Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Reversal of theta rhythm flow through intact hippocampal circuits

Nature Neuroscience volume 17, pages 1362–1370 (2014)Cite this article

Subjects

Abstract

Activity flow through the hippocampus is thought to arise exclusively from unidirectional excitatory synaptic signaling from CA3 to CA1 to the subiculum. Theta rhythms are important for hippocampal synchronization during episodic memory processing; thus, it is assumed that theta rhythms follow these excitatory feedforward circuits. To the contrary, we found that theta rhythms generated in the rat subiculum flowed backward to actively modulate spike timing and local network rhythms in CA1 and CA3. This reversed signaling involved GABAergic mechanisms. However, when hippocampal circuits were physically limited to a lamellar slab, CA3 outputs synchronized CA1 and the subiculum using excitatory mechanisms, as predicted by classic hippocampal models. Finally, analysis of in vivo recordings revealed that this reversed theta flow was most prominent during REM sleep. These data demonstrate that communication between CA3, CA1 and the subiculum is not exclusively unidirectional or excitatory and that reversed inhibitory theta signaling also contributes to intrahippocampal synchrony.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Two independent topographically organized oscillators in the subiculum and CA3.
Figure 2: Phase coupling between oscillators is mediated by the subicular rhythms.
Figure 3: Subicular activity influences intrinsic CA3 rhythms more than CA3 influences the subiculum.
Figure 4: Reversed GABAergic synchronization between the subiculum and CA3.
Figure 5: Slice-like hippocampi show forward flow from CA3 to the subiculum.
Figure 6: Subicular-CA3 coupling in vivo shares similar properties with the intact in vitro hippocampus.
Figure 7: Bidirectional communication between regions using theta and gamma rhythms.

Similar content being viewed by others

References

  1. Andersen, P., Morris, R.M., Amaral, D., Bliss, T. & O'Keefe, J. The Hippocampus Book (Oxford University Press, 2007).

  2. Squire, L.R. & Zola-Morgan, S. The medial temporal lobe memory system. Science 253, 1380–1386 (1991).

    Article  CAS  Google Scholar 

  3. Amaral, D.G., Dolorfo, C. & Alvarez-Royo, P. Organization of CA1 projections to the subiculum: a PHA-L analysis in the rat. Hippocampus 1, 415–435 (1991).

    Article  CAS  Google Scholar 

  4. Andersen, P., Blackstad, T.W. & Lomo, T. Location and identification of excitatory synapses on hippocampal pyramidal cells. Exp. Brain Res. 1, 236–248 (1966).

    Article  CAS  Google Scholar 

  5. Li, X.G., Somogyi, P., Ylinen, A. & Buzsaki, G. The hippocampal CA3 network: an in vivo intracellular labeling study. J. Comp. Neurol. 339, 181–208 (1994).

    Article  CAS  Google Scholar 

  6. Ishizuka, N., Weber, J. & Amaral, D.G. Organization of intrahippocampal projections originating from CA3 pyramidal cells in the rat. J. Comp. Neurol. 295, 580–623 (1990).

    Article  CAS  Google Scholar 

  7. Blackstad, T.W. Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. J. Comp. Neurol. 105, 417–537 (1956).

    Article  CAS  Google Scholar 

  8. Chrobak, J.J. & Buzsaki, G. High-frequency oscillations in the output networks of the hippocampal-entorhinal axis of the freely behaving rat. J. Neurosci. 16, 3056–3066 (1996).

    Article  CAS  Google Scholar 

  9. Behrens, C.J., van den Boom, L.P., de Hoz, L., Friedman, A. & Heinemann, U. Induction of sharp wave-ripple complexes in vitro and reorganization of hippocampal networks. Nat. Neurosci. 8, 1560–1567 (2005).

    Article  CAS  Google Scholar 

  10. Both, M., Bahner, F., von Bohlen und Halbach, O. & Draguhn, A. Propagation of specific network patterns through the mouse hippocampus. Hippocampus 18, 899–908 (2008).

    Article  Google Scholar 

  11. Rolls, E.T. & Kesner, R.P. A computational theory of hippocampal function, and empirical tests of the theory. Prog. Neurobiol. 79, 1–48 (2006).

