Electrical coupling regulates layer 1 interneuron microcircuit formation in the neocortex

Abstract

The coexistence of electrical and chemical synapses among interneurons is essential for interneuron function in the neocortex. However, it remains largely unclear whether electrical coupling between interneurons influences chemical synapse formation and microcircuit assembly during development. Here, we show that electrical and GABAergic chemical connections robustly develop between interneurons in neocortical layer 1 over a similar time course. Electrical coupling promotes action potential generation and synchronous firing between layer 1 interneurons. Furthermore, electrically coupled interneurons exhibit strong GABA-A receptor-mediated synchronous synaptic activity. Disruption of electrical coupling leads to a loss of bidirectional, but not unidirectional, GABAergic connections. Moreover, a reduction in electrical coupling induces an increase in excitatory synaptic inputs to layer 1 interneurons. Together, these findings strongly suggest that electrical coupling between neocortical interneurons plays a critical role in regulating chemical synapse development and precise formation of circuits.

Figure 1. Development of electrical and GABAergic connections between interneurons in neocortical layer 1.

(a) A schematic diagram of a quadruple whole-cell recording of four neurons in cortical layer 1. (b) An image of four layer 1 neurons filled with neurobiotin during recording and labelled with fluorescence-conjugated avidin. Scale bar, 100 μm. (c) A magnified image of b (dotted rectangular region). Red arrowheads indicate the four recorded neurons, and green arrowheads indicate dye-coupled neurons. Scale bar, 50 μm. (d) A DIC image of quadruple whole-cell recording of the four neurons in layer 1 shown in b and c. Scale bar, 50 μm. (e) Summary of the synaptic connections detected in this quadruple recording. The average traces of the postsynaptic responses are shown in the rectangle. Red traces indicate the existence of electrical synapses, and green traces indicate the existence of chemical synapses. Sample traces of AP and hyperpolarization potentials are shown to the left. Scale bar, 40 pA (green vertical scale bar), 20 ms (red horizontal scale bar), 200 mV (black vertical scale bar). ‘Pre-Stimu.', presynaptic potential; ‘Pre', presynaptic neuron; ‘Post', postsynaptic neuron. (f) A schematic diagram showing connections between the four neurons in e. Wavy red arrowheads indicate electrical connections, and green arrowheads indicate chemical connections. (g) Summary of proportion of electrical connections (red bars) and unidirectional/bidirectional chemical connections (light blue bars for unidirectional chemical connections and dark blue bars for bidirectional chemical connections) between interneuron pairs in neocortical layer 1 at different postnatal stages.

Figure 2. Electrical and GABAergic connections between layer 1 interneurons.

(a) Representative traces of voltage responses to 500 ms current pulse step injections recorded in the current-clamp mode. The red traces show the initial AP spike. Layer 1 interneurons were divided into two subtypes, LS and BS neurons, based on their AP firing patterns. The inset shows the ADP in the initial AP of a BS neuron. (b) Histogram showing that the majority of interneurons displayed LS firing pattern (LS, 79.7%, 374 cells; BS, 20.3%, 95 cells; n=63 mice). (c) Summary of the proportion of electrical coupling observed between layer 1 interneurons at P9–P25. The rate of electrical connections between LS interneurons is significantly higher than the rate between LS and BS interneurons and between BS interneurons. (d) The scatter plot of coupling coefficients revealed no significant difference between interneuron subtypes (LS–LS pairs, 1.8±0.12%, n=122; LS–BS pairs, 2.2±0.45%, n=22; BS–BS pairs, 1.1±0.15%, n=4). (e) Summary of the proportion of GABAergic connections between interneuron subtypes. (f) The unidirectional chemical synapses between LS and BS interneurons showed directional selectivity. (g) The amplitude of uIPSCs between LS interneuron pairs (37.98±2.13 pA, n=73) was significantly larger than those between LS–BS pairs (24.22±2.09 pA, n=39) and BS–BS pairs (21.44±1.73 pA, n=34) (not including failures). *P<0.05, **P<0.01, ***P<0.001, n.s., P>0.05, not significant. χ²-test, Fisher's exact test and Mann–Whitney rank sum test. Error bars in d and g represent mean±s.e.m. ADP, after depolarization.

Figure 3. Preferential existence of bidirectional chemical connections in electrically coupled pairs.

(a) A representative sample pair with bidirectional chemical connection and reciprocal electrical connection between two layer 1 interneurons. APs and hyperpolarization sequentially triggered in the presynaptic neurons (blue traces) and responses (red traces) of GABAergic connections (inward currents) and electrical connections (outward currents) recorded in the postsynaptic neurons. The bold traces represent the average and the grey traces represent the individual traces. A schematic diagram showing electrical and GABAergic connections between the two neurons (inset). Wavy red arrowheads indicate electrical connections, and green arrowheads indicate GABAergic connections. (b) Bidirectional chemical connections between layer 1 interneurons preferentially existed in electrically coupled pairs. (c) Layer 1 interneuron pairs connected by bidirectional chemical synapses preferentially formed electrical synapses. ***P<0.001, χ²-test.

Figure 4. The effect of bidirectional chemical synapse in controlling firing synchrony between layer 1 interneurons.

To test for firing synchrony in three interneuron groups (C-coupled, E-coupled and D-coupled), paired spike trains were elicited by suprathreshold current in the driver cell (red traces, 30–50 Hz) and by tonic suprathreshold depolarization in the follower cell (blue traces, ∼10 Hz), respectively. (a) Sample traces of paired spike trains from two neurons coupled by different patterns of connectivity in neocortical layer 1, the arrowheads indicate the synchronous spikes. (b) Normalized cross-correlogram (Z-score) for neuron pairs in (a). Bin size is 2 ms. Note that the cross-correlogram exhibited a peak ∼0 ms in all three groups. (c) Average Z-scores for different patterns of connectivity. Note that Z-scores of E- and D-coupled pairs were significantly higher than that of C-coupled pairs, and Z-scores of D-coupled pairs were slightly lower than that of E-coupled pairs. (d) Average JBSI values for different patterns of connectivity. C-coupled pairs=4, E-coupled pairs=6, D-coupled pairs=7, from 10 mice. *P<0.05, **P<0.01, two-tailed paired t-test. Error bars represent mean±s.e.m.

Figure 5. Synchronous synaptic activity between layer 1 interneurons.

(a) Sample traces of spontaneous activity of four layer 1 interneurons. High-temporal-resolution displays of a segment of recordings (thick black line) are shown at the bottom. Green asterisks and dotted lines indicate synchronized events between two neurons (cell 1 and 3, red traces). (b) Normalized cross-correlogram (Z-score) for neuron pairs in a. Bin size is 1 ms. Note that the frequency of events is significantly increased at ∼0 ms (inset) for the interneuron pair (1 versus 3, red), but not for any other neuron pairs. The grey region in the inset corresponds to −1 ms≤Δt≤1 ms. Similar symbols and displays are used in subsequent figures. (c) Proportion of layer 1 interneuron pairs exhibiting synchronized spontaneous activity at different developmental stages. Note that synchronous synaptic activity appeared to be correlated with the developmental process. (d) Proportion of layer 1 interneuron pairs of different subtypes from P6 to P20 exhibiting synchronized spontaneous activity. (e) Proportion of electrically coupled or not coupled interneuron pairs from P6 to P20 exhibiting synchronized spontaneous activity. n=47 mice. *P<0.05, ***P<0.001, χ²-test and Fisher's exact test.

Figure 6. Inducible lentiviral shRNA mediates effective knockdown of Connexin 36.

(a) Schematic illustration of the shRNA-expressing lentiviral constructs. The Ctrl-shRNA and the Cx36-shRNA inserts contained 21-nt sense and antisense strands. Sense and antisense strands were linked through a standard 5′- CTCAAGAGA -3′ loop structure specific to mammalian cells. (b) Western blot of protein samples isolated from cultured E18 mouse primary cortical neurons infected with Ctrl-shRNA and Cx36-shRNA lentiviruses or WT cells (uninfected cells) and probed with anti-Cx36 and anti-β-tubulin antiserum. β-Tubulin served as a loading control (n=4 pregnant mice). (c,d) Quantitative analysis of western blot (c) and real-time PCR (d) showed that both the protein (∼50%) and mRNA levels (∼68%) of Cx36 were significantly reduced in Cx36-shRNA-transfected cells as compared with WT cells and Ctrl-shRNA-transfected cells; no significant difference was noted between WT and Ctrl-shRNA groups (P>0.05). Cx36 mRNA expression levels normalized for GAPDH. (e) Schema showing the method of virus injection, using a tiny self-made glass injector to inject the virus into the gap between the dura and cortical layer 1 at P1 with CD-1 mice (also see Methods). (f) Image of virus injection into neocortical layer 1 at P1. The fast green injected along with virus helped in estimating the extent of diffusion of the virus in neocortical layer 1. (g) Representative distribution of lentivirus-labelled GFP-positive (GFP⁺) cells. GFP⁺ cells densely packed in neocortical layer 1, but sparsely distributed in deep layers, were detected at P15. Scale bar, 50 μm. *P<0.05, **P<0.01, ***P<0.001, two-tailed paired t-test. Error bars represent mean±s.e.m.

Figure 7. Connexin 36 is required for formation of electrical coupling between interneurons in neocortical layer 1.

