Some neural circuits are responsible for only one specialized function. Examples include the exquisitely tuned delay lines that barn owls use to locate sounds (Carr and Konishi, 1988) and the motor neurons that control the honeybee’s stinger (Ogawa et al., 1995). In the early days of neuroscience, those circuits received most attention, likely because they are in important ways more tractable. In most animals’ central nervous systems, however, many circuits respond to stimuli of multiple sensory modalities or control more than one behavior (Briggman and Kristan, 2008). Increasingly, even sensory brain areas once considered unimodal are found to be modulated by or to directly respond to other sensory modalities, or to be modulated by behavioral state. For instance, auditory stimuli can modulate activity in the human visual cortex (Plass et al., 2019) (in sighted as well as in blind subjects [Bedny et al., 2015]). Likewise, motor cortex activity can be modulated not just by visual presentation of images of relevant body parts, but even in a working memory task in the absence of immediate stimuli (Galvez-Pol et al., 2018). All these forms of multifunctionality likely contribute to the versatility of large brains like ours and may help smaller animals use their more constrained neural hardware more efficiently (Briggman and Kristan, 2008).

A particularly interesting situation occurs when multifunctional circuits converge onto a single motor neuron which plays distinct roles in different behaviors (Miroschnikow et al., 2018; Zarin et al., 2019). What functional and anatomical aspects of the converging pathways permit the output neuron to play its various roles? Control of locomotion in the medicinal leech (Hirudo verbana) is a prime example of a system organized in this manner. In each of the animal’s segmental ganglia, partially overlapping circuits generate the rhythms for swimming and crawling as well as the dynamics of an escape behavior known as the local bend (Kristan et al., 2005; Briggman et al., 2005; Tomina and Wagenaar, 2017). These circuits have a common output in a motor neuron called DE-3 that excites the animal’s dorsal longitudinal muscles. DE-3 plays markedly different dynamical roles in each of the behaviors: In swimming, it oscillates at a relatively rapid 2 Hz in antiphase to its Ventral Excitor counterpart (VE-4) and controls the local dorsal flexion phase of the swim rhythm. In crawling, it is active in phase with VE-4 and controls the contraction phase of the crawl rhythm. In ‘local bending’ it operates in tight concert with its contralateral homolog as well as VE-4 to precisely deform the local body wall away from a mechanical stimulus.

To study the mechanisms underlying the versatility of multifunctional circuits, one would ideally like to record from every single neuron in a nervous system during all of the behaviors the animal can execute, and then reconstruct the anatomical connections between those neurons. Once a mere dream, this is rapidly becoming practicable: Activity imaging using calcium dyes has advanced to the point where simultaneous recordings from the vast majority of individual neurons in smaller species can be accomplished, for instance in larval zebrafish (Ahrens et al., 2013). This technique has even been applied to behaving animals (Jiao et al., 2018). Likewise, anatomical imaging using electron microscopy (EM) has advanced to the point that brains as large as that of the fruit fly Drosophila melanogaster can be imaged—and substantial fractions of their circuitry reconstructed—at a synaptic resolution (Zheng et al., 2018; Bates et al., 2020; Scheffer et al., 2020; Maniates-Selvin et al., 2020). Even small vertebrate brains like that of larval zebrafish are yielding to this approach (Hildebrand et al., 2017).

To bring form and function together, an anatomical atlas based on electron microscopy can be combined with functional studies that use the split-GAL4 system to target individual neurons and identify them with neurons in the anatomical map (Eichler et al., 2017; Eschbach et al., 2020). This is an immensely powerful approach because a single (costly and time consuming) EM run can be used with an unlimited number of functional studies. However, cross-identification of neurons between EM and functional imaging in this approach is ultimately limited by the fact that all nervous systems, even the simplest ones, exhibit considerable variability in their connectomes (Bhattacharya et al., 2019). Accordingly, robustly linking function to connectivity requires a combined anatomical and functional assessment within the same animal (Bargmann and Marder, 2013). Nearly a decade ago, back-to-back papers in Nature described two-photon calcium imaging in the mouse visual cortex followed by serial transmission electron microscopy of a small volume of that same cortex (Bock et al., 2011) and two-photon calcium imaging in the mouse retina followed by serial block face electron microscopy of that same retina (Briggman et al., 2011). This approach, ‘Correlated Light and Electron Microscopy’ or CLEM (de Boer et al., 2015), has since allowed the quantification of synaptic connections between functionally identified neurons that control eye movement in the larval zebrafish (Vishwanathan et al., 2017), a reconstruction of the connections in the olfactory bulb of another larval zebrafish in which neural responses to eight different odors had been recorded prior to sectioning (Wanner and Friedrich, 2020), and a critical re-assessment of the mechanisms of selectivity to particular sensory features in neurons of the mammalian visual cortex (Scholl et al., 2019).

For the present study of multifunctional behavioral circuits, we chose to use a voltage-sensitive dye (VSD) (Grinvald and Hildesheim, 2004; Kulkarni and Miller, 2017) rather than a calcium dye because it can capture faster temporal dynamics, potentially even recording individual action potentials, as well as both excitatory and inhibitory synaptic potentials. We used a fast, high-sensitivity VSD (Woodford et al., 2015) to record the neuronal activity in a segmental ganglion of the medicinal leech Hirudo verbana while its nervous system expressed several (fictive) behaviors. Then, to map the circuits underlying those behaviors with synapse-level resolution, we re-imaged the same individual ganglion using serial blockface electron microscopy (SBEM) (Lippens et al., 2019). Finally, we traced the full arborization of the DE-3 motor neuron and the relevant parts of all its presynaptic partners, to enable comparison between functional and anatomical features of the circuit. Figure 1 illustrates the overall approach. To our knowledge, this is the first time VSDs and SBEM have been combined at this scale.

Figure 1

Download asset Open asset

We combined voltage-sensitive dye imaging with serial blockface electron microscopy to obtain a comprehensive recording of neuronal activity at the scale of an entire leech ganglion along with a full record of the anatomy of that same ganglion.

The use of a fast and sensitive VSD (Woodford et al., 2015) allowed us to record even small membrane potential fluctuations from somata. Those signals included both subthreshold excitatory and inhibitory synaptic potentials that could not have been detected by calcium imaging. Conversely, SBEM (Denk and Horstmann, 2004) allowed us to image the entire neuropil with sufficient resolution to visualize even thin neurites through much of the volume.

Within this vast dataset, we have focused on a neuron that plays three very distinct dynamic roles in different behaviors: motor neuron DE-3^R, a main excitatory motor neuron of the dorsal longitudinal muscle, as well as its presynaptic partners (Stent et al., 1978). We reconstructed its arborization and traced its presynaptic partners back to their somata (up to the limits enabled by our SEM images, see Figure 6 and supplements). Thus, we generated a detailed map of the output stage of a multifunctional circuit that controls many of the animal’s main forms of gross body movement, including swimming, crawling, and the local bend escape.

The reconstructed morphology of DE-3^R was in accordance with previous light microscopic studies in adult (Norris and Calabrese, 1987; Fan et al., 2005) and electron microscopy in a juvenile ganglion (Pipkin et al., 2016): its primary neurite emerged from the soma laterally and traveled toward the ipsilateral roots before making a 180° turn to run laterally across the ganglion. Our tracing revealed 531 synapses onto DE-3^R. This was in the same range as the numbers previously reported for a juvenile ganglion (437 and 650) (Pipkin et al., 2016). Unlike Pipkin et al., 2016, we did not find any output synapses from DE-3^R onto other cells in the ganglion. Several of the input synapses we found were from presynaptic partners that had previously been reported (Ort et al., 1974; Wagenaar, 2017) based on paired electrophysiological recordings. We focused here on chemical synaptic connections, since SBEM does not yet allow for the identification of gap junctions. However, electrical synapses between DE-3^R and their contralateral homologs as well as some interneurons do exist (Fan et al., 2005). Work on molecular markers of electrical synapses that do not require expression of a particular connexin (Shu et al., 2011) may overcome this hurdle in the foreseeable future.

In previous reports, 21 neurons (including bilateral homologues) were found to be monosynaptically connected to the DE-3^R motor neuron (Figure 12). Of those, four (bilateral DI-1, 107, and L) were matched to the presynaptic partners seen in the VSD imaging and identified with sufficient confidence. The other neurons previously reported were not confidently matched to presynaptic partners found here. Six other new partners were confidently identified (VI-2, 8, 101, 102, 109, and 117), and a further 41 synaptic partners of DE-3^R could not be confidently identified with any previously described neuron, mainly because our limited tracing did not allow a morphological identity match. With additional tracing, many of those remaining neurons could possibly be assigned to other previously reported presynaptic neurons shown in Figure 12.

