The Journal of Neuroscience, December 9, 2020 • 40(50):9701–9714 • 9701

Systems/Circuits Activation of Granule Cell Interneurons by Two Divergent Local Circuit Pathways in the Rat Olfactory Bulb

R. Todd Pressler and Ben W. Strowbridge Department of Neurosciences, Case Western Reserve University, Cleveland, Ohio 44106

The olfactory bulb (OB) serves as a relay region for sensory information transduced by receptor neurons in the nose and ulti- mately routed to a variety of cortical areas. Despite the highly structured organization of the sensory inputs to the OB, even simple monomolecular odors activate large regions of the OB comprising many glomerular modules defined by afferents from different receptor neuron subtypes. OB principal cells receive their primary excitatory input from only one glomerular channel defined by inputs from one class of olfactory receptor neurons. By contrast, interneurons, such as GABAergic granule cells (GCs), integrate across multiple channels through dendodendritic inputs on their distal apical dendrites. Through their inhibitory synaptic actions, GCs appear to modulate principal cell firing to enhance olfactory discrimination, although how GCs contribute to olfactory function is not well understood. In this study, we identify a second synaptic pathway by which principal cells in the rat (both sexes) OB excite GCs by evoking potent nondepressing EPSPs (termed large-amplitude, non- dendrodendritic [LANDD] EPSPs). LANDD EPSPs show little depression in response to tetanic stimulation and, therefore, can be distinguished other EPSPs that target GCs. LANDD EPSPs can be evoked by both focal stimulation near GC proximal dendrites and by activating sensory inputs in the glomerular layer in truncated GCs lacking dendrodendritic inputs. Using computational simulations, we show that LANDD EPSPs more reliably encode the duration of principal cell discharges than DD EPSPs, enabling GCs to compare contrasting versions of odor-driven activity patterns. Key words: brain slice; interneurons; olfactory bulb; patch clamp; short-term plasticity

Significance Statement The olfactory bulb plays a critical role in transforming broad sensory input patterns into odor-selective population responses. How this occurs is not well understood, but the local bulbar interneurons appear to be centrally involved in the process. Granule cells, the most common interneuron in the olfactory bulb, are known to broadly integrate sensory input through spe- cialized synapses on their distal dendrites. Here we describe a second class of local excitatory inputs to granule cells that are more powerful than distal inputs and fail to depress with repeated stimulation. This second, proximal pathway allows bulbar interneurons to assay divergent versions of the same sensory input pattern.

Introduction sensory neuron input to downstream cortical regions are mitral The classic view of the olfactory bulb (OB) that has evolved from cells (MCs) and tufted cells (TCs), the only known glutamatergic anatomic and functional studies is of an array of semi-independ- cell types in the OB aside for a small population of interneurons ent glomerular modules, each driven by a distinct subgroup of in the glomerular input layer (Fremeau et al., 2001; Herzog et al., olfactory sensory neurons. By interpreting the patterns of acti- 2001; Lein et al., 2007; Brill et al., 2009; Ohmomo et al., 2009; vated glomerular modules, animals can determine which odor is Tatti et al., 2014). The OB also contains a large abundance of in- present and initiate the complex processing steps that underlie hibitory interneurons, including small axonless granule cells olfactory learning. The principal neurons for conveying olfactory (GCs). These GABAergic interneurons vastly outnumber princi- pal neurons and are excited along their distal apical dendrites by dendrodendritic (DD) synapses formed in the external plexiform Received Apr. 26, 2020; revised Oct. 17, 2020; accepted Oct. 21, 2020. layer (EPL) (Rall et al., 1966; Isaacson and Strowbridge, 1998; Author contributions: B.W.S. and R.T.P. designed research; B.W.S. and R.T.P. analyzed data; B.W.S. and R.T.P. edited the paper; B.W.S. and R.T.P. wrote the paper; R.T.P. performed research. Shepherd et al., 2007; Pressler and Strowbridge, 2017). The distal This work was supported by National Institutes of Health Grant R01-DC04285. We thank Dr. Dan Wesson apical DD pathway provides a mechanism for individual GC to for helpful discussions related to this project; and Drs. Chris Ford and Joel Zylberberg for providing helpful sample a large number of glomerular input modules since these comments on this manuscript. connections are formed along the expansive lateral dendrites of The authors declare no competing financial interests. Correspondence should be addressed to Ben W. Strowbridge at [email protected]. both MCs and TCs. Multiple lines of evidence suggest that inhi- https://doi.org/10.1523/JNEUROSCI.0989-20.2020 bition mediated by GCs through these reciprocal DD synaptic Copyright © 2020 the authors local circuits plays a critical role when animals are making fine 9702 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons distinctions between similar odors or odor mixtures (Abraham and perfused with oxygenated ACSF at a rate of 1.5 ml/min. Recordings et al., 2010; Nunes and Kuner, 2015). Exactly how GC inhibition were made between 29°C and 31°C. ACSF consisted of the following (in onto principal cells facilitates odor discrimination is not known, mM): 124 NaCl, 3 KCl, 1.23 NaH2PO4,1.2MgSO4,26NaHCO3,10dex- but the mechanism likely involves modulation of principal cell trose, 2.5 CaCl2, equilibrated with 95% O2/5% CO2. Except where noted, firing patterns to enhance subtle differences evident at the glo- drugs were added to the submerged recording chamber by changing the external solution source. In some experiments, intracellular blockers merular input stage in responses to different stimuli (Fukunaga were included in the recording pipette loading solution. Whole-cell cur- et al., 2014). rent-clamp patch-clamp recordings were made using AxoPatch 1D Critical to understanding how GC activity facilitates distinct amplifiers (Molecular Devices) and borosilicate glass pipettes (3-7 MV) neural representations of sensory inputs is the question of how under IR-DIC video microscopy. Recording electrodes contained the fol- these interneurons are excited. Researchers have often studied lowing (in mM): 140 K-methylsulfate, 4 NaCl, 10 HEPES, 0.2 EGTA, 4 how neural circuits differentially respond to sensory stimuli MgATP, 0.3 Na3GTP3, 10 phosphocreatine di-Tris. Alexa-594 or Alexa- through plasticity in one class of orthodromic synaptic inputs 488 (100-150 mM; Invitrogen) was also included routinely to permit visu- (e.g., Fiete et al., 2010). Neurons in many brain regions, however, alization of the recorded neuron using live 2-photon imaging, as often compare activity in two or more excitatory pathways (e.g., described previously (Pressler and Strowbridge, 2006, 2017). Cell- parallel and climbing fiber inputs to Purkinje cerebellar neu- attached recordings used the same internal solution to enable subsequent breakthrough to whole-cell recording model. All internal solutions were rons), providing an alternate model for understanding the orga- adjusted to pH 7.3 and 290 mOsm. nization of circuits within the olfactory system. Simultaneously Intracellular recordings were low-pass filtered at 5 kHz (FLA-01, comparing several distinct representations of a sensory input Cygus Technology) and acquired at 10 kHz using an ITC-18 interface arriving via different local circuit pathways would enable GCs to (Instrutech/HEKA) and custom Visual Basic.NET and Python software reliably develop specific associations between odorants and the on a Windows 7-based personal computer. Intracellular recordings were diffuse patterns of activity across glomerular modules. The pres- not corrected for the liquid junction potential. All cell type classifications ence of a second excitatory synaptic input that complements were confirmed by intracellular dye filling followed by fluorescence visu- the widely sampling DD input could explain the paradoxical abil- alization. Unless noted, all drugs were obtained from Tocris Bioscience, ity of broadly sampling GCs to generate sparse and selective except for TTX, which was purchased from Calbiochem. In this study, we stimulated afferent inputs to GCs using three differ- responses to individual odors (Kato et al., 2012; Cazakoff et al., ent methods: focal stimulation near a visualized dendritic section, extrac- 2014). ellular stimulation within a specific layer (either the olfactory sensory Previous studies (Price and Powell, 1970a; Kishi et al., 1984; nerve [OSN] input layer or the GC layer [GCL]), or stimulation within Mori et al., 1999; Nagayama et al., 2004) have speculated that one glomeruli. Visualized focal stimulation (e.g., see Figs. 2D, 3B–E) local axon collaterals of MCs and/or TCs might synapse on GCs, (Balu et al., 2007; Gao and Strowbridge, 2009) was performed using live providing a potential second local excitatory connection to these 2-photon imaging to position a second patch pipette filled with 124 mM interneurons. By using a novel approach of eliminating the abun- NaCl, 3 mM KCl, and 10 mM HEPES, pH 7.4, and 100 mM Alexa-594 dant DD inputs from individual GCs that could obscure other within 5-10mm of an identified dendritic segment. We used sharpened types of connections (Pressler and Strowbridge, 2019), we monopolar tungsten electrodes (FHC) for layer- and glomerulus-specific stimulation. In MC recording experiments involving glomerular-specific have revealed this new local excitatory pathway to GCs that stimulation (e.g., see Fig. 5A,B), the glomerulus containing the MC api- selectively targets proximal dendrites and likely arises from cal dendritic tuft is indicated as “glom0” (confirmed in all experiments axons of OB principal cells. The large-amplitude, nondepress- by filling MCs with Alexa-594 dye via breaking through to whole-cell ing EPSPs evoked through this pathway contrast with both recordings). – depressing DD and facilitating inputs to the same interneur- In Figure 6H K, we used bath application of the GABAA receptor ons (Balu et al., 2007; Gao and Strowbridge, 2009; Pressler and agonist muscimol (1-2 mM) to test whether evoked synaptic responses Strowbridge, 2017). Both DD and axo-dendritic circuits appear were mediated by monosynaptic or polysynaptic pathways, paralleling to be at least partially maintained within in vitro brain slices, ena- previous work by others (Najac et al., 2011; Fukunaga et al., 2012; bling sensory inputs to excite GCs through both pathways. Budzillo et al., 2017). We found similar effects of 1 and 2 mM muscimol on OSN-evoked responses (conversion of prolonged EPSP barrages to Because DD and axo-dendritic inputs to GCs have contrasting single EPSPs in most responses). A third truncated GCs was only tested forms of short-term plasticity, proximal axo-dendritic inputs using 1 mM muscimol and also a similar conversion of an EPSP barrage more reliably reflect the duration of discharges in presynaptic to single evoked EPSPs in most episodes. principal cells. Such differential processing may facilitate detect- Data analysis and statistics. We selected for GCs by recording from ing aspects of the sensory input that modulate discharge dura- neurons with small, elongated (“egg-shaped”) cell bodies in the GCL tion, such as odor concentration (Cang and Isaacson, 2003; imaged using IR-DIC video microscopy and a 60Â water immersion Wienisch and Murthy, 2016). objective and confirmed GC-like morphology in all recordings using live 2-photon imaging. Some experiments involved recording from GCs with apical dendrites severed in the GCL (“truncated GCs”), as described Materials and Methods in a recent publication from our group (Pressler and Strowbridge, 2019). Animals. Horizontal OB slices (300mmthick)werepreparedfrom Truncated GCs appeared to be healthy neurons with high input resistan- postnatal day (P) 14-25 Sprague Dawley rats (both sexes) anesthetized ces (1450 MV vs 960 MV for intact GCs, as reported previously in with ketamine, as described previously (Pressler and Strowbridge, 2006, Pressler and Strowbridge, 2019) and long membrane time constants (49 2019) using a Leica Microsystems VT1200 vibrotome. Parasagittal ante- vs 43 ms in intact GCs). Truncated GCs were not specifically targeted for rior piriform cortex (APC) slices, 350mm thick, were prepared from rats recording; GC subtype (intact or truncated) was determined only after in the same age range anesthetized with isoflurane, as described by intracellular filling and visualization. Suzuki and Bekkers (2011). All experiments were conducted in accord- Summary results are presented as mean 6 SEM, except for the com- ance with the guidelines approved by the Case Western Reserve putation simulation (mean 6 SD). Statistical comparisons used Student’s University Animal Care and Use Committee. t test unless otherwise indicated and were performed in Python (version Electrophysiology. Both OB and APC slices were incubated initially 3.6) using the SciPy.Stats library. Cumulative distributions were compared at 30°C for 30 min and then maintained at room temperature until use. using the Kolmogorov–Smirnov test. The assay of two-dimensional clus- During recording sessions, slices were placed in a submersion chamber tering presented in Figure 3F used the Hopkins statistic (Banerjee and Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9703

