Aging increases calcium influx at motor nerve terminal

hr. 1.

Devl. Neuroscience.

1990 Vol. 8. No. 6. pp.655-666.

0736-5748/!Xl53.00+0.00 Pergamon Press plc 0 1990ISDN

Printed in Great Britain.

AGING INCREASES CALCIUM INFLUX AT MOTOR NERVE TERMINAL WALEED B. ALSHUAIB* and MOHAMED A. FAHIM$ *Andrus Gerontology Center and Department of Biological Sciences, University of Southern California, Los Angeles. CA 90089-0191. U.S.A. and Physiology Departmens, Faculty of Medicine and Health Sciences. UAE University, P.O. Box 17666, Al Ain. United Arab Emirates (Received

17 October

1989; in revised form

19 January

1990; accepted 22 January

1990)

Abstract-To determine whether increased transmitter release from soleus nerve terminals of old C57BL/6J mice is caused by an altered Ca*+ regulation, the time course of post-tetanic potentiation of miniature endplate potential (MEPP) frequency was used as an indicator of the kinetics of Ca*+ metabolism in young (10 months) and old (24 months) mice. Post-tetanic potentiation properties were studied in either (1) 0.2 mM Ca*+, 5.0 mM Mg ‘+ Krebs; or (2) Ca*+-free/EGTA Krebs to eliminate Ca*+ influx, and thereby isolated Ca’+ buffering. In the 0.2 mM Ca*+ Krebs, the time constants of decay of augmentation (TA) and potentiation (Tr) were longer in old ( TA = 10.3 % 1.0 set, Tp = 195.3 ? 5.4 set) than in young (TA = 7.0 -C0.7 set, Tr = 78.8 2 6.6 set) nerve terminals. Evoked transmitter release was measured in 6.4 mM Ca*+, 2.75 mM Mg’+ Krebs. Quanta1 content of the endplate potential was positivelv correlated with T, (r=O.95) and with T, (r=O.98). In the Ca’+-free/EGTA Krebs. there was no difference in post-tetanic potentiation properties between young and old terminals. These results suggest that Ca*+ influx into the soleus nerve terminal increases with aging. This may explain, at least in part, the increased quanta1 content observed at old terminals. Key words:

neuromuscular

transmission,

aging, post-tetanic

potentiation,

Ca homeostasis,

synaptic

plasticity.

The neuromuscular junction undergoes a lifelong continual process of remodelling in structure and However, there are two striking but contradictory changes that occur during aging at function. 2,3~13,39 nerve terminals of soleus and extensor digitorum longus muscles of the mouse. While the number of synaptic vesicles (quanta) in the nerve terminal decreases,6*13,14the mean number of acetylcholine (ACh) quanta liberated by the nerve impulse (quanta1 content, m) increases.‘.2.4.“S’9The quanta1 hypothesis states that m = np, where n is the number of quanta in the nerve terminal, and p is the average probability of a quantum being released. It has been shown previously that increased m is caused by an increase in p.43 Consequently, p may be greater in old than in young terminals. Perhaps, intraterminal free Ca*+ plays a role in modulating p, since Ca*+ has been implicated as an essential trigger of transmitter release. 32 Most probably, the action potential causes an increase in the conductance of Ca*+ and Ca*+ flows into the nerve terminal.38 Ca*+ is believed to combine with a specific site X inside the terminal forming Ca*+X (active complex). The number of ACh packets released is proportional to (Ca*+X)“, where s is the slope of the double logarithmic relationship between the endplate potential (EPP) and low [Ca2+].9 In view of the importance of Ca*+ to the process of transmitter release, the goal of the present study was to determine whether there is an age-related change in Ca*+ regulation in the mouse nerve terminal. The soleus muscle was selected for study, since it is believed that the increase of soleus quanta1 content is a ‘fundamental’ age change rather than a change associated with inactivity of advancing age.* It is plausible that intraterminal Ca*+ is regulated at a higher concentration in old than in young soleus terminals. 23,31.41 This high regulation of Ca*+ could be caused by either of two mechanisms: (a) The number or properties of Ca*+ channels in the pre-synaptic terminal membrane may be altered with aging, such that Ca*+ influx is increased during the action potential; or (b) Ca*+ clearance within the terminal may be reduced as a result of altered number or properties of Ca *+ binding sites (e.g. mitochondria, endoplasmic reticulum), or as a result of altered Ca*+ extrusion by pre-synaptic membrane transport systems (e.g. Na+-Ca*+ exchanger, ATP-dependent Ca*+ pump), *Author to whom correspondence should be addressed. PTP. post-tetanic potentiation; R, resting; T. tetanic; PT. post-tetanic; MEPP. miniature endplate potential; EEP. endplate potential, T,, time constant of decay of augmentation; T,, time constant of decay of potentiation; TAY,time constant of decay of augmentation in young; T,,, time constant of decay of augmentation in old; Tr,, time constant of decay of potentiation in young; Tro, time constant of decay of potentiation in old. Abbreviations:

