Prefrontal parvalbumin interneurons shape neuronal activity to drive fear expression


Synchronization of spiking activity in neuronal networks is a fundamental process that enables the precise transmission of information to drive behavioural responses1,2,3. In cortical areas, synchronization of principal-neuron spiking activity is an effective mechanism for information coding that is regulated by GABA (γ-aminobutyric acid)-ergic interneurons through the generation of neuronal oscillations4,5. Although neuronal synchrony has been demonstrated to be crucial for sensory, motor and cognitive processing6,7,8, it has not been investigated at the level of defined circuits involved in the control of emotional behaviour. Converging evidence indicates that fear behaviour is regulated by the dorsomedial prefrontal cortex9,10,11,12 (dmPFC). This control over fear behaviour relies on the activation of specific prefrontal projections to the basolateral complex of the amygdala (BLA), a structure that encodes associative fear memories13,14,15. However, it remains to be established how the precise temporal control of fear behaviour is achieved at the level of prefrontal circuits. Here we use single-unit recordings and optogenetic manipulations in behaving mice to show that fear expression is causally related to the phasic inhibition of prefrontal parvalbumin interneurons (PVINs). Inhibition of PVIN activity disinhibits prefrontal projection neurons and synchronizes their firing by resetting local theta oscillations, leading to fear expression. Our results identify two complementary neuronal mechanisms mediated by PVINs that precisely coordinate and enhance the neuronal activity of prefrontal projection neurons to drive fear expression.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Firing of distinct dmPFC INs oppositely correlates with fear expression.
Figure 2: Prefrontal type 2 PVINs control fear expression.
Figure 3: dmPFC PVINs disinhibit PNs during fear expression.
Figure 4: Inhibition of PVINs induces theta phase resetting.
Figure 5: Synchronization of dmPFC PNs during fear expression.


  1. 1

    Singer, W. Neuronal synchrony: a versatile code for the definition of relations? Neuron 24, 49–65,–111–125 (1999)

    Article  Google Scholar 

  2. 2

    Buzsáki, G. & Draguhn, A. Neuronal oscillations in cortical networks. Science 304, 1926–1929 (2004)

    ADS  Article  Google Scholar 

  3. 3

    Womelsdorf, T. et al. Modulation of neuronal interactions through neuronal synchronization. Science 316, 1609–1612 (2007)

    ADS  CAS  Article  Google Scholar 

  4. 4

    Royer, S. et al. Control of timing, rate and bursts of hippocampal place cells by dendritic and somatic inhibition. Nature Neurosci. 15, 769–775 (2012)

    CAS  Article  Google Scholar 

  5. 5

    Cobb, S. R., Buhl, E. H., Halasy, K., Paulsen, O. & Somogyi, P. Synchronization of neuronal activity in hippocampus by individual GABAergic interneurons. Nature 378, 75–78 (1995)

    ADS  CAS  Article  Google Scholar 

  6. 6

    Benchenane, K. et al. Coherent theta oscillations and reorganization of spike timing in the hippocampal- prefrontal network upon learning. Neuron 66, 921–936 (2010)

    CAS  Article  Google Scholar 

  7. 7

    Friedrich, R. W., Habermann, C. J. & Laurent, G. Multiplexing using synchrony in the zebrafish olfactory bulb. Nature Neurosci. 7, 862–871 (2004)

    CAS  Article  Google Scholar 

  8. 8

    Riehle, A., Grun, S., Diesmann, M. & Aertsen, A. Spike synchronization and rate modulation differentially involved in motor cortical function. Science 278, 1950–1953 (1997)

    ADS  CAS  Article  Google Scholar 

  9. 9

    Burgos-Robles, A., Vidal-Gonzalez, I. & Quirk, G. J. Sustained conditioned responses in prelimbic prefrontal neurons are correlated with fear expression and extinction failure. J. Neurosci. 29, 8474–8482 (2009)

    CAS  Article  Google Scholar 

  10. 10

    Tang, J. et al. Pavlovian fear memory induced by activation in the anterior cingulate cortex. Mol. Pain 1, 6 (2005)

    Article  Google Scholar 

  11. 11

    Vidal-Gonzalez, I., Vidal-Gonzalez, B., Rauch, S. L. & Quirk, G. J. Microstimulation reveals opposing influences of prelimbic and infralimbic cortex on the expression of conditioned fear. Learn. Mem. 13, 728–733 (2006)