(a,b,d,e) DIC (a,d) and GFP expression (b,e) images of a quadruple recording of four layer 1 interneurons in a Ctrl-shRNA (a,b) and Cx36-shRNA (d,e) neocortical slice. The red arrows indicate reciprocal electrical coupling (b,e). Scale bar, 20 μm (a,b), 40 μm (d,e). (c,f) Sample traces of voltage changes in the four neurons in response to sequential current injection into one of the four neurons in Ctrl-shRNA slices (c) and Cx36-shRNA slices (f). Green circles indicate GFP⁺ cells, and white circles indicate GFP-negative (GFP⁻) cells. The average traces are shown in individual table cells at varying scales. Scale bar, 3 mV (blue), 70 mV (red), 200 ms (horizontal, black) and 200 pA (vertical, black). ‘Pre-Stimu.', presynaptic stimulus protocol. (g) Summary of the proportion of electrical coupling observed between layer 1 GFP⁺ interneurons in Ctrl-shRNA or Cx36-shRNA slices at P9–P25. Green circles indicate GFP⁺ cells. n=21 mice for Ctrl-shRNA group, 37 mice for Cx36-shRNA group. ***P<0.001, χ²-test, Fisher's exact test.

Figure 8. Electrical coupling is required for the formation of bidirectional chemical synapses between layer 1 interneurons.

(a–f) Quadruple whole-cell recordings of four layer 1 interneurons in a Ctrl-shRNA (a–c) and Cx36-shRNA (d–f) slice. DIC (a,d) and fluorescence images (b,e) of the respective quadruple whole-cell recordings are shown. The green arrows in b and e indicate the direction of chemical synapses. A summary of the chemical synaptic connections detected in the quadruple recordings is also shown (c,f). Pink shading indicates the existence of chemical synapses. Scale bar, 20 μm in (a,b,d,e), 20 pA (red), 40 mV (blue) and 15 ms (black) in (c,f). ‘Pre-Stimu.', presynaptic potential; ‘Pre', presynaptic neuron; ‘Post', postsynaptic neuron. (g) Summary of the proportion of chemical synapse formation between layer 1 interneurons in Ctrl-shRNA (GFP⁺-GFP⁺ pairs) and Cx36-shRNA (GFP⁺-GFP⁺ pairs) slices at P9–P25. Green circle indicates GFP⁺ cells. (h) The proportion of bidirectional chemical connections in Cx36-shRNA group was significantly less than in Ctrl-shRNA groups. (i) The proportion of unidirectional chemical connections in Ctrl-shRNA and Cx36-shRNA groups showed no significant differences. n=21 mice for Ctrl-shRNA group, 37 mice for Cx36-shRNA group. *P<0.05, n.s., not significant, P>0.05. χ²-test, Fisher's exact test.

Figure 9. Modulation of miniature postsynaptic currents by electrical coupling in neocortical layer 1.

(a) Representative traces of inward mEPSCs (red traces, holding potential of −60 mV) and outward mIPSCs (blue traces, holding potential of +10 mV) recorded in Ctrl-shRNA and Cx36-shRNA groups in the presence of tetrodotoxin (5 μM). (b–e) Histograms (b,d) and cumulative distributions (c,e) of mEPSC amplitudes (b,c) and frequencies (d,e). Both the peak amplitude and frequency of mEPSCs in Cx36-shRNA group were significantly higher than those in Ctrl-shRNA group. (f–i) Histograms (f,h) and cumulative distributions (g,i) of mIPSC amplitudes (f,g) and frequencies (h,i). The peak amplitude and frequency of mIPSCs showed no significant differences between Cx36-shRNA and Ctrl-shRNA groups (n=12 in Cx36-shRNA group, n=8 in Ctrl-shRNA group, three mice for each group). *P<0.05, ***P<0.001, n.s., P>0.05, not significant. Two-tailed paired t-test. Error bars represent mean±s.e.m.

Figure 10. Electrical coupling promotes AP generation and synchronous firing of layer 1 interneurons.

(a) Sample traces of synchronous (1+2) or asynchronous (1 or 2) injection of subthreshold current pulses into electrically coupled interneurons in neocortical layer 1. Note that synchronous injection, but not asynchronous injection (arrowheads), results in AP generation (arrows). (b) Summary of the firing rate of the two interneurons shown in a responding to 80 pA current injection. (c) Summary of the firing rate in electrically coupled or non-coupled interneurons responding to simultaneous current injections. (d–g) Electrical coupling promoting synchronous firing of electrically coupled neurons in neocortical layer 1 in response to uncorrelated simulated neuronal activity (d,e) or uncorrelated native neuronal activity (f,g). (d,f) Sample traces of voltage changes in electrically coupled interneurons. Arrows indicate the spikes that occur in both neurons within a 1 ms window. (e,g) Normalized cross-correlogram (Z-score) analysis. The bin size is 1 ms. Note that the firing frequency is significantly increased near 0 ms for coupled interneuron pairs (red) but not for non-coupled interneuron pairs (black), indicating synchronous firing. The insets show the statistical analysis of Z-scores of coupled/non-coupled interneuron pairs. *P<0.05, **P<0.01, ***P<0.001. Mann–Whitney rank sum test and two-tailed paired t-test. Error bars represent mean±s.e.m.