    Article  CAS  Google Scholar 

  12. Buzsáki, G. The hippocampo-neocortical dialogue. Cereb. Cortex 6, 81–92 (1996).

    Article  Google Scholar 

  13. Marr, D. Simple memory: a theory for archicortex. Phil. Trans. R. Soc. Lond. B 262, 23–81 (1971).

    CAS  Google Scholar 

  14. Hasselmo, M.E., Bodelon, C. & Wyble, B.P. A proposed function for hippocampal theta rhythm: separate phases of encoding and retrieval enhance reversal of prior learning. Neural Comput. 14, 793–817 (2002).

    Article  Google Scholar 

  15. Patel, J., Fujisawa, S., Berenyi, A., Royer, S. & Buzsaki, G. Traveling theta waves along the entire septotemporal axis of the hippocampus. Neuron 75, 410–417 (2012).

    Article  CAS  Google Scholar 

  16. Lubenov, E.V. & Siapas, A.G. Hippocampal theta oscillations are travelling waves. Nature 459, 534–539 (2009).

    Article  CAS  Google Scholar 

  17. Amaral, D.G. & Witter, M.P. The three-dimensional organization of the hippocampal formation: a review of anatomical data. Neuroscience 31, 571–591 (1989).

    Article  CAS  Google Scholar 

  18. van Strien, N.M., Cappaert, N.L. & Witter, M.P. The anatomy of memory: an interactive overview of the parahippocampal-hippocampal network. Nat. Rev. Neurosci. 10, 272–282 (2009).

    Article  CAS  Google Scholar 

  19. Sik, A., Ylinen, A., Penttonen, M. & Buzsaki, G. Inhibitory CA1–CA3-hilar region feedback in the hippocampus. Science 265, 1722–1724 (1994).

    Article  CAS  Google Scholar 

  20. Klausberger, T. & Somogyi, P. Neuronal diversity and temporal dynamics: the unity of hippocampal circuit operations. Science 321, 53–57 (2008).

    Article  CAS  Google Scholar 

  21. Gulyás, A.I., Hajos, N., Katona, I. & Freund, T.F. Interneurons are the local targets of hippocampal inhibitory cells which project to the medial septum. Eur. J. Neurosci. 17, 1861–1872 (2003).

    Article  Google Scholar 

  22. Goutagny, R., Jackson, J. & Williams, S. Self-generated theta oscillations in the hippocampus. Nat. Neurosci. 12, 1491–1493 (2009).

    Article  CAS  Google Scholar 

  23. Jackson, J., Goutagny, R. & Williams, S. Fast and slow gamma rhythms are intrinsically and independently generated in the subiculum. J. Neurosci. 31, 12104–12117 (2011).

    Article  CAS  Google Scholar 

  24. Bland, B.H. & Whishaw, I.Q. Generators and topography of hippocampal theta (RSA) in the anaesthetized and freely moving rat. Brain Res. 118, 259–280 (1976).

    Article  CAS  Google Scholar 

  25. Montgomery, S.M., Betancur, M.I. & Buzsaki, G. Behavior-dependent coordination of multiple theta dipoles in the hippocampus. J. Neurosci. 29, 1381–1394 (2009).

    Article  CAS  Google Scholar 

  26. Schäfer, C., Rosenblum, M.G., Kurths, J. & Abel, H.H. Heartbeat synchronized with ventilation. Nature 392, 239–240 (1998).

    Article  Google Scholar 

  27. Tass, P. et al. Detection of n:m phase locking from noisy data: application to magnetoencephalography. Phys. Rev. Lett. 81, 3291–3294 (1998).

    Article  CAS  Google Scholar 

  28. deGuzman, G.C. & Kelso, J.A. Multifrequency behavioral patterns and the phase attractive circle map. Biol. Cybern. 64, 485–495 (1991).

    Article  CAS  Google Scholar 

  29. Palva, J.M., Palva, S. & Kaila, K. Phase synchrony among neuronal oscillations in the human cortex. J. Neurosci. 25, 3962–3972 (2005).

    Article  CAS  Google Scholar 

  30. Fujisawa, S. & Buzsaki, G. A 4 Hz oscillation adaptively synchronizes prefrontal, VTA and hippocampal activities. Neuron 72, 153–165 (2011).

    Article  CAS  Google Scholar 

  31. Nadim, F., Manor, Y., Nusbaum, M.P. & Marder, E. Frequency regulation of a slow rhythm by a fast periodic input. J. Neurosci. 18, 5053–5067 (1998).

