Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Dec 1.
Published in final edited form as: Neuroscience. 2020 Jun 6;450:184–191. doi: 10.1016/j.neuroscience.2020.05.050

Corticospinal Pathways and Interactions Underpinning Dexterous Forelimb Movement of the Rodent

Mark Basista 1,2, Yutaka Yoshida 1,2
PMCID: PMC7688482  NIHMSID: NIHMS1605393  PMID: 32512136

Abstract

In 2013, Thomas Jessell published a paper with Andrew Miri and Eiman Azim that took on the task of examining corticospinal neuron function during movement (Miri et al., 2013). They took the view that a combination of approaches would be able to shed light on corticospinal function, and that this function must be considered in the context of corticospinal connectivity with spinal circuits. In this review, we will highlight recent developments in this area, along with new information regarding inputs and cross-connectivity of the corticospinal circuit with other circuits across the rodent central nervous system. The genetic and viral manipulations available in these animals have led to new insights into descending circuit interaction and function. As species differences exist in the circuitry profile that contributes to dexterous forelimb movements (Lemon, 2008; Yoshida & Isa, 2018), highlighting important advances in one model could help to compare and contrast with what is known about other models. We will focus on the circuitry underpinning dexterous forelimb movements, including some recent developments from systems besides the corticospinal tract, to build a more holistic understanding of sensorimotor circuits and their control of voluntary movement. The rodent corticospinal system is thus a central point of reference in this review, but not the only focus.

Introduction

Many systems have been studied in the context of dexterous forelimb movement, but the corticospinal tract (CST) has been of particular interest, as it is a major source of supraspinal movement control in humans (Lemon, 2019). As such, corticospinal neuron (CSN) activity and connectivity within cortex have been extensively studied in a variety of model organisms and humans. Functional studies of CSN activity during movement have been wide-ranging, often using different learned and innate movements, from sequential ballistic movements like lever presses (Peters et al., 2017) to dexterous single-forelimb movements like grasping and retrieving objects (Wang et al., 2017). Additionally, the CSN circuit is one of many descending and horizontally-projecting circuits along the motor control axis (Lemon, 2008; Alstermark & Isa, 2012; Shepherd, 2013; Mohammed & Hollis, 2018; Isa et al., 2019; Guo et al., 2020).

It is thought that damage to the descending tracts of the CSNs underpins some of the lost motor control seen in patients with spinal cord injuries or some strokes, and that restoring functionality to this circuit could improve recovery from these injuries. However, the direct connections from sensorimotor cortex to the spinal cord are only one of many different routes for activity to control volitional movement in the mammalian brain (Fig. 1). The interconnectedness of these circuits suggests that damage to one circuit likely influences activity in the others. Even if some circuits are left physically intact – which often is not the case, for example, with an internal capsule stroke – their activity and output is undoubtedly changed by the loss of input from the damaged portion of the circuit.

Figure 1.

Figure 1.

Open in a new tab

Schematic of the circuitry with which CSNs interact directly and indirectly during dexterous forelimb movement. This figure illustrates the multitude of routes for motor output to reach effector motor pools in the spinal cord, and for ascending sensory information to influence motor encoding. The arrows designate routes and directions of communication. Note that CSNs have parallel outputs to the spinal cord, which may contribute in distinct manners to the traditionally-defined cortico-striatal, cortico-rubral, and cortico-reticular populations. Additionally, this circuitry likely exists in parallel for each muscle group innervated by different portions of the motor cortex, as defined by intracortical microstimulation. The degree of cross-talk between these parallel circuits is still being investigated. To aid interpretability, the complex interconnectivity, and potential paralleling of outputs, within the subcortical projection regions is not illustrated. This simplicity should not be taken to suggest that these regions act as simple relays for descending cortical signals, but their processing of these signals is outside of the scope of this review. Each cortical neural population is defined by its major projection patterns, or the traditional definitions used when defining the neural population. Layer II/III neurons may exist as separate, parallel populations that project to each cortical output population independently, or there could be significant cross-talk within layer II/III. This has yet to be established in the context of dexterous forelimb movements. The dotted line from striatum to thalamus indicates its indirect passage through the substantia nigra and globus pallidus. CC – cross-cortical; CStr – cortico-striatal; CT – cortico-thalamic; CRu – cortico-rubral; CSN – corticospinal neuron; CRet – cortico-reticular.

Much progress has been made in the study of descending motor control across model organisms and humans. Largely, the understanding is that, in primate species, corticospinal neurons play a major role in the execution of dexterous forelimb movements (Evarts, 1966), and that this is achieved through their direct connectivity with spinal motoneurons (Lawrence & Kuypers, 1968a; Bortoff & Strick, 1993). However, damage to other pathways, such as the rubrospinal and reticulospinal pathways, impairs hand movements, as well (Lawrence & Kuypers, 1968b), and many neural populations contribute to movement execution (Churchland et al., 2012). This implicates a need to carefully dissect the functions and interactions of pathways across all parts of the central nervous system sensorimotor axis. The advent of powerful tools for studying cell physiology and exceptional engineering of cellular recording and behavioral apparatuses has allowed for many new insights in the circuit-level control of movement in rodents, as well. While rodents normally lack direct corticomotoneuronal connections and do not generally achieve the same levels of dexterity seen in primates (Lemon, 2008), we are now afforded greater resolution of how each pathway is involved in learning and executing movements than ever before. To begin, we will look at recent developments regarding functional inputs to the corticospinal tract of the rodent.

