-plate potentials at the frog neuromuscular junction in Mn2+-, Co2+- and Ni’+-saline solutions. Brain Res. 205, 11l-121. 12. Kita H., Narita K. and Van der Kloot W. (1982). The relation between tonicity and impulse-evoked transmitter release in the frog. J. Physiol., Land. 325, 213-222. 13. Klinger R., Wetzker R., Fleischer I. and Frunder H. (1980) Effect of calmodulin, Ca2+ and Mg2+ on the (Ca2+ + Mg2+)ATPase of erythrocyte membranes. Cell Calcium 1, 229-240. 14. Lev-Tov A. and Rahamimoff R. (1980) A study of tetanic and post-tetanic notentiation of miniature end-date potentials at the frog neuromuscular junction. J. Physiol., Lond. 309, 247-273. _ 15. McCleskev E. W. and Almers W. (1985) The Ca channel in skeletal muscle is a laree Proc. natn. Acad. Sci. U.S.A. _ Y oore. 1 I
82, 7149-7153.
16. Meiri U. and Rahamimoff R. (1971) Activation of transmitter release by strontium and calcium ions at the neuromuscular junction. J. Physiol., Land. 215, 709726. 17. Meiri H., Zelingher J. and Rahamimoff R. (1986) A possible involvement of the NaCa exchanger in regulation of transmitter release at the frog neuromuscular junction. In Calcium, Neuronal Function and Transmitter Release (eds
Rahamimoff R. and Katz B.), pp. 2399256. Martinus Nijhoff, Boston, MA. 18. Misler S., Falke L. and Martin S. (1987) Cation dependence of posttetanic potentiation of neuromuscular transmission. Am. J. Physiol. 252, C55C62. 19. Narita K. (1984) A device for measuring miniature end-plate potential frequency. Kawasaki Igakkai Shi 10, 62-69 [in
Japanese]. 20. Narita K., Kawasaki F. and Kita H. (1990) Mn and Mg influxes through Ca channels of motor nerve terminals are prevented by verapamil in frogs. Bruin Res. 510, 289-295. 21. Sanchez-Armass S. and Blaustein M. P. (1987) Role of sodium-calcium exchange in regulation of intracellular calcium in nerve terminals. Am. J. Physiol. 252, C595C603. 22. Schatzmann H. J. (1989) The calcium pump of the surface membrane and of the sarcoplasmic reticulum. A. Rev. Physiol. 51, 473-485. 23. Wegenhorst U., Gratzl M., Fohr K. J. and Ahnert-Hilger
G. (1989) Millimolar concentrations of free magnesium enhance exocytosis from permeabilized rat pheochromocytoma (PC 12) cells. Neurosci. Left. 106, 300-304. 24. Zucker R. S. (1989) Short-term synaptic plasticity. A. Rev. Neurosci. 12, 13-31. (Accepted 9 August 1990)