    Article  CAS  Google Scholar 

  32. Siapas, A.G., Lubenov, E.V. & Wilson, M.A. Prefrontal phase locking to hippocampal theta oscillations. Neuron 46, 141–151 (2005).

    Article  CAS  Google Scholar 

  33. Hangya, B., Borhegyi, Z., Szilagyi, N., Freund, T.F. & Varga, V. GABAergic neurons of the medial septum lead the hippocampal network during theta activity. J. Neurosci. 29, 8094–8102 (2009).

    Article  CAS  Google Scholar 

  34. Deadwyler, S.A. & Hampson, R.E. Differential but complementary mnemonic functions of the hippocampus and subiculum. Neuron 42, 465–476 (2004).

    Article  CAS  Google Scholar 

  35. Zhang, Y. et al. Slow oscillations (≤1 Hz) mediated by GABAergic interneuronal networks in rat hippocampus. J. Neurosci. 18, 9256–9268 (1998).

    Article  Google Scholar 

  36. Lee, I., Rao, G. & Knierim, J.J. A double dissociation between hippocampal subfields: differential time course of CA3 and CA1 place cells for processing changed environments. Neuron 42, 803–815 (2004).

    Article  CAS  Google Scholar 

  37. Leutgeb, S., Leutgeb, J.K., Treves, A., Moser, M.B. & Moser, E.I. Distinct ensemble codes in hippocampal areas CA3 and CA1. Science 305, 1295–1298 (2004).

    Article  CAS  Google Scholar 

  38. Brovelli, A. et al. Beta oscillations in a large-scale sensorimotor cortical network: directional influences revealed by Granger causality. Proc. Natl. Acad. Sci. USA 101, 9849–9854 (2004).

    Article  CAS  Google Scholar 

  39. Cui, J., Xu, L., Bressler, S.L., Ding, M. & Liang, H. BSMART: a Matlab/C toolbox for analysis of multichannel neural time series. Neural Netw. 21, 1094–1104 (2008).

    Article  Google Scholar 

  40. Sik, A., Penttonen, M., Ylinen, A. & Buzsaki, G. Hippocampal CA1 interneurons: an in vivo intracellular labeling study. J. Neurosci. 15, 6651–6665 (1995).

    Article  CAS  Google Scholar 

  41. Chang, E.H. & Huerta, P.T. Neurophysiological correlates of object recognition in the dorsal subiculum. Front. Behav. Neurosci. 6, 46 (2012).

    Article  Google Scholar 

  42. Commins, S., Aggleton, J.P. & O′Mara, S.M. Physiological evidence for a possible projection from dorsal subiculum to hippocampal area CA1. Exp. Brain Res. 146, 155–160 (2002).

    Article  Google Scholar 

  43. Kim, S.M., Ganguli, S. & Frank, L.M. Spatial information outflow from the hippocampal circuit: distributed spatial coding and phase precession in the subiculum. J. Neurosci. 32, 11539–11558 (2012).

    Article  CAS  Google Scholar 

  44. Montgomery, S.M. & Buzsaki, G. Gamma oscillations dynamically couple hippocampal CA3 and CA1 regions during memory task performance. Proc. Natl. Acad. Sci. USA 104, 14495–14500 (2007).

    Article  CAS  Google Scholar 

  45. Montgomery, S.M., Sirota, A. & Buzsaki, G. Theta and gamma coordination of hippocampal networks during waking and rapid eye movement sleep. J. Neurosci. 28, 6731–6741 (2008).

    Article  CAS  Google Scholar 

  46. Colgin, L.L. et al. Frequency of gamma oscillations routes flow of information in the hippocampus. Nature 462, 353–357 (2009).

    Article  CAS  Google Scholar 

  47. Dragoi, G. & Buzsaki, G. Temporal encoding of place sequences by hippocampal cell assemblies. Neuron 50, 145–157 (2006).

    Article  CAS  Google Scholar 

  48. Sun, Y. et al. Cell type–specific circuit connectivity of hippocampal CA1 revealed through cre-dependent rabies tracing. Cell Reports 7, 269–280 (2014).

    Article  CAS  Google Scholar 

  49. Nakashiba, T., Young, J.Z., McHugh, T.J., Buhl, D.L. & Tonegawa, S. Transgenic inhibition of synaptic transmission reveals role of CA3 output in hippocampal learning. Science 319, 1260–1264 (2008).

