Comparison of the cholinergic synaptic responses in mature and old animals

In the present study, the effect of aging on synaptic function in the feeding neural circuit was explored. We focused on the inhibitory monosynaptic response in the JC motor neurons produced by the cholinergic MA neurons. To suppress the polysynaptic activity in these experiments, we used a solution with increased divalent cations in which Ca2+ and Mg2+ concentrations were increased, respectively, five- and twofold by substitution of Na+.

Figure 1a shows the typical inhibitory postsynaptic potentials (IPSPs) produced in JC motor neurons by firing of MA neurons in mature and old animals. For each preparation, the average size of the IPSPs was obtained from six trials of MA firing at 40 s intervals. This time interval was sufficient to prevent activity-dependent synaptic suppression. Figure 1b shows a comparison of the average size of the IPSPs obtained from all preparations in the mature (5.35 ± 0.76 mV, n = 11) and old animals (2.65 ± 0.37 mV, n = 11). The average size of the IPSPs was significantly smaller in the old animals than in the mature animals (P < 0.005). We also compared the resting membrane potentials of the JC motor neurons because the sizes of the postsynaptic potentials were affected by the membrane potentials of these neurons. In our analysis, we could not find a significant difference in the average resting membrane potential of the JC motor neurons between the mature (− 57.8 ± 1.1 mV, n = 18) and old animals (− 57.9 ± 1.5 mV, n = 18). These results indicate that the difference in IPSP size between the two age groups may be physiologically meaningful.

Fig. 1

Effects of aging on the inhibitory postsynaptic potentials (IPSPs) in the JC motor neurons produced by firing of MA neurons. a Typical MA-produced IPSPs in JC motor neurons in mature and old animals. b Comparison of the average sizes of the IPSPs in the mature (n = 11) and old (n = 11) animals. *** significant difference at P < 0.005

In voltage-clamp measurements, the MA-produced synaptic currents in the JC motor neurons were also compared in the mature and old animals. Then the membrane potential of the JC motor neurons was clamped at − 40 mV because the inhibitory effects of the MA neurons on the JC motor neurons during the rhythmic feeding response usually appeared near the firing threshold of the JC motor neurons (Nagahama et al. 1999). Figure 2a shows the typical inhibitory postsynaptic currents (IPSCs) produced in JC motor neurons by firing MA neurons in mature and old animals. For each preparation, the average size of the IPSCs was obtained from six trials of MA firing at 40 s intervals. Figure 2b shows a comparison of the average size of the IPSCs obtained from all preparations in the mature (8.02 ± 1.37 nA, n = 13) and old animals (3.24 ± 0.39 nA, n = 10). The average size of the IPSCs was significantly smaller in the old animals than in the mature animals (P < 0.005). We also explored the Cm of the same JC motor neurons, which increased in proportion to the surface area of the neurons, in the mature and old animals. As shown in Fig. 2c, the average Cm was significantly smaller (P < 0.05) in the old animals (10.7 ± 0.8 nF, n = 10) than in the mature animals (14.0 ± 1.3 nF, n = 13). These results suggest that the average surface area and average size of the JC motor neurons may decrease with aging. Therefore, to remove the effects of surface area on the size of the IPSC between the mature and old animals, we obtained the current density (IPSC divided by Cm) in each preparation. Figure 2d shows a comparison of the average current density in the mature and old animals. Even in this case, the average current density was significantly lower (P < 0.005) in the old animals (0.312 ± 0.038 pA/pF, n = 10) than in the mature animals (0.554 ± 0.058 pA/pF, n = 13). These results indicate that the cholinergic synaptic response in the JC motor neurons produced by the MA neurons may decrease with aging.

Fig. 2

Effects of aging on the inhibitory postsynaptic currents (IPSCs) and the synaptic current density in the JC motor neurons produced by firing of MA neurons when the membrane potential of the JC motor neurons was clamped at − 40 mV in voltage-clamp measurements. a Typical MA-produced IPSCs in JC motor neurons in mature and old animals. b Comparison of the average sizes of the IPSCs in the mature (n = 13) and old animals (n = 10). c Comparison of the average membrane capacitance (Cm) in the JC motor neurons in the mature (n = 13) and old animals (n = 10). d Comparison of the average current density (IPSC/Cm) in the mature (n = 13) and old animals (n = 10). *significant difference at P < 0.05; ***significant difference at P < 0.005

Rhythmic firing patterns in JC motor and MA neurons induced by electrical nerve stimulation in mature and old animals

In the central nervous system of Aplysia, synaptic activity in other regions has also been reported to decrease with aging (Southall et al. 1997; Chandhoke et al. 2001; Kempsell and Fieber 2015). Thus, we next explored the extent to which the reduction in cholinergic synaptic activity can affect the feeding response in old animals. As reported previously, the change in this synaptic activity largely affected the rhythmic firing pattern in the JC motor neurons between the ingestive and rejective responses in food preference behavior (Nagahama et al. 1999). Therefore, we focused on the change in the rhythmic firing pattern in the JC motor neurons during the feeding response with aging. We have also previously demonstrated that electrical stimulation of any nerve leaving the buccal ganglia can directly activate the feeding CPG in isolated preparations (Nagahama and Takata 1990; Kinugawa and Nagahama 2006; Narusuye et al. 2013). Therefore, in our experiments herein, the feeding-like rhythmic firing patterns in the JC motor and MA neurons were induced via electrical stimulation of the ipsilateral esophageal nerve with repetitive short current pulses (duration, 5 ms; intensity, 6 V; frequency, 2 Hz).