Figure 12

Download asset Open asset

The activity patterns observed during all three fictive behaviors exhibited by the single leech in this study (Figures 2, 3 and 4) matched observations from a larger group of animals reported before (Tomina and Wagenaar, 2017): The majority of neurons in the ganglion were phasically active during multiple behaviors, indicating substantial but not complete overlap between the circuits that govern the behaviors (Briggman and Kristan, 2006). There is considerable variation in the coherence values of specific cells between animals, but the data here fall within the range of values seen before. We conclude that the leech used in this study was representative of its species.

Overlaying the anatomical and activity images (Figure 5) and comparing with the established canonical maps of the ganglion (Wagenaar, 2017) allowed us to address form–function relationships. For instance, our own data allowed us to ask whether neurons that are more strongly associated with a particular behavior than with others formed synapses onto the output motor neuron that were spatially localized. Such an arrangement would indicate a modular organization of the motor neuron’s processes (London and Häusser, 2005). However, we found no evidence of such organization in DE-3^R (Figure 10c,d) other than a slight trend that swim-associated synapses tended to be located closer to the primary neurite (Figure 10—figure supplement 1b). Since most presynaptic partners are multifunctional, the pre-motor network undoubtedly plays an important role in generating distinct motor patterns.

We also asked whether synapses with a particular valence (excitatory or inhibitory) were differently distributed along the tree, which would likewise have implications for possible models of computation in the cell (Saha and Truccolo, 2019). Thanks to an extensive body of previously published electrophysiological recordings from the leech ganglion, the valence of many of DE-3^R’s presynaptic partners is known, allowing us to visualize the spatial distributions of excitatory and inhibitory synapses separately (Figure 10a). However, we found no obvious differences between the two classes of synapses (Figure 10b, Figure 10—figure supplement 1a).

The lack of macroscopic organization of functionally related synapses does not imply that the spatial arrangement of synapses on the tree has no functional relevance. Indeed, the limited experimental evidence currently available suggests that even clusters of two synapses are functionally relevant in the mammalian cortex (Takahashi et al., 2012; Fu et al., 2012). Findings from computational (Poirazi and Mel, 2001) and in vitro studies (Losonczy and Magee, 2006; Nevian et al., 2007) suggest that individual dendritic branches act as integrative compartments and that spatial synaptic clusters facilitate nonlinear dendritic integration. These results imply that connectivity has to be specific with respect to individual dendritic branches. Such a precision could arise through spontaneous neural activity where synaptic clusters are established by a branch-specific 'fire-together-wire-together' rule (Kleindienst et al., 2011).

We therefore applied a spatial clustering algorithm to the input synapses on DE-3^R and asked whether the clusters identified by that algorithm also stand out as functionally significant groupings. Indeed, we found a set of parameter values for which synapses within spatial clusters also clustered in the phase space of the swim rhythm and a different set of parameters for which synapses within spatial clusters also clustered in the phase space of the crawl rhythm (Figure 11b,c). Results were highly consistent between trials of the same behavior. (The results of clustering were not consistent among local bending trials. This may stem from a nature of this reflex behavior: the responsiveness of the population of neurons involving pressure sensation is variable, reflecting the plasticity of the circuit (Crisp and Burrell, 2009).)

Earlier work relied on light microscopy and could therefore not identify presynaptic partners. In contrast, our use of SBEM allowed us to assess such connection specificity based on individually identified synapses. Our findings confirm previous observations on the synchrony of proximal synapses (Kleindienst et al., 2011; Takahashi et al., 2012). An attractive interpretation of our results is that the clusters are the loci where inputs from synchronized presynaptic cell assemblies are integrated (Briggman and Kristan, 2008). In agreement with earlier reports (Kleindienst et al., 2011), the strongest correspondence between spatial clusters and activity patterns was observed when synaptic clusters were defined by a maximum distance between nearest neighbors (d_NN) of up to 10 µm.

Computational and in vitro studies indicate that 10–20 synchronized inputs are required to trigger dendritic spikes (Ujfalussy and Makara, 2020). Existing in vivo studies, however, have found synchrony in only up to five neighboring dendritic spines, making it unclear what the impact of the synaptic clusters on the membrane potential of the posysynaptic neuron might be. The smaller number derived from in vivo conditions might stem from the limitations of experimental methods. Indeed, our results indicate that clusters of up to 23 synaptic members can display synchronous activity. Further work is required, however, to elucidate the role of synaptic clusters in vivo or ex vivo on the membrane potential of the postsynaptic neuron under different experimental conditions.

Additional studies complementing our data are also required to more fully understand the mechanisms underlying synaptic integration in motor neurons like DE-3^R. For instance, single-cell voltage imaging (Kuhn and Roome, 2019) and voltage clamp using sharp electrodes on neurites (Laurent, 1993; Takashima et al., 2006) would help determine where in the neurites active propagation of membrane potentials is supported and thus lead to a better understanding of how postsynaptic potentials as well as action potentials propagate through the tree. As it stands, we only have voltage data from the soma. Leech neurons—along with most neurons in the central nervous systems of other annelids, arthopods, and molluscs—have a unipolar morphology with the somata strictly segregated from neuropil (Bullock and Horridge, 1965). With notable exceptions (including sensory neurons in the leech and select insect motorneurons [Hancox and Pitman, 1992]), these somata are thought to play a relatively minor role in the integration of synaptic inputs to action potential outputs (Andjelic and Torre, 2005).

Even so, the completeness of our functional dataset in combination with our anatomical data makes for an attractive basis for simulation studies to arrive at a computational understanding of multifunctional neuronal circuits (Real et al., 2017). Our data may also serve as a large-scale ground truth for EM segmentation algorithms (Plaza and Funke, 2018). This study not only represents an important step in a combined approach to studying multifunctional circuits at the synaptic level, but also lays the groundwork for a comprehensive neuronal mapping of a whole ganglion that semi-autonomously processes local sensory information and controls segmental movement.

The combination of anatomical methods with synaptic resolution and imaging techniques that can record from the entirety of the neurons of a circuit promises an extraordinary opportunity to assess neural computations at the level of circuit dynamics. Functional maps (Alivisatos et al., 2012) from recorded activity combined with anatomical connectomes (Denk et al., 2012; Bargmann and Marder, 2013; Morgan and Lichtman, 2013) are therefore poised to become a powerful tool not only to have a better understanding of complex behaviors, but also to predict the outcomes of new manipulations Devor et al., 2013; Kandel et al., 2013; Wang et al., 2013. This new addition with the annelid species to the developing field would also pave the way for comparative approaches by functional connectomics to study evolutionary aspects of neuronal design (Laurent, 2020).

The easiest way to access the raw electrophysiology and voltage-dye data as well as the tracing results used in this paper is through a series of Python modules that we made available at https://github.com/wagenadl/leechem-public (copy archived at https://archive.softwareheritage.org/swh:1:rev:73eee24e387e11c259a3f3fe0bd4e469048b25e6/). Included in the package is a file called "demo.py" that demonstrates the use of the modules. Table 4 lists the available VSD trials. The aligned EM volume may be accessed through the Neuroglancer instance at https://leechem.caltech.edu or by pointing SBEMViewer to https://leechem.caltech.edu/emdata. The API is documented at https://leechem.caltech.edu/emdata/help.

1. 1. Ashaber MA
   2. Tomina Y
   3. Kassraian P
   4. Bushong EA
   5. Kristan WB
   6. Ellisman MH
   7. Wagenaar DA

   (2020) Github

   ID wagenadl/leechem. Code and data access for Ashaber et al., 2021.

   https://github.com/wagenadl/leechem

1. 1. Ahrens MB
   2. Orger MB
   3. Robson DN
   4. Li JM
   5. Keller PJ

   (2013) Whole-brain functional imaging at cellular resolution using light-sheet microscopy

   Nature Methods 10:413–420.

   https://doi.org/10.1038/nmeth.2434

   • PubMed
   • Google Scholar
2. 1. Alivisatos AP
   2. Chun M
   3. Church GM
   4. Greenspan RJ
   5. Roukes ML
   6. Yuste R

   (2012) The brain activity map project and the challenge of functional connectomics

   Neuron 74:970–974.

   https://doi.org/10.1016/j.neuron.2012.06.006

   • PubMed
   • Google Scholar
3. 1. Andjelic S
   2. Torre V

   (2005) Calcium dynamics and compartmentalization in leech neurons

   Journal of Neurophysiology 94:4430–4440.

   https://doi.org/10.1152/jn.00695.2005

   • PubMed
   • Google Scholar
4. 1. Bargmann CI
   2. Marder E

   (2013) From the connectome to brain function

   Nature Methods 10:483–490.

   https://doi.org/10.1038/nmeth.2451

   • PubMed
   • Google Scholar
5. Preprint

   1. Bates AS
   2. Schlegel P
   3. Roberts RJ
   4. Drummond N
   5. Tamimi IFM
   6. Turnbull R
   7. Zhao X
   8. Marin EC
   9. Popovici PD
   10. Dhawan S
   11. Jamasb A
   12. Javier A
   13. Li F
   14. Rubin GM
   15. Waddell S
   16. Bock DD
   17. Costa M
   18. Jefferis G

   (2020) Complete connectomic reconstruction of olfactory projection neurons in the fly brain

   bioRxiv.

   https://doi.org/10.1101/2020.01.19.911453

   • Google Scholar
6. 1. Bedny M
   2. Richardson H
   3. Saxe R

   (2015) "Visual" Cortex Responds to Spoken Language in Blind Children

   The Journal of Neuroscience 35:11674–11681.

   https://doi.org/10.1523/JNEUROSCI.0634-15.2015

   • PubMed
   • Google Scholar
7. 1. Bhattacharya A
   2. Aghayeva U
   3. Berghoff EG
   4. Hobert O

   (2019) Plasticity of the electrical connectome of C. elegans

   Cell 176:1174–1189.

   https://doi.org/10.1016/j.cell.2018.12.024

   • PubMed
   • Google Scholar
8. 1. Bock DD
   2. Lee WC
   3. Kerlin AM
   4. Andermann ML
   5. Hood G
   6. Wetzel AW
   7. Yurgenson S
   8. Soucy ER
   9. Kim HS
   10. Reid RC

   (2011) Network anatomy and in vivo physiology of visual cortical neurons

   Nature 471:177–182.

   https://doi.org/10.1038/nature09802

   • PubMed
   • Google Scholar
9. 1. Briggman KL
   2. Abarbanel HD
   3. Kristan WB

   (2005) Optical imaging of neuronal populations during decision-making

   Science 307:896–901.

   https://doi.org/10.1126/science.1103736

   • PubMed
   • Google Scholar
10. 1. Briggman KL
    2. Helmstaedter M
    3. Denk W

    (2011) Wiring specificity in the direction-selectivity circuit of the retina

    Nature 471:183–188.

    https://doi.org/10.1038/nature09818

    • PubMed
    • Google Scholar
11. 1. Briggman KL
    2. Kristan WB

    (2006) Imaging dedicated and multifunctional neural circuits generating distinct behaviors

    Journal of Neuroscience 26:10925–10933.

    https://doi.org/10.1523/JNEUROSCI.3265-06.2006

    • PubMed
    • Google Scholar
12. 1. Briggman KL
    2. Kristan WB

    (2008) Multifunctional pattern-generating circuits

    Annual Review of Neuroscience 31:271–294.

    https://doi.org/10.1146/annurev.neuro.31.060407.125552

    • PubMed
    • Google Scholar
13. Book

    1. Bullock TH
    2. Horridge GA

    (1965)

    Structure and function in the nervous systems of invertebrates

    San Francisco: W. H. Freeman.

    • Google Scholar
14. 1. Bushong EA
    2. Johnson DD
    3. Kim KY
    4. Terada M
    5. Hatori M
    6. Peltier ST
    7. Panda S
    8. Merkle A
    9. Ellisman MH

    (2015) X-ray microscopy as an approach to increasing accuracy and efficiency of serial block-face imaging for correlated light and electron microscopy of biological specimens

    Microscopy and Microanalysis 21:231–238.

    https://doi.org/10.1017/S1431927614013579

    • PubMed
    • Google Scholar
15. 1. Carr CE
    2. Konishi M

    (1988) Axonal delay lines for time measurement in the owl's brainstem

    PNAS 85:8311–8315.

    https://doi.org/10.1073/pnas.85.21.8311

    • PubMed
    • Google Scholar
16. Book

    1. Crisp KM
    2. Burrell BD

    (2009) Cellular and behavioral properties of learning in leech and other annelids

    In: Shain D. H, editors. Annelids in Modern Biology. Wiley Blackwell. pp. 133–155.

    https://doi.org/10.1002/9780470455203.ch8

    • Google Scholar
17. 1. de Boer P
    2. Hoogenboom JP
    3. Giepmans BN

    (2015) Correlated light and electron microscopy: ultrastructure lights up!

    Nature Methods 12:503–513.

    https://doi.org/10.1038/nmeth.3400

    • PubMed
    • Google Scholar
18. 1. Deerinck TJ
    2. Shone TM
    3. Bushong EA
    4. Ramachandra R
    5. Peltier ST
    6. Ellisman MH

    (2018) High-performance serial block-face SEM of nonconductive biological samples enabled by focal gas injection-based charge compensation

    Journal of Microscopy 270:142–149.

    https://doi.org/10.1111/jmi.12667

    • PubMed
    • Google Scholar
19. 1. Denk W
    2. Briggman KL
    3. Helmstaedter M

    (2012) Structural neurobiology: missing link to a mechanistic understanding of neural computation

    Nature Reviews Neuroscience 13:351–358.

    https://doi.org/10.1038/nrn3169

    • PubMed
    • Google Scholar
20. 1. Denk W
    2. Horstmann H

    (2004) Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure

    PLOS Biology 2:e329.

    https://doi.org/10.1371/journal.pbio.0020329

    • PubMed
    • Google Scholar
21. 1. Devor A
    2. Bandettini PA
    3. Boas DA
    4. Bower JM
    5. Buxton RB
    6. Cohen LB
    7. Dale AM
    8. Einevoll GT
    9. Fox PT
    10. Franceschini MA
    11. Friston KJ
    12. Fujimoto JG
    13. Geyer MA
    14. Greenberg JH
    15. Halgren E
    16. Hämäläinen MS
    17. Helmchen F
    18. Hyman BT
    19. Jasanoff A
    20. Jernigan TL
    21. Judd LL
    22. Kim SG
    23. Kleinfeld D
    24. Kopell NJ
    25. Kutas M
    26. Kwong KK
    27. Larkum ME
    28. Lo EH
    29. Magistretti PJ
    30. Mandeville JB
    31. Masliah E
    32. Mitra PP
    33. Mobley WC
    34. Moskowitz MA
    35. Nimmerjahn A
    36. Reynolds JH
    37. Rosen BR
    38. Salzberg BM
    39. Schaffer CB
    40. Silva GA
    41. So PT
    42. Spitzer NC
    43. Tootell RB
    44. Van Essen DC
    45. Vanduffel W
    46. Vinogradov SA
    47. Wald LL
    48. Wang LV
    49. Weber B
    50. Yodh AG

    (2013) The challenge of connecting the dots in the b.r.a.i.n

    Neuron 80:270–274.

    https://doi.org/10.1016/j.neuron.2013.09.008

    • PubMed
    • Google Scholar
22. 1. Eichler K
    2. Li F
    3. Litwin-Kumar A
    4. Park Y
    5. Andrade I
    6. Schneider-Mizell CM
    7. Saumweber T
    8. Huser A
    9. Eschbach C
    10. Gerber B
    11. Fetter RD
    12. Truman JW
    13. Priebe CE
    14. Abbott LF
    15. Thum AS
    16. Zlatic M
    17. Cardona A

    (2017) The complete connectome of a learning and memory centre in an insect brain

    Nature 548:175–182.

    https://doi.org/10.1038/nature23455

    • PubMed
    • Google Scholar
23. 1. Eschbach C
    2. Fushiki A
    3. Winding M
    4. Schneider-Mizell CM
    5. Shao M
    6. Arruda R
    7. Eichler K
    8. Valdes-Aleman J
    9. Ohyama T
    10. Thum AS
    11. Gerber B
    12. Fetter RD
    13. Truman JW
    14. Litwin-Kumar A
    15. Cardona A
    16. Zlatic M

    (2020) Recurrent architecture for adaptive regulation of learning in the insect brain

    Nature Neuroscience 23:544–555.

    https://doi.org/10.1038/s41593-020-0607-9

    • PubMed
    • Google Scholar
24. 1. Fan RJ
    2. Marin-Burgin A
    3. French KA
    4. Otto Friesen W

    (2005) A dye mixture (Neurobiotin and Alexa 488) reveals extensive dye-coupling among neurons in leeches; physiology confirms the connections

    Journal of Comparative Physiology A 191:1157–1171.

    https://doi.org/10.1007/s00359-005-0047-8

    • PubMed
    • Google Scholar
25. 1. Fu M
    2. Yu X
    3. Lu J
    4. Zuo Y

    (2012) Repetitive motor learning induces coordinated formation of clustered dendritic spines in vivo

    Nature 483:92–95.

    https://doi.org/10.1038/nature10844

    • PubMed
    • Google Scholar
26. 1. Galvez-Pol A
    2. Forster B
    3. Calvo-Merino B

    (2018) Modulation of motor cortex activity in a visual working memory task of hand images

    Neuropsychologia 117:75–83.

    https://doi.org/10.1016/j.neuropsychologia.2018.05.005

    • PubMed
    • Google Scholar
27. Software

    1. Google

    (2016) Neuroglancer: Webgl-Based Viewer for Volumetric Data, version 0eb227b

    Neuroglancer: Webgl-Based Viewer for Volumetric Data.

    https://github.com/google/neuroglancer
28. 1. Grinvald A
    2. Hildesheim R

    (2004) VSDI: a new era in functional imaging of cortical dynamics

    Nature Reviews Neuroscience 5:874–885.

    https://doi.org/10.1038/nrn1536

    • PubMed
    • Google Scholar
29. 1. Hancox JC
    2. Pitman RM

    (1992)

    A time-dependent excitability change in the soma of an identified insect motoneurone

    The Journal of Experimental Biology 162:251–263.

    • Google Scholar
30. 1. Harris KM
    2. Perry E
    3. Bourne J
    4. Feinberg M
    5. Ostroff L
    6. Hurlburt J

    (2006) Uniform serial sectioning for transmission electron microscopy

    Journal of Neuroscience 26:12101–12103.

    https://doi.org/10.1523/JNEUROSCI.3994-06.2006

    • PubMed
    • Google Scholar
31. 1. Harvey CD
    2. Svoboda K

    (2007) Locally dynamic synaptic learning rules in pyramidal neuron dendrites

    Nature 450:1195–1200.

    https://doi.org/10.1038/nature06416

    • PubMed
    • Google Scholar
32. 1. Hildebrand DGC
    2. Cicconet M
    3. Torres RM
    4. Choi W
    5. Quan TM
    6. Moon J
    7. Wetzel AW
    8. Scott Champion A
    9. Graham BJ
    10. Randlett O
    11. Plummer GS
    12. Portugues R
    13. Bianco IH
    14. Saalfeld S
    15. Baden AD
    16. Lillaney K
    17. Burns R
    18. Vogelstein JT
    19. Schier AF
    20. Lee WA
    21. Jeong WK
    22. Lichtman JW
    23. Engert F

    (2017) Whole-brain serial-section electron microscopy in larval zebrafish

    Nature 545:345–349.

    https://doi.org/10.1038/nature22356

    • PubMed
    • Google Scholar
33. 1. Jiao ZF
    2. Shang CF
    3. Wang YF
    4. Yang Z
    5. Yang C
    6. Li FN
    7. Xie JZ
    8. Pan JW
    9. Fu L
    10. Du JL

    (2018) All-optical imaging and manipulation of whole-brain neuronal activities in behaving larval zebrafish

    Biomedical Optics Express 9:6154–6169.

    https://doi.org/10.1364/BOE.9.006154

    • PubMed
    • Google Scholar
34. 1. Kandel ER
    2. Markram H
    3. Matthews PM
    4. Yuste R
    5. Koch C

    (2013) Neuroscience thinks big (and collaboratively)

    Nature Reviews Neuroscience 14:659–664.

    https://doi.org/10.1038/nrn3578

    • PubMed
    • Google Scholar
35. Software

    1. Kassraian P
    2. Wagenaar DA

    (2021) leechem: Python code for the Leech SBEM Datasets, version swh:1:rev:73eee24e387e11c259a3f3fe0bd4e469048b25e6

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:9c8061831b15f6896d3a735dec50e68fcec1381a;origin=https://github.com/wagenadl/leechem;visit=swh:1:snp:a17f295e2354ceee5c65e59de8c2d4a5b10eac8c;anchor=swh:1:rev:73eee24e387e11c259a3f3fe0bd4e469048b25e6/
36. 1. Kleindienst T
    2. Winnubst J
    3. Roth-Alpermann C
    4. Bonhoeffer T
    5. Lohmann C

    (2011) Activity-dependent clustering of functional synaptic inputs on developing hippocampal dendrites

    Neuron 72:1012–1024.

    https://doi.org/10.1016/j.neuron.2011.10.015

    • PubMed
    • Google Scholar
37. 1. Kristan WB
    2. Calabrese RL
    3. Friesen WO

    (2005) Neuronal control of leech behavior

    Progress in Neurobiology 76:279–327.

    https://doi.org/10.1016/j.pneurobio.2005.09.004

    • PubMed
    • Google Scholar
38. 1. Kuhn B
    2. Roome CJ

    (2019) Primer to voltage imaging with ANNINE dyes and Two-Photon microscopy

    Frontiers in Cellular Neuroscience 13:321.

    https://doi.org/10.3389/fncel.2019.00321

    • PubMed
    • Google Scholar
39. 1. Kulkarni RU
    2. Miller EW

    (2017) Voltage imaging: pitfalls and potential

    Biochemistry 56:5171–5177.

    https://doi.org/10.1021/acs.biochem.7b00490

    • PubMed
    • Google Scholar
40. 1. Laurent G

    (1993) A dendritic gain control mechanism in axonless neurons of the Locust, Schistocerca americana

    The Journal of Physiology 470:45–54.

    https://doi.org/10.1113/jphysiol.1993.sp019846

    • PubMed
    • Google Scholar
41. 1. Laurent G

    (2020) On the value of model diversity in neuroscience

    Nature Reviews Neuroscience 21:395–396.

    https://doi.org/10.1038/s41583-020-0323-1

    • PubMed
    • Google Scholar
42. 1. Lippens S
    2. Kremer A
    3. Borghgraef P
    4. Guérin CJ

    (2019) Serial block face-scanning electron microscopy for volume electron microscopy

    Methods in Cell Biology 152:69–85.

    https://doi.org/10.1016/bs.mcb.2019.04.002

    • PubMed
    • Google Scholar
43. 1. London M
    2. Häusser M

    (2005) Dendritic computation

    Annual Review of Neuroscience 28:503–532.

    https://doi.org/10.1146/annurev.neuro.28.061604.135703

    • PubMed
    • Google Scholar
44. 1. Losonczy A
    2. Magee JC

    (2006) Integrative properties of radial oblique dendrites in hippocampal CA1 pyramidal neurons

    Neuron 50:291–307.

    https://doi.org/10.1016/j.neuron.2006.03.016

    • PubMed
    • Google Scholar
45. 1. Macagno ER

    (1980) Number and distribution of neurons in leech segmental ganglia

    The Journal of Comparative Neurology 190:283–302.

    https://doi.org/10.1002/cne.901900206

    • PubMed
    • Google Scholar
46. Preprint

    1. Maniates-Selvin JT
    2. Hildebrand DGC
    3. Graham BJ
    4. Kuan AT
    5. Thomas LA
    6. Nguyen T
    7. Buhmann J
    8. Azevedo AW
    9. Shanny BL
    10. Funke J
    11. Tuthill JC
    12. Lee W-C

    (2020) Reconstruction of motor control circuits in adult Drosophila using automated transmission electron microscopy

    bioRxiv.

    https://doi.org/10.1101/2020.01.10.902478

    • Google Scholar
47. 1. Miroschnikow A
    2. Schlegel P
    3. Schoofs A
    4. Hueckesfeld S
    5. Li F
    6. Schneider-Mizell CM
    7. Fetter RD
    8. Truman JW
    9. Cardona A
    10. Pankratz MJ

    (2018) Convergence of monosynaptic and polysynaptic sensory paths onto common motor outputs in a Drosophila feeding connectome

    eLife 7:e40247.

    https://doi.org/10.7554/eLife.40247

    • PubMed
    • Google Scholar
48. 1. Morgan JL
    2. Lichtman JW

    (2013) Why not connectomics?

    Nature Methods 10:494–500.

    https://doi.org/10.1038/nmeth.2480

    • PubMed
    • Google Scholar
49. Book

    1. Muller KJ
    2. Nicholls JG
    3. Stent GS

    (1981)

    Neurobiology of the Leech

    Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.

    • Google Scholar
50. 1. Muller KJ
    2. McMahan UJ

    (1976) The shapes of sensory and motor neurones and the distribution of their synapses in ganglia of the leech: a study using intracellular injection of horseradish peroxidase

    Proceedings of the Royal Society of London. Series B, Biological Sciences 194:481–499.

    https://doi.org/10.1098/rspb.1976.0090

    • PubMed
    • Google Scholar
51. 1. Nevian T
    2. Larkum ME
    3. Polsky A
    4. Schiller J

    (2007) Properties of basal dendrites of layer 5 pyramidal neurons: a direct patch-clamp recording study

    Nature Neuroscience 10:206–214.

    https://doi.org/10.1038/nn1826

    • PubMed
    • Google Scholar
52. 1. Norris BJ
    2. Calabrese RL

    (1987) Identification of motor neurons that contain a FMRFamidelike peptide and the effects of FMRFamide on longitudinal muscle in the medicinal leech, Hirudo medicinalis

    The Journal of Comparative Neurology 266:95–111.

    https://doi.org/10.1002/cne.902660108

    • PubMed
    • Google Scholar
53. 1. Ogawa H
    2. Kawakami Z
    3. Yamaguchi T

    (1995)

    Motor pattern of the stinging response in the honeybee Apis mellifera

    The Journal of Experimental Biology 198:39–47.

    • PubMed
    • Google Scholar
54. 1. Ort CA
    2. Kristan WB
    3. Stent GS

    (1974) Neuronal control of swimming in the medicinal leech. 2. identification and connections of motor neurons

    Journal of Comparative Physiology 94:121–154.

    https://doi.org/10.1007/BF00617838

    • Google Scholar
55. 1. Pipkin JE
    2. Bushong EA
    3. Ellisman MH
    4. Kristan WB

    (2016) Patterns and distribution of presynaptic and postsynaptic elements within serial electron microscopic reconstructions of neuronal arbors from the medicinal leech Hirudo verbana

    Journal of Comparative Neurology 524:3677–3695.

    https://doi.org/10.1002/cne.24120

    • PubMed
    • Google Scholar
56. 1. Plass J
    2. Ahn E
    3. Towle VL
    4. Stacey WC
    5. Wasade VS
    6. Tao J
    7. Wu S
    8. Issa NP
    9. Brang D

    (2019) Joint encoding of auditory timing and location in visual cortex

    Journal of Cognitive Neuroscience 31:1002–1017.

    https://doi.org/10.1162/jocn_a_01399

    • PubMed
    • Google Scholar
57. 1. Plaza SM
    2. Funke J

    (2018) Analyzing image segmentation for connectomics

    Frontiers in Neural Circuits 12:102.

    https://doi.org/10.3389/fncir.2018.00102

    • PubMed
    • Google Scholar
58. 1. Poirazi P
    2. Mel BW

    (2001) Impact of active dendrites and structural plasticity on the memory capacity of neural tissue

    Neuron 29:779–796.

    https://doi.org/10.1016/S0896-6273(01)00252-5

    • PubMed
    • Google Scholar
59. 1. Real E
    2. Asari H
    3. Gollisch T
    4. Meister M

    (2017) Neural circuit inference from function to structure

    Current Biology 27:189–198.

    https://doi.org/10.1016/j.cub.2016.11.040

    • PubMed
    • Google Scholar
60. Conference

    1. Saha D
    2. Truccolo W

    (2019) Stochastic point process models for multi-compartment dendritic-tree input-output transformations in spiking neurons

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference. pp. 4395–4399.

    https://doi.org/10.1109/EMBC.2019.8856996

    • Google Scholar
61. 1. Scheffer LK
    2. Xu CS
    3. Januszewski M
    4. Lu Z
    5. Takemura SY
    6. Hayworth KJ
    7. Huang GB
    8. Shinomiya K
    9. Maitlin-Shepard J
    10. Berg S
    11. Clements J
    12. Hubbard PM
    13. Katz WT
    14. Umayam L
    15. Zhao T
    16. Ackerman D
    17. Blakely T
    18. Bogovic J
    19. Dolafi T
    20. Kainmueller D
    21. Kawase T
    22. Khairy KA
    23. Leavitt L
    24. Li PH
    25. Lindsey L
    26. Neubarth N
    27. Olbris DJ
    28. Otsuna H
    29. Trautman ET
    30. Ito M
    31. Bates AS
    32. Goldammer J
    33. Wolff T
    34. Svirskas R
    35. Schlegel P
    36. Neace E
    37. Knecht CJ
    38. Alvarado CX
    39. Bailey DA
    40. Ballinger S
    41. Borycz JA
    42. Canino BS
    43. Cheatham N
    44. Cook M
    45. Dreher M
    46. Duclos O
    47. Eubanks B
    48. Fairbanks K
    49. Finley S
    50. Forknall N
    51. Francis A
    52. Hopkins GP
    53. Joyce EM
    54. Kim S
    55. Kirk NA
    56. Kovalyak J
    57. Lauchie SA
    58. Lohff A
    59. Maldonado C
    60. Manley EA
    61. McLin S
    62. Mooney C
    63. Ndama M
    64. Ogundeyi O
    65. Okeoma N
    66. Ordish C
    67. Padilla N
    68. Patrick CM
    69. Paterson T
    70. Phillips EE
    71. Phillips EM
    72. Rampally N
    73. Ribeiro C
    74. Robertson MK
    75. Rymer JT
    76. Ryan SM
    77. Sammons M
    78. Scott AK
    79. Scott AL
    80. Shinomiya A
    81. Smith C
    82. Smith K
    83. Smith NL
    84. Sobeski MA
    85. Suleiman A
    86. Swift J
    87. Takemura S
    88. Talebi I
    89. Tarnogorska D
    90. Tenshaw E
    91. Tokhi T
    92. Walsh JJ
    93. Yang T
    94. Horne JA
    95. Li F
    96. Parekh R
    97. Rivlin PK
    98. Jayaraman V
    99. Costa M
    100. Jefferis GS
    101. Ito K
    102. Saalfeld S
    103. George R
    104. Meinertzhagen IA
    105. Rubin GM
    106. Hess HF
    107. Jain V
    108. Plaza SM

    (2020) A connectome and analysis of the adult Drosophila central brain

    eLife 9:e57443.

    https://doi.org/10.7554/eLife.57443

    • PubMed
    • Google Scholar
62. Preprint

    1. Scholl B
    2. Thomas C
    3. Ryan MA
    4. Kamasawa N
    5. Fitzpatrick D

    (2019) Cortical neuron response selectivity derives from strength in numbers of synapses

    bioRxiv.

    https://doi.org/10.1101/2019.12.24.887422

    • Google Scholar
63. 1. Shu X
    2. Lev-Ram V
    3. Deerinck TJ
    4. Qi Y
    5. Ramko EB
    6. Davidson MW
    7. Jin Y
    8. Ellisman MH
    9. Tsien RY

    (2011) A genetically encoded tag for correlated light and electron microscopy of intact cells, tissues, and organisms

    PLOS Biology 9:e1001041.

    https://doi.org/10.1371/journal.pbio.1001041

    • PubMed
    • Google Scholar
64. 1. Stent GS
    2. Kristan WB
    3. Friesen WO
    4. Ort CA
    5. Poon M
    6. Calabrese RL

    (1978) Neuronal generation of the leech swimming movement

    Science 200:1348–1357.

    https://doi.org/10.1126/science.663615

    • PubMed
    • Google Scholar
65. 1. Takahashi N
    2. Kitamura K
    3. Matsuo N
    4. Mayford M
    5. Kano M
    6. Matsuki N
    7. Ikegaya Y

    (2012) Locally synchronized synaptic inputs

    Science 335:353–356.

    https://doi.org/10.1126/science.1210362

    • PubMed
    • Google Scholar
66. 1. Takashima A
    2. Hikosaka R
    3. Takahata M

    (2006) Functional significance of passive and active dendritic properties in the synaptic integration by an identified nonspiking interneuron of crayfish

    Journal of Neurophysiology 96:3157–3169.

    https://doi.org/10.1152/jn.00680.2006

    • PubMed
    • Google Scholar
67. 1. Tomina Y
    2. Wagenaar DA

    (2017) A double-sided microscope to realize whole-ganglion imaging of membrane potential in the medicinal leech

    eLife 6:e29839.

    https://doi.org/10.7554/eLife.29839

    • PubMed
    • Google Scholar
68. 1. Tomina Y
    2. Wagenaar DA

    (2018) Dual-sided voltage-sensitive dye imaging of leech ganglia

    Bio-Protocol 8:e2751.

    https://doi.org/10.21769/BioProtoc.2751

    • PubMed
    • Google Scholar
69. 1. Ujfalussy BB
    2. Makara JK

    (2020) Impact of functional synapse clusters on neuronal response selectivity

    Nature Communications 11:1413.

    https://doi.org/10.1038/s41467-020-15147-6

    • PubMed
    • Google Scholar
70. 1. Varga Z
    2. Jia H
    3. Sakmann B
    4. Konnerth A

    (2011) Dendritic coding of multiple sensory inputs in single cortical neurons in vivo

    PNAS 108:15420–15425.

    https://doi.org/10.1073/pnas.1112355108

    • PubMed
    • Google Scholar
71. 1. Vishwanathan A
    2. Daie K
    3. Ramirez AD
    4. Lichtman JW
    5. Aksay ERF
    6. Seung HS

    (2017) Electron microscopic reconstruction of functionally identified cells in a neural integrator

    Current Biology 27:2137–2147.

    https://doi.org/10.1016/j.cub.2017.06.028

    • PubMed
    • Google Scholar
72. 1. Wagenaar DA

    (2015) A classic model animal in the 21st century: recent lessons from the leech nervous system

    Journal of Experimental Biology 218:3353–3359.

    https://doi.org/10.1242/jeb.113860

    • PubMed
    • Google Scholar
73. Website

    1. Wagenaar DA

    (2017) An interactive map of the segmental ganglia of Hirudo verbana

    Accessed February 10, 2020.

    http://wagenaarlab.caltech.edu/ganglion/
74. Software

    1. Wagenaar DA

    (2021a) The SBEMALign Alignment Process, version swh:1:rev:d76dcc55e7dad3e7bca91de24d20d201696a5339

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:b60d1b1f1a0206f63431eb5d9e09292e400844cd;origin=https://github.com/wagenadl/sbemalign;visit=swh:1:snp:bc43bd59c083866b36b74793ae6cd0d52d4c919f;anchor=swh:1:rev:d76dcc55e7dad3e7bca91de24d20d201696a5339/
75. Software

