Functional autaptic neurotransmission in fast-spiking interneurons: a novel form of feedback inhibition in the neocortex

Abstract

Autapses are synapses made by a neuron onto itself. Although morphological evidence for existence of autapses has been reported in several brain areas, it is not known whether such self-innervation in the neocortex is functional and robust. Here we report that GABAergic autaptic activity is present in fast-spiking, but not in low-threshold spiking, interneurons of layer V in neocortical slices. Recordings made with the perforated-patch technique, in which physiological intracellular chloride homeostasis was unperturbed, demonstrated that autaptic activity has significant inhibitory effects on repetitive firing and increased the current threshold for evoking action potentials. These results show that autapses are not rudimentary nonfunctional structures, but rather they provide a novel and powerful form of feedback inhibitory synaptic transmission in one class of cortical interneurons.

Fig. 1.

Putative autaptic contacts in an FS interneuron.A1, Fluorescence micrograph of an FS interneuron filled with biocytin and processed with Texas Red-conjugated avidin.A2, A3, Enlargement of the rectangular areas shown in A1, revealing axonal swellings (arrowheads) in close proximity to a dendrite (A2) or intersection of axonal branches with the dendrites of the same cell (arrowhead inA3), indicating sites of putative autaptic connections. Scale bar (in A1): A1, 50 μm;A2, 4 μm; A3, 6 μm. B, Typical fast-spiking behavior of the cell of A.V_m before current injection, −64 mV. Current pulses: 600 msec; −300 and 1200 pA.

Fig. 2.

LTS interneurons do not show functional autapses.A, Fluorescence micrograph of an LTS interneuron filled with biocytin and processed with Texas Red-conjugated avidin.Arrowhead, Axon directed away from the somatodendritic compartment. No close appositions between axon and dendrites were detectable. Scale bar, 50 μm. B, Firing behavior of the LTS interneuron in A. A hyperpolarizing current pulse (600 msec, −200 pA) from a V_m of −60 mV (top) evokes a rebound burst of action potentials. A depolarizing current pulse (600 msec, 100 pA) results in an adapting firing pattern. A depolarizing current pulse (600 msec, 150 pA) from aV_m of −78 mV (bottom) evokes a burst followed by a single spike. C, In voltage clamp, the same cell did not show any GABA_A receptor-mediated current after fast inward Na currents (truncated) elicited as in Figure1C. [Cl⁻]_I, 72 mm; E_Cl = −16 mV.

Fig. 3.

Functional autapses in FS interneurons.A, Voltage steps (1 msec) to +10 mV from holding potential of −70 mV in cell of A elicits fast inward Na currents (truncated), followed by slower inward currents blocked by gabazine (10 μm). The traces represent superimposed single-trial responses, showing peak amplitude fluctuation. The dotted line indicates the peak of the response, showing fixed latency. B,Traces from single trials showing a response and a failure. C, Left, Average of 20 traces in control, gabazine, and after partial washout. Right, Trace resulting after subtracting the gabazine-averaged trace from the control-averaged trace. Traces in A–Care from the same neuron shown in Figure 1. D, Averages of 15 sweeps, from another neuron, in control, in the presence of 200 μm Cd²⁺, and after washout. Stimulus parameters as in A. E, Average of 10 sweeps, in control and in the presence of the GABA_Areceptor agonist CZP (100 nm). The recordings are from a different FS cell. F, Composite plot of the mean weighted decay time constant (τ_d,w) computed from the exponential fits of the current decays in control and in the presence of CZP (n = 5; *p < 0.04).

Fig. 4.

Effects of intracellular perfusion of BAPTA on amplitude of autaptically and synaptically evoked IPSCs in FS interneurons. A, Top, Autaptic currents are blocked by BAPTA perfusion. Representative traces of autaptic IPSCs recorded from an FS interneuron intracellularly perfused with 10 mm BAPTA (see diagram) 2 and 14 min after establishment of whole-cell configuration. Gabazine (10 μm) was applied after the 14th minute. A, Bottom, Time course of the decline of autaptic IPSC amplitudes in the same cell.Dots represent peak IPSC amplitudes elicited in each single sweep. Note that responses are on average stable during the first 4–5 min of recordings but begin to steadily decline after this point. B, Top, Representative traces of extracellularly evoked IPSCs in neuron in A at the same time points. B, Bottom, Time series showing peak IPSC amplitudes in the same cell. No decline of synaptic IPSCs is present during BAPTA perfusion. Gabazine reversibly blocks the IPSCs. C, Top, Representative traces of autaptic IPSCs recorded from another FS interneuron intracellularly perfused with 4 mm EGTA, 2 and 14 min after establishment of whole-cell configuration, and in the presence of gabazine (10 μm). C, Bottom, Time course of autaptic IPSC amplitudes in the same cell, showing absence of rundown during EGTA perfusion and block by gabazine. D, Summary plot of synaptic and autaptic IPSCs in six FS neurons intracellularly perfused with 10 mm BAPTA and six cells perfused with 4 mm EGTA. Autaptic currents in the BAPTA-perfused cells showed a progressive and substantial decline up to complete block (top), whereas no rundown occurred with EGTA (bottom). All IPSCs were blocked by gabazine. Allpoints shown are averages of 15–20 sweeps in each cell in each condition. Autaptic (autIPSCs) and synaptic (synIPSCs) IPSCs were elicited every 3 sec.Horizontal dotted lines indicate unitary IPSC values in control, just after whole-cell configuration was established.Horizontal bars indicate gabazine local perfusion.

Fig. 5.

Functional shunt operated by autaptic activation.A, Representative traces of perforated-patch recordings showing responses to intracellular injection of paired 1 msec depolarizing current pulses [interval, 10 msec in control (left) and in the presence of gabazine (right; gabaz)]. A suprathreshold conditioning current pulse was followed by variable amplitude test pulse. A current level that failed to elicit a second spike in control (left) evoked a spike in the presence of gabazine (right). Resting membrane potential, −70 mV. Calibration: 20 msec, 20 mV. A, Inset, Whole-cell (nonperforated-patch, high [Cl⁻]_i) recording of spike afterpotentials in control and in the presence of gabazine (superimposed). Control afterpotential contains a GABA_Areceptor-mediated depolarizing autaptic IPSP (E_Cl = −16 mV;V_rest = −65 mV) that is blocked by gabazine, unmasking a hyperpolarizing afterpotential. Action potentials have been truncated for display purposes. A 3.5 nA, 1 msec current pulse was intracellularly injected to reliably evoke an action potential in each sweep. Traces are average of 10 sweeps in each condition. The vertical dotted line marks membrane potential 10 msec after the action potential peak. Calibration: 10 msec, 5 mV. B, Plot of spike probability versus test current pulse amplitude for the spike after a previous action potential in cell in A. Each curveis the average of the responses obtained from three to four complete stimulation series, in either control (filled circles) or in gabazine (open circles). Note that the threshold current intensities necessary to evoke a second spike were affected by gabazine. C, Summary plot of the threshold current for evoking the second spike in control and during gabazine application (n = 7). Block of autaptic responses significantly shifts the threshold current necessary to generate a second action potential (**p < 0.01).

Fig. 6.

Modulation of action potential firing by functional autapses in FS interneurons. A, Representative traces of perforated-patch recordings from an FS interneuron, firing in response to a depolarizing current injection, in control (black trace) and in the presence of gabazine (gray trace; gabaz). Shown are the portions of the trains, including the intervals between first and second (1st doublet) and 10th and 11th (10th doublet) action potentials. Injected current, 500 pA. Resting membrane potential, −67 mV. Calibration: 10 msec, 25 mV. B, Plot of instantaneous frequency versus time in control (open symbols) and in the presence of gabazine (filled symbols) at different current–injection levels. The frequencies of the first 10 spike doublets are shown for each stimulus intensity. Same cell as in A. In gabazine, initial firing frequency is increased early in the train, at all stimulus intensities. C, Instantaneous frequency versus injected current plot (f–i), calculated in the same cell of A and B for the first (squares) and the 10th (circles) spike doublets in control (filled symbols) and in gabazine (open symbols). The lines are linear fits of the scatter plots (solid lines, control;dotted lines, gabazine). The f–i slope in gabazine deviates significantly from the control line for the first spike doublet but not for the 10th spike doublet frequency.D, Composite plot of the slopes, calculated from thef–i linear fits in six FS interneurons, for either the first or the 10th spike doublets in control and in the presence of gabazine. Gabazine significantly increased the f–islope for the first but not the 10th spike doublet (**p < 0.01; n.s., difference not statistically significant).