655

656

W. B. Alshuaib and M. A. Fahim

It is possible to investigate Ca2+ regulation in the nerve terminal by studying post-tetamc potentiation (PTP) of transmitter release, since the time constants of decay of augmentation and potentiation provide an indication of the kinetics of Ca*+ metabolism in the terminal.“” “’ .4r low quanta1 contents, high frequency stimulation of the motor nerve produces potentiation rn EPP amplitude’7,2R and MEPP frequency. ‘“vJ4The increased EPP amplitude consists of three phases. facilitation, augmentation and potentiation.27.-2y The increased MEPP frequency consist\ &’ at least two phases: augmentation and potentiation. “‘,44Both the EPP and MEPP are generally thought to be controlled by the same release mechanism. ” Electrophoretic application of C’a’+ into the nerve terminal causes an increase in MEPP frequency.3’ Thus it is presumed that high frequency stimulation of the motor nerve produces a great increase in the concentration cot intraterminal free Ca’+. and leads to an increase in the probability of release which appears a:, ,tn increase of tetanic MEPP frequency. The kinetics of PTP decay (T+,, T,,) represent the removal of Ca’+ from the terminal.24,‘h MEPP frequency potentiation makes it possible (using a Ca*+-free/EGTA Krebs solution) to eliminate Ca2+ influx during a tetanus. “‘,“.” Therefore, we used MEPP frequency potentiation to determine whether Ca’+ influx at the soleus prc-scrtaptic terminal membrane, and/or Ca*+ buffering within the terminal is altered during aging. A preliminary report of some of this work has appeared in abstract form.

EXPERIMENTAL

PROCEDURES

Materials

Methoxyflurane (Metofane) was purchased from Pitman-Moore Inc. (Washington Crossing, NJ, U.S.A.). and ethyleneglycol-bis-(B-amino-ethyl ether)-N,N’-tetra-acetic acid (EGTA) was purchased from the Sigma Chemical Company (St Louis, MO, U.S.A.). Animals

Experiments were performed on young adult (10 months) or old (24 months) male C57BL/6J mice. These mice were initially purchased as post-breeders (6 months of age) from Jackson Laboratories. They were randomly divided and raised in a university vivarium. There were 20 animals in each age group, and the mean body weights of the selected mice in the two groups were not significantly different. The C57BU6J strain was chosen because we have previously’ reported increased transmitter release during aging for this strain. In addition, this strain is widely used in aging studies. Animals were housed in groups of two in standard (28 X 14 X 12 cm) mouse cages, exposed to 12-hr light/dark cycles at 25°C and allowed food (Wayne Lab Blocks, Colton, CA, U.S.A.) and water ad Zibitum. Animals with gross organ pathology (large tumors, abscesses or ankylosed hindlimb joints) were excluded. Surgical procedure

and preparation

Animals were anaesthetized with methoxyflurane (Pitman-Moore). The soleus muscle was excised and pinned in a lucite chamber containing a low Ca*+lhigh Mg*+ Krebs solution (pH 7.2) at 23-25°C for quanta1 content experiments. Because transmitter release is affected by muscle stretch, the excised muscle was pinned at 1.1 x the rest length for all experiments. The solution was oxygenated (95% 02, 5% CO*) by a gas lifting device, which circulates freshly oxygenated solution, at a rate of 10-15 ml per minute, without agitating the recording chamber. Electrophysiological

recording

Glass capillary micro-electrodes filled with 3 M KCI and drawn to a tip yielding 8-15 megaohm resistance (measured in solution) were inserted into muscle fibers at the endplate region to record MEPPs and EPPs. A combination of oblique and trans-illumination, in conjunction with a Leitz-Wild microscope, were used to locate endplate regions. In addition, a MEPP rise time of less than 1 msec was used as a criterion for endplate regions. Three consecutive experiments were performed on the same endplate. First, quanta1 content; second, PTP in a Ca2+-containing Krebs solution; third, PTP in a Ca 2+ free/EGTA Krebs solution.

Calcium influx during aging

657

Quanta1 content

To determine the quanta1 content of the EPP, lOCL200 MEPPs and W-100 EPPs were recorded. MEPPs and EPPs were digitized on-line via an analog-to-digital conversion board (10 kHz sampling rate) in a Northstar Horizon microcomputer. A partial presynaptic block was administered. The Krebs solution with low Ca2+/high Mg 2+ had the following composition (in mM): NaCl(l12), KC1 (4.5), Ca gluconate (0.4), MgSO, (2.75), Na2HP0, (l), NaHC03 (15) and glucose (11). This 0.4 mM Ca2+ 2.75 mM Mg 2+ Krebs solution suppresses evoked release and the potential generated in the muscle fiber so that a regenerative action potential does not occur (hence, no muscle contraction), and an EPP can be recorded.’ Recordings were obtained from muscle fibers with resting membrane potentials more negative than -70 mV. ‘Giant’ MEPPs (those of an amplitude greater than twice the mean) were not included when calculating the mean MEPP amplitude or frequency. For EPP recording, the nerve supplying the muscle was stimulated via a suction electrode using a Grass S-44 stimulator and stimulus isolation unit. Supramaximal stimuli of 0.1 msec duration were delivered at a frequency of 0.5 Hz. No facilitation or depression of the EPP was seen at this stimulus frequency. Both the MEPP and EPP amplitudes were corrected to a standard RMP of -80 mV. However, EPP amplitude was not corrected for non-linear summation of the unit potentials which comprise the EPP. Non-linear summation results from the non-linear relationship between membrane conductance and potential changes at the endplate. EPP amplitudes were typically less than 5 mV, and at this level the correction factor for non-linear summation is negligible. Quanta1 content was calculated by the direct method (corrected mean EPP amplitude/corrected mean MEPP amplitude). PTP in Ca’+ containing Krebs

