Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jan 6.
Published in final edited form as: Curr Biol. 2019 Dec 12;30(1):31–41.e3. doi: 10.1016/j.cub.2019.11.006

Aversive Learning Increases Release Probability of Olfactory Sensory Neurons

Janardhan P Bhattarai 1, Mary Schreck 1, Andrew H Moberly 1, Wenqin Luo 1, Minghong Ma 1,1
PMCID: PMC6946845  NIHMSID: NIHMS1542115  PMID: 31839448

Summary

Predicting danger from previously associated sensory stimuli is essential for survival. Contributions from altered peripheral sensory inputs are implicated in this process, but the underlying mechanisms remain elusive. Here we use the mammalian olfactory system to investigate such mechanisms. Primary olfactory sensory neurons (OSNs) project their axons directly to the olfactory bulb (OB) glomeruli where their synaptic release is subject to local and cortical influence and neuromodulation. Pairing optogenetic activation of a single glomerulus with foot shock in mice induces freezing to the light stimulation alone during fear retrieval. This is accompanied by an increase in OSN release probability and a reduction in GABAB receptor expression in the conditioned glomerulus. Furthermore, freezing time is positively correlated with the release probability of OSNs in fear conditioned mice. These results suggest that aversive learning increases peripheral olfactory inputs at the first synapse, which may contribute to the behavioral outcome.

Graphical Abstract

graphic file with name nihms-1542115-f0001.jpg

eTOC Blurb

Bhattarai et al. show that pairing foot shock with optogenetic activation of a single glomerulus in the mouse olfactory bulb induces fear learning, which is accompanied by increased release probability of olfactory sensory neurons. This study highlights that the brain regulates peripheral sensory inputs based on predictive values.

Introduction

Associative learning, the process by which the brain assigns predictive values to sensory stimuli and adjusts behavioral responses, is essential for survival. This process is mediated by neuroplasticity along the sensory pathways as well as in associated cortical and limbic regions [1–3]. Although contributions from altered peripheral sensory inputs into the brain have been implicated [4–7], the underlying mechanisms remain poorly understood.

The mammalian olfactory system provides an experimentally accessible circuit to study learning-induced plasticity in relation to behavioral output. Odor detection and discrimination rely on a large family (>1000 in rodents) of G protein-coupled odorant receptors (ORs) expressed in olfactory sensory neurons (OSNs) in the nasal epithelium. Each OSN expresses a single OR type and all the OSNs with the same OR identity project their axons typically onto two glomeruli (one medial and one lateral) in each olfactory bulb (OB) [8, 9]. The axon terminals of OSNs make synaptic contacts with both glutamatergic projection neurons (mitral and tufted cells) and GABAergic interneurons in the OB. In the glomerular layer external tufted (ET) cells receive monosynaptic inputs from OSNs, while short-axon and periglomerular (PG) cells receive either direct or indirect OSN inputs [10, 11]. The PG cells provide presynaptic inhibition onto OSNs via GABAB receptors at the axonal terminals [12–17]. As PG cells also receive top-down centrifugal inputs, transmitter release from OSNs is subject to both local and cortical influence and neuromodulation [18, 19]. This strategy of modulating sensory inputs into the brain via presynaptic GABAB receptors and local inhibitory neurons is widely adapted by other sensory systems (e.g., visual and auditory) as well [20, 21].

The OSN axon terminals release glutamate onto postsynaptic cells and the release properties can be accessed by recording excitatory postsynaptic currents (EPSCs) in the postsynaptic cells. Olfactory nerve stimulation induces a form of short term plasticity called paired pulse depression; i.e. the EPSC evoked by the second pulse is smaller than the first one. The paired pulse ratio (PPR) of evoked EPSCs (second/first) is routinely used as indication of release probability: a higher (or lower) PPR corresponds with a lower (or higher) release probability of OSNs in the olfactory system [22–25] as well as in other systems [26].

Fear conditioning via odor-foot shock pairing alters odor-induced responses imaged in the OB [4, 5, 7], suggesting a potential change in neurotransmitter release from OSNs. However, direct investigation into this question is hindered by technical challenges in unequivocal identification of conditioned presynaptic OSNs and their postsynaptic partners for electrophysiological recordings. Furthermore, a single odor typically activates multiple glomeruli and the activity of a single glomerulus may be influenced by other glomerular units via the OB network [27–29]. Mice in which OSNs with the M72 receptor coexpress the light-activated cation channel channelrhodopsin 2 (ChR2) can perceive optical stimulation of a single glomerulus [30]. These mice provide an ideal experimental model to address the question whether and how fear conditioning alters neurotransmitter release from genetically identifiable OSNs associated with a single glomerulus.

In this study, we conducted fear conditioning by optogenetically activating a single M72 glomerulus paired with foot shock and demonstrated that mice readily freeze to optical stimulation alone during fear retrieval. We then performed patch clamp recordings from genetically-labeled postsynaptic ET cells innervating M72 glomeruli and found a significant increase in OSN release probability in the conditioned glomerulus compared to control glomeruli (unstimulated or stimulated but without foot shock pairing). The increased release probability is accompanied by reduced GABAB receptor expression in the conditioned glomerulus and is positively correlated with the animal’s freezing time. These results suggest that fear conditioning enhances peripheral olfactory inputs to the brain via GABAB-dependent modulation, which may directly contribute to the behavioral outcome.

Results

Pairing of monoglomerular activation and foot shock induces fear learning

To determine whether associative learning can be achieved by activating a single glomerulus, we used M72-ChR2/Vglut1-tdTomato mice (see STAR Methods for details) in which olfactory sensory neurons (OSNs) expressing the M72 (or Olfr160) receptor coexpress light-sensitive ChR2 and can be activated by blue light (473 nm) [30]. An optical fiber with a diameter of 400 (μm was implanted over a lateral M72 glomerulus of each mouse and the remaining three M72 glomeruli in the same mouse served as “unstimulated” control glomeruli in subsequent slice experiments (Figure 1A). We paired optical stimulation of a single glomerulus (conditioned stimulus, CS) with mild foot shock (unconditioned stimulus, US) in a conditioning session and associative learning was assessed 24 hours later in a retrieval session. The conditioning session consisted of five trials of light-foot shock pairing with an inter trial interval of five minutes, and each pairing consisted of five 150 ms light stimulation at 0.5 Hz followed by 0.5 sec, 0.5 mA foot shock (Figure 1A). During the retrieval session, mice were placed in a novel environment. After three minutes (i.e., pre light-stimulation period), three trials of light stimulation (without foot shock) were delivered with random intervals within 15 min while the animal was videotaped for behavioral assessment. The fear conditioned (FC) mice (n=19) showed robust freezing behavior, whereas the CS-Only mice (n=13) which underwent the same surgery and light stimulation procedure but without foot shock during the conditioning session did not (Figure 1B). To rule out that the freezing behavior displayed by the FC mice was due to general stress caused by foot shock, we included CS-US unpaired mice (n=9), which received optical stimuli and foot shock separately with one hour in between (Figure 1A). Unlike the FC mice, these mice did not show more freezing upon light stimulation during retrieval sessions (Figure 1B). These results suggest that aversive learning can be achieved by optogenetic activation of a single glomerulus paired with foot shock.

Figure 1. Aversive learning can be achieved by optogenetic activation of a single glomerulus paired with foot shock.

Figure 1.

