Amygdala circuitry mediating reversible and bidirectional control of anxiety

Abstract

Anxiety--a sustained state of heightened apprehension in the absence of immediate threat--becomes severely debilitating in disease states. Anxiety disorders represent the most common of psychiatric diseases (28% lifetime prevalence) and contribute to the aetiology of major depression and substance abuse. Although it has been proposed that the amygdala, a brain region important for emotional processing, has a role in anxiety, the neural mechanisms that control anxiety remain unclear. Here we explore the neural circuits underlying anxiety-related behaviours by using optogenetics with two-photon microscopy, anxiety assays in freely moving mice, and electrophysiology. With the capability of optogenetics to control not only cell types but also specific connections between cells, we observed that temporally precise optogenetic stimulation of basolateral amygdala (BLA) terminals in the central nucleus of the amygdala (CeA)--achieved by viral transduction of the BLA with a codon-optimized channelrhodopsin followed by restricted illumination in the downstream CeA--exerted an acute, reversible anxiolytic effect. Conversely, selective optogenetic inhibition of the same projection with a third-generation halorhodopsin (eNpHR3.0) increased anxiety-related behaviours. Importantly, these effects were not observed with direct optogenetic control of BLA somata, possibly owing to recruitment of antagonistic downstream structures. Together, these results implicate specific BLA-CeA projections as critical circuit elements for acute anxiety control in the mammalian brain, and demonstrate the importance of optogenetically targeting defined projections, beyond simply targeting cell types, in the study of circuit function relevant to neuropsychiatric disease.

Figure 1. Projection-specific excitation of BLA terminals in the CeA induces acute reversible anxiolysis

a) Mice were housed in a high-stress environment before behavioral manipulations and receive 5-ms light pulses at 20Hz for all light-on conditions. (b–c) ChR2:BLA-CeA mice (n=8) received selective illumination of BLA terminals in the CeA during the light-on epoch on the EPM; see ChR2:BLA-CeA representative path (b), which induced an increase in open-arm time upon photostimulation relative to eYFP:BLA-CeA (n=9) and ChR2:BLA(somata) (n=7) controls (c), and an increase in probability of open arm entry (see inset). (d–e) ChR2:BLA-CeA mice also increased center time on the OFT, as seen in a representative path (d), during light-on epochs relative to light-off epochs and eYFP:BLA-CeA and ChR2:BLA(somata) controls (e).

Figure 2. Projection-specific excitation of BLA terminals in the CeA activates CeL neurons and elicits feed-forward inhibition of CeM neurons

a) Two-photon images of representative BLA, CeL and CeM cells imaged from the same slice, overlaid on a brightfield image. (b–f) Schematics of the recording and illumination sites for the associated representative current-clamp traces (V_m=~−70 mV. b) Representative BLA pyramidal neuron trace expressing ChR2, all of which spiked for every pulse (n=4). c) Representative trace from a CeL neuron in the terminal field of BLA projection neurons, showing both sub- and supra-threshold excitatory responses upon photostimulation (n=16). Inset left, population summary of mean probability of spiking for each pulse in a 40-pulse train at 20Hz, dotted lines indicate s.e.m. Inset right, frequency histogram showing individual cell spiking fidelity; y-axis is the number of cells per each 5% bin. d) Six sweeps from a CeM neuron spiking in response to a current step (~60 pA; indicated in black) and inhibition of spiking upon 20Hz illumination of BLA terminals in the CeL. Inset, spike frequency was significantly reduced during light stimulation of CeL neurons (n=4; spikes/second before (49±9.0), during (1.5±0.87), and after (33±8.4) illumination; mean±s.e.m.). (e–f) Upon broad illumination of the CeM, voltage-clamp summaries show that the latency of excitatory postsynaptic currents (EPSCs) is significantly shorter than the latency of inhibitory postsynaptic currents (IPSCs), whereas there was a non-significant difference in the amplitude of EPSCs and IPSCs (n=11; *p=0.04, see insets). The same CeM neurons (n=7) showed either net excitation when receiving illumination of the CeM (e) or net inhibition upon selective illumination of the CeL (f).

Figure 3. Light-induced anxiolytic effects are attributable to activation of BLA-CeL synapses

(a–b) Schematic of the recording site and illumination positions, as whole-cell recordings were performed at each illumination location, in 100 μm increments away from the cell soma both over a visualized axon and in a direction that was not over an axon (inset). Normalized summary of spike fidelity and depolarizing current (a) to a 20Hz train delivered at various distances from the soma. (b) Representative traces upon ~125 μm diameter illumination at various locations within each slice (n=7). Illumination of BLA somata elicits high-fidelity spiking (top). Illumination of BLA terminals in CeL elicits strong excitatory responses shown in voltage-clamp in the postsynaptic CeL neuron (middle), but does not elicit reliable antidromic spiking in the BLA neuron itself (bottom), summarized in a frequency histogram (inset). (c,d) A separate group of ChR2:BLA-CeA mice (n=8) performed the EPM and OFT twice, one session preceded with intra-CeA infusions of saline (red) and the other session with glutamate receptor antagonists NBQX and AP5 (purple), counter-balanced for order. Glutamate receptor blockade in the CeA attenuated light-induced increases in both open arm time (c) and probability of open arm entry (inset) on the EPM and center time on the OFT (d, inset shows pooled summary), without altering baseline performance.

Figure 4. Selective inhibition of BLA terminals in the CeA induces an acute and reversible increase in anxiety

(a–e) Mice were group-housed in a low-stress environment and received bilateral constant 594 nm light during light-on epochs. a) Selective illumination of eNpHR3.0-expressing BLA terminals suppresses vesicle release evoked by electrical stimulation in the BLA. Schematic indicates the locations of the stimulating electrode, the recording electrode and the ~125 μm diameter light spot. Representative CeL EPSCs before (Off₁), during (On) and after (Off₂) selective illumination of eNpHR3.0-expressing BLA terminals. Normalized EPSC amplitude summary data from sections containing BLA neurons expressing eNpHR3.0 (n=7) and non-transduced controls (n=5) show that selectively illuminating BLA-CeL terminals reduces (*p=0.006) electrically-evoked EPSC amplitude in postsynaptic CeL neurons relative to non-transduced control slice preparations (inset). (b–c) Representative eNpHR3.0:BLA-CeA path (b) indicates reduced open arm time (c) and probability of open arm entry (inset) during illumination, relative to controls. (d–e) Representative eNpHR3.0:BLA-CeA path (d) reflects reduced center time on the OFT (e) for the eNpHR3.0:BLA-CeA group during light-on, but not light-off, epochs as compared to controls (inset shows pooled data).