Laminar Distribution of Interareal Connections Between Sensorimotor Cortex

CSNs are spread across different parts of the sensorimotor cortex, with some present in the primary motor (M1), secondary motor (M2), primary sensory (S1), and secondary sensory (S2) cortex in rodents. These areas are further subdivided by which limbs they are associated with. The rostral forelimb area (RFA) is within M2, the caudal forelimb area (CFA) is within M1 and S1, and another forelimb area is present in S2, all identified on the basis of retrograde labeling from axon terminals within the cervical spinal cord, where the forelimb motor pools are located (McKenna et al., 1993). RFA neurons with cell bodies located in layer Vb have been shown to project to layers Vb and I of CFA, while CFA neurons with cell bodies in layer II/III and Va project to RFA layers Va, II/III, and Vb (Hira et al., 2013b). Photostimulation revealed that RFA and CFA can drive activity in their respective projection targets in the other area.

This evidence for interareal interconnectedness was extended two years later. CSNs in M1 receive monosynaptic input from S2 and CSNs in S2 receive monosynaptic input from M1 (Suter & Shepherd, 2015). The authors first identified populations of neurons that projected from S2 to M1 and vice versa. The cell bodies of S2 ➔ M1 neurons were located most predominantly in layers Va, VI, and II/III. The cell bodies of M1 ➔ S2 neurons were located most predominantly in layer II/III and at the border of layer Va and Vb. The authors then used the sCRACM (subcellular ChR2-assisted circuit mapping) technique of optogenetic stimulation to excite axon terminals of either S2 neurons located in M1, or M1 axon terminals in S2, while recording from CSNs. In both cases they detected rapid depolarizations in CSNs, as well as other layer Vb neurons.

Layer-specific Inputs to CSNs Within Areas

Layer II/III has been determined to be a strong source of excitation to layer V neurons (Weiler et al., 2008). Uncaging glutamate in layer II/III in slices of forelimb cortex led to the strongest stimulation along the layer Va/Vb border. When the slices were disinhibited, in order to assess whether any inhibitory feedback networks might be suppressing inputs from other layers, the authors found the same pattern of excitation. This indicates that layer II/III holds a dominant position in a top-down hierarchy of input to the deeper output layers of motor cortex.

Subsequently, this route was shown to be organized into parallel pathways that differentially targeted the CSNs and corticostriatal neurons of layer V (Anderson et al., 2010). The authors again used glutamate uncaging to explore the excitatory connections from layer II/III to layer V while recording whole-cell activity in layer V. By labeling the neurons before the recordings with fluorescent beads, they were able to conduct paired recordings of labeled corticospinal and corticostriatal neurons. The most powerful currents in CSNs were elicited from stimulation of the terminals around the dendritic arbors located in layer II/III. CSNs located more than half the depth through layer Vb received little to no excitation from layer II/III neurons and appear to have a slightly elevated excitatory response to other layer Vb inputs. Contralaterally-projecting corticostriatal neurons in lower layer Va were most strongly driven by layer II/III stimulation. Vb corticostriatal neurons were much less responsive to all attempts at excitation. These results support the strong top-down hierarchy hypothesis of motor cortical circuitry. Further investigation is needed into how these hierarchical inputs target corticorubral and corticoreticular neurons.

Layer 2/3 Neurons as Substrate for Hierarchical Control of Dexterous Movements

A key difficulty in studying sensorimotor circuitry is the open question of how necessary different cortical areas are for the performance of specific motor tasks, and how they might interact across parallel hierarchies to facilitate dexterous movement. Nonetheless, correlations between neural activity within various portions of sensorimotor cortex with learned movements have been gathered. One group analyzed the activity of layer II/III neurons, as measured by calcium fluorescence, of the CFA during the learning of a cued lever-press task (Peters et al., 2014). The pool of task-associated neurons in layer II/III roughly doubled during learning, before being refined into a smaller ensemble comparable in size to the initial pool of neurons, but with sharper timing relative to the lever-press. This, coupled with the previous work on top-down pathways, indicates that task-learning may be associated with a refinement in the input from layer II/III to layer V.

In contrast to this work, another group found that a lever-pull task did not lead to an increase in the number of recruited neurons within layer II/III of CFA (Masamizu et al., 2014). Instead, a small ensemble of highly correlated neurons remained stable throughout learning of the lever-pull task, while an ensemble based in layer Va refined its activity to more tightly match lever-pull trajectories over learning. This result could be due to differences in the tasks between the two studies. One task was an extension away from the body while the other was flexion towards the body and may have involved muscle groups in different manners. Mice received rewards once they passed experimenter-defined thresholds, but no ceiling was placed on the task, meaning that mice were free to push the lever as far away or bring it as close to their body as they could physically. This lack of constraint on the apex of the movements may mean that the movements between animals were variable, complicating the detection of consistent movement-related activity. Additionally, as precise distance targets were not used, these may have not required as much consistent cortical control. This is similar to findings in monkeys that a power grip is not accompanied by cortical projection neuron activity, but that smaller-force movements are (Muir & Lemon, 1983). Thus, the actual function of the activity of these recorded neurons is still debatable.