Open in a new tab

A, Schematic drawing of a mouse head showing an optical fiber on one of the four M72 glomeruli and the fear conditioning (FC) and retrieval paradigm in M72-ChR2/Vglut1-tdTomato mice. Blue vertical bars represent light stimulation (CS) and red bars foot shock (US). Inset, a representative retrieval session showing freezing bouts upon light stimulation previously paired with foot shock. B, During retrieval sessions, fear-conditioned but not control mice (CS-Only or CS-US unpaired) spent more time freezing post light-stimulation than pre light-stimulation (one-way ANOVA with Tukey correction, F(5, 81)=13.21, p=2.89E–9. FC pre light 14.9±1.4% vs post light 35.43±4.1%; CS-Only pre light 11.5±3.1% vs post light 7.0±1.3%; CS-US unpaired pre light 14.9±3.3% vs post light 9.8±3.4%).

Release probability of M72 OSNs is assessed in postsynaptic ET cells

In order to investigate how fear conditioning alters neurotransmitter release from OSNs, we used ET cells in the glomerular layer as the postsynaptic readout of transmitter release from M72 OSNs for several reasons. First, mitral cells are located farther away from the glomerulus and may receive weaker and indirect inputs from OSNs [31]. Second, in the glomerular layer, there are multiple types of inhibitory PG cells which receive either direct or indirect inputs from OSNs [14, 28]. Third, ET cells, a major excitatory element in the glomerular layer, receive strong monosynaptic inputs from OSNs [31, 32] and are readily identifiable via genetic labeling in M72-ChR2/Vglut1-tdTomato mice (Figure 2A; See STAR Methods for details).

Figure 2. Transmitter release at M72-ChR2 glomeruli is assessed by measuring light-evoked EPSCs from Vglu1-tdTomato+ external tufted (ET) cells.

Figure 2.

Open in a new tab

A, (Left) Confocal image showing an M72-ChR2 glomerulus with Vglut1-tdTomato+ cells in different layers of the olfactory bulb. GL: glomerular layer. EPL, external plexiform layer. MCL, mitral cell layer. Individual glomeruli are outlined by dotted contours. (Right) Enlarged picture from the rectangle in (A) with two arrows pointing to ET cells associated with the M72 glomerulus. Scale: 50 μm. B, Representative firing patterns demonstrating two subtypes of ET cells associated with M72 glomeruli under current clamp mode. C, Light-evoked EPSCs from an ET cell increase with stimulation duration. Note that the maximum peak current was achieved when the stimulation duration > 6 ms in this cell. D, The peak EPSCs were plotted against the stimulation duration from multiple cells and fitted with the Boltzmann function for sigmoid curves. For clarity, only a subset of the recorded cells (n=21) are included. E, Light-evoked EPSCs were blocked by bath perfusion of CNQX (25 μM). F, Light-evoked EPSCs were blocked by bath perfusion of TTX (0.5 μM) and reappeared after addition of 4-AP (1mM), suggesting monosynaptic sensory neuron inputs.

Approximately 12 hours after the retrieval session, the OB slices with “green” M72 glomeruli were selected for patch clamp recordings. In the glomerular layer, we specifically targeted Vglut1-tdTomato+ ET cells with discernible primary dendrites innervating the M72 glomeruli (Figure 2A). Consistent with a previous study [33], two types of firing patterns were observed upon current injection: initial bursting (20 out of 64 or 31.3%; Figure 2B, upper panel) and adapting (44 out of 64 or 68.8%; Figure 2B, lower panel). Synaptic connection of a recorded ET cell with a M72-ChR2 glomerulus was confirmed by robust light-evoked excitatory postsynaptic currents (EPSCs) with little jitter (Figure 2C). The light evoked EPSCs increased with the pulse length and reached the maximum peak current typically at ~10 ms (Figure 2C, D). To compare OSN release properties across multiple glomeruli from different animals, we tested a series of pulse lengths to determine the minimum stimulus duration that evoked the maximum peak current in a given ET cell. This stimulus duration was then used for the release probability assessment in that cell. As expected, the light-evoked EPSCs were nearly eliminated by 25 μM CNQX (Control: 1473.3 ± 431.5 pA, CNQX: 22.7 ± 3.7 pA, and Washout: 770.7 ± 152.7 pA; n=4), an AMPA receptor antagonist, confirming glutamate mediated transmission (Figure 2E). Light-evoked EPSCs were blocked by tetrodotoxin (TTX; 0.5 (μM) and reappeared after application of TTX plus 1 mM 4-AP, a potassium channel blocker causing action potential independent release (n=3 out of 3), confirming monosynaptic connection (Figure 2F).

Fear conditioning increases release probability of OSNs

To assess transmitter release from OSNs, we measured the paired pulse ratio (PPR) [22–25] of light-evoked EPSCs in the ET cells innervating the M72 glomeruli with varying intervals (100, 200, and 300 ms). Compared to unstimulated M72 glomeruli in FC mice, the PPR of ET cells innervating the light-stimulated glomeruli was significantly reduced (Figure 3A, B, E). Because the peak amplitude of the first EPSC (i.e., evoked by the first light pulse) was not significantly different among these two groups (FC-stimulated: 1152.94±161.93 pA; FC-unstimulated: 1137.29±117.75 pA; student t test, p=0.939), the reduced PPR was mainly due to the reduced amplitude of the second EPSC, indicating an increased release probability. To rule out that the increased release probability was merely due to light activation rather than fear conditioning, we performed the same comparison in CS-Only mice with light stimulation but without foot shock. In these CS-Only control mice, the PPR of ET cells was similar between light-stimulated and unstimulated glomeruli (Figure 3C–E).

Figure 3. Fear conditioning increases the release probability of M72 OSNs.

Figure 3.

Open in a new tab

A-D, Light-evoked EPSCs from ET cells innervating different M72 glomeruli: stimulated (A, C) and unstimulated (B, D) in fear conditioned (FC) and CS-Only mice, respectively. Holding potential = −70 mV under voltage clamp mode. Blue arrows indicate onset of optical stimulation of associated M72 glomeruli and the inter-pulse interval (100, 200, or 300 ms) is defined from the end of the first pulse to the start of the second pulse. E, Summary of the paired pulse ratio (PPR) at varying intervals from ET cells innervating the four types of glomeruli: FC stimulated (100 ms interval: 31.4±2.9; 200 ms: 38.3±3.1; 300 ms: 42.7±3.4 ms; n=22), FC unstimulated (100 ms: 59.2±3.7; 200 ms: 64.7±3.7; 300 ms: 69.2±3.4; n=28), CS-Only stimulated (100 ms: 55.2±4.9; 200 ms: 59.4±4.6; 300 ms: 62.7±4.3; n=7) and CS-Only unstimulated (100 ms: 63.3±2.8; 200 ms: 69.2±2.4; 300 ms: 71.9±2.2; n=27). One-way ANOVA with Tukey correction (100 ms: F(3, 83)=18.28, p=3.9E−9; 200 ms: F(3,83)=17.94, p=5.3E−9; 300 ms F(3, 83)=18.71, p=2.7E−9). Each point represents the mean value of four to six trials from a single ET cell. Results from ET cells with different firing patterns were similar and thus grouped together. F, Relationship between the PPR (inter-pulse interval of 100 ms) and freezing behavior. In some mice, multiple ET cells associated with one glomerulus were recorded and the PPR was reported as mean ± SEM for that mouse.