    Article  Google Scholar 

  12. 12

    Corcoran, K. A. & Quirk, G. J. Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. J. Neurosci. 27, 840–844 (2007)

    CAS  Article  Google Scholar 

  13. 13

    Pape, H. C. & Pare, D. Plastic synaptic networks of the amygdala for the acquisition, expression, and extinction of conditioned fear. Physiol. Rev. 90, 419–463 (2010)

    CAS  Article  Google Scholar 

  14. 14

    Knapska, E. et al. Functional anatomy of neural circuits regulating fear and extinction. Proc. Natl Acad. Sci. USA 109, 17093–17098 (2012)

    ADS  CAS  Article  Google Scholar 

  15. 15

    LeDoux, J. E. Emotion circuits in the brain. Annu. Rev. Neurosci. 23, 155–184 (2000)

    CAS  Article  Google Scholar 

  16. 16

    Hartwich, K., Pollak, T. & Klausberger, T. Distinct firing patterns of identified basket and dendrite-targeting interneurons in the prefrontal cortex during hippocampal theta and local spindle oscillations. J. Neurosci. 29, 9563–9574 (2009)

    CAS  Article  Google Scholar 

  17. 17

    Freund, T. F. & Katona, I. Perisomatic inhibition. Neuron 56, 33–42 (2007)

    CAS  Article  Google Scholar 

  18. 18

    Blatow, M. et al. A novel network of multipolar bursting interneurons generates theta frequency oscillations in neocortex. Neuron 38, 805–817 (2003)

    CAS  Article  Google Scholar 

  19. 19

    Losonczy, A., Zemelman, B. V., Vaziri, A. & Magee, J. C. Network mechanisms of theta related neuronal activity in hippocampal CA1 pyramidal neurons. Nature Neurosci. 13, 967–972 (2010)

    CAS  Article  Google Scholar 

  20. 20

    Yoder, R. M. & Pang, K. C. Involvement of GABAergic and cholinergic medial septal neurons in hippocampal theta rhythm. Hippocampus 15, 381–392 (2005)

    CAS  Article  Google Scholar 

  21. 21

    Gabbott, P. L., Warner, T. A., Jays, P. R., Salway, P. & Busby, S. J. Prefrontal cortex in the rat: projections to subcortical autonomic, motor, and limbic centers. J. Comp. Neurol. 492, 145–177 (2005)

    Article  Google Scholar 

  22. 22

    Di Scala, G., Mana, M. J., Jacobs, W. J. & Phillips, A. G. Evidence of Pavlovian conditioned fear following electrical stimulation of the periaqueductal grey in the rat. Physiol. Behav. 40, 55–63 (1987)

    CAS  Article  Google Scholar 

  23. 23

    Herry, C. et al. Switching on and off fear by distinct neuronal circuits. Nature 454, 600–606 (2008)

    ADS  CAS  Article  Google Scholar 

  24. 24

    Sotres-Bayon, F., Sierra-Mercado, D., Pardilla-Delgado, E. & Quirk, G. J. Gating of fear in prelimbic cortex by hippocampal and amygdala inputs. Neuron 76, 804–812 (2012)

    CAS  Article  Google Scholar 

  25. 25

    Tierney, P. L., Degenetais, E., Thierry, A. M., Glowinski, J. & Gioanni, Y. Influence of the hippocampus on interneurons of the rat prefrontal cortex. Eur. J. Neurosci. 20, 514–524 (2004)

    Article  Google Scholar 

  26. 26

    McCartney, H., Johnson, A. D., Weil, Z. M. & Givens, B. Theta reset produces optimal conditions for long-term potentiation. Hippocampus 14, 684–687 (2004)

    Article  Google Scholar 

  27. 27

    Buzsáki, G., Grastyan, E., Tveritskaya, I. N. & Czopf, J. Hippocampal evoked potentials and EEG changes during classical conditioning in the rat. Electroencephalogr. Clin. Neurophysiol. 47, 64–74 (1979)

    Article  Google Scholar 

  28. 28

    Rizzuto, D. S. et al. Reset of human neocortical oscillations during a working memory task. Proc. Natl Acad. Sci. USA 100, 7931–7936 (2003)

