Input associativity underlies fear memory renewal

Abstract Synaptic associativity, a feature of Hebbian plasticity wherein coactivation of two inputs onto the same neuron produces synergistic actions on postsynaptic activity, is a primary cellular correlate of associative learning. However, whether and how synaptic associativity are implemented into context-dependent relapse of extinguished memory (i.e. fear renewal) is unknown. Here, using an auditory fear conditioning paradigm in mice, we show that fear renewal is determined by the associativity between convergent inputs from the auditory cortex (ACx) and ventral hippocampus (vHPC) onto the lateral amygdala (LA) that reactivate ensembles engaged during learning. Fear renewal enhances synaptic strengths of both ACx to LA and the previously unknown vHPC to LA monosynaptic inputs. While inactivating either of the afferents abolishes fear renewal, optogenetic activation of their input associativity in the LA recapitulates fear renewal. Thus, input associativity underlies fear memory renewal.

conditioning chamber for five consecutive days. The conditioning chambers (17 cm × 17 cm × 25 cm) equipped with stainless-steel shocking grids were connected to a precision-feedback current-regulated shocker (UGO BASILE srl). During fear conditioning, the chamber walls were covered with black-and-white checkered wallpaper, and the chambers were cleaned with 75% ethanol (context A). On day 0, mice were conditioned individually in context A with six pure tones (CS; 4 kHz, 76 dB, 20 s each) delivered at variable intervals (20-180 s), and each tone was co-terminated with a foot shock (US; 0.5 mA, 2 s each). ANY-maze software (Stoelting Co.) was used to automatically control the delivery of tones and foot shocks. Conditioned mice were returned to their home cages 30 s after the end of the last tone, and the floor and walls of the cage were cleaned with 75% ethanol for each mouse. For extinction learning, on days 1 and 2, mice trained in context A with six CS-US pairings on day 0 were presented with 12 CS presentations (4 kHz, 76 dB, 30 s each) without foot shock in a test chamber which had gray non-shocking plexiglass floor and dark gray wallpapers and was cleaned with 4% acetic acid solution (context B). On day 3, mice received four CSalone (30 s each) presentations in the extinction context (context B) for extinction test (also referred to as the ABB test) or in the conditioning context (context A) for fear renewal (also referred to as ABA renewal). For the fear generalization test, mice were presented with the unconditioned white noise (76 dB, 30 s each) in the context as indicated in the Figures. For additional extinction training in the renewal context, mice experienced with fear renewal were then presented with 12 CS presentations (4 kHz, 76 dB, 30 s each) without foot shock in the context A (also referred to as the ABA extinguished), while the control group was kept in the homecage without any treatment (also referred to as the ABA homecage). During behavioral testing, the chamber was placed in a sound-attenuating enclosure with a ventilation fan and a single house light (UGO BASILE srl). The movement of the mouse in the conditioning or test chamber was recorded using a near-infrared camera and analyzed in real-time with ANY-maze software. The ANY-maze behavior tracking software uses freezing score to represent the freezing status of the animal. The freezing score is a unit-less value as a result of rather complex calculations. When the software is calculating the freezing score, it looks for animal movements in the entire apparatus, which is accomplished by comparing every pixel of the current frame to earlier ones. If the software fails to find any movement (large number of flickering pixels) in the apparatus, the animal would be considered to be freezing. The software also includes 'noises' of the video when calculating the freezing score, for example, a breathing animal would cause some pixels to flicker. These kinds of noises coming with animals' physiological activities would have an influence on the value of the freezing score. Typically, louder video noise would result in lower freezing score of the animal at that frame. Finally, the software would give a result of freezing score at each frame, and periods that the animal is considered to be freezing according to the threshold setting. Freezing was defined as the absence of all movement, except respiration, for at least 2 s and was scored automatically using ANY-maze software. For animals connected to an optical fiber to the head, light stimuli during test sessions can interfere with the program's motion detection, so freezing of these sessions was scored independently for each video by an experienced experimenter in a double-blind manner. The time spent freezing during the tone (cue) was measured for each tone presentation.

Viral injection
Mice at 6-7 weeks old were anesthetized with 1% sodium pentobarbital via a single intraperitoneal injection per mouse (10 ml per kg body weight), after which each mouse was mounted in a stereotactic frame with non-rupture ear bars (RWD Life Science).
After making an incision to the midline of the scalp, small bilateral craniotomies were performed using a microdrill with 0.5-mm burrs. Glass pipettes, with tip diameter of 10-20 μm were made with a P-97 Micropipette Puller (Sutter Glass pipettes) for AAV microinjections. The microinjection pipettes were first filled with silicone oil and then connected to a microinjector pump (KDS 310, KD Scientific) with full air exclusion.
AAV-containing solutions were loaded into the tips of pipettes and injected at the following coordinates (2)

Optogenetic manipulations during extinction test or fear renewal
To investigate optogenetic-mediated effects during either extinction test or fear renewal, mice were subjected to auditory fear conditioning and to two consecutive days of was delivered in 10-ms pulses at 20 Hz during the presentation of each 30-s CS (with 5 s added before and after the CS to ensure that the light delivery covers the CS exposure). To photostimulate either vHPC  LA or ACx  LA inputs in the renewal context (same as conditioning), the light pulses were coordinated the same way as described above but the CS was omitted.

Cannula implantations and local drug injections
Mice were anaesthetized with 1% sodium pentobarbital and were then each fixed on a stereotaxic apparatus (RWD Life Science). Stainless-steel guide cannulae (RWD Life Science) were bilaterally implanted into the target brain areas, and the tips of cannulae were targeted at the following coordinates (in mm): LA: AP, -1.82 mm; ML, ±3.20 mm; DV, -3.20 mm. The cannulae were fixed to the skull using acrylic cement and two skull screws. Stainless-steel obturators (33 gauges) were inserted into guide cannulae to avoid obstruction until drug infusion. Mice were allowed to recover from surgery for two weeks before behavioral tests. Mice were handled and habituated to the infusion procedure several days before drug injections. During drug infusions, mice were briefly head-restrained, while the stainless-steel obturators were removed and injection cannulae (33 gauges, RWD Life Science) were inserted into the guide cannulae.
Injection cannulae protruded 1.00 mm from the tips of guide cannulae. Infusion cannulae were connected via PE20 tubing to a microsyringe driven by a microinfusion pump (KDS 310, KD Scientific). Drugs were infused bilaterally into the target brain areas at a flow rate of 0.15 μl per min. After finishing drug injections, the injection cannulae were left in place for 2 min to allow the solution to diffuse from each cannula tip. The stainless-steel obturators were subsequently reinserted into guide cannulae and the mice returned to their home cage for 30 min before behavioral tests. A mixture (0.5 μl per side) of CNQX (10 mM in aCSF) and D-APV (12.5 mM in aCSF), picrotoxin (100 μM in aCSF, 0.5 μl per side), or their respective vehicles were bilaterally microinfused into the LA ( Supplementary Fig. 3). The injection sites were examined at the end of the experiments, and mice with incorrect diffusion scopes were excluded from further data analysis.

DATA AVAILABILITY
All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Data. Additional data available from authors upon request.