As FC mice showed individual differences in the freezing time as well as in the release probability from the conditioned M72 glomeruli, we asked whether these two parameters were correlated. We quantified freezing behavior by calculating a freezing index, defined as percent freezing time post light-stimulation/percent freezing time pre light-stimulation during the retrieval session for each mouse. Out of the 14 FC mice in which the PPR was successfully measured from the ET cell(s) associated with the light-stimulated M72 glomeruli, the PPR was negatively correlated with the freezing index (Figure 3F), suggesting a positive correlation between release probability and freezing time. In comparison, the five CS-Only mice showed little freezing and ET cells associated with the light-stimulated M72 glomerulus had high PPRs (Figure 3F). These results suggest that fear conditioning increases the release probability from OSNs, which may contribute to the behavioral output.

Since we measured OSN transmitter release from the postsynaptic ET cells, the observed PPR changes could be due to pre- or postsynaptic mechanisms. To differentiate these two possibilities, we monitored the frequency and amplitude of spontaneous EPSCs (sEPSCs) in the ET cells innervating the M72 glomeruli from FC and CS-Only mice. While the amplitude of sEPSCs was similar among the four types of glomeruli (FC stimulated, FC unstimulated, CS-Only stimulated and CS-Only unstimulated), the frequency was significantly higher in the FC stimulated glomeruli (Figure 4), supporting a presynaptic change in the OSN release properties. We further compared the intrinsic membrane properties of ET cells associated with the stimulated and unstimulated M72 glomeruli in fear conditioned mice. No significant difference was observed in the “Sag” potential (mediated by the hyperpolarization activated current), input resistance, resting membrane potential, and action potential threshold (Figure 4E–H).

Figure 4. Fear conditioning increases the frequency but not amplitude of sEPSCs in the postsynaptic ET cells.

Figure 4.

Open in a new tab

A-B, Representative sEPSC traces recorded from an ET cell associated with the stimulated glomerulus in an FC (A) or CS-Only (B) mouse. Holding potential = −70 mV under voltage clamp mode. C, The frequency of sEPSCs from ET cells associated with different glomeruli: FC stimulated (25.3±3.7 Hz, n=12), FC unstimulated (15.8±2.1 Hz, n=21), CS-Only stimulated (8.9±3.0 Hz, n=5) and CS-Only unstimulated (14.6±1.8 Hz, n=25). There is an increase in the ET cells associated with the FC stimulated M72 glomeruli compared to the other three groups: F(3, 62)=4.47; p=0.006 (one-way ANOVA with Tukey correction). D, The amplitude of sEPSCs from the ET cells associated with different glomeruli: FC stimulated (12.8±1.5 pA, n=12), FC unstimulated (12.9±0.8 pA, n=21), CS-Only stimulated (12.9±0.7 pA, n=5), and CS-Only unstimulated (13.1±0.5 pA, n=25). There is no significant difference: F(3, 62)=0.02, P=0.99 (one-way ANOVA with Tukey correction). E-H, No significant differences observed for intrinsic properties of ET cells associated with FC stimulated and FC unstimulated M72 glomeruli: SAG potential (E), input resistance (F), resting membrane potential (G), and action potential (AP) threshold (H). In E, the Sag potential is plotted against the injected current. Inset shows a typical voltage trace from a hyperpolarizing step for sag potential calculation. NS represents no significant difference (p>0.05; student t test).

Presynaptic GABAB receptors may be involved in altered OSN release probability

A number of neurotransmitters and other molecules (GABA, dopamine, cyclic nucleotides, Zinc, serotonin, etc) in the glomerular layer have been implicated in modulating transmitter release from OSNs [reviewed in 18]. Among these possible mechanisms, the presynaptic GABAB receptor stands out as the most promising candidate as activation of this receptor tempers glutamate release from OSNs in a glomerulus-specific manner [12–17] and impairs olfactory aversive learning in young rats[34]. We hypothesized that GABAB receptors are involved in fear learning induced transmitter release changes from OSNs associated with the conditioned glomerulus. Because GABAB receptors are predominantly expressed in OSN axon terminals within glomeruli [35, 36], we examined the expression pattern of the GABAB receptor 1 (R1) via antibody staining in the glomerular layer of FC and CS-Only mice. Individual glomeruli have varying fluorescent intensity levels (Figure 5A–C). We speculate that each odorant receptor and its associated glomeruli may be stimulated to different extents by environmental odors, which sets a different baseline GABAB receptor expression level for each glomerulus. Nonetheless, the fluorescent intensity level of the stimulated M72 glomeruli from the FC mice was significantly lower than the control M72 glomeruli and non-M72 glomeruli (Figure 5A–D). These findings suggest that downregulation of GABAB receptor expression may play a role in the increased release probability of OSNs after fear conditioning.

Figure 5. Fear conditioning decreases GABAB receptor expression in the stimulated glomeruli.

Figure 5.

Open in a new tab

A-B, Confocal images of OB slices immuostained with GABAB R1 antibody on the FC stimulated (A) and unstimulated side (B). Scale bars = 50 μm. M72 glomeruli are outlined by the dotted contour. C, Histogram showing distribution of relative fluorescent levels of M72 and non-M72 glomeruli in FC and CS-Only mice. The intensity of each glomerulus is normalized to the mean of all non-M72 glomeruli from that image. Arrows pointing to three non-M72 glomeruli adjacent to the stimulated M72 glomeruli from FC mice. D, Summary of relative intensity of different groups of glomeruli. The control M72 group contains the FC unstimulated, and both stimulated and unstimulated glomeruli of CS-Only mice. Fluorescent intensity of FC stimulated M72 glomeruli is significantly lower than the control M72 glomeruli and non-M72 glomeruli (FC stimulated M72 glomeruli: 0.74±0.08, n=4; Control M72 glomeruli: 1.13±0.04, n=17; and non-M72 glomeruli 1.00±0.01, n=304 from four fear conditioned and four CS-Only mice). Independent samples Kruskal-Wallis test: ***p=0.001, **p=0.007, and *p=0.014.

If GABAB receptors are indeed involved, activation or inactivation of these receptors would alter transmitter release from OSNs, which has been supported by previous studies using odor or olfactory nerve stimulation combined with pharmacological manipulations [12–17]. Here we examined the effects of GABAB receptor activation or inactivation on spontaneous and optogenetically-evoked synaptic transmission from the ET cells associated with the M72 glomeruli. In the OB slices under control conditions, sEPSCs were evident from these ET cells (Figure 6A). Application of Baclofen (20 μM), a GABAB agonist, significantly decreased the frequency of sEPSCs (Figure 6A, B), while CGP 55485 (CGP for short; 10 μM), a GABAB antagonist, had the opposite effects (Figure 6A, B). In contrast, neither drug changed the sEPSC amplitude significantly (Figure 6C), supporting a role of GABAB receptors in presynaptic inhibition of OSN transmitter release. Consistent with this notion, Baclofen decreased while CGP increased the light-evoked EPSCs in the ET cells (insets of Figure 6D). Baclofen decreased the peak amplitude of the response to the first light stimulation by 50.1 ± 5.7% (paired t test, p=1.2E−8). When applied alone, CGP increased the peak amplitude by 18.3±4.4% (paired t test, p=0.002). When applied after Baclofen, CGP reversed the reduction caused by Baclofen. Furthermore, Baclofen increased the PPR of light-evoked EPSCs in the ET cells associated with the M72 glomeruli (Figure 6D) supporting that activation of GABAB receptors decreases the release probability of OSNs. The effect of Baclofen on the PPR was partially reversed by CGP, but application of CGP alone did not decrease the PPR as expected (Figure 6D). One potential reason for this outcome is that we titrated the light stimulation duration to induce the maximum peak response in control conditions to measure PPRs, which leaves little room for the peak current to increase in the presence of CGP. In fact, CGP caused only modest increase in peak currents, but substantial increase in the area under light-evoked EPSCs reflecting the net charge carried by synaptic currents (inset of Figure 6D). If we used the ratio of the area (second response/first response) to calculate PPR(area), CGP significantly decreased it (control 53.4±3.6% vs CGP 43.1±3.0%; paired t test p=0.020). Taken together, these immunostaining and pharmacological experiments support a role of GABAB receptors in modulating OSN transmitter release.

