Evidence that a defined population of neurons in lateral amygdala is directly involved in auditory fear learning and memory

https://doi.org/10.1016/j.nlm.2019.107139Get rights and content

Abstract

Memory is thought to be encoded within networks of neurons within the brain, but the identity of the neurons involved and circuits they form have not been described for any memory. Previously, we used fos-tau-lacZ (FTL) transgenic mice to identify discrete populations of neurons in different regions of the brain which were specifically activated following fear conditioning. This suggested that these populations of neurons form nodes in a network that encodes fear memory. In particular, one population of learning activated neurons was found within a discrete region of the lateral amygdala (LA), a key nucleus required for fear conditioning. In order to provide evidence that this population is directly involved in fear conditioning, we have analysed the expression of a key molecular requirement for fear conditioning in LA, phosphorylated Extracellular Signal Regulated Kinase 1 and 2 (pERK1/2). The only neurons in LA that specifically expressed pERK1/2 following auditory fear conditioning were in the ventrolateral nucleus of the LA (LAvl), in the same discrete region where we found learning specific FTL+ neurons. Double labelling experiments in FTL mice showed that a substantial proportion of the learning activated neurons expressed both pERK1/2 and FTL. These experiments provide clear evidence that the learning specific neurons we identified within LAvl are directly involved in auditory fear conditioning. In addition, learning specific expression of pERK1/2 was found in a dense network of dendrites contained within the border region of the LAvl. This network of dendrites may represent an activated dendritic field involved in fear conditioning in LA.

Introduction

A central concept in memory research is that of the engram, the hypothetical physical change in the brain responsible for storing, or encoding, a memory. Although there has been considerable effort to identify circuitry associated with particular learning and memory models (LeDoux, 2000, Thompson, 2005), no studies have identified all the changes which constitute an engram in vertebrates. A major reason for this is that it has been very difficult to identify the neurons which are directly involved in memory storage in the brain. There have some been real successes in this area in the classic studies of the marine snail, Aplysia (Kandel, 2001) and mollusc, Hermissenda (Blackwell, 2006). However, both of these models involve simple reflex motor behaviour and very simple circuitry, a long way from the complexities of the vertebrate brain.

In vertebrates, one method that enables identification of neurons involved in a certain task employs the immediate early gene c-fos as a marker of neuronal activation (Knapska, Radwanska, Werka et al., 2007). We previously utilized the transgenic fos-tau-LacZ (FTL) mouse to visualise neurons activated by fear conditioning (Butler et al., 2015, Trogrlic et al., 2011, Wilson and Murphy, 2009). The FTL transgene uses the c-fos promoter to drive expression of β-galactosidase into the cell bodies and processes of c-fos activated neurons (Murphy et al., 2007, Wilson and Murphy, 2009). Specific, spatially discrete populations of neurons within amygdala, hypothalamus and lateral septum were identified that were specifically activated following acquisition of conditioned fear, rather than by exposure to other sensory stimuli (Butler et al., 2015, Trogrlic et al., 2011, Wilson and Murphy, 2009). We proposed that these populations of neurons form nodes in a network that encodes fear memory (Butler et al., 2015, Trogrlic et al., 2011). In particular, one of these populations of neurons was found within the ventrolateral (vl) nucleus of the lateral amygdala (LA), a key region involved in fear conditioning (Diaz-Mataix et al., 2014, Izquierdo et al., 2016, Johansen et al., 2011, Josselyn and Frankland, 2018, Kim and Jung, 2006, Maren, 2003).

Other experiments have utilised c-fos regulated expression in transgenic mice to identify neuronal populations in dentate gyrus that are activated by exposure to specific contexts during fear conditioning (Garner et al., 2012, Liu et al., 2012, Ramirez et al., 2013, Reijmers et al., 2007). Subsequent inhibition or activation of these neurons demonstrated that these populations of c-fos+ neurons were required for memory of specific contexts (Garner et al., 2012, Liu et al., 2012, Ramirez et al., 2013). These experiments not only demonstrate that different hippocampal neurons are required in the memory of different contexts, they also validate the approach of using c-fos regulated expression as a marker of neurons involved in learning and memory. Nevertheless, and despite the functional strength of these experiments, the c-fos expressing neurons activated by learning could not be distinguished from those activated by different sensory stimuli during the context fear conditioning process. It could thus not be determined which of these neurons were directly involved in contextual memory formation.