In the mature animals, electrical stimulation induced early depolarization and successive stable rhythmic bursts of firing in both JC motor and MA neurons (Fig. 3a). However, the same stimulation did not always induce stable rhythmic bursts of firing of these neurons in the old animals, as shown in Fig. 3b1. Electrical stimulation was repeated more than 5 times in each preparation of all animals. Then, stable rhythmic responses were obtained in approximately two-thirds of all trials in each preparation of the old animals, although stable responses were always obtained in each preparation of the mature animals. Therefore, we analyzed the data representing the stable rhythmic bursts of firing in the neurons of the old animals, as shown in Fig. 3b2.

Fig. 3

Simultaneous recordings of the rhythmic bursts of firings in the JC motor and MA neurons during the responses induced by repetitive electrical stimulation of the ipsilateral esophageal nerve in the mature (a) and old animals (b2). In the old animals, the same stimulation could not always induce stable rhythmic bursts of firing in these neurons (b1). Upward and downward arrows indicate the start and end of the electrical stimulation, respectively

Comparison of the periods of rhythmic responses in mature and old animals

In the analysis of the rhythmic responses, we initially explored the period, i.e., the time length between the onset times of the adjacent depolarizing phases during the rhythmic depolarization of the JC motor neurons. The period tended to elongate later in a single rhythmic response (e.g. Fig 3b2), and we obtained the value by averaging several stable successive periods for each rhythmic response. For each preparation, the average period was obtained from four trials of stimulation. Figure 5a shows a comparison of the average periods obtained from all preparations in mature (6.66 ± 0.72 s, n = 10) and old animals (9.67 ± 1.02 s, n = 13). The average period in the old animals was significantly longer than that in the mature animals (P < 0.02), indicating that the period was prolonged with aging.

Comparison of firing patterns in JC motor neurons during rhythmic responses in mature and old animals

Simultaneous recordings of the typical bursts of firing in the JC motor and MA neurons during the rhythmic responses induced by electrical stimulation in the mature (a) and old animals (b) are shown in Fig. 4. In this figure, we found that the firing of the JC motor neurons was prominently stronger and lasted longer in old animals than in mature animals. This result can also be seen from a comparison of Fig. 3a and b2. Then, we quantitatively compared the firing patterns in the JC motor neurons between the two ages.

Fig. 4

Simultaneous recordings of the typical bursts of firing in the JC motor and MA neurons during the rhythmic responses induced by repetitive electrical stimulation in the mature (a) and old animals (b). c Illustration of the typical firing patterns in the JC motor and MA neurons at the same depolarizing phase, explaining the length of depolarization (Depo-length), delay time of firing onset (Delay) in the JC motor neurons, and burst length (Burst-length) in the MA neurons

The Depo-length at each depolarizing phase tended to fluctuate in a single rhythmic response. For each preparation, the average length was obtained from several stable successive depolarizing phases in each response and then from four trials of stimulation. Figure 5b shows a comparison of the average Depo-length obtained from all preparations in the mature (3.50 ± 0.39 s, n = 9) and old animals (3.80 ± 0.31 s, n = 9). There was no significant difference between the two age groups in the statistical analysis, although the average length was longer in the old animals.

Fig. 5

Analysis of the firing patterns in JC motor neurons during rhythmic responses in mature and old animals. a Comparison of the average periods of the rhythmic responses in the mature (n = 10) and old animals (n = 13). b Comparison of the average lengths of JC depolarization (Depo-length) in the mature (n = 9) and old animals (n = 9). c1 Relationship between the delay time of the JC firing onset (Delay) and the length of JC depolarization (Depo-length) at each depolarizing phase during the responses induced by three trials of stimulation (different symbols) in a single mature animal (open symbols) and a single old animal (filled symbols). c2 Comparison of the average normalized delay time of JC firing onset (normalized Delay) in mature (n = 9) and old animals (n = 9). NS not significant; **significant difference at P < 0.02; ****significant difference at P < 0.001

Next, we explored the relationship between the Delay and the Depo-length at each depolarizing phase during the rhythmic responses. Figure 5c1 shows the data plots of the Delay against the Depo-length during the responses induced by three trials of stimulation (different symbols) in a single mature animal (open symbols) and a single old animal (filled symbols). These data could almost be represented by two separate straight lines, both approaching the origin for the mature and old animals, although the Depo-length fluctuated. Therefore, to analyze these data, we obtained the normalized Delay by dividing the Delay by the Depo-length at each depolarizing phase. The average value for each preparation was obtained from several stable successive depolarizing phases in each response and then from four trials of stimulation. A comparison of the average normalized Delay obtained from all preparations in the mature (0.657 ± 0.030, n = 9) and old animals (0.307 ± 0.030, n = 9) is shown in Fig. 5c2. The average normalized Delay was significantly smaller in the old animals than in the mature animals (P < 0.001), indicating that the onset time of JC firing at each depolarizing phase may advance with aging.