Figure 6. Pharmacological manipulation of GABAB receptor activity changes the sEPSC frequency and OSN release probability measured in the ET cells.

Figure 6.

Open in a new tab

A, sEPSCs recorded from an ET cell under control conditions (upper panel), in the presence of Baclofen (20 μM; middle panel), and in the presence of Baclofen + CGP 55845 (10 (μM; lower panel). Holding potential = −70 mV under voltage clamp mode. B, Summary data showing decreased or increased sEPSC frequency after Baclofen (left panel: paired t test, p=1.0E−9; n=15) or CGP application, respectively (middle panel for CGP alone: paired t test, p=0.004, n=8; and right panel for CGP + Baclofen: paired t test, p=0.039, n=7). In the left panel, two different symbols denote that Baclofen was applied first or after CGP, and the latter group showed less reduction, which may reflect incomplete washout of CGP. Each data point was analyzed from 2 min recording (3 min after drug application) and normalized to the value right before the drug application. C, There was no significant difference in the sEPSC amplitude after Baclofen or CGP (paired t test, Baclofen: p=0.081; CGP: p=0.670). D, Insets, Baclofen decreased while CGP increased light-evoked EPSCs in the ET cells. Baclofen increased the PPR of light evoked EPSCs in the ET cells associated with M72 glomeruli (left panel: paired t test, p=5.2E−4, n=20), which was reversed by CGP application (right panel: paired t test, p=0.032, n=8). No significant change in PPR was observed when CGP was applied alone (middle panel: paired t test, p=0.737, n=10).

Fear conditioning decreases lateral inhibition from the conditioned glomeruli

In addition to increased OSN release probability, fear conditioning may change other aspects of the OB circuit to modulate information processing in the olfactory pathway. We next examined potential changes in lateral inhibition, which plays a pivotal role in shaping OB activity [37, 38]. Similar to a previous study [38], we assessed lateral inhibition by recording ET cells associated with adjacent, non-M72 glomeruli (typically within two glomeruli from M72) (Figure 7A). In contrast to ET cells associated with M72 glomeruli (Figures 2, 3), we did not observe light-evoked inward currents using normal intracellular solution in the non-M72 ET cells, suggesting minimal lateral excitation (Figure 7B, upper trace). We therefore used high [Cl−] pipette solution (see STAR method) so that GABAA receptor mediated Cl− currents would be inward at a holding potential of −70 mV to achieve more stable recordings (Figure 7B, middle trace). We confirmed that light-evoked currents in the non-M72 ET cells were completely blocked by bicuculline, a GABAA receptor antagonist (Figure 7B, lower trace). Compared to 65% non-M72 ET cells (13 out of 20) adjacent to the control M72 glomeruli (unstimulated in the FC mice, n=7) that showed lateral inhibition, only 20.7% (6 out of 29) ET cells adjacent to the conditioned M72 glomeruli from the same mice did (Figure 7C; p =0.003, χ2 test). These results suggest that fear conditioning induces multiple changes in the OB circuit to modulate how the brain processes the conditioned stimulus.

Figure 7. Fear conditioning decreases the incidence of lateral inhibition from the stimulated M72 glomeruli onto adjacent non-M72 ET cells.

Figure 7.

Open in a new tab

A, Confocal image showing an M72-ChR2 glomerulus (green) and non-M72 glomeruli (marked with white dotted contours). White arrows mark Vglut1-tdTomato+ ET cells associated with non-M72 glomeruli. B, Upper panel: representative whole-cell voltage-clamp traces from a non-M72 ET cell showing no lateral excitation upon optical stimulation of M72 glomeruli (n = 10 out of 10 cells). Middle panel: light-evoked inward currents recorded from non-M72 ET cells in high Cl− intracellular solution, which makes GABAA dependent Cl− currents inward at a holding potential of −70 mV. Lower panel: light-evoked inward currents completely blocked by bicuculline (20 μM), a GABAA receptor antagonist (n = 6 out of 6 cells). C, Stacked bar graph shows that a significantly lower percentage of non-M72 ET cells adjacent to the stimulated (conditioned) M72 glomeruli displaying lateral inhibition compared to the unstimulated (control) M72 glomeruli (χ2 test, p=0.003).

Discussion

Here we demonstrate that stimulating a single glomerulus in the OB paired with foot shock is sufficient to induce fear learning in mice. This fear conditioning paradigm leads to enhanced neurotransmitter release from OSNs that project to the conditioned glomerulus, which is accompanied by reduced GABAB receptor expression and reduced incidence of lateral inhibition. Furthermore, the release probability of OSNs is positively correlated with the animal’s freezing time. This study provides the first direct evidence that fear conditioning enhances release probability from primary sensory neurons in any mammalian sensory system.

Many species including rodents rely mostly on olfactory cues to avoid danger and their olfactory systems display robust associative plasticity in various fear conditioning paradigms. Aversive learning induces plasticity not only in cortical and limbic areas [39–42], but also in early stages of the olfactory system [4–7, 43], suggesting that the same sensory stimulus sends different inputs to the brain depending on its predictive value. OSNs regenerate throughout life [44], and fear conditioning using an M71 ligand increases the number of M71 OSNs in the olfactory epithelium three weeks later [6]. Exposure to certain odorants especially during early life promotes glomerular refinement and increases the glomerular size and mitral/tufted cell number, which may enhance specific information transmission [45–47]. Significant anatomical changes of the glomerular units are unlikely to take place during short-term odor-foot shock pairing paradigms, which nevertheless increase conditioned odor-induced activity in the OB [4, 5, 7]. In genetically modified mice in which OSNs express the fluorescent exocytosis indicator synaptopHluorin [48], discriminative fear conditioning (CS+ with US and CS− without US) increases the glomerular responses to CS+ odors but not to CS− odors [5]. Similarly, classical fear conditioning (CS-US pairing) also increases the activity from mitral and tufted cells imaged at their dendrites in individual glomeruli [4, 7] as well as from periglomerular cells in the OB [43]. In this study, we provide direct electrophysiological evidence to support that fear conditioning changes peripheral olfactory inputs from OSNs to the OB neurons. Pairing optogenetic activation of a single glomerulus with foot shock leads to successful fear learning in M72-ChR2/Vglut1-tdTomato mice (Figure 1). This is consistent with a previous study which shows that mice can perceive light activation of a single M72 glomerulus and detect small changes in odor-evoked glomerular spatial patterns [30]. This approach assigns learning to a single glomerulus with little interglomerular influence and permits further investigation of synaptic physiology in a cell-type specific manner, which would be otherwise impossible in classical odor-induced fear conditioning paradigms. The fear-conditioned M72 glomerulus and the associated postsynaptic ET cells are identifiable in OB slices, and the three unstimulated M72 glomeruli in the same animal or the four M72 glomeruli (one stimulated and three unstimulated) from CS-Only mice serve as controls (Figures 1–3). Although our monoglomerular fear conditioning paradigm may not fully replicate the involvement of the OB circuit in olfactory learning, it offers unprecedented precision in revealing synaptic transmission changes from the conditioned OSNs. It is very likely that what we have learned here will also apply to classical olfactory learning because both approaches elicit robust freezing responses to cued stimuli and the effects of pharmacological manipulations are similar between olfactory nerve stimulation- and light-evoked EPSCs (see below).