For repetitive nerve stimulation, a very low Ca’+/high Mg 2+ Krebs solution is commonly used to sufficiently reduce evoked release during stimulation, and thereby prevent synaptic depression. Similar to earlier experiments,’ a Krebs solution of 0.2 mM Ca2+/5.0 mM Mg2+ was used here. Since intracellular concentration of free Ca2+ is presumably less than lo-’ M, even at the low external concentration of 2 x lo-’ M Ca2+, an inward electrochemical gradient for Ca2+ will exist, leading to Ca2+ entry during the action potential. The Krebs solution of quanta1 content experiments was replaced by the 0.2 mM Ca2+ 5.0 mM Mg 2+ Krebs solution using a method of continuous drops (drops of new solution in, drops of old solution out). During the exchange of solutions, care was taken so that the micro-electrode remained inside the muscle fiber, at the same endplate. Five times the bath volume were used in the exchange of solutions. A chart recorder was used to record MEPPs. After a period of 20 min in the new Krebs solution, spontaneous resting (R) MEPPs were recorded for 120 sec. A tetanus (supramaximal stimuli, 0.1 msec duration) of 50 Hz for 40 set was delivered to the nerve, during which tetanic (T) MEPPs were recorded. Immediately after the tetanus, post-tetanic (PT) MEPPs were recorded for 500 sec. PTP in Ca’+-freelEGTA

Krebs

To determine any difference in Ca 2+ buffering (release and clearance) within the terminal between young and old mice, PTP experiments were performed in a Ca’+-free/EGTA Krebs solution (0 mM Ca2+, 2 mM Mg 2+, 1 mM EGTA). EGTA was used in preference to EDTA, because the binding constant for Mg 2+ by EGTA is much less than that by EDTA. The 0.2 mM Ca2+ 5.0 mM Mg2+ Krebs solution was replaced by the Ca 2+-free/EGTA Krebs solution using the same method of continuous drops. The [Ca’+],, in such EGTA Krebs solution is estimated to be less than lo-“ M.” Under these conditions, a reversed electrochemical gradient (from inside to outside the terminal) for Ca 2+ is expected to exist during the action potential. One hour equilibration in the EGTA Krebs solution was used in the present study to insure that there was no significant Ca2+ influx during the tetanus.“‘.2J.“h Using the same procedures as before, resting MEPPs were recorded, the tetanus was delivered to the nerve, and tetanic and post-tetanic MEPPs were recorded. Time course of MEPP frequency

To compute MEPP frequency (R, T and PT) as a function of time, a moving-interval technique was used to analyze the records of MEPPs. This technique computes MEPP frequency within a

658

W. B. Alshuaib and M. A. Fahim

small time interval moving along the time axis24 (i.e. a frequency value is computed every At set). The larger the interval, the greater the accuracy, since MEPPs are averaged over a longer time. The larger the At, the smaller the resolution, since a frequency value is computed every At, For the tetanic and post-tetanic MEPP frequency records, an interval size of 10 set and a Al ol 2 set were considered optimum for the desired accuracy and resolution. The complete MEPP frequency record was normalized by the pre-tetanus resting frequency to rule out any differences due to absolute frequencies of nerve terminals. Hereafter, normalized MEPP frequency 1s referred to as MEPP frequency. Statistical analysis

For a given property, one value was obtained from each mouse (one nerve terminal per mouse). These values from individual mice, in each age group, were used to calculate the mean for that age group. Data are presented as mean +. S.D., the means of the two age groups were compared using a two-tailed Student’s t-test for independent samples. To determine whether an observed correlation is real, the test statistic t = rb’n-2/d/1--? was used (r is the correlation coefficient and n is the sample size). A significance level of 0.01 (99% confidence) was used for all statistical tests. Regression analysis

Linear regression was performed on semi-logarithmic plots of normalized post-tetanic MEPP frequency against time to obtain a straight line of decay of MEPP frequency. The method of leastsquares was used to obtain the best-fitting line for data points. The line determined by this method is such that the sum of the squares of deviations of data points about the line is a minimum. The line of decay of MEPP frequency was used to calculate the time constant of decay of MEPP frequency (time required to reach l/e of the maximum initial value).