    1. Wagenaar DA

    (2021b) SBEMViewer, version swh:1:rev:8f8d3d2bcae39e165993d9e11ffe173640b940db

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:2d6602e667cb47db004a4267b6735bd005c554f1;origin=https://github.com/wagenadl/sbemviewer;visit=swh:1:snp:841e13d06021a6cfdebb9a73f681aa1ea6651c0e;anchor=swh:1:rev:8f8d3d2bcae39e165993d9e11ffe173640b940db/
76. Software

    1. Wagenaar DA

    (2021c) GVox - Visualization of and tracing in gigavoxel volumes, version swh:1:rev:5e7ccd2273caed49bac3e09ca39de68a0b182fc5

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:8e373277460920f26472735704cedb8a361de7ec;origin=https://github.com/wagenadl/gvox;visit=swh:1:snp:faa45f047be740d1d154afa3a8c6a0ae96ee24bb;anchor=swh:1:rev:5e7ccd2273caed49bac3e09ca39de68a0b182fc5/
77. 1. Walton J

    (1979) Lead asparate, an en bloc contrast stain particularly useful for ultrastructural enzymology

    Journal of Histochemistry & Cytochemistry 27:1337–1342.

    https://doi.org/10.1177/27.10.512319

    • PubMed
    • Google Scholar
78. 1. Wang Z
    2. Chen LM
    3. Négyessy L
    4. Friedman RM
    5. Mishra A
    6. Gore JC
    7. Roe AW

    (2013) The relationship of anatomical and functional connectivity to resting-state connectivity in primate somatosensory cortex

    Neuron 78:1116–1126.

    https://doi.org/10.1016/j.neuron.2013.04.023

    • PubMed
    • Google Scholar
79. 1. Wanner AA
    2. Friedrich RW

    (2020) Whitening of odor representations by the wiring diagram of the olfactory bulb

    Nature Neuroscience 23:433–442.

    https://doi.org/10.1038/s41593-019-0576-z

    • PubMed
    • Google Scholar
80. Conference

    1. Wetzel AW
    2. Bakal J
    3. Dittrich M
    4. Hildebrand DGC
    5. Morgan HL
    6. Lichtman JW

    (2016) Registering large volume serial-section electron microscopy image sets for neural circuit reconstruction using fft signal whitening

    Proc. IEEE Applied Imagery Pattern Recognition Workshop.

    https://doi.org/10.1109/AIPR.2016.8010595

    • Google Scholar
81. 1. Winnubst J
    2. Cheyne JE
    3. Niculescu D
    4. Lohmann C

    (2015) Spontaneous activity drives local synaptic plasticity in vivo

    Neuron 87:399–410.

    https://doi.org/10.1016/j.neuron.2015.06.029

    • PubMed
    • Google Scholar
82. 1. Woodford CR
    2. Frady EP
    3. Smith RS
    4. Morey B
    5. Canzi G
    6. Palida SF
    7. Araneda RC
    8. Kristan WB
    9. Kubiak CP
    10. Miller EW
    11. Tsien RY

    (2015) Improved PeT molecules for optically sensing voltage in neurons

    Journal of the American Chemical Society 137:1817–1824.

    https://doi.org/10.1021/ja510602z

    • PubMed
    • Google Scholar
83. 1. Zarin AA
    2. Mark B
    3. Cardona A
    4. Litwin-Kumar A
    5. Doe CQ

    (2019) A multilayer circuit architecture for the generation of distinct locomotor behaviors in Drosophila

    eLife 8:e51781.

    https://doi.org/10.7554/eLife.51781

    • PubMed
    • Google Scholar
84. 1. Zheng Z
    2. Lauritzen JS
    3. Perlman E
    4. Robinson CG
    5. Nichols M
    6. Milkie D
    7. Torrens O
    8. Price J
    9. Fisher CB
    10. Sharifi N
    11. Calle-Schuler SA
    12. Kmecova L
    13. Ali IJ
    14. Karsh B
    15. Trautman ET
    16. Bogovic JA
    17. Hanslovsky P
    18. Jefferis G
    19. Kazhdan M
    20. Khairy K
    21. Saalfeld S
    22. Fetter RD
    23. Bock DD

    (2018) A complete electron microscopy volume of the brain of adult Drosophila melanogaster

    Cell 174:730–743.

    https://doi.org/10.1016/j.cell.2018.06.019

    • PubMed
    • Google Scholar

Acceptance summary:

In this Research Advances article, the authors elegantly overlay an anatomical wiring diagram onto functional maps for the generation of distinct behaviors. They do this by combining voltage -sensitive dye imaging and EM reconstruction of the circuit to reveal functional synaptic locations onto a single motor neuron in 3 different behaviors. This study will be of great significance for bridging our understanding of neural basis of behaviors at the synaptic, cellular and circuit levels.

Decision letter after peer review:

Thank you for submitting your article "Anatomy and activity patterns in a multifunctional motor neuron and its surrounding circuits" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Ronald Calabrese as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Christel Genoud (Reviewer #1).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

As the editors have judged that your manuscript is of interest, but as described below that additional analyses are required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

In this manuscript, Ashabar et al. use a combination of voltage imaging and electron microscopy to explore the important question of how a specific motor neuron participates in distinct behaviors. The authors first perform voltage imaging of a leech ganglion to capture activity patterns of a large number of neurons during three fictive motor patterns: swimming, crawling, and local bending. Next, they perform x-ray tomography and serial block-face EM in order to reconstruct the inputs onto DE-3, a motor neuron that has different activity patterns during all three behaviors. The authors suggest that inputs correlated with distinct behaviors are widely distributed across the dendritic arbor of DE-3, but that synapses from presynaptic neurons with coherent activity cluster together more frequently than would be expected by chance. The addition of an EM dataset to the previous study on voltage imaging of the leech ganglion certainly constitutes a valuable advance.

Major Concerns:

1) The premise of this study is that mapping subcellular localization of synaptic inputs onto DE-3 may explain the multifunctionality of the DE-3 motor neuron. However, it is clear from the voltage imaging experiments that many of the premotor neurons are also active during multiple behaviors.

a) How do the authors assign neurons to distinct behaviors in Figure 8 when Figures 2-4 show that the same neurons are active across multiple behaviors?

b) If the premotor neurons are also multifunctional (i.e., active during multiple behaviors), then why would one expect there to be differences in subcellular localization on the motor neuron? Given the results from the voltage imaging data, it seems that the pre-motor network should be more important for generating distinct motor patterns, rather than integration of synaptic inputs in the DE-3 motor neuron.

2) The analysis of synapse distributions is confusing. The primary finding of the paper is that the structure of the functional clusters is different for different behaviors, as different clustering parameters gave optimal correspondence between clusters and the neural activity.

a) While useful as a statistical measure, it is not clear how the F-ratio score relates to strength of clustering or coherence of neural activity. The authors should provide additional measures that provide some indication of how much functional clustering would contribute to the activity of the motor neuron.

b) Additionally, it would be useful to see the distribution of how many neurons contribute synapses to each functional cluster. If the clusters arise from individual neurons, or a very small number of neurons, then it seems obvious that they would have a high coherence (with themselves).

c) Was an analysis of # of synapses versus known connectivity strength performed (to the extent this is known)? Or, for example, the strength of coherence with DE-3 versus number of synapses?

3) Another key analysis in the paper centers around measuring the distribution of synapses along the arbor of DE-3 (Figure 8). The authors find no differences between inhibitory/excitatory synaptic distributions nor neurons participating in different behaviors. It is unclear whether using the cell body as the reference point is a good choice for these analyses, given the unipolar nature of the motor neuron. It would be worthwhile to

a) Measure the distribution of synapses along the motor neuron arbor as a function of the distance from the primary neurite, as shunting inhibition could also occur in an arbor-specific manner, or

b) Use a point on the axon as a reference point rather than the soma, as the spike initiation zone is likely nowhere near the soma.

4) Generally an analysis of the known presynaptic connectivity onto DE-3 versus what was found in this study seems to be missing. Including a wiring diagram of the known monosynaptic presynaptic connections onto DE-3 would be extremely helpful along with an indication of which of these known connections were found in the volume and which were not (cells 27, 208, 204/205, etc…)

5) A reconstruction of a juvenile DE-3 was previously reported in Pipkin et al. How do the locations and distributions of synapses in the current volume compare to that reconstruction? At what developmental stage was the leech in this study?

6) The data quality in Pipkin et al. (e.g. the ability to resolve vesicles) seems to be higher than in this volume? Is this correct? Could data quality also be a reason for different synapse counts between the two datasets?

7) Why were 51-35 = 16 neurons not able to be matched to the VSD recording? Did they shift location during EM processing? Regarding the 35-10 = 25 neurons that were not included in the analysis because they could not be matched to known neurons: If the morphology of these 25 neurons had been completely traced, it seems at least possible they could have been matched based on morphology and previous light microscopy fills. It seems like this could have been done without too much effort given the size of leech neurons and the availability of programs to skeletonize neurons relatively rapidly. The number of synapses actually analyzed in Table 1 (~100 or so) is a small fraction of those actually identified onto DE-3 and it seems like such incompleteness could significantly impact the results in Figures 8 and 9.

8) It would be useful to see that the observed clustering on DE-3R is reproducible on DE-3^L. Given that the final number of neurons used in this analysis is small, is this feasible?

9) Is it possible to reconstruct the contralateral DE-3 homolog to examine contact locations with the analyzed DE-3 and whether there is an anatomical signature of an electrical synapse?

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Anatomy and activity patterns in a multifunctional motor neuron and its surrounding circuits" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Ronald Calabrese as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Christel Genoud (Reviewer #1).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

The manuscript is greatly improved after the first round of revisions. However, reviewers point out several places where clarifications and/or revising claims is essential. Please address these points.

Reviewer #1:

All the requests raised by the first review have been carefully considered. The new figures have positively improved the clarity of the results. The added analysis have strengthened the results and conclusions.

Reviewer #2:

The revisions made some of the analyses and conclusions more clear. We have four remaining concerns, intended to help explain and clarify the approach for a general scientific audience.

1) The revised Materials and methods section clarifies the process used to classify neurons as belonging to distinct behaviors. This does not, however, justify that it is appropriate to classify neurons in this manner. It appears as if most neurons that have been classified as participating in one behavior exhibit similar levels of activity during other behaviors. This is also clear from Figure 4J in the previous (original) paper. The authors should provide some justification for assigning neurons to specific behaviors. Wouldn't the tree analysis still assign neurons to single behaviors even if all neurons participated in all behaviors? Would the assignment be repeatable across individual animals?

2) In response to our previous comment about premotor neurons being multifunctional (ie, participating in multiple behaviors), the authors added a statement that "DE-3 may be a relatively passive integrator of its inputs." I find this confusing. The synapse clustering analysis seems to suggest that DE-3 is not a passive integrator of inputs. If it were, would one not expect there to be no clustering at all?

3) In response to our concern about the interpretability of the F-Ratio, the authors provide some additional description in the Results and Materials and methods. This is helpful, but it would still be useful to provide some context for readers unfamiliar with this ratio. This context could be provided with some examples of F-ratio values for highly clustered or randomly distributed synapses.

4) It would be helpful to compare the coherence analyses in this manuscript with that from the previous (original) paper. They do not always appear similar across the two manuscripts. This should be discussed, especially given the data quality concerns between this paper and the previous one.

Reviewer #3:

The authors addressed most of the issues pointed out in the initial review adequately. There are a few additional points I had:

1) I still cannot see the dataset at: https://leechem.caltech.edu/. It leads to a 502 Gateway Error.

2) I think the correlation analysis of synapse number vs coherence should be added as a supplement to the paper, despite it yielding a statistically insignificant correlation.

3) This sentence in the Discussion should be corrected: “We reconstructed its entire arborization and traced all of its presynaptic partners back to their somata.” This is not true based on the number of untraceable synapses, right? The Abstract contains a similar claim.

4) Including details of what limited the data quality would help future experimenters decide on imaging parameters. Perhaps adding a “Comment on data quality” to the Materials and methods section?

Major concerns

1) The premise of this study is that mapping subcellular localization of synaptic inputs onto DE-3 may explain the multifunctionality of the DE-3 motor neuron. However, it is clear from the voltage imaging experiments that many of the premotor neurons are also active during multiple behaviors.

a) How do the authors assign neurons to distinct behaviors in Figure 8 when Figures 2-4 show that the same neurons are active across multiple behaviors?

While it is true that many neurons are active across multiple behaviors, their degree of activity is not uniform across behaviors. We have added a Materials and methods subsection “Tree analysis” to explain the assignment.

b) If the premotor neurons are also multifunctional (i.e., active during multiple behaviors), then why would one expect there to be differences in subcellular localization on the motor neuron? Given the results from the voltage imaging data, it seems that the pre-motor network should be more important for generating distinct motor patterns, rather than integration of synaptic inputs in the DE-3 motor neuron.

We added a sentence to the Discussion to address this point: “Since most presynaptic partners are multifunctional, the pre-motor network undoubtedly plays an important role in generating distinct motor patterns. Our finding suggests that DE-3 may be a relatively passive integrator of its inputs.”

2) The analysis of synapse distributions is confusing. The primary finding of the paper is that the structure of the functional clusters is different for different behaviors, as different clustering parameters gave optimal correspondence between clusters and the neural activity.

a) While useful as a statistical measure, it is not clear how the F-ratio score relates to strength of clustering or coherence of neural activity.

The F-ratio does not directly relate to the strength of clustering, or to the coherence. Rather, the F-ratio captures the degree to which neurons within a spatial cluster also grouped together in the coherence plot for a given behavior. In other words, it quantifies the link between spatial clustering and functional relatedness. We have added text in Results and Materials and methods to clarify this, including a new Box that outlines the procedure.

The authors should provide additional measures that provide some indication of how much functional clustering would contribute to the activity of the motor neuron.

The role of synaptic clusters in vivo or ex vivo is a topic of ongoing research. In particular, how the background activity of in vivo conditions affects integrative dendritic properties is not yet fully understood. To answer the reviewer’s request would require a detailed biophysical model, which we deem out of the scope of the current work. Such a model was constructed in a recent study of mouse CA1 pyramidal cells, in which the authors benchmarked characteristics of Na⁺ spikes, the amplitude distribution of synaptic potentials, the integration of NMDA receptors, voltage thresholds, and the functional role of K⁺ channels as predicted by their model against the relevant values derived from in vitro conditions (Ujfalussy and Makara, 2020).

Our aim was rather to address another fundamental aspect underlying the functional role of synaptic clusters: If individual dendritic branches are to serve as single computational compartments, connectivity has to be not only specific between neurons, but precise with respect to individual dendritic branches. This precision could arise for instance through spontaneous neural activity or NMDA receptor signaling, such that synaptic clusters displaying synchronous activity are established through a branch-specific “fire-together-wire-together” rule (Kleindienst et al., 2011).

To assess this level of connection precision, it is fundamental to differentiate between clusters of synapses from the same presynaptic neuron vs. those from different presynaptic neurons, a feat which requires the precision of Electron Microscopy, and which could not been achieved in previous studies of synaptic clustering based on light microscopy (Druckmann et al., 2014).

Our finding that spatially clustered synapses display synchronized activity confirms observations on synchronous proximal synapses from previous studies (Kleindienst et al., 2011, Takahashi et al., 2012). Our analysis also serves as a “proof of concept”, illustrating what can be achieved by combining functional recording and EM reconstruction, which we hope will become standard in future connectomics studies.

We have added additional text in the Discussion section that seeks to convey the above ideas in a slightly more concise form.

b) Additionally, it would be useful to see the distribution of how many neurons contribute synapses to each functional cluster. If the clusters arise from individual neurons, or a very small number of neurons, then it seems obvious that they would have a high coherence (with themselves).

The F-ratio, like the F-statistic used in ANOVA, is actually immune to the size of the clusters.

c) Was an analysis of # of synapses versus known connectivity strength performed (to the extent this is known)? Or, for example, the strength of coherence with DE-3 versus number of synapses?

We have extensively analyzed the correlation between the number of synapses and the magnitude of the coherence of presynaptic partners with DE3-R during behaviors. However, even though almost all correlation coefficients were positive, none of them were significantly positive.

In particular, we calculated the Pearson as well as the rank-based Spearman correlation coefficients for swimming, crawling, and local bending. We found no statistically significant relationship between the number of synapses and the magnitude of the coherence of a presynaptic partner with DE3-R for any of the trials (all p > 0.05). As an example, we present here the results for swim trial 1 in Author response image 1. Results for all trials are summarized in Author response table 1.

Author response image 1

Download asset Open asset

In a second analysis, we weighted each synapse inversely proportional to its distance to the soma of DE-3^R. Again, none of the trials resulted in statistically significant correlation coefficients. As an example, we again present a visualization for swim trial 1 and a summary for all trials in Author response image 2 and Author response table 2.

Author response image 2

Download asset Open asset

In conclusion, no consistent relationship between synapse count and the strength of functional coherence of a presynaptic neuron and DE-3^R was found. This somewhat surprising result could be explained by variability in the strength of individual synapses or by other factors that affect the impact of individual synapses on the activity of the postsynaptic neuron, such as non-linearities associated with synaptic clusters.

3) Another key analysis in the paper centers around measuring the distribution of synapses along the arbor of DE-3 (Figure 8). The authors find no differences between inhibitory/excitatory synaptic distributions nor neurons participating in different behaviors. It is unclear whether using the cell body as the reference point is a good choice for these analyses, given the unipolar nature of the motor neuron. It would be worthwhile to

a) Measure the distribution of synapses along the motor neuron arbor as a function of the distance from the primary neurite, as shunting inhibition could also occur in an arbor-specific manner, or

b) Use a point on the axon as a reference point rather than the soma, as the spike initiation zone is likely nowhere near the soma.

These are excellent suggestions. We have added a supplemental figure with histograms of distances between synapses and the trunk. These did not reveal obvious structure either, which we noted in the Discussion. We created another set of histograms with distance measured from the point where the axon enters the contralateral dorsal posterior nerve, again with no significant results. In the interest of space, we have not included these histograms in the present submission, but we would not object to adding them.

4) Generally an analysis of the known presynaptic connectivity onto DE-3 versus what was found in this study seems to be missing. Including a wiring diagram of the known monosynaptic presynaptic connections onto DE-3 would be extremely helpful along with an indication of which of these known connections were found in the volume and which were not (cells 27, 208, 204/205, etc…)

Adding a wiring diagram is an excellent idea. We created a figure that shows monosynaptic connections onto DE-3 (new Figure 10) and added a paragraph related to the diagram to the Discussion section.

5) A reconstruction of a juvenile DE-3 was previously reported in Pipkin et al. How do the locations and distributions of synapses in the current volume compare to that reconstruction? At what developmental stage was the leech in this study?

The leech used in our study was an adult specimen. We have clarified that important point in both the Results and Materials and methods sections. We have inserted a sentence that explains that, unlike Pipkin et al., we did not find output synapses from DE-3 onto other cells. At a gross level, the input synapses onto DE-3 were distributed similarly in our sample as in the juvenile.

6) The data quality in Pipkin et al. (e.g. the ability to resolve vesicles) seems to be higher than in this volume? Is this correct? Could data quality also be a reason for different synapse counts between the two datasets?

Our image quality is indeed not as high as in Pipkin et al., 2016. There were three reasons for that: (1) To be able to image a much larger volume in a reasonable timespan, we had to accept a shorter per-pixel dwell time. (2) To reduce charging artifacts in the larger volume, we used gas injection (Deerinck et al., 2018) which had a small negative impact on signal to noise. (3) During our acquisition run, the microscope suffered from a hardware defect involving a bad cable. This last issue probably accounted for the majority of the difference. It is certainly possible that image quality had an effect on synapse counts, although it should be noted that our synapse count is right in between the counts Pipkin et al. report for the two DE-3 cells in their dataset.

7) Why were 51-35 = 16 neurons not able to be matched to the VSD recording? Did they shift location during EM processing? Regarding the 35-10 = 25 neurons that were not included in the analysis because they could not be matched to known neurons: If the morphology of these 25 neurons had been completely traced, it seems at least possible they could have been matched based on morphology and previous light microscopy fills. It seems like this could have been done without too much effort given the size of leech neurons and the availability of programs to skeletonize neurons relatively rapidly. The number of synapses actually analyzed in Table 1 (~100 or so) is a small fraction of those actually identified onto DE-3 and it seems like such incompleteness could significantly impact the results in Figures 8 and 9.

The neurons that could not be matched between EM and VSD did indeed shift during sample preparation. It is true that partial matching could have been done based on morphology. Unfortunately, given our image quality, complete tracing of all the neurons would have taken several years of effort. It is absolutely true that the incompleteness could have an impact on results. We are presently preparing for a second EM run, which should yield much higher image quality for future analysis.

8) It would be useful to see that the observed clustering on DE-3R is reproducible on DE-3^L. Given that the final number of neurons used in this analysis is small, is this feasible?

It would be very interesting to trace DE-3^L. Unfortunately, we lack the resources to do so, as tracing DE-3^R with its presynaptic partners took almost two years. Tracing DE-3^L would likely take another year of work.

9) Is it possible to reconstruct the contralateral DE-3 homolog to examine contact locations with the analyzed DE-3 and whether there is an anatomical signature of an electrical synapse?

We partially traced DE-3^L, but only insofar as needed to connect synapses to the soma. Our tracing includes three areas where the two neurons meet each other. Simply by inspecting relevant sections of the EM data, we could not confidently confirm any signature of electrical synapses (Fan et al., 2005). It is possible that future more sophisticated image analysis will change this picture.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Reviewer #2:

The revisions made some of the analyses and conclusions more clear. We have four remaining concerns, intended to help explain and clarify the approach for a general scientific audience.

1) The revised Materials and methods section clarifies the process used to classify neurons as belonging to distinct behaviors. This does not, however, justify that it is appropriate to classify neurons in this manner. It appears as if most neurons that have been classified as participating in one behavior exhibit similar levels of activity during other behaviors. This is also clear from Figure 4J in the previous (original) paper. The authors should provide some justification for assigning neurons to specific behaviors. Wouldn't the tree analysis still assign neurons to single behaviors even if all neurons participated in all behaviors? Would the assignment be repeatable across individual animals?

We added a sentence to the end of the “Tree analysis” Materials and methods subsection to address this concern.

2) In response to our previous comment about premotor neurons being multifunctional (ie, participating in multiple behaviors), the authors added a statement that "DE-3 may be a relatively passive integrator of its inputs." I find this confusing. The synapse clustering analysis seems to suggest that DE-3 is not a passive integrator of inputs. If it were, would one not expect there to be no clustering at all?

We agree that the sentence was confusing and removed it.

3) In response to our concern about the interpretability of the F-Ratio, the authors provide some additional description in the Results and Materials and methods. This is helpful, but it would still be useful to provide some context for readers unfamiliar with this ratio. This context could be provided with some examples of F-ratio values for highly clustered or randomly distributed synapses.

This is a great suggestion. We added a new figure supplement (Figure 11—figure supplement 3) to implement it, along with additional explanation in the Materials and methods section that now includes an explicit equation.

4) It would be helpful to compare the coherence analyses in this manuscript with that from the previous (original) paper. They do not always appear similar across the two manuscripts. This should be discussed, especially given the data quality concerns between this paper and the previous one.

We added a sentence to emphasize the considerable variation that always exists between animals in the coherence values of individual cells. It may also be worth noting that the coherence plots in the original paper were referenced to cells DI-1 (for swimming), AE (for crawling), or the stimulus (for local bending), rather than to DE-3 as in this paper.

Reviewer #3:

The authors addressed most of the issues pointed out in the initial review adequately. There are a few additional points I had:

1) I still cannot see the dataset at: https://leechem.caltech.edu/. It leads to a 502 Gateway Error.

Apologies. When the server rebooted, the software did not automatically restart. We have now fixed this shortcoming.

2) I think the correlation analysis of synapse number vs coherence should be added as a supplement to the paper, despite it yielding a statistically insignificant correlation.

We particularly appreciate this recommendation. Looking at the data again, we realized we had never tested whether on average across trials the correlation coefficients were statistically significant. We have now performed that test and found that, in fact, they were. These results are presented in the new Figure 9 and corresponding text in Results and in Materials and methods.

3) This sentence in the Discussion should be corrected: “We reconstructed its entire arborization and traced all of its presynaptic partners back to their somata.” This is not true based on the number of untraceable synapses, right? The Abstract contains a similar claim.

We removed the words “entire” and “all,” both here and in the Abstract and added a qualifying phrase.

4) Including details of what limited the data quality would help future experimenters decide on imaging parameters. Perhaps adding a “Comment on data quality” to the Materials and methods section?

We have added a figure (Figure 6—figure supplement 2) that estimates true image resolution based on power spectra of image pixel intensities and added text to the Materials and methods section to describe our findings.

Author details

1. Mária Ashaber

   Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, United States

   Contribution

   Formal analysis, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Contributed equally with

   Yusuke Tomina and Pegah Kassraian

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5586-9585

2. Yusuke Tomina

   Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, United States

   Present address

   Faculty of Science and Technology, Keio University, Yokohama, Japan

   Contribution

   Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Contributed equally with

   Mária Ashaber and Pegah Kassraian

   Competing interests

   No competing interests declared

3. Pegah Kassraian

   Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, United States

   Contribution

   Conceptualization, Software, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   Contributed equally with

   Mária Ashaber and Yusuke Tomina

   Competing interests

   No competing interests declared

4. Eric A Bushong

   Division of Biological Sciences, University of California, San Diego, San Diego, United States

   Contribution

   Investigation, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6195-2433

5. William B Kristan

   Division of Biological Sciences, University of California, San Diego, San Diego, United States

   Contribution

   Conceptualization, Funding acquisition, Methodology

   Competing interests

   No competing interests declared

6. Mark H Ellisman

   1. National Center for Microscopy and Imaging Research, University of California, San Diego, San Diego, United States
   2. Department of Neurosciences, UCSD School of Medicine, San Diego, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Project administration

   Competing interests

   No competing interests declared

7. Daniel A Wagenaar

   Division of Biology and Biological Engineering, California Institute of Technology, Pasadena, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   daw@caltech.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6222-761X

• 2,267

  Page views

• 263

  Downloads

• 2

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.