    Article  CAS  Google Scholar 

  50. Manseau, F., Goutagny, R., Danik, M. & Williams, S. The hippocamposeptal pathway generates rhythmic firing of GABAergic neurons in the medial septum and diagonal bands: an investigation using a complete septohippocampal preparation in vitro. J. Neurosci. 28, 4096–4107 (2008).

    Article  CAS  Google Scholar 

  51. Berens, P. CircStat: a MATLAB toolbox for circular statistics. J. Stat. Softw. 31, 1–21 (2009).

    Article  Google Scholar 

Download references

Acknowledgements

We are grateful to M. Sievers and the Integrated Program in Neuroscience at McGill for the continued support and encouragement and G. Ducharme for comments on the project and manuscript. This work was supported by the Canadian Institute of Health Research and the Natural Sciences and Engineering Research Council of Canada (NSERC). J.J. was supported by the Ann and Richard Sievers Innovation in Neuroscience award, the Sir James Lougheed Award of Distinction from the Government of Alberta, and the NSERC Michael Smith award. B.A. was a recipient of Fyssen and Fonds de la Recherche en Santé du Québec postdoctoral fellowships. R.G. was supported by the Projet International de Cooperation de Scientifique from CNRS and a Marie Curie reintegration grant from the European Research Council.

Author information

Author notes

  1. Jesse Jackson

    Present address: Present address: Department of Biological Sciences, Columbia University, New York, New York, USA.,

Authors and Affiliations

  1. Department of Psychiatry, Douglas Mental Health University Institute, McGill University, Montréal, Québec, Canada

    Jesse Jackson, Bénédicte Amilhon, Frédéric Manseau, Christian Kortleven & Sylvain Williams

  2. Laboratoire de neurosciences cognitives et adaptatives (LNCA)-CNRS UMR7364, Université de Strasbourg-CNRS, Strasbourg, France

    Romain Goutagny & Jean-Bastien Bott

  3. Center for Complex Systems and Brain Sciences, Florida Atlantic University, Boca Raton, Florida, USA

    Steven L Bressler

Authors
  1. Jesse Jackson
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Bénédicte Amilhon
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Romain Goutagny
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jean-Bastien Bott
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Frédéric Manseau
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Christian Kortleven
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Steven L Bressler
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Sylvain Williams
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

J.J., R.G. and S.W. designed the project. J.J. and R.G. collected in vitro data. B.A. designed and carried out the optogenetic experiments. R.G. and J.B.-B. performed the in vivo experiments. F.M. collected electrophysiological and pharmacological data. C.K. collected optogenetic data. S.L.B. assisted in data analysis and interpretation. J.J. analyzed the data. J.J. and S.W. wrote the paper.

Corresponding author

Correspondence to Sylvain Williams.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Oscillatory frequency of spiking and synaptic potentials in the subicular and CA3 oscillators

a: whole cell recordings were performed in CA3 together with local CA3 LFPs and subicular LFPs. Shown is a putative pyramidal cell characterized by hyperpolarizing and depolarizing current injection. b: example sub-threshold post-synaptic potentials (PSPs) in the cell. Note the PSPs co-vary with the LFP potentials. c: the oscillatory frequency of all multiunit/single unit (blue) and whole cell recordings (black) plotted against the CA3 LFP frequency showing a strong 1:1 relationship. d-f: the same as a-c but for the subiculum. g: Left: data from an example experiment, showing the power spectral density (PSD) of a simultaneously recorded subicular LFP, and the membrane potential oscillations in CA3 (Vm). The autocorrelations, cross correlations, and coherence between these signals are shown below. Note the communication (cross correlation and coherence) at the frequency of the subiculum. Right: CA3 spiking and Vm PSP frequency is plotted against the frequency of the simultaneously recorded subicular LFP. Note the points are nearly above or on the 1:1 line, indicating a faster subiculum oscillator.