After mice learned the same lever-pull task as above, Hira et al. (2013a) found that some layer II/III neurons were grouped into clusters approximately 70μm across. These clusters showed similar activity to each other during the pull movement. These layer II/III “pull cells” could function as a way to organize different movements. This interpretation is supported by the evidence that distinct neural signatures can be detected in layer II/III related to grooming versus running (Dombeck et al., 2009). The degree to which these neural signatures cause their associated movements is still debatable, but the distinct activity patterns in layer II/III that are present during different movements indicates that motor cortex is organized in such a way as to differentiate between movements.

Inputs to Motor Cortex and Influence on Movement

What about inputs to the other layers of motor cortex that could impact hierarchical processing? A set of recent studies have provided key insights into the necessity of motor cortex for different movements, and point toward a role for rodent motor cortex in dexterous movement that requires online sensory processing.

A study of motor cortex function found that rats trained to press a lever twice with a specific inter-press interval of 700ms could perform the task after learning it even after the motor cortex was entirely destroyed by aspiration or excitotoxic chemical lesion (Kawai et al., 2015). Rats could not learn the task, however, if they had not been previously trained before the lesion, pointing toward a necessary role for motor cortex in the learning and refinement of the movements. Muscimol inactivation and jamming produced by optical stimulation of the rat motor cortex also impaired performance of the lever-press task (Otchy et al., 2015). Together, these outcomes seem to suggest that while motor cortex may not be necessary for producing some isolated forelimb movements, its activity during the movement could override motor encoding by other circuits. This is consistent with previous studies of cat motor cortex function, which demonstrated that cortical stimulation could override otherwise non-cortical movements (Armstrong & Drew 1985a,b; Bretzner & Drew, 2005). These findings are consistent with the idea of hierarchical control of movement, with motor cortex capable of overriding non-cortical circuits when necessary. Hints as to when such overriding might become necessary arise from anatomical and functional studies of thalamic inputs to motor cortex.

Thalamus projects predominantly to M1 layers II/III and V in the mouse, and includes pyramidal populations (Hooks et al., 2013). What roles might these projections serve for movement? It was shown recently that scrambling activity from thalamic axons in motor cortex perturbs forelimb movements, and that inhibiting thalamic activity results in failed initiation or pausing of movements in progress (Sauerbrei, Guo, & Cohen et al., 2019). This was discovered through a surprising result of motor cortex inactivation during reaching and grasping. Sometimes, after optogenetic inactivation was lifted, mice would attempt a reach, and sometimes they would not. This was accompanied by predictive neural activity in motor cortex. When neural activity recapitulated activity that occurred during a normal reach, a reach was initiated, whereas when the activity failed to recapitulate reach-related activity, no reach occurred. The authors took this to mean that some other input was necessary for kick-starting the neural activity and producing a reaching movement. By optogenetically inhibiting the thalamus, the authors demonstrated that thalamic input was the source of this input. They additionally showed that modifying the pattern of thalamic input by optogenetically-stimulating thalamic terminals in motor cortex leads to movement perturbations.

In what cases would outside input guide motor cortex output? A recent study demonstrated that mouse motor cortex is crucial for mediating tasks that require online sensory processing (Heindorf et al., 2018). When mice were trained to run a virtual corridor on a trackball, the activity of their layer II/III and V neurons tracked with the magnitudes of turns down the corridor, and optogenetic inactivation of motor cortex impaired corrective turning induced through visual perturbations until the inactivation was removed. When the visual stimulus of the virtual corridor was unexpectedly turned in a way that did not match the motion of the mouse, layer II/III neurons became more active than usual, appearing to facilitate corrective turning. The findings of online sensory processing through cortex are consistent with evidence about cortical involvement during obstacle-avoidance in the cat (Drew, 1993; Widajewicz et al., 1994; Drew et al., 1996)

These findings add to our understanding of the variety of circuits providing input to the descending circuits controlling movement. Figure 1 summarizes our current understanding of the circuits that contribute to cortical, and particularly corticospinal, output during movement in the rodent brain. The figure is not meant to indicate any specific hierarchy of control, only to preserve approximate anatomical positions between regions. This input/output map indicates routes of communication, not specific axonal projections. For example, the CSNs extend large apical dendrites dorsally through layers Va and II/III. The synapses made on these dendrites are simply mapped as an arrow from layer II/III and layer Va to the corticospinal pool in layer Vb. As can be seen, a variety of pathways may be impacted, or even contribute to functional recovery, whenever manipulation of one pathway is undertaken.

Parallel CSN Output Pathways

Crucially, motor cortex output is much more complex than simply direct output from motor cortex to spinal cord. Detailed maps of the supraspinal descending tracts, and interconnections between them are now available (Wang & Maunze et al., 2018). These authors demonstrated through highly-resolved tracing studies that corticospinal neurons project collaterals to many of the other sources of input to the spinal cord, including the red nucleus and reticular formation. Tract-tracing combined with tissue clearing and light-sheet microscopy allowed for views of the intact pathways from and between each of the descending pathways. Collateralization from CSNs to the reticular formation have similarly been demonstrated in the monkey (Keizer & Kuypers, 1989), and the reticulospinal tract is involved in hand movements in the primate (reviewed by Baker, 2011). In addition to the CST, many other output neurons comingle with CSNs in layer V (Oswald et al., 2013), opening the way for parallel communication with CSNs.