    ADS  CAS  Article  Google Scholar 

  29. 29

    Livneh, U. & Paz, R. Amygdala-prefrontal synchronization underlies resistance to extinction of aversive memories. Neuron 75, 133–142 (2012)

    CAS  Article  Google Scholar 

  30. 30

    Chang, C. H., Berke, J. D. & Maren, S. Single-unit activity in the medial prefrontal cortex during immediate and delayed extinction of fear in rats. PLoS ONE 5, e11971 (2010)

    ADS  Article  Google Scholar 

Download references


We thank members of the Herry laboratory, K. Benchenane and D. Dupret for comments on the manuscript, K. Deisseroth and E. Boyden for generously sharing material, J. Bacelo, S. Wolff and P. Tovote for technical and computational assistance, the Bordeaux Imaging center of the University of Bordeaux, and C. Poujol and S. Marais for technical assistance with microscopy. This work was supported by grants from the French National Research Agency (ANR-2010-BLAN-1442-01; ANR-10-EQPX-08 OPTOPATH), the European Research Council (ERC) under the European Union’s Seventh Framework Program (FP7/2007-2013)/ERC grant agreement no. 281168, a Fonds AXA pour la recherche doctoral fellowship (J.C.) and the Conseil Regional d’Aquitaine. T.C.M.B is a fellow of Ecole de l’Inserm Liliane Bettencourt-MD-PhD program, France.

Author information




J.C., F.C., R.R.R., N.K., C.G.-C., H.W., A.A., J.B. and T.C.M.B. performed the experiments and analysed the data. J.C. and C.H. designed the experiments and wrote the paper.

Corresponding author

Correspondence to Cyril Herry.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Separation of putative principal neurons and putative interneurons.

a, Left panel, superimposed waveforms recorded from two different units. Right panel, spikes originating from individual units were sorted using three-dimensional principal-component analysis. b, Corresponding auto-correlograms, colour-coded as in a, displaying clear refractory periods. c, Among the population of dmPFC neurons displaying significant excitatory or inhibitory CS+-evoked responses (n = 493), 71.2% were classified as putative principal neurons (PNs, blue circles, n = 351) and 28.8% as putative interneurons (INs, red circles, n = 142) using an unbiased unsupervised cluster-separation algorithm based on three electrophysiological properties: firing frequency, spike half-width and spike area under waveform (AUP) peak. Inset, average waveform of a representative PN and IN illustrating the methodology used to quantify spike width (SW) and the spike segment used to calculate the AUP. d, Top panel, representative cross-correlogram performed between a putative inhibitory IN and a non-identified neuron showing a short-latency, presumably monosynaptic, inhibitory interaction (7 pairs identified among putative INs, no inhibitory interaction among putative PNs). Bottom panel, representative cross-correlogram between a putative PN and a non-identified neuron showing a short-latency, possibly monosynaptic, excitatory interaction (20 pairs identified among PNs, no excitatory interaction from putative INs). Reference events correspond to the spikes of the pre-synaptic neuron (dashed line at time 0, bins of 0.5 ms). Grey circles represent neurons that were not tone-responsive.

Extended Data Figure 2 CS+-evoked firing patterns and inhibitory interactions of putative INs.

