Proctolin enhances the pyloric rhythm. A. Top: Schematic diagram of the stomatogastric nervous system. The paired commissural ganglia (COGs) and esophageal ganglion (OG) contain modulatory projection neurons that send axons to the stomatogastric ganglion (STG) through the stomatogastric nerve (stn). Bottom: Simplified diagram of the pyloric network shows the pacemaker group consisting of the AB, PD and LPG neurons coupled strongly with electrical synapses (gap junctions) and synaptically inhibiting the follower neuron LP. The LP neuron in turn inhibits the two PD neurons, producing a reciprocally inhibitory sub-network. B. Intracellular traces recorded in the normal ongoing pyloric rhythm with intact descending neuromodulatory inputs show the rhythmic alternation of the PD and LP neurons (Control). Blocking or cutting the stn removes all descending neuromodulatory input to the STG and disrupts the pyloric rhythm (Decentralized). Bath application of 10⁻⁶ M proctolin recovers the pyloric rhythm, increases the amplitude of the slow wave oscillations in the PD and LP neurons and increases the spike frequency and number of spikes per burst (Proctolin). Line segments indicate the baselines for the membrane oscillations of PD (−56 mV) and LP (−59 mV) recorded in Control.

Proctolin strengthens the graded component of the LP to PD synapse. A. The LP neuron was voltage-clamped at −60 mV and 500 ms depolarizing square pulses with increasing amplitudes up to 0 mV (V_LP) were used to activate the LP to PD synapse in Control, 10⁻⁶ M Proctolin and Wash. The resulting graded IPSP (gIPSP) recorded in the PD neuron (V_PD) increased as the amplitude of the V_LP pulse increased for all conditions and was always larger in Proctolin. The enhancement of the synaptic strength was reversible after a 45 minute wash. The PD neuron was recorded at its resting potential which remains quite stable in TTX. Baseline V_PD (mV): Ctl: −60.5, Proc: −59.5 mV, Wash: −60 mV. B. The peak amplitudes of gIPSPs were plotted against the V_LP pulse values to obtain the synaptic input-output curves in control and in different doses of proctolin. The amplitudes of gIPSPs were augmented in proctolin in a dose dependent manner but saturated at ~10⁻⁶ M. C. The synaptic input-output curve showed a sigmoidal dependency on the presynaptic voltage for both control and proctolin (10⁻⁶ M). Proctolin significantly strengthened gIPSPs of the LP to PD synapse for presynaptic amplitudes ≥ 35 mV compared to control and wash (two-way ANOVA post hoc Tukey analysis, p < 0.01 for V_LP = −35 to −10 mV; N=6). D. The amplitude of the individual gIPSPs in each preparation was normalized to the maximum gIPSP in that condition. Each data point (black: control; red: proctolin; blue: wash) indicates a single measurement from six preparations and these measurements were fit with Boltzmann type equations. Proctolin shifted the midpoint of the input-output curve by ~5.5 mV to more negative membrane potentials (_halfV in mV—Ctrl: −29.0±0.11; Proc: −35.4±0.14; Wash: −28.0 ±0.44) but did not significantly change the slope of this curve, although the slope became shallower after wash (k_half in mV—Ctrl: −3.11±0.10; Proc: −2.83±0.12; Wash: −6.36 ±0.45).

Proctolin modulates the short-term dynamics of the LP to PD graded synapse in response to different amplitude depolarizations. A. LP neuron was voltage-clamped at −60 mV and stimulated with a series of five square pulses of 20 mV (V_LP: black trace) or 40 mV (V_LP: gray trace) amplitudes in control, proctolin and wash. In control and wash, the LP to PD synapse was depressing. In proctolin, in response to 20 mV presynaptic depolarization, the gIPSPs in the PD neuron showed facilitation. Darker color traces in each condition correspond to the lower pulse amplitude of 20 mV. Baseline VPD at −58 mV in control, −56.5 mV in proctolin and −58.5 mV in wash. B. The ratio of the 5^th to the 1^st mean gIPSP amplitude (A₅/A₁) calculated for all presynaptic depolarizations with an inter-pulse interval of 500 ms plotted against presynaptic pulse amplitudes (ΔV_LP). In control (black), the ratio A₅/A₁ was slightly less than one for all values of presynaptic pulse amplitudes (ΔV_LP), indicating that the synapse was always depressing. In contrast, in the presence of proctolin (red), the A₅/A₁ ratios in response to pulse amplitudes between 20 and 24 mV were greater than 1 while the ratios in response to larger pulse amplitudes were less than 1, indicating facilitation and depression, respectively (two-way ANOVA with post hoc Tukey test showing p<0.001 between ctrl and proctolin and p<0.008 between control and wash for ΔV_LP = 20, 22 and 24 mV; not significant at other ΔV_LP). C. A₅/A₁ measured for different inter-pulse interval (IPI) durations in a train of presynaptic pulses applied at low (20 mV) or high (40 mV) amplitudes. Proctolin produced significant facilitation synapse at IPI values ≤ 1000 ms with low-amplitude presynaptic pulses (two-way ANOVA with post hoc Tukey test showing p<0.01 marked by *; not significant at other IPI values). With high amplitude pulses there was no statistically significant difference in synaptic dynamics between control and proctolin, both of which resulted in depression (two-way ANOVA; p>0.5; dashed green curve shows exponential decay fit 1–0.603 exp(−t/434)).