A Average evoked DD EPSP B Spon. non-DD EPSPs GC

Evoked DD EPSP 1 mV 5 ms GC 1 mV (failure) 5 ms -70

MC (pre) GC 20 mV (post) MC -70

1 nA -71 Average evoked DD EPSP

C Fastest DD EPSPFastest Spon. EPSPD Largest DD EPSP Largest Spon. EPSP 759 1620 mV/s 5.1 8.9 mV

1 1

0.8 0.8

0.6 0.6

N = 85 Evoked DD EPSPs action N = 85 Evoked DD EPSPs N = 2695 Spon. EPSPs N = 2695 Spon. EPSPs 0.4 0.4

Cum. Fr Cum. Fraction 0.2 0.2

0 0 (6 MC/GC pairs with DD EPSPs)

0 500 1000 1500 2000 024 6810 EPSP Init. Slope (5-50%; mV/s) EPSP Amp. (mV)

Figure 1. Olfactory GCs receive large-amplitude spontaneous EPSPs. A, Diagram of experimental configuration for DD paired recording experiment with one intracellular recording in an MC and another in a postsynaptic GC. Thin red traces indicate repeated GC responses to evoked MC APs (bottom black trace). One trial failed to trigger a DD EPSP (labeled failure). Top thick red trace indicates average GC response. B, Example spontaneous EPSPs from the same postsynaptic GC shown in A with more rapid kinetics (all have .1000 mV/s rising phase slopes, 5%-50%) and larger amplitude than the average unitary-evoked DD EPSP shown at bottom. Other example spontaneous EPSPs recorded in the same GC (middle set of traces) resembled the evoked DD EPSP (bottom red trace). C, Plot of the cumulative distribution of spontaneous (black curve) and MC spike-evoked (red curve) EPSP rising phase slopes (5%-50%) in the 6 GCs with monosynap- tic DD connections with MCs. Arrows indicate the largest evoked DD EPSP and the largest spontaneous EPSP recorded in the same set of neurons. D, Cumulative plots of EPSP amplitudes from the same two datasets shown in C.

Dave (2004) with values .0.75 indicating significant clustering. Jitter esti- continuously increased the number of presynaptic MC discharges until mates represent the SD of the onset latency to the first EPSP (or action the GC reached firing threshold in 95% of trials. potential [AP]). Computational model. Computation simulations of EPSP summa- tion in GCs used the same methodology described in a recent publica- Results tion from our group (Pressler and Strowbridge, 2017). In brief, we Variation in spontaneous EPSP amplitudes and kinetics initially created a set of presynaptic firing patterns by either randomly Abundant previous work suggests that olfactory GCs receive selecting patterns from a library of previously recorded MC (or TC) dis- excitatory synaptic input through DD synapses they form – charges evoked by glomerular stimulation (similar to Fig. 5A D)orby with MC and TC dendrites (Rall et al., 1966; Isaacson and constructing synthetic discharges that matched the statistical properties Strowbridge, 1998; Schoppa et al., 1998). Our group recently of recorded MC/TC discharges (the same average interspike interval and reported properties of these synapses assayed using paired variability [coefficient of variation] of the interspike interval). Each spike within the MC/TC discharge was then processed to simulate probabilis- intracellular recordings (Pressler and Strowbridge, 2017) tic EPSP transmission (for details, see Pressler and Strowbridge, 2017) (example DD EPSPs shown in Fig. 1A). In that dataset, we to generate a new postsynaptic depolarization waveform. Estimates of found multiple examples of spontaneous EPSPs that were GC depolarization assume linear EPSP summation from each succes- larger than any evoked DD EPSP recorded in the 6 paired sive MC/TC spike train. The resting membrane potential in simula- recordings. (This initial 2017 report on paired MC/GC record- tions was À70 mV and firing threshold was set at À42 mV based on the ings focused exclusively on evoked DD EPSPs in GCs, not on most depolarized stable membrane potential we recorded in GCs the properties of spontaneous EPSPs.) Figure 1B illustrates À 6 ( 42.0 2.5 mV from Pressler and Strowbridge, 2017). Simulations examples of large-amplitude spontaneous EPSPs along with included a spontaneous EPSP rate of 13 Hz based on the mean rate of spontaneous putative DD EPSPs and evoked DD EPSPs, all spontaneous EPSPs in our intact GCs recordings. Estimates of the number of active, odor-triggered presynaptic MC/TC discharges recorded in the same GC. These large-amplitude spontaneous required to trigger GC APs (see Fig. 8A) reflect conditions in which EPSPs also had very rapid rising phase kinetics and reached 95% of simulations using that parameter set depolarized to at least their maximal amplitude ;4 ms before evoked DD EPSPs. À42 mV and, therefore, would have triggered at least one AP. We recorded 2695 spontaneous EPSPs in the same 6 postsy- Simulations performed to assay GC AP latency (see Fig. 8B,C) also naptic GCs with monosynaptic DD EPSPs evoked via paired 9704 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons intracellular recordings. The largest amplitude evoked DD EPSP stimulation experiments that evoked large-amplitude EPSPs, in this dataset was 5.1 mV, .40% smaller than the largest spon- most recordings revealed near neutral paired-pulse ratios (PPRs) taneous EPSP recorded in the cells (8.9 mV; cumulative histo- (neither depressing nor facilitating), a result that, while uncom- grams presented in Fig. 1C). We found a similar increase in the mon in CNS excitatory synapses, has been reported in a subset of range of the EPSP initial slopes with the fastest evoked DD EPSP glutamatergic synapses in piriform cortex (Suzuki and Bekkers, representing ;50% of the slope of the fastest spontaneous EPSP 2011). In the example recordings in Figure 3A–E,weevokeddif- recorded (Fig. 1D). When analyzed individually within each ferent monosynaptic EPSPs at two different focal stimulation paired recording, the fastest spontaneous EPSPs had initial slopes positions in the same intact GC: small-amplitude EPSPs that were 2.04 6 0.25 times greater than the fastest MC-evoked DD facilitated with paired stimulation when stimulating the proximal EPSP (spontaneous:evoked ratio ranged from 1.4 to 2.9, mean apical dendrite (Fig. 3B,D) and large-amplitude EPSPs with ratio significantly different from 1, p , 0.005, T =4.103). neutral PPRs when stimulating near a basal dendrite (Fig. 3C, This initial result suggests that the evoked DD EPSPs we E). Over 17 microstimulation experiments, plots of EPSP uni- assayed in our six monosynaptic paired recordings represented tary amplitude versus PPR showed two clusters: Hopkins (H) only a subset of the range of excitatory synaptic inputs GCs statistic = 0.94, greater than the 0.75 threshold for significant receive. While varying degrees of electrotonic attenuation and fil- clustering (Banerjee and Dave, 2004)(Fig. 3F), corresponding tering may contribute to the deviations between evoked and to small-amplitude facilitating (SAF) EPSPs (green symbols) spontaneous distributions, electrotonic effects are unlikely to be and large-amplitude EPSPs with near-neutral PPRs (orange the explanation for the excess large-amplitude spontaneous symbols). When analyzed solely by short-term plasticity, small- EPSPs recorded in GCs. The experiments described below will amplitude EPSPs had significantly larger PPRs than large-am- test the hypothesis that many of these large-amplitude EPSPs plitude EPSPs (2.24 6 0.21 vs 1.21 6 0.1 for large-amplitude represent a new type of synaptic response reflecting glutamater- EPSPs evoked at 20 Hz; p , 0.0002, T = 4.69). We found similar, gic contacts on the proximal apical and basal dendrites of GCs. near neutral PPRs when assaying large-amplitude EPSPs Our results are divided into three sections, beginning with test- evoked at 40 Hz (0.91 6 0.05; N = 7). ing whether large-amplitude EPSPs are distinct from classic Because they appear to reflect a distinct input class based on DD EPSPs by attempting to record them in GCs lacking distal both kinetics and short-term plasticity, we refer to the large-am- apical dendrites (“truncated GCs”). We then ask whether large- plitude, fast-rising EPSPs (initial slopes . 800 mV/s) synaptic amplitude EPSPs are distinct from previously demonstrated potentials as “large-amplitude nondendrodendritic” (LANDD) (Balu et al., 2007; Gao and Strowbridge, 2009; Sun et al., 2020) EPSPs. We used a primary classification metric based on EPSP facilitating inputs that also target proximal GC dendrites. And rising-phase slope, rather than amplitude, because of the diffi- finally, we ask whether local circuits evoke large-amplitude culty in assessing the baseline membrane potential during trains EPSPs. of summating EPSPs. Prior work suggests that small-amplitude facilitating EPSPs, referred to as “SAF EPSPs” in this study, are Large-amplitude EPSPs arise from proximal GC synapses likely to be heterogeneous, representing both cortical feedback If large-amplitude EPSPs arise from excitatory synapses on prox- (de Olmos et al., 1978; Uva et al., 2006; Balu et al., 2007; Gao and imal dendrites, then they should still be present in “truncated” Strowbridge, 2009) and a subset of intrinsic OB inputs (Sun et GCs that have no dendritic arbors in the EPL and, therefore, no al., 2020). Over our series of focal microstimulation experiments, DD synaptic inputs (Fig. 2A,B)(Price and Powell, 1970b; Balu et there was a trend for LANDD EPSPs to be evoked by stimulation al., 2007; Pressler and Strowbridge, 2017). We recorded from 27 near basal dendrites (in 5 of 7 of experiments) and a similar truncated GCs that lacked EPL dendrites and found spontaneous trend for SAF EPSPs to be evoked by stimulation near proximal large-amplitude EPSPs in 82% of those recordings. As shown in apical dendrites in the GCL (in 3 of 4 experiments). the example recordings in Figure 2C, spontaneously occurring The results presented thus far suggest that GCs receive three large-amplitude EPSPs in truncated GCs had similar large ampli- types of excitatory synaptic inputs: (1) weak and depressing DD tudes (.2 mV) and fast rising phase kinetics as spontaneous EPSPs on distal apical processes, (2) weak and facilitating EPSPs large-amplitude EPSPs in intact GCs (Fig. 1B). Focal stimuli near on proximal dendrites, and (3) large-amplitude LANDD EPSPs, truncated GCs elicited large-amplitude EPSPs with short, pre- also likely arising from excitatory synapses on proximal GC den- sumably monosynaptic latencies (1.8 ms in the example in Fig. drites. Initial rising phase slope in LANDD EPSPs differed signif- 2D) at threshold stimulus intensities that triggered both suc- icantly from both SAF and DD EPSPs (Fig. 3G). All three GC cesses and failures. We also could evoke large-amplitude EPSPs EPSP responses appeared to be monosynaptic based on short using extracellular GCL stimulation while recording from trun- (,3 ms) onset latencies (Fig. 3H). cated GCs (N = 2, not shown). Although SAF EPSPs facilitate with repeated stimulation, we Over 9 focal stimulation experiments, we evoked both small- find greater temporal summation in LANDD EPSPs, as shown in amplitude EPSPs that facilitated with repeated stimulation and Figure 4A–C. In the example recording in Figure 4A that com- large-amplitude EPSPs in truncated GCs. The mean amplitude pares two stimulation sites on the same intact GC, the maximal of evoked EPSPs in truncated GCs was 1.9 mV, significantly depolarization reached following a four-shock stimulus train was larger than the mean amplitude of evoked DD EPSPs assayed in larger following LANDD stimulation despite the absence of MC/GC paired recordings (p , 0.01, T = 2.38, unpaired t test). paired-pulse facilitation in this synapse. As shown in Figure 4B, We also found a large distribution of EPSP initial slope kinetics the average amplitude of DD EPSPs continuously depresses dur- (Fig. 2D,bottom)with.54% of evoked EPSPs in truncated GCs ing similar 40 Hz stimulus trains and SAF EPSPs begin to depress having more rapid initial slopes than found in all DD EPSPs following the third stimulus. Figure 4C summarizes the results assayed in our paired MC/GC recordings (Fig. 2E). of a series of similar 4-shock experiments across all three types of We next asked whether large-amplitude EPSPs also had a excitatory synapses onto GCs and demonstrates the degree of unique form of short-term plasticity that would help define a large temporal summation inherent in LANDD inputs. The min- functional signature for these inputs. Over 7 focal micro- imal depression following prolonged stimulus trains likely Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9705