The CST has been implicated in supportive functions for motor control, including gating of nociceptive ascending circuits. Work in the rat has illuminated its role in inhibiting C-fibers through stimulation of GABAergic interneurons in the spinal cord (Moreno-Lopez & Perez-Sanchez et al., 2013). The authors blocked potentiation of spinal dorsal horn responses to C-fiber stimulation by pairing it with stimulation of sensorimotor cortex in urethane-anesthetized rats. After abolishing GABAA signaling in the spinal cord with bicuculline, the authors observed that cortical stimulation no longer inhibited responses to C-fiber stimulation in the spinal cord. The question then arises of whether this effect is produced by a subset of CSNs, likely located in sensory cortices, or if an anatomical substrate exists for motor CSN influence on sensory neurons in the spinal cord.

This question was partially addressed by work showing differential population projections of motor and sensory CSNs into the spinal cord (Ueno et al., 2018). The authors used tracer injections along with several interneuron Cre-reporter lines to investigate the projection patterns of CSNs from the caudal forelimb area and forelimb sensory cortex. Sensory CSNs were found to predominantly project to touch, nociceptive, and proprioceptive interneurons, while primary motor CSNs projected to a variety of interneurons that then made synapses on spinal motor neurons. Notably, inhibiting Chx10 spinal interneurons impacted movement similarly to motor cortex lesions or inactivation, corresponding to their substantial innervation by motor CSNs. Inhibiting Vglut3 spinal interneurons induced deficits in food pellet release, demonstrating the need for sensorimotor integration for successful fluid movement execution. Many other types of interneurons and ascending neurons were found to be innervated by the CST, as well. Reviewing the literature supports the idea that the CST is subdivided into several parallel pathways (Moreno-Lopez et al., 2016), perhaps differentially regulating each of these spinal circuits.

In fact, evidence to this effect was discovered in the rat (Olivares-Moreno et al., 2017). Electrophysiological recordings in the spinal cord in response to CST stimulation showed differential latencies of excitation between dorsal and intermediate zones. To eliminate the possibility that this was being driven by other circuits, the authors antidromically stimulated the CSN axons that projected to the dorsal horn and found that different CSNs responded with different latencies. Finally, tracer injections into the dorsal and intermediate zones of the spinal cord confirmed different CSNs project to each area. CSNs projecting to the dorsal and intermediate lamina of the spinal cord in the same segmental zone are largely separate, though intermingled, populations of neurons with different conduction velocities. Furthermore, in slices of CFA, rat CSNs that project to the dorsal and intermediate/ventral zones of the spinal cord show activity that is more synchronous with other neurons that project to the same zone (Olivares-Moreno et al., 2019). This hints at the possibility of many parallel subdivisions of the CST along the projections to the spinal cord.

Time Course and Consistency of Descending Activity during Movement

A recent study observed for the first time what appears to be a tight spatiotemporal code of forelimb movement in CSNs (Wang et al., 2017). By imaging the calcium activity in apical dendrites of CSNs, the authors showed that some CSNs in anteromedial CFA begin firing before and during the reach toward a food pellet. The activity then shifts to RFA CSNs during the grasp of the pellet, followed by a final shift to the posterolateral CFA and sensory cortex in the post-grasp phase. While the timing of CSN activity in each region was not completely enforced – there were active CSNs during all phases of movement in all regions – there appeared to be preferred ensembles of activity based on total numbers of active neurons in each area. This is consistent with results from long-duration intracortical microstimulation and targeted cooling experiments (Ramanathan et al., 2006; Harrison et al., 2012; Brown & Teskey, 2014) which demonstrated that more anterior portions of motor cortex could drive reaching away from the body and grasping while the posterior-most portions of motor cortex could drive retrieval movements. This corresponds to the map, rotated approximately 90°, found for reaching, grasping, and retrieving-type movements in macaques (Graziano et al., 2005; Graziano, 2016). Wang et al. also demonstrated that selective CSN ablation in each cortical region resulted in selective deficits in the performance of each respective portion of the movement. In interpreting these results, however, it is important to recall the collaterals of CSNs to the red nucleus and reticular formation (Wang & Maunze et al., 2018), thus opening up the possibility that alterations to the inputs and outputs of these regions could have contributed to the behavioral findings.

An open question still exists in the consistency of CSN activity that underpins movement. Recently, Peters et al., 2017 observed inconsistent CSN activity during the learning and execution of a lever-press task. Successive days of recording showed that the same CSN might appear active at different times during the movement, and indeed, pairwise correlations of activity for CSNs in both early and late trials, despite increasing substantially over learning, were shown to be relatively low compared to layer 2/3 neuron activity. The results of Wang et al., 2017 would suggest that a consistent spatiotemporal code underlies forelimb movements, and this is supported by evidence in rats of sharpened M1 cell activity after learning a reaching task (Kargo & Nitz, 2004), and that recurrent connectivity is also strengthened after learning forelimb reaching (Biane et al., 2019). Mice and rats usually attain different levels of performance on reaching and grasping tasks (compare Xu et al., 2009; Whishaw et al., 1991). It is possible that species differences in the consistency of CSN activity during learned movements could underpin this difference in performance. It is also possible that differences in the level of dexterity required to perform the experimental tasks (lever-press versus food pellet reaching) could explain this difference in CSN activity.