RESULTS Quanta1 content

For each nerve terminal, quanta1 content and PTP properties were investigated to determine the relationship between the two. Quanta1 content was greater (P
659

Calcium influx during aging

-(

a) YOUNG

(b)

OLD

PT PT B

H

0 0

100

0 200

300

TIME

400

600

600

700

0

100

200

(set)

300

400

TIME

(set)

600

600

Fig. 1. Changes in normalized MEPP frequency during and after a tetanus (50 Hz. 40 set) in young (a) and old (b) soleus nerve terminals. Krebs solution contained 0.2 mM Ca’+, 5.0 mM Mg”+. Interruption of the time axis is due to the moving-interval technique which was performed to determine the time course of MEPP frequency. A frequency value was computed every 2 sec. from MEPPs in a IO set-interval moving along the time axis. The MEPP frequency was normalized by dividing by the average resting(R) frequency in the 120 set preceding the tetanus. (a): Young nerve terminal. 2 seconds before end of tetanus, normalized tetanic frequency (T) is 24.8; 2 set after end of tetanus, normalized post-tetanic frequency (PT) is 14.4, it decays back to the resting level after about 176 set post-tetanus. Quanta1 content at this nerve terminal was 0.44. (b) Old nerve terminal. 2 seconds before end of tetanus, normalized tetanic frequency is 28.7; 2 set after end of tetanus, normalized post-tetanic frequency is 15.4, it decays back to the resting level after about 468 set post-tetanus. Quanta1 content at this nerve terminal was 0.62. Table

I.

Post-tetanic potentiation properties of soleus nerve terminals in a Cal+-containing

Property Tetanic frequency Post-tetanic frequency TA (set) TP (=) Time to rest (set)

Young 21.l-c2.0 12.9 + 0.9 7.0*0.7 78.826.6 168.5 + 7.0

Old 26.8+ 1.0” 15.6?0.5* 10.3 + I .o* 195.3 + 5.4* 438.5 + 20.8’

young Krebs

and

old

% Change t 27 t 20 7 51 t 133 t

160

The Krebs solution contained 0.2 mM Ca2+, 5.0 mM Mg”+. A tetanus of 50 Hz, 40 set was used in all experiments. A movinginterval technique (interval size = IO set, At = 2 set) was performed on MEPP records to determine the time course of MEPP frequency. which was then normalized by the resting frequency. Tetanic frequency (T) is the MEPP frequency reached 2 set before the end of tetanus. Post-tetanic frequency (PT) is the MEPP frequency reached 2 set after the end of tetanus. One soleus nerve terminal per mouse was studied, and there were 20 mice in each age group. All values are means * S.D.; the means were compared using a two-tailed independent r-test. *Indicates significant difference (P
Time constants. For both young and old nerve terminals, semi-logarithmic plots of normalized post-tetanic MEPP frequency against time revealed two time courses of MEPP frequency decay. The first phase ends at 14 set post-tetanus, and it represents the overlap of augmentation and potentiation. It is not known how augmentation and potentiation are related during the overlap. However, similar to previous studies,“‘.2x an additive relationship between augmentation and potentiation was assumed in the present study to deduce the time course of augmentation. This means that both augmentation and potentiation begin at the end of tetanus (I= 0 set), augmentation ends before t= 14 set while potentiation continues for hundreds of seconds. Least-squares regression was used to determine the slopes of the augmentation and potentiation decay lines. The augmentation phase was much shorter than the potentiation phase (see Table l), for this reason, the augmentation phase was expanded in a separate figure, so that its time constant of

1

660

W.

Alshuaib and M. A. Fahim

B.

0

2

4

6

TIME

8

10

12

$4

bed

Fig. 2. Decay of the augmentation phase of the normalized post-tetanic MEPP frequency that was shown in Fig. 1 (Ca’+ Krebs). for young and old soleus nerve terminals. Semi-logarithmic plot. End of tetanus: r = 0. Least-squares regression was performed on points to determine the decay fines. The time constant of decay of augmentation for the young nerve terminal (TAY) was 7.8 set, and for the old nerve terminal (T,,) it was II .S sec. Note that these lines represent the MEPP frequency increase, above resting frequency, that augmentation provides. In the young nerve terminal, this contribution ends at 13.5 see post-tetanus. and in the old nerve terminal it ends at 13.2 set post-tetanus.

* - YOUNG *-OLD

Tpy = 87 set TPO = 203 set

--T---

560 TIME

WC)

Fig. 3. Decay of the potentiation phase of the normalized post-tetanic MEPP frequency that was shown in Fig. I (Ca’+ Krebs), for young and old nerve terminals. Semi-logarithmic plot. End of tetanus: I = 0. From I = 0 set to I = 14 set, augmentation and potentiation overiapped. Least-squares regression was performed on points beginning at I= I4 set to determine the potentiation decay lines. An additive relationship was assumed between augmentation and the extrapolation of the potentiation line to the ordinate in order to deduce the augmentation decay lines (Fig. 2). The time constant of decay of potentiation for the young nerve terminal (TV+.) was 87 set, and for the old nerve terminal ( T,c,) it was 203 sec. Note that these lines represent the MEPP frequency increase. above resting frequency, that potentiation provides. In the young nerve terminal. this contribution endsat 186secpost-tetanus; and in the old nerve terminal it ends at 4X4 set post-tetanus

decay could be readily calculated. The decay of this fast component of PTP, augmentation, is shown in Fig. 2, for the same young and old nerve terminals of Fig. 1. The time constant of the decay of augmentation in the young terminal, T,,. was 7.8 set, and in the old nerve terminal, TAO, it was 11.8 sec. The decay of the slow com~nent of PTP, ~tentiation, is shown in Fig. 3, for the same young and old nerve terminals of Fig. 1. The time constant of the decay of potentiation in the young nerve terminal, Tpy, was 87.0 set, and in the old nerve terminal, Tpo, it was 203.0 sec. Table 1 summarizes the PTP results obtained from 20 young and 20 old animals in the Ca*+ Krebs solution. Young nerve terminals were significantly different (PcO.01) from old nerve

