Phase-specific premotor inhibition modulates leech rhythmic motor output

Reviewer #1 (Public review):

The medicinal leech preparation is an amenable system in which to understand how the underlying cellular networks for locomotion function. A previously identified non-spiking neuron (NS) was studied and found to alter the mean firing frequency of a crawl-related motoneuron (DE-3), which fires during the contraction phase of crawling. The data are mostly solid. Identifying upstream neurons responsible for crawl motor patterning is essential for understanding how rhythmic behavior is controlled.

Review of Revision:

Reviewer: On a positive note, the rationale for the study is clearer to me now after reading the authors' responses to both reviewers, but that information, as described in the authors' responses, is minimally incorporated into the current revised paper. Incorporating a discussion of previous work on the NS cell has, indeed, improved the paper.

I suggested earlier that the paper be edited for clarity but not much text has been changed since the first draft. I will provide an example of the types of sentences that are confusing. The title of the paper is: "Phase-specific premotor inhibition modulates leech rhythmic motor output". Are the authors referring to the inhibition created by premotor neurons (e.g., on to the motoneurons) or the inhibition that the premotor neurons receive?

I also find the paper still confusing with regard to the suggested "functional homology" with the vertebrate Renshaw cells. When the authors set up this expectation of homology (should be analogy) in the introduction and other sections of the paper, one would assume that the NS cell would be directly receiving excitation from a motoneuron (like DE-3) and, in turn, the motoneuron would then receive some sort of inhibitory input to regulate its firing frequency. Essentially, I have always viewed the Renshaw cells as nature's clever way to monitor the ongoing activity of a motoneuron while also providing recurrent feedback or "recurrent inhibition" to modify that cell's excitatory state. The authors present their initial idea below on line 62. Authors write: "These neurons are present as bilateral pairs in each segmental ganglion and are functional homologs of the mammalian Renshaw cells (Szczupak, 2014). These spinal cord cells receive excitatory inputs from motoneurons and, in turn, transmit inhibitory signals to the motoneurons (Alvarez and Fyffe, 2007)."

[Reviewer (minor note): I suggest re-writing this last sentence as "these" is confusing. Change to: 'In the spinal cord, Renshaw interneurons receive excitatory inputs from motoneurons and, in turn, transmit inhibitory signals to them (Alvarez and Fyffe, 2007).']

Reviewer: Furthermore, the authors note that (line 69 on): "In the context of this circuit the activity of excitatory motoneurons evokes chemically mediated inhibitory synaptic potentials in NS. Additionally, the NS neurons are electrically coupled......In physiological conditions this coupling favors the transmission of inhibitory signals from NS to motoneurons." Based on what is being conveyed here, I see a disconnect with the "functional homology" being presented earlier. I may be missing something, but the Renshaw analogy seems to be quite different compared to what looks like reciprocal inhibition in the leech. If the authors want to make the analogy to Renshaw cells clearer, then they should make a simple ball and stick diagram of the leech system and visually compare it to the Renshaw/motoneuron circuit with regard to functionality. This simple addition would help many readers.

Reviewer: The Abstract, Authors write (line 19), "Specifically, we analyzed how electrophysiological manipulation of a premotor nonspiking (NS) neuron, that forms a recurrent inhibitory circuit (homologous to vertebrate Renshaw cells)...."
First, a circuit would not be homologous to a cell, and the term homology implies a strict developmental/evolutionary commonality. At best, I would use the term functionally analogous but even then I am still not sure that they are functionally that similar (see comments above). Line 22: "The study included a quantitative analysis of motor units active throughout the fictive crawling cycle that shows that the rhythmic motor output in isolated ganglia mirrors the phase relationships observed in vivo." This sentence must be revised to indicate that not all of the extracellular units were demonstrated to be motor units. Revise to: "The study included a quantitative analysis of identified and putative motor units active throughout the fictive crawling cycle that shows.....'

Line 187 regarding identifying units as motoneurons: Authors write, "While multiple extracellular recordings have been performed previously (Eisenhart et al., 2000), these results (Figure 4) present the first quantitative analysis of motor units activated throughout the crawling cycle in this type of recordings." The authors cannot assume that the units in the recorded nerves belong only to motoneurons. Based on their first rebuttal, the authors seem to be reluctant to accept the idea that the extracellularly recorded units might represent a different class of neurons. They admit that some sensory neurons (with somata located centrally) do, indeed, travel out the same nerves recorded, but go on to explain why they would not be active.