Task Dexterity and Involvement of Descending Circuits

If CSN output during reaching and grasping is a sparse spatiotemporal code (Wang et al., 2017), this would predict that the mapping of CSN synapses on the architecture of spinal circuitry (Ueno et al., 2018) is the primary computer of movement control, taking a timing input from CSNs and determining the pattern of – sometimes simultaneous – muscle contractions and co-contractions necessary to execute a learned movement. These patterns of CSN output onto the spinal cord would amount to (many) parallel pathways for CSN activity to sequentially activate appropriate muscle groups during the movement. These views are challenged by the fact that many movements can still be accomplished in model organisms even with extensive damage to or disconnection of motor cortex from the spinal cord (Kawai et al., 2015), and that CSN activity can be inconsistent during forelimb movement (Peters et al., 2017).

A potential answer to this impasse was recently suggested by work from Thomas Jessell’s group (Miri et al., 2017). Comparing two behaviors, treadmill walking and a precision joystick movement, the authors were able to show that cortical manipulations impacted the performance of the joystick movement but not treadmill walking, even as muscle activity was modulated in both tasks by cortical manipulations. The modulations of muscle activity were at different latencies during each task, supporting the existence of a direct pathway from cortex that is used in a dexterity-dependent manner. This points toward the dexterity requirements of the task as the primary determinant of cortical involvement in successful movement execution by rodents, similarly to primates (Lemon, 2008). How could we operationally define the dexterity requirements of a task in an animal that lacks individuated digit movement? A key feature of dexterity could be whether non-somatosensory online sensory integration is required for the task, as Heindorf et al., 2018 showed. This would also suggest a crucial role for CST integration of multimodal sensory information and modulation of spinal sensory feedback during dexterous forelimb tasks. More evidence for online sensory processing during dexterous movement was recently discovered in a water-reaching task (Galiñanes et al., 2018). The authors demonstrated that layer II/III neurons were selective for reach direction for the task, which required olfactory sensory information to perform accurately.

Additionally, in interpreting any effects of manipulations we observe, we must keep in mind that the interconnectedness of the circuits (Fig. 1) almost definitely means that manipulations, even when cell-population targeting is carefully controlled, may influence the activity in other descending and spinal circuits. Neurons in the red nucleus show plastic changes during movement refinement, and ablating these neurons negatively impacts reaching success (Rizzi et al., 2019).

Further recent work on other descending pathways and recovery of movement after CNS damage have helped to illuminate their roles in movement execution. While it has been known for some time that the rubro-and-reticulo-spinal tracts are involved in movement and recovery after certain injuries, recent studies in rats have helped to clarify the roles of cortico-rubral (Ishida et al., 2016, 2019) and cortico-reticular (Ishida et al., 2019) projections in recovery after injury. In the first study (Ishida et al., 2016), rats with a unilateral internal capsule stroke were forced to use their affected limb to retrieve food pellets and perform a ladder walk. Additional arborization of the cortico-rubral tract into the red nucleus was observed in these rats compared to rats that did not receive forced limb use treatment. Inactivation of these neurons, which could include surviving CSNs that project collaterals to the red nucleus, resulted in reduced successful task performance. This deficit was eventually compensated for by other descending pathways, presumably. Subsequently, the authors demonstrated that the cortico-rubral and cortico-reticular pathways were recruited sequentially in recovery. Again, if cortico-rubral activity was blocked after forced limb use rehabilitation, reaching was initially disrupted. After the rats recovered during cortico-rubral inhibition, subsequent inhibition of the cortico-reticular path also disrupted reaching. Together, these studies implicate these cortex to brainstem paths play important roles in recovery, which is complementary to their known involvement in forelimb use in the intact animal.

Conclusion

The emerging evidence points toward many parallel, but heavily interconnected pathways that contribute to dexterous forelimb movement in the rodent. Even careful manipulation of single pathways could impact activity across other pathways. Thus, it is necessary to maintain a holistic view of the ascending and descending pathways that could impact movement control. Movement is likely initiated through thalamic input to motor cortex, which in turn outputs particular temporally-patterned activity to several locations distributed throughout the entire CNS, down to the appropriate spinal level. Furthermore, interactions between descending circuits and separate populations of spinal interneurons are necessary for the proper execution of sequential movements such as grasping and releasing objects at appropriate times. Finally, task dexterity, defined by the degree of sensory integration necessary to accomplish the task, is a critical factor to consider when designing studies of cortical control of movement.

Highlights.

  • The interconnectedness and activity of the neural circuits involved in dexterous movement in the rodent is reviewed.

  • Recent experiments that shed light on the time course of neural activity along the motor control axis are summarized.

  • Task dexterity is proposed as a variable that could reconcile some recent experiments with seemingly contradictory results.

  • The degree of sensory processing in a task is likely a factor in the recruitment of the rodent corticospinal circuit.