A Intact granule cell B

s 20 mV

e

s

p 200 ms

a

n

y

s

D

D

25 pA Truncated granule cell

C Spon. non-DD EPSPs

1 mV trunc GC 5 ms -70

D E 1 mV Evoked non-DD Evoked DD EPSPs EPSPs

intact GC

-71 1 mV trunc GC 5 ms 40 mV -71 MC basal dend. stim * * 1 nA

Evoked EPSPs in trunc GCs Evoked DD EPSPs (paired MC/GC rec) 20 25 GL EPL 20 15 MCL GCL

t 15

t

n

n

u u 10

o

o rec

C C trunc GC 10 rec MC (pre) rec (post) focal stim 5 (2p-guided) 5 intact GC

0 0 0 1000 2000 3000 4000 0 1000 2000 3000 4000 EPSP Init. Slope (mV/s) EPSP Init. Slope (mV/s)

Figure 2. Large-amplitude EPSPs evoked by focal microstimulation of proximal GC dendrites. A, Left, 2-photon maximal intensity z stacks demonstrating extensive dendritic arborization of a recorded intact GC within the EPL, the lamina where DD synapses are located. Image was obscured by debris on the surface of the slice located on the MCL/EPL boundary. Right, z stacks of a truncated GC with apical dendrite severed in the GCL. B, Responses to depolarizing and hyperpolarizing current steps in the truncated GC shown in A (right). C, Spontaneous non-DD EPSPs recorded in the truncated GC shown in A. These example spontaneous EPSPs all have .1000 mV/s initial slopes (5%-50%). D, Top, Focal microstimulation near the basal dendrite of the trun- cated GC shown in A-C evoked large-amplitude non-DD EPSPs with fast kinetics. Dashed line indicates the average initial rising phase slope of EPSPs (5%-50%; all . 1000 mV/s). Asterisks indi- cate stimulus timing (and in subsequent figures). Bottom, Histogram of EPSP initial slopes from 9 truncated GC recordings with focal dendritic stimulation. Typical recording configuration shown in inset. E, DD EPSPs evoked in monosynaptic MC/GC recordings (top). Dashed line indicates the average EPSP slope. Bottom, Histogram of MC AP-evoked DD EPSP initial slopes paired MC/GC recordings. Diagram of MC/GC paired recording configuration shown in inset. explains the ability of summating LANDD EPSPs to generate LANDD EPSPs appeared to represent conventional iono- large postsynaptic depolarizations. Consistent with the large tropic glutamate receptor-mediated synaptic responses. Evoked degree of temporal summation we find with LANDD EPSPs, LANDD EPSPs were blocked by DNQX, an AMPAR antagonist trains of these EPSPs could reliably trigger GC spiking as shown (Fig. 4E; N =3; 10 mM). LANDD EPSPs could still be evoked 1 in the example recording in Figure 4D. when fast voltage-gated Na channels were blocked by perfusion 9706 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons

ABD Non-DD EPSPs (SAF)

avg of 11 GC -70

EPL rec 25 µm MCL * **apical dend. stim GCL 2 mV 50 µm C E Non-DD EPSPs (LANDD) 20 ms B rec GC avg of 14 rec 25 µm -70

C * * basal dend. stim (same GC)

F G H H SAF = 0.94 *** 3 2000 *** 3 **

)

s

) n.s.

o

/

i

s

t V 1660

a 1500 m 2.4

(

m

R

(

2 y 2.1 e 2

e

c

s

p

l

n

o

u e 1.7

l

1000 t

P

S

a

-

.

L

d

t

i

e 1

P

n

r

i

I

S a 1 500

P

287 P

P 500

S

E

P (6) LANDD E (9) (8) (9) (8) (6) 0 0 LANDD SAF DD LANDD SAF DD 0 Paired 0 1 2 3 4 Focal rec EPSP Unitary Amp. (mV) Microstim (6)

Figure 3. Two types of excitatory responses onto proximal GC dendrites. A,2-photonz-stack montage showing intact GC recorded in experiments in B–E. Same GC following placement of a dye-filled focal stimulating electrode near either the proximal apical dendrite (B) or a basal dendrite (C). D, Evoked monosynaptic EPSPs following trains of stimuli applied to the proximal apical dendritic stimulating electrode (trains of 3 shocks at 20 Hz). Individual responses superimposed at bottom; average response shown above (green trace). E, Responses in the GC to focal stimula- tion of a basal dendrite in the same GC. Individual responses superimposed at bottom; average response shown above (orange trace). Trains of two shocks at 20 Hz. These example spontaneous EPSPs all have . 1000 mV/s rising phase slopes (5%-50%). F, Summary plot of responses to GCL stimulation in visualized GCs. Filled green symbols represent SAF EPSPs. Filled orange symbols represent LANDD EPSPs. Diamond symbols represent 2-photon-guided microstimulation near visualized dendritic segments. Circles represent stimuli within the GCL but without visualization of the stimulation electrode in both groups. (Nonvisualized stimulation was only tested in truncated GCs to exclude activation of DD inputs.) Clusters were statistically different based on the Hopkins statistic . 0.75. Mean 6 SEM for PPR estimates indicated inside the y axis and unitary amplitude inside the x axis. G, Plot of EPSP initial slope (5%-50% of rising phase) for three dif- ferent classes of GC EPSPs. For LANDD versus SAF EPSPs, pppp , 1.2 Â 10À4, T = 5.03. For LANDD versus DD EPSPs, pppp , 1.3 Â 10À4, T = 4.75; both unpaired t tests. H, Plot of onset la- tency for the same three EPSP classes. ppp , 0.01, T =2.65(unpairedt test). n.s.; p . 0.05.

