Fear memories allow organisms to avoid danger, thereby increasing their chances of survival. Fear memories can be retrieved long after learning1,2, but little is known about how retrieval circuits change with time3,4. Here we show that the dorsal midline thalamus of rats is required for retrieval of auditory conditioned fear at late timepoints (24 h, 7 d, 28 d), but not early timepoints (0.5 h, 6 h) after learning. Consistent with this, the paraventricular subregion of the dorsal midline thalamus (PVT) showed increased cFos expression only at late timepoints, indicating that PVT is gradually recruited for fear retrieval. Accordingly, the conditioned tone responses of PVT neurons increased with time following training. The prelimbic (PL) prefrontal cortex, which is necessary for fear retrieval5–7, sends dense projections to PVT8. Retrieval at late timepoints activated PL neurons projecting to PVT, and optogenetic silencing of these projections impaired retrieval at late, but not early times. In contrast, silencing of PL inputs to the basolateral amygdala (BLA) impaired retrieval at early, but not late times, indicating a time-dependent shift in retrieval circuits. Retrieval at late timepoints also activated PVT neurons projecting to the central nucleus of the amygdala (CeA), and silencing these projections at late, but not early, times induced a persistent attenuation of fear. Thus, PVT may serve as a critical thalamic node recruited into cortico-amygalar networks for retrieval and maintenance of long-term fear memories.

Previous rodent studies have implicated the infralimbic (IL) subregion of the medial prefrontal cortex in extinction of auditory fear conditioning. However, these studies used pharmacological inactivation or electrical stimulation techniques, which lack temporal precision and neuronal specificity. Here, we used an optogenetic approach to either activate (with channelrhodopsin) or silence (with halorhodopsin) glutamatergic IL neurons during conditioned tones delivered in one of two phases: extinction training or extinction retrieval. Activating IL neurons during extinction training reduced fear expression and strengthened extinction memory the following day. Silencing IL neurons during extinction training had no effect on within-session extinction, but impaired the retrieval of extinction the following day, indicating that IL activity during extinction tones is necessary for the formation of extinction memory. Surprisingly, however, silencing IL neurons optogenetically or pharmacologically during the retrieval of extinction 1 day or 1 week following extinction training had no effect. Our findings suggest that IL activity during extinction training likely facilitates storage of extinction in target structures, but contrary to current models, IL activity does not appear to be necessary for retrieval of extinction memory.

SUMMARY The paraventricular nucleus of the thalamus (PVT) is thought to regulate behavioral responses under emotionally arousing conditions. Reward-associated cues activate PVT neurons, however, the specific PVT efferents regulating reward-seeking remain elusive. Using a cued sucrose-seeking task, we manipulated PVT activity under two emotionally distinct conditions: 1) when reward was available during the cue as expected, or 2) when reward was unexpectedly omitted during the cue. Pharmacological inactivation of the anterior PVT (aPVT), but not the posterior PVT, increased sucrose-seeking only when reward was omitted. Consistent with this, photoactivation of aPVT neurons abolished sucrose-seeking, and the firing of aPVT neurons differentiated reward availability. Photoinhibition of aPVT projections to the nucleus accumbens or to the amygdala increased or decreased, respectively, sucrose-seeking only when reward was omitted. Our findings suggest that PVT bidirectionally modulates sucrose-seeking under the negative (frustrative) conditions of reward omission.

Adult mammalian brains have largely lost neuroregeneration capability except for a few niches. Previous studies have converted glial cells into neurons, but the total number of neurons generated is limited and the therapeutic potential is unclear. Here, we demonstrate that NeuroD1-mediated in situ astrocyte-to-neuron conversion can regenerate a large number of functional new neurons after ischemic injury. Specifically, using NeuroD1 adeno-associated virus (AAV)-based gene therapy, we were able to regenerate one third of the total lost neurons caused by ischemic injury and simultaneously protect another one third of injured neurons, leading to a significant neuronal recovery. RNA sequencing and immunostaining confirmed neuronal recovery after cell conversion at both the mRNA level and protein level. Brain slice recordings found that the astrocyte-converted neurons showed robust action potentials and synaptic responses at 2 months after NeuroD1 expression. Anterograde and retrograde tracing revealed long-range axonal projections from astrocyte-converted neurons to their target regions in a time-dependent manner. Behavioral analyses showed a significant improvement of both motor and cognitive functions after cell conversion. Together, these results demonstrate that in vivo cell conversion technology through NeuroD1-based gene therapy can regenerate a large number of functional new neurons to restore lost neuronal functions after injury.

Infusing brain-derived neurotrophic factor (BDNF) into the infralimbic (IL) prefrontal cortex is capable of inducing extinction. Little is known, however, about the circuits mediating BDNF effects on extinction or the extent to which extinction requires BDNF in IL. Using local pharmacological infusion of BDNF protein, or an antibody against BDNF, we found that BDNF in the IL, but not prelimbic (PL) prefrontal cortex, is both necessary and sufficient for fear extinction. Furthermore, we report that BDNF in IL can induce extinction of older fear memories (14 days) as well as recent fear memories (1 day). Using immunocytochemistry, we show that BDNF is increased in the ventral hippocampus (vHPC), but not IL or PL, following extinction training. Finally, we observed that infusing BDNF into the vHPC increased the firing rate of IL, but not PL neurons in fear conditioned rats. These findings indicate that an extinction-induced increase in BDNF within the vHPC enhances excitability in IL targets, thereby supporting extinction memories.