Acknowledgements

This review was supported by grants from the NINDS (NS093002, NS100772, and NS115963).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Alstermark B, Isa T. Circuits for skilled reaching and grasping. Annu Rev Neurosci. 2012;35:559–78. doi: 10.1146/annurev-neuro-062111-150527. [DOI] [PubMed] [Google Scholar]
  2. Anderson CT, Sheets PL, Kiritani T, Shepherd GM. Sublayer-specific microcircuits of corticospinal and corticostriatal neurons in motor cortex. Nat Neurosci. 2010. June;13(6):739–44. doi: 10.1038/nn.2538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Armstrong DM, Drew T. Electromyographic responses evoked in muscles of the forelimb by intracortical stimulation in the cat. J Physiol. 1985a. October;367:309–26. doi: 10.1113/jphysiol.1985.sp015826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Armstrong DM, Drew T. Forelimb electromyographic responses to motor cortex stimulation during locomotion in the cat. J Physiol. 1985b. October;367:327–51. doi: 10.1113/jphysiol.1985.sp015827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baker SN. The primate reticulospinal tract, hand function and functional recovery. J Physiol. 2011. December 1;589(Pt 23):5603–12. doi: 10.1113/jphysiol.2011.215160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Biane JS, Takashima Y, Scanziani M, Conner JM, Tuszynski MH. Reorganization of Recurrent Layer 5 Corticospinal Networks Following Adult Motor Training. J Neurosci. 2019;39(24):4684–4693. doi: 10.1523/JNEUROSCI.3442-17.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bortoff GA, Strick PL. Corticospinal terminations in two new-world primates: further evidence that corticomotoneuronal connections provide part of the neural substrate for manual dexterity. J Neurosci. 1993. December;13(12):5105–18. doi: 10.1523/JNEUROSCI.13-12-05105.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bretzner F, Drew T. Contribution of the motor cortex to the structure and the timing of hindlimb locomotion in the cat: a microstimulation study. J Neurophysiol. 2005. July;94(1):657–72. doi: 10.1152/jn.01245.2004. [DOI] [PubMed] [Google Scholar]
  9. Brown AR, Teskey GC. Motor cortex is functionally organized as a set of spatially distinct representations for complex movements. J Neurosci. 2014. October 8;34(41):13574–85. doi: 10.1523/JNEUROSCI.2500-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Churchland MM, Cunningham JP, Kaufman MT, Foster JD, Nuyujukian P, Ryu SI, Shenoy KV. Neural population dynamics during reaching. Nature. 2012. July 5;487(7405):51–6. doi: 10.1038/nature11129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dombeck DA, Graziano MS, Tank DW. Functional clustering of neurons in motor cortex determined by cellular resolution imaging in awake behaving mice. J Neurosci. 2009. November 4;29(44):13751–60. doi: 10.1523/JNEUROSCI.2985-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Drew T Motor cortical activity during voluntary gait modifications in the cat. I. Cells related to the forelimbs. J Neurophysiol. 1993. July;70(1):179–99. doi: 10.1152/jn.1993.70.1.179. [DOI] [PubMed] [Google Scholar]
  13. Drew T, Jiang W, Kably B, Lavoie S. Role of the motor cortex in the control of visually triggered gait modifications. Can J Physiol Pharmacol. 1996. April;74(4):426–42. doi: 10.1139/y96-043. [DOI] [PubMed] [Google Scholar]
  14. Evarts EV. Pyramidal tract activity associated with a conditioned hand movement in the monkey. J Neurophysiol. 1966. November;29(6):1011–27. [DOI] [PubMed] [Google Scholar]
  15. Galiñanes GL, Bonardi C, Huber D. Directional Reaching for Water as a Cortex-Dependent Behavioral Framework for Mice. Cell Rep. 2018. March 6;22(10):2767–2783. doi: 10.1016/j.celrep.2018.02.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Graziano MS, Aflalo TN, Cooke DF. Arm movements evoked by electrical stimulation in the motor cortex of monkeys. J Neurophysiol. 2005. December;94(6):4209–23. doi: 10.1152/jn.01303.2004. [DOI] [PubMed] [Google Scholar]
  17. Graziano MS. Ethological Action Maps: A Paradigm Shift for the Motor Cortex. Trends Cogn Sci. 2016. Feb;20(2):121–32. doi: 10.1016/j.tics.2015.10.008. [DOI] [PubMed] [Google Scholar]
  18. Guo K, Yamawaki N, Barrett JM, Tapies M, Shepherd GMG. Cortico-Thalamo-Cortical Circuits of Mouse Forelimb S1 Are Organized Primarily as Recurrent Loops. J Neurosci. 2020. April 1;40(14):2849–2858. doi: 10.1523/JNEUROSCI.2277-19.2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Harrison TC, Ayling OG, Murphy TH. Distinct cortical circuit mechanisms for complex forelimb movement and motor map topography. Neuron. 2012. April 26;74(2):397–409. doi: 10.1016/j.neuron.2012.02.028. [DOI] [PubMed] [Google Scholar]
