Author response: Functional and anatomical specificity in a higher olfactory centre Article Swipe

MAKE IMPACT

Shahar Frechter , Alexander Shakeel Bates , Sina Tootoonian , Michael-John Dolan , James D. Manton , Arian R. Jamasb , Johannes Kohl , Davi D. Bock , Gregory S.X.E. Jefferis · YOU? · · 2019 · Open Access · · DOI: https://doi.org/10.7554/elife.44590.026

Article Figures and data Abstract Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Most sensory systems are organized into parallel neuronal pathways that process distinct aspects of incoming stimuli. In the insect olfactory system, second order projection neurons target both the mushroom body, required for learning, and the lateral horn (LH), proposed to mediate innate olfactory behavior. Mushroom body neurons form a sparse olfactory population code, which is not stereotyped across animals. In contrast, odor coding in the LH remains poorly understood. We combine genetic driver lines, anatomical and functional criteria to show that the Drosophila LH has ~1400 neurons and >165 cell types. Genetically labeled LHNs have stereotyped odor responses across animals and on average respond to three times more odors than single projection neurons. LHNs are better odor categorizers than projection neurons, likely due to stereotyped pooling of related inputs. Our results reveal some of the principles by which a higher processing area can extract innate behavioral significance from sensory stimuli. https://doi.org/10.7554/eLife.44590.001 Introduction In thinking about the transition from stimulus through perception to behavior, chemosensory systems have become increasingly studied due to their relatively shallow architecture: just two synapses separate the sensory periphery from neurons that are believed to form memories or instruct behavior (Wilson and Mainen, 2006; Masse et al., 2009). However, there are many shared organizational features with other sensory systems : for example, integration of information originating from distinct sensory receptors, and multiple levels of parallel and hierarchical processing. It is therefore possible that detailed studies of chemosensory systems may reveal general principles relevant to neurons that are considerably deeper in other sensory modalities. This study uses single cell neuroanatomy and genetically targeted in vivo electrophysiology to addresses two principal questions in the context of the Drosophila melanogaster olfactory system. First, can a higher olfactory center encode odors in a stereotyped way? More specifically, are there distinct cell types with reproducible odor responses across animals? Second, what is the nature of the population code in a higher olfactory area linked to innate behavior? How might the code relate to the behavioral requirements of the animal and how does it differ from neurons underlying learned responses? The olfactory systems of mammals and insects have many similarities, including the presence of glomerular units in the first olfactory processing center. Second order neurons then make divergent projections onto multiple higher olfactory centers. For example in both flies and mice there are separate projections to areas proposed to be specialized for memory formation (mushroom body and piriform cortex, respectively) and unlearned olfactory behaviors (lateral horn and e.g. cortical amygdala) (Heimbeck et al., 2001; Jefferis et al., 2007; Sosulski et al., 2011; Root et al., 2014). In insects in general and Drosophila melanogaster in particular, the anatomical and functional logic of odor coding in the mushroom body (MB) and its relationship to olfactory learning has been intensively studied. About 150 projection neurons (PNs) relay information to the input zone of the MB, the calyx, where they form synapses with the dendrites of ~2000 Kenyon cells, the intrinsic neurons of the MB (reviewed in Masse et al., 2009). There is limited spatial stereotypy in these projections (Jefferis et al., 2002; Wong et al., 2002; Tanaka et al., 2004; Jefferis et al., 2007; Lin et al., 2007) and each KC receives input from an apparently random sample of ~6 PNs (Caron et al., 2013). KC axons form a parallel fibre system intersected by the dendrites of 35 MB output neurons (Aso et al., 2014a); it is proposed that memories are stored by synaptic depression at these synapses. KC odor responses are very sparse (Perez-Orive et al., 2002; Turner et al., 2008), which is proposed to minimize interference between different memories, but do not appear stereotyped across animals (Murthy et al., 2008, but see Wang et al., 2004). Random PN-KC connectivity is consistent with the idea that KC responses acquire meaning through associative learning rather than having any intrinsic valence. In mice, pyramidal cells of the piriform cortex also integrate coincident inputs from different glomeruli (Miyamichi et al., 2011; Davison and Ehlers, 2011), but as in the MB the available evidence suggests that this integration is not stereotyped from animal to animal (Stettler and Axel, 2009; Choi et al., 2011). In contrast to extensive studies of the MB, there is much more limited information concerning anatomy and function of the lateral horn (LH); this is also true for higher olfactory centers of the mammalian brain that have been hypothesized to serve a similar functional role such as the cortical amygdala. Studies examining the axonal arbors of PNs showed that these have more stereotyped projections in the LH than in the MB (Marin et al., 2002; Wong et al., 2002; Tanaka et al., 2004; Jefferis et al., 2007). Since at least three classes of LH neurons (LHNs) appeared to have dendrites in stereotyped locations (Tanaka et al., 2004; Jefferis et al., 2007), it was hypothesized that these neurons have stereotyped odor responses that are conserved from animal to animal. The first studies of odor responses of LHNs focussed on pheromone responses of neurons suspected to be sexually dimorphic in number or anatomy owing to their expression of the fruitless gene (Ruta et al., 2010; Kohl et al., 2013). Kohl et al. (2013) characterized three neuronal clusters showing that they responded in a sex- and class-specific manner and ranged from narrowly tuned pheromone-specialists to more broadly responsive neurons. Pheromone responsive second order neurons project axons to a specialized subregion of the LH (Seki