Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2012;8(6):e1002550.
doi: 10.1371/journal.pcbi.1002550. Epub 2012 Jun 14.

Location-dependent Effects of Inhibition on Local Spiking in Pyramidal Neuron Dendrites

Free PMC article

Location-dependent Effects of Inhibition on Local Spiking in Pyramidal Neuron Dendrites

Monika Jadi et al. PLoS Comput Biol. .
Free PMC article


Cortical computations are critically dependent on interactions between pyramidal neurons (PNs) and a menagerie of inhibitory interneuron types. A key feature distinguishing interneuron types is the spatial distribution of their synaptic contacts onto PNs, but the location-dependent effects of inhibition are mostly unknown, especially under conditions involving active dendritic responses. We studied the effect of somatic vs. dendritic inhibition on local spike generation in basal dendrites of layer 5 PNs both in neocortical slices and in simple and detailed compartmental models, with equivalent results: somatic inhibition divisively suppressed the amplitude of dendritic spikes recorded at the soma while minimally affecting dendritic spike thresholds. In contrast, distal dendritic inhibition raised dendritic spike thresholds while minimally affecting their amplitudes. On-the-path dendritic inhibition modulated both the gain and threshold of dendritic spikes depending on its distance from the spike initiation zone. Our findings suggest that cortical circuits could assign different mixtures of gain vs. threshold inhibition to different neural pathways, and thus tailor their local computations, by managing their relative activation of soma- vs. dendrite-targeting interneurons.

Conflict of interest statement

The authors have declared that no competing interests exist.