a, Left panel, distribution of the fraction of variance for the 20 first principal components (PCs) obtained with principal component analysis (PCA). PC1, which accounted for more than 20% of variance of the data set, was used for the analysis. Middle panel, first principal-component coefficients representing the main firing pattern evoked by CS+ (CS+ onset at time 0) of the IN data set. Right panel, distribution of dmPFC IN correlation coefficients with PC1. The dashed lines indicate the levels of significance (P < 0.001). Among the 142 INs, 83 (58.5%) displayed a significant positive (n = 68, 48%, dark red bars) or negative (n = 15, 10.6%, light red bars) correlation with PC1, whereas 41.5% INs (n = 59, grey bars) did not. b, Raster plots and PSTH of individual INs negatively correlated (left part, type 2 IN), not correlated (middle part) or positively correlated (right part, type 1 IN) with PC1. Type 1 INs were excited, whereas type 2 INs were inhibited by CS+. Bins of 10 ms. c, PSTH of all type 1 (n = 68) and type 2 (n = 15) INs illustrating the CS+-evoked responses (Post-FC, Ext. or Ret. Sessions, CS+ 1–4). Bins of 10 ms. d, Individual (type 1 INs, dark red dots; type 2 INs, light red dots) and averaged (red dots) latencies of the first significant time bin (z score < −1.65 or > +1.65) following CS+ for type 1 and type 2 INs recorded simultaneously (n = 7 pairs recorded in 5 mice). CS+-evoked excitation in type 1 INs preceded CS+-evoked inhibition in type 2 INs (mean latency: type 1, 24.3 ± 2 ms; type 2, 38.6 ± 4.6 ms; paired t-test, *P < 0.05). Error bars, mean ± s.e.m. e, Cross-correlation analysis performed between a type 1 and a type 2 IN recorded simultaneously outside CS. The cross-correlogram shows a short latency, potentially monosynaptic, inhibitory interaction. Reference event, spikes of the type 1 IN (dashed line at time 0). Bins of 5 ms. f, Locations of recording sites and mean firing frequencies of type 1 (T1, n = 68) and type 2 (T2, n = 15) INs (Mann–Whitney test, **P < 0.01; Cg1, anterior cingulate cortex; PL, prelimbic area; IL, infralimbic area). g, Firing modulation of representative type 1 and type 2 INs with dmPFC theta oscillations filtered in the 8–12-Hz range (12-min recordings). Bins of 10°. h, Mean strength of firing synchronization to local theta oscillations as measured with the mean resultant length (MRL) vector (left panel, Mann–Whitney test, type 1 versus type 2, ***P < 0.001) and distribution of the preferred phases (right panels) for type 1 and type 2 INs significantly phase-locked to theta oscillations (type1, n = 29/68; type 2, n = 15/15).

Extended Data Figure 3 Anatomical characterization of AAV-mediated ArchT-GFP expression in PV-IRES-Cre transgenic mice and electrophysiological characteristics of ArchT , ChR2 and GFP PV-IRES-Cre-infected PV neurons.

a, Representative confocal micrographs used for PV and GFP co-localization assessment. Left panel, ArchT-GFP labelled with anti-GFP Alexa 488; middle panel, PV immunofluorescence; right panel, merge. Single optical slices, in the same focal plane. b, Quantitative analysis of viral infection specificity and efficacy. Pie charts show the numbers of neurons positive for GFP and/or PV in two mice (left and middle charts) and averaged proportions (right chart). c, Representative ChR2- (left) and ArchT-evoked (right) currents recorded from PVINs with somata located in layer 2/3 of the dmPFC (1 s light simulation). d, Representative optically evoked action potential firing and inhibition of PVINs expressing ChR2 (left, 500-ms blue light pulses) or ArchT, respectively (right, 250-ms yellow light pulse during a 250-pA current pulse injection). e, Left panel, changes in firing frequency of PVINs expressing GFP (white dots, n = 7), ChR2 (blue dots, n = 5) or ArchT (yellow dots, n = 8) upon injection of increasing current pulses (current pulses range, 0–400 pA). No significant differences were observed between groups. Right panel, resting membrane potentials of INs expressing GFP (white bar, n = 7), ChR2 (blue bar, n = 5) or ArchT (yellow bar, n = 8). No significant differences were observed between groups (unpaired t-tests).

Extended Data Figure 4 Type 2 PVINs mediate conditioned fear responses.

a, z-score transformation of CS+-evoked firing of a non-type-2 IN for sound pips outside (No freez.) or inside (Freez.) freezing periods during the extinction session (CS+1–12; No freez., 141 pips; Freez., 156 pips). This neuron was not classified as a type 1 or type 2 IN. b, Left panel, raster plot illustrating optogenetic identification of the same non-type-2 IN as ArchT-expressing (that is, PV-expressing). Right panel, mean z-score transformation of all non-type-2 INs identified as PV-expressing INs (n = 4; light-pulse duration, 250 ms; 108 stimulation trials). c, z-score transformation of CS+-evoked firing of a type 2 IN for No freez. and Freez. periods during the extinction session (CS+1–12; No freez., 141 pips; Freez., 156 pips). d, Raster plot illustrating optogenetic identification of the same type 2 IN as ArchT-expressing (that is, PV-expressing) (light-pulse duration, 250 ms; 108 stimulation trials). e, CS+-evoked changes in firing rate in two type 2 PVINs identified with optogenetic, and corresponding freezing scores of the two mice in which they were recorded (dots, mean z-score 150 ms post CS; bars, blocks of 4 CS+ presentation each, both during the second extinction session; light-pulse duration, 250 ms; 108 stimulation trials). Light-induced inhibition of PV, including type 2 INs, reinstated freezing behaviour. Error bars indicate mean ± s.e.m.