Supplementary Figure 2 Independence and interaction between oscillators

a: diagrammatic representation of the cut, separating the two regions. Below is a photomicrograph showing the cut hippocampus before being placed into the recording chamber. Colored regions indicate the locations of the recording electrodes. b: example raw LFPs simultaneously recorded from the subiculum (red) and CA3 (blue) in a preparation where the two regions were completely separated prior to placing the two oscillators in the recording chamber. Both regions oscillate independently. c: autocorrelation of the subicular LFP (left) and CA3 LFP (right). The oscillators retain their oscillatory properties and frequency difference when isolated. d: an example joint phase probability plot showing no distinct bands where constant phase coupling would normally occur. See the main text and methods and materials for details. e: the mean phase – phase synchrony between oscillators compared to all data from the intact preparation. f: the difference in peak frequency was maintained in these separated preparations. ** p < 0.01, * p < 0.05 two sample t-test and paired t-test respectively. g: peak (maximal value) coherence between all electrodes and the subiculum reference electrode. h: coherence between all electrodes and a CA3 reference electrode. Data from all cases are shown in gray lines and the mean ± s.e.m is shown with solid colored lines and error bars (n=14). i: The relationship between SUB- CA3 LFP phase coupling is plotted as a function of the ratio between the intrinsic SUB frequency and the intrinsic CA3 frequency. The negative linear correlation indicates that the greater the frequency differences the less cross region phase coupling. This is expected, as the greater the frequency difference, the less the regions can communicate. j: the same as (i) but for spike phase modulation (SPM) between subicular spikes and CA3 LFP. k: relationship between SPM for CA3 spikes and subicular LFP phase and the SUB:CA3 frequency ratio. l and m: the relationship between the endogenous regional frequency and the frequency ratio for CA3(l) and the subiculum (m), showing that frequency matching (value of 1) occurs within a limited range of in vitro network frequencies.

Supplementary Figure 3 Example Granger causality spectrums with separate recording configurations

a: Representative power spectrums for LFPs recorded in the subiculum (red) and CA3 (blue). b: Granger causality (GC) spectrums calculated for the same data as (a). the threshold for significance for this data is indicated by the dashed lines (red: threshold for SUB to CA3 GC, blue: threshold for CA3 to SUB GC). The peak from each GC spectrum was taken as a metric causal flow between regions in each direction. c: The GC peaks for SUB to CA3 (y axis) versus CA3 to SUB (x-axis) as shown in the main manuscript. Each point represents the mean GC peak for between 2-15 minutes of continuous stationary data. On the right is the GC peaks for data with the multichannel probe (n=14). The data from both plots in (c) were taken calculated from glass pipettes in similar CA3 and SUB regions, the only difference being the multi-probe was inserted through CA1 for the data on the right, whereas the data on the left had no probe in CA1. Qualitatively, there was a slight trend for towards a reduction in the SUB to CA3 GC peaks suggesting that the large 16 channel multi-probe likely creates a slight damage of the long range hippocampal networks and potentially weakens the reversed signaling.

Supplementary Figure 4 Full experiments showing subicular inactivation and CA3 changes.

Each panel (a-c) shows a preparation where multiple injections of procaine hydrochloride were made in the subiculum while recording activity in both the subiculum (LFPs, top panel), and CA3 (bottom panel). The white vertical lines and asterisk in each panel indicate the time that procaine was microinjected into the subiculum. Injections were performed with a broken micropipette so a small amount (estimated between 0.1-0.3 μL) is injected each time. Experiments were mainly performed in the intermediate subiculum, so not all of the subiculum is inactivated. Therefore, the effect is likely an underestimate. In (a), the subiculum entrains CA3, such that when the subiculum is inactivated the frequency undergoes a dramatic decrease. This entrainment was most obvious when frequencies were locked in a 1:1 ratio regime. Therefore, the 1:1 ratio is likely arising because of the subiculum. Note that in some cases particularly in (b) and (c), that CA3 LFP power does not completely wash back to baseline, and appears to increase with each subsequent subicular inactivation trial. In addition, in (c), subiculum inactivation increases CA3 power and the CA3 oscillator then begins to interact with the subiculum (slower frequency band). Therefore, highly complex bidirectional dynamics are governing the coupling between regions.

Supplementary Figure 5 CA3 electrical stimulation modulates subicular spiking.