  20. Heindorf M, Arber S, Keller GB. Mouse Motor Cortex Coordinates the Behavioral Response to Unpredicted Sensory Feedback. Neuron. 2018. September 5;99(5):1040–1054.e5. doi: 10.1016/j.neuron.2018.07.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hira R, Ohkubo F, Ozawa K, Isomura Y, Kitamura K, Kano M, Kasai H, Matsuzaki M. Spatiotemporal dynamics of functional clusters of neurons in the mouse motor cortex during a voluntary movement. J Neurosci. 2013a. January 23;33(4):1377–90. doi: 10.1523/JNEUROSCI.2550-12.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hira R, Ohkubo F, Tanaka YR, Masamizu Y, Augustine GJ, Kasai H, Matsuzaki M. In vivo optogenetic tracing of functional corticocortical connections between motor forelimb areas. Front Neural Circuits. 2013b. April 1;7:55. doi: 10.3389/fncir.2013.00055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hooks BM, Mao T, Gutnisky DA, Yamawaki N, Svoboda K, Shepherd GM. Organization of cortical and thalamic input to pyramidal neurons in mouse motor cortex. J Neurosci. 2013. January 9;33(2):748–60. doi: 10.1523/JNEUROSCI.4338-12.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Isa T, Mitsuhashi M, Yamaguchi R. Alternative routes for recovery of hand functions after corticospinal tract injury in primates and rodents. Curr Opin Neurol. 2019. December;32(6):836–843. doi: 10.1097/WCO.0000000000000749. [DOI] [PubMed] [Google Scholar]
  25. Ishida A, Isa K, Umeda T, Kobayashi K, Kobayashi K, Hida H, Isa T. Causal Link between the Cortico-Rubral Pathway and Functional Recovery through Forced Impaired Limb Use in Rats with Stroke. J Neurosci. 2016. January 13;36(2):455–67. doi: 10.1523/JNEUROSCI.2399-15.2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ishida A, Kobayashi K, Ueda Y, Shimizu T, Tajiri N, Isa T, Hida H. Dynamic Interaction between Cortico-Brainstem Pathways during Training-Induced Recovery in Stroke Model Rats. J Neurosci. 2019. September 11;39(37):7306–7320. doi: 10.1523/JNEUROSCI.0649-19.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kargo WJ, Nitz DA. Improvements in the signal-to-noise ratio of motor cortex cells distinguish early versus late phases of motor skill learning. J Neurosci. 2004. June 16;24(24):5560–9. doi: 10.1523/JNEUROSCI.0562-04.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kawai R, Markman T, Poddar R, Ko R, Fantana AL, Dhawale AK, Kampff AR, Ölveczky BP. Motor cortex is required for learning but not for executing a motor skill. Neuron. 2015. May 6;86(3):800–12. doi: 10.1016/j.neuron.2015.03.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Keizer K, Kuypers HG. Distribution of corticospinal neurons with collaterals to the lower brain stem reticular formation in monkey (Macaca fascicularis). Exp Brain Res. 1989;74(2):311–8. doi: 10.1007/bf00248864. [DOI] [PubMed] [Google Scholar]
  30. Lawrence DG, Kuypers HG. The functional organization of the motor system in the monkey. I. The effects of bilateral pyramidal lesions. Brain. 1968a. March;91(1):1–14. doi: 10.1093/brain/91.1.1. [DOI] [PubMed] [Google Scholar]
  31. Lawrence DG, Kuypers HG. The functional organization of the motor system in the monkey. II. The effects of lesions of the descending brain-stem pathways. Brain. 1968b. March;91(1):15–36. doi: 10.1093/brain/91.1.15. [DOI] [PubMed] [Google Scholar]
  32. Lemon RN. Descending pathways in motor control. Annu Rev Neurosci. 2008;31:195–218. doi: 10.1146/annurev.neuro.31.060407.125547. [DOI] [PubMed] [Google Scholar]
  33. Lemon R Recent advances in our understanding of the primate corticospinal system. F1000Res. 2019. March 11;8 pii: F1000 Faculty Rev-274. doi: 10.12688/f1000research.17445.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Masamizu Y, Tanaka YR, Tanaka YH, Hira R, Ohkubo F, Kitamura K, Isomura Y, Okada T, Matsuzaki M. Two distinct layer-specific dynamics of cortical ensembles during learning of a motor task. Nat Neurosci. 2014. July;17(7):987–94. doi: 10.1038/nn.3739. [DOI] [PubMed] [Google Scholar]
  35. McKenna JE, Prusky GT, Whishaw IQ. Cervical motoneuron topography reflects the proximodistal organization of muscles and movements of the rat forelimb: A retrograde carbocyanine dye analysis. J Comp Neurol. 2000. April 10;419(3):286–96. doi: [DOI] [PubMed] [Google Scholar]
  36. Miri A, Azim E, Jessell TM. Edging toward entelechy in motor control. Neuron. 2013. October 30;80(3):827–34. doi: 10.1016/j.neuron.2013.10.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Miri A, Warriner CL, Seely JS, Elsayed GF, Cunningham JP, Churchland MM, Jessell TM. Behaviorally Selective Engagement of Short-Latency Effector Pathways by Motor Cortex. Neuron. 2017. August 2;95(3):683–696.e11. doi: 10.1016/j.neuron.2017.06.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mohammed H, Hollis ER 2nd. Cortical Reorganization of Sensorimotor Systems and the Role of Intracortical Circuits After Spinal Cord Injury. Neurotherapeutics. 2018. July;15(3):588–603. doi: 10.1007/s13311-018-0638-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Moreno-López Y, Pérez-Sánchez J, Martínez-Lorenzana G, Condés-Lara M, Rojas-Piloni G. Cortical presynaptic control of dorsal horn C-afferents in the rat. PLoS One. 2013. July 30;8(7):e69063. doi: 10.1371/journal.pone.0069063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Moreno-López Y, Olivares-Moreno R, Cordero-Erausquin M, Rojas-Piloni G. Sensorimotor Integration by Corticospinal System. Front Neuroanat. 2016. March 9;10:24. doi: 10.3389/fnana.2016.00024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Muir RB, Lemon RN. Corticospinal neurons with a special role in precision grip. Brain Res. 1983. February 21;261(2):312–6. doi: 10.1016/0006-8993(83)90635-2. [DOI] [PubMed] [Google Scholar]