Our patch clamp analysis reveals that the PPR of light-evoked EPSCs in the ET cells from the conditioned M72 glomeruli is significantly lower than the controls (unstimulated or stimulated without foot shock pairing) and the sEPSC frequency is higher (Figures 3, 4), suggesting an increased release probability from presynaptic OSNs. This represents the first direct evidence that fear conditioning enhances release probability from primary sensory neurons in any mammalian sensory system. The yield of electrophysiological recordings from stimulated M72 glomeruli is relatively low as there is only one glomerulus/mouse with a small number of associated ET cells. Nonetheless, we were able to measure the PPR of light-evoked EPSCs in ET cells from 14 FC mice and found a positive correlation between the release probability of OSNs and the freezing duration (Figure 3). This finding does not imply that the increased release probability from OSNs associated with the conditioned glomeruli is sufficient to induce the fear response. Other changes within the OB, olfactory cortex, and fear circuits must be involved. In fact, we have observed reduced incidence of lateral inhibition onto the neighboring ET cells from the conditioned M72 glomeruli (Figure 7). Since there are no obvious changes in intrinsic membrane properties of ET cells (Figure 4), the reduced incidence of lateral inhibition is likely due to fear conditioning induced modification of the interglomerular circuit involving short-axon and PG cells. Further experiments are needed to determine the underlying mechanisms and to what extent the release probability change at the first synapse in the olfactory pathway contributes to subsequent plasticity along the pathway and to the behavioral response in fear retrieval.

To shed light on the cellular mechanism underlying this fast form of neuroplasticity, we explored possible involvement of GABAB receptors because presynaptic GABAB receptors temper transmitter release by suppressing calcium influx, activate inwardly rectifying K+ channels, and/or increase the energy barrier for vesicle fusion [49, 50]. Here we show that the GABAB receptor expression level in the conditioned M72 glomeruli is significantly lower than control M72 glomeruli (Figure 5), consistent with the enhanced release probability and sEPSC frequency in the ET cells associated with the conditioned glomeruli (Figures 3, 4). Fear conditioning may lead to internalization of GABAB receptors in the presynaptic terminals of OSNs, as reported in other synapses [51]. It remains to be determined what signals cause the downregulation of GABAB receptors in the conditioned glomeruli, but most likely it involves centrifugal inputs into the OB. For instance, the glomerular layer receives extensive top-down inputs from olfactory cortices and neuromodulatory centers (e.g., serotonin from raphe nuclei, acetylcholine from basal forebrain, and norepinephrine from locus ceruleus) [52–55].

We notice that the immediately adjacent glomeruli of the conditioned M72 glomerulus also show relatively lower GABAB receptor expression compared to other non-M72 glomeruli (Figure 5). This finding suggests that the mechanisms that are involved in downregulating GABAB receptors may not selectively target the conditioned glomerulus only, but rather a small area covering multiple glomeruli. Together with other fear conditioning induced changes in the OB circuit and neuromodulation [3, 5], this phenomenon may contribute to olfactory fear generalization [4, 43]. To test this possibility, future experiments would require activating adjacent glomeruli individually while recording from their postsynaptic cells.

Although the immunostaining experiments shown in Figure 5 do not differentiate GABAB receptors on the pre- or post-synaptic cells, previous studies using immuno-electron microscopy demonstrates that GABAB positive neuropils within glomeruli are predominantly axon terminals of the olfactory nerve [35, 36]. These GABAB receptors mediate presynaptic inhibition and suppress glutamate release from OSNs, supported by ex vivo as well as in vivo pharmacological studies [12–17]. In OB slices, the GABAB receptor agonist Balcofen decreases olfactory nerve stimulation evoked EPSCs in both PG and ET cells, while the GABAB antagonist CGP has the opposite effect [14]. Here we show that these pharmacological manipulations have similar effects on the evoked EPSCs in the ET cells upon optogenetic activation of OSNs associated with a single glomerulus (Figure 6). In addition, Baclofen decreases and CGP increases the sEPSC frequency in the ET cells, supporting a presynaptic action of GABAB receptors (Figure 6).

The effects of these drugs on the PPR of the evoked EPSCs are more complicated. Activating GABAB receptors by Baclofen increases the PPR mainly by reducing the response to the first stimulus, which is at least partially reversed by CGP (Figure 6). The potential role of GABAB receptors in mediating the OSN release probability is also supported by previous reports that CGP decreases the PPR evoked by olfactory nerve stimulation and measured in postsynaptic cells [14, 17]. However, in our slice recordings, blocking GABAB receptors by CGP alone does not change the PPR. This could be due to the use of light stimulation that induces near maximum response leaving little room for the peak current to increase. When we calculate the PPR based on the net charge carried by EPSCs, CGP decreases the PPR(area). Additionally, this could be due to a relatively low activity of GABAB receptors in the absence of agonists in control conditions. Moreover, our finding could mean that fear conditioning induced changes in light-evoked EPSCs involves both GABAB-dependent and GABAB-independent mechanisms, such as OSN synaptogenesis [56].

Olfactory aversive learning induced neuroplasticity has been reported in other species including humans. After aversive conditioning using two enantiomers indiscriminable to subjects, the evoked activity pattern in the piriform cortex becomes more distinct and the ability to discriminate the odors is enhanced [57]. It is plausible that plasticity observed in the piriform cortex reflects cumulative changes from the previous stages including the OSNs and the OB network as well as the cortical circuits.

What we have learned from the olfactory pathway may also apply to other sensory modalities as well. Presynaptic GABAB receptors and local inhibitory neurons are known to modulate visual and auditory glutamatergic inputs at their first synapses in the brain [20, 21]. Fear learning based on visual or auditory cues may increase the release probability of the neurons that respond to the conditioned stimulus via downregulation of presynaptic GABAB receptor expression. Our study reveals fear conditioning induced neuroplasticity at the first synapse in the olfactory pathway, highlighting the ability of the brain to adjust sensory inputs based on predictive values.

STAR Methods

LEAD CONTACT AND MATERIALS AVAILABILITY

This study did not generate new unique reagents. Further information and requests for resources and reagents should be directed to the Lead Contact, Minghong MA (minghong@pennmedicine.upenn.edu).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

M72-IRES-ChR2-YFP (in brief M72-ChR2) mice (JAX Stock No: 021206) in which OSNs expressing the M72 receptor coexpress fused ChR2-yellow fluorescent protein were provided by Drs. Thomas Bozza and Dmitry Rinberg [30]. In order to achieve fluorescent labeling of the OB projection neurons for electrophysiological recordings, M72-ChR2 mice were crossed with Vglut1-Cre mice (Vglut1-IRES2-Cre-D; JAX Stock No: 023527) [58] and the Cre-dependent tdTomato reporter mice (Rosa26-CAG-LSL-tdTomato-WPRE or Ai9 line; JAX Stock No: 007905) [59] to obtain triple transgenic M72-ChR2+/+;Vglut1Cre/+;RosatdTomatof/+ mice (in brief M72-ChR2/Vglut1-tdTomato) for experimental use. All mice were housed under 12 hr light-12 hr dark cycles with food and water ad libitum in a temperature- and humidity-controlled animal facility. Mice were randomly assigned for control or experimental groups. All behavioral and recording procedures were performed during the light phase of the cycle. Both male and female mice (2–3 months old, unless otherwise stated) were used. All procedures were approved by the University of Pennsylvania Institutional Animal Care and Use Committee.