a: experimental setup showing the bipolar stimulation electrode in the fibers emerging from CA3, and recording spiking and LFPs in CA3 and the subiculum. CA3 was stimulated at 0.1-0.2Hz, 0.1-0.3ms duration, 0.1– 0.7mA intensity. b: individual experiments (gray lines) and mean spike count (colored lines) in the brief period before (-200ms to 0ms) and after (0ms to +200ms) CA3 pulses showing that on average the subiculum spike rate is not changed by CA3 stimulation (110±6% of pre stimulation, n = 24, p = 0.2). CA3 spiking is massively increased by stimulation likely through both antidromic and polysynaptic activation of recurrent fibers (255±24%, n = 22, p < 0.001). c: the timing of subicular spiking is, however, changed in responses to CA3 inputs, provided that the subicular neurons did not recently spike. Raw data is shown for 10 CA3 pulses and a putative pyramidal cell recorded in the subiculum. d: the mean group CA3 triggered spike timing histogram for all subicular cells (the black line is the mean response probability, gray lines are the +- s.e.m.). Note how in the raw data subicular spikes are not "driven" by CA3 pulses, especially when they are in their refractory period just following spiking due to intrinsic subicular oscillations. The post-stimulus peak in the subicular spike timing histogram below is therefore not due to increase in rate, but to an effect on spike timing. The spike bin just after the stimulation (5ms) was removed due to the stimulus artifact.

Supplementary Figure 6 The subiculum modulates CA1 and CA3-CA1 coupling

Subicular theta rhythms restrict coupling between CA3 and CA1 in the isolated hippocampus in vitro. a: raw LFP data from 10 channels spanning CA3 to the subiculum, before (black) and after (gray) subiculum inactivation. Note how under baseline conditions the mid-CA1 is influenced by both the subicular and CA3 oscillators, and following subiculum inactivation, the slow CA3 generated rhythm is enhanced. b: spectrograms (left) and power spectrums (right) showing the time-frequency structure of CA3, CA1, and subiculum during inactivation. The faster theta in CA1 is removed, and replaced with the CA3 mediated 2.5Hz signal suggesting theta was arising in the subiculum. c: The GC index shows that following subicular inactivation, the influence of CA3 on CA1 is increased (t4 = 2.63, p = 0.029). d: The GC measurements in 11 hippocampi with simultaneous recordings in CA3, CA1, and subiculum under baseline conditions. The GC is significantly stronger in the direction SUB-> CA1, than CA1->SUB (t10 = 3.80, p = 0.0035). The GC is stronger from CA3->CA1 than from CA1->CA3 (t10 = 2.67, p = 0.024). Shown are mean±-s.d. These results highlight the complexity of interregional interactions. Although theta can flow backward to modulate CA1 and CA3, CA1 is dynamically controlled by both oscillators. One open question is the extent to which CA1 is required or involved mediating the theta frequency interactions between the subiculum and CA3.

Supplementary Figure 7 Subicular stimulation evokes GABAergic responses in CA3 neurons

a: Timing of GABAergic responses in CA3 neurons following stimulation of the subiculum in the intact hippocampus. Histograms (bottom, red) and raster plots (top, blue) showing the time of onset of IPSCs recorded during glutamatergic blockade (DNQX 10μM and APV 20μM) in six different cells. Subiculum stimulation is set at time 0. The probability of response is shown on the right. b: whole cell voltage clamp recording of CA3 neurons showing robust synaptic response to subiculum stimulation in the presence of DNQX + APV. Single (grey) and average traces (red) are shown for each cell. Note that a high Cl- solution was used in all recordings to maximize detection of IPSCs. In all cases CA3 IPSCs were evoked by subicular stimulation during DNQX+APV (left) but were abolished by DNQX+APV+Gabazine. c: Examples of CA3 neurons with current injection for spike characterization in the presence of DNQX + APV. The bottom three cells were classified as putative interneurons.

Supplementary Figure 8 Optogenetic driving of subicular PV cells and Granger causality analysis

a: Example immunohistochemical confirmation of eYFP virus in PV+ cells in the subiculum. b: the electrophysiological characterization of a representative PV+ interneuron during current injection, showing the fast spiking profile. c: Blue light generates strong inward current in voltage clamp. d: the PV+ cell responds reliably to 8Hz 5ms single pulses and to a 75Hz train of light. Experiments in b-d are from hippocampal slices. e: example power spectral density plots (PSDs) showing the LFP power in CA3 (left) and the subiculum (right) during baseline spontaneous activity, during 5Hz LED pulses, and during 5Hz LED pulses with the addition of Gabazine (GBZ, 5μM) to block GABAA receptors. Note the optical stimulation decreases the endogenous network power in both regions and entrains the LFP at the frequency of stimulation. f: Granger causality (GC) spectrums for an experiment with AAVdj-ChETA in the subiculum and an LED aimed to activate these infected GABAergic subicular neurons (see methods and materials). Each plot shows the GC spectrum for the optogenetic stimulation frequency used. The last plot on the bottom shows that the GBZ blocks this effect. SUB-> CA3 is shown in red, and CA3->SUB is blue. The GC correctly shows that the SUB drives CA3 in all cases. These data are also useful to validate the GC analysis.