The leech has a variety of sensory organs that are located in the periphery, and some of these sensory neurons do show rhythmic activity correlated with locomotor activity (see Blackshaw's early work). The numerous stretch receptors, in fact, have very large axons that pass through all the nerves recorded in the current paper. In Fig. 4, it is interesting that the waveforms of all the units recorded in the PP nerve exhibit a reversal in waveform as compared to those in the DP nerve, which might indicate (based on bipolar differential recording) that the units in the PP nerve are being propagated in the opposite direction (i.e., are perhaps afferent). Rhythmic presynaptic inhibition and excitation is commonly seen for stretch receptors within the CNS (see the work of Burrows) and many such cells are under modulatory control.

Most likely, the majority of the units are from motoneurons, but we do not really know at this point. The authors should reframe their statements throughout the paper as: 'While multiple extracellular recordings have been performed previously (Eisenhart et al., 2000), these results (Figure 4) present the first quantitative analysis of multiple extracellular units, using spike sorting methods, which are activated throughout the crawling cycle.' In cases where the identity of the unit is known, then it is fine to state that, but when the identity of the unit is not known, then there should be some qualification and stated as 'putative motor units'

Reviewer, the Methods section: needs to include the full parameters that were used to assess whether bursting activity was qualified in ways to be considered crawling activity or not. Typically, crawl-like burst periods of no more than 25 seconds have been the limit for their qualification as crawling activity. In Fig 2F, for example, the inter-burst period is over 35 seconds; that coupled with an average 5 second burst duration would bring the burst period to 40 seconds, which is substantially out of range for there to be bursting relevant to crawl activity. Simply put, long DE-3 burst periods are often observed but may not be indicative of a crawl state as the CV motoneurons are no longer out of phase with DE-3. A number of papers have adopted this criterion.

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1 (Public review):

Summary:

The Szczupak lab published a very interesting paper in 2012 (Rodriquez et al. J Neurophysiol 107:1917-1924) on the effects of the segmentally-distributed non-spiking (NS) cell on crawl-related motoneurons. As far as I can tell, the working model presented in 2012, for how the non-spiking (NS) cell impacts the crawling motor pattern, is the same functional model presented in this new paper. Unfortunately, the Discussion does not address any of the findings in the previous paper or cite them in the context of NS alterations of fictive crawling. Aside from different-looking figures and some new analyses, the results and conclusions are the same.

Reviewers #1 and #2 called our attention to our failure to cite the Rodriguez et al. 2012 article in the context of the main goal of the present work. We do now explain how the present study is framed by the published work. See lines 74-79.

In Rodriguez et al. 2012, we hypothesized that the inhibitory signals onto NS were originated in the motoneuron firing. We now cite this reference in line 104. In the current manuscript we further investigated the connection between the inhibitory signals onto NS and the motoneuron activity (Figure 2) and proved that the hypothesis was wrong. Thus, the model presented here differs from the one proposed in Rodriguez et al. 2012.

In Rodriguez et al. 2012, we speculated that the inhibitory signals received by NS were transmitted to the motoneurons, but an important control was missing in that study. In the current study depolarization of NS during crawling is tested against a control series that allows to properly examine the hypothesis (lines 138-147). But, most important, because NS is so widely connected with the layer of motoneurons it was necessary to test the effect on other motoneurons during the fictive crawling cycle. We now explain this rationale in lines 249-257.

Strengths:

The figures are well illustrated.

Weaknesses:

The paper is a mix of what appears to be two different studies and abruptly switches gears to examine how closely the crawl patterning is in the intact animal as compared to the fictive crawl patterning in the intact animal. Unfortunately, previous studies in other labs are not cited even though identical results have been obtained and similar conclusions were made. Thus, the novelty of the results is missing for those who are familiar with the leech preparation. The lack of appropriate citations and discussion of previous studies also deprives the scientific community of fully comprehending the impact of the data presented and the science it was built upon.

The main aim of the manuscript is to learn the role of premotor NS neurons in the crawling motor pattern studied using spike sorting in extracellular nerve recordings. This readout allows to simultaneously monitor a larger number of units than in any previous study. This approach aims to determine whether and how a recurrent inhibitory peripheral circuit is involved in coordinating or modulating the rhythmic motor pattern.

Our rationale was that the known effect of NS on one particular motoneuron (DE-3) may have overlooked a more general effect on crawling (lines 253-257). Moreover, we wanted to investigate whether this effect was due to the recurrent inhibitory circuit or if other elements were involved, and to study whether the modulation was mediated by the recurrent synapse between NS and the motoneurons.

In the context of this aim we studied the rhythmic activity of cell DE-3, together with motoneurons that fire in-phase and anti-phase, in isolated ganglia (Figure 4). To reveal the effect of NS manipulation we applied a quantitative analysis that showed the phase-specific effect of NS (Figure 6).

Given that this is the first study using a spike sorting algorithm to detect and describe the activity of motoneurons in nerve recordings we found it reasonable to compare these results with an in vivo study; thus, providing information to the general reader, that supports the correspondence between the ex vivo and the in vivo patterns.

(1) Results, Lines 167-170: "While multiple extracellular recordings have been performed previously (Eisenhart et al., 2000), these results present the first quantitative analysis of motor units activated throughout the crawling cycle. The In-Phase units are expected to control the contraction stage by exciting or inhibiting the longitudinal or circular muscles, respectively, and the Anti-Phase units to control the elongation stage by exciting or inhibiting the circular or longitudinal muscles, respectively."

Reviewer: The first line above is misleading. The study by Puhl and Mesce (2008, J. Neurosci, 28:4192- 420) contains a comprehensive analysis of the motoneurons active during fictive crawling with the aim of characterizing their roles and phase relationships and solidifying the idea that the oscillator for crawling resides in a single ganglion. Intracellular recordings from a number of key crawl-related motoneurons were made in combination with extracellular recordings of motoneuron DE-3, a key monitor of crawling. In their paper, it was shown that motoneurons AE, VE-4, DI-1, VI-2, and CV were all correlated with crawl activity, and fired repeatedly either in phase or out-of-phase with DE-3. They were shown to be either excitatory or inhibitory. At a minimum, the above paper should be cited.

The sentence in the submitted manuscript explicitly refers to the quantitative analysis of extracellular recordings, but we recognize that it may lead to confusion. We have now added a clarification (lines 197-199).

The article by Puhl and Mesce 2008 shows very nice intracellular recordings of the AE, CV, VE-4, DE-3, DI-1, and Vi-2, accompanied by extracellular recordings of DE-3 in the DP nerve. In all cases, there is only one intracellular recording paired with the DP nerve recording.

While it is possible to perform up to 3-4 simultaneous intracellular recordings, these are technically challenging, and more so when the recordings have to last 10-20 minutes. Due to this difficulty, and because our objective was to record multiple units simultaneously in order to comprehensively describe the different crawling stages, we implemented the spike sorting analysis on multiple extracellular recordings. This approach enabled us to reliably obtain multiple units per experiment and thus execute a quantitative analysis of the activity of each identified unit.

The article by Puhl and Mesce 2008 mentions several quantitative aspects of the neurons that fire in-phase or out-of-phase with DE-3, but, as far as we understand, there is no figure that summarizes activity levels and span in the way Figures 4 and 6 do in the current manuscript. To the best of our knowledge, no previous work renders this information.

It is very important for us to emphasize that the work by Puhl and Mesce was seminal for our research. We cited it four times in the original manuscript and 10 times in the present version. But, like any important discovery, it sets the ground for further work that can refine certain measurements that in the original discovery were not central.

This is why we believe that the cited sentence in our manuscript is not misleading. However, to comply with the requirement of Reviewer #1, we added a sentence preceding the mentioned paragraph (lines 185-187) that acknowledges the description made using intracellular recordings, and explains the need for implementing the approach we chose.

The submitted paper would be strengthened if some of these previously identified motoneurons were again recorded with intracellular electrodes and concomitant NS cell stimulation. The power of the leech preparation is that cells can be identified as individuals with dual somatic (intracellular) and axonal recordings (extracellular).

Most of the motoneurons mentioned by Reviewer #1 are located on the opposite side (dorsal) of the ganglion to NS (ventral), and therefore, simultaneous intracellular recordings in the context of fictive crawling are challenging.

In the publication of Rodriguez et al. 2009, Mariano Rodriguez did manage to record NS from the dorsal side together with DE-3 and MN-L (!) and this led to the discovery that these motoneurons are electrically coupled, but the recurrent inhibitory circuit masks this interaction. Repeating this type of experiments during crawling, which requires stable recordings for around 15 minutes, is not a reasonable experimental setting.

Rodriguez et al. 2012 shows intracellular recordings of motoneurons AE and CV during crawling in conjunction with NS, and their activity presented the expected correlation.

The shortfall of this aspect of the study (Figure 5) is that the extracellular units have not been identified here.

The Reviewer is right in that the extracellular units have not been identified in terms of cell identity. As we explained earlier, most motoneurons are on the opposite side (ventral/dorsal) of the ganglion relative to NS.

However, we do characterize the units in terms of the nerve through which they project to the periphery and their activity phase. In lines 345-349 we use this information and, based on published work, we propose possible cellular identities of the different units.

In xfact, these units might not even be motoneurons.

We are surprised by this comment. The classical work of Ort and collaborators (1974) showed that spikes detected in extracellular nerve recordings were emitted by specific motoneurons, and several previous publications have validated extracellular nerve recordings as a means to study fictive motor patterns (Wittenberg & Kristan 1992, Shaw & Kristan 1997, Eisenhart et al. 2000).

For further reassurance, we only took in consideration units whose activity was locked to DE3; any non-rhythmical activity was filtered out (see lines 433-435).

They could represent activity from the centrally located sensory neurons, dopamine-modulated afferent neurons or peripherally projecting modulatory neurons.

Peripheral nerves also contain axons from sensory neurons. However, in a previous article, we studied the activity of mechanosensory neurons (Alonso et al. 2020) and showed that they remain silent during crawling. Moreover, the low-threshold T sensory neurons are inhibited in phase with DE-3 bursts and NS IPSPs (Kearney et al. 2022). Alonso et al. 2000 showed that spiking activity of T cells affects the crawling motor pattern, revealing the relevance of keeping them silent.

What does the Reviewer mean by “dopamine-modulated afferents”? We are not aware of this category of leech neurons.

The neuromodulatory Rz neurons project peripherally through the recorded nerves, but intracellular recordings of these neurons from our lab show no rhythmic activity in those cells during dopamine-induced crawling.

Essentially, they may not have much to do with the crawl motor pattern at all.

Does the Reviewer consider that neurons engaged in a coherent rhythmic firing could be unrelated to the pattern? As indicated above, the units reported in our manuscript were selected because dopamine evoked their rhythmic activity, locked to DE-3.

Does the Reviewer consider that dopamine could evoke spurious neuronal activity?

(2) Results Lines 206-210: "with the elongation and contraction stages of in vivo behavior. However the isometric stages displayed in vivo have no obvious counterpart in the electrophysiological recordings. It is important to consider that the rhythmic movement of successive segments along the antero-posterior axis of the animal requires a delay signal that allows the appropriate propagation of the metachronal wave, and this signal is probably absent in the isolated ganglion."

Reviewer: The so-called isometric stages, indeed, have an electrophysiological counterpart due in part to the overlapping activities across segments. This submitted paper would be considerably strengthened if it referred to the body of work that has examined how the individual crawl oscillators operate in a fully intact nerve cord, excised from the body but with all the ganglia (and cephalic ganglion) attached. Puhl and Mesce 2010 (J. Neurosci 30: 2373-2383) and Puhl et al. 2012 (J. Neurosci, 32:17646 -17657) have shown that "appropriate propagation of the metachronal wave" requires the brain, especially cell R3b-1. They also show that the long-distance projecting cell R3b-1 synapses with the CV motoneuron, providing rhythmic excitatory input to it.

We would like to draw the Reviewer’s attention to the fact that Puhl and Mesce 2008, 2010 and Puhl et al. 2012 characterized crawling in intact (or nearly intact) animals considering the whole body. In our in vivo analysis, we studied the changes in length of the whole animal and of sections demarcated by the drawn points, as described in the Materials and Methods/Behavioral

Experiments. Because of this different analysis, we defined “isometric” stages as those in which a given section of the animal does not change its length. We now clarify this (line 230).

In the paragraph cited by the Reviewer, we intended to state that, in the context of our study, the intersegmental lag caused by the coordinating mechanisms has no counterpart “in the electrophysiological recordings of motoneurons in the isolated ganglia”. We have now completed this idea with the expression underlined in the previous sentence (line 231).

As the Reviewer indicates, in the intact nerve cord the behavioral isometric stages correspond to the “waiting time” between segments. We did refer to the metachronal order but did not cite the articles by Puhl and Mesce 2010 and Puhl et al. 2012; we now do so (lines 234).

For this and other reasons, the paper would be much more informative and exciting if the impacts of the NS cell were studied in a fully intact nerve cord. Those studies have never been done, and it would be exciting to see how and if the effects of NS cell manipulation deviated from those in the single ganglion.

The Reviewer may consider that a systematic analysis of multiple nerves in several ganglia along the whole nerve cord would have been a different enterprise than the one we carried out. The Reviewer is right in recognizing the interest of such study, but in our opinion, the value of the present work lies in presenting a thorough quantitative analysis of multiple nerves to demonstrate its usefulness for the study of the network underlying leech crawling. In this manuscript, we used it to analyze the role of the premotor NS neuron. Without the recording of units firing in-phase and out-ofphase with DE-3, we would have been unable to assess the span of NS effects.

(3) Discussion Lines 322-324. "The absence of descending brain signals and/or peripheral signals are assumed as important factors in determining the cycle period and the sequence at which the different behavioral stages take place."

Reviewer: The authors could strengthen their paper by including a more complete picture of what is known about the control of crawling. For example, Puhl et al. 2012 (J Neurosci, 32:17646-17657) demonstrated that the descending brain neuron R3b-1 plays a major role in establishing the crawlcycle frequency. With increased R3b-1 cell stimulation, DE-3 periods substantially shortened throughout the entire nerve cord. Thus, the importance of descending brain inputs should not be merely assumed; empirical evidence exists.

We now strengthen the concept using “known descending brain signals” (line 358) and cite Puhl et al. 2012. We believe that extending the discussion to cell R3b-1 does not contribute meaningfully to the focus of this manuscript.

(4) Discussion Lines 325-327: "the sequence of events, and the proportion of the active cycle dedicated to elongation and contraction were remarkably similar in both experimental settings. This suggests that the network activated in the isolated ganglion is the one underlying the motor behavior."

Reviewer: The results and conclusions drawn in the current manuscript mirror those previously reported by Puhl and Mesce (2008, J. Neurosci, 28:4192- 420) who first demonstrated that the essential pattern-generating elements for leech crawling were contained in each of the segmental ganglia comprising the nerve cord. Furthermore, the authors showed that the duty cycle of DE-3, in a single ganglion treated with dopamine, was statistically indistinguishable from the DE-3 duty cycle measured in an intact nerve cord showing spontaneous fictive crawling, in an intact nerve cord induced to crawl via dopamine, and in the intact behaving animal. What was statistically significant, however, was that the DE-3 burst period was greatly reduced in the intact animal (i.e., a higher crawl frequency), which was replicated in the submitted paper.

There is no doubt that the article by Puhl and Mesce 2008 is seminal to the work we present here. The Reviewer seems to suggest that we do not recognize the value of this work. The contrary is true, all our related papers cite this important breakthrough. We cite the paper very early in the article in the Introduction (see lines 51 and 52-53). Likely, we would like the Reviewer to recognize the novelty of the current report. To clarify what has been shown and what is new in our manuscript, considerer the following:

i. Figures 1-6 in Puhl and Mesce 2008 provide representative intracellular recordings that describe neurons that fire in phase and out of phase relative to DE-3. Some general measurements are given in the text, but none of these figures quantify the relative activity of neurons that fire in different stages; only DE-3 activity was quantified. A quantitative description of multiple units active in phase and out of phase with DE-3 is presented here for the first time, are we wrong? This quantification is particularly relevant when assessing how a treatment affects the function of the circuit.

ii. Regarding the cycle period, we referred to the work from the Kristan lab, which reported this value long before the requested reference. We now cite Puhl and Mesce 2008 in lines 222 regarding in vivo measurements, and in line 221 regarding isolated ganglia.

iii. Regarding the duty cycle:

Puhl and Mesce 2008 measured the duty cycle of DE-3 in three configurations: a. spontaneous whole cord, b. DA-mediated whole cord and c. DA mediated single ganglion crawling. However, it does not report the duty cycle of neurons out-of-phase with DE-3. Our current manuscript carried out this analysis. One could argue that the silence between DE-3 bursts captures that value, but this is a speculation that needed a proper measure.

Puhl and Mesce 2008 does not indicate the duty cycle of the contraction and elongation stages in vivo. Our current manuscript does.

Therefore, the sentence cited by the Reviewer refers to data presented in this manuscript, and not in any prior manuscript. It is true that Puhl and Mesce 2008 inspire the intuition that the sentence is true, but does not present the data that the current manuscript does.

Finally, our study focused only on the body sections corresponding to the same segmental range used in the ex vivo experiments, rather than the whole animal. The comparison was made only to validate that the duty cycles of neurons firing in phase and out of phase with DE-3 matched the dynamic stages in the studied sections of the leech (line 364).

In my opinion, the novelty of the results reported in the submitted manuscript is diminished in the light of previously published studies. At a minimum, the previous studies should be cited, and the authors should provide additional rationale for conducting their studies. They need to explain in the discussion how their approach provided additional insights into what has already been reported.

Throughout our reply, we have provided a detailed explanation of the rationale and necessity behind each experiment. Following the Reviewer’s suggestion, we have rephrased the research objectives, included what is known from our previously published work, and highlighted the substantial new data contributed by the present study. See lines 80-85.

Additionally, we further cite our published article in lines 93, 104, 138, 146 and 250.

Reviewer #2 (Public review):

The paper is well-written overall. The findings are clearly presented, and the data seems solid overall. I do have, however, a few major and some minor comments representing some concerns.

My major comments are below.

(1) This may seem somewhat semantic, yet, it has implications on the way the data is presented and moreover on the conclusions drawn - a single ganglion cannot show fictive crawling. It can demonstrate rhythmic patterns of activity that may serve in the (fictive) crawling motor pattern. The latter is a result of the intrinsic within single-ganglion connectivity AND the inter-ganglia connections and interactions (coupling) among the sequential ganglia. It may be affected by both short-range and long-range connections (e.g., descending inputs) along the ganglia chain.

Semantics is not a trivial issue in science communication. It entails metaphors that enter the bibliography as commonly used “shortcuts” to a complex concept that are adopted by a community of researchers. And yes, indeed, they can be misleading.

However, if recording the activity in an isolated ganglion shows that a wide group of motoneurons, that control known muscle movements, presents a rhythmic output that maintains the appropriate cycle period and phase relationships, the “shortcut” is incomplete but could be valid (Puhl and Mesce 2008). If we were to include the phase lag component, a single ganglion cannot generate the fictive motor output.

Because any new study builds knowledge on the basis of the cited bibliography, the way we name concepts is a sensitive point. Adopting the terminology used by previous publications (Puhl and Mesce 2008) seems important to allow readers to follow the development of knowledge. However, attending the observation made by Reviewer #2, we included a sentence clarifying that the concept “fictive crawling” does not include intersegmental connectivity (lines 54-57)

(2) The point above is even more critical where the authors set to compare the motor pattern in single ganglia with the intact animals. It would have made much more sense to add a description of the motor pattern of a chain of interconnected ganglia. The latter would be expected to better resemble the intact animal. Furthermore, this project would have benefitted from a three-way comparison (isolated ganglion-interconnected ganglia-intact animal.

As we answered to Reviewer #1, the present manuscript does not intend to present a thorough study on how the activity in the isolated nervous system compares with the animal behavior. To do so we would have needed to perform a completely different set of experiments. To better define the relevance of our comparison with the in vivo experiments we rephrased the objective of the behavioral analysis (lines 197-199).

The main aim of the manuscript is to learn the role of premotor NS neurons in the crawling motor pattern studied using a readout (spike sorting in extracellular nerve recordings) that allows simultaneous screening of a larger number of units than in any previous study, in order to determine whether and how a recurrent inhibitory peripheral circuit is involved in coordinating or modulating the rhythmic motor pattern.