METHOD DETAILS

Surgical Implantation

Mice were anesthetized via isoflurane at 3% (vol/vol) and secured in a stereotaxic system (Model 940, David Kopf Instruments). Isoflurane levels were maintained at 1.5–2% for the remainder of the surgery. Body temperature was maintained at 37 °C with a temperature control system (TC-1000, CWE Inc.). The overlying bone of the left olfactory bulb was thinned until the blood vessels over the OB were clearly visible and the lateral M72 glomerulus was located under a fluorescent dissecting microscope (Leica M80). A fiber optic cannula stub (400 (μm core, 2.5 mm fiber length, Thor Labs) was positioned above the glomerulus and fixed in place with Kwik-Sil adhesive (World Precision Instruments) and dental cement. After surgery, mice were caged individually and allowed to recover for 1 week prior to fear conditioning.

Fear Conditioning and Retrieval

In the fear conditioning session (Day 1), mice were acclimatized to the foot shock chamber (Modular Chamber, ENV-307A, Med Associates Inc.) for at least one hour. Fear conditioning was performed by pairing optical stimulation (conditioned stimulus or CS: five 150 ms pulses at 0.5 Hz, with each light pulse mimicking the length of an inhalation) with mild foot shock (unconditioned stimulus or US: 0.5 sec at 0.5 mA). The onset of the US coincided with the offset of the CS and such pairing repeated five times with an inter-trial interval of five minutes (Figure 1A). Optical stimulation was delivered from a 473nm laser (SLOC Lasers, BL473T8–150FC) through a fiber optic rotary joint patch cable (Thor Labs). The laser power was adjusted to 50 mW/mm2 measured from the optic fiber tip and the timing was controlled by an arbitrary waveform generator (Agilent 33201A). In CS-Only mice, the same optical stimuli were delivered but without foot shock. In CS-US unpaired mice, optical stimuli (CS) and foot shocks (US) were delivered separately with one hour in between (Figure 1A). In the fear retrieval session 24 hours later (Day 2), mice were put into a novel chamber (7 1/2” × 11 1/2” × 5”). After 3 min of free exploration period, three trials of light stimulation (five 150 ms pulses at 0.5 Hz) at a random interval of 3 to 5 minutes were presented without foot shock within 15 min. Before each conditioning or retrieval session, the mouse was briefly anesthetized via isoflurane at 3% (vol/vol) and the laser fiber was coupled to the implanted fiber stub with a mating sleeve (Thorlabs Inc.), which was further covered with an opaque tubing to prevent light coming out of the junction of the two ferrules. To prevent mice from using the leaked light as a visual cue for associative learning, bright blue LED was positioned over the fear conditioning and retrieval chamber to mask the dull leaked light.

Patch Clamp Recording

Approximately 12 hr after the retrieval session, the mice were sacrificed for electrophysiological recordings in acute brain slices. Mice were deeply anesthetized with ketamine-xylazine (200 and 15 mg/kg body weight, respectively) and decapitated. The brain was dissected out and immediately placed in ice-cold HEPES buffer. Using a Leica VT 1200S vibratome, relatively thin coronal OB slices (thickness=130 μm) were cut to facilitate visualization of the M72 glomeruli (diameter ~70 μm) and accessibility of surrounding cells. Slices were incubated in oxygenated artificial cerebrospinal fluid (ACSF) containing the following (in mM): 126 NaCl, 2.5 KCl, 2.4 CaCl2, 1.2 MgSO4, 11 D-glucose, 1.4 NaH2PO4 and 25 NaHCO3 (osmolality ~305 mOsm and pH 7.4, bubbled with 95% O2−5% CO2) for 1 hr at 31°C and kept in oxygenated ACSF in room temperature thereafter. Before recording, slices were transferred to a recording chamber and continuously perfused with oxygenated ACSF. M72-ChR2-YFP glomeruli and Vglut1-tdTomato+ cells were visualized with a 40× water-immersion objective under an Olympus BX51WI upright microscope equipped with epifluorescence. Blue light used to activate ChR2 in the OB slices was provided by a Lambda DG-4 (Sutter Instrument) through a band pass filter (450 to 490 nm).

Whole-cell patch-clamp recordings were made in both current and voltage-clamp mode. Recording pipettes were made from borosilicate glass (GC210F-10; Harvard Apparatus) with a Flaming-Brown P-97 puller (Sutter Instruments; tip resistance 5–8 MΩ). The pipette solution contained the following (in mM): 120 K-gluconate, 10 NaCl, 1 CaCl2, 10 EGTA, 10 HEPES, 5 Mg-ATP, 0.5 Na-GTP, and 10 phosphocreatine. For lateral inhibition experiments, a high Cl− pipette solution (120 mM KCl instead of K-gluconate) was used so that the reversal potential of [Cl−] was at ~ 0 mV and GABAA receptor mediated currents would be inward at a holding potential of −70 mV.

For a subset of neurons, intrinsic membrane properties were measured via current injections under current-clamp mode. The resting membrane potential was determined when there is no current injected. For the remaining properties, the cell membrane potential was kept at −70 mV. The input resistance was calculated as the slope of the current-voltage (I-V) curve (seven current steps from −100 pA with 20 pA increments). The “sag” potential, mediated by the hyperpolarization-activated current, was measured as the difference between the peak and steady-state potential at the end of the current steps as previously described [60]. The action potential (AP) threshold was determined as the point with the maximum curvature before the AP evoked by the minimum current step (out of a series of 10 ms, 2 pA depolarizing current steps).

Electrophysiological recordings were controlled by an EPC-10 amplifier combined with Pulse Software (HEKA Electronic) and analyzed using Igor Pro and mini-analysis (Synaptosoft, Inc.). Pharmacological reagents including tetrodotoxin (TTX) citrate (Abcam), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 4-Aminopyridine (4-AP), Baclofen (Sigma-Aldrich) and CGP 55845 (Tocris Bioscience) were bath perfused.

Immunohistochemistry

After electrophysiological recordings, brain slices were fixed in 4% paraformaldehyde (Sigma) for 15–20 min at 4 °C. Sections were washed in 1% Triton X-100 and 1% Tween 20 in PBS (3X 10 minutes), blocked for 60 min in 1% Triton X-100 and 1% Tween 20 in PBS with 3% bovine serum albumin, and then incubated at 4 °C with the mouse anti-GABAB-R1 (1:500, H00002550-M01, ABNOVA) primary antibody in the same solution overnight. Immunofluorescence was achieved by reaction with the secondary antibody goat anti-mouse, Alexa Fluor 633 (A21052, Molecular Probes, Invitrogen) at 1:1000 overnight at 4 °C. Tissues were washed in 1% Tween 20 in PBS (3X 20 min) and mounted with Fluoromount-G (Southern Biotech). Fluorescent images were taken under a SP5/Leica confocal microscope with LAS AF Lite software. The fluorescent intensity of individual glomeruli was quantified from unprocessed images using ImageJ. For each image, the same-size circle (~600 μm2) was placed in the middle of each glomerulus void of stained cell bodies. After subtracting the background intensity (measured using the same circle placed in an area void of GABAB staining), the fluorescent intensity of each glomerulus was calculated and normalized to the mean of all non-M72 glomeruli on the same image.