Whereas we have identified discrete populations of neurons activated specifically by fear conditioning (Butler et al., 2015, Trogrlic et al., 2011, Wilson and Murphy, 2009), we do not have clear evidence to establish that these neurons are functionally involved in fear learning and memory. To determine if the FTL expressing neurons in LAvl are required for fear learning and memory, it is theoretically possible to inhibit or lesion these neurons. Whereas there are effective methods for inhibiting or lesioning c-fos expressing neurons (Cruz et al., 2013, Garner et al., 2012, Koya et al., 2016, Liu et al., 2012, Ramirez et al., 2013), these approaches have been used for volumes of brain tissue which were much larger than those containing the spatially discrete populations of neurons we have identified (the volume containing these neurons in LAvl is ~0.1 mm3). Even for the techniques which use stereotaxic injections to target particular brain areas, for the very small fraction of mice which could be accurately and bilaterally targeted to LAvl, it may not be possible to inhibit most of the learning specific neurons and at the same time avoid c-fos expressing neurons in closely neighbouring parts of the brain, such as dorsal or ventromedial LA (LAd, LAvm) or the basolateral amygdala (BLA) subnuclei.

We have used an alternative approach to obtain evidence of functional involvement of neuronal populations in fear conditioning. A number of key signalling pathways required for fear conditioning have been identified, with particular focus on the LA. One such requirement is phosphorylation of Extracellular Signal Regulated Kinase 1 and 2 (ERK1/2), a mitogen activated protein kinase. Inhibition of ERK1/2 phosphorylation in LA in the period following fear conditioning results in attenuation of long term auditory fear memory and ERK1/2 is transiently phosphorylated in rat LA following auditory fear conditioning (Schafe, Atkins, Swank et al., 2000). Thus, phosphorylation of ERK1/2 in LA following fear conditioning is required for the formation of fear memory. This finding not only points to the involvement of ERK1/2 in the memory process, it also means that the cells in LA in which ERK1/2 is phosphorylated following fear conditioning are required for fear memory formation because the essential phosphorylation step is occurring in these cells. It follows that the cells in LA which are directly involved in fear memory can be identified based on their expression of phosphorylated pERK1/2.

Experiments assessing pERK1/2 expression after fear conditioning have generally shown an increase in number of pERK1/2 neurons in LA, but have not identified a clear pattern of learning specific expression (Besnard et al., 2014, Di Benedetto et al., 2009, Schafe et al., 2000). Some of these experiments were not necessarily designed to find learning specific populations of neurons. Thus, there may have been expression of pERK1/2 due to non-specific sensory stimuli or to some context learning in the control groups, which could have masked the identification of a discrete population of learning specific neurons. We modified the traditional auditory fear conditioning protocol to minimise effects of non-specific sensory stimuli and context learning (Butler et al., 2015, Nithianantharajah and Murphy, 2008). Using this refined approach, we successfully identified FTL+ neurons specifically activated by auditory fear conditioning (Butler et al., 2015). Here, we have employed this same protocol of auditory fear conditioning to identify neurons in LA that express pERK1/2 under these specific learning conditions, to provide evidence for their involvement in fear learning and memory.

Section snippets

Animals

All experiments were conducted with approval of the Animal Ethics Committee of the University of Melbourne and in accordance with the guidelines of the National Health and Medical Research Council (Australia). Male FTL+ mice aged 8–12 weeks were obtained from the Biomedical Sciences Animal Facility, University of Melbourne, singly housed in standard 15 cm × 30 cm × 12 cm cages on a 12 h light/dark cycle, and had food and water supplied ad libitum. Two days prior to the commencement of the

Auditory fear conditioning

Pavlovian or classical conditioning involves presentation of a neutral stimulus, the conditioned stimulus (CS), paired with a biologically significant stimulus, the unconditioned stimulus (US), such as pain or food. Once learning has occurred, presentation of CS alone is able to elicit a response, and this learned reaction to the previously neutral CS is known as the conditioned response. To identify neurons involved in auditory fear conditioning, we utilised a method that enables training of

Discussion

It is well established that activation of ERK1/2 by phosphorylation is part of a key signalling pathway in memory formation (Adams & Sweatt, 2002). In studies of fear memory, inhibitors of ERK1/2 phosphorylation infused either systemically (Atkins et al., 1998, Selcher et al., 1999) or locally into LA impaired long term memory consolidation (Di Benedetto et al., 2009, Schafe et al., 2000) and reconsolidation (Duvarci, Nader, & LeDoux, 2005). However, short term memory was unaffected (Schafe et

Acknowledgements

We would like to thank Ellie Hyun-Jung Cho from the Biological Optical Microscopy Platform, University of Melbourne; Maya Kesar, Sarah Taverner and staff of the Biomedical Sciences Animal Facility, University of Melbourne. This research was supported by funding from the National Health and Medical Research Council of Australia and the Department and Anatomy and Neuroscience, University of Melbourne.

References (80)

  • L.M. Mao et al.

    Amphetamine increases phosphorylation of MAPK/ERK at synaptic sites in the rat striatum and medial prefrontal cortex

    Brain Research

    (2013)
  • J. Nithianantharajah et al.

    Auditory specific fear conditioning results in increased levels of synaptophysin in the basolateral amygdala

    Neurobiology of Learning and Memory

    (2008)
  • S.M. Rodrigues et al.