Figure 1
Figure 1. Inhibitory location effects: electrophysiological recordings from brain slices and detailed compartmental model.
(A, C) Whole-cell somatic recording were carried out in a layer 5 pyramidal cell. Excitation was provided by UV laser uncaging of glutamate at a site 150 µm from the soma in a basal dendrite. Inhibition was delivered via GABA iontophoresis at the same site (A) or at the soma (C). Excitation was delivered at least 10 ms after the iontophoresis. Black traces show control case without inhibition, blue traces are in the presence of inhibition. (B, D) Input-output curves for peak somatic depolarization as a function of laser intensity. Spike thresholds indicated by asterisks were computed from sigmoidal fits to the i/o curves (see Methods); spike heights were computed from asymptotic values of sigmoidal fits, indicated by horizontal dashed lines. (E, G) Voltage traces at the soma generated by a detailed compartmental model of a layer-5 pyramidal cell. Excitatory synapses (NMDA+AMPA) were placed on a single basal dendrite 125 µm from the soma and inhibitory (GABAA) synapses were either co-localized with the excitation (E) or placed at the soma (G). Line colors and dashing are as in a–d. (F, H) Input-output curves for compartmental model as a function of activated excitatory synapses. Each excitatory synapse in this experiment had 6 nS peak AMPA conductance. Excitatory synapses with 1.5 nS peak AMPA conductance with similar distribution of density along the dendrite gave similar results. For the cases shown, peak inhibitory conductance was 10 nS in case of dendritic inhibition case and 90 nS in case of somatic inhibition.
Figure 2
Figure 2. Summary of location effects of inhibition.
(A) % reduction in somatic depolarization caused by dendritic vs. somatic inhibition at stimulus levels subthreshold for NMDA spike initiation, averaged over subthreshold part of i/o curves like those shown in Figure 1 B,D. Bars shown are standard errors. (B) Scatter plot showing changes in NMDA spike threshold (x-axis) and height (y-axis) in response to dendritic (green symbols) and somatic (red symbols) inhibition, expressed as joint % change in spike height and threshold relative to no-inhibition control (black square at origin). Peak conductance for dendritic inhibition cases shown here was 10, 20, 30, and 40 nS, while that for somatic inhibition was 30, 60, 90, 120 and 150 nS. Each excitatory synapse in this experiment had 6 nS peak AMPA conductance. Excitatory synapses with 1.5 nS peak AMPA conductance with similar distribution of density along the dendrite gave similar results. The figure includes data from in vitro experiments (circles), detailed compartmental model (squares) and the reduced (2-compartment) steady state model (triangles). Open circles show the means of the respective in vitro data. Green and red shaded regions highlight the predominance of threshold elevation in cases of dendritic inhibition, and height suppression in cases of somatic inhibition.
Figure 3
Figure 3. Dendritic spike height is not affected by somatic inhibition.
(A) Experimental setup for testing somatic inhibition. Red electrode shows dendritic site of stimulation, blue electrode shows somatic site of GABA iontophoresis. (B) Voltage traces (top) and dendritic calcium signal (bottom) for control case (black) and with somatic inhibition (blue). (C) Bar plots compare dendritic calcium signal peaks (control: black, GABA: blue) for EPSPs that were both subthreshold and suprathreshold to NMDA spikes. (D) Morphology and stimulation set up in detailed compartmental model. Red square indicates location of excitatory synapses on a single dendrite, while the blue square indicates somatic location of inhibitory synapses. (E) Membrane potential at the soma and dendritic location for increasing levels of excitation (6 nS per synapse). Black traces indicate control, while blue traces indicate co-stimulation of somatic inhibitory synapses (peak conductance = 90 nS). (F) I/O curves at the soma and at the dendritic location for peak Vm for control (black) and somatic inhibition (blue).
Figure 4
Figure 4. Dendritic vs. somatic inhibition in a 2-compartment model.
(A,C) Two-compartment models (see Methods for details) contained an NMDA conductance in the dendrite (node d) scaled by Nsyn and an inhibitory conductance either in the dendrite (A) or soma (C). (B, D) Input-output curves in the somatic compartment (node s) with and without inhibition. Curves reproduce main features of input-output curves from experiments and detailed compartmental modeling results (Figure 1).
Figure 5
Figure 5. Mechanisms underlying stereotypical NMDA spike height.
(A) A single voltage compartment containing NMDA and leak conductances. (B) I–V curves for two levels of leak and threshold NMDA conductances for the circuit shown in A. Increasing leak scales the leak I–V curve (shown in green, sign reversed and reflected below the x-axis). Levels of NMDA conductance shown (red I–V curves) correspond to two different values of channels Nsyn in equation 1 and were just suprathreshold for NMDA spike generation in control and increased leak cases. Spike heights are same in two cases (black and blue dashed arrows). (C) Equivalent circuits from perspective of dendritic compartment (node s) for cases with no inhibition (c1), with dendritic inhibition (c2) or with somatic inhibition (c3), (EL = EI). Shaded grey areas indicate all conductances contributing to total leak from the perspective of the dendritic compartment.
Figure 6
Figure 6. Effect of dendritic inhibition depends on location relative to excitation.
(A,B) Voltage traces and i/o curves for whole-cell somatic recordings in vitro. In A, uncaging site was 75 µm from the soma while inhibition was 120 µm from the soma. Sites of excitation and inhibition were reversed in B. Excitation was delivered at least 10 ms after the iontophoresis. (C,D) I/O curves from the detailed compartmental model. Excitatory synapses (containing NMDA+AMPA conductance) were placed on a basal dendrite 125 µm from the soma. Inhibitory synapses (GABAA) were placed either 80 µm more distal than the excitation (C) or 80 µm more proximal, i.e. on-the-path to the soma (D). The red and blue rectangles in the C and D insets illustrate the spread of E and I types of synapses at their respective locations on the dendrite. The synapses were placed 0.5 µm apart as illustrated in Figure 1. Same number of GABAA- type synapses were activated in C,D. Each excitatory synapse in the simulations had 6 nS peak AMPA conductance. For the cases shown, peak inhibitory conductance was 20 nS.
Figure 7
Figure 7. Summary of location-dependent effects of dendrite-targeting inhibition.
This is expressed as joint % change in spike height and threshold relative to no-inhibition control (black square at origin). Red and green shaded areas were carried over from Figure 2 to indicate general trends for dendritic vs. somatic inhibition. In vitro data (red and green circles) were collected from different dendrites at different distal or on-the-path locations; separation distances between excitation and site of GABA iontophoresis are indicated in figure next to each data point. Results from detailed compartmental model are shown to provide context, including one representative location of more distal inhibition (open green squares) and three locations of on-the-path inhibition (open orange squares). Iso-inhibition and iso-location lines are splines fitted to the data points from the detailed compartmental model. Co-localized (filled green squares) and somatic (filled red squares) inhibition locations are shown for reference. In case of data points from the detailed compartmental model, size is indicative of strength. Simulations were carried out on an un-branched dendrite, though the results were similar for other dendrites.

Similar articles

See all similar articles

Cited by 36 articles

See all "Cited by" articles


    1. Markram H, Toledo-Rodriguez M, Wang Y, Gupta A, Silberberg G, et al. Interneurons of the neocortical inhibitory system. Nat Rev Neurosci. 2004;5:793–807. - PubMed
    1. Burkhalter A. Many specialists for suppressing cortical excitation. Front Neurosci. 2008;2:155–167. - PMC - PubMed
    1. Chance F, Abbott L, Reyes A. Gain modulation from background synaptic input. Neuron. 2002;35:773–782. - PubMed
    1. Hao J, Wang XJ, Dan Y, Poo MM, Zhang XH. An arithmetic rule for spatial summation of excitatory and inhibitory inputs in pyramidal neurons. Proc Natl Acad Sci U S A. 2009;106:21906–21911. - PMC - PubMed
    1. Holt G, Koch C. Shunting Inhibition Does Not Have a Divisive Effect on Firing Rates. Neural Comput. 1997;9:1001–1013. - PubMed

Publication types