  42. Olivares-Moreno R, Moreno-Lopez Y, Concha L, Martínez-Lorenzana G, Condés-Lara M, Cordero-Erausquin M, Rojas-Piloni G. The rat corticospinal system is functionally and anatomically segregated. Brain Struct Funct. 2017. December;222(9):3945–3958. doi: 10.1007/s00429-017-1447-6. [DOI] [PubMed] [Google Scholar]
  43. Olivares-Moreno R, López-Hidalgo M, Altamirano-Espinoza A, González-Gallardo A, Antaramian A, Lopez-Virgen V, Rojas-Piloni G. Mouse corticospinal system comprises different functional neuronal ensembles depending on their hodology. BMC Neurosci. 2019. Sep 23;20(1):50. doi: 10.1186/s12868-019-0533-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Oswald MJ, Tantirigama ML, Sonntag I, Hughes SM, Empson RM. Diversity of layer 5 projection neurons in the mouse motor cortex. Front Cell Neurosci. 2013. October 16;7:174. doi: 10.3389/fncel.2013.00174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Otchy TM, Wolff SB, Rhee JY, Pehlevan C, Kawai R, Kempf A, Gobes SM, Ölveczky BP. Acute off-target effects of neural circuit manipulations. Nature. 2015. December 17;528(7582):358–63. doi: 10.1038/nature16442. [DOI] [PubMed] [Google Scholar]
  46. Peters AJ, Chen SX, Komiyama T. Emergence of reproducible spatiotemporal activity during motor learning. Nature. 2014. June 12;510(7504):263–7. doi: 10.1038/nature13235. [DOI] [PubMed] [Google Scholar]
  47. Peters AJ, Lee J, Hedrick NG, O’Neil K, Komiyama T. Reorganization of corticospinal output during motor learning. Nat Neurosci. 2017. August;20(8):1133–1141. doi: 10.1038/nn.4596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ramanathan D, Conner JM, Tuszynski MH. A form of motor cortical plasticity that correlates with recovery of function after brain injury. Proc Natl Acad Sci U S A. 2006. July 25;103(30):11370–5. doi: 10.1073/pnas.0601065103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Rizzi G, Coban M, Tan KR. Excitatory rubral cells encode the acquisition of novel complex motor tasks. Nat Commun. 2019. May 21;10(1):2241. doi: 10.1038/s41467-019-10223-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sauerbrei BA, Guo JZ, Cohen JD, Mischiati M, Guo W, Kabra M, Verma N, Mensh B, Branson K, Hantman AW. Cortical pattern generation during dexterous movement is input-driven. Nature. 2020. January;577(7790):386–391. doi: 10.1038/s41586-019-1869-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Shepherd GMG. Corticostriatal connectivity and its role in disease. Nat Rev Neurosci. 2013. April;14(4):278–91. doi: 10.1038/nrn3469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Suter BA, Shepherd GM. Reciprocal interareal connections to corticospinal neurons in mouse M1 and S2. J Neurosci. 2015. February 18;35(7):2959–74. doi: 10.1523/JNEUROSCI.4287-14.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Ueno M, Nakamura Y, Li J, Gu Z, Niehaus J, Maezawa M, Crone SA, Goulding M, Baccei ML, Yoshida Y. Corticospinal Circuits from the Sensory and Motor Cortices Differentially Regulate Skilled Movements through Distinct Spinal Interneurons. Cell Rep. 2018. May 1;23(5):1286–1300.e7. doi: 10.1016/j.celrep.2018.03.137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wang X, Liu Y, Li X, Zhang Z, Yang H, Zhang Y, Williams PR, Alwahab NSA, Kapur K, Yu B, Zhang Y, Chen M, Ding H, Gerfen CR, Wang KH, He Z. Deconstruction of Corticospinal Circuits for Goal-Directed Motor Skills. Cell. 2017. September 20 pii: S0092–8674(17)30939-X. doi: 10.1016/j.cell.2017.08.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wang Z, Maunze B, Wang Y, Tsoulfas P, Blackmore MG. Global Connectivity and Function of Descending Spinal Input Revealed by 3D Microscopy and Retrograde Transduction. J Neurosci. 2018. December 5;38(49):10566–10581. doi: 10.1523/JNEUROSCI.1196-18.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Weiler N, Wood L, Yu J, Solla SA, Shepherd GM. Top-down laminar organization of the excitatory network in motor cortex. Nat Neurosci. 2008. March;11(3):360–6. doi: 10.1038/nn2049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Whishaw IQ, Pellis SM, Gorny BP, Pellis VC. The impairments in reaching and the movements of compensation in rats with motor cortex lesions: an endpoint, videorecording, and movement notation analysis. Behav Brain Res. 1991. January 31;42(1):77–91. doi: 10.1016/s0166-4328(05)80042-7. [DOI] [PubMed] [Google Scholar]
  58. Widajewicz W, Kably B, Drew T. Motor cortical activity during voluntary gait modifications in the cat. II. Cells related to the hindlimbs. J Neurophysiol. 1994. November;72(5):2070–89. doi: 10.1152/jn.1994.72.5.2070. [DOI] [PubMed] [Google Scholar]
  59. Xu T, Yu X, Perlik AJ, Tobin WF, Zweig JA, Tennant K, Jones T, Zuo Y. Rapid formation and selective stabilization of synapses for enduring motor memories. Nature. 2009. December 17;462(7275):915–9. doi: 10.1038/nature08389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Yoshida Y, Isa T. Neural and genetic basis of dexterous hand movements. Curr Opin Neurobiol. 2018. October;52:25–32. doi: 10.1016/j.conb.2018.04.005. [DOI] [PubMed] [Google Scholar]

RESOURCES