with QX-314 (N =3; 5 mM). We also observed spontaneous mediated by local (bulbar) excitatory neurons, presumably MCs LANDD EPSPs in following both intracellular QX-314 perfusion and/or TCs. If LANDD EPSPs arise from presynaptic MCs or 1 and when Na channels were blocked extracellularly by bath TCs, then glomerular layer stimulation should evoke polysynap- application of TTX (see Fig. 7). In addition, we could evoke tic LANDD EPSPs (by activating distal MC/TC dendrites directly LANDD EPSPs in GCs filled with 10 mM BAPTA, which disrupts and/or stimulating OSN excitatory inputs to MCs and TCs). To 1 Ca2 -mediated intracellular signaling (N = 3). Together, these test this possibility, we first verified that focal stimulation in the results indicated that, despite their large amplitude and the pro- glomerular layer could bring MCs to threshold. As shown in pensity of GCs to generate intrinsic spikes (Egger et al., 2005; Figure 5A, graded single-shock stimulation applied to one glo- Bywalez et al., 2015), LANDD EPSPs resemble AMPAR-based merulus recruited repetitive firing assayed with a cell-attached classic cortical and hippocampal synaptic potentials that do not MC recording. Once the threshold stimulus level was reached, 1 1 appear to be dependent on voltage-gated Na channels or Ca2 - doubling the stimulus intensity did not recruit faster firing or activated conductances. prolong MC discharges. MC responses were highly selective to which glomeruli was stimulated. As shown in Figure 5B,ofthree Polysynaptic networks mediate LANDD EPSPs neighboring glomeruli tested with suprathreshold stimuli (3Â We next conducted a series of experiments to test whether threshold intensity), only one stimulus position elicited MC fir- LANDD EPSPs arise from centrifugal axons or whether they are ing (the same glomerulus we found contained the apical Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9707

A B 40 Hz stimuli

2 GC -72 apical dend. stim (SAF) **** SAF 2 mV 1 GC LANDD rec 20 ms DD GC Paired-pulse Ratio 0 1 2 3 4 -72 basal dend. stim (LANDD) **** Stim. Num.

C DE ** Control 8

) ***

V

m

( 6.8 6 n 20 mV

o

i

t a 20 ms -70 2 mV m 4

m

u 10 ms S 3.1 + DNQX

P 2

S 1.7

P

E (6) (6) (6) -70 0 basal dend. stim intact GC LANDD SAF DD basal dend. stim (intact GC) **

Figure 4. Contrasting forms of short-term plasticity in SAF and LANDD EPSPs. A, Comparison of short-term plasticity in responses to trains of 40 Hz stimuli applied near the proximal apical (top green trace) or basal (bottom orange trace) of the same intact GC. B, Summary plot of short-term plasticity assayed using trains of 40 Hz stimuli that triggered SAF EPSPs (green symbols; N = 6 experiments), DD (blue; N = 6), or LANDD (orange; N =8)symbols. C, Plot of EPSP summation following the final stimulus in the 4 Â 40 Hz stimulation trains illustrated in A. pppp , 1.03 Â 10À4, T =5.67;ppp , 3.3 Â 10À4, T = 4.86, both unpaired t tests. D, Example recordings in which trains of LANDD EPSPs triggered APs with low jitter (0.32 ms; latency SD). E, Blockade of AMPARs with 10 mM DNQX abolishes LANDD EPSPs recorded at À70 mV evoked by focal stimuli applied near the basal dendrite. dendritic tuft on subsequent dye filling). In both MC and TC The onset latencies and barrage patterns of polysynaptic cell-attached recordings, we observed a variety of responses to LANDD EPSPs evoked by glomerular stimulation closely focal stimuli in the glomerular layer stimulation with some cells resembled the discharge patterns we demonstrated in cell- responding to the stimuli with a single short-latency spike (Fig. attached MC and TC recordings, suggesting that LANDD 5C). Other principal cells responded to similar glomerular stim- EPSPs likely arise from axon collaterals of a subset of OB uli with a delayed discharge (Fig. 5D). principal cells. [Both MCs and TCs are polarized neurons We evoked polysynaptic LANDD EPSPs in 8 truncated GCs with separate laminar zones for their dendritic and axonal in a separate set of recordings using the same type of focal glo- arborization (Price and Powell, 1970a) with only the axons of merular stimulation used to assay MC and TC responses. We MCs and TCs entering the GCL.] We tested this hypothesis fur- observed LANDD EPSP patterns that matched all three MC/TC ther by quantifying the latency/jitter relationship for MC/TC discharge patterns we observed, including short-latency dis- discharges (Fig. 5H) and polysynaptic LANDD EPSPs (Fig. 5I, charges (Fig. 5E), short-latency single EPSP responses (Fig. 5F), blue symbols), both evoked by similar focal glomerular stimula- and long-latency discharges (Fig. 5G). In all cases, the latency to tion protocols. Consistent with previous studies using similar elicit LANDD EPSPs was .3 ms, suggesting that the focal glo- in vitro recording conditions (De Saint Jan et al., 2009; Gire and merular layer stimulation activated a local bulbar excitatory neu- Schoppa, 2009) and in vivo discharges evoked by odors ron that was presynaptic to the GC, generating a polysynaptic (Kashiwadani et al., 1999; Shusterman et al., 2011; Fukunaga et EPSP in the truncated GC. al., 2012), we find a wide range of onset latencies for both Since all LANDD EPSP recordings in this set of experiments MC and TC discharges. We also found a positive correlation were from truncated GCs that lacked distal apical dendrites in between mean onset latency and variability of train latency, dis- the EPL, these excitatory responses could not arise from DD syn- charges that began relatively long after the stimulus tended to apses. Projections from nearby olfactory cortical regions also have highly variable start times. As shown in Figure 5I, there were unlikely to generate polysynaptic LANDD EPSPs since was a similar range of onset latencies and latency variation in most OB slices had no cortical targets that projected back to the LANDD EPSPs recorded in truncated GCs providing additional OB (e.g., anterior olfactory nucleus [AON], APC, or posterior evidence that LANDD EPSPs likely arise from axon collaterals piriform cortex). While no slices in this study included piriform of MCs and/or TCs. All the EPSP barrages evoked glomerular cortex, which requires a special slicing configuration to obtain stimulation in truncated GCs appeared to be polysynaptic based OB-APC connections (Balu et al., 2007; Apicella et al., 2010), a on their onset latencies (all .3 ms). By contrast, monosynaptic minority of OB slices included a section of the AON. In two LANDD EPSPs evoked by focal stimulation near-proximal den- experiments, we dissected the AON region away from these slices dritic segments had relatively low latency jitter (Fig. 5I, orange and were still able to record polysynaptic LANDD EPSPs in symbols, all with latencies ,2.6 ms). LANDD EPSPs evoked truncated GCs (e.g., Fig. 5E). monosynaptically by focal stimulation and polysynaptically via 9708 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons

0.5 V A glom stim B C -1 0 +1 glom -1 1 V MC MC glom 0 GL * EPL D MCL 2 V glom +1 GCL 20 ms Rec * 20 ms 3* V (MC) *

E glom stim LANDD EPSPs (4 - 37 ms latency) F G LANDD EPSPs (48 - 104 ms latency) LANDD EPSPs (10 - 11 ms latency) GL

EPL 4 mV MCL 40 ms

GCL 2 mV rec trunc GC trunc GC 1 mV 20 ms trunc GC 10 ms * glom stim (no AON) * glom stim glom stim H I J MC/TC APs (glom layer stim) Evoked LANDD EPSPs Evoked LANDD EPSPs 100 100 3000 mono- synaptic

)

s

/

)

)

V D 10

D 10

m

S

S

(

n.s.

y

y 2000

e

c

c

p

n

n

o

l

e

e

t

t 1 1

S

a

a

l . 1530

l

(

t

i

( 1440

r

n

r

I

e

t

e

t

t

i

P 1000

t

i 0.1 J 0.1 Glom layer stim S J Mitral cells

P Tufted cells Focal GCL stim E

0.01 0.01 (7) (8) 1 10 100 1 10 100 0 GCL stim Glom stim First AP Latency (ms) First EPSP Latency (ms)

Figure 5. Sensory input evokes LANDD EPSPs. A, Left, Diagram of recording configuration with cell-attached MC recording and focal microstimulation of three nearby glomeruli. Subsequent whole-cell recording and 2-photon imaging was used to determine the “on-beam” glomerulus for that MC. Right, Example MC responses to on-beam (“glom 0”) stimulation recorded at three intensities. B, Only the stimulation of the on-beam glomerulus triggered MC discharges. Same recording as in A; stimulus intensity 3 times threshold level determined in A in glom0 and two adjacent glomeruli (glom-1 and glom1). C, Example single short-latency (7.7-9.0 ms) AP response to glomerular stimulation in a different MC. D, Example long-latency (53-71 ms) discharge fol- lowing glomerular layer stimulation in a different MC. E, Long-latency LANDD EPSPs evoked by glomerular stimulation in a truncated GC. Diagram of recording configuration shown on left. Experiment in an OB slice with the AON removed. F, Examples of short-latency single LANDD EPSPs evoked by glomerular stimulation in a different truncated GC. G, Long-latency clusters of LANDD EPSPs evoked by glomerular stimulation in a different truncated GC. H, Plot of onset latency versus onset latency jitter in cell-attached recordings from MCs (filled square symbols) and TCs (filled circles) following on-beam glomerular stimulation. I, Analogous plot of latency versus jitter in EPSP responses to glomerular layer (blue symbols) or GCL (orange symbols) stimulation in truncated GCs. All GCL stimuli experiments generated presumptive monosynaptic EPSPs with onset latencies ,3ms.J, Plot of initial slopes of LANDD EPSPs evoked by GCL (orange) or glo- merular layer (blue) stimulation. Mean initial slope values are not statistically different (n.s.; p . 0.05). glomerular layer stimulation had similar properties, including possibility that OSN stimulation activated long-range OB/corti- nearly identically average initial slopes (Fig. 5J). cal pathways mediated the responses. The individual EPSPs While we demonstrated that glomerular layer stimulation within the synaptic barraged evoked by OSN stimulation in trun- triggered long-latency LANDD EPSPs, the glomerular layer con- cated GCs resembled LANDD EPSPs with large amplitudes and tains a variety of external afferents fibers (Shipley and Adamek, rapid initial slopes. 1984; McLean and Shipley, 1987, 1991; Boyd et al., 2012), which All OSN-evoked responses recorded in truncated GCs might also contribute to evoked EPSPs we recorded in truncated appeared to be polysynaptic based on onset latencies .4 ms for GCs. To address this issue, we asked whether barrages of the initial EPSP evoked in the barrage. The similar pattern of LANDD EPSPs could also be evoked by stimulation OSN axons. delayed EPSPs evoked by OSN (Fig. 6A) and direct glomerular In 4 of 4 truncated GC recordings, OSN stimulation triggered stimulation (Fig. 5E–G) could be explained by axo-dendritic barrages of EPSPs. Since these GCs lacked distal dendrites in the MC/TC synapses onto GCs since both stimulation paradigms EPL, these EPSPs could not arise from the more common DD robustly activate OB principal cells. pathway. The AON was dissected off the OB slice in 3 of 4 We next performed additional experiments to determine experiments recording OSN responses in truncated GCs, includ- whether OSN-evoked EPSP barrages could reflect inputs from ing the example response shown in Figure 6A, eliminating the centrifugal glutamatergic axons. Centrifugal inputs in our OB Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9709