Calcium

influx

661

during aging

terminals in tetanic frequency, post-tetanic frequency, the time constants of decay of augmentation and potentiation, and the time for MEPP frequency to return to the resting level (time to rest). Relationship between quanta1 content and time constants. To determine whether the level of free intraterminal Caz+ (indicated by the length of PTP decay) is involved in the mechanism of greater release at old soleus terminals, data from both young and old nerve terminals were used to calculate correlation coefficients between quanta1 content and each of the time constants of augmentation and potentiation. Quanta1 content was positively correlated with the time constant of decay of au~entation (r = 0.95, significant at P = 0.01). Quanta1 content was also positively correlated with the time constant of decay of potentiation (r = 0.96, significant at P = 0.01). Furthermore, to determine whether there is a relationship between the augmentation and the potentiation phases, the correlation coefficient between their time constants of decay was calculated. The time constant of decay of augmentation was positively correlated with the time constant of decay of potentiation (r = 0.94, significant at P= 0.01). Periodic oscillations. Periodic oscillations were observed in post-tetanic MEPP frequency, they presumably reflect changes in intraterminal free Ca2+ level.37 The period of oscillation was greater (P
Krebs

PTP experiments were performed in Ca’+- free/EGTA Krebs solution (called hereafter EGTA Krebs solution) to eliminate Ca2+ influx, and thereby isolate Ca2+ buffering.‘0*35*36Typical examples of the changes in MEPP frequency during and after tetanic stimulation (50 Hz, 40 set) in the EGTA Krebs solution are shown in Fig. 4 for the same young and old nerve terminals of Fig. 3.5

55

3.0

a” E

2.5

ii

2.0

Y

1.5

cl

W E ii

1.0

$

0.5

z

0

100

TIME

200

WC)

300

O

100

TIME

200

300

WC)

Fig. 4. Changes in normalized MEPP frequency during and after a tetanus (50 Hz. 40 set) in the same young and old nerve terminals of Fig. 1, but after replacing the solution by Ca?+ free/EGTA Krebs solution (0 mM Ca’+, 2.0 mM Mg *+, 1.OmM EGTA). interruption of the time axis is due to the movinginterval technique which was performed to determine the time course of MEPP frequency. A frequency value was computed every 2 set from MEPPs in a 10 set-interval moving along the time axis. The MEPP frequency was normalized by dividing by the average resting (R) MEPP frequency in the 120 set preceding the tetanus. (a) Young nerve terminal. Normalized tetanic (T) frequency is 2.9 during the first 2 set of tetanus. 2 set after end of tetanus, normalized post-tetanic (PT) frequency is 1.7and it decays back to the resting level after about 58 set post-tetanus. (b) Old nerve terminal. Normalized tetanic frequency is 3.3 during the first 2 set of tetanus. 2 set after end of tetanus, normalized post-tetanic frequency is 1.5 and it decays back to the resting level after about 62 set post-tetanus. MI 8:6-C

662

W. B. Alshuaib

and M. A. Fahim

set set

!

(

0.50$-

20

40

80

60

TIME

loo

120

I40

tsec)

Fig. 5. Decay of potentation of the normalized post-tetanic MEPP frequency that was shown in Fig. 4 (EGTA Krebs), for young and old nerve terminals. Semi-logarithmic plot. End of tetanus: f= 0. There was no augmentation phase in the decay. Least-squares regression was performed on normalized posttetanic frequency to determine the decay lines. The time constant of decay of ~tent~ation for the young nerve terminal (Tpy) was 112 set, and for the old nerve terminal (‘Fro) it was 98 sec. i3ecause of considerable overlap between young and old data points, they were not shown, for clarity.

1. After the tetanus, no immediate drop was observed in MEPP frequency. For both young and old nerve terminals, semi-logarithmic plots of normafized post-tetanic MEPP frequency against time revealed only one time course of MEPP frequency decay, potentiation. Least-squares regression was used to determine the slope of the potentiation decay line. The decay of potentiation is shown in Fig. 5 for the same young and old nerve terminals of Fig. 4. The time constant of the decay of potentiation in the young nerve termina1, Tpv, was 112.0 set and in the old nerve termmal, Tpo, it was 98.0 sec. Due to occasiona failures to maintain the micro-electrode inside the muscle fibers, at the same nerve terminal, it was possible to obtain recordings from only 17 out of the 20 animals in each group. Table 2 summarizes the PTP results obtained from young and old animals in the EGTA Krebs solution. There was no significant difference between young and old terminals in tetanic MEPP frequency, post-tetanic MEPP frequency, time constant of decay of potentiation, or time to rest. It should be noted that, in the EGTA Krebs. normalized tetanic Table 2. Post-tetanic potentiation properties of young and old soleus nerve terminals in a Ca’+-free/EGTA Krebs Property