Proctolin caused a facilitation of the LP to PD graded inhibitory postsynaptic potentials in response to low-amplitude presynaptic depolarizing pulses, which is associated with the activation of a Ca²⁺-like inward current. A. The presynaptic currents in the LP neuron (I_LP) are measured in Ca²⁺ saline (green) and in Mn²⁺ saline (magenta) using a train of low-amplitude presynaptic voltage pulses. The difference between the two measured currents (ΔI_LP= I_LP:Ca2+ − I_LP:_Mn2+) reveals a new inward current activated in proctolin (red). A series of low amplitude presynaptic depolarizing pulses cause little response in postsynaptic neuron PD in control. Proctolin strengthened and caused facilitation in the PD neuron in response to low amplitude presynaptic depolarizing pulses. ΔI_LP also increased with each pulse. V_PD is not shown in Mn²⁺ because synaptic transmission was blocked (baseline V_PD at −59 mV control, −60 mV proctolin; dotted line indicates 0 nA). B. The mean value (area divided by pulse duration) of the presynaptic inward current ΔI_LP in response to each pulse plotted against the mean value of the postsynaptic potentials V_PD (IPSP area divided by pulse duration) for measurements in Ca²⁺ saline in proctolin. Linear regression fit shows the correlation between the presynaptic inward current and the postsynaptic potentials (R=0.69, P<0.001, N=9).

The strength of unitary spike-mediated IPSPs (sIPSPs) is enhanced by proctolin. A. The LP neuron was voltage-clamped at −60 mV and stimulated with one short square pulse of duration 10 ms and amplitude 30 mV. This resulted in a single action potential in the LP neuron. The resulting sIPSP was recorded in PD neuron (V_PD) in control and in proctolin (10⁻⁶ M). Baseline V_PD was at −58 mV in both conditions. B. Increasing the presynaptic 10-ms square pulse amplitude (to peak voltages V_LP = −40, −30, −20, −10 mV; shown schematically) always resulted in a single action potential in the LP neuron which was recorded on the nerve lvn. The resulting sIPSPs recorded in PD neuron (V_PD) in control and in proctolin (10⁻⁶ M) depended on the amplitude of the presynaptic stimulus. C. The amplitude of sIPSPs in response to the 10 ms voltage pulse in control and proctolin increased if the amplitude of this short presynaptic voltage pulse was increased. Bath application of proctolin significantly increased the sIPSP amplitude for all values of the presynaptic pulse amplitudes (two-way ANOVA post hoc Tukey analysis, p<0.05, N=8). D. Schematic diagram showing how antidromic spikes were elicited in the LP neuron. The LP neuron (blue) soma was voltage-clamped and the PD neuron (red) was recorded in current clamp. To elicit antidromic spikes, the lateral pyloric nerve (lpn) which contains the axon of the LP neuron but not that of the PD neuron was stimulated. E. The LP neuron was voltage-clamped at different holding potentials (V_LP=−70, −60, −55 and −50 mV). The sIPSPs were recorded in PD neuron in response to antidromic spikes in the LP neuron in control and proctolin (10⁻⁶ M). F. The peak sIPSP amplitude plotted against the holding potentials of the LP neuron in control and in proctolin. As the holding potential of the LP neuron was increased, the amplitude of sIPSPs in the PD neuron also increased in both control and 10⁻⁶ M proctolin. In the presence of proctolin, the sIPSPs were larger in comparison to control (two-way RM-ANOVA, p<0.05, N=5).

Proctolin does not have a significant effect on the short-term dynamics of the sIPSPs. A. The LP and PD neurons were voltage clamped at a holding potential of −50 mV. The resulting spike-mediated synaptic currents in the PD neuron in response to two successive antidromic spikes with a 100 ms interval were recorded and repeated 10 times. The averaged traces were shown in control (black) and proctolin (red). The spike-mediated synaptic current (sIPSC) amplitude was significantly strengthened by proctolin (Ctrl: 0.49±0.02 nA, Proc: 1.3±0.1 nA; Student's t-test, P<0.01, N=4). Baseline I_PD at 2.2 nA incontrol and 2.0 nA in proctolin. B. The ratio of the sIPSC amplitudes in response to the first two successive antidromic spikes plotted against inter-spike interval in control and in proctolin. The exponential fit curves (y = 1− Ae^−x/B) were plotted in control (black: A=−0.23, B=125) and in proctolin (red: A=−0.15, B=167). Proctolin did not cause a significant change in the short-term dynamics compared to control (Student's t-test on exponential fits, p=0.37).