Supplementary Figure 9 Peri oscillation triggered histograms show that the subiculum leads CA3 in the intact, but not slice-like hippocampus

a: example spike timing histograms centered on the peak of the CA3 LFP (time 0 s) for SUB (red) and CA3 (blue) spiking from four example experiments, each with slightly different frequency regimes. Spikes were binned in 5ms bins. The red and blue points indicate the time lag where the peak in the PETH occurred for SUB and CA3 spikes in four representative experiments. Note in the examples, the SUB oscillates at different frequencies, yet the PETH peak occurred at a constant latency before CA3 LFP peaks and CA3 spikes. Note the first peak in the SUB PETHs (> lag 0) occurs at latencies corresponding to the different intrinsic subicular frequencies and not at a fixed latency from CA3. b: the mean (solid) and s.e.m (dashed lines) PETH for the SUB and CA3 in the intact hippocampus. Note that the SUB spiking occurs before CA3 suggesting the temporal order of firing is SUB then CA3. Data are normalized to 1 which represents the overall mean spike count. c: histograms of the PETH peak lags for all experiments for the SUB and CA3. Again note that SUB peaks arise before CA3. d-f: the same as a-c but for slice-like hippocampus experiments. This analysis supports data in Figs 2 and 5.

Supplementary Figure 10 Polysynaptic excitatory synaptic transmission from the CA3 to the subiculum.

a: Left: spontaneous raw LFPs from the subiculum (SUB) and CA3 during blockade of GABAergic transmission after bath application of SR 95531 (Gabazine 5μM) and CGP46381 (5-10 μM) to block GABAA and GABAB receptors respectively. Right: CA3 spike triggered LFPs. Above is the heat map showing the amplitude of the z-scored LFPs from each region and below is the line vector of the same. Here, CA3 spiking drives LFPs through the hippocampus and synchronizes the CA1 and subiculum at a delay of 16ms consistent with two excitatory synapses. b: spontaneous spiking from CA3, CA1, and subiculum from another representative experiment under the same pharmacological conditions of GABAergic blockade. Each plot is a CA3-LFP peak triggered spike raster. Each CA3 peak was considered a "trial". Note the rapid and strong synchronization between all regions at fast delays. c: expanded view of the smoothed spike timing histograms from b. In addition, cross correlation between spikes in both regions confirmed that in the majority of cases (n = 9/11, CA3 leads subiculum by 22±7 ms). Simultaneous spike recordings from CA3, CA1 and subiculum suggest that this synchronization arises through a poly-synaptic relay in CA1 (n = 3).

Supplementary Figure 11 Spectral properties of CA3 and subicular LFPs during home cage (HC), REM sleep and novel open field exploration (OF).

a: example power spectra from the subiculum (top) and CA3 (bottom) during three theta states. Data from a representative animal are shown. Each black line is data from one theta epoch and the colored lines are the mean spectrum. Note the increase in theta power in both regions, and a decrease in delta in CA3 during REM. b: Plots of theta power as a function theta epoch number during OF exploration. Each line represents data from one rat. Note that the largest theta power arises at the start of OF exploration in all cases, suggesting the dynamics of coupling between regions is non-stationary in this behavioral state. Time was not used on the x-axis as theat epochs in different animals could arise at different times. Theta power was normalized to the mean overall theta power across all epochs. c: group data for delta, theta, and gamma (20-45 Hz) power in each behavioral state. Data were normalized to the mean power in each frequency band and averaged across states, thus a value of 1 indicates the mean power for a given frequency band. SUB delta was increased in OF versus HC (t187 = 3.9, p = 1.5 × 10−4) and REM (t189 = 4.3,p = 2.9 × 10−5). SUB theta power was increased during REM relative to HC (t268 = 7.6, p = 5 × 10−13), and theta in the OF was larger than HC (t189 = 4.6, p = 7.6178 × 10−6). SUB gamma was also higher during OF compared to HC (t189 = 3.1, p = 0.0024) and REM (t187 = 3.6, p = 3.6 × 10−4). For CA3, delta power was reduced during REM relative to HC (t268 = 3.4, p = 8.7 × 10−4) and OF (t189 = 4.1, p = 7.3 × 10−5). Theta in CA3 was increased during REM relative to HC (t268 = 3.6, p = 4.1 × 10−4), but not OF. The only gamma difference we could detect was a weak difference between REM and OF (t187 = 2.4, p = 0.0155). d: theta frequency in the SUB was increased during OF relative to REM (t187 = 2.8, p = 0.006). Data are expressed as mean±sem and averaged across epochs from all four rats. e: example coherence spectra showing coherence between CA3 and the subiculum mainly at theta and (6-9 Hz) and 20-45 Hz (slow gamma). Each gray line is the mean coherence for an individual theta epoch, whereas the black line is the group mean. All data from one animal during REM and HC were pooled for this plot. The group mean ± s.e.m. for HC, REM and OF GC and coherence shown in Fig. 7f of the main paper are SUB –> CA3 theta GC: (0.37±0.01, 0.30±0.01, 0.31±0.02) SUB-CA3 coherence (0.77±0.01, 0.83±0.01, 0.83±0.02); gamma CA3->SUB GC (0.14±0.01, 0.12 ± 0.01, 0.132±0.01); and gamma coherence (0.63±0.01, 0.59±0.01, 0.61±0.02).: * p < 0.05, ** p < 0.01, *** p<0.001.