B Monosynaptic circuit

A OSN stim

SL (APC) GL 2 mV ****LOT stim EPL 100 ms Polysynaptic circuit MCL C GCL rec trunc GC

trunc GC OSN stim (no AON) * intact GC ****OSN stim (no AON)

DEMonosyn. F G 150 ** 200 * 1 130 160 SL (APC) 120 110 80 0.8 1 mV

n ) 90 40

)

s

o

i trunc GC intact GC

s *** t

m 70 10 ***

5 ms c

( 0.6

m

SL a

(

r

y

(APC) 50 r

c Polysyn. 8 F

e

n

t

.

t

i e 30 0.4

t LOT m

J

a stim 6 u

L 16

* C 12 4 0.2 1 mV 8 5 ms (9) intact 2 (9) 4 0 GC 0 0 020406080100 OSN SL (APC) Intact GC Trunc GC SL (APC) Intact GC Trunc GC stim * Monosyn. Polysyn. Monosyn. Polysyn. EPSP Latency (ms)

HI J K Control (APC SL) Control (TC) Control (intact GC) Control (trunc GC) 5 mV 5 mV 50 ms 50 ms

2 mV

5 ms *

Muscimol * * Muscimol * 4 mV Muscimol 4 mV Muscimol 20 ms 20 mV 20 ms 20 ms SL (APC) TC intact t = 6-7 ms trunc t = 37-39 ms t = 2.2 ms t = 1.8 ms GC * GC * **LOT stim OSN stim (no AON) * OSN stim (no AON) * OSN stim (no AON)

Figure 6. Sensory input stimulation excites GCs through a polysynaptic circuit. A, A single OSN shock evokes an EPSP barrage recorded in a truncated GC from an OB slice without the AON. Latency to initial EPSP was 20.8 ms. Schematic of experimental configuration shown on left. B, Responses to 40 Hz tetanic stimulation of the LOT in APC SLs (EPSP latency = 2.4 ms). C, Analogous 40 Hz OSN stimulation evokes long-lasting EPSP barrages in OB GCs. D, LOT stimulation triggered monosynaptic EPSP onto an APC SL (top, 2.3 ms latency), while OSN stimulation triggered a polysynaptic EPSP onto an intact OB GC (bottom, 5.8 ms latency). Dashed line indicates onset of SL EPSP. Blue arrow indicates onset of GC EPSP. E, Summary plot of initial EPSP la- tency in SL, intact, and truncated GCs. ppp , 0.02, T =2.41;pppp , 0.001, T =4.638(y axis broken to illustrate range of data points). F, Summary plot of EPSP latency jitter in the same three cell types. pp , 0.05, T =2.0;pppp , 0.01, T =2.74(alsowithy axis broken). G, Plot of the cumulative distribution of EPSP latencies found in the first 100 ms of time following stim- ulation, triggered by LOT stimulation in SLs (black curve) and OSN stimulation in intact GCs (blue curve) and OSN stimulation in truncated GCs (orange curve). The plots of intact GC and trun- cated GC EPSPs are significantly different from the plot of EPSPs onto SLs (N =9)versusintactGC(N =6):p , 1.9 Â 10À17;SL(N = 9) versus truncated GC (N =4):p , 3.7 Â 10À13;both Kolmogorov–Smirnov tests. H, Activation of GABAA receptors with muscimol does not affect the latency to the initial EPSP in an SL cell. I, Muscimol abolishes late spikes in OSN-evoked TC response while preserving the initial, short-latency (1.8 ms) spike. J, Muscimol abolishes late EPSPs evoked by OSN stimulation in an intact GC while preserving the initial EPSP (enlarged in inset). K, Muscimol also abolishes late EPSPs evoked by OSN stimulation in a truncated GC. Inset, Initial muscimol-resistant EPSP. slice preparation primarily represent severed axons as we dissected via severed glutamatergic axons, the LOT input to semilunar cells the only nearby cortical region (the AON) away in most experi- (SLs) in APC (Suzuki and Bekkers, 2006, 2011), and for polysy- ments and more distant sources of excitatory centrifugal axons, naptic excitation, the DD input to intact OB GCs (Isaacson and such as piriform cortex, were not present in our OB slices (Balu et Strowbridge, 1998; Pressler and Strowbridge, 2017). As shown in al., 2007). the examples presented in Figure 6B–D, Each LOT stimuli typi- We initially compared OSN-evoked responses in truncated cally triggered one monosynaptic EPSP in SLs with short latency GCs with a “ground truth” standard for monosynaptic excitation (,3 ms) and with low jitter. By contrast, OSN stimuli triggered 9710 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons

A B Control C 3 ***

MCL TTX 1 mV 2 10 ms Control

2 mV 1 TTX 200 ms 1.0

25 µm Spon. EPSP Freq. (Hz) 0.4 (13) (13) 0 trunc GC Control TTX

Figure 7. Spontaneous LANDD EPSPs require Na1 spiking. A, Image of filled truncated GC. B, Spontaneous EPSPs before (top, black traces) and after (bottom, orange traces) bath application of 1 mM TTX. Recordings from the truncated GCs shown in A. C, Summary plot of the effect of TTX on spontaneous EPSP rates in 13 truncated GCs. pppp , 0.01, T =2.86(pairedt test). Inset, Examples of spontaneous LANDD EPSPs before (black) and after (orange) TTX recorded in the same truncated GC. polysynaptic EPSPs in intact GCs with latencies .4ms and (at ;2ms;Fig. 6I, bottom). In one MC experiment, muscimol larger jitter than monosynaptic EPSPs. abolished all OSN-evoked APs. As expected based on these MC/ OSN-evoked responses in truncated GCs more closely resem- TC responses, muscimol abolished most of the EPSP barrage in ble polysynaptic EPSPs with long latencies and high jitter than 9 of 9 intact GCs tested while preserving the initial EPSP (Fig. monosynaptic EPSPs (Fig. 6E,F; and for similar latency and jitter 6J). Truncated GCs behaved similarly to intact GCs with musci- results for EPSPs evoked by glomerular stimulation, see Fig. mol abolishing nearly all of the OSN-evoked EPSP barrage while 5G,H). The average latency and jitter of OSN-evoked EPSPs in leaving only the initial EPSP (Fig. 6K; 3 of 3 experiments). In truncated GCs were significantly larger than monosynaptic both intact and truncated GCs, the remaining initial EPSP in LOT-evoked EPSPs in SLs (Fig. 6E,F, orange symbols). Both sin- muscimol occurred at latencies associated with polysynaptic, not gle and repetitive OSN stimulation evoked prolonged barrages of monosynaptic, excitation (.6 ms in all GC experiments; see Fig. EPSPs in intact and truncated GCs (Fig. 6A,C;barrageEPSP 6J,K, insets). The ability of prolonged shunting with muscimol to latencies plotted in Fig. 6G). In SLs, similar repetitive stimulation abolish most of the OSN response in truncated GCs provides fur- of severed LOT axons failed to trigger barrages of EPSPs (Fig. ther evidence that LANDD EPSPs are mediated by local polysy- 6G, black curve). The mean EPSP frequency in the 1 s window naptic excitatory circuits rather than by severed centrifugal following tetanic LOT stimulation of SLs (2.0 6 0.3 Hz, exclud- afferents. ing the initial monosynaptic evoked EPSP) was not elevated Our ability to drive LANDD EPSPs at long, presumably poly- above spontaneous EPSP rates (2.0 6 0.4 Hz; p . 0.05; N =9; synaptic latencies with glomerular layer and OSN layer stimula- paired t test) in the same neurons. By contrast, EPSP frequency tion suggests that they originate from local MCs and/or TCs. during the initial 1 s following OSN stimulation was significantly Since the OB principal cells often are spontaneously active even elevated above the spontaneous EPSP frequency (p , 0.005 for in in vitro brain slices (Heyward et al., 2001; Schmidt and intact GCs; p , 0.02 for truncated GCs). These findings provide Strowbridge, 2014), we asked whether we could record AP-de- further evidence suggesting that OSN stimulation evoked EPSPs pendent spontaneous LANDD EPSPs in truncated GCs. We 1 barrages in truncated GCs by activating OB excitatory local cir- tested the effect of blocking voltage-gated Na spikes with TTX cuits rather than inducing spontaneous spiking in severed cen- in 13 truncated GCs. On average, TTX significantly reduced the trifugal axons. frequency of all EPSPs by 60% (Fig. 7A–C). Nearly all truncated We conducted one additional test for polysynaptic versus mono- GCs (12 of 13) still received spontaneous LANDD EPSPs in synaptic excitation in truncated GCs using the GABAA receptor TTX, presumably reflecting the random, quantal release events agonist muscimol to shunt responses in potential intermediate exci- at LANDD synapses. The frequency of spontaneous LANDD tatory neurons. Muscimol shunting should preferentially impact EPSPs was significantly reduced by TTX (from 0.25 6 0.095 to polysynaptic responses by attenuating intracellular depolarization 0.061 6 0.014 Hz; p , 0.05; T = 1.93, paired t test), consistent associated with EPSPs and therefore lowering the transmission with the hypothesis that at least some LANDD EPSPs originated probability through multisynapse excitatory networks (Najac et al., from spontaneously active local presynaptic excitatory neurons, 2011; Fukunaga et al., 2012; Budzillo et al., 2017). neurons contained within OB slices. In control experiments recording monosynaptic SL EPSPs, bath application of muscimol failed to block EPSPs evoked by Functional differences between DD and LANDD synaptic LOT stimulation in 9 of 9 experiments despite reducing SL input inputs resistance by 53%. As expected from the increasing shunting, These experimental results suggest that GCs receive two func- muscimol depressed the amplitude of LOT-evoked EPSPs (by tionally distinct classes of excitatory drive from OB principal ;30%; Fig. 6H). In MCs and TCs, muscimol reduced input re- cells: depressing DD inputs onto their distal apical dendrites and sistance to a similar degree as SLs (;70%) and abolished most nondepressing inputs via axo-dendritic synapses on their proxi- OSN-driven spiking. In most MCs/TCs tested (4 of 5 experi- mal dendrites. We used a computational approach similar to that ments, including the example shown in Fig. 6I), muscimol elimi- used in Pressler and Strowbridge (2017) to address the signifi- nated all APs, except the initial AP evoked by OSN stimulation cance of these two converging excitatory pathways. We first Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9711