Young

Old

Tetanic frequency Post-tetanic frequency

2.9-tO.3 1.9C0.5 109.5 + 10.6 60.7 -r 11.8

3. I 10.3

TP(set) Time to rest (set)

.-

I .h +0.-l 104.4+‘).1 69.12 13.6

The Krebs solution was free of Ca’+ and contained 2 mM Mg’+ and 1 mM EGTA. A tetanus of SO Hz, 40 set was used in all experiments. An interval of IO set and Arof 2 set were used :n the moving-interval technique. Tetanic freauencv CT) is the MEPP frequency reached 2 set before the’end’of tetanus. Post-tetanic frequency (PT) is the MEPP frequency reached 2 set after end of tetanus. One soleus nerve terminat (same nerve terminal of Table 1) per mouse was studied, and results were obtained from I7 mice in each age group. All values are means 2 S. D. ; the means were compared using a two-tailed independent r-test. For all properties shown, there was no significant difference between young and old nerve terminals.

Calcium influx during aging

667

MEPP frequency was only elevated to 2.920.3 (young) and 3.1 + 0.3 (old) which c~)tlIrA\ normalized tetanic MEPP frequency in the Ca2+ Krebs (young, 21.1 + 2.0; old, 26.X + I .O) fc.g. compare Figs 1 and 4).

DISCUSSION Tetanus-induced

changes in transmitter release

In the present study, the increase in MEPP frequency after repetitive nerve stimulation wa\ taken as an indication of the voltage-dependent increase in intraterminal free Ca2+ level.*” The decay of post-tetanic frequency was considered an indication of the removal of this residual Ca2+ from the terminal. In the presence of external Ca2+, the time constants of augmentation and potentiation are similar to those reported for EPP potentiation.2s The positive correlation between the time constant of augmentation and the time constant of potentiation suggests a coupling between the two processes responsible for augmentation and potentiation. Differences

between inward and reversed Ca2+ gradients

Tetanic and post-tetanic MEPP frequencies in the Ca ‘+ Krebs differed from those in the EGTA Krebs in several respects: (1) In the Ca2+ Krebs, tetanic frequency increased during the progression of the tetanus, while in the EGTA Krebs, it fluctuated towards smaller values during the tetanus. Ca2+ that causes transmitter release in the EGTA Krebs is believed to be from intracellular sources, such as mitochondria, endoplasmic reticulum and sites on the internal surface of the membrane.‘6,44 Coupling between nerve impulse activity and Ca2+ release from intracellular sources is not completely understood. However, release of Ca2+ from mitochondria has been suggested to be caused by Na+ entry.24,‘5 In the EGTA Krebs, the decrease in MEPP frequency during the tetanus could be due to the efflux of Ca2+ (reversed Ca2+ gradient) through voltage-dependent Ca2+ channels.5 (2) Tetanic and post-tetanic MEPP frequencies in the EGTA Krebs solution (Fig. 4, Table 2) were much smaller (about 10 fold) than those in the Ca2+ Krebs solution (Fig. 1, Table 1). This seems to be typical of PTP in the mammalian soleus nerve terminal, since tetanic frequency in the rat soleus terminal was reduced by 90% when Ca 2+ Krebs was replaced by EGTA Krebs.35 In the Ca2+ Krebs, the increase of intraterminal free Ca2+ is presumably much greater than that in the EGTA Krebs, this manifests the difference in tetanic and post-tetanic MEPP frequencies between the two solutions.35,44 (3) Facilitation and augmentation were present in the Ca2+ Krebs but not in the EGTA Krebs. This indicates that facilitation and augmentation are essential components only in high states of transmitter release (e.g. in the presence of Ca2+ influx). Absence of a post-tetanic jump

For the frog nerve terminal, in EGTA Ringer solution, post-tetanic MEPP frequency immediately after the tetanus was greater than end tetanic frequency (i.e. a post-tetanic jump).” It was suggested, that after the end of tetanus, Ca2+ efflux stops and Ca2+ release from intracellular sources leads to the jump in frequency. No post-tetanic jump was observed in any of the present EGTA experiments. This could be explained by: (a) Ca2+ release from intracellular sources is believed to be small in mammalian nerve terminals, such that most free calcium ions leave the terminal through open Ca2+ channels; and (b) Ca2+ release from intracellular sources may stop immediately after the end of the tetanus. Differences

between young and old terminals

Calcium Krebs. In the Ca2+ Krebs solution, young and old nerve terminals were compared with

regard to Ca*+ influx during the tetanus and the ensuing removal of accumulated Ca2+ from the terminal. All properties (Table 1) that were considered indicative of Ca2+ accumulation (tetanic frequency, post-tetanic frequency, TA, TP, and time to rest) increased during aging. Therefore, the tetanus led to greater free [Ca’+]i in old than in young terminals, possibly increasing the average probability of a quantum being released in old terminals.2*4.6 Based only on Ca2+ Krebs