Supplementary Figure 12 Granger causality in vivo suggests that theta in the subiculum and CA1 influence CA3.

a: histological sections showing the position of the recording electrodes – in subiculum, CA1, and CA3. b: a schematic showing electrode placements in all rats. c: The GC coefficients for all pairs of regions are compared and displayed as a function of behavioral state (home cage, REM, novel environment). Each gray point is the mean GC coefficient for a 15-45s theta epoch (analyzed in 5s increments and averaged to obtain one value per theta epoch), and the mean ± s.d. for each animal is shown in color. Note the overall stronger SUB→CA3 and CA1→CA3 GC, as well as the stronger CA1→SUB in 3/4 animals. The differences between states were not explored further. d: a summary table with p-values associated with paired t-tests for each electrode pair and behavioral state.

Supplementary Figure 13 Cross-frequency coupling between CA3 gamma and subicular theta

a: theta phase in the subiculum was calculated using the phase angle of the Hilbert transform on theta filtered data, and CA3 gamma amplitude was calculated by band pass filtering the LFP (20-45Hz) and taking the absolute value of the Hilbert transformed signal. To assess if there was a time lag at which cross frequency coupling (theta-gamma) was increased, we shifted the theta phase across different lags (-1s to +1s in 10ms steps) while keeping the gamma amplitude held constant. The modulation index of the CA3 amplitude – subicular phase histogram was calculated at each time step. One example is shown in which the CFC is maximal at negative shifts (top) suggesting CA3 gamma is acting to influence the future dynamics of subicular theta phase. In another theta epoch (bottom), maximal CFC arises when subicular theta phase is shifted forward in time suggesting subicular theta frequencies are influencing the future dynamics of CA3 gamma amplitude. The red line is the threshold for significance based on randomized subicular shuffled data. b: A histogram of the time lags of maximal CFC for all theta epochs with peaks between -450ms and 450ms (277/325 total theta epochs analyzed). Note the two peaks suggesting cross frequency interactions in both directions. Also, interestingly, these peaks occur at lags predicted by the in vitro spike cross correlations between CA3 and subiculum for "slice – like" (-15-30ms) and "intact" hippocampi (+60-90ms) as shown in Fig. 5 of the main paper. c: a breakdown of the CFC lags as a function of state (Homecage, HC; REM; novel open field, OF). d: the % of theta epochs showing backward (SUB -→ CA3) and forward (CA3 → SUB) CFC lags. The only state which appears to have a dominant direction is the OF state. Although some trends are emerging, there are too few data points to make definitive claims regarding state dependent cross frequency coupling dynamics. Likely this parameter is changing faster than we are measuring it here (15-45s for each theta epoch). However, these data do show that of the theta epochs analyzed, nearly half (43% across states) show a positive CFC shift suggesting that in many instances reversed theta flow plays a role in timing CA3 gamma.

Supplementary information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jackson, J., Amilhon, B., Goutagny, R. et al. Reversal of theta rhythm flow through intact hippocampal circuits. Nat Neurosci 17, 1362–1370 (2014). https://doi.org/10.1038/nn.3803

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.3803

This article is cited by

Search

Advanced search

Quick links

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