QUANTIFICATION AND STATISTICAL ANALYSIS

In behavioral experiments, freezing (≥ 1 sec immobility) was scored by ANY-maze behavioral tracking software and verified by manual scoring of the recorded videos. In patch clamp experiments, ET cells connected to M72-ChR2 glomeruli (Figure 2A), showed robust light-induced inward currents, which increased with stimulation duration (Figure 2C). An ET cell associated with a non-M72 glomerulus (Figure 7) was defined as showing lateral inhibition if the light-evoked current peak was more than thrice the root mean square noise (1.67 ± 0.07 pA; n=26 random traces) in more than 60% of the trials. All data were expressed as mean ± S.E.M. Student t-test was used for statistical comparison of two groups and independent samples Kruskal-Wallis test (SPSS 20) or one-way ANOVA followed by post hoc Tukey test for multiple groups (ORIGIN 8.0). Chi-square test was applied for percentage comparisons (SPSS 20). A p value <0.05 was considered to be significant (* p<0.05, ** p<0.01, *** p<0.001, and **** p<0.0001).

DATA AND CODE AVAILABILITY

Available from the Lead Contact upon reasonable request.

Highlights.

  • Monoglomerular activation paired with foot shock induces fear learning

  • Fear learning increases release probability of OSNs

  • Freezing time is positively correlated with release probability of OSNs

  • Fear learning causes reduction of presynaptic GABAb receptor expression in OSNs

Acknowledgements

We thank Drs. Thomas Bozza and Dmitry Rinberg for sharing the M72-IRES-ChR2-YFP mouse line and genotyping protocol. This work was supported by the National Institute of Health (F31 DC017054 and NIMH T32 MH017168 to M.S. and NIDCD R01 DC006213 to M.M.).

Footnotes

Declaration of Interests

The authors declare no competing interests.