In old animals, synaptic inputs to JC motor neurons can be produced as a result of complex changes in synaptic activities and cell properties of many neurons in the neural circuit of the feeding CPG. An advance of the JC firing onset may be caused by an increase in the excitatory inputs, a decrease in the inhibitory inputs, or sometimes both, to the JC motor neurons during the early phase of JC depolarization. Therefore, the age-related decline in the inhibitory synaptic response produced by MA neurons may partly contribute to an advance of JC firing onset. However, there is a possibility that the age-related changes in some other excitatory or inhibitory inputs to the JC motor neurons could also affect the change in the firing pattern in the JC motor neurons.

Analysis of the relationship between the firing patterns in JC motor and MA neurons during rhythmic responses in mature and old animals

The bursts of the MA firing may basically affect the firing pattern in the JC motor neurons. We next explored whether the firing pattern in the MA neurons changed with aging to affect the firing pattern in the JC motor neurons during the rhythmic responses. The Burst-length and/or firing frequency at each burst in the MA neurons at each depolarizing phase may affect the Delay in the JC motor neurons.

For the Burst-length, the normalized Burst-length was analyzed by dividing the Burst-length in the MA neurons by the Depo-length in the JC motor neurons at each depolarizing phase at the same time in these neurons. For each preparation, the normalized Burst-length was obtained from several stable successive phases for each response and then from four trials of stimulation. Figure 6a shows a comparison of the average normalized Burst-length obtained from all preparations in the mature (0.434 ± 0.031, n = 6) and old animals (0.355 ± 0.046, n = 6). There was no significant difference in the statistical processing between the two age groups, although the average value of the normalized Burst-length was smaller in the old animals. In contrast, the average normalized Delay in the same preparations (n = 6 in each age) was significantly smaller (P < 0.001) in the old animals (0.291 ± 0.043) than in the mature animals (0.606 ± 0.031) despite the relatively small number of preparations. To evaluate the effect of the normalized Burst-length on the decrease in the normalized Delay with aging, we also explored the relationships between the normalized Burst-length and the normalized Delay for each preparation in the same animals used for the average of the normalized Burst-length, and the results are shown in Fig. 6b. The data plots of the normalized Delay against the normalized Burst-length for each preparation showed two separate groups with large differences in the values of the normalized Delay in the mature (open circles) and old animals (filled circles). The decrease in the normalized Delay with the decrease in the normalized Burst-length in each group was not large enough to explain the large decrease in the normalized Delay in the two age groups. These results indicate that the Burst-length in MA neurons may have a little effect on the age-related decrease in the delay time of JC firing onset.

Fig. 6

Analysis of the firing patterns in MA neurons during rhythmic responses in mature and old animals. a Comparison of the average normalized burst length (normalized Burst-length) in the MA neurons in the mature (n = 6) and old animals (n = 6). b Relationships between the normalized burst length (normalized Burst-length) and the normalized delay time of the JC firing onset (normalized Delay) for each preparation in the mature (open circles, n = 6) and old animals (filled circles, n = 6). c Comparison of the average firing frequency of MA neurons (Freq at burst) in mature (n = 6) and old animals (n = 6). d Relationships between the firing frequency of the MA neurons (Freq at burst) and the normalized delay time of the JC firing onset (normalized Delay) for each preparation in the mature (open circles, n = 6) and old animals (filled circles, n = 6). NS not significant

Thereafter, we explored the frequency of MA firing at each burst of the depolarizing phase using the same preparations. For each preparation, the average MA firing frequency was obtained from several stable successive phases for each response and then from four trials of stimulation. Figure 6c shows a comparison of the average MA firing frequency at each burst (Freq at burst) obtained from all preparations in the mature (39.9 ± 4.4 spikes/s, n = 6) and old animals (43.9 ± 4.0 spikes/s, n = 6). There was no significant difference in the statistical processing between the two age groups. To evaluate the effect of the MA firing frequency on the decrease in the normalized Delay in the JC motor neurons with aging, we also explored the relationships between the firing frequency and the normalized Delay for each preparation in the same animals used for the average of the MA firing frequency. As shown in Fig. 6d, the data plots of the normalized Delay against the MA firing frequency for each preparation in the mature (open circles) and old animals (filled circles) showed two separate groups with large differences in the values of the normalized Delay values. Dependence on the firing frequency of the normalized Delay value was not found in either age group. These results indicate that the MA firing frequency may have a minimal effect on the age-related decrease in the delay time of the JC firing onset.