    Molecular mechanisms underlying emotional learning and memory in the lateral amygdala

    Neuron

    (2004)
  • J.M. Santos et al.

    Gabaergic mechanisms of hypothalamic nuclei in the expression of conditioned fear

    Neurobiology of Learning and Memory

    (2008)
  • G.P. Shumyatsky et al.

    stathmin, a gene enriched in the amygdala, controls both learned and innate fear

    Cell

    (2005)
  • A.P. Yiu et al.

    Neurons are recruited to a memory trace based on relative neuronal excitability immediately before training

    Neuron

    (2014)
  • J.P. Adams et al.

    The A-type potassium channel Kv4.2 is a substrate for the mitogen-activated protein kinase ERK

    Journal of Neurochemistry

    (2000)
  • J.P. Adams et al.

    Molecular psychology: Roles for the ERK MAP kinase cascade in memory

    Annual Review of Pharmacology and Toxicology

    (2002)
  • C.M. Alberini

    Transcription factors in long-term memory and synaptic plasticity

    Physiological Reviews

    (2009)
  • C.M. Atkins et al.

    The MAPK cascade is required for mammalian associative learning

    Nature Neuroscience

    (1998)
  • H.C. Bergstrom et al.

    An organization of visual and auditory fear conditioning in the lateral amygdala

    Neurobiology of Learning and Memory

    (2014)
  • H.C. Bergstrom et al.

    Pavlovian fear conditioning activates a common pattern of neurons in the lateral amygdala of individual brains

    PLoS ONE

    (2011)
  • A. Besnard et al.

    Comparative dynamics of MAPK/ERK signalling components and immediate early genes in the hippocampus and amygdala following contextual fear conditioning and retrieval

    Brain Structure & Function

    (2014)
  • K.T. Blackwell

    Subcellular, cellular, and circuit mechanisms underlying classical conditioning in Hermissenda crassicornis

    The Anatomical Record Part B: The New Anatomist

    (2006)
  • E.M. Boggio et al.

    Visual stimulation activates ERK in synaptic and somatic compartments of rat cortical neurons with parallel kinetics

    PLoS ONE

    (2007)
  • F. Bordi et al.

    Response properties of single units in areas of rat auditory thalamus that project to the amygdala. I. Acoustic discharge patterns and frequency receptive fields

    Experimental Brain Research

    (1994)
  • F. Bordi et al.

    Response properties of single units in areas of rat auditory thalamus that project to the amygdala. II. Cells receiving convergent auditory and somatosensory inputs and cells antidromically activated by amygdala stimulation

    Experimental Brain Research

    (1994)
  • C.W. Butler et al.

    Tracking the fear memory engram: Discrete populations of neurons within amygdala, hypothalamus, and lateral septum are specifically activated by auditory fear conditioning

    Learning & Memory

    (2015)
  • C.W. Butler et al.

    Neurons specifically activated by fear learning in lateral amygdala display increased synaptic strength

    eNeuro

    (2018)
  • B. Casar et al.

    Ras subcellular localization defines extracellular signal-regulated kinase 1 and 2 substrate specificity through distinct utilization of scaffold proteins

    Molecular and Cellular Biology

    (2009)
  • L. Chung

    A brief introduction to the transduction of neural activity into Fos signal

    Development & Reproduction

    (2015)
  • C. Cirelli et al.

    On the functional significance of c-fos induction during the sleep-waking cycle

    Sleep

    (2000)
  • F.C. Cruz et al.

    New technologies for examining the role of neuronal ensembles in drug addiction and fear

    Nature Reviews Neuroscience

    (2013)
  • B. Di Benedetto et al.

    Activation of ERK/MAPK in the lateral amygdala of the mouse is required for acquisition of a fear-potentiated startle response

    Neuropsychopharmacology

    (2009)
  • M.C. Diamond

    Response of the brain to enrichment

    Anais da Academia Brasileira de Ciências

    (2001)
  • N.N. Doron et al.

    Organization of projections to the lateral amygdala from auditory and visual areas of the thalamus in the rat

    The Journal of Comparative Neurology

    (1999)
  • S. Duvarci et al.

    Activation of extracellular signal-regulated kinase- mitogen-activated protein kinase cascade in the amygdala is required for memory reconsolidation of auditory fear conditioning

    European Journal of Neuroscience

    (2005)
  • C.R. Farb et al.

    NMDA and AMPA receptors in the lateral nucleus of the amygdala are postsynaptic to auditory thalamic afferents

    Synapse (New York, N. Y.)

    (1997)
  • S.W. Flavell et al.

    Signaling mechanisms linking neuronal activity to gene expression and plasticity of the nervous system

    Annual Review of Neuroscience

    (2008)
  • A.R. Garner et al.

    Generation of a synthetic memory trace

    Science

    (2012)
  • Cited by (0)

    1

    These authors contributed equally to this work.

    View full text