664

W. B. Alshuaib and M. A. Fahim

PTP experiments, the longer TA and Tp in old than in young terminals reflect increased C:a” influx and/or reduced Ca2+ clearance during aging. On the other hand. it is unlikely that the difference in PTP properties between young and old terminals (Table I ) is due to increased scns,rtivity of the release process to Ca’+, because in both young and old soleux nerve tertnin~~l~. the cooperative action of about 3 Cal+ 1s necessary for the release of one quantum * EGTA Krebs. In the EGTA Krebs solution, Ca’+ influx during thr tetanus was cii~nil~iltcd to test whether CaTf buffering (release, clearance) within the terminal is different” in young aivd old terminals. Since there was no significant difference between young and old terminals in Wanic frequency, post-tetanic frequency, Tp, or time to rest (Table 2). it can be concluded that old terminals do not seem to differ from young terminals in: (a) amount of Ca2+ released from internal sources; (b) Ca’+ clearance kinetics in response to an increase of C‘:P from internal sources. These kinetics include Ca’+ sequestering by intraterminal orpanelles and proteins. the Na”-Ca2+ exchanger (Naf in-Ca’+ out), and the ATP-dependent Ca’+ pump that extrude3 (Ya.31 out of the terminal; or (c) the sensitivity of the transmitter release process IO Ca’+. Calcium homeosrasis during normative uging

Although young and old terminals did not differ in Ca” clearance kinetics in response to internally released Ca’+, they could possibly differ in Ca’+ clearance kinetics in response to Ca’+ influx. However, even if a difference in Ca2+ clearance ex’isted, it is unlikely to lead to such large differences in PTP properties (Table 1) between young and old terminals, because Ca’+ clearance properties (Table 2) were basically the same in the EGTA Krebs. Apparently, the reduced number of mitochondria” in old terminals did not induce a difference in Ca2+ buffering between young and old terminals (EGTA Krebs experiments). Mitochondria serve as Ca2+ sequestering organelles when free cytosolic CaZf concentration is above 0.4 p.M.16 However, it is possible that mitochondrial uptake is physiologically unimportant in the maintenance of intracellular Ca*+ because of its unfavorable kinetic properties.” In the rat soleus nerve terminal, Ca2+ clearance does not change during aging.” Furthermore, the important role of the Ca2+ channel in the release process is well established.“.-” Increased Ca*+ influx could be caused by an increase in the number of Ca2+ channels, a change in their kinetics, or an impairment of the inactivation of Ca”+ channels during aging. Indeed, corticosteroids have been suggested to modulate voltagedependent Ca2+ conductance in hippocampal neurons during aging.2’ Hence, it is suggested that the difference in PTP properties, when Ca2+ is present externally, is due mainly to a difference in Ca2+ influx; a difference in Ca*+ clearance may or may not exist. Despite the presumed change in Ca”+ channels (increased Ca*’ influx), Ca 2+ efflux in the EGTA Krebs did not seem to change during aging (see Fig. 4 and Table 2). This may be explained by the great difference in the Ca’+ concentration gradient between the influx condition (approximately, 2 :x:IO’+ M/10-’ M .= 2000 fold) and the efflux condition (approximately, loo7 M/10e9 M = 100 fold). Recent evidence implicated altered Ca’” homeostasis in brain aging. C’a’+-dependent K’ conductance (after hyperpolarization) and Ca’+ action potentials arc prolonged in aged rat 23 Altered Ca2+ homeostasis can cause changes in neuronal function during hippocampal neurons. 1h%2” The present study demonstrated that altered Ca”+ homeostasis could modulate aging. Ca*+ homeostasis was assessed using the indirect method of synaptic function during aging.’ MEPP frequency potentiation. To understand Ca2+ homeostasis more completely. direct methods are required. Unfortunately, limited by the present knowledge, it is exceedingly difficult to precisely identify the changes in the different components of the Ca”+ regulating system in motor nerve terminals. Further information is needed about each of the foitowing components: the voltage-dependent Ca2+ channel, the mitochondrial Ca2+ uniporter, endoplasmic reticulum and plasma membrane ATP-dependent Ca’+ pumps, intraterminal Ca2*-binding proteins, and the plasma membrane Naf-Ca’” antiport. For instance, the development of selective antibodies to these Ca2+ regulating proteins could make it possible to identify any change in each of these components during aging.j” Calcium ~orneo~~~is during pathological aging

involved in age-related disorders as Altered Ca ‘+ homeostasis may be pathophysiologically suggested by the following evidence: (a) Nimodipine, a Ca*+ channel antagonist, has been

665

Calcium influx during aging

reported to enhance memory in older persons with chronic cerebrovascular disorders;’ (b) pyruvate dehydrogenase, an enzyme which is inactivated by Ca*+, is decreased in post-mortem Alzheimer’s brains;40 ( c ) accumulation of Ca*+ in neurofibrillary tangles (plaques) in Parkinsonism dementia of Guam;” (d) Ca*+-activated proteases (e.g. calpain) are thought to be causal agents in the development of brain degeneration and pathology.” Ca*+ homeostasis may actually be more important than is currently thought. In fact, changes in internal [Ca*+] can trigger the transcription of ‘immediate-early genes’ such as c-binding proteins which are thought to initiate or modulate the expression of other genes possibly leading to long term functional changes. The difference between old persons with normal brains and with brain disorders suggest that if Ca*+ homeostasis is only perturbed during normative aging, it may be extremely disrupted during pathological aging. In order to identify such pathological changes in Ca*+ homeostasis, it is necessary to recognize Ca*+ homeostasis changes that usually occur during healthy normative aging and determine what causes them.** Such an approach might lead to more effective treatments for age-related disorders. In this respect, study of the readily accessible mouse neuromuscular junction is essential because: (a) it can help in the development of hypotheses and techniques appropriate for studying central nervous system synapses; (b) it has the advantage of experimental manipulations, different factors could be introduced or eliminated to elucidate aging and pathology. Factors involved in the increased release at old terminals

Ultimately, a single or a combination of presynaptic and/or postsynaptic processes must be responsible for the observed increase in soleus quanta1 content during aging. Thus far, no morphological or physiological correlate has been linked to this increase of quanta1 content during These previous studies, however, revealed the following: (a) the greater quanta1 aging. 1,2,4,6,‘2~13,39 content during aging is observed whether or not muscle fiber input resistance is increased;2,4T6 and whether or not acetylcholinesterase (AChE) activity is reduced;4*42 (b) postsynaptic acetylcholine receptors (AChRs) are not increased in number despite the increased quanta1 content during aging;6 and (c) there is no concomitant trend towards larger nerve terminals3 to account for the progressive increase in quanta1 content.4 The present study was an initial step in examining Ca*+ regulation in the motor nerve terminal during aging. The strong positive correlations between quanta1 content and each of the TA and Tp suggest that an increase in the level of free intraterminal Ca*+ is somehow involved in the mechanism of greater release at old terminals. Other important physiological differences (e.g. in synaptic vesicle recycling, active zones) between young and old terminals may also exist, but this demonstration of age-related change in Ca*+ regulation provides a foundation for further study. Based on the present findings of a physiological correlate to the increase of transmitter release during aging, it is necessary to determine whether release per unit length (or area) of nerve terminal is increased during aging. Acknowledgements-This

research was supported by National Institute of Health grant AGO4755 to Dr M. A. Fahim.

REFERENCES 1. Alshuaib W. B. and Fahim M. A. (1989) Increased Ca-influx at old soleus nerve terminals of the mouse. Sot. neurosci. Abstr. 15, 1382. 2. Alshuaib W. B. and Fahim M. A. (1990) Effect of exercise on physiological age-related change at mouse neuromuscular junctions. Neurobiol. Aging (in press). 3. Andonian M. H. and Fahim M. A. (1989) Nerve terminal morphology in C57BU6NNia mice at different ages. J. Gerontol. 44, B43-51.

4. Anis N. and Robbins N. (1987) General and strain specific age changes at mouse limb neuromuscular Neurobiol.

junctions.

Aging 8, 319-318.

5. Baker P. F. (1972) Transport and metabolism of calcium ions in nerve. Prog. biophys. molec. Biol. 24, 177-223. 6. Banker B. Q., Kelly S.S. and Robbins N. (1983) Neuromuscular transmission and correlative morphology in young and old mice. J. Physiol. (London) 339, 355-375. 7. Bono G. (1985) In Nimodipine-pharmacological and clinical properties (eds Betz E., Deck K. and Hoffmeister F), pp. 275-287. Schattauer, Stuttgart. 8. Brinley F. J., Tiffert T. and Scarpa A. (1978) Mitochondria and other calcium buffers of squid axon in situ. J. gen. Physiol. 72, 101-127. 9. Dodge F. A. Jr and Rahamimoff R. (1967) Co-operative neuromuscular junction. J. Physiol. (London) 193, 41W32.

action of calcium ions in transmitter

release at the

666

W. B. Alshuaib

and M. A.

Fahim

IO. Erulkar S. D. and Rahamimoff R. (1978) The role ofcalciumions in tetanic and post temntc mctc;i~ <>Imiti;.i ,/I. .- :./ plate potential frequency. f. Physiul. {London) 278, 501-5 I 1. 11. Etienne P. and Baudry M. (1987) Calcium dependent aspects of synaptic plasttcity. excitatory anifil a .ttxi neurotransmission. brain aging and schizophrenia: a unifying hypothesis. Neurohiol Axing X. K?-%h 12. Fahim M. A., Holley J. A. and Robhins N. (19X3) Scanning and light microscopic Stud) \tt *ipc cb:~n~:c li neuromuscular junction in the mouse. J. Neurocyrol. 12, 13-25. studies of young and old mctusc IIL’~I~OI~I~ISCU~;II tij~. iii*n ! 13. Fahim M. A. and Robbins N. (1982) Ultrastructural

~~U~O~~i~l. 11,64I-656. 14. Fahim M. A.. Robbins N. and Price R. (1987) Fixation effects on synaptic vcsiclc density m I~~‘~I~~~IIIu~~LII:II ;:i:,&:;.>i! v of young and old mice. Neurohiol. Aging 8, 71-75. IS. Garruto R. M., Fukatsu R., Yanagihara R., Gajdusek D. C., Hook G. and Fiori c‘. E. (1’184) imaging ofcaic!r~m ,tmi aluminum in neurofibrillary tangle-bearing neurons in Parkinsonism-dementia of Guam I+rl( ,mor. l! ?
_