A B C 125 Sim. MC discharges (2 ms onset jitter) 125 Sim. MC discharges (20 ms onset jitter) 30

28.4 28.9 LANDD LANDD 100 100

20 DD 75 DD 75

50 50 10 10.7 10.5 Latency to GC spike (ms) 25 Latency to GC spike (ms) 25 Num. Presynaptic Cells Required for GC Spiking

0 0 0 DDLANDD DD LANDD 2 4 6 8 10 12 2 4 6 8 10 12 EPSP Class Presynaptic discharge length Presynaptic discharge length Mitral Cells Tufted Cells (num. MC APs) (num. MC APs)

Figure 8. Differential sensitivity to MC discharge duration in DD and LANDD pathways. A, Summary plot of the number of presynaptic MCs (or TCs assayed in separate simulations) required to bring a GC to firing threshold in 95% of model runs. Error bars indicate mean 6 SD. B, Plot of the length of the presynaptic MC discharge versus latency to the initial GC spike in simulations using DD-like (blue symbols) or LANDD-like (orange) short-term plasticity. Simulations used the same (DD-like) EPSP properties (e.g., amplitude, probability, response template). Results using simulated MC discharges based on mean interspike interval in recorded glomerular-evoked MC discharges (21.3 ms) and the measured coefficient of variation of the interspike interval (1.83). The number of active (discharging) presynaptic MCs required to trigger at least one GC AP in the DD-like STP simulations varied from 30 (3 spike discharges) to 22 (12 spike discharges). With LANDD-like STP, the number of active presynaptic MCs required to bring the GC to the AP threshold varied from 25 (3 spike discharges) to 12 (12 spike discharges). C, Similar results were obtained when increasing the onset latency jitter from 2 to 20 ms. repeated the computational assay our group used to estimate the GC spike latency was relatively insensitive to later MC spikes. In number of presynaptic odor-activated MCs and/or TCs required vivo, MC discharge duration triggered by the initial sniff is highly to bring a GC to spiking threshold. In these simulations, we used variable and often extends beyond 150 ms, including 10 spikes a large library of cell-attached recordings of MCs and TCs (e.g., Fukunaga et al., 2012, their Fig. 5; see also Bolding and responses to glomerular layer stimulation (similar to responses Franks, 2017, 2018; Wilson et al., 2017). Excitatory drive with shown in Fig. 5A–D; N = 120 TC responses and 84 MC LANDD-like short-term plasticity, by contrast, preserves more responses). We simulated EPSP generation and temporal sum- information related to the duration of presynaptic MC dis- mation triggered by each AP in the MC/TC discharge using syn- charges, and presumably related sensory information, such as aptic release probability, unitary amplitude, and short-term odor concentration. The divergence in the effect of variable MC plasticity parameters estimated in this report and earlier, related discharge duration was evident both when relatively little jitter studies (Pressler and Strowbridge, 2017). As expected based on was added to the onset of each MC discharge (Fig. 8B)andwhen the large unitary amplitude and absence of short-term depres- there was substantial random variation in when MCs responded sion, summating LANDD inputs from relatively few presynaptic to the same stimulus (Fig. 8C). These results suggest that GCs MC or TCs (;11) are required to bring a GC to threshold and may exploit differences in synaptic properties in multiple excita- trigger at least one AP (Fig. 8A). When GCs were excited exclu- tory pathways from OB principal cells to create parallel represen- sively through the DD pathway, GC spiking required sensory- tations of the same sensory input (e.g., with different degrees of evoked discharges in approximately 3 times as many MCs or sensitivity to odor concentration). TCs (;29). In addition to requiring different size active presynaptic ensembles, inputs through the DD and LANDD pathways Discussion appear specialized to convey different types of odor-related in- We make three conclusions based on the results presented in this formation. Increasing odor concentration prolongs the duration study. First, we identified a new excitatory synaptic input to ol- of both MC and TC discharges (Fukunaga et al., 2012), an effect factory GCs with rapid onset kinetics (termed LANDD EPSPs). that appears to be minimized in downstream GCs by the short- The novel LANDD EPSPs we identified are distinguished from term depression inherent in the DD pathway. Figure 8B,C illus- previously described DD (Pressler and Strowbridge, 2017)and trates the effect of varying the duration of synthetic MCs dis- SAF EPSPs (Balu et al., 2007; Gao and Strowbridge, 2009; Sun et charges (see Materials and Methods). In each simulation, the size al., 2020) by unitary EPSP properties and by markedly different of the presynaptic MC ensemble was increased until the GC forms of short-term plasticity. Second, we find that LANDD reached firing threshold in 95% of trials (see Materials and EPSPs likely arise from synaptic inputs onto the proximal den- Methods). The LANDD and DD curves represent the functional drites of GCs, especially the basal dendrites. Because of their effects of different short-term plasticity experimentally measured proximal location, we were able to record spontaneous and in both types of EPSPs (synaptic amplitude, probability, and evoked LANDD EPSPs in truncated GCs that lack distal apical response time course were kept identical [DD-like] in both sets dendrites, the primary site of DD synaptic inputs. The large am- of simulations). While GC AP latency was sensitive to the dura- plitude of LANDD EPSPs likely reflects minimal electrotonic tion of MC discharges for relatively brief trains (up to 6 APs) attenuation of AMPAR-mediated currents that are proximal to with activated synapses with DD-like short-term plasticity, the the somatic recording electrode rather than amplification by 9712 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons

Via LANDD input Via DD input These findings argue that LANDD EPSPs arise from axo-den- glom dritic excitatory synapses on GCs. Instead of reflecting an exter- layer nal input from glutamatergic neurons outside the OB, the final two lines of evidence suggest that at least some LANDD EPSPs DD arise from MCs and/or TCs, the only established glutamatergic cell types in the OB (Fremeau et al., 2001; Lein et al., 2007; MC/TC MC/TC Ohmomo et al., 2009). Approximately half of our truncated GC recordings had relatively frequent (.1 Hz) spontaneous EPSPs. All these recordings included spontaneous LANDD EPSPs and SAF blockade of spontaneous spiking with TTX decreased EPSP fre- GC quency. Because severed axons are typically not spontaneously active, this result implies that LANDD EPSPs originate from LANDD local (bulbar) excitatory neurons that are spontaneously active. Previous studies have demonstrated that both MCs and TCs can (to olfactory be spontaneously active in rodent brain slices (Heyward et al., cortex) 2001; Schmidt and Strowbridge, 2014). And finally, stimulation in the glomerular layer and OSN of OB slices drove polysynaptic Figure 9. Diagram of hypothesized routes of GC excitation leading to DD and LANDD EPSPs. Diagram of proposed local circuits responsible for MC/TC-to-GC excitation via the DD LANDD EPSPs in truncated GCs. The long latencies of sensory pathway (targeting the distal apical dendrite) and through the LANDD pathway (targeting afferent-driven LANDD EPSPs suggests that the stimulation acti- the basal dendrite). GCs are also excited by SAF EPSPs, shown terminating on the proximal vated intermediate excitatory neurons contained within the OB apical dendrite in the diagram. SAF EPSPs originate from both intrinsic bulbar circuits and slice. While our OB slices did not contain piriform cortex, we long-range feedback projections from piriform cortex. Blue text represents synaptic occasionally made slices that contained part of the AON. In those pathways. slices, we could dissect the AON region off the slice and still evoke polysynaptic LANDD EPSPs. Both MCs and TCs exhibit a intrinsic currents. And finally, LANDD EPSPs likely originate variety of spiking response patterns to focal glomerular stimula- from local MCs and/or TCs since we find the frequency of spon- tion (e.g., discharges and single spikes). The pattern of LANDD 1 taneous LANDD EPSPs was reduced when spontaneous Na EPSPs evoked by the same type of glomerular stimulation spiking was suppressed with TTX. Also, orthodromic stimulation matched both the variation in onset latency and latency jitter we in the glomerular layer and OSN triggered barrages of polysy- recorded in glutamatergic principal cells. naptic LANDD EPSPs, even in truncated GCs that lack DD syn- While our study does not demonstrate LANDD EPSPs in aptic inputs. Together, these results suggest that at least some monosynaptic paired recordings, such a study is likely to be MCs and/or TCs form axo-dendritic synapses on the proximal exceeding difficult given the large number of potential postsy- dendrites of GCs, providing a second othrodromic pathways to naptic GCs there are for each principal cell. With live 2-photon activate these interneurons (Fig. 9). Because of their different visualization of MC/TC processes (via intracellular recording forms of short-term plasticity, DD and LANDD EPSPs likely and dye filling), most filled MC and TC axons either were trun- reflect different aspects of MC/TC firing patterns with depressing cated at the slice surface or failed to bifurcate within the superfi- DD inputs primarily signaling the onset of an MC/TC spike dis- cial portion of the GCL, technical factors that accentuate the charge while temporally summating LANDD EPSPs enable GCs already sparse nature of the LANDD pathway. When this visual- to monitor the duration and frequency of the discharge. ization approach was successful as part of MC/GC and TC/GC paired recordings, we found GCL axon collaterals more fre- LANDD EPSPs likely reflect GC excitation from synapses quently in TCs than MCs (and see Igarashi et al., 2012). Another formed by MC/TC axon collaterals potential caveat of our work relates to truncated GCs, which Our results provide six lines of evidence that MCs/TCs activate have a higher input resistance than GCs with intact apical den- GCs via axo-dendritic synapses onto proximal dendrites. First, dritic arbors (Pressler and Strowbridge, 2019). While this effect we find that focal stimulation near proximal dendrites, especially may accentuate the amplitude of LANDD EPSPs recorded in basal dendrites, monosynaptically activates LANDD EPSPs. truncated GCs, we often recorded similar large-amplitude spon- LANDD synaptic responses appear to be unitary EPSPs with dis- taneous EPSPs in intact GCs. The higher input resistance (and tinct failures and successes with irreducible latencies .1.2 ms, longer membrane time constant) also would not explain very fast suggesting that they are not mediated by intrinsic currents acti- kinetics of LANDD EPSPs. In intact GCs, shunting associated vated by passive depolarization from electrical stimulation. with frequent spontaneous EPSPs and IPSPs may also attenuate Second, we evoked LANDD EPSPs by focal stimulation in both LANDD EPSPs. Finally, with our limited dataset of polysynaptic intact GCs and in truncated GCs that lack distal apical dendritic evoked LANDD EPSPs, we cannot yet determine the spatial arbors in the EPL and, therefore, DD EPSPs. This result suggests range of glomerular modules that can excite GCs through proxi- that LANDD EPSPs are not a large-amplitude variant of DD syn- mal inputs and whether individual presynaptic neurons excite aptic responses. Third, monosynaptic LANDD EPSPs are abol- GCs through both DD and LANDD pathways. ished by the non-NMDAR antagonist DNQX, demonstrated Two previous ultrastructural reports support the hypothesis they are driven by a glutamatergic presynaptic neuron. Fourth, that MC/TC axon collaterals synapse on GCs. Classic work by LANDD EPSPs have a distinctive functional signature, a neutral Price and Powell (1970b) suggested that the MC/TC axons con- PPR evident with repeated stimulation, that distinguishes them tacted GC dendrites within the GCL since they found asymmet- from both DD and SAF EPSPs. Excitatory synapses that neither ric synapses on GC dendrites within the GCL. The presynaptic facilitate nor depress with repeated stimulation are relatively terminals in these systems had vesicles that resembled vesicles in rare, although the association inputs to pyramidal cells in APC MC axon initial segments and in known synaptic terminals (Suzuki and Bekkers, 2006) show similar neutral PPRs. formed by MCs in piriform cortex. More recently, Schaefer and Pressler and Strowbridge · Activation of Granule Cell Interneurons J. Neurosci., December 9, 2020 • 40(50):9701–9714 • 9713 colleagues demonstrated synaptic contacts from a filled TC axon Dual LANDD/DD synaptic inputs may also facilitate learning collateral onto the proximal dendrite of a GC (Schwarz et al., the constellation of active glomerular modules activated by an 2018, their Supplemental Fig. 8A–C). Our results suggesting that odor through spike timing-dependent plasticity mechanisms if a local excitatory bulbar neurons form axo-dendritic synapses is subset of active principal cells drives GC spiking through proxi- also consistent with findings from a recent study from Sun et al. mal LANDD EPSPs. The GC could then use conventional long- (2020) who activated subsets of TCs optogenetically and evoked term plasticity mechanisms to associate the larger population of EPSPs in GCs with near-neutral PPRs (;1.3 average PPR). active principal cells that form DD synapses with the sparser set of “primary” glomeruli that initiated GC firing via LANDD inputs. Further studies will be necessary to determine whether Functional significance “ ” Principal cells in the OB respond to olfactory sensory input by such a teacher/learner learning model, reminiscent of classic models proposed by Marr (1969) and others, operates in the OB. generating an AP barrage at g-band frequencies that typically lasts for 30-100 ms (e.g., Eeckman and Freeman, 1990). Because of their different forms of short-term plasticity, excitatory input References through DD and LANDD synapses likely convey two divergent Abraham NM, Spors H, Carleton A, Margrie TW, Kuner T, Schaefer AT signals to GC interneurons in response to MC/TC discharges. (2004) Maintaining accuracy at the expense of speed: stimulus similarity defines odor discrimination time in mice. Neuron 44:865–876. Strongly depressing DD inputs function to detect the onset of a Abraham NM, Egger V, Shimshek DR, Renden R, Fukunaga I, Sprengel R, MC or TC discharge since the largest postsynaptic depolarization Seeburg PH, Klugmann M, Margrie TW, Schaefer AT, Kuner T (2010) occurs following the initial AP in the discharge. Synaptic depres- Synaptic inhibition in the olfactory bulb accelerates odor discrimination sion reduces the probability that GCs will continue to spike in in mice. Neuron 65:399–411. response to subsequent spikes within the MC/TC discharge. By Apicella A, Yuan Q, Scanziani M, Isaacson JS (2010) Pyramidal cells in piri- form cortex receive convergent input from distinct olfactory bulb glo- contrast, LANDD EPSPs temporally summate in response to – g-band tetanic stimulation (Fig. 4A–C), that mimic odor-evoked meruli. J Neurosci 30:14255 14260. Balu R, Pressler RT, Strowbridge BW (2007) Multiple modes of synaptic exci- MC/TC discharges, allowing this pathway to drive GC spiking tation of olfactory bulb granule cells. J Neurosci 27:5621–5632. later within the presynaptic MC/TC discharge. Banerjee A, Dave RN (2004) Validating clusters using the Hopkins statistic. Whether GCs spike primarily in response to the onset of prin- In: IEEE International Conference on Fuzzy Systems, pp 149–153, Vol 1, cipal cell discharges or throughout the discharge will have a large Budapest, Hungary. impact on both the neural signals that activate GCs and the result- Bolding KA, Franks KM (2017) Complementary codes for odor identity and ing recurrent inhibition GCs provide back to principal cells. intensity in olfactory cortex. Elife 6:e22630. Bolding KA, Franks KM (2018) Recurrent cortical circuits implement con- Through the LANDD excitatory pathway, GC responses are likely centration-invariant odor coding. Science 361:eaat6904. influenced by the duration of principal cell discharges. Odor con- Boyd AM, Sturgill JF, Poo C, Isaacson JS (2012) Cortical feedback control of centration modulates the duration of MC and TC discharges olfactory bulb circuits. Neuron 76:1161–1174. (Fukunaga et al., 2012), suggesting that GCs likely receive sensory Brill MS, Ninkovic J, Winpenny E, Hodge RD, Ozen I, Yang R, Lepier A, signals that are less sensitive to odor concentration through the DD Gascón S, Erdelyi F, Szabo G, Parras C, Guillemot F, Frotscher M, pathway, while LANDD EPSPs convey more information-related Berninger B, Hevner RF, Raineteau O, Götz M (2009) Adult generation of glutamatergic olfactory bulb interneurons. Nat Neurosci 12:1524– odor concentration. Therefore, LANDD inputs could facilitate odor 1533. concentration coding since they would convey excitatory drive to Budzillo A, Duffy A, Miller KE, Fairhall AL, Perkel DJ (2017) Dopaminergic GCs throughout the odor-evoked MC/TC barrage. In vivo record- modulation of basal ganglia output through coupled excitation-inhibi- ings (Cang and Isaacson, 2003; Wienisch and Murthy, 2016)also tion. Proc Natl Acad Sci USA 114:5713–5718. demonstrate that the intensity and duration of GC discharges are Bywalez WG, Patirniche D, Rupprecht V, Stemmler M, Herz AV, Pálfi D, modulated by odor concentration, suggesting that the nondepress- Rózsa B, Egger V (2015) Local postsynaptic voltage-gated sodium chan- nel activation in dendritic spines of olfactory bulb granule cells. Neuron ing LANDD pathway we identified onto GCs is likely active in vivo. 85:590–601. The ability of LANDD EPSPs to trigger late discharges in Cang J, Isaacson JS (2003) In vivo whole-cell recording of odor-evoked syn- GCs, compared with the transient DD excitatory input near the aptic transmission in the rat olfactory bulb. J Neurosci 23:4108–4116. start of MC/TC firing, is likely to influence olfactory function. Carleton A, Petreanu LT, Lansford R, Alvarez-Buylla A, Lledo PM (2003) Accurate discrimination between similar odorants requires more Becoming a new neuron in the adult olfactory bulb. Nat Neurosci 6:507– processing time than easier odor classifications (Uchida and 518. Cazakoff BN, Lau BY, Crump KL, Demmer HS, Shea SD (2014) Broadly Mainen, 2003; Abraham et al., 2004; Rinberg et al., 2006). tuned and respiration-independent inhibition in the olfactory bulb of Interneuron excitation via LANDD EPSPs provides a cellular awake mice. Nat Neurosci 17:569–576. mechanism for MC/TC inhibition to develop later in the odor de Olmos J, Hardy H, Heimer L (1978) The afferent connections of the main response, presumably reflecting longer periods of synaptic inte- and the accessory olfactory bulb formations in the rat: an experimental gration in GCs, that could help pattern principal cell discharges. HRP-study. J Comp Neurol 181:213–244. The large amplitude of LANDD EPSPs also provides a potential De Saint Jan D, Hirnet D, Westbrook GL, Charpak S (2009) External tufted cells drive the output of olfactory bulb glomeruli. J Neurosci 29:2043– mechanism to explain sparse GC odor responses (relative to 2052. MC/TC responses) (Kato et al., 2012, 2013; Cazakoff et al., 2014; Eeckman FH, Freeman WJ (1990) Correlations between unit firing and EEG and see Koulakov and Rinberg, 2011). GCs express fewer dendri- in the rat olfactory system. Brain Res 528:238–244. tic spines on their basal dendritic arbor, the likely site of LANDD Egger V, Svoboda K, Mainen ZF (2005) Dendrodendritic synaptic signals in inputs, than on their distal apical dendrites where DD inputs are olfactory bulb granule cells: local spine boost and global low-threshold – found (Price and Powell, 1970c; Carleton et al., 2003; Whitman spike. J Neurosci 25:3521 3530. and Greer, 2007; Kelsch et al., 2008). These anatomic differences Fiete IR, Senn W, Wang CZ, Hahnloser RH (2010) Spike-time-dependent plasticity and heterosynaptic competition organize networks to produce suggest that the LANDD input pathway may reflect a narrower long scale-free sequences of neural activity. Neuron 65:563–576. range of presynaptic glomerular modules and, therefore, sparser Fremeau RT, Troyer MD, Pahner I, Nygaard GO, Tran CH, Reimer RJ, odor coding. Bellocchio EE, Fortin D, Storm-Mathisen J, Edwards RH (2001) The 9714 • J. Neurosci., December 9, 2020 • 40(50):9701–9714 Pressler and Strowbridge · Activation of Granule Cell Interneurons