et al., 2005; Jefferis et al., 2007) and the extent to which pheromone responsive LHNs are typical of the whole LH has been questioned by other studies. In particular (Gupta and Stopfer, 2012) recorded from a random sample of neurons in the locust LH, reporting that all LHNs were extremely broadly tuned and without finding evidence of neurons with repeated odor profiles; they eventually concluded that generalist LHNs are unlikely to be stereotyped encoders of innate behavior. Fişek and Wilson (2014) carried out the first electrophysiological recordings of non-pheromone LH neurons in Drosophila. They recorded from two genetically identified cell types with reproducible response patterns: one LHN class responded to 1 out of 8 tested odorants, the other responded to all odorants. Although these results suggested that generalist LHNs can also have stereotyped odor responses, the limited number of neurons investigated precluded general conclusions about LH odor coding. Studies of analogous regions in the mammalian brain are even more challenging, but recent recordings by Lurilli and Datta (2017) from the cortical amygdala found no evidence for response stereotypy or encoding of the behavioral valence or chemical category of odors. We have taken a stepwise approach to understanding the organizational and functional logic of the LH. We first reasoned that it was essential to characterize its cellular composition and to develop approaches for reproducible access to different cell populations. We achieved this by screening and annotating genetic driver lines, hierarchically classifying single cell morphologies, and using whole brain electron microscopy (EM) data to place rigorous bounds on total cell numbers, which turn out to be much greater than anticipated. We then used genetically targeted single cell electrophysiology and anatomy to reveal principles of odor coding in genetically defined cell populations, finding that LHNs typically respond to more odors but with fewer spikes than their PN input. Since we found that LHNs show stereotyped odor responses across animals, we carried out a detailed analysis of neuronal cell types. We show that functional and morphological criteria can both be used to define cell type and that they are highly consistent. We then show that LHNs are better odor categorizers than their PN inputs, providing one justification for their distinct coding properties. Finally we use EM data to measure direct convergence of different olfactory channels onto both local and output neurons of the LH, providing an anatomical substrate for the response broadening. Taken together these results reveal some of the logic by which the nervous system can map sensory responses onto behaviorally relevant categories encoded in a population of stereotyped cell types in a higher brain area. Results Key anatomical features of the lateral horn Our key goal in this study was to understand the coding principles of third order neurons underlying innate olfactory behaviors. Nevertheless it is hard to understand the functional properties of a brain area without a basic understanding of the number and variety of constituents neurons. We used a wide variety of experimental/analytic approaches to obtain a comprehensive overview of the functional anatomy of the LH. We now present the observations most relevant to odor coding, organized hierarchically. We present further details in Materials and methods and online supplements including accompanying 3D data (jefferislab.org/si/lhlibrary). Neuropil volume is indicative of the energetic investment in particular sensory information (Sterling and Laughlin, 2015), and strongly correlated with length of neuronal cable and synapse numbers (e.g. Schlegel et al., 2017). We find that the volume of the LH is 70% of the volume of the whole MB. However many LHNs extend axons beyond the LH; we estimate that the total volume of LHN arbors is therefore actually (40%) greater than the MB. This large investment in neuropil volume argues for the significance of the LH in sensory processing, whereas the literature currently contains 13 fold more studies of the Drosophila MB than LH (Materials and methods section Neuropil Volumes). The number of neurons in a brain area is a key determinant of neuronal coding. A classic EM study cutting the MB’s parallel axon tract counted 2200 Kenyon cells (Technau and Heisenberg, 1982), while comprehensive genetic driver lines contain up to 2000 KCs (Aso et al., 2009). However the number of LHNs has remained undefined, since there is no single tract to cross-section, nor any driver line labelling all LHNs. In the locust, (Gupta and Stopfer, 2012) estimated that there are fewer LHNs than PNs. We combined light level image data with whole brain electron microscopy (EM) (Zheng et al., 2018) to address this question. Our anatomical screen (see below) identified 31 primary neurite tracts entering the LH (Figure 1B and Table 1); a random EM tracing procedure targeting 17 of the largest tracts yielded an estimate of 1410 LHNs (90% CI 1368–1454, see Experimental Procedures). Each tract consists predominantly of either output or local neurons giving an estimate of about 580 LH local neurons (LHLNs, 40%) and 830 LH output neurons (LHONs, 60%). These results show that LHONs are much more numerous than second order input PNs and within a factor of 2 of the number of third order MB Kenyon cells. The large number of KCs enables sparse odor coding, which is proposed to avoid synaptic interference during memory formation (reviewed by Masse et al., 2009). Why should the LH also contain so many neurons? Figure 1 with 1 supplement see all Download asset Open asset Screen for genetic driver lines labeling Lateral Horn Neurons. (A) Flow diagram following olfactory information to third order neurons of the LH and MB calyx. (B) Section through the PV5 primary neurite tract within the EM data; each profile was identified as an LHN (circles) or non LHN (triangles) by tracing to the first branch point. Scale bar 1 μm. (C) Sample Split-GAL4 intersection with parental lines inset. Cell body locations marked with white circle. (D) Summary table for the genetic screen. (E) Matrix summarizing which driver lines contained which anatomy groups (colored LHON = blue, LHLN = green). The vast majority of genetic driver lines labeled only a few LH anatomy groups. NB these data are also available as a supplementary spreadsheet. (F) Anterior (left) and posterior (right) views of the different LHN primary neurite tracts demonstrating the broad origin of LHNs. Grey dashed arrow indicates the order of increasing tract numbers for ventral PNTs (Lower arrow) and dorsal PNTs (above). The entry point into the brain rather than soma location was the point of reference for naming tracts. (G) Upper panels, cartoons summarizing the logic of the LHN naming system, lower panels, the PV5 primary neurite as an example. Note this includes only 3 out of 24 cell types in PV5. For each cell type one cell is highlighted (thick lines). https://doi.org/10.7554/eLife.44590.002 Table 1 LHN tracts characterized in electron microscopy data. Tracts¯ match the Primary Neurite Tract nomenclature defined in Figure 1. Type¯ indicates whether the tract contains output or local neurons or a mix of both. Profiles¯ indicates the total number of profiles within the tract. Est.LHNs¯ indicates the sampling based estimate for the number of LHNs in the tract. Range¯ gives a 90% confidence interval. https://doi.org/10.7554/eLife.44590.004 TractTypeProfilesEst. LHNsRangeRecordedAV4LHLN>>LHON324252244–259YesPV4LHLN>LHON158155152–158YesPV2LHLN>>LHON1939281–102YesPD3LHLN755943–75PD4LHLN882210–33–LHLNs–838578555–602AV3LHON>LHLN144140140PD2LHON193128128YesPV5LHON127119119YesAD1LHON286116102–130YesAV6LHON32310696–115YesAV2LHON>>LHLN986349–77YesAD3LHON595959AV7LHON1414825–70AV1LHON332525AV5LHON108177–27PV3LHON52120–25AD2LHON5200–LHONs–1616832797–868–Total–245414111368–1454 Driver lines and hierarchical naming system for LHNs Transgenic driver lines are the standard approach to label and manipulate neurons in Drosophila (Venken et al., 2011). Given the large number of LHNs, it seemed essential to identify lines targeting subpopulations to test our hypothesis that LHNs are stereotyped odor encoders. When we began our studies, the only relevant lines came from Tanaka et al. (2004), who identified four drivers labeling distinct populations of LHNs from a screen of ~4,000 GAL4 lines. We carried out an enhancer trap Split-GAL4 screen (Luan et al., 2006), eventually selecting 234 lines containing LHNs from 2769 screened (see Experimental Procedures for details). These lines gave us access to the majority of known classes of LHNs and were used for most functional studies in this paper. However they were rarely cell type-specific, making them less suitable for behavioral experiments. In addition to the previous absence of genetic reagents, LHNs are morphologically extremely diverse and do not innervate discrete glomeruli or compartments (Laissue et al., 1999; Tanaka et al., 2008; Aso et al., 2014a). For all these reasons, there was no prior system to name and classify different LHNs. We devised a hierarchical naming system with three levels of increasing anatomical detail to disambiguate neurons (Figure 1F and G): (1) primary neurite tract: the tract containing fibers connecting the soma to the rest of the neuron, (2) anatomy group: neurons sharing a common axon tract and broadly similar arborizations in the LH and target areas, (3) cell type: the finest level revealed by reproducible differences in precise axonal or dendritic arborization patterns. We use single cell data to illustrate these three levels for three closely related cell types (PV5a1, PV5b1, PV5b2) (Figure 1G, see Materials and methods for details). This system was key to successful genetic screening as well as planning and reporting functional experiments. Building on our initial screen, we also annotated (Figure 1D) the widely used FlyLight (often referred to as GMR lines, Jenett et al., 2012) and Vienna Tiles libraries (VT lines, Tirian and Dickson, 2017). These lines are now very widely used in Drosophila neurobiology, in part because co-registered 3D image data are publicly available (e.g. through virtualflybrain.org Milyaev et al., 2012; Manton et al., 2014). The vast majority of genetic driver lines labeled only a few LH anatomy groups (mean of 3.8) while just 21/422 lines contained more than 8; we did not find any lines specific to multiple LHN anatomy groups without labelling other neurons in the central brain (Figure 1D and E). Similarly, none of the primary neurite tracts proved LH specific, although some were highly LH enriched (Figure 1B). This demonstrates how hard it is to obtain LH selective lines that label most or even a large portion of the LHNs. At the conclusion of our screen we had identified 69 distinct LHN anatomy groups – that is, neurons with substantially different axonal tracts/arborisation patterns – each of which was consistently labeled by a subset of driver lines. This cellular and genetic diversity significantly exceeded our initial expectations and represented an almost order of magnitude increase over prior studies. It also contrasts very strongly with the seven genetically defined Kenyon cell types comprising the third order neurons of the mushroom body (Aso et al., 2014a). This second screen of GMR/VT lines provides a link between our LHN classification and experimentally valuable resources including further driver lines and co-registered 3D image data (see Materials and methods). Indeed, building on these annotations we went on to prepare a large collection of highly specific intersectional Split-GAL4 lines selectively targeting specific LH cell types; this facilitates many experiments including behavioral analysis for which our first generation split-GAL4 reagents were less suitable (see Dolan et al., 2019, sister manuscript). Single cell anatomy of the lateral horn The results presented so far provide principled estimates of the number of LHNs, identify genetic reagents for their study and develop a hierarchical nomenclature classification system. The final part of our neuroanatomical groundwork was to carry out a large scale single cell analysis of the LH in order to gain an initial understanding of the variety of cell types that it contains. Given our new estimate that there are ~1400 neurons LHNs, what is the anatomical and functional diversity amongst this large number of neurons? To address this, we co-registered FlyCircuit neurons (Chiang et al., 2011) and neurons recorded during this study, segmenting each into axonal and dendritic (Figure see Materials and methods). We an online 3D of LHNs as well as LH input neurons. We first LH inputs (Figure The principal olfactory inputs to the LH have been but we found new classes including many inputs (see Figure supplement and Materials and methods for details). olfactory neurons, and and neurons were all in a of the LH (Figure that the LH is a Figure 2 with 2 supplements see all Download asset Open asset Single cell anatomy of the Lateral (A) Sample single projection neurons with axonal projections in the LH, showing all axon tracts and sensory that provide input. (B) up of the LH with axonal arbors for all FlyCircuit neurons of each sensory (C) of our annotated LHN showing all with LH LHN cell types (see Figure supplement and neurons electrophysiological in the present by anatomy (D) of single for all cell types for which we have in the or from which we electrophysiological recordings in this neurons in blue, local neurons in (E) for each target the total axonal cable length by all LHONs as of for each identified cell show in 3 axonal and of neuropil to et al. lateral tract. To classify LHNs, neurons with dendrites in the LH, we neurons to anatomy groups and cell types (Figure using et al., by a of within and across cell type stereotypy in patterns (Figure supplement 1B). We found that there is no single for that is for all LH cell types (Figure supplement even these anatomical cell type can be by other cellular properties (see Nevertheless for cell types with more than one neuron, identified the cell type of the (Figure supplement We identified a total of LHN cell types into local cell types with arbors to the LH and LHON cell types with axons beyond the LH. Most cell types from the tracts identified by EM as containing the largest number of LHNs are with about the number of tracts as LHONs EM Table the and tracts for of the estimated 580 local neurons, whereas it LHON tracts to this LHONs are more originating from tracts as well as having a wide of axonal LHNs had dendrites in multiple We therefore focussed on a of LHON cell types with their dendrites within the LH (Figure supplement see Materials and methods for details). These LHONs project to a wide of target areas (Figure The and are the most they are the location of where direct olfactory output from the LH may be with learned olfactory information from the mushroom body (Aso et al., 2014a). The is the this area also receives extensive input from projection neurons originating in the (e.g. et al., and is by dendrites of neurons, including at least two now known to be of LHONs et al., LHONs also have both axonal and dendritic in the LH. We that few cell types project to the because most olfactory projection neurons in the information from both The most studied projections to the LH are projection neurons from the that through the tract 3D of these PNs have been based on and of single cell data (Jefferis et al., 2007; et al., there are numerous inputs to the LH. We annotated LH input neurons from FlyCircuit and them into different groups based on the axon tract they use to the LH and their of dendritic arborization (Figure supplement which we used as a for the of the sensory information they We the naming system of Tanaka et al. to types not identified (see Materials and methods). Tanaka et al. have three tract three lateral tract and three tract PN types that project to the LH from the PNs any tract can have or dendritic in the sampling broadly or from the available odor Tanaka et al. but as has been in the et al., we find that some of these olfactory projections do not in the MB not olfactory input is known to be PNs the (Wilson and 2005; et al., whereas the majority of projections through the and are to be (Tanaka et al., We were not to find a few PN types that had been in the literature to project to the LH, including (Tanaka et al., is not within the LH. and PNs project widely but a of the LH, which is the of olfactory neurons and other sensory this arborization is shared by two neurons et al., and et al., 2007). In contrast, projection neurons are in the LH, where their arbors with PNs from and glomeruli et al., et al., in addition projection neurons from the neuropil may carry input et al., 2009; and projection neurons also innervate this although these are in an to the LH (see also et al., 2017). In the ventral LH receives input and likely to be in integration while the is predominantly to sampling in the FlyCircuit it is very likely that some cell types are while may be It has been from electron microscopy that the contains the and the fibers from the of the (Tanaka et al., In order to more of these we have a number of data and code resources (see These a 3D at which also to the highly selective split-GAL4 reagents in our et al., sister as well as LHN cell types characterized by et al. responses of lateral horn neurons The neuroanatomical groundwork that we have just includes a of detail that be relevant for many studies. However as we turn our to olfactory coding one one Why are there so many LHN cell To this we began by the odor response properties of LHNs and them with their the PNs. We also to contrast LHN responses with of MB Kenyon cells, the other class of third order olfactory genetic driver lines in we were to carry out targeted recordings from LHNs (Figure Given that these cells had response properties and our previous was that in LHN are not a measure of LHN we carried out in vivo whole cell recordings as we have et al., 2013). We recorded cells of which and identified the criteria for in our population analysis (see Experimental Procedures). basic electrophysiological across different both LHONs and have a much higher input and lower cell than PNs (Figure and Figure supplement this suggests that the energetic of spikes be lower in LHONs than PNs. Figure 3 with 1 supplement see all Download asset Open asset odor responses of second and third order olfactory neurons. for two PNs and seven LHONs Each odor was presented times to each cell with a For each odor the response of the four was while show the response for each Note the in and between PNs and and on the can be identified from a supplementary spreadsheet. Figure Download asset Open asset of second and third order olfactory neurons. (A) of used responses (see Materials and methods). (B) Matrix showing responses of and LHONs match Figure to different odors. A and white

Type

peer-review

Language

en

Landing Page

https://doi.org/10.7554/elife.44590.026

OA Status

gold

Cited By

4

Related Works

10

OpenAlex ID

https://openalex.org/W2988441760

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W2988441760

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.7554/elife.44590.026

Digital Object Identifier

Title

Author response: Functional and anatomical specificity in a higher olfactory centre

Work title

Type

peer-review

OpenAlex work type

Language

en

Primary language

Publication year

2019

Year of publication

Publication date

2019-04-12

Full publication date if available

Authors

Shahar Frechter, Alexander Shakeel Bates, Sina Tootoonian, Michael-John Dolan, James D. Manton, Arian R. Jamasb, Johannes Kohl, Davi D. Bock, Gregory S.X.E. Jefferis

List of authors in order

Landing page

https://doi.org/10.7554/elife.44590.026

Publisher landing page

Open access

Yes

Whether a free full text is available

OA status

gold

Open access status per OpenAlex

OA URL

https://doi.org/10.7554/elife.44590.026

Direct OA link when available

Concepts

Mushroom bodies, Sensory system, Neuroscience, Antennal lobe, Olfactory system, Population, Stimulus (psychology), Perception, Neural coding, Biology, Odor, Communication, Psychology, Cognitive psychology, Drosophila melanogaster, Gene, Genetics, Sociology, Demography

Top concepts (fields/topics) attached by OpenAlex

Cited by

4

Total citation count in OpenAlex

Citations by year (recent)

2021: 1, 2020: 3

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload