Brief Communication

Excitatory connections between the prelimbic and infralimbic medial prefrontal cortex show a role for the prelimbic cortex in fear extinction

  • Nature Neurosciencevolume 21pages654658 (2018)
  • doi:10.1038/s41593-018-0137-x
  • Download Citation


We elucidated the intrinsic circuitry of the medial prefrontal cortex and its role in regulating fear extinction, using neuronal tracing and optogenetic stimulation in vitro and in vivo. We show that pyramidal neurons in layer 5/6 of the prelimbic medial prefrontal cortex project to pyramidal cells in layer 5/6 of the infralimbic cortex. Activation of this connection enhances fear extinction, redefining the role of the prelimbic cortex in extinction learning.

  • Subscribe to Nature Neuroscience for full access:



Additional access options:

Already a subscriber?  Log in  now or  Register  for online access.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Marek, R., Strobel, C., Bredy, T. W. & Sah, P. J. Physiol. (Lond.) 591, 2381–2391 (2013).

  2. 2.

    LeDoux, J. E. Annu. Rev. Neurosci. 23, 155–184 (2000).

  3. 3.

    Delamater, A. R. Q. J. Exp. Psychol. B 57, 97–132 (2004).

  4. 4.

    Sah, P., Faber, E. S., Lopez De Armentia, M. & Power, J. Physiol. Rev. 83, 803–834 (2003).

  5. 5.

    Dejean, C. et al. Nature 535, 420–424 (2016).

  6. 6.

    Milad, M. R. & Quirk, G. J. Nature 420, 70–74 (2002).

  7. 7.

    Sierra-Mercado, D., Padilla-Coreano, N. & Quirk, G. J. Neuropsychopharmacology 36, 529–538 (2011).

  8. 8.

    Hoover, W. B. & Vertes, R. P. Brain Struct. Funct. 212, 149–179 (2007).

  9. 9.

    Ji, G. & Neugebauer, V. Mol. Brain 5, 36 (2012).

  10. 10.

    van Aerde, K. I., Heistek, T. S. & Mansvelder, H. D. PLoS One 3, e2725 (2008).

  11. 11.

    Englund, C. et al. J. Neurosci. 25, 247–251 (2005).

  12. 12.

    Senn, V. et al. Neuron 81, 428–437 (2014).

  13. 13.

    Maren, S. & Quirk, G. J. Nat. Rev. Neurosci. 5, 844–852 (2004).

  14. 14.

    Bukalo, O. et al. Sci. Adv. 1, e1500251 (2015).

  15. 15.

    Do-Monte, F. H., Manzano-Nieves, G., Quiñones-Laracuente, K., Ramos-Medina, L. & Quirk, G. J. J. Neurosci. 35, 3607–3615 (2015).

  16. 16.

    Vidal-Gonzalez, I., Vidal-Gonzalez, B., Rauch, S. L. & Quirk, G. J. Learn. Mem 13, 728–733 (2006).

  17. 17.

    Burgos-Robles, A., Vidal-Gonzalez, I. & Quirk, G. J. J. Neurosci. 29, 8474–8482 (2009).

  18. 18.

    Vertes, R. P. Synapse 51, 32–58 (2004).

  19. 19.

    Paxinos, G. & Watson, C. The Rat Brain in Stereotaxic Coordinates, 6th ed. (New York, 2007)

  20. 20.

    Marek, R. et al. Nat. Neurosci. 21, 384–392 (2018).

  21. 21.

    Gabbott, P. L., Warner, T. A., Jays, P. R., Salway, P. & Busby, S. J. J. Comp. Neurol. 492, 145–177 (2005).

Download references


We thank R. Tweedale and A. Woodruff for comments on our manuscript and the Sah Laboratory for feedback on the experiments. This work was supported by grants to P.S. from the Australian Research Council (CE140100007) and the National Health and Medical Research Council. Imaging was performed at the Queensland Brain Institute’s Advanced Microscopy Facility.

Author information


  1. Queensland Brain Institute, The University of Queensland, Brisbane, Australia

    • Roger Marek
    • , Li Xu
    • , Robert K. P. Sullivan
    •  & Pankaj Sah


  1. Search for Roger Marek in:

  2. Search for Li Xu in:

  3. Search for Robert K. P. Sullivan in:

  4. Search for Pankaj Sah in:


P.S. supervised all experiments. P.S. and R.M. designed the experiments. R.M. collected and analyzed the data. L.X. generated some of the viral vectors. R.K.P.S. performed the immunohistochemistry. R.M. and P.S. wrote the manuscript; all authors read and edited the manuscript.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Pankaj Sah.

Integrated supplementary information

  1. Supplementary Figure 1 Intrinsic firing properties of mPFC pyramidal neurons.

    (a) Illustration of the intrinsic firing types for repetitive firing neurons (RS), accommodating cells (AC), bursting firing cells (BS) found in L2/3 and L5/6 of the Il and PL. The traces show hyperpolarizing currents and multiple spiking following a 600 ms long negative or positive current injection, respectively. Traces are shown for current injections at threshold for firing (1T, grey) and two times threshold firing (2T, black). The line graphs show the total number of spikes over time at 1T (black dots), 2T (black squares) and 3 times threshold firing (white dots). Note that initial bursting type neurons often show large after-hyperpolarizing currents following the burst. (b) The bar graph depicts the distribution of pyramidal neurons with specific firing patterns across L2/3 and L5/6 of the PL and IL. The dotted line shows the 50% mark. Quantification given in Supplementary Table 1.