Extended Data Figure 5 Optogenetic inhibition of prefrontal PVINs induces place aversion.

a, On day 1, GFP- and ArchT-infected mice (n = 11 and 13, respectively) were exposed to a two-compartment place aversion apparatus during 15 min. Following pre-exposure, the most visited compartment was selected for each animal. On day 2, systematic yellow-light-induced inhibition of PVINs was triggered only in the most visited compartment during a 15-min exposure session. On day 3, GFP and ArchT infected mice (n = 6 in both cases) were re-exposed to the place aversion apparatus during 15 min to evaluate the long-term effect of yellow-light stimulation during day 2. b, Time spent in the most and less visited compartments on day 1 for individual infected mice (GFP and ArchT ). c, Average percentage of time spent in the most visited compartment on days 1, 2 and 3 for GFP- and ArchT-infected mice. A one-way analysis of variance (ANOVA) repeated measures performed on values from the GFP or the ArchT group revealed a significant effect only for the ArchT group (ArchT, F2,10 = 4.234, P < 0.05; GFP, F2,10 = 0.950, P = 0.4191). Post-hoc analysis revealed that on day 2, light inhibition of PVINs induced an aversion of the most visited compartment for ArchT infected animals in comparison to day 1 (ArchT mice, day 1 versus day 2, paired t-test, **P < 0.01) and to GFP controls on day 2 (day 2, ArchT versus GFP, unpaired t-test, *P < 0.05; 250-ms pulses delivered at 0.9 Hz). On day 3, ArchT mice did not avoid the most visited compartment any more (ArchT mice, day 2 versus day 3, unpaired t-test, * P < 0.05). Error bars indicate mean ± s.e.m.

Extended Data Figure 6 Optogenetic activation of PVINs inhibits principal neurons and reduces freezing behaviour.

a, Raster plots and peristimulus time histograms illustrating the CS+-evoked excitation of a representative PN (left panel, Post FC, CS+ presentations 1–4, 108 pips) and its blockade upon optogenetic-induced activation of PVINs (right panel, CS+ presentations 5–8; light-pulse duration, 250 ms; 108 pips + stimulation trials) during the Post-FC session. b, z-score-transformed peristimulus time histogram showing PNs inhibition (n = 7) following optogenetic-evoked activation of PVINs during CS+ presentations (Post-FC session, CS+ presentations 5–8; light-pulse duration, 250 ms; 108 stimulation trials). c, Freezing behaviour (bars, n = 3 mice, block of 4 CS+) and CS+-evoked firing changes of PNs (red dots, n = 7 neurons, mean z-score 100 ms post CS) before and in response to light-induced activation of PVINs during Post-FC sessions (light pulse duration, 250 ms; 108 stimulation trials; CS+ 1–4 and 5–8, respectively). Optogenetic activation of PVINs inhibited PNs and reduced conditioned freezing behaviour (Wilcoxon signed-rank test, *P < 0.05). d, z-score transformed peristimulus time histogram showing CS+-evoked excitation of PNs (n = 3) exhibiting antidromic responses to BLA stimulations (Post-FC, CS and CS+ presentations 1–4, 108 pips each). These three neurons were included in the seven neurons for which CS+-evoked excitation was blocked by light excitation of PVINs (a and b). Error bars indicate mean ± s.e.m.