expression of vesicular glutamate transporters defines two classes of exci- Pressler RT, Strowbridge BW (2006) Blanes cells mediate persistent feedfor- tatory synapse. Neuron 31:247–260. ward inhibition onto granule cells in the olfactory bulb. Neuron 49:889– Fukunaga I, Berning M, Kollo M, Schmaltz A, Schaefer AT (2012) Two dis- 904. tinct channels of olfactory bulb output. Neuron 75:320–329. Pressler RT, Strowbridge BW (2017) Direct recording of dendrodendritic ex- Fukunaga I, Herb JT, Kollo M, Boyden ES, Schaefer AT (2014) Independent citation in the olfactory bulb: divergent properties of local and external control of gamma and theta activity by distinct interneuron networks in glutamatergic inputs govern synaptic integration in granule cells. J the olfactory bulb. Nat Neurosci 17:1208–1216. Neurosci 37:11774–11788. Gao Y, Strowbridge BW (2009) Long-term plasticity of excitatory inputs to Pressler RT, Strowbridge BW (2019) Functional specialization of interneuron – granule cells in the rat olfactory bulb. Nat Neurosci 12:731 733. dendrites: identification of action potential initiation zone in axonless ol- Gire DH, Schoppa NE (2009) Control of on/off glomerular signaling by a factory bulb granule cells. J Neurosci 39:9674–9688. – local GABAergic microcircuit in the olfactory bulb. J Neurosci 29:13454 Price JL, Powell TP (1970a) The mitral and short axon cells of the olfactory 13464. bulb. J Cell Sci 7:631–651. Herzog E, Bellenchi GC, Gras C, Bernard V, Ravassard P, Bedet C, Gasnier Price JL, Powell TP (1970b) The synaptology of the granule cells of the olfac- B, Giros B, El Mestikawy S (2001) The existence of a second vesicular glu- tory bulb. J Cell Sci 7:125–155. tamate transporter specifies subpopulations of glutamatergic neurons. J Price JL, Powell TP (1970c) The morphology of the granule cells of the olfac- Neurosci 21:RC181. tory bulb. J Cell Sci 7:91–123. Heyward P, Ennis M, Keller A, Shipley MT (2001) Membrane bistability in Rall W, Shepherd GM, Reese TS, Brightman MW (1966) Dendrodendritic olfactory bulb mitral cells. J Neurosci 21:5311–5320. synaptic pathway for inhibition in the olfactory bulb. Exp Neurol 14:44– Igarashi KM, Ieki N, An M, Yamaguchi Y, Nagayama S, Kobayakawa K, 56. Kobayakawa R, Tanifuji M, Sakano H, Chen WR, Mori K (2012) Parallel mitral and tufted cell pathways route distinct odor information to differ- Rinberg D, Koulakov A, Gelperin A (2006) Speed-accuracy tradeoff in olfac- – ent targets in the olfactory cortex. J Neurosci 32:7970–7985. tion. Neuron 51:351 358. Isaacson JS, Strowbridge BW (1998) Olfactory reciprocal synapses: dendritic Schmidt LJ, Strowbridge BW (2014) Modulation of olfactory bulb network signaling in the CNS. Neuron 20:749–761. activity by serotonin: synchronous inhibition of mitral cells mediated by – Kashiwadani H, Sasaki YF, Uchida N, Mori K (1999) Synchronized oscilla- spatially localized GABAergic microcircuits. Learn Mem 21:406 416. tory discharges of mitral/tufted cells with different molecular receptive Schoppa NE, Kinzie JM, Sahara Y, Segerson TP, Westbrook GL (1998) ranges in the rabbit olfactory bulb. J Neurophysiol 82:1786–1792. Dendrodendritic inhibition in the olfactory bulb is driven by NMDA Kato HK, Chu MW, Isaacson JS, Komiyama T (2012) Dynamic sensory rep- receptors. J Neurosci 18:6790–6802. resentations in the olfactory bulb: modulation by wakefulness and experi- Schwarz D, Kollo M, Bosch C, Feinauer C, Whiteley I, Margrie TW, Cutforth ence. Neuron 76:962–975. T, Schaefer AT (2018) Architecture of a mammalian glomerular domain Kato HK, Gillet SN, Peters AJ, Isaacson JS, Komiyama T (2013) revealed by novel volume electroporation using nanoengineered micro- Parvalbumin-expressing interneurons linearly control olfactory bulb out- electrodes. Nat Commun 9:183. put. Neuron 80:1218–1231. Shepherd GM, Chen WR, Willhite D, Migliore M, Greer CA (2007) The ol- Kelsch W, Lin CW, Lois C (2008) Sequential development of synapses in factory granule cell: from classical enigma to central role in olfactory dendritic domains during adult neurogenesis. Proc Natl Acad Sci USA processing. Brain Res Rev 55:373–382. 105:16803–16808. Shipley MT, Adamek GD (1984) The connections of the mouse olfactory Kishi K, Mori K, Ojima H (1984) Distribution of local axon collaterals of mi- bulb: a study using orthograde and retrograde transport of wheat germ tral, displaced mitral, and tufted cells in the rabbit olfactory bulb. J Comp agglutinin conjugated to horseradish peroxidase. Brain Res Bull 12:669– Neurol 225:511–526. 688. Koulakov AA, Rinberg D (2011) Sparse incomplete representations: a poten- Shusterman R, Smear MC, Koulakov AA, Rinberg D (2011) Precise olfactory – tial role of olfactory granule cells. Neuron 72:124 136. responses tile the sniff cycle. Nat Neurosci 14:1039–1044. Lein ES, Hawrylycz MJ, Ao N, Ayres M, Bensinger A, Bernard A, Boe AF, Sun X, Liu X, Starr ER, Liu S (2020) CCKergic tufted cells differentially drive Boguski MS, Brockway KS, Byrnes EJ, Chen L, Chen L, Chen TM, Chin two anatomically segregated inhibitory circuits in the mouse olfactory MC, Chong J, Crook BE, Czaplinska A, Dang CN, Datta S, Dee NR, et al. bulb. J Neurosci 40:6189–6206. (2007) Genome-wide atlas of gene expression in the adult mouse brain. – Suzuki N, Bekkers JM (2006) Neural coding by two classes of principal cells Nature 445:168 176. in the mouse piriform cortex. J Neurosci 26:11938–11947. Marr D (1969) A theory of cerebellar cortex. J Physiol 202:437–470. Suzuki N, Bekkers JM (2011) Two layers of synaptic processing by principal McLean JH, Shipley MT (1987) Serotonergic afferents to the rat olfactory neurons in piriform cortex. J Neurosci 31:2156–2166. bulb: I. Origins and laminar specificity of serotonergic inputs in the adult Tatti R, Bhaukaurally K, Gschwend O, Seal RP, Edwards RH, Rodriguez I, rat. J Neurosci 7:3016–3028. Carleton A (2014) A population of glomerular glutamatergic neurons McLean JH, Shipley MT (1991) Postnatal development of the noradrenergic controls sensory information transfer in the mouse olfactory bulb. Nat projection from locus coeruleus to the olfactory bulb in the rat. J Comp Commun 5:3791. Neurol 304:467–477. Mori K, Nagao H, Yoshihara Y (1999) The olfactory bulb: coding and proc- Uchida N, Mainen ZF (2003) Speed and accuracy of olfactory discrimination – essing of odor molecule information. Science 286:711–715. in the rat. Nat Neurosci 6:1224 1229. Nagayama S, Takahashi YK, Yoshihara Y, Mori K (2004) Mitral and tufted Uva L, Strowbridge BW, Curtis MD (2006) Olfactory bulb networks revealed cells differ in the decoding manner of odor maps in the rat olfactory bulb. by lateral olfactory tract stimulation in the in vitro isolated guinea-pig – J Neurophysiol 91:2532–2540. brain. Neuroscience 142:567 577. Najac M, De Saint Jan D, Reguero L, Grandes P, Charpak S (2011) Whitman MC, Greer CA (2007) Synaptic integration of adult-generated ol- Monosynaptic and polysynaptic feed-forward inputs to mitral cells from factory bulb granule cells: basal axodendritic centrifugal input precedes olfactory sensory neurons. J Neurosci 31:8722–8729. apical dendrodendritic local circuits. J Neurosci 27:9951–9961. Nunes D, Kuner T (2015) Disinhibition of olfactory bulb granule cells accel- Wienisch M, Murthy VN (2016) Population imaging at subcellular resolution erates odour discrimination in mice. Nat Commun 6:8950. supports specific and local inhibition by granule cells in the olfactory Ohmomo H, Ina A, Yoshida S, Shutoh F, Ueda S, Hisano S (2009) Postnatal bulb. Sci Rep 6:29308. changes in expression of vesicular glutamate transporters in the main ol- Wilson CD, Serrano GO, Koulakov AA, Rinberg D (2017) A primacy code factory bulb of the rat. Neuroscience 160:419–426. for odor identity. Nat Commun 8:1477.