References

  • 1.McGann JP (2015). Associative learning and sensory neuroplasticity: how does it happen and what is it good for? Learn Mem 22, 567–576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gallistel CR, and Matzel LD (2013). The neuroscience of learning: beyond the Hebbian synapse. Annu Rev Psychol 64, 169–200. [DOI] [PubMed] [Google Scholar]
  • 3.Ross JM, and Fletcher ML (2019). Aversive learning-induced plasticity throughout the adult mammalian olfactory system: insights across development. J Bioenerg Biomembr 51, 15–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ross JM, and Fletcher ML (2018). Learning-Dependent and -Independent Enhancement of Mitral/Tufted Cell Glomerular Odor Responses Following Olfactory Fear Conditioning in Awake Mice. J Neurosci 38, 4623–4640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kass MD, Rosenthal MC, Pottackal J, and McGann JP (2013). Fear learning enhances neural responses to threat-predictive sensory stimuli. Science 342, 1389–1392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Jones SV, Choi DC, Davis M, and Ressler KJ (2008). Learning-dependent structural plasticity in the adult olfactory pathway. J Neurosci 28, 13106–13111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Fletcher ML (2012). Olfactory aversive conditioning alters olfactory bulb mitral/tufted cell glomerular odor responses. Front Syst Neurosci 6, 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Murthy VN (2011). Olfactory maps in the brain. Annu Rev Neurosci 34, 233–258. [DOI] [PubMed] [Google Scholar]
  • 9.Mori K, and Sakano H (2011). How is the olfactory map formed and interpreted in the mammalian brain? Annu Rev Neurosci 34, 467–499. [DOI] [PubMed] [Google Scholar]
  • 10.Nagayama S, Homma R, and Imamura F (2014). Neuronal organization of olfactory bulb circuits. Front Neural Circuits 8, 98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wachowiak M, and Shipley MT (2006). Coding and synaptic processing of sensory information in the glomerular layer of the olfactory bulb. Semin Cell Dev Biol 17, 411–423. [DOI] [PubMed] [Google Scholar]
  • 12.Wachowiak M, McGann JP, Heyward PM, Shao Z, Puche AC, and Shipley MT (2005). Inhibition [corrected] of olfactory receptor neuron input to olfactory bulb glomeruli mediated by suppression of presynaptic calcium influx. J Neurophysiol 94, 2700–2712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Murphy GJ, Darcy DP, and Isaacson JS (2005). Intraglomerular inhibition: signaling mechanisms of an olfactory microcircuit. Nat Neurosci 8, 354–364. [DOI] [PubMed] [Google Scholar]
  • 14.Shao Z, Puche AC, Kiyokage E, Szabo G, and Shipley MT (2009). Two GABAergic intraglomerular circuits differentially regulate tonic and phasic presynaptic inhibition of olfactory nerve terminals. J Neurophysiol 101, 1988–2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Pirez N, and Wachowiak M (2008). In vivo modulation of sensory input to the olfactory bulb by tonic and activity-dependent presynaptic inhibition of receptor neurons. J Neurosci 28, 6360–6371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.McGann JP, Pirez N, Gainey MA, Muratore C, Elias AS, and Wachowiak M (2005). Odorant representations are modulated by intra- but not interglomerular presynaptic inhibition of olfactory sensory neurons. Neuron 48, 1039–1053. [DOI] [PubMed] [Google Scholar]
  • 17.Aroniadou-Anderjaska V, Zhou FM, Priest CA, Ennis M, and Shipley MT (2000). Tonic and synaptically evoked presynaptic inhibition of sensory input to the rat olfactory bulb via GABA(B) heteroreceptors. J Neurophysiol 84, 1194–1203. [DOI] [PubMed] [Google Scholar]
  • 18.McGann JP (2013). Presynaptic inhibition of olfactory sensory neurons: new mechanisms and potential functions. Chem Senses 38, 459–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Harvey JD, and Heinbockel T (2018). Neuromodulation of Synaptic Transmission in the Main Olfactory Bulb. Int J Environ Res Public Health 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Chen C, and Regehr WG (2003). Presynaptic modulation of the retinogeniculate synapse. J Neurosci 23, 3130–3135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Brenowitz S, David J, and Trussell L (1998). Enhancement of synaptic efficacy by presynaptic GABA(B) receptors. Neuron 20, 135–141. [DOI] [PubMed] [Google Scholar]
  • 22.Murphy GJ, Glickfeld LL, Balsen Z, and Isaacson JS (2004). Sensory neuron signaling to the brain: properties of transmitter release from olfactory nerve terminals. J Neurosci 24, 3023–3030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Tyler WJ, Petzold GC, Pal SK, and Murthy VN (2007). Experience-dependent modification of primary sensory synapses in the mammalian olfactory bulb. J Neurosci 27, 9427–9438. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Grubb MS, Nissant A, Murray K, and Lledo PM (2008). Functional maturation of the first synapse in olfaction: development and adult neurogenesis. J Neurosci 28, 2919–2932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Vaaga CE, and Westbrook GL (2017). Distinct temporal filters in mitral cells and external tufted cells of the olfactory bulb. J Physiol 595, 6349–6362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Zucker RS, and Regehr WG (2002). Short-term synaptic plasticity. Annu Rev Physiol 64, 355–405. [DOI] [PubMed] [Google Scholar]
  • 27.Aungst JL, Heyward PM, Puche AC, Karnup SV, Hayar A, Szabo G, and Shipley MT (2003). Centre-surround inhibition among olfactory bulb glomeruli. Nature 426, 623–629. [DOI] [PubMed] [Google Scholar]
  • 28.Kiyokage E, Pan YZ, Shao Z, Kobayashi K, Szabo G, Yanagawa Y, Obata K, Okano H, Toida K, Puche AC, et al. (2010). Molecular identity of periglomerular and short axon cells. J Neurosci 30, 1185–1196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Whitesell JD, Sorensen KA, Jarvie BC, Hentges ST, and Schoppa NE (2013). Interglomerular lateral inhibition targeted on external tufted cells in the olfactory bulb. J Neurosci 33, 1552–1563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Smear M, Resulaj A, Zhang J, Bozza T, and Rinberg D (2013). Multiple perceptible signals from a single olfactory glomerulus. Nat Neurosci 16, 1687–1691. [DOI] [PubMed] [Google Scholar]
  • 31.Gire DH, Franks KM, Zak JD, Tanaka KF, Whitesell JD, Mulligan AA, Hen R, and Schoppa NE (2012). Mitral cells in the olfactory bulb are mainly excited through a multistep signaling path. J Neurosci 32, 2964–2975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hayar A, Karnup S, Ennis M, and Shipley MT (2004). External tufted cells: a major excitatory element that coordinates glomerular activity. J Neurosci 24, 6676–6685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Antal M, Eyre M, Finklea B, and Nusser Z (2006). External tufted cells in the main olfactory bulb form two distinct subpopulations. Eur J Neurosci 24, 1124–1136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Okutani F, Zhang JJ, Otsuka T, Yagi F, and Kaba H (2003). Modulation of olfactory learning in young rats through intrabulbar GABA(B) receptors. Eur J Neurosci 18, 2031–2036. [DOI] [PubMed] [Google Scholar]
  • 35.Bonino M, Cantino D, and Sassoe-Pognetto M (1999). Cellular and subcellular localization of gamma-aminobutyric acidB receptors in the rat olfactory bulb. Neurosci Lett 274, 195–198. [DOI] [PubMed] [Google Scholar]
  • 36.Panzanelli P, Lopez-Bendito G, Lujan R, and Sassoe-Pognetto M (2004). Localization and developmental expression of GABA(B) receptors in the rat olfactory bulb. J Neurocytol 33, 87–99. [DOI] [PubMed] [Google Scholar]
  • 37.Liu S, Plachez C, Shao Z, Puche A, and Shipley MT (2013). Olfactory bulb short axon cell release of GABA and dopamine produces a temporally biphasic inhibition-excitation response in external tufted cells. J Neurosci 33, 2916–2926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Geramita MA, Burton SD, and Urban NN (2016). Distinct lateral inhibitory circuits drive parallel processing of sensory information in the mammalian olfactory bulb. Elife, 5, e16039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chen CF, Barnes DC, and Wilson DA (2011). Generalized vs. stimulus-specific learned fear differentially modifies stimulus encoding in primary sensory cortex of awake rats. J Neurophysiol 106, 3136–3144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Sevelinges Y, Gervais R, Messaoudi B, Granjon L, and Mouly AM (2004). Olfactory fear conditioning induces field potential potentiation in rat olfactory cortex and amygdala. Learn Mem 11, 761–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mouly AM, Fort A, Ben-Boutayab N, and Gervais R (2001). Olfactory learning induces differential long-lasting changes in rat central olfactory pathways. Neuroscience 102, 11–21. [DOI] [PubMed] [Google Scholar]
  • 42.Meissner-Bernard C, Dembitskaya Y, Venance L, and Fleischmann A (2019). Encoding of Odor Fear Memories in the Mouse Olfactory Cortex. Curr Biol 29, 367–380. [DOI] [PubMed] [Google Scholar]
  • 43.Kass MD, and McGann JP (2017). Persistent, generalized hypersensitivity of olfactory bulb interneurons after olfactory fear generalization. Neurobiol Learn Mem 146, 47–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Brann JH, and Firestein SJ (2014). A lifetime of neurogenesis in the olfactory system. Front Neurosci 8, 182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Todrank J, Heth G, and Restrepo D (2011). Effects of in utero odorant exposure on neuroanatomical development of the olfactory bulb and odour preferences. Proc Biol Sci 278, 1949–1955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Kerr MA, and Belluscio L (2006). Olfactory experience accelerates glomerular refinement in the mammalian olfactory bulb. Nat Neurosci 9, 484–486. [DOI] [PubMed] [Google Scholar]
  • 47.Liu A, Savya S, and Urban NN (2016). Early Odorant Exposure Increases the Number of Mitral and Tufted Cells Associated with a Single Glomerulus. J Neurosci 36, 11646–11653. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Bozza T, McGann JP, Mombaerts P, and Wachowiak M (2004). In vivo imaging of neuronal activity by targeted expression of a genetically encoded probe in the mouse. Neuron 42, 9–21. [DOI] [PubMed] [Google Scholar]
  • 49.Benke D, Balakrishnan K, and Zemoura K (2015). Regulation of cell surface GABA(B) receptors: contribution to synaptic plasticity in neurological diseases. Adv Pharmacol 73, 41–70. [DOI] [PubMed] [Google Scholar]
  • 50.Gassmann M, and Bettler B (2012). Regulation of neuronal GABA(B) receptor functions by subunit composition. Nat Rev Neurosci 13, 380–394. [DOI] [PubMed] [Google Scholar]
  • 51.Guetg N, Abdel Aziz S, Holbro N, Turecek R, Rose T, Seddik R, Gassmann M, Moes S, Jenoe P, Oertner TG, et al. (2010). NMDA receptor-dependent GABAB receptor internalization via CaMKII phosphorylation of serine 867 in GABAB1. Proc Natl Acad Sci U S A 107, 13924–13929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Petzold GC, Hagiwara A, and Murthy VN (2009). Serotonergic modulation of odor input to the mammalian olfactory bulb. Nat Neurosci 12, 784–791. [DOI] [PubMed] [Google Scholar]
  • 53.Eckmeier D, and Shea SD (2014). Noradrenergic plasticity of olfactory sensory neuron inputs to the main olfactory bulb. J Neurosci 34, 15234–15243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Boyd AM, Sturgill JF, Poo C, and Isaacson JS (2012). Cortical feedback control of olfactory bulb circuits. Neuron 76, 1161–1174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Fast CD, and McGann JP (2017). Amygdalar Gating of Early Sensory Processing through Interactions with Locus Coeruleus. J Neurosci 37, 3085–3101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Cheetham CE, Park U, and Belluscio L (2016). Rapid and continuous activity-dependent plasticity of olfactory sensory input. Nat Commun 7, 10729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Li W, Howard JD, Parrish TB, and Gottfried JA (2008). Aversive learning enhances perceptual and cortical discrimination of indiscriminable odor cues. Science 319, 1842–1845. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Harris JA, Hirokawa KE, Sorensen SA, Gu H, Mills M, Ng LL, Bohn P, Mortrud M, Ouellette B, Kidney J, et al. (2014). Anatomical characterization of Cre driver mice for neural circuit mapping and manipulation. Front Neural Circuits 8, 76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Madisen L, Mao T, Koch H, Zhuo JM, Berenyi A, Fujisawa S, Hsu YW, Garcia AJ 3rd, Gu X, Zanella S, et al. (2012). A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing. Nat Neurosci 15, 793–802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Angelo K, Rancz EA, Pimentel D, Hundahl C, Hannibal J, Fleischmann A, Pichler B, and Margrie TW (2012). A biophysical signature of network affiliation and sensory processing in mitral cells. Nature 488, 375–378. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Available from the Lead Contact upon reasonable request.

RESOURCES