Extended Data Figure 7 Transient amplitude increase and phase reset of local theta oscillations during fear expression.

a, Left panel, power spectrum of the non-filtered dmPFC LFPs recorded during Post-FC sessions (n = 28 mice) for non-freezing (No freez.) and freezing (Freez.) periods showing a prominent 8–12-Hz component (that is, theta) only during non-freezing periods. Right panel, normalized theta power (8–12 Hz) for freezing and non-freezing periods during Post-FC sessions (n = 28 mice, Wilcoxon signed-rank test ***P < 0.001). b, Top panels, non-filtered dmPFC LFP traces selected on the basis of prominent theta oscillations illustrating the transient increase in amplitude and phase reset of theta oscillations in response to CS+ (Post-FC, 1 trial). Bottom-left panel, representative dmPFC 8–12-Hz LFP traces illustrating the phase reset and transient amplitude increase of theta oscillations in response to CS+ or CS presentations (Post-FC, 27 pips each). Bottom-right panel, average ratio of LFP theta power (500 ms post CS or 500 ms pre CS) in response to CS and CS+ pips. This analysis revealed a larger transient increase in LFP upon CS+ presentations (Post-FC, n = 28 mice, CS versus CS+, paired t-test, ***P < 0.001). c, Left panel, representative dmPFC LFP traces filtered in the 8–12-Hz range, illustrating the phase resetting of theta oscillations during presentations of CS pips associated with no freezing or freezing behaviour (Post-FC, 27 pips). Right panel, quantification of the variance of the first theta peak occurrence following pip presentations in freezing and non-freezing periods (Post-FC, n = 28 mice, No freez. versus Freez., paired t-test, ***P < 0.001). A small variance corresponds to a strong theta phase resetting. d, Quantification of the time variance of the first theta peak following CS and CS+ presentations or No freez. and Freez. periods for extinction and retrieval sessions (extinction, CS presentations and CS+ presentations 1–4, n = 28 mice; retrieval, CS and CS+, n = 21 mice; CS versus CS+, paired t-test, ***P < 0.001; No freez. versus Freez., paired t-test, ***P < 0.001). Error bars indicate mean ± s.e.m.

Extended Data Figure 8 Targeted reversible inactivation of the medial septum does not block conditioned fear expression or theta phase resetting.

a, Locations of injection sites in the medial septum (MS) and composite transmission light and epifluorescence micrograph showing the MS targeted injection of muscimol (MUS.) covalently bound to a fluorescent tag (right, dipyrromethene boron difluoride (BODIPY), red). b, Experimental design and mean freezing values of fear conditioned mice (n = 6) before (Test Pre-MUS.), following (Test MUS.), and one day after (Post-MUS.) injections of MUS in the MS. Following fear conditioning, targeted inactivation of the MS had no effect on basal locomotor activity or CS+-evoked freezing responses (paired t-tests). c, Illustrative raw and filtered (8–120-Hz) LFP traces recorded in the dorsal CA1 (dCA1) before and following MUS injections in the MS. d, Left panel, power spectra of dCA1 LFPs before, following and 1 day after MS inactivation. Right panel, quantification of dCA1 LFP power (8–12 Hz) before, during and after MS inactivation. MS inactivation significantly reduced dCA1 theta power (n = 6 mice, Pre-MUS. versus MUS., paired t-test, **P < 0.01; Post-MUS. versus MUS., paired t-test, **P < 0.01, ns, not significant). e, Resetting of prefrontal theta oscillations. Left panel, representative dmPFC LFP traces filtered in the 8–12-Hz range (Test-MUS., first CS+). Right panel, quantification of the time variance of the first theta peak following CS+ presentations before, following and 1 day after MS inactivation (Pre-MUS., MUS., Post-MUS., CS+ presentations 1–4, paired t-tests). MS inactivation had no effect on dmPFC theta phase resetting upon CS+ presentations. Error bars indicate mean ± s.e.m.

Extended Data Figure 9 Optogenetic inhibition of prefrontal PVINs resets local theta phase.

Left panel, representative 8–12-Hz filtered LFP traces, showing the resetting of local theta phase upon optogenetic inhibition of PVINs (top part, light-pulse duration, 250 ms, 27 stimulation trials; bottom part, light-pulse duration, 500 ms, 27 stimulation trials). Right panel, quantification of the time variance of theta peaks (theta peaks 1–4) following presentations of 250-ms or 500-ms light pulses (n = 9 mice, paired t-tests, *P < 0.05, ***P < 0.001, NS, not significant). Interestingly, dmPFC theta oscillations were precisely entrained for as long as PVINs were silenced. This suggests that inhibition from PVINs masks an oscillatory process in PNs.

Extended Data Figure 10 PVINs control principal-neuron theta phase locking and spiking synchronization.

a, Normalized averaged ratio of changes in coincident activity between pairs of PNs induced by CS+ and CS and corrected for changes in firing rate (Post-FC, Ext. or Ret. sessions; n = 975 pairs from 308 PNs). Dashed line indicates significant z score (P < 0.05). Bins of 30 ms. b, Mean vector length (MRL) and concentration of Von Mises fit (κ) upon CS or CS+, two measures of modulation strength in phase with theta oscillations (Post-FC, Ext. or Ret. Sessions). Only neurons significantly phase locked to theta and for which at least 50 spikes were recorded during CS+ were included (n = 45) (CS versus CS+, Wilcoxon tests, *** P < 0.001). Error bars indicate mean ± s.e.m. CS+ entrains a stronger locking of PN spikes to ongoing theta oscillations. Together with the precise timing between CS+ onset (resetting) and subsequent theta cycles, this ensures robust, coincident and timed spiking of PNs. c, Distribution of log-transformed Rayleigh’s test Z values of PN theta modulation before and upon light-induced inhibition (top, n = 41 neurons) and light-induced activation (bottom, n = 18 neurons) of PVINs (light-pulse duration, 250ms; 108 trials for each; yellow light, stimulation at the end of the behavioural session; blue light, stimulation during Post-FC session, CS+ presentations 5–8). Dashed line indicates significant theta phase locking threshold (ln (Z) = 1.1, P = 0.05). d, Theta modulation of PNs significantly phase locked to theta and displaying at least 15 spikes during No light and Light conditions. Modulation with local theta was measured with the MRL (top-left panel, n = 8 neurons, yellow light stimulation, paired t-tests, No light versus Light, *P < 0.05; bottom-left panel, n = 8 neurons, blue light stimulation, No light versus Light, * P < 0.05) and κ (top-right panel, n = 8 neurons, yellow light stimulation, paired t-tests, No light versus Light, ***P < 0.001; bottom-left panel, n = 8 neurons, blue light stimulation, No light versus Light, ***P < 0.001). Error bars indicate mean ± s.e.m. These results show that inhibiting PVINs is both sufficient to increase PNs’ modulation with local theta, and necessary for theta entrainment of PNs evoked by CS+. e, Top panel, distribution of PNs’ preferred theta phase (n = 308) during cycles around CS. The phases of LFPs were aligned to the first theta peak following CS onset to mimic phase resetting of local theta (one theta cycle before, and three theta cycles following CS were included, bins of 45°). Bottom panel, distribution of individual PNs’ preferred theta phases during theta cycles around CS showing a synchronization of PNs around the peak of the LFP (Rayleigh’s test for circular uniformity, first theta cycle post CS, P < 0.001, indicating that the circular distribution is not uniform). f, Top panel, distribution of PNs’ preferred theta phase (n = 41) during theta cycles outside light stimulation (left part, 15.8% freezing) and upon light-induced resetting of theta oscillations (right part, 36.8% freezing; one theta cycle before, and 3 theta cycles following CS were included, bins of 45°). Bottom panel, distributions of individual PNs’ preferred theta phase outside and upon light stimulation. Despite a low number of neurons and a moderate freezing induced by light inhibition of dmPFC PVINs (36.8% freezing), this analysis revealed that light-induced reset of local theta oscillations promotes neuronal synchronization of PNs (Rayleigh’s test for circular uniformity, first theta cycle post CS; Light, P < 0.00; No light, P = NS).

Supplementary information

Supplementary Information

This file contains Supplementary Methods and a Supplementary Discussion. The Supplementary Methods contain additional information about the methodology used for in vitro electrophysiology, anatomical analyses, pharmacological inactivation, electrical extracellular stimulation, virus injection and optogenetics and field potential analyses. The Supplementary Discussion discusses the role of PV INs in aversive and appetitive behaviours and the origin and influence of the changes in neuronal activity observed in dmPFC PNs during fear behaviour. (PDF 413 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Courtin, J., Chaudun, F., Rozeske, R. et al. Prefrontal parvalbumin interneurons shape neuronal activity to drive fear expression. Nature 505, 92–96 (2014).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing