Persistent Firing Neurons in the Medial Septum Drive Arousal and Locomotion

  1. Endre Levente Marosi
  2. Karolina Korvasova
  3. Felix Ludwig
  4. Hiroshi Kaneko
  5. Liudmila Sosulina
  6. Tom Tetzlaff
  7. Stefan Remy
  8. Sanja Mikulovic

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This paper suggest that that intrinsically generated persistent firing activity of medial septal glutamatergic (VGluT2+) neurons underlies initiation of locomotor activity. In this work, the authors provide evidence for a non-canonical role for persistent firing in initiating locomotion by performing a series of technically difficult experiments to dissect the circuit mechanisms of the persistent firing. This manuscript will be of interest to readers in the field of spatial navigation, motor control, and neural network dynamics.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

The medial septum and diagonal band of Broca (MSDB) serve as a central hub in an ascending brainstem pathway that conveys sensory and motor signals to the limbic system. However, the cellular and circuit mechanisms underlying these functions remain unclear. Here, we show that transient optogenetic activation of MSDB VGluT2⁺ neurons initiates a structured arousal sequence -beginning with facial movements, followed by pupil dilation and locomotion. Neuropixels recordings reveal persistent MSDB neuronal activity that strongly correlates with arousal-related behaviors. We demonstrate that persistent firing (PF) is an intrinsic property of a subset of MSDB neurons, independent of ongoing synaptic input. PF neurons, along with putative GABAergic theta-bursting neurons activity, reliably predicted behavioral initiation. These findings suggest that PF neurons orchestrate the transition from preparatory movements to full behavioral engagement, thus bridging sensory input with locomotor arousal and supporting state transitions in the limbic system.

Article activity feed

  1. eLife

    Evaluation Summary:

    This paper suggest that that intrinsically generated persistent firing activity of medial septal glutamatergic (VGluT2+) neurons underlies initiation of locomotor activity. In this work, the authors provide evidence for a non-canonical role for persistent firing in initiating locomotion by performing a series of technically difficult experiments to dissect the circuit mechanisms of the persistent firing. This manuscript will be of interest to readers in the field of spatial navigation, motor control, and neural network dynamics.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)

  2. eLife

    Reviewer #1 (Public Review):

    Involvement of persistent neuronal activity in behavioral-cognitive functions represents an important research field of neuroscience. Persistent activity may be caused by intrinsic biophysical properties, neural circuit dynamics (i.e., synaptic reverberation in recurrent circuits) or a combination of both. Intrinsically generated persistent firing of excitatory neurons has been frequently observed in acute slices obtained from various brain areas. However, it is still unclear how cell-autonomous persistent firing is involved in cognitive functions (e.g. working memory) and whether intrinsically generated persistent firing underlies other types of behavior.

    The manuscript by Korvasová et al investigated glutamatergic (VGluT2+) neurons of medial septum and diagonal band of Broca (MSDB) and revealed that persistent activity driven by intrinsic excitability of these neurons controls locomotor activity. This group previously described how septo-hippocampal glutamatergic (VGluT2+) neurons control the initiation and velocity of locomotion as well as the entrainment of theta oscillations (Fuhrmann et al., 2015). The manuscript by Korvasová et al represents a continuation and important extension of their previously published research, providing important insights into the underlying cellular mechanisms.

    The experiments and data analysis have been carefully performed. The article is compact and well written. In vivo data and in vitro experiments show good coherence and data are represented in a well-structured, comprehensive manner. The main finding is novel and of significant importance.

    However, while the experiments have been carried out very competently and the paper is well written, I am a bit concerned that the manuscript in the present form is rather descriptive, without going deeper into the investigation of mechanisms underlying activation of intrinsic persistent activity. The authors point out that analysis of the intrinsic mechanisms and conductances underlying persistent firing in the MSDB is beyond the scope of the present paper and thoroughly debate possible mechanisms in the discussion section.

  3. eLife

    Reviewer #2 (Public Review):

    The medial septum is thought to be a central hub where generation of theta rhythm is coalesced with the regulation of movement. VGluT2-expressing (glutamatergic) MS neurons were identified as interdependently controlling both movement initiation and theta rhythm genesis. Stimulation of these neurons triggers movement and theta outlasting the duration of the stimulus. The Authors explored the mechanism whereby triggered activity persists beyond stimulation and whether movement-induction is independent of the emergence of theta. In behaving head-fixed VGluT2-Cre mice they demonstrate that specific activation of VGluT2 neurons initiated motion and theta paralleled by the persistent activity of MS neurons. Blocking synaptic transmission within the MS attenuated theta induction and reduced persistent neuronal activity without affecting movement initiation by brief VGluT2-activation. The latter manipulation reliably evoked persistent firing in MS slice preparations weakened by synaptic blockers. They conclude that movement is controlled by VGluT2 neurons independent of theta whereas for the latter interaction among the glutamatergic and other major MS neuron populations (cholinergic and especially GABAergic) is pivotal. They also claim that VGluT2 neurons' persistent activity depends on the intrinsic dynamics of these neurons modulated by the MS network. The study is nicely designed, and the Authors used well-established methods. The conclusion is in line with the major findings. However, the analysis falls short in several respects and there are some missed opportunities because of which this study is only an incremental step beyond what we already know about the "third" major neuron class of the MS.

    First, the Authors simultaneously registered the activity of multiple MS units, but they did not exploit the potential of multichannel data for separating and characterizing the response and single stimulus-triggered interaction of the major MS neuron types (seemingly, only multiunit activity was used). The timing (latency and duration) and dynamics (gradually accelerating, fluctuating or dampening alteration of activity) of their stimulus-triggered activity would reveal key details about what happens to the MS network following the injection of a brief excitatory pulse. What types of cells show persistent activity: only the regular, tonically firing (putative glutamatergic) neurons or even the theta bursting ones maintain their elevated activity outlasting the stimulus? A particularly important point would be to correlate the timing of theta and movement with that of the identifiable firing pattern types. Uncovering causal relationship among the activated interacting neurons would also be interesting: would it be possible to explain altered activity of a given type by the stimulus-evoked change of another type?

    Light stimulation would have given the opportunity of identifying the stimulated VGluT2 neurons, for example by applying a train composed of very short (1 ms) tagging light pulses at the end of a recording session for the later identification and isolation of VGluT2 units. Then, the response of these optically tagged VGluT2 neurons could have been compared to the other, unidentified neuron types.

    As stated, the in vitro experiments are especially suitable for exploring the mechanisms of persistent activity. Unfortunately, the question about the mechanism remains unanswered. While we are informed about the network-independence of persistent activity, no further attempts have been made to uncover cell-autonomous processes. We could also learn from the Results that the network dampens persistent activity probably by recurrent inhibition. Demonstrating the facilitated activity of putative inhibitory (fast, rhythmic spiking) neurons locked to the light-activation of VGluT2 neurons would disclose how stimulus-outlasting activity of VGluT2 neurons is controlled by inhibition.

    Sensory stimuli reliably evoke theta and movement comparable to what was detected in response to VGluT2 neurons' activation. Hence, an opsin-lacking reporter control should be added to the results for separating the animal's reaction to light from the effect elicited by selective VGluT2-stimulation.

  4. eLife

    Reviewer #3 (Public Review):

    In this study, the authors have discovered a novel activity type of activity within the MSDB - persistent activity. They show strong evidence that brief stimulation of glutamatergic neurons within the MSDB generates a sustained increase in neuronal firing that lasts long beyond the stimulus (and continues after animals stop running). They go on to determine the circuit mechanisms of this activity, and find that the persistent activity is maintained in the presence of synaptic blockers (in vivo and in vitro). There are subtle differences between the in vivo and in vitro results which slightly weaken the conclusions here. Overall, it seems true that synaptic connections within MSDB are not necessary for the persistent activity, therefore the following critique should be considered as minor. In vitro - the blocking of synapses reduces the magnitude of the firing rates during the persistent firing, whereas in vivo no reduction of magnitude is observed. It is unclear whether the difference between in vivo and in vitro data are because of slice dynamics or because the blockers are not as effective in vivo - there is no clear cut control that the blockers are working in vivo. Synaptic blockers in vivo do inhibit hippocampal theta (suggesting that the connections from glutamatergic -> PV interneurons are indeed blocked). More analysis of MSDB spiking in the blocked condition or more in depth presentation of the inhibited theta could bolster the claim that the synaptic block is effective in vivo - which would strengthen the conclusion that intra-septal circuitry is not necessary for the persistent activity. In lieu of that, it may be better to soften the conclusion (the in vitro data suggest that persistent activity does not require intra-septal circuitry (as concluded), however the magnitude of the activity is dependent on intra-septal circuitry).

    A more serious weakness of the work concerns whether the persistent firing occurs under normal physiological conditions (i.e. - with no optogenetic push). The title of the manuscript suggest that persistent activity is linked to locomotion and this suggests that the persistent activity is a physiologically relevant mechanism. There are some data presented that show that during voluntary running the MSDB neural activity is increased - however there is not a clear presentation of the data that shows us that increased activity during voluntary running is persistent. In the stimulation experiments, the persistent activity is sustained (with lower magnitude) well after the animal stops running. Is this the case for voluntary running? A clear presentation of persistent firing associated with voluntary running epochs would greatly strengthen the manuscript - in that it would prove that persistent firing occurs under physiological conditions.

  5. Version published to 10.1101/2021.04.23.441122 on bioRxiv