Local origin of excitatory–inhibitory tuning equivalence in a cortical network

The interplay between excitation and inhibition determines the fidelity of cortical representations. The receptive fields of excitatory neurons are often finely tuned to encoded features, but the principles governing the tuning of inhibitory neurons remain elusive. In this study, we recorded populations of neurons in the mouse postsubiculum (PoSub), where the majority of excitatory neurons are head-direction (HD) cells. We show that the tuning of fast-spiking (FS) cells, the largest class of cortical inhibitory neurons, was broad and frequently radially symmetrical. By decomposing tuning curves using the Fourier transform, we identified an equivalence in tuning between PoSub-FS and PoSub-HD cell populations. Furthermore, recordings, optogenetic manipulations of upstream thalamic populations and computational modeling provide evidence that the tuning of PoSub-FS cells has a local origin. These findings support the notion that the equivalence of neuronal tuning between excitatory and inhibitory cell populations is an intrinsic property of local cortical networks.


nature portfolio | reporting summary
April 2023 Sample size across cortical layers, angled implantation along cortical layers) was pooled, resulting in a large cohort of 32 mice and nearly 3000 single units.We did not observe any substantial differences in results across the two implant configurations.For mice implanted in the anterodorsal thalamic nucleus the total number of animals (n = 8) and number of single units (n = 228) due to the small size of this brain structure and the high targeting difficulty associated with it.
Samples included all available cells that matched the classification criteria for the relevant cell type.We reasoned that due to the high heterogeneity of tuning in PoSub-FS compared to HD cells it was critical to direct our resources to maximise the sample size of this cell group.The total number of FS cells in our dataset (n = 427) is high in comparison with other studies of cortical interneuron tuning.Since HD cells show much more homogenous tuning, we judge the sample size of 97 ADN-HD cells to be sufficient to quantify the tuning properties of these neurons.This sample size is comparable to, and often exceeds, those reported in other papers about ADN-HD cells.
In optogenetic experiments, the responses cells to the optogenetic manipulation were relatively homogenous within each cell group, which in our opinion justifies the smaller sample size.
Data exclusions For the main analysis, the only exclusion criterion was off-target implantation, as judged by absence of sharply-tuned HD cells and histological assessment.For the analysis of tuning stability across environments, only mice that underwent the additional exploration session in a triangular environment (n = 20) were included.For the cue rotation analysis, only the mice that underwent the cue rotation experiment and in which a substantial effect of cue rotation on receptive fields was observed were included (n = 6).Mice which underwent the cue rotation protocol but in which HD cells did not remap following cue rotation (n = 5) were excluded.All the above criteria were pre-established.
For REM sleep analysis, only mice in which at least 2 minutes of REM sleep was recorded (n = 28) were included, and those with less than 2 min of REM sleep (n = 6) were excluded.We observed that less than 2 min of REM sleep is not enough to reliably calculate the crosscorrelations between cell pairs or perform the manifold analysis.This criterion was not pre-established.
For the purpose of comparison with simulated data (Extended Data Figure 5) HD information for the recorded population of FS cells was calculated after exclusion of two outliers (out of 427 cells).Both outliers had narrow waveforms and high firing rates consistent with putative FS cells, but their tuning curves were indistinguishable from canonical HD cells, which resulted in HD information scores orders of magnitude higher than other putative FS cells (see Figure 1f).

Replication
For PoSub recordings, results from mice implanted vertically (n = 14) were reproduced in the next cohort of mice implanted parallel to cortical layers (n = 18).All other experiments were conducted as single cohorts and replication was not attempted.
Randomization Allocation of animals to either ADN or PoSub implant group was not random as these experiments were carried out in sequence and over a period of 3 years.For optogenetic experiments, animals were randomly allocated to either ArchT or control group.

Blinding
Experimenters were not blinded to the group allocation during data collection or analysis since group allocation was easily deducible due to different appearance of the probe implant, obvious differences in neural activity patterns between PoSub and ADN and strong effects of the optogenetic manipulation on neural activity.

Reporting for specific materials, systems and methods
We require information from authors about some types of materials, experimental systems and methods used in many studies.Here, indicate whether each material, system or method listed is relevant to your study.If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.