Our rationale was that the known effect of NS on one particular motoneuron (DE-3) may have overlooked a more general effect on crawling (lines 253-257). Moreover, we wanted to investigate whether this effect was due to the recurrent inhibitory circuit or if other elements were involved, and to study whether the modulation was mediated by the recurrent synapse between NS and the motoneurons.

In the context of this aim we studied the rhythmic activity of cell DE-3, together with motoneurons that fire in-phase and anti-phase, in isolated ganglia (Figure 4). To reveal the effect of NS manipulation we applied a quantitative analysis that showed the phase-specific effect of NS (Figure 6).

Given that this is the first study using a spike sorting algorithm to detect and describe the activity of motoneurons in nerve recordings we found it reasonable to compare these results with an in vivo study; thus, providing information to the general reader, that supports the correspondence between the ex vivo and the in vivo patterns.

(3) Two previous studies by the same group are repeatedly mentioned (Rela and Szczupak, 2003; Rodriguez et al., 2009) and serve as a basis for the current work. The aim of one of these previous studies was to assess the role of the NS neurons in regulating the function of motor networks. The other (Rodriguez et al., 2009) reported on a neuron (the NS) that can regulate the crawling motor pattern. LL 71-74 of the current report presents the aim of this study as evaluating the role of the known connectivity of the premotor NS neuron in shaping the crawling motor pattern. The authors should make it very clear what indeed served as background knowledge, what exactly was known about the circuitry beforehand, and what is different and new in the current study.

Rela and Szczupak 2003 and Rodriguez et al. 2009 analyze the interactions of motoneurons with NS. We believe that Reviewer #2 refers here to Rodriguez et al. 2012. A similar observation was made by Reviewer #1. Below, we copy the answer previously stated:

Following the Reviewer’s suggestion, we have rephrased the research objectives, included what is known from our previously published work, and highlighted the substantial new data contributed by the present study. See lines 80-85.

Additionally, we further cite our published article in lines 93, 104, 138, 146 and 250.

Reviewer #1 (Recommendations for the authors):

Please edit for correct word usage.

Reviewer #2 (Recommendations for the authors):

Minor Concerns

(1) LL33-36: These lines are somewhat vague and non-informative. Why is the functional organization of motor systems an open question? What are the mechanisms at the level of the nerve cord that are an open question? Maybe be more explicit?

We did as suggested (lines 30-32).

(2) L62: The homology between the NS neurons and the vertebrate Renshaw cells is mentioned already in the Abstract and here again. While a reference is provided (citing the lead author of this current work), the reader would benefit from some further short words of explanation regarding the alleged homology.

We included a description of Renshaw cell connectivity (lines 64-65).

(3) LL90-92: The NS recording in Figure 1 (similar to Figure 3 in Rodriguez et al.) demonstrates clear distinct IPSPs. Could these be correlated with DE-3 spikes?

We investigated this correlation in detail and the answer is that there is no strictly a 1:1 DE-3 spike to IPSP correlation. NS receives inputs from other dorsal and ventral excitors of longitudinal muscles, and the NS trace is too “noisy” to reflect any short-term correlation. Originally we proposed that the NS IPSPs were due to the polysynaptic interaction between the MN and NS (Rodríguez et al. 2012). However, the present work demonstrates that the IPSPs in NS are caused by a source upstream from the MNs.

(4) LL145-145: Do you mean - inhibitory signals FROM NS premotor neurons? Not clear.

We see the confusion, and we rewrote the sentence (lines 164). We hope it is clearer now: “…inhibitory signals onto NS premotor neurons were transmitted to DE-3 motoneurons via rectifying electrical synapses and counteracted their excitatory drive during crawling, limiting their firing frequency.”

(5) LL153-154: Why isn't AA included in Figure 4A?

Reading our original text, the Reviewer #1 is right in expecting to see the AA recording. We changed the sentence: “we performed extracellular recordings of DP along with AA and/or PP root nerves” (lines 171-172).

We dissected the three nerves but, unfortunately, we did not always obtain good recordings from the three of them.

(6) LL237-238: The statistical significance (B- antiphase) is not clear. Furthermore, with N of 7-8, I'm not sure the parametric tests utilized are appropriate.