Proctolin strengthens the total LP to PD synaptic IPSP measured using prerecorded realistic waveforms in the LP neuron. A. After decentralizing the preparation, the LP neuron was voltage-clamped at a holding potential of −60 mV and injected with a voltage profile constructed using a pre-recorded realistic LP neuron waveform (V_LP) with a total amplitude 39 mV (1.5 times the recorded amplitude). The resulting IPSP was measured in the PD neuron (V_PD). The simulated action potentials of the realistic waveform resulted in real action potentials in the presynaptic LP neuron which were recorded on the nerve lvn. Baseline V_PD at −59 mV in Normal Saline control and −60 mV in both TTX control and TTX proctolin. Voltage scalebar applies to panels A–D. B. Application of the realistic waveform in TTX with total amplitude of 26 mV (same as recorded amplitude; gray V_LP trace) resulted in synaptic facilitation of the gIPSP in proctolin (red V_PD trace) but not control saline (black V_PD trace). In contrast, application of the waveform at 39 mV (gray) resulted in slight depression of the gIPSP both in control (gray) and in proctolin (pink). C. In normal saline, the LP waveform was applied periodically with total amplitude of 26 mV or 39 mV. The resulting IPSP, a combination of the graded and spike-mediated components, was measured in the PD neuron in control saline or in the presence of proctolin. The responses shown are after at least four cycles of the presynaptic waveform (circled in A and B) when the IPSP has reached its stationary state. D. In TTX, the same LP waveform was applied periodically with total amplitude of 26 mV or 39 mV. The resulting IPSP in this case consists only of the graded component. All responses shown in C and D were measured in the same preparation. E. Mean values of the IPSP as shown in panel C indicate a significant increase with both amplitude and proctolin (Proc) application (two-way RM-ANOVA; p<0.05, N=5; * indicates p<0.05 in post hoc Tukey analysis). Ctrl indicates control saline. In panels B–E, darker colors in each condition correspond to the lower amplitude waveform (26 mV).

Proctolin effects on the LP to PD synapse in biological conditions. A. The LP neuron voltage waveform changes in shape and amplitude in proctolin. These waveforms were played back periodically at the original recorded amplitude (as shown) from a minimum holding potential of −60 mV and a cycle period of 1 s into the voltage clamped LP neuron and the postsynaptic response (mean value) was measured in the PD neuron after reaching stationary state (see ). B. Bath application of proctolin (Proc) significantly enhanced the IPSP amplitude measured with either waveform (shown in A) compared to control saline (Ctrl). The proctolin-recorded waveform produced a modestly larger response only in the presence of proctolin but not in control saline (two-way RM-ANOVA with post-hoc Tukey analysis; * indicates p<0.02; N=5). C. Proctolin enhances the LP to PD synapse during the ongoing pyloric rhythm. The traces of extracellular recordings from lvn show the ongoing rhythm activity of the pyloric network in control and in proctolin (10⁻⁶ M). One of the two PD neurons was voltage clamped at a holding potential of −55 mV; the resulting synaptic currents were recorded in this PD neuron in control and in proctolin. Baseline I_PD at −1 nA in control and proctolin. D. The synaptic response in PD neuron was integrated from the end of a PD burst to the start of the following PD burst (vertical dashed lines in A). When necessary, DC current was injected in the second PD neuron to bring the cycle period to 1 sec in all measurements. The IPSC area was significantly larger in proctolin than control or wash (one-way RM-ANOVA, p<0.01, N=5).

The enhancement of the LP to PD synapse by proctolin reduces the variability in the pyloric cycle period. A. The coefficient of variation (CV) of cycle period is significantly reduced in the presence of bath applied proctolin (Student's t-Test, p=0.030, N=8). B. The PD neuron was subjected to a noisy current input (0.5 nA, 20 ms voltage pulses applied at discrete time intervals following a Poisson distribution with mean frequency of 5 Hz). Functional removal of the LP to PD synapse (−5 nA DC current in LP) significantly increased the CV in both control and proctolin. The CV in proctolin was significantly smaller than control only when the LP to PD synapse was functionally intact (two-way RM-ANOVA with post-hoc Tukey analysis; * indicates p<0.05; N=6). C. The pyloric cycle period was perturbed with a single 2 nA, 50 ms pulse was injected into the PD neuron once very ten or so cycles at different phases of the cycle. The absolute value of the change in cycle period (|ΔP/P|), independent of the stimulus phase, was compared in different conditions. Functional removal of the LP to PD synapse significantly increased |ΔP/P| in both control and proctolin. The value of |ΔP/P| in proctolin was significantly smaller than control only when the LP to PD synapse was functionally intact (two-way RM-ANOVA with post-hoc Tukey analysis; * indicates p<0.05; N=7). D. Adding the proctolin-induced enhancement of the LP to PD synapse using dynamic clamp. Traces of extracellular recordings from lvn show the ongoing pyloric rhythm activity. Intracellular traces recorded in LP and PD neurons show the rhythmic alternation of the PD and LP neurons. Left panel (black) shows pyloric activity in Control. Using dynamic clamp, the LP to PD synapse was enhanced (right panel, red; see Materials and Methods for equations). E. The CV of cycle period was calculated over 40 consecutive cycles from the nerve pdn to measure the effect of the dynamic-clamp enhanced synapse. The variability of the pyloric rhythm period was significantly reduced when the LP to PD synapse was enhanced (Student's t-Test, p=0.022, N=5; experiments in A, B, C and E are non-overlapping).