  2. Supplementary Figure 2 PL targeting to backlabel IL neurons using retrogradely transported AAV-GPF.

    (a) Actual brightfield images (left) and corresponding atlas images that illustrate the infection spread (right; 2 weeks following stereotactic injections) of the retrograde AAV-GFP in dorsal (top left) to caudal sections (bottom right) of the mPFC containing the PL and IL. The atlas images also illustrate the core (dark yellow) and maximal spread (light yellow) of the infection. DAPI: blue; Scale bar: 1 mm. (b) Confocal image showing back-labeled neurons in the IL with most PL-projecting neurons located in L2/3 (Quantification in Fig. 1i). Scale bar: 200 μm.

  3. Supplementary Figure 3 Quantification of the PL infection spread confirms neuronal infection of the PL neurons without spread to the IL.

    (a) Example coronal sections (brightfield images) from dorsal (top left) to caudal sections (bottom right) of the mPFC that contains PL and IL. The atlas images illustrate the core (dark yellow) and maximal spread (light yellow) of the infection. DAPI: blue; scale bar: 1 mm. (b) Quantification of the infection area from rostral (left bar) to caudal sections (right bar). Note that most caudal sections only contain YFP fluorescence from fiber tract labeling (7 slices from 4 animals). Error bars show mean ± SEM.

  4. Supplementary Figure 4 Adeno-associated virus (AAV) injection of channelrhodopsin-YFP into PL transduces pyramidal cells in all layers.

    (a) Representative micrographs from a PL injection (green) showing neurons in layer 1 (L1), layer 2/3 (L2/3) and deep layers 5/6 (L5/6). Scale bar: 100 μm. Whole cell recordings were obtained from cells expressing ChR2-YFP and light stimulation (blue bar) evoked a spike train in current clamp (b) and a sustained inward current in voltage clamp (c). (d) Cell counts in L2/3 (black) and L5/6 (grey) show a high infection rate. Numbers at the bottom of the bars denote the number of recorded neurons (total data from 3 animals). Infected neurons showed repetitive firing to 20 Hz (e) and 50 Hz (f) train stimulation (20x 5 ms pulses).

  5. Supplementary Figure 5 Monosynaptic PL projections to the IL.

    (a) Example of evoked postsynaptic responses of PL innervated IL neurons, together with the corresponding jitter of EPSC onset (b). (c) Application of the sodium channel blocker tetrodotoxin (TTX) eliminated the response (red), which was restored by adding the potassium channel blocker 4-aminopyridine (4-AP; green) to enhance the neurotransmitter release confirming that the response was monosynaptic. Data replicated 3 times in 3 animals.

  6. Supplementary Figure 6 Optical high frequency adaptation in the PL to IL projection.

    Optical 30 Hz PL terminal stimulation (HFS) for 1 second (15 ms pulses) evoked adapting EPSPs in IL principal neurons. Blue bars denote the 470 nm light stimulation. Data replicated 6 times.

  7. Supplementary Figure 7 Optogenetic silencing of PL-to-IL projection impairs fear extinction learning in the early phase.

    (a) Schematic showing the angled injection of AAV-retro-cre combined with AAV-TdTomato in the IL and AAV-DIO-ArchT-GPF injection (or AAV-GFP for controls) in the PL. Optic fibers were also placed in the PL. (b) examples of viral injection sites in the IL (TdTomato, red), and the PL (ArchT-GFP, green) with the corresponding anterior/posterior location from Bregma. DAPI: blue). Scale bars: main panel: 1 mm; enlarged inset; 100 μm. (c) Scoring of the fear response (as lever pressing ratio) in control animals (n = 5) black bars) and ArchT-stimulated group (n = 4; red bars). Light stimulation significantly increased the fear response during early extinction. FC: after fear conditioning; Ext: extinction. Two-way ANOVA with repeated measures: F(4, 28) = 5.98; early extinction: unpaired one-sided t-test: * P = 0.043; full extinction over 30 CS was presented in both groups. Error bars show mean ± SEM. Dots represent data point from individual animals.

  8. Supplementary Figure 8 Circuit model for extinction of fear by the mPFC.

    Schematic of the investigated circuits, showing that predominantly L5/6 principal neurons in the PL target deep layer neurons in the IL. Some L5/6 IL neurons that receive input from the PL also project to the basolateral amygdala (BLA). Activation of this circuitry was found to enhance the extinction of learned fear (circuitry highlighted in green).

Supplementary information

  1. Supplementary Text and Figures

    Supplementary Figures 1–8

  2. Reporting Summary

  3. Supplementary Table 1

  4. Supplementary Video 1 - Illustration of the behavioral task.

    The first half of the video illustrates the behavior of control- (top) and IL-terminal stimulated animals (bottom) before (period without the tone symbol, 10 s.) and during the tone presentation (period with the tone symbol, 10 s.) following fear conditioning (initial period of extinction learning: After FC). Note that both example animals show a similar retraction from the lever during the tone presentation due to a high fear response. The second half shows the same time-points after extinction learning (test day; After Ext), with a 1 second light stimulation 100 ms after the tone onset. Note that the control animal still shows high levels of fear (retraction from lever), whereas the IL terminal-stimulated animal continued to press the lever during the tone presentation due to low fear levels.