Regarding the Reviewer's concern about the tests, please note that all the assumptions made for each model were tested (see now Materials and Methods lines 466-467).The information on each model is provided in Supplementary Table 2 under the column 'Model, random effect,' which specifies whether a Linear Mixed Model (LMM) or a Generalized Linear Mixed Model (GLMM) was implemented. For GLMMs, the corresponding distribution and link function are also specified. For the analysis of Max bFF of Anti-Phase motor units, we found a significant interaction between epoch and treatment, indicating a difference between treatments. This is indicated on the left of the y-axis (##). In control experiments, all three comparisons (pre-test, pre-post, test-post) show significant differences in Max bFF: this variable decreased (slightly but significantly) along the subsequent epochs, suggesting a change over time. We now corrected the text to indicate that these changes were small (line 268). In contrast, Max bFF in depo experiments remained stable between pre-test and pre-post, but significantly decreased between the depo and post epochs. Thus, in our view the comparison between control and the test supports the conclusion that NS depolarization was limited to counteracting this decrease (lines 270-273). Supplementary Table 2 provides the significance and modeled estimated ratio for each comparison in the column for pairwise simple contrasts.

Thanks to this question, we realized that the nomenclature used in the table for the epochs (pre - depo - post) needed to be changed to pre - test - post, and we have now corrected it.

(7) LL240-241: I fail to see a difference from Control.

For the Relative HW of In-Phase units, we also found a significant interaction between epoch and treatment, indicating a difference between treatments, as denoted to the left of the y-axis (#). Then, the significance of the comparisons across epochs within each treatment are shown in the figure (*). What is important to notice is that obtaining the same significance for each treatment does not imply identical results, but we failed to describe this in our original text and we do now in lines 275-279.

(8) LL244-245: I must admit that Table 2 is beyond me. Maybe add some detail or point out to the reader what is important (if at all).

We have now clarified what each column of the tables indicates in the corresponding legends.

Here, we also share an insight into how the experiments were designed and analyzed:

To account for possible temporal drifts of the variables during the recordings that could mask or confuse the results, we compared two experimental series: one in which NS was subjected to depolarizing current pulses (depo), and another series (ctrl) in which the neurons were not depolarized.

The statistical analysis was made using Linear Mixed Models (LMMs) or Generalized Linear Mixed Models (GLMMs). In these analyses treatments and epochs are used as explanatory variables to evaluate the interaction between these factors. These models allow us to determine whether changes in each variable across epochs differ depending on the treatment. For example, whether the variation in firing frequency from pre to test to post differs between control experiments and those in which NS was depolarized.

A significant interaction between treatment and epoch indicates that NS depolarization affected the variable. In such cases, we performed pairwise comparisons between epochs (pre-test, test-post, pre-post) within each treatment. In contrast, the absence of a significant interaction can result from two possibilities: either the variable did not change across epoch in either treatment, or a similar temporal drift occurred in both cases.

(9) LL245-256: Move this paragraph to the discussion.

Because we introduced a rationale for the experiments described in Figure 6 (lines 282-284) the paragraph was mostly removed, but the part that supports the methodological approach was left.

(10) LL259-260: see my second minor point above. This is explained in LL270-272 for the first time.

We amended according to comment (2).

(11) Figures: The quantitative analysis shown in Figure 3B is very useful. Why isn't this type of analysis utilized for the comparisons shown in Figures 4 and 6?

We chose different ways of plotting the data based on their nature. In Figure 3B, we present data from an identified neuron (DE-3) recorded in different experiments. In contrast, in Figure 6 we analyze data from neurons classified into the same group based on their activity during the fictive crawling cycle, but their individual identity was not ascertained. Therefore, we consider it important to plot the results for each unit individually, to assess the effect of temporal drift and NS depolarization.

(12) Figures: Figure 7 is meant to be compared to Figure 1C; the point being the addition of an inhibitory connection onto the NS neuron. Why are other details of the figure also different (different colored M)?

While Figure 1C illustrates the known connection between NS and both DE-3 and CV motoneurons, Figure 7 shows the connections between NS and the different groups of motor units described in this study. The units are represented in the circuit using the same colors that identify them in Figures 4 and 6. Since the CV motoneuron was not recorded in this study, the circuit represents the AntiPhase neurons but does not identify them with CV. Figure 7 legend now clarifies what the colors represent, and Figure 1C has been updated to match the same color scheme.